DNA methylation differences between female and male X chromosomes in human brain
TL;DR
The results provide chromosome-wide evidence that, in the examined human brain cell types, the combined female X chromosome methylation level is globally lower than that of the male X chromosome, and support a model where relative hypomethylation is a general feature of transcriptionally inactive genomic regions compared to active genomic regions.
Abstract
DNA methylation is a key epigenetic mechanism of sex chromosome regulation. However, DNA methylation studies often exclude X chromosomes. Moreover, widely used DNA methylation assays interrogate biased subsets of CpGs, rather than assessing CpGs across the whole chromosome. Consequently, our understanding of X chromosome DNA methylation lags behind that of autosomes. Here we addressed this gap of knowledge by investigating X chromosome DNA methylation in 89 whole genome methylomes from neurons and oligodendrocytes. Using this unbiased and comprehensive data, we show that the combined methylation of the female X chromosomes is globally reduced (hypomethylated) across the chromosome compared with the male X and autosomes. DNA methylation patterns of promoters are more heterogeneous; while many promoters are more highly methylated (hypermethylated) in female X than in male X, a substantial number of promoters exhibit female hypomethylation. Notably, promoters of genes with higher expression in females—used here as a proxy for those that escape X chromosome inactivation—are significantly enriched among female hypomethylated promoters. Differentially methylated regions across the whole X chromosome corroborate the pervasive chromosome-wide reduction in the combined female X methylation and reveal many cell-type and sex-differentiated regions. Mechanistically, we show that female hyper- and hypo-methylation of promoters in large part associate with low and high density CpG promoters that tend to be highly and lowly methylated, respectively. These results provide chromosome-wide evidence that, in the examined human brain cell types, the combined female X chromosome methylation level is globally lower than that of the male X chromosome. More broadly, our findings support a model where relative hypomethylation is a general feature of transcriptionally inactive genomic regions compared to active genomic regions. Furthermore, we show that DNA methylation-mediated silencing of X chromosomes may preferentially occur in a subset of promoters with distinctive genomic and epigenetic features. This study provides a comprehensive and novel list of sex differentiated genomic regions on the X chromosome and furthers our understanding of epigenetic regulation of X chromosomes.