One major contribution of this work was to provide a detailed molecular view (epigenomic "roadmap") of the earliest chromatin events following Xist RNA coating of the chromosome in cis in ESCs. This represents a unique resource for the XCI community and for scientists interested in chromatin and gene expression. This epigenomic roadmap revealed the earliest chromatin events during XCI, leading us to predict that specific chromatin modifiers (histone deacetylases) might be involved. We went on to show that this is indeed the case, with a histone deacetylase, HDAC3 ensuring rapid histone deacetylation. We also revealed the precise kinetics of Polycomb complex associated chromatin modifications during XCI, showing that PRC1-mediated H2A ubiquitination is a very early mark after Xist RNA coating, and this followed by PRC2-mediated H3K27me3. We also discovered that these Polycomb marks can only spread into genes, if and when they become transcriptionally silenced. Finally, we were able to visualize the dynamic appearance of chromatin changes including H3K27me3, H4K20me1 and Xist RNA itself using fluorescent tagging of all three in ESCs.
Another major contribution of this work was to decipher the roles of some of the key factors involved in the initiation of XCI. One such factor we explored was the SPEN protein, a previously identified partner of Xist RNA. We found that SPEN is essential for initiating gene silencing on the X chromosome in preimplantation mouse embryos and in embryonic stem cells. SPEN was found to be dispensable for maintenance of XCI in neural progenitors, but its depletion nevertheless led to increased expression of genes that escape XCI. We also elucidated SPEN’s mode of action and the other factors that it interacts with. We found that SPEN is recruited to the X chromosome by Xist RNA, but specifically targets only the enhancers and promoters of transcriptionally active genes, rapidly disengaging from chromatin as soon as gene silencing is achieved. This suggests that active transcription is required to tether SPEN to chromatin. We also defined the SPOC domain as the major effector of the gene-silencing function of SPEN. Indeed, artificial tethering SPOC to Xist RNA was shown to be sufficient to mediate silencing of many genes along the X when endogenous SPEN was depleted. We also identified the protein partners of SPOC, which include NCoR/SMRT, the m6A RNA methylation machinery, the NuRD complex, RNA polymerase II and factors involved in the regulation of transcription initiation and elongation. Based on our findings, we have proposed that SPEN acts as a molecular integrator for the initiation of XCI, bridging Xist RNA with the transcription machinery, as well as with nucleosome remodellers and histone deacetylases at active enhancers and promoters.
