Search

Background: Brominated flame retardants (BFRs), used in many types of consumer goods, are being studied because of concerns about possible health effects related to endocrine disruption, immunotoxicity, reproductive toxicity, and neurotoxicity. Tetrabromobisphenol A (TBBPA), the most widely used BFR, and human metabolites of certain congeners of polybrominated diphenyl ether (e.g., 3-OH-BDE-47) have been suggested to inhibit estrogen sulfotransferase, potentially affecting estrogen metabolism. Objectives: Our primary goal was to understand the structural mechanism for inhibition of the hormone-metabolizing enzyme estrogen sulfotransferase by certain BFRs. We also sought to understand various factors that facilitate the binding of flame retardants in the enzyme binding pocket. Methods: We used X-ray crystallography to obtain atomic detail of the binding modes of TBBPA and 3-OH-BDE-47 to estrogen sulfotransferase for comparison with binding of the endogenous substrate estradiol. Results: The crystal structures reveal how BFRs mimic estradiol binding as well as the various interactions between the compounds and protein residues that facilitate its binding. In addition, the structures provide insights into the ability of the sulfotransferase substrate binding pocket to accommodate a range of halogenated compounds that satisfy minimal structural criteria. Conclusions: Our results show how BFRs or their metabolites can bind to and inhibit a key hormone-metabolizing enzyme, potentially causing endocrine disruption. Citation: Gosavi RA, Knudsen GA, Birnbaum LS, Pedersen LC. 2013. Mimicking of estradiol binding by flame retardants and their metabolites: a crystallographic analysis. Environ Health Perspect 121:1194–1199; http://dx.doi.org/10.1289/ehp.1306902

Background: Diarylheptanoid (D3) isolated from the medicinal plant, Curcuma comosa, has estrogenic activity. Objective: We aimed to elucidate the mechanism(s) of D3 action and compare it with that of 17β-estradiol (E2) using both in vitro and in vivo uterine models. Methods: We used human uterine (Ishikawa) cells to determine the estrogenic action of D3 on the activation and nuclear translocation of estrogen receptor α (ERα). In addition, we further characterized the uterine response to D3 treatment in vivo. Results: D3 activated an estrogen responsive element (ERE) luciferase reporter through ERα, and molecular modeling suggested that D3 could be accommodated in the ERα binding pocket. Using modified ERα to assay ligand-dependent nuclear translocation, we observed D3-dependent ERα interaction and translocation. In mouse uteri, early- and late-phase estrogen-regulated gene responses were increased in D3-treated ovariectomized wild-type animals, in a manner similar to that of E2; no response was seen in ERα knockout animals. We observed a divergence in estrogen responses after D3 treatment: D3 induced robust DNA synthesis in uterine epithelial cells, linked to an increase in cell-cycle–related genes; however, no increase in uterine weight was observed 24 hr after treatment. D3 also affected uterine progesterone receptor expression patterns similar to E2. When D3 and E2 were administered together, we observed no additive or antagonistic effects of D3 on E2. Our findings suggest that D3 is a weak estrogenic agonist compound. Conclusion: D3 is a weakly acting phytoestrogen that mimics the mitogenic responses produced by E2 in an ERα-dependent manner, but it is unable to increase uterine weight or enhance or antagonize the effects of estrogen.

Previous studies have addressed the biological effects of brominated flame retardants (Birnbaum and Staskal 2004; Koike et al. 2013; Mariussen and Fonnum 2003; Ogunbayo et al. 2008), including a 2-year bioassay study performed by the National Toxicology Program (NTP), which demonstrated that tetrabromobisphenol A (TBBPA) can induce aggressive uterine tumors in rats (NTP 2013). As pointed out by Osimitz et al., TBBPA has been shown to bind poorly to the estrogen receptor, providing the impetus to study other pathways such as disruption of steroid transport and metabolism. Other groups have demonstrated the ability of TBBPA and flame retardant metabolites to inhibit estrogen sulfotransferase (SULT1E1), with IC50 (median inhibitory concentration) values near the Km for estradiol (Hamers et al. 2008; Kester et al. 2002; Zhang et al. 1998). Our work (Gosavi et al. 2013) was focused solely on understanding the structural mechanism by which these compounds bind to and inhibit SULT1E1’s ability to metabolize estradiol. The results of our work demonstrate that TBBPA and the 3-OH metabolite of BDE-47, although structurally different, bind in a similar manner at the estradiol binding site. This work suggests that these compounds could have an additive effect on the inhibition of this enzyme. We wholeheartedly agree with Osimitz et al. that the results of our work warrant future studies addressing the potential additive effect of these compounds on steroid metabolism in target tissues.