DOC
EPA
United States
Department of
Commerce
National Oceanic and
Atmospheric Administration
Seattle WA 98112
United States
Environmental Protection
Agency
Office of Environmental Processes and
Effects Research
Washington DC 20460
EPA-600/7-82-001
February 1982
Research and Development
Pathology of Two
Species of Flatfish
From Urban
Estuaries in
Puget Sound
Interagency
Energy/Environment
R&D Program
Report
-------
EPA-600/7-82-001
February 1982
PATHOLOGY OF TWO SPECIES OF FLATFISH
FROM URBAN ESTUARIES IN PUGET SOUND
by
Bruce B. McCain, Mark S. Myers, Usha Varanasi,
Donald W. Brown, Linda D. Rhodes, William D. Gronlund, Diane G. Elliott
Wayne A. Palsson, Harold 0. Hodgins and Donald C. Mai ins
Environmental Conservation Division
Northwest and Alaska Fisheries Center
National Marine Fisheries Service
National Oceanic and Atmospheric Administration
2725 Montiake Boulevard East
Seattle, Washington 98112
NOAA Project Officer: Douglas A. Wolfe (NOAA/Boulder, CO)
This is a final report of research conducted as part of the
Federal Interagency Energy/Environment Research and Development Program
Prepared for
OFFICE OF ENERGY, MINERALS, AND INDUSTRY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, D.C. 20460
-------
Prepared for the NOAA Energy Resources Project #3
in partial fulfillment of the
Environmental Interagency Agreement #EPA-AIG-E693
Work Unit #3-3-6
DISCLAIMER
This work is the result of research sponsored by the Environmental
Protection Agency and administered by the Environmental Research Laboratories
of the National Oceanic and Atmospheric Administration.
The National Oceanic and Atmospheric Administration (NOAA) does not
approve or endorse any proprietary product or proprietary material nentioned
in this publication. No reference shall be made to NOAA or to this
publication in any advertisin-g or sales promotion which would indicate or
imply that NOAA approves, recommends, or endorses any proprietary product
or proprietary material mentioned herein.
This report has been reviewed by the Office of Research and Development,
U.S. Environmental Protection Agency, and approved for publication.
Approval does not signify that the contents necessarily reflect the views
and policies of the U.S. Environmental Protection Agency, nor does mention
of trade names or commercial products constitute endorsement or
recommendation for use.
-------
FOREWORD
Hunan activities in or near coastal areas create a diverse variety
of waste materials and by-products, which may eventually reach the marine
environment and have negative impacts on marine resource values. In order
to develop an adequate understanding of the potential marine environmental
consequences of such pollution, the National Oceanic and Atmospheric
Administration has conducted studies on the Fate and Effects of Petroleum
Hydrocarbons and Toxic iietals in Selected Marine Organisms and Ecosystems
under Interayency Agreement with the Environmental Protection Agency.
The overall objectives of this effort have been to study experimentally
those specific processes underlying the distribution, transport, and
biological effects of petroleum hydrocarbons in coastal marine ecosystems.
The results are expected to facilitate the assessment of impacts of
contaminant releases, and thereby to improve the basis for developing
regulatory (Measures for suitable protection of the marine environment.
The effects of chemical pollutants on the ecosystem and animal
populations in oceans and estuaries are of concern, with respect to loss
of marine animals and to possible harmful effects on man as a result of
eating contaminated food or by direct contact with water. One means
of determining the environmental factors which may lead to carcinogenic
or toxic effects on marine animals is to investigate sensitive marine
animals for pathological conditions which are associated with polluted
areas. The report that follows describes disease prevalences in flatfish
from relatively contaminated and uncontaminated estuarine systems in
Puget Sound, along with results of experiments designed to test cause/
effect relations between sediment-associated chemicals and the observed
pathological abnormalities.
Douglas A. Wolfe
Office of Marine Pollution Assessment
NOAA
11
-------
ABSTRACT
Described in this report are the results of field and laboratory
investigations conducted between October 1978 and October 1980. The
field studies yielded data on the prevalence and geographical distribution
of diseased bottom-dwelling flatfish, specifically English sole (Parophyrys
vetulus) and starry flounder (PI at 1 cht hys stel1 at us), in the Duwamish
Waterway, Seattle, and other selected estuaries in Puget Sound. Levels
of toxic metals, chlorinated hydrocarbons, and polynuclear aromatic
hydrocarbons (AHs) in fish and botton sediments from these estuaries
were also determined. The laboratory studies were conducted in order
to evaluate cause-and-effect relationships between sediment-associated
chemicals and the abnormalities observed in English sole.
A variety of lesions were observed in the liver, kidney, gills,
gastrointestinal tract, and in the skin and fins of both species from
the urban estuaries. In addition to the Duwamish Waterway, these urban
estuaries included the Lake Washington Ship Canal, Seattle, and
the Snohomish River, Everett. Fish with some of these lesions were
also found in McAllister Creek, the reference estuary, near Olympia.
Fourteen principal types of idiopathic liver lesions were observed:
eight of the lesions were non-neoplastic, two were classified as
potentially preneoplastic and four were clearly neoplastic. English
sole and starry flounder with the latter two types of lesions were
found only in the Duwarm'sh Waterway and the Lake Washington Ship Canal.
The only liver lesions found in flatfish from the Snohomish River and
McAllister Creek were non-neoplastic lesions: nonspecific hepatocellular
necrosis, fatty change (an abnormal accumulation of lipid in hepatocytes),
hernosiderosis (an intracellular accumulation of iron-containing pigments),
and intrahepatic blood cysts. The prevalence of all these idiopathic
liver lesions was very low in the latter two estuaries.
The prevalences of English sole with either "preneoplastic" or
neoplastic lesions, or both, were 20.5% (113 of 551 fish) and 20.4%
(10 of 49 fish) for the Duwamish Waterway and Lake Washington Ship Canal,
respectively. The prevalence of English sole with liver neoplasms
in these two estuaries was 12.9% (71 of 551 fish) and 8.4% (4 of 49
fish), respectively. In the Duwamish Waterway, the prevalence of starry
flounder with liver neoplasms was 1.1% (3 of 279 fish), considerably
lower than that of English sole. (This is the first report of liver
neoplasms in starry flounder.)
The types of lesions found in other organ systems were either found
at very low prevalence or the geographical distribution of animals with
these lesions was such that clear-cut associations with urban areas
were tenuous. The four most common idiopathic lesions of the gills were
respiratory epithelial hyperplasia (a thickening of the respiratory
epithelium), lymphocytic infiltration (a chronic inflammatory response
-------
to injury), rricroaneurysns (small dilations of blood vessels, probably
a shock response induced by capture), and mucous cell hyperplasia (an
increase in the density of goblet cells in gill lamellae and filaments).
The two most prevalent kidney lesions were mesangiolysis and mesangio-
sclerusis, both being primary lesions of the gloneruli. Pathological
conditions of the gastrointestional tract generally were associated
with parasitic infestations, and affected animals were usually found
in all the estuaries. Lesions of the skin and fins included such
grossly visible conditions as parasitic infestation (neinatodes and
trematodes), lymphocystis (a virus-caused condition), skin tumors
(growths which may be caused by a parasitic amoeba), and fin erosion.
Concentrations of polychlorinated biphenyls (PCBs) in liver tissues
of both English sole and starry flounder reflected the concentrations
of these compounds in sediments from which the animals were captured.
In the Duwamish Waterway, the liver tissue levels of PCBs in English
sole ranged from 17.0 to 161.0 yg/g (dry weight), compared to 0.5 vg/g
(dry weight) in a composite of English sole livers from McAllister
Creek. Low concentrations of most AHs were detected in the livers of
flatfish from both the urban and reference estuaries. This observation
of concentrations of AHs in fish in the urban estuaries was almost
surely a reflection of the we! 1-documented ability of flatfish to
rapidly metabolize AHs to compounds not detectable by routine analytical
procedures. Nevertheless, one AH, naphthalene, was frequently detected
in liver tissue of English sole from the Uuwamish Waterway, but not in
sole from McAllister Creek, the reference estuary.
Three types of laboratory experiments were performed to evaluate
if English sole were adversely affected by exposure to contaminated
bottom sediments. In one type of experiment, sole were maintained
for up to 3 months in aquaria containing either sediment from the
Duwamish Waterway or from a minimally contaminated area. The mortality
rate of sole exposed to sediment from the Uuwamish Waterway was not
statistically different from that of sole on reference sediments.
Another type of experiment involved intraperitoneal injection of juvenile
English sole with extracts of contaminated or reference sediments and
observing the fish for possible effects for up to 3 weeks. Extracts
of sediment from the Uuwamish Waterway were more acutely toxic than
were extracts of reference sediment; significantly (p_<0.05) more fish
died when injected with extracts of Duwamish Waterway sediment than did
fish receiving reference extracts, carrier (corn oil) only, or no injection,
None of the above-mentioned major histopathological conditions observed in
English sole from the Uuwamish Waterway were observed in experimental
fish. The third type of laboratory experiment was intended to define the
uptake from sediment and fate of a carcinogenic polynuclear AH,
benzo[a]pyrene (BaP), by English sole. English sole were exposed for
24 hours to reference sediment contaminated with radioactively-labeled
BaP; concentrations of BaP in the blood, liver, muscle, and gall
bladder were measured. The highest levels of BaP and its metabolites
were detected in bile and in liver tissue.
VI
-------
This study constitutes an important step in further defining the
pathologic anatomy and geographical distribution of several urban-associated
abnormalities in two flatfish species which are widely distributed in
Puget Sound. Although laboratory experiments were initiated in an
attempt to elucidate the causes of these abnormalities, the causes
remain unknown. These experiments do provide evidence, however, that
sediments from the Duwarni sh Waterway contain factors which can be
toxic to English sole.
VI 1
-------
CONTENTS
FOREWORD iii
ABSTRACT V
FIGURES xii
TABLES xv
ACKNOWLEDGMENT xvi
1. INTRODUCTION 1
2. CONCLUSIONS 2
3. RECOMMENDATIONS 4
4. EXPERIMENTATION 5
METHODS AND MATERIALS 5
Field Studies 5
Sampling Areas and Sampling Stations 5
Exanination of Fish and Necropsy Procedures 5
Histological Procedures 9
Histopathological Procedures 9
Statistical Methods 9
Chemical Analyses 10
Laboratory Studies 10
Care of Experimental Fish 11
Exposures to Sediment Extracts 11
Preparation of extracts of sediment and
interstitial water 11
Exposure of fish to sediment extracts 11
Exposures to Bottom Sediments 14
Statistical Analysis of Effects Data 15
Exposure of English Sole to Aromatic Hydrocarbons
in Sediment 15
Analysis of sediment and sediment-associated
water 15
Analysis of tissues . 15
IX
-------
Page
RESULTS AMD DISCUSSION 16
Field Studies 16
Pathological Conditions of English Sole 16
Lesions of the skin and fins 16
Parasitic lesions 16
Viral lesions 17
Skin tumors 17
Fin erosion 17
Miscellaneous skin lesions 21
Liver lesions 21
Megalocytic hepatosis (MH) 22
Hepatocel lular regeneration (HR) 22
Hepatocellular eosinophilic hypertrophy (HEH) . 24
Minimum deviation nodules (MDN) ........ 26
Liver cell adenoma (LCA) 26
Hepatocellular carcinoma (HC) 29
Cholangiocel lular carcinoma (CC) 31
Cholangiofibrosis (CF) 33
Hernangioma 33
Hepatocellular necrosis (HN) 34
Miscellaneous non-parasitic liver lesions ... 34
Parasitic hepatic lesions 34
Prevalence of liver lesions 35
Interrelationships among idiopathic liver
lesions 38
Analysis of idiopathic hepatic lesions in
Duwamish River English sole with respect
to sex 41
Length-age regressions in Duwamish River
English sole with and without idiopathic
liver lesions 43
Prevalence of hepatic lesions related to length
in English sole and effect of hepatic lesion
presence on the length-weight relationship in
Duwamish River English sole . 45
Characteristics of gill lesions 46
Respiratory epithelial hyperplasia (REH). ... 46
Lymphocytic infiltration 46
Microaneurysm 49
Miscellaneous gill lesions 49
Parasitic conditions 49
-------
Prevalence of gill lesions 50
Characteristics and prevalence of cardiac
abnormalities .52
Characteristics and prevalence of
gastrointestinal abnormalities 52
Characteristics and Prevalence of
renal abnormalities 54
Mesangiolysis 54
Mesangiosclerosis 56
Miscellaneous idiopathic renal lesions .... 56
Association of idiopathic
kidney lesions with idiopathic liver lesions
in English sole 58
Parasitic conditions in the kidney. 58
Pathological conditions of Starry Flounder -58
Characteristics and prevalence of fin and skin
conditions 58
Characteristics and prevalence of gill anomalies. . 61
Characteristics and prevalence of liver lesions . . 61
Characteristics and prevalence of gastrointestinal
abnormalities 63
Characteristics and prevalence of renal
abnormalities 63
Ecology of English Sole and Starry Flounder 65
Abundance 65
Biological characteristics 69
Chemical Analyses 73
Sediment chemistry 73
Tissue chemistry 77
Interrelationships between chemistry and
pathology 77
Laboratory Studies 80
Exposures to Sediment Extracts 80
Exposures to Bottom Sediments 83
Benzo[a]pyrene Bioavailability Experiment 85
APPENDIX I 89
REFERENCES 91
XI
-------
FIGURES
Number Page
1 General map showing the four field sampling areas
in Puget Sound 6
2 Detailed map of field sampling stations in the
Duwarnish Waterway system 7
3 Micrograph of lymphocystis skin lesion in an
English sole IB
4 Electronmicrograph of intracytopl asrm'c lymphocystis
virus particles 19
5 Electron nicrograph of intracytoplasmic lymphocystis
virus particles and fibrillar structures showing
banding periodicity 20
6 Micrograph of the hepatic lesions megalocytic hepatosis
and hyperplastic hepatocellul ar regeneration in English
sole 23
7 Micrograph of nodular hepatocellular eosinophilic
hypertrophy in English sole 25
8 Micrograph of the hepatic neoplastic lesion termed a
minimum deviation nodule in English sole 27
9 Micrograph of a liver cell adenoma in English sole. ... 28
10 Micrograph of an hepatocellular carcinoma in English
sole 30
11 Micrograph of a metastasizing cholangiocellular carcinoma
in English sole 32
12 Prevalences of idiopathic liver lesions in English sole
from the Duwarnish Waterway, Lake Washington Ship Canal,
Snohonish River, and McAllister Creek sampling areas. . . 36
13 Prevalences (by station) of idiopathic liver lesions
in English sole from the Duwanish Waterway 37
14 Prevalences of English sole with suspected "preneoplastic"
liver lesions, liver neoplasms, and combined
preneoplastic/neoplastic liver lesions from the Duwamish
Waterway and Lake Washington Ship Canal 39
-------
Number Page
15 Length-age regression curves and mean length-at-age
values for Duwarni sh River English sole with and without
idiopathic hepatic lesions. 44
16 Length-weight regression curves for Duwanish River
English sole with and without idiopathic hepatic
lesions 47
17 Hicrograph of an English sole gill with severe
respiratory epithelial hyperplasia with fusion of
adjacent lamellae 48
18 Prevalences of gill and kidney lesions in English sole
from the Duwarnish Waterway, Lake Washington Ship Canal,
Snohomish River, and McAllister Creek sampling areas. . . 51
19 Micrograph of the renal lesion, mesangiolysis, in an
English sole 55
20 Micrograph of the renal lesion, mesangiosclerosis, in
an English sole 57
21 Hicrograph of severe thickening of the peripheral basal
lamina in the glomerular tuft in the kidney of an English
sole 59
22 Prevalences of gill and kidney lesions in starry flounder
from the Duwamish Waterway, Snohomish River, and
McAllister Creek sampling areas 62
23 Prevalences of idiopathic hepatic lesions in starry
flounder fron the Duwamish Waterway and McAllister Creek
sampling areas 64
24 Catch Per Unit Effort (CPUE) values for English sole fron
the Duwam'sh Waterway, shown by station and sampling
period 66
25 Catch Per Unit Effort (CPUE) values for starry flounder
from the Duwamish Waterway, shown by station and sampling
period 67
26 Catch Per Unit Effort (CPUE) values for English sole and
starry flounder from the Lake Washington Ship Canal (LWSC),
Snohomish River, and McAllister Creek sampling areas,
shown by sampling period 68
-------
Dumber Page
27 Length-frequency distributions of English sole from
the Ouwairtish Waterway (expressed as a percentage of
the catch), shown by sampling period 70
28 Length-frequency distributions of English sole from
the Lake Washington Ship Canal, Snohomish River, and
McAllister Creek sampling areas (expressed as a
percentage of the catch), shown by station and
sampling period 71
29 Length-frequency distributions of starry flounder
(expressed as a percentage of the catch) from the
Duwanish Waterway and McAllister Creek sampling areas,
shown by sampling period 72
30 Concentrations of toxic chemicals in liver tissue from
normal-appearing English sole from the Duwamish Waterway,
Lake Washington Ship Canal, and McAllister Creek .... 78
31 Tissue concentration of naphthalene (NPH) and z
polychlorinated biphenyls (PCBs) in English sole with
various liver lesions 79
xiv
-------
TABLES
Number Pa ge
1 English sole and starry flounder which were examined
for externally visible and microscopic abnorrnalites
(listed by sampling area and sampling period) 8
2 Lengths and weights of English sole used in laboratory
experiments 12
3 Concentrations of selected compounds in sediment from the
Duwami sh Waterway, the Snohornish River, and Port Madison
used in laboratory experiments 13
4 Temporal variation in the prevalence of English sole
with selected liver lesions in the Duwamish Waterway. . . 40
5 Matrix of idiopathic liver lesions which were compared
to determine the degree of association of lesions by
chi-square contingency analysis in individual English
sole from the Uuwamish Waterway system 42
6 Lesions in selected organs and tissues of English sole. . 53
7 Lesions in selected organs and tissues of starry
flounder 60
8 Prevalences of various length groups of English sole and
starry flounder with liver lesions 74
9 Age/length/weight characteristics of English sole with
various liver lesions 75
10 Concentrations of chemicals in sediment at sampling
stations 76
11 Effects on English sole of injections of extracts of
bottom sediments 81
12 Mortality after exposure of English sole to bottom
sediments 84
13 Concentrations of benzo[a]pyrene (BaP)-associated radio-
activity in sediment, sediment-associated water (SAW) and
tissues of English sole, and the bioconcentration factors
for the tissues relative to BaP concentrations in the
sediment and SAW 86
14 Proportions of the metabolites of benzo[a]pyrene (BaP)
in bile of English sole exposed to contaminated
sediment 87
xv
-------
ACKNOWLEDGMENT
We wish to acknowledge several scientists and research associates
who helped to complete the research reported here, namely, Trish Emry
and Warren Ames (fish pathology), Ton Horn (benzo[a]pyrene uptake
experiments), Doug Burroughs, Don Gennero, Patty Prohaska, Orlando
Maynes, Robert Clark, and Andrew Friedman (chemistry), and Gary Walters
and Dr. Bruce Miller (fish ecology). We also give our appreciation
to those persons involved in the production of this report, namely,
Carol Hastings, Jim Peacock, Melody Tune, Annette Hodyson, Marylyn West,
and Gerald Williams. Finally, we would like to thank Dr. Sin-Lam Chan
for reviewing this report.
xvi
-------
SECTION 1
INTRODUCTION
Pathological conditions have been reported in demersal fishes in
the Duwamish Waterway: (1) liver neoplasms in English sole (Parophrys
vetulus) (McCain et al. 1977, Pierce et al. 1978); (2) skin papillornas of
English sole and starry flounder (Platichthys stellatus) (Wingert et al.
1976, Wei lings et al. 1976b); (3) fin erosion of starry flounder and
English sole (Wei lings et al. 1976a); and (4) non-neoplastic liver
abnornalities (including severe hepatocellular lipid vacuolization,
necrosis and loss of liver structure) in 90% of both the starry flounder
and English sole (Pierce et al. 1978, Wingert et al. 1976). The Duwai;iish
estuary receives agricultural, industrial, and municipal wastes. Its
sediments are known to contain carcinogenic aromatic hydrocarbons (AHs),
such as benzo[a]pyrene (BaP) and benzanthracene (Halins et al. 1980),
heavy metals, arid high levels of polychl orinated biphenyls (PCBs) (Pavlou
et al. 1973, Pattie 1975, Sherwood and McCain 1976, Malins et al. 1980).
Although at present there is no unequivocal evidence that the above-
mentioned pathological conditions are caused by xenobiotic chemicals,
microbiological and histological evidence suggests that infectious
microorganisms are not the direct cause. Thus, environmental pollutants
seem to be the most likely etiological agents.
Field studies were performed over a 2-year period in the Duwamish
River estuary (Duwamish Waterway), Lake Washington Ship Canal, Snohorm'sh
River estuary (urban-associated areas), and in a reference area, McAllister
Creek to (1) describe the types and prevalences of histologically
detectable pathologic conditions in English sole and starry flounder
captured in these areas, (2) define the ecology of these species in the
respective sampling areas, and (3) investigate the potential relationship
of these pathologic conditions to levels of xenobiotic chemicals present
in the sediments from these areas and/or in tissues from affected fish.
Three types of laboratory studies were also conducted to evaluate
if English sole were adversely affected by exposure to contaminated
sediments. The long-term effects of such exposure were tested by
maintaining fish on sediment from the Duwamish Waterway for up to 3
nonths. The acutely toxic effects of components of sediment from this
waterway were tested by injecting sole with extracts of fractions of
sediment. The third type of study was intended to define the relative
uptake of BaP from contaminated sediments by English sole.
-------
SECTION 2
CONCLUSIONS
1. English sole and starry flounder with certain types of lesions
of the liver, gills, kidney, and skin were found primarily in urban
estuaries of Puget Sound, the lesions of the liver included neoplastic,
"preneoplastic", and a variety of non-neoplastic lesions (e.g., negalocytic
hepatosis, hepatocellular regeneration and hyperplasia, and
cholangiofibrosis). These idiopathic liver lesions were found
exclusively in fish from the Duwamish Waterway and the Lake Washington
Ship Canal, both located in Seattle. Flatfish with a gill lesion,
(respiratory epithelial hyperplasia), a kidney lesion, (peritubular
fibrosis) and a skin lesion, (epidermal necrosis), were found in these
two estuaries as well as in the estuary of Snohomish River in Everett.
Most idiopathic lesions in the liver were consistently found only in
English sole and starry flounder from the urban-associated areas.
2. In general, lesions in organs other than the liver failed to
show higher prevalences in the urban-associated estuaries compared to the
reference estuary (McAllister Creek). Individuals of the two flatfish
species with certain lesions of the liver (other than above), gills,
and kidney, as well as lesions of other tissues (skin, gastrointestinal
tract), were found in the urban estuaries as well as in the reference
estuary. The widely occurring lesions of the liver were fatty change
and hepatocellular necrosis in both species, and hemosiderosis in
English sole and blood cysts in starry flounder. English sole with
the gill lesion, respiratory epithelial hyperplasia, were found only
in urban estuaries; however, starry flounder with this lesion were
equally or more prevalent in the reference estuary. Widely observed
abnormalities of the skin and gastrointestinal tract were chronic
dermal necrosis and parasitic infections, respectively.
3. Most of the lesions observed in this study probably were the
result of long-term exposure to one or more environmental factors.
The length of exposure time is suggested by the average age of English
sole with liver lesions, 4 to 6 years. However, certain idiopathic
liver lesions were detected in fish as young as one year of age.
4. The prevalence of English sole with two types of liver lesions,
rnegalocytic hepatosis and hepatocel lular eosinophilic hypertrophy, was
significantly higher (p<0.05) in January, 1980 (24.3 and 14.8%,
respectively) than in January, 1979 (14.8 and 9.3%, respectively).
Whether this increase represents a seasonal fluctuation in prevalence
or an overall increase in prevalence, suggesting a worsening of the
health of English sole, is not presently known.
-------
5. The abundance of English sole and starry flounder in the
Duwainish Waterway and McAllister Creek was seasonally variable and,
in the case of the Waterway, varied according to the distance of the
sanpling stations from the mouth of the Waterway. The abundance of
English sole in both estuaries was lowest during the spring, and in
the Waterway the abundance was similar during the other three seasons.
For starry flounder, the abundance in the Duwamish Waterway was lowest
during the initial fall, winter, and spring sample collections, and 4
to 14 times higher in the summer and succeeding winter collections,
respectively. In the Waterway, English sole tended to be most abundant
near the mouth during the winter, when flow was highest, and near the
turning-basin (the upper portion of the sampling area) during the
summer and fall, when flow was lowest. A similar trend was observed
for starry flounder, except that, since this species prefers or tolerates
relatively low-salinity water, they were generally not collected at
stations at the mouth of the Waterway.
6. The concentrations of AHs and metals in sediment samples from
the Lake Washington Ship Canal and the Duwainish Waterway were similar;
however, the concentration of PCBs in sediment from the Lake Washington
Ship Canal was considerably lower than in sediment from the Ouwamish
Waterway. None of the AHs or PCBs measured in sediments from these
two urban estuaries were detected in sediment from the reference estuary,
McAllister Creek, while sediments from the Snohornish River had levels
of AHs and PCBs intermediate between the two types of estuaries.
Concentrations of toxic metals (cadmiun, mercury, chromium, lead, and
copper) were generally considerably higher in sediment from Duwainish
Waterway and Lake Washington Ship Canal than from McAllister Creek.
With the exception of cadmium, which was detected only in these two
urban estuaries, the levels of the other netals were 2 to 20 times
lower in the reference estuary.
7. The concentrations of PCBs and metals in tissues of flatfish
reflected the levels found in sediment samples collected from areas in
which the fish were captured. Although most AHs were not detected in
liver tissue of fish from any of the sampling areas, low concentrations
of naphthalene were routinely found in tissues of fish from the Lake
Washington Ship Canal and Duwainish Waterway. The almost complete
absence of AHs in flatfish tissue was almost surely due to the ability
of flatfish to metabolize AHs. The detection of naphthalene is consistent
with laboratory findings that flatfish do not metabolize this compound
as rapidly as higher molecular weight AHs.
8. The laboratory studies in which English sole were exposed to
chemical contaminants found in sediments from the Duwamish Waterway
did not induce pathological conditions observed in English sole from
the Duwamish Waterway. However, the experiments did demonstrate that
extracts of sediments from the Waterway have certain components that
are toxic to sole that were not present in reference sediments. Other
experiments clearly showed that BaP, a carcinogenic AH present in sediment
from the Duwamish Waterway, was readily taken up directly from contaminated
sediment by English sole.
-------
SECTION 3
RECOMMENDATIONS
The results of the field survey portion of this study demonstrate
that any comprehensive monitoring program intended to evaluate the
health of aquatic animals should include histopathology as well as
other scientific disciplines, such as analytical chemistry and ecology.
The histopathological procedures and diagnostic criteria should also
incorporate the standards employed by veterinary pathologists and
toxi cologists.
Additional research efforts need to be expended on elucidating
the causes of the lesions observed in flatfish in the Seattle waterways.
The laboratory studies reported here provide preliminary evidence that
sediment in the Duwamish Waterway contains factors which are toxic to
English sole. Further experimentation utilizing both controlled laboratory
exposures and semi-controlled field experiments will be needed to
verify these preliminary observations and to characterize the nature
of the etiological factors. The laboratory experiments could involve
long-term exposures of juvenile flatfish to various fractions, of sediment
extracts, administered orally or by injection. Field experiments
could employ long-term in situ exposures of flatfish (held in cages)
to bottom sediment in contaminated and uncontairrinated estuaries.
An important question left unanswered by this study was the effect
of the observed diseases in English sole and starry flounder on the
survival of these two species. In order to provide further information
for answering this basic, key question, a complex ecological investigation,
including tagging studies, should be performed. Alternatively or
additionally, flatfish could be captured from selected areas, biopsied
and pathological analysis of liver tissue performed, and transferred
to holding facilities in the laboratory where survivial could be monitored.
-------
SECTION 4
EXPERIMENTATION
METHODS AND MATERIALS
FIELD STUDIES
Sampling Areas and Sampling Stations
The four areas of Puget Sound which were sampled in this study were:
(1) the Duwamish Waterway in Seattle, Washington (seven stations, A-G,
Figs. 1 and 2); (2) the Lake Washington Ship Canal in Seattle, Washington
below the Hiram M. Chittenden Locks (one station, Fig. 1); (3) the
Snohonish River estuary near Everett, Washington (two stations, Fig. 1);
and (4) McAllister Creek which enters into the Nisqually delta about
twenty miles south of Tacoma, Washington (two stations, Fig. 1). Since
the first three areas sampled are within or close to urban industrial
centers, they are termed "urban-associated estuaries". The fourth area
served as a reference area due to its distance from urban and/or industrial
influences.
Target fish species (English sole and starry flounder) were collected
by an otter trawl with the following specifications: 7.5 m opening,
10.8 m total length, 3.8 cm mesh in the cod end. Individual trawls at
each station were of 5 minutes duration at approximately 2.5 knots,
sweeping approximately 0.2 nautical miles. Sediments were collected
with a modified Van Veen sediment grab (Word 1976). Temperature,
salinity and dissolved oxygen measurements at or near the bottom at
each sampling station were taken with a Hydro-lab.1 The seasons
and areas sampled during this study are described in Table 1.
Examination of Fish and Necropsy Procedures
At each station, after collection by otter trawl and separation by
species, the two target species were placed in holding tanks containing
fresh seawater. All fish were examined for externally visible abnormalities.
The final composition of the necropsy subsample for each station consisted
of up to 30 fish of each species. To gain a broad size distribution
of fish within the subsamples, the intent was to select randomly equal
numbers of fish for necropsy from three length groups: <10 cm, 10-20 cm,
and >20 cm.
Fish to be necropsied were measured for total length (to the nearest
mm), weighed (to the nearest g), and assigned an individual field
number. The spinal cord was then severed and otoliths were collected
from each fish for aging purposes. The gonads were examined and the
Use of brand or trade names of materials used in this study does
not constitute official endorsement by the National Marine Fisheries
Service.
-------
Everett
Snohomish
River
H Lake
Washington
Seattle ~$L, Ship Canal
Duwamish
River
. . -^McAllister
Olymp.a Creek
FIGURE 1. General nap showing the four field sampling areas in
Puget Sound (Snohomish River, Lake Washington Ship Canal,
Duwamish River Waterway and McAllister Creek).
-------
West waterway
Harbor
Island
East waterway
Duwamish waterway
SOUTH SEATTLE
,t£ 1st Ave. S. bridge
'^Dtlfe.s
y;14th Ave. S. bridge
-$^
1 km
FIGURE 2. Detailed nap of field sampling stations in the Duwamish
Waterway system.
-------
TABLE 1. English sole and starry flounder which were examined for externally visible and microscopic
abnormalities.
Number of Fish Examined For Each Sampling
Estuary
Duwamish
Waterway
Lake Washington
Ship Canal
Snohorni sh
Ri ver
McAllister
Creek
Species3
ES
SF
ES
SF
ES
SF
ES
SF
Oct. 1978
264(168)b
42 (27)
NSC
24(24)
1 (1)
4b(18)
26(16)
Jan. 1979
178(93)
52(51)
50(31)
2 (2)
NS
NS
Apr. 1979 July 1979
127(95) 311(121)
90(90) 76 (26)
NS 44(17)
0
19(19) NS
(8)(8)
22(19) NS
41(26)
Jan. 1980
209 (94)
187(103)
NS
NS
NS
a ES, English sole; SF, starry flounder.
b Number in parenthesis is the number of fish which were examined for external abnormalities,
necropsied, and examined for histopathology; numbers not in parenthesis were examined only for
externally visible abnormalities.
c NS, not sampled.
-------
sex was determined. Tissues were excised from the following organs:
liver, spleen, gastrointestinal tract, gall bladder, gills, kidney,
heart, gonads, skin, and fins. (Skin and fins were collected only when
gross lesions were present.) These tissues were placed in cassettes
individually labeled with the appropriate field number and placed
immediately in Dietrich's fixative (Malins et al. 1980). When a
grossly visible lesion was observed in any organ, a portion of that
tissue with the surrounding normal-appearing tissue was excised and
fixed with the other organs described above.
Histological Procedures
Tissue specimens fixed in Dietrich's were removed from the fixative
after a minimum of 48-hours fixation time, processed automatically with
an Autotechnicon tissue processor, embedded in paraffin with a Tissue
Tek embedding center, and sectioned at 5 microns (Preece 1972). Tissues
such as those of gills and fins were decalcified following fixation,
using a commercial decalcification solution (Scientific Products, Redmond,
WA) prior to tissue processing with the Autotechnicon. Sections were
routinely stained with Mayer's hematoxylin and eosin-phloxine (Luna
1968). All stained tissues from each fish were examined by light
microscopy, and descriptions of any observed lesions and/or parasites
were recorded. When necessary for further characterization of lesions,
additional sections were cut and stained with special stains (Thompson
1966, Luna 1968, Preece 1972).
Histopathological Procedures
All slides were examined employing a "blind" system. As mentioned
previously, each fish was first assigned a field number prior to necropsy.
After processing, embedding, and sectioning were completed, and prior to
microscopic examination, each fish was assigned an identification number
derived from a random number table. Therefore, examining histopathologists
were given information on the species, length, weight, sex, and presence
of gross pathology, but did not know the area or station of capture.
Sections with microscopic lesions were classified using standard
nomenclature.
Statistical Methods
The G-statistic was used to test the null hypothesis that the fish
lesion frequency within a given area or at a particular station did not
differ from the average lesion frequency (for the particular lesion
being analyzed) for all areas sampled. In other cases, the G-statistic
was also employed to assess the statistical significance of lesion
frequency variations between areas and different seasonal samplings.
Since the data consisted usually of two categories (fish with lesions
vs. normal fish, or two lesion frequencies being compared), the Yates
correction for continuity (Sokal and Rohlf 1969) was used. This method
takes into account both the number of fish examined and the number of
lesions observed.
-------
Chi-square tests were also employed in evaluating the independence
of two variables (Zar 1974). This analytical technique was useful in
assessing the degree of association between specific pathologic conditions
in a given species of fish.
Chemical Analyses
Samples of liver tissue from English sole and starry flounder,
and of bottom sediments were analyzed for AHs, PCBs, and metals. In
addition, liver tissues were analyzed for percent lipid.
Samples of liver tissue, collected as part of the necropsy procedure,
were placed in methylene chloride-rinsed glass vials, immediately frozen,
and kept in frozen storage until they were analyzed. The analytical
procedures for organic compounds involved tissue homogenization, tissue
extraction with organic solvents, and fractionation of the extracts
using column chrornatography. Appropriate fractions were analyzed
using glass capillary gas chrornatography employing flame ionization,
electron capture, or mass spectrometer detectors (Brown et al. 1980,
Ramos and Prohaska 1981). Analyses for metals were performed by plasma
emission spectroscopy as in Malins et al. (1980).
Five sediment sample cores (5.5 cm diameter x 2 cm height) were
collected from each of two grab samples at each sampling station. The
ten cores of sediment were thoroughly mixed in a Teflon beaker with a
Teflon stirring rod. Portions of the mixed sediment were placed in a
glass bottle for analysis of organic compounds and in a plastic container
for metals analysis. These samples were frozen in the field and kept
frozen until they were thawed for analysis. Analyses for metals were
performed in essentially the same manner as for liver tissue. Sediment
samples were also analyzed for total organic carbon and grain size
distribution. Sediment grain size was measured by screening samples
of sediment through a series of standard sieves according to methods
reported by Krunbein and Pettijohn (1938). Total organic carbon was
measured using a LICO (manufacturer) induction furnace.
LABORATORY STUDIES
Several laboratory experiments were performed to attempt to define
the possible relationship between pollutants present in the Duwamish
waterway and various lesions observed in English sole from this area.
In some tests, sole were held in aquaria containing sediment from
either the Duwamish Waterway or two relatively unpolluted areas, Port
Madison and the Snohomish River. Sole were also exposed to extracts
of sediment and interstitial water from the Duwamish Waterway and Port
Madison by intraperitoneal (i.p.) injection. As a means of evaluating
if the aromatic hydrocarbons present in the sediment extracts were
bioavailable to sole by natural means, sole were exposed to reference
sediment contaminated with radiolabeled BaP. The uptake and fate of
BaP was determined.
10
-------
Care of Experimental Fish
English sole collected by beach seine or bottom trawl from non-urban
areas of Puget Sound (principally Mutiny Bay, Whidbey Island) were used
for all experiments. The average lengths and weights of the fish used
in each experiment are shown in Table 2. Newly captured fish were
acclimated in holding tanks for at least 3 days prior to experimentation.
They were maintained in all-glass aquaria supplied with running seawater.
Water temperature during the experimentation period ranged from 8.5°C
to 12.4°C.
Fish used in the effects studies were fed clams dug from non-urban
beaches of Puget Sound and frozen until use, frozen euphausids taken
fran Alaskan waters and obtained from a commercial supplier, and earth-
worms from local suppliers. The fish were fed ad libitum, and the food
supply was replenished twice daily.
English sole used in the bioavailability experiment were kept in
flowing seawater (salinity of 28% and 13.0+0.5°C) and fed a diet of
minced clams, obtained as above, for 2 weeks. The feeding was stopped
3 days prior to the initiation of the experiment.
Exposures to Sediment Extracts
Preparation of extracts of sediment and interstitial water--
Uuwamish Waterway sediment and Port Madison sediment were extracted
using identical procedures. Briefly, the sediment extraction involved
the following procedure: (1) separating the sediment from associated
water by centrifugation; (2) extracting the sediment with rnethanol ;
(3) extracting the sediment with a 2:1 mixture of the sediment with a
2:1 mixture of methylene chloride and methanol; (4) pooling the extracts
and extracting the methanol-pooled extract with saline (2%); and
(5) storing at 4°C the resulting methanol-aqueous phase and concentrating
the methylene chloride phase by evaporation. This bulk concentrate was
dissolved in corn oil, the remaining methylene chloride was evaporated,
and the residual, undiluted material was injected with a corn oil
vehicle into English sole. The interstitial water was extracted in a
similar manner.
The chemical composition of the original sediments is shown in
Table 3.
Exposure of fish to sediment extracts--
Two experiments involving i.p. injection of sediment extracts were
completed. Fish were held in 30 x 50 x 30 cm aquaria. In Experiment
One, for each type of extract, i.e., Duwamish Waterway (test) and Port
Madison (reference), three dilutions (1/2, 1/20 and 1/200) of sediment
extract in corn oil were each injected (0.1 ml per fish) into one of
11
-------
TABLE 2. Lengths and weights of English sole used in laboratory experiments.
Experiment
Sediment Extract Injection
Experiment One
Experiment Two
Interstitial Water Injection
Experiment One
Sediment Contact
Experiment One
Experiment Two
Number of
Specimens
21
39
20
200
67
Mean
Length
(mm)
103
110
121
103
130
Length
Range
(mm)
87-118
78-143
94-173
73-132
91-175
Mean
Wei ght
(g)
10.4
10.3
13.0
9.8
18.1
Weight
Range
(g)
5.8-15.0
4.8-18.6
5.6-37.8
3.4-19.6
5.3-40.3
Bi oavailabi1ity
Experiment One
210
200-220 70.0
65-75
-------
TABLE 3. Concentrations of selected compounds in sediment from the
Duwamish Waterway, the Snohom'sh River, and Port Madison
used in laboratory experiments.
Compounds
Fluorene
Dibenzothiophene
Phenanthrene
Anthracene
Fluoranthene
Pyrene
Benz[a]anthracene
Chrysene
Benzo[e]pyrene
Benzo[a]pyrene
Total PCBs
Total chlorinated
butadienes
p,p'-DDl)/o,p-DnT
p,p'-DUE
Snohomi sh
Ri ver
49
43
250
22
450
330
150
150
43
26
71
1
3.5
<1
Port
Madison
(ng/g dry wt. )
5
7
35
4
57
63
44
30
33
24
56
21
1
<1
Uuwami sh
Waterway
95
110
510
94
1200
1000
1000
670
580
650
713
4
11
5
13
-------
three groups of three fish each. An additional group of three fish
received corn oil alone. Fourteen days following injection, surviving
fish were sacrificed and necropsied. The necropsy procedure was similar
to that described for the field studies, except that blood was collected
by syringe from the caudal vessels just ventral' to the vertebral column,
or in smaller fish by excising the caudal fin. Aliquots of whole blood
were used immediately after collection to determine hemoglobin concentrations
and hematocrit values.
In Experiment Two, fish were injected with extracts of Duwamish
Waterway and Port Madison sediments which were approximately three times
more concentrated than those used in Experiment One. For each type of
sediment extract, each of three dilutions (undiluted, 1/2 and 1/10)
were injected (0.1 ml per fish) into one of three groups of five fish
each. Control groups of five fish each were injected with corn oil
alone (vehicle control) or received no injection (untreated control).
Eighteen days after the first injection, surviving fish in each group
were reinjected with the same amount and dilution of the test or reference
substance as described for the first injection. Survivors were sacrificed
and necropsied five days after the second injection.
One experiment examining the effects of the injection of interstitial
water was performed. In this test, groups of five fish were injected
i.p. (0.1 ml per fish) with extracts of either Duwamish Waterway or
Port Madison interstitial water (undiluted and diluted 1/2 in corn oil).
Surviving fish were sacrificed and necropsied 16 days after injection.
Exposures to Bottom Sediments
Two experiments were performed in which fish were maintained for
2 or 3 months in aquaria containing test sediment (Duwamish Waterway)
or reference sediment (Snohomish River or Port Madison). The chemical
composition of each sediment is shown in Table 3.
In Experiment One, English sole were maintained for 92 days in 30 x
90 x 120 cm aquaria containing 3 cm of Duwamish Waterway or Snohomish
River sediment. Six fish were necropsied and examined hematologically
and histopathologically at the beginning of the experiment for comparison
with fish sampled at later times. One hundred fish were maintained
on each sediment type; each group (test or reference) was divided
equally between two aquaria. Fish were sacrificed and necropsied from
each group on days 29, 57 and 92.
In Experiment Two, English sole were maintained under conditions
similar to those described for Experiment One, except that the reference
sediment was from Port Madison. At the time the experiment was started
(0-time), five fish were necropsied for histopathological and hematological
examination. Thirty-five fish were placed on Duwamish River sediment
and 32 were placed on Port Madison sediment. Three fish were sacrificed
and necropsied from each group on day 37 and five were sacrificed and
necropsied from each group on day 65.
14
-------
Analysis of Effects Data
Standard statistical tests (Zar 1974) were used for analysis of
mortality, lesion frequency and hematological data. Mortality data for
the various treatment groups were arranged in 2 x 2 contingency tables
and a chi-square statistic was calculated for each table to determine
whether the distribution of deaths in a given experiment was independent
of the type of treatment administered. Similar tests were performed on
the histopathological data to ascertain whether the frequency of
microscopically-observed anomalies was independent of treatment type.
One-way analysis of variance and the Neuman-Keuls multiple range
test were employed for comparisons of mean hematocrit or hemoglobin
values among the various test and control groups in the experiments.
Apparent trends in hematological data were further investigated using
simple linear regression.
Exposure of English Sole to Aromatic Hydrocarbons in Sediment
Sediment collected from a reference area (Port Susan) was mixed
with [3H] BaP [3 mg (5480 uc) per 2 1 of sediment]. Mixing was
performed by first dissolving the BaP in acetone, adding this solution
to 0.5 1 of sediment, evaporating the acetone with a stream of nitrogen,
and combining and nixing the contaminated sediment with the remaining
1.5 1 of sediment. The [3H] BaP-sediment was placed in glass aquaria
(17.0 1 capacity), seawater was added at a flow rate of 20 I/day for
1 day. Three fish (see Table 2 for size data) were then placed in the
experimental tank and sampled 24 h after initiation of exposure.
Analysis of sediment and sediment-associated water--
Samples of sediment and sediment-associated water (SAW) were
taken from 2 cm below the sediment/water interface. The wide end of a
glass pipette was vertically inserted into the sediment while the tip
of the pipette was covered. After positioning the pipette, the tip
was uncovered to allow sediment and SAW to rise within the pipette.
The sample was carefully transferred to a vial and the SAW was decanted
after the suspended particles had settled. Samples of water from near
the air/water interface were also taken.
Samples of wet sediment and 1-ml aliquots of unfiltered SAW were
analyzed for total radioactivity [3H] using liquid scintillation
spectrometry as described by Varanasi et al. (1978, 1979).
Analysis of tissues--
Fish were killed by a blow to the head and samples of blood,
liver, muscle, and bile were analyzed for [3h] using methods detailed
by Varanasi and Gmur (1981a). Enzymatic hydrolysis of the conjugated
metabolites of BaP in the bile samples were analyzed for glucuronides
using Glucurase (Sigma Chemical Co., 500U per assay).
15
-------
RESULTS AND DISCUSSION
FIELD STUDIES
The field survey portion of this investigation was primarily
intended to document the prevalence, geographical distribution and
appearance of lesions in English sole and starry flounder from selected
estuaries in Puget Sound. As part of this effort, the catch data and
length/weight/age characteristics of these two species were also
obtained. Chemical analyses were performed on samples of sediment and
fish 1iver tissue.
Of the 1293 English sole examined for externally visible lesions,
673 were sacrificed, necropsied, and examined for microscopic lesions
in najor organs. In the sane nanner, 525 starry flounder were examined
externally and 350 were necropsied. Two major types of lesions were
observed, parasitic and idiopathic (lesions of unknown etiology). The
description of these lesions which follows will be divided according to
species and subdivided into affected organs.
Pathological Conditions of English Sole
Lesions of the skin and fins--
A broad spectrum of parasitic and idiopathic lesions were diagnosed
in English sole skin and fin sections. Several could be diagnosed grossly,
but sone could only be diagnosed definitively at the light microscopic
level. Because skin and fin sections were taken only from fish with
grossly visible anomalies, only the prevalence of fish with these anomalies
will be presented. Those lesions which were detectable only by micro-
scopic examination will be described, but prevalance data will not be given,
Parasitic lesionsThe most common parasitic condition was readily
visible on gross inspection, but prevalence data for this condition
was not compiled. In this parasitic infection, members of the nematode
genus Philometra were grossly visible in the subcutaneous tissue of the
skin and fins, and often parasitized the peritoneum, peritoneal cavity,
kidney, pericardial cavity, and body musculature. The typical host
response to these blood-feeding nematodes was chronic inflammation
with fibrosis. Microscopic examination showed that most of the parasites
observed were females bearing filariae. Free filariae were often
found within the subcutaneous tissue. Affected English sole were
found in all sampling areas.
Another helminth infection was grossly visible as small 3-4 mm
subcutaneous nodules in the skin and fins. A definitive diagnosis was
possible only by microscopic examination. This infection of the derrnis
by digenean trematode metacercariae was accompanied by dermal fibrosis,
chronic inflammation, and necrosis of the dermal connective tissue
16
-------
elements. The netacercariae were miniature adults lacking only a
developed reproductive system, and are presumably infective to the
definitive host (Chitwood and Lichtenfels 1972). English sole are,
therefore, very likely an intermediate host. Each metacercaria was
surrounded by a cyst, thought to be elaborated principally by the
parasite. However, in most sections there was significant fibrotic
host reaction as well, with occasional congestion of the subcutaneous
vasculature. Precise identification to family or genus was not attempted
for these digenean trematode metacercariae, but morphologic similarity
between cases indicates that a single genus was represented.
Skin metacercarial lesions were detected microscopically in a
few fish (10) from the Lake Washington Ship Canal, Duwamish River and
Snohorni sh River. No cases were found in English sole from McAllister
Creek.
Viral lesionsA single case of a dermal viral infection (lymphocystis
virus disease) was detected in an English sole from the Duwamish Waterway.
This condition is ubiquitous worldwide, and has to date been reported
from northwestern and western coastal marine waters in the U.S. (Alpers
et al. 1977a, McCain et al. 1978a, 1979, McCosker and Nigrelli 1971).
To our knowledge, this is the first report of lymphocystis virus disease
in Puget Sound, and the first report in English sole (Parophrys vetulus).
The light and electron microscopic features of this lesion are described
in Figures 3 to 5. (For a more detailed description of this lesion,
see Appendix 1.)
Skin tumors--Tumor-pus skin lesions were also infrequently detected
grossly in English sole. These tumors were of three main types, termed
angioepithelial nodules (AEN), transitional angioepithelial nodules
(TAEN) and epidermal papillornas (EP). All three are elements of a single
disease complex common in pieuronectids (McArn et al. 1968), the etiology
of which is controversial, but is thought to be caused by a parasitic
amoeba (Uawe 1980, Myers 1981). The tumors are known to progress from
AEN to TAEN to EP. These lesions are described in detail in Appendix 1.
AEN lesions were detected in all four sampling areas, with the
highest prevalence in McAllister Creek (10.3%, 4 of 39 fish) and the
lowest in the Duwamish River (0.2%, 1 of 565 fish). The prevalence in
Lake Washington Ship Canal was 4.7% (2 of 48 fish), and in the
Snohomish River the prevalence was 2.7%. Transitional tumors were found
only at McAllister Creek (5.1%, 2 of 39 fish) and the Duwamish Waterway,
(0.2%, 1 of 565 fish), and the single case of a wel1-developed EP was
from the Duwamish Waterway.
Fin erosion--An idiopathic fin lesion, termed fin erosion (Wellings
et a 1. 19760) was visible grossly as a loss, distortion or fusion of fin
rays accompanied by hemorrhage and fibrogranulation tissue on the fin
surfaces and at the juncture of the fin rays and body musculature. The
condition was most common on the anal and dorsal fins. The free margins
of damaged fins were often ulcerated, and scarification of fin tissue
17
-------
FIGURE 3. Micrograph of lyrnphocystis skin lesion in an English sole.
In the center is a 1 yrnphocysti s-i nfected cell with the
characteristic enlarged nucleus, densely basophilic
cytoplasmic inclusion bodies and external hyaline capsule,
H a E, 975X.
18
-------
100 nm
FIGURE 4. Electronmicrograph of intracytoplasmic lymphocystis virus
particles. The particles have an internal nucleoid
structure and outer capsids.
19
-------
100 nn
FIGURE 5. Electromicrograph of intracytoplasmic lymphocystis virus
particles and fibril lar structures showing banding
periodicity.
20
-------
was common, with folding and fusion of adjacent fin rays. In the most
severe cases, the fin rays were entirely absent and the residual tissue
was scarred, retracted, and lacked rigidity.
The microscopic characterisitcs of these lesions have been previously
described in detail (Wellings et al. 1976b), but the more salient features
will be described here. The condition is generally a chronic fibrosing
disease with epidermal hyperplasia, dermal fibroplasia and fibrosis,
infiltration of mononuclear inflammatory cells, increased numbers of
melanophores in the dermis, blood vessel sclerosis, and increased epidermal
eosinophilic granular cells. The hallmark of this condition is resorption
of the bony fin rays and replacement with fibroblasts and collagen.
Generally, this constellation of features could be found in any fully
developed case of fin erosion. Occasional ulceration of the epidermis
was present, and inclusion cysts lined by epidermal cells were often
observed in more severe cases. Pathogenic bacteria were not detected
in the Brown and Brenner Gram Stain (Luna 1968), nor were other
potential protozoan agents. Viral inclusions were not detected at the
light microscope level, and the condition was not commonly associated
with any parasitic infection.
Two microscopically definitive cases (0.4%, 2 of 565 fish) of
this condition in English sole were detected in the Duwamish Waterway,
and in none of the remaining sampling areas. Fin erosion is much more
common in starry flounder from the Duwamish River, but the lesion is
identical and prevalence data will be presented in the section describing
skin and fin lesions in starry flounder.
Miscellaneous skin 1esions--0ther lesions found in the areas
sampled were as follows: non-specific epidermal necrosis and spongiosis,
dermal chronic inflammation, exostosis of fin rays (found in only one
fish from the Duwamish Waterway), dermal necrosis, subcutaneous hemorrhage,
epidermal hyperplasia without true fin erosion, subcutaneous edema, and
the presence of dermal nematode filariae.
Liver lesions--
Both idiopathic and parasitic lesions were observed in the livers
of English sole. The major idiopathic liver lesions included megalocytic
hepatosis (MH), hepatocellular regeneration (HR), hepatocellular
eosinophilic hypertrophy (HEH), minimum deviation nodule (MDN),
hepatocellular nodular hyperplasia (HNH), liver cell adenoma (LCA),
hepatocellular carcinoma (HC), cholangiocellular carcinoma (CC), mixed
carcinomas (MC), cholangiofibrosis (CF), hepatocellular necrosis (HN),
fatty change (FC), hemosiderosis (HEM) and hemangiomas. The only parasites
consistently associated with lesions of the liver were members of the
sporozoan genus Myxidium (Butschli), family Myxidiidae (Thelohan).
The morphologic characteristics of these lesions will first be presented
in this section, followed by a section dealing with the geographical
distribution and prevalence of the lesions.
21
-------
Megalocytic hepatosis (MH)--Megalocytic hepatosis was characterized
by massive increases in the dfameters of hepatocytes and their nuclei
(i.e., megalocytosis) as a non-inflammatory lesion of the liver, i.e.,
hepatosis (Jones and Butler 1975). Other characteristics included
cytoplasmic acidophilia or a "ground-glass appearance" with occasional
fibrillar "fingerprint" cytoplasmic structures characteristic of
proliferation of the smooth endoplasrnic reticulum (Bannasch 1976), and
nuclear hyperchromasia. The enlarged nuclei were generally highly
vesicular and exhibited aberrent, dense distributions of chromatin,
often marginated, indicating an elevation in DNA content (Fig. 6).
The cytoplasm commonly showed signs of degeneration, including hydropic
degeneration and hyalinization. Excessive amounts of cytoplasmic iron-
containing pigment (hemosiderin) were commonly found in the affected
megalocytes by the Prussian Blue reaction (Preece 1972). No
proliferation of megalocytic hepatocytes was seen.
The morphology of this lesion closely parallels hepatotoxic changes
induced in fish and other vertebrate species by diverse types of
naturally occurring and xenobiotic chemical species. Exposure to
pyrrolizidine alkaloids (plant extracts) produces hepatic megalocytosis
in mammals (Jubb and Kennedy 1970, McLean 1970, Jago 1969, McLean 1974),
as do cyclopropenoid fatty acids in fish (Sinnhuber et al. 1968,
Struthers et al. 1975, Malevski et al. 1974, Hendricks et al. 1980),
PCBs in mammals and fish (Koller and Zinkl 1973, Nishizumi 1970, Hinton
et al. 1978, Kimbrough and Linder 1974), and phenobarbitone in mice
(Jones and Butler 1975).
The nuclear characteristics in MH suggest action by a mitotic
poison, resulting in polyploid hepatocytes which fail to divide. The
degenerative cytoplasmic changes suggest parallel action by a cytotoxin.
Quite possibly, megalocytic hepatosis is a subacute or chronic manifes-
tation of the cytotoxicity of hepatotoxin(s) and/or hepatocarcinogen(s).
In the early stages of lesion development, this possible effect may be
combined with the paradoxical but general antiproliferative effects of
most hepatocarcinogens (Farber et al. 1976).
Hepatocellular regeneration (HR)--Typical1y associated with MH,
but also found independently, were discrete foci of hepatocellular
regeneration, usually in hyperplastic form. The foci were of variable
size and were randomly distributed. These scattered islands were
composed of small, moderately basophilic hepatocytes with normal-sized
or smaller than normal-sized nuclei arranged in hyperplastic (more than
two cells thick) muralial cords. These foci did not contain iron
pigment but did contain periodic acid-Schiff (PAS)-positive material,
presumably glycogen. The islands were not well demarcated from the
surrounding parenchyma except in cases where the remainder of the
liver was homogeneously necrotic and/or affected with MH, and there
was no visible compression of adjacent parenchyma (Fig. 6). These
islands occurred in the presence of hepatocellular degenerative lesions,
including frank hepatocellular necrosis and were probably a compensatory
response to the degenerative effects and cytotoxic effects of potential
hepatotoxins and/or hepatocarcinogen(s) or non-specific necrogenic agents.
22
-------
FIGURE 6. Micrograph of the hepatic lesions megalocytic hepatosis
and hyperplastic hepatocellular regeneration in English sole.
In this liver section, numerous megalocytic hepatocytes
(megalocytic hepatosis) indicated by large arrows, were
accompanied by a hyperplastic proliferation of snail,
basophilic, regenerative hepatocytes in a nultifocal pattern
(small arrows). H & E, 480X.
23
-------
The pro!iterative capacity of cells comprising these islands
suggests that they have developed a selective resistance to the necrogenic
effects of putative hepatotoxin(s)/carcinogen(s), as proposed by Farber
(1976) in his raodel of experimental rat hepatocarcinogenesis. These
islands are probably metabolically immature as evidenced by their
inability to take up or store iron (Williams 1976). These regenerative,
undifferentiated hepatocytes may, in fact, be unable to transport and/or
bioactivate hepatotoxins or hepatocarcinogens as indicated by Solt et al.
(1977) and Farber (1976). Although our research has explored very
few indices of maturation, this interruption of maturation is thought
to be a constant characteristic of the pathway of differentiation
leading to neoplasia (Farber et al. 1977, Solt et al. 1977), and is a
functional manifestation of anaplasia (reversion to a more embryonic
form of the cell type in question). Therefore, these islands may
represent preneoplastic lesions, or at least precursors of preneoplasia,
in the stepwise selection process progressing towards the autonomous
growth seen in neoplasia (Pitot 1977).
On the other hand, these islands may be "remodeled" (Kitigawa
1976) back into the normal liver parenchyma by differentiating into
mature hepatocytes, or they may simply regress. They are not considered
to possess the property of autonomous growth (neoplasia), and therefore,
are considered reversible lesions. Such "remodeled" cells, however,
may later give rise to neoplasia if a proper stimulatory agent is
supplied, and are considered to be "initiated" hepatocytes in the two-
step initiation-promotion model of hepatic hyperplasia (Pitot 1977).
In our study, hepatocellular neoplasms and foci of hepatocellular
atypia were found adjacent to or within these regenerative, hyperplastic
islands.
Hepatocellular eosinophilic hypertrophy (HEH)--Also commonly
associated with MH, but often as an independent lesion, were discrete
hepatic foci or micronodules characterized by a dense cytoplasmic
eosinophilia, often with accompanying increase in cytoplasmic diameter,
termed eosinophilic hypertrophy or eosinophilic foci (Fig. 7). The
nuclei were vesicular, normal-appearing, or contained condensed chromatin
elements. Densely eosinophilic perinuclear inclusions were often
visible in the more pale eosinophilic cytoplasm. The significance of
or nature of these inclusions is not presently known. The cytoplasm
was rich in glycogen, with reduced basophilia and did not contain
stainable iron, indicating an inability to store or take up iron.
This absence of stainable iron-containing pigment parallels the findings
for regenerative, hyperplastic islands. Pyknotic nuclei and cytoplasmic
degeneration were rare. Occasionally, within these typically spherical,
multiple micronodules were groups of hepatocytes which were eosinophilic
and hyperplastic and lacked the normal muralial structure. These eosino-
philic foci were often found closely opposed to or even within larger
multifocal to diffuse areas of hepatocellular hyperplasia/regeneration.
There was no parenchymal compression associated with these foci.
24
-------
FIGURE 7. Micrograph of nodular hepatocellular eosinophilic hypertrophy
in English sole. In the right hand section of this micrograph
is a hepatic nodule, characterized by hypertrophied, eosinophilic
hepatocytes within an organized rnuralium, with a low degree
of cellular atypia. This lesion is diagnosed as nodular
hepatocellular eosinophilic hypertrophy. H & E, 240X.
25
-------
These foci are similar morphologically to the enzyme-deficient,
altered foci described as preneoplastic lesions in rats (Solt et al. 1977,
Scherer and Emmelot 1975, Bannasch 1976, 1978). Their rich glycogen
stores suggest a retarded rate of glycogen metabolism, also seen in
similar rat lesions (Solt et al. 1977, Bannasch 1976). Such foci nay be
precursors of neoplasia (Rabes et al. 1972), but are also considered
reversible. Some authors (Rabes et al. 1972) consider these enzyme-
deficient areas to be an essential but not sufficient prerequisite for
microcarcinoma formation. The fact that these eosinophilic foci are
found in livers with true hepatocel lular neoplasms suggests a role in
the progression towards neoplasia in the English sole. Similar lesions
have been detected in rats exposed to the PCB Aroclor 1260 (Kimbrough
et al. 1975).
Minimum deviation nodules (MDN)--Commonly found in English sole
livers bearing one or all of the previously described lesions were small
hepatocellular nodules. Termed minimum deviation nodules (MDN), they
had several morphologic features typical of neoplasia, including
cytoplasmic hyperbasophilia, changes in hepatocellular polarity, a
slightly increased nuclear to cytoplasmic ratio by virtue of nuclear
enlargement, prominent nucleoli, slight compression of the adjacent
parenchyma, absence of other hepatic elements and a total absence of
stainable cytoplasmic iron (Fig. 8). Hyperplastic hepatocytes were
often located within these nodules.
These proliferative nodules are thought to arise from the
regenerative, hyperplastic islands, as similar nodules emerge from areas
of hyperplasia in rats (Rabes et al. 1972, Pitot 1977, Farber 1976,
Solt et al. 1977, Squire and Levitt 1975, Newberne 1976, Karasaki 1976,
Bannasch 1976, Weisburger et al. 1972). These cells are putatively the
result of a complex sequential selection process (Scherer and Emmelot
1976) in the hepatocyte population. They most probably originate from
within the previously selected hyperplastic, regenerative foci, and
may be resistant to the antiproliferative, degenerative effects of
hepatotoxin(s)/carcinogen(s). MDNs correspond closely with the
hyperbasophilic foci (Bannasch 1976, Karasaki 1976, Rabes et al. 1972)
which are considered irreversible lesions in neoplastic progression,
capable of autonomous growth. This issue is controversial due to the
lack of consistency in both transplantation success with this type of
lesion (Ward and Vlahakis 1978) and cell culture growth characteristics
(Karasaki 1976) and the inadequate predictive ability of malignancy
provided by morphologic features (Newberne 1976).
Liver cell adenoma (LCA)--These supposedly benign neoplasms exhibited
heterogeneous microscopic features. They were usually spherical nodules
with clearly demarcated borders, and were composed of basophilic,
enlarged hepatocytes with intact liver cord architecture, but occasionally
showed an eosinophilic or amphophilic cytoplasm (Fig. 9). The cytoplasm
of individual hepatocytes composing the nodules was sometimes heavily
vacuolated but did not contain iron pigments. There was no appreciable
fibrosis, but sometimes the nodules were encapsulated. Tumor growth
was expansive and not overtly invasive, and the nodules showed an
26
-------
FIGURE 8. Micrograph of the hepatic neoplastic lesion termed a
minimum deviation nodule in English sole. A small nodule
composed of slightly hypertrophie, basophilic hepatocytes
is seen at the center of the micrograph. This type of
nodule is classified as a minimum deviation nodule.
H & E, 240X.
27
-------
FIGURE 9. Micrograph of a liver cell adenoma in English sole. The
upper portion of this micrograph contains a large hepatic
nodule composed of an organized proliferation of
basophilic hepatocytes, characteristic of a liver cell
adenoma. H & E, 240X.
28
-------
increased cellularity in comparison to the surrounding parenchyma.
When stained with Masson's Trichrome (Preece 1972), these neoplasms
exhibited a rich maroon-brown cytoplasmic staining. The nodules had
a relatively normal-appearing muralial architecture, and lacked of
bile ducts and rnelanomacrophage centers (MMCs) (Roberts 1975, Agius
1979). The nodules were sometimes slightly anaplastic; in which case,
the hepatocytes had prominent nucleoli. Monornorphic cell populations
were often observed within individual nodules, retaining polarity, as
was occasional hyperplasia and thickening of hepatic cords.
Extrahepatic metastases of this type of neoplasm have never been
found in English sole from Puget Sound, but multiple LCA were located
in individual livers, suggesting either an independent, multifocal
origin, or intrahepatic metastases. The classification of LCA was
based solely on histologic criteria typical of similar non-metastasizing,
benign hepatocellular neoplasms from human studies and mammalian models
(Squire and Levitt 1975, Edmondson 1958, Newberne and Butler 1978).
Hepatocellular carcinoma (HC)--Other hepatocellular neoplasms with
widely-accepted histologic criteria characteristic of hepatocellular
carcinomas in experimental animals were also found in English sole.
These tumors were typically multiple and/or adjoined, with irregular
borders, and displayed foci of invasion with considerable compression
in growth-by-expansion (Fig. 10). The muralial architecture of the
neoplasms was typically trabecular. They were composed of anaplastic,
polygonal, pleomorphic hepatocytes with an increased nuclear-to-
cytoplasmic ratio, vesicular nuclei with prominent nucleoli, and loss
of hepatocyte polarity in relation to the sinusoids. Extrahepatic
metastases, tumor emboli, and vascular invasion were not observed. The
degree of cytoplasmic basophilia in these presumptively malignant
neoplasms was greater than in any of the hepatic neoplasms encountered.
Glycogen levels were low, possibly due to development of a Warburg-type
of glycolysis present in most hepatocellular carcinomata (Bannasch
1976). Bile ducts and MMCs were absent and cytoplasmic iron was not
demonstrated in the neoplastic hepatocytes. As with all of the hepato-
cellular neoplasms in English sole, mitotic figures were rare. Often
up to 75% of the liver at a particular cross section was replaced by
the neoplasm. With Masson's trichrorne staining, the hepatocellular
cytoplasm exhibited a purple-green tint distinguished from the red-
maroon of normal hepatocytes and the denser maroon typical of hepato-
cellular adenomata.
Hepatocellular carcinomas exhibited morphologic heterogeneity.
Most were overtly trabecular, while some were more densely cellular and
solid, growing in broad sheets with little recognizable cordal pattern.
Others had a pseudotubular, pseudoacinar morphology. Varying degrees
of bile duct proliferation and fibrosis were seen associated with the
periphery of most of these neoplasms.
29
-------
FIGURE 10.
Micrograph of an hepatocellular carcinoma in English sole.
The upper section of this micrograph shows an hepatic nodule
with a disorganized muralial architecture, composed of
pleomorphic, anaplastic hepatocytes which have lost the usual
orientation (polarity) to the sinusoids. This morphology is
characteristic of an hepatocellular carcinoma. H & E, 240X.
30
-------
Similar neoplasms have been induced in mammalian models by a wide
variety of natural and xenobiotic hepatocarcinogens (Farber 1976).
These compounds include pyrrolizidize alkaloids, azo dyes, chlorinated
biphenyls (Kimura and Baba 1973, Ito et al. 1973a, Kimbrough and Under
1974, Kimbrough et al. 1975), and benzene hexachloride (Ito et al. 1973b).
Hepatocellular carcinomas are also associated with high hepatic PCS
levels in Atlantic hagfish (Myxine glutinosa) (Falkmer et al. 1977), and
have been found in Atlantic tomcod (Hicrogacfus tomcod) from polluted
waters of the Hudson River (Smith et al. 1979).
Cholangiocellular carcinoma (CC)--A separate class of intrahepatic
proliferative lesions in English sole was clearly cholangioproliferative.
The most common type was a Cholangiocellular carcinoma and was manifested
by a disorganized proliferation of cholangiolar epithelium assuming a
tubular pattern. These lesions were grossly visible as pale green
nodules with highly irregular borders, and were often fungiforrn with
numerous fingers insinuating throughout the hepatic parenchyma.
Microscopically, the tubules were usually very irregular in architecture,
or were simply absent, with the neoplasms consisting of a highly cellular
proliferation of squamous to cuboidal biliary epithelium, usually within
a fibrous stroma (Fig. 11). These tumors in English sole could be
differentiated from hepatocellular neoplasms by virtue of their tubular
structure and the distinct nuclear morphology of the turnor cells.
Nuclei of biliary epithelium are much smaller and more ovoid than
hepatocellular nuclei, and possess a less prominent nucleolus and less
nuclear chromophilia. The degree of cytoplasmic basophilia in Cholangio-
cellular neoplasms was considerably less, while the cytoplasmic
eosinophilia (H & E stain) was slightly greater than the hepatocellular
counterpart.
Although these neoplasms were not particularly anaplastic, they
generally were extremely disorganized and tubular structures resembling
mature bile ducts were infrequently observed. They were also often
highly invasive, replacing large portions of the liver.
Although for most of these neoplasms, malignancy was presumed from
their cytologic features, the malignancy of one individual neoplasm
was clearly demonstrated in an English sole from the Duwamish River.
This extremely invasive neoplasm similar in morphology to other liver
neoplasms in English sole diagnosed as Cholangiocellular carcinomas,
almost entirely replaced the posterior hepatic lobe. The invasive
capacity of this neoplasm was so great that the available blood supply
was outgrown, and vascular channels were invaded, resulting in massive
areas of both ischemic and hemorrhagic necrosis. The most significant
findings, however, were extrahepatic metastases in the spleen, kidney,
small intestine muscle wall, and ventricular myocardium. In the kidney,
heart and spleen, these metastases had seeded successfully and were
invading and replacing adjacent tissues. These nietastatic foci were
identical histologically to the massive primary carcinoma in the liver.
31
-------
FIGURE 11. Micrograph of a rnetastasizing cholangiocel lular carcinoma
in English sole. This disorganized neoplasm in the liver
was composed of pleomorphic, invasive cells closely resembling
biliary epithelium and had metastasized to the heart, spleen,
kidney and gut wall musculature. H & £, 240X.
32
-------
The cell of origin in cholangiocellular carcinoma observed in
English sole was most probably biliary epithelium from the vicinity of
the ducts of Hering (Moulton 1978). However, there is controversy
over this issue. Exposure to most hepatocarcinogens often produces a
proliferation of bile ducts which may lead to a cholangiocellular
carcinoma (Farber 1976). But with the carcinogenic azo dyes there is
an induction of both cholangiocellular and hepatocellular carcinoma
(Farber 1956). These carcinomas may arise from tubular stern cells
which can develop into either type of malignancy. Such a mechanism
may be operant in English sole since numerous examples of both separate
hepatocellular and cholangiocellular carcinomas were encountered in
individual livers, and mixed carcinomas were also seen. Both mature
epithelial cell types do arise from the same embryonic stem cell in
mammals. Similar tumors in rats and nice have been classified as
tubular variants of a hepatocellular carcinoma (Squire and Levitt 1975,
Jones and Butler 1975). The clear morphologic differences between
cholangiocellular carcinomas and hepatocellular carcinomas described
here permit such a distinction in the case of English sole. Similar
neoplasms diagnosed as cholangiocellular carcinomas as well as mixed
carcinomas have been reported in hagfish (Falkrner et al. 1977).
Cholangiofibrosis (CF)--A separate cholangioproliferative lesion,
thought to be non-neoplastic by some (Farber 1976) and preneoplastic by
others (Reuber 1968) was classified as Cholangiofibrosis (CF)/adenofibrosis
(Squire and Levitt 1975) in English sole. This lesion consisted of a
proliferation of dilated cystic tubular structures, probably bile ducts,
in a dense encapsulating and enveloping fibrous stroma, with only
occasional atypia in the bilary epithelium. Compression and invasion
were absent, and because of the fibrosis and delimiting encapsulation,
the lesion in English sole was considered to be a benign proliferation.
Reuber (1968) believes that CF precedes development of wel1-differentiated
cholangiocarcinoma. No evidence to support this has been found in
English sole. As Farber (1976) and Ward and Vlahakis (1978) suggest,
most hepatocarcinogens are also inducers of proliferation of bile ducts
in the liver early in carcinogenesis. With hepatocarcinogens, this
proliferative response regresses or may develop into Cholangiofibrosis,
but it may also lead to the development of a hepatocellular carcinoma.
This lesion has been induced in rats, mice, and hamsters by
xenobiotics including 2-acetamidofluorene and 2-diacetamidofluorene
(Reuber 1968), and PCBs (Kimbrough et al. 1972, Kimbrough and Linder
1974).
Hemangioma--Two cases of lesions which closely resembled a mammalian
hemangioma were found in the livers of Duwamish River English sole.
These neoplasms were characterized by a proliferation of vascular
endothelial cells surrounding blood-filled spaces, supported by a
thin, fibrous stroma. Often these cystic vessels were ruptured, with
necrosis in the hepatocellular parenchyma. There was no invasion of
peripheral tissues and the neoplastic cells showed no cellular atypia.
Similar benign, hemangiornatous neoplasms have been induced by urethane
33
-------
injection (Heston et al. 1960, Kawamato et al. 1961, Roe 1954) and
dimethylnitrosarnine (Jones and Butler 1975) in experimental mammals.
Hepatocel1ular necrosis (HN)--Other incidental hepatic lesions were
found in English sole.The most common was hepatocellular necrosis (HN)/
degeneration (not associated with incidental parasitism), a non-specific
response to toxins or pathogens.
In sone livers, necrosis was diffuse, affecting 75-80% of the liver,
with only a minimal cellular inflammatory response. If a biologic
agent were present, the cellular inflammatory response should have been
prominent. Most degenerative/necrotic foci were found in conjunction
with previously described lesions, and were associated with microinvasion
by neoplasms, or were equated with the degenerative changes found in
MH. In cases where benign neoplasms were present, or where HN was the
only visible lesion, the necrotic foci may have represented necrogenic
action by toxic chemicals or carcinogen(s).
Miscellaneous non-parasitic liver lesionsFatty change (FC), or
abnormal accumulation of lipid in groups of hepatocytes was also frequently
encountered, both in fish with and without neoplasms. This lesion is
a degenerative lesion, commonly associated with dietary deficiencies,
or toxic chemical adninistration. Its role, if any, in the progression
of lesions towards turior formation in the English sole is not presently
known.
Many English sole livers from the Duwamish Waterway exhibited
elevated levels of intracellular iron when stained by Prussian Blue
reaction. (This reaction is mainly exhibited by hemosiderin.) This
accumulation is especially common in the megalocytic cells of MH.
Iron accumulation can be associated with non-specific liver injury
(Robbins and Cotran 1979) but is also associated with a wide spectrum
of factors, including excessive dietary iron, anemia, liver cirrhosis,
and hemorrhage. The identification of this condition, termed
nemosiderosis (Hem), generally indicates the existence of some concurrent
condition responsible for the excess iron (Robbins and Cotran 1979).
The final hepatic lesion to be described is rarely found in the
English sole, but is more common in the starry flounder. A detailed
description will be presented in the section leading with hepatic
lesions in starry flounder. This lesion most closely resembles the
human hepatic condition peliosis hepatis (PH) and is basically
characterized by severe congestion/ectasia of the sinusoids with intra-
sinusoidal fibrin deposition (resembling cysts containing blood)
hemorrhage, and necrosis of the hepatic cords adjacent to the foci of
ectasia (dilatation).
Parasitic hepatic 1esions--The most frequent and widely distributed
parasitism of hepatic tissue was an infection by members of the sporozoan
genus flyxidium (Butschli), family Myxidiidae (Thelohan). The infection
was generally confined to the intrahepatic bile ducts and bile ductules
34
-------
(cholangioles) where plasmodial forms of this cnidosporidan were present.
Spores were also present adjacent, to the mucosa of the gallbladder.
Commonly, the plasmodia were of insufficient volume to cause compression
atrophy of the biliary epithelium, and the typical host response was a
mild peribiliary lymphoid infiltrate. In more severe cases, there was a
proliferation of bile ductules containing plasmodia and spores, and
variable degrees of fibrosis with chronic inflammation. In the most
severe cases, where the parasite could be found scattered throughout
the hepatic parenchyma in addition to the intrahepatic bile ducts and
bile ductules, severe fibrosis and fibroplasia were present along with
hepatocel lular necrosis affecting a significant proportion of the liver.
Varying degrees of cholestasis (bile stasis) were also observed in
severe cases. These severe infections, with significant hepatic fibrosis,
were interpreted as true pathologic lesions due to the amount of
parenchyma replaced by fibrous tissue. However, the functional effect
of these lesions is not known.
Prevalence of liver lesions--
The livers of English sole from McAllister Creek and the Snohomish
River sampling sites (a total sample of 73 fish) exhibited only low
prevalences of hemosiderosis, fatty change, and non-specific hepatocellular
necrosis, and contained none of the other degenerative, potential
preneoplastic, and true neoplastic lesions found in both the Duwamish
Waterway and Lake Washington Ship Canal (Fig. 12). The prevalences of
all three of these lesion categories were significantly higher (p_<0.001)
in the Duwamish Waterway than in any of the other areas. Twelve of
the 16 major types of hepatic lesions were found only in English sole
from the Duwamish Waterway and the Lake Washington Ship Canal (551 and
49 livers, respectively, were examined). These lesions included MH, HNH,
HR, CC, MC, MDN, LCA, HC, CF, PH, HER and hemangiomas. The prevalences
of fish with neoplastic lesions ranged from 2.0% (CC) to 7.1% (LCA) in
Duwamish River English sole (overall prevalence) while in the Lake
Washington Ship Canal the average prevalence of fish with hepatic neoplasms
ranged from 0% (CC) to 4.1% (HC). For fish with non-neoplastic hepatic
lesions, prevalences ranged from 0.5% (3 of 551 fish) for peliosis
hepatis to 18.5% (101 of 551 fish) for rnegalocytic hepatosis in the
Duwamish Waterway and in the Lake Washington Ship Canal from 0 (PH) to
12.2% (MH).
Within the Uuwamish River estuary, only Stations D and G had
idiopathic hepatic lesion frequencies significantly higher (p_<0.05)
than those found at the other Stations (Fig. 13). The overall prevalence
of HN and CC was highest at Station D, and the prevalence of HNH and
Hem was highest at Station G. In general, the prevalence of sole with
liver lesions tended to be lower at Stations A and B as compared to
the other sampling stations. This observation is suggested by the
fact that the prevalence of only 2 to 3 of the 13 lesions shown in
Figure 13 were higher than 5% in sole from Stations A and B, while
prevalences of between 6 and 11 of the 13 lesions were higher than 5%
in sole from the other stations. In view of this general trend, it is
surprising that the highest prevalence of MH was found at Station A.
35
-------
ENGLISH SOLE
20
10
Duwamish Waterway
(551)
1
1
20 r
10
Lake Washington Ship Canal
(49)
ofTl I
20
10
Snohomish River
(37)
Significantly higher than the
reference area (p< 0.05}
0 | | 0 0 0 0 0
0000
20
10
McAllister Creek
(36)
Megalocytic hepatosis (MH)
Fatty change (FC)
Hepatocellular nodular hyperplasia (HNH)
Hepatocellular necrosis (HIM)
Hemosiderosis (Hem)
Hepatocellular regeneration (HR)
Cholangiocellular carcinoma (CC)
Cholangiofibrosis (CF)
Minimum deviation nodule (MDIM)
Liver cell adenoma (LCA)
Hepatocellular carcinoma (HC}
Mixed carcinoma (MC}
ihc hypertrophy (HEH)
, , Mixed carcinoma (MC)
I Hepatocellular eosmophihc
I | oooooooo
MH FCHNHHNHemHR CF CCMDN LCAHC MC HEH
Liver lesions
FIGURE 12. Prevalences of idiopathic liver lesions in English sole
from the Duwamish Waterway, Lake Washington Ship Canal,
Snohomish River, and McAllister Creek sampling areas.
36
-------
ENGLISH SOLE
Duwamish Waterway
15
30
15
n
B
(62)
Significantly lower than the average
prevalence for all stations (p<0.05)
Significantly higher than the average
prevalence for all stations (p<0.05)
o oi I I Inm
30 r-'
_ 15
(108)
30
15
(95)
n
30 r-
15 -
H
(
77)
i 1
rnor i 1 I o
n
30 p
15
(56)
30
15
_G
(46)
I
RtlUfTTTfl
MH FCHNHHN Hem HR CF CC MDN LCAHC MC HEH
Liver lesions
MH FC HIMH HN Hem HR CF CC MDN LCA HC MC HEH
Liver lesions
Megalocytic hepatosis (MH)
Fatty change (FC)
Hepatocellular nodular hyperplasia (HNH)
Hepatocellular necrosis (HIM)
Hemosiderosis (Hem)
Hepatocellular regeneration (HR)
Cholangiocpllular carcinoma (CC)
Cholangiofibrosis (CF}
Minimum deviation nodule (MDN)
Liver cell adenoma (LCA}
Hepatocellular carcinoma (HC)
Mixed carcinoma (MC}
Hepatocellular eosmophiltc hypertrophy (HEH)
FIGURE 13. Prevalences (by station) of idiopathic liver lesions in
English sole from the Duwamish Waterway.
37
-------
Prevalences of hepatic neoplasms and "preneoplastic" lesions (HR
and HEH) are presented in Figure 14. The overall prevalences of sole
with one or more of these types of lesions were essentially the same
in the Duwamish Waterway (20.%) and the Lake Washington Ship Canal
(20.4%). The overall prevalences of sole with liver neoplasms alone
from the Duwamish Waterway and Lake Washington Ship Canal were 12.9%
and 8.2%, respectively. The close association between the suspected
preneoplastic lesions and neoplasms in English sole livers is reflected
by the much higher prevalence of fish having both types of lesions,
17.2% in the Duwamish Waterway and 8.4% in the Lake Washington Ship Canal
When the hepatic lesion prevalence data for English sole from the
Duwamish Waterway are analyzed from a temporal approach, some striking
trends emerge. Table 4 depicts the temporal changes in the prevalence
of English sole with MH, HEH, MDN, LCA, and HN. The great differences
that were observed between the lesion prevalences in two winter seasons
(winter 1979 and 1980) do not appear to be accounted for by seasonal
variation alone. However, due to the relatively short duration of this
survey, the possibility of a periodic multi-year fluctuation in the
prevalence of these conditions cannot be eliminated. As one means of
determining if the prevalences of the lesions represented in Table 4
were significantly higher during the last two samplings (summer 1979
and winter 1980) than the prevalences found in the previous three
samplings, statistical comparisons were performed. The mean prevalence
of each lesion for the first three samplings (fall 1978, winter 1979
and spring 1979) was statistically compared with the prevalences for
the summer 1979 and winter 1980 samplings. The prevalence of MH and HN
was significantly higher (p_<0.05) in both the last two samplings, while
the prevalence of HEH was significantly higher only during the winter
1980 sampling. These lesions which had increased prevalences during
the last two samplings were non-neoplastic lesions thought to be acute
or subacute responses by animals to relatively short-term exposure to
toxic chemicals.
In contrast, the prevalences of all types of liver neoplasms except
one (MDN) fluctuated only slightly over the length of the study (Table 4)
MDN were significantly more prevalent during the last two samplings.
Neoplasms are thought to be a chronic response to long-term exposure to
chemical carcinogens, while MDN are thought by some to be preneoplastic
or an early stage of a neoplastic lesion (Bannasch 1976). Thus, the
increased prevalence of MDN in English sole may be indicative of the
same factors responsible for the higher prevalences of the above-
mentioned non-neoplastic lesions.
Interrelationships among idiopathic liver 1esions--The chi-square
contingency analysis was used to test the independence of the occurrence
of 16 idiopathic liver lesions in Duwamish River English sole.
Pathologic conditions that were found to be not independent may have a
temporal and/or etiological relationship since lesions involved in a
progressive sequence or induced by the same agent(s) are more likely
38
-------
DUWAMISH WATERWAY LAKE WASHINGTON SHIP CANAL
English sole with both liver neoplasms and suspected
preneoplastic lesions
30 r
10
30
20
10
-
(162)
(88)
(95)
(112)
(94)
(551)
NS
(30)
NS
(19)
NS
(49)
c
OJ
o
CU
CJ
co
O)
i_
Q.
English sole with liver neoplasms
30 r
20 -
10 -
-
30
20
10
NS
NS
NS
English sole with suspected preneoplastic liver lesions
30
20
10
-
30
20
i 1 m
NS
NS
NS
I
II III IV V annual
average
Sampling period
NS= not sampled
I II III IV V annual
average
FIGURE 14. Prevalences of English sole with suspected "preneoplastic"
liver lesions, liver neoplasms, and combined preneoplastic/
neopl astic liver lesions from the Uuwainish Waterway and
Lake Washington Ship Canal.
39
-------
o
TABLE 4. Temporal variation in the prevalence of English sole with selected liver lesions in the Ouwamish
Waterway. (See Table 1 for the number of fish examined during each sampling period).
Liver
Lesion
Megalocytic
hepatosis (MH)
Hepatocel lular
eosinophilic
hypertrophy (HEH)
Liver cell
adenoma (LCA)
Minimum deviation
nodule (MDN)
Hepatocel lular
necrosis (HN)
Prevalence (%)
Oct. 1978 Jan. 1979 Apr. 1979 July 1979 Jan. 1980 Mean
14.8 14.8 14.7 22.3* 24.3* 18.5
9.3 6.8 6.3 10.7 14.8* 9.8
7.4 3.4 8.4 4.5 11.7 7.1
0.1 0 3.2 7.1* 8.5* 3.6
7.4 3.4 11.6 17.9* 17.0* 11.3
* Signifies that the prevalence was significantly higher (p<0.05) than the mean prevalence for the first
three sampling periods.
-------
to be observed concurrently within an animal. Nevertheless, in a study
such as this one, disease etiology and the process of progression towards
neoplasia can only be inferred by statistical analysis. Controlled
laboratory exposure experiments are essential for establishing such
relationships.
The results of the chi-square contingency tests are listed in
Table 5. This table also contains values for Pearson's coefficient of
contingency (PC), a measure of the degree of association between two
variables (Ostle 1972). The strongest associations were generally found
between the separate categories of neoplastic conditions. For example,
a fish liver with a HC was very likely to possess a CC (HCxCC, PC=.33,
p<0.n01), or a mixed carcinoma where the neoplastic elements were not
separable into distinct tumors (HCxMC, PC=.2, p<0.005). The neoplasms
judged to be malignant by histologic criteria were also highly associated
with their benign counterparts, (HCxLCA, PC=.27, p<0.001; CCxLCA PC=.16,
p<0.05).
Hepatocellular lesions which have been suspected to be preneoplastic
(HEH) were in fact highly associated with hepatocellular and non-hepato-
cellular neoplasms, (HEH vs. LCA, PC=.31, p<0.001; HEH vs. HC, PC=.22,
p<0.005; HEH vs. MC, PC=0.19, p<0.01; HEH vs. hemangioma, PC=.16,
p<0.025). HEH was also associated with a potential progenitor lesion,
fatty change or clear cell foci (Jones and Butler 1975, Squire and Levitt
1975; HEH vs. FC, PC=.25, p<0.001). Another potential preneoplastic
hepatocellular lesion, HNH, was found to occur in statistically significant
association with a neoplasm, the mixed carcinoma (HNH vs. MC, PC=.14,
p<0.05).
Among the degenerative lesions, several showed high associations
with other hepatic lesions. In MH, widespread degeneration of the
hepatocellular parenchyma was usually accompanied by compensatory HR
(MH vs. HR PC=.27, p<0.001). This relationship was the single significant
one for megalocytic hepatosis. In addition to MH, HR occurred with
parenchymal fibrosis (HR vs. fibrosis, PC=.23, p<0.001) and with
hemangiomatous lesions (HR vs. hemangioma, PC=.16, p<0.025).
Table 5 presents the entire matrix of liver lesions which were
compared by the chi-square test of contingency. Also of significance
in these comparisons are lesions which show the least association,
implying independent etiologies or events that are not closely related
within a certain time frame. Examples of lesions which show low
association and independence are degenerative conditions compared with
"preneoplastic" lesions (HN vs. HEH) and the specific degenerative
condition negalocytic hepatosis (MH) compared with the supposed chronic
lesion adenofibrosis/ chlolangiofibrosis (AF/CF).
Analysis of idiopathic hepatic lesions in Duwamish River English sole
with respect to sex--0f the total number of English sole captured in
the Duwamish Waterway, which were examined histologically, and for
whom the sex was determinate (421 in the total sample), 198 or 47%
41
-------
TABLE 5. Matrix of idiopathic liver lesions which were compared in order to determine the degree of association
of lesions by chi-square contingency analysis in individual English sole from the Duwamish Waterway
system. Values shown indicate Pearsons coefficient of contingency (PC) and the level of statistical
significance (p value) respectively.
-p.
ro
Lesion MH
Type
HR
(PC). 27
p<001*
CC
.11
.25
Hem
.10
.25
LCA
.10
.25
HNH
.08
.25
Hem
.06
.5
HC
.05
.5
PH
.04
.75
MDN
.04
.75
HN
.03
<.75
HR
MH
.27
.001*
Fibrosis
.23
.001*
Hem
.16
.025*
MC
.10
.25
HEM
.09
.25
AF/CF
.08
.5
PH
.07
.5
HC
.06
.5
HNH
.05
.5
LCA
.04
.75
HEH
LCA
.31
.001*
FC
.25
.001*
AF/CF
.23
.001*
HC
.22
.005*
PH
.20
.005
MC
.19
.01*
Hemangioma
.16
p<.025*
Fibrosis
.10
.25
HR
.09
.25
Hem
.08
.5
HN
Hemangioma
.15
.05*
AF/CF
.13
.10
Fibrosis
.11
.25
LCA
.05
.5
HC
.05
.5
HNH
.05
.5
MC
.04
.75
HR
.04
.75
MH
.03
.75
CC
.03
.75
FC
HEH
.25
.001*
Hem
.16
.05*
Fibrosis
.13
.10
LCA
.12
.10
HNH
.10
.25
HC
.10
.25
AF/CF
.09
.25
MC
.06
.5
Hem
FC
.16
.05*
HC
.11
.25
MC
.09
.25
MDN
.09
.25
LCA
.08
.25
Hemangioma
.08
.5
HEH
.08
.5
Fibrosis
.07
.5
Hemangioma CC
.06
.5
CC
.03
.75
.06
.5
MH
.06
.5
HNH
MC
.14
.05*
MDN
.12
.10
FC
.10
.25
MH
.08
.25
CC
.07
.5
HEH
.06
.5
HR
.05
.5
Fibrosis
.05
.5
HN
.05
.5
LCA
.04
.75
LCA
HEH
.31
.001*
HC
.27
.001*
CC
.16
.05*
PH
.15
.05*
FC
.12
.10
Fibrosi
.11
.25
AF/CF
.11
.25
MH
.10
.25
Hem
.08
.25
MC
.08
.25
HC
CC
.33
.001*
LCA
.27
.001*
HEH
.22
.005*
MC
.20
.005*
PH
.19
.01*
s AF/CF
.15
.05*
MDN
.11
.25
AF/CF
Hemangioma
.33
.001*
Fibrosis
.23
.001*
HEH
.23
.001*
MC
.20
.005*
HC
.15
.05*
HN
.13
.10
LCA
.11
.25
Hemangioma FC
.11
.25
Hem
.11
.25
FC
.10
.25
.09
.25
HR
.08
.5
MDN
.05
.5
CC
HC
.33
.001*
Fibrosis
.19
.01*
LCA
.16
.05*
PH
.15
.05*
MDN
.14
.1
MH
.11
.25
MC
.10
.26
HNH
.07
.5
Hem
.06
.5
HEH
.05
.5
* Signifies a significant level of association.
-------
were males, and 223 or 53% were females. In order to determine if any
hepatic lesion or lesion category was found disproportionately in fish
of either sex, the percentage of male and female English sole affected
with a particular hepatic lesion was computed for each lesion category
(combined hepatic lesions, MH, HEH, etc.), and this value was compared
to the percentage of males or females found in the general sample.
Testing the null hypothesis that all hepatic lesions are unrelated to
sex, the G-statistic was computed for each lesion category. Even though
the percentage of males and females affected by certain lesions deviated
from the sex ratio seen in the general sample, the null hypothesis
could not be rejected in any instance. Therefore, these deviations
from the percentages seen in the general sample are not statistically
significant, and it can be concluded that neither sex is affected
disproportionately by any of the hepatic lesions analyzed.
Length-age regressions in Duwamish River English sole with and without
idiopathic liver lesionsTesting the hypothesis that hepatic disease in
English sole may affect the length at a particular age, and therefore
growth, length-age regressions were developed for (1) Duwamish English
sole without hepatic lesions, (2) Duwamish English sole with hepatic
lesions (including all non-parasitic hepatic lesions), and (3) Duwamish
English sole with specific hepatic lesions. For each group, regression
curves were fitted for males, females, and for the entire sample composing
each group. A graphical representation of one of these power curve
regressions is presented in Figure 15.
Comparing Duwamish English sole without hepatic lesions to Duwamish
English sole with non-parasitic hepatic lesions (Fig. 15), it is
graphically evident that when both sexes are combined, there is a slight
reduction in length-at-age for fish affected with liver lesions. This
reduction in length is 2-3% for all ages. However, when the two samples
are compared using the paired sample t-test (comparing the mean length-
at-age for each age class in both samples), there was statistically no
significant difference between the two samples.
Considering the series of length-age plots for English sole with
specific hepatic lesions, there was no clear pattern of length reduction
at all ages. Fish with HN, Hem, FC, MH, HEH, LCA, CF, CC, and the
grouped "preneoplastic"/neoplastic lesions, graphically showed reduced
length in older fish, generally beyond age 4 or 5. This reduction
approached 16% in some lesions at the 10-year age level (as for HEH).
Only the regression curves for MH and CC showed reduction in length
for all ages. However, in comparing mean length-at-age for normal
Duwamish River English sole against mean-length-at-age for English
sole with specific hepatic lesions (at all overlapping ages) by the
paired sample t-test, no significant difference was observed in any of
the affected groups tested. Therefore, although the regression curves
suggested a reduction in length at a particular age for English sole
affected with idiopathic liver lesions, this apparent trend could not
be affirmed statistically over the entire age range in which English
sole were affected.
43
-------
4001-
350 -
300
250
cn
200
150
100
50
(4)
(13)
(7)
o _
(5)=
Regression curves
without lesions
with lesions
mean length-at-age
for normal sample
(13)
mean I
for diseased sample
number of specimens
composing mean
length-at-age value
10
Age (years)
FIGURE 15. Length-age regression curves and mean length-at-age values
for Uuwarnish River English sole with and without idiopathic
hepatic lesions (both sexes combined).
44
-------
Prevalence of hepatic lesions related to length (and indirectly to
age) in English sole and effect of hepatic lesion presence on the
length-weight relationship In Duwamish River English soleAlthough the
documentation of age data for our entire sample is not yet available,
the parameter of length is strongly related to age, and therefore hepatic
lesion prevalences in English sole were examined in three separate size
groups, _<200 mm, 201-300 mm, and _>301 mm. From the age data presently
available, English sole from the first size group (<200 mm) are generally
1-3 years of age; English sole from the second group (201-300 mm) are
2-6 years of age; and English sole from the third group (J>301 mm) are
4-13 years of age. The fact that idiopathic hepatic lesions are clearly
age-related is demonstrated in the hepatic lesion prevalence for these
size groups. The first group (_<200 rnrn) show a prevalence of 40/279 or
14.3% (affected animal prevalence), the second (201-300 mm), a hepatic
lesion prevalence of 55.3% (83 of 150), and the third size group (_>301 mm),
a prevalence of 67.3%.
To investigate the possible affect of these hepatic lesions on the
general health of English sole, simple length vs. weight regression
curves (based on the power function y=ax'3 where y=weight and x=length)
were fitted for Ouwamish English sole which were affected with idiopathic
liver lesions and for those that were not affected with idiopathic
liver lesions. Similar curves were fitted (again, by the least-squares
method) for English sole from the Lake Washington Ship Canal, Snohomish
River and McAllister Creek sampling sites. The goal of this analysis
was to find a type of "emaciation index" for each group of fish to
determine if fish affected with liver lesions tended to have a reduced
weight at a particular length.
The sample of normal fish was composed mainly of smaller fish
(_<200 mrn) and, thus, the length-weight regression was skewed because of
tTie disproportionate number of small normal fish compared to the sample
of English sole affected with liver lesions. Since this effect tends
to represent older, longer fish, separate length-weight regressions
were performed for the three separate length groups described previously.
In the first length group the relationship between length and
weight in normal fish and those with idiopathic liver lesions was obscure.
The curve for the normal Duwamish sample was almost linear up to 130 mrn,
indicating that normal sole had a higher weight than diseased Duwamish
English sole at corresponding lengths. For fish longer than 130 mm,
this relationship was reversed; diseased fish had higher weights than
their normal counterparts. Diseased Duwamish English sole and normal
Lake Washington Ship Canal English sole showed almost identical curves,
and there were the anomalous results of lower weights for normal Snohomish
and McAllister Creek English sole as compared to their diseased Duwamish
English sole counterparts. Because of the conflicting, shifting, and
anomalous patterns which emerged from these curves, there is probably
no effect upon weight in fish within this size range due to liver
lesion presence.
45
-------
In the second length group, again there was no clear distinction
between groups of normal fish from different sampling areas (except that
Snohomish normal group showed a reduced weight compared to all other normal
and diseased groups), nor was there any graphic difference between normal
and diseased Duwamish English sole.
The third length group represented the older fish sampled, and
therefore included a larger sample of English sole with liver lesions
(74 as opposed to 36 normal Duwamish English sole). Although the
differences in weight at a particular length between different curves
were not statistically significant, the weight of Duwamish English
sole with liver lesions at any length tended to be lower than any of
the other normal groups as determined by this simple regression
(Fig. 16). This suggests that liver lesions in English sole adversely
affect either food conversion efficiency, the ability to compete for
food, or general nutritional status, especially in older English sole.
It should be reemphasized here that this apparent reduction in weight-
at-length is only a trend visible in the larger length group and that
these weight differences may not be statistically significant. In
fact, when the data for English sole from the Duwamish River with
idiopathic liver lesions is compared to similar data for English sole
without these liver lesions, using the paired sample t-test comparing
weight-at-length between the two samples, no statistically significant
differences were found.
Characteristics of gill lesions--
Several morphologically separable gill lesions of an idiopathic,
non-parasitic nature, as well as parasitic infestations were found in
English sole.
Respiratory epithelial hyperplasia (REH)--The most common gill
lesion, REH, consisted of a clearly hyperplastic or thickened respiratory
epithelium of the gill lamellae, especially in the distal regions of
the lamellae. In sole with REH, the superficial respiratory epithelium
was markedly thickened, often with accompanying lamellar fusion,
hyperplasia of the filament epithelium, and edema in both the lamellae
and gill filaments (Fig. 17). The lesion was generally multifocal to
diffuse. In this condition, respiratory exchange may be compromised
due to the impaired capacity for gaseous exchange through the thickened
epithelium.
Lyrnphocytic infiltrationA second idiopathic lesion was characterized
by multifocal sites of lytnphocytic and histiocytic infiltration in the
subepithelial regions of the gill lamellae and filaments. There was
generally no accompanying necrosis, and this condition can be interpreted
as a non-specific chronic inflammatory response to injury. Response to
parasites, which were commonly observed in English sole gill tissue,
cannot be ruled out.
46
-------
7001-
650 -
with lesions
250
200
I
320 340 360 380 400
Length (mm)
420
440
460
FIGURE 16. Length-weight regression curves for Duwamish River
English sole with and without idiopathic hepatic lesions
(least squares method).
47
-------
FIGURE 17. Micrograph of an English sole gill with severe respiratory
epithelial hyperplasia with fusion of adjacent lamellae.
H 5 E, 240X.
48
-------
Microaneurysm--A third common idiopathic gill abnormality was
represented by multiple sites of intralamellar capillary dilations,
microaneurysms or telangiectasis. These occurred in absence of hemorrhage,
but these dilated capillaries often contained fibrin accumulations
indicative of a coagulation disorder or a tendency towards thrombosis,
possibly from endothelial injury. Most commonly, though, the lesion
consisted of simple capillary dilation and congestion, possibly associated
with the shock response induced in capture.
Miscellaneous gill 1esions--0ther idiopathic lesions found at low
levels only in the urban-associated estuaries were: (1) mucous cell
hyperplasia (an increase in the density of goblet cells in the superficial
surfaces of the lamellae and filaments); (2) lamellar necrosis (a non-
specific degeneration and necrosis of the lamellar epithelium, and the
capillary endotheliurn supporting the pillar cells and the basement
membrane); (3) mixed capillary dilation and congestion (a form of
lamellar microaneurysms, but without actual ballooning and severe
dilatation of lamellar capillaries seen in microaneurysms); (4) edema
and necrosis of filament epithelium; and (5) hyperplasia of the pillar
cells forming the supportive framework of the lamellae, usually with
thickening of the peripheral basement membrane.
Parasitic conditionsA broad spectrum of protozoan and metazoan
parasites were detected in and adjacent to the gill tissue of English
sole with varying degrees of host response. Due to the general
non-pathogenic character of these infestations (external presence) and
infections (internal presence) and the difficulty of identifying these
parasites in tissue section to genus or species, taxonomic classification
was usually limited to general taxonomic groups.
The most common parasite observed was the ciliate ectoparasite
Trichodina sp. present adjacent to the filament and lamellar epithelium.
The major host response was an excessive mucous secretion by the
epithelial goblet cells when the parasite was present in denser
concentrations. Generally, however, the infestation was light, with no
significant lesions or host response.
An infectious condition identical morphologically to epitheliocystis
in fresh water and marine fish (Hoffman et al. 1969, Zachary and Paperna
1977, Paperna 1977, Paperna et al. 1978) was found in the lamellar
respiratory epithelium. The condition is an intracellular infection of
epithelial cells by organisms resembling members of the Chlamydia/Bedsonia
groups (Hoffman et al. 1969) and Rickettsia (Zachary and Paperna 1977,
Paperna et al. 1978). Affected cells were greatly hypertrophied and
contained a large cytoplasmic inclusion body consisting of a nass of
uniformly sized basophilic particles. The nucleus, when visible, was
greatly enlarged with a prominent nucleolus. These affected cells
often reached a diameter exceeding 50 microns. The only visible host
response was occasional epithelial hyperplasia encircling the infected
cells in concentric layers. Heavy infections were rare. This condition
is considered to be a chronic, benign, ubiquitous, infectious disease
with no significant deleterious affect on the host.
49
-------
A trematode (blood fluke) infection of the filaments and lamellae
was also common in English sole from all regions sampled. Multiple
larval forms of schistosorna-1 ike, sanguinicolid trematodes were observed
within the subepithelial connective tissue and blood vessels of gill
filament shafts. These immature adults lacking reproductive organs
(Chitwood and Lichtenfels 1973) elicited a variable fibrotic response in
the filaments, but there was no serious pathology associated with this
infection. Other parasitic conditions of the gill were as follows:
(1) presence of unidentified helminth ova in the filament shaft, often
with hyperplasia of surrounding chondrocytes composing the skeletal
framework of the filament shaft; (2) ectoparasitism by a member of
the group monogenea, possibly Gyrodactylus sp. (Olson 1978); and
(3) infestation by copepods. None of these latter conditions were
associated with any significant lesions.
Prevalence of gill lesions--
The prevalences of five of the idiopathic gill lesions and two
parasitic conditions in English sole are shown in Figure 18. Respiratory
epithelial hyperplasia was observed only in sole from the three urban
estuaries; while lymphocytic infiltration was detected in sole from all
four estuaries, but the frequencies in the urban estuaries were
significantly higher than in McAllister Creek. Mucous cell hyperplasia
was observed only in the Uuwamish Waterway; and although microaneurysms
were found in sole from two of the urban estuaries as well as the
reference estuary, the prevalence in sole from Duwamish Waterway was
significantly higher than that in the reference estuary. Lamellar
necrosis affected only small numbers of English sole from the Duwamish
Waterway and Snohomish River.
Two of the most prevalent parasitic conditions, epitheliocystis
and infection by digenetic trematode larvae, were detected in English
sole from all four areas. Only sole from the Duwamish Waterway had
frequencies significantly higher than the sole from McAllister Creek.
Trichodinids were found in the gills of sole from all sampling areas
(Fig. 18).
A percentage of English sole from all areas sampled were free from
microscopically detectable idiopathic and parasitic lesions in the
sections examined. These "normals" (in terms of condition of the gill)
were encountered most often in McAllister Creek (64.7%), followed by
the Snohomish River (51.5%), Lake Washington Ship Canal (45.9%), and
Duwamish Waterway (36.5%). The Lake Washington Ship Canal and Duwamish
Waterway samples contained significantly fewer "normals" than the
reference area (p<0.05 and 0.001, respectively), while the Snohomish
River sample was not statistically different from the reference value.
Conversely, English sole from the Duwamish Waterway and Lake Washington
Ship Canal were significantly more frequently affected with idiopathic
and/or parasitic conditions than English sole from either the Snohomish
area or the McAllister Creek reference site.
50
-------
ENGLISH SOLE
Duwamish Waterway
30r- <542>
15
15
Lake Washington Ship Canal
(37)
0 0
Snohomish River
301-(33)
30
15
McAllister Creek
r<34)
50
25
Significantly higher than
the reference area (p< 0.05)
0 0
REH LI
MA MCH LN
Gill lesions
EC TL TRIC
Respiratory epithelial hyperplasia (REH)
Lymphocytic infiltration (LI)
Microaneurysm (MA)
Mucous cell hyperplasia (MCH)
Lamelar necrosis (LN)
Epitheliocystis (EC)
Trematode larval infection (TL)
Trichodmid infestation (TRIC)
n
50 r <47>
25
50r(16)
25
uu
25
- (^u;
-
n
0 00
ML MS TBPL IF
Kidney Lesions
TN
Mesangiolysis (ML)
Mesangiosclerosis (MS)
Thickened peripheral basal lamina (TBPL)
Interstitial fibrosis (IF)
Tubular necrosis (TN)
FIGURE 18. Prevalences of gill and kidney lesions in English sole from
the Duwamish kiaterway, Lake Washington Ship Canal,
Snohomish River, and McAllister Creek sampling areas.
51
-------
Characteristics and prevalence of cardiac abnormalities--
Cardiac tissues were not routinely sampled until the summer of
1979. Since only the Lake Washington Ship Canal and Duwamish Waterway
were sampled in this and the winter 1980 sampling, the following
descriptions will present data only from these areas, precluding
other inter-area comparisons.
The most frequently encountered cardiac lesion was multifocal
chronic inflammation of the myocardium (chronic myocarditis), typically
with a prominent mononuclear infiltrate, significant fibrosis and
occasional myocardial degeneration/necrosis. The overall prevalence for
this lesion in the Duwamish Waterway was 18.9% (Table 6), with a general
trend of increased prevalence from downstream stations to the upstream
station.
This lesion was most often found in conjunction with an infection
of sanguinicolid trematodes identical to those found in gill tissues.
Occasionally, no inciting agent was visible at the light microscopic
level. The prevalence of this trematode infection also tended to
increase from the lower to upper Duwamish Waterway sampling stations,
and showed an overall prevalence of 11.4% (Table 6). The infection was
variable in severity, with significant fibrosis and chronic inflammatory
cell infiltration peripheral to the parasites in the more severe cases.
Another common lesion was a non-specific idiopathic epicarditis,
evidenced by multifocal aggregations of mononuclear cells in the
epicardiurn, with increased rnelano-macrophage center populations, edema,
and occasional congestion. The overall prevalence in English sole from
the Duwamish Waterway was 7.0% (Table 6), with no general trend in the
prevalences at individual stations. The etiology of epicarditis is
not clear, but the occasional presence of philometrid nematodes in the
pericardial cavity in English sole at gross necropsy may provide some
insight into the cause of this incidental finding.
Characteristics and prevalence of gastrointestional abnormalities--
The major anomalies detected histologically in the gastrointestional
tract were parasitic or parasite-related, and since the bulk of the
gastrointestional tissues examined were from the small intestine
immediately distal to the pyloric caeca, this section will focus on the
anomalies found at that anatomical site. The most common parasitic
infection was a protozoan infection by an unclassified coccidian,
generally present adjacent to or embedded within the brush border of
the intestinal mucosa. Often, however, histozoic forms of this undefined
coccidian were present within the mucosa, and morphologies parallel to
life history stages of coccidians were encountered. The host response
in this infection was minimal or absent, and the agent is considered
non-pathogenic. At the light microscope level this coccidian closely
resembled members of the genus Eimeria, a common coccidian parasite of
marine and fresh water fish (Margolis and Arthur 1979, Hoffman 1970).
52
-------
01
GO
TABLE 6.
Lesions in selected organs and tissues of English sole.
Prevalence (%)
Organ
Heart
G.I.
Tract
Kidney
Lesion Description
Chronic myocarditis
Trematode infection
Idiopathic epicarditis
Coccidian infection
Nematode infection
Acanthocephala infection
Nonspecific nucosal necrosis
Ulcerative gastroenteritis
Thickening of the basement
membrane
Tubular protein casts
Mesangial cell proliferation
Myxosporidia infection
Duwamish
Waterway
18.1(105)a
11.4
7.0
45.0(398)
35.9
4.3
1.8
0.8
3.3(429)
3,3
0.5
3.7
Lake Wash.
Ship Canal
22.2(18)
22.2
0
28.0(25)
32.0
4.0
0
0
6.4(17)
0
2.1
14.9
Snohomish
River
NDb
ND
NO
11.1(18)
38.9
22.2
0
0
6.3(16)
6.3
0
0
McAllister
Creek
ND
ND
ND
38.1(21)
38.1
47.6
0
0
0(20)
0
0
5.0
a Number of fish examined.
b ND, no cardiac tissue from English sole were examined.
-------
Prevalences of sole with coccidian infections ranged from 11.1%
(2 of 18 fish) in the Snohornish River to 45% (179 of 398 fish) in the
Duwamish Waterway (Table 6). By the G-statistic, the prevalence
of this infection was significantly higher (p<0.01) in English sole
captured in the Duwamish Waterway.
Another protozoan infection, confined to the intestinal lumen and
also apparently non-pathogenic, was parasitism by flagellates of the
Hexamita sp., a member of the order Polymastigina (Hoffman 1970).
Affected English sole were found only in the Duwamish Waterway (2.5%,
20 of 398 fish).
Representatives of several helminth groups parasitized all layers
of the alimentary tract wall, and were often detected within the gut
lumen. The typical host response to the intramural helminths was
chronic inflammation, fibrosis, and granuloma formation. Unspecified
members of the class Nematoda were the most commonly encountered helminth
groups, and larval and adult forms were found. The larval forms tended
to be confirmed to the mucosa, lamina propria, and submucosa, while adult
forms were present in the lumen, submucosa, the muscular wall, and
supraserosally in the mesenteries. Undoubtedly several species were
represented, with the larval forms possibly Contracaecum sp. (Olson
1978). In the study of English sole reported by Olson (1978), adult
intestinal nematodes were described as Cucullanus annulatus and
Thynascaris sp. Infection by acanthocephalids was also encountered in
al1 four areas, generally with penetration of the spiny head through
the mucosa and into the muscular layers, accompanied by a
fibrogranulomatous response. If our findings are parallel to those of
Olson (1978), the probable species is Echinorhynchus lageniformis.
English sole with the above-mentioned helminth infestations were found
in all of the estuaries (Table 6).
Several lesions apparently unassociated with parasites were detected
only in the intestines of English sole from the Duwamish Waterway. These
were found at very low prevalences (0.8 to 1.8%) (Table 6), and included
non-specific necrosis of the mucosal epithelium, and three cases of chronic
ulcerative gastroenteritis.
Characteristics and prevalences of renal anormalities--
Mesangiolysi s--The most striking idiopathic renal condition found in
the English sole kidney was a primary, degenerative, glomerular lesion
which closely resembles mesangiolysis observed in humans and experimental
mammals (Morita et al. 1978). This lesion was characterized histologically
by varying degrees of cystic dilatation and ballooning of the glomerular
tuft, loss of architectural integrity, dilation and/or congestion of
the capillary component, with edema and lysis of the mesangium (the
structural framework of the glomerular tuft, Fig. 19). Often there
was a generalized necrosis of all glomerular tuft components, and
fibrin and protein casts could be detected in the Bowman's space. In
the more severe cases, only the visceral epithelium and peripheral
54
-------
FIGURE 19. Micrograph of the renal lesion, mesangiolysis, in an English
sole. This section of kidney shows two glorneruli with
ballooning of the glonierular tuft and loss of mesangial
architecture, both of which are characteristic of mesangiolysis.
H 8 E, 240X.
55
-------
basal lamina remained. The endothelial cells lining the capillaries
were often swollen and occasionally were exfoliated. This glomerular
lesion occurred in the absence of any visible tubular nephropathy, and
inflammatory cell infiltrates were rarely present. Mesangiolysis was
found at highest prevalences in English sole from McAllister Creek and
the Duwamish Waterway (Fig. 18). No affected fish were detected in
the Snohomish River. Fish as small as 100 run were affected, but this
condition was more prevalent in older fish.
Mesangiosclerosis--Hesangiosclerosis was commonly found concommitantly
with mesangiolysis. Mesangiosclerosis is defined as an increase in the
mesangial matrix of the glomerular tuft (Fig. 20). The matrix is composed
of fine fibrils including collagen, all embedded in a mucopolysaccharide
ground substance. When riesangiosclerosis and rnesangiolysis were found
together, one or several segments of a mesangiolytic glornerulus exhibited
mesangial thickening and increased matrix density in conjunction with
proliferation of mesangial cells. Rarely was the peripheral basal lamina
thickened. Occasionally, the visceral and parietal epithelial components
were proliferative in tnesangiosclerotic glomeruli, with formation of
synechiae or adhesions involving these inner and outer epithelial com-
ponents of the glomerulus. Sclerosis in Bowman's capsule, peripheral to
the parietal epithelium, was also sometimes observed. It is not possible
to definitely prove the temporal relationship between rnesangiolysis and
mesangiosclerosis in a static study such as this, but principles of
reaction to injury in the glomerulus suggest that mesangiolysis precedes
mesangiosclerosis.
Cases of coincident mesangiolysis and mesangiosclerosis were found
only in the Duwamish Waterway and McAllister Creek, representing over
20% of the total number of cases of mesangiolysis and mesangiosclerosis.
Fish affected by both of these lesions also tended to be older. Using
a chi-square contingency test of association, mesangiolysis and mesangio-
sclerosis were found to be highly associated (p<0.001).
Mesangiosclerosis was often detected independent of mesangiolysis,
and in these cases this lesion may represent a separate pathogenetic
entity. Sclerosis of the mesangium and the glomerular tuft is a common
response to injury (Schillings and Stekhoven 1980) in the glomerulus
and is seen in a variety of chronic renal disease states. Similar
increases in mesangial matrix with mesangial cell proliferation are also
associated with increasing age in humans (Wehner 1968). Mesangio-
sclerosis was found in all sampling areas, with highest prevalences in
the Duwamish Waterway, Lake Washington Ship Canal, and McAllister Creek
(Fig. 18).
Miscellaneous idiopathic renal 1esions--0ther renal lesions included
the following:(1) thickening of the basement membrane peripheral to the
neck segments and tubules, often with peritubular fibrosis; (2) degeneration
and necrosis of the tubular epithelium of the first and second proximal
segments of the nephron (Duwamish Waterway only, 3.3%, 14 of 429 fish);
(3) thickened peripheral basal lamina in the glomerular tufts (membranous
56
-------
FIGURE 20. Micrograph of the renal lesion, mesangiosclerosis, in an
English sole. This micrograph of a renal glomerulus demonstrates
a markedly increased density in the mesangial matrix characteristic
of mesangiosclerosis. H & E, 975X.
57
-------
glomeallopathy, Figs. 18 and 21); (4) presence of protein casts in the
renal tubular lumina (indicative of proteinuria) (Table 6); (5) mesangial
cell proliferation (not in association with other renal lesions, Table 6);
(6) fibrosis of the interstitial supporting framework of the kidney
including fibrosis peripheral to Bowman's capsule (Duwamish Waterway
only, 1.6%, 7 of 429 fish); (7) proliferation of the visceral and
parietal epithelium of the glomerulus (Duwamish Waterway only, 0.7%,
3 of 429 fish); and (8) squanous metaplasia of the tubular epithelium
(change from a columnar-cuboidal epithelial type to a more squamous
profile) (Duwamish Waterway only, 0.5%, 2 of 429 fish).
Association of idiopathic kidney lesions with idiopathic
liver lesions in English soleTesting the hypothesis that specific
idiopathic kidney lesions are associated with idiopathic liver lesions,
the chi-square contigency test (Zar 1974) was applied to the combined
data for all quarters, with the following results: Grouped idiopathic
liver lesions are highly associated with (1) mesangiolysis (p<0.01);
(2) rnesangiosclerosis (p
-------
"f ^ ^ As
**?* ^~'
^ *
*
FIGURE 21. Micrograph of severe thickening of the peripheral basal lamina
in the glomerular tuft in the kidney of an English sole.
The renal glomeruins in the center has a severely thickened,
dense peripheral basal lamina in the glomerular tuft,
characteristic of membranous glonerulonephritis. H & E, 975X.
59
-------
TABLE 7. Lesions in selected organs and tissues of starry flounder.
Organ
G.I.
Tract
Kidney
Lesion Description
Chronic inflammation
Submucosal congestion
Cocci di a infection
Hexamita infection
Neinatode infection
Trematode infection
Acanthocephala infection
Thickening of the basement
membrane
Tubular protein casts
Hesangial cell proliferation
Myxosporidia infection
Microsporidia infection
Uuwamish
Waterway
10.2(225)*
1.3
50.7
2.2
14.7
2.2
2.7
0.8(243)
2.1
0.4
57.6
2.5
Prevalence (%)
Snohomish
River
77.8(9)
0
22.2
0
44.4
11.1
55.6
0(8)
0
0
25.0
0
McAllister
Creek
12.1(33)
0
18.2
0
18.2
3.0
18.2
3.3(30)
0
0
6.7
0
* Number of fish examined.
-------
flounder from the Duwamish Waterway. A syndrome of the fins and skin,
referred to as fin erosion, was found only from the Uuwamish Waterway
specimens, and was characterized by all or a few of the following lesion
components: epidermal hyperplasia, subcutaneous chronic inflammation
with fibrosis, subcutaneous hemorrhage, epidermal inclusion cysts and
fin ray resorption (Wei lings et al. 1976b). In cases of actual fin
ray loss, the remaining unsupported fin tissue had folded back onto
the adjacent skin overlying the body musculature, creating a fusional
graft of tissue united by scar tissue. Microscopically confirmed cases
of this syndrome were found in 2.9% (8 of 280 fish) of the flounder
from the Duwamish Waterway..
Minor parasitic infestations of skin and fins were found and they
consisted of subcutaneous nematodal microfilariae and an intramuscular
myxosporidian infection resembling Kudoa sp.
The final lesion group has been discussed previously for English
sole, namely angioepithelial nodules (AEN), transitional AEN (TAEN),
and epidermal papillomas (EP). Aside from a single fish from McAllister
Creek with an EP, all other specimens affected by these lesions were
found in the Duwamish Waterway at the following frequencies: AEN, 1.1%
(3 of 280 fish); TAEN, 0.4% (1 of 280 fish); and EP, 4.6% (13 of 280 fish).
Characteristics and prevalence of gill anomalies--
Table 7 presents the types and prevalences of parasitic and non-
parasitic (idiopathic) conditions of gill tissue in starry flounder.
These conditions closely resembled the lesions detected in English sole.
The idiopathic lesions (REH, lamellar lymphoid infiltrates, lamellar
microaneurysms, respiratory epithelial degeneration and necrosis,
filament epithelial degeneration and necrosis, pillar cell hyperplasia
with increased basement membrane thickness, and mucous cell hyperplasia)
were generally found in starry flounder from all sampling areas. The
prevalence of flounder with lymphoid infiltrates and microaneurysms was
significantly higher (p<0.05) in the Duwamish Waterway, compared to
McAllister Creek (Fig. 22).
Flounder with infectious conditions (surface trichodiniasis,
epitheliocystis, intravascular microsporidia, sanguinicolid-1ike digenetic
trematodes and external monogenea) were found at low prevalences and
were generally widely distributed (Fig. 22). The exceptions were:
the absence of trichodiniasis in Snohomish River samples; absence of
epitheliocystis in McAllister Creek starry flounder; presence of
microsporidia only in specimens from the Duwamish Waterway; and absence
of external monogenea in Snohomish River specimens.
Characteristics and prevalence of liver lesions--
Many of the same types of lesions observed in livers of English
sole from the Duwamish Waterway were also found in livers of starry
flounder from this estuary. Those lesions not detected in starry
61
-------
en
TO
m
ro
IV)
^ 3 °> f -< «"
(C r "^ B~3^o~'^"n 30
3 0> ft a~ ^ "5 §"^"2 !J
^SfD 5'2£o="|".g ~
QJ ' ro ?i ^ ^ r~"O r? ^ ^
co ST *i H § ^ ^_-a CD" I
Ca 3J_ o "" -
??"= 0^ X ="2. 0 £
ID -J« ^ 1 . Q) j 2
2 ' iL m
O) 30 O
"^ X _,
Q- JJ
CO
_J A _|
O 3 °
LO
^- «^
!?!2 s-| s
< o' S"^^
^ = i o' 5'
"' 1"^ ?; P
* = -5'Ss: §
3 3- «' £ (D
Q- CO (0 ^ < g
rt 3 S CO
2C? " 3 ~*
' h =i |
co 5-
CT O W
? § H
Q- ?
o ro r-
-3-3
ro
ro -h
o
3
» |O CO Ji» C71
o o o o o
1 ' ' ' '2
w ^
CD
s
J 3 ">. re 5S
o 3 q; 3 5;
= o1 = 5'
8 § 8 3
O o> £. oi r*_
CD "< CD "<
Pri P'*
§1 2 =
cn o
1 ^^
2
p
Prevalence (percent)
-> NO CO -P* C71
O O O O O
co
o
0
o ^__ <'
Prevalence (percent)
in o
_
00
o
o
o
' NJ CO *> CJI
O o O O O
1 III I0 CO
C )i
N) S 30
^ ^1 i
^WS^ r? 0
^SSSSSNS^ ff £
o § 3
tJl o
1 l_
$
o
ff)
]
-------
flounder included CF and HC. With the exception of the intrahepatic
blood cysts (peliosis hepatis), the prevalences of all the lesions
common to both species were considerably lower in starry flounder
(Fig. 23). The blood cysts were found in only 0.5% (3 of 551 fish)
of the English sole but in 5.0% (14 of 279 fish) of the starry flounder
from the Duwamish Waterway.
Of the other three estuaries sampled in this study, sufficient
starry flounder for comparative purposes were captured only in McAllister
Creek. Only 3 of the 11 types of liver lesions observed in flounder
from the Duwamish Waterway were also detected in this species in
McAllister Creek (Fig. 23). The prevalences of two of these lesions,
FC and HN, were not significantly different, while the prevalence of
blood cysts in flounder from the Duwamish Waterway was significantly
higher (p<0.05).
Characteristics and prevalence of gastrointestinal abnormalities--
The lesions and parasitic conditions detected in the gastrointestinal
tract of starry flounder were similar to those already described for
English sole. Table 7 presents the types and prevalences of these conditions.
The most common non-parasitic condition was chronic inflammation
with lymphoid infiltrates, fibrosis, and granuloma formation generally
confined to the subrnucosa and lamina propria. Although this condition was
not associated with visible parasitic infection, some of the cases may
have represented a post-parasitic healing response. This may explain the
much higher prevalence found in starry flounder from the Snohomish River;
these fish also had a high prevalence of helminth infections (nematodes and
acanthocephalans). The only other non-parasitic lesions were submucosal
congestion (found only in the Duwamish Waterway at a low prevalence)
and a single case of mucosal necrosis in a Snohomish River specimen.
The parasitic conditions were represented by protozoan (unidentified
coccidia and the flagellate, Hexamita s_p.) and helminth (nematodes,
trematodes, acanthocephalans and cestodes) infections. The luminal and
mucusal coccidian infection was found at a much higher prevalence in
flounder from the Duwamish Waterway. Hexamita-infected flounder were
found only in the Duwamish Waterway (Table 7). Flounder with nematode,
trematode, and acanthocephala infestations were broadly distributed,
but the highest prevalence was inthe Snohomish River. A single cestode
infection was present in a starry flounder from the Duwarnish Waterway.
Characteristics and prevalence of renal abnormalities--
The lesions encountered through histopathologic examination of
kidney tissues in the starry flounder were similar to those already
described for English sole. These renal lesions included mesangiolysis,
mesangiosclerosis, coincident mesangiolysis, and mesangiosclerosis,
thickening of basement membrane peripheral to neck segments and tubules
including peritubular fibrosis, tubular epithelial degeneration and
63
-------
ro
OJ
CTl
-3 ~5
o re
3 <
Cu
O
Cu O
(/) i.
3- Q-
21 O
CU T3
n- cu
rD rt
-3 3"
cu n
Oj rt>
=3 -O
Q- Cu
r+
o" o
>
' (D
' vi
tn i.
n- o
o> 3
-3 in
~s
fl)
Cu -3
3 <<
a
Prevalence (percent)
CO
O
I -a
o
CD X.
II
If
o y
Ui a>
Ul
CD
co
cn
as
00 w'
CD
O
T
CD
CD
7T
D H
= 3>
«o =. -<
12
3 O
1C C
CU Q.
-i >T>
TO -3
-------
necrosis, hypennenbranous glomeruli, tubular protein casts, mesangial
cell proliferation (in the absence of mesangiosclerosis), and squamous
metaplasia of the tubular epithelium. Mesangiolysis was found only in
starry flounder from the Duwamish Waterway at a prevalence of 19.3% (47
of 243 fish, Fig. 22). However, varying degrees of mesangiosclerosis
were detected in flounder in all three areas where starry flounder were
captured, with the highest prevalence (37.5%, 91 of 243 fish) in the
Duwamish Waterway (Fig. 22). This prevalence value was significantly
higher statistically (p<0.01) than the prevalence values for either
the Snohomish River or McAllister Creek. Coincident mesangiolysis and
mesangiosclerosis were found only in Duwamish River starry flounder. By
the chi-square contingency test of independence, these two lesions were
found to be highly associated (p<0.01).
The remaining idiopathic kidney lesions were found at low prevalence
and are listed by area in Table 7. Most of these lesions were detected
in Duwamish Waterway starry flounder.
Parasitic infections were limited to infection of the excretory
tubular lumina with unidentified non-pathogenic myxosporidia, and
infection of the tubular and vascular epithelium by unidentified
intracellular microsporida. The parasitic infections were found
primarily in starry flounder from the Duwamish Waterway, but flounder
with myxosporidian infections were found at lower prevalences in the
Snohomish River and McAllister Creek (Table 7).
Ecology of English Sole and Starry Flounder
Abundance--
Catch Per Unit Effort (CPUE) is defined as the number of fish of a
particular species caught in a single 5-nrinute trawl. At the sampling
stations, CPUE values ranged from 0 to 86 for English sole and from 0 to
642 for starry flounder (Figs. 24 to 26). Average CPUE values were low
for starry flounder and English sole in the Lake Washington Ship Canal
and in the Snohomish River during all the sampling periods compared to
CPUE values in the Duwamish Waterway and McAllister Creek.
The average CPUE values for English sole and starry flounder by
season for sampling stations in the Duwamish Waterway ranged from 14.1
to 34.8 and 7.3 to 113.7, respectively (Figs. 24 and 25). The high
CPUE value (113.7) for starry flounder was obtained during the winter
of 1980. The Lake Washington Ship Canal, McAllister Creek, and Snohomish
River were sampled during two seasons. The CPUE values for English
sole for both seasons in the Lake Washington Ship Canal were 50.0 and
44.0 and in the Snohomish River the CPUE values were 12.0 and 9.8.
McAllister Creek had divergent seasonal catch rates for English sole,
with high CPUE values in the fall (45.0) and a nine-fold lower value
(5.5) in the spring. Few starry flounder were captured in the Lake
Washington Ship Canal and Snohonish River (CPUE values were less
than 2.5) while moderate catch rates of this species (26.0 and 10.3)
were obtained in McAllister Creek (Fig. 26).
65
-------
ENGLISH SOLE
Duwamish River
50.-
Fall 1978
(54)
nrinnllnn
Winter 1979
(65)
0
-
n
fln^l
Spring 1979
£ 50 (53)
c
3
O>
Q.
-5 0
n
n «
Summer 1979
50 r
(74) (57)
Winter 1980
501-
(86)
n
A B C D E F G mean
CPUE
Sampling station
FIGURE 24. Catch Per Unit Effort (CPUE) values for English sole from
the Duwamish Waterway, shown by station and sampling period.
66
-------
STARRY FLOUNDER
Duwamish River
50 r
Fall 1978
000000
Winter 1979
50 r
o o o_nn_
.
50 r~
Spring 1979
~
o>
Q.
S
+-*
ro
O
50 r
Summer 1979
(237)
0 0
0 r-i
ll
50
n
Winter 1980
0 0 _
n
(81)
««
(
342
M
(
113
I
A B C D E F G mean
CPUE
Sampling station
FIGURE 25. Catch Per Unit Effort (CPUE) values for starry flounder from
the Duwamish Waterway, shown by station and sampling period.
67
-------
50 r
ENGLISH SOLE
Fall 1978
STARRY FLOUNDER
50
MS
n
Winter 1979
50r ri
NS
a. Spring 1979
-5 5°r
NS
IMS
50 r
2 0
'E
&_
01
150
O
0
IMS
NS
NS
Summer 1979
NS
NS
LWSC Snohomish McAllister
River Creek
50 r
Sampling area
NS= not sampled
0
NS
NS
LWSC
Snohomish McAllister
River Creek
FIGURE 26. Catch Per Unit Effort (CPUE) values for English sole and
starry flounder from the Lake Washington Ship Canal (LWSC),
Snohomish River, and McAllister Creek sampling areas, shown
by sampling period.
68
-------
The geographical distribution of English sole in the Duwamish
Waterway was generally even throughout the seven Waterway stations
during the fall. However, few sole were captured in the upper stations
during the winter and spring (Fig. 24). This distribution was reversed
during the summer, with fewer English sole in the lower stations and
greater abundances in the middle and uppermost stations. Starry flounder
usually did not occur in the lower three Duwamish Waterway stations
during any season and had differing abundances in the upper four stations
(Fig. 25).
The intrusion and extent of the saltwater wedge into the mouth of
the Waterway probably strongly influences the observed patterns of
abundance of the target species. This dense wedge of saltwater underlies
the freshwater from the Duwamish River and has an up-river boundary
that depends upon the tides and the seasonal flow of freshwater from
the Duwamish River drainage (Dawson and Tilley 1972). There were high
catch rates of starry flounder, a species tolerant of lower salinities
(Hart 1973), at the uppermost portions of the Duwamish Waterway during
the summer and fall when river discharge was minimal and the saltwater
intrusion was maximal. In the Waterway this species was captured
almost exclusively from the uppermost portions during these seasons.
The winter and spring sampling was conducted when runoff was high and
flounder were commonly found at stations lower in the estuary. English
sole were less abundant in stations far up the estuary during the
periods of high river flow, i.e., during winter and spring, while
during the summer sole were more abundant at stations in the upper
estuary.
Biological characteristics--
In the Duwamish Waterway, young-of-the-year (less than 15 crn)
English sole were generally most abundant during the fall, winter, and
spring sampling seasons (Fig. 27). Young English sole also comprised
the majority of the English sole population in McAllister Creek during
the fall sampling (Fig. 28). Mid-sized English sole (15 to 24 cm)
routinely comprised over half of the Snohomish River catch. Mid-sized
and larger sole were in roughly equal proportions in the Duwamish
River during the first four sampling quarters, but a high proportion
of mid-sized fish were found in the winter of 1980. These two size
groups also dominated the catches of English sole from the Lake Washington
Ship Canal in the winter and summer. Mid-sized fish dominated the
proportion of all the sampled starry flounder populations in the Duwamish
River, but young-of-the-year starry flounder were most abundant in
McAllister Creek (Fig. 29).
The presence of small target fishes in the fall, winter, and spring
generally occurs with the expected sizes of young as related in Hart
(1973). Starry flounder grow to approximately 10.5 crn at one year of
age and English sole to approximately 15 cm. The young became available
to the trawl in the fall and showed growth to the one-year-old size by
the following winter.
69
-------
ENGLISH SOLE
Duwamish River
Fall 1978
50r ,-^
n
50 r-
Winter 1979
cj
CD
O
03
O)
1
o
-
n
Spring 1979
(51.6)
-
n
n
n
50 r- '^
50r-
Summer 1979
n
Winter 1980
50 r
(71.3)
<15cm 15-25cm > 25cm
Length groups
FIGURE 27. Length-frequency distributions of English sole from the
Duwamish Waterway (expressed as a percentage of the catch),
shown by sampling period.
70
-------
ENGLISH SOLE
Lake Washington Ship Canal
50 r-"
Winter 1979
(58.6)
n
Summer 1979
50 r
(54.6)
CO
CJ
(3 Snohomish River
Fall 1978
50 r
n
n
g, Spring 1979
| 50
0)
o
o>
Q-
n
McAllister Creek
Fall 1978
50 r r
Spring 1979
50 r- (7J^'
<15cm 15-25cm > 25cm
Length groups
FIGURE 28. Length-frequency distributions of English sole from the
Lake Washington Ship Canal, Snohomish River, and McAllister
Creek sampling areas (expressed as a percentage of the catch),
shown by station and sampling period.
71
-------
50 r
STARRY FLOUNDER
Duwamish Waterway
Fall 1978
(69.0)
n
n
McAllister Creek
50 r
n
Winter 1980
50 r
(53.7)
JI
o
50 r
'n T
Spring 1979
(72.2
II
n
i
n
50
* o
0
-t-J
ro
o
03
t 50
o
8.
03
4-i
c
03
p n
Not Sampled
(58.5)
n fl
03
Q_
50
Summer 1979
(645)
n
03
O.
50
Not Sampled
Winter 1979
50 r
(53.1)
-
"
50
n
Not Sampled
<11cm 11-20cm > 20cm
<11cm 11-20cm > 20cm
Length groups
FIGURE 29. Length-frequency distributions of starry flounder (expressed
as a percentage of the catch) from the Duwamish Waterway
and McAllister Creek sampling areas, shown by sampling period.
72
-------
Fewer juvenile English sole were captured in the winter of 1980,
compared to the winter of 1979, possibly indicating a weak year class or
a differential geographical distribution of young between the years.
Starry flounder young were generally not as numerous in the Duwamish
Waterway, compared to English sole for the same seasons. The young
starry flounder may have different settling patterns than the English
sole young. McAllister Creek, alternatively, had high proportions of
young-of-the-year English sole in the fall, and starry flounder in the
fall and spring, perhaps indicating its value as a nursery ground.
Younger (less than 15 cm) English sole and starry flounder tended
to have lower prevalences of liver lesions (11% and 12%, respectively),
than did mid-sized (15 to 25 cm) fish (26% and 21%, respectively) or
larger (greater than 25 cm) fish (67% and 44%, respectively, Table 8).
In general, the types of liver lesions observed in the younger (1 year
or less in age) English sole were the degenerative types, including MH,
HN, and CF (Table 9). The lesion types observed in sole as young as
2 years included CC, MDN, LCA, Hem and HEH; although the average age
of fish with these lesions ranged from 5.0 to 6.2 years. No English
sole younger than 3 years were observed with HNH or HC, with the average
age of affected sole being 5.1 and 5.6 years, respectively.
Chemical Analyses
Chemical analyses were performed on sediment samples from all
four estuaries, and on liver tissues from English sole and starry
flounder from the reference estuary and the two estuaries in Seattle.
Sediment chemist ry--
PCBs and selected AHs were found in all of the sediment samples
from the urban estuaries (Table 10). No PCBs or the AHs listed in
Table 10 were detected in sediment from McAllister Creek. The levels
of sediment-associated metals also tended to be lower in the reference
estuary. Within the Duwarnish Waterway, the concentrations of PCBs,
AHs and metals were generally highest in sediments near the mouth of
the estuary and gradually decreased in sediment samples taken progressively
upstream. A similar concentration gradient for PCBs in sediment within
the estuary of the Duwamish River was reported by Pavlou and Horn (1979).
The concentrations of PCBs and the lower molecular weight
AHs were quite similar in sediment samples from the Lake Washington
Ship Canal and the Snohomish River. However, the levels of higher
molecular weight AHs (those listed in Table 10 from anthracene to
perylene) were considerably higher in sediment from the Lake Washington
Ship Canal. In fact, the levels of these AHs in Lake Washington Ship
Canal sediment were quite similar to the levels found in sediment from
Station B in the Duwamish Waterway. The primary sources of these high
molecular weight AHs are probably combustion products resulting from
the use of fossil fuels.
73
-------
TABLE 8. Prevalences of various length groups of English sole and starry flounder with liver
lesions.
English Sole
Length
Group
<15cm
15 to
20 cm
Age
Range
0 to 1
2 to 3
No.
Exami ned
160
201
No. Prevalence
Affected (%)
18 11.3
52 25.8
Starry Flounder
No.
Examined
100
156
No.
Affected
12
32
Prevalence
12
21
>25 cm 3 to 10
192
129
67.2
34
15
44
-------
TABLE 9. Age/length/weight characteristics of English sole with various liver lesions,
en
Lesion
Type
Megal ocytic
hepatosi s
Eosinophi] ic
hypertrophy
Fatty Change
Nodular
hyperplasia
Hepatocel 1 ular
necrosis
Hemosiderosis
Hepatocel lular
regeneration
Cholangiocel lular
carci norna
Him'num deviation
nodule
Liver cell
adenoma
Hepatocel 1 ular
carcinoma
Mi xed
carcinomas
Number
of Fish
67
40
36
7
50
22
5
10
19
32
23
6
Average
Length
243.2
306.2
304.4
302.3
269.1
310.1
273.8
280.7
303.6
299.8
321.8
314.0
Length
Range
101-402
215-402
130-405
220-350
91-442
195-402
168-330
175-360
175-402
166-442
230-402
275-374
Average
Weight
167.6
267.6
268.9
245.3
213.9
269.8
195.2
206.6
272.0
256.2
303.8
270.0
Weight
Range
8-550
100-550
18-620
100-355
5-620
55-550
48-363
41-390
41-550
39-502
109-550
170-450
Average
Age
4.0
6.2
5.5
5.1
4.7
6.0
4.8
5.0
5.7
5.7
5.6
5.0
Age
Range
1-10
2-13
1-10
4- 7
1-10
2-13
1- 8
2- 8
2- 9
2-10
3- 9
4- 6
-------
TABLE 10. Concentrations of chemicals in sediment at sampling stations.
CTl
Compound
or
Element
Total PCBs
Naphtha! ene
2-Methyl naphthalene
1-Methyl naphthalene
Duwamish River
Aa
0.65
0.31
0.15
0.09
2,6-Dimethylnaphthalene 0.08
Fluorene
Dibenzothiophene
Phenanthrene
Anthracene
Fluoranthene
Pyrene
Benz[a]anthracene
Chrysene
Benz[e]pyrene
Benz[a]pyrene
Pery lene
Sil ver
Beryl 1 ium
Cadmi urn
Chromium
Mercury
Nickel
Phosphorous
Antimony
Strontium
Vanadium
Cobalt
Copper
Lead
% Carbon
Sand/Mud Ratio
0.35
0.14
2.60
0.63
3.70
3.20
2.70
2.40
1.70
1.50
0.53
2.5
0.9
11.3
63.9
0.8
38.9
1020.0
80.4
123.0
92.2
13.5
206.0
627.0
1.8
0.22
Bb
0.53
0.08
0.06
0.03
0.05
0.17
0.08
0.96
0.29
1.70
1.20
1.10
0.92
0.48
0.44
0.27
3.0
1.0
11.9
44.8
0.4
36.0
1174.0
73.4
124.0
95.9
17.3
131.0
265.0
1.8
0.18
C
0.18
0.04
0.05
0.02
0.05
0.05
0.08
0.33
0.10
0.72
0.40
0.56
0.42
0.25
0.19
0.20
2.4
0.8
10.0
46.0
0.3
35.6
1286.0
79.6
130.3
98.0
21.4
116.0
154.0
2.4
0.13
D
(ppm dry
0.40
0.06
0.08
0.03
0.13
0.10
0.07
0.60
0.19
0.73
0.46
0.52
0.40
0.24
0.19
0.13
1.6
0.6
7.0
33.7
0.3
28.4
881.0
54.2
92.4
82.3
14.0
52.4
63.2
2.0
0.61
E
wt)
0.29
0.04
0.05
0.02
0.04
0.05
0.04
0.40
0.11
0.79
0.45
0.30
0.37
0.20
0.18
0.21
2.3
0.8
8.6
39.4
0.2
30.6
1376.0
67.7
121.5
88.0
16.2
67.0
92.5
2.9
0.16
F
0.18
0.03
0.05
0.03
0.03
0.03
0.05
0.10
0.02
0.15
0.13
0.13
0.10
0.05
0.04
0.11
1.5
0.5
5.7
29.0
0.1
23.7
898.5
40.8
79.4
68.6
10.6
38.3
30.3
1.5
1.88
G
0.05
0.01
0.03
0.01
<0.01
0.03
0.04
0.27
0.04
0.15
0.34
0.29
0.24
0.12
0.27
0.30
2.3
0.8
8.6
37.3
0.2
30.0
133.3
67.9
131.0
90.8
14.3
59.7
77.0
2.9
0.12
Lake
Washington
Ship Canal
0.06
0.11
0.05
0.03
0.04
0.04
0.05
0.57
0.18
0.97
1.60
1.10
0.75
0.55
0.63
0.30
2.0
0.4
5.6
48.0
0.4
46.7
640.5
34.7
59.6
58.7
9.1
50.5
60.9
1.7
0.34
Snohomish
River
0.07
0.30
0.09
0.05
0.02
0.05
0.04
0.25
0.02
0.45
0.33
0.15
0.15
0.04
0.03
0.02
NDb
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
McAl 1 ister
Creek
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
ND
ND
<0.1
13.4
0.02
ND
ND
ND
ND
ND
ND
9.2
38.0
ND
ND
a Chemistry data for sediment samples from stations A and B were obtained as part of a companion study
(Maiins et al. 1980).
k ND, not determined.
-------
Tissue chemistry--
The concentrations of RGBs in liver tissues clearly corresponded
to the levels in the sediment from which the animals were captured
(Figs. 30 and 31). No such correspondence was observed with AHs or
metals. Regardless of the concentrations of AHs in the sediment,
levels in liver tissue were always very low or not detectable. The
most likely explanation for this observation is that even though
English sole and starry flounder can readily take up AHs from the
sediment (McCain et al. 1978b, Varanasi and Gmur 1981b), these species
rapidly convert these AHs to oxygenated compounds which are not
detectable by routine gas chromatographic techniques (Malins et al.
1979, Varanasi et al. 1979). Naphthalene (NPH) was the only AH that
was routinely detected in liver tissue of English sole from the Duwamish
Waterway and the Lake Washington Ship Canal. Two composite samples of
liver tissue from English sole taken from the Lake Washington Ship
Canal had concentrations of NPH of 0.06 and 0.04 yg/g (dry weight).
In the Duwamish Waterway, English sole had NPH concentrations ranging
from 0 to 0.48 yg/g (dry weight). The tendency for NPH to be present in
English sole livers while other aromatic hydrocarbons are not detectable
has also been reported by Varanasi and Gmur (1981a). They found that
NPH taken up from contaminated sediment was metabolized at a much
slower rate than was another AH, BaP.
The lack of correspondence between tissue levels and sediment
levels of metals may be due largely to the varying degrees of bio-
availability of many of the metals, or may be due to differential
excretion. Because a weak acid was used to extract the sediment samples,
the concentrations of sediment-associated metals measured in the extract
do not necessarily reflect the levels of metals that are bioavailable
to these fish species.
Interrelationships between chemistry and pathology--
Interrelationships between the occurrence of pathological conditions
in flatfish and the presence of xenobiotic chemicals in their environment
were reflected in the levels of chemicals in the tissues of fish with
certain types of lesions and/or in the levels of chemical contaminants
in sediments compared to the prevalence of fish with lesions. An
example of comparing lesions and tissue xenobiotic levels is demonstrated
in Figure 31. No consistent pattern was observed in the tissue
concentrations of PCBs and NPH in liver tissues of English sole with a
variety of liver lesions.
The second type of approach in evaluating interrelations between
chemistry and pathology involves comparing lesion prevalence with
sediment chemistry. Since English sole from the Lake Washington Ship
Canal and the Duwamish Waterway had similar prevalences of liver neoplasms,
8.2% (4 of 49 fish) and 12.7% (70 of 551 fish), respectively, it is of
interest to compare the chemical composition of the sediment samples
from both estuaries. The concentrations of aromatic hydrocarbons and
77
-------
TO
rn
GO
CO
1 3 C~)
01 C O
j^ 3 ^ Tissue concentration NJ -. INJ
Q> 1 r+ O O OO O O
^--S ^7 2
-3^ tj PJ n
-c n- ° M
3 (D - -iS
in cu o o £
rl- -s 3 S X
O - m Q.
33 ^
Id o f? M
?.?: i 2
T3 tQ C
1 S
O
2 >
D =:
r+
O
CD
CD
7T
' X I
o -..- o
O) 10 o Q. 11
33- U
CU O
-. O fD
_, 3 0
QI ro '
3 O
CX -d CU
~3 '
~? o co - .
^r^ ~- » '
O =3 c
i c~1- 3
i 3-
-1' I* ' -r,
rt-C<" ?
re c n>
~3 S ~5
a>
0 3 r+ >
-,-.._.. to
fD
Z
D
"~|
J
D
2
D
1 1 1-
Q)
CD
en
1 I
-I S
3
-
3-
5'
0
03
3
SL
5
O5 3- in
1 I a
1
a>
3
C/)'
r*
^
<
a>
S~
Q3
5'
o
~~|
_J
El fD
OJ
<"*" h
rc -j
-J O
£ 3
cu
-------
c;
Tissue concentration (,ug/g, dry weight)
K>
zr en
fis c.
' O
to o
O
fD
O
CD
LO
3 O
^
m i/>
3
IQ O
O
ro
S 0)
rt- fD
3-
O) -o
o "
(/) 3
Q.
H! M
-J O
n> o
o' o
3 ~5
to -".
~^
QJ
0>
CL
o
(-4-
Q.
2 CD
o <»
If
3 F+
Q O
3 CD
&) =
C
09
^V
GO
O
CJl
o
O)
o
T-
o
T~
00
O
CD
O
o
o
1
_1 \
ix
DS
2
""
\
AY
-------
metals in sediment in the Lake Washington Ship Canal were quite comparable
with the concentrations in sediment from many of the sampling stations
in the Duwamish Waterway (Table 10). However, with the exception of
sediment from Station G in the Waterway (a site that is annually dredged),
the levels of PCBs in the Lake Washington Ship Canal sediment were 3 to
11 times lower than in sediment from the Duwamish Waterway. Concentrations
of PCBs in the livers of sole from the Lake Washington Ship Canal were
generally 5 to 20 times lower than in livers of Duwamish Waterway sole.
These data suggest that exposure of English sole to PCBs may not
be directly related to liver neoplasms or other types of liver lesions
in English sole such as HEH and MH.
LABORATORY STUDIES
Exposures to Sediment Extracts
In experiments in which English sole were injected with extracts
of bottom sediments from the Duwamish Waterway, these sediments were
more acutely toxic than sediments from Port Madison (the reference area).
The results are consistent with the finding of lower concentrations
in the sediment from Port Madison (Table 3) of most of the chemical
compounds analyzed. However, none of the histopathological conditions
observed in English sole from the Duwamish Waterway were seen in injected
fish.
The failure to induce the same hepatic lesions seen in Duwamish
Waterway English sole may be because a longer exposure time is needed
to produce these lesions. Of course a number of other factors may be
responsible for this inability to induce the lesions seen in the field,
including: 1) exposure by injection does not readily induce these
lesions, and 2) fish were not of a susceptible age.
In a preliminary experiment (Experiment One), no mortalities were
observed in groups injected with Port Madison extracts, and although
mortalities were seen in fish injected with Duwamish River sediment
extract, the small number of fish injected probably accounted for the
finding of no statistically significant difference between the two
groups. In Experiment Two, significantly more of the fish injected
with extracts from Duwamish River sediments died than did those injected
with corn oil (p<0.001), Port Madison sediment extracts (p<0.005),
or uninjected control fish. Although Experiments One and Two are not
directly comparable because the sediment extracts used in Experiment
Two were about three times more concentrated than the extracts used in
Experiment One, they both demonstrated the acutely toxic effects of
Duwamish River sediment.
Histopathological changes were observed in the livers of sole
injected with extracts of either test or reference sediment (Table 11).
The changes were minor alterations in the appearance of hepatocytes;
one included a finely granular (rather than the usually-observed coarsely
80
-------
TABLE 11. Effects on English sole of injections of extracts of bottom sediments.
Experimental
Parameter
MORTALITY
Duwamish River Extract
Experiment Experiment
One Two
Dilutions of Extract
Undiluted - 100%(5/5)a,a
1/2 66%(2/3) 100%(5/5)a'al
1/10 - 20%(l/5)
1/20 33%(l/3)
Corn oil only
No injection
HISTOPATHOLOGY
Granular Hepatocel
Cytoplasm (GHC)
Hepatocel lular
Hypertrophy (HH)
Undiluted and 1/2
1/lQC or l/20d
Corn oil only
No injection
HEMATOCRIT
Undiluted
1/2
1/lOC
l/20d
Corn oil only
No injection
a Significantly
al Significantly
b ND, not done.
lular
GHC 100% No
(l/l)f Survivors
HH 100%
(1/D
GHC 100% 50% (2/4)
(2/2)f
HH 0% 50%(2/4)
(0/2)
-
-
ND No survivors
10.0%(l)e No survivors
ND 17.4(4)
jfl.9
21.05(2) ND
~
-
different from control values.
different from Port Madison values
c Fish were injected with a 1/10 dilution only in
Fish were injected with a 1/20 dilution only in
e Number of fish
Port Madison
Experiment
One
0%(0/3)
0%(0/3)
GHC 50%
(1/2)
HH 0%
(0/2)
GHC 50%
(1/2)
HH 0%
(1/2)
-
-
ND
26.6(2)
^5.9
ND
21.8(2)
_
-
.
Experiment Two.
Experiment One.
Extract
Experiment
Two
40%(2/5)
25%(l/4)
0%(0/5)
83% (5/6)
50% (3/6)
0% (0/5)
0% (0/5)
-
-
8.3
^3.7(4)3
15.2(3)
JK3.3
16.0(3)
±°
ND
_
-
Controls
Experiment Experiment
One Two
0%(0/3) 0%(0/5)
NDb 0%(0/5)
; ;
~ ~
GHC 0%(0/2) 0%(0/5)
HH 0%(0/2) 0%(0/5)
ND 0%(0/5)
0%(0/5)
-
-
-
-
23.3(2) 21.3(4)
+0.4 +6.4
ND 23.0(4)
+4.2
from which blood samples were tested are in parentheses.
. , _ . ,
, ... .
, .
nJ\
Not applicable.
81
-------
flocculent) cytoplasmic texture (GHC), and the other change consisted
of slight hepatocelluar hypertrophy (HH) (Table 11). In Experiment One,
granular changes in the hepatocellular cytoplasm were observed in fish
injected with Duwamish or Port Madison sediment extracts, with no
changes seen in either control group. Hepatocellular hypertrophy was
seen in only one fish injected with a 1/2 dilution of Duwamish River
sediment extract. In Experiment Two, the granular hepatocellular
changes were observed in both Duwamish and Port Madison sediment extract-
injected fish, with hepatocellular hypertrophy also seen in the same
groups. No hepatocellular changes were found in the control groups.
Experiment Three showed fish with granular cytoplasmic changes only
in the group injected with Duwamish River sediment extracts (Port
Madison extracts were not tested) with no changes in the livers of fish
from control groups.
In analyzing these results (Table 11) statistically by the chi-square
contingency test, the data for dilution groups was combined for each
general group (i.e., 1/2 dilution combined with 1/10 or 1/20 for Duwamish
River sediment extract-injected fish) and then the three general groups
(Duwamish River, Port Madison, and corn oil-injected controls) were
compared to test the null hypothesis that the presence of hepatic
anomalies is independent of treatment group. In Experiment One, granular
hepatocellular changes (GHC) were found at a significantly higher
frequency by this test than in the corn-oil injected controls. No
significant differences were found either between Duwamish River sediment-
injected fish and Port Madison test fish or Port Madison and control
fish affected with GHC. Hepatocellular hypertrophy (HH) was independent
of treatment group. Both HH and GHC were independent of treatment group
in Experiment Two. No other significant differences were found among
the groups tested. The importance of granular hepatocellular cytoplasmic
changes is not known, but it may represent an early degenerative change.
However, the inconsistencies among the three experiments with regard
to this subtle histological change makes it impossible to make any
conclusions as to the toxicity of Duwamish River sediment to English
sole liver relative to that of the Port Madison sediment.
Another histological change, characterized as chronic visceral
inflammation, was observed in most of the sole injected with either
sediment extracts dissolved in corn oil or with corn oil alone. Chronic
inflammation was observed most frequently in the mesenteries and the
stroma (connective tissue portion) of the pancreatic tissue scattered
throughout the mesenteries, but this response was also observed with a
lesser frequency in the serosal surfaces of both the liver and intestinal
tract. The inflammatory response was characterized by a marked
mononuclear cell infiltrate with or without associated fibrosis, or
necrosis, or both. This condition therefore appears to be a response
to the i.p. injection of corn oil.
32
-------
In an additional experiment, English sole were injected with
extracts of interstitial water from either Duwarnish River or Port Madison
sediment. Fish injected with the test extracts did not have significantly
higher mortality, lower hematocrits, or frequencies of hepatic lesions
than did the fish receiving the control extracts.
In one sediment-extract injection experiment (Experiment Two),
mortality was the most important assessed parameter, where a statistically
significant difference was observed between fish injected with extracts
of Duwamish River test sediment and those injected with extracts of
control sediment or with corn oil. Mortality results suggested the
presence of an acutely toxic substance (or substances) in the test
sediment, but further exposure studies would be required to confirm this
single finding and identify the responsible toxicant(s). Furthermore,
a significant difference in mortality between the two sedinent types
was only observed when the fish were injected twice with relatively
undiluted preparations of the sediment extract which was concentrated
three-fold suggesting that the toxicant(s) was present in relatively low
levels in the original sediment. Alternatively, the toxicants nay have
been detoxified by the extraction procedure, or the toxicity may have
been reduced during storage of the extract.
Exposures to Bottom Sediments
For Experiment One, mortality in the test (Duwamish River) sediment
group was 16% (16 of 100 fish), while that in the control (Snohomish River)
sediment group was 27% (27 of 100 fish, Table 12). Statistical tests
indicated that mortality was independent of the sediment type used for
the exposure (x2=3.58, p=0.059). No consistent histopathological or
hematological differences between the test and control groups were
observed through Day 92, the ternination date.
Chemical analysis of livers from fish exposed to Duwamish River
and Snohomish River sediments for approximately two months revealed
similar levels of PCBs in test (2.8 yg/g dry wt) and control (1.4 yg/g
dry wt.) fish. No AHs were detected in the livers of either groups of
fish. The concentrations of PCBs and AHs in Snohomish River sediment
were considerably lower than in sediment from the Duwamish Waterway.
Mortality for Experiment Two was 54% (19 of 35 fish) in the test
(Duwamish Waterway) sediment group and 38% (12 of 32 fish) in the
control (Port Madison) group prior to termination of the test. Mortality
was independent (x2=1.89, p=0.169) of the type of sediment to which the
fish were exposed. Microscopically, no differences between tissues of
test and control animals were observed through Day 65, the last sampling
date. Hematocrit hemoglobin values for test and control fish were not
consistently different.
The purpose of these laboratory experiments was to compare
histological and hematological findings for fish exposed to sediments
and interstitial water from polluted (test) and presumed non-polluted
83
-------
TABLE 1?. Mortality after exposure of English sole to bottom sediments.
Experimental
Parameter
Duwamish
Waterway
Value of Parameter
Snohorm'sh
River
Port
Madison
Mortalities
Experiment 1
(after 92 days)
Experiment 2
(after 65 days)
16%(16/100) 27%(27/ino)
54%(19/35)
38%(12/32)
84
-------
(control) areas. Nevertheless, the control sediments chosen for the
study contained certain pollutants which may have affected experimental
results. For example, the Port Madison sediment contained a higher
concentration of chlorinated butadienes than did the Duwarnish River
sediment (Table 3). Additionally, Snohomish River sediment contained
relatively high levels of pesticides and PCBs as indicated by the
concentrations detected in fish which had been exposed to this sediment
for approximately 2 months. If possible, a control sediment lacking
in such pollutants should be used if further sediment exposure experiments
are conducted. It may be necessary to search outside of Puget Sound
for an ideal control sediment.
None of the characteristic lesions observed in English sole from
the Duwamish Waterway were seen in fish in these laboratory experiments.
One possible explanation for this results was the relatively short
duration of the exposures.
The average age of Duwamish River English sole exhibiting pollution-
associated hepatic lesions in the field portion of this study was
4.54+2.46 years, which is consistent with an extended period of exposure
to a~particular pollutant or combination of pollutants occurring prior
to development of the lesions. The longest laboratory exposure to any
test or control substance was 92 days.
Benzo[a]pyrene Bioavailability Experiment
Blood, liver, muscle and bile of English sole contained considerable
radioactivity at 24 hours after the fish were placed on sediment (Table 13).
Bioconcentration values of BaP in tissues were calculated using both
the concentration of BaP in sediment and sediment-associated water (SAW).
The bioconcentration value for liver was greater than that for either
blood or muscle. Liver and nuscle contained 410 and 5.7 times, respectively,
as much BaP-derived radioactivity as that present in the SAW. The
bioconcentration in the liver was also higher than for blood or muscle
when the bioconcentration factor was based on sediment concentrations.
BaP was extensively metabolized and excreted in the bile as evidenced
by large concentrations of BaP-derived radioactivity in bile (Table 13)
and the presence of high proportions of metabolites in bile of these
fish (Table 14). Of the total radioactivity in bile, 94% was present
as BaP metabolites. Enzymatic hydrolysis revealed that 26% of the
radioactivity in bile was in the form of glucuronic acid conjugates of
BaP. Low specific activity of BaP prevented further analyses of the
samples to identify individual BaP metabolites in either liver or bile
of these fish.
This experiment clearly demonstrated that BaP associated with
sediment is bioavailable to English sole. Uptake of sediment-associated
BaP can occur via the gills or via the gastrointestional tract as a
result of fish contacting and ingesting particle-bound BaP together
with solubilized BaP present in SAW. Earlier results reported by
85
-------
TABLE 13. Concentrations of benzo[a]pyrene (BaP)-associated radioactivity in sediment, sediment-associated
water (SAW) and tissues of English sole3, and the bioconcentration factors for the tissues
relative to BaP concentrations in the sediment and SAW.
Time of
Exposures
(hours)
Sediment
(pmoles/g)
Sediment-Associated Water
Unfiltered
(pmoles/ml )
Blood Liver
[~pmo 1 e s BaP/g
Muscle
tissue dry
Bile
wt.)
n
24
2200+320
2700+ 97
12+10
6+5
c c c c
390+50 2400+1200 34+18 31,000+17,000
CO
01
Bioconcentration Factors
24 c c 0.15d
66e
0.92
410
0.13
5.7
12
5100
a Sediment and SAW samples were taken from the exposure tanks as described in the text. SAW was decanted
from the sediment and analyzed unfiltered. Three replicates of sediment were analyzed to get average
amount of tritium in each sample.
b Concentration values are reported as the mean +_ SD.
c Not done.
d Based on sediment concentrations.
e Based on SAW concentrations.
-------
TABLE 14. Proportions of the metabolites of benzo[a]pyrene (BaP) in bile of
English sole exposed to contaminated sediment.9
Percent of total radioactivity
Before enzymatic hydrolysis*3 After enzymatic hydro1ysisc
Ethyl actetate phase Aqueous phase Ethyl acetate phase Aqueous phase
6 94 26 74
a Two samples of bile from separate fish were pooled.
b Bile was extracted with ethyl acetate to separate unconverted BaP and its
non-conjugated metabolites from the conjugated metabolites.
c The remaining aqueous phase containing the conjugated metabolites was
treated with Glucurase and hydrolyzed. HPLC analysis was performed on
the ethyl acetate fraction but no conclusive results were reported due
to low activity of the sample.
87
-------
Varanasi and Gmur (1981a), in which English sole were exposed to BaP in
oiled sediment, showed that BaP in SAW was associated mainly with the
particles. Lakowicz et al. (1980) reported that adsorption of BaP to
participates results in an increased rate of membrane uptake compared
to that obtained with an aqueous suspension of BaP. The results of
Martin (1980) suggest that uptake of BaP via the gastrointestinal
tract in estuarine fishes may be the major route.
Large proportions of BaP metabolites in bile of English sole
indicate that these fish metabolize BaP and excrete BaP-derived compounds
efficiently. BaP, administered orally, is also metabolized extensively
by pleuronectid fish (Varanasi and Gmur 1981b). Moreover, as with
earlier studies (Varanasi et al. 1979, Varanasi and Gmur 1981a, 1981b)
in which bottom fish were force fed either naphthalene or BaP,
glucuronides were the major class of conjugates present in the bile.
Glucuronidation is also identified as a major detoxification pathway
for polynuclear aromatic hydrocarbons in mammals (Baird et al. 1977).
In this study, individual metabolites of BaP were not characterized
in liver or bile because of the low levels of radioactivity. However, in
a separate study (Varanasi and Gmur 1981b), BaP is biotransformed by
English sole liver into a number of metabolites which are known to be
mutagenic and carcinogenic in mammals (Brookes 1977). Notable among
these metabolites was 7,8-dihydro-7,8-dihydroxy BaP, a proximate carcinogen,
Studies with mammals (Sims et al. 1974) show that this dihydrodiol
is further metabolized to produce a diolepoxide which binds covalently
to DNA, and that such chemical modification of DNA is implicated as the
critical step in a series of events leading to some chemically-induced
neoplasms. In a recent report (Varanasi et al. 1981), it was shown
that when BaP is force fed to English sole, a significant portion of
the BaP intermediates become covalently bound to liver DNA. In view
of these findings, the present results showing the ability of English
sole to take up and bioconcentrate sediment-associated BaP may have
considerable environmental significance.
-------
APPENDIX I
Lymphocystis Disease
Lymphocystis disease in marine fishes of the eastern Pacific Ocean
was reported in yellow fin sole (Limanda aspera) from the Bering Sea
(Alpers et al. 1977a); in three species of Sebastes from the Steinhart
Aquarium in San Francisco (McCosker et al. 1976); in Hypsoblennius spp.
commonly found in intertidal and subtidal waters in bays of southern
California (Weissenberg 1955, Stephens et al. 1970); in gobiid and
diodontid fishes of the Gulf of Panama and Gulf of California (McCosker
and Nigrelli 1971); and in ling cod (Ophiodon elongatus) from the
Hecate Strait, British Columbia (Walker 1947).
The gross lesion found on English sole in this study was present
as multiple white to pale yellowish ovoid surface nodules 0.5 to 2 mm
in diameter, present only on the blind side pectoral fin; the fin most
often in contact with bottom sediments. The histopathologic features
of the nodules included massively hypertrophied cells of a fibroblastic
nature directly beneath the stratified epithelium of the fin epidermis.
Each individual hypertrophied cell was enclosed by a distinctive hyaline
capsule. These cells had a large nucleus undergoing karyolysis and a
granular cytoplasm containing basophilic and Feulgen-positive material
comprising the viral inclusion body (Fig. 3). Microscopic appearance
of this lesion was consistent with the features of lyrnphocystis disease
reported for other fish species (Wellings et al. 1977, Nigrelli et al.
1965).
A portion of a nodule from the fin of the affected English sole
was reprocessed for examination by electron microscopy. The viral
particles were well-preserved, and displayed the morphology typical of
that previously described for lymphocystis virus (Alpers et al. 1977a,
Walker and Weissenberg 1965, Fig. 4).
The diameter of the nucleocapsid ranged from 180 to 210 nrn. This
corresponds closely to the 180-220 nm nucleocopsid measurements reported
for lymphocystis virus particles in yellowfin sole (Alpers et al. 1977a).
Also present in the cytoplasm were numerous randomly distributed fibril lar
structures (Fig. 5) similar to those described by Alpers et al. (1977a).
These fibrillar structures displayed a banding periodicity of approximately
800 A periodicity also reported by Alpers et al. (1977a). Thisoperiodicity
does not coincide with the value associated with collagen (640 A), and
the tapered appearance is not typical of the appearance of collagen.
These structures were not found in normal fibroblasts, and their presence
may be associated with the process of viral infection.
89
-------
Angioepithelial Nodules (AENs)
The AEN was grossly visible as a hemispherical, smooth-surfaced,
red to pale pink nodule, on the skin furface, measuring 1-10 mm in
diameter. Multiple tumors were often observed. Microscopic features
were increased vascularization of the stratum spongiosum (subepidermal
layer) with hyperernia, prominent mononuclear infiltrate, increased
connective tissue elements, foci of coagulation necrosis, all indicative
of a typical pleuronectid inflammatory response. The overlying epidermis
was typically hyperplastic, but not papillomatous or folded. Within
these areas of dermal inflanimation and congestion were numerous
extravascular cells identical in appearance to the so-called X-cel1
associated with this disease complex (Brooks et al. 1969, Wellings et
al. 1976b). In the AEN, the X-cell had a prominent karysosorne within a
large nucleus, a granular, vacuolated cytoplasm, a cell diameter of
approximately 15 y, and was typically surrounded by reactive inacrophages
and extravasated red blood cells. Only rarely were X-cel Is visible in
the overlying epidermis. The swelling and redness in this tumorous
lesion was due to the increased vascularization, hyperemia, and
inflammatory response, and not to multiplication of" X-cells.
The transitional form of AEN was characterized by a similar but
reduced dermal hyperemia, rnononuclear (lymphocytes and macrophages)
infiltrate, and connective tissue proliferative response, but in contrast
to the AEN ttie epidermis was not only thickened but was folded (papil liform)
and appeared to contain multiple nests of X-cells surrounded by a
wavy, folded fibrovasular strorna. The stroma typically contained
large melanophores, and often a dense rnononuclear cell infiltrate. A
thin cap (2-3 cell layer thick) of epidermal cells overlay the multiple
clusters of X-cells.
Grossly, the fully developed epidermal papillomas were pink to
grey to brown in color depending on the vascularity of the tumor and
the density of melanophages in the stroma. The outer cutaneous surfaces
were papillomatous in appearance, composed of interdigitating, folding
ridges of tissue. These low lying tumors were found anywhere on the
external body surface, fins, and opercula, spreading in a sheet-like
fashion. They were quite variable in size and shape, but generally
were up to 5 cm in diameter, and 3-4 mm in depth. Often the tumors
were multipl e.
Microscopically, the epidermal papillomas were composed of folded
ridges of a highly thickened epidermis-like tissue underlain by the
stratum spongiosum of the dermis. Multiple nests of X-cells of variable
size, bordered by a basal lamina, and supported by a fibrovascular
stroma containing melanophores formed the bulk of the tumor tissue. The
thin cap of epidermal malphigian cells overlayed the papilliform folds.
The stroma was typically populated by mild to moderate infiltrate
of mononuclear leucocytes, which were also occasionally present within
the X-cell nests. When present as focal aggregates, these inflammatory
cells were usually associated with degenerate, vacuolated forms of X-
cells at a central position within the nest.
90
-------
REFERENCES
Agius, C. 1979. The role of melano-rnacrophage centers in iron storage
in normal and diseased fish. Journal of Fish Diseases _2:337-343.
Alpers, C.E., B.B. McCain, M.S. Myers and S.R. Wei lings. 1977a.
Lymphocystis disease in yellow in sole (Limanda aspera) in the
Bering Sea. Journal of the Fisheries Research Board of Canada.
34(5):611-616.
Alpers, C.E., B.B. McCain,'M.S. flyers, S.R. Wei lings, M. Poore, J. Bagshaw
and C.J. Dawe. 19775. Pathologic anatomy of pseudobranch tumors
in Pacific cod, G_ad_us macrocephalus. Journal of the National Cancer
Institute J59:377-398.
Baird, W.M., C.J. Chern and L. Diamond. 1977. Formation of benzo[a]
pyrene glucuronic acid conjugates in hamster embryo cell cultures.
Cancer Research _37:3190-3197.
Bannasch, P. 1976. Cytology and cytogenesis of neoplastic (hyperplastic)
hepatic nodules. Cancer Research 36:2555-2562.
Bannasch, P. 1978. Sequential cellular alterations during hepato-
carcinogenesis. In, Rat Hepatic Neoplasia, P.M. Newberne and W.H. Butler,
eds. MIT Press, Cambridge, pp. 58-93.
Brookes, P. 1977. Mutagenicity of polycyclic aromatic hydrocarbons.
Mutation Research ^:257-284.
Brooks, R.E., G.E. McArn and S.R. Wellings. 1969. Ultrastructural
observations on an unidentified cell type found in epidermal tumors
of flounders. Journal of the National Cancer Institute 43:97-110.
Brown, U.W., L.S. Ramos, M.Y. Uyeda, A.J. Friedman and W.D. MacLeod, Jr.
1980. Ambient-temperature extraction of hydrocarbons from marine
sediment: Comparison with boiling-solvent extractions. Advances
in Chemistry Series, American Chemical Society. 14:313-326.
Chitwood, M. and J.R. Lichtenfels. 1972. Identification of parasitic
metazoa in tissue sections. Experimental Parasitology 32:407-519.
Dawe, C.J. 1980. Polyorna tumors in mice and X-cell tumors in fish,
viewed through telescope and microscope. Eleventh International
Symposium of Princess Takamatsu Cancer Research Fund, Nakahara
Memorial Lecture.
91
-------
Uawson, W.A. and L..J. Tilly. 1972. Measurement of salt-wedge excursion
distance in the Ouwamish River estuary, Seattle, Washington, by
means of dissolved oxygen gradient. U.S. Geological Survey Water-
Supply Paper, 1873-d. Prepared in cooperation with the Municipality
of Metropolitan Seattle. 27 pp.
Edmondson, H.A. 1958. Tumors of the liver and intrahepatic bile ducts.
In, Atlas of Tumor Pathology, Armed Forces Institute of Pathology,
Section VII - Fasicle 25. 216 pp.
Falkmer, S., S. Marklund, P.E. Mattson and C. Rappe. 1977. Hepatomas
and other neoplasms in the Atlantic hagfish, Myxine glutinosa:
A histopathologic and chemical study. Annals of the New York
Academy of Sciences 298:342-355.
Farber, E. 1956. Similarities in the sequence of early histological
changes induced in the liver by ethionine, 2-acetylaminof1uorene
and 3' methyl-4-dimethyl aminoazo-benzene. Cancer Research 16:142-149.
Farber, E. 1976. On the pathogenesis of experimental hepatocellular
carcinoma. In, Hepatocellular Carcinoma, K.O. Kuda and R.L. Peters, eds.
John Wiley and Sons, New York. pp. 1-24.
Farber, E., S. Parker and M. Greenstein. 1976. The resistance of
putative premalignant liver cell populations, hyperplastic nodules,
to the acute cytotoxic effects of some hepatocarcinogens. Cancer
Research ^6:3879-3887.
Farber, E., D. Solt, R. Cameron, B. Laishes, K. Ogawa and A. Medline.
1977. Newer insights into the pathogenesis of liver cancer.
American Journal of Pathology ^9_:477-482.
Hart, J.L. 1973. Pacific Fishes of Canada. Fisheries Research Board
of Canada, Bulletin JL80. 740 pp.
Hendricks, J.D., R.O. Sinnhuber, J.E. Dixon, J.H. Wales, and M.S. Masri and
U.P.H. Hsieb. 1980. Carcinogenic response of rainbow trout, (Salmo
gairdneri). Aflatoxin Q. and synergistic effect of cyclopropenoid
fatty acids. Journal of the National Cancer Institute 64(3):523-527.
Heston, W.E., G. Vlahakis and M.K. Deringer. 1960. High incidence of
spontaneous hepatomas and the increase of this incidence with urethan
in CBH, C3Hf and C3He male mice. Journal of the National Cancer
Institute ^4:425.
Hinton, D.E., J.E. Klaunig and M.M. Lipsky. 1978. PCB-induced alterations
in teleost liver: A model for environmental disease in fish. Marine
Fisheries Review _40(10):47-50.
Hoffman, G.L. 1970. Parasites of North American Freshwater Fishes.
University of California Press, Berkeley. 486 pp.
92
-------
Hoffman, G.L., C.E. Dunbar, K. Wolf and L.O. Zwillenberg. 1969.
Epitheliocystis, a new infectious disease of the bluegill, Lepomi s
macrochirus. Antonie von Leeuwenhoek 35:146-158.
Houston, A.H. and M.A. Dewilde. 1969. Environmental temperature and the
body fluid system of the freshwater teleost. III. Hernatology and
blood volume of thermally acclimated brook trout, Salvelinus fontinalis.
Comparative Biochemistry and Physiology 28:877-885.
Ito, N., H. Nagasaki, M. Arai, S. Makiura, S. Sugihara and H. Kazuya.
1973a. Histopathologic studies of liver tumorigenesis induced in
mice by technical polychlorinated biphenyls and its promoting
effect on liver tumors induced by benzene hexachloride. Journal
of the National Cancer Institute jn:1637-1646.
Ito, N., H. Nagasaki, M. Arai, S. Sugihara and S. Makiura. 1973b.
Histologic and ultrastructural studies of the hepatocarcinogenicity
of benzene hexachloride in nice. Journal of the National Cancer
Institute J51:817-826.
Ito Y., I. Kimura and T. Miyake. 1976. Histopathological and virological
investigations of papillomas in soles and gobies in coastal waters
of Japan. Progress in Experimental Tumor Research, Karger, Basel.
^0:86-93.
Jago, M.V. 1969. The development of the hepatic megalocytosis of chronic
pyrrolizidine alkaloid poisoning. American Journal of Pathology
_56( 3): 405-421.
Jones, G. and W.H. Butler. 1975. Morphology of spontaneous and induced
neoplasia. In, Mouse Hepatic Neoplasia, W.H. Butler and P.M. Newberne,
eds. Elsevier Scientific Publishing Co., Amsterdam, pp. 21-60.
Jubb, K.V.F. and P.C. Kennedy. 1970. Pathology of Domestic Animals.
2nd Edition, Vol 2. Academic Press, New York. pp. 216-219.
Karasaki, S. 1976. Ultrastructural and cytochemical studies on
hyperbasophilic foci with special reference to the demonstration
of cells surface alterations in hepatocarcinogenesis. Cancer
Research _36:2567-2572.
Kawamoto, S., A. Kirschbaum, M.L. Ibanez, J.J. Trentin and H.G. Taylor.
1961. Influence of urethane on the development of spontaneous
leukemia and on the induction of hemangiomas in the AKR and C58
strains of mice. Cancer Research 21(1):71.
Kimbrough, R.U., R.E. Linder and T.B. Gaines. 1972. Morphological
changes in livers of rats fed polychlorinated biphenyls. Archives
of Environmental Health 25:354-364.
93
-------
Kimbrough, R.U. and R.E. Linden. 1974. Induction of adenofibrosis
and hepatomas of the liver in BALB/CJ mice by polychlorinated
biphenyls. Journal of the National Cancer Institute 53:547-552.
Kimbrough, R.D., R.A. Squire, R.E. Linden, J.D. Stnaudbeng, R.J. Montali
and V.W. Bunse. 1975. Induction of liven tumors in Sherman stnain
female nats by the polychlorinated biphenyl Anochlon 1260. Jounnal
of the National Cancen Institute _55:1453-1459.
Kimuna, N.T. and T. Baba. 1973. Neoplastic changes in the nat liven
induced by polychloninated biphenyls. GANN 64:105-108.
Kitigawa, T. 1976. Sequential phenotypic changes in hypenplastic
aneas duning hepatocancinogenesis in the nat. Cancer Research
^6:2534-2539.
Koller, L.D. and J.G. Zinkl. 1973. Pathology of polychloninated
biphenyls in nabbits. American Jounnal of Pathology 70(3) :363-377.
Knumbein, W.C. and F.J. Pettijohn. 1938. Manual of Sedimentary
Petrography. The Twentieth Century Earth Science Series,
Kirtley F. Mather, ed. Appleton-Century-Crofts, Inc., New York.
549 pp.
Lakowicz, J.R., D.R. Beran and S.C. Reimer. 1980. Transport of a
carcinogen benzo[a]pyrene from particulates to lipid bilayers.
A model for the fate of particle-absorbed polynuclear aromatic
hydrocarbons which are retained in the lungs. Biochinica
Biophysica Acta 629:243-258.
Lange E. and J.V. Johanessen. 1977. Histochernical and ultrastructural
studies of chemodectoma-like tumors in the cod, Gadus morhua L.,
Laboratory Investigation 37(1):96-104.
Larsson, A., B.E. Bengsston and 0. Svanbeng. 1976. Some hematological
and biochemical effects of cadmium on fish. In, Effect of Pollutants
on Aquatic Organisms, A.P.N. Lockwood, ed. Society for Experimental
Biology Seminar Vol. 2. Cambridge University Press, Cambridge.
pp. 35-45.
Luna, L.G. 1968. Manual of Histologic Staining Methods of the Armed
Forces Institute of Pathology.3rd Edition.McGraw-Hill Book
Company, New York. 258 pp.
Malevski Y., J.H. Wales and M.W. Montgomery. 1974. Liver damage in
rainbow trout, Sal mo gairdneri, fed cyclopropenoid fatty acids.
Journal of the Fisheries Research Board of Canada 31:1397-1400.
Malins, D.C., T.K. Collier and H.R. Sanborn. 1979. Disposition and
metabolism of aromatic hydrocarbons in marine organisms. In,
Pesticide and Xenobiotic Metabolism in Aquatic Organisms,
M.I.Q. Khan, J.J. Lech and J.J. Menn, eds. American Chemical
Society, Washington, D.C. pp. 57-75.
94
-------
Malins, D.C., B.B. McCain, D.W. Brown, A.K. Sparks and H.O. Hodgins.
1980. Chemical contaminants and biological abnormalities in
Central and Southern Puget Sound. NOAA Technical Memorandum
OMPA-2. 295 pp.
Margolis, L. and J.R. Arthur. 1979. Synopsis of the parasites of
fishes of Canada. Fisheries Research Board of Canada, Bulletin
199. 269 pp.
Martin, B.J. 1980. Effects of petroleum compounds on estuarine fishes.
Ecological Research Series EPA-600/3-80-01. pp. 1-31.
McArn, G.E., R.E. Chuinard, B.S. Miller, R.E. Brooks and S.R. Wellings.
1968. Pathology of skin tumors found on English sole and starry
flounder from Puget Sound, Washington. Journal of the National
Cancer Institute 4J:229-242
McCain, B.B., K.V. Pierce, S.R. Wellings and B.S. Miller. 1977.
Hepatomas in marine fish from urban estuary. Bulletin of
Environmental Contamination and Toxi cology JL8:1-2.
McCain B.B., M.S. Myers, W.D. Gronlund, S.W. Wellings and C.E. Alpers.
1978a. The frequency, distribution and pathology of three diseases
of demersal fishes in the Bering Sea. Journal of Fish Biology
Jj?:: 267-276.
McCain, B.B., H.O. Hodgins, W.D. Gronlund, J.W. Hawkes, D.W. Brown,
M.S. Myers and J.H. Vandermeulen. 1978b. Bioavailability of crude
oil from experimentally oiled sediments to English sole (Parophrys
vetulus), and pathological consequences. Journal of the Fisheries
Research Board of Canada _35(5) :657-664
McCain, B.B., W.D. Gronlund, M.S. Myers and S.R. Wellings. 1979.
Tumors and nicrobial diseases of marine fishes in Alaskan waters.
Journal of Fish Diseases _2:111-130.
McCain, B.B. and D.C. Malins. 1981. Effects of petroleum hydrocarbons
on selected demersal fish and crustaceans. In, Ecological Stress
and the New York Bight: Science and Management. Published by
New York Bight Marine Ecosystems Analysis (MESA), in press.
McCosker, J.E. and R.F. Nigrelli. 1971. New records of lymphocystis
disease in four eastern Pacific fish species. Journal of Fisheries
Research Board of Canada _28:1809-1810.
McCosker, J.E., M.D. Lagios and T. Tucker. 1976. Ultrastructure of
lymphocystis virus in the Quillback rock fish (Sebastes maliger)
with records of infection in other aquarium-held fishes. Transactions
of the American Fisheries Society 2:333-337.
95
-------
McLean, E.K. 1970. The toxic action of pyrrolizidine (Senecio)
alkaloids. Pharmacology Review 22:429.
McLean, E.K. 1974. Senecio and other plants as liver poisons.
In, International Symposium on Hepatotoxicity. Academic Press,
New York. pp. 142-146.
Morita, T., I. Kihara, T. Oite, T. Yamamoto a'nd Y. Suzuki. 1978.
Mesangiolysis: sequential ultrastructural study of Habu venom-
induced glomerular lesions. Laboratory Investigation 38(1):94-102.
Moulton, J.E. 1978. Tumors in Domestic Animals. University of
of California, Berkeley. 2nd Edition, pp. 279-281.
Myers, M.S. 1981. Pathologic anatomy of papilloma-1ike tumors in
Pacific Ocean perch (Sebastes alutus) from the Gulf of Alaska.
M.S. Thesis, University of Washington. 98 pp.
Newberne, P.M. 1976. Experimental hepatocellular carcinogenesis.
Cancer Research ^36_: 2573-2578.
Newberne, P.M. and W.H. Butler. 1978. Rat Hepatic Neoplasia. MIT
Press, Cambridge. 288 pp.
Nigrelli, R.F., K.S. Ketchen and S.J. Ruggieri. 1965. Studies on virus
diseases of fish. Epizootiology of epithelial tumors in skin of
flatfishes of the Pacific Coast; with special reference to the sand
sole from northern Hecate- strait, B.C., Canada. Zoologica (New
York) _5U(3):115-122.
Nishizumi, M. 1970. Light and electron microscope study of chlorobiphenyl
poisoning in mouse and monkey liver. Archives of Environmental
Health 2J.:620-632.
Olson, R.E. 1978. Parasitology of the English sole, Parophrys vetulus
Girard in Oregon, U.S.A. Journal of Fish Biology~13:237-238.
Ostle, B. 1972. Statistics in Research. Iowa State University Press,
Ames. 585 pp.
Paperna, I. 1977. Epitheliocystis infection in wild and cultured
seabream (Sparus aurata, Sparidae) and grey mullets (Liza ramada,
Mugilidae). Aquaculture 10:169-176.
Paperna, I., I. Sabnai and M. Castel. 1978. Ultrastructural study of
epitheliocystis organisms from gill epithelium of the fish Sparus
auraota and Liza ramada (Risso) and their relation to the host cell.
Journal of FTsTTDiseases 1:181-189.
96
-------
Pattie, B.H. 1975. Estuarine monitoring program (July 1, 1974 -
June 30, 1975). Project Completion Report, State of Washington
Department of Fisheries, Marine Program. 37 pp.
Pavlou, S.P., K.A. Korgslund, R.N. Dexter and J.R. Clayton. 1973.
Data report for the National Coastal Pollution Research Program,
University of Washington, Seattle. Special Report. No. 54. 178 pp.
Pavlou, S.P. and W. Horn. 1979. PCB removal from the Duwamish River
estuary: Implications to the management alternative for the
Hudson River PCB cleanup. In, Health Effects of Halogenated
Hydrocarbons, W. Nicholson and J. Moore, eds., Annals of the
New York Academy of Sciences, N.Y. 320:69-124.
Pierce, K.V., B.B. McCain and S.R. Wei lings. 1978. The pathology of
hepatomas and other liver abnormalities in English sole (Parophrys
vetulus) from the Duwamish River, Seattle, Washington. Journal
of the National Cancer Institute 60:1445-1453.
Pitot, H.C. 1977. The natural history of neoplasia. American Journal
of Pathology _89(2):402-411.
Preece, A. 1972. A Manual for Histologic Technicians. 3rd Edition.
Little, Brown and Co., Boston. 428 pp.
Rabes, H.M., P. Scholze and B. Jantsch. 1972. Growth kinetics of
diethylnitrosamine-induced enzyme deficient "preneoplastic" liver
cell populations in vivo and in vitro. Cancer Research 32:2577-2586.
Ramos, L.S. and P.G. Prohaska. 1981. Sephadex LH-20 chromatography
of extracts of marine sediment and biological samples for the
isolation of polynuclear aromatic hydrocarbons. Journal of
Chromatography 211:284-289.
Reuber, M.D. 1965. "Development of preneoplastic arid neoplastic lesions
of the liver in male rats given 0.025 percent N-2 fluorenydiacetamide.
Journal of the National Cancer Institute _34:697-724.
Reuber, M.D. 1968. Histogenesis of cholangiofibrosis and wel1-differentiated
cholangiocarcinoma in Syrian hamsters given Z-acetamidofluorene or
2-diacetamidoflourene. GANN _59:239-246.
Robbins, S.L. and R.S. Cotran. 1979. Pathologic Basis of Disease.
2nd Edition. W.B. Saunders Company, Philadelphia.1598 pp.
Roberts, R.J. 1975. Melanin containing cells of teleost fish and
their relation to disease. In, The Pathology of Fishes,
W.E. Ribelin and G. Migaki, eds. University of Wisconsin Press,
Madison, pp. 477-496.
97
-------
Roe, F.J.C. 1954. Liver changes in urethane-treated mice appearing
after a long latent interval. 32nd Annual Report of the British
Empire Cancer Campaign. 170 pp.
Salt, G. 1970. The cellular defense reactions of insects. Cambridge
monographs in experimental biology, #16. Cambridge University
Press. 118 pp.
Scherer, E. and P. Emnelot. 1975. Kinetics of induction and growth of
precancerous liver cell foci and liver tumor formation by diethyl -
nitrosamine in the rat. European Journal of Cancer 11:689-696.
Scherer, E. and P. tmrnelot. 1976. Kinetics of induction and growth of
enzyme-deficient islands involved in hepatocarcinogenesis. Cancer
Research 36_:2544-2554.
Schillings. P.H.H. and J.H.S. Stekhoven. 1980. Atlas of Glomerular
Histopathology. S. Karger, (Basel, New York"]"^ 143 pp.
Sherwood, M.J. and b.B. flcCain. 1976. Comparison of fin erosion
disease: Los Angeles and Seattle. Annual Report, Southern
California Coastal Water Research Project, El Segundo, California.
Sims, P., P.L. Grover, A. Swaisland, K. Pal and A. Hewer. 1974.
Metabolic activation of benzo[a]pyrene proceeds by a diol-epoxide.
Nature (London) 252:326-328.
Sinnhuber, R.O., D.J. Lee, H.H. Wales and J.L. Ayres. 1968. Dietary
factors and hepatoma in rainbow trout (Salmo gairdneri).
II. Cocarcinogenesis by cyclopropenoid fatty acids and the effect
of gossypol and altered lipids on aflatoxin-induced liver cancer.
Journal of the National Cancer Institute _41:1293-1301.
Sinnhuber, R.O., J.D. Hendricks, J.H. Wales and G.B. Putman. 1977.
Neoplasms in rainbow trout, a sensitive animal model for
environmental carcinogenesis. In, Aquatic Pollutants and Biologic
Effects with Emphasis on Neoplasia, H.F. Kraybill, C.J. Dawe,
J.C. Harshbarger, R.G. Tardiff, eds. Annals of the New York
Academy Sciences 298:389-408.
Smith, C., T.H. Peck, R.J. Klauda and J.B. McLaren. 1979. Hepatomas
in Atlantic torncod Microgadus tomcod (Walbaum) collected in the
Hudson River estuary in New York. Journal of Fish Diseases 2^:313-319.
Sokal, R. and F. Rohlf. 1969. Biometry. W.H. Freeman arid Co., San
Francisco. 776 pp.
Solt, D.B., A. Medline and E. Farber. 1977. Rapid emergence of carcinogen-
induced hyperplastic lesions in a new model for the sequential analysis
of liver carcinogenesis. American Journal of Pathology 88:595-618.
98
-------
Squire, R.A. and M.H. Levitt. 1975. Report of a workshop on classification
of specific hepatocellular lesions in rats. Cancer Research 35:3214-3223.
Stephens, J.S., Jr., R.K. Johnson, G.S. Key, and J.E. McCosker. 1970.
The comparative ecology of three sympatric species of California
blennies of the genus Hypsoblennius (Teleostomi Blenniidae).
Ecological Monographs 4-0(2)213-233.
Struthers, B.J., J.H. Wales3 D.J. Lee and R.O. Sinnhuber. 1975.
Liver composition and histology of rainbow trout fed cyclopropenoid
fatty acids. Experimental and Molecular Pathology 23:164-170.
Thompson, S.W.
Charles C.
1966. Selected Histochemical and Histopathological Methods.
Thomas, Springfield, Illinois. 1639 pp.
Varanasi, U., M. Uhler and S.I. Stranahan. 1978. Uptake and release of
napthalene and its metabolites in skin and epidermal mucus of
salmonids. Toxicology and Applied Pharmacology 44:277-289.
Varanasi, U., D.J. Gmur and P.A. Treseler. 1979. Influence of time
and mode of exposure on biotransformation of naphthalene by juvenile
starry flounder (Platichthys stellatus) and rock sole (Lepidopsetta
bilineata). Archives of Environmental Contamination and Toxicology
j3:673-692.
Varanasi, U. and D.J. Gmur. 1981a. Hydrocarbons and metabolites in
English sole (Parophrys yetulus) exposed simultaneously to [H^]
benzo[a] pyrene and [l^C] naphthalene in oil-contaminated sediment.
Aquatic Toxicology 1:49-67.
In vivo metabolism of naphtalene
In, Chemical Analysis and
and
Varanasi, U. and D.J. Grnur. 1981b.
and benzo[a]pyrene by flatfish.
Biological Fate: Polynuclear Aromatic Hydrocarbons, M. Cook
A.J. Dennis, eds.Fifth International Symposium on Polynuclear
Aromatic Hydrocarbons, Battelle Memorial Institute. Battelle Press,
Columbus, pp. 367-376.
Walker, R. 1947. Lyrnphocyctis disease and neoplasia in fish. Report
of lymphocystis in Ophiodon elongatus from the Straits of Georgia,
B.C. Anatomical Record J39:559-560 .
Walker, R. and R. Weissenberg. !965. Conformity of light and electron
microscopic studies on virus particle distribution in lymphocystis
tumor cells of fish. Archives of the New York Academy of Sciences
126:375-385.
Ward, J.M. and G. Vlahakis. 1978. Evaluation of hepatocellular
neoplasms in mice. Journal of the National Cancer Institute
_61:807-811.
Wehner, H. 1968. Stereologische untersuchungen am mesangium normaler
menschlicher nieren. Virchows Arch. Abt. A. Path Anat. 344:286-194.
99
-------
Weisburger, J.H., R.S. Yamamoto, 6.M. Williams, P.H. Grantham, T. Matsushima
and E.K. Weisburger. 1972. On the sulfate ester of N-hydroxy-N-2-
fluyorenylactanide as a key ultimate hepatocarcinogen in the rat.
Cancer Research _32_:491-50U.
Weissenberg, R. 1955. The third spontaneous case of lymphocystis
virus disease of fish from the Pacific Coast of North America.
Anatomical Record (Abstract) _122(3):434-435
Wellings, S.R., B.B. McCain, and B.S. Butler. 1976a. Epidermal
papillomas in Pleuronectidae of Puget Sound, Washington. Progress
in Experimental Tumor Research 20:55-74.
Wellings, S.R., C.E. Alpers, B.B. McCain, and B.S. Miller. 1976b. Fin
erosion disease of starry flounder (Platichthys stellatus) and
English sole (Parophrys vetulus) in the estuary of the Duwamish
River, Seattle, Washington. Journal of the Fisheries Research
Board of Canada 3^:2577-2586
Wellings, S.R., C.E. Alpers, B.B. McCain and M.S. Myers. 1977. Fish
disease in the Bering Sea. In, Aquatic Pollutants and Biologic
Effects with Emphasis on Neoplasia, H.F. Kraybill, C.J. Dawe,
J.C. Harshbarger, R.G. Tardiff, eds. Annals of the New York
Academy of Sciences 298:290-304.
Williams, G.M. 1976. Functional markers and growth behavior of
preneoplastic hepatocytes. Cancer Research 36:2540-2543.
Wingert, R.C., B.B. McCain, K.V. Pierce, S.F. Borton, D.T. Griggs and
B.S. Miller. 1976. Ecological and disease studies of demersal
fishes in the vicinity of sewage outfalls. 1976 Research in
Fisheries. Annual Report of the College of Fisheries, University
of Washington, Seattle.
Word, Jack Q. 1976. Biological comparison of grab sampling devices.
Annual Report, Southern California Coastal Water Research Project,
Long Beach, California, pp. 189-194
Zachary, A. and I. Paperna. 1977. Epitheliocystis disease in the
striped bass Morpne saxatilis from the Chesapeake Bay. Canadian
Journal of Microbiology 23:T4"04-1414.
Zar, J.H. 1974. Biostatistical Analysis. Prentice-Hall, Inglewood
Cliffs New Jersey.620 pp.
100
.S. GOVERNMENT PRINTING OFFICE:1982 576-001 / 1226
------- |