Pleiotropic combinatorial transcriptomes of human breast cancer cells exposed to mixtures of dietary phytoestrogens.

Combinations of estrogen receptor agonists have been shown to exert more potent effects than single compounds in many single-endpoint bioassays. However, to our knowledge, it has never been tested how genome-wide expression programs are shaped by the interplay of multiple estrogenic stimuli. In view of the abundance of dietary phytoestrogens, we selected binary mixtures of these phytochemicals to determine their global impact using high-density DNA microarrays. MCF7 cells, a frequent in vitro model for molecular processes associated with breast cancer, were exposed to a sub-saturating concentration of coumestrol either alone or in combination with analogs that exhibit 1000-fold lower estrogen receptor activity. As expected, in the presence of coumestrol, the induction of many estrogen-sensitive genes was not further increased by the addition of resveratrol or enterolactone. However, it was surprising to find that these weak phytoestrogens, when combined with coumestrol in equal concentrations, were able to more than double the number of significantly regulated transcripts. Thus, phytoestrogens with low receptor affinity interact with other estrogenic agonists to generate more widespread expression fingerprints. This effect involving the number of susceptible transcripts instead of their amplitude of induction remains undetected if mixtures are evaluated with conventional bioassays.