key: cord-0704454-69vuc6l9
authors: Kruse, Susanne; Zhong, Silin; Bodi, Zsuzsanna; Button, James; Alcocer, Marcos J. C.; Hayes, Christopher J.; Fray, Rupert
title: A novel synthesis and detection method for cap-associated adenosine modifications in mouse mRNA
date: 2011-10-24
journal: Sci Rep
DOI: 10.1038/srep00126
sha: 90cc93e5eac1e836f5cbce50fe92ded531045abf
doc_id: 704454
cord_uid: 69vuc6l9

A method is described for the detection of certain nucleotide modifications adjacent to the 5' 7-methyl guanosine cap of mRNAs from individual genes. The method quantitatively measures the relative abundance of 2'-O-methyl and N(6),2'-O-dimethyladenosine, two of the most common modifications. In order to identify and quantitatify the amounts of N(6),2'-O-dimethyladenosine, a novel method for the synthesis of modified adenosine phosphoramidites was developed. This method is a one step synthesis and the product can directly be used for the production of N(6),2'-O-dimethyladenosine containing RNA oligonucleotides. The nature of the cap-adjacent nucleotides were shown to be characteristic for mRNAs from individual genes transcribed in liver and testis.

I n most eukaryotes, polymerase II (Pol II) dependent transcripts are modified co-transcriptionally at their 5'end by the addition of a 7-methyl guanosine (m 7 G) cap to the first nucleoside of the nascent transcript. If no additional modifications are made to the cap-adjacent nucleotides, the structure is referred to as a cap0. In yeast and plants, only cap0 structures are found, however, in animals, modifications of the two nucleotides adjacent to the m 7 G are possible The methylation of the first nucleotide on the ribose residue ( Fig. 1a) will form a cap1 structure 1 . Cap1 messages can be converted to cap2 structures if a further 2'ribose methylation takes place on the next nucleotide following the cap1 (Fig. 1b) . These methylation steps are sequential and carried out by nuclear located methylases [1] [2] [3] . In a transcript where the first nucleotide is an adenosine in a cap1 structure, a further methylation of the 2'-O-dimethyladenosine (Am) at the N 6 position of the adenine can take place to give N 6 ,2'-O-dimethyladenosine (m 6 Am) (Fig. 1c) 4 . The methyltransferase that carries out this modification has been partially characterized and appears to be predominantly located in the cytoplasm 5 . Transcription start sites of focused promoters are usually contained within an initiator motif (Inr) of general sequence YYANWYY (Y5pyrimidine, N-any nucleotide, W5A or T) where the A is the principal start site and neighbouring nucleotides may be used to varying degrees 6 . Since, A is often the first nucleotide after the m 7 G cap, its modification may have a functional role. In order to study the effect of m 6 Am present in cap1 messages, the chemical synthesis of RNA oligonucleotide sequences that contain m 6 Am in well-defined positions, is necessary. A number of methods exist in the literature for the preparation of N 6 -methylated adenosine derivatives, with the Dimroth rearrangement being perhaps the most well known [7] [8] [9] . This transformation relies upon an initial N1-methylation of the adenine ring followed by an alkali-mediated rearrangement to give the N6-methylated adenine product. An alternative method for accessing N6-alkylated adenines has been developed, which involves the nucleophilic aromatic substitution of adenine derivitives, that are activated at the 6-position, with amine nucleophiles (i.e. MeNH 2 ) [10] [11] [12] [13] . In order to obtain the N6-methylated phosphoramidite reagents from these adenine derivatives, a number of additional synthetic steps are required, which results in a 6 to 8 step synthesis being needed to produce each phosphoramidite reagent [10] [11] [12] [13] . Such approaches are not realistic other than in synthetic chemistry laboratories.

An interaction of m 7 G capped transcripts with the nuclear cap-binding complex (CBC) is important for promoting correct splicing 14 and 3'end formation 15 . After export from the nucleus, translation of most mRNAs is initiated with the recognition of the m 7 G cap by eukaryotic translation initiation factor 4E (eIF4E) 16 . This cap binding step is usually rate-limiting for translation. The m 7 G cap also protects the message from 5' to 3' exonuclease digestion. Compared to the well characterised role of m 7 G, relatively little is known regarding the function of 2' ribose methylation on the following nucleotides. The presence of a cap1 structure may promote binding to ribosomes 17 and can increase translational efficiency in vitro 18, 19 . Many animal viruses encode their own cap1 methylases 20, 21 and this may also help them to limit host immune responses 22, 23 . Whilst viral cap1 methylases are common viral cap2 methylases and m 6 Am methylases have not been reported. However, m 6 Am and cap2 structures can be found in viral mRNAs in vivo, presumably as a result of the action of the host cell's enzymes 24 . The functional consequences for an mRNA of possessing m 6 Am in a cap1 or a cap2 structure are not known.

In order to investigate the roles of these cap-associated modifications, we have developed enzymatic techniques, in combination with thin layer chromatography (TLC) that allow the detection of m 6 Am and other cap nucleotides. We also describe a direct synthetic method for the preparation of the desired m 6 Am phosphoramidite reagent in a single step from commercially available starting material.

Through the use of this synthetic substrate we have demonstrated that a thin layer chromatography (TLC) method is quantitative for determining m 6 Am:Am ratios, and allows the relative proportions of Am and m 6 Am to be rapidly determined for mRNAs from different cell types and also for caps of messages from individual genes. We further give evidence that the relative proportion of m 6 Am varies in a characteristic way for mRNA from different mouse organs.

Synthesis and detection of N 6 , 2'-O-dimethyladenosine. We have previously used TLC assays to detect and quantify N 6 -methyladenosine (m 6 A) 25 , a modification found internally in the mRNA of many eukaryotes [26] [27] [28] [29] . This method was modified to label and detect only the first nucleotide after the cap. In order to use this approach to compare the relative proportions of m 6 Am and Am in cap structures from different sources, it was first necessary to establish the mobility of m 6 Am in the TLC assay and to demonstrate that T4 polynucleotide kinase does not preferentially label either of these nucleotides. As neither m 6 Am triphosphate or m 6 Am phosphoramidite are commercially available, a novel method for its synthesis and incorporation into RNA oligonucleotides was developed.

In order to get quick access to the required m 6 Am phosphoramidite building block for solid-phase oligoribonucleotide synthesis, we performed a selective one-step methylation of the commercially available 2'-OMe-Bz-A-CE phosphoramidite. Due to its high reactivity and sensitivity towards acids and bases, Aritomo's relatively mild phase transfer catalysis (PTC) method was used 30 . Under these conditions (iodomethane, sodium hydroxide, tetrabutylammonium bromide) the N 6 -methylated product was obtained as the major product in 56% yield, with the N 1 -methylated isomer being formed as a byproduct in 25% yield. Due to their different Rf values, the N 6 -and N 1 -methylated products were easily separated via column chromatography, and the purified m 6 Am phosphoramidite was used directly for RNA synthesis.

All oligoribonucleotides were synthesized using the standard protocol for solid-phase RNA synthesis with a 15 min coupling time per nucleotide. The last coupling step was followed by a standard DMTroff procedure, the oligoribonucleotides were removed from the solid support, deprotected, precipitated and desalted. MALDI mass spectrometry was used to confirm the mass of all RNA oligonucleotides produced Thin layer chromatography of modified adenosines. Radioactively 5' end labelled oligonucleotides SK-526 (m 6 AmGGGCUGCU) and SK-524 (AmGGGCUGCU) were digested to release pm 6 Am and pAm then mixed with a combination of 5' radioactive labelled pm 6 A and unmodified nucleotide monophosphates for which relative mobilities are well characterised. This mixture was separated by 2D TLC using the solvent systems previously described 25 to establish the position of pm 6 Am (Fig. 2) .

To show that T4 polynucleotide kinase does not preferentially label m 6 Am or Am, the RNA oligonucleotides SK-524 and SK-526 were mixed in different ratios, end labelled, digested with P1 nuclease and separated by TLC. The spots corresponding to pm 6 Am and pAm were then quantified using phosphorimaging. The results demonstrated that both nucleotides are labelled with equal efficiency by T4 polynucleotide kinase (Fig. 3) Labelling and analysis of the first cap adjecent nucleotide. To label the first nucleotide following the m 7 G, poly(A) RNA was prepared from various mouse organs then digested with tobacco acid pyrophosphatase to remove the m 7 G cap. The exposed 5' ends were dephosphorylated with alkaline phosphatase and after phenol/chloroform extraction and ethanol precipitation, the mRNA transcripts were radiolabelled at their 5' end using T4 polynucleotide kinase in the presence of .20 fold excess of [c-32 P] ATP. The labelled RNA was digested to monophospho-nucleotides by P1 nuclease prior to TLC separation. Using this method, new spots corresponding to the 2'-O methylated nucleotides are apparent in the labelled samples after cap removal (Fig 4A) . A spot corresponding to pm 6 Am is readily detectable in all mRNA samples tested, and with pm 6 Am:pAm ratios of between 15:1 (brain) and 2:1 (liver) it appears that m 6 Am is more prevalent at the cap1 than is Am (Fig. 4A ). Under these labelling conditions, where both ATP and polynucleotide kinase are in excess, the intensity of the spots corresponding to the unmodified Analysis of cap structures for mRNA transcripts from individual genes. In order to assay modifications on the first transcribed nucleotide for messages from individual genes, mRNA from liver and testis was de-capped and end-labelled as described above then fragmented to ,120 nt. This was then hybridised to single stranded DNA targets corresponding to the 5' region of selected messages. These DNA targets were first cross-linked to 2 mm 3 2 mm teeth cut from a Hybond N 1 membrane. After hybridization and washing, the membrane was subjected to phosphorimaging (Fig. 4B) . The individual teeth containing the labelled mRNA gene-specific fragments were then removed and digested to nucleotide 5' monophosphates using P1 nuclease. These samples were individually spotted onto TLC plates and developed as described (Fig. 4B ). Four mRNAs were chosen for analysis, apolipoprotein A-I (Apoa1, BC012253), albumin (Alb, BC024643), protamine 2 (Prm2, BC049612) and poly(A) binding protein cytoplasmic 1 (Pabpc1, BC046233). Alb and Apoa1 are predominantly liver expressed messages, whereas Prm2 shows testis specific expression and Pabpc1 is expressed highly in both organs. Pabpc1 may be subjected to translational regulation, but unlike Pabpc2, it is present in actively translating polyribosomes of mouse testicular cells 31 . For both Alb and Apoa1, the labelled nucleotide in the cap adjacent position included A, G, U and C, as well as m 6 Am and Am. This is consistent with multiple alternative transcription start sites at and around the Inr 32 . In both cases, Am and m 6 Am appeared as the major nucleotide modifications; unlike the mixed mRNA starting material, 2'-O-methylcytosine was not present (compare Fig. 4A, B) . The m 6 Am:Am ratios were 1.65:1 and 1.4:1 for Apoa1 and Alb respectively, which is only slightly lower compared to that seen for the liver mRNA population as a whole. The Prm2 mRNAs from testis tissues almost exclusively had m 6 Am at the cap adjacent position (Fig. 4B) . With an m 6 Am:Am ratio of 40:1, this was four times larger than the m 6 Am:Am ratio for the testis mRNA population as a whole. In contrast the cap adjacent nucleotides of Pabpc1 transcripts were predominantly unmodified, although a pronounced 2'-O-methyluridine was apparent in the sample from testis (Fig. 4B ).

Core promoters of Pol II transcribed genes may be focused, directing transcription initiation at a single site or cluster of adjacent sites, or they may be dispersed and have multiple start sites over a region of 50-100 nucleotides 6 . In case of focused promoters the transcription start sites are contained within and around the Inr (YYANWYY), where the A is the principal start site and surrounding nucleotides are utilised to varying degrees 6 . Recent high-throughput sequencing from full length mRNAs, suggests that purine instead of A may be a better representation of the Inr consensus. However, most core promoters have several closely arrayed transcription start sites with different initiation rates rather than initiating at a single nucleotide 32 . Thus, most genes will give rise to a population of mRNAs that differ in their starting nucleotides. For example the Prm2 message initiates predominantly at the central A of the Inr as well as a second A 3 nucleotides upstream, whereas, Apoa1 strongly initiates at the A of the Inr but also uses surrounding U, G, C and sites at a significant level (Supplementary Figure S2 online) and (Fig. 4B) .

Conversion of an adenosine following the m 7 G cap to Am (cap1) can be followed by further conversion to m 6 Am (Fig. 1a, and 1c) . It is not known if this Am to m 6 Am conversion preferentially targets certain messages, or if the Am:m 6 Am ratio for messages from any one gene is merely a reflection of the general m 6 Am methylase activity of a particular cell type. The enzyme responsible for the formation of m 6 Am has been partially characterised 5 . It appears to be a cytosolic enzyme and is distinct from METTL3/MT-A70, the nuclear enzyme responsible for the formation of m 6 A at internal positions within mRNA 33 .

Whilst there is evidence that cap1 methylation may increase translation efficiency 18, 19 , the consequences of converting a cap1 Am to m 6 Am are largely unknown. Some crystallographic observations suggest that the first nucleotide after the m7G may influence interaction with eIF4E or other cap binding proteins. Crystallographic analysis of eIF4E in a complex with the dinucleotide m 7 GpppG, show the G in a disordered arrangement 34 . However, with m 7 GpppA as substrate, the A interacts with a C-terminal loop of the protein 35 . A key interaction of this C-loop is a hydrogen bond between Thr 205 and the amino group in the N 6 position of the adenosine. It is this amino group that is targeted for methylation by the m 6 Am methylase. Thus, N 6 methylation of a cap1 adenosine might influence the flexibility and orientation of the eIF4E C-terminal loop region. The ability to synthesize RNA oligonucleotides that containe m 6 Am in welldefined positions should facilitate future structure-function analysis. Therefore a direct synthetic method for the preparation of the desired m 6 Am phosphoramidite reagent, in a single step from commercially available starting material via direct N 6 -methylation, dramatically improves access to these materials for biological evaluation.

Preparation of 2'-OMe-Bz-m 6 A-CE phosphoramidite. 59-(4,49-Dimethoxytrityl)-6-N-benzoyl-6-N-methyl-adenosine,29-O-methyl-39-[(2-cyanoethyl)-(N,N-diisopropyl)]-phosphoramidite (2'-OMe-Bz-m 6 A-CE phosphoramidite) was prepared as follows: tetrabutylammonium bromide (260 mg, 807 mmol) and aq. NaOH (1M, 7 ml) were added to a stirring solution of 2'-OMe-Bz-A-CE phosphoramidite (Link Technologies, 708 mg, 797 mmol) and iodomethane (200 ml, 3.21 mmol) in dichloromethane (7 ml). After vigorous stirring for 30 min, ether and water were added and the resulting layers were separated. The aqueous layer was extracted with ether, the combined organic extracts were dried over Na 2 SO 4 and the volatiles were removed in vacuo. 2'-OMe-Bz-m 6 A-CE phosphoramidite was readily purified by column chromatography (1:1 petrol/etherR9:1 ether/methanol) affording 2'-OMe-Bz-m 6 A-CE phosphoramidite (403 mg, 56%, white foam, 1:1 mixture of diastereoisomers) and 2'-OMe-Bz-m 1 A-CE phosphoramidite (182 mg, 25%, white foam, 1:1 mixture of diastereoisomers). Data for 2'-OMe-Bz-m 6 A-CE phosphoramidite and for 2'-OMe-Bz-m 1 A-CE phosphoramidite are presented in Supplementary Table S1. online.

Synthesis of oligoribonucleotides SK-524 and SK-526. Oligoribonucleotides SK-524(AmGGGCUGCU) and SK-526(m 6 AmGGGCUGCU) were synthesized on an ABI 394 DNA/RNA synthesizer and mass spectra were recorded using a Bruker Ultraflex III mass spectrometer via MALDI-TOF. Columns (SynBaseTM CPG 1000Å , RNA: 0.2 mmol), apart from the 2'-OMe-Bz-m 6 A-CE phosphoramidite, standard RNA-phosphoramidites and reagents for the synthesizer were purchased from Link Technologies Ltd., MeNH 2 solution (33 wt.% in ethanol) was obtained from Fluka, NEt 3 N3HF, N-methylpyrrolidinone (NMP), 3-hydroxypicolinic acid (HPA) and DowexTM 50WX8-200 were purchased from Aldrich, illustra Nap TM -10 columns were obtained from GE Healthcare Europe GmbH. Dichloromethane and acetonitrile were freshly distilled from CaH 2 before use on the synthesizer.

The RNA oligonucleotides were synthesized using a standard 0.2 mM scale protocol, but with a 15 min coupling time for each nucleotide addition step. The polymer-bound oligoribonucleotide was transferred from the synthesis column to a 1.5 ml microfuge tube and suspended in MeNH 2 solution (1 ml). The mixture was heated to 65 uC for 10 min, cooled to room temperature (water/icebath) and centrifuged for 1 min (10,000 g). The supernatant was separated from the CPG beads, the beads were washed with RNase free water (2 3 0.25 ml), all supernatants were combined and dried (2 h under nitrogen stream, then freeze dried). The oligoribonucleotide was resuspended in anhydrous NEt 3 N3HF/NEt 3 /NMP solution (250 ml of a solution of 1.5 ml NMP, 750 mlNEt 3 and 1.0 ml NEt 3 N3HF), heated to 65 uC for 1.5 h, cooled to room temperature and quenched with 3M NaOAc solution (25 ml). n-BuOH (1 ml) was added to the mixture, which was then thoroughly mixed, cooled to 270 uC for 1 -2 h to encourage further precipitation and centrifuged for 30 min (4 uC, 13 000 g). The supernatant was removed, the pellet washed with 70% EtOH (2 3 500 ml) and then dried in vacuo (30 min). The dry precipitate was dissolved in RNase free water (1 ml) and desalted using a Nap TM -10 column following the standard protocol. The resulting solution was freeze dried over night leaving the oligoribonucleotide as a white foam/powder. Samples for MALDI-mass spectrometry were prepared as follows 36 : Dowex TM ion-exchange beads were rigorously cleaned with dilute HCl, washed with water, then treated with dilute NH 3 and finally washed with water again to generate Dowex-NH 4 1 . Diammonium citrate (DAC) (100 mg) was dissolved in water (1 ml) and HPA (34.8 mg) was dissolved in 1:1 acetonitrile/ water (1 ml). The HPA solution was filtered through Dowex-NH 4 1 and the DAC solution (100 ml) was added to prepare the matrix stock. Prior to MALDI-MS acquisition, matrix stock (20 ml) was mixed with Dowex-NH 4 1 (5 ml) and each oligoribonucleotide sample (1 ml) was mixed with Dowex-NH 4 1 (19 ml). After 30 min the matrix, and after drying, the sample solution (0.5ml) were spotted onto the sample well and allowed to dry prior to confirmatory analysis by MALDI-MS.

RNA purification. Total RNA was prepared from mouse (C57BL/6) tissues using Trizol reagent (Invitrogen). The poly(A) RNA was purified twice using oligo(dT) cellulose (Fluka) followed by oligo(dT) magnetic beads (Invitrogen). All samples were quantified and assayed for poly(A) purity using an Agilent Bioanalyser.

Analysis of gene-specific mRNA cap structures. Regions corresponding to the first 300 to 500 nucleotides of the target messages were PCR amplified from cDNAs using the following oligonucleotides:

Apoa1forward GCTCCGGGGAGGTCACCCACACCT and Apoa1reverse CAATGGGCCCAGCCGTTCCTGCAGC; Albforward CCCCACTAGCCTCTGGCAAAATGAAGTG and Albreverse GGCTGGGGTTGTCATCTTTGTGTTGCAG; Prm2forward GCTGGGTGTGCGCGAGTCAGGGGCTC and Prm2reverse CTTGTGGATCCTATGTAGCCTCTTACG; Pabpc1forward CGGCGGTTAGTGCTGAGAGTGCGGAG and Pabpc1reverse GAAGTTCACGTACGCGTAGCCCAAGG. Prior to amplification, the forward oligonucleotides were 5' phosphorylated, the amplification products were subsequently digested with lambda nuclease (New England BioLabs) to leave the single stranded antisense DNA strand. 100 ng of this ssDNA (2 ml) was spotted onto 2 mm 3 2 mm teeth cut from a Hybond N 1 membrane ( Amersham) (Fig. 2B,C) , and UV cross-linked (Stratalinker). Membranes were prehybridised at 42 uC in 40 % formamide with 5 3 Denhardt's, 3 % SDS, 0.3 M NaCl, 50 mM sodium phosphate buffer (pH 7.0) and 0.1 mg ml 21 sonicated salmon sperm DNA. 600 ng of poly(A) RNA was digested with 20 units of Tobacco Acid Pyrophosphatase (Epicentre) for 30 minutes at 37 uC. The 5' phosphate of the exposed cap adjacent nucleotide was removed by the addition of 10 units of Alkaline Phosphatase (Fermentas) and incubation for a further 15 minutes at 37 uC. After phenol-chloroform extraction and ethanol precipitation, RNA samples were resuspended in 20 ml of sterile distilled water and 5' ends were labelled using 30 units T4 polynucleotide kinase (PNK, Fermentas) and 7.4 MBq [c-32 P] ATP at 37 uC for 30 minutes. The PNK was heat inactivated (70 uC for 15 min) and the reaction made up to 60 ml with sterile distilled water then passed through a P-30 spin column (Bio-Rad) to remove unincorporated isotope. A 1 ml aliquot was taken, added to 9 ml of nuclease P1 buffer and digested with P1 (Sigma) for one hour at 37 uC. 1.5 ml of the released 5' monophosphates from this digest was then analysed by 2D TLC as described previously 25 . The remaining end labelled RNA was fragmented to lengths of approximately 120 nt by the addition of Na 2 CO 3 to a final concentration of 60 mM and NaHCO 3 to a final concentration of 40 mM followed by incubation for one hour at 60 uC.12 ml 3M sodium acetate pH 5.2 was then added and the RNA precipitated with ethanol. The pellet was resuspended in 200 ml of pre-hybridisation buffer then added to the membranes (final volume 2 ml) and hybridised overnight at 42 uC. Membrane washings were carried out with 2 3, 1 3 and 0.2 3 SSC, 0.1% SDS. Two final washes were carried out at 60 uC, with 0.2 3 SSC but with the SDS omitted. Hybridised membranes were exposed to storage phosphor screens (K-screen; KODAK) and imaged using Bio-Rad Molecular Imager FX in combination with Quantity One 4.6.3 software (Bio-Rad). Individual teeth containing the target mRNAs end labelled at the cap adjacent position were cut off and digested with P1 nuclease (Sigma) in a final volume of 3 ml. All of this digestion mix was applied to a cellulose coated TLC plate (20 3 20 cm, MERCK) and developed as described previously 25 .

Post-transcriptional modifications of mRNA. Purification and characterization of cap I and cap II RNA (nucleoside-2'-)-methyltransferases from HeLa cells

Characterization of hMTr1, a Human Cap1 2'-O-Ribose Methyltransferase

2'-O-ribose methylation of cap2 in human: function and evolution in a horizontal mobile family

N 6 ,O 2 -dimethyladenosine a novel ribonucleoside next to the 5'terminal of animal cell and virus mRNAs

HeLa cell RNA(2'-O-methyladenosine -N 6 -)-methyltransferase specific for the capped 5'-end of messenger RNA

The RNA polymerase II core promoter -the gateway to transcription

Purines 6. Rate study and mechanism of Dimroth rearrangement of 1-alkoxy-9-alkyladenines and 1-alkyl-9-methyladenines

Synthesis of 9-alkyl-2-deuterio-N6-methoxyadenines and 2-deuterio-N6,9-dimethyladenine -tautomerism in 9-substituted N6-alkoxyadenines

Molecular and crystal structures of dialkylated adenines

The synthesis of 2 '-O-[(triisopropylsilyl)oxy] methyl (TOM) phosphoramidites of methylated ribonucleosides m(1)G, m(2)G, m

A Substrate analogues for an RNA-editing adenosine deaminase: Mechanistic investigation and inhibitor design

Synthesis of 2 '-O-[(triisopropylsilyl)oxy]methyl (5tom)-protected ribonucleoside phosphoramidites containing various nucleobase analogues

Identification of recognition residues for ligation-based detection and quantitation of pseudouridine and N-6-methyladenosine

A nuclear cap binding protein complex involved in pre-mRNA splicing

Participation of the nuclear cap binding complex in pre-mRNA 30 processing

Origins and evolution of the mechanisms regulating translation initiation in eukaryotes

Influence of 5'-terminal m 7 G and 2'-O-methylated residues on messenger ribonucleic acid binding to ribosomes

Cap ribose methylation of c-mos mRNA stimulates translation and oocyte maturation in Xenopus laevis

Improve the translation efficiency of any 5'-capped mRNA with the new ScriptCap TM 2'-O-methyltransferase

In vitro reconstruction of SARS-coronavirus cap methylation

Recognition of capped RNA substrates by VP39, the vaccinia virus-encoded mRNA cap-specific 2 '-O-methyltransferase

2'-O methylation of the viral mRNA cap evades host restriction by IFIT family members

Ribose 29-O-methylation provides a molecular signature for the distinction of self and non-self mRNA dependent on the RNA sensor Mda5

5-Terminal cap structures in eukaryotic messenger ribonucleic acids

MTA is an Arabidopsis messenger RNA adenosine methylase and interacts with a homolog of a sex-specific splicing factor

5'-terminal and internal methylated nucleotide-sequences in Hela-cell messenger-RNA

Induction of sporulation in Saccharomyces cerevisiae leads to the formation of N 6 -methyladenosine in mRNA: a potential mechanism for the activity of the IME4 gene

Precise localization of m 6 A in Rous sarcoma virus RNA reveals clustering of methylation sites: implications for RNA processing

N 6 -methyladenosine in maize poly(A)-containing RNA

Alkylation of 6-N-acylated adenosine derivatives by the use of phase transfer catalysis

Characterization of two cytoplasmic poly(A)-binding proteins, PABPC1 and PABPC2, in mouse spermatogenic cells

A code for transcription initiation in mammalian genomes

Purification and cDNA cloning of the AdoMet-binding subunit of the human mRNA (N 6 -adenosine)-methyltransferase

Biophysical studies of eIF4Ecap-binding protein: recognition of mRNA 5' cap structure and synthetic fragments of eIF4G and 4E-BP1 proteins

Crystal structures of 7-methylguanosine 5'-triphosphate (m(7)GTP)-and P-1-7-methylguanosine-P-3-adenosine-5',5'-triphosphate (m7GpppA)-bound human full-length eukaryotic initiation factor 4E: biological importance of the C-terminal flexible region Biochem

Simplified sample preparation for the analysis of oligonucleotides by matrix-assisted laser desorption/ionisation time-of-flight mass spectrometry

Z. Bodi was supported by a BBSRC Daphne Jackson Fellowship and a Welcome Trust VIP award, J. Button was supported by a BBSRC Studentship.

SK and CH developed and carried out the novel chemical synthesis of phosphoramidites. SZ, JB and RF developed the cap labeling protocols. ZB and RF wrote the manuscript. MA provided materials and useful comments. All authors reviewed the manuscript.