Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2018 Jan 15.
Published in final edited form as: Cancer Res. 2016 Nov 22;77(2):470–481. doi: 10.1158/0008-5472.CAN-16-0834

Transcriptional Selectivity of Epigenetic Therapy in Cancer

Takahiro Sato 1, Matteo Cesaroni 1, Woonbok Chung 1, Shoghag Panjarian 1, Anthony Tran 1, Jozef Madzo 1, Yasuyuki Okamoto 1, Hanghang Zhang 1, Xiaowei Chen 2, Jaroslav Jelinek 1, Jean-Pierre J Issa 1
PMCID: PMC5243171  NIHMSID: NIHMS832366  PMID: 27879268

Abstract

A central challenge in the development of epigenetic cancer therapy is the ability to direct selectivity in modulating gene expression for disease-selective efficacy. To address this issue, we characterized by RNA-seq, DNA methylation and ChIP-seq analyses the epigenetic response of a set of colon, breast and leukemia cancer cell lines to small molecule inhibitors against DNA methyltransferases (DAC), histone deacetylases (Depsi), histone demethylases (KDM1A inhibitor S2101), and histone methylases (EHMT2 inhibitor UNC0638 and EZH2 inhibitor GSK343). We also characterized the effects of DAC as combined with the other compounds. Averaged over the cancer cell models employed, we found that DAC affected 8.6% of the transcriptome and that 95.4% of the genes affected were upregulated. DAC preferentially regulated genes that were silenced in cancer and that were methylated at their promoters. In contrast, Depsi affected the expression of 30.4% of the transcriptome but showed little selectivity for gene upregulation or silenced genes. S2101, UNC0638, and GSK343 affected only 2% of the transcriptome, with UNC0638 and GSK343 preferentially targeting genes marked with H3K9me2 or H3K27me3, respectively. When combined with histone methylase inhibitors, the extent of gene upregulation by DAC was extended while still maintaining selectivity for DNA methylated genes and silenced genes. However, the genes upregulated by combination treatment exhibited limited overlap, indicating the possibility of targeting distinct sets of genes based on different epigenetic therapy combinations. Overall, our results demonstrated that DNA methyltransferase inhibitors preferentially target cancer-relevant genes and can be combined with inhibitors targeting histone methylation for synergistic effects while still maintaining selectivity.

Keywords: Epigenetic Therapy, DNA Methylation, Histone Methylation, Histone Acetylation

Introduction

Cancers often display aberrant epigenetic modifications in DNA methylation or histone modifications, which can lead to silencing of critical tumor suppressor genes (1). In contrast to genetic mutations in the DNA sequence which permanently alter the gene in an irreversible fashion, genes modified by aberrant epigenetic modifications are still intact and the epigenetic state can be reversed by reactivating expression of the silenced genes (1). Supporting the promise of epigenetic therapy, low dose treatments with DNA methyltransferase (DNMT) inhibitors azacitidine (AZA) and decitabine (DAC) have been approved by the FDA for the treatment of myelodysplastic syndrome (2–4). DNMT inhibitors can have long term effects on reactivation of silenced gene expression even after removal of the drug (5,6), and demonstrate significant inhibition of cancer initiating cells in various cancers (5). DNMT inhibitors can also stimulate the immune response against cancer, which can lead to sensitization of these cancers to immunotherapy (7–9). In addition to DNMT inhibitors, the histone deacetylase (HDAC) inhibitors vorinostat (SAHA) and romidepsin (depsipeptide) have been approved by the FDA for the treatment of cutaneous T-cell lymphoma (10,11).

Despite these successes, epigenetic therapy still faces major questions about the selectivity of gene reactivation (12), since the drugs affects enzymes involved in day to day cellular regulation. A second major issue is that many patients with hematological malignancies relapse or have no initial response to epigenetic therapy (4,13,14), and patients with solid tumors have limited responses (15). Therefore, an important question is how to increase efficacy, possibly by combining therapies that target different epigenetic marks. Previous studies have demonstrated that combination treatments with DNMT and HDAC inhibitors have synergy in activating expression of silenced tumor suppressor genes (16,17). However, this combination has demonstrated disappointing results in clinical trials (18–21), indicating the need for additional strategies of combining epigenetic therapies.

Recent progress has led to the promising development of inhibitors targeting enzymes that regulate histone methylation. These include inhibitors for KDM1A (22,23), a histone demethylase that catalyzes removal of H3K4 mono and di methylation, EHMT2 (24,25), a histone methyltransferase that catalyzes mono and di methylation of H3K9, and EZH2 (26–29), a histone methyltransferase that catalyzes mono, di, and tri methylation of H3K27. All three enzymes have been shown to play critical roles in cancer (23,24,27,30). Importantly, the efficacy and selectivity of combining DNMT inhibitors with these histone methylation inhibitors (inhibitors that affect histone methylation) has yet to be fully addressed. In this study, we report that DNMT inhibitors have a high degree of selectivity for cancer relevant genes, HDAC inhibitors are much more non-selective, and combining DNMT with histone methylation inhibitors leads to synergistic induction of gene expression while maintaining selectivity.

Materials and Methods

Cell Culture

YB5, MCF7, and HL60 cells were cultured in L-15 medium with 10% FBS, DMEM with 10% FBS, and IMDM with 20% FBS, respectively. Cell lines were obtained from ATCC (2005–2007) and were authenticated at MD Anderson Cancer Center genomic core facility by DNA fingerprinting in 2011. Cells were treated daily with 100 nM decitabine (5-aza-CdR, Sigma-Aldrich), 10 µM S2101 (Millipore), 1 µM UNC0638 (Sigma-Aldrich), and 1 µM GSK343 (Sigma-Aldrich) for 96 hours, or 20 nM depsipeptide (Romidepsin, Sigma-Aldrich) was added for 24 hours. Primary normal breast epithelial cells were isolated as and cultured as described previously (31). RNA isolation, DNA isolation, qPCR, western blot, and pyrosequencing were carried out as described previously (6). Primer sequences are listed in Table S3.

RNA-seq

RNA was isolated using Rneasy Mini Kit (Qiagen) from experiments with biological triplicates. Strand-specific RNA libraries were generated from 1 µg of RNA using TruSeq stranded total RNA with Ribo-Zero Gold (Illumina). Sequencing was performed using single end reads (50 bp, average 50 million reads per sample) on the HiSeq2500 platform (Illumina). Sequenced reads were aligned to the hg19 genome assembly using TopHat2 (32). The expression level and fold change of each treatment group was evaluated using EdgeR (33). Genes that had 0 reads across all samples were excluded. Ingenuity Pathway Analysis (Ingenuity® Inc, Redwood city, CA, USA, http://www.ingenuity.com) was used to determine functional enrichment for “Diseases and Biofunctions”.

ChIP-seq

Chromatin immunoprecipitation (ChIP) was performed as described previously (6) using antibodies listed in Table S3. Biological duplicates were performed for each antibody. Indexed libraries were prepared using NEB Next ChIP-Seq Library Prep (Illumina), then sequenced using single end reads (50 bp, average 30 million reads per sample) on the HiSeq2500 platform (Illumina). Sequenced reads were aligned to the hg19 genome assembly using Bowtie2 (34). Significant peaks for H3K4me2 were obtained using MACS2 (compared to input, no model, q<0.01) (35). Significant peaks for H3K9me2 and H3K27me3 marks were obtained using SICER (compared to input, Window=200, Gap=600, q<0.01) (36). We associated genes with histone marks by defining whether each gene had significant H3K4me2 peaks present in the promoter (−1kb to +500bp of TSS), or H3K9me2 and H3K27me3 peaks present in the promoter or the gene body. ChIP-seq enrichment plots over the gene body were generated using NGS plot (37).

Global DNA methylation analysis

Genome-wide DNA methylation analysis was carried out using digital restriction enzyme analysis of methylation (DREAM) as described previously (38). To identify genes with promoter DNA methylation, we limited the analysis to genes that had at least 10 sequencing reads per CpG site and methylation values in the promoter region (−1000bp to +500bp).

Results

Genome wide effects of epigenetic therapy

In order to study the selectivity of epigenetic therapy, we performed RNA-seq on colon cancer (YB5 (39)), breast cancer (MCF7), and myeloid leukemia cells (HL60) treated with epigenetic inhibitors that target DNMTs (100 nM decitabine for YB5 and HL60, 1 µM for MCF7), KDM1A (10 µM S2101) (22), EHMT2 (1 µM UNC0638) (24), EZH2 (1 µM GSK343) (26), and HDACs (20nM depsipeptide). We performed each inhibitor treatment in triplicate and used the highest non-toxic doses of the drugs when given daily for four consecutive days, except for the HDAC inhibitor desipeptide (Depsi), which was given for 24h. qPCR experiments on MCF7 cells treated with varying doses of DAC showed that 100 nM DAC resulted in small to no induction of silenced tumor suppressor genes while 1 µM DAC led to reactivation, so the slightly higher dose of DAC was used for MCF7 (Figure S1A). The same doses of UNC0638 and GSK343 were able to repress their target histone methylation marks in all three cell lines (Figure S1B–D). S2101 did not affect global levels of H3K4me2 (Figure S1B–D), but this is consistent with previous reports (23), so the highest non-toxic dose was used.

DAC and Depsi led to the highest number of genes being significantly regulated in YB5, MCF7, and HL60 (Fold Change (FC) > 2, False Discovery Rate (FDR) < 0.1, Fig. 1A). When combining the data from all three cell lines, DAC modulated 8.6% of the transcriptome and 95.4% of the effect was on gene upregulation (Fig. 1B). By contrast, Depsi modulated 30.4% of the transcriptome and only 62.8% of the effect was on gene upregulation (Fig. 1B). The histone methylation inhibitors had minimal effects on gene expression (S2101 affects 1.7%, UNC0638 affects 1.4%, and GSK343 affects 1.2% of transcriptome) (Fig. 1B), and had limited overlap (Figure S1E). The epigenetic inhibitors regulated different genes in each of the three cell lines (DAC regulated genes had 3.4% overlap and the histone methylation inhibitors had close to 0%), supporting the selectivity of epigenetic therapy, with the exception of Depsi regulated genes, which had a slightly higher gene overlap (6.4%) (Fig. 1C).

Figure 1. Genome wide effects of epigenetic therapy.

Figure 1

Open in a new tab

A) Number of upregulated and downregulated genes by each inhibitor in YB5, MCF7, and HL60 cells (FC > 2, FDR < 0.1). B) Percentage of the transcriptome regulated by each inhibitor when combining data from all three cell lines. C) Venn diagram showing the overlap of regulated genes between the three cell lines. D) Distribution of RPKM values of normal colon (NC), YB5, normal breast epithelial cells (NBE), MCF7, promyelocytes (PME), and HL60, in all genes and genes regulated by each inhibitor. E) Average RPKM of normal (NC, NBE, and PME) vs. cancer (YB5, MCF7, HL60) of all genes and genes regulated by each inhibitor. * indicates p < 0.05 by Student’s t-test. F) Number of genes regulated by each inhibitor (FC > 2, FDR < 0.1) that belong in the categories of genes repressed in cancer and upregulated by inhibitor (RPKM at least 2 fold less in cancer compared to normal), and other (any other regulated genes that do not fall into the first category).

We then analyzed whether the inhibitors were targeting genes that were expressed in normal tissue and silenced in cancer (Fig. 1D and 1E). For this purpose, we generated RNA-seq data from normal breast epithelial cells (NBE) grown briefly in culture (GSE74417), and downloaded published RNA-seq data from normal colon (NC, GSM835561 (40)), and promyelocytes (PME, GSM1704844 (41)). Genes regulated by DAC had lower expression in cancer (YB5, MCF7, HL60) compared to its expression in normal tissue (NC, NBE, PME) (Student’s t-test, p < 0.05), while genes regulated by Depsi did not show this decrease (Student’s t-test, p > 0.05) (Fig. 1E). The histone methylation inhibitors regulated genes silenced in cancer similar to DAC (Student’s t-test, p < 0.05) (Fig. 1E).

To analyze this further, we classified the genes regulated by the epigenetic inhibitors into specific and nonspecific categories. For comparative purposes, specific effects were defined as genes expressed in normal and repressed in cancer (> 2 fold decrease in cancer compared to normal, 7007 genes in YB5, 5218 genes in MCF7, and 4322 genes in HL60) that are upregulated by inhibitor, and nonspecific effects were defined as any other gene regulation induced by the inhibitor, though some of these genes may be downstream effects of specific gene induction (Fig. 1F). DAC led to similar number of specific and nonspecific gene regulation (average of 816 specific genes and 1004 nonspecific genes), while Depsi had a much larger effect on regulating nonspecific genes (average of 4643 genes) compared to specific genes (1790 genes) (Fig. 1G). Histone methylation inhibitors showed similar specificity as DAC, but the number of overall genes regulated is very low (Fig. 1F and G).

Pathway analysis using Ingenuity Pathway Analysis (IPA) on “Diseases and Biofunctions” showed that the top pathway for DAC regulated genes in all three cell lines was “Cancer” (Table S1). Collectively, these results demonstrate that DAC is selective for upregulating genes silenced in cancer with minimal effects on gene downregulation, in contrast to Depsi which has equal effects on gene up and downregulation as well as on genes that do not change expression in cancer.

Baseline DNA methylation of genes regulated by epigenetic therapy

To study the baseline DNA methylation of genes affected by epigenetic therapy, we used methylation values generated using digital restriction enzyme analysis of methylation (DREAM) (38) on SW48 cells and normal colon samples (NC) (GSE66296, (42)). We also generated new DREAM data on MCF7 (GSE74910), NBE (GSE74910), HL60 (GSE78103), and normal white blood cells (WBC) (GSE75274). To evaluate correlation with gene expression, we focused on promoter DNA methylation values (Fig. 2A). Cancer samples (YB5, MCF7, and HL60) had higher mean methylation values than the corresponding normal tissues (NC, NBE, WBC, respectively), as expected (Fig. 2A). Genes upregulated by DAC were highly methylated, although the extent of the methylation of the target genes varied by cell line (Fig. 2A). When combining the data from all three cell lines, 39.2% of DAC target genes had methylation of over 50%, and 60.8% had methylation values of over 10% (Fig. 2B). In addition, DAC had a larger effect on regulating genes with promoter DNA methylation (10.2% of methylated genes upregulated in YB5, 27.2% in MCF7, 19.3% in HL60) compared to genes with no or low methylation (0.8% of unmethylated genes upregulated in YB5, 7.1% in MCF7, and 10.3% in HL60) in all three cell lines (Chi-square test, p < 0.05) (Fig. 2C and 2D). There was no preference for gene body methylation on genes upregulated (Fig. S2A) or downregulated (Fig. S2B) by DAC in YB5 cells.

Figure 2. DNA methylation of genes regulated by epigenetic therapy.

Figure 2

Open in a new tab

A) Distribution of DNA methylation of genes regulated by epigenetic therapy with DNA methylation values in the gene promoter (−1000bp to +500bp of TSS) in YB5 cells (total of 8564 genes), MCF7 cells (total of 7019 genes), and HL60 cells (total of 8263 genes). Promoter methylation values of normal colon (NC, total of 8117 genes), normal breast epithelial cells (NBE, total of 6303 genes), and white blood cells (WBC, 6559 genes) are shown as comparison. B) Percentage of genes regulated by each inhibitor when combining data from all three cell lines that have low (0–10%), moderate (10–50%) or high (50–100%) promoter DNA methylation. C) Percentage of genes with high promoter DNA methylation (50–100%) and low promoter DNA methylation (0–50%) that are regulated by each inhibitor. * indicates p < 0.05 in a Chi-square test. D) Percentage of each category regulated by inhibitor when averaging the data from all three cell lines.

Surprisingly, genes regulated by UNC0638 also had high promoter DNA methylation (Fig. 2A–D). We therefore tested whether UNC0638 could be inducing DNA demethylation but found no effect on DNA methylation levels of two of its most upregulated genes (Fig. S2C). By contrast, genes regulated by Depsi, S2101, or GSK343 did not show higher promoter DNA methylation (Fig. 2A and B), and did not have a larger effect on regulating genes with promoter DNA methylation in all three cell lines (Chi-square test, p > 0.05 except for YB5) (Fig. 2C and D). The ability of Depsi to have some effect on methylated genes is consistent with a previously published study (6).

We next factored in DNA methylation in normal tissues with respect to the effects of drugs. DAC had a larger effect on genes that gain at least 20% methylation in cancer (8.9% of genes that gain methylation upregulated in YB5, 27.6% in MCF7, 16.9% in HL60) compared to genes that do not gain methylation (0.4% of unmethylated genes upregulated in YB5, 5.4% in MCF7, and 10.4% in HL60) (Fig. S2D). Collectively, these data indicate that DAC shows relatively high target selectivity for genes with promoter DNA methylation.

Synergy and specificity of combining DAC with other epigenetic therapies

We next set out to determine whether combining DNA methylation inhibitors with histone methyltransferase or demethylase inhibitors could lead to increased efficacy in reactivating epigenetically silenced genes in cancer. To study this, we initially used the YB5 cell line, which was derived from the SW48 colon cancer cell line and contains a DNA methylated and silenced CMV promoter driving GFP expression (39). Treatment of YB5 with DAC led to GFP expression as reported previously (39) (Fig. S3F for time course), while treating YB5 with S2101, UNC0638, GSK343 at multiple doses had no effect on GFP levels (Fig. S3A). DAC in combination with S2101 showed synergistic reactivation of GFP, but DAC in combination with UNC0638 or GSK343 did not show any increased GFP activation (Fig. S3A for simultaneous treatment, Fig. S3E for sequential treatment). The synergistic effects of S2101 were not due to changes in DNA methylation, as none of the inhibitors affected methylation of the CMV promoter or LINE1 elements (Fig. S3B).

Next, we studied 10 known DNA methylated and silenced tumor suppressor genes in YB5 cells that can be activated by DAC. When we tested the combination of DAC with histone methylation inhibitors, we observed that DAC+S2101 synergistically reactivated only GFP, DAC+UNC0638 synergistic effects were mainly on CDH13, and DAC+GSK343 synergistic effects were mainly on TIMP3 (Fig. S3C, S3D for MCF7). By contrast, the combination of DAC and HDAC inhibitor synergistically reactivated 7 out of 11 silenced tumor suppressor genes (Fig. S3C).

To study the synergy and selectivity of combination epigenetic therapy on a genome-wide scale, we performed RNA-seq on YB5, MCF7, and HL60 cells treated with the drug combinations. Combining DAC with S2101, UNC0638, or GSK343, led to a higher number of genes being upregulated compared to DAC alone in all three cell lines (11.6%, 10.5%, 13.2% of transcriptome affected when combining data from all three cell lines, compared to 8.6% with DAC alone), while maintaining their preference for gene upregulation (91.2%, 93.6%, 90.8% of modulated genes were upregulated, compared to 95.4% for DAC alone) (Fig. 3A and B). Combining DAC with Depsi led to changes in 40% of the transcriptome and only 68.8% of the target genes were upregulated. The combination therapies showed limited overlap in the genes regulated between the three cell lines (4.8%, 6%, and 5.5%), except for DAC+Depsi (15.4%) (Fig. 3C). Principal component analysis of the entire transcriptome of each cell line after inhibitor treatment showed that individual samples clustered together based on cell line, except for Depsi and DAC+Depsi which seemed to cluster towards each other due to the extreme changes in gene expression that these inhibitors induce (Fig. 3D).

Figure 3. Genome wide effects of combination epigenetic therapy.

Figure 3

Open in a new tab

A) Number of upregulated and downregulated genes by each inhibitor combination in YB5, MCF7, and HL60 cells (FC > 2, FDR < 0.1). B) Percentage of the transcriptome regulated by each inhibitor when combining data from all three cell lines. C) Venn diagram showing the overlap of regulated genes between the three cell lines. D) Principal component analysis on the RPKM values of the entire transcriptome of each cell line after treatment with epigenetic inhibitors. E) Average RPKM of normal (NC, NBE, and PME) vs. cancer (YB5, MCF7, HL60) of all genes and genes regulated by each inhibitor. * indicates p < 0.05 by Student’s t-test. F) Number of genes regulated by each inhibitor (FC > 2, FDR < 0.1) that belong in the categories of genes as defined in Figure 1F.

Genes regulated by DAC and histone methylation inhibitors were expressed in normal tissue and repressed in cancer similar to DAC alone (Student’s t-test, p < 0.05, Fig. 3E has data averaged from the three cell lines and Fig. S4A has data from individual cell lines). By contrast, DAC+Depsi regulated many genes that had similar expression in both normal tissue and cancer. When dividing the genes into specific and nonspecific categories as done in Fig. 1F, combination therapy of DAC and histone methylation inhibitors still had a similar effect on specific genes (average of 1050, 945, and 1236 genes) compared to nonspecific genes (1174, 1134, and 1305 genes) (Fig. 3F and 3G). Collectively, these results show that combining DAC and histone methylation inhibitors can lead to increased effects with minimal loss of specificity, in contrast to the combination of DAC and HDAC inhibitors.

DNA methylation of genes regulated by combination epigenetic therapy

We next studied the promoter DNA methylation level of genes regulated by combination therapy. Combining DAC with S2101, UNC0638, or GSK343 led to regulation of genes with similar methylation as DAC alone in all three cell lines, while combining DAC with Depsi led to loss of specificity, especially in YB5 cells (Fig. 4A). When combining the data from all three cell lines, 56.9% of DAC+S2101 target genes, 60.8% of DAC+UNC0638 target genes, and 56.5% of DAC+GSK343 target genes had promoter DNA methylation over 10%, similar to DAC alone (59.6%) (Fig. 4B). In addition, combination therapy of DAC with histone methylation inhibitors maintained the larger effect on regulating genes with promoter DNA methylation (24.4%, 21.4%, 27.9%) compared to genes with no or low methylation (9.2%, 7.3%, 10.6%) (Fig. 4C and 4D). By contrast, DAC+Depsi had large effects on both categories of genes (56% of methylated genes and 39.8% of unmethylated genes) (Fig. 4C and 4D). DAC in combination with histone methylation inhibitors also maintained the larger effect on genes that gain at least 20% methylation in cancer (22.7%, 20.2%, 19.1%) compared to genes that do not gain methylation (7.2%, 5.7%, 8.3%) (Fig. S4B). These results demonstrate that combination therapy of DAC and histone methylation inhibitors maintains the selectivity for upregulating genes with promoter DNA methylation, while combining DAC with histone deacetylase inhibitors leads to a loss of this selectivity.

Figure 4. DNA methylation of genes regulated by combination epigenetic therapy.

Figure 4

Open in a new tab

A) Distribution of DNA methylation of genes regulated by combination epigenetic therapy with DNA methylation values in the gene promoter (−1000bp to +500bp of TSS) in YB5 cells, MCF7 cells and HL60 cells. Promoter methylation values of normal colon (NC), normal breast epithelial cells (NBE), and white blood cells (WBC) are shown as comparison. B) Percentage of genes regulated by each inhibitor combination when combining data from all three cell lines that have low (0–10%), moderate (10–50%) or high (50–100%) promoter DNA methylation. C) Percentage of genes with high promoter DNA methylation (50–100%) and low promoter DNA methylation (0–50%) that are regulated by each inhibitor. All conditions except DAC+Depsi in HL60 had p < 0.05 in a Chi-square test. D) Percentage of each category regulated by inhibitor when averaging the data from all three cell lines.

Histone methylation of genes regulated by epigenetic therapy

To determine the impact of baseline histone methylation patterns on gene reactivation, we performed ChIP-seq on YB5 cells using antibodies for H3K4me2 (target of KDM1A), H3K9me2 (target of EHMT2), and H3K27me3 (target of EZH2). Genes upregulated by DAC had low levels of all three modifications, S2101 target genes had relatively high H3K4me2 levels with low H3K9me2 and H3K27me3 levels, UNC0638 target genes had low H3K4me2 with high levels of H3K9me2 and modestly high levels of H3K27me3, and GSK343 target genes had medium levels of H3K4me2 and H3K9me2 with high levels of H3K27me3 (Fig. 5A). Genes upregulated by combination therapy other than DAC+Depsi had low enrichment of all three marks similar to DAC alone (Fig. 5B).

Figure 5. Histone methylation of genes regulated by epigenetic therapy.

Figure 5

Open in a new tab

A) Average read count per million mapped reads of genes upregulated by each epigenetic monotherapy in YB5 cells plotted around gene bodies. B) Average read count per million mapped reads of genes upregulated by each epigenetic combination therapy in YB5 cells plotted around gene bodies. C) Fold change of the percentage of genes with a significant histone mark when comparing genes upregulated by inhibitor and all genes (Baseline). D) Percentage of genes with only histone methylation (H3K9me2 and/or H3K27me3 with < 50% promoter DNA methylation), only DNA methylation (> 50% promoter DNA methylation with no H3K9me2 or H3K27me3), both histone and DNA methylation (> 50% promoter DNA methylation with H3K9me2 and/or H3K27me3), and neither (< 50% promoter DNA methylation with no H3K9me2 or H3K27me3). E) Percentage of genes in each category that are significantly upregulated by each inhibitor.

Next, we associated genes with histone marks by defining whether each gene had significant H3K4me2 peaks present in the promoter (−1kb to +500bp of transcriptional start site), or H3K9me2 and H3K27me3 peaks present in the promoter or the gene body (Fig. S5A, B, and C). In the entire genome, 57% of genes had H3K4me2, 18% had H3K9me2, 17% had H3K27me3, and 27% of genes had none of these three marks (Other) (Fig. 5C). As expected, H3K4me2 marked genes were expressed while the others were silenced (Fig. S5B). When we compared the percentage of genes targeted by the inhibitors that had these marks to the percentage of genes with these marks in the entire genome, DAC targets had 2.1 fold higher percentage of genes that have none of the three marks, UNC0638 targets had 2.9 fold higher percentage of genes that have H3K9me2, and GSK343 targets had 2.6 fold higher percentage of genes that have H3K27me3 (Fig. 5C). 57% of S2101 target genes had H3K4me2 present on the gene promoter, consistent with previous reports that KDM1A preferentially binds to regions already containing H3K4 methylation (23).

We then studied the effect of the inhibitors on genes with both DNA and histone methylation by grouping genes into four categories: genes that were marked only by a silencing histone mark (H3K9me2 and/or H3K27me3), genes that were marked only by DNA methylation (>50% DNA methylation in promoter), genes that had neither silencing mark (most likely expressed genes which could be used to measure nonspecific effect of inhibitors), and genes that were marked by both silencing histone and DNA methylation marks (Fig. 5D). DAC had little effect on genes marked only by histone methylation (0.7% of genes in category) or genes with no silencing marks (0.8%), but had much larger effects on genes marked by only DNA methylation (12.4%) or genes with both marks (6.6%) (Fig. 5E). Combining DAC with S2101, UNC0638, or GSK343 led to an increased percentage of genes induced in genes with only histone methylation (2.5%, 2.3%, and 6.8%, respectively), only DNA methylation (18%, 19.6%, and 22.5%, respectively), and both histone and DNA methylation (9.2%, 14.9%, and 17.3%, respectively), while still having limited effects on genes with no silencing marks (1.7%, 1.1%, and 2.5%, respectively) (Fig. 5E). Depsi and DAC+Depsi had large effects on genes in all four categories, including those with no silencing marks (Fig. S5D and E).

Overall, these data demonstrate that DAC is selective towards genes with DNA methylation with little effect on genes with histone methylation, UNC0638 preferentially targets genes marked by H3K9me2, GSK343 preferentially targets genes marked by H3K27me3, and combination therapy of DAC and histone methylation inhibitors upregulates the highest number of genes that are marked by both DNA and histone methylation.

Synergistic effects of combining DAC with histone methylation inhibitors

To study the synergistic effects of combining DAC with histone methylation inhibitors, we studied the overlap between genes regulated by DAC alone, histone methylation inhibitor alone, and combination therapy (Fig. 6A). In YB5 cells, DAC+S2101 (583 genes), DAC+UNC0638 (641 genes), and DAC+GSK343 (1018 genes) were all able to regulate hundreds of synergistic genes that were not regulated by monotherapy (Fig. 6A). The ability of combination therapy to regulate synergistic genes also occurred in MCF7 and HL60 cells (Fig. 6B, Fig. S6A and B). Pathway analysis of synergistic genes induced in each cell line using Ingenuity Pathway Analysis (IPA) on “Diseases and Biofunctions” showed that the top pathway in each case was “Cancer” (Fig. 6B). In addition, the synergy genes had very limited overlap between the three cell lines (Fig. 6C).

Figure 6. Synergistic effects of combining DAC with histone methylation inhibitors.

Figure 6

Open in a new tab

A) Overlap of genes upregulated by monotherapy (S2101, UNC0638, GSK343, or DAC) and combination therapy in YB5 cells. B) Number of synergistically regulated genes by each inhibitor combination in YB5, MCF7, and HL60 cells, and top 5 “Diseases and Biofunctions” of unique synergy genes as defined by Ingenuity Pathway Analysis. C) Overlap of synergy genes for each combination therapy. D) Heatmap and hierarchical clustering based on RPKMs of all unique synergy genes (1611 genes) in YB5 cells. E) Average RPKM of normal (NC, NBE, and PME) vs. cancer (YB5, MCF7, HL60) of synergy genes by each combination therapy. F) Heatmap and hierarchical clustering based on RPKMs of genes regulated by DAC (715 genes) in YB5 cells.

When we performed hierarchical clustering analysis based on all genes that were synergistically regulated by any of the three combinations in YB5 cells (1611 genes), we observed that each of the combinations clustered separately and had distinct effects (Fig. 6D, Fig. S6C for MCF7 and HL60). Synergy genes induced by the combinations were genes that were silenced in cancer (Fig. 6E). Further demonstrating the distinct effects induced by each combination therapy, treatment of HL60 cells with DAC and S2101 led to the highest upregulation of the differentiation marker CD11b (Figure S6E), treatment with DAC and GSK343 in YB5 cells reactivated the highest number of tumor suppressor genes (as defined in (43)) that have decreased expression in colon cancer (277 genes, Fig. S6F and Table S2), and treatment with DAC and UNC0638 in YB5 cells induced the highest expression of AZA immune genes (367 genes defined in (7) with RNA-seq data in YB5 cells) (Fig. S6G). Consistent with all of these analyses, combination therapy was able to reduce cell proliferation in YB5 cells more effectively than monotherapy (Fig. S6H), and IPA analysis of the top “Molecular and Cellular Functions” shared by the synergy genes in the three cell lines was “Cellular Growth and Proliferation” (Fig. S6I)

Adding histone methylation inhibitors on top of DAC was also able to further upregulate DAC monotherapy target genes, and this effect was again distinct based on each combination as observed by hierarchical clustering (Fig. 6F has data on YB5 cells and Fig S6D has data on MCF7 and HL60 cells). These results demonstrate that combining DAC with histone methylation inhibitors leads to synergistic effects on expression of genes silenced in cancer, and the effects are distinct based on which combination is used, which is consistent with the initial results we obtained from performing qPCR on 10 tumor suppressor genes (Fig. S3E).

Discussion

In this study, we have provided a comprehensive analysis of the selectivity of epigenetic therapy and demonstrated the potential for combining DNA methyltransferase inhibitors with histone methyltransferase and demethylase inhibitors. We report that DAC had high target selectivity in cancer cells, as it had effects on gene upregulation with little effects on gene downregulation, it preferentially upregulated genes with promoter DNA methylation, and it preferentially targeted genes that are silenced in cancer. On the other hand, Depsi demonstrated much less target selectivity with regulation of one-third of the transcriptome, as well as effects on both up and downregulation of gene expression and modulation of many genes that did not change expression in cancer. KDM1A, EHMT2, and EZH2 inhibitors had limited effects on their own, although EHMT2 and EZH2 inhibitors were selective in upregulating genes with high levels of H3K9me2 and H3K27me3, respectively. Combining DAC with KDM1A, EHMT2, or EZH2 inhibitors led to a synergistic increase in gene regulation while still maintaining similar selectivity to DAC, in contrast to the combination of DAC with a HDAC inhibitor.

Epigenetic inhibitors led to different genes being regulated based on the cell line. One reason for this variability could be due to the different baseline gene expression patterns seen in each cell line, as each cell line clustered separately from each other in a principal component analysis (Fig. 3D) and in a hierarchical cluster analysis (Fig. S8A). Another reason for this could be differential expression of epigenetic regulators between the three cell lines, though these were relatively minor (Fig. S8B). Despite these differences, DAC still led to preferential regulation of genes silenced in cancer in all three cell lines. This is in contrast to Depsi and DAC+Depsi, which shared a higher percentage of genes regulated between the cell lines due to its non-selectivity and clustered separately, especially in YB5 and HL60 cells (Fig. 3D and Fig. S8A).

Although DAC was more selective than Depsi in all three cell lines, the magnitude of selectivity varied based on the cell line. The number of genes that have high promoter DNA methylation (32% genes have >50% methylation in YB5 21% in MCF7, 7% in HL60) are different between the cell lines, which could be one of the reasons for the discrepancy in selectivity of DAC for methylated genes. Another possible reason for the discrepancy is that HL60 and MCF7 cells may be more sensitive to DAC and differentiate in a shorter time period, which would lead to more unmethylated downstream genes being upregulated, which is supported by the fact that DAC led to a high induction of the differentiation marker CD11b in HL60 cells (Figure S6E).

The lack of selectivity that we observe in this study from HDAC inhibitors is consistent with previous studies. HDAC inhibitors have been shown to affect 0.5–20% of the transcriptome and have an equal effect on up and downregulation of gene expression (44–46). The non-selectivity that has been observed in previous studies as well as ours may explain the multitude of adverse effects seen in patients treated with HDAC inhibitors (14), and the limited effects in clinical trials of combining DNMT inhibitors with HDAC inhibitors (18–21). We have yet to determine why MCF7 cells seem unique in their response to HDAC inhibitors compared to other cell lines, although Depsi was still non-selective in this cell line since it led to the highest number of genes being downregulated.

DNMT inhibitors cause global hypomethylation throughout the entire genome (5,47–49), although only a small percentage of genes that undergo promoter hypomethylation in response to DNMT inhibitors actually gain chromatin accessibility and become expressed (19% of methylated genes in our study). One possible reason behind this disconnect may be due to other factors such as changes in histone acetylation, histone methylation, or chromatin remodeling being required on top of DNA demethylation for reactivation of gene expression (6,39,48,50). This hypothesis is consistent with our study, where combining DNA demethylation agents with inhibitors that affect histone methylation led to many synergistic gene expression changes (average of 10–12% of transcriptome regulated compared to 8% of transcriptome by DAC alone).

Distinct synergistic genes were induced based on which histone methylation inhibitor was added to DAC, which raises the possibility of targeting specific tumor types based on their gene expression profiles. Indeed, it is noteworthy that DAC+S2101 induced the most differentiation in HL60 cells, while in YB5 cells DAC+GSK343 induced the most tumor suppressor genes and DAC+UNC0638 induced the highest immune response. The focus of this paper was on selectivity and synergy of epigenetic therapy on gene expression, and future studies should analyze the biological impact of combination therapy on cancer growth as well as whether different cancers will have different sensitivities to each combination therapy, which will require long-term experiments that will be able to factor in various effects mediated by epigenetic therapy such as stem cell exhaustion and immune response.

Supplementary Material

1
2
3
4

Acknowledgments

Financial Support: JP.J. Issa is an American Cancer Society Clinical Research professor supported by a generous gift from the F. M. Kirby Foundation, National Institutes of Health grants CA100632, CA046939, CA158112, DE022015, and a grant from the Ellison Medical Foundation. T. Sato was supported by a grant from the Brody Family Medical Trust Fund Fellowship in “Incurable Diseases” of The Philadelphia Foundation.

Conflicts of Interest: JP.J. Issa receives grant support from and is a consultant for Astex (CA), which is developing DNMT inhibitors for clinical use.

References

  • 1.Baylin SB, Jones PA. A decade of exploring the cancer epigenome - biological and translational implications. Nat Rev Cancer. 2011;11(10):726–734. doi: 10.1038/nrc3130. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Kantarjian H, Issa JP, Rosenfeld CS, Bennett JM, Albitar M, DiPersio J, et al. Decitabine improves patient outcomes in myelodysplastic syndromes: results of a phase III randomized study. Cancer. 2006;106(8):1794–1803. doi: 10.1002/cncr.21792. [DOI] [PubMed] [Google Scholar]
  • 3.Silverman LR, Demakos EP, Peterson BL, Kornblith AB, Holland JC, Odchimar-Reissig R, et al. Randomized controlled trial of azacitidine in patients with the myelodysplastic syndrome: a study of the cancer and leukemia group B. J Clin Oncol. 2002;20(10):2429–2440. doi: 10.1200/JCO.2002.04.117. [DOI] [PubMed] [Google Scholar]
  • 4.Issa JP, Kantarjian HM. Targeting DNA methylation. Clin Cancer Res. 2009;15(12):3938–3946. doi: 10.1158/1078-0432.CCR-08-2783. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Tsai HC, Li H, Van Neste L, Cai Y, Robert C, Rassool FV, et al. Transient low doses of DNA-demethylating agents exert durable antitumor effects on hematological and epithelial tumor cells. Cancer Cell. 2012;21(3):430–446. doi: 10.1016/j.ccr.2011.12.029. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Raynal NJ, Si J, Taby RF, Gharibyan V, Ahmed S, Jelinek J, et al. DNA methylation does not stably lock gene expression but instead serves as a molecular mark for gene silencing memory. Cancer Res. 2012;72(5):1170–1181. doi: 10.1158/0008-5472.CAN-11-3248. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Li H, Chiappinelli KB, Guzzetta AA, Easwaran H, Yen RW, Vatapalli R, et al. Immune regulation by low doses of the DNA methyltransferase inhibitor 5-azacitidine in common human epithelial cancers. Oncotarget. 2014;5(3):587–598. doi: 10.18632/oncotarget.1782. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Chiappinelli KB, Strissel PL, Desrichard A, Li H, Henke C, Akman B, et al. Inhibiting DNA Methylation Causes an Interferon Response in Cancer via dsRNA Including Endogenous Retroviruses. Cell. 2015;162(5):974–986. doi: 10.1016/j.cell.2015.07.011. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Roulois D, Loo Yau H, Singhania R, Wang Y, Danesh A, Shen SY, et al. DNA-Demethylating Agents Target Colorectal Cancer Cells by Inducing Viral Mimicry by Endogenous Transcripts. Cell. 2015;162(5):961–973. doi: 10.1016/j.cell.2015.07.056. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Whittaker SJ, Demierre MF, Kim EJ, Rook AH, Lerner A, Duvic M, et al. Final results from a multicenter, international, pivotal study of romidepsin in refractory cutaneous T-cell lymphoma. J Clin Oncol. 2010;28(29):4485–4491. doi: 10.1200/JCO.2010.28.9066. [DOI] [PubMed] [Google Scholar]
  • 11.Duvic M, Talpur R, Ni X, Zhang C, Hazarika P, Kelly C, et al. Phase 2 trial of oral vorinostat (suberoylanilide hydroxamic acid, SAHA) for refractory cutaneous T-cell lymphoma (CTCL) Blood. 2007;109(1):31–39. doi: 10.1182/blood-2006-06-025999. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Rius M, Lyko F. Epigenetic cancer therapy: rationales, targets and drugs. Oncogene. 2012;31(39):4257–4265. doi: 10.1038/onc.2011.601. [DOI] [PubMed] [Google Scholar]
  • 13.Navada SC, Steinmann J, Lubbert M, Silverman LR. Clinical development of demethylating agents in hematology. J Clin Invest. 2014;124(1):40–46. doi: 10.1172/JCI69739. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.West AC, Johnstone RW. New and emerging HDAC inhibitors for cancer treatment. J Clin Invest. 2014;124(1):30–39. doi: 10.1172/JCI69738. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Ahuja N, Easwaran H, Baylin SB. Harnessing the potential of epigenetic therapy to target solid tumors. J Clin Invest. 2014;124(1):56–63. doi: 10.1172/JCI69736. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Cameron EE, Bachman KE, Myohanen S, Herman JG, Baylin SB. Synergy of demethylation and histone deacetylase inhibition in the re-expression of genes silenced in cancer. Nat Genet. 1999;21(1):103–107. doi: 10.1038/5047. [DOI] [PubMed] [Google Scholar]
  • 17.Kalac M, Scotto L, Marchi E, Amengual J, Seshan VE, Bhagat G, et al. HDAC inhibitors and decitabine are highly synergistic and associated with unique gene-expression and epigenetic profiles in models of DLBCL. Blood. 2011;118(20):5506–5516. doi: 10.1182/blood-2011-02-336891. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Issa JP, Garcia-Manero G, Huang X, Cortes J, Ravandi F, Jabbour E, et al. Results of phase 2 randomized study of low-dose decitabine with or without valproic acid in patients with myelodysplastic syndrome and acute myelogenous leukemia. Cancer. 2015;121(4):556–561. doi: 10.1002/cncr.29085. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Blum W, Klisovic RB, Hackanson B, Liu Z, Liu S, Devine H, et al. Phase I study of decitabine alone or in combination with valproic acid in acute myeloid leukemia. J Clin Oncol. 2007;25(25):3884–3891. doi: 10.1200/JCO.2006.09.4169. [DOI] [PubMed] [Google Scholar]
  • 20.Lin J, Gilbert J, Rudek MA, Zwiebel JA, Gore S, Jiemjit A, et al. A phase I dose-finding study of 5-azacytidine in combination with sodium phenylbutyrate in patients with refractory solid tumors. Clin Cancer Res. 2009;15(19):6241–6249. doi: 10.1158/1078-0432.CCR-09-0567. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Prebet T, Sun Z, Figueroa ME, Ketterling R, Melnick A, Greenberg PL, et al. Prolonged administration of azacitidine with or without entinostat for myelodysplastic syndrome and acute myeloid leukemia with myelodysplasia-related changes: results of the US Leukemia Intergroup trial E1905. J Clin Oncol. 2014;32(12):1242–1248. doi: 10.1200/JCO.2013.50.3102. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Mimasu S, Umezawa N, Sato S, Higuchi T, Umehara T, Yokoyama S. Structurally designed trans-2-phenylcyclopropylamine derivatives potently inhibit histone demethylase LSD1/KDM1. Biochemistry. 2010;49(30):6494–6503. doi: 10.1021/bi100299r. [DOI] [PubMed] [Google Scholar]
  • 23.Mohammad HP, Smitheman KN, Kamat CD, Soong D, Federowicz KE, Van Aller GS, et al. A DNA Hypomethylation Signature Predicts Antitumor Activity of LSD1 Inhibitors in SCLC. Cancer Cell. 2015;28(1):57–69. doi: 10.1016/j.ccell.2015.06.002. [DOI] [PubMed] [Google Scholar]
  • 24.Vedadi M, Barsyte-Lovejoy D, Liu F, Rival-Gervier S, Allali-Hassani A, Labrie V, et al. A chemical probe selectively inhibits G9a and GLP methyltransferase activity in cells. Nature chemical biology. 2011;7(8):566–574. doi: 10.1038/nchembio.599. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Liu F, Barsyte-Lovejoy D, Li F, Xiong Y, Korboukh V, Huang XP, et al. Discovery of an in vivo chemical probe of the lysine methyltransferases G9a and GLP. Journal of medicinal chemistry. 2013;56(21):8931–8942. doi: 10.1021/jm401480r. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Verma SK, Tian XR, LaFrance LV, Duquenne C, Suarez DP, Newlander KA, et al. Identification of Potent, Selective, Cell-Active Inhibitors of the Histone Lysine Methyltransferase EZH2. Acs Med Chem Lett. 2012;3(12):1091–1096. doi: 10.1021/ml3003346. doi. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.McCabe MT, Ott HM, Ganji G, Korenchuk S, Thompson C, Van Aller GS, et al. EZH2 inhibition as a therapeutic strategy for lymphoma with EZH2-activating mutations. Nature. 2012;492(7427):108–112. doi: 10.1038/nature11606. [DOI] [PubMed] [Google Scholar]
  • 28.Knutson SK, Warholic NM, Wigle TJ, Klaus CR, Allain CJ, Raimondi A, et al. Durable tumor regression in genetically altered malignant rhabdoid tumors by inhibition of methyltransferase EZH2. Proc Natl Acad Sci U S A. 2013;110(19):7922–7927. doi: 10.1073/pnas.1303800110. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Xu B, On DM, Ma A, Parton T, Konze KD, Pattenden SG, et al. Selective inhibition of EZH2 and EZH1 enzymatic activity by a small molecule suppresses MLL-rearranged leukemia. Blood. 2015;125(2):346–357. doi: 10.1182/blood-2014-06-581082. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Lehnertz B, Pabst C, Su L, Miller M, Liu F, Yi L, et al. The methyltransferase G9a regulates HoxA9-dependent transcription in AML. Genes Dev. 2014;28(4):317–327. doi: 10.1101/gad.236794.113. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Gao C, Devarajan K, Zhou Y, Slater CM, Daly MB, Chen X. Identifying breast cancer risk loci by global differential allele-specific expression (DASE) analysis in mammary epithelial transcriptome. BMC genomics. 2012;13:570. doi: 10.1186/1471-2164-13-570. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Kim D, Pertea G, Trapnell C, Pimentel H, Kelley R, Salzberg SL. TopHat2: accurate alignment of transcriptomes in the presence of insertions, deletions and gene fusions. Genome Biol. 2013;14(4):R36. doi: 10.1186/gb-2013-14-4-r36. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Robinson MD, McCarthy DJ, Smyth GK. edgeR: a Bioconductor package for differential expression analysis of digital gene expression data. Bioinformatics. 2010;26(1):139–140. doi: 10.1093/bioinformatics/btp616. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Langmead B, Salzberg SL. Fast gapped-read alignment with Bowtie 2. Nat Methods. 2012;9(4):357–359. doi: 10.1038/nmeth.1923. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Zhang Y, Liu T, Meyer CA, Eeckhoute J, Johnson DS, Bernstein BE, et al. Model-based analysis of ChIP-Seq (MACS) Genome Biol. 2008;9(9):R137. doi: 10.1186/gb-2008-9-9-r137. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Zang C, Schones DE, Zeng C, Cui K, Zhao K, Peng W. A clustering approach for identification of enriched domains from histone modification ChIP-Seq data. Bioinformatics. 2009;25(15):1952–1958. doi: 10.1093/bioinformatics/btp340. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Shen L, Shao N, Liu X, Nestler E. ngs.plot: Quick mining and visualization of next-generation sequencing data by integrating genomic databases. BMC genomics. 2014;15:284. doi: 10.1186/1471-2164-15-284. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Jelinek J, Liang S, Lu Y, He R, Ramagli LS, Shpall EJ, et al. Conserved DNA methylation patterns in healthy blood cells and extensive changes in leukemia measured by a new quantitative technique. Epigenetics : official journal of the DNA Methylation Society. 2012;7(12):1368–1378. doi: 10.4161/epi.22552. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Si J, Boumber YA, Shu J, Qin T, Ahmed S, He R, et al. Chromatin remodeling is required for gene reactivation after decitabine-mediated DNA hypomethylation. Cancer Res. 2010;70(17):6968–6977. doi: 10.1158/0008-5472.CAN-09-4474. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Wu Y, Wang X, Wu F, Huang R, Xue F, Liang G, et al. Transcriptome profiling of the cancer, adjacent non-tumor and distant normal tissues from a colorectal cancer patient by deep sequencing. PloS one. 2012;7(8):e41001. doi: 10.1371/journal.pone.0041001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Chang KH, Sengupta A, Nayak RC, Duran A, Lee SJ, Pratt RG, et al. p62 is required for stem cell/progenitor retention through inhibition of IKK/NF-kappaB/Ccl4 signaling at the bone marrow macrophage-osteoblast niche. Cell reports. 2014;9(6):2084–2097. doi: 10.1016/j.celrep.2014.11.031. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Qin T, Si J, Raynal NJ, Wang X, Gharibyan V, Ahmed S, et al. Epigenetic synergy between decitabine and platinum derivatives. Clinical epigenetics. 2015;7(1):97. doi: 10.1186/s13148-015-0131-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Zhao M, Sun J, Zhao Z. TSGene: a web resource for tumor suppressor genes. Nucleic Acids Res. 2013;41:D970–D976. doi: 10.1093/nar/gks937. (Database issue) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Chueh AC, Tse JW, Togel L, Mariadason JM. Mechanisms of Histone Deacetylase Inhibitor-Regulated Gene Expression in Cancer Cells. Antioxidants & redox signaling. 2015;23(1):66–84. doi: 10.1089/ars.2014.5863. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.LaBonte MJ, Wilson PM, Fazzone W, Groshen S, Lenz HJ, Ladner RD. DNA microarray profiling of genes differentially regulated by the histone deacetylase inhibitors vorinostat and LBH589 in colon cancer cell lines. BMC medical genomics. 2009;2:67. doi: 10.1186/1755-8794-2-67. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46.Peart MJ, Smyth GK, van Laar RK, Bowtell DD, Richon VM, Marks PA, et al. Identification and functional significance of genes regulated by structurally different histone deacetylase inhibitors. Proc Natl Acad Sci U S A. 2005;102(10):3697–3702. doi: 10.1073/pnas.0500369102. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 47.Klco JM, Spencer DH, Lamprecht TL, Sarkaria SM, Wylie T, Magrini V, et al. Genomic impact of transient low-dose decitabine treatment on primary AML cells. Blood. 2013;121(9):1633–1643. doi: 10.1182/blood-2012-09-459313. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48.Pandiyan K, You JS, Yang X, Dai C, Zhou XJ, Baylin SB, et al. Functional DNA demethylation is accompanied by chromatin accessibility. Nucleic Acids Res. 2013;41(7):3973–3985. doi: 10.1093/nar/gkt077. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49.Lund K, Cole JJ, VanderKraats ND, McBryan T, Pchelintsev NA, Clark W, et al. DNMT inhibitors reverse a specific signature of aberrant promoter DNA methylation and associated gene silencing in AML. Genome Biol. 2014;15(8):406. doi: 10.1186/s13059-014-0406-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 50.Kondo Y, Shen L, Yan PS, Huang TH, Issa JP. Chromatin immunoprecipitation microarrays for identification of genes silenced by histone H3 lysine 9 methylation. Proc Natl Acad Sci U S A. 2004;101(19):7398–7403. doi: 10.1073/pnas.0306641101. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1
NIHMS832366-supplement-1.xlsx (18KB, xlsx)
2
NIHMS832366-supplement-2.pdf (3.9MB, pdf)
3
NIHMS832366-supplement-3.xlsx (21.7KB, xlsx)
4
NIHMS832366-supplement-4.xlsx (13.4KB, xlsx)

RESOURCES