- Email: [email protected]
Effects of Gonadal Neoplasms on Oogenesis in Softshell Clams,Mya arenaria
JOBNAME: 67#2 96 PAGE: 1 SESS: 2 OUTPUT: Mon Jun 10 12:40:57 1996 /xypage/worksmart/tsp000/67596b/11 JOURNAL OF INVERTEBRATE PATHOLOGY ARTICLE NO.
67, 161–168 (1996)
0024
Effects of Gonadal Neoplasms on Oogenesis in Softshell Clams, Mya arenaria BRUCE J. BARBER Department of Animal, Veterinary and Aquatic Sciences, University of Maine, Orono, Maine 04469 Received May 25, 1995; accepted November 27, 1995
The average prevalence of gonadal neoplasms in softshell clams, Mya arenaria, from Whiting Bay, Washington County, Maine, was 19.4% in 1994. Monthly prevalences ranged from 10 to 26.7%. Neoplasms ranged in intensity from few, small foci of undifferentiated germ cells (Stage 1), to 50–100% of gonadal follicles being involved (Stage 2), to invasion and metastasis with loss of tissue architecture (Stage 3), indicating that the disease is progressive and lethal. There was no relationship (P > 0.05) between prevalence of neoplasms and clam size between 45.7 and 60.7 mm mean shell length. Clams of both sexes were affected, but females were significantly more likely (P ≤ 0.025) to have neoplasms than males. Female clams with neoplasms produced significantly fewer (P ≤ 0.001) gametes than healthy clams. Overall reduction of gamete number was 66%, resulting from direct displacement of gametes by tumor cells in affected gonadal follicles. Compared to healthy clams, clams with neoplasms exhibited a significantly lower (P ≤ 0.001) mean oocyte diameter before spawning and a significantly greater (P ≤ 0.001) mean oocyte diameter after spawning, as the result of a general (throughout the entire gonad) inhibition of normal oogenesis and spawning. It is concluded that gonadal neoplasms have a negative impact on the reproductive output of the Whiting Bay clam population. © 1996 Academic Press, Inc. KEY WORDS: Mya arenaria; gonadal neoplasm; prevalence; intensity; oogenesis effects. INTRODUCTION
Gonadal neoplasms (germinomas, tumors) have been reported in several species of marine bivalve molluscs, including the softshell clam, Mya arenaria (Peters et al., 1994). Germinomas, ranging from noninvasive to metastatic, in both male and female clams, are char1 Representative cases of M. arenaria with gonadal neoplasms have been assigned Accession Nos. RTLA 6071, 6072, and 6073 in the Registry of Tumors in Lower Animals, Department of Pathology, George Washington University Medical Center, 2300 I Street, NW, Washington, DC 20037.
acterized by the presence of monomorphic, basophilic, undifferentiated germ cells that proliferate rather than mature (Gardner et al., 1991). The nucleii of these cells are often eccentric and the nucleolus is frequently difficult to distinguish in the abundantly clumped chromatin (Brown et al., 1977). Disease progression has been described, based on histopathology, by Yevich and Barszcz (1977). In early cases, small foci of neoplastic cells occupy a small portion of one or more follicles. Development of oocytes or spermatocytes apparently continues, even as the neoplastic cells increase in number within follicles. In some cases, however, follicles are completely packed with neoplastic cells and as a result, sexual identity is lost. Metastasis, in which tumor cells have broken through basement membrane and become established at a remote site, is commonly noted. Cell types at the metastatic sites are the same as those seen in the follicles, except that mitotic figures are more prevalent. These neoplasms have been termed “germinomas” and are similar to those reported in the quahog, Mercenaria mercenaria and other species (Barry and Yevich, 1972; Hesselman et al., 1988; Peters et al., 1994). Based on the above description, gonadal neoplasms are considered malignant. Even though M. arenaria is distributed from Labrador to South Carolina on the east coast of North America (Hidu and Newell, 1989), gonadal neoplasms have only been reported from locations in Maine. This condition was initially described by Barry and Yevich (1975) and Yevich and Barszcz (1976, 1977) in clams from Long Cove, Searsport, where 1 to 22% of clams examined between 1971 and 1975 had tumors. Similar neoplasms were present in 3.6% of clams examined from Long Cove in 1976 (Brown et al., 1977). Germinomas were also present in 3% of clams taken from Roque Bluffs (near Machiasport) and in 35% of clams collected from Dennys Bay (near Dennysville) collected in March 1980 (Gardner et al., 1991). In these cases, occurrence of neoplasms was correlated with environmental contamination from either hydrocarbons (Barry and Yevich, 1975; Yevich and Barszcz, 1976, 1977) or herbicides (Gardner et al., 1991). Harshbarger et al. (1979)
161 0022-2011/96 $18.00 Copyright © 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.
JOBNAME: 67#2 96 PAGE: 2 SESS: 2 OUTPUT: Mon Jun 10 12:40:57 1996 /xypage/worksmart/tsp000/67596b/11 162
BRUCE J. BARBER
reported the presence of intranuclear inclusions in tumor cells from clams examined by Yevich and Barszcz (1976, 1977), but a viral nature was never confirmed. Thus the etiology of this disease is presently unknown. Since 1982, the production of clams from flats in Washington County, Maine, has declined from 240,000 bushels to 24,000 bushels (Maine Department of Marine Resources). Although there are several potential reasons for this decline, the impact of disease on reproductive development has not been previously examined. A survey of several clam flats in Washington County, Maine, conducted in September 1993 revealed that a population in Whiting Bay had a 43% prevalence of gonadal neoplasms (Barber, 1995). In this study, the prevalence and intensity of gonadal neoplasms in this population was determined throughout 1994 and the impact of the disease on oogenesis was quantified.
parametric mean square successive difference test (Zar, 1984, p. 419). The effect of gonadal neoplasms on oogenesis was assessed using an image analysis system consisting of a video camera mounted on a Nikon Labophot microscope and connected to a Truevision Targa+ videographics adapter. Images of five gonadal fields from each female clam (600×) were enhanced, and data were collected using Image Pro Plus software. In each of the five fields, the number of oogonia, oocytes, and ova were counted, and the diameter of those having a visible nucleus was measured. This method has been found to accurately reflect oogenesis in marine bivalves (Barber and Blake, 1991; Barber et al., 1988). The mean number of gametes per field and the mean diameter of those gametes for clams with and without neoplasms were compared for each sampling date using two-way ANOVA (Wilkinson et al., 1992).
MATERIALS AND METHODS
Adult clams (n 4 11–30) were collected using a hand rake from an intertidal flat (Whiting Bay, latitude 44° 499 180 N; longitude 67° 099 460 W) located near Whiting, in Washington County, Maine. Samples were collected in all months in 1994 except February, when the flats were covered with ice. Water temperature (mercury thermometer, °C) and salinity (refractometer, ppt) were recorded at the time of collection (low tide) beginning May 1994. In the laboratory, clams were measured (shell length, mm), removed from their shells, and placed in Helly’s fixative (Barszcz and Yevich, 1975). Transverse sections (3–4 mm) through kidney, digestive gland, gonad, gills, and foot were removed, placed in cassettes, and dehydrated and cleared in a Fisher Histomatic Tissue Processor (Model 166). Tissues were embedded in Paraplast, sectioned (6 mm), and stained with Shandon instant hematoxylin and eosin Y before being coverslipped. Finished slides were examined with a Nikon Labophot microscope (100×) for the presence of gonadal neoplasms. In clams having neoplasms, the degree of disease development was assigned using criteria similar to those defined by Peters et al. (1994). Stage 1 neoplasms consisted of undifferentiated germs cells partially or totally filling one or more (but less than half of the total) follicles; loss of gonadal architecture was rare. Stage 2 neoplasms were characterized by the presence of undifferentiated germ cells in over 50% of visible follicles; usually there was some loss of tissue architecture, but invasion and metastasis were not evident. In Stage 3 neoplasms, a few to all follicles were involved, loss of tissue architecture was common, and invasion or metastasis was evident. The prevalence (percentage of clams having neoplasms) and the mean intensity (Stage) of neoplasms for each monthly sample were examined for serial randomness using the non-
RESULTS
Mean shell length of clams examined ranged from 45.7 to 60.7 mm (Table 1). Although these means were significantly different (ANOVA, P # 0.001), there was no relationship (correlation, P 4 0.729) between clam size and the prevalence of gonadal neoplasms over this range. Water temperature at the collection site increased from 12.1°C in May to a maximum of 19.1°C in July and then declined to 0.6°C in December (Table 1). Salinity ranged from 28 to 34 ppt between May and November, but was 20 ppt in December (Table 1). Thirty clams were collected each month in 1994 except February, May, and July (Table 2). In all, 310 clams from Whiting Bay were examined histologically over the course of the study. Of these, 60 had gonadal neoplasms, for an overall prevalence of 19.4%.1 The 60 neoplasms included 22 Stage 1 cases, 33 Stage 2 cases, and 5 Stage 3 cases. At the gross level, it was possible TABLE 1 Dates That Softshell Clams, M. arenaria, Were Collected from Whiting Bay, Mean and SE of Shell Length (mm), Water Temperature (°C), and Salinity (ppt) Shell length
Collection date
Mean
SE
Temperature (°C)
Salinity (ppt)
1/11/94 3/17/94 4/19/94 5/20/94 6/15/94 7/18/94 8/14/94 9/11/94 10/17/94 11/13/94 12/13/94
49.8 45.7 48.7 52.2 54.7 57.1 57.2 60.7 52.9 52.8 52.7
1.0 1.0 1.0 1.0 1.0 1.7 1.0 1.0 1.0 1.0 1.0
— — — 12.1 14.5 19.1 16.5 14.3 11.1 8.8 0.6
— — — 28 29 32 32 33 34 31 20
JOBNAME: 67#2 96 PAGE: 3 SESS: 3 OUTPUT: Mon Jun 10 12:40:57 1996 /xypage/worksmart/tsp000/67596b/11 EFFECTS OF GONADAL NEOPLASMS ON OOGENESIS IN Mya
163
to distinguish clams with advanced neoplasms (Stages 2 or 3) from clams without neoplasms by the shrunken, darkened, mottled appearance of their gonads. Gonadal neoplasms were found in all months sampled and in both male and female clams (Table 2). In May, June, and July, all neoplasms occurred in female clams. The sex ratio of all clams without neoplasms was not significantly different from 1:1, but the sex ratio of all clams with neoplasms was significantly different from 1:1 (x2, P ø 0.025), with predominantly more female clams having neoplasms (Table 2). Overall, monthly prevalences of neoplasms ranged from 10% in June and August to 26.7% in both September and October (Fig. 1). Mean monthly intensity was greatest (2.2) in April and May and lowest (1.0) in June (Fig. 1). Seasonal variability of both prevalence and mean intensity, however, was random (not significant, P > 0.25). Gonadal neoplasms had a negative impact on the
number of eggs produced per individual (Fig. 2). In both healthy (without neoplasms) and diseased (with neoplasms) clams, mean egg number per field (for an individual) was greater in the months (March–June) preceeding spawning; this was especially apparent in healthy clams. The mean number of female gametes (oogonia, oocytes, ova) per field was lower in diseased clams than in healthy clams in all months. Two-way ANOVA revealed that there were significant (P # 0.001) differences in the number of female gametes both between samples (dates) and between healthy and diseased clams; there was also a significant (P # 0.001) interaction between date and disease status (Table 3). The overall effect of gonadal neoplasia was a 66% reduction in the number of female gametes produced per individual. Gonadal neoplasms also affected the size of eggs produced, but the effect varied over the course of the oogenic cycle (Fig. 3). Mean egg diameter in healthy clams was greatest in the months of June and July. A sharp decline in mean oocyte diameter between July and August indicated that most healthy clams spawned during this period. In clams with neoplasms, mean egg diameter was greatest in August, after healthy clams had spawned. Two-way ANOVA revealed that mean egg size varied significantly (P # 0.001) with date but not with disease status over the entire study; there was also a significant (P # 0.001) date–disease interaction (Table 4). On closer examination of Fig. 3, it can be seen that in the months preceeding spawning (January–July), mean egg diameter was lower in clams with neoplasms than in healthy clams; the magnitude of this difference generally increased in relation to oogenic activity. The opposite was true for the months immediately after spawning (August–November), when clams with neoplasms had greater mean egg diameters than healthy clams; this difference decreased with time after spawning. When separate two-way ANOVA were run on these two sub-
FIG. 1. Monthly percentage prevalence (Y1; bars) and mean intensity (Y2; line) of gonadal neoplasms in clams, M. arenaria, from Whiting Bay.
FIG. 2. Mean (+SE) number of female gametes per microscopic field from clams, M. arenaria, with and without gonadal neoplasms.
TABLE 2 Numbers of Clams Examined Histologically at Each Sampling Date and the Distribution of Neoplasms by Stage and Gender Stage
Gender
Collection date
No. examined
1
2
3
M
F
U
1/11/94 3/17/94 4/19/94 5/20/94 6/15/94 7/18/94 8/14/94 9/11/94 10/17/94 11/13/94 12/13/94 Totals
30 30 30 29 30 11 30 30 30 30 30 310
4 2 1 1 3 1 1 3 4 2 0 22
2 4 4 4 0 1 2 5 3 2 6 33
0 0 2 2 0 0 0 0 1 0 0 5
1 2 3 0 0 0 1 4 3 2 3 19
4 3 3 6 3 2 2 4 5 2 2 36
1 1 1 1 0 0 0 0 0 0 1 5
Note. M, male; F, female; U, undetermined.
JOBNAME: 67#2 96 PAGE: 4 SESS: 2 OUTPUT: Mon Jun 10 12:40:57 1996 /xypage/worksmart/tsp000/67596b/11 164
BRUCE J. BARBER
TABLE 3 Two-Way Analysis of Variance for Effects of Collection Date and Disease on Mean Egg Number per Microscopic Field of Individual Female Clams, M. Arenaria Source of variation
Sum of squares
Degrees of freedom
Mean square
F value
Probability
Date Disease Date × Disease Error
5921.54 3327.67 1670.54 4685.57
10 1 10 143
592.15 3327.67 167.05 32.77
18.07 101.56 5.10
0.000 0.000 0.000
sets of data, there were highly significant (P # 0.001) differences by date and by disease status (P # 0.004), and for the postspawning months, there was a significant date–disease interaction (P # 0.002). Thus in the months preceeding spawning, clams with neoplasms were producing fewer and smaller gametes than clams without neoplasms (Fig. 4). After spawning, clams with neoplasms contained fewer, larger gametes than healthy clams (Fig. 5). DISCUSSION
There are two major types of neoplasms that occur in marine bivalves, including M. arenaria (Yevich and Barszcz, 1976, 1977; Brown et al., 1977; Peters, 1988; Peters et al., 1994). The first type, hematopoietic neoplasms, or disseminated sarcomas, are characterized by unusual cells in the connective tissue, blood vessels, and sinuses of the visceral mass, muscle, and mantle tissue (Peters, 1988). The cells are hypertrophied, usually enlarged two to four times the diameter of normal hemocytes, and have a hyperchromatic nucleus that is often pleomorphic. Proportions of abnormal cells in affected clams generally increase over time and result in death (Cooper et al., 1982; Barber, 1990; Brousseau and Baglivo, 1991). A viral etiology has been suggested for this disease, but other factors, such as chemical pollution and genetics may be involved (Oprandy et al., 1981; Oprandy and Chang, 1983; Peters, 1988). Al-
FIG. 3. Mean (+SE) diameter of female gametes from clams, M. arenaria, with and without gonadal neoplasms.
TABLE 4 Two-Way Analysis of Variance for Effects of Collection Date and Disease on Mean Egg Diameter of Individual Female Clams, M. arenaria Source of variation
Sum of squares
Degrees of freedom
Mean square
F value
Probability
Date Disease Date × Disease Error
3022.71 9.14 1395.58 3555.64
10 1 10 141
302.27 9.14 139.56 25.22
11.99 0.36 5.53
0.000 0.548 0.000
though hematopoietic neoplasms are epizootic throughout the range of M. arenaria (Brown et al., 1977; Farley et al., 1986; Brousseau, 1987a; Reinisch et al., 1984; Morrison et al., 1993), none were seen in the clams from Whiting Bay examined in the present study. The second major type of proliferative disorder reported to occur in marine bivalves, including M. arenaria, is the gonadal neoplasm or germinoma (Hesselman et al., 1988; Peters et al., 1994), the focus of this study. Based on histological examination, the proliferation of undifferentiated germ cells first in the gonad and then into surrounding tissues suggests a progressive and lethal nature, although this has not been demonstrated (Barry and Yevich, 1975; Yevich and Barszcz, 1977; Peters et al., 1994). Initial reports of this disease were correlated with hydrocarbon pollution (Barry and Yevich, 1975; Yevich and Barszcz, 1976, 1977; Brown et al., 1977), but subsequent attempts to induce the disease by exposure to hydrocarbons were unsuccessful (Gardner et al., 1991). More recently, higher prevalences of gonadal neoplasms in Washington County, Maine, have been correlated with a history of herbicide application on adjacent land areas, suggesting that there could be a chemical etiology for this disease (Gardner et al., 1991). The fact, however, that gonadal neoplasms have only been found in Maine despite the occurrence of M. arenaria in more chemically contaminated areas along the east coast than Maine (Reinisch et al., 1984), suggests instead that the etiological agent may be of a localized, genetic nature. Gonadal neoplasms were epizootic at the Whiting Bay site, with an overall prevalence of 19.4%. Although neoplasms were identified at the cellular level by examination of histological preparations, it was possible to identify clams with advanced neoplasms by the gross appearance of their gonad. In a separate sample of 100 clams collected from Whiting Bay in November 1994, 12 clams that had shrunken, darkened, mottled gonads were all found to have advanced (Stage 2 or 3) neoplasms upon histological examination; among 20 clams that appeared to be healthy at the gross level, 2 were found to have very early (Stage 1) neoplasms when examined histologically. Thus, gross examination of
JOBNAME: 67#2 96 PAGE: 5 SESS: 2 OUTPUT: Mon Jun 10 12:40:57 1996 /xypage/worksmart/tsp000/67596b/11 EFFECTS OF GONADAL NEOPLASMS ON OOGENESIS IN Mya
165
FIG. 4. Gonad of female clams, M. arenaria, collected from Whiting Bay in May 1994. (A) Gonad without neoplasm (175×) and numerous developing oocytes (OC). (B) Gonad (175×) with neoplasm (N); note that oocyte (OC) growth is inhibited throughout the gonad. Bar, 100 mm.
clams might be useful for determining the approximate prevalence of gonadal tumors at a site without the time and expense associated with histology. Based on histological observations and the distribution of cases among all three intensity stages, this study also suggests that this disease is progressive and possibly lethal. Early cases are confined to one or more gonadal follicles with little apparent effect on surrounding follicles, tissues, or their physiological func-
tion (Stage 1). As the cells proliferate and spread, more follicles become involved, and gonadal architecture and function are affected (Stage 2). Once the tumor cells break through the follicle wall and metastasize, adjacent and distant (gill) tissues become involved (Stage 3). There were considerably fewer Stage 3 cases than Stage 2 cases, suggesting that few clams survived once the disease reached an advanced stage. The rate at which this disease progresses and results in death is
JOBNAME: 67#2 96 PAGE: 6 SESS: 2 OUTPUT: Mon Jun 10 12:40:57 1996 /xypage/worksmart/tsp000/67596b/11 166
BRUCE J. BARBER
FIG. 5. Gonad of female clams, M. arenaria, collected from Whiting Bay in August 1994. (A) Gonad without neoplasm (175×) after spawning; visible cells are primary oocytes (OC) or spherical products of cytolysis (C). (B) Gonad (175×) with neoplasm (N); note that follicles without tumor cells contain unspawned ova (OV). Bar, 100 mm.
unknown. For M. arenaria with early cases of hematopoietic neoplasia, about 60 days are required for disease progression and death to occur (Barber, 1990). At the Whiting Bay site, it was not unusual to find dead clams with intact shells in the sediment and no visible sign of predation. In a few instances, decomposed clam tissue was still associated with the shells. Future studies will attempt to more clearly determine the mortality rate associated with this disease.
All the clams examined in this study were adults (reproductively mature), and there was no evident relationship between prevalence of gonadal neoplasia and shell height over the size range of 45.7 to 60.7 mm. Clams of this size from Whiting Bay are estimated to be 7–15 years old (B. Beal, pers. commun.). The minimum age/size at which clams develop gonadal neoplasms is unknown, but it would presumably be after attaining sexual maturity, which occurs at age >1 year
JOBNAME: 67#2 96 PAGE: 7 SESS: 2 OUTPUT: Mon Jun 10 12:40:57 1996 /xypage/worksmart/tsp000/67596b/11 EFFECTS OF GONADAL NEOPLASMS ON OOGENESIS IN Mya
and shell height >20 mm (Brousseau, 1987b). One sample of 20 clams with a mean size of 25.4 mm was obtained in November from a site near Whiting Bay where neoplasms occur. These smaller clams had developed gonads that appeared normal for the time of year (spawned out); no gonadal neoplasms were found. Thus the minimum size at which M. arenaria develops gonadal neoplasia is probably between 25.4 and 45.7 mm shell height (age 2–7 years). This may represent the time required for the etiological agent to manifest itself. Previous reports of gonadal neoplasms in M. arenaria indicated that both male and female clams were affected (Yevich and Barszcz, 1976; Peters, 1994), but the finding that gonadal neoplasms occur almost twice as often in female clams (60%) than male clams (32%) has not been previously been noted for this species. In contrast, healthy clams from Whiting Bay (this study) as well as those from Long Island Sound (Brousseau, 1987b) had a 1:1 ratio of males to females. A similar predominance of neoplasms in females over neoplasms in males, however, has been reported for the quahog, M. mercenaria. Barry and Yevich (1972) examined 539 individuals and found 12 neoplasms in females and only 2 neoplasms in males. Hesselman et al. (1988) examined 1263 individuals and of the 147 with gonadal neoplasms, 61% were females, 33% were males, and 6% were not recognizable as either male or female. The fact that gonadal neoplasms occur primarily in females of both clam species suggests that a common disease mechanism might be involved. Healthy clams at Whiting Bay underwent a normal gametogenic cycle. Mean number of eggs per microscope field increased from January to May and then declined as oocytes continued to grow. Most oocyte growth occurred in May and June, in conjunction with increasing water temperature. Spawning began as early as June, based on histological observation. The sharp decrease in mean oocyte diameter seen in August, however, indicated that most spawning activity occurred at this time, in conjunction with decreasing water temperature. M. arenaria at Whiting Bay exhibited one spawning period, which is in agreement with previous studies of clam populations north of Cape Cod (Ropes and Stickney, 1965). Gonadal neoplasms had a direct negative effect on the number of eggs produced. The more advanced the disease, the greater the proportion of gonadal follicles containing tumor cells and the lower the proportion of normal follicles containing oocytes. The magnitude of reduction in mean egg number per field between normal and diseased clams at any date was thus related to the intensity of the disease. The interaction that occurred between month and disease status most likely was the result of the increase in disease intensity that occurred in April and May (and hence decrease in egg number) at the same time that healthy clams were
167
experiencing an increase in egg number per field. Overall, female clams with neoplasms produced 66% fewer gametes than clams without neoplasms. The effect of gonadal neoplasms on the size of eggs produced appeared to be indirect. That is, the effect on oocytes was seen throughout the gonads of affected clams, not merely in follicles containing tumor cells. The net effect was an inhibition of the normal oogenic process. The impact of neoplasms on oocyte development was related to the timing of disease acquisition, which accounts for the interaction seen between date and disease status. For clams in which the disease appeared early in the oogenic process, an overall reduction in mean oocyte size resulted, as growth of oocytes was stunted (Fig. 4). If neoplasms are progressive and lethal, these clams would most likely not have lived to produce normal ova. If any ova were produced, they would be smaller and less likely to produce viable larvae than ova from healthy clams (Kraeuter et al., 1982). For clams in which the disease appeared late in the developmental phase, the effect was an increase in mean oocyte diameter, as the otherwise normal appearing ova were not spawned along with ova from healthy clams (Fig. 5). The mean diameter of diseased clams in August was even larger than that of healthy clams in July, indicating that while mature ova were released from healthy clams as early as June, they were retained in diseased clams. Whether or not diseased clams with oocytes ultimately spawned and what the fate of these gametes was are unknown. It is clear that diseased clams, even if they spawned would not produce as many eggs, and the retarded timing of release would reduce the likelihood of fertilization by healthy sperm and larval survival as water temperature decreased. This interaction between disease and oogenesis is complicated by differing rates of disease initiation and progression throughout the year. The observation of a general inhibition of oogenesis in clams with neoplasms suggests that the disease affects the reproductive process at some basic level, perhaps by disrupting or blocking normal neurohormonal cycles, which are involved in the endogenous regulation of gametogenesis in marine invertebrates (Schroeder, 1987). Disseminated sarcoma cells in M. arenaria produce proteins that, besides promoting tumor growth, could be cytotoxic (Sunila, 1992), thus possibly inhibiting oocyte development. Additionally, the high respiration rates of mitotically active tumor cells could possibly divert energy away from oogenesis (Sunila, 1991). Estimation of the impact of gonadal neoplasms on the reproductive capacity of the Whiting Bay clam population depends on several factors. This study has established the high prevalence of this disease at this location and the serious impact that it has on oogenesis. The disproportionate number of female clams having gonadal neoplasms has further implications, especially if egg production is a factor limiting overall
JOBNAME: 67#2 96 PAGE: 8 SESS: 2 OUTPUT: Mon Jun 10 12:40:57 1996 /xypage/worksmart/tsp000/67596b/11 168
BRUCE J. BARBER
reproductive success. The rates of disease progression and mortality remain to be explored. Disease is one potential factor limiting clam production in Whiting Bay. ACKNOWLEDGMENTS I thank the following people for their help: Brian Beal, Ken Vencile, Robert Bayer, Bruce Wiersma, Neil Greenberg, and Brian Barber for field assistance and Dawna Beane for the histological preparations. Many thanks to Paul Yevich and Esther Peters for verifying neoplasms. Stephen Fegley and Esther Peters provided thoughtful comments on an earlier version of the manuscript. Finally, I would like to thank Bob and Jane Bell for access to the study area. This project was supported in part by the Maine Aquaculture Innovation Center. This is Maine Agricultural and Forestry Experiment Station external publication 1955. REFERENCES Barber, B. J. 1990. Seasonal prevalence and intensity and disease progression of neoplasia in soft shell clams, Mya arenaria, from the Shrewsbury River, New Jersey. In “Pathology in Marine Science” (F. O. Perkins and T. Cheng, Eds.), pp. 377–386. Academic Press, New York. Barber, B. J. 1995. Seasonal prevalence of gonadal neoplasms in clams, Mya arenaria, in Maine and impact on oogenesis. J. Shellfish Res. 14, 238. Barber, B. J., and Blake, N. J. 1991. Reproductive physiology. In “Scallops: Biology, Ecology and Aquaculture” (S. E. Shumway, Ed.), pp. 377–428. Elsevier, New York. Barber, B. J., Getchell, R., Shumway, S., and Schick, D. 1988. Reduced fecundity in a deep-water population of the giant scallop Placopecten magellanicus in the Gulf of Maine, USA. Mar. Ecol. Prog. Ser. 42, 207–212. Barry, M. M., and Yevich, P. P. 1972. Incidence of gonadal cancer in the quahog Mercenaria mercenaria. Oncology 26, 87–96. Barry, M., and Yevich, P. P. 1975. The ecological, chemical and histopathological evaluation of an oil spill site. Part III. Histopathological studies. Mar. Poll. Bull. 6, 171–173. Barszcz, C. A., and Yevich, P. P. 1975. The use of Helly’s fixative for marine invertebrate histopathology. Comp. Pathol. Bull. 7, 4. Brousseau, D. J. 1987a. Seasonal aspects of sarcomatous neoplasia in Mya arenaria (soft-shell clam) from Long Island Sound. J. Invertebr. Pathol. 50, 269–276. Brousseau, D. J. 1987b. A comparative study of the reproductive cycle of the soft-shell clam, Mya arenaria in Long Island Sound. J. Shellfish Res. 6, 7–15. Brousseau, D. J., and Baglivo, J. A. 1991. Field and laboratory comparisons of mortality in normal and neoplastic Mya arenaria. J. Invertebr. Pathol. 57, 59–65. Brown, R. S., Wolke, R. E., Saila, S. B., and Brown, C. W. 1977. Prevalence of neoplasia in 10 New England populations of the soft-shell clam (Mya arenaria). Ann. NY Acad. Sci. 298, 522–534. Cooper, K. R., Brown, R. S., and Chang, P. W. 1982. The course and mortality of a hematopoietic neoplasm in the soft-shell clam, Mya arenaria. J. Invertebr. Pathol. 39, 149–157. Farley, C. A., Otto, S. V., and Reinisch, C. L. 1986. New occurrence of epizootic sarcoma in Chesapeake Bay soft shell clams, Mya arenaria. Fish. Bull. 84, 851–857. Gardner, G. R., Yevich, P. P., Hurst, J., Thayer, P., Benyi, S., Harshbarger, J. C., and Pruell, R. J. 1991. Germinomas and teratoid
siphon anomalies in softshell clams, Mya arenaria, environmentally exposed to herbicides. Environmental Health Perspectives 90, 43–51. Harshbarger, J. C., Otto, S. V., and Chang, S. C. 1979. Proliferative disorders in Crassostrea virginica and Mya arenaria from the Chesapeake Bay and intranuclear virus-like inclusions in Mya arenaria with germinomas from a Maine oil spill site. Haliotis 8, 243–248. Hesselman, D. M., Blake, N. J., and Peters, E. C. 1988. Gonadal neoplasms in hard shell clams Mercenaria spp., from the Indian River, Florida: Occurrence, prevalence, and histopathology. J. Invertebr. Pathol. 52, 436–446. Hidu, H., and Newell, C. R. 1989. Culture and ecology of the softshelled clam, Mya arenaria. In “Clam Mariculture in North America” (J. J. Manzi and M. Castagna, Eds.), pp. 277–292. Elsevier Science, Amsterdam. Kraeuter, J. N., Castagna, M., and van Dessel, R. 1982. Egg size and larval survival of Mercenaria mercenaria (L.) and Argopecten irradians (Lamarck). J. Exp. Mar. Biol. Ecol. 56, 3–8. Morrison, C. M., Moore, A. R., Marryatt, V. M., and Scarratt, D. J. 1993. Disseminated sarcomas of soft shell clams, Mya arenaria Linnaeus 1758, from sites in Nova Scotia and New Brunswick. J. Shellfish Res. 12, 65–69. Oprandy, J. J., and Chang, P. W. 1983. 5-bromodeoxyuridine induction of hemopoietic neoplasia and retrovirus activation in the softshell clam, Mya arenaria. J. Invertebr. Pathol. 42, 196–206. Oprandy, J. J., Chang, P. W., Pronovost, A. D., Cooper, K. R., Brown, R. S., and Yates, V. J. 1981. Isolation of a viral agent causing hematopoietic neoplasia in the soft-shell clam, Mya arenaria. J. Invertebr. Pathol. 38, 45–51. Peters, E. C., Yevich, P. P., Harshbarger, J. C., and Zaroogian, G. E. 1994. Comparative histopathology of gonadal neoplasms in marine bivalve molluscs. Dis. Aquat. Org. 20, 59–76. Peters, E. C. 1988. Recent investigations on the disseminated sarcomas of marine bivalve molluscs. Am. Fish. Soc. Spec. Publ. 18, 74–92. Reinisch, C. L., Charles, A. M., and Stone, A. M. 1984. Epizootic neoplasia in soft shell clams collected from New Bedford Harbor. Hazardous Waste 1, 73–81. Ropes, J. W., and Stickney, A. P. 1965. Reproductive cycle of Mya arenaria in New England. Biol. Bull. 128, 315–327. Schroeder, P. C. 1987. Endogenous control of gametogenesis. In “Reproduction of Marine Invertebrates” (A. C. Giese, J. S. Pearse, and V. B. Pearse, Eds.), Vol. IX, pp. 179–249. Blackwell Scientific, Palo Alto, CA. Sunila, I. 1991. Respiration of sarcoma cells from the soft-shell clam Mya arenaria L. under various conditions. J. Exp. Mar. Biol. Ecol. 150, 19–29. Sunila, I. 1992. Serum–cell interactions in transmission of sarcoma in the soft shell clam, Mya arenaria L. Comp. Biochem. Physiol. 102A, 727–730. Wilkinson, L., Hill, M., Welna, J. P., and Birkenbeuel, G. K. 1992. “SYSTAT for Windows: Statistics,” Version 5 ed. SYSTAT, Inc., Evanston, IL. Yevich, P. P., and Barszcz, C. A. 1976. Gonadal and hematopoietic neoplasms in Mya arenaria. Mar. Fish. Rev. 38, 42–43. Yevich, P. P., and Barszcz, C. A. 1977. Neoplasia in soft-shell clams (Mya arenaria) collected from oil-impacted sites. Ann. NY Acad. Sci. 298, 409–426. Zar, J. H. 1984. “Biostatistical Analysis.” Prentice–Hall, Englewood Cliffs, NJ.
67, 161–168 (1996)
0024
Effects of Gonadal Neoplasms on Oogenesis in Softshell Clams, Mya arenaria BRUCE J. BARBER Department of Animal, Veterinary and Aquatic Sciences, University of Maine, Orono, Maine 04469 Received May 25, 1995; accepted November 27, 1995
The average prevalence of gonadal neoplasms in softshell clams, Mya arenaria, from Whiting Bay, Washington County, Maine, was 19.4% in 1994. Monthly prevalences ranged from 10 to 26.7%. Neoplasms ranged in intensity from few, small foci of undifferentiated germ cells (Stage 1), to 50–100% of gonadal follicles being involved (Stage 2), to invasion and metastasis with loss of tissue architecture (Stage 3), indicating that the disease is progressive and lethal. There was no relationship (P > 0.05) between prevalence of neoplasms and clam size between 45.7 and 60.7 mm mean shell length. Clams of both sexes were affected, but females were significantly more likely (P ≤ 0.025) to have neoplasms than males. Female clams with neoplasms produced significantly fewer (P ≤ 0.001) gametes than healthy clams. Overall reduction of gamete number was 66%, resulting from direct displacement of gametes by tumor cells in affected gonadal follicles. Compared to healthy clams, clams with neoplasms exhibited a significantly lower (P ≤ 0.001) mean oocyte diameter before spawning and a significantly greater (P ≤ 0.001) mean oocyte diameter after spawning, as the result of a general (throughout the entire gonad) inhibition of normal oogenesis and spawning. It is concluded that gonadal neoplasms have a negative impact on the reproductive output of the Whiting Bay clam population. © 1996 Academic Press, Inc. KEY WORDS: Mya arenaria; gonadal neoplasm; prevalence; intensity; oogenesis effects. INTRODUCTION
Gonadal neoplasms (germinomas, tumors) have been reported in several species of marine bivalve molluscs, including the softshell clam, Mya arenaria (Peters et al., 1994). Germinomas, ranging from noninvasive to metastatic, in both male and female clams, are char1 Representative cases of M. arenaria with gonadal neoplasms have been assigned Accession Nos. RTLA 6071, 6072, and 6073 in the Registry of Tumors in Lower Animals, Department of Pathology, George Washington University Medical Center, 2300 I Street, NW, Washington, DC 20037.
acterized by the presence of monomorphic, basophilic, undifferentiated germ cells that proliferate rather than mature (Gardner et al., 1991). The nucleii of these cells are often eccentric and the nucleolus is frequently difficult to distinguish in the abundantly clumped chromatin (Brown et al., 1977). Disease progression has been described, based on histopathology, by Yevich and Barszcz (1977). In early cases, small foci of neoplastic cells occupy a small portion of one or more follicles. Development of oocytes or spermatocytes apparently continues, even as the neoplastic cells increase in number within follicles. In some cases, however, follicles are completely packed with neoplastic cells and as a result, sexual identity is lost. Metastasis, in which tumor cells have broken through basement membrane and become established at a remote site, is commonly noted. Cell types at the metastatic sites are the same as those seen in the follicles, except that mitotic figures are more prevalent. These neoplasms have been termed “germinomas” and are similar to those reported in the quahog, Mercenaria mercenaria and other species (Barry and Yevich, 1972; Hesselman et al., 1988; Peters et al., 1994). Based on the above description, gonadal neoplasms are considered malignant. Even though M. arenaria is distributed from Labrador to South Carolina on the east coast of North America (Hidu and Newell, 1989), gonadal neoplasms have only been reported from locations in Maine. This condition was initially described by Barry and Yevich (1975) and Yevich and Barszcz (1976, 1977) in clams from Long Cove, Searsport, where 1 to 22% of clams examined between 1971 and 1975 had tumors. Similar neoplasms were present in 3.6% of clams examined from Long Cove in 1976 (Brown et al., 1977). Germinomas were also present in 3% of clams taken from Roque Bluffs (near Machiasport) and in 35% of clams collected from Dennys Bay (near Dennysville) collected in March 1980 (Gardner et al., 1991). In these cases, occurrence of neoplasms was correlated with environmental contamination from either hydrocarbons (Barry and Yevich, 1975; Yevich and Barszcz, 1976, 1977) or herbicides (Gardner et al., 1991). Harshbarger et al. (1979)
161 0022-2011/96 $18.00 Copyright © 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.
JOBNAME: 67#2 96 PAGE: 2 SESS: 2 OUTPUT: Mon Jun 10 12:40:57 1996 /xypage/worksmart/tsp000/67596b/11 162
BRUCE J. BARBER
reported the presence of intranuclear inclusions in tumor cells from clams examined by Yevich and Barszcz (1976, 1977), but a viral nature was never confirmed. Thus the etiology of this disease is presently unknown. Since 1982, the production of clams from flats in Washington County, Maine, has declined from 240,000 bushels to 24,000 bushels (Maine Department of Marine Resources). Although there are several potential reasons for this decline, the impact of disease on reproductive development has not been previously examined. A survey of several clam flats in Washington County, Maine, conducted in September 1993 revealed that a population in Whiting Bay had a 43% prevalence of gonadal neoplasms (Barber, 1995). In this study, the prevalence and intensity of gonadal neoplasms in this population was determined throughout 1994 and the impact of the disease on oogenesis was quantified.
parametric mean square successive difference test (Zar, 1984, p. 419). The effect of gonadal neoplasms on oogenesis was assessed using an image analysis system consisting of a video camera mounted on a Nikon Labophot microscope and connected to a Truevision Targa+ videographics adapter. Images of five gonadal fields from each female clam (600×) were enhanced, and data were collected using Image Pro Plus software. In each of the five fields, the number of oogonia, oocytes, and ova were counted, and the diameter of those having a visible nucleus was measured. This method has been found to accurately reflect oogenesis in marine bivalves (Barber and Blake, 1991; Barber et al., 1988). The mean number of gametes per field and the mean diameter of those gametes for clams with and without neoplasms were compared for each sampling date using two-way ANOVA (Wilkinson et al., 1992).
MATERIALS AND METHODS
Adult clams (n 4 11–30) were collected using a hand rake from an intertidal flat (Whiting Bay, latitude 44° 499 180 N; longitude 67° 099 460 W) located near Whiting, in Washington County, Maine. Samples were collected in all months in 1994 except February, when the flats were covered with ice. Water temperature (mercury thermometer, °C) and salinity (refractometer, ppt) were recorded at the time of collection (low tide) beginning May 1994. In the laboratory, clams were measured (shell length, mm), removed from their shells, and placed in Helly’s fixative (Barszcz and Yevich, 1975). Transverse sections (3–4 mm) through kidney, digestive gland, gonad, gills, and foot were removed, placed in cassettes, and dehydrated and cleared in a Fisher Histomatic Tissue Processor (Model 166). Tissues were embedded in Paraplast, sectioned (6 mm), and stained with Shandon instant hematoxylin and eosin Y before being coverslipped. Finished slides were examined with a Nikon Labophot microscope (100×) for the presence of gonadal neoplasms. In clams having neoplasms, the degree of disease development was assigned using criteria similar to those defined by Peters et al. (1994). Stage 1 neoplasms consisted of undifferentiated germs cells partially or totally filling one or more (but less than half of the total) follicles; loss of gonadal architecture was rare. Stage 2 neoplasms were characterized by the presence of undifferentiated germ cells in over 50% of visible follicles; usually there was some loss of tissue architecture, but invasion and metastasis were not evident. In Stage 3 neoplasms, a few to all follicles were involved, loss of tissue architecture was common, and invasion or metastasis was evident. The prevalence (percentage of clams having neoplasms) and the mean intensity (Stage) of neoplasms for each monthly sample were examined for serial randomness using the non-
RESULTS
Mean shell length of clams examined ranged from 45.7 to 60.7 mm (Table 1). Although these means were significantly different (ANOVA, P # 0.001), there was no relationship (correlation, P 4 0.729) between clam size and the prevalence of gonadal neoplasms over this range. Water temperature at the collection site increased from 12.1°C in May to a maximum of 19.1°C in July and then declined to 0.6°C in December (Table 1). Salinity ranged from 28 to 34 ppt between May and November, but was 20 ppt in December (Table 1). Thirty clams were collected each month in 1994 except February, May, and July (Table 2). In all, 310 clams from Whiting Bay were examined histologically over the course of the study. Of these, 60 had gonadal neoplasms, for an overall prevalence of 19.4%.1 The 60 neoplasms included 22 Stage 1 cases, 33 Stage 2 cases, and 5 Stage 3 cases. At the gross level, it was possible TABLE 1 Dates That Softshell Clams, M. arenaria, Were Collected from Whiting Bay, Mean and SE of Shell Length (mm), Water Temperature (°C), and Salinity (ppt) Shell length
Collection date
Mean
SE
Temperature (°C)
Salinity (ppt)
1/11/94 3/17/94 4/19/94 5/20/94 6/15/94 7/18/94 8/14/94 9/11/94 10/17/94 11/13/94 12/13/94
49.8 45.7 48.7 52.2 54.7 57.1 57.2 60.7 52.9 52.8 52.7
1.0 1.0 1.0 1.0 1.0 1.7 1.0 1.0 1.0 1.0 1.0
— — — 12.1 14.5 19.1 16.5 14.3 11.1 8.8 0.6
— — — 28 29 32 32 33 34 31 20
JOBNAME: 67#2 96 PAGE: 3 SESS: 3 OUTPUT: Mon Jun 10 12:40:57 1996 /xypage/worksmart/tsp000/67596b/11 EFFECTS OF GONADAL NEOPLASMS ON OOGENESIS IN Mya
163
to distinguish clams with advanced neoplasms (Stages 2 or 3) from clams without neoplasms by the shrunken, darkened, mottled appearance of their gonads. Gonadal neoplasms were found in all months sampled and in both male and female clams (Table 2). In May, June, and July, all neoplasms occurred in female clams. The sex ratio of all clams without neoplasms was not significantly different from 1:1, but the sex ratio of all clams with neoplasms was significantly different from 1:1 (x2, P ø 0.025), with predominantly more female clams having neoplasms (Table 2). Overall, monthly prevalences of neoplasms ranged from 10% in June and August to 26.7% in both September and October (Fig. 1). Mean monthly intensity was greatest (2.2) in April and May and lowest (1.0) in June (Fig. 1). Seasonal variability of both prevalence and mean intensity, however, was random (not significant, P > 0.25). Gonadal neoplasms had a negative impact on the
number of eggs produced per individual (Fig. 2). In both healthy (without neoplasms) and diseased (with neoplasms) clams, mean egg number per field (for an individual) was greater in the months (March–June) preceeding spawning; this was especially apparent in healthy clams. The mean number of female gametes (oogonia, oocytes, ova) per field was lower in diseased clams than in healthy clams in all months. Two-way ANOVA revealed that there were significant (P # 0.001) differences in the number of female gametes both between samples (dates) and between healthy and diseased clams; there was also a significant (P # 0.001) interaction between date and disease status (Table 3). The overall effect of gonadal neoplasia was a 66% reduction in the number of female gametes produced per individual. Gonadal neoplasms also affected the size of eggs produced, but the effect varied over the course of the oogenic cycle (Fig. 3). Mean egg diameter in healthy clams was greatest in the months of June and July. A sharp decline in mean oocyte diameter between July and August indicated that most healthy clams spawned during this period. In clams with neoplasms, mean egg diameter was greatest in August, after healthy clams had spawned. Two-way ANOVA revealed that mean egg size varied significantly (P # 0.001) with date but not with disease status over the entire study; there was also a significant (P # 0.001) date–disease interaction (Table 4). On closer examination of Fig. 3, it can be seen that in the months preceeding spawning (January–July), mean egg diameter was lower in clams with neoplasms than in healthy clams; the magnitude of this difference generally increased in relation to oogenic activity. The opposite was true for the months immediately after spawning (August–November), when clams with neoplasms had greater mean egg diameters than healthy clams; this difference decreased with time after spawning. When separate two-way ANOVA were run on these two sub-
FIG. 1. Monthly percentage prevalence (Y1; bars) and mean intensity (Y2; line) of gonadal neoplasms in clams, M. arenaria, from Whiting Bay.
FIG. 2. Mean (+SE) number of female gametes per microscopic field from clams, M. arenaria, with and without gonadal neoplasms.
TABLE 2 Numbers of Clams Examined Histologically at Each Sampling Date and the Distribution of Neoplasms by Stage and Gender Stage
Gender
Collection date
No. examined
1
2
3
M
F
U
1/11/94 3/17/94 4/19/94 5/20/94 6/15/94 7/18/94 8/14/94 9/11/94 10/17/94 11/13/94 12/13/94 Totals
30 30 30 29 30 11 30 30 30 30 30 310
4 2 1 1 3 1 1 3 4 2 0 22
2 4 4 4 0 1 2 5 3 2 6 33
0 0 2 2 0 0 0 0 1 0 0 5
1 2 3 0 0 0 1 4 3 2 3 19
4 3 3 6 3 2 2 4 5 2 2 36
1 1 1 1 0 0 0 0 0 0 1 5
Note. M, male; F, female; U, undetermined.
JOBNAME: 67#2 96 PAGE: 4 SESS: 2 OUTPUT: Mon Jun 10 12:40:57 1996 /xypage/worksmart/tsp000/67596b/11 164
BRUCE J. BARBER
TABLE 3 Two-Way Analysis of Variance for Effects of Collection Date and Disease on Mean Egg Number per Microscopic Field of Individual Female Clams, M. Arenaria Source of variation
Sum of squares
Degrees of freedom
Mean square
F value
Probability
Date Disease Date × Disease Error
5921.54 3327.67 1670.54 4685.57
10 1 10 143
592.15 3327.67 167.05 32.77
18.07 101.56 5.10
0.000 0.000 0.000
sets of data, there were highly significant (P # 0.001) differences by date and by disease status (P # 0.004), and for the postspawning months, there was a significant date–disease interaction (P # 0.002). Thus in the months preceeding spawning, clams with neoplasms were producing fewer and smaller gametes than clams without neoplasms (Fig. 4). After spawning, clams with neoplasms contained fewer, larger gametes than healthy clams (Fig. 5). DISCUSSION
There are two major types of neoplasms that occur in marine bivalves, including M. arenaria (Yevich and Barszcz, 1976, 1977; Brown et al., 1977; Peters, 1988; Peters et al., 1994). The first type, hematopoietic neoplasms, or disseminated sarcomas, are characterized by unusual cells in the connective tissue, blood vessels, and sinuses of the visceral mass, muscle, and mantle tissue (Peters, 1988). The cells are hypertrophied, usually enlarged two to four times the diameter of normal hemocytes, and have a hyperchromatic nucleus that is often pleomorphic. Proportions of abnormal cells in affected clams generally increase over time and result in death (Cooper et al., 1982; Barber, 1990; Brousseau and Baglivo, 1991). A viral etiology has been suggested for this disease, but other factors, such as chemical pollution and genetics may be involved (Oprandy et al., 1981; Oprandy and Chang, 1983; Peters, 1988). Al-
FIG. 3. Mean (+SE) diameter of female gametes from clams, M. arenaria, with and without gonadal neoplasms.
TABLE 4 Two-Way Analysis of Variance for Effects of Collection Date and Disease on Mean Egg Diameter of Individual Female Clams, M. arenaria Source of variation
Sum of squares
Degrees of freedom
Mean square
F value
Probability
Date Disease Date × Disease Error
3022.71 9.14 1395.58 3555.64
10 1 10 141
302.27 9.14 139.56 25.22
11.99 0.36 5.53
0.000 0.548 0.000
though hematopoietic neoplasms are epizootic throughout the range of M. arenaria (Brown et al., 1977; Farley et al., 1986; Brousseau, 1987a; Reinisch et al., 1984; Morrison et al., 1993), none were seen in the clams from Whiting Bay examined in the present study. The second major type of proliferative disorder reported to occur in marine bivalves, including M. arenaria, is the gonadal neoplasm or germinoma (Hesselman et al., 1988; Peters et al., 1994), the focus of this study. Based on histological examination, the proliferation of undifferentiated germ cells first in the gonad and then into surrounding tissues suggests a progressive and lethal nature, although this has not been demonstrated (Barry and Yevich, 1975; Yevich and Barszcz, 1977; Peters et al., 1994). Initial reports of this disease were correlated with hydrocarbon pollution (Barry and Yevich, 1975; Yevich and Barszcz, 1976, 1977; Brown et al., 1977), but subsequent attempts to induce the disease by exposure to hydrocarbons were unsuccessful (Gardner et al., 1991). More recently, higher prevalences of gonadal neoplasms in Washington County, Maine, have been correlated with a history of herbicide application on adjacent land areas, suggesting that there could be a chemical etiology for this disease (Gardner et al., 1991). The fact, however, that gonadal neoplasms have only been found in Maine despite the occurrence of M. arenaria in more chemically contaminated areas along the east coast than Maine (Reinisch et al., 1984), suggests instead that the etiological agent may be of a localized, genetic nature. Gonadal neoplasms were epizootic at the Whiting Bay site, with an overall prevalence of 19.4%. Although neoplasms were identified at the cellular level by examination of histological preparations, it was possible to identify clams with advanced neoplasms by the gross appearance of their gonad. In a separate sample of 100 clams collected from Whiting Bay in November 1994, 12 clams that had shrunken, darkened, mottled gonads were all found to have advanced (Stage 2 or 3) neoplasms upon histological examination; among 20 clams that appeared to be healthy at the gross level, 2 were found to have very early (Stage 1) neoplasms when examined histologically. Thus, gross examination of
JOBNAME: 67#2 96 PAGE: 5 SESS: 2 OUTPUT: Mon Jun 10 12:40:57 1996 /xypage/worksmart/tsp000/67596b/11 EFFECTS OF GONADAL NEOPLASMS ON OOGENESIS IN Mya
165
FIG. 4. Gonad of female clams, M. arenaria, collected from Whiting Bay in May 1994. (A) Gonad without neoplasm (175×) and numerous developing oocytes (OC). (B) Gonad (175×) with neoplasm (N); note that oocyte (OC) growth is inhibited throughout the gonad. Bar, 100 mm.
clams might be useful for determining the approximate prevalence of gonadal tumors at a site without the time and expense associated with histology. Based on histological observations and the distribution of cases among all three intensity stages, this study also suggests that this disease is progressive and possibly lethal. Early cases are confined to one or more gonadal follicles with little apparent effect on surrounding follicles, tissues, or their physiological func-
tion (Stage 1). As the cells proliferate and spread, more follicles become involved, and gonadal architecture and function are affected (Stage 2). Once the tumor cells break through the follicle wall and metastasize, adjacent and distant (gill) tissues become involved (Stage 3). There were considerably fewer Stage 3 cases than Stage 2 cases, suggesting that few clams survived once the disease reached an advanced stage. The rate at which this disease progresses and results in death is
JOBNAME: 67#2 96 PAGE: 6 SESS: 2 OUTPUT: Mon Jun 10 12:40:57 1996 /xypage/worksmart/tsp000/67596b/11 166
BRUCE J. BARBER
FIG. 5. Gonad of female clams, M. arenaria, collected from Whiting Bay in August 1994. (A) Gonad without neoplasm (175×) after spawning; visible cells are primary oocytes (OC) or spherical products of cytolysis (C). (B) Gonad (175×) with neoplasm (N); note that follicles without tumor cells contain unspawned ova (OV). Bar, 100 mm.
unknown. For M. arenaria with early cases of hematopoietic neoplasia, about 60 days are required for disease progression and death to occur (Barber, 1990). At the Whiting Bay site, it was not unusual to find dead clams with intact shells in the sediment and no visible sign of predation. In a few instances, decomposed clam tissue was still associated with the shells. Future studies will attempt to more clearly determine the mortality rate associated with this disease.
All the clams examined in this study were adults (reproductively mature), and there was no evident relationship between prevalence of gonadal neoplasia and shell height over the size range of 45.7 to 60.7 mm. Clams of this size from Whiting Bay are estimated to be 7–15 years old (B. Beal, pers. commun.). The minimum age/size at which clams develop gonadal neoplasms is unknown, but it would presumably be after attaining sexual maturity, which occurs at age >1 year
JOBNAME: 67#2 96 PAGE: 7 SESS: 2 OUTPUT: Mon Jun 10 12:40:57 1996 /xypage/worksmart/tsp000/67596b/11 EFFECTS OF GONADAL NEOPLASMS ON OOGENESIS IN Mya
and shell height >20 mm (Brousseau, 1987b). One sample of 20 clams with a mean size of 25.4 mm was obtained in November from a site near Whiting Bay where neoplasms occur. These smaller clams had developed gonads that appeared normal for the time of year (spawned out); no gonadal neoplasms were found. Thus the minimum size at which M. arenaria develops gonadal neoplasia is probably between 25.4 and 45.7 mm shell height (age 2–7 years). This may represent the time required for the etiological agent to manifest itself. Previous reports of gonadal neoplasms in M. arenaria indicated that both male and female clams were affected (Yevich and Barszcz, 1976; Peters, 1994), but the finding that gonadal neoplasms occur almost twice as often in female clams (60%) than male clams (32%) has not been previously been noted for this species. In contrast, healthy clams from Whiting Bay (this study) as well as those from Long Island Sound (Brousseau, 1987b) had a 1:1 ratio of males to females. A similar predominance of neoplasms in females over neoplasms in males, however, has been reported for the quahog, M. mercenaria. Barry and Yevich (1972) examined 539 individuals and found 12 neoplasms in females and only 2 neoplasms in males. Hesselman et al. (1988) examined 1263 individuals and of the 147 with gonadal neoplasms, 61% were females, 33% were males, and 6% were not recognizable as either male or female. The fact that gonadal neoplasms occur primarily in females of both clam species suggests that a common disease mechanism might be involved. Healthy clams at Whiting Bay underwent a normal gametogenic cycle. Mean number of eggs per microscope field increased from January to May and then declined as oocytes continued to grow. Most oocyte growth occurred in May and June, in conjunction with increasing water temperature. Spawning began as early as June, based on histological observation. The sharp decrease in mean oocyte diameter seen in August, however, indicated that most spawning activity occurred at this time, in conjunction with decreasing water temperature. M. arenaria at Whiting Bay exhibited one spawning period, which is in agreement with previous studies of clam populations north of Cape Cod (Ropes and Stickney, 1965). Gonadal neoplasms had a direct negative effect on the number of eggs produced. The more advanced the disease, the greater the proportion of gonadal follicles containing tumor cells and the lower the proportion of normal follicles containing oocytes. The magnitude of reduction in mean egg number per field between normal and diseased clams at any date was thus related to the intensity of the disease. The interaction that occurred between month and disease status most likely was the result of the increase in disease intensity that occurred in April and May (and hence decrease in egg number) at the same time that healthy clams were
167
experiencing an increase in egg number per field. Overall, female clams with neoplasms produced 66% fewer gametes than clams without neoplasms. The effect of gonadal neoplasms on the size of eggs produced appeared to be indirect. That is, the effect on oocytes was seen throughout the gonads of affected clams, not merely in follicles containing tumor cells. The net effect was an inhibition of the normal oogenic process. The impact of neoplasms on oocyte development was related to the timing of disease acquisition, which accounts for the interaction seen between date and disease status. For clams in which the disease appeared early in the oogenic process, an overall reduction in mean oocyte size resulted, as growth of oocytes was stunted (Fig. 4). If neoplasms are progressive and lethal, these clams would most likely not have lived to produce normal ova. If any ova were produced, they would be smaller and less likely to produce viable larvae than ova from healthy clams (Kraeuter et al., 1982). For clams in which the disease appeared late in the developmental phase, the effect was an increase in mean oocyte diameter, as the otherwise normal appearing ova were not spawned along with ova from healthy clams (Fig. 5). The mean diameter of diseased clams in August was even larger than that of healthy clams in July, indicating that while mature ova were released from healthy clams as early as June, they were retained in diseased clams. Whether or not diseased clams with oocytes ultimately spawned and what the fate of these gametes was are unknown. It is clear that diseased clams, even if they spawned would not produce as many eggs, and the retarded timing of release would reduce the likelihood of fertilization by healthy sperm and larval survival as water temperature decreased. This interaction between disease and oogenesis is complicated by differing rates of disease initiation and progression throughout the year. The observation of a general inhibition of oogenesis in clams with neoplasms suggests that the disease affects the reproductive process at some basic level, perhaps by disrupting or blocking normal neurohormonal cycles, which are involved in the endogenous regulation of gametogenesis in marine invertebrates (Schroeder, 1987). Disseminated sarcoma cells in M. arenaria produce proteins that, besides promoting tumor growth, could be cytotoxic (Sunila, 1992), thus possibly inhibiting oocyte development. Additionally, the high respiration rates of mitotically active tumor cells could possibly divert energy away from oogenesis (Sunila, 1991). Estimation of the impact of gonadal neoplasms on the reproductive capacity of the Whiting Bay clam population depends on several factors. This study has established the high prevalence of this disease at this location and the serious impact that it has on oogenesis. The disproportionate number of female clams having gonadal neoplasms has further implications, especially if egg production is a factor limiting overall
JOBNAME: 67#2 96 PAGE: 8 SESS: 2 OUTPUT: Mon Jun 10 12:40:57 1996 /xypage/worksmart/tsp000/67596b/11 168
BRUCE J. BARBER
reproductive success. The rates of disease progression and mortality remain to be explored. Disease is one potential factor limiting clam production in Whiting Bay. ACKNOWLEDGMENTS I thank the following people for their help: Brian Beal, Ken Vencile, Robert Bayer, Bruce Wiersma, Neil Greenberg, and Brian Barber for field assistance and Dawna Beane for the histological preparations. Many thanks to Paul Yevich and Esther Peters for verifying neoplasms. Stephen Fegley and Esther Peters provided thoughtful comments on an earlier version of the manuscript. Finally, I would like to thank Bob and Jane Bell for access to the study area. This project was supported in part by the Maine Aquaculture Innovation Center. This is Maine Agricultural and Forestry Experiment Station external publication 1955. REFERENCES Barber, B. J. 1990. Seasonal prevalence and intensity and disease progression of neoplasia in soft shell clams, Mya arenaria, from the Shrewsbury River, New Jersey. In “Pathology in Marine Science” (F. O. Perkins and T. Cheng, Eds.), pp. 377–386. Academic Press, New York. Barber, B. J. 1995. Seasonal prevalence of gonadal neoplasms in clams, Mya arenaria, in Maine and impact on oogenesis. J. Shellfish Res. 14, 238. Barber, B. J., and Blake, N. J. 1991. Reproductive physiology. In “Scallops: Biology, Ecology and Aquaculture” (S. E. Shumway, Ed.), pp. 377–428. Elsevier, New York. Barber, B. J., Getchell, R., Shumway, S., and Schick, D. 1988. Reduced fecundity in a deep-water population of the giant scallop Placopecten magellanicus in the Gulf of Maine, USA. Mar. Ecol. Prog. Ser. 42, 207–212. Barry, M. M., and Yevich, P. P. 1972. Incidence of gonadal cancer in the quahog Mercenaria mercenaria. Oncology 26, 87–96. Barry, M., and Yevich, P. P. 1975. The ecological, chemical and histopathological evaluation of an oil spill site. Part III. Histopathological studies. Mar. Poll. Bull. 6, 171–173. Barszcz, C. A., and Yevich, P. P. 1975. The use of Helly’s fixative for marine invertebrate histopathology. Comp. Pathol. Bull. 7, 4. Brousseau, D. J. 1987a. Seasonal aspects of sarcomatous neoplasia in Mya arenaria (soft-shell clam) from Long Island Sound. J. Invertebr. Pathol. 50, 269–276. Brousseau, D. J. 1987b. A comparative study of the reproductive cycle of the soft-shell clam, Mya arenaria in Long Island Sound. J. Shellfish Res. 6, 7–15. Brousseau, D. J., and Baglivo, J. A. 1991. Field and laboratory comparisons of mortality in normal and neoplastic Mya arenaria. J. Invertebr. Pathol. 57, 59–65. Brown, R. S., Wolke, R. E., Saila, S. B., and Brown, C. W. 1977. Prevalence of neoplasia in 10 New England populations of the soft-shell clam (Mya arenaria). Ann. NY Acad. Sci. 298, 522–534. Cooper, K. R., Brown, R. S., and Chang, P. W. 1982. The course and mortality of a hematopoietic neoplasm in the soft-shell clam, Mya arenaria. J. Invertebr. Pathol. 39, 149–157. Farley, C. A., Otto, S. V., and Reinisch, C. L. 1986. New occurrence of epizootic sarcoma in Chesapeake Bay soft shell clams, Mya arenaria. Fish. Bull. 84, 851–857. Gardner, G. R., Yevich, P. P., Hurst, J., Thayer, P., Benyi, S., Harshbarger, J. C., and Pruell, R. J. 1991. Germinomas and teratoid
siphon anomalies in softshell clams, Mya arenaria, environmentally exposed to herbicides. Environmental Health Perspectives 90, 43–51. Harshbarger, J. C., Otto, S. V., and Chang, S. C. 1979. Proliferative disorders in Crassostrea virginica and Mya arenaria from the Chesapeake Bay and intranuclear virus-like inclusions in Mya arenaria with germinomas from a Maine oil spill site. Haliotis 8, 243–248. Hesselman, D. M., Blake, N. J., and Peters, E. C. 1988. Gonadal neoplasms in hard shell clams Mercenaria spp., from the Indian River, Florida: Occurrence, prevalence, and histopathology. J. Invertebr. Pathol. 52, 436–446. Hidu, H., and Newell, C. R. 1989. Culture and ecology of the softshelled clam, Mya arenaria. In “Clam Mariculture in North America” (J. J. Manzi and M. Castagna, Eds.), pp. 277–292. Elsevier Science, Amsterdam. Kraeuter, J. N., Castagna, M., and van Dessel, R. 1982. Egg size and larval survival of Mercenaria mercenaria (L.) and Argopecten irradians (Lamarck). J. Exp. Mar. Biol. Ecol. 56, 3–8. Morrison, C. M., Moore, A. R., Marryatt, V. M., and Scarratt, D. J. 1993. Disseminated sarcomas of soft shell clams, Mya arenaria Linnaeus 1758, from sites in Nova Scotia and New Brunswick. J. Shellfish Res. 12, 65–69. Oprandy, J. J., and Chang, P. W. 1983. 5-bromodeoxyuridine induction of hemopoietic neoplasia and retrovirus activation in the softshell clam, Mya arenaria. J. Invertebr. Pathol. 42, 196–206. Oprandy, J. J., Chang, P. W., Pronovost, A. D., Cooper, K. R., Brown, R. S., and Yates, V. J. 1981. Isolation of a viral agent causing hematopoietic neoplasia in the soft-shell clam, Mya arenaria. J. Invertebr. Pathol. 38, 45–51. Peters, E. C., Yevich, P. P., Harshbarger, J. C., and Zaroogian, G. E. 1994. Comparative histopathology of gonadal neoplasms in marine bivalve molluscs. Dis. Aquat. Org. 20, 59–76. Peters, E. C. 1988. Recent investigations on the disseminated sarcomas of marine bivalve molluscs. Am. Fish. Soc. Spec. Publ. 18, 74–92. Reinisch, C. L., Charles, A. M., and Stone, A. M. 1984. Epizootic neoplasia in soft shell clams collected from New Bedford Harbor. Hazardous Waste 1, 73–81. Ropes, J. W., and Stickney, A. P. 1965. Reproductive cycle of Mya arenaria in New England. Biol. Bull. 128, 315–327. Schroeder, P. C. 1987. Endogenous control of gametogenesis. In “Reproduction of Marine Invertebrates” (A. C. Giese, J. S. Pearse, and V. B. Pearse, Eds.), Vol. IX, pp. 179–249. Blackwell Scientific, Palo Alto, CA. Sunila, I. 1991. Respiration of sarcoma cells from the soft-shell clam Mya arenaria L. under various conditions. J. Exp. Mar. Biol. Ecol. 150, 19–29. Sunila, I. 1992. Serum–cell interactions in transmission of sarcoma in the soft shell clam, Mya arenaria L. Comp. Biochem. Physiol. 102A, 727–730. Wilkinson, L., Hill, M., Welna, J. P., and Birkenbeuel, G. K. 1992. “SYSTAT for Windows: Statistics,” Version 5 ed. SYSTAT, Inc., Evanston, IL. Yevich, P. P., and Barszcz, C. A. 1976. Gonadal and hematopoietic neoplasms in Mya arenaria. Mar. Fish. Rev. 38, 42–43. Yevich, P. P., and Barszcz, C. A. 1977. Neoplasia in soft-shell clams (Mya arenaria) collected from oil-impacted sites. Ann. NY Acad. Sci. 298, 409–426. Zar, J. H. 1984. “Biostatistical Analysis.” Prentice–Hall, Englewood Cliffs, NJ.