br Materials and methods br Results
Materials and methods
Discussion The oestrogen agonist properties and androgen antagonist activities of parabens have been well documented by a variety of in vitro and in vivo assay systems over the past decade (Chen et al., 2007, Darbre and Harvey, 2008, Golden et al., 2005). To the best of our knowledge, this is the first study to explore the binding activities of parabens to ERRγ. ERRγ is the most recently identified member of the ERR subfamily by yeast two-hybrid screening, using SRC2 as bait (Hong et al., 1999). It is known as a constitutive regulator that can interact with transcriptional coactivators, resulting in transcriptional activation in the absence of ligands (Hong et al., 1999, Xie et al., 1999). In this study, the developed in vitro ERRγ coactivator recruiting assay clearly showed the constitutive transcriptional activity of ERRγ, as evidenced by the BAP activity (Fig. 2A). Through antagonist-competitive binding, this in vitro assay also demonstrated the binding activity of BPA with ERRγ (Fig. 2D), suggesting that the method could be used to analyse the binding activity of chemicals on ERRγ. Recent studies have shown strong ERRγ binding activity of compounds containing phenolic hydroxyl groups in a para position, such as BPA, BPZ, 43M and 1OH, indicating the potential and intrinsic importance of these bindings (Abad et al., 2008, Matsushima et al., 2008, Okada et al., 2008, Takayanagi et al., 2006). Furthermore, the concentrations of these known ERRγ binding compounds were much lower than those found in the human body. In this study, we found that parabens possessed high inverse antagonist activities on ERRγ that were close to that of BPA (Fig. 3), whereas salicylates and benzoates possessed lower and no activities, respectively (Fig. 3). Automated molecular docking revealed that parabens fitted well into the active site of ERRγ, and that hydrogen bonds could be formed between the p-hydroxyl groups in parabens and the Glu275/Arg316 of ERRγ. Although the PMF values of MP, EP, PP, BuP and BzP were less than that of BPA, they were higher than that of 43M. This confirmed the strong ERRγ binding activities of parabens and the importance of the phenolic hydroxyl group in a para position for ERRγ agonists (Fig. 4). Recent studies have shown that parabens are commonly detectable in human breast cancer GW4064 weight at relatively high levels (Barr et al., 2012, Darbre et al., 2004). These findings have raised concern about the involvement of parabens in the increased incidence of human breast cancer, as all of these chemicals possess oestrogenic properties and oestrogen is known to play a central role in the development, growth and progression of breast cancer (Miller, 1996). However, it has also been argued that the effect of parabens may not be important (Golden and Gandy, 2004, Golden et al., 2005) because the most widely used parabens have been found to be weakly oestrogenic, with the most potent (BuP) being 10,000 times less potent than E2 (Routledge et al., 1998). As ERRγ is expressed in most human breast cancers (Ijichi et al., 2011), there is concern over the role that parabens may play in breast cancer via ERRγ (Darbre and Harvey, 2008). By converting the molar concentration into ppb (ng/g), we found that the LOEL values of parabens (10M≈15ppb for MP) observed from the ERRγ coactivator recruiting assay were lower than the levels detected in breast cancer tissues. Barr et al. (2012), for instance, reported a median value of 85.5 (range 0–5134.5) ppb for five measured parabens in breast tissue samples collected from mastectomies for primary breast cancer in England between 2005 and 2008. These values suggest that the levels in most breast cancer patients are sufficient to have an effect via ERRγ. Worse yet, the levels detected in breast cancer tissues were often higher than the REC50s observed in this study. Although ERRγ is known as a constitutively active orphan nuclear receptor, the agonist may not be able to alter ERRγ function. However, Yamamoto et al. (2012) showed recently that a selective ERRγ agonist, DY131, promoted the growth of ERα-negative uterine endometrial cancer cells. ERRγ is known as a modulator of oestrogen signalling in breast cancer cells, and shares target genes, coregulatory proteins, ligands and sites of action with ERs (Giguere, 2002). Ijichi et al. (2011) showed that ERRγ can stimulate oestrogen response element (ERE)-driven transcription in MCF-7 cells and 293T cells, and that exogenously transfected ERRγ can increase MCF-7 cell proliferation. Charles and Darbre (2013) very recently investigated the extent to which proliferation of MCF-7 cells could be increased by exposure to MP, EP, PP, BuP and isobutyl paraben at concentrations in 160 human breast tissue samples reported by Barr et al., 2012, and demonstrated that parabens, either alone or in combination, were present in human breast tissue at concentrations sufficient to stimulate the proliferation of MCF-7 cells in vitro. Therefore, we believe that parabens play some role in the carcinogenesis of human breast cancers. Parabens may have significant effects in breast cancer patients who are taking tamoxifen because ERRγ, in addition to ER, is considered as a treatment target for tamoxifen (Coward et al., 2001). Additionally, although to date there has been no report on paraben levels in uterine tissue, they undoubtedly exist in most uteri due to the high levels of parabens detected in human urine, serum/plasma, seminal plasma and milk samples (Meeker et al., 2011, Sandanger et al., 2011, Ye et al., 2008, Ye et al., 2009). As ERRγ agonists, parabens may possess similar effects to those of DY131 on uterine endometrial cancers, and a detailed study is required.