Skip to main content
PMC home page
Molecules and Cells logoLink to Molecules and Cells
. 2012 Jan 2;33(2):127–133. doi: 10.1007/s10059-012-2182-8

Expression Profiling after Induction of Demethylation in MCF-7 Breast Cancer Cells Identifies Involvement of TNF-α Mediated Cancer Pathways

Ju Hee Kim 1,3, Seongeun Kang 1,3, Tae Woo Kim 1, Lihong Yin 2, Ran Liu 2,*, Sun Jung Kim 1,*
PMCID: PMC3887717  PMID: 22228181

Abstract

Epigenetic methylation change is a major process that occurs during cancer development. Even though many tumor-related genes have been identified based on their relationship between methylation and expression, few studies have been conducted to investigate the relevant biological pathways involved in these changes. To identify essential pathways likely to be affected by methylation in breast cancer, we examined a pool of genes in which expression was upregulated after induction of demethylation by 5-Aza-2′-deoxycytidine (Aza) in the MCF-7 breast cancer cell line. Genome-wide demethylation was confirmed by monitoring the demethylation of a previously known gene, SULT1A1. Overall, 210 and 213 genes were found to be upregulated and downregulated (fold change ≥ 2), respectively, in common in cells treated with 5 and 10 μM of Aza. Network analysis of these 423 genes with altered expression patterns identified the involvement of a cancer related network of genes that were heavily regulated by TNF-α in breast tumorigenesis. Our results suggest that epigenetic dysregulation of cellular processes relevant to TNF-α-dependent apoptosis may be intimately involved in tumorigenesis in MCF-7 cells.

Keywords: 5-Aza-2′-deoxycytidine, breast cancer, genome-wide expression, MCF-7, TNF-α

INTRODUCTION

Methylation of CpG at the promoter has been shown to play a key role in regulating gene activity in cellular development, differentiation, apoptosis, and carcinogenesis (Kulis and Esteller, 2010; Liu et al., 2010; Maegawa et al., 2010). Indeed, many tumor suppressors and proto-oncogenes undergo activation or inactivation during carcinogenesis via altered methylation. RASSF1A, BRCA1, and 15-Hydroxyprostaglandin dehy-dogenase are tumor suppressors with CpGs that are frequently methylated in breast cancer (Matros et al., 2005; Wolf et al., 2006; Yan et al., 2003). Genome-wide screening of genes with altered methylation has been conducted in breast cancer tis-sues (Tommasi et al., 2009; Van der Auwera et al., 2010) and cell lines (Rodenhiser et al., 2008a) with the goal of identifying genes en masse. Through these approaches, novel tumor suppressors including HOXB13, HNF1B, and MOGAT2 have been newly identified. Systematic screening has made it possible to elucidate the network of genes that are epigenetically regulated in breast cancer. For example, Li et al. conducted an Illumina methylation array for ER+ and ER− tissues of breast cancer and found that the inflammatory response, connective tissue disorders pathway was overwhelming in the cancer (Li et al., 2010).

CpG islands are frequently found at or near the promoter, and methylation of the majority of CpGs has commonly been observed in inactivated genes (Shao et al., 2007). However, methylation in only a portion of the CpGs (Kim et al., 2010) or even a single site could have induced inactivation (Sohn et al., 2010). Even though methylation at the promoter has mainly been linked to gene inactivation, methylation in other regions such as introns or the 3′-UTR have also been shown to induce inactivation (Laux et al., 1999; Malumbres et al., 1999). The methylation of CpG is closely associated with the chromatin structure; therefore, demethylation alone could not have induced gene activation often. A methyltransferase inhibitor, 5-Aza-2′-deoxycytidine (Aza) has been widely applied to induce demethylation of the methylated CpG (Xiong et al., 2009). For example, this compound was utilized to confirm the restoration of gene activity that was caused by methylation. In some cases, treatment with Trichostatin A, a histone deacetylase inhibitor, together with Aza induced a synergistic upregulation, implying that methylation and chromatin structure are tightly linked (Rivenbark et al., 2006). Aza can also be utilized to identify novel genes that are epigenetically regulated. Examples in breast cancer include APAF-1, CEACAM6 and ZC3HDC1, which were found to be upregulated after treatment with Aza against the MCF-7 cancer cell; therefore, their expression was further examined in cancer tissues (Rivenbark et al., 2006; Xiong et al., 2009).

Genome-wide approaches have been confined to monitoring the methylation status using techniques such as differential methylation hybridization (Tommasi et al., 2009) and genotyping-based sequencing methods such as the Illumina methylation array (Li et al., 2010). The genome-wide approach using Aza has an advantage in that it can directly identify a pool of genes that are inactivated by methylation. However, only a few cases have been reported in cancer, and no network analysis has been conducted to the best knowledge of the authors.

In this study, we induced global demethylation in MCF-7 cells by Aza and conducted expression array analysis using a system harboring 30,968 genes. For 423 genes showing statistical significance, Ingenuity pathway analysis was conducted to reveal pivotal pathways that are epigenetically regulated. A cancer pathway in which TNF-α plays the central role was at the top pathway.

MATERIALS AND METHODS

Cell culture and 5-Aza-2′-deoxycytidine treatment

A human breast cancer cell line, MCF7, was purchased from the American Type Culture Collection (ATCC; USA) and grown in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% fetal bovine serum (FBS). Demethylation of the cytosine residues was achieved by exposing the cell to culture media containing a methyltransferase inhibitor, 5-Aza-2′-deoxycytidine (Aza) (Sigma, USA), at 5 and 10 μM for 72 h. Culture medium with or without treatment was changed every 24 h.

Bisulfite sequencing

Chromosomal DNA extracted from the tissues was subjected to bisulfite treatment using an EpiTect Bisulfite kit (Qiagen, USA). The bisulfite-treated DNA was subjected to poly-merase chain reaction (PCR) to amplify the 372-bp promoter region of SULT1A1 (nucleotides −503 to −131 of GenBank accession no. AB062428; transcriptional start site, +1) containing six CpG sites. The PCR conditions were as follows: 94°C for 2 min, followed by 30 cycles of 94°C for 20 s, 55°C for 20 s and 72°C for 30 s, with a final extension at 72°C for 5 min. The resulting products were purified using a Qiaex II gel extraction kit (Qiagen) and then sub-cloned into the pGEM-T vector. The DNA sequences were confirmed by analyzing each plasmid clone in both directions.

Methylation-specific PCR (MSP)

Chromosomal DNA was isolated from the cell culture in a 75 cm2 culture flask using a genomic DNA purification kit (Promega, USA) according to the manufacturer’s protocol. The extracted DNA was then eluted with 250 μl of distilled water. Sodium bisulfite modification of genomic DNA was conducted using an EpiTect Bisulfite kit (Qiagen) according to the manufacturer’s protocol using 0.1 mg of purified DNA. PCR primers for the selected genes (Supplementary Table S1) were designed using the Methprimer program (http://www.urogene.org/methprimer/index1.html). One primer set was specific for amplification of methylated DNA and one was designed for amplification of unmethylated DNA. Quantitative PCR was conducted using a Power SYBR Green Kit (Applied Biosystems, USA) according to the protocols of the manufacturer. To assign a quantitative measure to the level of methylation, a methylation index was calculated for each sample using the following formula: methylation index = [1 / (1 + 2−(CTu − CTme))] × 100%, as pre-viously described (Lu et al., 2007), where CTu is the average cycle threshold (CT) obtained from duplicate quantitative PCR analyses using the unmethylated primer set and CTme is the average CT obtained using the methylated primer pair.

Real time RT-PCR

Isolation of Total RNA from cell culture and reverse transcription were conducted as described previously (Kim et al., 2010). Expression levels of TNF-α was measured by real-time quantitative RT-PCR analysis. Duplicate reactions were performed for each sample using a Kapa SYBR Fast qPCR Kit (Kapa Biosystems, USA) with TNF-α-specific primers on an ABI 7300 instrument (Applied Biosystems). RNA quantity was normalized to GAPDH content, and gene expression was quantified according to the 2−ΔCt method.

Microarray analysis

Approximately 5 × 106 proliferating MCF-7 cells treated with or without Aza were harvested and total RNA was extracted using an RNeasy mini kit (Qiagen). The RNA was then used for microarray analysis in duplicate using a Phalanx Human OneArray chip (Digital-Genomics, Korea) containing 30,968 human cDNA samples. Differential expression values, calculated as +Cy3/Cy5, where Cy3Signal > Cy5Signal, or -Cy3/Cy5, where Cy3Signal < Cy5Signal, were compared between the duplicate experiments. Clones differentially regulated in both experiments that had a significant ratio of Cy3 to Cy5 (defined as > 2) were selected for further analysis. All array data have been uploaded to the Gene Expression Omnibus (GEO) database, and can be accessed via their website (http://www.ncbi.nlm.nih.gov/geo/; accession number GSE31441).

Pathway analysis

To identify pathways displaying methylation-specific altered expression patterns with potential roles in breast carcinogenesis, functional categorization and pathway construction were conducted using the Ingenuity Pathway Analysis (IPA) software tool produced by Ingenuity Systems. IPA utilizes an extensive database of functional interactions that are drawn from peer-reviewed publications and are manually maintained (Calvano et al., 2005). P-values for individual networks were obtained by comparing the likelihood of obtaining the same number of transcripts or greater in a random gene set as are actually present in the input file (i.e., the set of genes differentially expressed in the Aza treated MCF-7 cells) using Fischer’s exact test based on the hypergeometric distribution. The highest confidence functional network was designated as the top network.

Statistical analysis

The heatmap data were obtained using the TreeView software (http://taxonomy.zoology.gla.ac.uk/rod/treeview.html). A pool of genes that fit the top IPA network was submitted to the program. In the microarray analysis, observations with adjusted p-values equal to or greater than 0.05 were removed and thus precluded from further analysis. Following adjustment, remaining genes were defined as differentially expressed if they displayed at least a two-fold difference in expression levels between the control and Aza-treated groups to further reduce the number of false positive observations, and to enrich for biologically relevant expression changes. A student’s t-test was used to detect differences in the mean levels of methylation for selected genes, as well as the expression level between the control and Aza-treated cells. Statistical analyses were conducted using SPSS for Windows, release 17.0 (SPSS Inc., USA). A p < 0.05 was considered to be statistically significant.

RESULTS

Aza induced genome-wide demethylation in MCF-7 cells

To induce genome-wide demethylation of the MCF-7 cells, they were treated with a methyltransferase inhibitor, 5-Aza-2′-deoxycytidine (Aza), at 5 and 10 μM for 72 h. The efficacy of Aza was then monitored by measuring the methylation status of CpGs at the promoter of SULT1A1, which was previously identified as a breast cancer methylation marker (Kwon et al., 2006). To accomplish this, 372 bp of the promoter region containing six CpGs was amplified by PCR after bisulfite treatment of the chromosomal DNA. The PCR product was then subcloned into the pGEM-T plasmid and the methylation status of ten clones from the 10 μM Aza treated samples were determined. Overall, the methylation level of the Aza-treated cells was 38%, while the control cells showed 96% methylation (Fig. 1). These results implied that Aza successfully induced global hypomethylation in the MCF-7 cells.

Fig. 1.

Fig. 1.

Open in a new tab

Induction of demethylation by Aza in MCF-7 cells. The methylation status of CpG sites at the SULT1A1 promoter was determined by sequencing after subcloning of the bisulfate-PCR product into the pGEM-T plasmid. Schematic diagram of the CpG sites at the promoter is drawn on the top. Methylated and unmethylated CpGs are indicated as black and open circles, respectively.

TNF-α mediated pathways are involved in the network of methylation-altered genes

Genome-wide expression analysis was conducted using the RNA as probes isolated from MCF-7 cells treated with Aza. Overall, 577 genes and 846 genes from cells treated with 5 μM and 10 μM of Aza were found to be upregulated, while 967 genes and 1180 genes were found to be downregulated, respectively, with a cutoff of a two-fold change in expression. Overall, 210 upregulated and 213 downregulated genes fit the significance criteria and were observed in both Aza treatments. Top regulated genes were listed in Table 1. All of these genes were examined for functional interrelatedness using the Ingenuity Pathway Analysis software tool. The highest functional network (Score = 61) resulting from differential gene expression was designated as “Cellular Movement, Cellular Development, Cancer” (Figs. 2 and 3). Interestingly, TNF-α, the key extrinsic mediator of the TNF-induced model of cellular apoptosis, was most prominent in the network and appeared to function as a master signaling molecule of a network of all other gene transcripts potentially relevant to breast tumor development and progression. To examine whether the alteration of methylation and expression of TNF-α is involved in the breast tumorigenesis, they were monitored in the normal MCF-10A breast cell and the cancerous MCF-7 cell (Fig. 4). The methylation was low and the expression was upregulated in MCF-10A. These patterns are reminiscent of those in the Aza-treated MCF-7 cell, implying treatment of Aza to the cancer cell mimicked the normal cell. Cluster analysis of the 423 genes included in the top two networks is shown in Fig. 5.

Table 1.

Genes in top network displaying differential expression after treatment of the MCF-7 cells with Aza

Symbol Accession Description Fold change
5 μM Aza 10 μM Aza
CCL20 NM_004591.1 Chemokine (C-C motif) ligand 20 2.73 2.41
CFB NM_001710.3 Complement factor B 3.07 2.53
CNN1 NM_001299.3 Calponin 1, basic, smooth muscle −6.00 −10.0
CTSS NM_004079.3 Cathepsin S −2.50 −2.25
CXCL1 NM_001511.1 Chemokine (C-X-C motif) ligand 1 (melanoma growth stimulating activity, alpha) 2.44 3.33
CXCL2 NM_002089.1 Chemokine (C-X-C motif) ligand 2 2.28 2.19
DLEU1 AW303167 Deleted in lymphocytic leukemia, 1 −2.25 −10.0
EBF1 CR627349 Early B-cell factor 1 2.50 2.60
FBXO32 NM_058229.2 F-box protein 32 2.02 2.62
GBP1 NM_018284.1 Guanylate binding protein 1, interferon-inducible, 67 kDa 2.34 2.83
HSD11B1 NM_005525.2 Hydroxysteroid (11-beta) dehydrogenase 1 −3.50 −2.25
ICAM1 NM_000201.1 Intercellular adhesion molecule 1 (CD54), human rhinovirus receptor 2.26 2.26
IRF7 NM_004029.1 Interferon regulatory factor 7 2.01 2.21
ISG15 NM_005101.1 ISG15 ubiquitin-like modifier 2.28 2.42
LCN2 NM_005564.2 Lipocalin 2 (oncogene 24p3) 4.29 2.71
LXN NM_020169.2 Latexin 3.21 2.61
MEOX2 NM_005924.3 Mesenchyme homeo box 2 (growth arrest-specific homeo box) −2.67 −2.50
MVP NM_005115.3 Major vault protein 2.50 2.54
MX2 NM_002463.1 Myxovirus (influenza virus) resistance 2 (mouse) 2.32 2.72
NEFH NM_021076.2 Neurofilament, heavy polypeptide 200 kDa 2.22 2.20
NOTCH4 NM_004557.3 Notch homolog 4 (Drosophila) −3.00 −5.00
OAS1 NM_002534.1 2′,5′-oligoadenylate synthetase 1, 40/46 kDa 3.47 2.56
ORM1/ORM2 NM_000607.1 Orosomucoid 1 5.56 4.52
POSTN NM_006475.1 Periostin, osteoblast specific factor −3.00 −3.00
S1PR3 NM_001001938.1 Endothelial differentiation, sphingolipid G-protein-coupled receptor, 3 2.66 2.05
SAMD4A BX092102 Sterile alpha motif domain containing 4A −2.50 −2.50
SERPINF1 NM_002615.4 Serpin peptidase inhibitor, clade F (alpha-2 antiplasmin, pigment epithelium derived factor), member 1 3.34 2.95
SP100 AK160379 SP100 nuclear antigen 2.86 2.43
ST6GAL1 NM_173216.1 ST6 beta-galactosamide alpha-2,6-sialyltranferase 1 4.60 2.87
SYNPO AB028952 Synaptopodin 2.91 2.76
TFPI2 NM_006528.2 Tissue factor pathway inhibitor 2 2.01 2.82
TIMP3 NM_000362.3 Tissue inhibitor of metalloprotei--se 3 (Sorsby fundus dystrophy, pseudoinflammatory) 4.60 4.41
TNF NM_000594.2 Tumor necrosis factor (TNF superfamily, member 2) 2.08 2.75
TNFSF10 NM_003810.2 Tumor necrosis factor (ligand) superfamily, member 10 2.07 2.13
TP53INP1 NM_033285.2 Tumor protein p53 inducible nuclear protein 1 3.99 6.85
Open in a new tab

Fig. 2.

Fig. 2.

Open in a new tab

Highest confidence network of genes displaying altered expression levels in response to demethylation in MCF-7 cells. According to IPA, the network is relevant to ‘Cellular Movement, Cellular Development, Cancer’. Genes that were upregulated in the Aza-treated MCF-7 cells are red, while those that were downregulated are green, and the intensity of the color reflects the magnitude of expression change. Each interaction is supported by at least one literature reference, with solid lines representing direct interactions, and dashed lines representing indirect interactions.

Fig. 3.

Fig. 3.

Open in a new tab

Pathways most strongly associated with the significantly altered genes in the Aza-treated MCF-7 cells. (A) Top functional categories. (B) Canonical pathways. The Ingenuity software assigns a P-value based on the likelihood of obtaining the observed number of category or pathway-related molecules in a given data set by chance alone. The threshold line denotes the P = 0.05 level. The line graph represents the ratio of affected genes to the total number of genes in a pathway.

Fig. 4.

Fig. 4.

Open in a new tab

Differential methylation and expression of TNF-α in normal and cancer cell lines. Real-time MSP (A) and real-time RT-PCR analysis (B) of TNF-α were performed in the normal cell line, MCF-10A, and the cancer cell line, MCF-7. Each sample was examined in duplicate and the average relative methylation and expression levels are presented.

Fig. 5.

Fig. 5.

Open in a new tab

Gene expression profiles of genes with altered expression in response to treatment of the MCF-7 cells with Aza. Heatmap analysis was conducted for the 59 genes that were included in the top two networks of IPA pathway analysis. Each row represents a gene, and each column represents the MCF-7 cells treated with 5 or 10 μM of Aza. As shown in the color bar, black represents no change, red represents upregulation, and green represents downregulation of gene expression.

The transcripts displaying the highest levels of altered expression within this network were TIMP3 (4.6-fold increase in 5 μM Aza treated cells and 4.4-fold increase in 10 μM Aza treated cells) and TP53INP1 (4.0-fold increase in 5 μM treated cells and 6.9-fold increase in 10 μM Aza treated cells) (Table 1). TIMP3 has been shown to inhibit tumor growth, invasion and angiogenesis, and has previously been established as a hypermethylated and downregulated gene in breast cancer (Lui et al., 2005). TP53INP1, which plays a role in both cell cycle arrest and p53-mediated apoptosis (Weng et al., 2011), has been shown to be decreased in melanoma and its reduced expression was correlated with hypermethylation of a promoter (Bonazzi et al., 2009). Notable cancer-related genes within the TNF-α-dependent apoptosis pathways that were upregulated in the Aza-treated MCF-7 cells included TNFSF10 (2.1-fold increase in both 5 and 10 μM Aza treated cells) and ICAM1 (2.3-fold increase in both 5 and 10 μM Aza treated cells). TNFSF10, which is known to be a TNF-related apoptosis-inducing ligand, has been found to induce apoptosis in cancer cells, and its receptors (DcR1 and DcR2) had an aberrant methylation pattern in breast cancer (Shivapurkar et al., 2004). ICAM1 has been shown to play a role as a cell adhesion molecule and is related to immune responses. TNF-α has been found to reinduce expression of ICAM1, which was downregulated in breast cancer (Budinsky et al., 1997). Significantly downregulated cancer-relevant genes included POSTN (3.0-fold decrease in both 5 and 10 μM Aza treated cells) and NOTCH4 (3.0-fold decrease in 5 μM Aza treated cells and 5.0-fold decrease in 10 μM Aza treated cells). POSTN has been shown to be overexpressed in breast cancer (Zhang et al., 2010) and is included in pathways to promote cancer cell survival, epithelial-mesenchymal transition (EMT), invasion, and metastasis (Ruan et al., 2009). Expression of NOTCH4 was found to lead to tumor formation in mammary glands (Politi et al., 2004) and its signaling was shown to be induced by TNF (Ando et al., 2003).

Genes in the TNF-α mediated pathway are frequently hypermethylated

To determine if genes in the pathways displayed consistency between mRNA levels and DNA methylation in the Aza-treated MCF-7 cells, ten genes were randomly selected and their methylation status was examined by MSP analysis. Seven of these were upregulated genes and three were downregulated genes. Four (ICAM1, TIMP3, TNF and TNFSF10) were previously established as methylation markers in cancer or other diseases, while six (CNN1, HSD11B1, ISG15, NOTCH4, SERPINF1, and S1RP3) were novel genes for which the methylation status is not known. As shown in Fig. 6, six of the seven upregulated genes showed hypomethylation in the Aza-treated MCF-7 cells in a concentration-dependent manner. The six downregulated genes did not show any consistent methylation profile.

Fig. 6.

Fig. 6.

Open in a new tab

MSP analysis of selected genes displaying altered expression in response to demethylation in the MCF-7 cells. Seven of the upregulated and six of the downregulated genes were randomly selected from the top network of IPA pathway and their methylation was examined by methylation-specific PCR. Each sample was analyzed in duplicate and the average relative methylation level is presented.

DISCUSSION

This study was carried out to screen differentially expressed genes conferred by CpG methylation and thereby elucidate essential regulatory pathways in MCF-7 cells. 210 upregulated and 213 downregulated genes were observed in common from 5 μM and 10 μM Aza-treated cells. These relatively small numbers of genes are possibly due to the fact that Aza not only induces the re-expression of silenced genes through the demethylation of CpG islands, but it increases the expression of unmethylated genes. The agent also induces cell stress by DNA damage (Xiong and Epstein, 2009).

It is notable that the TNF-α pathway, which was most prominent in the MCF-7 cells, also ranks among the top networks in other breast cancer cells and other cancer tissues. For example, in a metastatic variant of the breast cancer cell line, MDA-MB-468LN, the genome-wide methylation profile identified the TNF-pathway as the second highest network (Chambers and Goss, 2008). In the case of prostate cancer, the TNF-α pathway was the top network in a genome-wide methyaltion profile (Kim et al., 2011). TNF-α, a pleiotropic cytokine whose expression has been detected in breast tissue, is the key mediator of TNF-α-dependent apoptosis pathways, and has been found to both contribute to and protect against breast tumorigenesis (Yin et al., 2009). The TNF-α gene promoter frequently displayed DNA methylation in T-cell lymphoma (Zhang et al., 2009). In addition, the cytokine is involved in apoptosis, cell survival, inflammation, and immunity, and acts via two receptors (Dempsey et al., 2003; van Horssen et al., 2006). Its apoptotic activity is mediated through activation of death domain proteins such as TRADD and FADD, and caspases. For cell proliferation, TNF-α signals to Fos/Jun through the MAPK and JNK pathways, or to the NF-κB pathway. Our pathway analysis indicated that all three TNF-α-mediated apoptosis-relevant pathways (the MAPK pathway, NF-κB pathway, and the TNF-α-induced caspase-dependent death signaling pathway) are epigenetically regulated in MCF-7 cells, MDA-MB-468LN cells and prostate cancer. This fact implies that there is a precise signaling mechanism that determines the cellular fate as either proliferation or death.

It should be emphasized that the composites of interacting proteins with TNF-α were somewhat different, even though their pathways all included TNF-α as the core regulatory element in various tissues and the pathways contained all three TNF-α-related pathways. This may imply that TNF-α is linked to different networks depending on the cancer type. In fact, in MDA-MB-468LN cells, genes involved in chromatin remodeling such as ADNP, which is an element of the SWI/SNF chromatin remodeling complex, were uniquely involved (Rodenhiser et al., 2008b). In case of the prostate cancer, BIK and BAK1, which are BCL-2 antagonists, comprised the central elements (Kim et al., 2011). In the present study of MCF-7 cells, ICAM1 and TNFSF10 uniquely appeared in the network. These findings imply that TNF-α is linked to different networks in different cancer tissues, and even to different subtypes in single types of cancer tissue.

It should be noted that our study was limited in that the epigenetic regulation was based on just one cell type. Extending the examination further to various cell types and tissues might provide insight into common and/or pivotal TNF-α-related pathways. It should be noted that the epigenetic dysregulation could be caused by events other than changes in methyaltion such as histone modification. Co-treatment of the cell together with a histone acetylation inhibitor, Tricostatin A, might overcome the gene silencing effect that occurred when only Aza was used.

A key concern of our study was whether Aza might induce altered expression in a way other than methylation, thus spoiling the pathway analysis. Nevertheless, we have successfully shown that the majority of genes in the pathway underwent demethylation by confirming that eight of ten selected genes in the TNF-α pathway were appropriately demethylated.

In summary, we present here a genome-wide expression profile of MCF-7 breast cancer cells, which was compiled based on promoter methylation and the most comprehensive pathway assignment method currently available. Our findings lend credence to the strong link between epigenetic dysregulation of important cancer-relevant molecules, including those pertinent to TNF-α-dependent apoptosis. Furthermore, our findings indicate that the relevant alteration of expression of these molecules may be involved in breast carcinogenesis. Moreover, tumor-specific alterations in methylation patterns were shown to have the potential to serve as prognostic bio-markers in breast cancer development and progression. Further investigations into the mechanisms leading to the differential methylation observed in the present study, as well as associations of methylation status with breast tumor stage and progression may provide additional insights that could prove useful in estimating prognosis and determining treatment options.

Supplementary Material

Acknowledgments

This study was supported by the National Research Foundation of Korea (NRF) Grant (NRF-2011-C00056) and by the National Science Foundation of China (NSFC) Grant (30800891 & 81111140396).

Note:

Supplementary information is available on the Molecules and Cells website (www.molcells.org).

REFERENCES

  1. Ando K., Kanazawa S., Tetsuka T., Ohta S., Jiang X., Tada T., Kobayashi M., Matsui N., Okamoto T. Induction of Notch signaling by tumor necrosis factor in rheumatoid synovial fibroblasts. Oncogene. 2003;22:7796–7803. doi: 10.1038/sj.onc.1206965. [DOI] [PubMed] [Google Scholar]
  2. Bonazzi V.F., Irwin D., Hayward N.K. Identification of candidate tumor suppressor genes inactivated by promoter methylation in melanoma. Genes Chromosomes Cancer. 2009;48:10–21. doi: 10.1002/gcc.20615. [DOI] [PubMed] [Google Scholar]
  3. Budinsky A.C., Brodowicz T., Wiltschke C., Czerwenka K., Michl I., Krainer M., Zielinski C.C. Decreased expression of ICAM-1 and its induction by tumor necrosis factor on breast-cancer cells in vitro. Int. J. Cancer. 1997;71:1086–1090. doi: 10.1002/(sici)1097-0215(19970611)71:6<1086::aid-ijc27>3.0.co;2-a. [DOI] [PubMed] [Google Scholar]
  4. Calvano S.E., Xiao W., Richards D.R., Felciano R.M., Baker H.V., Cho R.J., Chen R.O., Brownstein B.H., Cobb J.P., Tschoeke S.K., et al. A network-based analysis of systemic inflammation in humans. Nature. 2005;437:1032–1037. doi: 10.1038/nature03985. [DOI] [PubMed] [Google Scholar]
  5. Chambers A.F., Goss P.E. Putative growth characteristics of micrometastatic breast cancer. Breast Cancer Res. 2008;10:114. doi: 10.1186/bcr2197. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Dempsey P.W., Doyle S.E., He J.Q., Cheng G. The signaling adaptors and pathways activated by TNF superfamily. Cytokine Growth Factor Rev. 2003;14:193–209. doi: 10.1016/s1359-6101(03)00021-2. [DOI] [PubMed] [Google Scholar]
  7. Kim T.W., Kim Y.J., Lee H.J., Min S.Y., Kang H.S., Kim S.J. Hs.137007 is a novel epigenetic marker hypermethylated and up-regulated in breast cancer. Int. J. Oncol. 2010;36:1105–1111. doi: 10.3892/ijo_00000592. [DOI] [PubMed] [Google Scholar]
  8. Kim S.J., Kelly W.K., Fu A., Haines K., Hoffman A., Zheng T., Zhu Y. Genome-wide methylation analysis identifies involvement of TNF-α mediated cancer pathways in prostate cancer. Cancer Lett. 2011;302:47–53. doi: 10.1016/j.canlet.2010.12.010. [DOI] [PubMed] [Google Scholar]
  9. Kulis M., Esteller M. DNA methylation and cancer. Adv. Genet. 2010;70:27–56. doi: 10.1016/B978-0-12-380866-0.60002-2. [DOI] [PubMed] [Google Scholar]
  10. Kwon M.S., Kim S.J., Lee S.Y., Jeong J.H., Lee E.S., Kang H.S. Epigenetic silencing of the sulfotransferase 1A1 gene by hypermethylation in breast tissue. Oncol. Rep. 2006;15:27–32. [PubMed] [Google Scholar]
  11. Laux D.E., Curran E.M., Welshons W.V., Lubahn D.B., Huang T.H. Hypermethylation of the Wilms’ tumor suppressor gene CpG island in human breast carcinomas. Breast Cancer Res. Treat. 1999;56:35–43. doi: 10.1023/a:1006222803788. [DOI] [PubMed] [Google Scholar]
  12. Li L., Lee K.M., Han W., Choi J.Y., Lee J.Y., Kang G.H., Park S.K., Noh D.Y., Yoo K.Y., Kang D. Estrogen and progesterone receptor status affect genome-wide DNA methylation profile in breast cancer. Hum. Mol. Genet. 2010;19:4273–4277. doi: 10.1093/hmg/ddq351. [DOI] [PubMed] [Google Scholar]
  13. Liu B., Tahk S., Yee K.M., Fan G., Shuai K. The ligase PIAS1 restricts natural regulatory T cell differentiation by epigenetic repression. Science. 2010;330:521–525. doi: 10.1126/science.1193787. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Lu L., Katsaros D., de la Longrais I.A., Sochirca O., Yu H. Hypermethylation of let-7a-3 in epithelial ovarian cancer is associated with low insulin-like growth factor-II expression and favorable prognosis. Cancer Res. 2007;67:10117–10122. doi: 10.1158/0008-5472.CAN-07-2544. [DOI] [PubMed] [Google Scholar]
  15. Lui E.L., Loo W.T., Zhu L., Cheung M.N., Chow L.W. DNA hypermethylation of TIMP3 gene in invasive breast ductal carcinoma. Biomed. Pharmacother. 2005;59(Suppl 2):S363–365. doi: 10.1016/s0753-3322(05)80079-4. [DOI] [PubMed] [Google Scholar]
  16. Maegawa S., Hinkal G., Kim H.S., Shen L., Zhang L., Zhang J., Zhang N., Liang S., Donehower L.A., Issa J.P. Widespread and tissue specific age-related DNA methylation changes in mice. Genome Res. 2010;20:332–340. doi: 10.1101/gr.096826.109. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Malumbres M., Pérez de Castro I., Santos J., Fernández Piqueras J., Pellicer A. Hypermethylation of the cell cycle inhibitor p15INK4b 3′-untranslated region interferes with its transcriptional regulation in primary lymphomas. Oncogene. 1999;18:385–396. doi: 10.1038/sj.onc.1202299. [DOI] [PubMed] [Google Scholar]
  18. Matros E., Wang Z.C., Lodeiro G., Miron A., Iglehart J.D., Richardson A.L. BRCA1 promoter methylation in sporadic breast tumors: relationship to gene expression profiles. Breast Cancer Res. Treat. 2005;91:179–186. doi: 10.1007/s10549-004-7603-8. [DOI] [PubMed] [Google Scholar]
  19. Politi K., Feirt N., Kitajewski J. Notch in mammary gland development and breast cancer. Semin. Cancer Biol. 2004;14:341–347. doi: 10.1016/j.semcancer.2004.04.013. [DOI] [PubMed] [Google Scholar]
  20. Rivenbark A.G., Jones W.D., Risher J.D., Coleman W.B. DNA methylation-dependent epigenetic regulation of gene expression in MCF-7 breast cancer cells. Epigenetics. 2006;1:32–44. doi: 10.4161/epi.1.1.2358. [DOI] [PubMed] [Google Scholar]
  21. Rodenhiser D., Andrews J., Kennette W., Sadikovic B., Mendlowitz A., Tuck A., Chambers A. Epigenetic mapping and functional analysis in a breast cancer metastasis model using whole-genome promoter tiling microarrays. Breast Cancer Res. 2008a;10:R62. doi: 10.1186/bcr2121. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Rodenhiser D.I., Andrews J., Kennette W., Sadikovic B., Mendlowitz A., Tuck A.B., Chambers A.F. Epigenetic mapping and functional analysis in a breast cancer metastasis model using whole-genome promoter tiling microarrays. Breast Cancer Res. 2008b;10:R62. doi: 10.1186/bcr2121. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Ruan K., Bao S., Ouyang G. The multifaceted role of periostin in tumorigenesis. Cell Mol. Life Sci. 2009;66:2219–2230. doi: 10.1007/s00018-009-0013-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Shao L., Cui Y., Li H., Liu Y., Zhao H., Wang Y., Zhang Y., Ng K.M., Han W., Ma D., et al. CMTM5 exhibits tumor suppressor activities and is frequently silenced by methylation in carcinoma cell lines. Clin. Cancer Res. 2007;13:5756–5762. doi: 10.1158/1078-0432.CCR-06-3082. [DOI] [PubMed] [Google Scholar]
  25. Shivapurkar N., Toyooka S., Toyooka K.O., Reddy J., Miyajima K., Suzuki M., Shigematsu H., Takahashi T., Parikh G., Pass H.I., et al. Aberrant methylation of trail decoy receptor genes is frequent in multiple tumor types. Int. J. Cancer. 2004;109:786–792. doi: 10.1002/ijc.20041. [DOI] [PubMed] [Google Scholar]
  26. Sohn B.H., Park I.Y., Lee J.J., Yang S.J., Jang Y.J., Park K.C., Kim D.J., Lee D.C., Sohn H.A., Kim T.W., et al. Functional switching of TGF-beta1 signaling in liver cancer via epigenetic modulation of a single CpG site in TTP promoter. Gastroenterology. 2010;138:1898–1908. doi: 10.1053/j.gastro.2009.12.044. [DOI] [PubMed] [Google Scholar]
  27. Tommasi S., Karm D.L., Wu X., Yen Y., Pfeifer G.P. Methylation of homeobox genes is a frequent and early epigenetic event in breast cancer. Breast Cancer Res. 2009;11:R14. doi: 10.1186/bcr2233. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Van der Auwera I., Yu W., Suo L., Van Neste L., van Dam P., Van Marck E.A., Pauwels P., Vermeulen P.B., Dirix L.Y., Van Laere S.J. Array-based DNA methylation profiling for breast cancer subtype discrimination. PLoS One. 2010;5:e12616. doi: 10.1371/journal.pone.0012616. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. van Horssen R., Ten Hagen T.L., Eggermont A.M. TNF-alpha in cancer treatment: molecular insights, antitumor effects, and clinical utility. Oncologist. 2006;11:397–408. doi: 10.1634/theoncologist.11-4-397. [DOI] [PubMed] [Google Scholar]
  30. Weng W., Yang Q., Huang M., Qiao Y., Xie Y., Yu Y., jing A., Li Z. c-Myc inhibits TP53INP1 expression via promoter methylation in esophageal carcinoma. Biochem. Biophys. Res. Commun. 2011;405:278–284. doi: 10.1016/j.bbrc.2011.01.028. [DOI] [PubMed] [Google Scholar]
  31. Wolf I., O’Kelly J., Rubinek T., Tong M., Nguyen A., Lin B., Tai H., Karlan B., Koeffler H. 15-hydroxyprostaglandin dehydrogenase is a tumor suppressor of human breast cancer. Cancer Res. 2006;66:7818–7823. doi: 10.1158/0008-5472.CAN-05-4368. [DOI] [PubMed] [Google Scholar]
  32. Xiong J., Epstein R.J. Growth inhibition of human cancer cells by 5-aza-2′-deoxycytidine does not correlate with its effects on INK4a/ARF expression or initial promoter methylation status. Mol. Cancer Ther. 2009;8:779–785. doi: 10.1158/1535-7163.MCT-08-0926. [DOI] [PubMed] [Google Scholar]
  33. Xiong H., Qiu H., Zhuang L., Jiang R., Chen Y. Effects of 5-Aza-CdR on the proliferation of human breast cancer cell line MCF-7 and on the expression of Apaf-1 gene. J. Huazhong Univ. Sci. Technolog. Med. Sci. 2009;29:498–502. doi: 10.1007/s11596-009-0421-9. [DOI] [PubMed] [Google Scholar]
  34. Yan P.S., Shi H., Rahmatpanah F., Hsiau T.H., Hsiau A.H., Leu Y.W., Liu J.C., Huang T.H. Differential distribution of DNA methylation within the RASSF1A CpG island in breast cancer. Cancer Res. 2003;63:6178–6186. [PubMed] [Google Scholar]
  35. Yin Y., Chen X., Shu Y. Gene expression of the invasive phenotype of TNF-alpha-treated MCF-7 cells. Biomed. Pharmacother. 2009;63:421–428. doi: 10.1016/j.biopha.2009.04.032. [DOI] [PubMed] [Google Scholar]
  36. Zhang Q., Wang H.Y., Bhutani G., Liu X., Paessler M., Tobias J.W., Baldwin D., Swaminathan K., Milone M.C., Wasik M.A. Lack of TNFalpha expression protects anaplastic lymphoma kinase-positive T-cell lymphoma (ALK+ TCL) cells from apoptosis. Proc. Natl. Acad. Sci. USA. 2009;106:15843–15848. doi: 10.1073/pnas.0907070106. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Zhang Y., Zhang G., Li J., Tao Q., Tang W. The expression analysis of periostin in human breast cancer. J. Surg. Res. 2010;160:102–106. doi: 10.1016/j.jss.2008.12.042. [DOI] [PubMed] [Google Scholar]

Associated Data

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

Supplementary Materials

molcell-33-2-127-4-supplementary.pdf (19.4KB, pdf)

Articles from Molecules and Cells are provided here courtesy of Korean Society for Molecular and Cellular Biology

RESOURCES