Interrogation of H3.3K27me3 in mESC pluripotency and differentiation through ChIP-seq, RNA-seq, and H3.3K27A RNA-seq analysis

Histone variant H3.3 is a critical regulator of cell differentiation and development. Likewise, methylation on lysine 27 of the canonical histone H3 (H3K27me3) has been previously characterized in development. However, the effect of methylation on lysine 27 on the variant histone H3.3 remains unclear. By applying proteomics and genomics techniques, we investigate the role of lysine 27 tri-methylation specifically on the histone variant H3.3 (H3.3K27me3) in the context of mouse embryonic stem cell pluripotency and differentiation as a model system for development. Using a custom-made antibody, we identify through ChIP-seq experiments and corresponding RNA-seq experiments that H3.3K27me3 does not show the same strong negative correlation with gene expression as is observed for canonical H3K27me3, nor the same strong positive correlation with gene expression as observed for H3.3. However, a unique enrichment was detected of H3.3K27me3 at lineage-specific genes, such as olfactory receptor genes, and at binding motifs for the transcription factors FOXJ2/3. REST, a predicted FOXJ2/3 target that acts as a transcriptional repressor of terminal neuronal genes, was identified with H3.3K27me3 at its promoter region. H3.3K27A mutant cells confirmed an upregulation of FOXJ2/3 targets upon the loss of methylation at H3.3K27 both at the transcriptomic and proteomic level. Thus, while canonical H3K27me3 has been characterized to regulate the expression of transcription factors that play a general role in differentiation, our work suggests H3.3K27me3 is essential for regulating distinct terminal differentiation genes. Overall design: Mouse embryonic stem cells (mESC) were grown in pluripotency with leukemia inhibitory factor (LIF), and then differentiated to embryoid bodies upon removal of LIF with the addition of 10 uM of retinoic acid (RA). Cells were collected at day 0 (pluripotency), and day 6 of differentiation. ChIP-seq was performed on sonicated chromatin extracted from the nuclei of previously frozen cell pellets that had been cross-linked with formaldehyde (in biological duplicates for each condition). For ChIP-seq, the pooled sample was sequenced three times to ensure enough total reads for analysis. RNA-seq was performed on polyA-enriched RNA in duplicate for each condition, with an additional day 3 of differentiation time point. The H3.3K27A and H3.3A knockout cells were generated using the CRISPR/Cas9 system. For the H3.3K27A mutant cells analysis by RNA-seq, cells were grown the same and samples were prepared the same as described above, with the only difference being that only day 0 and day 6 were collected for analysis.