(Anti)estrogenic effects of phytochemicals on human primary mammary fibroblasts, MCF-7 cells and their co-culture

In the public opinion, phytochemicals (PCs) present in the human diet are often considered beneficial (e.g. by preventing breast cancer). Two possible mechanisms that could modulate tumor growth are via interaction with the estrogen receptor (ER) and inhibition of aromatase (CYP19). Multiple in vitro studies confirmed that these compounds act estrogenic, thus potentially induce tumor growth, as well as aromatase inhibitory, thus potentially reduce tumor growth. It is thought that in the in vivo situation breast epithelial (tumor) cells communicate with surrounding connective tissue by means of cytokines, prostaglandins and estradiol forming a complex feedback mechanism. Recently our laboratory developed an in vitro co-culture model of healthy mammary fibroblasts and MCF-7 cells that (at least partly) simulated this feedback mechanism (M. Heneweer et al., TAAP vol. 202(1): 50-58, 2005). In the present study biochanin A, chrysin, naringenin, apigenin, genistein and quercetin were studied for their estrogenic properties (cell proliferation, pS2 mRNA) and aromatase inhibition in MCF-7 breast tumor cells, healthy mammary fibroblasts and their co-culture. The proliferative potency of these compounds in the MCF-7 cells derived from their EC(50)s decreased in the following order: estadiol (4*10(-3) nM)>biochanin A (9 nM)>genistein (32 nM)>testosterone (46 nM)>naringenin (287 nM)>apigenin (440 nM)>chrysin (4 microM). The potency to inhibit aromatase derived from their IC(50)s decreased in the following order: chrysin (1.5 microM)>naringenin (2.2 microM)>genistein (3.6 microM)>apigenin (4.1 microM)>biochanin A (25 microM)>quercetin (30 microM). The results of these studies show that these PCs can induce cell proliferation or inhibit aromatase in the same concentration range (1-10 microM). Results from co-cultures did not elucidate the dominant effect of these compounds. MCF-7 cell proliferation occurs at concentrations that are not uncommon in blood of individuals using food supplements. Results also indicate that estrogenicity of these PCs is quantitatively more sensitive than aromatase inhibition. It is suggested that perhaps a more cautionary approach should be taken for these PCs before taken as food supplements.