Chemical Modifications to RNA: A New Layer of Gene ... · Chemical Modifications to RNA: A New Layer of Gene Expression Regulation Jinghui Song† and Chengqi Yi*,†,‡ †State

Chemical Modifications to RNA: A New Layer of Gene ExpressionRegulationJinghui Song† and Chengqi Yi*,†,‡

†State Key Laboratory of Protein and Plant Gene Research, School of Life Sciences, and Peking-Tsinghua Center for Life Sciencesand ‡Department of Chemical Biology and Synthetic and Functional Biomolecules Center, College of Chemistry and MolecularEngineering, Peking University, Beijing 100871, China

ABSTRACT: The first chemical modification to RNA wasdiscovered nearly 60 years ago; to date, more than 100chemically distinct modifications have been identified in cellularRNA. With the recent development of novel chemical and/orbiochemical methods, dynamic modifications to RNA have beenidentified in the transcriptome, including N6-methyladenosine(m6A), inosine (I), 5-methylcytosine (m5C), pseudouridine (Ψ),5-hydroxymethylcytosine (hm5C), and N1-methyladenosine(m1A). Collectively, the multitude of RNA modifications aretermed epitranscriptome, leading to the emerging field ofepitranscriptomics. In this review, we primarily focus on recentlyreported chemical modifications to mRNA; we discuss theirchemical properties, biological functions, and mechanisms withan emphasis on their high-throughput detection methods. Wealso envision that future tools, particularly novel chemical biology methods, could further facilitate and enable studies in the fieldof epitranscriptomics.

Besides the four regular nucleosides (adenosine, guanosine,cytidine, and uridine), natural RNA molecules contain

over 100 chemically modified nucleosides in the threekingdoms of life (http://mods.rna.albany.edu/mods/). MostRNA modification studies have focused on the abundantnoncoding RNAs (ncRNA), such as rRNA, tRNA, and smallnuclear RNA (snRNA). These studies reveal that RNAmodifications are important for the functions of these ncRNAsin translation and splicing.1−3 Messenger RNA (mRNA) andlong noncoding RNA (lncRNA) are also decorated bymodifications, such as the 5′ cap and RNA editing. Ineukaryotes, the 5′ cap, which is a 7-methylguanosine linkedto the 5′ terminus of mRNA (and some lncRNA) via anunusual 5′ to 5′ triphosphate linkage, functions importantly inRNA stability and translation.4,5 RNA editing can change thecoding message in mRNA and diversify protein synthesis.6,7

Benefiting from recently developed high-throughput detec-tion strategies, diverse chemical modifications have beenidentified in mRNA, including N6-methyladenosine (m6A),inosine (I), 5-methylcytosine (m5C), pseudouridine (Ψ), 5-hydroxymethylcytosine (hm5C), and N1-methyladenosine(m1A). The multitude of dynamic modifications in mRNAhave diverse biological functions. Understanding the distribu-tion, mechanisms, regulation, and function of these dynamicRNA modifications will greatly expand our knowledge ofepitranscriptomics. Because inosine has been reviewedextensively elsewhere,8,9 we will focus on five recently identifiedchemical modifications (m6A, m5C, hm5C, Ψ, and m1A) inmRNA. We will discuss their chemical properties, modification

enzymes, high-throughput detection methods, and biologicalfunctions.

■ CHEMICAL PROPERTIES AND BIOGENESIS OF RNAMODIFICATIONS

As the most prevalent internal modification in all highereukaryote mRNA and lncRNA, m6A (Figure 1) was discoveredin the 1970s and occurs on an average of ∼3 sites per mRNAmolecule.10 The methyl group of m6A can exist in both syn andanti conformations: although the syn conformation is moreenergetically favored than the anti conformation in single-stranded RNA (ssRNA),11,12 the methyl group in m6A mustadopt a high-energy anti orientation in paired RNA toaccommodate normal Waston−Crick base pairs.13 Thus, N6methylation in m6A could destabilize the base pairs of its localRNA regions.13,14 The formation of m6A is catalyzed by“writers”, which is a methyltransferase complex consisting ofmethyltransferase like 3 (METTL3), methyltransferase like 14(METTL14), and regulatory subunit Wilms’ tumor 1-associating protein (WTAP).15−19 In the complex, METTL3serves as the catalytically active subunit, METTL14 enhancessubstrate recognition and RNA binding, and WTAP mayfacilitate the translocation into nuclear speckles.19−21 Bio-chemical analyses have revealed that the consensus substrate

Received: October 31, 2016Accepted: January 4, 2017Published: January 4, 2017

Reviews

pubs.acs.org/acschemicalbiology

© 2017 American Chemical Society 316 DOI: 10.1021/acschembio.6b00960ACS Chem. Biol. 2017, 12, 316−325

sequence of the m6A methyltransferase is RRm6ACH (R= A, G;H = A, C, U).22 m6A can be demethylated by “erasers”, fat massand obesity-associated protein (FTO) and alkylation repairhomologue protein 5 (ALKBH5), which belong to the Fe(II)-dependent and 2-oxoglutarate-dependent oxygenase super-family.23,24 FTO oxidatively removes m6A modification throughintermediates N6-hydroxymethyladenosine (hm6A) and N6-formyladenosine (f6A).25 However, hm6A and f6A have notbeen detected in the demethylation process by ALKBH5.26 Thediscovery of the “erasers” sheds light on the reversible nature ofm6A, indicating its potential regulatory function.In addition to methylation at the N6 position, it can also

occur at the N1 position of adenosine, forming m1A (Figure 1).m1A was first discovered in the 1960s27,28 and is prevalent intRNA and rRNA where it plays a role in maintaining tRNAtertiary structure, ribosome biogenesis, and translation.29−31

Recent studies report that m1A is also a dynamic and prevalentmodification in mRNA.32,33 Notably, the N1 methyl group inm1A can introduce a positive charge under physiologicalconditions; thus, it influences the strength of hydrogen bondsand facilitates the interaction with negatively chargedphosphates in the backbone.34−36 Furthermore, because ofthe methyl group in the N1 position of m1A at the Watson−Crick interface, m1A can interfere with normal base pairing.37

In 28S rRNAs of human and mouse cells, m1A is modified bynucleomethylin (NML; also known as RRP8);38 in humantRNA, m1A formation is catalyzed by TRMT61B, TRMT10C,and the complex of TRMT6 and TRMT61A.39−41 Veryrecently, m1A in tRNA was reported to be reversible byALKBH1 in mammalian cells;42 m1A in mRNA could also betargeted by ALKBH3.32,33

m5C is another methylation in mRNA (Figure 1). Previously,it has been reported that m5C exists in highly abundant tRNAand rRNA in both prokaryote and eukaryote organisms,43

where it stabilizes the RNA secondary structure and affectstranslational fidelity.34,44−46 Through recent high-throughputstrategies, m5C has been found in mRNA and otherncRNA.47−50 The formation of m5C is catalyzed by themethyltransferases DNMT2 and NSUN2 in human cells.47−49

Although m5C has not been reported to be reversible in RNA,it can be further oxidized to hm5C (Figure 1) by ten-eleventranslocation (TET) family enzymes,51−53 suggesting that m5Cmay be regulated. The discovery of m5C and hm5C in mRNAfurther expand the alphabet of RNA epigenetics and makeepitranscriptome more intricate and diverse.Modifications in RNA are diverse, and the chemical

functionalities can certainly go beyond methylation. Knownas the “fifth nucleotide” of RNA, Ψ (Figure 1) is a C−Cglycosidic rotation isomer of uridine, which was first found in1951 in RNA.54,55 Although Ψ has the same molecular weightand base pairing pattern with regular U, an additional H-bonddonor in the N1H of Ψ can bind a water molecule to bridge theinteractions of this N1H and the preceding phosphate groupsand stabilize the RNA structure.56 Ψ is the most abundant RNAmodification, present in tRNA, rRNA, and snRNA; untilrecently, it had not been found in mRNA and lncRNA.57−61 Ψplays important roles in folding and translation fidelity ofrRNA, stability and decoding of tRNA, function of snRNA, andsplicing and translation of mRNA.3 There are two pathways forthe formation of Ψ in eukaryotes: one is catalyzed by the RNA-dependent pseudouridine synthases (PUSs) that require thecofactor box H/ACA ribonucleoproteins as guides to recognizedifferent substrates; another is catalyzed by the RNA-

Figure 1. Chemical structures of m6A, m1A, m5C, hm5C, and Ψ.

Figure 2. Dynamic RNA modifications in mRNA and lncRNA affect a wide range of cellular processes. In the nucleus, RNA modifications affect pre-mRNA and pri-miRNA processing and nuclear export. In the cytoplasm, RNA modifications regulate mRNA translation and stability. Red and blacklines represent RNA and DNA, respectively. RNA modifications are indicated by blue filled circles.

ACS Chemical Biology Reviews

DOI: 10.1021/acschembio.6b00960ACS Chem. Biol. 2017, 12, 316−325

317

independent PUSs that require no cofactor.57,62,63 Ten PUSswere identified in yeast: Cbf5 is an RNA-dependent PUS, andthe rest (Pus1−9) are RNA independent.64 Thirteen PUSswere identified in human: only DKC1 (a homologue of Cbf5)is an RNA-dependent PUS.57 The study of Ψ is complicated bymultiple PUSs and potential functional redundancy. The inertC−C glycosidic bond of Ψ makes it difficult to be reversible.Interestingly, Ψ can be further methylated to 1-methylpseu-douridine (m1Ψ) by EMG1,65 potentially offering a mechanismto reduce the level of Ψ.The abundance of m6A, m1A, m5C, hm5C, and Ψ in mRNA

are quite different as quantified by the same method (LC-MS/MS). As the most abundant internal mRNA modification ineukaryotes, the m6A/A ratio of human mRNA is approximately0.4−0.7%.16,23,24 The ratio of Ψ/U in mammalian mRNA isapproximately 0.2−0.6%, which is comparable in content tom6A.61 Thus, m6A and Ψ are two widespread chemicalmodifications in mRNA. In human mRNA, the m5C/C ratiois approximately 0.02−0.09%, and the hm5C/C ratio isapproximateyl 0.001−0.003%, only 1/25 of m5C.52 The m1A/A ratio in mammalian mRNA is approximately 0.02−0.06%,32,33 approximately 1/10 of m6A. Among the fivemodifications, m6A and Ψ are modifications with highabundance in mRNA, and hm5C is the one with the lowestabundance.

■ BIOLOGICAL FUNCTIONSWith the development of high-throughput technologies of m6A,more and more functions of m6A have been discovered in awide range of cellular processes. For instance, right aftertranscription, processing of pre-mRNA and primary micro-RNAs (pri-miRNAs) are critical steps to generate maturemRNA and miRNA. Many investigations have provided strongevidence for the regulatory role of m6A in pre-mRNA and pri-miRNA processing (Figure 2). For pri-miRNA processing, m6Ais present in pri-miRNA and promotes miRNA biogenesis inwhich HNRNPA2B1 is involved.66,67 For pre-mRNA process-ing, m6A can alter its RNA local structure to regulate theinteractions of RNA and HNRNPC, which can further affectthe alternative splicing of m6A-containing mRNAs mediated byHNRNPC.68 YTHDC1 is a nuclear reader of m6A that candirectly regulate inclusion of targeted exons through recruitingSRSF3 while blocking the binding of SRSF10 to mRNA.69 Theincreased m6A level could also promote the RNA bindingability of SRSF2, leading to increased inclusion of targetexons.70 Besides splicing, m6A may affect alternative poly-adenylation (APA) as well.71,72 After pre-mRNA processing inthe nucleus, it is crucial that mature mRNAs are exported to thecytoplasm. Researchers have found that nuclear mRNA exportis accelerated upon ALKBH5 knockdown, indicating that m6Amight be involved in mRNA export (Figure 2).24

After export out of nucleus, mature mRNA can be eithertranslated or degraded in the cytoplasm. mRNA turnover is animportant and fast process that regulates mRNA abundance,allowing for rapid cellular adjustment in various environments.Researchers found that adjacent m6A methylation might blockthe binding of the human antigen R (HuR) to destabilizemRNA.17 Moreover, YTH domain family 2 (YTHDF2, the firstcomprehensively established m6A reader) selectively binds tom6A-decorated mRNA to promote the decay of thousands ofmRNAs via the translocation of bound mRNA from thetranslatable pool to processing bodies.73 These results revealedthe tight regulation between m6A modification and mRNA

decay (Figure 2). On the other hand, the m6A mark can alsoaffect the translation efficiency of mRNA. It has been reportedthat the translation efficiency was increased modestly uponMETTL3 knockout in mouse embryonic stem cells (mESCs)and embryoid bodies (EBs), suggesting a negative regulatoryrole of m6A in translation.74 However, when m6A is recognizedby another cytoplasmic m6A reader YTHDF1 (which interactswith initiation factors and ribosomes), these m6A-markedtranscripts are preferentially translated.75 More recently, it hasbeen reported that m6A in the 5′ UTR of mRNA can promotecap-independent translation initiation.76,77 The researchersfurther found that eukaryotic initiation factor 3 (eIF3) directlybinds m6A-containing 5′ UTR and recruits the 43S complex toinitiate translation in the absence of the cap-binding factoreIF4E.76 These data provide direct evidence for a function ofm6A in translational regulation (Figure 2), although METTL3can also directly promote translation independently of itscatalytic activity and m6A reader proteins.78 In addition, theregulatory role of m6A in gene expression has been reported toaffect ESC pluripotency in human and mouse.17,74,79−82 Veryrecently, it was also found that m6A is required for XIST-mediated transcriptional repression of genes on the Xchromosome.83

Besides the functions in host cells, m6A was also found inviral RNA genomes. Several groups independently reported thefunction of m6A during virus infection. It has been reportedthat m6A in both host and viral mRNAs were increased duringHIV-1 infection of human CD4 T cells.84 The researchersfurther found that m6A in the stem loop II region of HIV-1 Revresponse element (RRE) RNA modulates the interactionbetween Rev protein and RRE RNA and influences the nuclearexport of RNA. Moreover, one study found that binding ofYTHDF1−3 proteins to m6A-methylated HIV-1 RNApromotes viral protein and RNA expression,85 whereas anotherrecent study found that YTHDF1−3 proteins bind m6A-modified HIV-1 RNA and inhibit viral reverse transcription(RT).86 Very recently, m6A has also been found in the RNAgenomes of flavivirus Zika (ZIKV), hepatitis C virus (HCV),and other Flaviviridae family members and modulates infectionby the Flaviviridae virus.87,88 These data suggested theimportant role of m6A in viral infection. Given that cells candiscriminate self and nonself by RNA modifications,89−92 RNAmodifications may globally participate in immune response.In addition to m6A, Ψ in mRNA might also affect mRNA

splicing, stability, and translation (Figure 2). First, thereplacement of U with Ψ in the polypyrimidine tract (nearthe 3′ splice site) of artificial pre-mRNA inhibits pre-mRNAsplicing by affecting U2 auxiliary factor (U2AF) binding inXenopus oocytes.93 However, whether or not Ψ is present inpre-mRNA remains to be investigated. Second, one study inyeast has demonstrated heat shock-induced Ψ sites. Interest-ingly, the expression level of the heat shock induced, Pus7-dependent pseudouridylated transcripts are higher in WT yeastthan in Pus7-deletion yeast, suggesting that Ψ could contributeto RNA stability.58 Third, pseudouridylation appears to impactmRNA translation in a complicated manner. When Ψ-containing mRNA was translated in rabbit reticulocyte lysates,the presence of Ψ stimulates translation. However, in wheatgerm extracts and bacterial cell lysates, Ψ represses trans-lation.94 When Ψ was incorporated in different positions andcodons of mRNA, different effects on translation were alsoobserved. One study has reported that the incorporation of asingle Ψ at each position of the codon “UUU” of bacterial

ACS Chemical Biology Reviews

DOI: 10.1021/acschembio.6b00960ACS Chem. Biol. 2017, 12, 316−325

318

mRNA represses translation in vitro with the strongestrepression at the third codon position.95 In a second study,the replacement of U with Ψ at stop codons suppressestranslation termination and converts nonsense codons intosense codons both in vitro and in vivo.96,97 However, all theexperiments regarding translation were performed usingartificial mRNA carrying Ψ. Whether natural Ψ in mRNAcould also affect translation in vivo is worth furtherinvestigation.In addition to m6A and Ψ, other internal mRNA

modifications might affect mRNA translation as well (Figure2).98 For instance, a single m5C in mRNA could reduce ∼40%of the amount translation product using bacterial whole cellextract, and m5C at the second nucleotide of the codon “CCC”could induce recoding.95 Additionally, uncapped mRNAs thatare randomly incorporated by 50% m5C significantly decreasetranslation in HeLa cell extracts.76 In contrast to m5C, whichnegatively impacts translation, hm5C is shown to associate withactive translation in Drosophila.53 Furthermore, m1A might alsocontribute to translation. It has been reported that the genescarrying m1A around the start codon correlate with higherprotein level, indicating the positive effect of m1A ontranslation.33 All these phenomena favor the regulatory roleof internal mRNA modifications on translation.

■ HIGH-THROUGHPUT DETECTION METHODS

The first transcriptome-wide m6A maps with resolution of100−200 nt were generated in 2012 by two independentgroups based on an m6A-specific antibody (m6A/MeRIP-seq).99,100 Because of the lack of chemical methods todifferentiate m6A from A, these approaches used an m6A-specific antibody to pull down m6A-containing RNA (but notnon-m6A RNA), which allows pre-enrichment of m6A-

containing RNA for high-throughput sequencing. Theseapproaches were quickly adopted by many other laboratoriesand are used frequently today. Such methods have identifiedthousands of m6A peaks in mammalian mRNA and lncRNA:the m6A peaks are enriched in the 3′ UTR and near stopcodons. Moreover, m6A is also a widespread modification inother species mRNA: in yeast, m6A is dynamically regulatedmethylation during meiosis;101 in plants, m6A peaks areabundant around the start codon (16%).102 To reach single-base-resolution m6A methylomes and quantify the m6Astoichiometry, several groups have independently reportedthe optimized strategies based on the m6A/MeRIP-seqprotocols. In m6A-CLIP and miCLIP strategies, 254 nm UVlight was used to induce covalent cross-links between antibodyand m6A-containing RNA.72,103 During RT, the antibody−RNAcross-linking sites can cause the mutational or truncationalsignatures of cDNA so that base-resolution m6A sites can beidentified. In PA-m6A-seq, photoreactive nucleoside analogue(4-thiouridine, 4SU), which can improve the cross-linkingefficiency, was added to the growth medium and thenincorporated into newly synthesized RNA molecules; afterimmunoprecipitation, covalent cross-links between antibodyand RNA were formed under UV irradiation at 365 nm(instead of UV 254 nm).104 Notably, the incorporated 4SU canlead to T to C transition at cross-linking sites; thus, it ispossible to identify m6A at base-resolution. Recently, a newmodified strategy, called m6A-LAIC-seq, was developed toquantify the m6A stoichiometry on a transcriptome-widescale.71 It revealed that most genes exhibit less than 50%m6A modification levels in human ESC. The development ofoptimized detection methods would help us to betterunderstand the intricate m6A methylome.

Figure 3. Dimroth rearrangement and demethylase can eliminate the RT signatures of m1A. (a) m1A in RNA causes stop or misincorporation duringRT, whereas the conversion of m1A to m6A or regular A can remove the RT signatures of m1A. (b) Dimroth rearrangement of m1A under alkalineconditions. (c) Demethylation of m1A by AlkB.

ACS Chemical Biology Reviews

DOI: 10.1021/acschembio.6b00960ACS Chem. Biol. 2017, 12, 316−325

319

Unlike m6A, which still pairs to thymidine during RT, m1Acan interfere with Watson−Crick base pairing. Recently, we andothers independently developed two approaches (called m1A-ID-seq and m1A-seq) to map the m1A methylomes.32,33 Bothapproaches are based on m1A-specific antibody, which permitsthe pre-enrichment of m1A-containing RNA. After immuno-precipitation, the m1A-containing RNA was subjected to anadditional treatment step to remove the RT signatures of m1A.Therefore, by comparing the sequencing profiles of treated anduntreated samples, high-confidence m1A peaks can beidentified. In particular, m1A-seq exploited Dimroth rearrange-ment to convert m1A to m6A under alkaline conditions becausem6A neither changes the base-pairing properties nor inhibitsRT;105 the conversion of m1A to regular A by demethylase wasused in m1A-ID-seq to eliminate the RT signatures of m1A(Figure 3). Such methods first report that m1A is a new,dynamic, and reversible epitranscriptomic mark in mammalianmRNA and lncRNA, which is enriched in the 5′ UTR.

The detection of m5C in RNA could draw lessons fromm5dC detection in DNA for which bisulfite treatment is widelyused. A modified bisulfite strategy, which converts unmodifiedcytosines (but not m5C) to uracils (Figure 4), was reported todevelop m5C sequencing in RNA.106 In 2012, the first profile ofm5C was obtained by coupling bisulfite conversion with high-throughput sequencing, indicating the presence of m5C inhuman mRNA and lncRNA.47 Several groups have also madeefforts to develop new approaches to identify m5C sites.48−50

Aza-IP exploited the formation of 5-azacytidine-methyltransfer-ase adducts to enrich and subsequently sequence m5C targets;this approach was applied to identify the direct targets ofNSUN2 and DNMT2 in the human tanscriptome. In miCLIPof m5C (different from miCLIP of m6A), the mutation(C271A) of NSUN2, which could form a stable covalentbond with RNA, was exploited to identify the transcriptome-wide targets of NSUN2. In the m5C-RIP approach, an antibodyagainst m5C was used to validate m5C sites in archaea mRNA,

Figure 4. Bisulfite-mediated conversion of cytosine to uracil. (a) Principle of RNA bisulfite sequencing. (b) Cytosine deamination via bisulfitetreatment.

Figure 5. CMC specifically labels Ψ. (a) Work flow of CMC labeling. (b) Chemical structure of CMC. (c) CMC acylation of U, Ψ, and G. (d) Afteralkaline treatment, only the adducts of CMC-Ψ remain.

ACS Chemical Biology Reviews

DOI: 10.1021/acschembio.6b00960ACS Chem. Biol. 2017, 12, 316−325

320

but m5C-RIP has not been applied to mammalian cells as yet.Such an antibody-based approach could in principle be appliedto other modifications as well. For instance, a method utilizinga hm5C-antibody (called hMeRIP-seq) was reported to identifymore than 3,000 hm5C peaks in the Drosophila transcriptomeby using an hm5C specific antibody.53 In Drosophila S2 cells,hm5C presents preferentially in polyadenylated RNA and is notdetected in rRNA or small RNA by dot blotting.The transcriptome-wide mapping of Ψ sites relies on the

chemical N-cyclohexyl-N′-β-(4-methylmorpholinium) ethylcar-bodiimide (CMC), which specifically labels Ψ (Figure5).58−61,107 Initially, CMC can acylate guanosine (G) at theN1 position, uridine (U) at the N3 position, and Ψ at the N1and N3 positions in RNA (Figure 5c); subsequently, theadducts of CMC-G and CMC-U are hydrolyzed under alkalinetreatment (pH 10.4), whereas CMC remains specifically at theN3 position of Ψ (Figure 5d).108 The formation of CMC-Ψadducts can cause stops at one nucleotide 3′ to the labeled Ψsites during RT, thereby achieving base-resolution detection ofΨ (Figure 5a). Three independent CMC-based profilingmethods of Ψ (called Ψ-Seq, Pseudo-Seq, and PSI-Seq) haveidentified approximately 50−100 Ψ sites in yeast mRNA and100−400 sites in human mRNA at single-base resolution.58−60

However, these methods cannot pre-enrich Ψ-containing RNAof interest. We developed a chemical pull-down method for Ψ,termed CeU-Seq, which allows pre-enrichment of Ψ-containingRNA as well as Ψ detection at single-base resolution.61 In CeU-Seq, CMC derivative azido-CMC (N3-CMC) was used to labelΨ, and the N3-CMC-Ψ adduct can be coupled with biotin via“click” chemistry.109 Then, the Ψ-containing RNA was pulleddown through biotin and subsequently sequenced. ThroughCeU-Seq, we identify over 2,000 Ψ sites in human mRNA and∼1,500−1,700 Ψ sites in mouse mRNA.

■ EXISTING CHALLENGES FOR DETECTION OF RNAMODIFICATIONS

From the discussions above, we could briefly divide sequencingtechnologies of RNA modifications into two categories: One isantibody-based, which allows the pre-enrichment to increasethe signal-to-noise ratio. The other is chemical labeling-based,which generally enables detection of RNA modifications atsingle-base resolution. Each category has certain advantagesover the other, and the two categories are not mutuallyexclusive. Major concerns for detecting RNA modifications arethe precision, sensitivity, quantitative ability, and resolution ofhigh-throughput detection methods, and many challengesremain to be solved.For m6A detection, single-base-resolution and quantitative

detection technologies of m6A have been developed to facilitatethe functional studies, but all these sequencing technologies areantibody-based. Indeed, it has been shown that the m6A-specificantibody could have intrinsic bias on RNA sequences andsecondary structures and could not distinguish m6A from N6,2′-O-dimethyladenosine (m6Am);101,103 therefore, antibody-in-dependent methods are still desired to dissect the distributionof m6A in an unbiased way. For m5C detection, RNA bisulfitesequencing can directly detect site-specific endogenous m5Csites, but it still has several disadvantages: It did not pre-enrichm5C sites of interest, so extremely high depths of sequencingare required to detect methylated RNAs of low abundance. Inaddition, bisulfite treatment could lead to significant RNAdegradation owing to harsh chemical and thermal conditions;incomplete conversion of cytosines may be included, and other

cytosine modifications are resistant to bisulfite treatment, bothof which result in false positives.110−112 Although otherapproaches (Aza-IP and miCLIP of m5C) are bisulfite-freeand can pre-enrich m5C targets, overexpression of methyl-transferase is required, which may lead to some nonspecifictargets due to the high expression and potential mis-localizationof the enzyme within the cell. More sensitive and accurate m5Cdetection methods need to be developed in the future. For Ψdetection, CMC-based Ψ sequencing methods have successfullyidentified Ψ sites in the transcriptome at single-base resolution,but the alkaline treatment step gives rise to RNA degradation.Further improvements in future Ψ profiling are needed, e.g.,finding new chemicals that specifically react with Ψ. For hm5Cand m1A detection, current sequencing technologies have notreached single-base resolution, which hinders functional studiesof RNA modification. Learning from the success of single-baseand quantitative m6A sequencing technologies, optimizedsequencing methods of hm5C and m1A are expected in thefuture.Other considerations are also required for functional studies

of RNA modifications. First, it is important to quantify theabsolute stoichiometry of RNA modifications. Modificationswith high stoichiometry in a certain transcript are more likely toaffect the fate of the RNA. At present, except for m6A-LAIC-seq, most sequencing technologies cannot quantify thepercentage of transcripts containing modifications. Second, itis still unknown whether any correlations exist among differentRNA modifications. Multiple RNA modifications may togethercontribute to regulate the biological consequences for a certaintranscript. Therefore, it will be helpful to develop high-throughput technologies that can identify multiple modifica-tions at the same molecule simultaneously. Third, all existingRNA modification sequencing methods cannot be applied tosingle cells directly. It is known that m6A has a regulatory rolein ESC pluripotency; thus, single-cell sequencing of m6A, aswell as other modifications, is desired to uncover thecontributions of RNA modifications in cell fate determinationduring development and differentiation. In summary, sequenc-ing technologies of RNA modifications are just beginning. Weenvision that future tools, particularly novel chemical biologymethods, could be applied to further facilitate studies in thefield of epitranscriptomics.

■ CONCLUDING REMARKSGreat progress on RNA modifications has been achieved overthe past five years due to the rapid development of novel RNAmodification sequencing technologies. m6A has been revealedto affect a broad set of biological functions, including mRNAsplicing, export, translation, stability, structure, and miRNAbiogenesis; emerging evidence also suggests m6A as a regulatorin embryonic stem cell state transition and immune response.In recent studies, other dynamic chemical modifications ofmRNA (m5C, Ψ, hm5C, and m1A) were also identified in thetranscriptome. However, the biological function of thesemRNA modifications is poorly understood due to the lack ofoptimized detection methods and necessary knowledge. First,similar to that of m6A, optimized detection methods are neededto reach single-base resolution and to quantify the stoichiom-etry. Furthermore, identifying the specific modificationenzymes for a given RNA modification is urgent. For m1A, itis still unclear which methyltransferase could catalyze theformation of m1A in mRNA. For Ψ and m5C, modificationenzymes are shared between mRNA and tRNA, which make it

ACS Chemical Biology Reviews

DOI: 10.1021/acschembio.6b00960ACS Chem. Biol. 2017, 12, 316−325

321

difficult to study the function in mRNA. Moreover, identifyingthe reader proteins that specifically recognize a given RNAmodification will contribute further to our understanding of thefunctions of RNA modifications. We are still at the verybeginning of the epitranscriptomics epoch, and we believe thatvital input from the chemical biology field will significantly andcontinuously enrich our understanding of RNA biology.

■ AUTHOR INFORMATIONCorresponding Author*E-mail: [email protected].

ORCIDChengqi Yi: 0000-0003-2540-9729NotesThe authors declare no competing financial interest.

■ ACKNOWLEDGMENTSWe thank X. Shu and C. Zhu for their insights and discussions.This work was supported by the National Basic ResearchFoundation of China (No. MOST2016YFC0900300), theNational Natural Science Foundation of China (No.21522201) and the Beijing Natural Science Foundation (No.5162012). We apologize for not being able to cite allpublications related to this topic due to space constraints ofthe journal.

■ KEYWORDSRNA modification: natural RNA molecules contain variouschemically modified nucleosides, which are derived from thefour standard nucleosides (adenosine, guanosine, cytidine, anduridine), and over 100 chemically distinct modifications inRNA have been identified in cellular RNA to date; N6-methyladenosine: the most prevalent internal modification inmRNA is methylation at the N6 position of adenosine; 5-methylcytosine: the methylation at the C5 position of cytosineis another RNA modification deposited into a wide range oftRNA, rRNA, mRNA, and lncRNA; pseudouridine: known asthe “fifth nucleotide” of RNA with the most abundant RNAmodification being a C−C glycosidic rotation isomer of uridine;N1-methyladenosine: modification in RNA that has a methylgroup in the N1 position of adenosine and is present in tRNA,rRNA, and recently shown in mRNA and lncRNA; 5-hydroxymethylcytosine: an oxidation derivative of 5-methyl-cytosine in RNA catalyzed by ten-eleven translocation (TET)family enzymes; epitranscriptomics: booming and expandingresearch area of RNA modifications

■ REFERENCES(1) Song, X., and Nazar, R. N. (2002) Modification of rRNA as a’quality control mechanism’ in ribosome biogenesis. FEBS Lett. 523,182−186.(2) Agris, P. F. (2004) Decoding the genome: a modified view.Nucleic Acids Res. 32, 223−238.(3) Karijolich, J., Yi, C., and Yu, Y. T. (2015) Transcriptome-widedynamics of RNA pseudouridylation. Nat. Rev. Mol. Cell Biol. 16, 581−585.(4) Mamane, Y., Petroulakis, E., Rong, L., Yoshida, K., Ler, L. W., andSonenberg, N. (2004) eIF4E–from translation to transformation.Oncogene 23, 3172−3179.(5) Topisirovic, I., Svitkin, Y. V., Sonenberg, N., and Shatkin, A. J.(2011) Cap and cap-binding proteins in the control of geneexpression. Wiley Interdiscip. Rev. RNA 2, 277−298.

(6) Fossat, N., Tourle, K., Radziewic, T., Barratt, K., Liebhold, D.,Studdert, J. B., Power, M., Jones, V., Loebel, D. A., and Tam, P. P.(2014) C to U RNA editing mediated by APOBEC1 requires RNA-binding protein RBM47. EMBO Rep. 15, 903−910.(7) Nishikura, K. (2016) A-to-I editing of coding and non-codingRNAs by ADARs. Nat. Rev. Mol. Cell Biol. 17, 83−96.(8) Slotkin, W., and Nishikura, K. (2013) Adenosine-to-inosine RNAediting and human disease. Genome Med. 5, 105.(9) Wulff, B. E., Sakurai, M., and Nishikura, K. (2011) Elucidating theinosinome: global approaches to adenosine-to-inosine RNA editing.Nat. Rev. Genet. 12, 81−85.(10) Desrosiers, R., Friderici, K., and Rottman, F. (1974)Identification of methylated nucleosides in messenger RNA fromNovikoff hepatoma cells. Proc. Natl. Acad. Sci. U. S. A. 71, 3971−3975.(11) Engel, J. D., and von Hippel, P. H. (1974) Effects of methylationon the stability of nucleic acid conformations: studies at the monomerlevel. Biochemistry 13, 4143−4158.(12) Engel, J. D., and von Hippel, P. H. (1978) Effects of methylationon the stability of nucleic acid conformations. Studies at the polymerlevel. J. Biol. Chem. 253, 927−934.(13) Roost, C., Lynch, S. R., Batista, P. J., Qu, K., Chang, H. Y., andKool, E. T. (2015) Structure and thermodynamics of N6-methyladenosine in RNA: a spring-loaded base modification. J. Am.Chem. Soc. 137, 2107−2115.(14) Spitale, R. C., Flynn, R. A., Zhang, Q. C., Crisalli, P., Lee, B.,Jung, J. W., Kuchelmeister, H. Y., Batista, P. J., Torre, E. A., Kool, E. T.,and Chang, H. Y. (2015) Structural imprints in vivo decode RNAregulatory mechanisms. Nature 519, 486−490.(15) Bokar, J. A., Rath-Shambaugh, M. E., Ludwiczak, R., Narayan, P.,and Rottman, F. (1994) Characterization and partial purification ofmRNA N6-adenosine methyltransferase from HeLa cell nuclei.Internal mRNA methylation requires a multisubunit complex. J. Biol.Chem. 269, 17697−17704.(16) Liu, J., Yue, Y., Han, D., Wang, X., Fu, Y., Zhang, L., Jia, G., Yu,M., Lu, Z., Deng, X., Dai, Q., Chen, W., and He, C. (2014) AMETTL3-METTL14 complex mediates mammalian nuclear RNA N6-adenosine methylation. Nat. Chem. Biol. 10, 93−95.(17) Wang, Y., Li, Y., Toth, J. I., Petroski, M. D., Zhang, Z., and Zhao,J. C. (2014) N6-methyladenosine modification destabilizes devel-opmental regulators in embryonic stem cells. Nat. Cell Biol. 16, 191−198.(18) Schwartz, S., Mumbach, Maxwell, R., Jovanovic, M., Wang, T.,Maciag, K., Bushkin, G. G., Mertins, P., Ter-Ovanesyan, D., Habib, N.,Cacchiarelli, D., Sanjana, Neville, E., Freinkman, E., Pacold, Michael,E., Satija, R., Mikkelsen, Tarjei, S., Hacohen, N., Zhang, F., Carr,Steven, A., Lander, Eric, S., and Regev, A. (2014) Perturbation of m6AWriters Reveals Two Distinct Classes of mRNA Methylation atInternal and 5′ Sites. Cell Rep. 8, 284−296.(19) Ping, X. L., Sun, B. F., Wang, L., Xiao, W., Yang, X., Wang, W. J.,Adhikari, S., Shi, Y., Lv, Y., Chen, Y. S., Zhao, X., Li, A., Yang, Y.,Dahal, U., Lou, X. M., Liu, X., Huang, J., Yuan, W. P., Zhu, X. F.,Cheng, T., Zhao, Y. L., Wang, X., Rendtlew Danielsen, J. M., Liu, F.,and Yang, Y. G. (2014) Mammalian WTAP is a regulatory subunit ofthe RNA N6-methyladenosine methyltransferase. Cell Res. 24, 177−189.(20) Wang, X., Feng, J., Xue, Y., Guan, Z., Zhang, D., Liu, Z., Gong,Z., Wang, Q., Huang, J., Tang, C., Zou, T., and Yin, P. (2016)Structural basis of N(6)-adenosine methylation by the METTL3-METTL14 complex. Nature 534, 575−578.(21) Wang, P., Doxtader, K. A., and Nam, Y. (2016) Structural Basisfor Cooperative Function of Mettl3 and Mettl14 Methyltransferases.Mol. Cell 63, 306−317.(22) Harper, J. E., Miceli, S. M., Roberts, R. J., and Manley, J. L.(1990) Sequence specificity of the human mRNA N6-adenosinemethylase in vitro. Nucleic Acids Res. 18, 5735−5741.(23) Jia, G., Fu, Y., Zhao, X., Dai, Q., Zheng, G., Yang, Y., Yi, C.,Lindahl, T., Pan, T., Yang, Y. G., and He, C. (2011) N6-methyladenosine in nuclear RNA is a major substrate of the obesity-associated FTO. Nat. Chem. Biol. 7, 885−887.

ACS Chemical Biology Reviews

DOI: 10.1021/acschembio.6b00960ACS Chem. Biol. 2017, 12, 316−325

322

(24) Zheng, G., Dahl, J. A., Niu, Y., Fedorcsak, P., Huang, C. M., Li,C. J., Vagbo, C. B., Shi, Y., Wang, W. L., Song, S. H., Lu, Z., Bosmans,R. P., Dai, Q., Hao, Y. J., Yang, X., Zhao, W. M., Tong, W. M., Wang,X. J., Bogdan, F., Furu, K., Fu, Y., Jia, G., Zhao, X., Liu, J., Krokan, H.E., Klungland, A., Yang, Y. G., and He, C. (2013) ALKBH5 Is aMammalian RNA Demethylase that Impacts RNA Metabolism andMouse Fertility. Mol. Cell 49, 18−29.(25) Fu, Y., Jia, G., Pang, X., Wang, R. N., Wang, X., Li, C. J., Smemo,S., Dai, Q., Bailey, K. A., Nobrega, M. A., Han, K. L., Cui, Q., and He,C. (2013) FTO-mediated formation of N6-hydroxymethyladenosineand N6-formyladenosine in mammalian RNA. Nat. Commun. 4, 1798.(26) Zheng, G., Dahl, J. A., Niu, Y., Fu, Y., Klungland, A., Yang, Y. G.,and He, C. (2013) Sprouts of RNA epigenetics: the discovery ofmammalian RNA demethylases. RNA Biol. 10, 915−918.(27) Dunn, D. B. (1961) The occurrence of 1-methyladenine inribonucleic acid. Biochim. Biophys. Acta 46, 198−200.(28) Hall, R. H. (1963) Method for isolation of 2′-O-methylribonucleosides and N1-methyladenosine from ribonucleicacid. Biochim. Biophys. Acta, Spec. Sect. Nucleic Acids Relat. Subj. 68,278−283.(29) El Yacoubi, B., Bailly, M., and de Crecy-Lagard, V. (2012)Biosynthesis and function of posttranscriptional modifications oftransfer RNAs. Annu. Rev. Genet. 46, 69−95.(30) Sharma, S., Watzinger, P., Kotter, P., and Entian, K. D. (2013)Identification of a novel methyltransferase, Bmt2, responsible for theN-1-methyl-adenosine base modification of 25S rRNA in Saccha-romyces cerevisiae. Nucleic Acids Res. 41, 5428−5443.(31) Peifer, C., Sharma, S., Watzinger, P., Lamberth, S., Kotter, P.,and Entian, K. D. (2013) Yeast Rrp8p, a novel methyltransferaseresponsible for m1A 645 base modification of 25S rRNA. Nucleic AcidsRes. 41, 1151−1163.(32) Li, X., Xiong, X., Wang, K., Wang, L., Shu, X., Ma, S., and Yi, C.(2016) Transcriptome-wide mapping reveals reversible and dynamicN(1)-methyladenosine methylome. Nat. Chem. Biol. 12, 311−316.(33) Dominissini, D., Nachtergaele, S., Moshitch-Moshkovitz, S.,Peer, E., Kol, N., Ben-Haim, M. S., Dai, Q., Di Segni, A., Salmon-Divon, M., Clark, W. C., Zheng, G., Pan, T., Solomon, O., Eyal, E.,Hershkovitz, V., Han, D., Dore, L. C., Amariglio, N., Rechavi, G., andHe, C. (2016) The dynamic N(1)-methyladenosine methylome ineukaryotic messenger RNA. Nature 530, 441−446.(34) Helm, M. (2006) Post-transcriptional nucleotide modificationand alternative folding of RNA. Nucleic Acids Res. 34, 721−733.(35) Helm, M., Giege, R., and Florentz, C. (1999) A Watson-Crickbase-pair-disrupting methyl group (m1A9) is sufficient for cloverleaffolding of human mitochondrial tRNALys. Biochemistry 38, 13338−13346.(36) Agris, P. F., Sierzputowska-Gracz, H., and Smith, C. (1986)Transfer RNA contains sites of localized positive charge: carbon NMRstudies of [13C]methyl-enriched Escherichia coli and yeast tRNAPhe.Biochemistry 25, 5126−5131.(37) Hauenschild, R., Tserovski, L., Schmid, K., Thuring, K., Winz,M. L., Sharma, S., Entian, K. D., Wacheul, L., Lafontaine, D. L.,Anderson, J., Alfonzo, J., Hildebrandt, A., Jaschke, A., Motorin, Y., andHelm, M. (2015) The reverse transcription signature of N-1-methyladenosine in RNA-Seq is sequence dependent. Nucleic AcidsRes. 43, 9950−9964.(38) Waku, T., and Nakajima, Y. (2016) NML-mediated rRNA basemethylation links ribosomal subunit formation to cell proliferation in ap53-dependent manner. J. Cell Sci. 129, 2382−2393.(39) Chujo, T., and Suzuki, T. (2012) Trmt61B is a methyltransfer-ase responsible for 1-methyladenosine at position 58 of humanmitochondrial tRNAs. RNA 18, 2269−2276.(40) Ozanick, S., Krecic, A., Andersland, J., and Anderson, J. T.(2005) The bipartite structure of the tRNA m1A58 methyltransferasefrom S. cerevisiae is conserved in humans. RNA 11, 1281−1290.(41) Vilardo, E., Nachbagauer, C., Buzet, A., Taschner, A., Holzmann,J., and Rossmanith, W. (2012) A subcomplex of human mitochondrialRNase P is a bifunctional methyltransferase–extensive moonlighting inmitochondrial tRNA biogenesis. Nucleic Acids Res. 40, 11583−11593.

(42) Liu, F., Clark, W., Luo, G., Wang, X., Fu, Y., Wei, J., Wang, X.,Hao, Z., Dai, Q., Zheng, G., Ma, H., Han, D., Evans, M., Klungland, A.,Pan, T., and He, C. (2016) ALKBH1-Mediated tRNA DemethylationRegulates Translation. Cell 167, 816−828.(43) Klagsbrun, M. (1973) An evolutionary study of the methylationof transfer and ribosomal ribonucleic acid in prokaryote and eukaryoteorganisms. J. Biol. Chem. 248, 2612−2620.(44) Motorin, Y., and Helm, M. (2010) tRNA stabilization bymodified nucleotides. Biochemistry 49, 4934−4944.(45) Squires, J. E., and Preiss, T. (2010) Function and detection of 5-methylcytosine in eukaryotic RNA. Epigenomics 2, 709−715.(46) Chow, C. S., Lamichhane, T. N., and Mahto, S. K. (2007)Expanding the nucleotide repertoire of the ribosome with post-transcriptional modifications. ACS Chem. Biol. 2, 610−619.(47) Squires, J. E., Patel, H. R., Nousch, M., Sibbritt, T., Humphreys,D. T., Parker, B. J., Suter, C. M., and Preiss, T. (2012) Widespreadoccurrence of 5-methylcytosine in human coding and non-codingRNA. Nucleic Acids Res. 40, 5023−5033.(48) Hussain, S., Sajini, A. A., Blanco, S., Dietmann, S., Lombard, P.,Sugimoto, Y., Paramor, M., Gleeson, J. G., Odom, D. T., Ule, J., andFrye, M. (2013) NSun2-mediated cytosine-5 methylation of vaultnoncoding RNA determines its processing into regulatory small RNAs.Cell Rep. 4, 255−261.(49) Khoddami, V., and Cairns, B. R. (2013) Identification of directtargets and modified bases of RNA cytosine methyltransferases. Nat.Biotechnol. 31, 458−464.(50) Edelheit, S., Schwartz, S., Mumbach, M. R., Wurtzel, O., andSorek, R. (2013) Transcriptome-wide mapping of 5-methylcytidineRNA modifications in bacteria, archaea, and yeast reveals m5C withinarchaeal mRNAs. PLoS Genet. 9, e1003602.(51) Fu, L., Guerrero, C. R., Zhong, N., Amato, N. J., Liu, Y., Liu, S.,Cai, Q., Ji, D., Jin, S. G., Niedernhofer, L. J., Pfeifer, G. P., Xu, G. L.,and Wang, Y. (2014) Tet-mediated formation of 5-hydroxymethylcy-tosine in RNA. J. Am. Chem. Soc. 136, 11582−11585.(52) Huber, S. M., van Delft, P., Mendil, L., Bachman, M., Smollett,K., Werner, F., Miska, E. A., and Balasubramanian, S. (2015)Formation and abundance of 5-hydroxymethylcytosine in RNA.ChemBioChem 16, 752−755.(53) Delatte, B., Wang, F., Ngoc, L. V., Collignon, E., Bonvin, E.,Deplus, R., Calonne, E., Hassabi, B., Putmans, P., Awe, S., Wetzel, C.,Kreher, J., Soin, R., Creppe, C., Limbach, P. A., Gueydan, C., Kruys, V.,Brehm, A., Minakhina, S., Defrance, M., Steward, R., and Fuks, F.(2016) RNA biochemistry. Transcriptome-wide distribution andfunction of RNA hydroxymethylcytosine. Science 351, 282−285.(54) Cohn, W. E. (1951) Some results of the applications of ion-exchange chromatography to nucleic acid chemistry. J. Cell. Comp.Physiol. 38, 21−40.(55) Cohn, W. E., and Volkin, E. (1951) Nucleoside-5′-Phosphatesfrom Ribonucleic Acid. Nature 167, 483−484.(56) Arnez, J. G., and Steitz, T. A. (1994) Crystal structure ofunmodified tRNA(Gln) complexed with glutaminyl-tRNA synthetaseand ATP suggests a possible role for pseudo-uridines in stabilization ofRNA structure. Biochemistry 33, 7560−7567.(57) Li, X., Ma, S., and Yi, C. (2016) Pseudouridine: the fifth RNAnucleotide with renewed interests. Curr. Opin. Chem. Biol. 33, 108−116.(58) Schwartz, S., Bernstein, D. A., Mumbach, M. R., Jovanovic, M.,Herbst, R. H., Leon-Ricardo, B. X., Engreitz, J. M., Guttman, M., Satija,R., Lander, E. S., Fink, G., and Regev, A. (2014) Transcriptome-widemapping reveals widespread dynamic-regulated pseudouridylation ofncRNA and mRNA. Cell 159, 148−162.(59) Carlile, T. M., Rojas-Duran, M. F., Zinshteyn, B., Shin, H.,Bartoli, K. M., and Gilbert, W. V. (2014) Pseudouridine profilingreveals regulated mRNA pseudouridylation in yeast and human cells.Nature 515, 143−146.(60) Lovejoy, A. F., Riordan, D. P., and Brown, P. O. (2014)Transcriptome-wide mapping of pseudouridines: pseudouridinesynthases modify specific mRNAs in S. cerevisiae. PLoS One 9,e110799.

ACS Chemical Biology Reviews

DOI: 10.1021/acschembio.6b00960ACS Chem. Biol. 2017, 12, 316−325

323

(61) Li, X., Zhu, P., Ma, S., Song, J., Bai, J., Sun, F., and Yi, C. (2015)Chemical pulldown reveals dynamic pseudouridylation of themammalian transcriptome. Nat. Chem. Biol. 11, 592−597.(62) Kiss, T., Fayet-Lebaron, E., and Jady, B. E. (2010) Box H/ACAsmall ribonucleoproteins. Mol. Cell 37, 597−606.(63) Hamma, T., and Ferre-D’Amare, A. R. (2006) Pseudouridinesynthases. Chem. Biol. 13, 1125−1135.(64) Spenkuch, F., Motorin, Y., and Helm, M. (2014) Pseudouridine:still mysterious, but never a fake (uridine)! RNA Biol. 11, 1540−1554.(65) Wurm, J. P., Meyer, B., Bahr, U., Held, M., Frolow, O., Kotter,P., Engels, J. W., Heckel, A., Karas, M., Entian, K. D., and Wohnert, J.(2010) The ribosome assembly factor Nep1 responsible for Bowen-Conradi syndrome is a pseudouridine-N1-specific methyltransferase.Nucleic Acids Res. 38, 2387−2398.(66) Alarcon, C. R., Goodarzi, H., Lee, H., Liu, X., Tavazoie, S., andTavazoie, S. F. (2015) HNRNPA2B1 Is a Mediator of m(6)A-Dependent Nuclear RNA Processing Events. Cell 162, 1299−1308.(67) Alarcon, C. R., Lee, H., Goodarzi, H., Halberg, N., and Tavazoie,S. F. (2015) N6-methyladenosine marks primary microRNAs forprocessing. Nature 519, 482−485.(68) Liu, N., Dai, Q., Zheng, G., He, C., Parisien, M., and Pan, T.(2015) N(6)-methyladenosine-dependent RNA structural switchesregulate RNA-protein interactions. Nature 518, 560−564.(69) Xiao, W., Adhikari, S., Dahal, U., Chen, Y. S., Hao, Y. J., Sun, B.F., Sun, H. Y., Li, A., Ping, X. L., Lai, W. Y., Wang, X., Ma, H. L.,Huang, C. M., Yang, Y., Huang, N., Jiang, G. B., Wang, H. L., Zhou, Q.,Wang, X. J., Zhao, Y. L., and Yang, Y. G. (2016) Nuclear m(6)AReader YTHDC1 Regulates mRNA Splicing. Mol. Cell 61, 507−519.(70) Zhao, X., Yang, Y., Sun, B. F., Shi, Y., Yang, X., Xiao, W., Hao, Y.J., Ping, X. L., Chen, Y. S., Wang, W. J., Jin, K. X., Wang, X., Huang, C.M., Fu, Y., Ge, X. M., Song, S. H., Jeong, H. S., Yanagisawa, H., Niu, Y.,Jia, G. F., Wu, W., Tong, W. M., Okamoto, A., He, C., RendtlewDanielsen, J. M., Wang, X. J., and Yang, Y. G. (2014) FTO-dependentdemethylation of N6-methyladenosine regulates mRNA splicing and isrequired for adipogenesis. Cell Res. 24, 1403−1419.(71) Molinie, B., Wang, J., Lim, K. S., Hillebrand, R., and Lu, Z. X.(2016) m(6)A-LAIC-seq reveals the census and complexity of them(6)A epitranscriptome. Nat. Methods 13, 692−698.(72) Ke, S., Alemu, E. A., Mertens, C., Gantman, E. C., Fak, J. J.,Mele, A., Haripal, B., Zucker-Scharff, I., Moore, M. J., Park, C. Y.,Vagbo, C. B., Kussnierczyk, A., Klungland, A., Darnell, J. E., Jr., andDarnell, R. B. (2015) A majority of m6A residues are in the last exons,allowing the potential for 3′ UTR regulation. Genes Dev. 29, 2037−2053.(73) Wang, X., Lu, Z., Gomez, A., Hon, G. C., Yue, Y., Han, D., Fu,Y., Parisien, M., Dai, Q., Jia, G., Ren, B., Pan, T., and He, C. (2014)N6-methyladenosine-dependent regulation of messenger RNA stabil-ity. Nature 505, 117−120.(74) Geula, S., Moshitch-Moshkovitz, S., Dominissini, D., Mansour,A. A., Kol, N., Salmon-Divon, M., Hershkovitz, V., Peer, E., Mor, N.,Manor, Y. S., Ben-Haim, M. S., Eyal, E., Yunger, S., Pinto, Y., Jaitin, D.A., Viukov, S., Rais, Y., Krupalnik, V., Chomsky, E., Zerbib, M., Maza,I., Rechavi, Y., Massarwa, R., Hanna, S., Amit, I., Levanon, E. Y.,Amariglio, N., Stern-Ginossar, N., Novershtern, N., Rechavi, G., andHanna, J. H. (2015) Stem cells. m6A mRNA methylation facilitatesresolution of naive pluripotency toward differentiation. Science 347,1002−1006.(75) Wang, X., Zhao, B. S., Roundtree, I. A., Lu, Z., Han, D., Ma, H.,Weng, X., Chen, K., Shi, H., and He, C. (2015) N(6)-methyladenosineModulates Messenger RNA Translation Efficiency. Cell 161, 1388−1399.(76) Meyer, K. D., Patil, D. P., Zhou, J., Zinoviev, A., Skabkin, M. A.,Elemento, O., Pestova, T. V., Qian, S. B., and Jaffrey, S. R. (2015) 5′UTR m(6)A Promotes Cap-Independent Translation. Cell 163, 999−1010.(77) Zhou, J., Wan, J., Gao, X., Zhang, X., Jaffrey, S. R., and Qian, S.B. (2015) Dynamic m(6)A mRNA methylation directs translationalcontrol of heat shock response. Nature 526, 591−594.

(78) Lin, S., Choe, J., Du, P., Triboulet, R., and Gregory, R. I. (2016)The m(6)A Methyltransferase METTL3 Promotes Translation inHuman Cancer Cells. Mol. Cell 62, 335−345.(79) Zhao, B. S., and He, C. (2015) Fate by RNA methylation: m6Asteers stem cell pluripotency. Genome Biol. 16, 43.(80) Chen, T., Hao, Y. J., Zhang, Y., Li, M. M., Wang, M., Han, W.,Wu, Y., Lv, Y., Hao, J., Wang, L., Li, A., Yang, Y., Jin, K. X., Zhao, X.,Li, Y., Ping, X. L., Lai, W. Y., Wu, L. G., Jiang, G., Wang, H. L., Sang,L., Wang, X. J., Yang, Y. G., and Zhou, Q. (2015) m(6)A RNAmethylation is regulated by microRNAs and promotes reprogrammingto pluripotency. Cell Stem Cell 16, 289−301.(81) Batista, P. J., Molinie, B., Wang, J., Qu, K., Zhang, J., Li, L.,Bouley, D. M., Lujan, E., Haddad, B., Daneshvar, K., Carter, A. C.,Flynn, R. A., Zhou, C., Lim, K. S., Dedon, P., Wernig, M., Mullen, A.C., Xing, Y., Giallourakis, C. C., and Chang, H. Y. (2014) m(6)A RNAmodification controls cell fate transition in mammalian embryonicstem cells. Cell Stem Cell 15, 707−719.(82) Aguilo, F., Zhang, F., Sancho, A., Fidalgo, M., Di Cecilia, S.,Vashisht, A., Lee, D. F., Chen, C. H., Rengasamy, M., Andino, B.,Jahouh, F., Roman, A., Krig, S. R., Wang, R., Zhang, W., Wohlschlegel,J. A., Wang, J., and Walsh, M. J. (2015) Coordination of m(6)AmRNA Methylation and Gene Transcription by ZFP217 RegulatesPluripotency and Reprogramming. Cell Stem Cell 17, 689−704.(83) Patil, D. P., Chen, C. K., Pickering, B. F., Chow, A., Jackson, C.,Guttman, M., and Jaffrey, S. R. (2016) m6A RNA methylationpromotes XIST-mediated transcriptional repression. Nature 537, 369−373.(84) Lichinchi, G., Gao, S., Saletore, Y., and Gonzalez, G. M. (2016)Dynamics of the human and viral m(6)A RNA methylomes duringHIV-1 infection of T cells. Nat. Microbiol. 1, 16011.(85) Kennedy, E. M., Bogerd, H. P., Kornepati, A. V., Kang, D.,Ghoshal, D., Marshall, J. B., Poling, B. C., Tsai, K., Gokhale, N. S.,Horner, S. M., and Cullen, B. R. (2016) Posttranscriptional m(6)AEditing of HIV-1 mRNAs Enhances Viral Gene Expression. Cell HostMicrobe 19, 675−685.(86) Tirumuru, N., Zhao, B. S., Lu, W., Lu, Z., He, C., and Wu, L.(2016) N(6)-methyladenosine of HIV-1 RNA regulates viral infectionand HIV-1 Gag protein expression. eLife 5, e15528.(87) Lichinchi, G., Zhao, B. S., Wu, Y., Lu, Z., Qin, Y., He, C., andRana, T. M. (2016) Dynamics of Human and Viral RNA Methylationduring Zika Virus Infection. Cell Host Microbe 20, 666−673.(88) Gokhale, N. S., McIntyre, A. B., McFadden, M. J., Roder, A. E.,Kennedy, E. M., Gandara, J. A., Hopcraft, S. E., Quicke, K. M.,Vazquez, C., Willer, J., Ilkayeva, O. R., Law, B. A., Holley, C. L., Garcia-Blanco, M. A., Evans, M. J., Suthar, M. S., Bradrick, S. S., Mason, C. E.,and Horner, S. M. (2016) N6-Methyladenosine in Flaviviridae ViralRNA Genomes Regulates Infection. Cell Host Microbe 20, 654−665.(89) Kariko,́ K., Buckstein, M., Ni, H., and Weissman, D. (2005)Suppression of RNA Recognition by Toll-like Receptors: The Impactof Nucleoside Modification and the Evolutionary Origin of RNA.Immunity 23, 165−175.(90) Hull, C. M., and Bevilacqua, P. C. (2016) Discriminating Selfand Non-Self by RNA: Roles for RNA Structure, Misfolding, andModification in Regulating the Innate Immune Sensor PKR. Acc.Chem. Res. 49, 1242−1249.(91) Kariko, K., Muramatsu, H., Ludwig, J., and Weissman, D. (2011)Generating the optimal mRNA for therapy: HPLC purificationeliminates immune activation and improves translation of nucleo-side-modified, protein-encoding mRNA. Nucleic Acids Res. 39, e142.(92) Kariko, K., Muramatsu, H., Keller, J. M., and Weissman, D.(2012) Increased erythropoiesis in mice injected with submicrogramquantities of pseudouridine-containing mRNA encoding erythropoie-tin. Mol. Ther. 20, 948−953.(93) Chen, C., Zhao, X., Kierzek, R., and Yu, Y. T. (2010) A flexibleRNA backbone within the polypyrimidine tract is required forU2AF65 binding and pre-mRNA splicing in vivo. Mol. Cell. Biol. 30,4108−4119.(94) Kariko, K., Muramatsu, H., Welsh, F. A., Ludwig, J., Kato, H.,Akira, S., and Weissman, D. (2008) Incorporation of pseudouridine

ACS Chemical Biology Reviews

DOI: 10.1021/acschembio.6b00960ACS Chem. Biol. 2017, 12, 316−325

324

into mRNA yields superior nonimmunogenic vector with increasedtranslational capacity and biological stability. Mol. Ther. 16, 1833−1840.(95) Hoernes, T. P., Clementi, N., Faserl, K., Glasner, H., Breuker,K., Lindner, H., Huttenhofer, A., and Erlacher, M. D. (2016)Nucleotide modifications within bacterial messenger RNAs regulatetheir translation and are able to rewire the genetic code. Nucleic AcidsRes. 44, 852−862.(96) Karijolich, J., and Yu, Y. T. (2011) Converting nonsense codonsinto sense codons by targeted pseudouridylation. Nature 474, 395−398.(97) Fernandez, I. S., Ng, C. L., Kelley, A. C., Wu, G., Yu, Y. T., andRamakrishnan, V. (2013) Unusual base pairing during the decoding ofa stop codon by the ribosome. Nature 500, 107−110.(98) Hoernes, T. P., and Erlacher, M. D. (2017) Translating theepitranscriptome. Wiley Interdiscip. Rev. RNA 8, e1375.(99) Dominissini, D., Moshitch-Moshkovitz, S., Schwartz, S., Salmon-Divon, M., Ungar, L., Osenberg, S., Cesarkas, K., Jacob-Hirsch, J.,Amariglio, N., Kupiec, M., Sorek, R., and Rechavi, G. (2012) Topologyof the human and mouse m6A RNA methylomes revealed by m6A-seq. Nature 485, 201−206.(100) Meyer, K. D., Saletore, Y., Zumbo, P., Elemento, O., Mason, C.E., and Jaffrey, S. R. (2012) Comprehensive analysis of mRNAmethylation reveals enrichment in 3′ UTRs and near stop codons. Cell149, 1635−1646.(101) Schwartz, S., Agarwala, S. D., Mumbach, M. R., Jovanovic, M.,Mertins, P., Shishkin, A., Tabach, Y., Mikkelsen, T. S., Satija, R.,Ruvkun, G., Carr, S. A., Lander, E. S., Fink, G. R., and Regev, A.(2013) High-resolution mapping reveals a conserved, widespread,dynamic mRNA methylation program in yeast meiosis. Cell 155,1409−1421.(102) Luo, G. Z., MacQueen, A., Zheng, G., Duan, H., Dore, L. C.,Lu, Z., Liu, J., Chen, K., Jia, G., Bergelson, J., and He, C. (2014)Unique features of the m6A methylome in Arabidopsis thaliana. Nat.Commun. 5, 5630.(103) Linder, B., Grozhik, A. V., Olarerin-George, A. O., Meydan, C.,Mason, C. E., and Jaffrey, S. R. (2015) Single-nucleotide-resolutionmapping of m6A and m6Am throughout the transcriptome. Nat.Methods 12, 767−772.(104) Chen, K., Lu, Z., Wang, X., Fu, Y., Luo, G. Z., Liu, N., Han, D.,Dominissini, D., Dai, Q., Pan, T., and He, C. (2015) High-resolutionN(6) -methyladenosine (m(6) A) map using photo-crosslinking-assisted m(6) A sequencing. Angew. Chem., Int. Ed. 54, 1587−1590.(105) Macon, J. B., and Wolfenden, R. (1968) 1-Methyladenosine.Dimroth rearrangement and reversible reduction. Biochemistry 7,3453−3458.(106) Schaefer, M., Pollex, T., Hanna, K., and Lyko, F. (2009) RNAcytosine methylation analysis by bisulfite sequencing. Nucleic Acids Res.37, e12.(107) Gilham, P. T. (1962) An Addition Reaction Specific forUridine and Guanosine Nucleotides and its Application to theModification of Ribonuclease Action. J. Am. Chem. Soc. 84, 687−688.(108) Bakin, A., and Ofengand, J. (1993) Four newly locatedpseudouridylate residues in Escherichia coli 23S ribosomal RNA are allat the peptidyltransferase center: analysis by the application of a newsequencing technique. Biochemistry 32, 9754−9762.(109) Li, X., Ma, S., and Yi, C. (2015) Pseudouridine ChemicalLabeling and Profiling. Methods Enzymol. 560, 247−272.(110) Gilbert, W. V., Bell, T. A., and Schaening, C. (2016) MessengerRNA modifications: Form, distribution, and function. Science 352,1408−1412.(111) Hussain, S., Aleksic, J., Blanco, S., Dietmann, S., and Frye, M.(2013) Characterizing 5-methylcytosine in the mammalian epitran-scriptome. Genome Biol. 14, 215.(112) Shafik, A., Schumann, U., Evers, M., Sibbritt, T., and Preiss, T.(2016) The emerging epitranscriptomics of long noncoding RNAs.Biochim. Biophys. Acta, Gene Regul. Mech. 1859, 59−70.

ACS Chemical Biology Reviews

DOI: 10.1021/acschembio.6b00960ACS Chem. Biol. 2017, 12, 316−325

325