A comprehensive evaluation of organic contamination was performed in sediments sampled in two reference and three impacted small streams where endocrine disruptive (ED) effects in fish have been evidenced. The approach combined quantitative chemical analyses of more than 50 ED chemicals (EDCs) and a battery of in vitro bioassays allowing the quantification of receptor-mediated activities, namely estrogen (ER), androgen (AR), dioxin (AhR) and pregnane X (PXR) receptors. At the most impacted sites, chemical analyses showed the presence of natural estrogens, organochlorine pesticides, parabens, polycyclic aromatic hydrocarbons (16 PAHs), bisphenol A and alkylphenols, while synthetic steroids, myco-estrogens and phyto-estrogens were not detected. Determination of toxic-equivalent amounts showed that 28-96% of estrogenic activities in bioassays (0.2-6.3 ng/g 17beta-estradiol equivalents) were explained by 17beta-estradiol and estrone. PAHs were major contributors (20-60%) to the total dioxin-like activities. Interestingly, high PXR and (anti)AR activities were detected; however, the targeted analysed compounds could not explain the measured biological activities. This study highlighted the presence of multiple organic EDCs in French river sediments subjected to mixed diffuse pollution, and argues for the need to further identify AR and PXR active compounds in the aquatic environment.

We used an array of in vitro cell-based bioassays to assess dioxin-like, estrogenic and (anti-)androgenic activities in organic extracts of sediments from the Bizerta lagoon, one of the largest Tunisian lagoons subjected to various anthropogenic and industrial pressures. The sediments were sampled both in winter and summer 2006 in 6 stations differently impacted and in one reference station located in the seawards entrance of Ghar el Melh lagoon. Chemical analyses of the 16 priority PAHs showed that the sediments were low to moderately contaminated (2-537 ng/g dry weight). By using the estrogen- (MELN) and androgen-responsive (MDA-kb2) reporter cell lines, significant estrogenic and anti-androgenic activities were detected only in the Menzel Bourguiba (MB) site, the most contaminated site, both in winter and summer. By using 7-ethoxyresorufin-O-deethylase (EROD) induction in the fish PLHC-1 cell line after both 4 and 24 h of cell exposure, dioxin-like activities were detected in all analysed samples. Dioxin-like activities were higher after 4 h exposure, and varied according to the sites and the sampling season. While highly significant correlation was observed between bioassay- and chemical analyses-derived toxic equivalents (TEQs), PAHs accounted for only a small part (up to 4%) of the detected biological activities, suggesting that other readily metabolised EROD-inducing compounds were present. This study argues for the use of short time exposure to assess biological TEQs in low contaminated samples and provides new induction equivalent factors (IEF(4h)) for 16 PAHs in the PLHC-1 cell line. Finally, our results stress the need to further characterise the nature of organic chemical contamination as well as its long-term impacts on aquatic wildlife in the Bizerta lagoon.