Investigation of (anti)estrogenicity of triclosan using in vitro and in vivo zebrafish bioassays
Résumé
Triclosan (TCS) is a wide spectrum antimicrobial agent used in many personal care products (PCPs) that has been detected in surface water and biota. TCS is suspected to be an endocrine disrupting chemical (EDC) and to interact with the estrogen and androgen signaling pathways. However, data available regarding its estrogenic or anti-estrogenic potency provide contradictory information, highlighting the uncertainty regarding its endocrine disruptive potential. The few studies published so far on chronic reproductive effects of TCS on fish report ambiguous results, i.e. a male bias sex ratio in Japanese medaka but a vitellogenin induction in male mosquito fish. In this context, our study aimed to investigate whether TCS acts as an agonist or an antagonist of the estrogen receptor (ER) in human (in vitro) and in zebrafish (in vitro and in vivo). For this purpose, we assessed luciferase-dependent ER transactivation in vitro using a human breast cancer reporter cell line expressing hERα (MELN cells), and three zebrafish (zf) liver cell lines (ZELH) stably transfected with zfER (i.e. ZELH-zfERα, -zfERß1 and -zfERß2). In addition, cyp19a1b gene induction, which is strictly ER-regulated, was assessed in transgenic cyp19a1b-GFP zebrafish embryos (EASZY assay). Our results showed that TCS alone from 3nM to 10µM was unable to induce luciferase activity whatever the in vitro models used. At higher concentrations, cytotoxic effects were observed. Co-exposure of cells with E2 0.1nM (ZELH-zfERß1, ZELH-zfERß2 and MELN) or E2 3nM (ZELH-zfERα) and increasing concentrations of TCS had no effect on the E2-induced luciferase activity in either human or zebrafish cells, suggesting that TCS did not elicit any antagonist activity. In vivo, TCS alone did not induce GFP in transgenic zebrafish embryos, but decreased the basal cyp19a1b expression at 1µM. This effect could reflect the developmental toxicity of TCS as revealed by lethargy and swimming alteration observed in exposed-fish at 1µM. Co-exposure with TCS 1µM and E2 resulted in ambiguous responses: TCS 1µM decreased E2-induced cyp19a1b expression by 24% as compared to E2 10nM alone, whereas it slightly increased E2-induced cyp19a1b expression at 0.625 nM. All together, our data indicate that TCS does not alter the ER signaling in vitro in either human and zebrafish cell lines but seems to alter the E2-induced response in vivo through unknown mechanisms.