METTL1 Promotes let-7 MicroRNA Processing via m7G Methylation

摘要

class="head no_bottom_margin" id="sec1title">IntroductionPost-synthesis covalent modification of biological molecules is a key aspect of intracellular signaling, and it is critically important in many biological processes. RNA molecules, similar to proteins, are subject to a vast array of post-synthesis covalent modifications, which together constitute the epitranscriptome. To date, >100 RNA modifications have been identified, which are spread throughout every class of RNA and are evolutionarily conserved throughout all kingdoms of life (, ).RNA modifications have the potential to affect all RNA processes, including splicing, stability, and localization (). Many RNA modifications have been identified by mass spectrometry (MS), and complex epitranscriptomes of tRNA and rRNA have been thoroughly studied. However, this represents a mere snapshot of a much bigger picture, with the clear majority of modifications remaining uncharacterized. This is predominantly due to a lack of sensitive methodologies with which to detect the modifications at a high resolution. Even now, MS methodologies are largely unable to generate transcriptome-wide modification profiles. However, a few very recent analyses have used anti-modification antibodies (e.g., against N1-methyladenosine [], N6-methyladenosine [], and 5-hydroxymethylcytosine []) or chemical reactivity of the modification (for pseudouridine [, ], m5C [], and 2′-O-methylation []). Their results clearly suggest that many of the modifications identified on rRNA and tRNA are also present on other RNA classes. Therefore, the development of epitranscriptomic methodologies (e.g., new antibody and chemical methods coupled to next-generation sequencing [NGS]) represents a bottleneck in deciphering the function of new RNA modifications.Certain nucleotides, such as 7-methylguanosine (m7G), display specific modification-dependent chemistries that can be exploited to study their prevalence and transcript location. m7G is present in eukaryotic mRNA 5′ caps and at defined internal positions within tRNAs and rRNAs across all domains of life. The best-characterized enzyme mediating internal m7G methylation is the TRMT8 yeast enzyme homolog METTL1 (methyltransferase-like 1), which, together with its co-factor WDR4 (WD repeat domain 4), catalyzes m7G at G46 of specific tRNAs, such as tRNA^Phe ().In contrast to deoxy-m7G, m7G in RNA is highly stable in neutral aqueous solution (). The methylation significantly alters the charge density of RNA, potentially serving as a molecular handle, but it does not impair Watson-Crick G:C base complementarity. It does, however, interfere with non-canonical base pairing (i.e., Hoogsteen pairs), possibly affecting the secondary structure of RNA. Although relatively abundant, m7G has proved very difficult to study so far. Being neutral to Watson-Crick base pairing, it does not interfere with reverse transcription, rendering it invisible to detection by standard sequencing-based technologies.microRNAs (miRNAs) are short single-stranded RNA molecules (18–24 nucleotides [nt]) that target the RNA interference silencing complex (RISC) to specific mRNAs. Their specificity is mediated by partial base pairing to sequences predominantly found in the 3′ UTR of mRNAs (). This interaction results in the decreased translation of the proteins they encode and/or in the degradation of the mRNAs themselves (, ). To date, >1,000 human miRNAs have been identified, and they are key regulators of numerous physiological and pathological processes.miRNA biosynthesis is complex and involves a multistep pathway that can be regulated at many levels (href="#bib6" rid="bib6" class=" bibr popnode">Bartel, 2018), including post-transcriptional modification of miRNA precursors (href="#bib1" rid="bib1" class=" bibr popnode">Alarcón et al., 2015, href="#bib63" rid="bib63" class=" bibr popnode">Xhemalce et al., 2012). miRNAs are synthesized from larger transcripts by RNA polymerase II or III. These primary miRNA transcripts (pri-miRNAs) are then cleaved by DROSHA to release hairpin-shaped RNAs called pre-miRNAs (href="#bib37" rid="bib37" class=" bibr popnode">Lee et al., 2003), and further cleaved by DICER to generate a miRNA duplex (href="#bib14" rid="bib14" class=" bibr popnode">Chendrimada et al., 2005). Certain miRNAs can form alternative secondary structures, such as G-quadruplexes, that can interfere with their processing (href="#bib48" rid="bib48" class=" bibr popnode">Mirihana Arachchilage et al., 2015, href="#bib50" rid="bib50" class=" bibr popnode">Pandey et al., 2015). However, little is known about the biological relevance of these structures in a physiological context.Here, we develop two different but complementary high-throughput sequencing strategies to identify miRNAs harboring internal m7G modification. We show that METTL1 methylates a specific subset of tumor suppressor miRNAs, including let-7, to promote their processing from primary transcript to precursor miRNA. Depletion of METTL1 causes gene expression and phenotypic changes in a miRNA-dependent manner. We show that m7G-modified miRNAs have a propensity to form G-quadruplexes. We identify guanosine 11 as the m7G methylated residue within let-7e-5p, and we show that methylation at this position affects G-quadruplex formation, thereby promoting processing of the precursor miRNA.