Skip to main content
NCBI home page
Molecular Oncology logoLink to Molecular Oncology
. 2013 Aug 30;7(6):1129–1141. doi: 10.1016/j.molonc.2013.08.008

Long range epigenetic silencing is a trans‐species mechanism that results in cancer specific deregulation by overriding the chromatin domains of normal cells

Marta Forn 1, Mar Muñoz 1, Daniele VF Tauriello 2, Anna Merlos-Suárez 2, Verónica Rodilla 3, Anna Bigas 3, Eduard Batlle 2,4, Mireia Jordà 1, Miguel A Peinado 1,✉
PMCID: PMC5528435  PMID: 24035705

Abstract

DNA methylation and chromatin remodeling are frequently implicated in the silencing of genes involved in carcinogenesis. Long Range Epigenetic Silencing (LRES) is a mechanism of gene inactivation that affects multiple contiguous CpG islands and has been described in different human cancer types. However, it is unknown whether there is a coordinated regulation of the genes embedded in these regions in normal cells and in early stages of tumor progression. To better characterize the molecular events associated with the regulation and remodeling of these regions we analyzed two regions undergoing LRES in human colon cancer in the mouse model. We demonstrate that LRES also occurs in murine cancer in vivo and mimics the molecular features of the human phenomenon, namely, downregulation of gene expression, acquisition of inactive histone marks, and DNA hypermethylation of specific CpG islands. The genes embedded in these regions showed a dynamic and autonomous regulation during mouse intestinal cell differentiation, indicating that, in the framework considered here, the coordinated regulation in LRES is restricted to cancer. Unexpectedly, benign adenomas in ApcMin/+ mice showed overexpression of most of the genes affected by LRES in cancer, which suggests that the repressive remodeling of the region is a late event. Chromatin immunoprecipitation analysis of the transcriptional insulator CTCF in mouse colon cancer cells revealed disrupted chromatin domain boundaries as compared with normal cells. Malignant regression of cancer cells by in vitro differentiation resulted in partial reversion of LRES and gain of CTCF binding. We conclude that genes in LRES regions are plastically regulated in cell differentiation and hyperproliferation, but are constrained to a coordinated repression by abolishing boundaries and the autonomous regulation of chromatin domains in cancer cells.

Keywords: Gene silencing, DNA methylation, Colorectal cancer, Chromatin marks, Histone modification, Long range epigenetic silencing, Cell differentiation

Highlights

  • Long range epigenetic silencing (LRES) is a mechanism of gene inactivation in cancer.

  • Human and murine intestinal cancers show LRES with similar epigenetic features.

  • Early adenomas show upregulation of genes that will be silenced in advanced stages.

  • LRES overrides autonomous gene regulation by loss of chromatin architecture.

  • Malignant regression results in partial reversion of LRES.

Abbreviations

LRES

Long range epigenetic silencing

ChIP

Chromatin immunoprecipitation

ISC

Intestinal stem cells

1. Introduction

Colorectal cancer is the third most common cancer worldwide, being more prevalent in developed countries. Multistep transformation of normal intestine into high‐grade malignancies is due to genetic and epigenetic alterations in the networks governing homeostasis in multicellular organisms. Both mechanisms cooperate in cancer progression allowing the acquisition of essential hallmarks of the neoplastic state. Developmental pathways are often misregulated in cancer cells, suggesting that tumor‐propagating cells have takeover cellular networks that control the behavior of normal stem cells. This model is supported by recent studies showing that a stem cell/progenitor cell hierarchy is maintained in mouse small intestine early adenomas (Barker et al., 2009; Schepers et al., 2012). Indeed, the intestinal stem cell program defines a cancer stem cell niche within colorectal tumors which may play a central role in cancer relapse (Merlos‐Suarez et al., 2011).

The cancer epigenome is characterized by alterations in DNA methylation (global hypomethylation and specific hypermethylation), and changes in the patterns of histone modification and chromatin modifying‐enzymes, which play important roles in tumor initiation and progression (reviewed in Jones and Baylin, 2007; Portela and Esteller, 2010). Identification of discrete hypermethylated CpG islands associated with gene silencing has contributed to a better knowledge of tumor progression and a promise of novel biomarkers for cancer detection and prognosis (Esteller, 2008). Conversely, recent studies stress the fact that colorectal cancer cells present differentially methylated domains and that hypomethylated domains are prone to interact with nuclear lamina, which links the epigenetic landscape disruption to perturbations in cellular architecture (Berman et al., 2012; Hansen et al., 2011). In addition, other mechanisms of regional coordination have been found in tumors, such as Long Range Epigenetic Silencing (LRES), which was initially described in colorectal cancer as a heterochromatinization process, affecting a 4 Mb gene rich region in 2q14.2 (Frigola et al., 2006). This phenomenon involves the hypermethylation of specific CpG islands, down‐regulation of most of the genes in the region and enrichment in H3K9me2. In colorectal cancer, a similar silencing mechanism has been described in other chromosomal regions including 3q22.2 (Hitchins et al., 2007), 5q35.2 (Rodriguez et al., 2008), the protocadherin cluster at 5q31 (Dallosso et al., 2012) and the Ikaros gene family (Javierre et al., 2011). Moreover, hypermethylation of genes located in LRES regions has been associated with worse outcome and its detection has been proposed as a diagnostic tool (Mayor et al., 2009). In addition to colorectal cancer, LRES has been also detected in prostate (Coolen et al., 2010), breast (Hsu et al., 2010; Novak et al., 2008), gastric (Park et al., 2011), bladder (Stransky et al., 2006) and Wilms' kidney tumors (Dallosso et al., 2009). The most comprehensive study regarding LRES was a genome wide analysis of prostate cancer that revealed 47 LRES regions, typically spanning about 2 Mb and harboring approximately 12 genes, with a relatively high prevalence of tumor suppressor and miRNA genes (Coolen et al., 2010).

The analysis of the LRES chromatin landscape in cancer cells treated with epigenetic drugs points out that this phenomenon may result from the overtaking of chromatin boundaries determining the autonomous regulation of domains with different functional properties (Mayor et al., 2011; Rodriguez et al., 2008). However, the regulation of these regions in normal cells and the dynamics of the concurrent silencing during tumor progression remain to be elucidated. The limited access to normal human tissue and especially at initial stages of carcinogenesis led us to exploit the mouse as a model of study. Here we demonstrate that genomic regions affected by LRES in human cancer are also silenced in mouse carcinogenesis by the same mechanism. These regions are dynamically expressed during intestinal stem cell (ISC) differentiation and in benign Apc Min/+ adenomas without significant changes in the epigenetic profiles. On the other hand, the epigenetic landscape of these regions appears to be remodeled in advanced stages of tumorigenesis resulting in DNA hypermethylation and downregulation of some genes. Furthermore, reorganization of chromatin domains is also affected during differentiation and malignant processes, as uncovered by changes in the insulator factor CTCF associated with boundaries, revealing a crosstalk between regional co‐regulation and chromatin architecture.

2. Materials and methods

2.1. Tissue samples

All mice used were in a C57BL/6J background and kept under pathogen‐free conditions and in accordance with ethical committee guidelines. Small intestine cellular fractions were obtained from wild type animals as previously described (Merlos‐Suarez et al., 2011). Briefly, serial EDTA incubation/shaking steps of whole small intestines were performed to obtain fractions enriched in villi (first incubations) or crypts (last incubations). Crypt cell populations were further purified by FACS based on their EphB2 surface levels. Adenomas and control tissues were obtained from Apc Min/+ mice (Jackson Laboratories) as previously described (Rodilla et al., 2009).

2.2. Cell lines

Mouse colon carcinoma cell line CT26, human colon adenocarcinoma cell line CaCo‐2 and human colorectal carcinoma cell line HCT116 were obtained from the American Type Culture Collection (ATCC). In order to analyze CaCo‐2 during differentiation, these cells were cultured to confluence (day 0), or for 10, 20 or 40 days post‐confluence, with medium changed every second or third day. CaCo‐2 differentiation was also achieved using BIOCOAT® HTS Caco‐2 Assay System (BD Bioscience, Bedorf, MA USA), in which cells grown on Fibrillar Collagen Cell Culture Inserts were treated with butyric acid for 48 h according to manufacturer instructions. Finally, differentiated cells were cultured in normal conditions for 18 days (post‐treated cells).

2.3. Mouse colon carcinomas

C57BL/6J mice with genotype Lgr5GFP‐CreERt2/wt; Apcflox/wt; Tgfbr2flox/flox; Trp53flox/flox were treated with 3% DSS for 5 days and received two injections with 10 mg/ml tamoxifen at day 1 and 5 of the DSS treatment. After approximately 150 days, mice showed advanced symptoms of morbidity, pain and weight loss. Upon sacrifice, mice presented with large intestinal tumors, mainly in the colon, that histologically resemble carcinomas. Tumor samples were chopped into small pieces, part of which was taken up in Trizol for mRNA extraction using the PureLink mRNA mini kit (Ambio/Life Technologies) and the rest was digested with 15 U/ml collagenase IV (Sigma), filtered through consecutive 100 um, 70 um and 40 um filters, and single cells were sorted by high GFP expression using a BD FACSAria III cytometer (BD). These Lgr5‐positive tumor stem cells were cultured with Advanced DMEM/F12 with 2 mM GlutaMAX, 10 mM Hepes, N2 and B27 without retinoic acid supplements (Gibco/Life technologies), and 50 ng/ml hEGF (Peprotech), in matrigel (basement membrane matrix low concentration, BD). After expansion, genomic DNA was obtained using the GenElute kit (Sigma).

2.4. Bisulfite sequencing

Genomic DNA was obtained by conventional phenol‐chloroform extraction and ethanol precipitation. Bisulfite treatment was performed using the EZ DNA methylation kitTM (Zymo Research). Specific genomic regions were amplified by nested PCR (primers are listed in Supplementary Table 3 and Mayor et al., 2011). Methylation analysis was carried out on pooled PCR products by direct PCR sequencing with the BigDye Terminator v3.1 Cycle Sequencing Kit (Applied Biosystems, Foster City, CA, USA). Bisulfite clonal analysis was performed as described (Frigola et al., 2006).

2.5. RNA extraction and quantitative real‐time RT‐PCR

RNA and the corresponding cDNA were obtained using standard protocols. Gene expression was quantified using the ABI PRISM 7900 HT sequence detection system (Applied Biosystems) and primers described in references (Frigola et al., 2006; Mayor et al., 2011), or Light Cycler 2.0 real time PCR system (Roche Diagnostics), and primers listed in Supplementary Table 1. The reactions were performed in triplicate. Gene expression values were normalized using three to six housekeeping genes (Supplementary Table 1).

2.6. Chromatin immunoprecipitation (ChIP) and sequential ChIP assays

Histone marks and chromatin proteins were analyzed in cells and tissues as described in Supplementary Methods. Briefly, chromatin was sonicated into 200–500 bp fragments (Bioruptor, Diagenode, Liège, Belgium) and immunoprecipitated using the Immunoprecipitation Assay Kit (Millipore) or the Low Cell ChIP Kit (Diagenode, Liège, Belgium) for low amounts of material. Total H3, CTCF, EZH2 and the following histone modifications were analyzed by ChIP: histone 3 lysine 4 dimethylation (H3K4me2) and trimethylation (H3K4me3), histone 3 lysine 9 acetylation (H3K9ac), histone 3 lysine 27 trimethylation (H3K27me3), histone 3 lysine 9 dimethylation (H3K9me2), and trimethylation (H3K9me3). Rabbit IgG serum was used as negative control (see Supplementary Methods).

2.7. Western blot

The lysates for Western blot analysis were prepared as described (Andrews and Faller, 1991). Western blot analyses were conducted under denaturing conditions using the primary antibodies anti‐CTCF (Millipore) and anti‐α‐Tubulin (Sigma) as control. The conjugated secondary antibodies, anti‐rabbit and anti‐mouse horseradish peroxidase, were purchased from Dako (Glostrup, Denmark). Detection was done with enhanced chemiluminescence Pierce® Supersignal West Pico Kit (Thermo Scientific, Rockford, IL, USA) according to the manufacturer's instructions.

3. Results

3.1. LRES affects orthologous human–mouse chromosomal regions in colon cancer

Previous studies have shown that most human colorectal cancers display LRES in chromosomal regions 2q14.2 and 5q35.2 (Frigola et al., 2006; Mayor et al., 2009; Rodriguez et al., 2008). The corresponding mouse orthologous chromosomal regions (1qE2.3 and 13qB1, respectively) are highly conserved, as gene order is identical and the CpG islands in promoter regions are preserved, although additional CpG islands appear in the human genome (Supplementary Figure 1). In order to determine the occurrence of LRES in mouse cancer, we analyzed gene expression and epigenetic profiles of 10 genes embedded in each one of these two chromosomal regions in mouse normal colon tissue and the colon cancer cell line CT26.

Genes embedded in both regions exhibited variable levels of expression in normal mouse colon (Figure 1A) mimicking profiles reported in human colon cells (Frigola et al., 2006; Mayor et al., 2009; Rodriguez et al., 2008). Most genes were downregulated in CT26 cells (Figure 1B). Genes Inhbb, Gli2, Cplx var2 and Sncb exhibited full silencing accompanied by hypermethylation of the respective CpG island (Figure 1C), while Insig2, Ptpn4, Ralb, Tsn, Cplx var 1, Cltb, Rnf44 and Tspan17 were downregulated without changes in the DNA methylation profile (Figure 1B). The promoter CpG island of genes inactive in normal colon (En1, Drd1 and Hrh2) was hypermethylated in cancer cells, as well as the Sctr CpG rich promoter region (Figure 1C). Marco and Cdhr2 did not contain CpG island in the promoter and only Cdhr2 was expressed in normal colon but fully silenced in CT26 cancer cells (Figure 1B).

Figure 1.

Figure 1

Open in a new tab

A, Normalized gene expression levels in genes along the chromosomal regions 1qE2.3 and 13qB1 in normal colon tissue and the CT26 colon cancer cell line. Values for Ptpn4, Ralbb, and Rfn44 in normal colon are shown on top of the corresponding bar because they are out of scale. B, Differential gene expression profiles of chromosomal regions 1qE2.3 and 13qB1 in mouse CT26 colon cancer cell line compared with normal colon tissue. Values are normalized against three housekeeping genes (Sdha, Actb, B2m) and represented as the log2 of the ratio CT26/colon. C, DNA methylation status of promoter CpG islands associated to genes. Black squares represent fully methylated CpG islands and empty squares unmethylated CpG islands. SCTR does not contain a canonical CpG island but the methylation value for a CpG rich region in the promoter is shown. Marco and Cdhr2 promoters were CpG poor and were not analyzed for DNA methylation.

The global downregulation of the genes in the two regions was accompanied by a loss of H3K9ac levels analyzed by ChIP (Figure 2 and Supplementary Figure 2A). Moreover, CT26 cells showed increases of the inactive chromatin epigenetic marks H3K9me2 and H3K9me3 (in less extent) in the promoter regions of low expressed genes. This heterochromatinization was accompanied by an increase of EZH2 binding, suggesting a rise of the PRC2 in silenced genes (Figure 2 and Supplementary Figure 2A).

Figure 2.

Figure 2

Open in a new tab

Epigenetic marks along the 1qE2.3 region in murine normal colon and CT26 colon cancer cells. Histone modification marks are represented relative to total H3 values while EZH2 is relative to input.

In addition to DNA methylation, mouse CT26 cancer cells also presented active (H3K4me3) and inactive (H3K27me3) histone modifications at the promoters of silenced genes (Figure 2 and Supplementary Figure 2). Interestingly, the coexistence of active and inactive marks, initially identified in stem cells and known as bivalent domains (Azuara et al., 2006; Bernstein et al., 2006), was also detected in normal tissue (Figure 3A and Supplementary Figure 3). Moreover, we observed a reduction of both H3K4me3 and H3K27me3 in CT26 as compared with normal cells (Figure 3B and Supplementary Figure 3B), which reinforces the concept that bivalence is actively maintained by a balanced regulation of active and inactive marks. Indeed, the bivalent nature of these genes in normal tissue and the colon cancer cell line CT26 was confirmed by Sequential Chip (Figure 3C,D and Supplementary Figure 3).

Figure 3.

Figure 3

Open in a new tab

Bivalent chromatin marks along the 1qE2.3 chromosomal (data on 13qB1 region are represented in Supplementary Figure 3). A, Relative occupancy of H3K27me3 and H3K4me3 marks in promoter regions in colon normal tissue. The chromatin was considered bivalent when the values of both marks were between 20 and 80% (dotted lines). Bivalent genes are denoted with an asterisk. B, Relative enrichment of H3K4me3 and H3K27me3 marks in CT26 compared to normal colon tissue. C, Sequential ChIP analysis of bivalent genes in normal colon tissue. Relative values are calculated respect the input. Bars are colored according to the combination and sequence of antibodies (see the legend in panel D). Results are grouped according to the first immunoprecipitation (IP) mark. D, Sequential ChIP along the 1qE2.3 chromosomal region in CT26 cells. Data using H3K4me3 in the first IP, are not shown due to the low levels of detection. NA, not analyzed.

As a whole, we have shown that two chromosomal regions undergoing LRES in human colon cancer cells (Frigola et al., 2006; Rodriguez et al., 2008) go through the same phenomenon in a murine cancer cell line, mimicking the transcriptional and epigenetic patterns of the human counterpart: This is down‐regulation of genes with active chromatin promoters and complete silencing of genes with bivalent promoters. Importantly, sequential ChIP experiments dismiss any possible doubt about the bivalent nature of the chromatin in the genes analyzed here in normal cells.

3.2. LRES in early and advanced intestinal tumors

Gene expression and epigenetic profiles of 1qE2.3 and 13qB1 chromosomal regions were analyzed in Apc Min/+ adenomas and compared with their adjacent non‐tumor tissue and normal small intestine. Genes expressed at high levels in normal tissue (Ddx18, Insig2, Ptpn4, Ralbb, Cplx var 1, Thoc3, Cltb and Rnf44) were upregulated in adenomas, except for Cdhr2 that was downregulated and Sncb, that showed both up‐ and downregulation depending on the adenoma. On the other hand, low expressed genes with bivalent (En1, Marco, Sctr, Inhbb, Gli2, Hrh2, Cplx var 2) or not bivalent promoters (Tsn, Ddr1, Sfxn1, Tspan17), displayed a variable pattern among adenomas (Figure 4A). CpG island promoters remained unmethylated in all the adenomas (data not shown). Changes in histone modification profiles were modest in adenomas as compared with both normal intestine and the adjacent tissue, yet were consistent with the changes in gene expression (enrichment of active marks H3K4me3 and H3K9ac in overexpressed genes) (Supplementary Figure 4). Genes with bivalent promoters (coexistence of H3K4me3 and H3K27me3) in normal colon and cell line CT26 retained this signature in adenomas (Supplementary Figure 4).

Figure 4.

Figure 4

Open in a new tab

Gene expression in early adenomas and advanced tumors. A, Relative values of gene expression in 1qE2.3 and 13qB1 regions in ApcMin/+ adenomas compared with small intestine tissue from normal mice. B, Relative values of gene expression in 1qE2.3 and 13qB1 regions in colon carcinomas obtained from C57BL/6J transgenic mice with floxed Apc, Tgfbr2, and Trp53 compared with normal colon tissue. C, DNA methylation status of promoter CpG islands associated to genes in intestinal tumors. Black squares represent fully methylated CpG islands and empty squares unmethylated CpG islands. SCTR does not contain a canonical CpG island but the methylation value for a CpG rich region in the promoter is shown. Marco and Cdhr2 promoters were CpG poor and were not analyzed for DNA methylation. Asterisks denote bivalent genes. NA, not analyzed.

To investigate the presence of LRES in advanced stages of intestinal carcinogenesis we analyzed DNA methylation and gene expression in colon carcinomas obtained from C57BL/6J transgenic mice with floxed Apc, Tgfbr2, and Trp53 (see Section 2.3). Gene expression changes were similar in all the tumors (Figure 4B), which underwent downregulation of ten genes and reversing the global trend towards upregulation observed in adenomas (Figure 4A). Indeed, gene promoters with bivalent domains showed variable degrees of DNA hypermethylation (Figure 4C). Surprisingly, the two genes consistently upregulated (Hrh2 and Inhbb) also exhibited gains of DNA methylation. To get insights into the distribution of DNA hypermethylation we performed bisulfite clonal analysis of Hrh2 and En1 genes. The presence of diverse DNA methylation profiles in all the cases examined indicated the existence of multiple cell populations (Supplementary Figure 5). Therefore, the overexpression concurrent with DNA hypermethylation of the promoter observed in these two genes might be explained by the coexistence of cells retaining the “activating” phenotype observed in adenomas (Figure 4A) and the epigenetic inactivation observed in advanced tumors and cell lines (Figures 1C and 4C).

In summary, 1qE2.3 and 13qB1 show a dynamic regulation during carcinogenesis: while most of the genes tend to be overexpressed in early stages (represented by Apc Min/+ adenomas), more advanced tumors gain hypermethylation in promoters associated with bivalent genes and shown downregulation of most genes, although full silencing is not observed as it could be masked by the presence of cells in earlier stages of transformation.

3.3. Regions silenced by LRES are dynamically expressed during small intestine differentiation

The regional epigenetic architecture and the regulation of LRES have been well characterized in human cancer cells and also its coordinated response to drug induced reactivation (Frigola et al., 2006; Mayor et al., 2011; Rodriguez et al., 2008). We wondered about the dynamics of epigenetic profiles in these regions under a physiological process. Therefore, we compared expression and epigenetic profiles of 1qE2.3 and 13qB1 genes in crypt and villus fractions from the mouse small intestine. Furthermore, focusing within the crypt compartment, we analyzed cells at different levels of differentiation as determined by the EphB2 receptor levels: intestinal stem cells (ISC, EphB2high), transient‐amplifying cells (TA, EphB2medium and EphB2low) and differentiated cells (EphB2negative) using previously established criteria (Merlos‐Suarez et al., 2011).

Along the differentiation axis, highly expressed genes tended to have diminished levels (similarly to what it is observed in cancer cells), while bivalent genes remained invariable, except for Cdhr2 and the variant 2 of Cplx2 that were increased (Supplementary Figure 6). Strikingly, Cplx2 variant 1 was downregulated in all stages of differentiation as compared with ISC (Supplementary Figure 6). The two Cplx2 variants also exist in human and rat (Raevskaya et al., 2005) and, although both are associated with bivalent promoters in normal tissue (Supplementary Figure 3), according to our results, they appear to be autonomously regulated during cell differentiation (Supplementary Figure 6). Regarding Cdhr2, it is of note that its overexpression during differentiation is reversed in most adenomas (Figure 4A), resembling the pattern found in CT26 (Figure 1A) and advanced tumors (see below).

At the epigenetic level, all analyzed CpG islands remained unmethylated during differentiation and CpG island shores (2 kbp regions flanking CpG islands (Irizarry et al., 2009)) showed low levels of methylation, with no differences between ISC and differentiated cells (data not shown). Moreover, minimal or no changes were observed in histone marks (H3K9ac, H3K4me2 and H3K27me3) (Supplementary Figure 7). Thus, genes embedded in 1qE2.3 and 13qB1 regions were dynamically expressed and autonomously regulated along ISC differentiation without significant changes at the epigenetic level.

3.4. CTCF binding is diminished in LRES regions in cancer

The transcriptional repressor CTCF plays an important role as insulator in the maintenance of chromatin boundaries (Kim et al., 2007). We wondered whether CTCF binding could be affected in LRES by comparing the profiles of normal colon tissue and the CT26 cancer counterpart.

Along the 1qE2.3 region several predicted CTCF binding sites have been described (Insulator Data Base, University of Tennessee Health Science Center, http://insulatordb.uthsc.edu/home.php). The presence of CTCF was determined by ChIP analysis in six predicted binding sites. In normal colon, all six were positive for CTCF binding, although two of them (PRE07749 and PRE07752) produced a lower signal (Supplementary Figure 8A). Moreover, CTCF was virtually undetectable in six close regions not predicted as CTCF binding sites and therefore considered negative controls (Supplementary Figure 8A). Cell fractionation of intestinal mucosa cells revealed CTCF binding (Figure 5A) in the crypts, but we were not able to determine CTCF binding in villi due to technical limitations. In CT26 cells the binding was also confirmed in all the predicted sites, although the colon cancer cell line exhibited lower levels in most of the sites located between Tsn‐Gli2 (PRE07749 and PRE07750), Sctr‐Marco (PRE07757) and Insig2‐Ddx18 (PRE07758) (Figure 5B).

Figure 5.

Figure 5

Open in a new tab

CTCF binding along LRES region and global content in normal and cancer cells. A, CTCF binding in predicted sites along the 1qE2.3 region in mouse small intestine crypts. B, CTCF binding in predicted sites along the 1qE2.3 region in CT26 cancer cells was diminished relative to colon normal tissue. C, On the contrary, CT26 cells exhibited higher levels of total CTCF protein as revealed by Western Blot. D to F, LRES reversion during CaCo‐2 differentiation in 2q14.2 region is revealed by upregulation of gene expression and increase of H3K9ac and CTCF signals in CaCo‐2 differentiated cells for 11 and 20 days post‐confluence (dpc) as compared with CaCo‐2 cells (0 days post‐confluence). Asterisks denote bivalent genes. G, Western blot of CTCF also demonstrated that total levels of CTCF protein are reduced during the cell differentiation.

Next, we investigated whether the decrease in binding could be due to changes in the total amount of CTCF. However, Western blot analysis revealed that global levels of CTCF are higher in CT26 cells (Figure 5C). Furthermore, most of CTCF in this colon cancer cell line corresponded to the faster migrating form of CTCF (CTCF‐130), whereas in normal tissue the slower migration form (CTCF‐180, polyADP‐ribosylated) was prevalent (see Section 4.4 for more information). In summary, although overall CTCF levels are increased in CT26 cells as compared with normal colon, its binding in the 1qE2.3 region is slightly depleted, which is consistent with a disturbance of the chromatin boundaries in LRES‐affected regions.

3.5. CTCF in LRES regions during CaCo‐2 cell differentiation

To further characterize the role of CTCF in the definition of chromatin domains along LRES regions we have used the CaCo‐2 differentiation to enterocyte‐like cell model. CaCo‐2 is a human colon cancer cell line and therefore we analyzed the 2q14.2 region, which is orthologous to mouse 1qE2.3. LRES of 2q14.2 was not fully achieved in CaCo‐2 cells as genes embedded in this region were slightly more expressed than in other cell lines (i.e. HCT116), and hypermethylation of CpG islands was incomplete: the SCTR CpG island was fully methylated, yet the EN1 and GLI2 CpG islands were only partially methylated (Supplementary Figure 9A) as revealed by clonal analysis (data not shown), and INHBB remained unmethylated (Supplementary Figure 9A).

In a first setting, differentiation of CaCo‐2 cells was induced by confluence and the molecular profiles were analyzed after 11, 20 and 40 days post‐confluence. In a second setting, CaCo‐2 differentiation was induced by treatment with butyric acid for 48 h, and cells were maintained in culture for 18 more days. Differentiated cells exhibited higher expression levels for most genes in both settings (Figure 5D and Supplementary Figure 9B–C). An exception was INHBB, the downregulation of which has been previously reported in differentiated cells (Saaf et al., 2007). Nevertheless, INHBB was also upregulated 28 days after butyric acid treatment (Supplementary Figure 9C). Throughout our experiments, no changes in DNA methylation were observed (Supplementary Figure 10). ChIP analyses revealed that the gain of expression was accompanied by an increase of H3K9 acetylation in the studied region (Figure 5E), yet no changes in other histone marks were observed (H3K4me3, H3K27me3, H3K9me2 and H3K9me3) except for INHBB, which showed increased levels of H3K27me3 in consistence with its downregulation (Supplementary Figure 11).

Then, we analyzed CTCF binding in six sites previously described (Barski et al., 2007; Kim et al., 2007) and distributed along the 4 Mb region. Neighboring regions not predicted as CTCF binding sites were considered as negative controls, and produced almost undetectable signals (Supplementary Figure 8B). As the enrichment was often at the limit of the technical detection, these results must be interpreted with caution. A gain of insulator binding was observed in differentiated CaCo‐2 cell in sites located between: INSIG2‐EN1 (ZHAO2821), EN1‐MARCO (REN1951), SCTR‐PTPN4 (ZHAO2829), RALBB‐INHBB (REN1964) and GLI2‐TSN (REN1972) (Figure 5F). On the contrary, Western blot analysis revealed that global levels of CTCF‐130 decrease during this process (Figure 5G).

In summary, regression of the malignant phenotype of CaCo‐2 cells by inducing cell differentiation was accompanied by reactivation of gene expression together with a moderate gain of CTCF binding along the 2q14.2 region, which could be interpreted as a partial reversion of the LRES and most probably a reinforcement of chromatin domain barriers.

4. Discussion

4.1. LRES is a trans‐species mechanism of gene inactivation in cancer

LRES has been well characterized in different types of cancers (Coolen et al., 2010, 2009, 2012, 2006, 2007, 2010, 2011, 2008, 2008, 2006), pointing out that it is a common mechanism in human tumorigenesis. Here we demonstrate for the first time the occurrence of LRES in mouse, establishing the feasibility of this model to get insights into the mechanisms and biological implications of this epigenetic alteration. Moreover, as the regions affected by LRES are orthologous to the human counterparts, it is highly possible that these epigenetic alterations share mechanisms and functional implications in mouse and human. This represents another example of epigenetic landscape conservation, similarly to histone patterns of the Hox region in lung fibroblasts (Bernstein et al., 2005), methylation changes in breast adenocarcinomas (Acosta et al., 2011) and the colocalization of different epigenetic marks in pluripotent stem cells (Xiao et al., 2012).

The existence of murine LRES with identical features than the human phenomenon allows us to study in depth the epigenomic landscape of the target regions in normal and tumor samples. Sequential ChIP experiments in normal tissue showed that low expressed genes maintained the chromatin bivalent marks described in embryonic stem cells (Mikkelsen et al., 2007) and tumor cells (Mayor et al., 2011; Ohm et al., 2007; Rodriguez et al., 2008). This implies that they are not resolved during differentiation, as it has been proposed for some bivalent domains during neural differentiation (Mohn et al., 2008), and hence retaining the regulatory potential. The low levels of DNA methylation observed in the En1 CpG island shore in normal intestine cells (progenitors and differentiated) could sustain a methylation seeding process for the initiation of LRES in early stages of carcinogenesis as previously hypothesized (Clark, 2007).

Several works have described the resolution of bivalent domains during the differentiation process (Meissner et al., 2008; Mohn et al., 2008) or tumor progression (Iliou et al., 2011) correlating DNA methylation with the loss of histone active marks. In the CT26 colorectal cell line we observed that DNA methylation was accompanied by a decrease in H3K4me3 and H3K27me3, but the histone mark ratio was maintained. Indeed, we could confirm the presence of bivalent domains by sequential ChIP. DNA hypermethylation was correlated with an increase of H3K9me2, in agreement with other studies (Fahrner et al., 2002; Kondo and Issa, 2003). Moreover, those genes presented an increase of EZH2 levels despite the lower levels of H3K27me3, which could be explained by a global decrease of H3K27me3 levels described in tumors (Wei et al., 2008). The increase of EZH2 binding may also drive the recruitment of DNMT and the subsequent DNA methylation (Vire et al., 2006).

4.2. LRES is restricted to advanced stages of malignancy

LRES has been well characterized in a few types of human tumors and especially in cancer cell lines. Few studies have investigated this phenomenon in early stages of carcinogenesis. Hence, we addressed this issue by characterizing the epigenetic landscape of LRES regions in tumors produced in transgenic mice with different genotypes.

Apc Min/+ mice develop benign adenomas displaying fewer chromosomal alterations than sporadic adenomas (Tarafa et al., 2003). The most striking finding in the murine intestinal adenomas was that low expressed bivalent genes exhibited a global increase in gene expression as compared with the normal intestine or the tissue adjacent to the adenoma (Figure 4A) and no changes in epigenetic marks (histone modifications and DNA methylation). A recent study has also reported that epigenetic changes are rare in these adenomas, although reproducing in part the tumor‐specific signature of human colon cancer (Grimm et al., 2013). However no epigenetic changes were observed in the two chromosomal regions we analyzed. Therefore, LRES does not appear to occur in very initial stages of carcinogenesis. On the contrary, most of the region appears to be activated, which could be considered a response (or perhaps a defense) of the cell against some of the biological processes associated with malignant transformation. This hypothesis is supported by data showing that overexpression of Inhbb could represent an antioncogenic cellular reaction related with its putative role as tumor suppressor (Choi and Han, 2011). Interestingly, another study found that the gene PCDHGC3, embedded in the 5q31 region undergoing LRES in human colorectal cancer, was overexpressed in adenomas even though epigenetic silencing of other genes was already achieved in these early stages (Dallosso et al., 2012), which also illustrates a dual deregulation during tumor progression.

More advanced stages of the disease were obtained in transgenic mice with floxed Apc, Tgfbr2, and Trp53. Polyps and carcinomas generated in the colon and small intestine of these mice presented gene expression profiles and epigenetic signatures consistent with LRES. However, some of the hypermethylated genes were not silenced, which is likely to result from the presence of mixed cell populations in developing tumors. Thus, DNA hypermethylation and gene silencing may occur preferentially in the more malign cells, whereas genes are likely to be overexpressed in the adenoma‐like cells, similar to the Apc Min/+ model. This interpretation is supported by the heterogenous methylation profiles of clones (Supplementary Figure 5).

The parallelism of the gene expression and the epigenetic profiles of these two regions between human and mouse intestinal cancers illustrates that common mechanisms are likely to be involved in the malignant transformation in both species. The lack of genome‐wide epigenetic studies in murine colon tumors precludes further conclusions and future studies should clarify the extent and frequency of LRES in this model.

Our studies demonstrate that genes located in these regions are not uniformly regulated during normal intestinal differentiation, which indicates that the coordinated downregulation observed in LRES is a cancer specific phenomenon, at least in the model considered here. The dynamic expression profile is coherent with the participation of many of these genes in the regulation of different functions related with the control of proliferation and cell differentiation. For example, Cplx2 encodes a cytosolic protein involved in the exocytosis of synaptic vesicles (Raevskaya et al., 2005) and the opposite expression pattern of Cplx2 variants could suggest a possible role in the signal transmission required for maintaining the intestinal architecture. Furthermore, Cdhr2 (also known as Protocadherin 24 or pcLKC) overexpression may facilitate cellular polarization (Krahn et al., 2010) and contact inhibition through the β‐Catenin pathway (Ose et al., 2009).

4.3. Regression of malignant phenotype results in partial reversion of LRES

Although the mouse 1qE2.3 chromosomal region (orthologous to human 2q14.2 region) exhibited an uneven regulation along the crypt/villus axis, an apparently coordinated overexpression was observed during in vitro differentiation of the human CaCo‐2 cell line to an enterocyte‐like population (Saaf et al., 2007). CaCo‐2 DNA methylation patterns of the 2q14.2 chromosomal region resembled that found in colon cancers with incomplete LRES (Mayor et al., 2009). Gene expression and histone modification profiles showed a partial reactivation of the region during the differentiation process (Figure 5), which would imply a reversion of the LRES process. The lack of changes in DNA methylation result in an apparent inconsistency, as SCTR exhibited a fully and stably methylated promoter CpG island, but the gene was reexpressed and showed increased levels of H3K9ac in enterocyte‐like cells. It should be noted that there is an alternative isoform of SCTR with the TSS downstream of the full transcript and not associated with a CpG island, it remains to be determined whether the activation of this alternative promoter could explain the observed changes. The functional relevance of this isoform is unknown (Long et al., 2007).

4.4. Chromatin insulators are involved in LRES

CTCF binding inside the region silenced by LRES was very low in both mouse and human colorectal cancer cell lines (CT26 and CaCo‐2) as compared with their normal counterparts (normal colon and enterocyte like‐cells, respectively), in which LRES was not acting or was partially reversed (Figure 5). Thus, LRES is accompanied by the loss or weakening of regional boundaries, which contrasts with the increase in the global content of CTCF. Most of CTCF in colon cancer cell line CT26 corresponded to the faster migrating form of CTCF (CTCF‐130), whereas in normal tissue, the slower migration form (CTCF‐180, polyADP‐ribosylated) was prevalent (Figure 5C). High levels of total CTCF‐130 have been also reported in breast tumors and an anti‐apoptotic role has been postulated (Docquier et al., 2005, 2009, 2005). On the other hand, CaCo‐2 differentiation is characterized by a global decrease of CTCF‐130, which is consistent with results obtained in differentiated layers of epidermis (Rosa‐Garrido et al., 2012) and in myeloid leukemia cells (Delgado et al., 1999). In mouse samples, our results concur with previous studies that have reported the predominance of the CTCF‐180 form (contains more than 20 ADP‐ribose residues) in normal cells, whereas CTCF‐130 (non‐modified or few ADP‐ribose residues) is the foremost form in cell lines and tumor tissues (Docquier et al., 2009; Farrar et al., 2010).

5. Final considerations

The biological consequences of LRES in cancer cells are mostly unknown. Silencing of two genes embedded in LRES (INHBB and PCDHGC3) has been correlated with poor prognosis in colorectal cancer (Dallosso et al., 2012; Mayor et al., 2009), but this association does not necessarily imply a direct impact of these alterations on cellular biology. Data reported here reinforce the role of LRES as a mechanism of gene inactivation in cancer and illustrate a panorama in which these regions are plastically regulated during physiological processes, but are constrained to a coordinated repression by abolishing the autonomous regulation in cancer progression (Figure 6). The disruption of regional boundaries, probably by loss of CTCF binding as our data show, suggests that chromatin remodeling during LRES is accompanied by nuclear restructuring of the affected chromosomal regions. Further insights into the 3D chromatin organization should contribute to a better understanding of mechanisms underlying epigenetic regulation of these regions in physiological and pathological conditions.

Figure 6.

Figure 6

Open in a new tab

A model illustrating the epigenetic landscape and dynamics of chromosomal regions affected by LRES. Genes embedded in these regions are independently and plastically regulated during physiological processes, as cell differentiation, due to the maintenance of autonomous chromatin domains by CTCF. We postulate that in cancer cells, the coordinated silencing is achieved by a local reduction of CTCF, chromatin remodeling and accompanied by nuclear reorganization of LRES regions into suppressive domains. In Caco‐2 cells, regression of malignancy by induced cell differentiation partially restores the active compartmentalization of chromatin resulting in the LRES region.

Conflict of interest

Authors declare no conflict of interest.

Supporting information

The following is the supplementary data related to this article:

Supplementary data

Click here for additional data file. (838.6KB, pdf)

Acknowledgments

We thank Marta Vives and Rosa Ampudia (Laboratori d'Immunobiologia per a la Recerca i les Aplicacions Diagnòstiques ‐LIRAD, Badalona) and Pura Muñoz‐Cànoves (Universitat Pompeu Fabra, Barcelona) for kindly providing mouse tissues; and Marcus Buschbeck for helpful advices and Sergio Palomo for technical assistance. MF was supported by a fellowship from the Ministry of Science and Innovation (FPI BES2007‐16580), DVFT holds a Juan de la Cierva postdoctoral fellowship and AM‐S held an AECC postdoctoral fellowship. This work was supported by grants from the Ministry of Science and Innovation (SAF2008/1409, SAF2011/23638 and CSD2006/49), and the Generalitat de Catalunya (2009 SGR 1356) to MAP; and an ERC Starting grant to EB.

Supplementary data 1.

Supplementary data related to this article can be found at http://dx.doi.org/10.1016/j.molonc.2013.08.008.

Forn Marta, Muñoz Mar, Tauriello Daniele V.F., Merlos-Suárez Anna, Rodilla Verónica, Bigas Anna, Batlle Eduard, Jordà Mireia, Peinado Miguel A., (2013), Long range epigenetic silencing is a trans‐species mechanism that results in cancer specific deregulation by overriding the chromatin domains of normal cells, Molecular Oncology, 7, doi: 10.1016/j.molonc.2013.08.008.

References

  1. Acosta, D. , Suzuki, M. , Connolly, D. , Thompson, R.F. , Fazzari, M.J. , Greally, J.M. , Montagna, C. , 2011. DNA methylation changes in murine breast adenocarcinomas allow the identification of candidate genes for human breast carcinogenesis. Mamm. Genome. 22, 249–259. [DOI] [PubMed] [Google Scholar]
  2. Andrews, N.C. , Faller, D.V. , 1991. A rapid micropreparation technique for extraction of DNA-binding proteins from limiting numbers of mammalian cells. Nucleic Acids Res.. 19, 2499 [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Azuara, V. , Perry, P. , Sauer, S. , Spivakov, M. , Jorgensen, H.F. , John, R.M. , Gouti, M. , Casanova, M. , Warnes, G. , Merkenschlager, M. , Fisher, A.G. , 2006. Chromatin signatures of pluripotent cell lines. Nat. Cell Biol.. 8, 532–538. [DOI] [PubMed] [Google Scholar]
  4. Barker, N. , Ridgway, R.A. , van Es, J.H. , van de Wetering, M. , Begthel, H. , van den Born, M. , Danenberg, E. , Clarke, A.R. , Sansom, O.J. , Clevers, H. , 2009. Crypt stem cells as the cells-of-origin of intestinal cancer. Nature. 457, 608–611. [DOI] [PubMed] [Google Scholar]
  5. Barski, A. , Cuddapah, S. , Cui, K. , Roh, T.Y. , Schones, D.E. , Wang, Z. , Wei, G. , Chepelev, I. , Zhao, K. , 2007. High-resolution profiling of histone methylations in the human genome. Cell. 129, 823–837. [DOI] [PubMed] [Google Scholar]
  6. Berman, B.P. , Weisenberger, D.J. , Aman, J.F. , Hinoue, T. , Ramjan, Z. , Liu, Y. , Noushmehr, H. , Lange, C.P. , van Dijk, C.M. , Tollenaar, R.A. , 2012. Regions of focal DNA hypermethylation and long-range hypomethylation in colorectal cancer coincide with nuclear lamina-associated domains. Nat. Genet.. 44, 40–46. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Bernstein, B.E. , Kamal, M. , Lindblad-Toh, K. , Bekiranov, S. , Bailey, D.K. , Huebert, D.J. , McMahon, S. , Karlsson, E.K. , Kulbokas, E.J. , Gingeras, T.R. , 2005. Genomic maps and comparative analysis of histone modifications in human and mouse. Cell. 120, 169–181. [DOI] [PubMed] [Google Scholar]
  8. Bernstein, B.E. , Mikkelsen, T.S. , Xie, X. , Kamal, M. , Huebert, D.J. , Cuff, J. , Fry, B. , Meissner, A. , Wernig, M. , Plath, K. , 2006. A bivalent chromatin structure marks key developmental genes in embryonic stem cells. Cell. 125, 315–326. [DOI] [PubMed] [Google Scholar]
  9. Clark, S.J. , 2007. Action at a distance: epigenetic silencing of large chromosomal regions in carcinogenesis. Hum. Mol. Genet.. 16, (1) R88–R95. [DOI] [PubMed] [Google Scholar]
  10. Coolen, M.W. , Stirzaker, C. , Song, J.Z. , Statham, A.L. , Kassir, Z. , Moreno, C.S. , Young, A.N. , Varma, V. , Speed, T.P. , Cowley, M. , 2010. Consolidation of the cancer genome into domains of repressive chromatin by long-range epigenetic silencing (LRES) reduces transcriptional plasticity. Nat. Cell Biol.. 12, 235–246. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Choi, S.C. , Han, J.K. , 2011. Negative regulation of activin signal transduction. Vitam. Horm.. 85, 79–104. [DOI] [PubMed] [Google Scholar]
  12. Dallosso, A.R. , Hancock, A.L. , Szemes, M. , Moorwood, K. , Chilukamarri, L. , Tsai, H.H. , Sarkar, A. , Barasch, J. , Vuononvirta, R. , Jones, C. , 2009. Frequent long-range epigenetic silencing of protocadherin gene clusters on chromosome 5q31 in Wilms' tumor. PLoS Genet.. 5, e1000745 [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Dallosso, A.R. , Oster, B. , Greenhough, A. , Thorsen, K. , Curry, T.J. , Owen, C. , Hancock, A.L. , Szemes, M. , Paraskeva, C. , Frank, M. , 2012. Long-range epigenetic silencing of chromosome 5q31 protocadherins is involved in early and late stages of colorectal tumorigenesis through modulation of oncogenic pathways. Oncogene. 31, 4409–4419. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Delgado, M.D. , Chernukhin, I.V. , Bigas, A. , Klenova, E.M. , Leon, J. , 1999. Differential expression and phosphorylation of CTCF, a c-myc transcriptional regulator, during differentiation of human myeloid cells. FEBS Lett.. 444, 5–10. [DOI] [PubMed] [Google Scholar]
  15. Docquier, F. , Farrar, D. , D'Arcy, V. , Chernukhin, I. , Robinson, A.F. , Loukinov, D. , Vatolin, S. , Pack, S. , Mackay, A. , Harris, R.A. , 2005. Heightened expression of CTCF in breast cancer cells is associated with resistance to apoptosis. Cancer Res.. 65, 5112–5122. [DOI] [PubMed] [Google Scholar]
  16. Docquier, F. , Kita, G.X. , Farrar, D. , Jat, P. , O'Hare, M. , Chernukhin, I. , Gretton, S. , Mandal, A. , Alldridge, L. , Klenova, E. , 2009. Decreased poly(ADP-ribosyl)ation of CTCF, a transcription factor, is associated with breast cancer phenotype and cell proliferation. Clin. Cancer Res.. 15, 5762–5771. [DOI] [PubMed] [Google Scholar]
  17. Esteller, M. , 2008. Epigenetics in cancer. N. Engl. J. Med.. 358, 1148–1159. [DOI] [PubMed] [Google Scholar]
  18. Fahrner, J.A. , Eguchi, S. , Herman, J.G. , Baylin, S.B. , 2002. Dependence of histone modifications and gene expression on DNA hypermethylation in cancer. Cancer Res.. 62, 7213–7218. [PubMed] [Google Scholar]
  19. Farrar, D. , Rai, S. , Chernukhin, I. , Jagodic, M. , Ito, Y. , Yammine, S. , Ohlsson, R. , Murrell, A. , Klenova, E. , 2010. Mutational analysis of the poly(ADP-ribosyl)ation sites of the transcription factor CTCF provides an insight into the mechanism of its regulation by poly(ADP-ribosyl)ation. Mol. Cell Biol.. 30, 1199–1216. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Frigola, J. , Song, J. , Stirzaker, C. , Hinshelwood, R.A. , Peinado, M.A. , Clark, S. , 2006. Epigenetic remodeling in colorectal cancer results in coordinate gene suppression across an entire chromosome band. Nat. Genet.. 38, 540–549. [DOI] [PubMed] [Google Scholar]
  21. Grimm, C. , Chavez, L. , Vilardell, M. , Farrall, A.L. , Tierling, S. , Bohm, J.W. , Grote, P. , Lienhard, M. , Dietrich, J. , Timmermann, B. , 2013. DNA-methylome analysis of mouse intestinal adenoma identifies a tumour-specific signature that is partly conserved in human colon cancer. PLoS Genet.. 9, e1003250 [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Hansen, K.D. , Timp, W. , Bravo, H.C. , Sabunciyan, S. , Langmead, B. , McDonald, O.G. , Wen, B. , Wu, H. , Liu, Y. , Diep, D. , 2011. Increased methylation variation in epigenetic domains across cancer types. Nat. Genet.. 43, 768–775. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Hitchins, M.P. , Lin, V.A. , Buckle, A. , Cheong, K. , Halani, N. , Ku, S. , Kwok, C.T. , Packham, D. , Suter, C.M. , Meagher, A. , 2007. Epigenetic inactivation of a cluster of genes flanking MLH1 in microsatellite-unstable colorectal cancer. Cancer Res.. 67, 9107–9116. [DOI] [PubMed] [Google Scholar]
  24. Hsu, P.Y. , Hsu, H.K. , Singer, G.A.C. , Yan, P.S. , Rodriguez, B.A.T. , Liu, J.C. , Weng, Y.I. , Deatherage, D.E. , Chen, Z. , Pereira, J.S. , 2010. Estrogen-mediated epigenetic repression of large chromosomal regions through DNA looping. Genome Res.. 20, 733–744. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Iliou, M.S. , Lujambio, A. , Portela, A. , Brustle, O. , Koch, P. , Andersson-Vincent, P.H. , Sundstrom, E. , Hovatta, O. , Esteller, M. , 2011. Bivalent histone modifications in stem cells poise miRNA loci for CpG island hypermethylation in human cancer. Epigenetics. 6, 1344–1353. [DOI] [PubMed] [Google Scholar]
  26. Irizarry, R.A. , Ladd-Acosta, C. , Wen, B. , Wu, Z. , Montano, C. , Onyango, P. , Cui, H. , Gabo, K. , Rongione, M. , Webster, M. , 2009. The human colon cancer methylome shows similar hypo- and hypermethylation at conserved tissue-specific CpG island shores. Nat. Genet.. 41, 178–186. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Javierre, B.M. , Rodriguez-Ubreva, J. , Al-Shahrour, F. , Corominas, M. , Grana, O. , Ciudad, L. , Agirre, X. , Pisano, D.G. , Valencia, A. , Roman-Gomez, J. , 2011. Long-range epigenetic silencing associates with deregulation of Ikaros targets in colorectal cancer cells. Mol. Cancer Res.. 9, 1139–1151. [DOI] [PubMed] [Google Scholar]
  28. Jones, P.A. , Baylin, S.B. , 2007. The epigenomics of cancer. Cell. 128, 683–692. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Kim, T.H. , Abdullaev, Z.K. , Smith, A.D. , Ching, K.A. , Loukinov, D.I. , Green, R.D. , Zhang, M.Q. , Lobanenkov, V.V. , Ren, B. , 2007. Analysis of the vertebrate insulator protein CTCF-binding sites in the human genome. Cell. 128, 1231–1245. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Kondo, Y. , Issa, J.P. , 2003. Enrichment for histone H3 lysine 9 methylation at Alu repeats in human cells. J. Biol. Chem.. 278, 27658–27662. [DOI] [PubMed] [Google Scholar]
  31. Krahn, M.P. , Rizk, S. , Alfalah, M. , Behrendt, M. , Naim, H.Y. , 2010. Protocadherin of the liver, kidney, and colon associates with detergent-resistant membranes during cellular differentiation. J. Biol. Chem.. 285, 13193–13200. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Long, S.H. , Berna, M.J. , Thill, M. , Pace, A. , Pradhan, T.K. , Hoffmann, K.M. , Serrano, J. , Jensen, R.T. , 2007. Secretin-receptor and secretin-receptor-variant expression in gastrinomas: correlation with clinical and tumoral features and secretin and calcium provocative test results. J. Clin. Endocrinol. Metab.. 92, 4394–4402. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Mayor, R. , Casadome, L. , Azuara, D. , Moreno, V. , Clark, S.J. , Capella, G. , Peinado, M.A. , 2009. Long-range epigenetic silencing at 2q14.2 affects most human colorectal cancers and may have application as a non-invasive biomarker of disease. Br. J. Cancer. 100, 1534–1539. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Mayor, R. , Munoz, M. , Coolen, M.W. , Custodio, J. , Esteller, M. , Clark, S.J. , Peinado, M.A. , 2011. Dynamics of bivalent chromatin domains upon drug induced reactivation and resilencing in cancer cells. Epigenetics. 6, 1138–1148. [DOI] [PubMed] [Google Scholar]
  35. Meissner, A. , Mikkelsen, T.S. , Gu, H. , Wernig, M. , Hanna, J. , Sivachenko, A. , Zhang, X. , Bernstein, B.E. , Nusbaum, C. , Jaffe, D.B. , 2008. Genome-scale DNA methylation maps of pluripotent and differentiated cells. Nature. 454, 766–770. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Merlos-Suarez, A. , Barriga, F.M. , Jung, P. , Iglesias, M. , Cespedes, M.V. , Rossell, D. , Sevillano, M. , Hernando-Momblona, X. , da Silva-Diz, V. , Munoz, P. , 2011. The intestinal stem cell signature identifies colorectal cancer stem cells and predicts disease relapse. Cell Stem Cell. 8, 511–524. [DOI] [PubMed] [Google Scholar]
  37. Mikkelsen, T.S. , Ku, M. , Jaffe, D.B. , Issac, B. , Lieberman, E. , Giannoukos, G. , Alvarez, P. , Brockman, W. , Kim, T.K. , Koche, R.P. , 2007. Genome-wide maps of chromatin state in pluripotent and lineage-committed cells. Nature. 448, 553–560. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Mohn, F. , Weber, M. , Rebhan, M. , Roloff, T.C. , Richter, J. , Stadler, M.B. , Bibel, M. , Schubeler, D. , 2008. Lineage-specific polycomb targets and de novo DNA methylation define restriction and potential of neuronal progenitors. Mol. Cell. 30, 755–766. [DOI] [PubMed] [Google Scholar]
  39. Novak, P. , Jensen, T. , Oshiro, M.M. , Watts, G.S. , Kim, C.J. , Futscher, B.W. , 2008. Agglomerative epigenetic aberrations are a common event in human breast cancer. Cancer Res.. 68, 8616–8625. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Ohm, J.E. , McGarvey, K.M. , Yu, X. , Cheng, L. , Schuebel, K.E. , Cope, L. , Mohammad, H.P. , Chen, W. , Daniel, V.C. , Yu, W. , 2007. A stem cell-like chromatin pattern may predispose tumor suppressor genes to DNA hypermethylation and heritable silencing. Nat. Genet.. 39, 237–242. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Ose, R. , Yanagawa, T. , Ikeda, S. , Ohara, O. , Koga, H. , 2009. PCDH24-induced contact inhibition involves downregulation of beta-catenin signaling. Mol. Oncol.. 3, 54–66. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Park, J.H. , Park, J. , Choi, J.K. , Lyu, J. , Bae, M.G. , Lee, Y.G. , Bae, J.B. , Park, D.Y. , Yang, H.K. , Kim, T.Y. , Kim, Y.J. , 2011. Identification of DNA methylation changes associated with human gastric cancer. BMC Med. Genomics. 4, 82 [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Portela, A. , Esteller, M. , 2010. Epigenetic modifications and human disease. Nat. Biotechnol.. 28, 1057–1068. [DOI] [PubMed] [Google Scholar]
  44. Raevskaya, N.M. , Dergunova, L.V. , Vladychenskaya, I.P. , Stavchansky, V.V. , Oborina, M.V. , Poltaraus, A.B. , Limborska, S.A. , 2005. Structural organization of the human complexin 2 gene (CPLX2) and aspects of its functional activity. Gene. 359, 127–137. [DOI] [PubMed] [Google Scholar]
  45. Rodilla, V. , Villanueva, A. , Obrador-Hevia, A. , Robert-Moreno, A. , Fernandez-Majada, V. , Grilli, A. , Lopez-Bigas, N. , Bellora, N. , Alba, M.M. , Torres, F. , 2009. Jagged1 is the pathological link between Wnt and Notch pathways in colorectal cancer. Proc. Natl. Acad. Sci. U S A. 106, 6315–6320. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Rodriguez, J. , Muñoz, M. , Vives, L. , Frangou, C.G. , Groudine, M. , Peinado, M.A. , 2008. Bivalent domains enforce transcriptional memory of DNA methylated genes in cancer cells. Proc. Natl. Acad. Sci. U S A. 105, 19809–19814. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Rosa-Garrido, M. , Ceballos, L. , Alonso-Lecue, P. , Abraira, C. , Delgado, M.D. , Gandarillas, A. , 2012. A cell cycle role for the epigenetic factor CTCF-L/BORIS. PLoS one. 7, e39371 [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Saaf, A.M. , Halbleib, J.M. , Chen, X. , Yuen, S.T. , Leung, S.Y. , Nelson, W.J. , Brown, P.O. , 2007. Parallels between global transcriptional programs of polarizing Caco-2 intestinal epithelial cells in vitro and gene expression programs in normal colon and colon cancer. Mol. Biol. Cell. 18, 4245–4260. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Schepers, A.G. , Snippert, H.J. , Stange, D.E. , van den Born, M. , van Es, J.H. , van de Wetering, M. , Clevers, H. , 2012. Lineage tracing reveals Lgr5+ stem cell activity in mouse intestinal adenomas. Science. 337, 730–735. [DOI] [PubMed] [Google Scholar]
  50. Stransky, N. , Vallot, C. , Reyal, F. , Bernard-Pierrot, I. , de Medina, S.G. , Segraves, R. , de Rycke, Y. , Elvin, P. , Cassidy, A. , Spraggon, C. , 2006. Regional copy number-independent deregulation of transcription in cancer. Nat. Genet.. 38, 1386–1396. [DOI] [PubMed] [Google Scholar]
  51. Tarafa, G. , Prat, E. , Risques, R.A. , Gonzalez, S. , Camps, J. , Grau, M. , Guino, E. , Moreno, V. , Esteller, M. , Herman, J.G. , 2003. Common genetic evolutionary pathways in familial adenomatous polyposis tumors. Cancer Res.. 63, 5731–5737. [PubMed] [Google Scholar]
  52. Torrano, V. , Chernukhin, I. , Docquier, F. , D'Arcy, V. , Leon, J. , Klenova, E. , Delgado, M.D. , 2005. CTCF regulates growth and erythroid differentiation of human myeloid leukemia cells. J. Biol. Chem.. 280, 28152–28161. [DOI] [PubMed] [Google Scholar]
  53. Vire, E. , Brenner, C. , Deplus, R. , Blanchon, L. , Fraga, M. , Didelot, C. , Morey, L. , Van Eynde, A. , Bernard, D. , Vanderwinden, J.M. , 2006. The Polycomb group protein EZH2 directly controls DNA methylation. Nature. 439, 871–874. [DOI] [PubMed] [Google Scholar]
  54. Wei, Y. , Xia, W. , Zhang, Z. , Liu, J. , Wang, H. , Adsay, N.V. , Albarracin, C. , Yu, D. , Abbruzzese, J.L. , Mills, G.B. , 2008. Loss of trimethylation at lysine 27 of histone H3 is a predictor of poor outcome in breast, ovarian, and pancreatic cancers. Mol. Carcinog.. 47, 701–706. [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Xiao, S. , Xie, D. , Cao, X. , Yu, P. , Xing, X. , Chen, C.C. , Musselman, M. , Xie, M. , West, F.D. , Lewin, H.A. , 2012. Comparative epigenomic annotation of regulatory DNA. Cell. 149, 1381–1392. [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

The following is the supplementary data related to this article:

Supplementary data

Click here for additional data file. (838.6KB, pdf)

Articles from Molecular Oncology are provided here courtesy of Wiley

RESOURCES