In studies to determine whether Saccharomyces cerevisiae produced estrogens, the organism was grown in culture media prepared using distilled water autoclaved in polycarbonate flasks. The yeast-conditioned media showed the presence of a substance that competed with [3H]estradiol for binding to estrogen receptors (ER) from rat uterus. However, it soon became clear that the estrogenic substance in the conditioned media was not a product of the yeast grown in culture, but was leached out of the polycarbonate flasks during the autoclaving procedure. [3H]Estradiol displacement activity was monitored by ER RRA, and the active substance was purified from autoclaved medium using a series of HPLC steps. The final purified product was identified as bisphenol-A (BPA) by nuclear magnetic resonance spectroscopy and mass spectrometry. BPA could also be identified in distilled water autoclaved in polycarbonate flasks without the requirement of either the organism or the constituents of the culture medium. Authentic BPA was active in competitive RRAs, demonstrating an affinity approximately 1:2000 that of estradiol for ER. In functional assays, BPA (10-25 nM) induced progesterone receptors in cultured human mammary cancer cells (MCF-7) at a potency of approximately 1:5000 compared to that of estradiol. The BPA effect on PR induction was blocked by tamoxifen. In addition, BPA (25 nM) increased the rate of proliferation of MCF-7 cells assessed by [3H]thymidine incorporation. Thus, BPA exhibited estrogenic activity by both RRA and two functional bioresponse assays. Finally, MCF-7 cells grown in media prepared with water autoclaved in polycarbonate exhibited higher progesterone receptor levels than cells.grown in media prepared with water autoclaved in glass, suggesting an estrogenic effect of the water autoclaved in polycarbonate. Our findings raise the possibility that unsuspected estrogenic activity in the form of BPA may have an impact on experiments employing media autoclaved in polycarbonate flasks. It remains to be determined whether BPA derived from consumer products manufactured from polycarbonate could significantly contribute to the pool of estrogenic substances in the environment.