Skip to main content

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

Candidate DNA methylation drivers of acquired cisplatin resistance in ovarian cancer identified by methylome and expression profiling

Abstract

Multiple DNA methylation changes in the cancer methylome are associated with the acquisition of drug resistance; however it remains uncertain how many represent critical DNA methylation drivers of chemoresistance. Using isogenic, cisplatin-sensitive/resistant ovarian cancer cell lines and inducing resensitizaton with demethylating agents, we aimed to identify consistent methylation and expression changes associated with chemoresistance. Using genome-wide DNA methylation profiling across 27 578 CpG sites, we identified loci at 4092 genes becoming hypermethylated in chemoresistant A2780/cp70 compared with the parental-sensitive A2780 cell line. Hypermethylation at gene promoter regions is often associated with transcriptional silencing; however, expression of only 245 of these hypermethylated genes becomes downregulated in A2780/cp70 as measured by microarray expression profiling. Treatment of A2780/cp70 with the demethylating agent 2-deoxy-5′-azacytidine induces resensitization to cisplatin and re-expression of 41 of the downregulated genes. A total of 13/41 genes were consistently hypermethylated in further independent cisplatin-resistant A2780 cell derivatives. CpG sites at 9 of the 13 genes (ARHGDIB, ARMCX2, COL1A, FLNA, FLNC, MEST, MLH1, NTS and PSMB9) acquired methylation in ovarian tumours at relapse following chemotherapy or chemoresistant cell lines derived at the time of patient relapse. Furthermore, 5/13 genes (ARMCX2, COL1A1, MDK, MEST and MLH1) acquired methylation in drug-resistant ovarian cancer-sustaining (side population) cells. MLH1 has a direct role in conferring cisplatin sensitivity when reintroduced into cells in vitro. This combined genomics approach has identified further potential key drivers of chemoresistance whose expression is silenced by DNA methylation that should be further evaluated as clinical biomarkers of drug resistance.

Introduction

Ovarian cancer often presents at an advanced stage with limited chances for curative treatment (Ozols, 2004). Common treatment strategies include debulking surgery in combination with chemotherapy, which usually consists of a platinum-based compound, such as cisplatin/carboplatin and a taxane, for example, paclitaxel. Despite responses to first-line chemotherapy in a high proportion of patients, many will relapse and eventually develop resistance to currently available treatment options, making the acquisition of clinical drug resistance one of the major challenges in ovarian cancer therapy and a limiting factor in patient survival (Cannistra, 2004).

Platinum drugs are DNA cross-linking agents exerting their effect mainly via the formation of intra-strand adducts between adjacent guanosines that interfere with transcription and replication, eventually leading to cell death (Kartalou and Essigmann, 2001; Wang and Lippard, 2005). Several mechanisms have been suggested to participate in conferring platinum-resistant properties to a tumour cell such as genetic alterations in genes involved in DNA repair, drug uptake, apoptosis, cell cycle control and IGF signalling pathways (Johnstone et al., 2002; Luqmani, 2005; Edwards et al., 2008; Broxterman et al., 2009; Eckstein et al., 2009). More recently, it has been proposed that, in addition to genetic changes, aberrant epigenetic marks can critically contribute to the acquisition of drug resistance (Sharma et al., 2010). In cisplatin-resistant cancer cells, in particular, multiple DNA methylation changes at promoter CpG islands and associated transcriptional gene silencing have been reported (Teodoridis et al., 2005; Dai et al., 2008; Chang et al., 2010). For instance, methylation at MLH1, a mismatch repair gene, is acquired in about 25–35% of ovarian cancer patients following platinum-based chemotherapy and has been shown to be associated with poor patient survival (Strathdee et al., 1999; Gifford et al., 2004). Importantly, reversal of MLH1 epigenetic silencing by demethylation or re-expression of the gene was demonstrated to resensitise tumour cells to subsequent chemotherapeutic treatment in vitro and in vivo (Plumb et al., 2000; Papouli et al., 2004; Steele et al., 2009). MLH1 might, therefore, represent one of the key genes driving chemoresistance in ovarian cancer cell lines. However, although the role of methylation changes at MLH1 has been well characterized, for the majority of aberrant DNA methylation events it is not particularly clear whether they are associated with response to chemotherapy or are just occurring by chance due to a methylator phenotype or simply as random methylation events (Issa, 2004) during platinum selection or DNA damage induction. In analogy to the concept of ‘driver and passenger’ mutations emerging during carcinogenesis, methylation changes could either represent ‘drivers’ of chemoresistance based on their potential to provide the cell with a selective advantage or ‘passenger’ events, with no substantial impact on chemosensitivity (Greenman et al., 2007).

We hypothesized that, in analogy to driver mutations, there might be a subset of epigenetic changes that are causally associated with the acquisition of chemoresistance. In order to identify the proportion of epigenetically altered genes driving platinum resistance in ovarian cancer, we have analysed acquired DNA methylation changes in a human ovarian cancer cell line model of cisplatin resistance and expression changes associated with acquired resistance or following resensitization with demethylating agents. We then examined the generality of the changes observed in ovarian tumours at relapse following chemotherapy.

Results

Cisplatin selects preferentially for DNA hypermethylation in A2780-chemoresistant cell lines

To identify DNA methylation changes associated with changes in gene expression and differential cisplatin chemosensitivity, we used the human A2780 ovarian cell line model that consists of sensitive and matched isogenic platinum-resistant lines derived from A2780 cells by repeated exposures to increasing levels of cisplatin (Anthoney et al., 1996). We performed genome-wide DNA methylation profiling of resistant A2780/cp70, A2780/MCP1 and A2780/MCP6 compared with non-selected A2780p5 and A2780p6 clones employing Infinium HumanMethylation27 BeadArrays that comprise 27 578 CpG sites across more than 14 000 genes. In a first step, differentially methylated genes were extracted as showing a significantly increased or decreased difference of beta (Δβ) corresponding to a false discovery rate (FDR) <0.05 estimated from biological replicates within the study (Δβ0.1) between A2780/cp70 versus A2780 at 1 associated CpG site (see Materials and methods). Using these criteria, we identified multiple methylation changes between A2780/cp70 versus A2780 (Figure 1): 1 CpG sites at 4092 genes were hypermethylated, whereas only 1289 genes became hypomethylated following exposure to cisplatin, suggesting that hypermethylation occurs more frequently than hypomethylation during the process of selection for acquired cisplatin chemoresistance.

Figure 1
figure1

Hypermethylation is prevalent in cisplatin-resistant A2780 cell clones. Scatter plots of β-values show differentially methylated loci between A2780-sensitive (average of A2780p5/A2780p6) versus -resistant derivatives A2780/cp70, A2780/MCP1 and A2780/MCP6 (yellow) as measured by Infinium HumanMethylation27 BeadArrays. The differential methylation cutoff (red and green solid line) was estimated from the difference between biological replicates of cell lines PEO1 and PEO4 (blue) by controlling FDR<0.05. A full colour version of this figure is available at the Oncogene journal online.

Methylation changes in cisplatin-resistant lines associate with gene expression changes in only a subset of genes

In a second step, we examined the expression profiles of sensitive A2780 and resistant A2780/cp70 cell clones using Affymetrix HG-U133 Plus 2.0 Arrays and identified differentially expressed genes by applying rank products analysis (Breitling et al., 2004). Aberrant DNA methylation at CpG island (CGI) has been shown to critically affect gene expression and is strongly associated with transcriptional repression (Jones and Baylin, 2002; Esteller, 2008). We assumed that expression changes associating with hypermethylation in chemoresistant A2780/cp70 as compared with A2780 would likely represent silencing events associated with the drug-resistant phenotype. We identified a total of 1370 genes showing significant gene expression changes, with 687 genes going up and 683 genes going down in resistant versus sensitive cell lines. We further filtered for those genes where hypermethylation associated with reduced expression in chemoresistant A2780/cp70 cells. Of the 4092 genes hypermethylated in A2780/cp70, 3823 (93%) genes are present on the HG-U133 Plus 2.0 Array. Combined analysis revealed that only a small proportion of changes in methylation correlated with expression changes, with 245 genes becoming hypermethylated and downregulated following selection for cisplatin resistance in A2780/cp70.

Only a small proportion of hypermethylated genes that are downregulated become re-expressed following treatment with decitabine

In order to identify DNA methylation changes associating with changes in gene expression and differential cisplatin chemosensitivity in this cell line model, we combined our data with expression profiles of A2780/cp70 cells treated with either the DNA methyltransferase inhibitor (DNMTi) 2-deoxy-5′-azacytidine (decitabine, DAC), the histone deacetylase inhibitor (HDACi) Belinostat or a combination of both at conditions and time point known to induce chemosensitization in this cell line (Steele et al., 2009). We reasoned that genes affecting drug sensitivity via hypermethylation-associated silencing in chemoresistant cells would become switched backed on following pharmacological intervention of methylation and/or deacetylation. DNMTi treatment can function synergistically with an HDACi in restoring gene expression (Cameron et al., 1999), and sequential exposure of cells to DNMTi and HDACi might, therefore, lead to re-expression of additional genes.

Treatment with DAC resulted in re-expression of 41 out of the 245 genes that were hypermethylated in addition to being downregulated in A2780/cp70 cells (Figure 2, Supplementary Table S1). A total of 45 genes became switched back on following combined treatment with DAC and the HDACi Belinostat. Exposure to Belinostat alone reinduced the expression of only 10 genes that were downregulated in addition to being hypermethylated in untreated A2780/cp70 cells. Comparison of the gene lists showed a big overlap between the 41 (DAC) and 45 (DAC+Belinostat) re-expressed genes, with 40/41 genes being contained within the list of 45 genes, suggesting that addition of the HDACi does not add substantially to the number of genes reactivated by DAC alone. We further compared DNA methylation profiles of A2780/cp70 cells following drug treatment with untreated A2780/cp70. Using a differential methylation cutoff of |Δβ|0.1, we validated that all of the 41 DAC-responsive genes became demethylated at 1 associated CpG site following treatment with DAC. This substantiates that re-expression of the 41 DAC-induced genes was truly mediated by DNA demethylation (Supplementary Figure S2).

Figure 2
figure2

Re-expression of genes in chemoresistant A2780/cp70 cells following treatment with epigenetic remodelling agents. In all, 41 genes (DAC), 45 genes (DAC+Belinostat) or 10 genes (Belinostat) become reactivated following treatment respectively. Venn diagrams show the overlap between number of genes being hypermethylated in untreated A2780/cp70 (green circle), being downregulated in untreated A2780/cp70 (pink circle) and genes with increased expression following drug exposure in A2780/cp70 as compared with A2780 cells (blue circle). A full colour version of this figure is available at the Oncogene journal online.

A small set of 13 genes are strong candidates as epigenetically silenced cisplatin-resistant drivers

In order to identify drug resistance-associated methylation changes that were independent of clonal effects, we further filtered for genes showing consistent hypermethylation of reinduced genes in independently selected cisplatin-resistant A2780 derivatives. Out of the 4092 hypermethylated genes initially identified in A2780/cp70, 1824 genes were also hypermethylated in the two independent cisplatin-resistant derivatives, A2780/MCP1 and A2780/MCP6 (Figure 3).

Figure 3
figure3

A small percentage of consistently hypermethylated genes is affected following resensitization with epigenetic remodelling agents. Out of 1824 genes hypermethylated in independent cisplatin-resistant A2780 derivatives, only 13 genes become re-expressed following DAC treatment, 14 genes become reactivated following combined DAC and Belinostat treatment and only 1 gene becomes re-expressed following Belinostat treatment. Venn diagrams show the overlap between genes being hypermethylated in chemoresistant A2780/cp70, A2780/MCP1 and A2780/MCP6 (green circle), being downregulated in untreated A2780/cp70 (pink circle) and genes with increased expression following drug exposure in A2780/cp70 as compared with A2780 cells (blue circle). A full colour version of this figure is available at the Oncogene journal online.

Among those 1824 hypermethylated genes, a small proportion of only 13 genes (ARHGDIB, PSMB9, HSPA1A, ARMCX2, MEST, FLNC, MLH1, MDK, GLUL, FLNA, NTS, COL1A1 and NEFL) became re-expressed in A2780/cp70 following treatment with DAC (Table 1, Figures 3 and 4a). Epigenetic silencing of MLH1 has previously been shown to be a clinically relevant mechanism of acquired cisplatin resistance in ovarian cancer (Gifford et al., 2004), and re-expression of MLH1 confers increased cisplatin chemosensitivity in xenograft models (Plumb et al., 2000). MLH1 was identified among the list of 13 genes supporting our approach and substantiating the notion that aberrant methylation of a small set of genes can putatively drive the acquisition of cisplatin resistance.

Table 1 DAC-induced genes being hypermethylated and downregulated in A2780/cp70
Figure 4
figure4

Methylation and expression analysis of genes getting re-expressed following DAC treatment in A2780-resistant cell lines. (a) Heatmap of the 13 DAC-regulated genes showing increased methylation in resistant A2780/cp70, A2780/MCP1 and A2780/MCP6 cell clones as compared with A2780-sensitive lines (Δβ0.1). (b) Validation of hypermethylation of DAC-regulated candidate genes in A2780-resistant cell lines by bisulphite pyrosequencing. Average methylation values across A2780-sensitive and -resistant cell lines are shown for three genes. Pyrosequencing assays show average methylation values across two CpG sites for MEST (CpG site no. 1 relates to Illumina ID cg08077673), seven CpG sites for FLNC (CpG site no. 6 relates to Illumina ID cg02661879) and three CpG sites for ARHGDIB (CpG site no. 2 relates to Illumina ID cg10925082), covering the differentially methylated CpG site shown in methylation profile (a). (c) Validation of downregulation of DAC-regulated candidate genes in A2780-resistant cell lines by quantitative real-time PCR. The fold reduction in A2780/cp70 cells versus A2780-sensitive cells is shown for three genes. A full colour version of this figure is available at the Oncogene journal online.

Combined DAC and Belinostat treatment did not markedly increase the number of re-expressed genes, with 14 commonly hypermethylated genes becoming switched back on in the resistant cell line (ARHGDIB, PSMB9, HSPA1A, ARMCX2, MEST, FLNC, MLH1, MDK, GLUL, NTS, COL1A1, NEFL, SERPINB2 and HIST1H2BF). Again, we observed a strong overlap between genes being reinduced following DAC and combined DAC/Belinostat treatment. Only SERPINB2 and HIST1H2BF were exclusively DAC/Belinostat responsive, suggesting that synergistic demethylation and acetylation are necessary for their reactivation. In contrast, only one gene (GLUL) was being reinduced following Belinostat treatment alone in addition to being consistently hypermethylated across resistant A2780 cell lines. Interestingly, GLUL was also reactivated following DAC and combined DAC/Belinostat treatment. However, the effect of Belinostat alone on GLUL expression was very subtle with a low fold change, suggesting that DAC-induced demethylation is still required for full restoration of gene activity.

We further validated our array-based findings in the A2780 cell line model by using bisulphite pyrosequencing and quantitative real-time PCR. We arbitrarily chose the three genes MEST, FLNC and ARHGDIB that were commonly hypermethylated in independent cisplatin-resistant cell lines (Figure 4) and became re-expressed following DAC as well as combined DAC/Belinostat treatment. FLNC and MEST are both associated with CGIs, whereas ARHGDIB does not contain a CGI. Acquired hypermethylation was confirmed for MEST, FLNC and ARHGDIB in resistant versus sensitive A2780 cell lines (Figure 4b). The observed differences in CpG methylation between sensitive and resistant lines at 39%, 42% and 27% corresponded to a Δβ of 0.38, 0.43 and 0.23 for MEST, FLNC and ARHGDIB, respectively (Figure 4). Hypermethylation of the three genes was associated with downregulation in A2780/cp70-resistant cells as compared with A2780 (Figures 4b and c). We also tested mRNA expression of the three genes in independent cisplatin-resistant A2780/MCP1 and A2780/MCP6 and found downregulation to be associated with hypermethylation in these cell lines (data not shown).

Candidate drug-resistant genes commonly acquire methylation in independent in vivo derived ovarian chemoresistant cells and in tumours at relapse

In order to address how the small set of epigenetically inactivated candidate resistance drivers impacts on clinical drug resistance, we evaluated their methylation status in three independent pairs of in vivo derived chemonaive and chemoresistant ovarian tumour cell lines derived from patients before chemotherapy and at the time of developing clinical resistance (Langdon et al., 1988). Using a difference of Δβ0.1 we observed acquired methylation in 6 out of the 13 genes (ARHGDIB, ARMCX2, COL1A1, FLNA, MEST and MLH1) in chemoresistant, post-chemotherapy cell lines (Table 2).

Table 2 Acquired methylation (Δβ0.1) of candidate drug-resistant drivers in drug-resistant in vivo derived ovarian cancer cell lines, SPs and tumours at relapse

One hypothesis of how drug resistance evolves implies the expansion of the so-called tumour-initiating or cancer-sustaining cells (CSCs; Rajasekhar et al., 2008). This population of cells is thought to share many characteristics with normal stem cells including the capacity to self-renew and to survive drug treatment (Dean et al., 2005). According to this paradigm, the acquisition of drug resistance results from repopulation of the tumour with inherently drug-resistant CSCs following chemotherapeutic treatment. We examined whether candidate drug-resistant drivers showed altered methylation in CSCs obtained from an ovarian cancer cell line by comparing the β-values of the set of 13 genes in independently sorted side populations (SPs) of IGROV1 cells (n=2) versus non-SPs. We have previously shown that IGROV1 SPs have tumour stem cell-like properties, including the ability to form tumours in NOD/SCID mice at low cell number, enhanced ability to grow as spheroids, ability to repopulate SP and non-SP cells and expression patterns with significant enrichment for known stem cell markers (Rizzo et al., 2011). We have also previously shown that IGROV1 SP are more resistant to cisplatin and carboplatin than non-SP IGROV1 cells (Rizzo et al., 2011). Although we did attempt to isolate SPs from A2780, the numbers of SP cells present were too low to allow detailed phenotypic and molecular analysis.

Applying a Δβ0.1 we observed five genes as having higher methylation in IGROV1 CSC populations (ARMCX2, COL1A1, MEST, MDK and MLH1; Table 2). Notably, four of these genes had acquired methylation in in vivo derived cisplatin-resistant cell lines (ARMCX2, COL1A1, MEST and MLH1), supporting enrichment of methylation at these loci in chemoresistant cell populations.

We further evaluated the methylation status of the 13 candidate genes in clinical specimens using seven pairs of matched primary ovarian tumours and tumours at the time of relapse. Methylation of 8/13 genes was acquired in tumours following platinum-based chemotherapy (Table 2). Three of these genes (ARMCX2, MEST and MLH1) were found to have higher methylation in chemoresistant in vivo derived cell lines, CSC populations and tumours at relapse, supporting the clinical relevance of changes at these loci.

Discussion

Substantial DNA methylation changes have previously been reported to occur during the acquisition of platinum resistance. Among those changes, hypermethylation at CGI promoters and associated epigenetic silencing are prevalent in various cell line models (Dai et al., 2008; Li et al., 2009; Chang et al., 2010) and are thought to, either alone or in combination with genetic changes, account for the loss of expression of key genes in the platinum-responsive phenotype. However, our knowledge of whether these DNA methylation changes and aberrantly inactivated genes are actually driving chemoresistance is limited. By integrating drug induced resensitization, methylation and expression profiling in an isogenic, cisplatin-resistant ovarian cancer cell line model, we present evidence that the proportion of key DNA methylation and linked expression changes associating with response to chemotherapy is surprisingly small. Our results indicate that <1% of hypermethylated genes in independent platinum-resistant ovarian cancer cell lines account for the acquisition of platinum resistance. Indeed, by comparing methylation changes in multiple independent cisplatin-resistant derivatives of A2780, we identify as few as 13 key genes that acquire methylation at associated CpG sites. These are strong candidates for drivers of resistance in this genetic background, although other genes may be more important in other ovarian tumour genetic backgrounds. However, we show that a high proportion of the loci identified acquire methylation in ovarian tumours at relapse (8/13) and are differentially methylated in drug-resistant ovarian tumour sustaining (stem) cells (5/13). Therefore, methylation at CpG sites at these genes are key candidate epigenetic drivers of acquired drug resistance independent of genetic background.

Few previous studies established hypermethylation as a marker of platinum-resistant ovarian cancer cell lines on an epigenome-wide basis (Dai et al., 2008; Li et al., 2009). Although the mechanism of how drug-resistant cells accumulate methylation is unresolved, there are clear examples showing an association between epigenetic silencing of specific genes and drug response. For example, acquired MLH1 CGI hypermethylation and associated gene inactivation are observed following platinum-based chemotherapy in ovarian cancer and is associated with poor clinical outcome (Gifford et al., 2004). Re-expression of MLH1 in chemoresistant cell lines has been shown to partially restore their sensitivity to subsequent cisplatin therapy (Plumb et al., 2000; Papouli et al., 2004). Notably, our list of methylation-dependent genes included MLH1, substantiating the hypothesis that we have identified a specific subset of genes that drive the acquisition of drug resistance. It has to be noted, however, that full resensitization of cells to platinum-based drugs could not be achieved via MLH1 re-expression alone in cell line models, in contrast to reversal of methylation using demethylating agents that fully restore chemosensitivity and induce re-expression of multiple genes (Plumb et al., 2000; Papouli et al., 2004). These observations support the idea that the induction of additional repressed genes to MLH1 is vital for full restoration of drug sensitivity in these models. However, this also illustrates the challenges in demonstrating direct effects of a gene on chemosensitivity in vitro based on gene reintroduction, where the effects may be small or require inactivation of multiple genes. Our results indicate that several methylation changes are directly associated with chemosensitization induced by demethylating agents combined with HDAC inhibitors. Similar observations have been reported in other tumour types (Chang et al., 2010); however, our data also reveal that epigenetic alterations associated with sensitivity to cisplatin occur at a few selected genes rather than at large numbers of loci in this cell line model (Glasspool et al., 2006).

The group of epigenetically inactivated genes was upregulated by DAC and also the combined DAC and Belinostat treatment, but remained unaffected by Belinostat alone, with the exception of GLUL. Consistent with previous reports, the inhibition of HDAC did not add extensively to the number of re-expressed genes, with only two additional genes (SERPINB2 and HIST1H2BF) being reinduced (Cameron et al., 1999; Suzuki et al., 2002). Addition of an HDACi can markedly increase the level of gene re-expression such as MLH1 in vitro and in vivo (Steele et al., 2009), but may not convert further genes to re-express. Although other studies reported lack of methylation of genes responsive to HDACi and, partially, to DNMTi (Suzuki et al., 2002), our analysis specifically extracted genes that were hypermethylated in chemoresistant cell lines. One explanation for this discrepancy could be the method that we used for identification of differentially expressed genes. We applied the rank products method based on our aim to detect differences that consistently rank highly across any number of replicate experiments and that are most likely to have biological significance. This form of analysis avoids identifying a fold change that exceeds an arbitrary threshold. Conversely, this method can identify subtle changes and these subtle changes may be missed by other methods.

Our subset of candidate drug-resistant drivers was identified using an isogenic cell line model. Although this model allows identifying changes ultimately linked to cisplatin exposure, detection of genes in identical rather than in diverse genetic settings, as well as issues surrounding in vitro selection, may be a limitation to the value of this gene set. However, we observed increased methylation of 6 out of the 13 candidate genes in at least one of the three cisplatin-resistant in vivo derived cell lines. We further observed acquired methylation of five out of these six genes in tumours at relapse. Although our set of matched tumour pairs before and after chemotherapy was small due to the clinical challenges in obtaining such matched samples, this nevertheless suggests that the acquisition of methylation at these genes may be a common event occurring during acquired cisplatin resistance and may be clinically relevant. Notably, the low frequency at which epigenetic changes are acquired at candidate drug-resistant drivers shows analogies to driver mutations that often present at a low frequency (Wood et al., 2007) and might, therefore, be missed within a small set of tumours. Interestingly, we also observed increased methylation at three (ARMCX2, MEST and MLH1) out of these five commonly hypermethylated candidate genes in drug-resistant sustaining (SP) cells isolated from the human ovarian tumour cell line IGROV1 as compared with the bulk of the tumour cells. Our observation could indicate that hypermethylation and epigenetic silencing of this subset of three genes are already present in stem cells. As SPs are believed to contain the cells responsible for maintenance of long-term growth of ovarian cancer (Bapat et al., 2005; Szotek et al., 2006), regrowth of these cells could be contributing to manifesting a drug-resistant phenotype (Agarwal and Kaye, 2003).

Silencing of MLH1 leads to the loss of DNA mismatch repair, which in turn has been suggested to lead to platinum DNA damage tolerance due to translesion synthesis, reduced replication stalling or reduced signalling of cell death pathways (Stojic et al., 2004; O’Brien and Brown, 2006; Yoshioka et al., 2006). However, little is known about the function in drug resistance of the other genes found to commonly acquire methylation in tumours and cell lines at relapse and in drug-resistant SP populations. ARMCX2 might have a role in tumour suppression based on the presence of an armadillo repeat motif, which is found in other proteins fulfilling functions in cell proliferation, migration, maintenance of tissue integrity and tumourigenesis, and has been involved in development (Smith et al., 2005). MEST, a maternally imprinted gene, has been implicated in embryonic growth and maternal behaviour (Lefebvre et al., 1998), and loss of MEST imprinting has been reported in breast and lung cancers (Pedersen et al., 2002; Nakanishi et al., 2004). Interestingly, it has recently been shown that MEST is a negative regulator of the Wnt pathway and that MEST knockdown might activate Wnt signalling (Jung et al., 2011). Epigenetic regulation of Wnt pathway genes has previously been shown to be associated with patient survival following platinum-based chemotherapy and chemoresistance in ovarian cancer (Dai et al., 2011; Peng et al., 2011).

Taken together, we have identified a key subset of genes potentially driving acquired drug resistance in ovarian cancer from the large number of epigenetic changes occurring following chemotherapy. As well as providing novel insight into mechanisms of drug resistance, this has identified candidate biomarkers for further evaluation in future clinical studies, including potential stratification biomarkers in clinical trials of epigenetic therapies that reverse the acquired resistance phenotype.

Materials and methods

Cell lines

In vitro derived A2780p5, A2780p6, A2780/cp70, A2780/MCP1 and A2780/MCP6 cell lines (Anthoney et al., 1996) were obtained from Dr Robert F Ozols (Fox Chase Cancer Centre, Philadelphia, PA, USA). PEO1, PEO4, PEO14, PEO23, PEA1 and PEA2 in vivo derived cell lines (Langdon et al., 1988) were obtained from Professor Hani Gabra (Imperial College). Isolation of SPs and non-SPs of IGROV1 cells was performed as described previously (Rizzo et al., 2011). The identity of the cell lines was verified by DNA genotyping. All cell lines were grown in RPMI-1640 supplemented with glutamine (2 mM) and 10% fetal bovine serum.

Tumour samples

Seven pairs of matched serous epithelial ovarian tumours before chemotherapy and at relapse were obtained from the University Medical Center Groningen. Appropriate ethical approval was obtained for all the collected samples.

Methylation profiling

All array-based methylation profiling was performed using the Infinium HumanMethylation27 BeadChip (Illumina, San Diego, CA, USA), except for DNA methylation profiling following treatment with epigenetic remodelling agents, which was performed using Infinium HumanMethylation450 BeadChips (Illumina). A total amount of 1 μg genomic DNA was bisulphite modified using the EZ DNA Methylation Kit (Zymo Research, Irvine, CA, USA). In all, 200 ng of converted DNA was further processed to run BeadArrays according to the manufacturer's instructions. Each locus is represented by fluorescent signals from two bead types corresponding to the methylated (M) and unmethylated (U) alleles, respectively. The raw signals of unmethylated and methylated bead types were background corrected and computed into a β-value using the BeadStudio software Methylation Module (version 1.0.5; Illumina). The β-value represents the ratio of the intensity of the methylated bead type to the combined locus intensity: β=max(M, 0)/(max(M,0)+max(U,0)+100) and reflects the methylation status of a specific CpG site. Subsequent analysis was carried out in R (version 2.10.1), http://www.r-project.org/.

The reproducibility of the BeadChips was evaluated using biological (independent bisulphite modifications of two independently prepared DNAs) and technical replicates (bisulphite conversion of identical DNA) of matched chemosensitive/chemoresistant ovarian cancer cell lines PEO1/PEO4. β-Values of biological and technical replicates of PEO1 and PEO4 were highly correlated with r2 values >0.99 (Supplementary Figure S1).

Gene annotation was carried out using the Mar. 2006 (NCBI/36/hg18) assembly at the UCSC database (http://genome.ucsc.edu/). Individual CpG sites located within a CGI as defined by Gardiner-Garden and Frommer (1987) were linked to an associated gene if the CGI was within 2 kb distance from the respective transcription start site (Saxonov et al., 2006). Non-CGI-related CpG sites were linked to a gene if they were located within 2 kb from the transcription start site.

For pairwise differential methylation analysis, the difference of two biological replicates of the ovarian cancer cell lines PEO1 and PEO4 was calculated as the expected difference (null distribution). The observed difference was calculated from the paired samples. Using bootstrap resampling method, we extracted the same number of CpG sites (n=27 578) from the null distribution 200 times. Each time, a FDR was calculated using the formula below. The median of FDRs was used as an estimation of the probability that CpG sites with differential methylation above the selected cutoff were identified by chance. This median FDR was set at 0.05 corresponding to Δβ0.1.

In order to filter for differentially methylated CpG sites in the A2780 cell line model mean β-values across sensitive cell clones were used.

Bisulphite modification and pyrosequencing

A total amount of 1 μg of genomic DNA was bisulphite modified using the EpiTect Bisulphite Kit (Qiagen, West Sussex, UK) according to the manufacturer's instructions. Pyrosequencing primer sets covering differentially methylated CpG sites in the ARHGDIB, FLNC and MEST gene promoters were ARHGDIB_PYRO_F: 5′-biotin-IndexTermTGGGAATAGAAGTGAGTGGTATAA-3′, ARHGDIB_PYRO_R: 5′-IndexTermCCTATTCCTTTACACTACCTATCT-3′, ARHGDIB_PYRO_S: 5′-IndexTermCAACATTCTTATCAATTAATAACAC-3′, FLNC_PYRO_F: 5′-IndexTermTGGAGGGAGAGAGAGTTAG-3′, FLNC_PYRO_R: 5′-biotin-IndexTermCTTACCCACCCACTTAAAATACTCATTAC-3′, FLNC_PYRO_S:5′- IndexTermAGAAGTTGGAGAGGA-3′, MEST_PYRO_F: 5′-biotin-IndexTermGTGGGTTATATTAGTTTTAGGGGTAG-3′, MEST_PYRO_R: 5′-IndexTermCCTTTCCAACCTCCAAAACTAACTAT-3′ and MEST_PYRO_S: 5′-IndexTermAAATTATATAACTTTTATATTCTC-3′. Pyrosequencing PCR was performed in duplicate for each sample in a 25 μl volume containing 0.2 μl Faststart Taq polymerase (Roche, Welwyn Garden City, UK), 2.5 μl Faststart Buffer (Roche), 0.2 mM dNTPs (Applied Biosystems, Warrington, UK), 75 ng primers (each) and 1 μl of modified DNA template using the following conditions: 95 °C for 6 min, 35 cycles of 95 °C for 30 s, 60 °C for 30 s, 72 °C for 30 s, followed by 72 °C for 5 min. Pyrosequencing of PCR products was performed using the PyroGold Reagent Kit (Qiagen) according to the manufacturer's instructions. The methylation percentage of CpG sites for individual genes was calculated by using the Pyro Q-CpG software (version 1.0.9), Biotage (Uppsala, Sweden), and then averaged across sensitive and resistant A2780 derivatives, respectively.

Treatment of cells with DNMT inhibitor and/or HDAC inhibitor

Cells were treated at 50% confluence with 0.1. μM 5-aza-2′-deoxycytidine (DAC) (Sigma-Aldrich, St Louis, MO, USA) or 0.1 μM Belinostat or mock treated for 48 h. DAC was replaced after 24 h. For combined DAC/Belinostat treatment, DAC (0.1 μM) was added for 48 h, with Belinostat (0.1 μM) being added for an additional 24 h following the initial 24 h DAC treatment. Cells were harvested 96 h following drug removal.

RNA extraction and microarray analysis

Total cellular RNA was isolated using the RNAeasy kit (Qiagen) according to the manufacturer's instructions. RNA quality was verified using the Agilent 2100 Bioanalyser (Agilent, Wokingham, UK). Expression profiling was carried out using the Human Genome U133 Plus 2.0 Array (Affymetrix, High Wycombe, UK) following the manufacturer's recommendations. Three biological replicates were used for each sample. The raw expression data were normalized as previously described (Irizarry et al., 2003). Subsequent analysis was performed in R (version 2.10.1). Data were log2 transformed, and signal intensity and statistical significance were established for each transcript. Rank products (Breitling et al., 2004) was applied to identify differentially expressed genes using individual probes, and significance was set at a FDR <0.05.

Data deposition

Methylation and expression profiling data are available in NCBI's Gene Expression Omnibus (GEO accession number: GSE28648).

Reverse transcription and real-time PCR

A total amount of 2 μg RNA was reverse transcribed into complementary DNA using SuperScript II (Invitrogen, Paisley, UK) and subsequently used as a template in quantitative real-time experiments to amplify products for MEST, FLNC, ARHGDIB and GAPDH. Primer sequences were MEST_qRT_2F: 5′-IndexTermCGGCCATGGTGCGCCGAGAT-3′, MEST_qRT_2R: 5′-IndexTermACGCAGCAAGCAGGGGCACG-3′, FLNC_qRT_1F: 5′-IndexTermGTGCCCAAGGTCGCTGGGTTACA-3′, FLNC_qRT_1R: 5′-IndexTermTCCCAGGGCCATGCCCAC&GTT-3′, ARHGDIB_qRT_1F: 5′-IndexTermAACGCTGCTGGGAGATGGTCCTGT-3′, ARHGDIB_qRT_1R: 5′-IndexTermACCAGGGTGAGCCGGGTGACAA-3′, GAPDH_qRT_F: 5′-IndexTermCCTGTTCGACAGTCAGCCG-3′ and GAPDH_qRT_R: 5′-IndexTermCGACCAAATCCGTTGACTCC-3′. Quantitative real-time PCR was carried out in triplicate with the Biorad CFX96 Real-Time system using SYBR Green JumpStart Taq ReadyMix (Sigma-Aldrich) and amplification parameters: 95 °C for 3 min, 42 cycles of 95 °C for 10 s, 60 °C for 10 s and 72 °C for 30 s, followed by 95 °C for 10 s and temperature increments from 72–95 °C for 5 s. Relative expression was determined by applying the comparative Ct method (Schmittgen and Livak, 2008) using GAPDH as the internal control gene.

Accession codes

Accessions

Gene Expression Omnibus

References

  1. Agarwal R, Kaye SB . (2003). Ovarian cancer: strategies for overcoming resistance to chemotherapy. Nat Rev Cancer 3: 502–516.

    CAS  Article  Google Scholar 

  2. Anthoney DA, McIlwrath AJ, Gallagher WM, Edlin AR, Brown R . (1996). Microsatellite instability, apoptosis, and loss of p53 function in drug-resistant tumor cells. Cancer Res 56: 1374–1381.

    CAS  PubMed  Google Scholar 

  3. Bapat SA, Mali AM, Koppikar CB, Kurrey NK . (2005). Stem and progenitor-like cells contribute to the aggressive behavior of human epithelial ovarian cancer. Cancer Res 65: 3025–3029.

    CAS  Article  Google Scholar 

  4. Breitling R, Armengaud P, Amtmann A, Herzyk P . (2004). Rank products: a simple, yet powerful, new method to detect differentially regulated genes in replicated microarray experiments. FEBS Lett 573: 83–92.

    CAS  Article  Google Scholar 

  5. Broxterman HJ, Gotink KJ, Verheul HM . (2009). Understanding the causes of multidrug resistance in cancer: a comparison of doxorubicin and sunitinib. Drug Resist Updat 12: 114–126.

    CAS  Article  Google Scholar 

  6. Cameron EE, Bachman KE, Myohanen S, Herman JG, Baylin SB . (1999). Synergy of demethylation and histone deacetylase inhibition in the re-expression of genes silenced in cancer. Nat Genet 21: 103–107.

    CAS  Article  Google Scholar 

  7. Cannistra SA . (2004). Cancer of the ovary. N Engl J Med 351: 2519–2529.

    CAS  Article  Google Scholar 

  8. Chang X, Monitto CL, Demokan S, Kim MS, Chang SS, Zhong X et al. (2010). Identification of hypermethylated genes associated with cisplatin resistance in human cancers. Cancer Res 70: 2870–2879.

    CAS  Article  Google Scholar 

  9. Dai W, Teodoridis JM, Graham J, Zeller C, Huang TH, Yan P et al. (2008). Methylation linear discriminant analysis (MLDA) for identifying differentially methylated CpG islands. BMC Bioinformatics 9: 337.

    Article  Google Scholar 

  10. Dai W, Teodoridis J, Zeller C, Graham JS, Hersey JM, Flanagan JM et al. (2011). Systematic CpG islands methylation profiling of genes in the Wnt pathway in epithelial ovarian cancer identifies biomarkers of progression-free survival. Clin Cancer Res 17: 4052–4062.

    CAS  Article  Google Scholar 

  11. Dean M, Fojo T, Bates S . (2005). Tumour stem cells and drug resistance. Nat Rev Cancer 5: 275–284.

    CAS  Article  Google Scholar 

  12. Eckstein N, Servan K, Hildebrandt B, Politz A, von Jonquieres G, Wolf-Kummeth S et al. (2009). Hyperactivation of the insulin-like growth factor receptor I signaling pathway is an essential event for cisplatin resistance of ovarian cancer cells. Cancer Res 69: 2996–3003.

    CAS  Article  Google Scholar 

  13. Edwards SL, Brough R, Lord CJ, Natrajan R, Vatcheva R, Levine DA et al. (2008). Resistance to therapy caused by intragenic deletion in BRCA2. Nature 451: 1111–1115.

    CAS  Article  Google Scholar 

  14. Esteller M . (2008). Epigenetics in cancer. N Engl J Med 358: 1148–1159.

    CAS  Article  Google Scholar 

  15. Gardiner-Garden M, Frommer M . (1987). CpG islands in vertebrate genomes. J Mol Biol 196: 261–282.

    CAS  Article  Google Scholar 

  16. Gifford G, Paul J, Vasey PA, Kaye SB, Brown R . (2004). The acquisition of hMLH1 methylation in plasma DNA after chemotherapy predicts poor survival for ovarian cancer patients. Clin Cancer Res 10: 4420–4426.

    CAS  Article  Google Scholar 

  17. Glasspool RM, Teodoridis JM, Brown R . (2006). Epigenetics as a mechanism driving polygenic clinical drug resistance. Br J Cancer 94: 1087–1092.

    CAS  Article  Google Scholar 

  18. Greenman C, Stephens P, Smith R, Dalgliesh GL, Hunter C, Bignell G et al. (2007). Patterns of somatic mutation in human cancer genomes. Nature 446: 153–158.

    CAS  Article  Google Scholar 

  19. Irizarry RA, Hobbs B, Collin F, Beazer-Barclay YD, Antonellis KJ, Scherf U et al. (2003). Exploration, normalization, and summaries of high density oligonucleotide array probe level data. Biostatistics 4: 249–264.

    Article  Google Scholar 

  20. Issa JP . (2004). CpG island methylator phenotype in cancer. Nat Rev Cancer 4: 988–993.

    CAS  Article  Google Scholar 

  21. Johnstone RW, Ruefli AA, Lowe SW . (2002). Apoptosis: a link between cancer genetics and chemotherapy. Cell 108: 153–164.

    CAS  Article  Google Scholar 

  22. Jones PA, Baylin SB . (2002). The fundamental role of epigenetic events in cancer. Nat Rev Genet 3: 415–428.

    CAS  Article  Google Scholar 

  23. Jung H, Lee SK, Jho EH . (2011). Mest/Peg1 inhibits Wnt signaling via regulation of LRP6 glycosylation. Biochem J 436: 263–269.

    CAS  Article  Google Scholar 

  24. Kartalou M, Essigmann JM . (2001). Recognition of cisplatin adducts by cellular proteins. Mutat Res 478: 1–21.

    CAS  Article  Google Scholar 

  25. Langdon SP, Lawrie SS, Hay FG, Hawkes MM, McDonald A, Hayward IP et al. (1988). Characterization and properties of nine human ovarian adenocarcinoma cell lines. Cancer Res 48: 6166–6172.

    CAS  PubMed  Google Scholar 

  26. Lefebvre L, Viville S, Barton SC, Ishino F, Keverne EB, Surani MA . (1998). Abnormal maternal behaviour and growth retardation associated with loss of the imprinted gene Mest. Nat Genet 20: 163–169.

    CAS  Article  Google Scholar 

  27. Li M, Balch C, Montgomery JS, Jeong M, Chung JH, Yan P et al. (2009). Integrated analysis of DNA methylation and gene expression reveals specific signaling pathways associated with platinum resistance in ovarian cancer. BMC Med Genomics 2: 34.

    Article  Google Scholar 

  28. Luqmani YA . (2005). Mechanisms of drug resistance in cancer chemotherapy. Med Princ Pract 14 (Suppl 1): 35–48.

    Article  Google Scholar 

  29. Nakanishi H, Suda T, Katoh M, Watanabe A, Igishi T, Kodani M et al. (2004). Loss of imprinting of PEG1/MEST in lung cancer cell lines. Oncol Rep 12: 1273–1278.

    CAS  PubMed  Google Scholar 

  30. O'Brien V, Brown R . (2006). Signalling cell cycle arrest and cell death through the MMR System. Carcinogenesis 27: 682–692.

    CAS  Article  Google Scholar 

  31. Ozols RF . (2004). Advanced ovarian cancer: a clinical update on first-line treatment, recurrent disease, and new agents. J Natl Compr Canc Netw 2 (Suppl 2): S60–S73.

    CAS  PubMed  Google Scholar 

  32. Papouli E, Cejka P, Jiricny J . (2004). Dependence of the cytotoxicity of DNA-damaging agents on the mismatch repair status of human cells. Cancer Res 64: 3391–3394.

    CAS  Article  Google Scholar 

  33. Pedersen IS, Dervan P, McGoldrick A, Harrison M, Ponchel F, Speirs V et al. (2002). Promoter switch: a novel mechanism causing biallelic PEG1/MEST expression in invasive breast cancer. Hum Mol Genet 11: 1449–1453.

    CAS  Article  Google Scholar 

  34. Peng C, Zhang X, Yu H, Wu D, Zheng J . (2011). Wnt5a as a predictor in poor clinical outcome of patients and a mediator in chemoresistance of ovarian cancer. Int J Gynecol Cancer 21: 280–288.

    Article  Google Scholar 

  35. Plumb JA, Strathdee G, Sludden J, Kaye SB, Brown R . (2000). Reversal of drug resistance in human tumor xenografts by 2′-deoxy-5-azacytidine-induced demethylation of the hMLH1 gene promoter. Cancer Res 60: 6039–6044.

    CAS  PubMed  Google Scholar 

  36. Rajasekhar VK, Dalerba P, Passegue E, Lagasse E, Najbauer J . (2008). The 5th International Society for Stem Cell Research (ISSCR) Annual Meeting, June 2007. Stem Cells 26: 292–298.

    CAS  Article  Google Scholar 

  37. Rizzo S, Hersey JM, Mellor P, Dai W, Santos-Silva A, Liber D et al. (2011). Ovarian cancer stem cell-like side populations are enriched following chemotherapy and overexpress EZH2. Mol Cancer Ther 10: 325–335.

    CAS  Article  Google Scholar 

  38. Saxonov S, Berg P, Brutlag DL . (2006). A genome-wide analysis of CpG dinucleotides in the human genome distinguishes two distinct classes of promoters. Proc Natl Acad Sci USA 103: 1412–1417.

    CAS  Article  Google Scholar 

  39. Schmittgen TD, Livak KJ . (2008). Analyzing real-time PCR data by the comparative C(T) method. Nat Protoc 3: 1101–1108.

    CAS  Article  Google Scholar 

  40. Sharma SV, Lee DY, Li B, Quinlan MP, Takahashi F, Maheswaran S et al. (2010). A chromatin-mediated reversible drug-tolerant state in cancer cell subpopulations. Cell 141: 69–80.

    CAS  Article  Google Scholar 

  41. Smith CA, McClive PJ, Sinclair AH . (2005). Temporal and spatial expression profile of the novel armadillo-related gene, Alex2, during testicular differentiation in the mouse embryo. Dev Dyn 233: 188–193.

    CAS  Article  Google Scholar 

  42. Steele N, Finn P, Brown R, Plumb JA . (2009). Combined inhibition of DNA methylation and histone acetylation enhances gene re-expression and drug sensitivity in vivo. Br J Cancer 100: 758–763.

    CAS  Article  Google Scholar 

  43. Stojic L, Brun R, Jiricny J . (2004). Mismatch repair and DNA damage signalling. DNA Repair (Amst) 3: 1091–1101.

    CAS  Article  Google Scholar 

  44. Strathdee G, MacKean MJ, Illand M, Brown R . (1999). A role for methylation of the hMLH1 promoter in loss of hMLH1 expression and drug resistance in ovarian cancer. Oncogene 18: 2335–2341.

    CAS  Article  Google Scholar 

  45. Suzuki H, Gabrielson E, Chen W, Anbazhagan R, van Engeland M, Weijenberg MP et al. (2002). A genomic screen for genes upregulated by demethylation and histone deacetylase inhibition in human colorectal cancer. Nat Genet 31: 141–149.

    CAS  Article  Google Scholar 

  46. Szotek PP, Pieretti-Vanmarcke R, Masiakos PT, Dinulescu DM, Connolly D, Foster R et al. (2006). Ovarian cancer side population defines cells with stem cell-like characteristics and Mullerian inhibiting substance responsiveness. Proc Natl Acad Sci USA 103: 11154–11159.

    CAS  Article  Google Scholar 

  47. Teodoridis JM, Hall J, Marsh S, Kannall HD, Smyth C, Curto J et al. (2005). CpG island methylation of DNA damage response genes in advanced ovarian cancer. Cancer Res 65: 8961–8967.

    CAS  Article  Google Scholar 

  48. Wang D, Lippard SJ . (2005). Cellular processing of platinum anticancer drugs. Nat Rev Drug Discov 4: 307–320.

    CAS  Article  Google Scholar 

  49. Wood LD, Parsons DW, Jones S, Lin J, Sjoblom T, Leary RJ et al. (2007). The genomic landscapes of human breast and colorectal cancers. Science 318: 1108–1113.

    CAS  Article  Google Scholar 

  50. Yoshioka K, Yoshioka Y, Hsieh P . (2006). ATR kinase activation mediated by MutSalpha and MutLalpha in response to cytotoxic O6-methylguanine adducts. Mol Cell 22: 501–510.

    CAS  Article  Google Scholar 

Download references

Acknowledgements

We thank Lisa McMillan for her valuable contribution to the initial analysis of the Affymetrix expression data. We also thank Nahal Masrour for preparation of samples and Kerra Pearce for running the Infinium HumanMethylation450 BeadChips at the UCL Genomics Center, London. This work was supported by a Cancer Research UK (CZ, WD and RB) grant (C536/A6689), Imperial Experimental Cancer Medicine Centre, Imperial Biomedical Research Centre and Ovarian Cancer Action (SR and CSMWB).

Author information

Affiliations

Authors

Corresponding author

Correspondence to R Brown.

Ethics declarations

Competing interests

The authors declare no conflict of interest.

Additional information

Supplementary Information accompanies the paper on the Oncogene website

Supplementary information

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Zeller, C., Dai, W., Steele, N. et al. Candidate DNA methylation drivers of acquired cisplatin resistance in ovarian cancer identified by methylome and expression profiling. Oncogene 31, 4567–4576 (2012). https://doi.org/10.1038/onc.2011.611

Download citation

Keywords

  • drug resistance
  • cisplatin
  • DNA methylation
  • ovarian cancer
  • DNMTi
  • HDACi

Further reading

Search

Quick links