Constanze Zeller

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

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.

Systematic CpG Islands Methylation Profiling of Genes in the Wnt Pathway in Epithelial Ovarian Cancer Identifies Biomarkers of Progression-Free Survival

Purpose: Wnt pathways control key biological processes that potentially impact on tumor progression and patient survival. We aimed to evaluate DNA methylation at promoter CpG islands (CGI) of Wnt pathway genes in ovarian tumors at presentation and identify biomarkers of patient progression-free survival (PFS). Experimental Design: Epithelial ovarian tumors (screening study n = 120, validation study n = 61), prospectively collected through a cohort study, were analyzed by differential methylation hybridization at 302 loci spanning 189 promoter CGIs at 137 genes in Wnt pathways. The association of methylation and PFS was examined by Cox proportional hazards model. Results: DNA methylation is associated with PFS at 20 of 302 loci (P < 0.05, n = 111), with 5 loci significant at false discovery rate (FDR) less than 10%. A total of 11 of 20 loci retain significance in an independent validation cohort (n = 48, P ≤ 0.05, FDR ≤ 10%), and 7 of these loci, at FZD4, DVL1, NFATC3, ROCK1, LRP5, AXIN1, and NKD1 genes, are independent from clinical parameters (adjusted P < 0.05). Increased methylation at these loci associates with increased hazard of disease progression. A multivariate Cox model incorporates only NKD1 and DVL1, identifying two groups differing in PFS [HR = 2.09; 95% CI (1.39–3.15); permutation test P < 0.005]. Methylation at DVL1 and NFATC3 show significant association with response. Consistent with their epigenetic regulation, reduced expression of FZD4, DVL1, and ROCK1 is an indicator of early-disease relapse in an independent ovarian tumor cohort (n = 311, adjusted P < 0.05). Conclusion: The data highlight the importance of epigenetic regulation of multiple promoter CGIs of Wnt pathway genes in ovarian cancer and identify methylation at NKD1 and DVL1 as independent predictors of PFS. Clin Cancer Res; 17(12); 4052–62. ©2011 AACR.

Chromatin H3K27me3/H3K4me3 histone marks define gene sets in high-grade serous ovarian cancer that distinguish malignant, tumour-sustaining and chemo-resistant ovarian tumour cells

In embryonic stem (ES) cells, bivalent chromatin domains containing H3K4me3 and H3K27me3 marks silence developmental genes, while keeping them poised for activation following differentiation. We have identified gene sets associated with H3K27me3 and H3K4me3 marks at transcription start sites in a high-grade ovarian serous tumour and examined their association with epigenetic silencing and malignant progression. This revealed novel silenced bivalent marked genes, not described previously for ES cells, which are significantly enriched for the PI3K (P<10−7) and TGF-β signalling pathways (P<10−5). We matched histone marked gene sets to gene expression sets of eight normal fallopian tubes and 499 high-grade serous malignant ovarian samples. This revealed a significant decrease in gene expression for the H3K27me3 and bivalent gene sets in malignant tissue. We then correlated H3K27me3 and bivalent gene sets to gene expression data of ovarian tumour ‘stem cell-like’ sustaining cells versus non-sustaining cells. This showed a significantly lower expression for the H3K27me3 and bivalent gene sets in the tumour-sustaining cells. Similarly, comparison of matched chemo-sensitive and chemo-resistant ovarian cell lines showed a significantly lower expression of H3K27me3/bivalent marked genes in the chemo-resistant compared with the chemo-sensitive cell line. Our analysis supports the hypothesis that bivalent marks are associated with epigenetic silencing in ovarian cancer. However it also suggests that additional tumour specific bivalent marks, to those known in ES cells, are present in tumours and may potentially influence the subsequent development of drug resistance and tumour progression.

CpG-island methylation study of liver fluke-related cholangiocarcinoma

Background:Genetic changes have been widely reported in association with cholangiocarcinoma (CCA), while epigenetic changes are poorly characterised. We aimed to further evaluate CpG-island hypermethylation in CCA at candidate loci, which may have potential as diagnostic or prognostic biomarkers.Methods:We analysed methylation of 26 CpG-islands in 102 liver fluke related-CCA and 29 adjacent normal samples using methylation-specific PCR (MSP). Methylation of interest loci was confirmed using pyrosequencing and/or combined bisulfite restriction analysis, and protein expression by immunohistochemistry.Results:A number of CpG-islands (OPCML, SFRP1, HIC1, PTEN and DcR1) showed frequency of hypermethylation in >28% of CCA, but not adjacent normal tissues. The results showed that 91% of CCA were methylated in at least one CpG-island. The OPCML was the most frequently methylated locus (72.5%) and was more frequently methylated in less differentiated CCA. Patients with methylated DcR1 had significantly longer overall survival (Median; 41.7 vs 21.7 weeks, P=0.027). Low-protein expression was found in >70% of CCA with methylation of OPCML or DcR1.Conclusion:Aberrant hypermethylation of certain loci is a common event in liver fluke-related CCA and may potentially contribute to cholangiocarcinogenesis. The OPCML and DcR1 might serve as methylation biomarkers in CCA that can be readily examined by MSP.

Promoter CpG island methylation of genes in key cancer pathways associates with clinical outcome in high-grade serous ovarian cancer

Purpose: We aimed to identify DNA methylation biomarkers of progression-free survival (PFS) to platinum-based chemotherapy in high-grade serous ovarian cancer (HGSOC) within biologically relevant ovarian cancer-associated pathways. Experimental Design: Association with PFS of CpG island (CGI) promoter DNA methylation at genes in the pathways Akt/mTOR, p53, redox, and homologous recombination DNA repair was sought with PFS as the primary objective in a prospectively collected ovarian cancer cohort (n = 150). Significant loci were validated for associations between PFS, methylation, and gene expression in an independent The Cancer Genome Atlas (TCGA) data set of HGSOC (n = 311). Results: DNA methylation at 29 CGI loci linked to 28 genes was significantly associated with PFS, independent from conventional clinical prognostic factors (adjusted P < 0.05). Of 17 out of the 28 genes represented in the TCGA data set, methylation of VEGFB, VEGFA, HDAC11, FANCA, E2F1, GPX4, PRDX2, RAD54L, and RECQL4 was prognostic in this independent patient cohort (one-sided P < 0.05, false discovery rate < 10%). A multivariate Cox model was constructed, with clinical parameters (age, stage, grade, and histologic type) and significant loci. The final model included NKD1, VEGFB, and PRDX2 as the three best predictors of PFS (P = 6.62 × 10−6, permutation test P < 0.05). Focussing only on known VEGFs in the TCGA cohort showed that methylation at promoters of VEGFA, VEGFB, and VEGFC was significantly associated with PFS. Conclusions: A three loci model of DNA methylation could identify two distinct prognostic groups of patients with ovarian cancer (PFS: HR = 2.29, P = 3.34 × 10−5; overall survival: HR = 1.87, P = 0.007) and patients more likely to have poor response to chemotherapy (OR = 3.45, P = 0.012). Clin Cancer Res; 19(20); 5788–97. ©2013 AACR.

Methylation Linear Discriminant Analysis (MLDA) for identifying differentially methylated CpG islands

BackgroundHypermethylation of promoter CpG islands is strongly correlated to transcriptional gene silencing and epigenetic maintenance of the silenced state. As well as its role in tumor development, CpG island methylation contributes to the acquisition of resistance to chemotherapy. Differential Methylation Hybridisation (DMH) is one technique used for genome-wide DNA methylation analysis. The study of such microarray data sets should ideally account for the specific biological features of DNA methylation and the non-symmetrical distribution of the ratios of unmethylated and methylated sequences hybridised on the array. We have therefore developed a novel algorithm tailored to this type of data, Methylation Linear Discriminant Analysis (MLDA).ResultsMLDA was programmed in R (version 2.7.0) and the package is available at CRAN [1]. This approach utilizes linear regression models of non-normalised hybridisation data to define methylation status. Log-transformed signal intensities of unmethylated controls on the microarray are used as a reference. The signal intensities of DNA samples digested with methylation sensitive restriction enzymes and mock digested are then transformed to the likelihood of a locus being methylated using this reference. We tested the ability of MLDA to identify loci differentially methylated as analysed by DMH between cisplatin sensitive and resistant ovarian cancer cell lines. MLDA identified 115 differentially methylated loci and 23 out of 26 of these loci have been independently validated by Methylation Specific PCR and/or bisulphite pyrosequencing.ConclusionMLDA has advantages for analyzing methylation data from CpG island microarrays, since there is a clear rational for the definition of methylation status, it uses DMH data without between-group normalisation and is less influenced by cross-hybridisation of loci. The MLDA algorithm successfully identified differentially methylated loci between two classes of samples analysed by DMH using CpG island microarrays.

Aberrant DNA Methylation at Genes Associated with a Stem Cell-like Phenotype in Cholangiocarcinoma Tumors

Genetic abnormalities of cholangiocarcinoma have been widely studied; however, epigenomic changes related to cholangiocarcinogenesis have been less well characterized. We have profiled the DNA methylomes of 28 primary cholangiocarcinoma and six matched adjacent normal tissues using Infiniums HumanMethylation27 BeadChips with the aim of identifying gene sets aberrantly and epigenetically regulated in this tumor type. Using a linear model for microarray data, we identified 1610 differentially methylated autosomal CpG sites, with 809 hypermethylated (representing 603 genes) and 801 hypomethylated (representing 712 genes) in cholangiocarcinoma versus adjacent normal tissues (false-discovery rate ≤ 0.05). Gene ontology and gene set enrichment analyses identified gene sets significantly associated with hypermethylation at linked CpG sites in cholangiocarcinoma including homeobox genes and target genes of PRC2, EED, SUZ12, and histone H3 trimethylation at lysine 27. We confirmed frequent hypermethylation at the homeobox genes HOXA9 and HOXD9 by bisulfite pyrosequencing in a larger cohort of cholangiocarcinoma (n = 102). Our findings indicate a key role for hypermethylation of multiple CpG sites at genes associated with a stem cell-like phenotype as a common molecular aberration in cholangiocarcinoma. These data have implications for cholangiocarcinogenesis, as well as possible novel treatment options using histone methyltransferase inhibitors. Cancer Prev Res; 6(12); 1348–55. ©2013 AACR.

Therapeutic modulation of epigenetic drivers of drug resistance in ovarian cancer.

Epigenetic changes in tumours are associated not only with cancer development and progression, but also with resistance to chemotherapy. Aberrant DNA methylation at CpG islands and associated epigenetic silencing are observed during the acquisition of drug resistance. However, it remains unclear whether all of the observed changes are drivers of drug resistance, causally associated with response of tumours to chemotherapy, or are passenger events representing chance DNA methylation changes. Systematic approaches that link DNA methylation and expression with chemosensitivity will be required to identify key drivers. Such drivers will be important prognostic or predicitive biomarkers, both to existing chemotherapies, but also to epigenetic therapies used to modulate drug resistance.

A Critical Evaluation of the PAXgene Tissue Fixation System: Morphology, Immunohistochemistry, Molecular Biology, and Proteomics.

OBJECTIVES To evaluate the PAXgene tissue fixation system. METHODS Clinical biospecimens (n = 46) were divided into PAXgene-fixed paraffin-embedded (PFPE), formalin-fixed paraffin-embedded (FFPE), and fresh-frozen (FF) blocks. PFPE and FFPE sections were compared for histology (H&E staining) and immunohistochemistry (14 antibodies) using tissue microarrays. PFPE, FFPE, and FF samples were compared in terms of RNA quality (RNA integrity number, polymerase chain reaction [PCR] amplicon length, and quantitative reverse transcription PCR), DNA quality (gel electrophoresis and methylation profiling) and protein quality (liquid chromatography-mass spectrometry [LC-MS/MS]). RESULTS PFPE protocol optimization was required in most cases and is described. RNA extracted from PFPE sections was considerably less degraded than that from FFPE sections but more degraded than that from FF blocks. Genomic-length DNA was extracted from PFPE and FF biospecimens, and methylation profiling showed PFPE and FF biospecimens to be almost indistinguishable. Only degraded DNA was extracted from FFPE biospecimens. PFPE sections yielded peptides that were slightly less amenable to LC-MS/MS analysis than FFPE sections, but FF gave slightly better results. CONCLUSIONS While it cannot be envisaged that PAXgene will replace formalin in a routine clinical setting, for specific projects or immunodiagnostics involving biospecimens destined for immunohistochemical or histologic staining and DNA or RNA analyses, PAXgene is a viable option.

The DNA Methylomes of Serous Borderline Tumors Reveal Subgroups With Malignant- or Benign-Like Profiles

Serous borderline tumors (SBOTs) are a challenging group of ovarian tumors positioned between benign and malignant disease. We have profiled the DNA methylomes of 12 low-grade serous carcinomas (LGSCs), 19 SBOTs, and 16 benign serous tumors (BSTs) across 27,578 CpG sites to further characterize the epigenomic relationship between these subtypes of ovarian tumors. Unsupervised hierarchical clustering of DNA methylation levels showed that LGSCs differ distinctly from BSTs, but not from SBOTs. Gene ontology analysis of genes showing differential methylation at linked CpG sites between LGSCs and BSTs revealed significant enrichment of gene groups associated with cell adhesion, cell-cell signaling, and the extracellular region, consistent with a more invasive phenotype of LGSCs compared with BSTs. Consensus clustering highlighted differences between SBOT methylomes and returned subgroups with malignant- or benign-like methylation profiles. Furthermore, a two-loci DNA methylation signature can distinguish between these SBOT subgroups with benign- and malignant-like methylation characteristics. Our findings indicate striking similarities between SBOT and LGSC methylomes, supporting a common origin and the view that LGSC may arise from SBOT. A subgroup of SBOTs can be classified into tumors with a benign- or a malignant-like methylation profile that may help in identifying tumors more likely to progress into LGSCs.