Using genome-wide methylation mapping to find oligodendroglial tumour suppressor genes

Oligodendrogliomas (ODGs) make up 5-30% of gliomas, the most common type of primary brain tumour. Most have allelic loss of the entire 1p and 19q chromosomal arms. This suggests that a tumour suppressor gene or genes, involved in the early stages of ODG formation, may be present in 1p and/or 19q. According to Knudsen's "two-hit" hypothesis, tumourigenesis requires the inactivation of both alleles of a tumour suppressor gene; despite extensive searches, mutations in the remaining copy of 1p and 19q are rarely found in ODG with 1p/19q loss. Our hypothesis is that aberrant epigenetic silencing (epimutation), characterised by DNA methylation, is the mechanism by which the remaining allele of tumour suppressor gene(s) on 1p and/or 19q is silenced. DNA methylation occurs at CpG dinucleotides, which are often concentrated at gene promoters, and called CpG islands. Aberrant CpG island methylation is strongly associated with transcriptional silence, thus an epimutation creates the functional equivalent of an inactivating mutation. Using methylated DNA immunoprecipitation (MeDIP) and Nimblegen CpG island-plus-promoter arrays, we profiled the epigenetic state of all CpG islands and gene promoter regions in seven ODGs, and compared them to four control brain samples. We have now obtained a comprehensive view of the methylation of all gene promoters on 1p and 19q. We are currently validating these array data by Combined-Bisulphite Restriction Analysis (COBRA) which provides a quantitative assessment of gene promoter methylation, and correlating our findings with expression data from these same tumours. A number of genes have been identified as aberrantly epigenetically silenced, and hence are candidate tumour suppressor genes for validation in a larger cohort of ODGs.