- EXECUTIVE SUMMARY
- PREFACE
- INTRODUCTION
- MATERIALS AND METHODS
- Field Sampling
- Laboratory
Analyses
- Fish Age Determination
- Histopathological
Analyses
- Chemical Analyses of
Sediments, Stomach Contents,
Liver Tissue, and Bile
- Stomach Contents Taxonomy
- Statistical Analyses
- Heterogeneity in Lesion
Prevalence
- Logistic Regression and Risk
Factor Analysis
- RESULTS
- Chemical Characterization
of
Sampling Sites
- Sediment Contaminant
Levels at
Sampling Sites
- Patterns of Occurrence of
Contaminants Within
Sediment
- Contaminant Levels in
Sediments, Stomach Contents,
and Tissues
- Stomach Contents Taxonomy
- Hepatic Lesions
- Hepatic Lesion
Prevalences
and Intersite Comparisons
- Risk Factors Associated
with
Hepatic Lesion
Occurrence in Individual Fish
- Relationships Between
Chemical
Contaminants and
Hepatic Lesion Prevalences
- Renal Lesions
- Renal Lesion
Prevalences
- Risk Factors
Associated
with Renal Lesion
Occurrence in Individual Fish
- Relationships Between
Chemical Contaminants and
Renal Lesion Prevalences
- Fin Erosion
- Fin Erosion
Prevalences
- Risk Factors
Associated
with Fin Erosion Occurrence
in Individual Fish
- Relationships Between
Chemical Contaminants and
Fin Erosion Prevalences
- DISCUSSION
- Patterns of Lesion
Occurrence
- Toxicopathic Response to
Contaminant Exposure in
Winter Flounder
- Biological Risk
Factors
Associated With Lesion
Occurrence
- Age
- Sampling Season
- Gender
- Relationships Between
Chemical Contaminants and
Idiopathic Lesions
- Hydropic Vacuolation
- Nonspecific
Hepatocellular
Necrosis
- Neoplasms and FCA
- Proliferative Lesions
- Spongiosis Hepatis
- Megalocytic
Hepatosis/Nuclear Pleomorphism
- Renal Lesions
- Fin Erosion
- PCBs and Idiopathic
Disease in
Winter Flounder
- CONCLUSIONS
- ACKNOWLEDGMENTS
- CITATIONS
- TABLES
- FIGURES
- APPENDIX
EXECUTIVE SUMMARY
This report presents and interprets the results of pathology studies
conducted on winter flounder
(Pleuronectes americanus) between 1987 and 1989 in
conjunction with NOAA's National
Status
and Trends (NS&T) Program. In these studies, a variety of
potentially contaminant-associated
disease conditions were monitored in winter flounder collected from
22 Northeast Coast sites, and
the relationship between disease occurrence and levels of organic
chemical contaminants in
sediment, stomach contents, and tissues was examined. Sampling was
conducted primarily in
spring of 1988 and 1989 (Cycles 5 and 6) as part of the National
Benthic Surveillance Project
(NBSP) of the NS&T Program, but to provide a more comprehensive view
of pathological
conditions in winter flounder from the Northeast Coast, data on
winter flounder collected in Long
Island Sound during 1987 and in Boston Harbor and adjacent embayments
during the winter of
1988 were also included in this memorandum. In the three studies
combined, a total of more than
1,500 fish were examined.
Embayments sampled included Salem Harbor, Boston Harbor,
Massachusetts Bay, New Bedford
Harbor, and Buzzards Bay in Massachusetts; Narragansett Bay in Rhode
Island; Niantic Bay in
Connecticut; several sites within Long Island Sound including New
Haven, Norwalk, Bridgeport,
and Rocky Point in Connecticut and Lloyd Point in New York; and
Raritan Bay and Great Bay in
New Jersey. The levels and types of chemical contaminants present in
sediments from sampling
sites within these embayments differed substantially. Among the most
heavily contaminated sites
were Mystic River and Quincy Bay in Boston Harbor, and Gravesend,
West Reach, and East
Reach sites in Raritan Bay. Sediments from these sites had elevated
levels of all or most of the
contaminants measured in this study, including aromatic hydrocarbons
(AHs), polychlorinated
biphenyls (PCBs), DDTs (DDT and its derivatives), and chlordanes.
The New Bedford Harbor
sampling site had very high concentrations of PCBs in sediments, but
relatively low
concentrations of other classes of contaminants. Lowest
concentrations of nearly all classes of
contaminants in sediment were found at the Plymouth Entrance site in
Massachusetts Bay and at
the Rocky Point site in Long Island Sound. The Rocky Point site
served as a reference site for
analyses of intersite differences in lesion prevalences. Detailed
information on concentrations of
contaminants in sediments from these sampling sites will be reported
in a forthcoming Technical
Memorandum.
Winter flounder were examined for necrotic, sclerotic, and
proliferative lesions in the kidney, fin
erosion, and a number of pathological conditions in the liver,
including:
- neoplastic lesions (e.g., hepatocellular and cholangiocellular
carcinomas, liver cell adenomas
and cholangiomas);
- putatively preneoplastic lesions (i.e., foci of cellular
alteration (FCA));
- several unique, presumably degenerative conditions (hydropic
vacuolation, nuclear
pleomorphism and megalocytic hepatosis, and spongiosis hepatis) which
have been shown to be
associated with toxicant exposure in previous studies with winter
flounder or other fish species ;
- nonneoplastic proliferative lesions such as hepatocellular or
biliary regeneration or
hyperplasia;
and
- a variety of nonspecific hepatocellular or biliary necrotic or
degenerative conditions.
Some of these lesions, such as liver neoplasms, FCA, and biliary or
hepatocellular degenerative
and regenerative lesions, are well established as histopathologic
biomarkers of contaminant stress
in fish and show strong evidence of a contaminant-associated etiology
on the basis of previous
field or laboratory studies with winter flounder or other fish
species. Other lesions, such as
necrotic and proliferative conditions in the kidney and fin erosion,
have shown some promise as
bioindicators of chemical pollution but require additional testing.
Prevalences of the lesion types described above were determined in
winter flounder from each of
the Northeast Coast sampling sites. In addition, logistic regression
was used to examine the
effects of selected biological and contaminant-associated risk
factors (e.g., age, gender, site and
season of capture, and levels of chemical contaminants in sediments
and fish) on lesion
occurrence. Logistic regression is a multivariate statistical method
which is similar to stepwise
multiple regression, but is uniquely suited to binomial or
proportional data such as lesion presence
or absence or lesion prevalence. This technique is frequently
applied in epidemiological and
epizootiological studies and has a distinct advantage over simple
intersite comparison of lesion
prevalences because it allows biological factors such as age and
gender to be taken into account
when evaluating relationships between disease and site of capture or
other parameters associated
with contaminant exposure.
Relationships between lesion prevalences and the following major
classes of chemical compounds
were examined: high molecular weight AHs containing 4-6 benzene
rings (HAHs), low molecular
weight AHs containing 1-3 benzene rings (LAHs), total AHs (LAHs +
HAHs), PCBs, DDTs, and
chlordanes. These classes of chemical compounds were chosen as
potential risk factors because
1) they are found at high concentrations in sediments from a number
of Northeast Coast sites; 2)
they are generally present in estuaries located near metropolitan
areas throughout the United
States and thus represent an index of urban pollution; and 3) they
are generally recognized as
toxicants in fish or mammals. Moreover, concentrations of these
contaminants in sediments were
highly correlated with levels of the same compounds or their
derivatives in stomach contents,
liver, and bile of winter flounder, indicating that these compounds
were being taken up and
accumulated or metabolized by the fish. Concentrations of organic
contaminants in all
compartments were correlated with disease prevalences to evaluate the
consistency of observed
relationships; however, contaminant levels in liver and bile were
considered to have the greatest
relevance as toxicologic risk factors because they correspond most
closely to actual uptake of
toxicants by animals.
Significant Findings
- Overall, hepatic lesions in winter flounder proved to be
extremely reliable histological
markers
of chemical contaminant exposure. Hydropic vacuolation and
nonspecific necrotic lesions showed
particular utility as indicators of anthropogenic stress. Both
lesions occurred at high prevalences
(25 to 50%) in winter flounder from heavily contaminated sites, and
showed a clear gradient in
prevalence from heavily to minimally contaminated sites. In
addition, they were closely correlated
with levels of several classes of organic contaminants in fish liver
and bile, sediments, and stomach
contents.
- Hepatic neoplasms, foci of cellular alteration, and
nonneoplastic proliferative lesions such as
biliary hyperplasia were also found in winter flounder from several
heavily contaminated sites, but
these lesions were not highly correlated with chemical contaminant
levels in liver, bile, stomach
contents, or sediments. The lack of correlation can be partly
attributed to the relatively low
prevalences (2 to 10%) of neoplastic, preneoplastic, and
proliferative lesions in winter flounder
sampled in this study. Also, there is a long latency period for the
development of neoplasms and
related lesions. Consequently, even if contaminant exposure is a
critical factor in their etiology,
they may not be highly correlated with contaminant levels in sediment
at the sampling site, or with
levels of contaminants in the animals at the time of collection.
This is especially likely to be the
case in younger animals, or in a relatively mobile species such as
winter flounder. The problem is
compounded by the fact that for certain key environmental carcinogens
such as AHs, current
methodologies estimate only short-term, not chronic exposure. There
is substantial evidence from
both field and laboratory studies that neoplastic and preneoplastic
lesions in fish are associated
with exposure to environmental toxicants. However, because of their
relatively low prevalence in
many marine species and their long latency period, they may be less
suitable as biomarkers of
anthropogenic stress in field monitoring programs than
earlier-occurring and more prevalent
toxicopathic lesions.
- The compounds most frequently identified as risk factors for
pathological conditions in
winter
flounder were AHs, DDTs, and chlordanes. In addition to these
chemical contaminants, several
biological factors, particularly age and sampling season, had a
significant influence on disease
prevalence. However, because AHs, DDTs, and chlordanes generally
occurred together in
sediments, and animals were exposed to them simultaneously, it was
difficult to evaluate the
relative importance of these toxicants in the development of
pollution-associated disease in winter
flounder.
- PCBs did not emerge as significant risk factors for any
pathological condition observed in
winter flounder. Not only did PCB levels in sediment, stomach
contents, and liver fail to correlate
with lesion prevalences, but perhaps more importantly, fish sampled
from one site, New Bedford
Harbor, that had high levels of PCBs in sediments but relatively low
levels of other contaminants
including AHs, had low prevalences of all pathological conditions
monitored in this study. These
data, however, do not exclude PCBs from a role as secondary etiologic
agents for neoplastic
lesions or regenerative and degenerative conditions in winter
flounder.
- Prevalences of renal lesions were not highly correlated with
levels of organic contaminants in
sediments, stomach contents, bile or liver of winter flounder. Of
the three classes of lesions
monitored in this study (proliferative, necrotic, and sclerotic),
only necrotic lesions showed any
evidence of a relationship to contaminant exposure.
- Fin erosion prevalences were relatively high (e.g., 25 to 30%)
at certain heavily
contaminated
sites especially within Boston Harbor, and were significantly
correlated with contaminant
concentrations in sediments and stomach contents, and with biliary
FAC concentrations.
However, because biotic factors such as fungal or bacterial pathogens
as well as xenobiotic
contaminants may be associated with the development of fin erosion,
it is not as specific a
biomarker as other toxicopathic lesions.
- The toxicopathic response (e.g., hydropic vacuolation of
hepatocellular and biliary cells and
biliary hyperplasia, and a predominance of cholangiocellular
neoplasms over hepatocellular
neoplasms) typically exhibited by winter flounder sampled from urban
sites is quite different from
patterns of lesion occurrence associated with the histogenesis of
neoplasia in mammals and fish
species such as medaka, rainbow trout, and English sole. The types
of lesions found in winter
flounder in contaminated areas have also been noted in other
bottomfish (e.g., starry flounder
(Platichthys stellatus) and white croaker (Genyonemus
lineatus) ) at contaminated
sites on the
West Coast sampled in the NBSP. These differences between starry
flounder, white croaker, and
winter flounder and species such as English sole suggest that
toxicopathic response, and possibly
histogenesis of neoplasia, may follow more than one pathway in
teleost fish.
In these studies we were not able to evaluate the effects of a number
of potential toxicological
risk factors such as certain metals, especially biologically-active
forms of metals including
organometals, dioxins, and coplanar PCBs, because appropriate
cost-effective methodologies
amenable for large-scale biomonitoring are not currently available.
Development of such
methodologies and their incorporation into the NBSP could bring about
significant advances in
our understanding of the toxicopathic effects of such contaminants
and their importance in the
development of pollution-associated disease. Our ability to identify
relationships between
environmental toxicants and disease conditions in benthic fish would
be further enhanced by the
utilization of indicators (e.g., xenobiotic-DNA adducts) that measure
chronic exposure to labile
toxicants, such as AHs, and by determination of concentrations of
selected contaminants,
including CHs, in individual fish. Such data may be used to develop
predictive dose-response
models linking contaminant levels in individuals with the risk of
disease.
Although biomonitoring studies such as those reported here could
benefit greatly by inclusion of a
broader suite of contaminants and biological endpoints, the data
gathered thus far provide
significant indications that certain environmental contaminants are
potential risk factors for
pathological conditions observed in winter flounder, a benthic fish
used as an indicator species for
the assessment of environmental quality on the Northeast Coast. This
correlational evidence,
while being circumstantial, provides valuable insights into the
potential involvement of certain
groups of contaminants in disease initiation and progression, and
will serve as a basis for the
formulation of definitive cause-and-effect studies with potential
etiologic agents present in urban
environments.
PREFACE
The National Benthic Surveillance Project (NBSP) was initiated in
1984 as a component of the
National Oceanic and Atmospheric Administration's (NOAA's) National
Status and Trends
(NS&T) Program, which was designed to assess the status of, as well
as long-term changes in, the
environmental quality of the Nation's coastal and estuarine waters.
The NBSP is a cooperative
effort between the National Marine Fisheries Service and the Coastal
Monitoring and Bioeffects
Division of the Office of Ocean Resources Conservation and Assessment
of NOAA's National
Ocean Service. The project's specific objectives are to measure
concentrations of chemical
contaminants in sediment and in bottom-dwelling fish species at
selected sites in urban and
nonurban embayments, to determine the prevalences of pathological
conditions and other
bioindicators of contaminant exposure in these same fish species, and
to evaluate temporal trends
in the above-mentioned parameters. One hundred sites in embayments
along the Atlantic, Gulf
and Pacific coasts, including Alaska, have been sampled since 1984,
with each annual sampling
referred to as a "cycle" (e.g., 1984 = Cycle 1). Prior to Cycle 5,
the Northeast Fisheries Center
was responsible for Northeast Coast NBSP sampling, and results of
these studies are presented in
Zdanowicz et al. (1986) and NOAA (1987, 1988).
This Technical Memorandum presents and interprets the results of
histopathological studies
conducted on winter flounder (Pleuronectes americanus formerly
Pseudopleuronectes
americanus)
sampled from Northeast Coast sites between 1987 and 1989 as part of
the NS&T Program.
Pathological conditions described include a variety of idiopathic
liver, kidney, and fin lesions.
Some of these lesions, such as liver neoplasms and related lesions,
are well established as
histopathologic biomarkers of contaminant stress, while others have
shown some promise as
indicators but require additional testing. Prevalences of these
lesions in winter flounder sampled
from sites with different degrees of chemical contamination are
compared. In addition, lesion
prevalences are correlated with chemical exposure data to better
understand the relationship
between environmental contaminants and fish disease. This
memorandum is not meant to
comprehensively review the marine pollution literature; however,
pertinent references are
included in discussions of the most significant findings.
In addition to this publication, a separate Technical Memorandum is
in preparation to provide
detailed information on contaminant levels in winter flounder tissues
and in sediments at the
Northeast Coast sampling sites (Brown et al. in prep.). While data
on sediment and tissue
contaminant concentrations will be utilized in this pathology
memorandum, the chemistry report
should be consulted for more comprehensive information on tissue and
sediment chemistry for
Northeast Coast NBSP studies conducted during 1988 and 1989.
INTRODUCTION
The National Benthic Surveillance Project (NBSP), a component of the
National Oceanic and
Atmospheric Administration's (NOAA's) Status and Trends (NS&T)
Program, is designed to
assess the status of the Nation's coastal and estuarine waters and
monitor long-term changes in
marine environmental quality. A key component of the NBSP is the
utilization of histopathologic
alterations in fish tissues as indicators of the effects of exposure
to environmental contaminants on
marine organisms. These histopathologic biomarkers are pathological
conditions that either
morphologically resemble lesions induced by the experimental exposure
of mammals or fish to
toxicants, or show evidence of a contaminant-associated etiology in
field studies because they
occur either exclusively or at significantly increased prevalences in
fish from contaminated sites.
Probably the most commonly used histopathologic biomarkers are
neoplastic and preneoplastic
liver lesions, although recent studies indicate that certain
associated degenerative and regenerative
conditions may be equally or even more useful as biomarkers of
anthropogenic stress (Myers et al.
in press, Varanasi et al. in press). Liver cancer and related
lesions have been reported in several
species of bottomfish (e.g., English sole (Pleuronectes
vetulus, formerly Parophrys
vetulus)
(McCain et al. 1977; Malins et al. 1984, 1985; Myers et al. 1987),
white croaker
(Genyonemus
lineatus) (Malins et al. 1987; Myers et al. 1991, 1992); Atlantic
tomcod (Microgadus
tomcod)
(Smith et al. 1979; Cormier 1986)) from chemically contaminated
coastal areas. Associations
between concentrations of chemical contaminants in sediments and fish
tissues and certain of
these lesions have been observed in field surveys (Malins et al.
1984, 1985; McCain et. al. 1988;
Vogelbein et al. 1990), and have been further substantiated by
statistical analyses and
epizootiological models (Rhodes et al. 1987, Myers et al, 1990, 1991,
1992; Landahl et al. 1990).
Moreover, in some species, including English sole, cause-and-effect
relationships between
selected contaminants and hepatic lesions similar to those observed
in field-sampled animals have
been demonstrated in long-term laboratory studies (Varanasi et al.
1987, Schiewe et al. 1991).
Other pathological conditions, including fin erosion and renal
lesions, have also shown some
promise as bioindicators of contaminant exposure in fish (Dethlefsen
1980, Cross 1985, Wellings
et al. 1976, Sherwood and Mearns 1977, Cross et al. 1985, Sindermann
1990, Reimschuessel et
al. 1990, Hawkins et al. 1989, McCain et al. 1992), but require
additional testing in both the field
and the laboratory to confirm their chemical etiology.
Previous field studies suggest that liver lesions and other disease
conditions in winter flounder
(Pleuronectes americanus formerly Pseudopleuronectes
americanus) may be
reliable bioindicators
of environmental degradation on the Northeast Coast. The occurrence
of neoplastic and
non-neoplastic liver disease in fish from contaminated areas,
especially from sites within Boston
Harbor, is well-documented, both in earlier NBSP sampling conducted
by the Northeast Fisheries
Center (Zdanowicz et al. 1986, NOAA 1987, 1988) and in studies by
other investigators
(Murchelano and Wolke 1985, 1991; Moore 1991, Gronlund et al. 1991).
Elevated prevalences
of fin erosion have also been reported in flounder from contaminated
sites along the Northeast
Coast (Williams 1981, Ziskowski and Murchelano 1975). At present,
however, there is relatively
little information available on how levels of specific contaminants
in sediments and winter
flounder tissues are related to disease occurrence, or on the effects
of biological factors such as
age and gender on disease prevalence in this species.
This report presents the results of pathological studies conducted on
winter flounder collected
from Northeast Coast sites between 1987 and 1989 as part of the NS&T
Program. The major
objectives of the study were 1) to determine the prevalences of fin
erosion and selected hepatic
and renal lesions in winter flounder from the Northeast Coast
sampling sites; 2) to examine the
effects of biological variables such as age, gender, and sampling
season on lesion occurrence; and
3) to assess the relationship between levels of chemical contaminants
in sediments and fish and
lesion prevalences.
Sampling was primarily conducted during Cycles 5 and 6 (1988 and
1989) of the Northeast Coast
NBSP, but to provide a more comprehensive view of pathological
conditions in winter flounder
from the Northeast Coast, data from winter flounder collected in Long
Island Sound in 1987
(Gronlund et al. 1991) and in Boston Harbor and adjacent embayments
during the winters of 1988
and 1989 (Johnson et al. 1992, in press) are also included in this
report. Histopathological
examination of tissues and chemical analyses were carried out in our
laboratory in a uniform
fashion for all three studies, so all data included here are
comparable.
Lesion types monitored in winter flounder included fin erosion;
necrotic, sclerotic, and
proliferative lesions in the kidney; and a number of suspected
toxicopathic liver lesions such as
neoplasms, putatively preneoplastic foci of cellular alteration, and
nonneoplastic proliferative and
degenerative conditions. A variety of other pathological conditions,
such as parasitic infections,
were noted in flounder examined in this study, but they will not be
described in detail in this report
as there is little evidence that they have a chemical etiology
(Hinton et al. 1992).
The following classes of chemical contaminants were included in the
analysis as potential risk
factors : high molecular weight aromatic hydrocarbons (AHs)
containing 4-5 benzene rings
(HAHs), low molecular weight AHs containing 2-3 benzene rings (LAHs),
total AHs (LAHs +
HAHs), polychlorinated biphenyls (PCBs), DDT and its derivatives, DDD
and DDE (DDTs), and
chlordanes. These classes of chemical compounds were chosen because
1) they were present at
high concentrations in sediments from a number of Northeast Coast
urban sites; 2) they are
among the most prevalent contaminants in estuaries adjacent to
metropolitan areas throughout the
United States, and thus represent an index of urban pollution
(Varanasi et al. 1988, 1989a); and 3)
they are generally recognized as toxicants in fish or mammals
(Klaassen et al. 1986).
Concentrations of these organic contaminants or their metabolites
were measured in several
compartments (i.e., sediments, stomach contents, liver, and bile) to
give a comprehensive picture
of exposure and relationships between contaminant levels in all three
compartments and idiopathic
disease. However, contaminant levels in liver and bile were
considered to have the greatest
relevance as toxicologic risk factors, because they are most
representative of the actual uptake of
toxic compounds by fish. It should be noted that other types of
contaminants, such as dioxins or
organometals, may also play a role in the development of disease in
winter flounder and other fish
species. However, because methodologies for measuring concentrations
of these compounds in
fish and sediments that are cost-effective and appropriate for a
large-scale monitoring program are
not currently available, it was not possible to include them in the
NBSP at this time.
Trace metal analyses of sediments, liver tissue, and stomach contents
were conducted by the
Southeast Fisheries Science Center, and data on metal concentrations
will not be presented here.
Analyses of metals from previous NBSP samplings on both the West
Coast (Varanasi et al. 1988,
1989a) and the Northeast Coast (NOAA 1988) indicated that there was
little relationship between
tissue concentrations of metals and concentrations of metals in
sediment in any of the fish species
examined. These findings suggest that metals may not be bioavailable
to fish, or that their uptake
and accumulation are governed by complex processes that make tissue
concentrations of metals a
poor indicator of environmental exposure. Consequently, no attempt
was made to correlate metal
concentrations with fish disease.
Logistic regression (Breslow and Day 1980, Schlesselman 1982) was
used to assess the influence
of biological and contaminant-associated risk factors on lesion
occurrence. The logistic function
is considered a basic model for dose-response relationships and
logistic regression analysis is
frequently applied in epidemiological and epizootiological studies.
This technique is similar to
commonly-used stepwise multiple regression procedures, but is
uniquely suited to binomial or
proportional data (e.g., lesion prevalence or lesion presence or
absence). Results of these
analyses and their relationship to other studies of
pollution-associated disease in winter flounder
will be discussed below.
Table of Contents
