Pharmaceuticals are found in the aquatic environment but their potential effects on non-target species like fish remain unknown. This in vitro study is a first approach in the toxicity assessment of human drugs on fish. Nine pharmaceuticals were tested on two fish hepatocyte models: primary cultures of rainbow trout hepatocytes (PRTH) and PLHC-1 fish cell line. Cell viability, interaction with cytochrome P450 1A (CYP1A) enzyme and oxidative stress were assessed by using 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrasodium bromide tetrazolium (MTT), 7-ethoxyresorufin-o-deethylase (EROD) and dichlorofluorescein (DCFH-DA) assays, respectively. The tested drugs were clofibrate (CF), fenofibrate (FF), carbamazepine (CBZ), fluoxetine (FX), diclofenac (DiCF), propranolol (POH), sulfamethoxazole (SFX), amoxicillin (AMX) and gadolinium chloride (GdCl(3)). All substances were cytotoxic, except AMX at concentration up to 500 microM. The calculated MTT EC(50) values ranged from 2 microM (CF) to 651 microM (CBZ) in PLHC-1, and from 53 microM (FF) to 962 microM (GdCl(3)) in PRTH. CF, FF, and FX were the most cytotoxic drugs and induced oxidative stress before being cytotoxic. Compared to hepatocytes from human and dog, fish hepatocytes seemed to be more susceptible to the peroxisome proliferators (PPs) CF and FF. In PLHC-1 cells none of the tested drugs induced the EROD activity whereas POH appeared as a weak EROD inducer in PRTH. Moreover, in PRTH, SFX, DiCF, CBZ and to a lesser extend, FF and CF inhibited the basal EROD activity at clearly sublethal concentrations which may be of concern at the biological and chemical levels in a multipollution context.

No abstract

Cross-species differences between human and fish estrogen receptor (ER) binding by environmental chemicals have been reported. To study ER transactivation in a fish cellular context, we stably co-transfected the PLHC-1 fish hepatoma cell line with a rainbow trout estrogen receptor (rtER) and the luciferase reporter gene driven by an estrogen response element (ERE). This new cell model, called PELN-rtER (for PLHC-1-ERE-Luciferase-Neomycin), responded to 17beta-estradiol (E2) in a both concentration- and temperature-dependent manner, as well as to environmental ER ligands from different chemical classes: natural and synthetic estrogens, zearalenone metabolites, genistein, alkyphenoles and benzophenone derivatives. The comparison with other in vitro models, i.e. human reporter cell lines (HELN-rtER, MELN) and vitellogenin induction in primary cultures of rainbow trout hepatocytes, showed an overall higher sensitivity of the human cells for a majority of ligands, except for benzophenone derivatives which were active at similar or lower concentrations in fish cells, suggesting species-specificity for these substances. Correlation analyses suggest that the fish cell line is closer to the trout hepatocyte than to the human cell context, and could serve as a relevant mechanistic tool to study ER activation in fish hepatic cellular context.