BackgroundMercury is a pervasive environmental pollutant whose toxic effects have not been studied in sea turtles in spite of their threatened status and evidence of immunosuppression in diseased populations.ObjectivesIn the present study we investigate mercury toxicity in loggerhead sea turtles (Caretta caretta) by examining trends between blood mercury concentrations and various health parameters.MethodsBlood was collected from free-ranging turtles, and correlations between blood mercury concentrations and plasma chemistries, complete blood counts, lysozyme, and lymphocyte proliferation were examined. Lymphocytes were also harvested from free-ranging turtles and exposed in vitro to methylmercury to assess proliferative responses.ResultsBlood mercury concentrations were positively correlated with hematocrit and creatine phosphokinase activity, and negatively correlated with lymphocyte cell counts and aspartate amino-transferase. Ex vivo negative correlations between blood mercury concentrations and B-cell proliferation were observed in 2001 and 2003 under optimal assay conditions. In vitro exposure of peripheral blood leukocytes to methylmercury resulted in suppression of proliferative responses for B cells (0.1 μg/g and 0.35 μg/g) and T cells (0.7 μg/g).ConclusionsThe positive correlation between blood mercury concentration and hematocrit reflects the higher affinity of mercury species for erythrocytes than plasma, and demonstrates the importance of measuring hematocrit when analyzing whole blood for mercury. In vitro immunosuppression occurred at methylmercury concentrations that correspond to approximately 5% of the individuals captured in the wild. This observation and the negative correlation found ex vivo between mercury and lymphocyte numbers and mercury and B-cell proliferative responses suggests that subtle negative impacts of mercury on sea turtle immune function are possible at concentrations observed in the wild.

Perfluorinated compounds (PFCs) have been measured in blood of humans and wildlife and are considered globally distributed contaminants. We examined 12 PFCs in the plasma of 73 loggerhead sea turtles (Caretta caretta) and 6 Kemp's ridley sea turtles (Lepidochelys kempii) captured from inshore waters of Core Sound, North Carolina (NC), and offshore waters of South Carolina, Georgia, and Florida (SC-FL). Perfluorooctanesulfonate (PFOS) and perfluorooctanoic acid (PFOA) were the dominant compounds, with respective mean concentrations of 11.0 ng/mL and 3.20 ng/mL for loggerhead turtles and 39.4 ng/mL and 3.57 ng/mL for Kemp's ridley turtles. Mean PFOS concentrations were 2- to 12-fold higher than typical mean sigmaPCB concentrations (approximately 5 ng/g wet mass) measured previously in sea turtle blood. More than 79% of the samples had detectable levels of perfluorocarboxylates (PFCAs) with 8-12 carbons, whereas only 17% or less of samples had detectable levels of PFCAs with 6 or 7 carbons. No samples had detectable levels of PFCAs with 4 or 5 carbons. In loggerhead turtles, sigmaPFC concentrations were not influenced by sex (p > 0.05), but were higher in turtles captured from inshore waters of NC than in turtles from offshore waters of SC-FL (p = 0.009). A backward stepwise multiple regression model showed that sigmaPFC concentrations were (1) significantly higher in Kemp's ridley turtles than loggerhead turtles (p < 0.0001), (2) higher in larger turtles (p = 0.018; carapace length used as a proxy for age), and (3) higher in turtles captured toward the north (p = 0.006). These findings suggest that bioaccumulation of PFCs in sea turtles is influenced by species, age, and habitat.