While we’ve been fascinated by how epigenetic alterations during development SET the stage for adult life, in order to fully understand this relationship, we also have to examine how the stage for the development is SET by our epigenome. The epigenome of the oocyte is reprogrammed during early embryonic development; however, some marks remain at important sites. For example, DNA methylation is maintained at imprinted loci and histone modification appears to be important in X chromosome inactivation. But the features of the epigenetic landscape during our developmental journey don’t end there.
H3K36me is associated with transcribed regions and functions in transcription fidelity, RNA splicing, and DNA repair. H3K36me3 is primarily catalyzed by the SETD2 methyltransferase and SETD2 deficiency is embryonic lethal. H3K36me3 participates in cross-talk with other chromatin marks, including antagonizing H3K4me3 and H3K27me3, and promoting de novo DNA methylation. However, how maternal SETD2 and H3K36me3 regulate early development, including the oocyte epigenome, remains poorly understood. To tackle this question labs of Wei Li (Chinese Academy of Sciences), Li Li (Shanghai Jiao Tong University), and Wei Xie (Tsinghua University) created mouse embryos and oocytes with homozygous Setd2 deletion using a Cre-lox system. They used a low-input ChIP-seq approach that they previously developed to assess histone mark distribution (H3K4me3, H3K27me3, & H3K36me3), STEM-seq (small-scale TELP-enabled methylome sequencing) to assess DNA methylation, and RNA-seq to assess expression in oocytes and other early cell type and found that:
- H3K36me3 peaks correlate with gene expression in
growing oocytes and with DNA methylation in fully grown oocytes
- H3K4me3 and H3K27me3 largely reside in untranscribed regions
- In Setd2 deficient oocytes, H3K36me3 is strongly reduced and DNA methylation is reduced
at regions that should have been marked with H3K36me3
- Regions outside former H3K36me3 territories gain ectopic DNA methylation
- H3K4me3 and H3K27me3 are globally increased in Setd2 null oocytes
- All maternal imprints fail to be properly established in Setd2 null oocytes, acquiring H3K4me3 in the absence of wild-type H3K36me3 and DNA methylation profiles
- Setd2 null oocytes arrest cell division at the one-cell stage after fertilization
The authors propose a model in which H3K36me3 and DNA methylation occur in gene bodies in normal oocytes, but in the absence of maternal SETD2, H3K4me3 and H3K27me3 occur ectopically in H3K36me3-lacking regions. DNA methylation is lost at transcribed regions including imprinted regions. Overall, this study establishes SETD2 as a critical regulator of the epigenetic maturation of oocytes and fertility. Furthermore, the skillful combination of low-input ChIP-seq and genetic knockout models provides a much clearer picture of the complex cross-talk between histone marks. The team definitely achieved what they SET out to do, and have SET the stage for more work on the complex world or histone modifications in early development.
Get all the details in Nature Genetics, May 2019.