This article is mentioned in:

Introduction

Breast cancer is the most prevalent type of cancer in females (1). There are >200,000 new cases each year in the USA, and the number is not decreasing (2). In total, >12% of females are diagnosed with breast cancer in their lifetime. In addition to local treatments, including surgery and radiation therapy, breast cancer patients are often systematically treated with hormone therapy, chemotherapy or targeted therapy (3). The biomarkers most commonly used to guide the choice of treatment are estrogen receptor (ER), progesterone receptor (PR) and human epidermal growth factor receptor 2 (HER2). For example, ER- and/or PR-positive breast cancer can be treated with hormone therapy, including tamoxifen, which blocks the binding of estrogen with its cognate receptor. Patients with HER2-positive breast cancer benefit from targeted therapies aimed at blocking HER2 activation (4). However, basal-like triple negative breast cancer (TNBC) does not express any of the above biomarkers, and therefore, it does not respond well to existing systemic therapies and exhibits a high rate of recurrence (5).

Enhancer of zeste homolog 2 (EZH2) belongs to the polycomb group (PcG) protein family (6). Along with other PcG proteins, EZH2 forms a protein complex named polycomb repressive complex 2 (PRC2), which suppresses gene transcription (7). EZH2 is the only enzymatic component of the PRC2 complex that catalyzes histone H3 Lys27 trimethylation (H3K27me3), a histone repression mark that mediates close chromatin and epigenetic silencing of genes associated with cell proliferation, migration and invasion.

EZH2 is often overexpressed in numerous types of human cancer, including TNBC (8–10), and has been revealed to promote tumor growth by repressing the expression of tumor-suppressor genes (11–13). Mutations with increased activity of EZH2 have also been detected in various human types of cancer, including lymphomas (7). Due to aberrant activation of EZH2 in numerous cancer subtypes, EZH2 has been an attractive target for drug development. A few promising small molecule inhibitors of EZH2, including GSK126, have been developed, and their antitumor efficacy has actively been tested in pre-clinical and clinical settings (14,15).

It has been demonstrated that PRC2 components interact with histone deacetylases (HDACs), and EZH2-mediated transcriptional silencing is known to be blocked by HDAC inhibitors (10,16). H3K27me3 induces closed chromatin and gene repression, whereas histone H3 Lys27 acetylation (H3K27ac) does the opposite by promoting open chromatin and gene transcription activation (17). It has been hypothesized that HDACs remove the acetyl group from Lys27 and make it available for methylation by EZH2, and thereby EZH2 and HDAC proteins work cooperatively to mediate gene silencing by acting on the same nucleosome(s) (8).

EZH2 has been identified as an oncoprotein that is crucial to the growth of TNBC (9); therefore, the present study aimed to determine the cooperative action of EZH2 and HDACs on TNBC by treating TNBC cells with EZH2 or HDAC inhibitors alone or combined. The results of the present study demonstrated that co-treatment with both inhibitors resulted in greater antitumor effects in TNBC, compared with each drug alone.

Materials and methods

Cell lines, cell culture, chemicals and antibodies

The TNBC cell lines MDA-MB-231 and MDA-MB-436 were purchased from the American Type Culture Collection (Manassas, VA, USA) and maintained in Dulbecco's modified Eagle's medium (Thermo Fisher Scientific, Inc., Waltham, MA, USA) supplemented with 10% fetal bovine serum (Thermo Fisher Scientific, Inc.) and 100 µg/ml penicillin-streptomycin-glutamine (Thermo Fisher Scientific, Inc.) at 37°C with 5% CO2. The EZH2 inhibitor GSK126 and the HDAC inhibitor LBH589 were purchased from Selleck Chemicals (Houston, TX, USA). Dimethyl sulfoxide (DMSO) was purchased from Sigma-Aldrich; Merck KGaA (Darmstadt, Germany). The antibodies used for western blotting were: Anti-B cell lymphoma-2 like 11 (BIM; (ChemiCon International; Thermo Fisher Scientific, Inc.; catalog no. AB17003; dilution, 1:1,000), anti-poly(ADP-ribose) polymerase (PARP; Cell Signaling Technology, Inc., Danvers, MA, USA; catalog no. 9532; dilution, 1:1,000), anti-forkhead box O1 (FOXO1; Cell Signaling Technology, Inc.; catalog no. 9462L; dilution, 1:1,000), anti-protein kinase B (AKT; Cell Signaling Technology, Inc.; catalog no. 9272; dilution, 1:1,000), anti-phosphorylated (p)-AKT (Ser473 phosphorylation; Cell Signaling Technology, Inc.; catalog no. 9471L; dilution, 1:1,000), anti-p27 (Santa Cruz Biotechnology, Inc., Dallas, TX, USA; catalog no. sc-71813, 1:1,000 dilution) and anti-extracellular-signal-related kinase 2 (ERK2; Santa Cruz Biotechnology, Inc.; catalog no. sc-1647; dilution, 1:10,000). The antibodies used for chromatin immunoprecipitation (ChIP) were as follows: Anti-H3K27ac (Abcam, Cambridge, MA, USA; catalog no. ab4729; 5 µg/ChIP assay) and non-specific rabbit immunoglobulin G (IgG; Vector laboratories, Inc., Burlingame, CA, USA; catalog no. I-1000; 5 µg/ChIP assay). Horseradish peroxidase (HRP)-conjugated secondary antibodies (donkey anti-mouse IgG HRP, catalog no. NA931V, dilution 1:5,000; and donkey anti-rabbit IgG HRP, catalog no. NA934V, dilution 1:5,000) for western blotting were purchased from GE Healthcare Life-Sciences (Pittsburgh, PA, USA).

Cell morphology analysis and imaging

MDA-MB-231 and MDA-MB-436 cells (1×105 cells/well) were plated into 6-well plates at ~20% confluence. At 24 h after plating, incubated at 37°C (~40% confluence), cells were treated with vehicle (DMSO), GSK126, LBH589 or both GSK126 and LBH589 for 24 h, and morphological images were captured using a DMI3000 B microscope (magnification, ×400; Leica Microsystems GmbH, Wetzlar, Germany) from ≥5 random fields.

Analysis of apoptotic cells using fluorescence-activated cell sorting (FACS)

Apoptotic cell death (DNA content <2 copy number of unreplicated genome (N) was examined using FACS, as previously described (18). MDA-MB-231 cells were treated with vehicle (DMSO), GSK126, LBH589 or GSK126 and LBH589 for 24 h at 37°C. Cells were then harvested by centrifugation at 1,000 × g for 5 min. Cells were washed with 1X PBS once, fixated with 70% ethanol on ice for 30 min and stored at −20°C. Following washing twice with 1X PBS, cells were stained with propidium iodide (PI) in a staining solution, which was supplemented with 20 µg/ml PI (Sigma-Aldrich; Merck KGaA) and 50 µg/ml RNase A (Thermo Fisher Scientific, Inc.), for 30 min at room temperature, and their DNA content profiles were then determined by flow cytometry using FACScan (BD Biosciences, Franklin Lakes, NJ, USA) and apoptotic cells (PI intensity, <2 N) were determined using CellQuest Pro software (version 5.1; BD Biosciences).

Western blot analysis

MDA-MB-231 and MDA-MB-436 cells (1.5×105 cells/well) were plated into 6-cm dishes at ~30% confluence. At 24 h after plating, incubated at 37°C (~50% confluence), cells were treated with vehicle (DMSO), GSK126, LBH589 or both GSK126 and LBH589 for 24 h at 37°C. Cells were then harvested and lysed for 1 h on ice in modified radioimmunoprecipitation assay buffer (1X PBS, 1% Nonidet P-40, 0.1% SDS and 1X protease inhibitor cocktail; Sigma-Aldrich). The protein concentration in the samples was determined using a bicinchoninic acid assay (Thermo Fisher Scientific, Inc.). Equal amounts (80–100 µg protein) of samples were separated using SDS-PAGE (6–10% gels) and transferred onto nitrocellulose membranes. Membranes were pre-blocked with 5% non-fat dry milk (Bio-Rad Laboratories, Inc., Hercules, CA, USA) in 1X TBS for 1 h at room temperature and subsequently incubated overnight at 4°C with primary antibodies against PARP, BIM, FOXO1, (p)-AKT, AKT or p27. ERK2 was used as a loading control (12). Membranes were washed three times with 1X TBS containing 0.1% Tween-20 and incubated with HRP-conjugated secondary antibodies, goat anti-mouse IgG or anti-rabbit IgG as aforementioned, for 1 h at room temperature. Proteins were detected by chemiluminescence (Thermo Fisher Scientific, Inc.).

RNA extraction and reverse transcription-quantitative polymerase chain reaction (RT-qPCR)

MDA-MB-231 and MDA-MB-436 cells (2×105 cells/well) were plated into 6-cm dishes. At 24 h after plating, and incubated at 37°C (~50% confluence), cells were treated with vehicle (DMSO), GSK126, LBH589 or both GSK126 and LBH589 for 24 h at 37°C. Cells were lysed with TRIzol® reagent (Thermo Fisher Scientific, Inc.) for 10 min at room temperature, and total RNA was extracted following the manufacturer's protocol. cDNA was synthesized using SuperScript II reverse transcriptase (Thermo Fisher Scientific, Inc.) according to the manufacturer's protocol. RT-qPCR was performed using iQ SYBR® Green Supermix (Bio-Rad Laboratories, Inc.) and an iCycler iQ™ Real-Time PCR Detection System (Bio-Rad Laboratories, Inc.), according to the manufacturer's protocol. The PCR cycles (denaturing at 95°C for 20 sec, annealing at 58°C for 20 sec and extending at 72°C for 25 sec) for BIM and GADPDH genes were between 26 and 30 and between 17 and 19, respectively, in different cell lines and different treatment conditions. The experiments were repeated three times. All the signals were normalized by GAPDH, and the 2−ΔΔCq method (19) was used to determine the fold changes in expression of mRNAs. The sequences of the primers used for RT-qPCR were as follows: BIM forward, 5′-AGACAGAGCCACAAGCTTCC-3′ and reverse, 5′-CAGGCGGACAATGTAACGTA-3′; and GAPDH forward, 5′-ACCCACTCCTCCACCTTTGAC-3′ and reverse, 5′-TGTTGCTGTAGCCAAATTCGTT-3′.

Analysis of public chromatin immunoprecipitation sequencing (ChIP-seq) data

ChIP-seq signals for the promoter histone mark histone H3 Lys4 trimethylation (H3K4me3), the enhancer histone mark histone H3 Lys4 monomethylation (H3K4me1) and transcriptionally active histone mark histone H3 Lys27 acetylation (H3K27ac), obtained from LNCaP prostate cancer cells (20), were analyzed and displayed using the University of California at Santa Cruz genome browser (genome.ucsc.edu), as reported previously (21).

ChIP assay

MDA-MB-231 and MDA-MB-436 cells (3×106 cells/well) were cultured in 10 cm dishes, for 24 h at 37°C, and treated with vehicle (DMSO), GSK126, LBH589 or both GSK126 and LBH589, for 24 h at 37°C. Following treatment, ~5×106 cells in each treatment group were collected and sonicated using a Bioruptor (Diagenode, Inc., Denville, NJ, USA), according to the manufacturer's protocol. ChIP was performed according to a previously described protocol (22). The soluble chromatin was incubated with 5 µg of non-specific control rabbit IgG or anti-H3K27ac antibodies overnight at 4°C. Immunoprecipitated and input DNA were subjected to reverse cross-linking by incubating at 65°C overnight. Following treatment with proteinase K at 55°C for 2 h, DNA was purified using the PureLink Quick PCR Purification kit (Qiagen, Inc., Valencia, CA, USA). ChIP and input samples were analyzed using qPCR using the iQ SYBR® Green Supermix and an iCycler iQ™ Real-Time PCR Detection System (Bio-Rad Laboratories, Inc.), according to manufacturer's protocol. The 2−ΔΔCq method (19) was used to determine the enrichment of ChIP signals. The experiments were performed three times. The sequences of the PCR primers were as follows: BIM promoter forward, 5′-GCGGACGTGAGTTTCGGTGTG-3′ and reverse, 5′-GGTGCACATCTCTAAATGGGGACGG-3′; BIM enhancer-1 forward, 5′-CCCGTTTGTAAGAGGCCAGGC-3′ and reverse, 5′-CCTCACTGCTGCCTCGTGGT-3′; and BIM enhancer-2 forward, 5′-GGCTATTGGTAAAGGCTAGGTAGCG-3′ and reverse 5′-CCGGTACATGCGCTCACACAG-3′.

Statistical analysis

Experiments were performed in ≥3 replicates unless indicated otherwise. The results are presented as the mean ± standard deviation. Statistical analyses were performed by two-tailed Student's t-test. P<0.05 was considered to indicate a statistically significant difference.

Results

EZH2 and HDAC inhibitors induce morphological changes in TNBC cells

To determine the effectiveness of EZH2 and HDAC inhibitors in TNBC, the present study first examined their effects on cell morphology using microscopic analyses. MDA-MB-231 and MDA-MB-436 TNBC cells were treated with vehicle (DMSO), EZH2 inhibitor GSK126 (15 mM), HDAC inhibitor LBH589 (2.5 nM) or both and LBH589 for 24 h. Cells grown under the control condition (0.1% DMSO in complete culture medium) were spindle-like, abundant and well attached to the culture dish. Conversely, following treatment with the above inhibitors, certain cells were detached and became round, a possible indicator of apoptotic cell death (23) (Fig. 1). With the addition of GSK126 alone there was only a slight change, namely a decrease in rigidity and size, in the morphology of the cultured cells. As expected, inhibition of the EZH2 methyl-transferase alone may have little impact on the malignant cells, as it would not address the deacetylation at the H3K27 position (8). With the treatment of LBH589 alone, there was a greater number of cells that were shriveled and no longer attached to the culture dish, compared with mock-treated cells (Fig. 1). The HDAC inhibitor therefore was observed to have a greater impact on tested TNBC cells compared with that of the EZH2 inhibitor alone. Of note, it was revealed that cells treated with both inhibitors were affected the most. A large majority of cells lost their normal morphology and were not able to adhere well to the culture dish (Fig. 1). This distinctly higher level of morphological change, caused by EZH2 and HDAC inhibitors relative to vehicle, suggested an association with an increase in the rate of apoptosis.

Figure 1. EZH2 and HDAC inhibitors induce morphological changes in MDA-MB231 and MDA-MB-436 cells. (A) MDA-MB-231 and (B) MDA-MB-436 cells were treated with GSK126 (15 µM), LBH589 (2.5 nM) or both for 24 h. Cells growing under normal conditions (DMSO) were spindle-like and attached well to the culture dish. Conversely, following treatment with these inhibitors, certain cells were detached and became round, which is indicative of apoptotic cell death. Scale bars, 100 µm. EZH2, enhancer of zeste 2 polycomb repressive complex 2 subunit; HDAC, histone deacetylase; DMSO, dimethyl sulfoxide; GSK, GSK126; LBH, LBH589.

Inhibition of EZH2 and HDAC proteins induces apoptotic cell death

To validate the morphological observation, FACS analysis was performed on the cultures of both TNBC cell lines with or without treatment. The cells were stained with PI and analyzed by flow cytometry to determine the amount of DNA in the cells. Cells with a PI intensity <2N were considered apoptotic, as the DNA was fragmented (24). Consistent with the morphological images, there were increasingly higher amounts of fragmented DNA, and thus apoptotic cells, in the cultures treated with GSK126, LBH589 or both, compared with vehicle (Fig. 2A), which is quantitatively presented in Fig. 2B. Specifically, there was a significant increase in the percentage of dead or dying cells in the culture treated with GSK126 and LBH589, compared with mock (DMSO) treatment. Cells treated with the vehicle (DMSO), GSK126 or LBH589 alone exhibited cell death rates of 6.00, 11.57 and 15.52%, respectively. Cells treated with GSK126 and LBH589 demonstrated a 40.31% apoptotic rate, which was over double the apoptotic level in the culture treated with LBH589 alone (Fig. 2B). The results of simultaneous treatment of GSK126 and LBH589 corroborate the hypothesis that inhibiting methylation by EZH2 and deacetylation by HDACs may induce an additive/synergistic effect in cancer therapy.

Figure 2. EZH2 and HDAC inhibitors induce apoptotic death in triple negative breast cancer cells. (A) MDA-MB-231 cells were treated with vehicle (DMSO), GSK126, LBH589 or both GSK126 and LBH589 for 24 h. Cells were harvested and stained with PI, followed by fluorescence-activated cell sorting analysis. Apoptotic cells (PI intensity <2N) were determined by CellQuest software (version 5.1; BD Biosciences, Franklin Lakes, NJ, USA). (B) The percentage of apoptotic cells in the indicated groups was determined from two replicates. Data are represented as the mean ± standard deviation (n=2). *P<0.05, **P<0.01, ***P<0.001 compared with the control (DMSO) group. EZH2, enhancer of zeste 2 polycomb repressive complex 2 subunit; HDAC, histone deacetylase; DMSO, dimethyl sulfoxide; PI, propidium iodide; N, copy number of unreplicated genome; GSK, GSK126; LBH, LBH589.

Increase in BIM protein expression level induces high levels of apoptosis

As another method to determine apoptosis, the present study evaluated the level of PARP cleavage, an apoptotic marker (25), and a key pro-apoptotic protein, BIM (encoded by a gene termed BCL2L11) (26), using western blot analysis. Full-length PARP is a protein involved in DNA repair and chromatin structure formation. During apoptosis, this full-length protein is cleaved into small fragments that are detected as cleaved PARP (25). PARP protein has a unique fragmentation as a result of apoptosis, which allows for convenient detection of the levels of programmed cell death (25). In both MDA-MB-231 and MDA-MB-436 cells treated with either GSK126 or LBH589, cleaved PARP expression level increased inversely to the expression level of uncleaved full-length PARP (Fig. 3). Consistent with the hypothesis of the present study, the combined treatment produced the highest levels of cleaved PARP, therefore resulting in the highest levels of apoptosis detected by FACS analysis (Fig. 2).

Figure 3. Treatment of triple negative breast cancer cells with EZH2 and HDAC2 inhibitors induces PARP cleavage and BIM protein expression. (A) MDA-MB-231 and (B) MDA-MB-436 cells were treated with vehicle (DMSO), GSK126, LBH589 or both GSK126 and LBH589 as indicated for 24 h. Cells were then harvested for western blot analysis using the indicated antibodies. ERK2 expression was used as a protein loading control. Similar results were obtained in two independent experiments. EZH2, enhancer of zeste 2 polycomb repressive complex 2 subunit; HDAC, histone deacetylase; PARP, poly (ADP-ribose) polymerase; BIM, B cell lymphoma-2 like 11; DMSO, dimethyl sulfoxide; ERK2, extracellular signal-related kinase 2; IB, immunoblot.

The reason for the increased apoptotic levels may be attributed to the overexpression of pro-apoptotic proteins. One such protein, BIM, often serves a key role in apoptosis in all cancer cells (26). Western blotting revealed that treating the cancer cells with GSK126 induced a slight increase in BIM expression. However, following treatment with LBH589, BIM protein expression levels were higher in both cell lines, compared with mock treatment (Fig. 3A and B). Thus, the significant increase in apoptotic levels evident in the cell count analysis (Fig. 2) was in agreement with the immunoblotting results (Fig. 3). The culture treated with both drugs demonstrated the highest levels of BIM expression, again displaying the effectiveness of both inhibitors on malignant cells (Fig. 3).

BIM mRNA expression levels are upregulated by inhibition of EZH2 and HDAC proteins

It was revealed that treatment with EZH2 and HDAC inhibitors induced increased levels of apoptosis in TNBC cells, as demonstrated by the increased levels of BIM protein expression. However, it was unclear whether the promotion of the pro-apoptotic protein and the increased levels of cell death were due to increased mRNA expression of BIM. Thus, the effect of GSK126 and LBH589 on the expression levels of BIM mRNA was determined by RT-qPCR. Quantitation of the steady-state mRNA expression levels confirmed that EZH2 and HDAC inhibitors upregulated the mRNA expression levels of BIM in the MDA-MB-231 and MDA-MB-436 cell lines (Fig. 4). Consistent with the results of the FACS analysis, treatment with GSK126 induced slight increases in BIM mRNA expression levels. Also similar to the results presented in Fig. 3, LBH589 induced the second-highest expression levels of BIM mRNA (Fig. 4). Finally, addition of both inhibitors consistently resulted in the highest level of BIM mRNA expression, which was associated with the level of apoptotic cell death (Fig. 4). Since both inhibitors were successful in inducing BIM mRNA expression, it was suggested that the expression of BIM mRNA may be regulated by EZH2 and HDACs in TNBC cells.

Figure 4. EZH2 and HDAC inhibitors induce BIM mRNA expression in triple negative breast cancer cells. (A) MDA-MB-231 and (B) MDA-MB-436 cells were treated with vehicle (DMSO), GSK126, LBH589 or both GSK126 and LBH589 as indicated for 24 h. Cells were then harvested, and BIM mRNA expression was analyzed by reverse transcription-quantitative polymerase chain reaction, using GAPDH as an internal control. All data are presented as the mean ± standard deviation (error bar) from three replicates. **P<0.01, ***P<0.001 compared with the control (DMSO) group. EZH2, enhancer of zeste 2 polycomb repressive complex 2 subunit; HDAC, histone deacetylase; BIM, B cell lymphoma-2 like 11; DMSO, dimethyl sulfoxide; GSK, GSK126; LBH, LBH589.

BIM regulation by EZH2 and HDAC inhibitors is independent of FOXO1 expression

BIM is a well-established target gene of the transcription factor FOXO1 (27). It has been demonstrated that HDAC inhibitors can augment FOXO1 expression in phosphatase and tensin homolog (PTEN)-positive and -negative cells (28), but can only induce BIM expression by FOXO1 in PTEN-positive cancer cells (29). The expression of FOXO3, a homologue of FOXO1, is also induced by the EZH2 inhibitor GSK126 in breast cancer 1-mutated breast cancer cells (30). It has been shown previously that MDA-MB-231 cells are PTEN-positive and MDA-MB-436 cells are negative for PTEN (31). Also, PTEN loss induces AKT activation, which phosphorylates and inhibits FOXO1 (32). Therefore, this should result in an increase in BIM expression in MDA-MB-231 cells only if HDAC and/or EZH2 inhibitor-induced BIM expression were mediated by FOXO1 upregulation. Of note, treatment with the GSK126 and LBH589 inhibitors increased BIM expression levels in both cell lines (Figs. 3 and 4). This suggested that, whereas the expression of FOXO1 may be repressed by EZH2, FOXO1 may not cause the elevated expression level of BIM observed in the present study. To investigate whether this is the case, the present study examined FOXO1 expression levels in comparison with BIM expression levels in both cancer cell lines. BIM expression levels increased upon treatment with EZH2 or HDAC inhibitor alone or together, and there was no clear indication of higher expression levels of FOXO1 in cells treated with these inhibitors, relative to mock treatment. Conversely, when treated with GSK126, FOXO1 expression was inhibited in AKT phosphorylation-positive (MDA-MB-436) and -negative (MDA-MB-231) cell lines (Fig. 5). Treatment of LBH589 alone induced FOXO1 expression in both cell lines, and was particularly pronounced in the cells with high p-AKT levels (Fig. 5). It is worth noting that increased expression of BIM (Figs. 3 and 4) was not entirely consistent with the apoptosis levels (Fig. 2). The present study also analyzed expression of p27, a different target gene of the FOXO transcription factors (33), as another functional readout to monitor the transcriptional activities of FOXO1. However, there was no positive association between FOXO1 and p27 protein levels in PTEN-positive MDA-MB-231 and PTEN-negative MDA-MB-436 cell lines (Fig. 5). These results suggested that BIM regulation by EZH2 and HDAC inhibitors may be independent of FOXO1 expression in these TNBC cell lines.

Figure 5. Effect of EZH2 and HDAC2 inhibitors on FOXO1 protein expression in triple negative breast cancer cells. (A) MDA-MB-231 and (B) MDA-MB-436 cells were treated with vehicle (DMSO), GSK126, LBH589 or both GSK126 and LBH589 as indicated for 24 h. Cells were then harvested for western blot analysis using the indicated antibodies. ERK2 expression level was used as the protein loading control. Non-specific western blotting bands are indicated by *, a phenomenon observed in HeLa cells (http://www.cellsignal.com/products/primary-antibodies/akt-antibody/9272?N=4294956287&Ntt=akt&fromPage=plp). Similar results were obtained in two independent experiments. EZH2, enhancer of zeste 2 polycomb repressive complex 2 subunit; HDAC, histone deacetylase; DMSO, dimethyl sulfoxide; ERK2, extracellular signal-related kinase 2; AKT, protein kinase B; p-, phosphorylated; FOXO1, forkhead box O1; PTEN, phosphatase and tensin homolog.

H3K27ac in the promoter and enhancers of the BIM gene is differentially upregulated by EZH2 and HDAC inhibitors

Having ruled out the possibility of FOXO1 regulation, the present study instead hypothesized that BIM expression via EZH2 and HDAC inhibitors was directly regulated by H3K27ac levels at the promoter and/or the enhancers. Therefore, a ChIP assay was performed using anti-H3K27ac antibodies. ChIP-qPCR results demonstrated increased expression levels of H3K27ac at the BIM promoter when both cell lines were treated with GSK126 (Fig. 6A and B). Of note, LBH589 treatment alone resulted in a smaller increase in H3K27ac in the BIM promoter compared with the effect of GSK126 alone, whereas treatment with both drugs revealed no change (Fig. 6A and B), thus contradicting the increased apoptosis levels observed under these two conditions (Figs. 2 and 3). Subsequently, the present study detected an increase in H3K27ac levels at the enhancer-1 following treatment with GSK126, LBH589 or both inhibitors, and revealed that H3K27ac expression levels were highest following LBH589 treatment alone (Fig. 6A and C). At enhancer-2, the results were opposite to the effects of GSK126 or LBH589 alone or together on the BIM promoter (Fig. 6A and D).

Figure 6. EZH2 and HDAC inhibitors increase H3K27ac expression levels in the BIM gene promoter and enhancers in triple negative breast cancer cells. (A) University of California at Santa Cruz genome browser screen shots display ChIP-seq signals of H3K27ac, the enhancer histone mark H3K4me1 and the promoter histone mark H3K4me3 obtained from the public data generated from a different cancer cell type (prostate cancer LNCaP cells). Reverse transcription-quantitative polymerase chain reaction analysis of ChIP DNA using (B) primers for the promoter, (C) putative enhancer-1 and (D) enhancer-2. DNA was immunoprecipitated by control IgG or anti-H3K27ac antibody from MDA-MB-231 and MDA-MB-436 cells treated with vehicle (DMSO), GSK126, LBH 589 or both GSK126 and LBH589. Cells were harvested for ChIP assay at 24 h after treatment. *P<0.05, **P<0.005 compared with the control (DMSO) group. H3K27me1, histone H3 Lys27 methylation; H3K27me3, histone H3 Lys27 trimethylation; H3K27ac, histone H3 Lys27 acetylation; EZH2, enhancer of zeste 2 polycomb repressive complex 2 subunit; HDAC, histone deacetylase; DMSO, dimethyl sulfoxide; BIM/BCL211, B cell lymphoma-2 like 11; GSK, GSK126; LBH, LBH 589; G+L, GSK126 and LBH 589; ChIP, chromatin immunoprecipitation; IgG, immunoglobulin G.

Discussion

EZH2 has been identified as an important oncoprotein associated with the growth of TNBC cells (34). Thus, the present study hypothesized that treatment of TNBC cells with EZH2 inhibitors may result in apoptosis. Following initial treatment of cells with EZH2 inhibitor alone, notably, there was no marked effect on apoptosis, which lead to the hypothesis that the combination of EZH2 and HDAC inhibitors may have a greater effect on apoptosis, compared with each individual agent. The present study assumed that EZH2, a methyl-transferase enzyme, would cooperate with HDACs at the H3K27ac level. As their name implies, HDACs may remove the acetyl group and allow EZH2 to transfer a methyl group to Lys27 on histone H3, thereby repressing gene expression by creating a closed chromatin. Indeed, this hypothesis was supported by the observation of the present study that there was an increased pro-apoptotic effect in TNBC cells co-treated with EZH2 and HDAC inhibitors.

BIM is a pro-apoptotic protein. High expression levels of BIM were associated with the inhibition of EZH2 and HDAC proteins. FOXO1 has been reported, in certain cellular settings, to induce BIM expression, thus presenting another possible target for treatment (35). Notably, expression of FOXO family proteins is repressed by EZH2 (30). The present study hypothesized that EZH2 inhibition elevates BIM via a mechanism of increased expression of the FOXO1 transcription factor. However, despite higher expression levels of BIM, there was a lack of a corresponding increase in the FOXO1 expression level associated with inhibition of EZH2 and/or HDAC in both TNBC cell types.

The present study assumed that both inhibitors regulated BIM transcription at the promoter, but repeated experiments demonstrated that this was not the case. Instead, further data established that BIM was directly regulated by the EZH2 inhibitor at the promoter, and by the HDAC inhibitor at the enhancers. Furthermore, the death of the malignant cancer cells may be more greatly affected by regulation of the enhancers compared with that of the promoter. The apoptotic effects of treatment with the EZH2 inhibitor, regulated at the promoter, were invariably lower compared with those induced by treatment with both inhibitors, and this observation was associated with increased expression levels of H3K27ac at the BIM enhances, particularly enhancer-2.

In conclusion, the results of the present study demonstrated that treating TNBC cells with an EZH2 inhibitor (GSK126) or a HDAC inhibitor (LBH589) alone produced limited effects on cell morphology, apoptotic cell death and BIM expression levels. However, a combination of both inhibitors resulted in increased tumor-killing effects which supported the hypothesis that the combination of EZH2 and HDAC inhibitors have an increased effect on apoptosis, compared with each individual inhibitor. Additionally, the results of the present study revealed the possibility of a novel treatment strategy via additive agents.

Acknowledgements

The authors would like to thank Dr Yu Zhao (laboratory of Dr Kun Ling, the Department of Biochemistry and Molecular Biology at Mayo Clinic, Rochester, MN, USA) for his assistance with designing primers for RT-qPCR and ChIP-qPCR. The authors thank the Mayo Clinic Institutional for their support and assistance. The present study was supported in part by the National Institutes of Health/National Cancer Institute (grant no. R01CA149039).

References