PB82-2491?
Usefulness of the Self-Fertilizing
Cyprinodontid Pish, "Rivulus marmoratus" as an
Experimental Animal in Studies Involving
Carcinogenesis, Teratogenesis and Mutagenesis
Charleston Coll., SC
Prepared for
Environmental Research Lab.
Gulf Breeze, PL
Jul 82
taraw
8J, UerrartmcBt cf Cwwiwrca
Slataeal TooVicci kftmtK&n Scrt'.ce
Ldv*J m
-------
EPA 600/3-82-075
July 1982
pb02-2«yiy«
USEFULNESS OF THE SELF-FERTILIZING CYPRINODONTID FISH,
RIVULUS MARMQRATUS AS AN EXPERIMENTAL ANIMAL IN STUDIES INVOLVING
CARCINOGENESIS, TERATOSENESIS AND MUTAGENESIS
by
Christopher C. Koenlg (P.I.)
Daniel C. Abel
Courtney W. Klingensmith
Michael B. Maddock
Grice Marine Biological laboratory
College of Charleston
Charleston, SC 29412
Grant Mo. R805469 01
Project Officer
William P. Davis
Bears Bluff Field Station
Environmental Research Laboratory
U. S. Environmental Protection Agency
Gulf Breeze, Florida 32561
ENVIRONMENTAL RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U. S. ENVIRONMENTAL PROTECTION AGENCY
GULF BREEZE, FLORIDA 32561
-------
NOTICE
Mention of trade names or commercial products does not
constitute endorsement or recommendation for use.
COLOR ILLUSTRATIONS REPRODUCE'.:
IN BLACK AMD WHITE
-------
DISCLAIMER
Although the research described In this report has been funded wholly
or in part by the United States Environmental Protection Agency under
Grant agreement number R805469 to Christopher C. Koenig, Grice Marine
Biological Laboratory College of Charleston, Charleston, SC, it has not
been subjected to the Agency's required peer and administrative review
and, therefore, does not necessarily reflect the view of the Agency and no
official endorsement should be inferred.
11
-------
FfREWORD
The protection of our estuarine and coastal areas from damage caused
by toxic organic pollutants required that regulations restricting the
introduction of these compounds Into the environment be formulated on a
sound scientific basis. Accurate infc-nation describing dose-response
relationships for organisms and ecosystems under varying conditions is
required. The EPA Environmental Research Laboratory, Gulf Breeze (ERL.GB),
contributes to this information through research prjgraris aimed at
determining:
the effects of toxic organic pollutants on individual species and
communities of organisms;
the effects of toxic organics on ecosystem processes and
components; *
the significance of chemical carcinogens in the estuarine and
marine environnent*
Use of fishes as models for carcinogenic responses is an important
step in the continued assessment of environmental contaminants. In this
study are important data on a fish species particularly suited for
laboratory screening of knovm and suspected agents. Roth methodology for
culture of the fish and responses to selected agents are reported. This
Information is presently being considered in on-going governmental programs
dealing with broader questions of determination and evaluation of
significance of agents in aquatic ecosystems.
Henry F. Enos
Director
Environmental Research Laboratory
Gulf Breeze, Florida
-------
ABSTRACT
Rival us marmoratus is a naturally self-fertilizing cyprinodontid fish
inhabiting mangrove marshes throughout the Caribbean. As a result of inter-
nal self-fertilization this oviparous species is composed of a number of
isogenic, homozygous lines (clones)» several of which have been identified
by histocompatibility experiments and maintained in laboratory culture for
over 30 years.
Simplified culture and handling methods are given and data are pre-
sented on the reproduction, growth and development of rivulus under
laboratory culture as a prelude to the evaluation of ics potential as a
bioassay animal. Several types of bioassays were run and evaluated using
rivulus: behavioral, carcinogenicity, teratogenicity, toxicity, and muta-
genicity. Advantages and disadvantages of using rivulus for such bioassays
are discussed. Behaviorally, rivulus is capable ,of detecting and avoiding
water contaminated with H^S. They respond (EC50 <* 123.6 ppb H^S) by "leaping
from the water and remaining emergent for various periods of time while
respiring cutaneously. Hepatocellular carcinoma among other pathologic
changes were observed in livers of rivulus a year after exposure of adults
and larvae to diethylnitrosamine {45, 30 and 15 ppm in water) for 5 weeks
and 12 weeks, respectively. No pathologic changes were found in embryos
exposed similarly. High rates of various skeletal malformations resulted in
offspring of adults exposed to dibutyl phthalate (D3P) and 2,3,4,6-tetra-
chlorophenol (TECP) at concentrations of 20, 10 and 5% (UBP - 0.740, 0.370,
0.185 mg/s.; TECP - 0.220, 0.110, 0.055 mgft) of the larval 96-hr LC50. No
dose-response relationships of skeletal malformations were found for similar
exposures to pentachlorophenol, 2,3,5-trichlorophenol o1* bromoform, How-
ever, chronic exposure of developing hatch!ings to TECP resulted in fin and
gill erosion and chronic exposure to bromoform produced dorsal fin abnormali-
ties. As part of a mutagenesis bioassay 14 enzyme systems representing
28 loci were screened for the three laboratory clones and one wild-caught
clone but no electrophoretic differences were found. Attempts to culture
rivulus cells failed. Also, the karyotype of rivulus is not suitable for
short-term cytogenetic assays such as the sister chromatid exchange (SCE)
assay. Alternatively, however, the toadfish (Onsanus tau) possesses a suit-
able karyotype for SCE analysis and we have been successful in culturing
toadfish cells to fourth passage. Increased rates of SCE were obtained when
toadfish cells were exposed in vitro to the mutagen ethyl methanesulfonate
but not bromoform. Another set of experiments is presented which involves
characterization of the nature of the toadfish cytochrome P450 system.
iv
-------
CONTENTS
Foreward i 1 i
Abstract L iv
/
Figures vii
Tables ix
Acknowledgments xi
1. General Introduction ............ 1,
2. Conclusions and Recommendations . 4
3. Laboratory Culture of Rivalus marmoratus .......... 6
4. Reproduction, Growth and Development of Rivulus marmoratus
under Laboratory Culture . . ......... 9
General observations . 9
Fecundity ... 9
Growth 11
Embryonic development 14
Comparison of egg incubation techniques ....... 23
5. Behavioral bioassay using Rivulus marmoratus . . 26
Introduction ...» 26
Materials and methods .... 27
Results ........ . 28
Discussion 28
6. Carcinogenesis assay using Rivulus marmoratus ... 33
Introduction 33
Materials and methods . 33
Results 35
Discussion . . ....... 46
v
-------
CONTENTS (con't)
7, Teratogenicity Bioassay and Chronic Full Life Cycle
Exposure of Rivulus marmoratus to Sublethal Levels of
Selected Toxicants 50
Intieduction ..... 50
Materials and methods ..... 51
Results ...... 53
Discussion ................ . 87
8. Mutagenesis Bioassay: Investigations with Rivulus marmoratus
and other Selected Fish Species . , 100
' Introduction 100
Materials and methods 104
Results ........ 106
Discussion ....... » , . 116
References 121
\
vi
-------
FIGURES
Number Page
1 Reproduction chambers for Rivulus marmoratus ..... 8
2 Rivulus marmoratus with egg cluster 10
3 Growth of Rivulus marmoratus under laboratory conditions .... 13
4 Photographs of early embryonic stages of Rivulus marmoratus . . is
5 Photographs of late embryonic stages of Rivulus marmoratus ... 21
6 Log-probit plot of emergence response of Rivulus marmoratus
vs HgS concentration . 29
7 Photonicrographs of histological sections of the livers of
Rivulus marmoratus ....... ......... 37
8 Photomicrographs of histological sections of the livers of
Rivulus marmoratus . . .......... 39
9 Photomicrographs of histological sections of the livers of
Rivulus marmoratus 41
10 Photomicrographs of histological sections of the livers of
Rivulus marmoratus ... 43
11 Photomicrographs of histological sections of the livers of
Rivulus marmoratus ... 45
12 Late stage embryo of Rivulus marmoratus showing edematous
pericardial cavity 63
13 Photomicrographs of caudal skeleton of cleared and stained
Rivulus marmoratus 76
14 Photomicrographs of caudal vertebrae of cleared and stained
Rivulus marmoratus , 73
15 Photomicrographs of trunk vertebrae of cleared and stained
Rivulus marmoratus . ..... 80
16 Photomicrographs of cleared and stained Rivulus marmoratus . . . 82
Vi1
-------
Number Page
17 Photomicrographs of cleared and stained frivolus marmoratus . , 84
18 Photomicrographs of pectoral girder of cleared and stained
Rivulus marmoratus 86
19 Photographs of cleared and stained Rivulus marmoratus .... 89
2(1 Photomicrographs of a cross section through the gill arches
°f Rivulus marmoratus 91
21 Histogram of percent of various skeletal abnormalities in
offspring of Rivulus marmoratus exposed to DBP 93
22 Histogram of percent of various skeletal abnormalities in
offspring of Rivulus marmoratus exposed to DBP ...... 95
23 Metaphase chromosomes of various fish species 108
24 Metaphase chromosomes of various fish species no
25 Metaphase chromosomes of toadfish leukocytes showing sister
chromatid exchanges; also, photomicrographs of monolayers
of toadfish ovary cells ....... 113
26 Effects of pretreatment of fish and rats with MC and AC on the
activation of BAP and AA to Salmonella mutagens ns
viii
-------
TABLES
Number Page
1 Growth of Rivulus marmoratus in three container sizes 12
2 Summary of Rivulus marmoratus developmental stages 16
3 Comparison of rates of embryonic development of Rivulus
Marmoratus in air incubation vs water incubation ...... 25
4 Total time emergent for Rivulus marmoratus relative to H2S
concentration 30
5 Pathologic changes in livers of Rivulus marmoratus larvae
exposed to DEN for 12 weeks 47
6 Pathologic changes in livers of Rivulus marmoratus adults
exposed to DEN for 5 weeks ; 48
7 Static 96-hr LC50 data for postlarvae of Funrfulus
heteroclitus and prescribed concentrations for chronic
study with Rivulus marmoratus ............... 54
8 Static 96-hr LC50 data for postlarvae of Rivulus marmoratus . . 55
9 Summary of mortality and skeletal abnormality data for
Rivulus marmoratus used as controls for chronic exposure
study 56
10 Summary of mortality and skeletal abnormility data for
Rivulus marmoratus from chronic exposure to PCP 57
11 Summary of mortality and skeletal abnormality data for
Rivulus marmoratus from chronic exposure to TECP 58
12 Summary of mortality and skeletal abnormality data for
Rivulus marmoratus from chronic exposure to TRCP ...... 59
13 Summary of mortality and skeletal abnormality data for
Rivulus marmoratus for chronic exposure to DBP 60
14 Summary of mortality and skeletal abnormality data for
Rivulus marmoratus from chronic exposure to bromoform ... 61
ix
-------
Number Page
15 Distribution of heart abnormalities in embryos of Rivulus
maraioratus exposed chronically to various toxicants .... 64
16 Distribution of skeletal abnormalities of offspring of
Rivulus marmoratus exposed to PCP 65
17 Distribution of skeletal abnormalities of offspring of
Rivul.js marmoratus; parental adults only were exposed
to PCF" 66
18 Distribution of skeletal abnormalities of offspring of
Rivulus marmoratus exposed to TECP 67
19 Distribution of skeletal abnormalities of offspring of
Rivulus marmoratus; parental adults only were
exposed to TECP 68
20 Distribution of skeletal abnormalities of offspring of
Rivulus marmoratus' exposed to TRCP 69
21 Distribution of skeletal abnormalities of offspring of
Rivulus marmoratus; parental adults only were
exposed to TRCP 70
22 Distribution of skeletal abnormalities of offspring of
Rivulus marmoratus exposed to DBP 71
23 Distribution of skeletal abnormalities of offspring of
Rivulus marmoratus; parental adults only were exposed
to DBP . 72
24 Distribution of skeletal abnormalities of offspring of
Rivulus marmoratus exposed to bromoform . . » 73
25 Distribution of skeletal abnormalities of offspring of
Rivulus marmoratus; parental adults only were '
exposed to bromoform 74
26 Screening of clonal and wild-caught Rivulus marmoratus
for biochemical variation 102
27 Sister chromatid exchange rates following in vitro
exposure of dividing toadfish leukocytes to EMS and
bromoform ' 115
28 Culture of tissues from Rivulus marmoratus and
Opsanus tau ng
x
-------
ACKNOWLEDGEMENTS
We thank many people, students and colleagues, for contributions in
time and effort towards the completion of this project. Drs. J. C.
Harshbarger (Registry of Tumors, Smithsonian Institution), A. B. Garvin
(Medical University of South Carolina), and J. A, Couch (EPA, Gulf Breeze,
FL) for analysis of histological sections, Drs. N, A. Chamberlain (Grice
ferine Biological Laboratory, Charleston, SC) and W. P. Davis (EPA, Gulf
Breeze, FL) for good advice during all phases of this study, Dr. M. P.
Weinstein, Mr. C. Courtney and Mr. B. Yokel for help in locating wild
populations of rivulus* Dr. T. C. Lewis and Ms. F, C. Coleman for assistance
in photography and printing, Ms. C. McLean and Mr. M. P. Chasar for help in
maintaining the rivulus stocks and carrying out various experiments,
Mr. C. R. Cripe for donating an electronic device to the behaviort.1 study,
Hs. J. M. Van Oolah, Ms. H. N. Patrick, Ms. J. J. Kelly, and Mr. D. M.
Milling for assistance in the mutagenesis study. Dr. J. Bishop for providing
toadfish for the mutagenesis study and Mrs. J. S. Koenig for expertly
typing a beautiful manuscript. Special thanks are given to Mrs. Eleanor
Harrington for donating the laboratory clones of rivulus and for encourage-
ment during all phases of this study. Among the authors, Dr. H. Maddock
was responsible for the mutagenesis work, Ms. C. Klingensmith for embryonic
development of rivulus, and Mr. D. C. Abel for the behavioral bioassay.
xi
-------
SECTION 1
GENERAL INTRODUCTION
The concept of a relationship between cancer and environmental
contaminants has become prominent in tie minds of many medical and
environmental researchers. The enormous array of synthetic organic chemicals
in the environment, many of which are quite persistent, and are expected t3
or have been shown to bioaccumulate, constitute a potential environmental
hazard. The assessment of this potential hazard is, however, very
difficult. A conference sponsored by the New York Academy of Sciences
(Kraybill et al.> 1977) relating pollutants and neoplasia in aquatic animals
pointed out tEe considerable confusion that exists, not only in the
assessment of the potential hazards but even in defining the potential
effects.
It fs important to recognize that the term cancer encompasses a group
of diseases caused by a multitude of intrinsic and extrinsic factors acting
singly or in combination. Carcinogenesis was defined by Miller (1970)
as an heritable and at least quasipermanent loss of control of cell
multiplication. Several mechanisms have been suggested to account for
such t change in cell behavior: 1) interaction of carcinogens with DMA
resulting in genie or chromosomal mutation, 2) alteration of specific
protein or RNA to produce heritable changes in genome expression, 3) the
activation of latent carcinogenic virus genome and 4) the selection of
preneoplastic cells by conditions that favor the multiplication and
survival of these cells. The latter two probably constitute indirect
mechanisms operating through the first two (Fishbein, 1975).
The assessment of genetic damage then, is an important aspect of the
assessment of potential carcinogenicity. Indeed, a number of screening
tests for chemical carcinogens are dependent upon the correlation between
carcinogenicity and mutagenicity, Fishbein (1975) states that it appears
that many, if not all, chemical carcinogens are potential mutagens and
probably most mutagens are potential carcinogens. This suggestion is
credible because carcinogenesis is thought to be a multistep process with
genetic alteration as the initial step. The relatively long induction
periods for carcinogenesis may then be dependent upon the remaining steps
in this process which in turn are related to a number of extrinisic and
Intrinsic factors.
Alterations in. the genetic material may also cause teratogenic effects
which could be mediated either by way of the parental adults or as a
direct effect on the developing embryos. Although teratogenesis may be
1
-------
induced by many agents other than chemical mutagens (e.g., physical trsuna,
dietary deficiency, etc.) the relationship to mutagenesis and therefore
carcinogenesis is evident.
Teleostean fishes show great promise as experimental animals in
mutation research (Schroder, 1973). Their potential value in cancer
research was clearly brought out at the conference on "Aquatic Pollutants and
Biological Effects with Emphasis on Neoplasia" (Kraybill et al_., 1977) and
the "Symposium on the Use of Small Fish Species in Carcinogenicity Testing"
(sponsored by NCI and EPA, December, 1981). A number of papers given at
these conferences also pointed out the high incidence of neoplasia in
certain marine and freshwater fishes associated with areas of high industrial
pollution. The consensus was the some fishes appear quite sensitive to
tumorogenic agents and that an urgent need, voiced by Dr. M, F. Stanton
(National Cancer Institute, Bethesda, Maryland), exists for a species
"supersensitive to carcinogens."
Such a need is readily recognized and a possible candidate, which is the
focus of this report, is the oviparous cyprinodontid fish, Rivulus
marmoratus. It is the only fish known that exhibits natural syncronous
hermaphroditism with internal self-fertilization (Harrington, 1971). This
ultimate mode of inbreeding apparently has rendered these fishes homozygous
and three isogenic clones (designated NA, DS, and M) have been identified
by intraclcnal, interclonal and interclonal hybrid offspring tissue trans-
plants (tollman and Harrington, 1964; Harrington and tollman, 1968). These
clones were carefully maintained in the laboratory of Dr. Robert W.
Harrington, Jr. for nearly 20 years until his untimely death in 1975. The
majority of the biological information known about this remarkable species
is due to the meticulous efforts of Dr. Harrington (Harrington and Rivas,
1958; Harrington, 1961; Harrington, 1963; tollman and Harrington, 1964;
Harrington, 1967; Harrington and tollman, 1968; Harrington, 1968; 1971;
Llndsey and Harrington, 1972; Harrington, 1975; Massaro and Harrington, 1975;
Harrington and Grossman, 1976). A complete pedigree was kept for each
individual fish in the collection.
E* marmoratus is small (20-50 urn at maturity), easily maintained in
the laboratory, has wide tolerances of salinity and temperature and oviposits
up to ten fertilized eggs per week (during daylight hours, modally at noon).
Generation time is on the order of 4-6 months. Eggs are large (1.5-2.0 mm
diam) with a transparent chorion (shell). Embryonic development is of
sufficient length (12-15 days at 25°C) to make careful observations on
developmental rate and sequence and embryonic stages are somewhat similar
to those of other cyprinodont fishes, for example, Fundulus heteroclitus
(Armstrong and Child, 1965) and Adinia xenica (Koenig and Livingston, 1976).
The natural habitat of R, marmoratus is shallow estuarine mangrove
marshes. Although the species is widely distributed throughout the
Caribbean, it is somewhat rare in the continental United States.
Males of R. marmoratus were extremely rare in Dr. Harrington's field
collections an? females were never found. Dr. Harrington, however,
demonstrated that males could readily be induced in his laboratory strains.
2
-------
Two types of males exist, primary males, in which ovarian development never
takes place and secondary males, which develop from hermaphrodites
(ovarian degeneration). Primary males may be produced in the laboratory by
submitting the terminal stages of embryonic development to low temperatures.
Secondary males may result frcnt hermaphrodites spontaneously or from
exposure of hermaphrodites to short day length.
It is obvious from the preceding discussion that R, marmoratus
possesses numerous attributes that make it particularly attractive as an
experimental animal for carinogenicity, teratogenicity and mutagenicity
studies. Its hardiness, ease of maintenance, developmental aspects, short
generation time, fecundity and unique genetic aspects coupled with the fact
'that it is an estuarine fish with wide salinity tolerances all serve to
increase its attractiveness, especially in light of recent evidence of
increased incidence of neoplasms and other pathologic changes 1n marine
organisms which may reflect accelerated deterioration of coastal marine
environments.
The purpose of this work is to investigate the potential of JR.
marmoratus as an experimental animal useful in standard bioassay studies
and particularly in the study of carcinogenicity and teratogenicity. Related
objectives include the establishment of simplified culture methods and
basic biological data while fish are held under laboratory conditions.
3
-------
SECTION 2
CONCLUSIONS AND RECOMMENDATIONS
Large numbers of rivulus (and all stages of the life cycle) may be held
efficiently in the laboratory in stackable glass culture dishes. Three
feedings of brine shrimp per week and monthly water changes are sufficient
to ensure health. As with any laboratory animal success in mcintainlng a
healthy colony lies with the culturist and is directly proportional to TLC.
We recommend that raw seawater not be used in the culture of rivulus to
reduce chances of pathogen introduction.
Embryonic development of rivulus is similar to that of other cyprinodon-
tid fishes. However, early development occurs to varying degrees within the
ovotestis. This could cause problems in a study in which observation of
early development (bl&stulation) is required. Also, because eggs are not
fertilized synchronously it is difficult to get eggs in the same stags of
development from a small group of egg-layers. A healthy egg-layer produces
about an egg a day; however, trauma may bring about extended refractory
periods. Problems associated with extended delays in hatching can be
averted by mechanically removing the chorion, chemically changing the incu-
bation medium which is moderately successful in stimulating hatching.
Rivulus avoids dissolved H?S (EC50 = 123.50, ppb HgS) by leaping from
the water and remaining emergent for various periods of time while respiring
cutaneously. Further development of a bioassay based on this quanta!
response is recanmended as it could be useful in water quality management.
Hepatocellular carcinoma, cholangioma, adenofibrosis and granuloma were
induced in the livers of rivulus one year after exposure as adults and larvae
to diethylnitrosamine (DEN) in water concentrations of 45, 30, and 15 ppm for
5 weeks and 12 weeks, respectively. No pathologic changes were observed in
embryos treated similarly. It is recommended that further studies be done
on the susceptibility of rivulus to various carcinogens. In the long-term,
this should be extended to clones more divergent from those used in this
report, laboratory produced or derived from wild-caught fish,from a number of
areas throughout the Caribbean. Such research may produce another inroad
into the relationship between genetics and cancer.
Exposure of reproducing rivulus to the teratogen dibutyl phthalate (DBP)
in water concentrations of 0.740, 0.370 and 0.185 mg/A (20, 10 and 5% of
larval 96-hr LC50) induced a variety of skeletal abnormalities in a dose-
response fashion. A similar but less dramatic effect was induced by 2,3,4,
6-tetrachlorophenol (TECP) but not by pentachlorophenol (PCP)» 2,3,5-tri-
4
-------
chlorophenol (TRCP) or bromoform. Although rivulus appears quite sensitive
to agents which induce skeletal malformations/the control rate (30X) is quite
high. Preliminary work suggests that other clones, derived from recent
wild-caught fish have spontaneous rates much lower than this. Because no
aquatic teratogenicity assay such as this exists we strongly recommend that
it be further developed.
Chronic exposure of rivulus parental adults and offspring through adult
stage to 2,3,4,6-tetrachlerophenol (TECP) resulted in gill and fin erosion
in offspring which showed a dose-response relationship. Exposure levels were
20, 10, and 5X of the larval 96-hr LC50 (0.220, 0.110 and 0.055 mg/t). It
Is not known whether a pathogen is secondarily involved; however, the
mechanisms of fin and gill erosion may be studied using rivulus as a model
as such effects have been noted in several estuaries in the United States.
Chronic exposure of rivulus parental adults and offspring through adult
stage to bromoform resulted in abnormal dorsal fin development in offspring.
Exposure levels were 20, 10, and 5% of the larval 96-hr LC50 (8.4, 4.2 and
2.1 mg/t). Such a highly specific effect was not expected. Further
research is recommended to corroborate these findings and if corroborated to
study the mechanism.
Fourteen enzyme systems representing about 28 loci were screened.
However, no electrophoretic differences were found among the three laboratory
clones or between clones and wild-caught rivulus taken near Naples, Florida.
Rivulus was found unsuitable for sister chromatid exchange (SCE) assay
because chromosomes are small and numerous. The toadfish (Opsanus tau),
however, has a suitable karyotype and in vitro exposure of toadfish
leukocytes to ethyl methanesulfonate, a mutagen, induced increased rates of
SCE.
Culture of rivulus cells has not been successful; however, toadfish cells
have been cultured up to fourth passage.
The effects of hepatic microsomal enzyme (S-9) preparations from rats
and toadfish pretreated with enzyme inducers 3-methyl-cholanthrene (MC) and
arochlor (AC) and untreated on the metabolism of benzo (a) pyrene (BP) and
2-aminoanthracene (AA) to Salmonella mutagens demonstrate the similarities
between these two species in their ability to metabolize promutagenic
xenobiotics.
The toadfish shows promise for these types of research and we recommend
that studies continue. Maddock is presently pursuing this work under
another grant.
5
-------
SECTION 3
LABORATORY CULTURE OF RIVULUS MftRMORATUS
The following culture methods have been found optimum for reproduction,
development and growth of R. marmoratus. These methods apply throughout
this report. Adults are held singly in 4.5 in (11.4 cm) diain stackable
culture dishes filled with 200 ml of 14 o/oo svnthetic seawater prepared by
dissolving approximately 14 g of Instant Ocean* sea salts per liter of
distilled water. Adults are fed frozen brine shrimp (Artemia) three times
a week taking care not to overfeed. A mature adult should not be fed more
than about 9 large brine shrimp per feeding and small adults about 3. All
uneaten food should be removed within 1 hour after feeding. Culture dishes
are washed and seawater is changed monthly. Under a daily feeding regime,
typically used when maximum growth rates are desired, seawater should be
changed twice a month.
Larvae and juveniles are held singly in 2 in (5 cm) diam stackable
culture dishes filled with synthetic 14 o/oo seawater. Young fish less than
3 weeks old may be held in groups; however, older fish become quite
aggressive often resulting in death for most members of a group. Larvae
and juveniles are fed freshly hatched brine shrimp nauplii three times a
week. To prepare the brine shrimp culture 2-6 ml of brine shrimp eggs
(San Francisco Brand) are added to a liter of synthetic 14 o/oo seawater in
a 1 liter glass separatory funnel outfitted with an air stone, hose and
air pump for aeration. After 24 hours of incubation at 25 C the air stone
1s removed for several minutes. The nauplii fall to the bottom and the
egg capsules rise to the surface. Nauplii may then be tapped off in
concentrated form and fed directly to rivulus larvae and juveniles. If
concentrated nauplii are allowed to stand most will die within twenty
minutes due to oxygen depletion. However, live nauplii wi11 remain alive
in rivulus culture dishes for several days providing a constant supply of
food. Rivulus larvae and juveniles should not be fed dead nauplii or
nauplii from a culture much older than 24 hrs.
Juveniles are transferred to 11.4 cm diam culture dishes about 2
months after hatching at which time the diet may be changed from brine shrimp
nauplii to frozen adult brine shrimp or both may be included in the diet
(see section on "Growth").
Eggs are incubated in 5 cm diam stackable culture dishes filled to 1
cm depth with synthetic 14 o/oo seawater. During the final stage of
embryonic development hatching must often be induced otherwise embryos may
delay hatching for considerable periods of time which may result in the
6
-------
death of the embryo. Hatching can be induced by changing daily the
incubation water to either freshly collected seawater or naturally stained
freshwater (see "Embryonic Development" section for discussion of delayed
hatching).
Eggs can be collected from hermaphrodites by holding there singly in
specially designed reproduction chambers (Fig. 1}. These consist of 11.4 cm
culture dishes with TeflonR screens (3-4 mm openings) suspended off the
bottom of each dish by a glass ring ( 1 cm thick, 10 cm diam). Teflon
screens are tied to the glass rings with fine monofilament line which
effectively confines the fish to the vnter space above the screen. Most
eggs accumulate between the walls of the culture dish and the glass ring,
thus preventing rivulus from eating its own eggs. Eggs and any feces or
uneaten food should be pipetted from the reproduction chambers ?-3 times
per week. This can be most efficiently done immediately after feeding. Fish
in reproduction chambers are fed individually by dangling thawed brine shrimp
just in front of them with fine tissue forceps. Six or more can be fed at
a time by proceeding from one to another repetatively until the entire group
is fed to satiation. The optimum temperature and photoperiod for reproduc-
tion as well as all other phases of the life cycle is 25 C and 14 hrs,
respectively.
There are numerous advantages to this culture system for research
purposes that are available from no other fish culture system, marine or
fresh water. The system is inexpensive, easy to maintain, records may be
kept for each individual fish and large stocks may be kept in a relatively
•ismall space. In addition, ail culture materials are made of either Teflon
or glass, which is essential to avoid contamination by such ubiquitous
substances as phthalate ester plasticizers.
7
-------
Figure 1. Photograph of Reproduction Chambers for
Rivillus marmoratus.
&
-------
SECTION 4
REPRODUCTION, GROWTH AND DEVELOPMENT OF RIVULUS MARMORATUS
UNDER LABORATORY CULTURE
GENERAL OBSERVATIONS
The process of internal self-fertilization in R. marmoratus causes
many unusual characteristics. The most readily noticed characteristic is
that eggs are released in various stages of development. Eggs may be
released from the parental fish singly or in clusters held together in a
clump or long chain by sticky filaments (Fig. 2). Usually several eggs in
ta cluster are at different stages of development. The differences in the
stages can range from a few hours to 24 to 48 hours. There is no particular
sequence to the release of eggs; eggs in an early stage may be released prior
to eggs in a more mature stage. As Harrington (1963) noted in his
observations, the eggs are probably released in random order after newly
fertilized eggs become mixed with older developing eggs within the lumen of
the ovotestis. v
Why eggs are held within the parental fish for various periods of time
and what triggers the release of one or several eggs are unknown.
Harrington (1963), ir. his study on the sexual rhythms of rivulus, found that
only It of the eggs observed were emitted within one hour of fertilization
but that 80% were emitted within 24 hours of fertilization. The mean
iintraparental incubation time of all eggs observed was 10 hours. Therefore,
eggs were usually retained within the body of the parental fish until the
late blastula stage.
FECUNDITY
To estimate fecundity of R. marmoratus held singly in reproduction
chambers under the conditions described inthe section "Laboratory
Culture of Rivulus marmoratus" of this report, seven fish were observed
over a four month period. These fish had been acclimated to the conditions
in the reproduction chambers for four months (Nov.-Feb.) prior to the
four months (Mar-June 1979) observation period. All were between 30-35 mm
in length (standard length, SL), healthy proven egg layers. Three belonged
to clone NA, two to clone DS and too to clone M.
Egg production for the seven fish averaged 25.3 eggs per month
(SE = 2,27, N « 28). Clone NA fish averaged 26.8 eggs per month (SE - 3.9,
N = 12), clone DS,20.5 eggs per month (SE = 3.8, N = 8) and clone
9
-------
Vjr'
! J'1 P";
>,V;
AW-^SI
Figure 2. Photograph of Rivulus marooratus with a cluster of eggs.
10
-------
produced 27.9 eggs per month (SE = 3.8, N = 8). No statistical difference
in egg production could be shown between clones (ANOVA; F « 0.9, p >.05).
Overall 22.5% of the eggs laid were non-viable. Most of these were
probably unfertilized although to make that statement with certainty
necessitates collection of the eggs Immediately upon oviposition (see
Harrington, 1963). Harrington (1963) carefully observed 6 fish, 4 large
(40-50 mm SL) and 2 small (25-30 ram SL) over a 4 month period (Aug-Nov).
Eggs were collected immediately upon oviposition. Average egg production was
41.5 eggs per fish per month. However, the cjlture system, food and feeding
regime were different as well as the average size of the fish. Our fish
produced more eggs the longer they were held in the reproduction chambers.
For example, the average egg production in March was 11.3 eggs per fish but
the average production in June was 38.0 eggs per fish. The average egg
production of Harrington's (1963) fish in August was 44.5 eggs per fish but
in November was 30.0 eggs per fish. However, his fish were under a natural
light regime. The effect of season on our fish, held under constant
conditions 1s unknown. As a rule of thumb, a researcher working with
medium sized, healthy adult rivulus using the culture system described in
this report can expect about 1 egg per fish per day.
GROWTH
To determine the growth rate of rivulus relative to container size, a
rearing experiment was set up with three container sizes. All containers
were stackeble culture dishes and there were four replicates for each dish
size. Dish diameters were 20.3, 11.4 and 5.1 cm and all were filled to a
depth of 2 cm with synthetic 14 o/oo seawater. Uniform lighting was
provided by 4-40 watt cool-white flourescent light bulbs suspended about
1 m above the fish. Standard 14 hr photoperiod and 25.0 - 0.5 C were
maintained throughout except for a 24 hr period in February when a power
failure caused the temperature to drop to 17 C. Growth rate was recorded
by periodically photographing each fish in its container next to a metric
ruler. At the beginning of the study (4 Sept 1979) 12 larvae, all hatching
on that day and all of the clone OS, were randomly assigned to the various
containers. All dishes were washed and water changed weekly. All fish
were fed equal amounts (in random order) of brine shrimp nauplii three times
a week for the first 85 days. The quantity of nauplii was adjusted so that
there were some remaining at each successive feeding. The young fish could
thus feed ad libitum. All fish were switched from brine shrimp nauplii to
frozen adult brine shrimp on the 86th day and they remained wn this diet for
the next 140 days at which time the study was ended. Frozen brine shrimp
were also fed in random order with equal numbers of shrimp to each fish.
Summary data from the experiment are presented in Table 1 and the
growth rate of fish in the 11.3 cm dishes is depicted in Figure 3. The
growth rate was rapid during the first 2-3 weeks then became almost linear
with time until about 160 days at which time it began leveling off.
Rate of growth was related to the size of dish (Table 1). The greatest
difference in mean size among the dishes occurred 85 days after hatching
and the effects of the smallest dish were greatest. It is possible that
-------
Table 1. Growth of Rivulus marmoratus in Three Container Sizes
Date
Day
After
Hatch
Diameter
of
Dish(cm)
TL(mm)
SE
N
Significance3
4 Sept 1978
0
5.1
5
0
4
NSb
It
11.4
5
0
4
It
20.3
5
0
4
21 Sept 1978
17
5.1
9.0
0
4
II
11,4
9.5
0.29
4
NS
II
20.3
10.1
0.43
4
27 Oct 1978
53
5.1
10.8
0.25
4
11 ,
11.4
12.3
0.63
4
p<0.05
It
20.3
13.3
0.25
4
28 Nov 1978
85
5.1
13.5
0.29
4
ll
11.4
15.6
0.24
4
p <0.001
11
20.3
17.3
0.25
\4
24 Jan 1979
142
5.1
21.3
0.33
3C
¦
II
11.4
22.3
0.25
4.
NS
II
20.3
23.7
1.2
3d
27 Mar 1979
204
5.1
25.0
*-->»
lc
11
11.4
26.5
0.29
4
US
II
20.3
26.3
0.44
3
17 Apr 1979
225
5.1
25.0
......
1
ft
11.4
26.8
0.25
4
NS
11
20.3
26.7
0.33
3
aDetermlned by single factor ANOVA
bNS = not significant » p >0.05
cNumbers reduced by death of fish
d0ne fish jumped out of container and was lost
12
-------
30
160
200
240
120
40
80
Days From Hatching
Figure 3, Growth of Rjvulus marmoratus in the laboratory under controlled conditions: 11.4 cm
d1am culture dishes, 25.0 ± 0.5 C, 14 o/oo salinity and 14 hr photoperiod. Arrows
represent time in which diet was changed from live brine shrimp nauplii to frozen adult
brine shrimp (f) and time when first egg was laid (e).
-------
feeding nauplii to older fish may have influenced the divergence in size.
The diet was changed on the 86th day to frozen brin? shrimp and all
subsequent measurements were not significantly different among the dish sizes
One fish jumped out of a large dish 28 Nov 1978 and was lost. All
other mortalities occurred in the smallest dishes; 1 died 26 Dec 1978
(113 days after hatching}, 2 died 26 Feb 1979 (175 days after hatching), and
1 died 17 Apr 1979 (225 days after hatching). The first egg was laid 28
Feb 1979 (177 days after hatching) and other fish followed during the sub-
sequent weeks.
The optimum size culture dish is 11.4 cm diarn because growth rate was
nearly the same as in the largest dishep. Also, the 11.4 cm dish is much
easier to handle and store. The smallest dish (5.1 cm) is clearly too
small for full life cycle culture tut is convenient for eggs, larvae and
juveniles up to about 2 months old. Food type should also be switched from
nauplii to frozen brine shrimp at about 2 months after hatching.
EMBRYONIC DEVELOPMENT
This description of the embryonic stages of Rivulus marmoratus was
patterned after Armstrong and Child's (1965) description of the embryonic
development of cyprinodontid fish Fundulus heteroclitus and Koeriig and
Livingston's (1976) description of the embryonic development of the
cyprinodontid fish Adinia xenica. Embryonic development is described as
occurring in stages. The term stage refers to a discrete event or series
of events in the developmental sequence. In this study, each stage is
centered around an easily recognized event and also includes various minor
changes which occur during the same time period. All stages are observable
with the aid of a dissecting microscope. All times given are midpoints of
stages.
Eggs usea for determination of stages were obtained after they were
naturally emitted from the parental fish. These eggs were collected at
least once a day. A minimum of 15 eggs were observed for determination of
each stage.
Eggs were handled with flexible forceps or eye droppers and observed
on depression microscope slides or in 5.1 cm diam culture dishes. For ease
in observation the later stages were dechorionated using stainless steel
watchmaker forceps.
Observations were made using a Wild M-5 binocular microscope at 25 X
magnification with transmitted and/or rsflected light. All photographs of
developmental stages were taken at 25 X or 10 X magnification with a Nikon
cawera attached to the microscope.
The following description of the stages of development of £. marmoratus
under.the controlled conditions of 25.0 * 0,5 C.14.0 t 0.5 o/oo salinity,
and 14 hours of light per day. The stages include development from
fertilization to post-hatching. Stage one was observed only once; the
14
-------
description and time from fertilisation until the midpoint of stage one
is from Harrington (1963). A summary of the stages including the time
from fertilization to the midpoint of each stage is given in Table 2,
Stage 1 (Figure 4a); The egg is fertilized and the perivitelline space
Is present. The blastodisc is present, resulting from the migration of
cytoplasm to the animal pole of the egg. The yolk is transparent and contains
i mass of oil droplets near the surface. This stage is rarely observed in
rivulus because eggs are not released from the parental fish until after
internal self-fertilization.
Stage 2; The first cleavage is meridional; the one cell dividing into
two cells of equal size. Cleavage is meroblastic, thus the yolk mass does
not divide. The surface of the egg has a speckled appearance which remains
throughout development.
Stage 3: The second cleavage is meridional and at right angles to the
first producing four clastomeres of approximately equal size.
Stage 4: The four blastomeres divide vertically to form eight
blastomeres.
Stage 5: The blastomeres again divide vertically producing sixteen
cells which, when viewed from the animal pole, are in four irregular rows of
four cells each.
Stage 6: Cleavage is horizontal, producing thirty-two cells arranged
in an upper and lower layer. The eel Is become progressively smaller as
cleavage continues.
Stage 7: The dividing blastomeres have formed a low dome-shaped mass
upon the yolk, the early blastula. Blastomeres are piled upon each other
and it is hard to distinguish the separate cells.
Stage 8 (Figure 4b); The high blastula stage is recognizable by the
distinct, bulging, bun-iike dome of cells on top of the yolk mass.
Stage 9; During the late or flat blastula stage the blastomeres are
still in a mass, but the margins of the blastoderm are less distinct and
the dome-shaped mass is not as high as previously noted.
Stage 10: The expanding blastula begins gastrulation by epiboly and
convergence. The margins of the blastoderm are ragged in appearance.
Stage 11 (Figure 4c): Gastrulation continues; the cell layer involved
in epiboly, the epiblast, covers 1/3 of the yolk.
Stage 12: Gastrulation continuesl the epiblast covers. 1/2 of the yolk.
It is apparent that there is a massing of cells around the leading edge of
migrating blastoderm, the germ ring. Also, there is a massing of cells in
one area of the germ ring, the embryonic shield. The germ ring and the
embryonic sheild are embryonic structures formed by the two movements of
IE
-------
1
2
3
4
5
6
7
8
8
10
11
12
13
14
15
16
17
18
19
20
21
22
23
Summary of the Development Stages of Rivulus maraioratus at
25,0 t 0.5 C, 14.0 ± 0.5 o/oo Salinity and 14 hours of Light per
Day. The Times given are from Fertilization to the Midpoint of
the Stages.
Time (hours) Description
2.5 1 cell
3.5 2 cells
4 4 cells
4.5 8 cells
5.5 16 cells
6.5 32 cells
8 Early blastula
9.5 High blastula
10.5 Late or flat blastula
15 Gastrulation begins, expanding blastula
19 Epiblast covers 1/3 yolk
22 Epiblast covers 1/2 yolk, germ ring and
embryonic shield are forming
24 Epiblast covers 2/3 yolk
25.5 Large yolk plug, embryonic shield enlarging
31 Epiboly complete
34.5 Head and tail regions recognizable
36 Optic vesicles appear, somite formation begins
43.5 Optocoeles are prominent, auditory vesicles form
53 Optic cup and lens formation
55.5 Heart beats, no circulation
58 Body movement
62 Circulation
71 Increased vitelline circulation
73 ¦ Urinary bladder and otoliths first appear, pig-
mentation on brain, trunk, and yolk near an bryo
-------
Table 2 (continued)
Stage Time (hours) Description
25
77
Pectoral fins appear, otoliths are prominent
26
90
Liver first appears macroscopically, pigmentation
on optic cup
27
105
Increased pigmentation and body movement
28
140
Pigmentation of the optic cup obscures the
lens, circulation in pectoral fin and liver,
caudal fin developing
29
180
Gas bladder and anal fin formation
30 .
211
Rays in caudal fin form, dorsal fin develops,
the jaw appears
31
240
Pectoral fin movement, circulation in the caudal
fin, a delay in hatching can occur
32
310
Hatching, rays in the dorsal and anal fin form;
hatching may be delayed for days
17
-------
T1
*n ft *™3ls
Figure 4a-f. Photographic representations of some early stages in the
embryonic development of Rivulus marmoratus. Stages corre-
sponding to figures 4a-f are: 4a-l, 4b-8, 4c-ll» 4d-13,
4e-16, 4f-19.
18
-------
gastrulation, epiboly, and convergence, respectively. The mass of large oil
droplets is present at the yolk surface near the germ ring.
Stage 13 (Figure 4d}: Gastrulation continues; the embryonic shield
enlarges; and tha epiblast covers 2/3 of the yolk.
Stage 14: As gastrulation continues, the yolk material not yet covered
by migrating epiblast cells bulges out forming a large yolk plug. Continued
convergence of eel Is of the germ ring increases the embryonic shield
longitudinally.
Stage 15: Gastrulation is complete; the epiblast encircles the entire
yolk. The embryo, lying on the yolk surface, appears to be a furrow running
through the mass of oil droplets located directly under the embryo. In
oblique view, the embryo appears to be a translucent ridge curving around
the surface of the egg.
Stage 16 (Figure 4e): The embryo continues to lengthen and amass tissue
laterally and the head and tail regions of the embryo are recognizable. In
this stage the embryo is seen as a prominent ridge on the surface of the
egg.
Stage 17: The three regions of the brain, the forebrain, the midbrain,
and the hindbrain, become visible and the optic vesicles appear as bulges
off the forebrain. Late in this stage the optocoeles can be distinguished
as cavities within the optic vesicles. The first few pairs of somites form,
but are difficult to see due to the oil droplets massed below the embryo.
The tip of the tail is bulbous, and there is a mass of tiny droplets directly
posterior to the tail. The pericardial cavity forms extending on the yolk
surface anterior to the head.
Stage 18: The embryo is increasing in size, and somite formation
continues. The optocoeles are prominent within the optic vesicles and the
auditory vesicles form lateral to the hindbrain. The pericardial cavity has
increased in size and now extends anterior to and lateral to the head. The
oil droplets are still beneath the embryo, but the head region now extends
beyond the mass of oil droplets.
Stage 19 (Figure 4f): The optic region is developing with the formation
of the optic cup and lens. The optic cup forms from the optic vesicle by
invagination of the lateral surface of the vesicle; thus the optic cup
appears first bean-shaped and then cup-shaped. The lens forms within the
ccup-shaped vesicle. The buloous tip of the tail is slightly elevated off
the yolk. The auditory vesicles are prominent, and the olfactory vesicles
are visible. The expansion of the pericardial cavity, now extending from
the tip of the head to the first somite, elevates that head. However, the
tip of the forebrain is still flush with the yolk. There is increased brain
development: the optic lobes appear as lateral enlargements of the midbrain
and a furrow forms down the entire length of the brain.
19
-------
Stage 20: The heart, visible within the pericardial cavity, extends
from the yolk surface anterior to the embryo to the anteroventral surface of
the embryo. This tiny tubular heart beats, but as yet no blood circulates,
Stage 21: The interior somite area has sporadic contractions.
Approximately 1/2 of the tail area is free of connection with the yolk,
but the tail still lies close to the yolk. There is still a mass of tiny oil
droplets posterior to the tail. Prior to circulation, anterior and posterior
to the embryo, there is a massing of blood cells.
Stage 22: Circulation begins through the dorsal blood vessel and is
observable anterior to the heart and posterior to the junction of the tail
and the yolk. The vitelline vessels develop slowly; complete posterior-to-
anterior vitelline circulation is established approximately two hours after
the onset of circulation. As the vitelline circulation encompasses the
entire yolk, the oil droplets, previously clumped together, spread outwards
until they are envenly dispersed around the yolk. The optic lobes are
enlarging, ventricles of the brain are observable, and the brain is taking
on a coiled, convoluted appearance. There is a fine network of pigmentation
covering the surface of the brain.
Stage 23 (Figure 5a): Circulation is well established now. The network
of vitelline vessels is increasing; the vessels emerging from the embryo
anterior to the somites are especially prominent. The brain continues to
enlarge, and the optic lobes become massive and protrude anteriorly, over-
hanging the posterior portion of the optic cup. Pigmentation covering the .
embryo increases. The tail occasionally swings sideways.
Stage 24: Organ development increases now that circulation is
established. A tiny urinary bladder forms dorsal to the hindgut and ventral
to the dorsal blood vessel. The otoliths first appear as tiny specks in the
auditory vesicles. Pigmentation on the brain and trunk region increases as
well as on the yolk lateral to the embryo.
Stage 25: The otoliths are dense aggregations now, and the pectoral
fins appear as tiny triangles protruding from the yolk surface immediately
anterior to the first somite.
Stage 26 (Figure 5b): The pectoral fins lengthen until they project
halfway up to the dorsal surface of the trunk. There is a general increase
in pigmentation on the head, trunk, and tail of the embryo and on the yolk
surface. Pigmentation shades the optic cup a light grey, but the cup is
still transparent. The liver bud forms and can be seen on the left side of
the embryo immediately posterior to the pectoral fin. Liver bud formation
was not followed histologically; these are macroscopic observations on whole
embryos.
Stage 27: The liver increases in size and becomes globular in
appearance. Somatic pigmentation continues to increase, especially down the
dorsal surface of the tail. The tail occasionally swings vigorously enough
to touch the head.
20
-------
. ^^-jsISssks tfc&ite&iaafo.iafoS
Figure 5a-d. Photographic representations of some late stages in the
embryonic development of Rivulus marmoratus. Stages
corresponding to figures 5a-d are: 5a-23, 5b-26, 5c-28,
5d-31. Chorions were removed from embryos 5c and 5d.
21
-------
Stage 28 (Figure 5c): Pigmentation of the optic cup Is so dark that
it obscures the lends, and pigmentation of the peritoneum 1s obvious. The
body cavity and liver are enlarging. The pectoral fins enlarge and circu-
lation is -observable in them. The caudal fin is forming, appearing as a
transparent membrane around the posterior end of the caudal peduncle.
Stage 29: The body cavity 1s large, lifting the embryo off the yolk
surfate, but the grey twisting gut does not fill the body cavity. The gas
bladder forms dorsal to the gut, but because of dark pigmentation it is hard
to distinguish. The pectoral fins project as high as the dorsal surface of
the trunk, and the anal fin begins developing. In the tail, circulation is
observable between the caudal artery and vein.
Stage 30: The rays form in the caudal fin and the dorsal fin begins
developing. Jaw formation is observable and the developing jaws are
elevated off the yolk. Pigmentation is heavy, guanine is located along the
vertebrae, on the bases of the pectoral fins, and on the surfaces of the
optic cups.
Stage 31 (Figure 5d): The pectoral fins flutter slightly, but have no
coordinated movement. Pigmentation is heavy on the dorsal surfaces of the
head, trunk, and tail, on the yolk sac, and on the opercula. Circulation
Is established in the caudal fin. The anal fin is more pronounced than the
dorsal fin, but no rays are present in either fin. A delay in further
development sometimes occurs at this stage*, the embryo can enter "a period
showing little or no movement and an irregular heart beat.
Stage 32: The mouth opens and closes with accompanying opercular
movements, and the pectoral fins are motile. Rays develop in the dorsal and
anal fins, and pigmentation appears along the rays of the caudal and
pectoral fins. Hatching occurs at this stage.
The young fish actively swims from the moment of hatching. There is
still a small amount of yolk observable on the ventral surface of the body.
Although most eggs of rivulus hatch within 16 days after fertilization,
some eggs may take up to 14 additional days after reacning stage 31 to
hatch. Eggs of other oviparous c^prinodonts also have varying times of
hatching. In different species the eggs may undergo either a period of
latency or delay (Harrington, 195S) or a diapause (Wourms, 1972a, b, c).
In diapause the embryo appears to be in a state of near suspended animation
with little or no growth, development, movement, or yolk utilization. This
may be an extended stage, the usual duration being two to three months for
the annual cyprinodontid fish Austrofundulus myersi—Wourms (197?a). On the
other hand, in a period of latency the embryo continues growth arid utili-
zation of the yolk. This delay is usually of short duration because the
limited amount of yolk, among other things, would curtail an extensive
delay. Koenig and Livingston (1976) showed that embryos of the cyprinodontid
fish Adinia xenica could successfully delay hatching for ten days, the
delay disrupted by a change in environmental conditions. Because rivulus
exhibits a relatively short interruption of final development and continues
to utilize the yolk during this period, it appears that it exhibits e delay
or latency rather than a diapause. However, rivulus shows only a limited
-------
amount of activity during this latency period. Often during this period,
the pulsations of the heart are so infrequent that no signs of circulation
are observable; for example, blood cannot be seen moving through the
vitelline vessels, usually sites of easily observed circulation. Often the
heart will resume regularity after stimulus from the probe used in handling
the eggs. However, hatching will not occur immediately after this increase
in activity. The control of hatching or the cause of delayed hatching in
rivulus is not known.
Sane of the most interesting characteristics of rivulus cause
difficulties when this fish and its eggs are used in research. Due to the
fact that eggs are emitted in different stages and often not until after 10
hours of development (Harrington, 1963), it is difficult co observe the
entire sequence of development, especially repeatedly, among a series of
eggs. In addition, any study using rivulus as the test animal :ould run
into difficulties if it were necessary to study the effects of sane factor
upon a certain age group. A large number of parental fish are necessary to
obtain a large number of eggs or juveniles in the same stage of development.
Comparing the development of R. marmoratus to that of the commonly
studied cyprinodont Fundu1us heteroc1i tus--(Sol berg, 1938; Armstrong and
Child, 1965) shows that the overall sequence of development stages is the
same, but that rivulus has a slower rate of development through some stages.
In particular, brain and circulatory development in rivulus takes longer than
it does in F. heteroc!itus. * Another developmental difference between these
species is that the embryos of rivulus tend to acquire heavy pigmentation in
the later stages of development. The pigmentation is noticeably darker than
that of £, heteroc1itus. Because of this heavy pigmentation it is sometimes
difficult to distinguish development of internal organs such as the liver
and swim bladder.
COMPARISON OF EGG INCUBATION TECHNIQUES
Because many estuarine cyprinodontid fish deposit their eggs so that
development takes place in moist air (Harrington, 1959; Hastings and
Yerger, 1971; Able and Castagna, 1975; Taylor et_ al_., 1977; Johnson, 1980}
a comparison of air incubation and water incubation was made. In a previous
study (Shealy, 1979) air incubation was found superior to water Incubation
for embryos of Fundulus heteroc!itus. Development was more rapid and more
uniform in air incubated funtfulus eggs»
Twelve rivulus were held singly in reproduction chambers (25 i 0.5 C,
14 hr photoperiod). Eggs were collected daily and a random number table was
used to select the appropriate culture technique (odd • air, even = water).
Water incubated eggs were held singly in 5 cm culture dishes filled to 1 cm
with synthetic seawater (14 o/oo). Air incubated eggs were held singly in
5 cm culture dishes on three layers of WhatmanR no. 2 filter paper which had
been moistened with synthetic seawater (14 o/oo). When embryos reached the
terminal stage of embryonic development (Stage 32) eggs were inundated for
1 hr with freshly collected seawater. If hatching did not occur within
23
-------
1 hr, incubation conditions were resumed and the procedure was repeated
the following day and so on until hatching occurred. The approximate age of
each embryo was estimated by observing its stage of development upon or
shortly after oviposition. All embryos were observed and staged daily under
low light on a dissecting microscope.
Comparison of the terminal stages of embryonic development (Table 3)
shows that development in water was more rapid than development in air
although uniformity was approximately the same for both. The method chosen
for egg development was therefore water incubation as described in the
¦section "Laboratory Culture of Rivulus marmoratus".
24
-------
Table 3. Comparison of Rates of Embryonic Development of Rivulus marmoratus
in Air Incubation vs. Water Incubation
Staqe
Air
Time(hrs)a
Water
t«
26
115,4 (4.66,Up
110.5 (4.37,IT)
0.75
NSd
27
141.6 (5.17,15)
135.5 (4.61,8 )
0.77
NS
28
167.3 (6.43,8 )
149.8 (4.78,12)
2.22
p < ,05
29
186.7 (4.95,22)
176.0 (3.97,18)
1.88
NS
30
236.3 (4.98,27)
210.5 (5.22,12)
3.12
p < .01
31
277.6 (7.89,11)
241.6 (5.59,16)
3.84
p < .01
32
201.2 (7.67,13)
264.8 (9.15,12)
2.22
p « .05
Hatch
368.6 (6.98,13)
355.4 (7.80,12)
1.24
NS
aAverage time from fertilization to midpoints of stages 26 through 31,
and to the beginning of stage 32.
^Numbers In parentheses are standard errors and numbers of observations,
respectively,
cStudent's t
dNS ¦= not significant
25
-------
SECTION 5
BEHAVIORAL BIOASSAY USING RIVULJS MARMORATUS
INTRODUCTION
A recent review article by Cairns and van der Schalie (1980) sunmarized
and discussed the concept of early warning systems in biological monitoring.
These systems are charged with the responsibility of detecting environmental
contamination at a very early stage so that further perturbation can be
prevented and clean-up operations begun. Central to a biological early
warning system is an organism possessing an observable physiological or
behavioral trait which is rapid, quantifiable, and, most important,
sensitive to harmful environmental variation. Of practical Importance, the
organism must be easy and inexpensive to culture if it is going to achieve
widespread use. Numerous species are presently being employed in this
capacity (see Table 1 in Ca\rns and van der Schalie, 1980). Yet, there is a
constant search for more suitable species to replace or supplement these
because, in most cases, one or more of the above conditions is not met
satisfactorily. The objectives here are to describe a behavioral response
by the cyprinodont fish Rivulus marmoratus to aquatic hydrogen sulfide
contamination, and to evaluate its potential for use in a biological early
warning system.
Recently R. marmoratus has been implicated as having terrestrial habits
(Kristensen, 1?70; Brockman, 1975; Courtney, 1975; Abel, 1981). Abel (1981)
proposed a possible ecological basis for the terrestrial behavior which
involves toxic levels of H?S. In the laboratory, upon exposure to H2S,
R. marmoratus leaps from the water and either sticks to or lies on emergent
objects, apparently respiring cutaneously for up to several hours before
returning to the water. The unique ability of R. marmoratus to detect and
avoid contaminated water in this way, plus the advantages of using geneti-
cally indentical individuals, serve as the bases for this bioassay.
Sulfurous effluents from industrial sources such as pulp mills,
chemical and gas plants, and oil refineries burden many aquatic environments
(Ellis, 1937; Van Horn et al_., 1949; Dorris et ai., 1960). In addition,
H|S is produced from decomposition of sewageTCoTby and Smith, 1967;
Ziebell et al., 1970). Since H2S is an extremely potent metabolic poison
of widespread occurrar.je, and because it has been demonstrated that
Rivulus marmoratus detects and avoids toxic levels, HgS was selected as a
reference toxicant for this study.
26
-------
MATERIALS AND METHODS
Fish used in this study were maintained throughout in synthetic sea-
water {Instant OceanR and distilled water) adjusted to 14 + 1 ppt, with a
14 hr photoperiod, and at a constant temperature of 25 ± 1 C, Additional
information on culture methods is given under the section "Laboratory
Culture of Rivulus marmoratus" of this report.
Assays of H2S were conducted in covered, 400-ml PyrexR beakers
modified for flow-through capacity. An oyster shell attached to the side
of each beaker slightly above the level of the overflow tube served as a
platform onto which the fish could jump and lie. The experimental design
contained numerous precautions to eliminate bias. Five beakers, each fed
by a different head tank, were used in each run. To three randomly
Cby random number table) selected tanks different solutions of HgS were
added. Two other randomly chosen tanks served as controls. Water pH was
7.9 i 0.05 in all tanks. Except for one control which contained air-
equilibrated water dissolved oxygen concentration was 2.0 i o.05 ppm in all
tanks. Assays were run at lew oxygen tension to duplicate conditions:
(1) possibly associated with conversion to aerial respiration in the
natural habitat of Rivulus mannoratus and, (2) frequently associated with
the presence of H2S in other areas.
Five fish from 0.3 - 0.5 g, starved for three days, were randomly
selected and individually added to a separate beaker containing 25 ma of
normoxic water. The experimental solutions were then introduced at a rate
of 2.0 i 0.05 a hr-1, which was maintained throughout each run. The fish
were allowed to acclimate for 1 hr, followed by 1 hr of observation. Each
fish was used only once. Since preliminary experiments suggested that the
ability of a fish to flop onto the oyster shell required learning, a
positive response to the test solution was chosen to be not only (1) leaping
from the water onto the shell, but also (2) demonstration of an attempt to
leave the water as illustrated by a least one jump completely from the
water. In addition to this quanta! response, total time emergent and
number of jumps were recorded. Eleven runs utilizing fifty-three fish were
performed. A single observer scored all runs without knowledge of the
relative or absolute.H2S concentrations of the solutions.
Hydrogen sulfide and oxygen tensions were measured in water samples
taken from the beakers at the beginning of acclimation, at least once
during an experiment, and at the conclusion. Additionally, blanks ware run
and measured. Close monitoring is necessary since H2S oxidizes rapidly.
However, since HgS concentrations varied i 152 and O2 did not change
significantly over the time span, values or variables measured at the
beginning of acclimation were used in all calculations. Solutions of HgS
were prepared by adding calculated amounts of sodium sulfide-to deoxygen-
ated water. Hydrogen sulfide was assayed by the spectrophotometry method
(American Public Health Association, 197C). Seawater was deoxygenated by
bubbling nitrogen through it while monitoring oxygen tension with a Yellow
Springs Instrument Corporation Model 51-A O2 meter, calibrated prior to
each use by the azide modification of the Winkler method (American Public
-------
Health Association, 1976).
Probit analysis (Finney, 1971) was used to analyze the quanta! response
of jumping. Data were grouped into intervals of 100 ppb HgS and the
midpoints used in the calculations. Spearman's Coefficient of Rank
Correlation (Steel and Torrie, 1380) was used to examine the relationship
between actual HgS concentration and the amount of time fish responding
positively remained emergent, that is, on the oyster shell.
RESULTS
The concentration of ^5 that elicited a positive response in 50% of
the fish (Median effective concentration = EC50) was 123.S9 ppb. The 95$
confidence limits were 63.68 to 181.97 ppb. Figure 6 depicts the results
of the assay. Of the twenty fish responding positively, sixteen actually
jumped on the oyster shell. Analysis of the amount of time spent emergent
(fran 0 to 60 rain) for the twenty fish by Spearman's Coefficient of Rank
Correlation showed a significant positive correlation {r » 0.52, p < 0.05)
between H2S concentration and time emergent. Data are presented in Table 4.
There was no discernible relationship between H2S concentration and number
of jumps. Of additional significance was an alternate strategy of
avoidance exhibited by fish in the lower concentrations of H2S. These fish
aligned themselves at the air-water interface, apparently using the oxygen -
microlayer, for as long as the duration cf the run; controls did not exhibit
such behavior. This behavior was not quantified. Although data on time
to first leap were not collected since close observations were not made
during acclimation, three fish were observed emergent at the beginning of
the observation period. The solutions in which this occurred were 233,
252, and 579 ppb H2S,
DISCUSSION
This experiment demonstrates the ability of Rivulus marmoratus to
detect an avoid water contaminated with HgS. Although the EC50 detennined
by this investigation (123.59 ppb) was higher than those reported in the
literature for other sub-lethal responses for example, Lepomis
macrochirus exhibited decreased swimming endurance at 1.5 ppb' HgS (Oseid
and Smith, 1972) , it is sensitive enough to warrant more comprehensive
testing of the leaping behavior as an indicator of H2S pollution.
Quantification of ether reactions, such as alignment at the surface or
activity patterns, could result in a much more sensitive response. Also,
since R. marmoratus is preadapted to still conditions (see below), static
assays might lower the EC50.
Analysis of the amount of zime emergent for those fish responding
positively to the treatments demonstrated that at higher H?S concentrations,
fish generally remained emergent longer. The high variability at the
higher concentrations and the low correlation coefficient may be due to
toxic effects of the H2S.
28
-------
YfPROBITr -1-794 +3548 logtttX
* 4
¦" .11 // I 1 1 I I I I—
0 0 // 30 1SO 250 350 S50
MjS (ppb)
Figure 6. Log-probit plot of t positive response vs. H2S
concentration. Solid dots represent experimental fish.
Solid block is control run in air-equilibrated water.
Open block is control run at 2.0 ppm 02 concentration.
Downward and upward arrows are directed at points
representing 0 and 1001 response, respectively. H|S
concentrations are mid-points of 100 ppb Intervals.
Other conditions are described in the text.
29
-------
Table 4. Total Time Emergent in Minutes for Each of the Twenty Fish
Responding Positively to the Treatments. Positive Correlation
Between the Variables was Significant (p < 0.05)
H?S Concentration Total Time Emergent
(ppb) (min)
25 0*
108 15
147 14
172 17
172 0*
195 0*
200 47
209 0*
211 38
232 45
233 51
252 60
270 39
304 31
339 27
399 31
525 54
564 28
565 4 15
579 ' 52
~Represents response judged positive based on jumping criterion
(see text).
30
-------
Koenig and McLean (1980) discussed the manifold advantages of using
the Rivulus marmoratus bioassay system relative to chronic marine bioassays.
Here, the characteristics of the system will be represented as they pertain
to biological early warning systems, using as guidelines the requirements
for such systems as given by Cairns and van der Schalie (1980).
1. The behavior used in this bioassay is easily quantifiable through
electronic devices, such as the one described by Cripe et al_., (1975),
which has been successfully tested for use with Rivulus marmoratus.
Since avoidance removes the fish from physical contact with the
toxicant, the fish is not noticeably stressed during the bioassay
except at highly toxic levels of contaminant, such as possibly
occurred in this experiment when some fish in the higher concentrations
of H2S remained emergent less than fish in medium or low concentrations.
2. Although rapidity of response was not measured in this experiment, the
response has been observed previously to be instantaneous to such
toxicants as H2S, anmonia, formalin, and rotenone, all in dilute
solutions. That a large number of fish responded within 2 hours of the
initial exposure in this study places the reactions in the upper 401
for the fish listed in Table 1 of Cairns and van der Schalie (1980).
Although it is not known how wide a range of toxicants will elicit the
jumping behavior, those which affect respiration or otherwise irritate
the fish are obvious possibilities.
3. The response of Rivulus marmoratus to so-called non-harmful variation
in water quality is not well-known. The species typically inhabits
very shallow areas ( < 15 cm deep) which are subject to wide variations
in water level and salinity. In the laboratory, this species has been
held in salinities of 0 (distilled water) to 50 ppt without noticeable
avoidance behavior (it is therefore suitable for both freshwater and
marine bioassays). As demonstrated here, environmental hypoxia does
not induce jumping, since R. marmoratus possesses adaptations which
have been shown by Lewis (T970J for other cyprinodontoids to facilitate
respiratory use of the oxygen-n'ch surface microlayer.
4. An important characteristic of the system is genetic uniformity.
Genetic variation should therefore not enter as a source of experimental
variability. However, the data presented, particularly those in
Table 4, indicate considerable variability! although behavioral
responses of this nature are generally quite variable. As a future
study it would be interesting to investigate directly the relationships
between genetic uniformity and uniformity of experimental data.
5. Because Rivulus marmoratus inhabits still waters, it adapts well and
quickly to living in culture dishes in the laboratory. Therefore,
large numbers of fish can be maintained in small areas at low cost. In
addition, fish older than 4 months produce a constant supply of eggs
throughout the year.
31
-------
Clearly, Rlvulus marmoratus possesses qualities equaled by no other
single fish species, freshwater or marine, for use 1n bioassays. Of course,
there tre also a number of possible disadvantages in the use of any single
test animal and this depends on the particular experimental design. For
example, homozygosity and isogenicity may be disadvantages to the experi-
ment purporting to extrapolate to responses of heterozygous populations.
Also, rivulus is quite aggressive and thus it 1s important that adults be
held singly or be given refuges when held communally.
Although more comprehensive testing is required and appropriate
interfacing techniques applied to complete the system, the use of R.
marmoratus, including the novel application here, has promise in water
quality management, and this species merits consideration for adoption as
a standard behavioral bioassay organism.
32
-------
SECTION 6
CARCINOGENESIS BIOASSAY USING RIVULUS MARMORATUS
INTRODUCTION
Much early work on neoplasms in fish grew out of an unexpected out-
break of liver cancer in cultured rainbow trout, Salmo gairdneri.from
1957 to 1960 (reviewed by Sinnhuber et al., 1977aJ! The cause of the
disease was traced to aflatoxin contaminated diet. Subsequent studies
demonstrated not only the great cancer producing potential of aflatoxins
but also the high sensitivity of certain strains of rainbow trout to
chemical carcinogens general.
Many recent studies have focused on establishing the sensitivities of
certain species of small aquarium fishes to known chemical carcinogens.
Stanton (1965) was the first to use a small aquarium fish, the zebrafish
Brachydanio rerio, in this type of experiment and he found that hepatic
neoplasms could be induced in as short a time as eight weeks after diethy-
Initrosamine (DEN) was added to the aquarium water. Later, hepatocellular
carcinomas were induced by various carcinogens in two more species of
small aquarium fish, the guppy Poecilia reticulata and the medaka Oryzias
latipes (Sato et al., 1973, Pliss and Khudoley, 1975, Ishikawa et aV. 1975,
Ishikawa and Takayama, 1979). In reviewing a number of studies involving
the effects of chemical carcinogens on small aquarium fish Matsushima and
Sugimura (1976) concluded that their use in studies involving chemical
carcinogenesis has certain advantages over manmalian systems.
However, much more work must be done before small aquarium fish test
systems are demonstrated superior to established rat and mouse systems and
included in standard protocols.
The purpose of this study is to establish the sensitivity of Rivulus
marmoratus to known chemical carcinogens so that its potential as a test
animal for carcinogenic studies may be evaluated.
MATERIALS AND METHODS
Routine culture methods and conditions were described under section 2,
"Laboratory Culirre of Rivulus marmoratus". The actual experimental setups
differed somewhat from the routine methods to facilitate handling of the
fish and to minimize chances of exposure of laboratory personnel to
carcinogenic compounds. Several experiments were devised in which adults
33
-------
(> 6 mcs.), larvae (< 1 mo.) or embryos were exposed via water to various
concentrations of diethylnitrosamine (DEN). Other experiments involving
water exposure of rivulus to 2-acetylaminoflourene (AAF) (with ethanol used
as a carrier) and injection (i.p. with cod liver oil carrier) of benzo-a-
pyrene (EaP) and AAF were inconclusive. Therefore, only the results of the
DEN exposures will be reported here. DEN was purchased from Sigma Chemical
Co. and BaP and AAF from Aldrich Chemical Co.
In all experiments fish were assigned to various exposure levels
randomly (random number table). Temperature, photoperiod and salinity were
25 + i c, 14 hr arid 14 + 1 ppt, respectively. Also, containers were cleaned
and solutions were renewed weekly. Fish were fed brine shrimp nauplii or
frozen adults three times a week. In all cases control fish were treated
identically as exposed fish but without the inclusion of the carcinogen.
All exposures were carried out in a 1ight and temperature controlled
..¦environator outfitted with an exhaust fan. Contaminated air from the
environator was periodically purged via the laboratory hood and all trans-
fers and cleanings were done under the hood. As an added precaution
rubber gloves and a carbon filter gas mask were worn during feeding or
cleaning. Any concentrated waste was stored in a stoppered glass jar under
the hood and disposed of through the hazardous waste disposal procedure at
the Medical University of South Carolina (MUSC), Charleston, SC. Because
of our intent to compare the response of rivulus to other fish exposed
similarly, concentrations were chosen on the basis of similar published
experiments for DEN (Uhikawa and Takayama, 1979). Exposures continued
until fish showed signs of toxic stress. In the case of the embryo
exposures experimental data were lacking in the literature. Exposure levels
were therefore designated in a log series up to the approximate lethal
level. The three exposure experiments are described individually:
(1) Adult rivulus (30 fish per concentration) were exposed to DEN in the
culture water in concentrations of 45, 30, 15, and 0 mgit for 5 weeks, then
held in uncontaminated water for one year before being killed, examined
grossly and prepared for histological examination, fish were held in
groups of 10 in 20.3 cm diam culture dishes which were kept covered in the
environator. (2) Larval rivulus (30 fish per concentration) were exposed
to DEN in the culture water in concentrations of 45, 30, 15, and 0 mg/i.
for 12 weeks, then held in uncontaminated water for 1 year before being
killed, examined, and prepared for histological examination. Each group
of 30 larvae were held together in 20.3 cm diam culture dishes which were
kept covered in the environator. (3) Embryos of rivulus (56 per concentra-
tion) were exposed to DEN in the culture water in concentrations of 1,000,
320, 100, 32, 10, and 0 mg/i for 1 week. The resulting hatchlings were
then held for approximately one year in uncontaminated water before being
killed, examined, and prepared for histological examination. Prior to
exposure embryos were staged and assigned to one of five groups:
(1) stages 1-9, (2) stages 10-14, (3) stages 15-17, (4) stages 18-23,
(5) stages 24-31. See under section 3 "Embryonic Development". During
exposures embryos were held in 20 ml screw top test tubes sealed with teflon
liners. Not more than 5 embryos were held in each test tube. All exposures
were carried out in the environator.
34
-------
The major emphasis in this study was on induction of hepatocellular
carcinoma. Therefore, fish were killed in 102 buffered formalin solution
and the livers were dissected from the fish and processed according to
standard histological procedures. Only fish that survived to the end of the
experiment were included in the analysis. All livers and occasionally whole
bodies were sectioned at several levels. Sections (5-6 pm) were then
stained with hematoxylin and eosin (H+E). Slides were examined and tumors
described by Dr. J. C. Harshbarger (Registry of Tumors in Lower Animals,
Smithsonian Institution) and confirmed by Dr. A. B. Garvin (Medical
University of South Carolina, Charleston, SC),
A single randan section through all livers was used to determine tumor
incidence. It was assumed that the source of error using this method
(as opposed to serial sections} was small compared to experimental error.
RESULTS
Histologically, the 11 vers of rivulus resemble those of other bony
fishes (Figs. 7a, 7b). Liver cords are two cells thick, rather than one as
in mammals, and pancreatic tissue occurs within the liver along vessels
rather than as a separate orgar.. No pathologic changes were observed in
the livers of the control fish.
In DEf! treated fish from the larval and adult exposure experiments,
neoplastic changes originated from both hepatocytes and cholangiocytes.
Hepatocellular neoplasms appealed mostly as basophilic nodules, but
eosinophilic nodules were also present. Neoplastic nodules showed several
patterns: solid (Figs. 7c, 7d), trabecular (Figs. 8a, 8b), and glandular
patterns (Fig-.. 8c, Bd). Different patterns occasionally occurred together
in the same liver. Boundries between tumor nodules and normal liver were
distinct in some cases (Figs. 9a, 9b) but no thick fibrous capsules were
seen. Other nodules lacked a distinct boundary with tongues of tumor
extending into the normal tissue (Figs. 9c, 9d). In the adult exposures,
patterns of hepatocellular carcinoma were of all three types of about equal
proportions, however, sol id and glandular patterns were much more common
than the trabecular pattern in the larval exposures.
Tumor cells were enlarged with enlarged nuclei and distinct nucleoli.
Size varied from only slightly to considerably larger than normal. Most
tumor cells were fair to well-differentiated except in one case where many
tumor eel 1s were spindle-shaped (Figs. 10a, 10b). No tumor cells were so
poorly differentiated that all resemblance to normal liver was absent.
Two types of ductule lesions were seen. One type, cholangioma
(Figs. 10c, lOd) consisted of multiple, thick walled, various-sized,
randomly oriented ducts or tubules formed by a single layer of tall
columnar epithelium. In some ducts, epithelial papillae extended into the
lumens. Fibrous stroma was minimal. The other type of ductule lesion had
multiple, thin walled ducts or tubules formed by simple cuboidal or squamous
epithelium (Figs. 11a, lib). They ter.ded to have larger ducts surrounded
35
-------
Figures 7a-d. Photomicrographs of histological sections
(6 pm» H & E) of the livers of Rivulus
marrooratus. (a & b) Control fish showing
normalliver cells with varying degrees of
staining from dark (a) to light (b), x400.
{c & d) Fish exposed to 45 ppm DEN for 12
weeks with solid pattern hepatocellular
carcinoma, x200 and 400, respectively.
Note enlarged cells and nuclei with
prominent nucleoli.
36
-------
It-" w«»- -Jr6Srv«^V'»V5 i* ; *•*•
¦*V "
w. * 1:'¦ ; JT -r- *i
•> » * »* - * * * * •
'"'¦*_ ^' O ' '• „ C
*#•
i .-i*,
v -->-
**"* ..•' '
* tt, ' "f • ?.
* .t * ' m. • 4
* '.fg ' *• '
«#* ' * ¦ ¦ '
*>
#< ^
>«V** —
ki*», «- ¦' -
: • e.U^ * * • , ., *%* • • <*-+¦ v -«.-
i , . , C-' * wi **• ' ~ +&t-1. *«#£-' «¦
«* V.*- •#*> *nf-^r i •. .<2 -r — .
w. **1 ,£ igf. * ' V "* *-(?' '\$L~ "' ••''
*»r- «•*£:-- ,% J
&i'2
%• K
JOB
fi.'- - . • :"3-, ^ , - |
"; '••" a^J.k*-> "'•«•»• «•"'•%, '.'- v i:5 «"„ - *»a3f 1^.« /j
'~». t jP&Jr'i /;'•*»' " i .* •. ' : .5»^";»r^L(
S .W,A^'.'A; -. -r I" V^J» SSt **%W^.t
m
Wei*
; i*» w.x.. % f r. , ^ ». >/¦ ¦ I ^ \
wv'W •..."sv»i-..•"• ^-»-«..« '»? * .< -_< .«• -
-------
Figures 8a-d. Photomicrographs of histological sections
(6 W,H & E) of the livers of Rivulus
warmoratus; larvae were exposed for 12
weeks to 45, 30 or 15 ppm DEN then held 1n
uncontaminated conditions for 1 year,
(a & b} Trabecular pattern hepatocellular
carcinoma, x20G and 400, respectively,
(c & d) Glandular pattern hepatocellular
carcinoma, x2QG and 400, respectively
33
-------
ali'S i:SKw.
? r5 ?«.;»«;•
a"»*£y
*ah5"¦*'(* & 'f^Si&Wc *zaz ~ •
s&V
fUidfek
%***vpi
R ' aJb A1
£'<>*~
¦A**®* •
W
M»y ^ V x •mstnn*. mm
'i^-rw*"'
I?
*«?*>* «i««< •
V£*"i #
* .Tut!*" pi^'
>.~ ,M? 4.wjs»
' «L* A.*¦ V **. *- -~ r • '?*
*€*:-*r MisZWtsi mkJ
I>^»
S% *' V?f *4^ '¦% *¦" ¥v '*
«.. « £ #,V :C/-A.- V"'. -*' •»
• %%»«-Jf ;*«».\»'«*k«v IV,.'-i** ¦ -.», -4, . .
sfe* .t3.1i-Muv/:' ' •{•i^"" *"- *¦ '*'
jvSV4v3* *'iA«
r
•:V-. •..•--- ...
<¦<*!'.JZM
it# *i & v ?"/
v*"/v £>
'/?K
• *«»•• a*" ».• r
'$ »«*»¦ '*¦*i»»«t* ?
w'Wm
>*3.0>,u
¦/»¦? A " r-V»- > *; *-- $:•*>*.•-'/.'% \ fi
»-»>•#.«?** s;y,w,<¦ -"v. .J•¦¦«•'. t -A
J> *V »*C<«V. ;,;'/* SV»1 W«> ~» V '•»<«
^r:.-C'n:-;vv^s ^ •*.•
' V
LfCM
§^cp
»»•; »*\ bv. ?•»?•' <»•
r#; vr•SA' ?c /*2S» J i»« ¦*,
!- •;..,V-JirVv^''¦ A-TO"' '
»»¦*i 't,^*
>* (r-*e:
-------
Figures 9a-d. Photomicrographs of histological sections
(6 pm, H & E) of the livers of Rivulus
marmoratus; larvae were exposed to 45, 30
or 15 ppm DEN then held for 1 year in
uncontaninated water. Sections show
distinct boundries between solid pattern
tumor (a) arid normal tissue{above)and
trabecular pattern tumor (b) and normal
tissue {above} x2Q0, Indistinct boundries
occur between glandular pattern tumor
(c I d) and surrounding normal
parenchyma, x4G0 and 200, respectively.
40,
-------
t'tH.
jp • i, ¦ *> *« , v *« If* 4 * ,* "
.«»; \ .«kV»l*,i' . f ift V t **.- i»'« w *••• :.,uiW.-
f«^ip
s ssa
!rV.
«*«W
41
-------
Figures 10a-d. Photomicrographs of histological sections
(6 H & E) of the 11 vers of Rivulus
marmoratus; larvae were exposed to 45, 30,
15 ppra DEN for 12 weeks then held for 1 year
in uncontarinated water. (a S b) tumor
composed chiefly of spindle-shaped cells,
Ml00 and 400 respectively, (c & d)
cholangioma, x200 and 400, respectively.
42
-------
* «¦ £.
?gipr rt-H4- vi* ~ ^
•« ,.:'.^.vs*,& »*/*. ?»c*.-
43
-------
Figures lla-c. Photomicrographs of histological sections
(6 H & E) of the livers of Rivulus
marmoratus; larvae were exposedto 45, 30,
or 15 ppn DEN for 12 weeks then held for
1 year in uncontaninated water, (a & b)
Adenofibrosis, x 100 and 200, respectively,
(c) Granuloma.
44
-------
L****. U:s:
' .4.1 V &S4 «frU> VTV*'*A*v** *-.v Vi-I #_
v:**.P,-;.
•A /,>->»»***. ¦ ' ^r>^*sk:
/'//': >;~ • ^\a'• *£r£V^
1
Ju f
ft
^•^SSSsfi^f
45
-------
by smaller satellite ducts. The satellite ducts are considered metaplastic
duct formations from the liver parenchyma rather than bile duct prolifer-
ations. Most ducts were ensheathed by fibrous tissue which was also some-
times abundant in the interductule spaces. These lesions were interpreted
as adencfibrosis. Ducts within both types of ductule lesions, cholangioma
and adenofibrosis, contained occasional concretions of unknown composition.
Chronic inflammatory lesions included granuloma (Fig. lie), focal
collections of epithelioid leukocytes with a fibrous sheath. In sane
livers, chronic inflammatory lesions were diffuse rather than focal as in
granuloma. These often contained larger numbers of eosinophils. Macrophage
centers composed of foamy macrophages containing a golden pigment were
also fairly common.
There was no clear dose-response among the various pathologic changes;
however, the incidence of hepatocellular carcinoma generally increased with
exposure level for both larvae (Table 5) and adults (Table 6). Generally,
1n the two highest exposure levels of the larval experiment tumors tended to
be large, occupying nearly all of the liver on the sections. Also, the
numbers of tumors per liver section generally increased with dose level.
No pathologic changes were observed in the livers of fish exposed to
DEN as embryos.
No neoplastic changes were found in organs other than the 1ivers
although a complete examination has not been done.
DISCUSSION
This study clearly shows that DEN, a potent liver carcinogen in
experimental mammals and fish, induces hepatocellular carcinoma, among other
pathologic changes, in the liver of rivulus. Because of limited numbers
of animals available for this study a staggered termination of the experi-
ment was not done. A latency period of one year was chosen to avoid the
ambiguity involved in the characterization of early neoplastic foci.
Therefore, it is not known whether clear determination of hepatocellular
carcinoma could be made with a shorter induction period.
The lack of hepatocellular carcinoma in rivulus treated with DEN as
embryos is difficult to explain as fish embryos are considered more sensitive
to hepatocarcinogens than are larvae or adults. For example, Wales et al.
(1978} induced liver cancer in rainbow trout by exposing fertilized eggs
to aflatoxin Bi (0.5 ppm) solutions for 1 hour. The tumor incidence,
determined 1 year after exposure, increased as the age of the embryo,
during which the exposure was conducted, increased with the most dramatic
rise occur ring near the time of liver bud formation. Perhaps the lack of
response of rivulus in this study relates to the possible impermeability of
fish eggs tu DEN, as suggested by Hendricks et_al_. (1980). However, in a
later experiment liver tumors were induced in trout by increasing the embryo
exposure level (0. D. Hendricks, personal communication).
46
-------
Table 5, Pathologic Changes in Livers of Rivulus mamoratus larvae Exposed to Diethylnitrosamine
for 12 Weeks
Nominal
Water
Cones.
M/ )
Incidence**
No. No. Fish Hepato-
Fish Fish with cellular Incipient Adeno-
(Start) (End)3 Neoplasms" Carcinoma Neoplasms Choiangioma fibrosis Granuloma
0
30
23
0/23(0)
0/23(0) 0/23(0)
0/23(0)
0/23(0)
0/23(0)
IS
30
18
17/18(94.4)
12/18(66.7) 7/18(38.9)
4/18(22.2)
16/18(88.9)
11/18(61.1)
30
30
14
12/14(85.7)
9/14(64.3) 4/14(28.6)
1/14(7.1)
6/14(42.9)
7/14(50.0)
45
30
17
16/17(94.1)
13/17(76.5) 4/17(23.5)
7/17(41.1)
14/17(82.4)
7/17(41.2)
aFish were killed one year after end of exposure period.
bNo. of fish with neoplasms (hepatocellular carcinoma, cholangioma or incipient neoplasms)/total
no. of fish.
cNo. of fish with pathologic change/total no. fish.
dNos. in parentheses are percents.
-------
Table 6. Pathologic Changes in Livers of Rivulus marmoratus Adults Exposed to Diethylnitrosamine
for 5 Weeks
Incidence0
Nominal
Water
Cones.
(mi/ )
No.
Fish
(Start)
No.
Fish3
(End)
Fish
with .
Neoplasms"
Hepato-
cellular
Carcinoma
Incipient
Neoplasms
Cholanqioma
Adeno-
fibrosis
Granuloma
0
30
23
0/23(0)d
0/23(0)
0/23(0)
0/23(0)
0/23(0)
0/23(0)
15
30
20
0/20(0)
0/20(0)
0/20(0)
0'20(0)
1/20(5.0)
0/20(0)
30
30
21
7/21(33.3)
6/21(28.6)
2/21(9.5)
l/i i (4.8)
13/21(61.9)
5/21(23.8)
45
30
16
5/16(31.3)
5/16(31.3)
0/16(0)
0/16(0)
8/16(50.0)
2/16(12.5)
aFish were killed one year after end of exposure period,
bNo. of fish with neoplasms (hepatocellular carcinoma, cholangloma, or Incipient neoplasms)/tota1
no. of fish.
cNo. of fish with pathologic change/total no. of fish.
^Nos. in parentheses are percents.
-------
The response of rivulus to DEN was not nearly as dramatic as that of the
raedaka. Our exposure levels (45, 30, 15 ppm) were the same as Ishikawa and
Takayama's (1979) and the exposure durations and conditions were somewhat
comparable; however, the medaka in their experiment showed turners as early
as 5-8 weeks after exposure to DEN ended and tumors were obvious on gross
examination after 13 weeks. Medaka usually died of massive abdominal tumors
after 20 weeks.
Although the DEN induced pathologic changes in Jivers of rivulus are
similar to those of other experimental animals (Schmahl and Osswald, 1967),
rivulus has several distinct advantages as a test animal for studies
involving carcinogenesis. First, genetic uniformity of "lones eliminates
genetic variability as a source of experimental error thus giving better
comparability of data between experiments and between laboratories
(Sonntag, 1977). Because self-fertilization is the natural mode of repro-
duction, no special techniques are necessary to maintain isogenicity.
Second, studies involving tissue transplants are possible. Third, the full
life cycle is carried out
-------
SECTION 7
TERATOGENICITY BIOASSAY AND CHROMIC FULL LIFE CYCLE EXPOSURE OF
RIVULUS MARMORATUS TO SUBLETHAL LEVELS OF SELECTED TOXICANTS
INTRODUCTION
The coastal and marine environments have become contaminated with a
wide variety of organic and inorganic pollutants (Whittle et aK, 1977).
Many of these substances are not only toxic but are known carcinogens,
mutagens, and teratogens! however, the potential effects (or the interfiling
effects) of the vast majority are unknown (Kraybill et al.» 1977). One way
in which to gain insight into the potential effects of tFese contaminants
on marine organisms (and indirectly on the human population) is to establish
a number of sensitive yet simple bioassays. The purpose of this study is to
explore the feasibility of using the marine fish Rivulus irarmoratus as a
bioassay animal in the evaluation of the teratogenic and chronic sublethal
effects of selected organic pollutants.
The advantages of using the self-fertilizing rivulus in this type of
bioassay are; 1, homozygous, isogenic lines (Kallman and Harrington, 1964;
and Harrington and Kallnan, 1968) should allow a high degree of uniformity
and reproducability in experimental results; 2, sensitivity to environmen-
tally (temperatures) induced skeletal abnormalities (Harrington and
Crossman, 1976) provides a convenient en^point for the determination of
teratogenic potential of various pollutants; 3, the extreme hardiness of
rivulus under simplest of culture conditions (stackable glass culture
dishes) allows full life cycle exposures to moderately high levels of various
pollutants without excessive mortality, thus allowing observation of
pathologic changes (Koenig and McLean, 1980). This combination of sensiti-
vity and hardiness makes rivulus particularly attractive as a bioassay
animal.
The selected chemicals used in this study are pentachlorophenol, 2, 3,
4, {j-tetrachl orophenol, 2, 3, b-trichlorophenoV, di-n-butyl phthalate and
bromoform. Pentachlorophenol and 2, 3, 4, 6-tetrachlorophenol and their
salts are widely used as biocides in the paint, wood and textile industries.
They enter the marine environment directly as biocides in drilling and
completion fluids for oil well drilling (see Rao, 1978). Also many
chlorophenols are inadvertently produced in the chlorination of sewage
effluents (Jolley, 1975).. The toxicity, bioaccumulation potential and
various other effects of pentachlorophenol and related compounds on aquatic
organisms have been investigated (Tagatz et al_., 1977; Pruitt et al_., 1977;
Glickman et ah , 1977; Rao, 1978; Iwama and Greer, 1979; Manumante and
50
-------
Kulkarni, 1979). Di-n-butyl phalate (DBF) is used abundantly as a
plasticizer in the manufacture of polyvinyl chloride and other plastics.
It has been found in marine (Giam et ah , 1978) and freshwater (Mayer et ah,
1972) environments and has been shown to be teratogenic in rats (Singh'et al.,
1972;. The toxicity, bi©accumulation potential and various other effects of
DBP on aquatic organisms have been investigated (Sanders et al., 1973;
Pfuderer and Francis, 1975; Laugh! in et ah, 1978; see ZitRbT^972 and
Autian, 1973 for reviews). Marine bromoform contamination results primarily
from the chlorination of seawater such as waters in cooling power plant
condensers (Carpenter et al», 1980). Bi©accumulation potential and toxicity
to various aquatic organisms has been investigated (Gibson et ah• 1980;
Trabalka et ah, 1980).
MATERIALS AND METHODS
All culture methods and conditions were as described under Section 2
"Laboratory Culture of Rivulus marmoratus".
Preliminary LC50 Tests
Sublethal water concentrations for the full life cycle chronic study
were chosen as fractions (0.20, 0.10, 0.05) of static 96-hr LC50 tests
(American Public Health Association, 1976) on newly hatched postlarvae of
Fundulus heteroclitus and Rivulus mannoratus. The term postlarvae as used
here refers to hatch!ings less than two weeks old. Because of the diffi-
culties initially in obtaining large numbers of postlarval rivulus in
similar stages of development F. heteroclitus, a species in the same family,
was substituted for the LC50 tests and chronic exposure levels were based
on these. Later, smaller numbers of rivulus were tested to corroborate the
IC50 data obtained initially on IF. heteroclitus. Postlarvae were chosen
because fish larvae are generally more sensitive to toxic effects than are
either adults or eggs (except in the sense of Danil 'chenko, 1977). Toxic
concentrations based on toxicity to larval stages would therefore probably
be sublethal for full life cycle exposures yet high enough to establish a
quantifiable frequency of chronic effects. Postlarval F. heteroclitus were
obtained from adults collected during the breeding season (March September).
After capture, the fish were brought to the laboratory and injected at the
base of the dorsal fin with 50 ui (50 IU) of human chorionic gonadotropin
in saline solution to stimulate gonad development. The fish were then
separated by sex and held under a 14 hr photoperiod at about 25 C. Five
to six days after injection the testes of several males were removed and
minced in 10 mfc of sea water (28 o/oo) in a 11.4 cm diam culture dish.
Eggs v/ere immediately stripped from females directly onto the sperm
suspension. After about 10 min of gentle agitation the eggs were washed with
fresh seawater and placed on moistened filter paper pads (5 layers of
Whatman"* no. 2) in 11.4 cm diam culture dishes. Dishes were covered to
prevent dessication and kept in an environator at 25 C and a 14 hr photo-
period. After about 10-12 days of incubation embryos were hatched by
flooding them with freshly collected water from Charleston harbor. Larvae
were fed brine shrimp nauplii until they were used for the toxicity experi-
ments.
51
-------
Pentachlorophenol (99%) was purchased from Eastman-Kodak Chemical
Company. 2, 3, 4, 6-tetracnlorophenol (technical grade), 2, 3, 5-trichloro-
phenol (971) and dibutyl phthalate (99%) were purchased from Aldrich Chemical
Company and bromoform (99%) was purchased from Fisher Scientific Company,
Static LC50 tests were run in 11,4 cm diam culture dishes for non-
volatile chemicals or air-tight specimen jars of about the same volume for
bromoform. Preliminary gas chromatographic analyses were performed to
determine loss of bromoform from a 10 mg/t solution over time. In tightly
stoppered containers with little air space there ! ss less than 102 loss over
24 hrs. The high concentrations of bromoform nec^sary for LC50 determina-
tions were formed by sonification. All test chemical solutions were made
fresh and changed every 24 hrs and postlarvae were fed brine shrimp nauplii
daily to simulate conditions in the chronic study and to avoid cannabalism.
A maximum of 10 postlarvae were randomly assigned to each container; however,
there were as many as 5 containers per concentration of toxicant. Bromoform
was most difficult to assess in terms of static LC50 because of its
volatility and also because it narcotized the fish; that is, postlarvae
judged as dead would regain equilibrium and activity soon after being placed
in clean water.
Whenever possible LC50 tests were done on rivulus postlarvae from the
same clone;however, it was later noted that differences in response between
clones were insignificant relative to other variables involved with the
tests. Probit analysis (Finney, 1971) was used when possible to evaluate
dose-mortality curves otherwise graphic methods were used.
Chronic Exposures
For chronic full life cycle studies, six adult rivulus (2 from each
clone) were randomly (random number table) selected for each of three sub-
lethal concentrations for each of the five chemical toxicants (90 fish).
Also, twelve were randomly chosen for the uncontaminated water control and
six for the acetone contaminated control (8.4 mg/£ acetone). Acetone was
used as a carrier and never exceeded 8.4 mg/*. Except for bromoform
exposed fish all were held singly in reproduction chambers (see Section 2)
and solutions (contaminated and control) were made fresh and changed weekly.
Bromoform exposed fish were held singly in air-tight reproduction chambers
fashioned from specimen jars. Bromoform solutions were made fresh and
changed every time the air-tight containers were opened for feeding or egg
collection (3 times per week). Offspring held at the same exposure levels
as parental adults were treated the same.
Parental adults were exposed from 20 October 1978 until 20 June 1979.
Eggs were collected three times a week and were randomly assigned to uncon-
taminated culture water or to culture water of the same type and degree of
contamination as that of parental adults. This was done primarily to allow
distinction between congenital effects (teratogenesis) and post-hatching
developmental effects or other chronic effects such as pathologic changes
resulting from direct sublethal exposure. On 20 September 1979 the
experiment was terminated and all offspring were preserved in either 10%
52
-------
seawater buffered formalin or Bouins fixative (about half in each). Thus,
the youngest offspring were 3 months old. After one week fixed fish were
transferred permanently to ethyl alcohol (70%) after several changes to wash
out fixative. In retrospect, 101 formalin is the superior fixative for it
allows histological procedures as well as clearing and staining procedures.
Standard histological procedures were used. Paraffin sections were cut 6 u»n
thick and stained with hematoxylin and eosin (H&E). Five to ten Fj fish
were randomly chosen for histological analysis from each of the various
contaminated and uncontaminated conditions. All remaining specimens fixed
in formalin were cleared and stained by the method of Dingerkus and Uhler
(1977). Any histopathologic changes and skeletal abnormalities were
recorded. During the course of the chronic study, records were kept on
adult, larval and embryo mortalities and on gross abnormalities. All embryos
were staged and any developmental abnormalities were described and recorded.
RESULTS
A summary of the static 96-hr LC50 data for Fundulus heteroclitus post-
larvae is presented in Table 7. Pentachlorophenol was most toxic followed
by tetrachlorophenol, trichlorophenol and dibutyl phthalate, respectively.
Bromoform was least toxic and had a narcotic effect on postlarvae. For
this reason the I.C50 was difficult Co determine. Narcotized fish trans-
ferred from bromoform solution to uncontaminated water would often revive.
A summary of the static 96-hr IG50 data for Rivulus marmoratus post-
larvae is given in Table 8. The same order of toxicity occurred; however,
in all cases rivulus LC50's were lower than F. heteroclitus LC501s. This
suggests that rivulus ir, more sensitive to the toxic effects of these
chemicals; however, the difference is not statistically significant (p >.05).
Thus, the LC50's for rivulus corroborate those for F. heteroclitus.
It is apparent from Table 8 that the response of rivulus postlarvae is
highly uniform; that is, groups of individuals behave as if they were a
single individual and genetically they are. A series of concentrations must
be very tight in order to observe percent responses other than 0 and 100.
In other words, probit lines are nearly vertical. This is a desirable
characteristic because precision is high.
Tables 9 through 14 summarize the results of chronic exposures of
rivulus: controls, pentachlorophenol (POP), 2,3,4,6-tetrachlorophenol
(TECP), 2,3,5-trichlorophenol (TRCP), dibutyl phthalate (DBF) and bromoform
(BRO) are given, respectively. The number of viable eggs produced was quite
variable among the various exposures. This reflects both the variability of
total egg production among individual fisn and variability due to production
of presumably infertile eggs. Productlon or infertile eggs varied from 0%
to 701 among different individuals but overall averaged 22.6%. Another
source of variability relates to the original design of the reproduction
chambers which was faulty and allowed fish to eat a larger proportion of
their eggs. The improved design (given in Section 2 of this report)
eliminated this as a significant problem.
53
-------
Table 7. Static 96-Hr LC50 Data for Postlarvae of Fundulus heterocHtus and Prescribed Concentrations
for Chronic Study with Rivalus marmoratus
Chemical
LC50(mg/£)a,c
Nb
95% Confidence
Limits®
Chronic Exposure
Levels (mg/i)
Pentachlorophenol
0.370
50
0.195 - 0.535
0.074, 0.037, 0.0185
2,3,4,6-Tetrachlorophenol
1.10
50
0.540 - 1.87
0.220, 0.110. 0.055
2,3,5-Trichlorophenol
1.80
50
1.41 - 2.40
0.360, 0.180, 0.090
Dibutyl Phthalate
3.70
50
1.50 - 4.73
0.740, 0.370, 0.185
Bronoform
42.0
30
23.0 - 60.0
8.40, 4.20, 2.10
3Dose-mortality evaluated by'problt analysis.
h
Number of postlarvae per concentration; minimum of 5 concentrations per chemical.
cAcetone controls and uncontaminated controls (50 postlarvae each) had no mortalities.
-------
Table 8. Static 96-Hr LC50 Data for Postlarvae of Rivulus marmoratus
Concentration
Mortality
LC50a
Chemical
(mg/£)
N
{%)
Pentachlorophenol
0.70
10
100
0.50
10
100
0.35
10
100
0.25
10
70
0.24
10
10
0.22
10
0
0.245
2,3,4,6-Tetrachlorophenol
1.44
10
100
1.00
10
100
0.91
10
100
0.84
10
100
0.79
10
90
0.70
10
0
0.50
10
0
0.76
2,3,5-Trichlorcphenol
2.6
10
100
1.8
10
100
1.7
10
100
1.6
10
20
1.5
10
10
1.3
10
0 .
1.62
Dibutyl Phthalate
5.0
8
100
3.7
8
75
2.7
8
0
3.6
Bromoform^
60.0
10
,100
42.0
10
100
30.0
10
100
21.0
10
0
15.0
10
0
11.0
10
0
25.0
Control - 1
0.0
10
0
Control - 2
Acetone
10
0
aLC50 determined graphically
bTest run in air-tight containers
55
-------
Table9. Summary of Mortality and Skeletal Abnormality Data for
Rivulus marmoratus Used as Controls for Chronic
Exposure Study
Nominal Water Concentration
(mg/i)
Number of Adults
(20 Oct 1978)
Number of Adults
(20 June 1979)*
Number of Viable Eggs'*
Number Hatched
Hatch (%)
Number of Offspring"
Number of Mortalities
Mortality {%)
Number of Skeletal
Abnormalities per Fish
Examined
Skeletal Abnormalities
(X)
Acetone
0.00
8.4
12
6
10
6
250
172
167
131
66.8
76.2
143
42
29.4
44
18
x40.9
Abnormalities
14/47
29,8
2/10
20.0
8.4
Acetone-C
87,
42
48.3
1/8
12.5
aParental adults were reared from 20 October 1978 through 20 June 1979
but offspring were reared until, 20 September 1979.
bEggs were judged viable or non-viable (unfertilized?) at time of
collection.
cParental adults were exposed to concentrations indicated but offspring
were reared in uncontaminatsd water.
^Offspring not accounted for were damaged while handling, used for
histology or lost while changing solutions.
56
-------
Table 10. Summary of Mortality and Skeletal Abnormality Data for
Rivulus marmorutus from Chronic Exposure to
Pentachlorophenol(PCP)
Nominal Water Concentrations
(mg/K)
Number of Adults 0,074 0,037 0,0185
(20 Oct 1978) 6 6 6
Number of Adults
(20 June 1979}® S 4 6
Number of Viable Eggsb 160 128 234
Number Hatched 138 112 186
Hatch (%) 86.3 87.5 79.5
0.074 0.037 0.0185
PCP-C c PCP-C c PCP-C c
Number of Offspring1* 57 54 36 34 103 67
Number of Mortalities 34 15 10 21 45 40
Mortality {%) 59.6 27.8 37.8 61.8 43.7 59.7
Abnormalities
Number of Skeletal
Abnormalities per
Fish Examined 4/11 8/14 4/17 3/8 20/33 10/15
Skeletal Abnormalities
{«) 36.4 57.1 23.5 , 37.5 60.6 66,7
aParental adults were exposed from 20 October 1978 through 20 June 1979
but offspring were exposed until 20 September 1979,
bEggs were judged viable or non-viable (unfertilized?) at time of
collection.
cParental adults exposed to concentrations indicated but offspring
reared in uncontaminated water.
^Offspring not accounted for were damaged while handling, used for
histology or lost while changing solutions.
57
-------
Table 11, Surmary of Mortality and Skeletal Abnormality Data for
Rivulus marnoratus from Chronic Exposure to 2,3,4,
6-Tetrachlorophenol(TECP)
Nominal Water Concentration
(mg/2.)
0.H20
o.no
0.055
Number of Adults
(20 Oct 1978)
Number of Adults
(20 June 1979)a
6
5
6
5
6
5
Number of Viable Eggsb
Number Hatched
Hatch {%)
119
99
83.2
0.220 „
TECP-C C
261
211
80.8
0.110 r
TECP-C
203
166
81.8
0.055
TECP-C
Number of Offspring^
Number of Mortalities
Mortality {%)
52
33
63.5
29
8
27.6
101
58
57.4
81
29
35.8
62
20
32.3
74
34
45.9
Abnormalities
Number of Skeletal
Abnormalities Per
Fish Examined 7/8 7/14 16/21 13/28 10/20 4/20
Skeletal Abnormalities©
(35) 87.5 50.0 7S.2 46.4 50.0 20.0
®Parental adults were exposed from 20 October 1978 through 20 June 1979
but offspring were exposed until 20 September 1979.
bEggs were judged viable or non-viable (unfertilized?) at time of
collection.
cParental adults were exposed to concentrations indicated but these
offspring were reared in uncontaminated water.
^Offspring not accounted for were damaged while handling, used for
histology or lost while changing solutions.
eFin erosion included under skeletal anomalies.
58
-------
Table 12.Summary of Mortality and Skeletal Abnormality Data for
Rivulus marmoratus fforo Chronic Exposure
to 2,3,5-Trichlorophenol(TRCP)
Nominal Water Concentration
(rag/*)
0.36
0.18
0.09
Number of Adults
(20 Oct 1978}
6
6
6
Number of Adults
(20 June 1979)a
3 ,
5
6
Number of Viable Eggs'1
105
204
196
Number Hatched
92
183
184
Hatch {%)
87.6
89.7
93.9
0.36
0.18
0.09
TRCP-C c
TRCP-C c
TRCP-C c
Number of Offspring*1
52
40
112
71
87
91
Number of Mortalities
8
7
18
8
22
23
Mortality (X)
15.4
17.5
16.1
11.3
25.3
25.3
Abnormalities
Number of Skeletal
\
Abnormalities per
Fish Examined
8/21
1/15
15/79
8/29
12/54
11/33
Skeletal Abnormalities
{%)
38.1
33.3
19.0
27.6
22.2
33.3
aParental adults were exposed from 20 October 1978 through 20 June 1979
but offspring were exposed until 20 September 1979.
''Eggs were judged viable or non-viable (unfertilized?) at time of
collection.
*
cParental adults ware exposed to concentrations Indicated but these
off sprite.' were reared in uncontarainated water.
^Offspring not accounted for were damaged while handling, used for
histology or lost while changing solutions.
59
-------
Table 13, Summary of Mortality and Skeletal Abnormality Data for
Rivulus mannoratus from Chronic Exposure
to 01 butyl Phthalate (DSP)
Nominal Water Concentrations
(mg/a)
0.740 0,370 0.185
Number of Adults
(20 Oct 1978) 6 g 6
Number of Adults
(20 June 1979)a 4 5 5
Number of Viable Egosb 186 288 222
Number Hatched 139 240 171
Hatch {%) 74.7 83.3 77.0
0,740 0.370 0.185
DBP-C c DBP-C c D8P-C c
Number of Offspring** 74 47 114 96 90 55
Number of Mortalities 38 30 35 41 51 31
Mortality (%) 51,4 63.8 30.7 42.7 56.7 56.4
Abnormalities
Number of Skeletal
Abnormalities per
Fish Examined 7/10 7/9 32/60 ¦ 19/34 10/20 3/13
Skeletal Abnormalities
(*> 70.0 77.7 53.3 55.9 50.0 23.1
aParental adults were exposed from 20 October 1978 through 20 June 1979
but offspring were exposed until 20 September 1979.
^Eggs were judged viable or non-viable (unfertilized?) at time of
collection.
cParental adults exposed to concentrations indicated but offspring
reared in uncontaminated water.
^Offspring not accounted for were damaged while handling, used for
histology or lost while changing solutions.
60
-------
Table 14. Summary of Mortality and Skeletal Abnormality Data for
Rivulus marmoratus from Chronic Exposure
to Bronoform(BRO)
Nominal Water Concentration
(nig/*)
8,4
4.2
2.1
Number of Adults
<20 Oct 1978)
6
6
6
Number of Adults
(20 June 1979)®
6
5
4
Number of Viable Eggsb
2S0
169
238
I'umber Hatched
197
109
171
Hatch {%)
78,8
64.5
71.8
8.4 r
4.2 c
2.1 _
BRO-C c
BRO-C
BRO-C C
Number of Offspring**
88
100
45
60
68
101
Number of Mortalities
26
23
19
15
34
38
Mortality {%)
29,5
23.0
42.2
25.0
50.0
37.6
Abnormalities
Number of Skeletal
Abnormalities per
Fish Examined
9/24
11/33
12/34
5/19
7/14
12/27
Skeletal Abnormalities
(%)
37. S
33.3
35.3
26.3
50.0
44.4
aParental adults were exposed from 20 October 1978 through 20 June 1979
but offspring were exposed until 20 September 1979.
%ggs were judged viable or non-viable (unfertilized?) at time of
collection,
c ^
Parental adults exposed to concentrations indicated but offspring
reared in uncontaminated water.
^Offspring not accounted for were damaged while handling, used for
histology or lost while changing solutions.
61
-------
The percent of viable eggs that hatched varied from about 70 to 90%
overall and was toward the lower end of this range in controls.
By far the dominant abnormality among embryos consisted of, or was
related to, an edematous pericardial cavity (Fig. 12), In its most
exaggerated form the pericardium ballooned around the head and often con-
tained a flocculent material of unknown composition. Also the heart appeared
string-like and was inefficient in pumping blood. Less exaggerated examples
had a slight bulge in the pericardium with no discernable alteration in the
heart. This abnormality always resulted in abnormal growth and development
of the embryo. Extreme cases never hatched whereas moderately affected
embryos were often stunted with curved or bent nonfunctional tails. There
was no particular pattern of this expression apparent among the various
chemical exposures or controls (Table 15); therefore} this abnormality did
not result from the exposures. Other abnormalities included blastulae with
irregular blastomeres, bubble-like expansions of the vitelline membrane and
twinning (occurred once) but these were quite infrequent and showed no
particular pattern among the various exposures or controls.
Mortality rates among post-hatch offspring were also variable.
Mortality in controls varied from about 30 to 50% as did mortalities of off-
spring from the bromoform and dibutyl phthalate exposure. Mortality rates
of offspring of some of the groups from the PCP and TECP exposures went as
high as about 60%. In contrast the mortality rate of offspring from the
TRCP exposures were consistently low (ca. 10-2556).
The frequency of skeletal abnormalities varied somewhat over the various
exposure series; however, certain trends are evident. Rates of skeletal
abnormality in controls were about 30%. Clear dose-response rates were
evident in offspring of TECP and D8P exposures.
Skeletal abnormalities observed in cleared and stained fish were
divided into six categories (Table 16 to 25): vertebral fusion, deformed
centra, abnormal neural and/or hemal spines, abnormal dorsal fins, abnormal
pectoral fins, and abnormal pelvic fins. Vertebral fusion is the joining
of adjacent centra into a solid immovable structure without discernable
articulation faces (compare Figs. 13a and 13b). Deformed centra or kinks
appear to result from the fusion of adjacent neural or hemal spines, arches
or zygapophyses. The articulation faces of the vertebrae are thus held askew
and a sharp bend or kink in the spinal column results (Figs. 14a-c). Neural
and hemal spine deformities include bilateral splits, fusions, disorienta-
tions or total lack of spines (see Figs. 14a-c and 15a, b). In cleared and
stained specimens the distinction between abnormal fins and fin erosion is
clear. Parts or all of the supporting skeleton are lacking in deformed fins
whereas eroded fins lack parts or all of the fin rays. The erosion process
generally proceeds from the tips of the rays toward the base (Figs, 16a-c).
Pterygiophores of abnormal dorsal fins may be lacking, disoriented or fused
(Figs. 17a-c). Abnormal pectoral fins generally have a withered appearance
with few rays. Also, supporting radials, scapula and the dorsal portion of
the coracoid bones may be partially or completely absent (Figs. 18a, b).
Pelvic fins develop last among fins, approximately two months after hatching,
62
-------
12
Figure 12. Photomicrograph of a late stage embryo of rivulus
Note edematous pericardial cavity (arrow).
63
-------
Table 15. Distribution of Heart Abnormalities in Embryos of Rivulus marrooratus
Exposed Chronically to Various Toxicants
Concentration .
Chemical faq/Jtlr Heart Abnormalities (%)•
Pentachlorophenol 0.074 6.25
0.037 2.34
0.0185 9.40
2,3,4,6-Tetrachlorophenol 0.22 2.52
0.11 3.45
0.05 7.88
2,3,5-Trichlorophenol 0.36 0.00
0.18 0.98
0.09 2.04
D1butyl phthalate 0.74 6.45
0.37 5.90
0.185 8.56
Bromoform 8.4 1.20
4.2 4.14
2.1 2.52
Control 0.00 6.80
Acetone 8.40 2,91
determined at time of egg collection; consisted of expanded peri-
cardial cavity with string-like heart in varying degrees.
64
-------
Table 16, Distribution of Skeletal Abnormalities of Offspring of Rival us
marmoratus Exposed to Pentachlorophenol.a
Vertebral
Fusion
Deformed
Centra
(kink)
Abnormal
Neural and/
or Heraal
Spines
Abnormal
Dorsal Fin
Abnormal
Pectoral
Fins
Abnormal
Pelvic
Fins
Erosion of
Fin Rays
0.0740
l/ll(9.1)c
0./ll{0)
2/11(18.2)
0/11(0)
0/11(0)
0/11(0)
0/11(0)
Skeletal
Abnormalities
(Total) 2/11(18.2)
Nominal Water Concentrations (mgft)
0,0370 0.0185 0.00b
3/17(17.6) 9/31(29.0) 5/65(7.70
0/17(0)
0/17(0)
0/17(0)
0/17(0)
0/17(0)
7/31(22.6) 6/65(9.2)
3/17(17.6) 13/31(41.9) 9/65(13.8)
0/31(0)
0/31(0)
0/65(0)
4/31(12.9) 2/65(3.1)
4/31(12.9) 2/65(3.1)
0/65(0)
3/17(17.6) 20/31(64.5) 17/65(26.2)
Parental adults and offspring reared in indicated concentrations of PCP.
bClean water controls and acetone controls combined.
cNos. in parentheses are percents.
65.
-------
Table 17. Distribution of Skeletal Abnormalities of Offspring of Rivulus
marnoratus; Parental Adults Only were Exposed to Pentachloro-
phenol®.
Vertebral
Fusion
Deformed
Centra
(kink)
Abnormal
Neural and/
or hemal
spines
Abnormal
Dorsal Fins
Abnormal
Pectoral
Fins
Abnormal
Pelvic
Fins
Erosion of
Fin Hays
Skeletal
Abnormalities
(Total)
Horainal Water Concentrations (tug/Si)
0.0740 0.0370 0.0185
3/14(21.4)C 0/8(0) 8/15(53.3)
3/14(21.4) 1/8(12.5) 1/15(6.7)
5/14(35.7) 3/8(37.5) 5/15(33.3)
0/14(0)
0/14(0)
0/14(0)
0/14(0)
0/8(0) 0/15(0)
0/8(0) 0/15(0)
0/8(0) 0/15(0)
0/8(0) 0/15(0)
7/14(50.0) 3/8(37.5) 8/15(53.3)
0.00°
5/65(7.7)
6/65(9.2)
9/65(13.8)
0/65(0)
2/65(3.1)
2/65(3.1)
0/65(0)
17/65(26.2)
Parental adults were held in indicated concentrations of PCP but
offspring were reared in uncontaminated water.
Clean water controls and acetone controls combined.
"Nos. in parentheses are percents.
66
-------
Table 18. Distribution of Skeletal Abnormalities of Offspring of Rivulus
marmoratus Exposed to 2,3,4,6-Tetrachlorophenol.a
0/8(0)
1/21(4.8) 2/20(10.0) 6/65(9.2)
2/8(25.0) 12/21(57.1) 5/20(25.0)
Nominal Water Concentrations (mgfi)
0/220 0.110 0.055 0.00b
Vertebral
Fusion 1/8(12.5)c 10/21(47.6) 5/20(25.0) 5/65(7.7)
Deformed
Centra
(kink)
Abnormal
Neural and/
or Hemal
Spines
Abnormal
Dorsal Fins
Abnormal
Pectoral
Fins
Abnormal
Pelvic
Fins
Erosion of
Fin Rays
Skeletal
Abnormalities
(Total)
0/8(0)
0/21(0)
1/20(5.0)
1/8(12.5) 5/21(23.8) 1/20(5.0)
0/8(0)
1/21(4.8) 0/20(0)
5/8(62.5) 10/21(47.6) 6/20(30.0
9/65(13.8)
0/65(0)
2/65(3.1)
0/65(3.1)
0/65(0)
5/8(62.5) 16/21(76.2) 10/20(50.0) 17/65(26.2)
Parental adults and offspring reared in indicated concentrations of
TECP.
bClean water controls and acetone controls combined.
cHos. in parentheses are percents.
67
-------
Table 19. Distribution of Skeletal Abnormalities of Offspring of Rivulus
marmoratus; Parental Adults Only were Exposed to 2,3,4,6-
Tetrachlorophenol,a
Vertebral
Fusion
Deformed
Centra
(kink)
Abnormal
Neural arid/
or Hemal
Spines
Abnormal
Dorsal Fins
Abnormal
Pectoral
fins
Abnormal
Pelvic
Fins
Erosion of
Fin Rays
Skeletal
Abnormalities
(Total)
Nominal Water Concentrations (mg/t)
0,220 0.110 0.055
6/14(42.9)c 9/28(32,1) 3/20(15.0)
0/14(0)
0/14(0)
0/14(0)
0/14(0)
1/28(3.6) 0/20(0)
4/14(28,6) 9/28(32.1) 3/20(15.0)
1/14(7.1) 0/28(0)
0/28(0)
1/20(5.0)
3/28(10.7) 0/20(0)
1/28(3.6) 0/20(0)
0/20(0)
0.00°
5/65(7.7)
6/65(9.2)
9/65(13.8)
0/55(0)
2/65(3.1)
2/65(3.1)
0/65(0)
7/14(50.0) 13/28(46.4) 4/20(20.0) 17/65(26.2)
aParental adults were held in indicated concentrations of TECP but
offspring were reared in uncontaminated water.
L «
Clean water controls and acetone controls combined.
"Nos. in parentheses are percents.
68
-------
Table 20. Distribution of Skeletal Abnormalities of Offspring of Rivulus
marmoratus Exposed to 2,3,5-Trichlorophenol.
Nominal Water Concentration (rag/i)
0.360
0.180
U.090
0.00b
Vertebral
Fusion
3/21(14.3)c
/79(3.8)
6/54(11.1)
1/65(7.7)
Deformed
Centra
{kink)
1/21(4.8)
1/79(1.3)
0/54(0)
6/65(9.?.)
Abnormal
Neural and/
or Hemal
Spines
6/71(23.8)
15/79(19.0)
9/54(16,7)
9/65(13.8)
Abnormal
Dorsal Fins
0/21(0)
0/79(0)
0/54(0)
(765(0)
Abnormal
Pectoral
Fins
0/21(0)
0/79(0)
0/54(0)
2/65(3.1)
Abnormal
Pelvic
Fins
3/21(14.3)
1/79(1.3)
0/54(0) •
2/65(3.1)
Frosion of
Fin Rays
0/21 (0)
0/79(0)
0/54(0)
0/65(0)
Skeletal
Abnormalities
(Total)
8/21(38.1)
15/79(19.0)
12/54(22.2)
17/65(26.2)
aParental adults and offspring reared in indicated concentrations of
TRCP.
bClean water controls and acetone controls combined.
cNos. in parentheses are percents.
69
-------
Table 21, Distribution of Skeletal Abnormalities of Offspring of Rival us
marmoratus; Parental Adults Only were Exposed to 2,3,5-
frichlorophenol .d
Nominal Water Concentration (mg/i)
0.360 .
0.180
0.090
0.00b
Vertebral
Fusion
3/15(20.0)C .
7/30(23.3)
5/33(15.2)
5/65(7.7)
Deformed
Centra
(kink)
1/15(6.7)
0/30(0)
4/33(12.1)
6/65(9.2)
Abnormal
Neural and/
or Hemal
Spines
0/15(0)
7/30(23.3)
7/33(21.2)
9/65(13.8)
Abnormal
Dorsal Fins
0/15(0)
0/30(0)
0/33(0)
0/65(0)
Abnormal
Pectoral
Fins
2/15(13.3)
2/30(6.7)
\
1/33(3.0)
2/65(3.1)
Abnormal
Pelvic
Fins
0/15(0)
0/30(0)
0/33(0)
2/65(3.1)
Erosion of
Fin Rays
0/15(0)
0/30(0)
0/33(0)
0/65(0)
Skeletal
Abnormalities
(Total)
5/15(33.3)
9/30(30.0)
11/33(33.3)
17/65(26.2)
aParental adults were held in indicated concentrations of TRCP but
offsprine were reared in uncontaminated water.
^Clean water controls and acetone controls combined.
cNos. in parentheses are percents.
70
-------
Table 22, Distribution of Skeletal Abnormalities of Offspring of Rivulus
marmoratus Exposed to Dibutyl Phthalate.a
Nominal Water Concentration, (rag/a)
'
0.740
0.370
0.185
0.00b
Vertebral
Fusion
6/17(35.3)c
15/60(25.0)
5/20(25.0)
5/65(7.7)
Deformed
Centra
(kink)
1/17(5.9)
10/60(16.7)
1/20(5.0)
6/65(9.2)
Abnormal
Neural and/
or Heinal
Spines
4/17(23.5)
13/60(21.7)
9/20(45.0)
9/65(13.8)
Abnormal
Dorsal Fins
0/17(0)
2/60(3.3)
0/20(0)
0/65(0)
Abnormal
Pectoral
Fins
1/17(5.9)
2/60(3.3)
2/20(10.0)
2/65(3.1)
Abnormal
Pelvic
Fins
0/17(0)
0/60(0)
0/20(0)
2/65(3.1)
Erosion of
Fin Rays
0/17(0)
0/60(0)
0/20(0)
0/65(0)
Skeletal
Abnormalities
(Total)
10/17(58.8)
32/60(53.3)
•10/20(50.0)
17/65(26.2)
aParental adults and offspring reared in indicated concentrations of
DBP.
^Clean water* controls and acetone controls combined.
cNos. in parentheses are percents..
71
-------
Table 23, Distribution of Skeletal Abnormalities of Offspring of Rivulus
marffloratus; Parental Adults Only were Exposed to DibutyT ~~
Phthalate.d
Nominal Water Concentration (mg/a)
0.740
0.370
0.185
0.00b
Vertebral
Fusion
5/9(55.5)c
12/34(35.5)
3/13(23.1)
5/65(7.7)
Deformed
Centra
(kink)
0/9(0)
1/34(2.9)
0/13(0)
6/55(9.2)
Abnormal
Neural and/
or Hemal
Spines
3/9(33.3}
10/34(29.4)
2/13(15.4)
9/65(13.8)
Abnormal
Dorsal Fin
0/9(0)
0/34(0)
0/13(0)
0/65(0)
Abnormal
Pectoral
Fins
1/9(11.1)
4/34(11.8)
1/13(7.8)
2/65(3.1)
Abnormal
Pelvic
Fins
0/9(0)
2/34(5.9)
0/13(0)
2/65(3.1)
Erosion of
Fin Pays
0/9(0)
0/34(0)
0/13(0)
0/65(0)
Skeletal
Abnormalities
(Total)
7/9(77.8)
19/34(55.5)
3/13(23.1)
17/65(26.2)
aParental adults were held in indicated concentrations of DBP but
offspring were reared in uncontaminated water.
^Clean water controls and acetone controls combined.
cNos. in parentheses are percents.
72
-------
Table 24. Distribution of Skeletal Abnormalities of Offspring of Rivulus
marmoratus Exposed to Bromoform.3
Vertebral
Fusion
Deformed
Centra
(kink)
Abnormal
Neural and/
or Hemal
Spines
Abnormal
Dorsal Fin
Abnormal
Pectoral
Fins
Abnormal
Pelvic
Fins
Erosion of
Fin Rays ,
Skeletal
Abnormalities
(Total)
Nominal Water Concentration (mg/t)
8.40 4.20 2.10
2/24(8.35° 3/34(8.8) 3/14(21.4)
0/24(0)
0/24(0)
1/34(2.9) 0/14(0)
4/24(16.7) 3/34(8.8) 4/14(28.6)
4/24(16.7) 3/34(8.8) 2/14(14.3)
1/24(4.2) 1/34(2.9) 0/14(0)
1/24(4.2) 6/34(17.6) 0/14(0)
0/34(0)
0/14(0}
0.00°
5/65(7.7)
6/65(9.2)
9/65(13.8)
0/65(0)
2/65(3.1)
2/65(3.1)
0/65(0)
9/24(37.5) 12/34(35.3) 7/14(50.0) 17/65(26.2)
Parental adults and offspring reared in indicated concentration of
bromoform.
Vlean water controls and acetone controls combined.
"Nos. in parentheses are percents.
73
-------
Table 25. Distribution of Skeletal Abnormalities of Offspring of Rivulus
marmoratus; Parental Adults Only were Exposed to Bromofom.3
Nominal Water Concentration (mg/i)
-
8.40
4.20
2.10
0.00b
Vertebral
Fusion
4/33(12.1)c
1/20(5.0)
7/27(25.9)
5/65(7.7)
Deformed
Centra
(kink)
0/33(0)
1/20(5.0)
1/27(3.7)
6/65(9.2)
Abnormal
Neural and/
or Hemal
Spines
6/33(18.2)
5/20(25.0)
5/27(18.5)
9/65(13.8)
Abnormal
Dorsal Fin
0/33(0)
0/20(0)
0/27(0)
0/65(0)
Abnormal
Pectoral
Fins
0/33(0)
1/20(5.0)
0/27(0)
2/65(3.1)
Abnormal
Pelvic
Fins
0/33(0)
0/20(0)
1/27(3.7)
2/65(3.1)
Erosion of
Fin Rays
, 0/33(0)
0/20(0)
0/27(0)
0/65(0)
Skeletal
Abnormalities
(Total)
11/33(33.3)
5/20(25.0)
12/27(44.4)
17/65(26.2)
aParental adults were held in Indicated concentrations of bromoform
but offspring were reared in uncontaminated water.
bClean water controls and acetone controls combined.
cNos. in parentheses are percents.
74
-------
Figure 13 a,b. Photomicrograph of caudal skeletal of
cleared and stained Rivul us marmoratus.
a-Normal caudal skeletal from control
fish. b-Specimen reared in uncontamin-
ated water from parental adult exposed
to 0,740 mg/* dibutyl phthalate (DBP),
Note fusion of four vertebrae (arrow)
and deformed centrum of penultimate
vertebra. Scale applies to a and b.
75-
-------
—-
r';^-.' £$>
>V.;>'% ^rlfe/- S -
»»*r '-¦ ^ ' * ^ssii
,r; %^vl'\5fw'- v "^Ssf
^;£41-,;7^?,..-.a
'*
l"-
"' "L" i/i»ifo»AM'iaii
^,i+>L-..~-*&*,:ii
«rti»*.
? - - \9|- "\v-;- ^ •;^';" %
ft i
,. sVy^
^¦&:- srrf,.-;.
.* •¦¦¦
i :^^SKSSi
76
-------
Figure Ka-c. Photomicrographs of caudal vertebrae of
cleared and stained Rivulus marmoratus.
a-Specimen (and parental adult) was
continuously exposed to 0.370 mg/z
dibutyl phthalate {DBP) for entire life.
Note medial split in neural spine (sns)
and vertebral fusion (vf). b-Specimen
(and parental adult) was continuously
exposed to 0.370 mg/i DBP for entire
life. Note "kinked" vertebrae (deformed
centra) held in position by fused neural
spines (nf) and fused hemal spines (hf).
c-Specimen reared in uncontaminated
water from parental adult exposed to
0,740 usg/a. DIP. Note "kinked" vertebrae
held in position .by fused hemal spines
(hf). Scale applies to a, b and c.
77
-------
~ —' ¦ -¦ — v- "* *7""~ r
f
If -
1 ¦ *" - - " 'Or' ' ¦' ¦
fwir
_
«„ *» -Ml, —. «**-».
><4' \'^V.^t£v-
1 • --/''.- % \
?%.'''
;e»?j ,fw- 7 .y^«i. 1
* > i l *J fj '*• ' ¦*. , '*' « • ,. • ,* \' ' * I, i, » •/.*'¦ -'•~Cljjt * ' . •• •
• J1? «¦&* «*, ,*w* • •>-. >/••*, »• . i-%.-!$?• - •"
". .-"«t A- .f® * * „,.'S«* ««.,4X J ."-. •' jfoWfir % -»
rMfer:-
/, ? .#
'y
78
-------
Figure iga.b. Photomicrographs of trunk vertebrae
of cleared and stained Rivulys
marmoratus. a-Normal vertebrae of
control fish. b-Specimen (and parental
adult) was continuously exposed to
0.185 mg/s. dibutyl phthalate (DEP) for
entire life. Note grossly abnormal
neural spines and poorly formed hemal
spines in exposed fish. Scale applies
to a and b»
79
-------
m-a"hA
-------
Figure 16 a-c. Photomicrographs of cleared and
Stained Rivulus marmoratus. Specimens
(and parental adults) were continuously
exposed to the following concentrations
of 2,3,4,6-tetrachlorophenol for entire
life: a-0.055 rag/a, b and c (same
fish)-0.110 mg/£. Note erosion of
fin rays (arrows), also deformed neural
and hemal spines and vertebral fusion
( a and b). Scale appl1es to a, b
and c.
81
-------
82
I
-------
F1gurel7a-c. Photomicrographs of cleared and stained
Rivulus marmoratus. Specimens (and
their parental adults) were continuously
exposed to the following concentrations
of bromoform for entire life:
a-t.l mg/£» b-4.2 mg/n, c-8.4 mg/fc.
Note partial to complete loss of dorsal
fin elements. a-proximal and distal
pteryglophores (arrow) are in disarray
and fin rays are absent. b-Pterygio-
phores (arrow) are reduced to two and fin
fiys are absent. c-Al.l dorsal fin
elements are lost, neural spines are
deformed and some vertebrae are fused.
83
-------
I l>Omm I
i
I
3
v
i:
' 1 "" » ¦ ¦ ,.a¦> '"¦*' "**- t - r~ *1
¦yi a-» 'i '&v>k9&^vi£K&jt&h-? £&
*&fi :;tK;tv5v.*v>'"j
mi,-1 ¦jUkM.M' UWf
I 2 Q mm t
- -s^*r -%§- •
/ r' r*'^'"'
A» ¦J
' 4
\W\YmHAA-
,/* f. \ \ *•• -< ¦ •
. \
„.-1
- »
: a
«P1
WW!
•<* -rfc
mrm
\ x\ x \ -¦ N a «:
• Y\\. Vr-iiV v.;.::-, S-. *S-:\ *
'rf^T'SP
l*iX*
84
-------
Figure 18a,b. Photomicrographs of pectoral girdles of
cleared and stained Ri vulus Marmoratus,
a-Normal pectoral girdle of control
fish. b-Specimen [and parental adult)
was continuously exposed to 0.370 nig/1
d1butyl phthalate {DBP) for entire life.
Note complete absence of fin rays and
radials and partial loss of scapula and
coracoid bones in exposed fish. Scale
applies to a and b.
85
-------
86
-------
under the culture conditions. Older specimens may lack one or both fins
and their supporting girdles. Severely effected fish often showed multiple
skeletal abnormalities (Figs. 19a-d).
The distribution of skeletal abnormalities in control as compared with
PCP exposures (Tables 16 and 17) indicate no clear dose-response relation-
ship. Although rates of abnormal neural and/or hemal spines were consis-
tently high in offspring from PCP exposed fish reared in clean water
(Table 17), low sample numbers preclude a clear relationship.
Erosion of fin rays was clearly the result of TECP exposure (Table 18).
There was a clear dose-response relationship and fin ray erosion occurred
in no other exposures or controls. That this response was the result of
direct chronic exposure to TECP »s demonstrated by the fact that offspring
from TECP treated adults which were reared in clean water did not show the
response (Table 19). Also, gill damage was evident in TECP treated off-
spring. In most severe cases there was loss of secondary lamellae and
partial to nearly complete loss of gill filaments and a marked epithelial
hyperplasia (compare Figs. 20 a & b). In less severe cases the terminal
portions of the filaments were stunted and the lamellae were irregular in
shape. Degree of gill damage showed a dose-response. Offspring of controls
and offspring reared in clean water from TECP treated adults had normal gill
There are also strong indications that vertebral fusion and abnormal
neural and/or hemal spine rates follow a dose-response. This is especially
evident in Table 19 (i.e., offspring reared in uncontaminated water)
because rates of congenital abnormalities are not altered by the toxic
effects of chronic exposure to TECP (see mortality rate in Table 11).
There are no clear dose-response relationships in the distributions of
skeletal abnormalities in TRCP exposures (Tables 20 and 21) and rates of
abnormalities are similar to control rates.
A teratogenic response from the DBF exposures is evident in the rates
of vertebral fusion and neural and hemal spine deformity (Tables 22 and 23).
Histograms of such effects (Figs. 21 and 22) dramatically illustrate a
dose-response relationship.
There are no clear .ose-response relationships in the distribution of
skeletal abnormalities in bromofor.n exposures (Tables 24 and 25). Except
for dorsal fins, rates of abnormalities were similar.to control rates.
Abnormal dorsal fins were fairly common among offspring exoosed to bromoform
(Table 24); however, there were no abnormal dorsal fins in offspring reared
in uncontaminated water (Table 25).
DISCUSSION
The basic design of this study was aimed at distinguishing between
teratogenic effects, that is, malformations resulting from aonormal
embryological development, and chronic exposure effects, or abnormalities
87
-------
Figure 19a-d. Photomicrographs of cleared and stained
Rivulus rarrooratus. a-Specimen (and
parental adult)was continuously exposed
to 0.740 mg/1 di butyl phthalate {DBP)
for entire life. Note extreme vertebral
fusion, "kinks" and abnormal neural and
hemal spines. b-Speeimen reared in
uncontaminated water from parental adult
exposed to 0.370 mg/i DBP. Note
vertebral fusion in caudal region,
abnormal neural and hemal spines and
reduced pectoral fins. c-Specimen
(and parental adult) was continuously
exposed to 4.2 mg/i bromoform for entire
life. Note vertebral fusion, abnormal
neural and hemal spines, abnormal dorsal
fin pterygiopbores and reduced pectoral
fins. d-Specimen reared in uncontamin-
ated water from parental adult exposed
to 2.1 mg/i bromoform. Skeleton normal.
Scale applies to a,b,c and d.
88
-------
-IT,- "*-- ft
SjSoJ^SV ..
imR&n?iaswit
.'¦i.t.o.'UVw;
. ' ' • < ¦"U>1 lv «F '•'*„ •*!• . .
wm ir» V
m
.AW- *
¦Ss^kJkl,/. ' ¦¦>•¦¦¦¦:¦ f-v.;^
,vV/3Ms£
x ¦ r~ ¦tTi^"-o^
JsiSs^cfs
--'•^«95!Hap«¥- ¦ -W "?'-T
PIS
pjw»j",y.M,S';'«yv^
EUWM&J^iiUdtCHiAy^ -0>
i 2
-------
Figure 20. (a) Photomicrograph of a cross section
through the gill arches (ga) of Rivulus
marmoratus showing normal gill filaments
(gf) and gill lamellae (gl)» and (b) R^.
mamoratus exposed to 2,3,4,6-tetrachloro-
phenol (0.220 mg/jt) for entire life
(4 months). Note complete absence of gill
filaments and lamellae. xlOO.
90
-------
mmm&g
-------
Figure 21. Histogram of percent of various skeletal
abnormalities in offspring of Rivulus
marmoratus» Parental adults and offspring
were exposed chronIcally to dibutyl
phthalate (DBP) in their culture water in
the following concentrations (rag/i):
0.740(h), 0.37O(m), 0.185(1). Parental
adults and offspring of controls (c) were
reared in uncontaminated water.
92
-------
Total Skeletal Vertebral Deformed Abnormal
Abnormalities Fusion Neural Spines Pectoral Fins
-------
Figure 22. Histogram of percent of various skeletal
abnormalities In offspring of Rivulus
mtL-moratus. Parental adults only were
exposed chronically to dibutyl phthalate
(DBP-C) in their culture water in the
following concentrations (ng/z):
0.740(h), 0.370(m), C.185(1 ). Offspring
of exposed parental adults and parental
adults and offspring of controls (c) were
all reared in uncontaminated water.
94
-------
m I c
Total Skeletal
Abnormalities
C
3
Deformed Abnormal
Neural Spines Pectoral Fi ns
-------
caused by continuous exposure of post-embryonic developing offspring to
chemical insult. Exposure concentrations were kept relatively low {20, 10,
and 5X of the larval 96 hr LC50) and all phases of the life cycle were
impacted to simulate contaminated conditions in natural waters. However,
exposure levels were high enough to produce observable frequencies of
effects using a limited number of animals.
During the course of the experiments actual chemical concentrations in
the exposure containers were not deterninod. Solutions of test chemicals
for the LC50 experiments were renewed daily; however, solutions for the
chronic exposures were undoubtedly lower than nominal due to adsorption,
volatilization, and/or breakdown of the test chemicals. For example,
laughi in et al_. (1978) showed nearly complete loss of DBP from a seawater
solution over a 24 hr period. Our exposures can therefore be considered more
pulse-like than continuous. Future experiments of this nature might gain
greater precision by exposure of test animals in a continuous-flow
dilution apparatus.
The percent hatch in control fish was about 672. This was the lowest
hatching rate (except for the 4,2 mg/£ bromoform exposure) among all
experiments. The highest consistent hatching rate was from those fish
exposed to TRCP. The most plausible explanation for this pattern is that
the chemicals had a stimulatory effect on hatching. Many cypHnodontid
fishes delay hatching (see pp.20 and 21 Section 4 of this report) and
rivulus is no exception. Often rivulus embryos would delay hatching to the
point of exhausting yolk reserves and would die within the chorion.
Harrington and Grossman (1976) routinely mechanically dechorionated stage 32
rivulus embryos to avoid this problem. Often embryos could be stimulated
to hatch by changing the incubating medium from the synthetic seawater to
natural seawater. Also, a weak ethanol solution stimulates hatching
(W. P. Davis, personal communication). Rates of skeletal abnormalities and
mortality among TRCP exposed fish and controls were comparable (Tables 10
and 13) but hatching rates differed greatly. At the levels used TRCP may
have been an irritant to the point of stimulating hatching.
Mortality of offspring was high and somewhat variable in all exposures
except TRCP where it was consistently lower than that of controls. In
addition to a possible stimulatory effect on hatching, TRCP may have had a
bacteriocidal effect which could have actually improved culture conditions.
The rate of skeletal abnormality in controls was 29.8% (Table 10).
•This compares well with the data of Lindsey and Harrington (1972) and
Harrington and Crossman (1976). In their studies of temperature induced
meristic variation in rivulus they reported rates of skeletal abnormalities
between 25 and 38% in fish held constantly throughout life at 25-26°C. In
their study, rivulus held at 31°c had skeletal abnormality rates between 87
and 1001. Also, low temperature (ca. 19°C) induced skeletal abnormality
rates between 53 and 882. It is clear then that the spontaneous rate of
skeletal abnormality in the three clones of rivulus held at standard culture
conditions is around 30% and abnormally high and low temperatures Induce
skeletal abnormalities beyond this rate. It is difficult to compare types of
abnormalities between this study and those cited because detailed
-------
descriptions of the abnormalities were not given. Preliminary results with
wild-caught rivulus from near Naples, FL indicate that the spontaneous rate
of skeletal abnormality is near zero; however, the response to temperature
Increase is similar to that of the fish derived from Harrington's clones
(personal observations). Future studies of this nature which utilize wild-
caught clones could be even more fruitful.
In some of the exposures it is difficult, because of low sample
numbers, to discern an increased abnormality rate over controls. Because
skeletal abnormalities ware among several possible responses, too few
animals were retained, in many cases, for clearing and staining procedures.
Also, a number of fish were preserved In Bouin's fluid which decalcifies bone
and precludes effective clearing and staining. However, overall the data
demonstrated the sensitivity of rivulus to environmentally (chemically)
induced skeletal malformations. For example, DBF has been shown to produce
teratogenesis in rats when injected (1p) into pregnant females on the 5th,
10th, and 15th dav of gestation at rates of 333, 20%, and 10% of the LD50
(Singh et al., 1Si2), Rates of skeletal abnormalities in treated rats were
33.3%, 24.Tf and 20.72, respectively; control rats showed no abnormalities.
In rivulus this known teratogen induces several types of skeletal abnormali-
ties (Tables 22 and 23, Figs. 21 and 22). Considering its widespread
occurrance in marine (Giam et al.» 1978) and freshwater (Mayer et al., 1972)
environments, these results couTd be significant, especially in view of the
fact that DBP bioaccumulates in certain aquatic animals to as much as
thousands of times the water concentration (Sanders et al.,1973).
Rates of skeletal abnormalities resulting from TRCP exposure (Tables 20
and 21) are approximately the same as control rates, indicating no terato-
genic effect. Also, there were no overt chronic effects or pathologic
changes resulting from continuous exposure to TRCP.
There are indications that TECP induces skeletal abnormalities,
specifically vertebral fusion and abnormal neural and hemal spines (Tables 18
and 19). This is clearly a teratogenic response because elevated rates of
abnormality appear in both exposed and unexposed offspring from exposed
parental adults. However, erosion of fin rays and gills (Table 18, Figs.
16a-c and 20b) with associated epithelial hyperplasia occurred as the result
of chronic exposure of the offspring to TECP. Fin erosion continued in
some exposed fish until the complete fin structure was lost. Pectoral fins
usually first showed signs of erosion, followed by median fins. TECP was
clearly the primary causative agent; howeveij it is not known whether
secondary causative agents (bacterial, fungal or viral) were also involved.
Gill erosion in its most severe form (Fig. 20b) involved nearly complete
loss of gill filaments with loss of associated lamellae. These fish
undoubtedly survived to this state of extreme loss of primary respiratory
surface by utilizing cutaneous respiration. Under natural conditions, in
situations of drought, rivulus is capable of surviving out of water under
moist leaf Utter (personal observations). In this state of semi-dormancy
the fish maintain a reduced metabolic rate and apparently utilize the skin
as the major site of respiratory gas exchange (Abel, 1981).
-------
Fin erosion has been observed In natural populations of other species
of marine fish in several coastal locations: in the Hew York Bight
(Mahoney et al., 1973), off southern California fMearns and Sherwood, 1974)
and in the estuary of the Dim-amish River, Seattle, Washington (Wellings et_
al., 1976). In each case the highest incidence of fin erosion was observed
IF areas heavily contaminated with industrial and domestic pollution.
Although the investigators were unable to determine the primary causative
agent, a high correlation between tne occurrence of the disease and the
presence of various pollutants (heavy metals and halogenated organics) led
them to suggest these as probable eteological factors. TECP induced fin
erosion in rivulus strongly corroborates the contention that certain
halogenated organics play a dominant role in the cause of the disease. The
mechanism of TECP induced fin and gill erosion warrants further investiga-
tion.
An unusual result in this study was the development of abnormal dorsal
fins in rivulus exposed chronically to bronu,form (Table 24 and Figs. 17a-c).
No dorsal fin abnormalities were noted in offspring of bromoform exposed
adults which were transferred to uncontaminated water (Table 25). This
indicates that bromoform exerts its effect during later development. The
stage of development at which the environmental insult takes place deter-
mines, in large part, the resulting abnormality (Stockard, 1921). Embryos
from bromoform exposed adults that were transferred to uncontaminated water
{Table 25) may have been transferred before the stage at which dorsal fin
development took place (stage 32; see Section 2 of this report). Because
bromoform depurates from organisms rapidly (Gibson et a]_., 1930) it probably
had little or no effect on development after eggs were removed from the
contaminated solutions. In future experiments of this type embryos should
be transferred to the respective conditions at a fixed stage, probably the
terminal stage of embryonic development, to avoid ambiguity.
There have been a number of studies on fish teratogenesis (for review
see Laale and Lerner, 1981). Most have involved direct exposure of eggs
to chemical or physical perterbations at specific stages or during the
entire period of embryo!ogical development. Although many such studies are
Involved with the elucidation of the mechanisms of teratogenesis in fish,
some purport to ascertain environmental relevance arid therefore involve
the exposure of embryos of various species to known pollutants. In many of
these studies egg exposure levels are chosen at or near the lethal level
of the embryo without regard for other life stage susceptabilities. In
developing a realistic evaluation of the potential of a chemical to induce
congenital malformations or pathologic changes due to chronic exposure, the
physico-chemical properties of the pollutant and levels of exposure are most
important considerations. In nature, all stages of the life cycle would
likely be exposed to tne pollutants in question in varying degrees, depending
upon the ecology of the species. Many studies involving direct exposure of
eggs ignore the importance of the natural environment. Organic pollutants
with high partition coefficients tend to bioaccumulate (Kopperman et al.,
1979) and may be transferred (unchanged or after metabolic alteration J-
directly to the developing oocytes. There the xenobiotic may become
incorporated into specific portions of the yolk and its ultimate effect on
98
-------
the developing embryo will depend upon the time and extent to which it is
mobilized from the yolk reserves. For example, DOT and mirox, two highly
lipophilic, persistent pesticides have been shown to accumulate 1n the eggs
of the diamond killiflsh, Adinia xenica, when incorporated into the maternal
test diet (Koenig, 1975). Negligible effects were observed during
embryological development; however, dose-related mortalities occurred soon
after hatching. Thfe vast majority of these two pesticides were apparently
confined to the triglyceride fraction of the yolk which was metabolized by
the larvae soon after hatching. In cases such as these the embryo is not
the target. Laboratory studies should be designed with the natural route
of exposure in mind.
It 1s also possible that various pollutants may alter the maternal
physiological state or the biochemistry of egg maturation which, in turn,
could result in abnormal or less fit offspring. An analogous situation in
certain birds is the well-known example of egg shell thinning resulting from
altered calcium metabolism due to DDE exposure.
It is thus apparent that before a meaningful assessment of the
teratologics! potential of a particular pollutant can be made, the maternal
component and its indirect effects on early ontogeny must be included.
Direct effects of a potential teratogen should be evaluated only at dosage
levels at or lower than those tolerated by other life stages, otherwise data
will lose environmental relevance.
-------
SECTION 8
MUTAGENESIS BIOASSAY: INVESTIGATIONS WITH RIVULUS MARMORATUS AND OTHER
SELECTED FISH SPECIES
INTRODUCTION
The alms of the investigations reported in this section were twofold:
1) to characterize the genetic system or Rivulus marmoratus and to evaluate
the consequences of this genetic system for long-term, whole body exposure
bioassays for the mutagenic effects of chemical';, and 2) to investigate
possible short-term procedures for assessing mutagenic effects of chemicals
in the Rivulus mamioratus system. These aims were subsequently modified to
include alternate species when we were unable to identify genetic markers in
laboratory clones and wild populations of kivillus marmoratus and when the
karyotype of R. marmoratus proved to be unsuitable for cytogenetic analysis.
Investigations with Rivulus marmoratus
The genetic system represented by Rivulus marmoratus is unique among
vertebrate species. Individuals are predominantly self-f^rti1izing
hermaphrodites (Harrington, 1961) although some individus':^ occur as primary
or secondary males (Harrington, 1967). The individuals available in this
laboratory (clones DS, NA» and M) were descended through at least 30 uni-
parental generations from individuals collected from the wild by Harrington
(1963) near Vero Beach (DS and NA) and near Miami (M). Wild-caught
individuals used in this study were collected near Marco Island, Florida in
October, 1978. Laboratory clonal individuals are presumably self-fertilizing
hermaphrodites which are homozygous for most genes (Harrington and
Kail roan, 1968) and are isogenic within but not between clones (Kailman and
Harrington, 1964). These conclusions are based upon histocompatibility
criteria applied to parents and offspring of interclonal crosses. Na+'.val
and artificial cross fertilization resulted in amphimixis and suggest that
subsequent segregation in hybrid animals is random with respect to
paternally- and maternally-derived chromosomes (Harrington and Kallman,
1968).
The purpose of the present investigation was to identify gene markers
by starch gel electrophoresis to assess the genetic distance between clones
and to verify the occurrance of self-fertilization and segregation of
maternally- and paternally-derived alleles during subesequent generations of
uniparental reproduction. A secondary objective was to utilize genetic
markers to assess the relative importance of self and cross-fertilization
and homozygosity and heterozygosity in natural populations. Our approach was
... . ...100
-------
to screen 1-3 fish of each clone and at least 20 wild fish for 14 enzyme
systems representing an estimated 28 loci.
The fish utilized in this study were adult and the largest available at
the time of this study. Laboratory clones had been maintained for 20 years
by uniparental reproduction and isolated from other individuals as eggs.
Wld Rivulus marmoratus were captured by dip net from an impounded area near
Marco Island on the West Coast of Florida. Fish were captured over a two
day period within 500 yards of each other. All fish were maintained in
isolation and fed frozen brine shrimp.
Individual fish were sacrificed by cooling and skeletal muscle, liver
and nervous tissue (eye and brain) were separated and homogenized in 2
volumes of distilled water. Homogenates were centrifuged and supernates
were frozen at -20°C in capillary tubes until they were used. Electrophore-
tic separations were performed on a horizontal starch gel system. Four
buffer systems (Shaw and Prasad, 1970) were utilized as follows; lithium
hydroxide, pH s 8,2 (LiOH), at 350V for 3.5 hours; trls-versene-boric acid,
pH = 8.0 (TVB), at 200V for 3.0 hours; tris-maleic acid - EDTA, pH = 7.4
(TME) at 100V for 5 hours and tris-citrate, pH = 6.7 (TC6.7) at 180V for
3.0 hours. Gels were sliced norizontolly and stained for the following
enzymes (Shaw and Prasad, 1970): adenylate kinase (AK), acid phosphatase
(AP), esterase (EST), glutamate dehydrogenase (GDH), alpha-glycerophosphate .
dehydrogenase (oGPDH), glutamate-oxaloacetate transaminase (GOT), glucose-6-
phosphate dehydrogenase (G5PDH), isocitrate dehydrogenase (IDH), lactate
dehydrogenase (LDH), peptidase (PEP), 6-phosphogluconate dehydrogenase (6PGd),
glucose phosphate isomerase (GPI), and phosphoglucomutase (PGM). Indole-
phenole oxidase was scored on gels stained for dehydrogenases.
Skeletal muscle, 1iver and neural"homogenates were run for 3-7
individuals on one gel. Samples for laboratory clones were run simultaneous-
ly for all three clones. One sample per tissue of a previously run
laboratory clonal fish was included with samples from wild-caught individuals
for at least one run for each enzyme. Isozyme patterns obtained for a
particular tissue were judged identical for all fish. The absence of genetic
differences among the individuals screened presented difficulty in the
identification of if/zyme bands with specific gene loci. Consequently the
number of loci that were screened was estimated on the assumption that all
fish were homozygous and each band thus represented a product of a separate
gene locus. This tends toward an overestimation of the number of loci as
some bands may represent heterodimers or heterotetramers of products from
different loci (Table 26).
The lack of detectable electrophoretic differences between individuals
separated by 20 years and hundreds of miles supports the view that the
North American populations of Rivulus marmoratus were recently derived from
a small founder population and are now expanding at a rapid rate (Fowler,
1928; Rivas, 1945; Harrington and Rivas, 1958; Tabb and Manning, 1961).
The surprising absence of genetic variation between laboratory clones and
within natural populations has resulted in the postponement of further
analyses of the Rivulus marmoratus genetic system until suitable genetic
markers can be identified perhapsfrom individuals collected in other parts
-------
Table 26. Screening of Clonal and Wild-Caught Rivulus marmoratus for
Biochemical Variation
Maximum Estimated
Number Number
Tissues" of
of
# of Irilmals Screened
Clones
Enzyme3 Buffer® Examined Isozymes Loci/Fish DS NA
Wild-
Mi 0 Caught
AK
THE
M,
L, N
I
1
2
2
2
25
AP
THE
H,
L, N
1
1
2
2
2
18
EST
LiOH
M
L
4
2
N
1
4
3
2
3
28
GDH
TME
H
3
L,
N
2
3
1
1
1
30
GPDH
TVB
H
1
1
2
2
2
29
SOT
TVB
H,
L, N
2
2
1
1
I
32
G6P0H
TVB
H,
L
1
N
2
2
2
2
2
21
IPO
TVB
M,
L, N
1
1
2
2
2
18
IDH
TC6.7
M,
L, N
2
2
0
0
0
36
LDH
TVB
M
L
2
1
N
3
3
3
3
3
29
PEP
LiOH
M»
L, N
2
2
2
2
2
28
6PGD
TC6.7
H.
t, N
1
I
2
2
2
15
GPI
LiOH
M,
L, N
2(+>
m
0
0
0
25
PGM
TC6.7
H,
N
3
L
1
3
2
2
2
17
See methods section for abbreviations of enzymes and buffer systems.
^Abbreviations for tissues: M=skeletal muscle; L=1iver, N=brain and eyes
102
-------
of the range of this species. The absence of electrophoretic variation
across 20-28 loci is surprising in view of the histocompatibility differences
reported by Harrington and Kallman (1968). Perhaps the use of histocom-
patibility criteria screens for a very much larger number of possible loci
{subject to mutation) than are screened for by elactrophoretic techniques.
Rivulus marmoratus: Short Term Assays
The utilization of short term in vitro procedures to assess the effects
of potentially mutagenic and carcinogenic chemicals and the investigation of
mechanisms by which these chemicals exert their toxic effects has progressed
rapidly in the past 10 years. The abi1ity to employ in vitro procedures
to investigate the effects or lack of effects of specific toxic chemicals in
whole animal bioassays is important in the overall evaluation of a test
organism. To this end we have attempted to establish a permanent line of
cultured cells derived from Rivulus marmoratus to provide sufficient
quantities of material to investigate the metabolism of mutagens and
carcinogens and to evaluate sensitivity to induction of point mutations,
chromosomal mutations, DNA repair and neoplastic transformation by these
chemicals. An evaluation of the Rivulus marmoratus karyotype for cytogenetic
analyses was also conducted. Procedures and results are discussed in the
following section.
Investigations of Alternate Species
Recent progress in the use of cellular and subcellular effects to assess
the toxicity of chemicals in a variety of genetic systems from bacteria to
man have greatly extended our ability to screen large numbers of chemicals
for potential sublethal mutagenic and carcinogenic activity. Concurrent
development of similar procedures for teleost fish systems might be expected
to provide aquatic toxicologists with useful assays for the detection,
identification, and assessment of the effects of waterborne mutagenic and
carcinogenic hazards in aquatic ecosystems. The development of short term
assay procedures for a suitable teleost species could also be expected to
enhance the ability of aquatic toxicologists using whole animal exposure
assays to investigate at the cellular and subcellular levels the causes of
differences between teleost fish systems and mammalian systems. Potential
sources of interspecific differences in response to exposure to genotoxic
chemicals include differences in metabolism, pharmacokinetics, DNA repair
and epigenetic phenomena associated with neoplastic transformation. The
purpose of this series of investigations were: 1) to survey a number of
teleost species, including Rivulus marmoratus, as possible test organisms
for the development of short-term procedures to assess the effects of
genotoxic chemicals in laboratory and field situations, 2) to evaluate the
feasibility of these procedures for use by aquatic toxicologists, and 3) to
develop in vitro procedures for investigating metabolism of xenobiotics,
DNA repair and neoplastic transformation in fish cells.
Among the short tern assays that have attained credibility as indicators
of the genetic activity of chemicals, sister chromatid exchange (SCE)
analysis is possibly the most sensitive and versatile (Perry and Evans, 1975;
103
-------
Perry, 1980). SCE analysis is a cytogenetic assay that is more sensitive
than conventional analysis of cnromosomal aberrations for assessing the
genotoxic activity of chemcals and SCE analysis can be adapted to a variety
of in vitro test protocols that are essential for applications in aquatic
toxicology. The primary criterion that we have applied in selecting a
suitable species for further development was the suitability of the karyotype
for cytogenetic analysis. A second feature considered desirable for a test
organism for sublethal effects is a tolerance for extensive manipulation in
the laboratory and for adverse conditions that are expected to be associated
with exposures to waterborne chemicals, A fourth consideration is that a
test organism be readily available to laboratories which would potentially
have use for the assays to be de/cloped. A final consideration for a test
organism to be used in field studies is preference for association with
bottom sediments where many chemicals tend to accumulate.
MATERIALS AND METHODS
Collection of Fish
The Rivulus marmoratus used for karyotype examination and for tissue
culture were obtained from laboratory stocks maintained as described in an
earlier section. Other species that were examined were obtained from fresh-
water and estuarine systems in the Charleston area. The freshwater species
Umbra pygmaea (DeKay), Fundulus chrysotus (Gunther) and Chologaster cornuta
(Agassiz) were collected with a dip net. The estuarine species Symphurus
plagiusa (Linnaeus), Paralichth.yes dentatus {Linnaeus) and Citharichthyes
spjlopterus (Gunter) were collected in trawls. Opsanus tau (Linnaeus) were
collected in crab traps.
Cytogenetic Methods
Numerous procedures have been described for the preparation of metaphase
chromosomes for cytogenetic analysis in fish. The technique with greatest
potential for application to mutagenesis studies, including SCE analysis,
is that of Kligerman e_t al_., (1975), Kligerman and Bloom (1976) and
Kligerman (1979) and was used for the initial cytogenetic screening of
potential test organisms. Fish were injected ip with 25 ug of colchicine per
gram body weight and held in well-aereated aquaria for 6 hours before
sacrificing. Gills, stomach and intestine were excised, cut into short
segments, treated with hypotonic KC1_, fixed with three changes of methanol-
acetic acid (3:1) fixative and stored at 5°C until used. "Slides were
prepared by mincing tissue pieces in 50% acetic acid until a slurry of cells
was visible (about 1 minute). The cell suspension was drawn into a 100 u£
pipet, expelled onto a microscope slide at 50°C and withdrawn from the slide
to leave a ring of cells 1 cm in diameter. Slides were air dried and
stained for 8 minutes in 41 Giemsa in distilled water.
The most effective technique for obtaining abundant, well spread, and
clearly defined metaphase chromosomes in many mammalian systems has been
the culture of peripheral leukocytes in the presence of a mitogen. Few
reports, however, have demonstrated the successful extension of this
104
-------
procedure to teleost fish systems {Gold, 1979). We were able to culture
leukocytes from the oyster toadfish. Subsequent cytogenetic analyses in
this species were done with peripheral leukocyte cultures (Haddock and
Kelly, 1980), Blood samples were obtained by cardiac puncture using a
heparinized syringe. Whole blood (0.5-1.0 ml) was expressed into 5 ml of
RMPI 1640 (GIBCO) medium supplemented with 15£ fetal calf serum, antibiotics,
sodium chloride end phytohemagglutinin (GIBCO). The pH of cultures was
adjusted by gassing tubes with 5% CO? in air. At harvest 6-8 days later
cultures were treated with elolcemid for 5 hours, centrifuged, resuspended ir
hypotonic KCL for 10-15 minutes and fixed with three changes of methanol-
acetic acid (3:1). Cells suspended in fixative were dropped onto glass
slides and air dried. For conventional staining, slides were placed in 4»
Glemsa for 8 minutes.
To obtain sister chromatid differentiation 5-bromo-2'-deoxyuridine
(Sigma) was added to culture medium before seeding with blood. Differential
staining was accomplished by staining in the fluorescent dye Hoechst 33258,
exposure to a 40W blacklight fluorescent bulb and restaining for 10 min. ir.
2% Giemsa in phosphate buffer (pH = 6.8) after the method of Goto et al.
(1975). In vitro exposures of dividing leukocytes to the test chemicals
ethyl methanesulfonate, EMS (Sigma) and bromoform, BF (Fisher) were con-
ducted by additions within 24 hours of culture Initiation for the duration
of the culture period. All cultures were maintained at 26°C.
Assays for Samonella Mutagenesis
Preparation of S-9 and plate assays for Salmonella mutagenesis were
conducted with minor modification of the methods described by Ames et al.
(1975). The concentration of S-9 protein per gram of toadfish hepatic
tissue is lower than that of rat hepatic S-9, Consequently liver tissue
from toadfish was homogenized in KC1 at a 1:2 dilution and final protein
concentrations of rat and toadfish liver S-9 were prepared on a protein for
protein basis (Lowry et al., 1951). Plate assays were incubated for 48 hrs
at 37°c when rat S-9 was used and for 12 hrs at 26°C then for 48 hrs at
370C when fish S-9 was used. Animals (rats and fish from 200-300 g) from
which S-9 was to be prepared received either no pretreatment or were
injected IP with 200 rug/gram body weight of Aroclor (AC) in corn oil five
days prior to sacrifice or with 0.02 mg/g body weight of 3-nethylcholanthrene
(MC) 5, 3, and 1 days before sacrifice. Liver S-9 fractions were frozen
at -80°C until use. The precarcinogens utilized in these assays, benzo(a)
pyrene (BP), and 2-aminoanthracene (AA) (Aldrich) were dissolved in DMSO
and stored at 5°C in DMSO between experiments.
Tissue Culture
In Initial attempts to initiate primary cell cultures from tissues of
Rivulus marmoratus and Opsanus tau we have used two methods: direct
planting (explant) and trypsin dispersion (Wolf, 1973). Tissues to be
cultured by direct planting methods were excised into sterile Hank's BBS
and minced with scissors for 7-10 minutes. The resulting suspension of
small tissue pieces was added (4-6 drops) to 4 ml of growth medium in a
105 _
-------
culture flask with 25 cir£ surface area. Medium and unattached tissue pieces
were drained and replaced by fresh growth medium after 1-1.5 hours. Medium
was changed at weekly intervals until a monolayer of cells was obtained.
Tissues to be dispersed by trypsin were minced and washed with Hank's BSS
and tissue pieces placed in a trypsinization flask in a refrigerator at
5°C and stirred with a magnetic stirrer for 30-90 minutes in ATV dispersion
mixture (Nadin and Darby, 1958). The resulting fluid and dispersed cells
were discarded and remaining tissue pieces were resuspended in ATV and
stirred several hours to overnight at 5°C. The resultant suspension of
cells was centrifuged, the supernate discarded and the pelleted cells washed
and resuspended in growth medium to give 1-3 x 105 cells per ml. Culture
flasks were seeded with 4-5 ml of cell suspension. Medium was changed
weekly until cultures showing new growth reached monolayer. Monolayer
cultures from either of the procedures outlines above were subcultured by
discarding old medium, rinsing with Hank's BSS, treating with three changes
of ATV for 15 minutes, !5 minutes, and 30 minutes. Harvested cells were
washed and then resuspended in growth medium and used to seed 2 new culture
flasks. The growth medium used in all experiments was 1-15 medium (GIBCO)
supplemented with 15% fetal calf serum, antibiotics, MEM amino acids and MEM
vitamins. Additional NaCl was added to increase the osmotic pressure of the
medium. Cultures were maintained in sealed flasks at 26°C.
RESULTS
Cytogenetic Survey
There are a number of teleost species available in the Charleston area
that from previously published reports could be expected to possess karyo-
types that were suitable for detecting chromosome damage.- The freshwater
species Umbra pygmaea was closely related to the species Umbra 1imi that had
been used by Kligerman et al_. (1975) and at the time U, pygmaea had not been
examined cytogenetics!ly. The karyotype of U. pygmaea is similar to that
of U. 1imi (Figure 23a) and is excellent for mutagenesis studies. U. pygmaea
has subsequently been used to detect mutagenic activity in Rhine River
water by SCE analysis (A1ink et aK, 1980). The freshwater species
Fundulus chrysotus had been previously examined (Chen, 1971) and possesses
a karyotype suitable for cytogenetic analysis (Figure 23b). Further studies
of the two species were postponed because of their limited availability in
the area and our greater interest in identifying a marine species for further
study. Another freshwater fish examined was Chologaster cornuta (Figure 23C).
The primitive teleost karyotype is comprised of 24 pairs of small
chromosomes (Denton, 1973). Many species of marine teleosts, including
Rivulus marmoratus (Figure 24b) possess this cytogenetic structure and are
not suitable for use as cytogenetic models in mutagenesis studies. Our
approach was to examine local representatives of families that contain
demersal marine species. The flatfish (PIeuronectiformes) and the toadfish
(Batrachoidiformes) were surveyed. The blackcheek tonguefish (Cynoglossidae:
Symphurus plagiusa) karyotype consists of 46 chromosomes that are quite
small (FTgure 23d7. The summer -.'loMnder (Bothidae: Paralichthys dentatus)
karyotype was comprised of 46 small chromosomes (Figure 24a). The oyster
106
-------
Figure 23. Metaphase chromosomes of various fish species
prepared by the solid tissue technique
(K11 gentian and Bloom, 1977):
a. Umbra pygmaea (DeKay)
b. Fundulus chrysotus (Sunther)
c. Choloqaster cornuta (Agassiz)
d. S.ymphurus plagiusa (Linnaeus)
107
-------
*
* I
A
*&/-$:>
* n
#;
$
%
D
108
-------
Figure 24. Metaphase chromosomes of various fish species
prepared by the solid tissue technique (a-c)
and by air drying of cultured leukocytes (d):
a. Paralichthys dentatus (Linnaeus)
b. Rivulus mannoratus (Poey)
c. Opsanus tau (Linnaeus) .
d. Opsanus tau (Linnaeus)
-------
%
%
ft
w
* I
*S *
« #«*• I • O'.-v
it, A
:
«• *
*
f*£#
«h» 'J-
1-m%
' W
* '.
* «£ ^ *
«* r
* ^ +
%
»?£i <*
%*|W
* ;# "*
M*« .
4,^ K*.
» » i <»<
.
• * - t»
, »L ¦
»
)
»
n
.*>»
ax
j*
*
a:
»
*
% m ¦»
%/*.*', »«
I #
D
1
:»
no
-------
toadfish (Batrachoididae; Opsanus tau) karyotype was comprised of 46
chromosomes, many quite large, and was quite suitable for cytogenetic
analysis (Figure 24c). The promising features of the toadfish karyotype and
its potential as a test organism by all criteria that were discussed earlier
led to the discontinuation of the cytogenetic survey and increased efforts
to evaluate the toadfish as a cytogenetic model.
Culture of peripheral leukocytes Is one of the most effective methods
for preparing metaphase chromosomes for cytogenetic analysis in mammalian
species. Metaphase chromosomes from leukocyte cultures arc abundant, well
spread and have a wall defined morphology. Leukocyte preparations are
applicable to the assessment of mutagenesis in protocols with in vivo and
with in vitro exposure to chemicals with genotoxic nativity. Our attempts
to culture toadfish leukocytes were succ#»?tful figure 24d) and leukocyte
culture has been the basis for cur buosequent investigations.
Sitter Chromatid Exchange (SCE) Analysis in Dividing Toadfish Leukocytes
Sister chromatid exchange assays involve the counting of reciprocal
exchanges between .the genetically identical chromatids of replicated
chromosomes at metaphase. A dose-response relationship between the con-
centration of a variety of known mutagens and carcinogens and the number of
sister chromatid exchanges was demonstrated by Perry and Evans (1975) using
Chinese hamster ovary cells in vitro. A requisite procedure in sister
chromatid assays is the incorporation of bromodeoxyuridine (BUdr), an
analog of thymidine, into DNA*, this is usually done for two rounds of
replication to provide chemical differentiation of DNA and subsequent
visualization of exchange between otherwise identical sister chromatids.
Presumably, one double-stranded DNA molecule comprises one chromatid of a
metaphase chromosome. After two rounds of replication in the presence of
BUdr, every metaphase chromosome has one chromatid with one unsubstituted
and one BUdr-substituted strand of DNA and another chromatid containing two
BUdr-substituted strands of DMA. Subsequent staining with a fluorescent
dye (usually Hoechst 33258} or with a fluorescent dye followed by exposure
to light and staining in Giemsa (fluorescent plus Giemisa or FPG technique)
results in cytologically detectable sister chromatid exchanges.
A representative toadfish karyotype (Figure 24d) contains 46 chromo-
somes, including one pair of large metacentrics, eight pairs of large
acrocentrics, six pairs of subtelocentricsj four pairs of small metacentrics
and four pairs of small acrocentrics. These chromosomes are larger than
those of all other marine species we have examined and are suitable for SCE
analysis (Figures 25 a & b). In addition, the ease with which toadfish
leukocytes may be cultured, a quality shared by few other fish species
(Gold, 1979), their tolsrance to unfavorable conditions and to repeated
blood collection, and their association with estuarine sediments, where many
pollutants tend to accumulate, combine to recommend the toadfish as a useful
organism to explore the possibility of an SCE assay system for both field
and laboratory applications.
Ill
-------
Figure 25. a, A second devision metaphase from a control
culture of toadfish leukocytes showing
sister chromatid differentiation of cell
chromosomes and sister chromatid exchanges
(10 exchanges were counted).
b. A second division metaphase from a toadfish
leukocyte culture exposed in^ vitro to
62 ug/ml of EMS for 4 days prior to harvest-
ing (22 exchanges were counted).
c. Monolayer of cells from a primary culture
of Opsanus ovary tissue.
d. Monolayer of cells from a culture of Opjanus
ovary tissue in the fourth passage.
112
-------
#1
>4*7 » '
\
1
r <¦*
/ «c
i
, * \Z
,* 0 %
V , * - '
i ^
.V t#
» %
I
J V
I
*
«S^C%bvJj
¦"•• ¦¦¦» • ¦, f* t ¦"
fcVv' 4j
WW
BP1
UN
M
lillilMlaiwB
ii»
Ei£-;iS^8
III
wmmmW^
wsimKm
Wfm£
¦HH
MMM
¦»
113
-------
Exposure of dividing toadfish leukocytes to the mutagen-carcinogen EMS
at dosages comparable to those reported for mammalian test systems indicate
that SCE analysis of toadfish leukocytes is of comparable sensitivity
(Table 27). Nine metaphases from two cultures exposed to 62 pg/ml of EMS
yielded 192 exchanges for an average of 21.3 exchanges per metaphase. This
is a significant increase over the control rate of 7.0 exchanges per meta-
phase obtained from 20 metaphases from two cultures. Exposure to 124 ug/ml
of EMS yielded an average of 35.1 exchanges per metaphase. Indicating a
possible dose-dependent response of SCE rate to EMS exposure. Although these
results ire based on small samples of metaphases, they nevertheless
demonstrate the feasibility of SCE analysis using a toadfish in vitro system.
Exposure of dividing leukocytes to bromoform at low concentrations did
not result in increased rates of SCE (Table 27). Bromoform has been shown
to be mutagenic In Salmonella and carcinogenic in mice. Perhaps the genetic
activity of bromoform, if any, may prove to be detectable at higher concen-
trations or in tests in vivo, allowing for metabolic activation of compounds
not mutagenic before activation. The volatility of bromoform and the
necessity of periodic gassing of cultures with 5X cOg in air may have
reduced the degree of exposures of leukocytes below the level of detection
by SCE analysis. In addition, our lack of success in obtaining metaphases
in glass culture tubes with Teflon-lined caps necessitated the exposure of
cells to bromoform in polystyrene culture tubes (Falcon 3033). We are not
aware of any previous reports of SCE analysis following exposure to bromo-
form in other SCE assay systems.
Metabolism of Benzo(a) pyrene (BP) and 2-Am1noanthracene (AA) to Salmonella
Hutagens by 10ad11sh S-H 1 ^ ~ "
The relative proportions of functionally distinct forms of cytochrome
P450 enzymes which populate the endoplasmic reticulum of mammalian cells
appear to be a major determinant of the degree to which promutagens and
carcinogens are activated or detoxified. Although recent investigations
indicate th?t metabolism of xenobiotics in fish is similar to that in
mammals (Schoor and Couch, 1979; James and Bend, 1980) there are definite
species differences even among rodents in the ability to activate specific
mutagens and carcinogens. The nature of P450 systems of a teleost species
proposed for use as a test organism for potential mutagens and carcinogens
should be characterized. To this end we report a series of comparative
studies to determine the effects of hepatic microsomal enzyme (S-9) pre-
parations from rats and toadfish pretrevced with standard enzyme inducers
(3 methyl-cholarithrene (MC) and Aroclor (AC) and untreated on the metabolism
of a polycyclic aromatic hydrocarbon and an aromatic amine to Salmonella
mutagens.
Six concentrations of each test chemical (0.2-20.0 ug/plate) were
tested in the presence of six different S-9 types (from untreated, MC-pre-
treated or AC-pretreated fish or rats). Each of the 432 data points were
done in duplicate. Strain TA 98 was provided by Bruce Ames.
114
-------
Table 27. Sister Chromatid Exchange (SCE) Rates Following In Vitro Exposure of Dividing Toadfish
Leukocytes to Ethyl Methanesulphonate (EMS), Bromofonm and Hank's (control)
Concentration
Number of
Kumber of
No. SCEs
vs Control
Chemical
(pg/ml) (molarity)
Fish
Cultures
No. Metaphases
t-test
Control (Hank's)
2
2
139/20
EMS
62(503 yM)
1
2
192/9
? < 0.01®
124(1000 vM)
1
1
211/6
p < 0.01®
BF
0.10(0.40 yM)
2
5
45/7
p > 0.05
100.00(400 jM)
1
3
58/8
p > 0.05
Three metaphases from one control culture were scored for 24, 23, sad 28 SCE. Because they
represented an SCE rate at least twice as large as the largest from any other control culture,
they were not included in this table. If they are included, the mean rate of SCE in control
cultures becomes 9.3 SCE per metaphase, and the difference between SCE rate in control and
EMS-treated cultures is not statistically significant, probably because of the small sample
size.
-------
The concentrations of S-9 protein from fish liver (90 mg protein/g wet
liver) was lower than that from rat liver (160 mg protein/g wet liver) and,
thus, the concentrations of S-9 per plate of fish and rat S-9 were adjusted
to be equivalent on a mg protein/plate basis. Neither BP nor AA were
mutagenic in absence of S-9 protein. The activation of AA was considerable
for all fish and rat S-9's tested. All fish S-9 and S-9 from MC- and AC-
pretreated rats activated BP but S-9 from untreated rats resulted in little,
if any, activation of BP. The extent of activation of BP and AA under
optimal conditions of S-9 type and concentration for fish (BP:937 revertants/
plate and AA:6330 revertants/piate) and rat (BP:847 revertants/plate and
AA:6692 revertants/plate) were comparable. The data summarized in Figure 26
show the effects of Increasing concentrations of S-9 protein on a single
concentration of test chemical. The curves are representative of results
obtained for other concentrations of test chemical. It is apparent from the
forms of the curves that pretreatment with MC but not with AC had a
detectable effect on the activation of these test chemicals. As expected
both MC and AC pretreatment had discernable effects on the metabolism of
these test chemicals in the rat. These results support the growing evidence
that the capacity of fish enzymes systems to metabolize promutagenic
xenobiotics are similar in fish and mammals although the effects of inducers
in these systems may not be similar. The lack of a detectable effect of
phenobarbital pretreatment on xenobiotic metabolism in fish has been
reported (Bend et a!., 1973).
Tissue Culture
The initiation of tissue cultures from small species is hampered by
difficulty in obtaining gram quantities of any tissue needed for primary
culture. Consequently, the results we have obtained in limited attempts to
culture tissues from Rivulus marmoratus have been discouraging (Table 28).
On the other hand, the ready availability of tissues from toadfish has
resulted in encouraging prospects for primary culture of toadfish tissues
(Table 23). Primary cultures of toadfish liver, ovary, and skin have given
monolayer cultures. Liver and ovary primary cultures have been successfully
subcultured. We have had no success in culturing cells from the toadfish
beyond the fourth passage. The toadfish cultures are of mixed cell type
in the primary culture (Figure 25c) but appear to be predominantly
epithelioid in later passages (Figure 25d). These results suggest that the
development of assays for mutagenesis and carcinogenesis with toadfish cells
be based upon primary cultures as are a number of assays for cell trans-
formation and unscheduled DNA synthesis that utilize mammalian material.
The use of primary cultures to assay chemicals for genotoxic activity
confers certain advantages. Primary cultures appear to retain a greater
capacity for xenobiotic metabolism than do cells which have been in culture
for a longer period of time.
DISCUSSION
This series of preliminary studies collectively suggest that the oyster
toadfish is a suitable organism with which to pursue investigations of the
116
-------
Figure 26.. The effects of no treatment and pretreatment
of fish and rats with a 3-methylcholanthrene
(MC) and Aroclor 12S4 (AC) on the activation
of benzo (a) pyrene (BP) and 2-aminoanthracene
(AA) to Salmonella (TA98) mutagens.
Four graphs: a. Rat 5-9, BP; b. rat S-9, AA
c. fish S-9, BP; d. Fish S-9, AA.
\
117
-------
RAT S9 Cmg protein)
BW - 20.0 uo/plol.
FISH S9 Cmg prole in)
M - B.e uo/plol
RAT S9 Cmg prote i rO
¦
(04
I—
z
<
H i
a
K
Id
>
ia,
a
©
o(
FISH S9 Cmg protein)
d
-------
Table 28. Culture of .Tissues from Rivulus marmoratus and Opsanus tau
Number
Success-
Maximum
Number
Number
ful
Number
Method3
Cultures
Cultures
Sub-
Sub-
Species
Tissue
Initiated
Growing
cultures
cultures
Rivulus
marmoratus
liver
E» T
6
1
0
0
gonad
E, T
2
0
0
0
fin
E
2
0
0
0
Opsanus tau
liver
E, T
37
15
1
1
ovary
E, T
26
18
5
4
skin
E, T
25
3
0
0
spleen
E, T
9
1
0
0
other
E, T
28
1
0
0
aE = explant; T * trypsin dispersion
119
-------
effects of mutagens and carcinogens on marine teleost fish. Furthermore,
the toadfish is a suitable organism to use as a model for evaluating the use
of cytogenetic effects on the detection, identification and issessment cf
hazards of water bcrne chemicals in estuarine ecosystems. The following
summarizes the features that recommend the toadfish for further investi-
gation;
1} The toadfish has a karyotype suitable for cytogenetic
analysis.
2) Leukocytes obtained from toadfish can be cultured and
provide a system suitable for SCE analysis following in
vitro exposure to mutagenic chemicals (Maddock ard KelTy,
1980).
3) The toadfish is sufficiently large to provide gram
quantities of tissue needed for metabolic evaluation
for primary culture initiation and for pharmacokinetic
investigation.
4) The toadfish is tolerant of handling out of water for
extended periods and will survive adverse conditions
that are not tolerated by many other species.
5) Toadfish are available in abundance from Massachusetts to
Florida and are available to any laboratory on the
Atlantic coast.
6) The toadfish is a sedentary, bottom-dweller and is
associated with the sediments where many potentially
toxic chemicals tend to accumulate.
7) The capacity for xenobiotic metabolism in toadfish is
similar to that of mammals.
120
-------
REFERENCES
Abel, D. C. 1981. An ecological basis and adaptations for emergence in
the cyprinodontld fish Rivulus marmoratus. Masters Thesis. College
of Charleston, Charleston, SC
A1ink, G, M., E. M. H. Frederix-Walters, M. A. van der Gaag, J. F. J. van
der kerkhoff and C. L. M. Poels. 1980. Induction of sister-chromatid
exchanges in fishes exposed to Rhine water. Mutation Res. 78:369-374.
American Public Health Association. 1976. Standard methods for the
examination of water and wastewater, 14th ed., Washington, O.C.
Ames, R. N,, J. McCann, and £. Yamasakl. 1975. Methods for detecting
carcinogens and mutagens with the Salmonel1 a/mammalian-microsome
mutagenicity test. Mutation Res. 31:347-364.
Armstrong, P, B. and J. S. Child. 1965. Stages in the normal development
of Fundulus heteroclitus. Biol. Bull. 128:143-168.
Autian, J. 1973. Toxicity and health threats of phthalate esters:
review of the literature. Environm. Health Persp., Experimental Issue
No. 4, pp. 3-26.
Bend, J. R., R. J. Pohl, and J. R. Fouts. 1973. Further studies of the
microsomal mixed-function oxidase system of the little skate, Raja
erinacea, including its response to some xenobiotics. Bull. Mt, Desert
Island Biol. Lab. 13:9-13.
Brockman, F.W. 1975. An unusual habitat for the fish Rivulus marmoratus.
Florida Sci. 38:35-36.
Cairns, J. and H. W. van der Schalie. 1980. Biological monitoring.
Part 1 - early warning systems. Water Res. 14:1179-1196.
Carpenter, J. H., C. A. Smith, and R. G. Zika. 1980. Reaction products
from the chlorination of seawater. Pages 379-385 in R. L. Jolley,
W. A. Brungs and R. B. Cumming, eds. Water chlorination: environmental
impact and health effects, Vol. 3. Ann Arbor Science, Ann Arbor, M1ch.
Chen, T. R. 1971. A comparative study of twenty killifish species of the
genus Fundulus (Teleosteii Cyprinodontidae). Chromosoma, 29:255.
121
-------
Colby, P. and L. L. Smith, Jr. 1967. Survival of walleye eggs and fry on
paper fiber sludge deposits in Rainy River, Minnesota. Trans. Amer.
Fish. See. 96:278-296.
Courtney, C. M. 1975. The production of organic detritus in an impounded
mangrove forest. Ann. Rep. Marco Appl. Mar. Ecol. Sta. July 1, 1974-
July 1, 1975. pp. 197-244.
Cripe, C. R., J. H. Cripe and R. J. Livingston. 1975. Apparatus for
quantitative determination of locomotor activity patterns of aquatic
organisms using infrared light-emitting diodes. J. Fish. Res. Bd Can.
32:1884-1886.
Danil'Chenko, 0. P. 1977. The sensitivity of the fish embryo to the action
of toxic substances. J. Ichthyol. 17:455-463.
Denton, T. A. 1973. Fish chromosome methodology. Charles C. Thomas,
Springfield, 111.
Dingerkus, G. and L. D. Uhler. 1977. Enzyme clearing of alcian blue
stained whole small vertebrates for demonstration of cartilage. Stain
Techno!. 52(4):229-232.
Dorris, T. C., tf. Gould and C. R. Jenkins. 1960. Toxicity bioassay of oil
refinery effluents in Oklahoma. Pages 276-285 in Biological problems
in water pollution. Trans. 1959 Semin. R. A. Taft Sanit. Eng. Cent.
Ellis, M. M. 1937. Detection and measurement of stream pollution. Bull.
Bur. Fish. 48:365-437.
Finney, D. J. 1971. Probit analysis, Cambridge University Press,
Cambridge.
Fishbein, L. 1975. Teratogenic, mutagenic, and carcinogenic effects of
insecticides. Pages 555-603 in C. F. Wilinson, ed. Insecticide
biochemistry and physiology. Plenum Press, N.Y.
Fowler, H. W. 1928, Fishes from Florida and the West Indies. Proc. Nat.
Acad. Sci. Philadel. 80:451-473.
Giam, C. S., H. S. Chan, G. S. Neff, and E. L. Atlas. 1978, Phthlate ester
plasticizers: a new class of marine pollutant. Science 199:419-421.
Gibson, C. I., F. C. Tobe, R. E. Schirmer and J. W. Blaylock. 1980.
Bioaccumulation and deparation of bromoform in five marine species.
Pages 517-533 in. R. L. Jolley, W. A. Brungs and R. B. Cumming, eds.
Water ch1orination: environmental impact ahd health effects, vol. 3,
Ann Arbor Science, Ann Arbor, Mich.
122
-------
Glickman, A. H., C. N. Statham, A. Hu and J. Lech. 1977. Studies on the
uptake, metabolism and disposition of pentachlorophenol and penta-
chloroanisole in rainbow trout. Toxicol, Appl. Pharmacol. 41:649-658,
Gold, J. R. 1979. Cytogenetics. Pages 353-405 i£ W. S. Hoar, D, S. Randall
and J. I?. Britt, eds. Fish physiology, vol. 8. Academic Press, H.Y.
Goto, K., T. Akematsu, H. Shlmazu and T, Suqiyama. 1975, Simple
differential Giemsa staining of sister chromatids after treatment with
photosensitive dyes and exposure to light and the mechanism of staining.
Chromosoma. 53:223-230.
Hanumnnte, M. M. and S. S. Kulkarn. 1979. Acute toxicity of two mollusci-
cides, mercuric chloride and pentachlorophenol to a freshwater fish
{Channa qachua). Bull. Environm. Contam. Toxicol. 23:725-727.
Her.c'ricks, J. D., J. H. Wales, R. 0. Sinnhuber, J. E. Nixon, P. M.
Loveland, and R. A. Scanlan. 1980, Rainbow trout (Salmo gairdneri)
embryos: a sensitive animal model for experimental carcinogenesis.
Federation Pr'oc. 39:3222-3229,
Harrington, R. W., Jr. 1959. Delayed hatching in stranded eggs of the
marsh killifish, Fundulus confluentus. Ecology 40:430-437.
Harrington, R. H., Jr. 1961. Oviparous hermaphroditic fish with internal
self-fertilization. Science 134:1749-1750.
Harrington, R. W., Jr. 1963. Twenty-four-hour rhythms of internal self-
fertilization and of oviposition by hermaphrodites of Rivulus
roartHOratus. Physiol. Zool. 36:325-241.
Harrington, R. W., Or. 1967. Environmentally controlled induction of
primary male gonochorists from eggs fo the self-fertllizing hermaphro-
ditic fish, Rivulus marmoratus(Poey). Biol. Bull. 132:174-199,
Harrington, R. M., Jr. 1968. Delimitation of the thermolabile pheno-
critical period of sex determination and differentiation in the
ontoqeny of the normally hermaphroditic fish Rivulus marmoratus (Poey).
Physiol. Zool. 41:447-460.
Harrington, R. W., Jr. 1971. How ecological and genetic factors interact
to determine when self-fertilizing hermaphrodites of Rivulus marmoratus
change into functional secondary males, with a reappraisal of the modes
of intersexual ity among fishes. Copeia 1971: 389-432.
Harrington, R. W., Jr. 1975. Sex determination and differentiation among
uniparental homozygotes of the hermaphroditic fish Rivulus marmoratus
(Cyprinodontidae: Atheriniformes). Pages 249-262 in^ R. Reinboth, ed.
Intersexuality in the animal kingdom. Springer-Verlag, N.Y.
123
-------
Harrington, R. VI., Or. and R. A. Crossman, Jr. 1976. Teiroerature Induced
meristic variation among three homozygous genotypes (clones) of the
slef-fertilizing fish Rivulus marmoratus. Can. J. Zool. 54:1143-1155.
Harrington, R. H., Jr. and K. D. tollman. 19b8. The homozygosity of clones
of the self-fertilizing hermaphroditic fish Rivulus mimioratus Pocy
{Cyprinodonticlae, Atheriniformes}. Amer. Nat. 102:337-343.
Harrington, R. W., Jr. and L. R. Rivas. 1958. The discovery in Florida
of the cyprinodont fish. Rivulus marmoratus, with a redescription and
ecological notes. Copeia 1953: 1*25-130.
Hastings, R. N. and R. W. Verger. 1971. Ecology and life history of the
diamond killifish, Adenia xenica. Amsr. Midi. Nat. 86:276-291.
Ishikawa, T., T. Shimamine and S, Takayama. 1975. Histologic and
electron microscopy observations oii diethylnitrosamine-induced
hepatomas in small aquarium fish (Oryzias latipes). J. Natl. Cancer
Inst. 55:901-912.
Ishikawa, T. and S. Takayama. 1979. Importance of hepatic neoplasms
in lower vertebrate animals as a tool in car.cer research. J. Toxicol.
and Environ. Health 5:537-550.
Iwama, G. K. and G. 1. Greer, 1979. Toxicity of sodium pentachlorophenol
to juvenile chinook salmon under conditions o* hicth loading density and
continuous flow exposure. Bull. Environin. Contam. Toxicol. £3:711-716.
Janes, M. 0. and J. R. Bend. 1980. Polycyclic aromatic hydrocarbon
induction of cytochrome P-450 dependent mixed function oxidases in
marine fish. Toxicol, Appl. Pharmacol. 54:117-133,
Johnson, I. S. 1980. The reproductive biology of Fundulus majalis
(Walbaum). Masters Thesis. College of Charleston. Charleston, S.C.
Jolley, R. L. 1975.. Chlorine-containing organic constituents in sewage
effluents. J. Water Poll. Control Fed. 47:601.
Kallman, K. D. and R. W. Harrington, Jr. 1964. Evidence for the existence
of homozygous clones in the self-fertilizing hermaphroditic fish,
Rivulus marmoratus Foey. Biol. Bull, 126:101-114,
Kligerman, A. 1979. Induction of sister chromatid exchanges in the central
mudminnow following in vivo exposure to mutagenic agents. Mutation
Res. 64:205-217. " "
Kligerman, A. and S. E. Bloom. 1976. Sister chromatid differentiation and
exchanges iii adult mudminnows (Umbra limi) after in vivo exposure to
5-brotnodeoxyurieline. Chromosoraa 56:101-109.
124
-------
Kligerman, A., S. E. Blow, arid W. H. Howell. 1975. Umbra limi: a model for
the study of chromosome aberrations in fishes. Mutation Res, 31:225-233.
Koenig, C. C. and R. J. Livingston. 1976. The embryo!agical development of
the diamond killifish (Adinia xenica). Copeia 1976; 435-445.
Koenig, C. C. and McLean. 1980. Rivulus marmoratus: A unique fish useful
in chronic marine bioassays of halogenated organics. Pages 827-833 in.
R. L. Jolley, W. A. Brungs and R. B. Cummings, eds. Water chlorination:
environmental impact and health effects, vol. 3. Ann Arbor Science,
Ann Arbor, Mich.
Kopperman, H. L., 0. W. Kuehl and G. E. Glass. 1979. Chlorinated compounds
found in water treatment effluents and their capacity to bioaccumulate.
Pages 311-328 in R. L. Jolley, ed. Water chlorination: environmental
impact and health effects, vol. 1. Ann Arbor Science, Ann Arbor, Mich.
Kraybill, H. F., C. J. Dawe, J. C. Harshbarger and R. G. Tardoff, eds. 1977.
Aquatic pollutants and biologic effects with emphasis on neoplasia.
Annals of the New York Academy of Sciences. Vol. 298, N.Y.
Kristensen, t. 1970. Competition in three cyprinodont fish species in the
Netherlands Antilles. Stud. Fauna Curacao 32:82-101.
laale, H. W. and W. Lerner. 1981. Teratology and early fish development.
Amer. Zool. 21:517-533.
\
Laughlin, R, B., Jr., J. M. Neff, Y. C. Hrung, T. C. Goodwin, and C. S. Giam.
1978. The effects of three phthalate esters on the larval development of
the grass shrimp Palaemonetes pugio(Holthuis). Water, Air, and Soil
Pollut. 9:323-336.
Lewis, W. M., Jr. 1970. Morphological adaptations of cyprinodontoids for
inhabiting oxygen deficient water. Copeia 1970: 319-326.
Lindsey, C. C. and R. W. Harrington, Jr. 1972. Extreme vertebral variation
induced by temperature in a homozygous clone of the self-fertilizing
fish Rivulus marmoratus. Can. J. Zool. 50:733-744.
Lowry, 0. M., N. J. Rosenbrough, A. L. Farr and R. J. Randall. 1951. Pro-
tein measurement with the Folin phenol reagent. J. Biol. Chan. 193:265-
275.
Haddock, M. B. and J. J. Kelly. 1980. A sister chromatid exchange assay for
detecting genetic damage to fish exposed to mutagens and carcinogens.
Pages 835-844 in. R. L. Jolley, W. A. Brungs and R. B. Cunning, eds. Water
chlorination: environmental impact and health effects, vol. 3. Ann
Arbor Science, Ann Arbor, Mich.
Madin, S. H. and N. B. Darby. 1958. Established kidney cell lines of normal
adult bovine origin. Proc, Soc. Exp. Biol. Med. 98:574-576.
125
-------
Mahoney, J. E., F. H. Mldlige and D. G. Denel. 1973. A fin rot disease
of marine and euryhaline fishes in the New York Bight. Trans. Amer,
Fish. Soc, 102:596-605.
Hassero, E. J.» J. C. Hassaro and R. W. Harrington, Jr. 1975. Biochemical
comparison of genetically different homozygous clones (isogenic, uni-
parental V.nes) of the self-fertilizing fish Rivulus marrcoratus Poey.
Pages 439-453 in. C. L. Markert, ed. Isozymes.Academic Press, N.Y.
Matsushima, T. and T. Sugimura. 1976. Experimental carcinogenesis in small
aquarium fishes. Prog, Exp. Tumor Res. 20:367-379.
Mayer, F. L., D. L. Stalling and J. L. Johnson. 1972. Phthalate esters as
environmental contaminants. Nature 238:411-413.
Mearns, A. J, anc M. Sherwood. 1974. Environmental aspects of fin erosion
and tumors in southern California dover sole. Trans. Amer. Fish. Soc.
103:799-810.
Miller, J. A. 1970, Carcinogenesis by chemicals: an overview, G.H.A.
Clowes memorial lecture. Cancer Res. 30:559-576.
Oseid, 0. and L. L. Smith, Jr. 1972. Swimming endurance and resistance to
copper and ma lathi on of bluegills treated by long-term exposure to sub-
lethal levels of hydrogen sulfide. Trans. Amer. Finh. Sco. 101:620-625.
Perry, P. E. 1980. Chemical mutagens and sister chromatid exchange.
Payes 1-39 in F. 6. de Serres and A. Hollaender, eds. Chemical mutagens,
vol. 6, Plenum Press, N.Y.
Perry, P. and H, J, Evans. 1975. Cytological detection of mutagen-
carcinogen exposure by sister chromatid exchange. Nature (London)
258:121-125.
Pfuderer, P. and A. A. Francis. 1975. Phthalate esters: heart rate
depressors in the goldfish. Ball. Environm. Contain. Toxicol. 13:275-279.
Pli ss, G. B. and V. V. Khudoley. 1975. Tumor induction by carcinogenic
agents 1n aquarium fish. J. Nat. Cancer Inst. 55:129-133.
* ' <
Pruitt, G. W., B. J. Grantham and R. H. Pierce, Jr. 1977. Accumulation
and elimination of pentachlorophenol by the bluegill, lepomis macrochirus
Trans. Amer. Fish. Soc, 106:462-265,
Rao, K. R., ed. 1978. Pentachlorophenol. Plenum Press, N.Y.
Rivas, L. R. 1945. The discovery and redescription of the types of
Rivulus marmoratus Poey, a cyprinodor.t fish from Cuba. J. Wash. Acad.
Sci. 35:95-97.
126
-------
Sanders, H. 0., F» C, Mayer and D, F. Walsh. 1973. Toxicity, residue
dynamics, and reproductive effects of phthalate esters in aquatic
Invertebrates. Environm. Res. 6:84-90.
Sayo, S., T, Matsushima, N. Tanaka, T. Sugimura and F. Takashima. 1973.
Hepatic tumors in the guppy (Lebistes reticulatus) induced by
aflayoxin B., Dimethylnitrosamine, and 2-acetylaminoflourene. J. Natl.
Cancer Inst. 50:765-778.
Schmahl, D. and H. Osswald. 1967. Carcinogenesis in different animal
species by diethylnitrosamine. Experientia 15:497-498.
Schroeder, J. H. 1973. Teleosts as a tool in mutation research. Pages
91-99 in J. H. Schroeder, ed. Genetics and mutagenesis of fish.
Springer-Verlag, N.Y.
Shaw, C. R. and R. Prasad. 1970, Starch gel electrophoresis of enzymes -
A compilation of recipes. Biochemical Genetics 4:297-320.
Shealy, R. I. 1979. Daily variation in the ther.nal adaptation of post-
1arval Fundulus heteroclitus. Masters Thesis. College of Charleston,
Charleston, S C.
Shoor, U. P. and J. A. Couch. 1979. Correlation of mixed-function
oxidase activities with ultrastructural changes in the liver of a marine
fish. Cancer Bioehem. Biophys. 4:95-103.
Singh, A. R., W. H. Lawrence and J. Autian. 1972. Teratogenicity of
phthalate esters in rats. J. Pharm. Sci. 61:51-55.
Sinnhuber, R. D., 0. D. Hendricks, J. H. Wales and G. B. Putnam. 1977.
Neoplasms in rainbow trout, a sensitive animal model for environmental
'carcinogei.esis. Ann. N.Y. Acad. Sci. 298:389-408.
Sinnhuber, R. 0., J. H. Wales, J. D. Hendricks, G. B. Putnam, J. E. Nixon
and N. E. Pawlowski. 1977. Trout bioassay of mycotoxins. Pages 731-
744 in Mycotoxins in human and animal health. Pathotox Publishers.
Sol berg, A. N. 1938. The development of a bony fish. The Prog. Fish Cult.
Mem. 1-131 No. 40:1-19.
Sontag, J. H. 1977. Aspects in carcinogen bioassay. Pages 1327-1338 jm
H, H. Hiatt, J. 0. Watson and J. A. Winsten, eds. Origins of human
cancer, book C, human risk assessment. Cold Spring Harbor conferences
on cell proliferation, vol. 4.
Stanton, M. F. 1965. Diethylnitrosamine-induced hepatic degeneration and
neoplasia in the aquarium fish, Brachydanio rerio. J. Natl. Cancer
Inst. 34:117-123.
127
-------
Wourms, J. P. 1972a. Developmental biology of annual fishes. I. Stages
in the normal development of Austrofundulus myersi Dahl. J. Exp, Zool.
182:143-153, "
Wourms, J. P. 1972b. The developmental biology of annual fishes. II.
Naturally occurring dispersion and reaggregation of blastomeres during
the development of annual fish eggs. J. Exp. Zool. 182:169-200.
Wourms, J. P. 1972c. The developmental biology of annual fishes, ill.
Pre-embryomc :nd embryonic diapause of variable duration in the eggs
of annual fishes. J. Exp. Zool, 182:389-414.
Ziehen» C. D., R. E. Pine, A. D. mills and R. K. Cunningham. 1970. Field
toxicity studies and juvenile salmon distribution in Port Angeles
Harbor, Washington. J. Water Pollut. Contr. Fed. 42:229-236.
ZitkO, V. 1972. Determination, toxicity, and environmental levels of
phthlete plasticizers. Fish. Res. Bd. Canada, Tech. Rep. No. 344.
129
-------
|