Unique bisphenol a transcriptome in prostate cancer: novel effects on ER[beta] expression that correspond to androgen receptor mutation status

Environmental Health Perspectives, Nov, 2007 by Janet K. Hess-Wilson, Siobhan L. Webb, Hannah K. Daly, Yuet-Kin Leung, Joanne Boldison, Clay E.S. Comstock, Maureen A. Sartor, Shuk-Mei Ho, Karen E. Knudsen

Accruing evidence indicates that exposure to environmental compounds that affect the function of the endocrine system may adversely impact human health. These substances deemed "endocrine disrupting compounds," or EDCs, are agents that disrupt or enhance known regulatory functions of the endocrine system and function through multiple mechanisms, including alteration of hormone receptor function (Henley 2006; Welshons et al. 2003). While mechanisms by which these effects occur and the level of risk posed to humans have yet to be fully elucidated, the biological significance of EDC exposure can be significant. In humans, a putative link has been established between increased abundance of estrogenic EDCs in the environment and both rising hormonedependent cancer incidence and reduced fertility (Huff et al. 1996). Thus, recent investigations have placed particular emphasis on delineating the consequence of estrogenic EDC exposure on reproductive tissues.

One such agent, 4,4'-isopropylidene-2-diphenol (bisphenol A, BPA), has been identified as a mitogen for a subset of prostate cancers (Wetherill et al. 2002, 2005, 2006). BPA is a nonplaner plasticizer, which is leached in microgram quantities from polycarbonate plastics and epoxy resins into food and water supplies (Welshons et al. 2003). More than 800 million kilograms of this compound are generated annually in the United States, and up to 95% of adults in the United States have detectable BPA in their urine (Calafat et al. 2005), with adult serum concentrations reported to range in nanomolar concentrations [reviewed by Welshons et al. (2006)]. BPA is known to harbor estrogenic activity; this compound is a weak agonist of both estrogen receptor alpha and beta (ER[alpha]and ER[beta], respectively) (Kuiper et al. 1997) and is capable of stimulating moderate estrogen-independent proliferation in breast cancer cells (Hess-Wilson et al. 2006; Olsen et al. 2003). The ERs are members of the steroid hormone nuclear receptor family of transcription factors, which are activated by specific steroid hormones and control numerous molecular pathways in hormone responsive tissues, including differentiation and proliferation (Ascenzi et al. 2006; Shang and Brown 2002). Although these reports highlighted a potentially influential role of BPA on ER activity in breast cancer, recent studies reveal BPA as an agonist for mutant androgen receptor (AR) activity in recurrent prostate cancer (Wetherill et al. 2002, 2005, 2006).

Prostatic adenocarcinomas are uniquely dependent on AR activity for growth and proliferation (Trapman and Brinkmann 1996). In prostate cancer cells, androgen [testosterone or dihydrotestosterone (DHT) binding activates the receptor to bind DNA at androgen-responsive 0elements (AREs), recruit co-activators, and initiate a program of gene transcription that induces cellular proliferation (Lee et al. 1995). Given the reliance of prostate cancer cells on this signal, nullification of AR activity is the first line of therapeutic intervention for disseminated disease, as achieved through either ligand depletion (androgen ablation strategies) or through the use of direct AR antagonists that prevent formation of active AR transcriptional complexes (Klotz 2000; Reid et al. 1999). Although these strategies are initially effective at inducing cell death or cell cycle arrest, recurrent tumors arise within a median of 2-3 years, wherein AR activity has been restored (Feldman and Feldman 2001). One mechanism of AR re-activation is somatic mutation of the AR (estimated to occur in 8-25% of recurrent tumors), where specific mutations of the ligand binding domain render the receptor responsive to an expanded host of ligands (Suzuki et al. 1993; Veldscholte et al. 1990, 1992a, 1992b; Zhao et al. 2000). Remarkably, it has been shown that one of the most common tumor-derived mutant forms of the AR, AR-T877A, is responsive to activation by BPA (Wetherill et al. 2002, 2005). Specifically, it was shown that at environmentally relevant levels of BPA (1 nM), this agent is capable of binding and activating the mutant receptor to induce endogenous expression of PSA (prostate specific antigen, a known direct AR target), which is used clinically to monitor prostate cancer development and progression. In cancer cells that express this mutant, lowlevel BPA exposure induced androgen-independent cellular proliferation, thereby indicating that in the context of AR-T877A, BPA exposure could potentially reduce therapeutic efficacy. This concept was recently validated in vivo, wherein BPA accelerated tumor growth after androgen ablation and significantly reduced PSA doubling times (Wetherill et al. 2006). Thus, there is significant evidence that BPA can bolster AR-T877A activity and as a result potentially alter the course of therapeutic response in prostate cancer.

Given the potent effects of BPA on mutant AR activation and prostate cancer progression, it is imperative to determine the molecular underpinning of BPA action. There are differences in prostatic cellular response to BPA compared with DHT, including magnitude of proliferative response, distinctions in ligand binding, and different ligand-induced kinetics of AR recruitment to gene regulatory regions (Wetherill et al. 2002, 2005). Moreover, it is known that differential ligands dictate target gene specificity with regard to the nuclear receptors. Although substantial in vitro analyses have revealed the direct binding to and activation of mutant ARs by BPA, it is also well documented that BPA can activate ER[alpha]and ER[beta], which are expressed in several prostate cancer cell lines [e.g., LNCaP (human prostatic adenocarcinoma cell line)]. However, the BPA-induced cellular proliferation is dependent on AR activation, as blocking of AR function (using the specific AR antagonist, Casodex) reversed all effects of BPA on tumor cell proliferation (Wetherill et al. 2002). Therefore, it is believed that the adverse proliferative effect of BPA is directly through the AR. In our present study, only the proliferation-inducing dose of BPA (1 nM) was used for analysis.