Methylation-Mediated Proviral Silencing Is Associated with MeCP2 Recruitment and Localized Histone H3 Deacetylation

Abstract

The majority of 5-methylcytosine in mammalian DNA resides in endogenous transposable elements and is associated with the transcriptional silencing of these parasitic elements. Methylation also plays an important role in the silencing of exogenous retroviruses. One of the difficulties inherent in the study of proviral silencing is that the sites in which proviruses randomly integrate influence the probability of de novo methylation and expression. In order to compare methylated and unmethylated proviruses at the same genomic site, we used a recombinase-based targeting approach to introduce an in vitro methylated or unmethylated Moloney murine leukemia-based provirus in MEL cells. The methylated and unmethylated states are maintained in vivo, with the exception of the initially methylated proviral enhancer, which becomes demethylated in vivo. Although the enhancer is unmethylated and remodeled, the methylated provirus is transcriptionally silent. To further analyze the repressed state, histone acetylation status was determined by chromatin immunoprecipitation (ChIP) analyses, which revealed that localized histone H3 but not histone H4 hyperacetylation is inversely correlated with proviral methylation density. Since members of the methyl-CpG binding domain (MBD) family of proteins recruit histone deacetylase activity, these proteins may play a role in proviral repression. Interestingly, only MBD3 and MeCP2 are expressed in MEL cells. ChIPs with antibodies specific for these proteins revealed that only MeCP2 associates with the provirus in a methylation-dependent manner. Taken together, our results suggest that MeCP2 recruitment to a methylated provirus is sufficient for transcriptional silencing, despite the presence of a remodeled enhancer.

ACKNOWLEDGMENTS

This work was supported by NIH fellowship GM 19767/01to M.C.L., a fellowship from the Rett Syndrome Research Foundation to D.S., and NIH grants DK44746 and HL57620 to M.G.

We thank M. Bender for mouse brain cDNA; Eric Bouhassira and the members of the Groudine laboratory for suggestions; Claire Francastel and Tomoyuki Sawado for comments on the manuscript; and Joan Hamilton, David Scalzo, Jennifer Stout, and Urszula Maliszewski for technical assistance.