There’s no denying that understanding a new language can be tricky. Without a complete alphabet, the most important messages can be lost in translation. When it comes to complex languages, the epigenome is no exception. There are a number of RNA base modifications involved in post transcriptional regulation, including pseudouridine (Ψ) and 5hmC, and there’s no Rosetta Stone to interpret them.
Adenine Methylation
Unlike the cytosine modifications favored on DNA, N6-methyladenosine (m6A) is an RNA modification linked to the RNA demethylase FTO. Recently, m6A, also known as 6mA, was found to be a conserved epigenetic DNA modification. Interestingly, as an epigenetic modification of RNA and DNA, m6A appears to take on a unique array of functions related to dynamic gene regulation and development. In RNA, m6A is enriched for at the stop codon and 5’UTR. Among several other functions, m6A plays a role in stem cell differentiation.
Like m6A, N1-methyladenosine (m1A) was discovered decades ago, but its epitranscriptomic implications are just being discovered. Now, the m6A-pioneering lab of Chuan He at the University of Chicago is investigating this second form of adenine methylation. The team carried out a number of clever experiments to examine both m1A abundance and regulatory potential.
m1A Abundance
Quantitative mass spectrometry (LC-MS/MS) revealed that m1A is present at lower levels than other modifications, but it is still present at a high enough level to have regulatory potential.
The team then utilized methylated RNA immunoprecipitation sequencing (MeRIP-seq) in human cells lines and found that:
- Under stringent conditions, m1A is present in 4,151 coding and 63 non-coding expressed transcripts.
- Most of the modified transcripts are methylated only once and are typically highly expressed.
- ncRNAs are surprisingly under-enriched for m1A.
- mRNAs with m1A have functions related to translation and RNA processing.
The team also used an immunodepletion and microarray approach to determine transcript abundance in humans, which revealed that m1A is present in ~20% of transcripts that contain a single m1A modification.
m1A Function
The team also used MeRIP-Seq to get an idea of what m1A was up to and found that:
- m1A is enriched at translation initiation sites as well as initial splice sites in the human genome.
- m1A is found in both canonical and alternative translation initiation sites, and the amount of m1A per gene is related to the number of alternative translation initiation sites.
- m1A typically occurs around the start codon, in GC rich sequence, upstream of the first splice site, and can be in exons. m1A is also found in genes with structured 5’UTRs.
- m1A patterns are evolutionarily conserved in mRNA and present in a number of human and mouse cell lines. m1A is also present in yeast mRNA, but not with the same pattern.
- m1A levels vary across tissues in humans and mice, with the brain and kidney showing the highest levels.
Using LC-MS/MS to analyze yeast, they found that stress conditions, such as glucose or amino acid starvation or heat shock, could alter m1A mRNA levels in human cells. Interestingly, m6A does not respond as robustly to these stress conditions.
At the same time as the above study, another group from Peking University in China analyzed m1A. This independent group also found that m1A is prevalent in humans, with ~600 coding and non-coding genes enriched for m1A at their 5’ UTRs. The team also found that m1A is involved in a dynamic response to heat stress and identified ALKBH3, a previously known DNA/RNA demethylase, as being capable of removing m1A.
Overall, it appears that m1A has an important role in promoting translation across dynamic conditions.
The Future of m1A and The Epitranscriptome Code
Chuan He shares that, “mRNA is the perfect place to regulate gene expression, because [mRNA] can code information from transcription and directly impact translation; you can add a consensus sequence to a group of genes and use a modification of the sequence to readily control several hundred transcripts simultaneously. If you want to rapidly change the expression of several hundred or a thousand genes, this offers the best way.”
He concludes, “The discovery of m1A is extremely important, not only because of its own potential in affecting biological processes, but also because it validates the hypothesis that there is not just one functional modification. There could be multiple modifications at different sites where each may carry a distinct message to control the fate and function of mRNA.”
Christopher Mason opines that, “This study represents a breakthrough discovery in the exciting, nascent field of the epitranscriptome. What is important about this work is that m6A was recently found to enrich at the ends of genes, and now we know that m1A is what is helping regulate the beginning of genes, and this opens up many questions about revealing the ‘epitranscriptome code’ just like the histone code or the genetic code.”
Learn more about this part of the epitranscriptomic alphabet In Nature, February 2016