Background
Appropriate regulation of genes is important in maintaining normal cell growth, and disruption of gene regulation is associated with human cancer. Changes in gene expression can distinguish types of breast tumors and predict response to therapies [
1‐
3]. Tremendous effort, therefore, has been devoted to dissecting pathways that regulate transcription. For example, understanding the mechanisms of gene activation by estrogen receptor-alpha (ERα) was foundational in the development of hormonal therapy [
4]. Interestingly, microarray analyses on estrogen-treated breast cancer cells show that the number of repressed genes is greater than or near the number of activated genes [
5‐
8]. Although these experiments show that estrogen-mediated repression of genes is clearly biologically important, the mechanisms responsible for repression are not fully understood. We previously showed that the Sin3A transcriptional repressor protein is a regulator of estrogen-induced repression of the ERα gene,
ESR1, in breast cancer cells [
9]. Furthermore, it was found that Sin3A and ERα exist in an endogenous estrogen-responsive complex. These data suggested that Sin3A may play a broader role in ERα-positive breast cancer cells.
The role of Sin3A in breast cancer is virtually unexplored, but studies suggest that Sin3A is important in normal growth and may be a player in other neoplastic model systems. Homozygous deletion of Sin3A in mice is embryonic lethal, demonstrating that Sin3A serves essential developmental functions [
10,
11]. Studies using conditional Sin3A knockout in mouse embryonic fibroblasts (MEFs) find that Sin3A deletion leads to decreased proliferation and increased apoptosis of cells [
10,
11]. In cancer models, Sin3A function is less clear. Lymphoma and sarcoma cell lines derived from primary tumors arising in a p53
-/- background exhibit proliferative arrest and increased apoptosis upon Cre-mediated deletion of Sin3A, suggesting that Sin3A has oncogenic functions [
11]. However, another report suggests that Sin3A functions as a tumor suppressor in non-small cell lung cancer (NSCLC), as down-regulation of Sin3A mRNA occurs in several cases of NSCLC [
12]. These few reports with disparate findings highlight a fundamental lack of understanding of the role of Sin3A in growth and cancer.
At the molecular level, Sin3A functions as the scaffolding component of the multi-protein Sin3 repressor complex that mediates transcriptional repression of several genes. The Sin3 complex was identified in yeast but is conserved in species through mammals [
13,
14]. The characteristic catalytic activity associated with Sin3A is histone deacetylation via its interactions with HDAC1/2 [
15,
16]. Additional components of the complex consist of SAP18/30, which stabilize the Sin3A-HDAC interaction, and RbAp46/48, which anchor the Sin3 complex on nucleosomes [
15‐
17]. Sin3A does not possess intrinsic DNA-binding activity, so it must be targeted via interaction with other DNA-bound factors. Interactions for numerous DNA-binding factors and Sin3A have been identified, including Mad, p53, MeCP2, NRSF, CTCF, and ERα as examples [
9,
18‐
22]. Sin3A can also interact with other enzymatic proteins, including those capable of histone methylation, DNA methylation, chromatin remodeling, and N-acetylglucoseamine transferase activity [
20,
23‐
28].
In this report, we expand our previous findings and identify the function of Sin3A in gene expression, survival, and growth of breast cancer cells. Gene expression analysis identified a specific subset of Sin3A-responsive genes that were regulated by both HDAC1/2-dependent and -independent mechanisms. Importantly, decreased Sin3A expression led to an increase in apoptosis and increased expression of several apoptotic genes, which translated into attenuation of cell growth of ERα-positive and not ERα-negative breast cancer cells. This study identifies Sin3A as an essential regulator of growth and survival of ERα-positive breast cancer cells, which may have important translational implications for breast cancer patients.
Discussion
Previous studies had suggested a role for Sin3A in growth of normal and neoplastic cells, but the function of Sin3A in breast cancer had not been fully explored [
10,
11]. Prior research from our lab identified Sin3A as a regulator of ERα gene,
ESR1, expression and found an estrogen-responsive interaction between ERα and Sin3A [
9]. This led us to further determine Sin3A regulation of gene expression and growth in breast cancer cells. We find that Sin3A regulates a subset of genes in ERα-positive MCF7 cells through both HDAC1/2-dependent and -independent activities. Maximum growth and survival of ERα-positive MCF7 and T47D cells requires expression of Sin3A. Interestingly, we also find that estrogen causes an increase in Sin3A protein levels in ERα-positive cells, suggesting the involvement of Sin3A in a feedback circuit regulating estrogen-dependent growth of breast cancer cells. Further, Sin3A represses important apoptotic genes in ERα-positive cell lines, consistent with our finding that decreased Sin3A levels leads to cellular apoptosis.
This study identifies the transcriptional repressor, Sin3A, as a necessary survival factor in ERα-positive breast cancer cells. Our data further support the idea that Sin3A promotes growth and survival of cells proposed in previous studies [
10,
11]. Together, these results raise the intriguing possibility that gene repression is as important of a determinant for cell growth as gene activation, as Sin3A primarily functions as a repressor. Other chromatin modifying repressor proteins, including the MTA components of the Mi-2/NuRD complex and EZH2, are also associated with breast cancer growth and progression [
32‐
35]. Our identification of Sin3A as a prosurvival factor is further interesting in that it highlights the importance of estrogen-mediated survival of breast cancer cells. Sin3A knockdown increased apoptosis but had no effect on the cell cycle (Figure
3). Estrogen is commonly viewed as a mitogenic agent that increases growth of breast cancer cells through cell cycle progression, but our data support the notion that estrogen-mediated repression of apoptosis also has a large impact on overall growth of cells [
29,
36].
Clinical trials for HDAC inhibitors in breast cancer treatment, such as vorinostat, are still in early phases and often involve patients with advanced disease [
37‐
40]. These studies have seen only partial efficacy that often increases when in combination with other agents, such as chemotherapy, and there are often issues of toxicity [
37‐
40]. Our data suggest that developing therapeutic agents to target the scaffolding component of HDAC-complexes, such as Sin3A, may be of value, particularly because Sin3A affected only a subset of genes, but its loss still caused cell death. An agent that could disrupt Sin3A would target both its HDAC-dependent and -independent activities, possibly expanding the efficacy beyond that of HDAC inhibitors. Other reports support the finding that Sin3A has both HDAC1/2-dependent and -independent capabilities. For example, in stem cells, Sin3A is the key member of the Sin3 complex involved in the regulation of
NANOG gene expression, not HDAC1/2 [
41]. Several studies have also shown that Sin3A can interact with histone methylases (Smyd2, Set1/Ash2, and ESET), DNA methylation proteins (MeCP2), chromatin remodeling enzymes (ISWI, Brg1, hBrm, and BAF155), and O-linked N-acetylglucosamine transferase (OGT), demonstrating that Sin3A has the potential to serve as an integrator of broad transcriptional and epigenetic changes in cells [
20,
23‐
28]. Furthermore,
in vitro transcription reactions on reconstituted nucleosomal templates find that addition of the HDAC inhibitor, trichostatin A (TSA), abolishes Sin3A-mediated repression of an acetylated histone H3 template, but not acetylated histone H4 template [
42]. This
in vitro experiment shows that Sin3A, even in the absence of other repressor molecules and enzymatic proteins, possesses some intrinsic HDAC1/2-independent capabilities.
Our data show that loss of Sin3A increases apoptosis of ERα-positive MCF7 cells (Figure
3). Upon further mechanistic experiments, we find that several genes with known roles in apoptosis are increased with Sin3A knockdown. This suggests that Sin3A normally represses their expression in MCF7 cells to aide in preventing apoptosis, and subsequently, promote cell growth. The apoptotic gene targets we identified fall into both the extrinsic death receptor and intrinsic mitochondrial apoptotic signaling pathways [
31]. Interestingly, we find that Sin3A regulates genes involved in all steps of the extrinsic pathway in MCF7 cells - ligands, death receptors, adaptors, and caspases (Figure
5A and Additional File
2A) [
43]. Specifically, levels of the
TRAIL ligand, and its receptor,
TRAILR1, are increased in MCF7 cells with Sin3A siRNA.
TRAIL is a member of the tumor necrosis factor (TNF) superfamily of cytokines which can induce apoptosis by binding to extracellular domains of one of its receptors, which includes
TRAILR1, a member of the TNF receptor superfamily [
44‐
46]. Death receptors further use interactions with intracellular adaptor proteins to mediate signals from the extracellular environment, and loss of Sin3A increased levels of both
TRADD and
TRAF4 adaptors [
47,
48]. Lastly, we identify
CASP10 as a Sin3A-responsive gene. In addition to caspase 8, caspase 10 has been shown to act as an initiator caspase in the death receptor signaling pathway, which can lead to activation of downstream executioner caspases to cause apoptosis [
49].
Three genes connected to the intrinsic mitochondrial apoptotic-inducing pathway are also regulated by Sin3A in MCF7 cells -
APAF1,
CASP9, and
BNIP3L (Figure
5A and Additional File
2A). The involvement and connection of Apaf-1 and caspase 9 in stress-induced apoptosis has been studied in great detail. Briefly, cellular stress stimulates release of cytochrome c from the mitochondria where it can then bind to Apaf-1, inducing conformational changes, ATP hydrolysis, and multimerization of Apaf-1 [
50]. The complex, referred to as the "apoptosome", then recruits and activates procaspase 9, and active caspase 9 can cleave executioner caspases to cause apoptosis [
51,
52]. Finally,
BNIP3L (also known as
NIX) is a proapoptotic member of the Bcl-2 family of proteins that function upstream of the apoptosome to regulate the release of cytochrome c from the mitochondria [
53,
54]. Our findings that Sin3A regulates key genes from both the death receptor and mitochondrial stress apoptotic-inducing pathways emphasize the importance of this transcriptional repressor.
Our data find that Sin3A differentially regulates the expression of the apoptotic genes discussed above in ERα-positive (MCF7) and ERα-negative (MDA-MB-231) cell lines.
TRAIL,
TRAILR1,
TRAF4,
CASP10, and
APAF1 increase upon Sin3A knockdown in MCF7, but not MDA-MB-231, breast cancer cells (Figure
5). Three genes,
TRADD,
BNIP3L, and
CASP9, increase in both cell lines with loss of Sin3A, demonstrating that Sin3A possesses some overlapping gene regulation between breast cancer cell lines, as may be expected (Additional File
2). However, it is of note that the increase seen in these three genes upon Sin3A knockdown is greater in the MCF7 cells. Differences in apoptotic gene regulation by Sin3A in ERα-subtypes can mechanistically explain the discrepancies seen in effects of Sin3A on cell growth. Induction of apoptotic genes in the ERα-positive MCF7, and not ERα-negative MDA-MB-231 cells, could lead to increases in apoptosis and a resulting decrease in cell growth, as we observe. Furthermore, Sin3A protein itself is increased by estrogen in the ERα-positive breast cancer cell lines, discussed below.
To our knowledge, this is one of the first studies to identify a regulator of Sin3A levels - estrogen. Most studies concerning Sin3A have focused on its ability to regulate expression of other genes, and little knowledge exists about how levels of Sin3A itself are modulated. Another study has shown that Sin3A can be sumoylated by TOPORS, but other modulators of Sin3A are virtually unknown [
55]. We observe an estrogen-induced increase in Sin3A protein levels that occurs independent of effects on Sin3A mRNA, demonstrating that regulation of Sin3A occurs via nongenomic actions (Additional File
1). This suggests that differences in Sin3A expression would not be found in microarray studies, possibly explaining why the role of Sin3A in breast cancer has not been appreciated until now. The mechanism by which estrogen treatment increases Sin3A protein levels is likely via a secondary effect since elevated levels are not seen before 24 hours (Figure
1A and
5C). Different ERα-positive cell types also seem to have a greater dependency on Sin3A levels for survival and growth than others. For example, we find a greater induction of Sin3A protein in MCF7 than T47D cells, and subsequently, a greater effect of Sin3A on growth of MCF7 cells (Figure
4).
While this manuscript was under revision, another group published that the Sin3 complex represses the ERα gene,
ESR1, in ERα-negative breast cancer cell lines [
56]. Consistent with this finding, we showed Sin3A regulation of estrogen-induced repression of
ESR1 in MCF7 cells [
9]. However, unlike the other publication, we did not observe reexpression of either
ESR1 mRNA (data not shown) or ERα protein (Figure
4F) in MDA-MB-231 in our current studies. These discrepancies may be due to different experimental conditions and techniques. Importantly, the authors in [
56] disrupted Sin3A and Sin3B function by using the Sin3 interaction domain (SID) of the MAD protein, while our experiments focused only on Sin3A. Additionally, the SID from MAD may participate in other protein interactions beyond the Sin3 proteins. Together, these reports suggest that components besides Sin3A are necessary to mediate the repression of
ESR1 in ERα-negative cells.
Finally, our data show several converging points between Sin3A and the estrogen signaling pathway. As described above, estrogen increases protein levels of Sin3A, suggesting a feedback loop to control estrogen-dependent growth. Our previous report shows that Sin3A controls expression of the ERα gene itself,
ESR1, and Sin3A can interact with ERα in ERα-positive breast cancer cells [
9]. We further show here that Sin3A controls expression of
NCOA2, a member of the p160 coactivator family involved in mediating ERα transcriptional activation (Figure
1D) [
57]. The estrogen-induced activation of
PGR, which encodes the progesterone receptor (PR), also increases upon Sin3A knockdown (Figure
1D). PR status is often used as a marker of estrogen sensitivity and predictor of response to endocrine therapy in breast cancer [
58,
59]. Additionally, knockdown of Sin3A only prevents growth of ERα-positive MCF7 and T47D cells, not ERα-negative MDA-MB-231 and Hs578T cells, further supporting the notion that components intrinsic to the ERα signaling pathway are involved in mediating the ability of Sin3A to promote survival.
Methods
Cell Culture and Hormone Treatments
MCF7, MDA-MB-231, and Hs578T cells were maintained at 37°C and 10% CO2 in Dulbecco's modified Eagle's medium (DMEM; Mediatech, Inc., Manassas, VA, USA) with phenol red and L-glutamine, supplemented with 10% fetal bovine serum (FBS; Biowest, Miami, FL, USA), 100 units/ml penicillin, and 100 μg/ml streptomycin (GIBCO/Invitrogen, Carlsbad, CA, USA). T47D cells were maintained at 37°C and 5% CO2 in RPMI 1640 medium with phenol red and L-glutamine (GIBCO), supplemented with 10% FBS, penicillin, and streptomycin as above. For hormone treatments, all cell lines were incubated at 37°C and 5% CO2 for at least three days in the media described above but without phenol red and containing six-times charcoal dextran stripped FBS. 17-β-estradiol (estrogen, E2; Steraloids, Inc., Newport, RI, USA) was added to a final concentration of 10 nM in all experiments for the length of time indicated in the figures. Ethanol (EtOH) vehicle control was 0.1% in all samples.
Transfection of siRNA
One day prior to transfection, cells were plated in 10 cm plates at a density of 2 × 106 cells in antibiotic free media. 800 pmol of siRNA was diluted in Lipofectamine reagent (Invitrogen) and Opti-MEM (GIBCO) and added to appropriate plates for five hours. Three days later, cells were transfected with siRNA again as above in order to achieve maximum silencing. siRNA duplexes for Sin3A, HDAC1, HDAC2, and a scrambled negative control were predesigned and purchased from Sigma (Saint Louis, MO, USA).
RNA Isolation and Quantitative RT-PCR
RNA isolation and quantitative reverse transcriptase real-time PCR (qRT-PCR) were carried out as previously detailed [
9]. Primer sequences are available upon request. Ribosomal protein P0 mRNA was used as the internal control. Relative mRNA levels were calculated using the ΔΔCt method [
60]. For initial screening of candidate Sin3A-regulated genes, two complimentary trial RT
2 Profiler Human Breast Cancer and Estrogen Receptor Signaling PCR arrays were used (SABiosciences now a Qiagen company, Frederick, MD, USA).
Western Blot Analysis
Western blot analysis was carried out as previously described [
61]. The primary antibodies used in this study were Sin3A (K-20, sc-994, Santa Cruz Biotechnology, Santa Cruz, CA, USA), ERα (VP-E613, Vector Laboratories, Burlingame, CA, USA), HDAC1 (H-11, sc-8410, Santa Cruz), HDAC2 (C-8, sc-9959, Santa Cruz), β-actin (A5441, Sigma), and α-tubulin (DM1A, CP06, Calbiochem, San Diego, CA, USA).
Cell Growth Assays
Cells were transfected with scrambled or Sin3A siRNA as detailed above, changing the media to phenol-red free media the day before the second transfection. The day after the second transfection, cells were harvested and plated in 6-well plates at a density of 4 × 105 live cells, as determined by trypan blue exclusion and counting on a hemacytometer. Cells were then treated with either 10 nM E2 or EtOH. At 24 hour intervals, cells were harvested and resuspended in media. The number of live cells at each time point was determined by hemacytometer counting and trypan blue exclusion, taking the average of two counts for each sample in each experiment.
Flow Cytometry for Cell Cycle and Apoptosis Analysis
Knockdown of Sin3A and hormone treatments were performed as described above. 72 and 96 hours post-treatment, media and cells were harvested and diluted to 1 × 105 cells in 1 ml of media. Hoescht (Invitrogen) was added at 10 μg/ml to the cells, and samples were incubated for 30 minutes at 37°C. Cells were spun down and resuspended in 100 μl of annexin binding buffer (10 mM HEPES, 140 mM NaCl, and 2.5 mM CaCl2, pH 7.4) containing Hoescht. 5 μl of the annexin V- Alexa Fluor 647 conjugate (Invitrogen) was added to the cell suspension and incubated for 15 minutes at room temperature. After the incubation, an additional 400 μl of annexin binding buffer was added, followed by propidium iodide (Sigma) to a concentration of 5 μg/ml. Dye intensities of 10,000 events were measured on the LSRII machine from BD Biosciences (San Jose, CA, USA) equipped with a UV laser. Apoptosis levels were analyzed using FlowJo software (Tree Star, Inc., Ashland, OR, USA), and cell cycle data were analyzed using ModFit software (Verity Software House, Topsham, ME, USA).
Statistical Analysis
Error bars in all figures are the standard error of the mean of a minimum of three independent experiments. Statistics were performed using OrginLab (OrginLab Corp., Northampton, MA, USA), with the exact test described in the corresponding figure legend. Data were considered significant if p < 0.05 and are indicated in the figures by asterisks.
Competing interests
The authors declare that they have no competing interests.
Authors' contributions
SE designed and performed experiments, analyzed data, and wrote the manuscript. EA assisted with experimental design, data analysis, and writing of the manuscript. All authors read and approved the final manuscript.