An estuarine mudsnail in situ toxicity assay based on postexposure feeding

Abstract

In situ assays provide more realistic exposure scenarios than laboratory assays, which is particularly pertinent for estuaries because exposure conditions are difficult to simulate. Traditionally, sublethal toxicity testing endpoints, such as growth, emergence, and reproduction, imply time-delayed extrapolations from individuals to populations, communities, and ecosystems. Sublethal responses mechanistically linked to ecosystem functions have been largely neglected. Feeding is an unequivocal ecologically meaningful response because its impairment has direct and immediate effects on ecosystems, by hampering key functions such as organic matter decomposition, long before its effects at the individual level have consequences at successively higher levels of biological organization. The ultimate goal of the present study was to widen the range of ecosystem functions for estuarine quality assessments. Specifically, a short-term in situ assay based on the postexposure feeding of the mudsnail Hydrobia ulvae is presented. Methodologies to quantify precisely postexposure egestion as a surrogate of feeding were achieved. A multiple regression model from laboratory experiments was successfully applied to an in situ assay at reference (Mira River) and contaminated Portuguese estuaries (Sado River) for predicting reference results and allowing unraveling confounding factors during exposure (temperature, salinity, sediment grain size). Sensitivity comparisons of postexposure feeding with survival and growth, after Cu exposure, were carried out and used for a first preliminary appraisal of the relative consequences of ecosystem-level immediate effects.

A laboratory and in situ postexposure feeding assay with a freshwater snail

“…Temperature did not influence postexposure feeding, but the range among reference sites was very narrow. This parameter was found to be significant in other studies [7,13,16]; thus, caution is advised if sites with different temperatures are to be assayed.…”

“…After 48 h of exposure, feeding of T. fluviatilis was marginally more sensitive to cadmium than survival, with EC50 and EC20 values being 1.2 times and 1.9 times lower than LC50 and LC20 values, respectively. The former difference-1.2 times-was somehow unexpected, given that previous studies found larger differences between both endpoints for other invertebrates, namely, the estuarine crab Carcinus maenas (6.5 times [10]) also with cadmium and, with copper, the polychaete Hediste diversicolor (4.6 times [7]), the estuarine mudsnail H. ulvae (over 2.7 times [13]), and the freshwater amphipod Echinogammarus meridionalis (2.2 times [19]). The possibility that physiological recovery was responsible for such a small difference found in the present study cannot be ruled out and, thus, further studies are needed to unravel this issue.…”

“…First, the feeding rate at each concentration was multiplied by the respective proportional survival and a logistic model was fitted to estimate popEC x values (i.e., concentrations causing x % inhibition in the immediate population consumption). Second, the concentration–response relationships of feeding and of survival were integrated as x % of the population consumption inhibition induced by the concentration that simultaneously caused a y % mortality (i.e., the LC y ) and a w % depression on feeding (i.e., the EC w ; therefore, LC y = EC w ), x (in %) being quantified as …”

“…For instance, if organisms more sensitive to feeding depression are those dying earlier, then the survivors will feed at a higher rate than the average of the initial population, thus partially compensating at the population level for the loss of some individuals (the dead ones) [13,19]. Both techniques led to very close 48-h population consumption EC50 estimates (69 mg/L and 67 mg/L of cadmium), corroborating the findings of Krell et al [13] (75 mg/L and 68 mg/L of copper) and Agostinho et al [19] (187 mg/L and 168 mg/L of copper) for a mudsnail and an amphipod, respectively. At the 48-h LC20, the immediate inhibition of population consumption would be 53%, which corresponds to an expectedly strong disruption of ecosystem functioning.…”

“…The decrease in feeding rate because of contamination is expected to be fast, general, and quantifiable [2,[5][6][7]. It has been used as a toxicity test endpoint with cladocerans [6,8], amphipods [3,9], decapods [10,11], snails [12,13], polychaetes [7,14,15], midges [14,16], and fish [17,18]. If feeding inhibition occurs at contaminant concentrations close to those affecting growth or reproduction, then predicting ecosystem-level effects-taking into account only time-delayed extrapolations from population growth rates-can strongly underestimate risk.…”