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

BACKGROUND: Hypermethylation 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). RESULTS: MLDA 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. CONCLUSION: MLDA 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.

CpG island methylation of DNA damage response genes in advanced ovarian cancer.

We have determined the methylation frequencies of 24 CpG islands of genes associated with DNA damage responses or with ovarian cancer in 106 stage III/IV epithelial ovarian tumors. We have analyzed this data for whether there is evidence of a CpG island methylator phenotype or associations of CpG island methylation with response to chemotherapy in advanced ovarian cancer. Frequent methylation was observed for OPCML, DCR1, RASSF1A, HIC1, BRCA1, and MINT25 (33.3%, 30.7%, 26.4%, 17.3%, 12.3%, and 12.0%, respectively), whereas no methylation was observed for APAF-1, DAPK, FANCF, FAS, P14, P21, P73, SOCS-3, and SURVIVIN. The remaining genes showed only a low frequency of methylation, <10%. Unsupervised gene shaving identified a nonrandom pattern of methylation for OPCML, DCR1, RASSF1A, MINT25, HIC1, and SFRP1, supporting the concept of concordant methylation of these genes in ovarian cancer. Methylation of at least one of the group of genes involved in DNA repair/drug detoxification (BRCA1, GSTP1, and MGMT) was associated with improved response to chemotherapy (P = 0.013). We have examined the frequency of a polymorphism in the DNA methyltransferase gene DNMT3b6, which has been previously reported to affect gene transcription and cancer risk. The genetic polymorphism in the DNMT3b6 gene promoter (at position -149) is not significantly associated with the concordant methylation observed, but is weakly associated with the overall frequency of methylation at the genes examined (P = 0.04, n = 56). This supports the hypothesis that genetic factors affecting function of DNMT genes may underlie the propensity of tumors to acquire aberrant CpG island methylation.

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.

CpG island methylator phenotype (CIMP) in cancer: causes and implications.

Strong evidence exists for a subgroup of tumours, from a variety of tissue types, exhibiting concordant tumour specific DNA methylation: the "CpG island methylator phenotype" (CIMP). Occurrence of CIMP is associated with a range of genetic and environmental factors, although the molecular causes are not well-understood. Both increased expression and aberrant targeting of DNA methyltransferases (DNMTs) could contribute to the occurrence of CIMP. One under-explored area is the possibility that DNA damage may induce or select for CIMP during carcinogenesis or treatment of tumours with chemotherapy. DNA damaging agents can induce DNA damage at guanine rich regions throughout the genome, including CpG islands. This DNA damage can result in stalled DNA synthesis, which will lead to localised increased DNMT1 concentration and therefore potentially increased DNA methylation at these sites. Chemotherapy can select for cells which have increased tolerance to DNA damage due to increased lesion bypass, in some cases by mechanisms which involve inactivation of genes by CpG island methylation. CIMP has been associated with worse patient prognosis, probably due to increased epigenetic plasticity. Therefore, further clinical testing of the diagnostic and prognostic value of the current CIMP markers, as well as increasing our understanding of the molecular causes underlying CIMP are required.

Epigenetic silencing mediated by CpG island methylation: potential as a therapeutic target and as a biomarker.

Many genes become transcriptionally silenced during the development of cancer. As well as affecting disease progression, gene silencing has the potential to influence drug resistance and clinical outcome following therapy. In addition to silencing due to gene mutations, covalent epigenetic modifications such as DNA hypermethylation and histone post-translational modifications are associated with transcriptional inactivation of many genes and are an important early event during carcinogenesis and tumour development. Aberrant methylation of CpG islands in promoters is associated with transcriptional inactivation of genes involved in all aspects of tumour development. Genes involved in key DNA damage response pathways, such as cell cycle control, apoptosis signalling and DNA repair, can frequently become methylated and epigenetically silenced in tumours. This may lead to differences in intrinsic sensitivity of tumours to chemotherapy, depending on the specific function of the gene inactivated. Furthermore, it is proposed that chemotherapy itself can exert a selective pressure on epigenetically silenced drug sensitivity genes present in subpopulations of cells, leading to acquired chemoresistance. Since the DNA sequence of epigenetically inactivated genes are not mutated but rather subject to reversible modifications via DNA methyltransferases (DNMTs) or histone modification, it is possible to reverse silencing using small molecule inhibitors. Such compounds show anti-tumour activity and can increase the sensitivity of drug resistant preclinical tumour models. Clinical trials of epigenetic therapies are now underway. Epigenetic profiling, using DNA methylation and histone analysis, will provide guidance on optimisation of these therapies with conventional chemotherapy and will help identify patient populations who may particularly benefit from such approaches.

During differentiation of the monocytic cell line U937, Pur alpha mediates induction of the CD11c beta 2 integrin gene promoter.

CD11c is a member of the beta(2) integrin family of adhesion molecules that, together with CD18, forms a heterodimeric receptor on the surface of myeloid, NK, dendritic, and certain leukemic, lymphoma, and activated lymphoid cells. Monocytic differentiation is associated with an induction of both CD11c and CD18 gene expression. The resulting CD11c/CD18 receptor mediates firm adhesion to the vascular endothelium, transendothelial migration, chemotaxis, and phagocytosis. Monocytic differentiation can be mimicked in vitro by treatment of the promonocytic cell line U937 with PMA. Recently, we reported that in U937 cells, expression of the CD11c gene is controlled by an unidentified transcription factor that binds ssDNA. This finding suggested that DNA secondary structure plays an important role in controlling the CD11c gene and prompted us to search for additional ssDNA-binding activities with which this gene interacts. In this study, we report that in U937 cells, expression of the CD11c gene is mediated by the ssDNA-binding protein Puralpha. During PMA-induced differentiation, the ability of Puralpha to activate the CD11c promoter in U937 cells increases, as does that of Sp1. Together, these increases in the functional activity of both Puralpha and Sp1 combine to induce CD11c expression.

CD11c gene expression in hairy cell leukemia is dependent upon activation of the proto-oncogenes ras and junD.

Hairy cell leukemia (HCL) is a chronic lymphoproliferative disease, the cause of which is unknown. Diagnostic of HCL is abnormal expression of the gene that encodes the beta2 integrin CD11c. In order to determine the cause of CD11c gene expression in HCL the CD11c gene promoter was characterized. Transfection of the CD11c promoter linked to a luciferase reporter gene indicated that it is sufficient to direct expression in hairy cells. Mutation analysis demonstrated that of predominant importance to the activity of the CD11c promoter is its interaction with the activator protein-1 (AP-1) family of transcription factors. Comparison of nuclear extracts prepared from hairy cells with those prepared from other cell types indicated that hairy cells exhibit abnormal constitutive expression of an AP-1 complex containing JunD. Functional inhibition of AP-1 expressed by hairy cells reduced CD11c promoter activity by 80%. Inhibition of Ras, which represents an upstream activator of AP-1, also significantly inhibited the CD11c promoter. Furthermore, in the hairy cell line EH, inhibition of Ras signaling through mitogen-activated protein kinase/extracellular signal-regulated kinase kinases 1 and 2 (MEK1/2) reduced not only CD11c promoter activity but also reduced both CD11c surface expression and proliferation. Expression in nonhairy cells of a dominant-positive Ras mutant activated the CD11c promoter to levels equivalent to those in hairy cells. Together, these data indicate that the abnormal expression of the CD11c gene characteristic of HCL is dependent upon activation of the proto-oncogenes ras and junD.