Final Report Summary - CHROMATINRNA (The role of CpG island RNAs and Polycomb-RNA interactions in developmental gene regulation) Polycomb repressive complex 2 (PRC2) modifies chromatin to maintain genes in a repressed state and is essential for cell differentiation during development and throughout life. As cells differentiate, PRC2 becomes lost from genes important for the new cell identity, but remains associated with those specific for other cell types. PRC2 binding to chromatin is often dysregulated in cancer, causing ectopic silencing of tumour suppressor genes. In addition to interacting with chromatin, PRC2 also binds RNA but the function of this, and how it relates to the observed patterns of PRC2 chromatin binding, remained unclear. To address this, we identified the RNAs directly bound by PRC2 in embryonic stem cells. We found that PRC2 directly binds nascent RNA at essentially all active genes. Although interacting with RNA promiscuously, analysis of sequences enriched at PRC2 RNA binding sites revealed preferential binding to G-tract sequences, especially those located 50 nucleotides into the first intron that are predicted to form G-quadruplex (G4) structures.Contrary to prevailing models, we demonstrated that RNA inhibits the interaction of PRC2 with chromatin. In vitro, the interaction of PRC2 with RNA or nucleosomes is mutually antagonistic. The inhibitory effect of RNA on nucleosome binding is enhanced if the RNA is folded into a G4 structure and can be traced to the PRC2 catalytic core. G4 RNA blocks the interaction of the PRC2 catalytic core with nucleosomes and can evict PRC2 from a pre-formed PRC2-nucleosome complex. In cells, RNA degradation triggers PRC2 recruitment to active genes. Furthermore, tethering G-tract RNA, but not other RNA sequences, to polycomb-target genes removes PRC2 from chromatin and depletes H3K27me3. This effect of G-tract RNA is specific to PRC2; other histone modifications are unaffected. This RNA tethering method also provides the means to reverse specific deleterious PRC2 recruitment events, demonstrated by the targeting of G-tract RNA to the tumour suppressor CDKN2A p16INK4A) in malignant rhabdoid tumour cells, which reactivates the gene and induces cell senescence. Our data thus support a model in which nascent pre-mRNA removes PRC2 during gene activation and prevents inappropriate silencing of active genes. G-tract RNA tethering also provides the means to selectively remove PRC2 from specific genes. This may have advantages over small molecule EZH2 inhibitors that reactivate polycomb-target genes across the genome. These data support consideration of nascent pre-mRNA as a regulatory molecule and not the passive intermediary that it is normally assumed to be. Further analysis of regulatory protein-nascent RNA interactions will likely reveal additional links between transcription and chromatin state fundamental for the control of cell identity.
