DEVELOPMENT OF A CARCINOGEN ASSAY  SYSTEM
       UTILIZING ESTUARINE FISHES
                   by
              B. J. Martin
          Department of Biology
 The University of Southern Mississippi
     Hattiesburg, Mississippi  39406
   EPA Cooperative Agreement CR 806212
             Project Officer
              John A. Couch
    Environmental Research Laboratory
       Gulf Breeze, Florida   32561
    ENVIRONMENTAL RESEARCH LABORATORY
                      !VND
 OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL^PROTECTION AGENCY
     GULF BREEZE,/FLORIDA   32561

-------
                                 DISCLAIMER

   The information in this document has been  funded  wholly  or  in part by the
U.S. Environmental Protection Agency  under cooperative agreement number
CR 806212 to B. J. Martin, the University of  Southern  Mississippi/
Hattiesburg, Mississippi,  it has been subject to  the Agency's  peer  and
administrative review,  and it has been approved for  publication. The
contents reflect  the views and policies of the Agency.
                                      11

-------
                                   FOREWORD

     The protection of our estuarine and coastal areas from  damage  caused by
toxic organic pollutants requires that regulations restricting  the  introduction
of these compounds into the environment be formulated on a sound  scientific
basis.  Accurate information describing dose-response relationships for
organisms and ecosystems under varying conditions is required.  The EPA
Environmental Research Laboratory, Gulf Breeze, contributes  to  this information
through research programs aimed at determining:

       the effects of toxic organic pollutants on individual species and
       communities or organisms;

       the effects of toxic organics on ecosystem processes  and components;

       the significance of chemical carcinogens in the estuarine  and marine
       environments.

     The development and use of select species of fishes as  carcinogen assay
subjects offer new tools for the assessment of carcinogenic  risks in the
environment.  This report summarizes data on the sheepshead  minnow  (Cyprinodon
variegatus) as such a test organism.  The results concern the response to known
carcinogens from subcellular to organismic levels, and thus  should  help provide
a rational basis for use of fishes in carcinogen risk assessment.
                                           Henr  F. Enos
                                           Director
                                           Environmental Research  Laboratory
                                           Gulf Breeze, Florida
                                     111

-------
                       ABBREVIATIONS AND SYMBOLS
BaP       — benzo(a)pyrene
BEN       — benzidine dihydrochloride
cm        — centimeter or centimeters
DCM       — dichloromethane
DENA or
  DEN     — diethylnitrosamine
DOC       — dissolved organic carbon
HBSS      — Hank's Balanced Salt Solution
HCG       — human chorionic gonadotropin
i.p.      — intraperitoneal
LC-50     — concentration resulting in 50% mortality in 96 hours
M         — molar
mg        — milligram or milligrams
ml        — milliliter or milliliters
MS222     — tricaine methane sulfonate, Sigma
ng        — nanogram or nanograms
PAS       — periodic acid-Schiff
PFU       — Plaqiae-Forming Units
ppb       — parts per billion
ppm       — parts per million
ppt       -- parts per thousand
pi        — microliter or microliters
ym        -- micrometer or micrometers
                                   IV

-------
                               CONTENTS
Foreword 	
Abbreviations and Symbols 	  1V
Figures and Tables 	  vii
Acknowledgment 	vm
Abstract 	  ^x

   1.  Introduction 	   -'-
   2.  Conclusions and Recommendations 	   2
   3.  Materials and Methods 	   4
           Facilities 	   4
           Specimen Care and Collection 	   4
           Benzidine Analysis 	   5
           Toxicity Tests 	   5
           Chronic And Pulse Exposures 	   5
           Food Contamination 	   5
           Injection Experiments 	   6
           Benzidine Exposure, C_. variegatus Embryos 	   6
           SHF-1 Cell Culture Exposures 	   6
           Early Embryogenesis, C_. variegatus	   7
           Gross And Histological Anatomy of the
             Post-Pharyngeal Digestive Tract, C_. variegatus	   7
           Peripheral Blood Cell Morphology, C_. variegatus 	   7
           Aseptic Embryo Preparation 	  10
           Aseptic Embryo Technique 	  10
           Embryo-Primary Cell Culture Technique 	  10
           Primary Hepatocyte Cell Culture Technique 	  11
           Immunological Studies 	  11
           Effects of Salinity, DOC, and Light on
             DENA Extractability 	  13
   4.  Results and Discussion 	  15
           Toxicity Tests 	  15
           Chronic Benzidine Exposures, C_. variegatus	  15
           Pulse Benzidine Exposures, C_. variegatus  	  16
           Diethylnitrosamine Exposures 	  19
           Benzo(a)pyrene Exposures 	  19
           Benzidine Feeding Experiments  	  19
           Diethylnitrosamine Feeding Experiments  	  20
           Benzidine Injection Experiments  	  20
           Diethylnitrosamine Injection Experiments  	  20
           Benzidine Exposures, C_. variegatus Embryos  	  21
           SHF-1 Cell Culture Exposures 	  21
           Early Embryogenesis, C. variegatus  	  23
                                   v

-------
           Gross and Histological Anatomy of the
             Post-Pharyngeal Digestive Tract, C_. variegatus 	   24
           Peripheral Blood Cell Morphology, C_. variegatus 	   29
           Aseptic Embryo Technique 	   33
           Embryo-Primary Cell Culture Technique 	   36
           Primary Hepatocyte Cell Culture Technique 	   37
           Immunological Studies 	   38
           Effects of Salinity, DOC, and Light on
             DENA Extractability 	   42

References 	   46
                                  VI

-------
                                   FIGURES
Number
                                                                        Page
   1   Penetration of the chorion 	    ^




   2   Rotation and closure of forceps	    8




   3   Splitting of the chorion 	    8




   4   Normal liver tissue 	   17




   5   Transition from normal liver to lesion 	   17




   6   BEN-induced liver lesion 	   17




   7   Tubular profiles in liver lesion 	   17




   8   Comparison of serum electrophoresis profiles 	   40




   9   Effect of salinity on DENA recovery 	   43




  10   Effect  of DOC on DENA recovery 	   44




  11   Effect of light on DENA recovery 	   45




                                   TABLES




Number                                                                  Page




   1   LC-50 of BEN, £. variegatus 	   15




   2   Exposure of SHF-1 cells to BaP 	   22




   3   Survival times of aseptic embryos 	   35




   4   Comparison of blood volumes collected 	   39




   5   Effects of serum MS2 bacterophage titer 	   42
                                     Vll

-------
                               ACKNOWLEDGMENT
     This report was supported by cooperative agreement (#CR 806212)  in
the NCI/EPA Collaborative Program, Project No. 3 "Effects of Carcinogens,
Mutagens and Teratogens in Non-human Species  (Aquatic Animals)" admin-
istered by the Gulf Breeze Environmental Research Laboratory.

     Appreciation is expressed to the following individuals who
contributed significantly to this project:  W. Greenwich, S. Bennett,
Dr. R. D. Ellender,- R. Fricke, M. Guess, S. Hillebert, R. Long,
B. Meador, Dr. K. McMurtrey, Dr. B. L. Middlebrooks, M. Moreno,
Dr. R. Pierce, Jr., Dr. R. Porter and J. Winstead.
                                    Vlll

-------
                                  ABSTRACT
     The overall objective of this research was the development of systems
previously devised in our laboratory to assay the effects of chemical car-
cinogens on marine teleosts.  The results include the following.

     The LC-50  (96 hours) for benzidine dihydrochloride (BEN) with respect
to Cyprinodon variegatusllwas determined to be ca. 64 ppm.
     Exposure of C_. variegatus to weekly contaminations of 1 ppm BEN caused
some individuals to develop liver lesions at 25-29 weeks.  The livers of
these individuals contain large fibrotic regions within which a prolif-
eration of various types of tubular profiles may be observed.

     Efforts to accomplish long-term exposure of teleosts to BEN or
diethylnitrosamine  (DENA) via contaminated food met with limited success
because of the toxicity of these compounds at the concentrations used.
Intraperitoneal injections of these compounds also proved to be toxic over
a wide range of concentrations.

     Exposure of early C_. variegatus embryos to BEN at various concentrations
produced abnormalities at concentrations of 50 ppm and above.  Anomalies in
the order of frequency of occurrence were :  tubed heart syndrome with
distended pericardia, poor circulation, sparse distribution of melanophores
around yolk, inability to hatch, abnormal head morphology, scoliosis, and
faint RBC pigmentation.

     Acute toxicity concentrations were established for benzo (a)pyrene  (BaP) ,
BEN and DENA with respect to a cell line from Archosargus probatoc ephalus
(the sheepshead) .  Long-term exposures provided evidence that BaP and BEN
have mutagenic effects on this cell line.

     A dechorionation technique was developed to observe better detailed
cellular and subcellular activities during early embryonic development
of £. variegatus.  Employment of this technique to observe inverted
blastoderms provided evidence that the ectodermal cells that cover the
yolk travel from the superficial blastoderm via a pathway along the
blastoderm floor.

     Detailed studies of the gross and histological structure of digestive
tract and histological studies of the peripheral blood cells of C. variegatus
were conducted.
                                     IX

-------
     Three novel techniques were developed to study the effects of car-
cinogens on C_. variegatus at the cellular level:  an aseptic embryo tech-
nique that provides the opportunity to study embryos in a sterile
environment, and embryo-primary cell culture technique that incorporates
in one system characteristics of both intact embryos and primary cell
cultures, and a primary hepatocyte cell culture technique that will be
employed to study the effects of carcinogens on teleost hepatocytes.

     In order to study the immune system of C. variegatus, standard
immunological techniques were miniaturized.  Serum electrophoresis
disclosed considerable variation between BEN-exposed and unexposed fish,
and the presence of antibody-forming cells in spleen suspensions from
£. variegatus immunized with human type 0 erythrocytes were demonstrated
by a modified immune rosette procedure.

     Finally, an improved method for the extraction of DENA from water was
developed.  This method was used to observe significant decreases in the ex-
tractability of DENA with increasing salinity.  A loss of DENA by
photodegradation in the presence of dissolved organic carbon was demon-
strated.

-------
                                  SECTION 1

                                INTRODUCTION

     In recent years, the public has become increasingly aware of the
importance of environmental pollutants as cancer-causing agents.  With
this awareness has come the realization that a major portion of the
world's cancer incidence is environmentally related and, consequently, pre-
ventable.  Thus, the public sector has begun to demand more rigorous environ-
mental safety assessment (1).   Although, we have made progress in recent
years in our methods for detecting chemical carcinogens and in our under-
standing of the basic mechanisms of neoplasia, our capability of accurately
predicting the human health hazards that may result from the myriad of
chemical pollutants entering the environment has not kept pace with our
increasing needs in this respect.

     Murine assay systems continue to be the mainstay for testing chemical
carcinogens; however, there is a recognized need for alternate systems.
Since existing systems are not particulary amenable to assessment of the
aquatic environment, an environment that is experiencing an ever increasing
quantity and a variety of potentially dangerous pollutants, it is imperative
that we develop valid carcinogen test systems appropriate for this environ-
ment.

     Teleost fishes are obvious candidates for this role, and there is
considerable evidence for the susceptibility of teleosts to carcinogenic
agents (2, 3).  Thus, in a previous project, whose purpose was the study of
the effects of certain carcinogenic compounds that are components of
petroleum products, we developed systems for long-term exposures of
Cyprinodon variegatus (the sheepshead minnow) at laboratories in inland
locations (4).  This species was selected because it is a common species in
the Gulf Region, its biological characteristics seemed ideal, and it had
already been employed extensively in toxicologic assays.  This initial
project demonstrated that C_. variegatus was a suitable model for chronic
testing and provided evidence that such a system, when fully developed,
could be employed extensively for carcinogen assay.

     Therefore, the principal objective of the present project was the
comprehensive development of the C_. variegatus assay system as a tool for
the biological assessment of suspect carcinogens in the aquatic environment.
An additional goal was to conduct a search for biochemical or cellular
changes that might indicate the usefulness of this species as an ''early
warning detector'1 of teratogenic, mutagenic, or carcinogenic substances in
the estuarine environment.

-------
                                   SECTION 2

                        CONCLUSIONS AND RECOMMENDATIONS

     This project demonstrates the feasibility of employing small estuarine
teleosts in laboratory assays of suspect carcinogens.  The relatively low cost
of maintaining such fish and their apparent short latency period (possibly only
5 to 6 months) provide an economic advantage over traditional murine bioassay
systems.  Experience with £. variegatus indicates that a requirement for
success with such systems is that the fish must be parasite-free and healthy,
and their state of health must be carefully maintained during the course of the
exposures.  The fish normally tolerate the stress resulting from the experi-
mental conditions; however, any additional stress due to poor health or to less
than an optimum environment, or both, may produce a sufficient amount to
quickly compromise the system.  Thus, feral fish brought into the laboratory
must be treated for ectoparasites and be in an excellent state of health prior
to their use in experiments.  Furthermore, all experiments must be designed and
conducted in a manner that insures the constant maintenance of a high quality
environment.  In fact, our experience suggests that the best approach may be to
use feral fish only as breeders and to conduct all laboratory experiments with
laboratory-bred individuals.

      The toxicity problems encountered in this project indicate that one
 should always determine the LC-50 of a compound as a point of reference
 prior to making decisions concerning concentrations to be used in long-
 term chronic exposures.

      Food contamination did not produce observable lesions and the i.p.
 injections, even over a wide range of concentrations, proved, in most
 cases, to be toxic;  therefore, additional research will be necessary to
 accurately assess these modes of exposure of C^. variegatus to carcinogens.

      The repeatable induction of a liver lesion, though its pathological
 classification is currently problematic, by weekly I ppm contaminations of
 the water with BEN demonstrates the usefulness of this mode of exposure.
 Thus,  it is recommended that this type of experiment be continued until the
 lesion is completely characterized, its incidence determined, and its
 latency period accurately established.  Additionally, experiments with
 other known mammalian carcinogens should be conducted to provide comparative
 data relative to this system.

      The variety of anomalies resulting from the exposure of early C_.
 variegatus embryos to BEN indicates that this system is likely to be an
 effective tool in the study of carcinogenesis and teratogenesis.  The

-------
studies of the early embryonic development of normal C.  var'iegatus provide
comparative data that is needed to accomplish detailed analyses of the
carcinogen induced anomalies.  In addition to the continued study of these
BEN-induced anomalies, it would seem logical to conduct exposures of this
type with a number of other known mammalian carcinogens.

     The occurrence of multilayered foci in SHF-1 cell cultures exposed to
both BaP and BEN indicates that teleost cell cultures, like mammalian cell
cultures, can be employed in this type test.  The effectiveness of this
system is enhanced by the fact that such exposures are less costly than
exposures of mammalian cells.
     The anatomy of the digestive tract of C. variegatus proved to be
similar to other Cyprinodontid species; however, the detailed study
accomplished in this project provides the background data needed to
evaluate any lesions induced in the digestive tract.  The studies of the
morphology of peripheral blood cells provide baseline knowledge  that will
allow the use of hematology, a traditional method for detecting pathology,
in evaluating the effects of carcinogens.
     The three cellular techniques (aseptic embryo, embryo-primary cell
culture, and primary hepatocyte cell culture) are all innovative techniques
that possibly can be developed into useful rapid test methods for carcinogens.
Because of the obvious need for development of improved rapid test
procedures, and the likelihood that these techniques may have even broader
applications, their continued development is recommended.

     The miniaturization and adaptation of a variety of standard
immunological techniques accomplished in this project provides the
tools needed to conduct sophisticated studies of the immune system of
C_. variegatus.  Thus, it is now possible to use this species in the study
of the intriguing relationship between carcinogenesis and the immune
system.  The significant effects of low-level chronic exposure to BEN
on the serum profiles of C^. variegatus provides a preliminary indication
that these techniques are likely to be quite valuable, therefore, it is
recommended that considerable effort be devoted to studies of the
immune systems of normal and carcinogen-exposed C. variegatus.

-------
                                  SECTION 3

                            MATERIALS AND METHODS

Facilities

     The long-term exposures of this project were conducted in a
laboratory located four miles from the main campus of the University
of Southern Mississippi.  The remote location and design of the
facility provides safe conditions for working with carcinogens.  The
laboratory allows control of the ambient environment and provides for
the safe handling and disposal of contaminated water.  Specimens are
maintained in closed-circulating systems, 950-liter cylindrical fiber-
glass tanks, 185-liter fiberglass-coated plywood tanks (4), or 10-gallon
glass aquaria.


Specimen Care and Collection

      Approximately 9600 Cyprinodon variegatus (sheepshead minnows)  were
utilized during the course of the study.  About 86% of these specimens were
seined from a marshy tidal entrance at Range Point on Santa Rosa Island
near the U.S. EPA Environmental Research Laboratory, Sabine Island,  Gulf
Breeze, Florida.  Approximately 13% were collected from a similar tidal
stream on the north side and at the east end of Horn Island on the Mississippi
Gulf Coast near Pascagoula, Mississippi.  All collected specimens were
examined for gross lesions and treated for 30 minutes with 1:4000 formalin to
remove parasites.  The collected fish were acclimated to 5-15 ppt artificial
seawater (Rila Mix, Rila Products, Teaneck, NJ) at least 2 weeks prior to
their use in experiments.  About 800 of the C_. variegatus were hatched and
raised in the laboratory.

     The 2100 Ictalurus punctatus (channel catfish) utilized were obtained
as 1- to 3-inch fingerlings from local commercial fish hatcheries.  These
fish were also formalin-treated and maintained in the laboratory for at
least 2 weeks prior to their use in experiments.

     The C_. variegatus were fed commercial fish flakes (TetraMin or Monolake
Flakes, F.A. Martiny and Sons, New Orleans, LA).  The diets of immature
specimens were supplemented with live or frozen brine shrimp  (Artemia).
The channel catfish were fed commercial catfish food  (Purina).

-------
Benzidine Analysis

     The HPLC methodology employed for the chemical analysis of BEN
dihydrochloride, Sigma, B-3383) was a modification of the techniques described
by Riggins and Howard  (5).   Liquid chromatography on reversed phase columns
was employed with detection of BEN by electrochemical oxidation (HPLC-ED) and
absorption of light at 214 nm  (HPLC-UV).
Toxicity Tests

     The toxicity of BEN to C_. variegatus was determined by using routine
methods for determination of the LC-50.  A ten-gallon aquarium was used for
each experimental condition, and ten fish were placed in each aquarium.  The
fish were first acclimated and then BEN was added to provide concentrations
of 16, 25, 40, 64, and 102 ppm.  The experiment was repeated four times and
the data were combined.
Chronic and Pulse Exposures

     Stock solutions of BEN were prepared in distilled water.  An appropriate
amount of the stock solution was then introduced into the systems to obtain
the final concentrations desired.  Chemical analysis disclosed that when the
systems were contaminated with 1 ppm BEN, the concentration began to drop
rather rapidly after 2-3 days.  It was determined that weekly contaminations
at a theoretical concentration of 1 ppm would maintain a concentration of
BEN that varied from a high  of about 1.5 ppm to a low after 7 days of about
0.5 ppm.  All pulse exposures involved a "one time" introduction of BEN into
the systems.

     Stock solutions of DENA  (N-Nitrosodiethylamine, Sigma, N-0756) were
prepared in distilled water and introduced directly into the systems.

     Acetone was used as a carrier for the introduction of BaP
(benzo(a)pyrene, Sigma, B-3500) into the systems because of the low water
solubility of BaP.  A stock solution was prepared by placing 4.35 mg of
BaP in 10 ml of acetone.  An appropriate amount of the stock solution was
then placed in the systems to obtain the theoretical concentrations desired.
Appropriate acetone controls were conducted.
Food Contamination

     BEN-contaminated food was prepared by adding 1 gm BEN per 100 gm dry
food.  Distilled water was added and the food was thoroughly mixed.  The
contaminated food was stored frozen and thawed prior to feeding.

     The DENA-contaminated food was prepared by mixing 1 ml of DENA per
100 gm dry food.  This food was prepared and stored utilizing the same
procedure as that employed for the BEN-contaminated food.

-------
Injection Experiments

     A 27-guage hypodermic needle was employed to make i.p. injections of  the
specimens along the ventral surface of the abdomen.  The volume of  the
injection fluid was maintained at 50 or  100 yl to avoid producing a trauma-
tizing pressure in the abdominal cavity.  Control specimens were sham-injected
with equal volumes of distilled water.   BEN was injected as an aqueous
solution, and DENA was injected as concentrated DENA or as an aqueous solution.


Benzidine Exposure, C_. variegatus Embryos

     Spawning was induced in  female C_. variegatus with three i.p. injections
of 50 IU of HCG  (human chorionic gonadotropin, Sigma) at 48-hour intervals.
The experiments were conducted at 13 parts per thousand salinity at 21°C.

     BEN exposures were accomplished in  finger bowls  (10 cm diameter) in
100 ml volumes.  The required dilutions  were prepared from a stock  solution
of 5 gm BEN per liter of 13 parts per thousand artificial sea water.
Experiments have been conducted at 5, 10, 25, 50, 75, 100 and 500 ppm
BEN.  All experiments were conducted at  17°-19°C, and the finger bowls
were covered with parafilm to reduce fluid loss due to evaporation.
SHF-1 Cell Culture Exposures

     A teleost cell line  from the  sheepshead  (Archosargus probatocephalus),
designated SHF-1  (6) was  employed  in  studies to determine its suitability
for carcinogen assays.  SHF-1 was  grown in Leibovitz's L-15 medium
supplemented 10% with fetal calf serum and 1% with L-glutamine.  The cells
were cultured in Falcon tissue culture flasks and subcultured weekly.

     BaP, BEN, and DENA were used  in  initial studies.  BEN and DENA were
guantitated by dilution in distilled  water or growth medium.  A stock
solution of 20% acetone in distilled  water containing 0.435 mg BaP per ml of
acetone was used to dissolve and quantitate the BaP.  In order to
determine their toxicity  to SHF-1  cells, various concentrations of each compound
were added to the growth  medium immediately following subcultivation of
the cells.  A particular  concentration of a compound was considered acutely
toxic if the cells failed to become attached to the floor of the culture
flask, or if the cells showed no growth within seven days and failed to
survive the subsequent subcultivation.  BaP and BEN exposures were becrun
at passage 18, and DENA exposures  were begun at passage  33.  Once acute
exposure concentrations for the three compounds were determined, subsequent
experiments were performed to observe the response of SHF-1 cells to
repeated  (one passage after another)  exposure.  Periodically, chromosome
spreads were performed to detect any  aberrations that might have occurred
in either control or exposed cultures.

-------
  Early Embryogenesis ,  C.  variegatus

       Spawning was induced naturally to  obtain the  approximately 2500  embryos
  utilized in the course of this study by maintaining  the  breeding stock at
  28°-30°C in a 12-hour light - 12-hour dark photoperiod.   After collection,
  eggs were placed in  10 cm diameter  finger bowls  and  maintained at 28°-30 C.
  Prior to microscopic observation, fertilized eggs  were cleaned by rolling them
  on lens paper and then washing them with tap water.   Observations were made
  both with the chorion intact and after  dechorionation.   Up to 5 eggs  were
  placed in saline water in each depression of a depression slide and covered
  with a coverslip.

O £-.   DechprT^nation ,  though extremely difficult  to accomplish, provides
  increased1 cellular resolution.   Therefore,  a technique was developed that
         dec^rl^Tation without causing significant damage to the embryonic
  and extraembfyonic membranes.  A number 11  scalpel blade is used to hold
  the egg in place.   The sharpened end of a pair of Dumont forceps is then
  pushed through the chorion (Figure 1) .   The f creep tip is then rotated,
  pushed into the perivitelline space, and closed against the opposite tip
  (Figure 2) .  The chorion will begin to split when it is rubbed at the
  lateral edge of the forceps by a scalpel blade (Figure 3) .  This process is
  continued until the chorion is excised.  Although the technique is quite
  tedious, with a reasonable amount of practice one can consistently obtain
  undamaged dechorionated eggs.


  Gross and Histological Anatomy of_ the Post-Pharyngeal Digestive
  Tract, C . variegatus

       For gross anatomical observations, the fish were pithed and abdominal
  incisions were made to expose the digestive organs.  The entire fish was then
  placed in Davidson's fixative (7)  for 5 minutes at 4°C.  The visceral walls
  were cut away to allow examination of the viscera in situ.

       For histological observations, the entire post-pharyngeal digestive
  tract was cut into 1 cm lengths and placed  in Davidson's fixative for
  24 hours, washed,  dehydrated, and embedded  in Paraplast.  The tissue blocks
  were serially sectioned at ca.  6 ym and consecutive sections were stained with
  one of three staining techniques;  routine Mayer's hematoxylin and eosin,
  Mallory's connective tissue stain (8), or periodic acid-Schiff  (PAS) counter-
  stained in aqueous fast green.


  Peripheral Blood Cell Morphology ,  C. variegatus

       Blood Collection:  Blood must be taken very carefully because it clots
  quite rapidly and excessive stress may damage or distort thrombocytes.  Two
  methods were used to remove blood for light and electron microscopy.  In the
  first method, samples were collected by allowing blood to flow  freely into
  the top of a heparinized capillary tube from the aorta after severing the
  caudal peduncle posterior to the anal vent.  Blood to be used for electron
  microscopy was pooled in a centrifuge tube  containing a cold solution of 2%

-------
                                 Figure 1
                                 Figure 2
                                Figure 3
Figure 1.  Diagramed penetration of the chorion by the tip of a
           pair  of  Dumont forceps.   Note the chorion  (C), perivitelline
           space (PS) ,  yolk (Y) ,  and blastomeres  (B) .

Figure 2.  Diagramed rotation and closure of the forceps.

Figure 3.  Splitting the  chorion with a scalpel blade.

-------
glutaraldehyde in 0.1 M cacodylate buffer, pH 7.2.  In the second method
developed by Winstead et al.  (9) , the caudal peduncle was severed and the
live  fish was placed in a centrifuge tube with the caudal end down.  The fish
was secured with a Kimwipe which was wrapped about the body of the fish.
Masking  tape was used to secure the upper portion of the Kimwipe, which
extended beyond the head of the fish to the outer side of the centrifuge
tube.  The posterior portion of the fish hung free over a solution of 2%
cacodylate buffered glutaraldehyde.  The fish was centrifuged at the lowest
speed on the clinical centrifuge.  No anticoagulant was needed, as blood
from the severed aorta dripped directly into the glutaraldehyde fixative
during centrifugation.

     Light Microscopy:  Blood to be used for light microscopy was taken
from heparinized capillary tubes and placed on slides or coverslips for
viewing  fresh preparations or stained smears.  The blood was viewed on an
Olympus  Vanox  light microscope, using either bright field or phase  contrast
microscopy.

     Blood smears were made by placing a small drop of blood from a
heparinized capillary tube on a #1 thickness coverslip.  Another identical
coverslip was placed on the original coverslip and the blood between the
coverslips was allowed to spread evenly to the margins.  The coverslips
were then rapidly pulled apart in a manner which caused the coverslips to
slide across one another in the plane of the coverslips.  The smears were
subsequently air dried and stained with either Wrights, Wrights-Giemsa or
May-Gruenwald blood stains for two to three minutes before washing in
distilled water.  Care was taken not to use an excess of blood as this
necessitated a longer drying time in air which allowed some breakdown of
cells and caused morphological changes in thrombocytes.

     Living cells were observed by placing a drop of blood on a slide,
covering it with a #1 thickness coverslip, and rapidly sealing it around
the edges with stopcock grease.  Blood kept in this manner could be
viewed at room temperature for as long as two hours and would remain
viable in the refrigerator for up to eight hours.  Congo red was used for
in vitro observations of the uptake properties of the blood cells.

     Mean red blood cell counts were made with an improved Neubauer's
hemacytometer.   A Hayem's solution was used which was made with 0.1 ml
ammonium salt heparin (1000 USP units/ml) in 20 ml of Hayem's diluting
solution.  Both red and white blood cell counts were taken.  An average
of twenty counts was used for total red and white blood cell counts.

     Electron Microscopy:  Blood from either of the collection methods
was centrifuged in a clinical centrifuge for three minutes at maximum
speed.  After twenty minutes in the 2% buffered glutaraldehyde solution,
the blood was washed three times at ten minutes per wash in the cacodylate
buffer at room temperature.   The glutaraldehyde was then replaced by a
1% osmium tetroxide in cacodylate buffer.  After twenty minutes in this
fixative at room temperature, the tissue was washed three times at ten
minutes per wash in cacodylate buffer at room temperature and dehydrated
through a graded ethanol series to absolute ethanol.  The pelleted blood

-------
sample was then removed and the buffy coat was cut from the top of the
pellet and sectioned into approximate 1 mm cubes.  These were infiltrated
with Spurr's embedding medium and placed in Beem capsules.  Thin sections
were cut with a diamond knife using a Sorvall Porter-Blum ultramicrotome
and examined on a Siemens Elmiskop 1A transmission electron microscope
at 80 Kv.
Aseptic Embryo Preparation

     Egg development was induced by the HCG injection method.  During the
harvest of eggs, the females were sacrificed or the eggs were milked into a Petri
dish containing artificial sea water by applying pressure along the lateral
surfaces of the gravid female.  Fertilization was accomplished by placing
macerated testes into a Petri dish containing the eggs.

     The fertilized eggs were maintained in Petri dishes with two daily
changes of artificial sea water.  On the third or fourth day, the eggs
were washed with several changes of artificial sea water and, under a
laminar flow hood, placed in Petri dishes containing sterile Hank's
Balanced Salt Solution  (HBSS) and 5% antibiotics (penicillin-streptomycin-
fungizone, Gibco).  The HBSS was changed three times before the eggs
were transferred to a fresh Petri dish containing sterile medium.  The
medium consisted of 9 ml of Leibovitz's L-15 medium plus 1 ml of anti-
biotics.  With sterile forceps or sterile wide-mouthed Pasteur pipettes,
the eggs were transferred to a second Petri dish containing sterile medium,
and incubated overnight at 37°C.  The eggs were then placed in a fresh
dish of sterile medium and incubated until all were dechorionated.
Aseptic Embryo Technique

     Five to fifteen embryos prepared according to the aseptic embryo
procedure were placed into 150 cm2 plastic flasks  (Corning) containing
50 ml of sterile medium.  The embryos were maintained at 25°C, and
monitored daily for viability.  Experiments were conducted in which
embryos were maintained in basal L-15 medium plus various combinations
of supplements.  In some experiments, the embryos were maintained in
sterile artificial sea water.  Pilot LD50 experiments were performed in
which the medium was contaminated with various concentrations of BEN.
Embryo-Primary Cell Culture Technique

     One embryo prepared according to the aseptic embryo procedure was
placed in each well of a 24-well Linbro plate with a sterile wide-mouthed
Pasteur pipette, and four drops of growth medium were added to each well.
This low volume of medium is important to the attachment process.  The
developing embryos were incubated at 25°C, and observed daily until they
were attached to the wells  (usually 2-3 days).  The growth medium consisted
of 10% fetal calf serum in L-15 medium plus 0.02% antibiotics and 0.01%
L-glutamine.  When the embryos became securely attached, 1 ml of growth

                                     10

-------
medium was added to each well.  Throughout the experimental -period, the
growth medium should be changed when  necessary, but no more often than once
every 2-3 days.


Primary Hepatocyte Cell Culture Technique

     Livers were excised from 2-4 large adult C_. variegatus and placed
in a Petri dish containing sterile HBSS plus 5% antibiotics at 25°C.  The
livers were transferred to a second Petri dish containing the same  solution
and a small gauge hypodermic needle was used to inject them with digestion
buffer.  The digestion buffer was the HBSS plus 0.05 % collagenase  (Sigma,
Type I).  The procedure was a modification of that of Berry and Friend  (10).
After 5 minutes, the livers were transferred to fresh digestion buffer  and
minced into 2-3 mm pieces with a razor blade.  The minced tissue was trans-
ferred to a digestion flask (Bellco)  containing fresh digestion buffer  and
gently swirled for 20 minutes.  Approximately 7 ml fractions were collected
in centrifuge tubes containing 3 ml of tissue culture medium precooled  to
4°C.  The tissue culture medium consisted of 100 ml L-15, 2.5 ml HEPES
 (Gibco), plus 15% fetal calf serum and 1% antibiotics at pH 7.4.  Following
centrifugation at 600 x g for 2-3 minutes, the resulting cellular pellet was
resuspended in fresh tissue culture medium at 4°C.  The suspended cells were
plated out into 25 cm2 T-flasks which had been coated with a collagen
substrate  (11).  A minimum of 500,000 cells per ml of medium was desired,
and about 3 ml of cell suspension was placed into each flask.  The  cultures
were maintained at 25°C, and the cells can be expected to become attached
to the substrate in about 24-48 hours.
 Immunological Studies

     Serum collection:  Blood was obtained initially by completely  severing
 the caudal fin, making a cut from the dorsal side to the ventral.   Modi-
 fications to the procedure included the combined use of an anticoagulant
 followed by an anaesthetic and by a modification of the cutting procedure.
 Fish were exposed to 1% sodium citrate  (Sigma) in water for  10 minutes.
 They were then exposed to 0.001% tricaine methane sulfonate  (MS222)  (Sigma)
 in water for 5 minutes.  A cut was made  (using a scalpel blade) on  the
 ventral side of the caudal peduncle, but the caudal fin was  not removed
 entirely.  Rather the cut was terminated at the backbone, so that the
 caudal artery and vein were severed, but not the spinal cord.  Blood was
 collected in heparinized capillary tubes, then centrifuged, and the  serum
 separated and frozen at -20°C pending use.

     Collection of leucocytes:  Two techniques were used to  separate and
 collect leucocytes from whole blood of c:. variegatus.  In the simpler
technique,  blood was collected in heparinized capillary tubes, then cen-
 trifuged to sediment the cells.  The tubes were then broken  just above
the cells and the buffy coat (containing the leucocytes) removed" with a
capillary syringe.  If desired, collected leucocytes from several fish were
pooled in a capillary tube and again centrifuged to reduce contamination with
erythrocytes.   In the second technique, 20 yl of collected blood was layered

                                     11

-------
onto 20 yl of Ficoll-Paque solution in a capillary tube and centrifuged at
400 X g for 30 minutes.  Erythrocytes were sedimented at the bottom of the
Ficoll-Paque layer, whereas leucocytes formed a loose layer at the inter-
face between the serum and Ficoll-Paque layer.  Cells collected by either
method were placed on microscope slides, stained by Wright's stain, and
examined to determine cell types present.  Cell viability was determined by
exposing cell suspensions to trypan blue and observing dye exclusion by
viable cells.

     Serum electrophoresis:  Serum was collected from fish exposed to
1 ppm BEN for seven weeks and from normal control fish.  Serum was
electrophoresed, using the Corning Agarose Electrophoresis system which
requires a total serum volume per fish of 0.6 yl.  Electrophoresis was
carried out for 30 minutes at a pH of 8.6, using barbital buffer.  Bands
were stained using amido-black 10B for 15 minutes and scanned on a Corning
Densitometer at 500 nm.

     Immune rosette formation:  Human type 0, Rh+ cells were used as the
immunogen.  Cells were collected, washed three times in phosphate-buffered
saline  (PBS), pH 7.2 and adjusted to a 20% suspension according to the
method of Ingram and Alexander  (12).  Fish were injected intraperitoneally
with 0.1 ml of the cell suspension.  Control fish received only PBS.  Fish
spleens were collected after 30 days and disrupted, using a conical tube cell
homogenizer with a loosely fitting pestle to minimize cell rupture.  Cells
from the resulting pulp were resuspended in tissue culture medium  (L-15,
Gibco) and passed through fiber mesh to remove large masses.  The resulting
suspension, containing predominantly lymphocytes, was collected in a 0.4 ml
microcentrifuge tube.  Lymphocytes were gently sedimented by centrifugation
at 10 X g  and washed three times with L-15 medium.  After the last wash, the
cells were resuspended in 0.2 ml of a 5% suspension of human type 0, Kh+ cells
This mixture was centrifuged for three minutes at 10 X g and allowed to
incubate for 1 hour at room temperature or overnight at 4°C.  The cell pellet
was then gently resuspended in PBS and a drop added to a microscope slide.
The smear was allowed to dry, and examined microscopically for the presence
of rosette-forming cells, then stained with Wright's stain and re-examined.

     Bacteriophage neutralization assay:  The semi-micro bacteriophage
neutralization assay of O'Neill  (13) was employed.  The neutralization titer
of fish sera is defined as the amount of serum required to produce 50%
inactivation of the bacteriophage  (SD50)  (14).  Two-fold serial dilutions
of C. variegatus serum were prepared, using 25 yl amounts per dilution in
Falcon microtest wells.  MS2 phage, obtained from American Type Culture
Collection and grown by the soft agar overlay method of Eisenstark  (15),
was diluted in R medium for phage lysates  (Bacto tryptone, 1.0%; Bacto yeast
extract, 0.1%; NaCl, 0.8%; CaCl2, 0.02%; glucose, 4.0%) to contain
approximately 150 plaque-forming units  (PFU)/25 yl.  Equal volumes of
dilutions of serum (or non-serum controls) and phage suspensions were mixed
and incubated at 25°C for 30 minutes.  The dilutions were then added to
0.75 ml aliquots of R medium containing 0.8% agar  (Difco) maintained at 40°C.
Next, 0.05 ml of an overnight culture of E_. coli K12 was added and the
mixture was then transferred to overlay an R medium agar  (1.5%) base in 60
mm plastic petri dishes (Falcon).  After an incubation of 6 hours at 37°C,

                                     12

-------
the numbers of PFU/plate were counted.


Effects £f Salinity, DOC, and Light o_n DENA Ex tract ability

     Extraction of DENA at trace levels from the various water matrices was
accomplished by Method 607 (16)  approved by the Environmental Protection
Agency.  The technique involves a simple extraction of a pre-cooled 25 ml
water sample with 2 x 20 ml volume of cold dichloromethane  (DCM) .  The DCM
extract was collected and rinsed with 10 ml of 10% HCl.  (All distilled,
deionized water  and HCl were previously extracted with DCM before use.)
Subsequent filtration through anhydrous NaaSOu removed residual water.  The
volume was then reduced in a Kuderna-Danish concentration apparatus until
approximately 5 ml of extract remained.   Further concentration was performed
in a Synder microcolumn apparatus to a volume of 0.5 ml.  Residual DCM was
removed by addition of hexane and repeated concentration to 0.5 ml.  The
concentrated hexane sample was immediately analyzed by gas chromatography
to minimize degradation effects during storage.  Triplicate analyses were
performed on all samples.

     Due to the inefficient recoveries at 1 mg/1 DEN by the EPA method, a
revised method for extraction of 1 mg/1 was developed.  The method consisted
of extraction of a 5 ml water sample with 3 x 10 ml volumes of hexane.  Both
the water samples and hexane used were kept at room temperature.  The hexane
extracts were combined and immediately analyzed by gas chromatography.
Triplicate analyses were performed on all samples.

     Gas chromatographic analysis was performed on a Perkin-Elmer Sigma-3B
Gas Chromatograph equipped with a   Ni electron capture detector interfaced
with a Sigma-lOB Analyzer.  The column used to separate DENA was the Supelco
10% SP-1000 on 100/120 Chromosorb W AW.   The injector, oven, and detector
temperatures were 200°C, 110°C, and 350°C, respectively.  The nitrogen gas
flow rate was 30 ml/minute for both the carrier and detector make-up gas.
An attenuation of 3 was used.

     The recovery efficiencies for both the EPA and revised methods in
distilled, deionized water were determined.  The water was spiked with DENA
at less then 100 yg/1 or 1 mg/1.  Since DENA is a liquid at room temperature,
the water samples were spiked using a 10 yl or 50 yl syringe.  The weight
delivered was calculated by using the density of DENA  (0.9422).  There was
difficulty in obtaining 2.5 yl consistently under red light  (and minimize
personal exposure at the same time) for trace level spiking, so actual
concentrations of DENA ranged from 55-85 yg/1.

     Immediately after addition of DENA, each extraction volume was collected
and taken through the corresponding extraction and clean-up procedures.
Subsequently, the individual solvent extracts were analyzed by gas chromato-
graphy.  Percentage recoveries of DENA partitioned in the individual extracts
were calculated and compared.

     The effect of salinity on DENA extractability was obtained by using
various dilutions (S parts per thousand = 0.5-24) of I.A.P.S.O. Standard

                                     13

-------
seawater of known chlorinity, spiked with DENA at 1 mg/1 or less than
100 yl/1.  Immediate extraction by the appropriate method and gas chromato-
graphic analysis were carried out as described above.  Percentage recoveries
of DENA at different salinities were then compared to recovery from distilled,
deionized water.

     To determine the possible effects of DOC (dissolved organic carbon) on
DENA extractability, serial dilutions  (1-22 mg/1) of humic acid  (Aldrich
Chemical Company) were made in standard seawater with a salinity of 10 parts
per thousand.   [Humic acid is 50% elemental carbon (17),  thus DOC was
0.5-11 mg/1.]   Each water sample was spiked with DENA at trace levels and
1 mg/1 followed by immediate extraction and gas chromatographic analysis as
before.  The percentage recoveries of DENA from these water matrices were
compared to recovery from standard seawater.

     The loss of DENA due to photodegradation was monitored in brackish water
(DOC = 6 mg/1, S parts per thousand =  10).  Water samples spiked with DENA
at 1 mg/1 were irradiated by two fluorescent lights  (General Electric, F04CW,
Cool White, 40 watts each) at room temperature.   The photoperiod was 12 hr/day
for one week.  In addition, blanks and control samples constantly kept in the
dark were monitored to determine the direct effect of light and DOC on DENA
loss.  All experimental conditions were reproduced in triplicate, except for
the blanks where only one sample was monitored.   Each water sample was
contained in a screw cap Erlenmeyer flask to eliminate water loss and for
safety reasons.  To reduce liquid waste, the total volume of each water sample
was kept to a minimum so that little or no sample was left after the one-week
period of extractions.  Extraction of the water samples and immediate gas
chromatographic analysis were performed every other day,  as described above.

     An attempt was made to extrapolate the laboratory data to the natural
environment at trace levels of DENA and to chronic fish exposure experiments
where high concentrations of DENA were used.  A surface water sarple was
taken from a local stream, Crane Creek, near its entrance to the Indian
River  (Melbourne, Florida).  The entire water sample was suction-filtered
through a 0.45 ym Millipore glass fiber filter to remove particulates prior
to the addition of trace levels of DENA and subsequent extraction.  Salinity
was determined by optical refractometry.  The percentage recovery of DENA
was compared to previous laboratory data.

     A control sample from a chronic exposure study was taken from an
aquarium in which a population of C_. variegatus had been maintained for two
months.  This artificial saltwater sample was a mixture of dechlorinated
tapwater and Rila salts.  No DENA had  been added to this water at any time.
The aliquots of the water sample were  spiked to 1 mg DENA/1, extracted, and
analyzed by gas chromatography.  There was no filtration prior to DENA
addition as performed on the above creek water sample.  Both the experimental
samples and control water blanks were  carried out in triplicate.  DENA
recovery was compared to the recovery  of DENA in standard seawater.
                                     14

-------
                                  SECTION 4

                           RESULTS AND DISCUSSION
Toxicity Tests
     Prior to the conduction of long-term exposures in water contaminated
chronically with BEN, the LC-50 (96 hrs)  was determined for BEN with respect to
C_. variegatus.  Based on these tests, the LC-50 for BEN was determined
to be ca. 60 ppm  (Table 1).  The data are the cumulative results of four
experiments, and the numbers represent cumulative mortality at the times
indicated.

TABLE 1.  The toxicity of BEN with respect to C_.  variegatus.
          6HR

102 PPM

 64 PPM

 40 PPM

 25 PPM

 16 PPM

CONTROL

Benzidine Exposures, C. variegatus

     Twelve  experiments were conducted in which approximately  400
C_.  variegatus were maintained in water contaminated weekly with  BEN
at  1 ppm.  Five of these experiments were terminated  after brief periods
of  time because of mechanical failures, disease, or other problems
believed to  be unrelated to the BEN contamination.  In  the remaining
experiments, exposures were conducted for periods  ranging
from 8 to 11 months.  A relatively high mortality occurred in  some of  these
experiments, such that the total number of  individuals  surviving to
the termination of the experiment was somewhat reduced.

     In one  of the more successful experiments, in which the fish were
exposed for  approximately eleven months, four individuals developed
histopathologically similar lesions within  a one-month  period.

12HR
4






24HR 48HR 72HR
7 27 36
1 8 13
2
1

1
g EH <
W H EH
96HR
40
21
7
3
6
2
EH O
O PJ
EH O
40
40
40
40
40
40
1^ M-)
EH U
OJ QJ
0 W
S ft
100
53
18
8
15
5
                                     15

-------
Nine individuals were sacrificed and examined histologically during the
first 22 weeks of the exposure and no unusual lesions were observed.  However
at 25 weeks, one individual became moribund, was sacrificed, and a liver
lesion was discovered.  During the next 4 weeks of this experiment, 3 of the
remaining 20 fish developed what appeared to be the same lesion.  The
experiment was eventually terminated after 18 more weeks when only one
individual remained.  Thirteen fish died during this period and were so
deteriorated that they could not be observed histologically.  Four
individuals were sacrificed and observed histologically, but none had
developed the lesion.

     The livers from the fish with the lesion contained large regions in
which the typical parenchyma had been displaced by regions of randomly
arranged collagenous tissue within which there were numerous zones that
contained a proliferation of different sized and types of tubules
(Figures 4 through 7).  The time of occurrence of these lesions suggested
a 5 to 6 month latency period.  We are in the process of repeating these
exposures and have already reproduced the lesion in a considerable number
of BEN-exposed fish.

     Three experiments were attempted in which £. variegatus were exposed
to weekly contaminations at 10 ppm; however, the fish did not survive
beyond 24 to 30 days.  Thus, this level of exposure, although significantly
less than the LC-50  for BEN with respect to £. variegatus, seems to produce
a toxic effect.  Even a 5 ppm chronic BEN exposure seemed to be toxic since
in two experiments in which the weekly contamination was reduced to 5 ppm,
most of the fish did not survive beyond two months.

     Although the basic chemical parameters normally used to indicate water
quality remained within acceptable ranges during our chronic exposures, our
concern about the mortality occurring during the exposures caused us to
install bottom filters  (oyster shell) in most of the exposure systems within
the last few months.  This change seems to have brought about a significant
reduction in mortality.  In fact, in 4 experiments involving over 75 fish
in which these bottom filters have been installed, no deaths have occurred
in over 4 months.
Pulse Benzidine Exposures, C. variegatus

     A pulse  (one time) exposure of 27 £. variegatus to 50 ppm BEN resulted
in 70% mortality in 5 days.  A 40 ppm concentration of exposure seemed less
toxic, since  the exposure of 30 £. variegatus at this concentration did not
begin to cause deaths until about ten days.  However, the deaths continued
until 50% had died by 60 days.  Six of the remaining individuals survived the
exposure and were alive and appeared to be in good health over one year after
the exposure  date.  Similar results were obtained when £. variegatus were
pulse-exposed at 25 ppm and 11.5 ppm.  In each of these experiments, over 50%
of the fish had died by 4 months, while some individuals were maintained for
7 to 12 months with no apparent ill effects.
                                     16

-------
Figure 4.  Typical morphology of the livers of non-exposed control fish.
            Mag. 180X.

Figure 5.  A region of transition from normal liver tissue to the
            BEN-induced fibrous lesions.   Arrows outline an extension
            of normal tissue down into the area of the lesion.
            Mag. 180X.

Figure 6.  High magnification of the lesion illustrating its fibrous
            nature with numerous tubular  profiles.  Some of these
            profiles have a duct-like morphology  (d)  and others are
            more vascular (v)  in appearance.   Mag. 660X.

Figure 7.  A higher magnification of an area from Figure 6 demon-
            strating the characteristic random arrangement of
            fibroblasts and collagen fibers.   Note the duct cell
            morphology of the cells of the duct-like (d) profiles
            and the endothelial morphology of cells forming the
            vascular type profiles (v) .   Mag. 1056X.
                                   17

-------

              *i^ i " '^ '- •'^' vS^*




              ^;*&^-v-5s^s^
- 6

-------
Diethylnitrosamine Exposures

     Weekly contaminations at 100 ppm DENA proved toxic to l_. punctatus_/
and all 15 of the fish in the experiment were dead by 14 weeks.  Exposure
of C. variegatus at 20 ppm DENA was quite toxic/ and all 21 fish were dead
by 8 days.  Two experiments were then conducted in which £. variegatus were
exposed to 10 ppm DENA.  In one of these experiments involving 15 individuals,
only 13% were alive at 40 days and similarly in the other experiment involving
75 £. variegatus, only ca. 15% were alive at 60 days.  When the concentration
of exposure was reduced to 1.2 ppm in an exposure of 75 £. variegatus, 50%
were dead by 4 months and the remaining individuals had died by 6 months.
Thus, DENA at these concentrations was toxic to both of these species.  Because
of this factor,- further attempts to conduct long-term exposure experiments
with DENA were terminated.
Benzo(a)pyrene Exposures

     These experiments were conducted early in the project, according to
methods previously published  (4).  In one experiment, 65 £. variegatus were
maintained in a system that was contaminated weekly with BaP at a concen-
tration of 50 ppb.  The fish died rather continuously throughout the course
of the experiment due to unknown causes, and when the experiment was
terminated after 8 months, only 7 individuals remained.  No tumors or other
lesions were observed when these specimens were necropsied.  In another
experiment, 20 _!. punctatus were exposed to weekly contaminations of BaP
at 1 ppb.  After one year, this experiment was terminated and all 20
specimens were necropsied and no lesions were observed.  While these
experiments were in progress, it became obvious that, because of the low
solubility of BaP in sea water and its tendency to absorb to solid surfaces  (18) ,
it was not possible to conduct valid experiments in this manner.  No further
attempts were made to expose fish to BaP.


Benzidine Feeding Experiments

     In three experiments involving ca. 125 £. variegatus, no lesions were
observed that could be associated with BEN contamination of the food.  The
mortality seemed to be higher in these experiments than in control fish or
fish maintained in BEN-contaminated water.  Thus, BEN contamination of the
food seemed to provide a stress factor.  It is interesting in this respect
that in BEN feeding experiments which have been in progress for over 3 months
using the systems with oyster shell bottom filters, no deaths have occurred.
Therefore, it seems likely that the bottom filters may provide a less stress-
ful environment in which this type of experiment could be conducted with
increased probability of success.

     Three experiments-involving a total of 105 fish were conducted in which
!_. punctatus were fed BEN-contaminated food.  One of these experiments invol-;
ving 30 fish was terminated at 12 days when a mechanical malfunction resulted
in the death of the fish due to oxygen deficiency.  In the two remaining
experiments, the fish began to die in significant numbers during

                                     19

-------
the first month, and none of them survived beyond the second month.  It was
concluded that this feeding regime was toxic to I_. punctatus( and no further
experiments of this type were attempted with the species.


Diethylnitrosamine Feeding Experiments

     In an experiment in which 20 C_. variegatus were fed DENA-contaminated
food,  50% died within 1 month, and only 25% were alive at 6 months when the
experiment was terminated.  However, when I_. punctatus were fed this same
regime, the fish exhibited no ill effects and remained in a good state of
health for one year.


Benzidine Injection Experiments

     In an experiment in which 8 I_. punctatus were injected with 2.5% BEN,
initially none of the fish displayed any ill effects from the injection.
They continued to feed and looked well for about two weeks; however, at this
point feeding stopped and on day 16, three individuals died and the remaining
fish were moribund at the bottom of the aquarium.

     In a similar experiment in which 7 C_. variegatus were injected with
2.5% BEN, 6 individuals died within 24 hours.  When the concentration of the
injection fluid was reduced to 1% BEN and 15 C_. variegatus were injected,
3 individuals were dead at 48 hours, and 4 more died at 10 days.  However,
in this experiment the surviving fish remained healthy for 12 months, at
which time the experiment was terminated.  Necropsies of these fish
disclosed no lesions.
Diethylnitrosamine Injection Experiments

     When 50 I. punctatus were injected with DENA, 18 died within 24 hours
and all were dead by 48 hours postinjection.  In a subsequent experiment,
the concentration of DENA was reduced to 10% DENA in distilled water and
10 !_. punctatus were injected.  In this experiment, the first deaths
occurred in 2 days and continued to occur until all were dead by 60 days.
Thus, DENA appeared to be toxic to I_. punctatus, even at a concentration of
10%.  An experiment was attempted in which  85 I_. punctatus were injected with
1% DENA.  At this concentration, no toxic effects were observed and there
were no deaths for one month; however, at this point, all the fish were
killed due to accidental chlorine poisioning.

     Injection of C_. variegatus with 20% DENA proved to be toxic, resulting
in the death of all 18 individuals within 24 hours.  Ten percent DENA produced
essentially the same effect.  Five percent  DENA was less toxic, but still
resulted in 50% mortality within three months.  Attempts to inject C_.
variegatus with 1% DENA produced essentially the same effect.
                                     20

-------
Benzidine Exposures,  C. variegatus Embryos

     Exposures of early C_. variegatus embryos to BEN conducted at concen-
tractions ranging 5 - 500 ppm produced abnormalities only at BEN concen-
trations of 50 ppm and higher.  Experiments are in progress to establish the
effective dose (ED-50), and initial results suggest that this value is  likely
to be ca. 35 ppm.  During this project, about 600 C_. variegatus early embryos
have been exposed to BEN at 50 ppm or higher.  Only about 13% of these  embryos
developed normally beyond the hatching stage.  Thus, about 85% of the embryos
exhibited some type of developmental anomaly, and only 13% of the 600 were
capable of development beyond the hatching stage.

     When these data are compared to the results of the control experiments
in which 70% of the 300 embryos observed developed normally beyond the
hatching stage, the detrimental effects of exposure to BEN at these concen-
trations are quite obvious.

     Anomalies that were observed in embryos exposed to BEN at 50 ppm
or higher were: 1) Tubed heart syndrome with distended pericardia,
2) Poor circulation, 3) Sparse distribution of melanophores around yolk,
4) Inability to hatch, 5) Abnormal head morphology, 6) Scoliosis, and
7) Faint RBC pigmentation.

     Anomalies 1 through 4 occurred most frequently and anomalies 2, 3,
and 4 occurred only if embryos were exposed after somite and lense
development.  Interestingly, some of the exposed embryos survived for up to
30 days, and yet, did not hatch.  Histologic examination of the anomalous
embryos is in progress, and scanning electron microscopic examination of
the surface of the chorion is being attempted.  Efforts are also in progress
to quantitate the penetration of the chorion by BEN and determine its
distribution in the developing embryonic tissues by use of radioactively
labeled BEN and autoradiographic techniques.
SHF-1 Cell Culture Exposures

     Benzo(a)pyrene, benzidine, and diethylnitrosamine were all  found  to be
acutely toxic to SHF-1 cells.  BaP was acutely toxic at 2.0 yg per ml  of
growth medium (7.9 x 10~6M) , BEN at 0.1-0.2 mg per ml  (4 x lO"1*  to 7.9 x
10~'+M) , and DENA at 2.0 mg per ml  (7.9 x 10"3M) .  In cells exposed to  subacute
concentrations, toxicity was evidenced by cell vacuolization  and a general
stress response of SHF-1 cells.  BaP, even at low concentrations, produced
this response.  SHF-1 cells were chronically exposed to subacute concen-
trations of each compound through as many as five subcultivations.  After
subcultivation, the time required for the cells to become confluent gradually
increased with each pass.  Foci of multilayered cells  (approximately 1 mm
in diameter)  exhibiting a lack of contact inhibition were observed in  several
cell cultures exposed to low concentrations of BaP and BEN.

     Exposure to BaP in the amounts of 100 ng per ml  (4 x 10"7M), 50 ng
per ml (2 x 10~7M), 40 ng per ml (1.6 x 10~7M), 30 ng per ml  (1.2 x 10"7M),
and 20 ng per ml (7.9 x 10~8M)  were carried through three passages  (28-30)

                                     21

-------
Benzidine Exposures, C. variegatus Embryos

     Exposures of early C_. variegatus embryos to BEN conducted at concen-
tractions ranging 5 - 500 ppm produced abnormalities only at BEN concen-
trations of 50 ppm and higher.  Experiments are in progress to establish the
effective dose (ED-50), and initial results suggest that this value is likely
to be ca. 35 ppm.  During this project, about 600 C_. variegatus early embryos
have been exposed to BEN at 50 ppm or higher.  Only about 15% of these embryos
developed normally beyond the hatching stage.  Thus, about 85% of the embryos
exhibited some type of developmental anomaly, and only 13% of the 600 were
capable of development beyond the hatching stage.

     When these data are compared to the results of the control experiments
in which 70% of the 300 embryos observed developed normally beyond the
hatching stage, the detrimental effects of exposure to BEN at these concen-
trations are quite obvious.

     Anomalies that were observed in embryos exposed to BEN at 50 ppm
or higher were: 1) Tubed heart syndrome with distended pericardia,
2) Poor circulation, 3) Sparse distribution of melanophores around yolk,
4) Inability to hatch, 5) Abnormal head morphology, 6) Scoliosis, and
7) Faint RBC pigmentation.

     Anomalies 1 through 4 occurred most frequently and anomalies 2, 3,
and 4 occurred only if embryos were exposed after somite and lense
development.  Interestingly, some of the exposed embryos survived for up to
30 days, and yet, did not hatch.  Histologic examination of the anomalous
embryos is in progress, and scanning electron microscopic examination of
the surface of the chorion is being attempted.  Efforts are also in progress
to quantitate the penetration of the chorion by BEN and determine its
distribution in the developing embryonic tissues by use of radioactively
labeled BEN and autoradiographic techniques.
SHF-1 Cell Culture Exposures

     Benzo(a)pyrene, benzidine, and diethylnitrosamine were all found to be
acutely toxic to SHF-1 cells.  BaP was acutely toxic at 2.0 yg per ml of
growth medium (7.9 x 10~6M) , BEN at 0.1-0.2 mg per ml  (4 x 10""* to 7.9 x
ICT^M), and DENA at 2.0 mg per ml  (7.9 x 10"3M).  In cells exposed to subacute
concentrations, toxicity was evidenced by cell vacuolization and a general
stress response of SHF-1 cells.  BaP, even at low concentrations, produced
this response.  SHF-1 cells were chronically exposed to subacute concen-
trations of each compound through as many as five subcultivations.  After
subcultivation, the time required for the cells to become confluent gradually
increased with each pass.  Foci of multilayered cells  (approximately 1 mm
in diameter) exhibiting a lack of contact inhibition were observed in several
cell cultures exposed to low concentrations of BaP and BEN.

     Exposure to BaP in the amounts of 100_ng per ml (4 x 10~7M), 50 ng
per ml (2 x 10~7M), 40 ng per ml (1.6 x 10 7M), 30 ng per ml (1.2 x 10~7M),
and 20 ng per ml  (7.9 x 10~8M) were carried through three passages  (28-30)

                                     21

-------
 TABLE  2.   Exposure of SHF-1 cells to benzo(a)pyrene.
REPEATED EXPOSURE
                                  EXPOSURE FOR P-28 ONLY
p-28 P-29 P-30

100 ng2


50 ng
.-

40 ng


30 ng


20 ng

100 ng*1 —
100 ng*
100 ng*
50 ng
50 ng
50 ng
40 ng*
40 ng*
40 ng*
30 ng —
30 ng
100 ng*
100 ng*
100 ng*
100 ng*
100 ng*
100 ng*
100 ng*
100 ng*
100 ng*
50 ng*
50 ng*
50 ng*
50 ng*
50 ng*
50 ng*
50 ng*
50 ng*
' 50 ng*
40 ng*
40 ng*
40 ng*
40 ng*
40 ng*
40 ng*
40 ng*
40 ng*
40 ng*
30 ng
30 ng
30 ng
30 ng
30 ng
30 ng
130 ng
30 ng
30 ng
20 ng
20 ng
20 ng
20 ng*
20 ng*
20 ng*
20 ng*
20 ng*
20 ng*
20 ng*
20 ng*
20 ng*
                                  P-28
                                              P-29
                                                       P-30
                                  100 ng
                                              0 ng
                                              0 ng
                                              0 ng
         0 ng*
         0 ng*
         0 ng*
         0 ng*
         0 ng*
         0 ng*
        I 0 ng*
         0 ng*
        I 0 ng*
                                   50 ng
                                              0 ng
                                              0 ng
                                              0 ng
         0 ng*
         0 ng*
         0 ng*.

         0 ng*
         0 ng*
         0 ng*

         0 ng*
         0 ng*
         0 ng*
                                   40 ng
                                              0 ng*  —
0 ng*  —
                                              0 ng*  —
0 ng
0 ng
0 ng
0 ng
0 ng
0 ng-
0 ng
0 ng
0 ng
                                    0 ng
                                              0 ng
0 ng
                                              0 ng
0 ng
0 ng
0 ng

0 ng
o ng
0 ng

0 ng
0 ng
0 ng
                                    0 ng
                                              0 ng
0 ng
                                              0 ng
0 ng*
0 ng*
0 ng*
0 ng*
0 ng*
0 ng*

0 ng*
0 ng*
0 ng*
 1*  Indicates  foci of multilayered cells  appeared in
     the  culture.

 2   Concentrations given are ng  per ml  of growth
     medium in  tissue culture flasks.
                        22

-------
(Table 2).  During passage 28, the cells appeared unaffected and were
subcultured within the normal seven day period.  In passage 28, exposure
was continued in some cultures and discontinued in others.  During passage 29,
multilayered foci appeared in a few cultures: those exposed to 100 rig BaP per
ml in both passages 28 and 29, those exposed to 40 r\g BaP per ml in both
passages 28 and 29, and those exposed to 40 r|g per ml in passage 28 only.
Passage 29 reached confluency in the normal amount of time  (7 days).  In
passage 30, multilayered foci appeared in essentially all cultures except
those exposed to 40 ng BaP per ml and 20 f|g BaP per ml in passage 28 only.
All of the experimental cultures of passage 30 failed to reach confluency.

     Cultures at passage 30 were exposed to 0.6 yg BEN per ml  (2.4 x 10~6M),
0.5 yg BEN per ml  (2 x 10~6M), 0.4 yg BEN per ml (1.6 x 10~6M), 0.3 yg BEN
per ml (1.2 x 10~6M), 0.2 yg BEN per ml  (7.9 x 10~7M), and 0.1 yg BEN per ml
(4 x 10~7M).  These cells were subcultured through passages 31 and 32 , using
the same scheme as outlined for BaP exposures in Table 2.  Multilayered
foci appeared only in the cultures of passage 32 which had been exposed to
0.4 yg BEN per ml in all three passages  (30, 31, and 32).

     Foci of multilayered cells have not been observed in any DENA
exposures, and have not appeared with BaP or BEN exposures after passage 33.

     In any cellular assay system, certain criteria or events indicate
a "transformation" of cells to a neoplastic state.  Transformation is
defined in terms of hereditary morphological changes, accelerated growth
rate, abnormal karyotypic shifts, and tumor promotion in vivo  (19, 20).
Thus far,  not all  of the above  listed criteria for transformation have
been observed in SHF-1 cells exposed to benzo(a)pyrene, benzidine, and
diethylnitrosamine.  However, the occurrence of multilayered cell foci
in cultures with an initial exposure to BaP, but no exposure in two sub-
sequent passages,  suggests a  mutagenic effect.  The explanation for the
appearance of multilayered foci in cultures before passage 33, but not
after, is not apparent at present.  It is possible that the
sensitivity of the cells may undergo variations throughout the life of
the cell line.
Early Embryogenesis, C. variegatus

     A search of the literature revealed that knowledge of the early
embryogenesis of C_. variegatus did not exist in the cellular detail
needed as baseline  information for our planned studies concerning
teratogenesis.  Consequently, a study of early embryonic development
of C. variegatus was conducted (21).

     Because of the nature of the chorion, it was necessary to develop
a technique for observation of excised blastoderms in order to observe
the details of cellular and subcellular activities that occur during
embryogenesis.
                                     23

-------
     The following are observations resulting from the use of this technique
that extend our knowledge of teleost embryology.

     Fertilization:   The mechanism or mechanisms that initiate the polar
concentration of protoplasm has been in question for many years (22).   In
the current study, this concentration of protoplasm was never observed in
an egg in which subsequent cellular cleavage did not occur.  Therefore, it
appears that the penetration of the micropyle by a spermatozoon is at least
partially responsible for initiating this phenomenon in C_. variegatus.

     Blastulation;  Wedge-like structures were observed on two adjacent
cells of the four-cell stage.  These wedges likely function in the rapid
elongation of the cells in one plane, causing the eight-cell stage to be
oval in shape.  The wedges are no longer apparent by the rounded
thirty-two-cell stage.

     Gastrulation;  Early in epiboly, superficial blastoderm cells move onto
the yolk, establish a leading edge which surrounds the yolk, and eventually
the entire yolk becomes covered by superficial ectodermal cells of the
blastoderm.  The exact source of these cells has been a long-standing
question in teleost embryology.  The technique of excision and inversion of
blastoderms used in this study made possible observations that provide
evidence that the ectodermal cells covering the yolk come from
the superficial blastoderm.  Observation of an inverted blastoderm reveals
a channel passing along the floor of the blastoderm providing a cellular
pathway along the blastoderm floor leading to its edge.  Our observations
indicate that somatic ectoderm, endoderm, and mesoderm cells likely pass
out of this channel and along the pathway mentioned.  This pathway leads
to the edge of the blastoderm where a divergence of its walls occur.
Observations made over a period of time clearly revealed individual
cells moving along this route to the edge of the blastoderm.
These gastrular activities strongly support the hypothesis of Ballard  (23),
i.e., an outward movement of deep internal blastodermal cells.

     Scanning electron microscope observation of intact eggs at this stage
reveals numerous pores on the floor of the subgerminal cavity.  These pores
seem to lead to the yolk proper and may function in the conduction of
nutrients to the blastoderm.

     Attempts were made to study the earlier stages of neurulation with
this procedure,- however, pigmentation and growth begins to obscure the
fine details of development from this point to hatching, and it appears
that it will be necessary to prepare serial histological sections to
study the details of histogenesis and organogenesis of C_. variegatus.


Gross and Histological Anatomy of_ the Post-Pharynageal Digestive Tract,
C. variegatus

     This study was accomplished because a review of the literature indicated
that a histological study of the structure of the normal digestive tract of
C_. variegatus adequate for our needs had not been accomplished.  Although

                                     24

-------
the histology of the digestive tract of some species of  cyprinodontid  fishes
had been studied, it was felt that a thorough study of the digestive tract of
this particular species was needed as baseline data, since C_. variegatus was
being employed extensively in our long-term exposures  (24).

Gross Anatomy:

     The digestive tract of C_. variegatus is similar to other cyprinodontid
fishes grossly, being composed of an esophagus, intestine and rectum.  The
RLG (relative length of the gut) of this species is 2.8, which classifies
it as an onmivore; according to  the method of Al-Hussaini  (25).

Histological Anatomy:

     Esophagus — The esophagus  can be divided into three distinct histological
regions; the anterior, middle, and posterior regions.  All of the regions
have a mucosa, submucosa, muscularis, and serosa.

     The esophageal mucosa is composed of an epithelium, a basement
membrane, and a stratum granulosum.  The mucosa is thrown into prominent
longitudinal folds which run the length of the esophagus.  In the anterior
esophagus, the folds are broad and flattened at their apices.  The cross-
sectional area of the lumen in the anterior esophagus is fairly small.   In
the middle esophagus, the mucosal folds are continuations of those folds of
the anterior esophagus.  The area of the lumen is slightly greater in this
region.  Posteriorly, the esophageal lumen reaches its maximum diameter as
the longitudinal folds become thinner and the thickness of the mucosa
decreases.  The longitudinal arrangement of esophageal mucosal folds allows
for optimum distensibility when  large food items are encountered.

     The mucosal folds of teleostean esophagi are typically covered by a
stratified epithelium and mucus-secreting cells.  The stratified cells are
polyhedrally shaped and are 10-12 cells deep at the apices of the folds.
The width of the mucosa thins to 2-3 cells at the bases of the folds.
The epithelial cells measure 10  ym across.  Each nucleus contains at least
one darkly staining nucleolus and a thin chromatin network when stained with
hematoxylin and eosin.  Saccular mucus-secreting cells are present in all
three regions of the esophagus.  Most are on the luminal boundaries of the
folds, but some are seen deeper  in the mucosal lining.  Mucous cells are
present in the anterior esophagus, but are more numerous in the middle and
posterior regions.  The mucous cells measure 10 urn deep by 9 urn wide and are
more numerous on the sides and at the bases of the folds.  The apices of the
mucous cells are covered by a single layer of squamous cells.  In the
posterior region of the esophagus, the stratified polyhedral cells are
gradually replaced by simple columnar epithelium.  Saccular mucus-secreting
cells are still present and are  numerous.  The columnar cells that have
replaced the polyhedral cells are slightly shorter than those lining the
intestine, but are otherwise identical.  They begin replacing the polyhedral
cells on the sides of the folds, covering increasingly more of the mucosal
folds posteriorly-
                                      25

-------
    A stratum granulosum is present in the esophagus.  It is. composed of a
more or less, continuous layer of granular cells that  lies  adjacent to  the
basement membrane.  The stratum granulosum is seen as.a band of darkly
staining cells at the base of the epithelium.

     The tunica (lamina) propria is a layer of fine connective tissue fibers
that lies beneath the basement membrane of the mucosa and extends into the
cores of the mucosal folds.  A true tunica propria is absent from the
esophagus of C_. variegatus.  The cores of the folds are formed by striated
muscle.  As there is no muscularis present in fish (26), the boundary between
the tunica propria and the submucosa is indistinct.

     The most obvious component of the esophageal submucosa is the irregularly
arranged striated muscle that forms the cores of the mucosal folds.  Copious
amounts of adipose tissue are present in this area in the anterior esophagus.
No blood vessels can be observed in the submucosa, although occasional- red
blood cells are seen.

     The muscularis of the esophagus is composed solely of an unusually thick
layer of circularly arranged striated muscle.  Longitudinally arranged muscle
underlying this circular layer is considered a part of the submucosa.  At the
junction of the esophagus  and intestine, the muscularis triples in thickness,
while internal muscle fibers are no longer seen.  Granulocytes are commonly
seen among the fibers of the muscularis.

     The anterior esophagus is found in the cephalic region of C_. variegatus
and is bound to adjacent connective tissue by fibrous connective tissue.
The serosa is continuous with the parietal peritoneum and is applied to
the esophagus as it enters the visceral cavity.  The serosa is composed
of a layer of simple squamous cells that are darkly pigmented.  Pigmen-
tation disappears at the junction of the esophagus and intestine.  Very
little subserosal connective tissue is present.

     Eosinophilic granular cells are numerous in the esophagus, especially
in the area of the mucosa.  At times, the granulocytes appear to have passed
through the epithelium  into the esophageal lumen.  The granular cells form
a stratum granulosum 2-3 cells thick adjacent to the basement membrane.
Granulocytes measure 3  ym  in diameter and contain obvious eosinophilic
granules.  Similar granulocytes may be observed throughout  the digestive
system and its accessory organs.

     Intestine and Rectum  — The anatomy of the remainder of the digestive
tract of C_. variegatus, the intestinal swelling, intestine proper, and
rectum, is basically the same histologically.  Each  region  is composed  of
four layers: mucosa, submucosa, muscularis, and serosa.

     The intestinal mucosa is composed of an epithelium of  columnar absorp-
tive cells, a basement membrane, a stratum granulosum, and  a tunica propria.
Neither a stratum compactum nor a muscularis mucosae is present.   Secondary
folding of the mucosal  folds increases the absorptive  surface area.   In the
intestinal swelling, mucosal folds are long and narrow.  Secondary folds  are
obvious along the sides of the primary folds.  Folds are broader and  shorter

                                     26

-------
in the intestine proper, and near the rectum, the intestinal mucosal folds
shorten and secondary folding is reduced.  The columnar absorptive cells of
the intestinal swelling and intestine proper have essentially the same
structure and size, measuring 24 ym by 4 ym.  Columnar cell nuclei average
6 ym by 4 ym.  One to two (rarely three) nucleoli are present.  Hematoxylin
and eosin or Mallory's connective tissue stain reveals fine strands of
chromatin in the columnar cell nuclei.  Nuclei near the bases of the folds
are sometimes compressed to 8 ym by 2 ym.  Nuclei at the tips of the mucosal
folds are mainly spherical and 4-5 ym in diameter.

     The second type of cell common to the intestinal epithelium of C_.
variegatus is a mucus-secreting goblet-type cell.  Goblet cells are generally
most numerous at the bases of the mucosal folds, but do occur on the sides and
at the tips.  The "goblets" stain darkly with PAS.  Also, PAS positive mucus
may be seen in the lumen adjacent to the goblet cells.  The dimensions of
the "goblet" portion of the goblet cells average 12 ym by 8 ym.  The nuclei
of goblet cells are seen near the basement membrane below each "goblet".
Often the enlarged part of a goblet cell is located deep in the epithelium and
connects to the lumen by a duct 10-15 ym long.

     The basement membrane is conspicuous when stained with PAS or Mallory's
connective tissue stain.  It appears as an unbroken line forming the basal
border of the columnar absorptive cells.

     A stratum granulosum is continuous throughout the intestine and rectum,
running roughly parallel to the basement membrane in these regions.

     A distinct tunica propria is seen in the intestine of C_. variegatus,
but is not clearly separate from the submucosa as no muscularis mucosae is
present.  The tunica propria is composed of very fine fibrous connective
tissue that forms the cores of the mucosal folds and extends down into the
submucosa.  The connective tissue stains intensely with PAS.  The tunica
propria is well vascularized and contains fibroblasts, lymphocytes, red
blood cells, and granulocytes.

     The intestinal submucosa of teleosts is composed of loose areolar
connective tissue that extends from the mucosa to the inner layer of
the muscularis.  In C_. variegatus, collagen is the most conspicuous
type of connective tissue, staining with both hematoxylin and eosin and
PAS.  Fibroblasts are the most prominent cell type present.  Other cell
types present include lymphocytes, granulocytes, and red blood cells.
Blood vessels are usually seen in the triangles of submucosa formed when
the basal portions of the mucosal folds approach the muscularis.

     The intestinal muscularis is generally composed of two layers, an
inner circular layer and an outer longitudinal layer.  The muscle bundles
are composed of smooth muscle , except at the  junction of the esophagus  and
the intestine.  In C_. variegatus, the muscularis at the intersection of
the esophagus and the intestine is composed exclusively of striated muscle
arranged in a circular pattern.  An external layer of longitudinally
arranged striated muscle extends posteriorly from the junction, over-
lying the small amounts of circularly arranged striated muscle that

                                     27

-------
persists into the intestine.  About 300 ym back from the junction an
internal circular layer appears.  Smooth muscle fibers gradually replace
the striated muscle in the intestinal swelling.  Isolated bundles of
striated muscle are seen as far back as the junction between the intestine
and the rectum.  The muscularis is about 30 ym thick in the intestinal
swelling and continues at that thickness posterior to the rectum.  The
layers of the muscularis are equally thick, about 15 ym.

     The serosa of the intestine is continuous with that of the esophagus.
It is composed of a single layer of squamous cells extending to the recto-
anal junction.  A layer of subserosal connective tissue (loose areolar) is
present in the intestinal swelling but thins posteriorly.

     Granular cells (granulocytes) are a common feature of teleostean
alimentary canals.  While they are present in the esophagus, they are
more obvious in other regions of the digestive tract, including the
accessory organs.  Granular cells form a stratum granulosum in £. variegatus
that is near, and parallel to, the basement membrane.  Granular cells are
often seen in the epithelium of the intestine, and some appear to have
passed into the lumen.  The majority of the granulocytes that are
present in the epithelium are confined to the infranuclear zone of the
columnar cells.  The granular cells measure about 4 ym in diameter and
contain several acidophilic granules.  Secretory and storage functions
(26) have been suggested for the granular cells, but no specific function
has yet been conclusively demonstrated.

     The rectum is a continuation of the intestine and is similar in
structure.  Histologically, several features distinguish
the rectum from the intestine.  The diameter of the digestive tube in
the rectal region is slightly reduced from that of the intestine; however,
the transition from the intestine to the rectum is gradual, and mucosal folds
shorten and widen during the transition.  The columnar cells of the rectum
are the same size as those of the intestine.  Rectal columnar cell nuclei are
slightly smaller, 6 ym by 3 ym.  Both the basement membrane and the stratum
granulosum are continuous with those of the intestine.  Goblet cells become
numerous in the anterior rectum and increase in number posteriorly, staining
intensely with PAS.  The most striking feature of the rectum is the presence
of PAS-positive granules in the sub-border region of the columnar cells
that compose a continuous band across the epithelial lining.  The granules
stain more intensely in the middle and posterior regions and are located
deep within in the columnar cells.  The rectal tunica propria is more
extensive than that of the intestine, forming the cores of short, wide mucosal
folds.

     The rectal submucosa is identical to that of the intestinal region,
although there is a general reduction in submucosal tissue in the
posterior portions of the rectum.

     The rectal muscularis in  C_. variegatus continues from the intestine
as a 30 ym thick layer.  Both the outer longitudinal layer and the inner
circular layer are composed of smooth muscle and are about 15 ym thick.
                                     28

-------
In the most posterior aspects of the rectum, the muscularis thins to
15-20 ym.

     The rectal serosa is continuous with that of the intestine, and still
composed of a layer of simple squamous cells.  Subserosal connective
tissue all but disappears in the posterior rectum.

     Granular cells are most numerous in the rectal region of C. variegatus,
forming the stratum granulosum that continues from the intestine.  Also,
many more granulocytes are seen in the rectal lumen than were present in
the intestinal lumen.  The morphology of the granular cells is identical
to  that of other regions of the digestive tract.

     Rodlet cells are more numerous in the rectum of C_. variegatus than in
the intestine.  They are  visible when stained with Mallory's connective
tissue stain, but are most obvious when stained with PAS and counter-
stained with fast green.  While most are located in the epithelial region,
some  are  seen  as deep  as  the  submucosa.  The rodlet cells
average 8-10 ym in diameter, and each cell contains several rodlets that
are 3-4 ym in diameter that stain intensely with PAS.   The number of
rodlets per cell ranges from 0 to 12.  The precise function of the
rodlet cells is yet to be determined; however, it has been suggested
that these bodies may be  sporocysts of a sporozoan parasite (27).
Peripheral Blood Cell Morphology, C_. variegatus

     This study was accomplished because of obvious importance of
hematological data in the diagnosis of diseases, and the fact that
data concerning teleost peripheral blood cell morphology is inadequate
in the literature  (28).

Light Microscopy:

     Erythrocytes — In fresh preparations, the definitive erythrocyte
is biconvex and ellipsoid with a centrally located oval nucleus.  The
cytoplasm is somewhat opaque when a minimal exposure to air has occurred.
Nuclei are visible in fresh preparations, but not prominent due to the
partial masking effect of the cytoplasmic hemoglobin.

     In Romanowsky-stained preparations, the homogeneous cytoplasm of
erythrocytes is eosinophilic and opaque, staining a pale brownish pink.
The staining properties of the cytoplasm are seen to be similar to those
of mammalian red blood cells stained in the same manner.  The nuclei of
Romanowsky-stained erythrocytes are basophilic.  The chromatin stains a
deep magenta and appears as a network of cross hatched patches against
the lighter nucleoplasm.  There is often a light concentration of
chromatin present at the periphery of the nucleus.

     In both fixed and living in vitro preparations, one can often
see erythrocytes which are teardrop-shaped due to a pointed pole at
one end of the cell and a rounded pole at the other.

                                     29

-------
    Erythrocytes that have been stained in fixed smears range from
8 x 6 ym to 11 x 8 ym, with a mean dimension of 10 x 6.3 ym.  In fresh
preparations, erythrocytes range from 10 x 6 ym, to 13 x 9 ym, with
average dimensions of 12 x 7 ym.

     The number of erythrocytes recorded in counts in females varies from
1.26 to 3.78 x 106/cu mm with a mean of 2.64 x 106/cu mm.  In males,
the counts ranged from 2.64 to 3.62 x 106/cu mm, with a mean of
3.12 x 106/cu mm.

     In fresh preparations, the erythrocytic cytoplasm often clears
when exposed to the air.  This results from clumping of the cytoplasm
into small globules which adhere to the nuclear and plasma membranes.
The nuclear material also appears to clump at times.  Usually nuclear
clumping involves larger globules than cytoplasmic clumping.

     In vitro preparations seem to indicate that the clearing and
clumping are related since these two phenomena can almost always be
observed to occur within a cell simultaneously.  Often after a period
of time in which clearing and clumping have occurred in a cell, globules
can be observed outside of the cell, with some of them adhering to the
exterior of the plasmalemma.

     Erythroid Cells — Several formed elements can be observed in the
peripheral hematocrit which apparently either give rise to, or are
derived from erythrocytes:  1) the erythroblast, an immature erythrocyte.
This cell is larger than the definitive erythrocyte and has a more
rounded nuclear and cellular shape. It is lightly basophilic and
becomes pale blue with Romanowsky stain.  2) the erythroplastid, an
anucleate, membrane-enclosed volume of cytoplasm derived from a definitive
erythrocyte.  Its elements possess the same eosinophilic, homogeneous
cytoplasm as the mature erythrocyte.  Erythroplastids are spherical,
ovoid or teardrop-shaped.  The diameters of these structures range from
3-5 ym.  Definitive erythrocytes with pointed pseudopodia are often seen
in areas where erythroplastids are found.  3) the senile erythrocyte.,
referred to in the  literature as "basket" or "smudge" cells.
It is seen in various stages of degeneration in fixed smears.  The
cytoplasm is more lightly stained, and the plasmalemma distended in
comparison to younger definitive erythrocytes.  The nuclei of this
cell are larger than normal mature red blood cell nuclei.  At times,
the nuclear membrane ruptures, as chromatin is often found dispersed
throughout some of the cells.  Senile erythrocytes with cleared cytoplasm
are often seen in fixed smears.

     A certain amount of extracellular chromatin, presumably  from
disintegrating "smudge" cells, is almost always seen in fixed preparations.
This extracellular chromatin is seen in increased amounts in  thick smears,
which require a longer drying period.

     In fresh preparations, senile erythrocytes are not readily dis-
tinguishable due to cytoplasmic opaqueness of red blood cells.  However,
there is a degree of chromatin and other cellular debris present in these

                                     30

-------
 in vitro preparations, which is more noticeable  in  older preparations  a
 few  hours  following  blood  collection.

     Acidophilic Granulocytes — In stained smears, eosinophilic granulocytes
are the only type of mature granulocytes found in the peripheral blood of
C_. variegatus.  Immature cells vary according to the staining properties
of the granules and the shape of the nuclei.  Nondefinitive cells are seen
to be basophilic, basophilic with some acidophilic properties, or completely
acidophilic.  The overwhelming majority of the granulocytes contain only
eosinophilic granules.  Indented or segmented nuclei are not observed in
cells having basophilia.  Truly lobed nuclei are not observed in any of the
granulocytic cells.  With the understanding that basophilia precedes
acidophilia in blood cell ontogeny, it is reasonable to suspect that those
granulocytic cells exhibiting basophilia are likely immature forms of
eosinophilic granulocytes.

     In fresh preparations, the granulocytes exhibit ameboid movement.
Pseudopodia are formed which may be fingerlike or blunt projections.
The fingerlike projections appear to initiate movement of these cells.
These first projections do not contain granules.  Following the initial
movement, a large portion of the cytoplasm containing granules streams
into the first, smaller projection, forming a second, large, blunt pseudo-
podium.  During this movement, the nucleus usually has a peripheral
location at a point most distant from the advancing end.  Often a trail-
ing nucleus can be seen at the end of a narrow, extended cytoplasmic
isthmus at the end opposite from the advancing cytoplasm.

     Most granulocytes move constantly among the formed elements of the
blood and thrombocytic networks.  These cells often attach to the slide
or coverslip and move over a planar surface.  Frequently, when free
floating in plasma streams under the coverslip, the granulocytes are
observed to round up into spheres.  After coming to a stop, a fingerlike
projection usually emerges from the sphere and the ameboid movement
commences again.

     The average diameter of the rounded form of eosinophilic granulocyte
is 8 um when stained and is slightly larger in fresh preparations.  Ex-
tended living cells observed during times of motility attain a length
many times the diameter of the spherical forms.

     Nongranular Leucocytes — At the light level, three morphologically
distinct nongranular leucocytes are observed which have been classified
into two separate groups, thromboid cells and lymphoid cells.

     The thromboid cells are subclassified into "lone nucleus" forms and
"extended"forms.The nuclei of the "lone nucleus" cells are round to
ovoid with little or no visible cytoplasm surrounding the nuclei.  In
stained smears, the nuclei in these cells are a deeply basophilic
magenta to dark purplish blue.  The cytoplasm, when observed, is
pink.  In fresh preparations, the nuclei are opaque with little or
no visible cytoplasm.  These cells are often found in clusters with
long, thin pseudopodia-like structures forming extended branching

                                     31

-------
networks among the cells.  These structures are actually long fibrin
strands radiating from ruptured plasma membranes.  They are often
several times the length of the cells.  The "lone nucleus'1 form of
thromboid cells are numerous in fish that have been unduly stressed
before or during blood collection, or if the blood is not quickly
heparinized or fixed.

     The "extended" forms of thromboid cells are either oval or they
have one or two spiked poles.  The nuclei in stained preparations are
magenta, with a pointed or somewhat rounded indentation usually present
in the nucleus.  In both stained and fresh preparations, the nuclear
chromatin has a crosshatched and somewhat patchworked appearance.  These
cells were designated "spindle cells" in the earlier literature because
of their characteristically elongated shape.

     Lymphoid cells are usually slightly larger than the "lone nucleus"
forms of thromboid cells, but resemble them to the extent that reliable
differential thromboid-lymphoid cell counts could not be made.  A
narrow rim of cytoplasm is seen surrounding the nucleus of these cells
in stained and fresh preparations.  In Romanowsky-stained smears, the
nucleus is basophilic, having a magenta or violet coloration, but rarely
containing the dark, purplish blue coloration of the "lone nucleus" forms.
The eosinophilic cytoplasm is a light pink and short pseudopodia are
often seen around the cell periphery.  There are usually one or more deep
indentations in the rounded nucleus.  These cells resemble mammalian
small lymphocytes morphologically.

Electron Microscopy:

     Erythrocytes -- At the ultrastructural level, the cytoplasm of the
erythrocytes  is  seen to possess a finely granular composition that does
not appear to contain free or membrane"bound ribosomes.  Scarce smooth
endoplasmic reticulum and a few thin mitochondria are seen.

     In the nucleus, dark chromatin and lighter interchromatin material
may be observed.  Lining the nuclear periphery are dense chromatin
patches which often penetrate deep into the nuclear interior and some-
times extend to the opposite edge of the nucleus.  The two perinuclear
membranes are not contiguous, and a prominent perinuclear space is
normally visible.  The perinuclear membranes unite to form the nuclear
pores.  At these sites, the cytoplasm is continuous with the  lighter
nuclear matrix.

     Lymphoid Cells — This cell type normally appears unevenly circular
to elliptical with a relatively large, irregularly round to oval nucleus.
The nucleus usually contains one or more deep clefts and is centrally
located within a surrounding, thin ring of cytoplasm.  At times, deep
nuclear indentations divide the nucleus providing the appearance of two
nuclei in some sections.  Areas of nucleoplasm and dense chromatin mate-
rial provide a light and dark patchwork pattern in the nucleus.  Nuclear
pores are present, providing continuity between the cytoplasm and the
lighter nuclear matrix.

                                     32

-------
     The most conspicuous organelles in the cytoplasm are the large
mitochondria.  These are usually elongated and  contain well-developed
internal cristal structures.  Extensive rough endoplasmic reticulum and
many free ribosomes are visible in the cytoplasmic matrix, but smooth
endoplasmic reticulum is not abundant.  A Golgi apparatus is often seen.
The cytoplasmic membrane is plicated and usually contains many small
pseudopodia as well as areas of pinocytotic vesicle formation.  These
areas of pinocytosis are discerned by the presence of adjacent vacuoles
in the cytoplasm.  A few dense granules are apparent within the
cytoplasm.

     Thromboid Cells — Electron microscopy reveals a superficial re-
semblance between thromboid and lymphoid cells.  Both have a centrally
located nucleus surrounded by a relatively thin ring of cytoplasm.  The
nuclei of both cells are composed of clearly defined areas of chromatin
and interchromatin material.  Both cells have a relative abundance of
ribonucleoprotein.

     Unlike the lympoid cells, however, the thromboid cells possess
nuclei which are not deeply indented.  In the cytoplasm, numerous, dark,
medium-sized to large granules are usually present, and mitochondria are
fewer and smaller in thromboid cells.  Also, fewer small pseudopodia
are present at the periphery of thromboid cells.  The most prominent
cytoplasmic feature of the thromboid cells are  the numerous electron-
lucent vesicles.

     An important physiologic mechanism peculiar to thromboid cells is
the clotting process.  Groups of these cells seen in sections of clotted
blood show long, thin fibrin strands.  During clotting, these strands
radiate from the interior of the cells through  newly formed openings in
the disintegrating plasmalemma.  The strands'form dense, interconnecting
networks between the thromboid cells and attach to nearby erythrocytes
and any other cells in the immediate vicinity.  These radiant strands
appear to be continuous with an extracellular flocculent material seen
throughout the sections and may, in fact, contribute to it.

     During the clotting process, the thromboid cell nuclei become
pyknotic and the cells lyse.  Many large electron-lucent vacuoles appear
in the cytoplasm of the cells.  Electron-opague granules are often
seen inside and immediately outside the ruptured cells.  Cytoplasmic
organelles in thromboid cells are difficult to  recognize in sections
in which fixation occurred during the process of coagulation.  Partially
enclosed, membrane-bound cytoplasmic portions of some of the cells
participating in the clotting process become separated from their
original cell mass.
Aseptic Embryo Technique

     Our initial experiments  indicated that C_. variegatus embryos
could be maintained  for up  to 18 days under a variety of aseptic
conditions  (Table  3).  Fry  that were fed  sterile particulate  food

                                     33

-------
tended to survive longer than those that were not fed.  Cannibalistic
behavior was observed when large numbers of individuals were placed
in the same flask.  It was determined that this behavior did not occur
'if the population density was kept below one individual per 10 ml of
medium.  The goal of experiments currently in progress is to establish
the specific aseptic medium that will provide maximum survival time.
Also, the morphology of the aseptic embryos is being compared with embryos
maintained for the same period of time in a septic environment.  The
effort includes size and weight comparisons, light microscopy, and
scanning and transmission electron microscopy.  Preliminary observations
of the external gross anatomy of the two types of embryos suggest that
they are quite similar.

     Experiments are currently in progress to determine BEN
toxicity of both septically and aseptically maintained embryos.  Once
these experiments are completed, both types of embryos will be exposed
chronically to subtoxic levels of BEN.
                                     34

-------
TABLE  3.  —  Survival  times of  aseptic  embryos  under  different
                 culture  conditions.
                                                 Ave. No.
                                Number            embryos
         Culture                   of                per          Ave.  Survival
        Conditions            Experiments        experiment       time (days)
A.     L-15 + 1%  Abs1              177
 B.     L-15  +  1%  Abs
       +  Supplements               4                  9                13


 C.     L-15  +  1%  Abs
       +  Supplements
       +  TM2                       1                 10                18
 D.    Sterile  artificial
      seawater +  sterile
      Artemia  eggs                1                  2                10
 E.     Sterile  artificial
       seawater +  1%  Abs
       + TM                        1                10                16


            1Antibiotics are  from Gibco.  Cat. No. 6005240, 100X
               Penicillin  10,000  U/ml-Fungizone, 25 meg/ml-Streptomycin,
               10,000 meg/ml.
            "TetraMin  Tropical  Fish  Food.
     A major advantage of the aseptic embryo technique is that it
provides an opportunity to study the effects of a contaminant on the
organism during its highly susceptible embryonic period entirely free
of any influence by bacteria or other organisms normally present.
Thus, the system should be particularly valuable in carcinogen studies
related to enzyme induction and the pathways by which procarcinogens are
metabolized into active carcinogenic agents.

     Although the survival times already attained are sufficient for
many of the studies anticipated for this system, it is reasonable to
assume that continued study of the system will lead to the development
of a procedure that will allow normal development of the organisms to
an adult state.  Experiments designed to attain this goal are currently
underway.  They involve observing the effects of changes in basal media

                                     35

-------
and substrate supplements,  determining food preference and optimum
nutrition, understanding the effects of different population densities,
and establishing the best photoperiod and level of aeration for embryonic
development.

     Laale and Lerner (29),  in a recent review, have indicated a
need for continued study of specific teleost teratogens and suggest
that these studies must involve the determination of specific metabolic
targets.  They state that this will require detailed studies of the
responding embryos at all levels of organization.  Obviously, the
present system, in which embryonic development in both contaminated
and uncontaminated sterile environments can be carefully observed,
provides an excellent system for such ichthyoteratological studies
as well as general studies of teleost embryonic development.
Embryo-Primary Cell Culture Technique

     With 75-80% efficiency, these embryos become attached to the
surface of the Linbro wells within 2-3 days and a mixed population
of cells begins migrating from the attached region of the embryo.
As cell migration continues, the attached region of the embryo
becomes progressively disorganized.  Three morphological types of
migrating cells are commonly observed: fibroblast-like cells, pigmented
cells, and ovoid-shaped cells.  The pigmented cells often develop
interconnecting processes.  Frequently, the pigmented cells form
rather extensive interconnecting networks.
     Since the embryo-primary cell culture technique blends the use
of an aseptic embryo with cell cultures into one system, it is possible
to observe simultaneously the effects of a specific carcinogen on a
relatively intact organism, and primary cell cultures from that same
organism.  The technique provides numerous advantages.  It provides
better genetic continuity; i.e., the cells of the primary cell culture
can be expected to be more genetically similar to the cells of the
intact organism than cells of an established cell line.  The major
portion of the embryo remains intact with relatively normal organ
arrangement and function, and the system can be removed for histo-
logical examination.  Also, the carcinogens make direct contact with
the various cells of the system in a chemically defined medium.
Not only does the system provide for exposure of the organism during
the highly susceptible embryonic development period, but the embryo
likely "conditions" the medium, bringing about increased viability
and differentiation in the cells of the primary culture.  Such
conditioning factors have been reported (30, 31), and a conditioning
phenomenon is consistent with the observations in the present study
that the cells in the primary cell culture region change their
appearance and behavior after the intact portion of the embryo becomes
disattached or is removed.  Finally, the system bridges the gap
between in vitro systems, such as established cell lines, and in vivo
systems that are currently in use.

                                     36

-------
     The technique may be used  for  a variety  or  tests  and  assays ,
such as:  1) tests for toxicity, teratogenicity,  and cell  transformation
2) karyologic alterations,  3) detection  of  antigenic or  enzymatic
changes, 4) metabolic activation/cocultivation,  5)  effects on
macromolecule synthesis  (collagen,  mRNA,  etc,) and, 6) viral
genome activation.

     Factors that may be considered limitations  are:   1) the
embryos do not contain the  normal microbial flora of their
septically reared counterparts,  2)  the feeding mechanisms  of
the cells, and to some degree,  the  intact embryos are  likely different
from normally reared embryos, and 3) the restricted embryo
movement and partial disorganization may bring about metabolic
differences as well as other subtle effects.  However, even with
these limitation, the model provides a useful system that  should
be reproducible, standardizable, and easily comparable to  in vivo
systems.  Laale  (31) has been able  to achieve 40-day survival times
with a similar system that  employs  blastoderm isolates from the the
zebrafish, Brachydanio rerio, so it is reasonable to assume that
equal or greater survival times  can ultimately be obtained with the
C_. variegatus system.  This should  provide  adequate time to perform
assays.  It is noteworthy here  that Hillebert and Martin (32) were
able to obtain multilayered foci in an established cell line from
the sheepshead, Archosargus probatocephalus,  exposed to benzo(o)pyrene
after only three passes.
Primary Hepatocyte Cell Culture Technique

     Data obtained from primary vertebrate hepatocyte cultures have
demonstrated excellent correlation with bacterial mutagenicity assays
(33) , and it follows  that  vertebrate  cells should be more relevant
than the procaryote assay  systems often used  to determine the mutagenic
and carcinogenic potential of  suspect chemicals  (34).  Recent studies
employing rat primary hepatocyte monolayers  (35, 36) have demonstrated
the usefulness of these methods in providing  a better understanding
of the metabolic processes responsible for the conversion of procarcin-
ogens to the active molecules.  Additionally, rat liver culture systems
have proven useful in studying mechanisms of  tumor promotion  (37, 38).

     Since primary hepatocyte  cultures can be expected to be more
genetically similar to the cells of the intact organism than an
established cell line, data derived from C_. variegatus primary hepato-
cyte cultures should  be more directly comparable to the whole animal
system.  Once the methods  have been optimized for maintaining these
primary hepatocyte cultures and have  been adequately standardized,
it should be relatively easy to conduct assays using numerous dupli-
cates with positive and negative controls.  In addition to  its value
in extending the usefulness of C. variegatus  as an assay system, the
system may provide advantages  over mammalian systems currently in
use.  For example, the systems can be operated more economically
because the cultures  can be maintained at room temperature, thus

                                      37

-------
avoiding the use of expensive incubators with gas-controlled envir-
onments .
Immunological Studies

     The immune system of fish is known to be very complex involving
many substances ranging from specific "acute phase" enzymes to
interferon, to immunoglobulin production (39).   The primary problem
in studying the immune response of C_. variegatus lies in its size.
The immune response to specific antigens has been routinely studied
in many larger species of fish, such as brown trout (Salmo trutta)  (40)
and sockeye salmon (Oncorhynchus nerka) (41), in attempts to increase
production of economically important species.  In studies of the
phylogeny of the immune response, large species of shark have also
been studied (42).  In order to study the humoral immune response
of any organism, serum must be easily obtainable in reasonable
volumes, which is not a problem in larger fish where blood may be
obtained in milliliter amounts via heart puncture or caudectomy.
The blood volumes of the largest C_. variegatus, however, is measured
in microliters. 'It has been necessary therefore to miniaturize or
modify techniques used to study the immunology of the fish and to
develop a bleeding procedure that significantly enhances blood
recovery from these organisms.

     Serum collection — Using fish averaging 5.4 cm in length, the
volume of blood collected averaged 18.5 yl per fish when the caudal
fin was completely severed and no treatment with anticoagulant or
anaesthetic was used.  When fish of similar size (averaging 5.5 cm)
were treated with sodium citrate and MS222, then bled using the
modified cut, the volume of blood collected averaged 33.7 yl/fish.
Thus, the modifications introduced to the bleeding procedure sign-
ificantly increased  (almost doubled) the blood volume obtainable
(Table 4).  Omission of any of the additional steps in the modified
procedure resulted in collection of smaller blood volumes.

     Collection of leucocytes — Both techniques employed were
successful in separating leucocytes from whole blood.  Collection
of cells from the buffy coat resulted in relatively more contami-
nation with erythrocytes than collection by the Ficoll-Paque pro-
cedure.  Leucocyte preparations collected by either method contained
cell types identified as normally present in fish blood, based on
comparison of cells in Wright-stained suspensions with descriptions
of previously described cell types  (43).  A high percentage of cells
were viable.
                                     38

-------
TABLE 4. -- Comparison of Blood Volumes Collected By Different
                Bleeding Procedures.
                                     AVERAGE LENGTH       AVERAGE BLOOD
       BLEEDING PROCEDURE               OF FISH          VOLUME COLLECTED
       NO PRETREATMENT,
       CAUDAL TOTALLY                  5.4±0.6              18.5±3.4
       SEVERED.
       PRETREATMENT WITH
       MS222 AND CITRATE,
       CAUDAL PARTIALLY                5'5±0'8             33.7±16.0
       CUT.
     Serum electrophoresis — Serum electrophoresis was performed to
provide baseline data for serum immunoelectrophoresis as well as for
its own intrinsic value for comparison of normal and carcinogen-exposed
fish.  The small volume of serum required for each electrophoresis run
allowed replicate samples of individual fish sera to be tested.  Densit-
ometer scans of duplicate runs showed good reproducibility.  Figure 8
compares the results of densitometer scans of serum from normal
C. variegatus and from fish exposed to 1 ppm BEN for seven weeks.  A scan
of normal human serum is included as a procedural control, and shows a
typical pattern.  There are obvious differences between the serum
profiles of the normal and BEN-exposed fish.  The normal serum shows
eight peaks, whereas the serum from the BEN-exposed fish shows only
five peaks, with some components evidently present only in very weak
concentrations or missing altogether.  Sera from different normal
individuals showed  some variation but were always similar, producing
seven to nine peaks in densitometer scans, with most producing eight.
Sera from BEN-exposed fish, on the other hand, showed wide variations
in serum profiles, both in size and number of peaks.
                                     39

-------
                              c.
FIGURE 8. — Comparison of serum electrophoresis profiles of  (a) normal
serum of _C. variegatus, (b)  serum of C_. variegatus exposed to 1 ppm
benzidine for 7 weeks and (c)  normal human serum.  Fastest migrating
peak is at right of profiles.
                                     40

-------
     Immune rosette formation — Immune rosettes were  successfully
produced and identified in spleen cell suspensions from C_. variegatus.
Complete "halos" of erythrocytes were frequently observed with no
leucocyte evident inside.  Careful examination of these preparations
led to the conclusion that the mounting or  staining procedure often
caused the destruction of the rosette-forming cells, which evidently
are rather fragile.  Destruction of  such  cells within  a complete ring
of erythrocytes left only the ring with little or no evidence of the
rosette-forming cell itself.  If only a few erythrocytes were bound
to a rosette-forming cell which lysed, it is likely that the rosettes
would not have been scored.  It is necessary to eliminate this problem
if the rosette-forming technique is  to be of use in quantitating
antibody-forming cells in spleen suspensions of C_. variegatus.  Use of
a mild fixative prior to mounting the suspension and use of wet mount
preparation are being explored as possible  solutions.

     Bacteriophage neutralization assay —  Prior to carrying out the
bacteriophage neutralization assay with serum from fish immunized with
MS2 phage, preliminary experiments were done with serum from normal non-
immune fish and from non-immune fish exposed to 1 ppm  BEN for seven weeks.
These experiments were done to determine  if any natural neutralizing
substances were present.  Table 5 presents  partial results.  Serum from
normal fish had no significant effect on  the virus titer, as may be seen
by comparing number of plaque-forming units (PFU) in normal serum-exposed
phage inoculum with the number in phage inoculum not exposed to serum.
By contrast, significant reduction of PFU/inoculum was observed if the
inoculum was exposed to serum from BEN-exposed fish.   At a serum
dilution of 1/32 to 1/256 or higher, the  plaque reduction became less
marked, but still significant.  Further experiments are necessary to
demonstrate the dilution necessary to eliminate plaque reduction
completely.  The results suggest that there are at least two components
of serum from the benzidine-exposed  fish  which reduce  plaque titer.
One dropped below the minimal concentration for effect at a dilution
of 1/16 to 1/32 and the other still  exerted an effect  at a dilution of
at least 1/256.  The results further suggest that the  plaque reduction
may not be dose-dependent.  Additional studies are planned to determine
the identity of the serum components.

     The bacteriophage neutralization assay has thus been easily
adapted to the C_. variegatus system.  Normal and carcinogen-exposed fish
are currently being immunized with MS2 phage.  Upon completion of the
immunization process, the assay will be applied to comparison of
neutralization titers, with non-immune serum from normal and exposed
fish included as controls.
                                     41

-------
TABLE 5. — Effects of  serum  from"normal and benzidine-exposed
             Cyprinodon variegatus on the titer of MS2 bacteriophage.
Source of serum
Bcniidinc-trcatcd fisli
Normal fish
Phage control (no scrum used)
Serum dilution
1/2 1/4 1/3 1/16 1/32 1/64 1/128 1/256
6Sa 65 62 89 113 106 110 117
153 159 132 1S8 1S4 152 153 169


No serum
---
---
141
         a.  Numbers represent number of plaque-forming units (PFU) .  A dilution of

            phage stock was used which should have produced approximately ISO PFU/

            inoculum volume.
Effects of Salinity,  DOC,  and Light on DENA Extractability

     Nitrosamines,  known chemical carcinogens, have been found in  a variety
of foods  (44,  45),  and  recently there have been some efforts to
detect these compounds  in the environment (46, 47, 48).

     However,  studies that identify the factors affecting the
chemical behavior of  nitrosamines in the environment are limited.
Tate and Alexander  (49)  monitored degradation of nitrosamines in
soil, water, and  sewage using extremely high concentration spikes
of nitrosamines  (100-2000 mg/1).   Also, the chemical behavior of
nitrosamines in super-acid solutions have been studied  (50, 51).
Extrapolations of data  from such studies, however, may not reflect
the behavior of nitrosamines at trace levels in natural aquatic
systems.  Thus, in  order to further our ability to interpret data  from
experiments exposing  fish to parts-per-million concentrations of
nitrosamines (3,  4, 52,  53), this study was conducted in collaboration
with Mr. Robert A.  Fricke and Dr. Richard H. Pierce, Jr. at the
Florida Institute of  Technology (54).

     In this study, the effects of salinity dissolved organic
carbon (DOC), and light on the extraction of diethylnitrosamine
(DENA)  from water at  trace concentrations (55-85 yg/1) and at I mg/1,
were investigated.

     Method 607 (16)  approved by the U.S. Environmental Protection
Agency, when used for trace concentration extraction of DENA, provided
rather inefficient  recoveries of DENA at 1 mg/1.  For example,
                                      42

-------
concentration of the extraction  solvent,  dichloromethane,  employed
by the EPA method caused as much as  75% of  the  originally  extracted
DENA to be lost by evaporation.   The  revised  extraction method
that was developed/ employing hexane  as an  extraction solvent, requires
no concentration of extracting solvent when simple  water extraction
of 1 mg/1  (or greater) of DENA is being performed,  thus, avoiding the
volatilization loss of the previous method.

     Varying the salinity  (0.5 - 24 parts per thousand)  and
DOC  (0.5 - 11 mg/1) had no effect on  the  extractability of DENA
at 1 mg/1  (Figure 9).  However,  a significant decrease in  extrac-
tability of DENA  (at 55-85 yg/1)  with increasing salinity  was
found  (Figure 9).  A decrease in trace  level  extractability was
also found at high concentrations (11 mg/1)  of  DOC  (Figure 10).
 IOC
  9C
  70
  60

o 50
o
UJ
o:
  40
 30
 20
  10
•-Img/L
• -55-85>jg/L
D -DISTILLED
   WATER
                                                 i  i   i   i   I   i   i  i
                               10
                                             15
                                                           20
                                                                         2 5
                               SALINITY  (%o)

   Figure  9.   The effects of salinity on DENA extractability.
                                     43

-------
                100 -
                90
                 80
                 70
               ui
               o
               O 60
               IT
               IU
               >
               O 50
               o
               UJ
               rr
                 30
                 20
                  10
a- I mg/L
• - 55-85;jg/L
                            23456783
                            DISSOLVED ORGANIC  CARBON   (mg/l)
        10
Figure 10.   The  effects  of dissolved organic carbon on DENA extractability.


     No significant loss of DENA in brackish water occurred over
a 7-day period (Figure 11).  However, the addition of DOC to the
brackish water enhanced the photodegradation of DENA, resulting  in
a 30% loss in 7 days  (Figure 11).

     In order to determine the recovery of DENA in the  1 mg/l
range from natural brackish water, samples were taken from  a brack-
ish creek  (Salinity ,  13 parts per thousand)  and spiked  with 1  mg/l
DENA.  Control water  for these experiments was artificial seawater
                                      44

-------
taken from aquaria  in which  C_.  variegatus  had been maintained for two
months.  These experiments indicate  that  the  laboratory data from this
study could be validly  extrapolated  to  the natural environment.   Finally,
the revised method  for  DENA  extraction  devised in the present study is a
feasible approach to monitoring DENA at the concentrations normally present
in the chronic exposures  of  teleosts conducted in this project.
    100 -
     90 -
     80
     70
   °60
   re
   UJ
   >30
   O
   o
   UJ
   E40
     30
     20
     10
                                                                        4.7
                                                                         o
                                                                         c
                                                                         o
                                                                         m
                                                                         Z
 n
2 o
                |        234        567
                                    DAYS
Figure  11.   The effects of light on extractability  of DENA from brackish
water  (Salinity = 10 ppt)  with added DOC (6 mg/1) during a 7-day period.

-------
                                  REFERENCES
 1.   Page,  N.  P.   Concepts of a bioassay program in environmental carcin-
     ogenesis.   In:   Environmental Cancer,  Kraybill and Mehlman, eds.,
     Advances  in Modern Toxicology,  Wiley & Sons,  New York.   3:87-171.  1977.

 2.   Wellings,  S.  R., B.  B.  McCain,  and B.  S.  Miller.  Epidermal papillomas in
     Pleuronectidae  of Puget Sound,  Washington.   Review of the Current Status
     of the Problem.   In:   Progress in Experimental Tumor Research.  20:55-74.
     1976.

 3.   Pliss, G.  B.  and V.  V.  Khudoley.   Tumor induction by carcinogenic agents
     in aquarium fish.  J.  Natl.  Can.  Inst.  55:129-136.  1975.

 4.   Martin, B. J.  Effects of petroleum compounds on estuarine fishes.  U. S.
     Environmental Protection Agency.   Ecol. Res.  Series.  EPA-600/3-80-019.
     1980.

 5.   Riggins,  R.  M.  and C.  C.  Howard.   Determination of benzidine, dichlorobenz-
     idine, and diphenylhydrazine in aqueous media by high performance liquid
     chromatography.   Anal.  Chem.  51:210-214.   1979.

 6.   Law, W. M.,  R.  D. Ellender,  J.  H. Wharton,  and B. L. Middlebrooks.  Fish
     cell culture:  propoerties of a cell line from the sheepshead, Archosargus
     probatocephalus.  J.  Fish. Res. Board Can.   35:470-473.  1978.

 7.   Couch, John A.,  George Gardner, John Harshbarger, M. R. Tripp, and Paul
     Yevich.  Histological and physiological evaluations in some marine fauna.
     In;  Marine Bioassays,  The Marine Technology Society, Washington, DC, pp.
     156-173.   1974.

 8.   Luna,  L.  G.,  ed.  Manual of histological staining of the Armed Forces
     Institute of Pathology.  McGraw-Hill Book Co., New York.  1968.

 9.   Winstead,  J.  T.   A simple method to obtain serum from small fish.  U. S.
     Fish Wildl.  Serv. Fish. Bull.  77:509-511.   1979.

10.   Berry, M.  N.  and D.  S.  Friend.   High-yield preparation of isolated rat
     liver parenchymal cells.   J. Cell Biol.  43:506-520.  1969.

11.   Michalopoulos,  G. and H.  C.  Pitot.  Primary cultures of parenchymal liver
     cells on  collagen membranes.  Exp. Cell Res.   94:70-78.  ]975.
                                      46

-------
12.   Ingram, G. A. and J. B. Alexander.  The primary immune response ot brown
     trout  (Salma trutta) to injection with cellular antigen.  J. Fish 2iol.
     10 :63-72~!1977^

13.   O'Neill, J. R.  The immune response of the brown trout  (Salmo trutta, L.),
     to MS2 bacteriophage:  immunogen concentration and adjuvants.  J. Fish.
     Biol.  15:237-248.  1979.

14.   Stashak, P. W., P. J. Baker, and B. S. Roberson.  The serum antibody
     response to bacteriophage 4>X174 in germ-free and conventionally reared
     mice. I.  Assay of neutralizing antibody by a 50 percent neutralization
     method.  Immunology   18:295-305.  1970.

15.   Eisenstark, A.  Bacteriophage techniques, In:  Virology Vol. I, K.
     Maramorosch and H. Koprowski, eds., Academic Press, New York, pp. 449-524.
     1967.

16.   Federal Register.  44:69496-69499.  1979.

17.   Duursma, E.  The dissolved organic constituents of sea water.  In:
     Chemical Oceanography.  J. Riley and G. Skirrow, eds., Academic Press.
     Vol. I, pp. 433-477.  1965.

18.   Schoor, W. P.  Personal Communication.

19.   Igel, H. J., A. E. Freeman, J. E. Spiewak, and K. L. Kleinfeld.  Carcino-
     genesis in vitro.  II.  Chemical transformation of diploid human cell
     cultures:  a rare event.  In Vitro 11 (3) :117-129.  1975.

20.   Freeman, A. E. and R. J. Huebner.  Problems in interpretation of
     experimental evidence of cell transformation.  J. Natl.  Cancer Inst.
     50:303-305.  1973.

21.   Moreno, M.  Embryologic development of the sheepshead minnow, Cyprinodon
     variegatus.  M.S. Thesis, University of Southern Mississippi, Hattiesburg,
     Mississippi.  1979.

22.   Kagan, B. M.  The fertilizable period of the eggs of Fundulus
     heteroclitus and some associated phenomena.  Biol. Bull. Vol. 68-69:
     185-200.1935.

23.   Ballard, W. W.  Origin of the hypoblast in Salmo.  II.  Outward
     movement of deep central cells.  J. Exp. Zool.  161:211-220.  1966.

24.   Porter, R. C.  The gross anatomy and histology of the post-pharyngeal
     digestive tract of the sheepshead minnow, Cyprinodon variegatus.  Ph.D.
     Dissertation, University of Southern Mississippi, Hattiesburg, MS.  1981.

25   Al-Hussaini, A. H.  On the functional morphology of the alimentary tract
     of some fish in relation to differences in their feeding habits:  anatomy
     and histology.  Q. J- Microsc. Sci.  90:109-139.  1949.


                                     47

-------
26.   Harrington,  E.  J.  W.   The alimentary canal and digestion.   In:  The
     Physiology of Fishes,  M.  E.  Brown,  ed.,  Academic Press., New York.
     Vol.  1:109-161.   1957.

27.   Hale, P.  A.   The morphology and histology of the digestive systems of two
     freshwater teleosts,  Poecilia reticulata and Gasterosteus aculeatus.
     J.  Zool.   146:132-149.

28.   Long, R.  L.   Morphology of the peripheral blood cells of Cyprinodon
     variegatus.   M.S.  Thesis, University of Southern Mississippi,
     Hattiesburg, MS.  1980.

29.   Laale, H. W. and W. Lerner.   Teratology and early fish development.
     Amer. Zool.   21:517-533.   1981.

30.   Trinkaus, J. P.  and J. W. Drake.  Enhancement of morphogenesis in
     Fundulus blastoderms by fusion and crowding in vitro.  Dev. Biol.
     1:377-395.  1959.

31.   Laale, H. W.  In vitro secondary spreading from dissociating blastoderm-
     derived embryonic tissues of the zebrafish Brachydanio rerio.  Can. J.
     Zool.  59:812-822.  1981.

32.   Hillebert, S. A.,  B.  J. Martin and R. D. Ellender.  Chronic exposure of
     a teleost cell line to suspected carcinogens.  J. Miss. Acad. Sci.
     25:71-75.  1980.

33.   Probst, G. S. and L.  E. Hill.  Chemically-induced DNA repair synthesis
     in primary rat hepatocytes:  A correlation with bacterial mutagenicity.
     Ann.  N.Y. Acad.  Sci.   349:405.  1980.

34.   Butterwerth, E.  B.  Predictive assays for mutagens and carcinogens.  In:
     Scientific Considerations in Monitoring and Evaluating Toxicological
     Research.  E. J. Gralla,  ed., Hemisphere Publ. Co. ,  Washington, DC,
     pp.  157-177.  1980.

35.   Okano, P., J. P. Whitlock, Jr., H.  V. Gelboin.  aryl hydrocarbon
     hydroxylase and benzo(a)pyrene metabolism in rodent liver and human
     cells.  Ann. N.  Y. Acad.  Sci.  349:232-247.  1980.

36.   Newnow, S., J. Hwsingh, H. Bergman, R.  Easterling, J. Inmon, L. King,
     M.  Morris, and K.  Williams.  Carcinogen metabolism in rat liver cell
     cultures.  Ann.  N. Y.  Acad. Sci.  349:247-264.  1980.

37.   Weinstein, I. B., M.  Wigler, and V. Stadler.  Analysis of the mechanism
     of chemical carcinogenesis in epithelial cell cultures.  In:  Screening
     Tests in Chemical Carcinogenesis.  R. Montesano, H.  Bartsch, and
     L.  Tomatis, eds.  IARC Scientific Publication.  12:355-381.  1976.

38.   Huberman, E. and C. A. Jones.  The use of liver cell cultures in muta-
     genesis studies.  Ann. N.Y. Acad. Sci.   349:264-273.  1980.
                                     48

-------
39-  Ingrain, G. A.
     infection—a review.  J. Fish. Biol.   16:23-60.   1980.

40.  Ingrain, G. A. and J. B. Alexander.  The  immune response  of brown  trout
     (Salmo trutta, L.) to injection with  soluble  antigens.   Acta Biologica
     et Medica Germanica.  35:1561-1570.   1976.

41.  Evelyn, T. P. T.  The agglutination response  in  sockeye  salmon vaccinated
     intraperitoneally with  a heat-killed  preparation of  the  bacterium
     responsible for  salmonid kidney disease.  J.  Wildl.  Dis.  7:325-328.  1971.

42.  Clem, L. W., and P. A.  Small.  Phylogeny of immunoglobulin structure  and
     function.  I.  Immunoglobulins of  the lemon shark.   J. Exp. Med.
     125:893-920.  1967.

43.  Gardner, G. R. and P. P. Yevich.   Studies on  the  blood morphology of
     three estuarine  Cyprinodontiform  fishes.  J.  Fish. Res.  Bd. Can^
     27:2185-2196.  1969.

44.  Telling, G. G.   A gas-liquid  chromatographic  procedure for the detection
     of volatile N-nitrosamines  at the  ten parts per  billion  level in
     foodstuffs after conversion to their  corresponding nitramines.  J. Chrom.
     73:79-87-  1972.

45.  Cox, G.  Estimation of  volatile N-nitrosamines by high performance
     liquid chromatography.  J.  Chrom.  83:471-481.   1973.

46.  Fine, D., F. Rufen, and D.  Lieb.   Group  analysis  of  volatile and  non-
     volatile N-nitroso compounds.  J.  Chromatogr.  107:351-357.  1975.

47.  Fine, D. , D. Rounbehler, E. Sawicki,  and K. Krost.   Determination of
     dimethylnitrosamine in  air  and water  by  thermal  energy analysis:
     validation of analytical procedures.   Environ. Sci.  Technol.  11:577-580.
     1977.

48.  Fine, D. , D. Rounbehler,- A. Rounbehler,  A. Silvergield,  E. Sawicki,
     K. Krost, and G. Demarrais.   Determination of dimethylnitrosamine in
     air, water, and  soil by thermal energy analysis:  measurements in
     Baltimore, MD.   Environ. Sci. Technol.   11:581-584.  1977.

49.  Tate, R., and M. Alexander.   Stability of nitrosamines in samples of
     lake water, soil, and sewage.  J.  Natl.  Cancer Inst.  54:327-330.  1975.

50.  Chow, Y.  Photochemistry of nitroso compounds in  solution.  V.
     Photolysis of N-nitrosodialkylamines.  Can. J. Chem.  45:53-62.   1967.

51   Chow, Y., M. Lau, R. Perry, and J. Tarn.  Photoreactions  of nitroso
     compounds in solution.  XX.   Photoreduction,  photoelimination, and
     photoaddition of nitrosamines.  Can.  J.  Chem.  50:1044-1050.  1972.
                                     49

-------
52.   Ishikawa, T.,  T.  Shiamine, and S.  Takayama.   Histologic and electron
     microscopy observation on diethylnitrosamine-induced hepatomas in small
     aquarium fish (Oryzias latipes).   J.  Natl.  Can.  Inst.  55:906-916.
     1975.

53.   Draper, A., and W.  Brewer.  Measurement of the aquatic toxicity of
     volatile nitrosamines.  J. Toxicol.  Environ.  Health.  5:985-993.  1979.

54.   Fricke, R. A.   The  effects of salinity and dissolved organic carbon on
     the extractability  and degradation of diethylnitrosamine.   M.S. Thesis,
     Florida Institute of Technology,  Melbourne,  FL.   1981.
                                     50

-------
                                   TECHNICAL REPORT DATA
                            (Please read Instructions on the reverse before completing}
1. REPORT NO.
                             2.
                                                           3. RECIPIENT'S ACCESSION NO.
 (.TITLE AND SUBTITLE

 DEVELOPMENT OF A  CARCINOGEN ASSAY SYSTEM UTILIZING
 ESTUARINE FISHES
            5. REPORT DATE
            6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
 B. J. Martin
            8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
 Department of Biology
 The University  of  Southern Mississippi
 Hattiesburg, Mississippi 39406
             10. PROGRAM ELEMENT NO.
             11. CONTRACT/GRANT NO.

              CR  806212
12. SPONSORING AGENCY NAME AND ADDRESS
 U.S. Environmental Protection Agency
 Environmental  Research Laboratory
 Office of  Research and Development
 Gulf Breeze, Florida 32561	
             13. TYPE OF REPORT AND PERIOD COVERED
             14. SPONSORING AGENCY CODE
15. SUPPLEMENTARY NOTES
16. ABSTRACT
 The objective  of this project was  the  development of systems  to  assay
 the effects  of chemical carcinogens  on marine teleosts.  It was  de-
 termined  that  the LC-50 for benzidine  with respect to Cyprinodon
 variegatus was ca.  64 ppm.  Weekly contaminations of 1 ppm benzidine
 caused some  individuals to develop proliferative liver lesions.   Exposure
 of C. variegatus early embryos produced the following anomalies  at
 concentrations of 50 ppm and above:   tubed heart syndrome with distended
 pericardia,  poor circulation, sparse distribution of melanophores,  in-
 ability to hatch, abnormal head morphology.- scoliosis, and faint RBC
 pigmentation.   Chronic exposure of a cell line from Archosargus
 probatocephalus to  benzidine and benzo (a)pyrene produced mutagenic  effects.
 Two novel techniques were developed  to study the effects of carcinogens on
 C.  variegatus  at the cellular level  — an aseptic embryo technique  and  an
 embryo-primary cell culture technique.   Standard immunological techniques
 were miniaturized to study the immune  system of C_. variegatus.   Serum
 electrophoresis disclosed that the serum proteins of benzidine-exposed  fish
 differed  from  unexposed controls.
                                KEY WORDS AND DOCUMENT ANALYSIS
                  DESCRIPTORS
                                              b. IDENTIFIERS/OPEN ENDED TERMS
                                                                         C. COSATI
                                   eld/Group
 8. DISTRIBUTION STATEMENT
 Release to  public
•nnrlassif ied
                                              20 SECURITY CLASS i Tins page /
                                               unclassified
                                                                         22 PRICE
EPA
   Form 2220-1 (Rev. 4-77)   PREVIOUS ED.
                                TION IS OBSOLETE

-------