NEW YORK (GenomeWeb News) – A new study of mouse post-natal brain stem cells appearing online today in Science suggests DNA methylation can suppress or enhance the expression of genes during mouse brain development, depending on where these epigenetic marks occur.
A University of California at Los Angeles-led research team mapped binding sites for the methyltransferase enzyme Dnmt3a across the genome of mouse post-natal neural stem cells and looked at how mutating the enzyme affected gene expression in — and development of — these cells. Their findings suggest Dnmt3a-related DNA methylation not only keeps some genes silent, but may also act outside of promoter regions to spur the activity of certain neuronal genes by silencing a specific repressive protein complex.
"Our study provides the first big picture of how these enzymes may work in mammalian cells," lead author Hao Wu, a post-doctoral researcher in molecular and medical pharmacology researcher Yi Sun's UCLA lab, told GenomeWeb Daily News, explaining that the team was able to selectively alter Dnmt3a's activity to uncover its potential regulatory roles.
Past research suggests that although mutant mice missing Dnmt3a seem normal at birth, they typically show abnormal early development, the researchers explained. To date, though, Dnmt3a has mainly been studied from a biochemical or genetic point of view, Wu explained, rather than a genomic perspective.
As such, the team decided to explore how the cytosine-methylating enzyme might influence post-natal development — specifically development within the mouse brain.
Indeed, Wu noted, they found that they could not efficiently generate neurons from post-natal mouse neural stem cells lacking Dnmt3a.
When the researchers used chromatin immunoprecipitation coupled with whole-genome tiling arrays to map the methyltransferase Dnmt3a's occupancy at sites across the genome, they detected binding sites within or between genes near promoters with high levels of cytosine nucleotides neighboring guanine (CpG islands) as well as at low CpG promoters.
"In general, if the methylation is very close to the transcriptional start sites of annotated protein-coding or non-coding genes, it works as predicted: it generally represses transcription," Wu explained.
But Dnmt3a's binding and related methylation patterns also seemed to vary depending on genomic context and gene function. For example, the researchers found some non-promoter methylation corresponding to genes implicated in brain and nervous system development.
When they compared gene expression profiles in normal mouse neural stem cells with expression in mouse neural stem cells lacking Dnmt3a, the team found increased expression of 1,253 genes in cells missing Dnmt3a. Among them: genes involved in non-neuronal differentiation.
On the other hand, another 1,022 genes — including several neurogenesis genes — were down regulated in the absence of the enzyme.
Based on these and other findings, the researchers speculated that Dnmt3a activity outside of promoters might curb Polycomb complexes that would otherwise repress some genes — an idea that's consistent with the observation that Polycomb-targeted genes often have CpG rich promoter sequences, Wu noted. "These genes, potentially, could be regulated by this unexpected mode of regulation from non-promoter methylation."
The researchers' subsequent ChIP-chip experiments, focusing on Polycomb complex components themselves, further supported this notion.
"Together, these results suggest that the DNA methyltransferase Dnmt3a not only mediates repression in self-renewing post-natal [neural stem cells] by methylating proximal promoters, but also promotes transcription of targets, including neurogenic genes, by antagonizing Polycomb repression through non-proximal promoter methylation," the researchers explained.
The team is continuing to do studies looking at the potential interplay between DNA methyltransferase activity, methylation, and Polycomb repression, Wu said, including studies of mouse embryonic stem cells and various mouse adult tissues. While they have not yet examined human cells, he said such studies are likely coming.
"Given that both DNA methylation and Polycomb pathways are indispensable for normal development and are implicated in diseases, including neurological disorders and cancer, it will be of interest to fully elucidate mechanisms by which these two epigenetic machineries are targeted to specific genomic loci and are cross-regulated," the researchers concluded.