Estrogens maintain female sexual health. The hormone also drives the growth of estrogen receptor (ER) positive breast tumors, and ER modulators, like tamoxifen, are used to reduce tumor recurrence. To identify phytoestrogens with possible health benefits, we screened several Traditional Chinese Medicines and encountered an extract from the leaves of Epimedium brevicornum (EB), with strong (EC50: 1.3 microg/mL) and specific ER-stimulatory activity. It increased estrogen-responsive human breast cancer cell proliferation at low doses, but paradoxically caused profound inhibition of growth at higher doses. Using bioassay-guided fractionation, we isolated and characterized a new prenylflavone, breviflavone B, which exerted biphasic stimulatory and inhibitory effects on breast cancer cell proliferation, mimicking the effects of EB. In contrast to estradiol and genistein, high doses (> 2 microM) of breviflavone B almost eliminated ERalpha protein; a process that may be mediated through increased proteasome degradation. Pre-clinical studies are needed to explore whether these prenylflavones are of value in estrogen-deficiency states and for prophylaxis of breast cancer.

No abstract

Flavonoids present in food, botanicals, and body fluids occur as complex mixtures, and data on their combinatorial estrogenic effects are sparse. Human cell lines that permanently express estrogen receptor (ER) alpha and ERbeta proteins were developed for the measurement of the global estrogenicity of flavonoids in such complex mixtures. The presence of estrogenic ligands, known and unknown, in these mixtures can be detected by activation of an ER-driven luciferase reporter gene. We also examined the effect of hydroxylation on the estrogenic activities of four common flavonoids-apigenin, kaempferol, luteolin, and quercetin, alone and in combination. An inverse relationship was observed between the number of hydroxyl groups in flavonoids and ERalpha bioactivity. When submaximal doses of apigenin, luteolin, kaempferol, genistein, and estradiol were combined in binary and higher order mixtures, the experimental estrogenic effects matched those obtained by summing effects extrapolated from dose-response curves of individual compounds. The estrogenic activities of mixtures containing quercetin were observed to deviate from additivity, suggesting that it was a partial agonist/antagonist. Our assay reveals superagonistic, additive, and antagonistic ERalpha or ERbeta actions of flavonoids and adds to our understanding of the estrogenic effects of phytoestrogens in complex mixtures.