Introduction
Estrogen receptors (ERs) are members of a nuclear hormone receptor superfamily. ERs exist in two isoforms, ERα and ERβ, which have highly conserved DNA binding domains and ligand binding domains [
1,
2]. Although these receptors display similar binding affinities for 17β-estradiol, they have distinct roles in the regulation of gene expression and different interactions with unique sets of transcriptional factors [
2]. Activation of ERα is considered a risk factor for the development of breast cancer, since the activation leads to cellular proliferation [
3,
4]. Cumulative data from tumor biopsies in the clinic have shown that two-thirds of breast cancers are ER-positive [
5,
6]. Tamoxifen, which regulates ERα activity, reduces the recurrence and death rate of ERα-positive breast cancer [
7]. Breast cancer patients with expression of ERα are seven to eight times more likely to benefit from selective estrogen receptor modulators such as tamoxifen than ERα-negative patients [
5]. ERα expression is therefore considered a significant outcome predictor for breast cancer patients to endocrine therapy.
The function of ERα is regulated by post-translational modifications such as phosphorylation [
8,
9], acetylation [
10,
11], sumoylation [
12], and ubiquitination [
13]. Among these modifications, acetylation is emerging as a central process in transcriptional activation of ERα [
14]. ERα is directly acetylated by p300 at lysine 302 and 303 in the absence of ligand, and its acetylation regulates transcriptional activation and ligand sensitivity [
10]. ERα is also acetylated at lysine 266 and 268 in the presence of coactivators p160 and p300, which enhances not only DNA binding but also transactivation activities. This acetylation was reversed by native cellular deacetylases, including trichostatin A (TSA)-sensitive class I and II histone deacetylases (HDACs), and nicotinamide adenine dinucleotide-dependent HDACs (class III, such as Sirt1) [
11].
Generally, TSA is known to modify the balance between histone acetyltransferase and HDAC activities that induce histone hyperacetylation and regulate gene expression. Recently, the effect of TSA in acetylation/deacetylation of nonhistone proteins has been demonstrated as a diverse regulatory event, including ubiquitination/proteasomal degradation [
15]. TSA effectively represses the mRNA and protein level of ERα in the ERα-positive breast cancer cells [
16,
17]. Although several previous studies have demonstrated the role of TSA-dependent HDACs in regulation ERα activity [
18‐
20], the precise mechanism of TSA-induced activation of ERα remains unclear. We therefore explored whether TSA induces acetylation of ERα and increases stability of ERα in the present investigation.
Discussion
TSA not only inhibits growth of ERα-positive breast cancer cells
in vitro but also inhibits breast tumor growth
in vivo [
16,
17,
30]. TSA may exert these beneficial effects against tumor growth by blocking deacetylation of histones and transcriptional factors, which subsequently alters transcriptional activity of target genes [
15,
31]. Here we report that TSA induces stability of ERα protein by enhancing acetylation and stability of p300, which may contribute to pharmacological effects of TSA.
Previous studies demonstrated that ERα is acetylated at multiple lysine residues, which may have different functions in the regulation of ERα activity. ERα was acetylated at lysine 266 and 268 in the presence of ligand in a steroid receptor coactivator-dependent manner [
11]. ERα was also acetylated at lysine 302 and 303 in the presence of p300, and thereby regulated transcriptional activation and ligand sensitivity of ERα [
10]. Ubiquitination at the same lysine residues was shown to regulate degradation of ERα [
13]. The TSA-induced acetylation of ERα was accompanied with increased protein level of ERα (Figure
1), and p300 protected ubiquitination of ERα in our investigation (Figure
4b) - supporting the hypothesis that acetylation of ERα, probably at lysine 302 and 303 residues, is important for maintaining protein stability.
This observation is similar to the p300-induced acetylation of p53 or that of Smad7, which blocks ubiquitination and degradation of the protein [
32,
33]. Other nuclear receptors such as LXRα and AR are also present as acetylated forms that are involved in transactivation and other post-translational modifications such as ubiquitination [
34,
35]. Our finding contrasts, however, with previous reports that TSA downregulated the protein and mRNA level of ERα in ERα-positive breast cancer cells [
16,
17]. When MCF7 or T47D cells were treated with TSA for a prolonged period, we also observed a similar downregulation of ERα, indicating that TSA may affect ERα activity through at least two different mechanisms: transcriptional repression of the ERα promoter, and protein stability of ERα at the post-translational level. TSA may accomplish its beneficial effects against breast cancer by inducing sequential and/or divergent modifications of ERα at different regulation levels.
p300 was originally identified as E1A, which is a transcriptional co-activator for various transcription factors, including ERα [
36]. The intrinsic histone acetyltransferase activity of p300 catalyzes acetylation of histone, which induces chromatin remodeling and subsequent transcriptional activation of target genes. p300 also acetylates nonhistone proteins such as p53 and Smad7, which leads to stabilization of target proteins [
33,
37]. Interestingly, the histone acetyltransferase domain of p300 acetylates itself [
28]. Blanco-García and colleagues demonstrated recently that PCAF was acetylated by itself and by p300. Deacetylation of PCAF was catalyzed mainly by HDAC3, which affected subcellular localization of PCAF [
38]. In the case of p300, acetylation was shown to increase transactivation activity and protein-protein interactions [
28,
29].
In the present study, we found another role for acetylation of p300 in the stabilization of p300 protein itself. Stabilization of p300 is induced within 3 hours of TSA treatment, which is similar to the TSA-induced acetylation and stabilization of ERα (Figure
2). Our finding, however, contradicts previous observations that autoacetylation of p300 did not alter its stability [
29]. We believe this conflict may be due to different experimental conditions such as the expression level of p300 and the cell lines examined. Since p300 mediates acetylation of many proteins including histone and ERα, acetylation and subsequent stabilization of p300 may regulate the pharmacological effects of TSA through activation of diverse cellular transcription factors in breast cancer cells.
Our results together with those of other studies strongly suggest that the TSA-dependent HDACs are involved in acetylation of ERα [
18‐
20]. The TSA-sensitive HDACs are class I and class II, which form a multiprotein repressor complex to remove the acetyl group from lysine residues of histones [
39]. Indeed, HDAC1 and HDAC4, which belong to class I and class II, respectively, interacted with ERα and suppressed the transcriptional activity and expression of ERα [
18,
19]. On the contrary, nicotinamide adenine dinucleotide-dependent HDACs such as Sirt1 have been demonstrated to deacetylate ERα, probably at lysine 266 and 268, which enhanced the DNA binding and transactivation of ERα [
11,
40]. It would be interesting to know the unique function of each HDAC subtype as well as of each lysine residue in the regulation of ERα activity, such as protein stability and transactivation function. Especially, resveratrol - a Sirt1 activator - caused inhibition of estrogen-dependent cell proliferation, further supporting the notion that modifying ERα acetylation strongly influences epithelial cell growth in breast tissue [
41]. Further molecular details of the acetylation of ERα and the resulting estrogen signaling could contribute to a novel therapeutic strategy against breast cancer.
Competing interests
The authors declare that they have no competing interests.
Authors' contributions
M-OL conceived of the study and its design, and interpreted the data. S-HK contributed to the study's conception and design, data collection and interpretation, and manuscript writing. H-JK and HN partly contributed to the study's design and data collection. Both authors read and approved the final manuscript.