DNA methylation modulates nucleosome retention in sperm and H3K4 methylation deposition in early mouse embryos

DNA methylation (DNAme) serves a stable gene regulatory function in somatic cells ([1][1]). In the germ line and during early embryogenesis, however, DNAme undergoes global erasure and re-establishment to support germ cell and embryonic development ([2][2]). While de novo DNAme acquisition during male germ cell development is essential for setting genomic DNA methylation imprints, other intergenerational roles for paternal DNAme in defining embryonic chromatin after fertilization are unknown. To approach this question, we reduced levels of DNAme in developing male germ cells through conditional gene deletion of the de novo DNA methyltransferases DNMT3A and DNMT3B in undifferentiated spermatogonia. We observed that DNMT3A serves a DNAme maintenance function in undifferentiated spermatogonia while DNMT3B catalyzes de novo DNAme during spermatogonial differentiation. Mutant male germ cells nevertheless completed their differentiation to sperm. Failing de novo DNAme in Dnmt3a / Dnmt3b double deficient spermatogonia is associated with increased nucleosome occupancy in mature sperm, preferentially at sites with higher CpG content, supporting the model that DNAme modulates nucleosome retention in sperm ([3][3]). To assess the impact of altered sperm chromatin in the formation of embryonic chromatin, we measured H3K4me3 occupancy at paternal and maternal alleles in 2-cell embryos using a newly developed transposon-based tagging assay for modified chromatin. Our data show that reduced DNAme in sperm renders paternal alleles permissive for H3K4me3 establishment in early embryos, independently of possible paternal inheritance of sperm born H3K4me3. Together, this study provides first evidence that paternally inherited DNAme directs chromatin formation during early embryonic development. ### Competing Interest Statement The authors have declared no competing interest. The datasets produced in this study are available in the following database: Gene Expression Omnibus GSE229246 (). Publicly available datasets used in this study can be found at Gene Expression Omnibus database with the following accession IDs: GSE148150 ([47][4]), GSE56697 ([25][5]) and GSE73952 ([66][6]). [1]: #ref-1 [2]: #ref-2 [3]: #ref-3 [4]: #ref-47 [5]: #ref-25 [6]: #ref-66