Combined effects of oestrogen receptor antagonists on in vitro vitellogenesis

Some environmental compounds are known to have anti-oestrogenic activity and their modes of action (MoA) are believed to include competitive inhibition of 17β-estradiol (E2) binding to the oestrogen receptor (ER) or interference with ER-dependent processes. The presence of multiple compounds having the same MoA may cause concern, as exposure to multiple compounds at concentrations below their threshold for effect can interact with cellular targets to cause effects in combination. The combined effect of mixtures can be assessed using prediction models such as concentration addition (CA) and independent action (IA). The objective of the present study was to determine if the CA and IA prediction models could accurately characterise the combined effects of mixtures of ER antagonists in rainbow trout (Oncorhynchus mykiss) hepatocytes using the ER-mediated production of the oestrogenic biomarker vitellogenin (Vtg) as a screening assay. Model anti-oestrogens (4-hydroxytamoxifen and ZM 189.154) and environmentally relevant compounds (PCBs and PAHs) were tested to ensure inclusion of compounds from different chemical classes and with different MoAs. All eleven tested compounds had the ability to reduce the in vitro E2-induced production of Vtg in a concentration-dependent manner. The potency of the tested compounds differed by four orders of magnitude based on the concentrations for 50% inhibition (IC(50)). The observed order of potency was 2,3,7,8-tetrachlorodibenzo-p-dioxin>4-hydroxytamoxifen>3,3',4,4',5-pentachlorobiphenyl>benzo(k)fluoranthene>3,3',4,4'-tetrachlorobiphenyl>β-naphthoflavone>ZM 189.154>indeno[1,2,3-cd]pyrene>benzo(b)fluoranthene>benzo(a)pyrene>benzo(a)anthracene. The CA and IA models were able to predict the combined effects of mixtures of ER antagonists with similar MoA. The mixtures of certain ER-antagonists with different and/or complex MoA caused deviations from both the CA and the IA model by causing higher anti-oestrogenic activity than predicted from the potency of the compounds alone. The rationale for these deviations warrants additional studies to assess the potential impacts on the health of organisms.