StableDNMT3LOverexpression in SH-SY5Y Neurons Recreates a Facet of the Genome-Wide Down Syndrome DNA Methylation Signature

Abstract

Down syndrome (DS) is characterized by a genome-wide profile of differential DNA methylation that is skewed towards hypermethylation in most tissues, including brain. The molecular mechanisms involve the overexpression of genes related to DNA methylation on chromosome 21. Here, we stably overexpressed the chromosome 21 gene DNA methyltransferase 3L (DNMT3L) in the human SH-SY5Y neuroblastoma cell line and assayed DNA methylation at over 26 million CpGs by whole genome bisulfite sequencing at three different developmental phases (undifferentiated, differentiating, and differentiated). DNMT3L overexpression resulted in global CpG and CpG island hypermethylation as well as thousands of differentially methylated regions (DMRs). The DNMT3L DMRs were skewed towards hypermethylation and mapped to genes involved in neurodevelopment, cellular signaling, and gene regulation. Merging the DMRs into a consensus profile where the cell lines clustered by genotype and then phase demonstrated that different regions of common genes are affected. The hypermethylated DMRs from all pairwise comparisons were enriched for regions of bivalent chromatin marked by H3K4me3 as well as differentially methylated CpGs from previous DS studies of diverse tissues. In contrast, the hypomethylated DMRs from all pairwise comparisons displayed a tissue-specific profile enriched for regions of heterochromatin marked by H3K9me3 during embryonic development. Taken together, we propose a mechanism whereby regions of bivalent chromatin that lose H3K4me3 during development are targeted by excess DNMT3L and become hypermethylated, while excess DNMT3L also evicts DNMT3A from heterochromatin, resulting in hypomethylation. Overall, these findings demonstrate that DNMT3L overexpression during neurodevelopment recreates a facet of the DS DNA methylation signature.

Long-term effects of wildfire smoke exposure during early-life on the nasal epigenome in rhesus macaques

“…The p-values from DMRichR were empirical p-values calculated from permutation tests (64,65). Although this puts our study at a higher risk of false-positive findings, these permutation p-values calculated by DMRichR were used to determine DMR significance in multiple published studies in combination with effect size (64)(65)(66)(67). Given that our analysis of chromatin states relied on human hg38 annotations, we compared our macaque rheMac10 annotations to the hg38 annotations to make sure they were similar enough.…”