Shall we crosstalk? – The relationship between DNA methylation and histone H3 lysine 27 trimethylation (comment on DOI 10.1002/bies.201300130)

Abstract

Cancer cells usurp silencing mechanisms to extinguish functional pathways by epigenetic processes. There are two important epigenetic silencing mechanisms, polycomb repressive complex 2 (PRC2)-based histone H3 lysine 27 trimethylation (H3K27me3), and DNA methylation, which are associated with a variety of cellular functions. Studies have shown that these two epigenetic marks seem to be dependent on one another, although the precise mechanistic link between them remains to be fully characterized 1. This crosstalk between DNA methylation and PRC2-H3K27me3 has implications for understanding stem cell differentiation, somatic cell reprogramming, and tumorigenesis as well: dysregulation of these two epigenetic mechanisms has frequently been observed in many types of cancers. It is well appreciated that stem cell PRC2 targets – which are generally associated with embryonic development – are more likely to have cancer-specific promoter DNA methylation than non-targets 1. During tumorigenesis, these vulnerable genes undergo replacement of dynamic H3K27me3 modification with static DNA methylation. By contrast, DNA methylation in CpG sequences interferes with PRC2 recruitment 2. H3K27me3 and DNA methylation are mutually exclusive in CpG islands (CGIs), as ascertained via precise genome-wide analyses 3. These data lead to an idea that during tumorigenesis, aberrant DNA methylation patterns affect the affinity of PRC2 binding and drive redistribution of the PRC2 pattern. Recently, Meehan and coworkers 4, 5 reported that DNA methylation could be a mark for appropriate polycomb-mediated gene repression by constraining PRC2 targeting. Since dense DNA methylation in CGIs interferes with the binding of polycomb protein components, altered DNA methylation patterns in cancer cells influence the PRC2 binding landscape. Meehan et al. uncovered an unappreciated plausible role of DNA methylation in gene regulation. Consistently, studies have shown that artificial removal of DNA methylation induces the accumulation of PRC2 in those genomic loci 3. Fascinatingly, data from these studies evoke another plausible feature of PRC2. PRC2 is mostly enriched in CGIs, although precise mechanisms of its recruitment are not fully understood so far. It is possible that PRC2 stochastically and preferentially modifies DNA at certain loci with low transcription activity but lack of DNA methylation. Since DNA methylated CGIs generally confer low transcription activity, removal of DNA methylation in the CGIs may provide preferential targets for PRC2 binding. Consequently, high levels of DNA methylation and H3K27me3 are rarely found at the same silenced loci, as was reported in whole epigenome analyses. Hence, a direct mechanistic link between PRC2 and DNA methylation may not be required in this regard. An aberrant epigenome is a hallmark of cancer. Understanding such complex epigenetic abnormalities may open the door for developing better therapeutic strategies for cancer treatment.