In vitro and in vivo evaluation of the potential estrogenic effects of polycyclic aromatic hydrocarbons

A large number of exogenous compounds have been found to possess the ability to bind to the α and β isoforms of the estrogen receptor (ER). The identification and characterization of these compounds is typically achieved through in vitro assays and, when concern is warranted, by assessing the ability of the compound to stimulate uterine proliferation in the rodent uterus. Benzo[a]pyrene (B[a]P) is a ubiquitous pollutant and shares some structural similarity with endogenous estrogens, particularly when hydroxyl groups are introduced during metabolism. Specific B[a]P metabolites, as well as other compounds of interest, were identified that were able to bind to ERα and ERβ in vitro, and to induce ERα- and ERβ-mediated reporter gene expression in MCF-7 breast cancer cells. B[a]P was also shown to interact in an additive manner in vitro when cotreated with estrogen. However, administration of B[a]P or of its most active metabolites to immature, ovariectomized mice was not sufficient to stimulate uterine proliferation or induction of uterine lactoferrin mRNA expression. Because the transcriptional targets of estrogen in the uterus are numerous and many are still undiscovered, a GeneChip microarray approach was used in order to characterize the temporal transcriptional responses to estrogen in the uterus. This approach was able to confirm previously characterized effects as well as to identify novel responses. For example, the arginine and ornithine utilization pathway was found to be particularly highly regulated in the uterus in response to estrogen, and while induction of some of these enzymes had been shown previously, this provided the first evidence that numerous aspects of this pathway could be monitored at the transcription level in the estrogen-stimulated uterus. Expression of arginase 1 was found to be particularly high in response to prolonged estrogen exposure. The ability of B[a]P to alter the expression of arginase 1 and other chosen genes was then assessed in the rodent uterus. While B[a]P was recently reported to antagonize the ability of estrogen to stimulate the proliferation of MCF-7 breast cancer cells in vitro, we showed that B[a]P was not able to interfere with estrogen-induced uterine proliferation or with expression of estrogen-regulated genes including arginase 1. Finally, the extensive literature searching required in the analysis of the microarray data made it clear that the lack of a comprehensive review of previously characterized estrogen-responsive genes is a great impediment to the analysis of new large-scale transcriptional profiling studies. Accordingly, a comprehensive review of estrogen-regulated transcription in the uterus is provided.