Bioassay-based profiling of steroid- and dioxin-like activities in French river waters using grab and integrative sampling
Résumé
Effect-based methods (EBM) are relevant tools to monitor chemical quality of aquatic systems as they provide integrative assessment of contamination components and inform on hazard associated with complex mixtures. In the frame of a French national program aiming at demonstrating the operational performance of passive sampling to monitor priority chemicals in surface waters, we applied a panel of in vitro bioassays to investigate 1) the occurrence of steroid- and dioxin- /PAH-like activities, 2) their distribution in the water column using grab sampling and integrative passive sampling. In addition, we looked into the relevance of in vitro estrogenicity in the water as compared to established trigger values and in vivo assessment in zebrafish. Twenty surface water sites of different anthropogenic pressures were studied. For each of them, both grab (dissolved phase and particulate matter) and integrative sampling (POCIS and Silicone Rubber membranes) were performed over a 15 days period and organic extract of each sample was analyzed with the bioassays. In vitro toxicity profiles in grab water organic extracts showed contrasted contamination patterns depending on the time of sampling (T0 and T15) and the studied sites. Estrogenic and PAH-like activities were the most occurent activities, while glucocorticoidic and anti-androgenic ones were detected at specific sites. No androgenic activity was detected. PAH-like activity was associated with both the dissolved and particulate matter phases while endocrine activities were only found in the dissolved phase. At some sites, in vitro estradiol-equivalents were found above the generic trigger value considering environmental risk (0.4 ng E2Eq/L). For 7 sites out of the 10 tested in vivo, activities were coherent with those predicted by in vitro bioassays. The present study highlights the relevance of combining a set of effect-based tools and smart sampling to monitor site-specific contamination of surface water chemical quality and the partitioning of measured activities in the water column. Our approach further includes in vivo mechanism-based assays, which enhances the toxicological relevance of the assessment. Comparison of our data with established trigger values for some activities will be discussed. Finally, ongoing chemical analyses of more than 100 priority substances, including estrogenic hormones, will allow the determination of their contribution to the biological activities detected.