ABSTRACT-Lysosomal destabilization indices (based on a neutral red retention assay) were evaluated for digestive gland cells of oysters Crassostrea virginica. Laboratory studies were conducted to evaluate the effects of variable salinity regimes (a natural stressor) and copper exposures (a pollutant stres-SOT) on lysosomal destabilization. Field studies were also conducted with native oysters and hatcheryreared juvenile oysters deployed in situ at reference and polluted sites. Lysosomal integrity was not affected by either short-term or longer-term variations in salinity during laboratory experiments, and the destabilization indices were similar to those of deployed and native oysters from reference sites characterized by a range of salinlties. However, laboratory Cu exposures (ranging from 2.5 to 20 pg Cu 1-') caused significant adverse effects on lysosomal destabilization after only 18 h, and the eiiecis were sustained or worsened with increasing exposure time and concentration. Hatchery-reared oysters deployed at or native oysters collected from polluted sites had significantly higher lysosomal destabilization indices. Estuanne habitats are characterized by variable sal~nity regimes, so cellular responses that are sensitive to salinity stress as well as contaminant stress would be difficult to interpret if salinity effects cannot be distinguished from pollutant effects. A robust biomarker of anthropogenic effects should be insensitive to natural stressors such as salinity, but should be sensitive to pollutants. The salinity and contaminant studies described in this paper indicate that lysosomal destabilization responses in oysters are insensitive to salinity variations and sensitive to pollutants, and are therefore potentially valuable biomarkers of anthropogenic stress.

Abstract-Selective serotonin reuptake inhibitors (SSRIs) are widely prescribed drugs that are present in sewage effluents and surface waters. The objective of the present study was to determine whether low environmentally relevant concentrations of the SSRIs fluoxetine and sertraline could impair growth and development in tadpoles of the African clawed frog (Xenopus laevis) and to evaluate if such effects may be caused by a disruption of the neuroendocrine system. Tadpoles were exposed to SSRIs at concentrations of 0.1, 1, and 10 mg/L for 70 d throughout metamorphosis. No effects on deformities were observed. Tadpoles exposed to fluoxetine (10 mg/L) and sertraline (0.1, 1, and 10 mg/L) exhibited reduced growth at metamorphosis. Tadpoles exposed to sertraline (0.1 and 1 mg/L) exhibited an acceleration of development as indicated by an increase in the time to tail resorption. The effects of SSRIs on growth and development in tadpoles were likely driven by reduced food intake. Reduced feeding rates were observed in SSRI-exposed tadpoles, and nutritional status can influence growth and development in amphibians via effects on the neuroendocrine system. Only sertraline was capable of causing developmental toxicity in tadpoles at environmentally relevant concentrations. These data warrant additional research to characterize the risks to human health and wildlife from pharmaceutical exposures.

Many chemicals, including fertilizers, herbicides, and insecticides, are routinely applied to turf in the care and maintenance of lawns. These chemicals have the potential to leach into nearby surface waters and adversely affect aquatic biota. In this study, we evaluated the lethal and genotoxic effects of chemicals used in lawn care on an early life stage of freshwater mussels (Utterbackia imbecillis). The chemicals tested were copper and commercial formulations of atrazine, glyphosate, carbaryl, and diazinon. Mussel glochidia were exposed to chemicals singly or in combination (equitoxic and environmentally realistic mixtures) for 24 h and toxic interactions were evaluated with Marking's additive index. Genotoxicity was quantified with the alkaline single-cell gel electrophoresis assay (Comet assay). In acute tests, copper was the most toxic of all chemicals evaluated (LC50 = 37.4 microg/L) and carbaryl was the most toxic of all pesticides evaluated (LC50 = 7.9 mg/L). In comparison to other aquatic organisms commonly used in toxicity tests (e.g., amphipods, cladocerans, and chironomids), mussel glochidia were as or more sensitive to the chemicals evaluated with the exception of diazinon, where mussels were observed to be less sensitive. The combined toxicity of equitoxic and environmentally realistic mixtures to mussels was additive. Genotoxic responses were observed in mussels exposed to copper, atrazine and diazinon at levels below their respective no-observed-effect concentrations. Together, these data indicate that freshwater mussels are among the most sensitive aquatic organisms tested for some chemicals commonly used in lawn care and that DNA damage may be useful as a screening tool to evaluate potential sublethal effects of lawn care products on non-target aquatic organisms.

Hatchery-reared juvenile oysters (Crassostrea virginica) were deployed in situ for approximately 1 month from mid-May to mid-June of 1996 at sites that were classified as reference, agricultural, suburban, or urban/industrial. Cellular responses (lysosomal destabilization, glutathione concentrations, lipid peroxidation, heat shock proteins, metallothioneins, and multi-xenobiotic resistance proteins) were analysed, and their efficacy as biomarkers of stress was evaluated. Increased lysosomal destabilization, glutathione depletion, increased lipid peroxidation, and induction of heat shock proteins and metallothioneins were observed at many of the polluted sites, but increases in multixenobiotic resistance proteins were not. Significant correlations between sediment contaminants and lysosomal destabilization or glutathione concentrations were observed. Similarly, there were significant correlations between sediment cadmium and copper levels and metallothioneins. Although elevated lipid peroxidation products and heat shock proteins were observed at some of the contaminated sites, there were no significant correlations with contaminants. These studies suggest that lysosomal destabilization and glutathione depletion are sensitive, robust indicators of contaminant stress. Although lipid peroxidation and heat shock protein responses were not correlated with contaminants, they are still regarded as valuable indicators of stress. These studies demonstrate the value of using a suite of cellular biomarkers to identify and characterize stress responses related to anthropogenic perturbations.

Biochemical indicators and in vitro models, if they mimic in vivo responses, offer potentially sensitive tools for inclusion in toxicity assessment programs. The purpose of this study was to determine whether the HepG2 cell line would mimic known in vivo or in vitro (or both) responses of mammalian systems when confronted with cadmium (Cd2+). Uptake and compartmentalization of Cd2+, metallothionein (MT) compartmentalization, and glutathione (GSH) depletion were examined. In addition, several cytotoxic and stress effects, e.g., viability (neutral red [NR] uptake, 3-[4,5-dimethylthiozole-2-yl]-2,5,-biphenyl tetrazolium bromide [MTT] dye conversion, and live/dead [L/D]), membrane damage (lactate dehydrogenase leakage), metabolic activity (adenosine triphosphate levels), and detoxification capabilities (GSH content, cytochrome P4501A1/2 [EROD (ethoxyresorufin-o-deethylase)] activity, and MT induction), were measured in both naive (no previous exposure) and Cd2+ preexposed cells. Cadmium uptake increased during a 24-h period. Metallothionein induction occurred in response to both Cd2+ and ZnCl2; however, Cd2+ was the more potent inducer. Both Cd2+ and MT were localized primarily in the cytoplasmic compartment. All biochemical responses, except EROD, showed concentration- response relationships, after 24-h exposure to Cd2+ (ranges 0-3 ppm [26.7 microM]). Cadmium effects were reduced in preexposed cells, indicating adaptive tolerance or increased resistance had occurred. Twenty-four-hour LC50, dose causing death of 50% of the test subjects, values were 0.97, 0.69, and 0.80 ppm (8.7, 6.2, and 7.2 microM) for naive cells and 1.45, 1.21, and 1.39 ppm (12.9, 10.7, and 12.3 microM) for preexposed cells based on the NR, MTT, and L/D assays, respectively. These data indicate that this carcinoma cell line is a useful in vitro model for cadmium toxicity studies.