It’s well established that there’s more to gene regulation than sequence alone. There’s also epigenetics, which fine-tunes transcription through chemical modification of DNA and histones. Messenger RNA itself, though, has largely been thought free of epigenetic modification.
Now a research team at the Weill Medical College of Cornell University has demonstrated that not only are mammalian mRNAs chemically modified, but that that modification is widespread, evolutionarily conserved, and likely plays a role in mRNA post-transcriptional regulation (1).
That observation, says Jaffrey, “links FTO to regulation of gene and protein expression through demethylation of protein-coding transcripts.”
The question was, just how abundant is m6A in mammalian mRNA? Finding out wouldn’t be trivial; m6A is invisible to most molecular techniques, even those used to study DNA methylation. So Jaffrey’s team, led by postdoctoral researcher Kate Meyer, developed a new method called MeRIP-Seq, a variant of chromatin immunoprecipitation that purifies m6A-containing transcripts with an antibody recognizing the modification, and then sequences what that antibody pulls down.
When they applied this approach to both mouse brain and human cultured cells, they found that m6A is a highly abundant modification in mammalian mRNA. They counted more than 7900 transcripts with the modification present in either one or both species, plus more than 300 non-coding RNAs.
The modification was not evenly distributed across the length of mRNA transcripts, however; it tended to be localized towards the 3´ end of genes, specifically near the stop codon and into the 3´ untranslated region, at sites that are relatively evolutionarily conserved. When they examined those sites more carefully, they identified an apparent loose consensus sequence, GAC, which could signal methylases to modify the transcripts at those locations.
This consensus, notes Meyer, is ubiquitous, suggesting there’s more to the process of m6A methylation than sequence recognition alone—that is, “that m6A is specifically occurring in distinct parts of transcripts for a reason,” she says.
“There are clearly additional layers of information than merely recognizing the consensus sequence, and how the cell recognizes which GAC should be methylated and which should not is clearly something we are working on.”
The team also uncovered evidence of a relationship between methylation and microRNA. Two-thirds of the mRNAs containing m6A in their 3´ UTRs were also candidate genes for miRNA regulation. But the two sites did not typically overlap; instead, the methylated site was frequently located 5´ of the presumptive miRNA-binding site.
The “absolute biggest question,” says Jaffrey, is one of function: What does methylation actually do to modified transcripts?
The modification appears to be dynamic, being abundant, for instance, in adult neurons but present at lower levels during embryogenesis. As methylation likely influences the recruitment of RNA-binding proteins that mediate stability, localization, or translational efficiency, shifting m6A patterns could have a profound effect on gene expression as cellular conditions change.
Weill colleague and coauthor Chris Mason calls this new regulatory paradigm the “epitranscriptome,” and it, he says, will be transformative. m6A methylation, “opens up an entirely new realm of regulation in biology,” he says.
1. K.D. Meyer et al., “Comprehensive analysis of mRNA methylation reveals enrichment in 3’ UTRs and near stop codons,” Cell, 150:1-12, July 6, 2012.