Optimization of an estrogen receptor-alpha transcriptional activation assay for testing a diverse spectrum of chemicals

Estrogen receptor (ER) transactivation assays were initially designed to st udy endogenous mechanisms of steroid hormone action, but more recently have been used to assess industrial chemicals for potential estrogenic activity . Given the diverse spectrum of physicochemical properties of these chemica ls, we examined the effects of pH (a weak organic and strong inorganic acid and base), hyperosmolality (NaCl, mannitol) and two different vehicles (DM SO, Triton X-100),on responses to estradiol-17 beta (E-2) in an ER transact ivation assay. MCF-7 human breast cancer cells were transiently transfected with a chimeric estrogen receptor (Gal4-HEG0) and a GaI4-regulated lucifer ase reporter gene (17m5-G-Luc), treated with E-2 under various test conditi ons, and then assessed for ER-mediated luciferase activity. Maximal E-2-ind uced reporter activity was observed at pH 7.8 (pre-incubation), but was mar kedly reduced at pH less than or equal to7.5 or greater than or equal to8.0 (P < 0.001), even though there was no evidence of cytotoxicity, Hyperosmol ality induced by addition of mannitol (<greater than or equal to>25 mM) res ulted in significant decreases in overall assay responsiveness, whereas NaC l (greater than or equal to 80 mM) decreased the sensitivity of the assay b y increasing the no-observed-effect level for E-2 compared to central cultu res (330 mOsm), The maximal DMSO concentration that resulted in consistentl y high E-2-induced reporter activity was 0.1%, whereas concentrations of Tr iton X-100 above 1 ppm inhibited E-2-induced reporter responses and were cy totoxic above 10 ppm. These results indicate that various physicochemical f actors have the potential to confound assay data if not kept within predefi ned operational limits. Copyright (C) 2000 John Whey & Sons, Ltd.