Red wine is enriched in resveratrol, trans-3,5,4'-trihydroxystilbene, a compound in grape skin that inhibits the development of pre-neoplastic lesions in mouse mammary tumor cells in culture and inhibits cancer cell proliferation in vitro. Grapes also contain other bioactive compounds including flavonoids, flavans, and anthocyanins. The estrogenic activities of extracts prepared from one white (Freie Weingärtner Wachau, Grüner Veltliner, Austria) and two red wines (Woodbridge, Cabernet Sauvignon, California; and Lenz Moser Prestige, Blaufränkisch Barrique, Austria) were examined and compared with those induced by estradiol (E(2)) and trans-resveratrol. First, the estrogenic activity of the wine extracts was evaluated in a yeast estrogen screen (YES) assay, in which yeast express copper-inducible estrogen receptor alpha (ERalpha) and an estrogen-response-element (ERE)-driven beta-galactosidase reporter. In YES, the white wine extract showed no estrogenic activity. In contrast, both of the red wine extracts showed estrogenic activity equivalent to that of 0.2 nM E(2). Similarly, the white wine extract showed no transcriptional activity with either ERalpha and ERbeta in transiently transfected CHO-K1 cells. In contrast, both red wine extracts stimulated ERE-reporter activity in a concentration-dependent manner that was inhibited by 4-hydroxytamoxifen (4-OHT), indicating that the observed transcriptional activity was ER-mediated. The red wine extracts showed significantly higher ERbeta versus ERalpha agonist activity. Resveratrol showed no agonist activity in YES but activated ERalpha and ERbeta in CHO-K1 cells in a concentration-dependent manner that was inhibited by 4-OHT. This indicates that resveratrol requires mammalian cell components that are absent in yeast for estrogen agonist activity, whereas the estrogenic activity of wine extracts is directly through ERalpha and does not require mammalian cell factors such as coactivators. The estrogenic activity in red wine found by using YES indicates that estrogenic compounds other than resveratrol are present. Chemical analysis clearly showed that the trans-resveratrol content of the red wine extracts was 1 order of magnitude below the detection limit for YES assay.