Evaluation of sensitivity and specificity of two crustacean biochemical biomarkers

Biochemical biomarkers are increasingly used for environmental assessment. Although the emphasis has been on vertebrate biomarkers, invertebrates biom arkers have been developed as well. This study evaluated the usefulness of biomarker responses of freshwater invertebrates by comparing the sensitivit y and specificity of endpoints at three levels of biological organization: biochemical, physiological, and individual. The study focused on the epiben thic amphipod Gammarus pulex L., and the end points were cholinesterase (Ch E) and glutathione-S-transferase (GST) activity, feeding inhibition, and mo rtality. Chemicals representing Five major classes of toxic chemicals were assessed, including zinc, linear alkylbenzene sulphonate (LAS; surfactant), lindane (organochlorine), pirimiphos-methyl (organophosphorus), and permet hrin (pyrethroid). Lethality was the least sensitive endpoint, with 96-h LC 50 values ranging from 2.78 mu g/L for permethrin to 6.31 mg/L for LAS. Com parison of the biochemical biomarkers and the sublethal feeding rate assay indicated that whereas ChE inhibition was a specific indicator of organopho sphate exposure, the biochemical assay was more than 13-fold less sensitive than the feeding rate assay. The GST biomarker performed with greater sens itivity but with lower specificity compared with the ChE biomarker. However , only on exposure to lindane did the GST biomarker marginally outperform t he feeding rate assay in terms of sensitivity. Feeding inhibition is both a general and a sensitive (LC50, <3%) indicator for exposure to a range of c hemicals. The Gammarus sp. ChE biomarker may have utility in providing a di agnostic and rapid indicator of organophosphate exposure, but evidence from this and other studies questions the sensitivity of this biomarker in pred icting sublethal, higher-order effects. The GST biomarker may provide a rap id and sensitive indicator for toxicant exposure, hut it has limited use as a diagnostic tool and provides only limited improvement in sensitivity ove r more ecologically relevant sublethal end points (e.g., feeding rate, grow th rate).