THE RNA CODE COMES INTO FOCUS As Researchers Open up to the Reality of RNA Modification, an Expanded Epitranscriptomics Toolbox Takes Shape
Total Page：16
File Type：pdf, Size：1020Kb
TECHNOLOGY FEATURE THE RNA CODE COMES INTO FOCUS As researchers open up to the reality of RNA modification, an expanded epitranscriptomics toolbox takes shape. LAGUNA DESIGN/SPL LAGUNA A molecular model of a bacterial ribosome bound to messenger RNA, a complex that is formed during protein synthesis. BY KELLY RAE CHI response in check. “It sounds simple, but in mRNAs harbour chemical tags — decorations real life it was really complicated,” Rechavi to the A, C, G and U nucleotides — that are n 2004, oncologist Gideon Rechavi at recalls. “Several groups had tried it before invisible to today’s sequencing technologies. Tel Aviv University in Israel and his col- and failed” because sequencing mistakes (Similar chemical tags, called epigenetic mark- leagues compared all the human genomic and single-nucleotide mutations had made ers, are also found on DNA.) Researchers aren’t IDNA sequences then available with their cor- the data noisy. But using a new bioinformat- sure what these chemical changes in RNA do, responding messenger RNAs — the molecules ics approach, his team uncovered thousands but they’re trying to find out. that carry the information needed to make a of sites in the transcriptome — the complete A wave of studies over the past five years — protein from a gene. They were looking for set of mRNAs found in an organism or cell many of which focus on a specific RNA mark signs that one of the nucleotide building population — and later studies upped the called N6-methyladenosine (m6A) — have blocks in the RNA sequence, called adenosine number into the millions1. mapped these alterations across transcrip- (A), had changed to another building block Inosine is something of a special case: tomes and demonstrated their importance called inosine (I). This ‘A-to-I editing’ can alter researchers can readily detect this chink in to health and disease. But the problem is vast: a protein’s coding sequence, and, in humans, the armour by comparing DNA and RNA these marks coat not only mRNA but other is crucial for keeping the innate immune sequences. But at least one-quarter of our RNA transcripts as well, and they cut across ©2017 Mac millan Publishers Li mited, part of Spri nger Nature. Al23l ri g FEBRUARYhts reserved. 2017 | VOL 542 | NATURE | 503 TECHNOLOGY EPITRANSCRIPTOMICS all the domains of life and beyond, marking That said, epitranscriptomics researchers from Active Motif (go.nature.com/2kqgzu8), even viruses with their presence. are excited about the direction their field is MilliporeSigma (go.nature.com/2kw39m3) The modifications themselves are not new. taking. “Just as you wouldn’t think of DNA and New England BioLabs (go.nature. What has given them meaning and driven without thinking about how DNA is packaged, com/2kqjjaz), among others. Researchers epitranscriptomics into the spotlight is the or epigenetically modified,” says geneticist think that the modification could control the discovery of enzymes that can add, remove Chris Mason at Weill Cornell Medical College way cells develop into different types, a pro- and interpret them. In 2010, chemical biolo- in New York City, who has led m6A-mapping cess that goes awry in cancer. Indeed, the first gist Chuan He at the University of Chicago, efforts, “I think now and in the future, no links between the epitranscriptome and cancer Illinois, proposed that these chemical tags one will think of RNA without thinking ‘How have already emerged. He’s group, for example, could be reversible and important regulators is it modified?’” showed that in some forms of acute myeloid of gene expression. Not long afterwards, his leukaemia, FTO is present in higher-than- group demonstrated2 the first eraser of these MAPPING WITH ANTIBODIES normal levels and seems to remove m6A from marks on mRNA, an enzyme called FTO. That In the early 1970s, scientists first showed that certain transcripts6, which could spur cells to discovery meant that m6A wasn’t just a passive mRNA was chemically modified by using differentiate. mark — cells actively controlled it. And this radioisotope labelling of m6A. But because A parallel line of research has turned that realization came at about the same time that those studies enriched the mRNA transcripts finding on its head. Using an antibody- global approaches, harnessing the power of by selecting their 3ʹ ends, which contain strings mapping method called miCLIP, which is next-generation sequencing, made it possible of adenosines, researchers worried that those higher in resolution than its predecessors, to map m6A and other modifications across preparations might contain trace amounts of Jaffrey’s team showed that its m6A antibod- the transcriptome. other classes of RNA molecules, as well. “Peo- ies also bind to N6, 2ʹ-O-dimethyladenosine Today, epitranscriptomics is blossoming. Yet ple stopped working on this because it was so (m6Am), a modifica- its toolbox remains a work in progress. Cur- difficult to get clear insights into whether the “There’s an tion of the chemical 6 rent methods lack the sensitivity required for m A in mRNA was a contaminant,” says Samie urgent demand structures that cap use with rare and precious samples. It’s also Jaffrey, a chemical biologist at Weill Cornell for additional the 5ʹ end of mRNAs. not possible to quantify the amount of a given Medical College. technology At the time, Jaffrey modification in the transcriptome, nor to map Also difficult was working out where in the didn’t know if m6Am 6 developments more than one modification in a single experi- transcriptome m A was located, which could for all kinds carried any biologi- ment. “There’s an urgent and high demand for provide clues to its function. Conventional of RNA cal meaning. But his additional technology developments for all sequencing approaches involve reverse tran- team has since shown modifications.” 6 kinds of RNA modifications,” says molecular scription — converting RNA into complemen- that m Am (and not biologist Tao Pan at the University of Chicago, tary DNA (cDNA), which is then amplified m6A) is in fact the who collaborated on He’s FTO studies. and sequenced. The problem is that the reverse major target of the FTO eraser, and that it transcriptase enzyme used to make cDNA affects the stability and subcellular location erases the modifications. “There was no way of mRNAs7. To Jaffrey, that suggests that He’s to see m6A,” Jaffrey says. “When you reverse- findings linking FTO to acute myeloid leukae- REF. 18/ACS REF. transcribe it, it behaves exactly like an A.” mia mean that m6Am, not m6A, is now impli- Despite the technical challenges, the discov- cated in the origin and development of cancer. ery of unexpected bacterial RNA modifi- Such discrepancies are par for the course in cations3 piqued Jaffrey’s interest, and he an emerging field, says cancer biologist How- decided to look for them in mamma- ard Chang at Stanford University, California. lian RNA. Working with Mason, “This particular issue is not that different from his team sheared RNA into tiny the early days of the histone-modification pieces, pulled out those that field,” he says, referring to the study of chemi- contained m6A using antibod- cal alterations to the histone ‘spools’ around ies, and sequenced the RNAs4. which DNA is tightly packaged in cells. “We were clearly seeing labelling of mRNAs and that was remark- MANY MODIFICATIONS able. It was not a contaminant,” Other RNA modifications have also attracted Jaffrey says. A similar study5 by researchers’ attention. In 2016, teams led by Rechavi’s group unearthed a hilly chemist Chengqi Yi at Peking University in landscape of m6A peaks, roughly Beijing and by Rechavi and He used antibody- 12,000 sites in 7,000 human genes. The based methods to map N1-methyladenosine modifications, Rechavi’s team dis- (m1A) in mouse and human cell lines and tis- covered, tended to be concentrated sues8,9. Using different approaches to prevent on the protein-coding sequences m1A from interfering with reverse transcrip- called exons and on stop codons, tion, the two teams showed that m1A, which the three-letter codes in mRNA was discovered in total RNA in the early 1960s, that signify the end of the pro- is present on mRNA at the position at which tein-coding sequence. the translation machinery initiates protein The methods these research- production. Stress conditions alter the maps, ers used, m6A-seq and MeRIP- suggesting that they are dynamic. seq, have since been broadly The researchers don’t yet know what m1A used to map m6A in different does, but they have a tantalizing clue: most A methylated RNA duplex (adenine disease contexts and organ- transcripts have only one m1A site, and N6 methyl groups in major groove isms. Antibodies and rea- these seem to be translated more often than highlighted in light blue). gents targeting m6A are available those that lack the modification. “This is 504 | NATURE | VOL 542 | 23 FEBRUARY©2017 M 2017ac millan Publishers Li mited, part of Spri nger Nature. All ri ghts reserved. EPITRANSCRIPTOMICS TECHNOLOGY very exciting — and of course challenging — modification sites, Yi says. Even the length of HHMI because we are dealing with a new regulatory time for which cells are kept alive in culture mechanism for translation of messenger RNA,” could influence modification levels, so it is cru- Rechavi says. An antibody that targets m1A is cial (and not necessarily trivial) to capture base- available from MBL International (go.nature. line maps for comparison, Mason says. com/2kvqpfs). But in any event, it’s not enough to show that Other global mapping strategies rely on the a sample contains a particular modification. fact that some RNA modifications can serve Instead, it will be crucial to quantify all RNA as handles for attaching chemical tags.