In vertebrates an inverse relationship between gene expression and DNA methylation is now well established but its mechanism is incompletely understood. Housekeeping genes, which are expressed in every cell are characterized by relatively high densities of unmethylated CpG islands in the promoter region; whenever these CpG residues become methylated the gene become transcriptionally inert. By contrast in tissue specific genes that are normally repressed, Cpg sequences are methylated but become hypomethylated where, or when, the specific genes are expressed. It is noteworthy that the effect of promoter methylation on gene repression can b best demonstrated in vivo. We have investigated two aspects of this multifaceted problem: a) the biochemical mechanism underlying conversion of mCpG residues to CpG residues, and b) the mechanism linking promoter methylation and inhibition of gene expression. Conversion of mCpG to CpG by removal of the methyl group from methyl cytidine is biochemically impossible as this reaction would involve cleavag of a C-C bond; in order to account for the loss of mCpG during gene expression two different models have been proposed: i) we and others have shown that hypomethylation occurring concomitantly with cell duplication ca be explained by inhibition of maintenance methylase through two cycles of DNA replication ii) in the absence of DNA duplication hypomethylation involves replacement of a large fraction of mC in mCpG sequences with cytidine as first demonstrated by Cantoni, Razin and collaborators. The mechanism by which the presence of methyl groups at the promoter region inhibit gene activity has been pursued by use of engineered constructs in transient transfections. The most effective suppression was observed when methylation was in the pre-initiation domain. Our recent results support earlier suggestion that a mediator protein is involved in the mechanism of promoter inhibition.
