- Email: [email protected]
Sublethal effects of chromium-VI in the Asian clam (Potamocorbula amurensis)
Marine Environmental Research 50 (2000) 295±300 www.elsevier.com/locate/marenvrev
Sublethal eects of chromium-VI in the Asian clam (Potamocorbula amurensis) S.J. Teh *, I. Werner, D.E. Hinton Department of Anatomy, Physiology, and Cell Biology, School of Veterinary Medicine, University of California-Davis, Davis, CA 95616, USA Received 29 April 1999; received in revised form 30 November 1999; accepted 16 February 2000
Abstract Previously, we have shown that Asian clams (Potamocorbula amurensis) with highest metallic body burdens have highest prevalence of disease and lowest reproduction. The present study was designed to assess and validate potential sublethal toxicity of hexavalent chromium (Cr-VI) in clams under controlled laboratory exposure. For 7 days, three replicates of clam (n=10 per replicate) were exposed to aqueous solution containing 0.00, 0.92, 8.40, or 25.6 mg lÿ1 of Cr-VI at 15 C and 15 g lÿ1 salinity. Mortality reached 100% in the 25.6 mg lÿ1 group within 7 days. There was no signi®cant dierence in mortality among the control, 0.92, and 8.40 mg lÿ1 groups. Western blot analyses revealed signi®cantly elevated stress protein hsp70 levels in the 8.40 mg lÿ1 treatment group. Histopathologic analyses revealed mild digestive gland (DG) atrophy in the control group. Clams exposed to 0.92 mg lÿ1 Cr-VI showed moderate DG atrophy, moderate granulomatous in¯ammation and necrosis in DG, ovary and testis. Lesions observed in the 8.40 mg lÿ1 treatment group included severe DG atrophy, severe granulomatous in¯ammation and necrosis in byssal gland, DG, gill, kidney, ovary and testis. In gills and testes of treated groups, apoptotic cells outnumbered mitotic cells. In addition, gills from clams in the 8.40 mg lÿ1 group showed enhanced hsp70 staining. Our studies support a cause±eect relationship between contaminants and reduced health in Asian clams and indicate the DGs, gills, and reproductive organs are principal targets of CrVI toxicity at sublethal concentrations. Results from this study suggest that Cr-VI may have played a role in the increased incidence of diseased clams seen in previous studies and these adverse eects may be working to decrease clam populations at sites with highest metallic contamination in the San Francisco Bay Estuary. # 2000 Elsevier Science Ltd. All rights reserved. Keywords: Clams; Histopathology; Metals; Sublethal eects; Toxicity
* Corresponding author. Tel.: +1-916-752-1174; fax: +1-916-752-7690. E-mail address: [email protected] (S.J. Teh). 0141-1136/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved. PII: S0141-1136(00)00086-6
296
S.J. Teh et al. / Marine Environmental Research 50 (2000) 295±300
1. Introduction The Asian clam (Potamocorbula amurensis) has spread widely in benthic communities of the San Francisco Bay Estuary and its ubiquitous occurrence makes it a potential sentinel species (Carlton, Thompson, Schemel & Nichols, 1990). For example, metal bioaccumulation, condition index, glycogen content, and population parameters of Asian clams were monitored in the northern San Francisco Bay area between 1990 and 1998 (Brown & Luoma, 1995, 1998; Thompson, Parchaso, Brown & Luoma, 1996). Results showed decreased glycogen content and lower condition index in clams from sites with signi®cantly higher levels of nickel, vanadium, cadmium, zinc, copper, and chromium compared to clams at sites with lowest metal body burden concentrations. Furthermore, clams diagnosed with the high prevalence of diseases and enzyme alterations also showed a lower condition index and glycogen content (Teh, Clark, Brown, Luoma & Hinton, 1999). Higher prevalences of digestive gland (DG), gill, kidney, ovarian and testicular alterations in clams from sites with highest metal body burden suggest these are the target organs for metal toxicity. The speci®c metal or combination of metals inducing such organ-speci®c toxic eects is unknown. This laboratory study investigated the sublethal eects of hexavalent chromium (Cr-VI) exposure. By using biochemical, histopathological, and immunohistochemical approaches, we extended and veri®ed interpretations of contaminant exposure and eects seen in previous and concurrent ®eld studies (Teh et al., 1999). 2. Materials and methods 2.1. Animal acquisition, acclimation and exposure Asian clams were collected from a relatively uncontaminated site west of Martinez Marina, California. Clams were not fed and were maintained in the dark for 3 days before being exposed for 7 days to either control, 0.92, 8.40, or 25.6 mg lÿ1 of Cr-VI solution at 15 C and 15 g lÿ1 salinity (n=10 clams per replicate for three replicates). Solutions were changed every 2 days, and pH, temperature, salinity and ammonium concentrations monitored. 2.2. Tissue preparation, biochemical and histological analysis Visceral masses from 30 clams were divided into three groups. The ®rst group (n=10) was ®xed in 10% buered formalin for histopathologic and immunohistochemical analysis. The second group (n=12) was frozen in liquid nitrogen for western blot analysis of hsp70 proteins. The third group (n=6±8) was frozen in liquid nitrogen and stored at ÿ80 C. The severity of the histopathologic alterations was semi-quantitatively ranked on a scale of 0±3 (Teh et al., 1999). Tissue sections for immunohistochemical analysis were immunolocalized separately with each primary antibody, mouse anti-PCNA
S.J. Teh et al. / Marine Environmental Research 50 (2000) 295±300
297
and mouse anti-metallothionein (Zymed, California, USA); mouse anti-hsp 70 (Anity Bioreagent MA3-001, Colorado, USA); and ApopTag in situ Apoptosis detection kit (Dako, California, USA). All Sections were then counterstained with hematoxylin and examined under a light microscope. Hsp70 proteins were analyzed using Western blotting techniques as described in Werner and Hinton (1999). 2.3. Statistical analysis Stress protein (hsp70) data was analyzed using Bartlett's test for homogeneity. If data had heterogeneous variance, it was transformed to relative ranks then analyzed by analysis of variance (ANOVA) and Tukey-Kramer multiple comparison test. 3. Results and discussion Mortality was seen in all Cr-VI-exposed clams and followed a concentration± response relationship: 100% (25.6 mg lÿ1), 7% (8.40 mg lÿ1), 3% (0.92 mg lÿ1). At the end of 7 days, no mortality was seen in the control group. Asian clams exposed to increasing concentrations of Cr-VI demonstrated a corresponding increase in cytochemical and cytological alterations. Histopathologic ®ndings revealed mild or moderate digestive gland atrophy in two of 10 control clams (20%). Clams (n=10; ®ve females and ®ve males) exposed to 0.92 mg lÿ1 Cr-VI had moderate tubular atrophy (90%) and granulomatous in¯ammation (20%) in DG, mild or moderate necrosis of primary oocytes (100%; Fig. 1A) and spermatogonia (100%), and moderate granulomatous in¯ammation in ovary (20%) and testis (20%). In the 8.40 mg lÿ1 Cr-VI exposure (n=10; four females and six males), severe DG and kidney atrophy were seen in 60% of the clams. Two females and three males observed had severe granulomatous in¯ammation and necrosis in the byssal gland, DG, gill, and kidney. All females had severe follicular necrosis (Fig. 1B). Five males had mild or moderate and one male had severe necrosis of the spermatogonia. Based on the histopathologic analysis, there is a high possibility that most of the clams at 8.40 mg lÿ1 exposure would succumb if allowed to grow in clean water for an additional 7 days. Extensive necrosis of primordial germ cells and immature oocytes, but not the mature oocytes, were seen in clams exposed to 0.92 mg lÿ1 CrVI. This suggests that initial recruitment of new ospring may not show eect, while later recruitment may be diminished (Fig. 1A). Even if clams survive the 8.40 mg lÿ1 level, histologic analysis suggests that they will be reproductively inactive since there were no or minimal germ cells that survived the chromium toxicity (Fig. 1B). Therefore, results of this study provide strong evidence that chromium, at sublethal concentrations, would cause reproductive failure and diminished fertility. Immunohistochemical analysis of apoptosis, hsp70, and PCNA showed dierences among the control and exposure groups. Enhanced cell turnover was detected by an increase in cells reacting positively with PCNA concurrent with enhanced cells reacting positively for apoptosis in gill and testis of both the 0.92 and 8.40 mg lÿ1
298
S.J. Teh et al. / Marine Environmental Research 50 (2000) 295±300
Fig. 1. (A) Ovary of Asian clam exposed to 0.92 mg lÿ1 of Cr-VI with moderate necrosis of the primordial germ cells and primary oocytes (arrows). The secondary and mature oocytes were not aected at this concentration. (B) Ovary of Asian clam exposed to 8.40 mg lÿ1 of Cr-VI showing severe alteration (necrosis) in the primordial germ cells, immature and mature oocytes. Arrows point to few primary oocytes surviving the exposure (H&E; bar=30 mm).
treatments. Enhanced hsp70 staining intensity was seen in gill of both the 0.92 and 8.40 mg lÿ1 exposure groups (Fig. 2). There were no changes in metallothionein staining intensity observed between the control and exposure groups. Biochemical analysis revealed signi®cantly increased levels of hsp70 proteins in the 8.40 mg lÿ1 group when compared to control and 0.92 mg lÿ1 groups (P<0.001).
S.J. Teh et al. / Marine Environmental Research 50 (2000) 295±300
299
Fig. 2. Gill of Asian clam exposed to 8.40 mg lÿ1 of Cr-VI shows enhanced positive hsp70 staining (arrow). Hsp70 with hematoxylin counterstain (bar=30 mm).
Hsp70 in the 8.40 mg lÿ1 group (mean relative band densityS.E.=495004329, n=12) was approximately 10-fold that of the 0.92 mg lÿ1 (relative density= 37871908, n=11) and control (relative density=48681594, n=12) groups. Due to severe pathological alterations seen in clams, immunohistochemical analysis was only compared in organs that were not severely aected after Cr-VI exposure. Induction of cell proliferation and cell death was suggested by the number of cells staining positively for PCNA and apoptosis. However, since the number of apoptotic cells was higher than the number of PCNA-positive cells in clams exposed at 0.92 and 8.40 mg lÿ1 Cr-VI, sublethal deleterious eects of chromium surpassed proliferative capacity and likely led to a net reduction in cell number. Hsp70 stress proteins were detected both biochemically and histochemically. Although, the immunostaining intensity was determined visually, enhanced staining was localized in gill suggesting that this organ was most responsive with respect to hsp70 production upon exposure to chromium. Future Western blot biochemical analysis of hsp70 could focus on gill and achieve some enhanced sensitivity and decrease of eort. In summary, Cr-VI was shown to be acutely toxic to Asian clams at 25.6 mg lÿ1 exposure. Histopathologic and immunohistochemistry results indicated sublethal deleterious eect at 0.92 and 8.40 mg lÿ1, whereas a signi®cant increase in hsp70 stress proteins was detected only at 8.40 mg lÿ1 Cr-VI exposure. The results support a cause±eect relationship between contaminants exposure and physiologic changes, and that the DG, gill and reproductive organs are the principal targets of Cr-VI toxicity at sublethal concentrations. Cr-VI may have also played a role in the increased incidence of disease seen in clams during concurrent previous ®eld studies (Teh et al., 1999). Finally, we suggest the adverse eects on reproduction seen in
300
S.J. Teh et al. / Marine Environmental Research 50 (2000) 295±300
these studies may have contributed to the decrease in clam populations at sites with highest metal contamination in the San Francisco Bay Estuary (Brown & Luoma, 1995, 1998; Thompson et al., 1996). Acknowledgements Funding for these investigations was provided, in part, by US Public Health Service Grants CA-45131 from the National Cancer Institute, ES-04699 from the Superfund Basic Science Research Project, the US Environmental Protection Agency (USEPA) Grant Nos. R823297 and R825298, and the USEPA Center for Ecological Health Research at UC Davis Grants CR 819658. Although the information in this document has been partially funded by the USEPA, it may not necessarily re¯ect the views of the Agency and no ocial endorsement should be inferred. References Brown, C. L., & Luoma, S. N. (1995). Marine Ecology Progress Series, 124, 129±142. Brown, C. L., & Luoma, S. N. (1998). Interagency Ecological Program Newsletter, 11(N2), 33±35. Carlton, J. T., Thompson, T. K., Schemel, L. E., & Nichols, F. H. (1990). Marine Ecology Progress Series, V66(N1±2), 81±94. Teh, S. J., Clark, S. L., Brown, C. L., Luoma, S. N., & Hinton, D. E. (1999). Biomarkers, V4(N6), 497± 509. Thompson, J. K., Parchaso, F., Brown, C. L., & Luoma, S. N. (1996). U.S. Geological Survey Open File Report 96-437. Werner, I., & Hinton, D. E. (1999). Biomarkers, V4(N6), 473±484.
Sublethal eects of chromium-VI in the Asian clam (Potamocorbula amurensis) S.J. Teh *, I. Werner, D.E. Hinton Department of Anatomy, Physiology, and Cell Biology, School of Veterinary Medicine, University of California-Davis, Davis, CA 95616, USA Received 29 April 1999; received in revised form 30 November 1999; accepted 16 February 2000
Abstract Previously, we have shown that Asian clams (Potamocorbula amurensis) with highest metallic body burdens have highest prevalence of disease and lowest reproduction. The present study was designed to assess and validate potential sublethal toxicity of hexavalent chromium (Cr-VI) in clams under controlled laboratory exposure. For 7 days, three replicates of clam (n=10 per replicate) were exposed to aqueous solution containing 0.00, 0.92, 8.40, or 25.6 mg lÿ1 of Cr-VI at 15 C and 15 g lÿ1 salinity. Mortality reached 100% in the 25.6 mg lÿ1 group within 7 days. There was no signi®cant dierence in mortality among the control, 0.92, and 8.40 mg lÿ1 groups. Western blot analyses revealed signi®cantly elevated stress protein hsp70 levels in the 8.40 mg lÿ1 treatment group. Histopathologic analyses revealed mild digestive gland (DG) atrophy in the control group. Clams exposed to 0.92 mg lÿ1 Cr-VI showed moderate DG atrophy, moderate granulomatous in¯ammation and necrosis in DG, ovary and testis. Lesions observed in the 8.40 mg lÿ1 treatment group included severe DG atrophy, severe granulomatous in¯ammation and necrosis in byssal gland, DG, gill, kidney, ovary and testis. In gills and testes of treated groups, apoptotic cells outnumbered mitotic cells. In addition, gills from clams in the 8.40 mg lÿ1 group showed enhanced hsp70 staining. Our studies support a cause±eect relationship between contaminants and reduced health in Asian clams and indicate the DGs, gills, and reproductive organs are principal targets of CrVI toxicity at sublethal concentrations. Results from this study suggest that Cr-VI may have played a role in the increased incidence of diseased clams seen in previous studies and these adverse eects may be working to decrease clam populations at sites with highest metallic contamination in the San Francisco Bay Estuary. # 2000 Elsevier Science Ltd. All rights reserved. Keywords: Clams; Histopathology; Metals; Sublethal eects; Toxicity
* Corresponding author. Tel.: +1-916-752-1174; fax: +1-916-752-7690. E-mail address: [email protected] (S.J. Teh). 0141-1136/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved. PII: S0141-1136(00)00086-6
296
S.J. Teh et al. / Marine Environmental Research 50 (2000) 295±300
1. Introduction The Asian clam (Potamocorbula amurensis) has spread widely in benthic communities of the San Francisco Bay Estuary and its ubiquitous occurrence makes it a potential sentinel species (Carlton, Thompson, Schemel & Nichols, 1990). For example, metal bioaccumulation, condition index, glycogen content, and population parameters of Asian clams were monitored in the northern San Francisco Bay area between 1990 and 1998 (Brown & Luoma, 1995, 1998; Thompson, Parchaso, Brown & Luoma, 1996). Results showed decreased glycogen content and lower condition index in clams from sites with signi®cantly higher levels of nickel, vanadium, cadmium, zinc, copper, and chromium compared to clams at sites with lowest metal body burden concentrations. Furthermore, clams diagnosed with the high prevalence of diseases and enzyme alterations also showed a lower condition index and glycogen content (Teh, Clark, Brown, Luoma & Hinton, 1999). Higher prevalences of digestive gland (DG), gill, kidney, ovarian and testicular alterations in clams from sites with highest metal body burden suggest these are the target organs for metal toxicity. The speci®c metal or combination of metals inducing such organ-speci®c toxic eects is unknown. This laboratory study investigated the sublethal eects of hexavalent chromium (Cr-VI) exposure. By using biochemical, histopathological, and immunohistochemical approaches, we extended and veri®ed interpretations of contaminant exposure and eects seen in previous and concurrent ®eld studies (Teh et al., 1999). 2. Materials and methods 2.1. Animal acquisition, acclimation and exposure Asian clams were collected from a relatively uncontaminated site west of Martinez Marina, California. Clams were not fed and were maintained in the dark for 3 days before being exposed for 7 days to either control, 0.92, 8.40, or 25.6 mg lÿ1 of Cr-VI solution at 15 C and 15 g lÿ1 salinity (n=10 clams per replicate for three replicates). Solutions were changed every 2 days, and pH, temperature, salinity and ammonium concentrations monitored. 2.2. Tissue preparation, biochemical and histological analysis Visceral masses from 30 clams were divided into three groups. The ®rst group (n=10) was ®xed in 10% buered formalin for histopathologic and immunohistochemical analysis. The second group (n=12) was frozen in liquid nitrogen for western blot analysis of hsp70 proteins. The third group (n=6±8) was frozen in liquid nitrogen and stored at ÿ80 C. The severity of the histopathologic alterations was semi-quantitatively ranked on a scale of 0±3 (Teh et al., 1999). Tissue sections for immunohistochemical analysis were immunolocalized separately with each primary antibody, mouse anti-PCNA
S.J. Teh et al. / Marine Environmental Research 50 (2000) 295±300
297
and mouse anti-metallothionein (Zymed, California, USA); mouse anti-hsp 70 (Anity Bioreagent MA3-001, Colorado, USA); and ApopTag in situ Apoptosis detection kit (Dako, California, USA). All Sections were then counterstained with hematoxylin and examined under a light microscope. Hsp70 proteins were analyzed using Western blotting techniques as described in Werner and Hinton (1999). 2.3. Statistical analysis Stress protein (hsp70) data was analyzed using Bartlett's test for homogeneity. If data had heterogeneous variance, it was transformed to relative ranks then analyzed by analysis of variance (ANOVA) and Tukey-Kramer multiple comparison test. 3. Results and discussion Mortality was seen in all Cr-VI-exposed clams and followed a concentration± response relationship: 100% (25.6 mg lÿ1), 7% (8.40 mg lÿ1), 3% (0.92 mg lÿ1). At the end of 7 days, no mortality was seen in the control group. Asian clams exposed to increasing concentrations of Cr-VI demonstrated a corresponding increase in cytochemical and cytological alterations. Histopathologic ®ndings revealed mild or moderate digestive gland atrophy in two of 10 control clams (20%). Clams (n=10; ®ve females and ®ve males) exposed to 0.92 mg lÿ1 Cr-VI had moderate tubular atrophy (90%) and granulomatous in¯ammation (20%) in DG, mild or moderate necrosis of primary oocytes (100%; Fig. 1A) and spermatogonia (100%), and moderate granulomatous in¯ammation in ovary (20%) and testis (20%). In the 8.40 mg lÿ1 Cr-VI exposure (n=10; four females and six males), severe DG and kidney atrophy were seen in 60% of the clams. Two females and three males observed had severe granulomatous in¯ammation and necrosis in the byssal gland, DG, gill, and kidney. All females had severe follicular necrosis (Fig. 1B). Five males had mild or moderate and one male had severe necrosis of the spermatogonia. Based on the histopathologic analysis, there is a high possibility that most of the clams at 8.40 mg lÿ1 exposure would succumb if allowed to grow in clean water for an additional 7 days. Extensive necrosis of primordial germ cells and immature oocytes, but not the mature oocytes, were seen in clams exposed to 0.92 mg lÿ1 CrVI. This suggests that initial recruitment of new ospring may not show eect, while later recruitment may be diminished (Fig. 1A). Even if clams survive the 8.40 mg lÿ1 level, histologic analysis suggests that they will be reproductively inactive since there were no or minimal germ cells that survived the chromium toxicity (Fig. 1B). Therefore, results of this study provide strong evidence that chromium, at sublethal concentrations, would cause reproductive failure and diminished fertility. Immunohistochemical analysis of apoptosis, hsp70, and PCNA showed dierences among the control and exposure groups. Enhanced cell turnover was detected by an increase in cells reacting positively with PCNA concurrent with enhanced cells reacting positively for apoptosis in gill and testis of both the 0.92 and 8.40 mg lÿ1
298
S.J. Teh et al. / Marine Environmental Research 50 (2000) 295±300
Fig. 1. (A) Ovary of Asian clam exposed to 0.92 mg lÿ1 of Cr-VI with moderate necrosis of the primordial germ cells and primary oocytes (arrows). The secondary and mature oocytes were not aected at this concentration. (B) Ovary of Asian clam exposed to 8.40 mg lÿ1 of Cr-VI showing severe alteration (necrosis) in the primordial germ cells, immature and mature oocytes. Arrows point to few primary oocytes surviving the exposure (H&E; bar=30 mm).
treatments. Enhanced hsp70 staining intensity was seen in gill of both the 0.92 and 8.40 mg lÿ1 exposure groups (Fig. 2). There were no changes in metallothionein staining intensity observed between the control and exposure groups. Biochemical analysis revealed signi®cantly increased levels of hsp70 proteins in the 8.40 mg lÿ1 group when compared to control and 0.92 mg lÿ1 groups (P<0.001).
S.J. Teh et al. / Marine Environmental Research 50 (2000) 295±300
299
Fig. 2. Gill of Asian clam exposed to 8.40 mg lÿ1 of Cr-VI shows enhanced positive hsp70 staining (arrow). Hsp70 with hematoxylin counterstain (bar=30 mm).
Hsp70 in the 8.40 mg lÿ1 group (mean relative band densityS.E.=495004329, n=12) was approximately 10-fold that of the 0.92 mg lÿ1 (relative density= 37871908, n=11) and control (relative density=48681594, n=12) groups. Due to severe pathological alterations seen in clams, immunohistochemical analysis was only compared in organs that were not severely aected after Cr-VI exposure. Induction of cell proliferation and cell death was suggested by the number of cells staining positively for PCNA and apoptosis. However, since the number of apoptotic cells was higher than the number of PCNA-positive cells in clams exposed at 0.92 and 8.40 mg lÿ1 Cr-VI, sublethal deleterious eects of chromium surpassed proliferative capacity and likely led to a net reduction in cell number. Hsp70 stress proteins were detected both biochemically and histochemically. Although, the immunostaining intensity was determined visually, enhanced staining was localized in gill suggesting that this organ was most responsive with respect to hsp70 production upon exposure to chromium. Future Western blot biochemical analysis of hsp70 could focus on gill and achieve some enhanced sensitivity and decrease of eort. In summary, Cr-VI was shown to be acutely toxic to Asian clams at 25.6 mg lÿ1 exposure. Histopathologic and immunohistochemistry results indicated sublethal deleterious eect at 0.92 and 8.40 mg lÿ1, whereas a signi®cant increase in hsp70 stress proteins was detected only at 8.40 mg lÿ1 Cr-VI exposure. The results support a cause±eect relationship between contaminants exposure and physiologic changes, and that the DG, gill and reproductive organs are the principal targets of Cr-VI toxicity at sublethal concentrations. Cr-VI may have also played a role in the increased incidence of disease seen in clams during concurrent previous ®eld studies (Teh et al., 1999). Finally, we suggest the adverse eects on reproduction seen in
300
S.J. Teh et al. / Marine Environmental Research 50 (2000) 295±300
these studies may have contributed to the decrease in clam populations at sites with highest metal contamination in the San Francisco Bay Estuary (Brown & Luoma, 1995, 1998; Thompson et al., 1996). Acknowledgements Funding for these investigations was provided, in part, by US Public Health Service Grants CA-45131 from the National Cancer Institute, ES-04699 from the Superfund Basic Science Research Project, the US Environmental Protection Agency (USEPA) Grant Nos. R823297 and R825298, and the USEPA Center for Ecological Health Research at UC Davis Grants CR 819658. Although the information in this document has been partially funded by the USEPA, it may not necessarily re¯ect the views of the Agency and no ocial endorsement should be inferred. References Brown, C. L., & Luoma, S. N. (1995). Marine Ecology Progress Series, 124, 129±142. Brown, C. L., & Luoma, S. N. (1998). Interagency Ecological Program Newsletter, 11(N2), 33±35. Carlton, J. T., Thompson, T. K., Schemel, L. E., & Nichols, F. H. (1990). Marine Ecology Progress Series, V66(N1±2), 81±94. Teh, S. J., Clark, S. L., Brown, C. L., Luoma, S. N., & Hinton, D. E. (1999). Biomarkers, V4(N6), 497± 509. Thompson, J. K., Parchaso, F., Brown, C. L., & Luoma, S. N. (1996). U.S. Geological Survey Open File Report 96-437. Werner, I., & Hinton, D. E. (1999). Biomarkers, V4(N6), 473±484.