Next Article in Journal / Special Issue
Cross-Talk between Dnmt2-Dependent tRNA Methylation and Queuosine Modification
Previous Article in Journal / Special Issue
The Escherichia coli COG1738 Member YhhQ Is Involved in 7-Cyanodeazaguanine (preQ0) Transport
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Biomolecules
Volume 7
Issue 1
10.3390/biom7010013
biomolecules-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Next‐Generation Sequencing‐Based RiboMethSeq Protocol for Analysis of tRNA 2′‐O‐Methylation

by
Virginie Marchand
1,2,
Florian Pichot
1,
Kathrin Thüring
3,
Lilia Ayadi
1,
Isabel Freund
4,
Alexander Dalpke
4,
Mark Helm
3 and
Yuri Motorin
1,2,*
1
IMoPA UMR7365 CNRS‐UL, Biopole Lorraine University, 54505 Vandoeuvre‐les‐Nancy, France
2
Next‐Generation Sequencing core facility, FR3209 BMCT CNRS‐UL, Biopole Lorraine University, 54505 Vandoeuvre‐les‐Nancy, France
3
Institute of Pharmacy and Biochemistry, Johannes Gutenberg University Mainz, 55128 Mainz, Germany
4
Department of Infectious Diseases, Medical Microbiology and Hygiene, Ruprecht‐Karls University Heidelberg, 69120 Heidelberg, Germany
*
Author to whom correspondence should be addressed.
Biomolecules 2017, 7(1), 13; https://doi.org/10.3390/biom7010013
Submission received: 29 December 2016 / Accepted: 2 February 2017 / Published: 9 February 2017
(This article belongs to the Collection RNA Modifications)
Browse Figures
Versions Notes

Abstract

:
Analysis of RNA modifications by traditional physico-chemical approaches is labor intensive, requires substantial amounts of input material and only allows site-by-site measurements. The recent development of qualitative and quantitative approaches based on next-generation sequencing (NGS) opens new perspectives for the analysis of various cellular RNA species. The Illumina sequencing-based RiboMethSeq protocol was initially developed and successfully applied for mapping of ribosomal RNA (rRNA) 2′-O-methylations. This method also gives excellent results in the quantitative analysis of rRNA modifications in different species and under varying growth conditions. However, until now, RiboMethSeq was only employed for rRNA, and the whole sequencing and analysis pipeline was only adapted to this long and rather conserved RNA species. A deep understanding of RNA modification functions requires large and global analysis datasets for other important RNA species, namely for transfer RNAs (tRNAs), which are well known to contain a great variety of functionally-important modified residues. Here, we evaluated the application of the RiboMethSeq protocol for the analysis of tRNA 2′-O-methylation in Escherichia coli and in Saccharomyces cerevisiae. After a careful optimization of the bioinformatic pipeline, RiboMethSeq proved to be suitable for relative quantification of methylation rates for known modified positions in different tRNA species.

1. Introduction

RNA modification is a complex step in the post-transcriptional maturation of cellular RNAs and includes numerous chemical alterations of initially incorporated parental nucleotides. Among these various chemical reactions, the most frequent is the transfer of a methyl (CH3) group from a methylation donor (generally S-adenosyl-l-methionine, abbreviated as SAM or AdoMet) to different positions on nucleobases or to the 2′-OH of the ribose [1,2,3,4,5]. Methylated nucleotides are found in every studied RNA species, and their presence allows fine modulation (tuning) of RNA properties in different cellular processes [6,7,8,9]. One of the best-known RNA species is the transfer RNAs (tRNAs), whose properties and functions have now been studied for over 50 years. These relatively small RNA molecules have been extensively analyzed for their RNA modification profile, and a wealth of information is now available for tRNA from bacteria, lower eukaryotes (mostly yeasts and fungi) and mammals (human, rat, rabbit, and beef) [10]. This already served as a basis for extensive search of the corresponding enzymes and enzymatic machineries involved in the modification of tRNA species. Thanks to these efforts, most of the enzymes for tRNA modification in model organisms were identified and their specificity characterized [11,12,13]. It is noteworthy that archaeal species somehow escaped this thorough analysis, and only a few tRNA species are included in available databases. Archaeal enzymes are mostly characterized by sequence homology followed by functional tests to confirm the initial hypothesis on their predicted activity, as archaeal tRNA methyltransferase (aTrm56) [14,15].
One very common methylation in many RNA species is 2′-O-ribose methylation (2′-O-Me). This type of methylations may occur in all four canonical nucleotides in RNA and even many base-modified residues, the most frequent probably being 2′-O-methylated pseudouridine (ψm).

1.1. Bacterial tRNA 2′-O-methylation

Early on, the analysis of tRNA species from the model bacterial organism Escherichia coli demonstrated the presence of three possible positions for 2′-O-methylation (Table 1). A major site is located in the D-loop at position 18, where the highly conserved G residue is frequently converted to 2’-O-methylated guanine (Gm), while two others are located in the anticodon loop at positions 32 and 34 (wobble) where Cm and Um/cmnm5Um residues are frequently present (Figure 1). E. coli tRNA has no other reported sites, while other bacterial species, like Bacillus subtilis and Mycoplasma capricolum, seem to lack Gm18, but may contain Gm34 (B. subtilis) or cmnm5Um34 (M. capricolum) at the wobble position of the anticodon.
The formation of Gm18 in bacterial tRNA is catalyzed by specific tRNA:Gm18-methyltransferases (MTases) from the SpoU (now renamed to TrmH) family, which had already been predicted to be an RNA:MTase [16,17,18], prior to experimental validation. This enzyme was found in the gmk-rpoZ-spoT-spoU-recG operon in E. coli [19] and was also conserved in related bacterial species and in the extreme thermophile Thermus thermophilus [20].
Recognition of tRNA substrates by TrmH was extensively investigated and the enzymatic mechanism elucidated by a structure-based site-directed mutagenesis study [21]. The TrmH protein from the thermophilic bacterium Aquifex aeolicus has been crystallized and its structure studied at high resolution [22].
Two other important sites of 2′-O-methylation in E. coli tRNA are located in the anticodon loop, at positions 32 and 34. The formation of Cm32 and Um32 tRNASer1 and tRNAGln2 is catalyzed by YfhQ (renamed now to TrmJ or TrMet(Xm32)) [23], which belongs to the SPOUT family, in contrast to eukaryotic TRM7 from the RFM group (see below) [24].
Finally, the enzyme encoded by yibK, named TrmL, is a part of a more complex pathway [25] designed for the biosynthesis of a complex, multistep modification of the nucleotide at wobble position 34 in two leucyl isoacceptor tRNAs. Depending on the tRNA species, TrmL catalyzes the formation of Cm34 or Um34 in tRNALeu(CmAA) and tRNALeu(cmnm5UmAA) [26]. TrmL methylates pyrimidines, but not purine residues at the wobble position, and the 2′-O-methylation depends on prior N6-isopentenyladenosine modification at position 37.
In other bacteria, enzymatic tRNA modification systems targeting the ribose-2′-OH are only partially characterized. Recently, a tRNA MTase TrmJ in Pseudomonas aeruginosa PA14 was found to be involved in resistance to oxidative stress. TrmJ catalyzes the formation of Cm34, Um34 and, in contrast to E. coli, an additional Am34 in tRNAPro(GGG) [27].

1.2. Archaeal tRNA 2′-O-Methylation

Archaeal tRNA 2′-O-methylation is only partially characterized up to now. In contrast to Bacteria, where only proteinaceous enzymes modify tRNAs, archaeal organisms use both stand-alone protein enzymes and C/D-box small nucleolar RNA (C/D snoRNA) guided machinery (Figure 1 and Table 1), sometimes with overlapping substrate and site specificity [28].
Methylation of the universally-conserved methylated position at position 32 in the tRNA anticodon in Archaea Sulfolobus acidocaldarius is catalyzed by a homolog of the bacterial TrmJ. However, while all four canonical nucleosides are substrates of the E. coli enzyme, the archaeal TrmJ can only methylate the ribose of a cytidine [24].
In contrast to Eubacteria, the methylation of the wobble residue C34 of Haloferax volcanii elongator tRNAMet is guided by a box C/D RNA targeting the tRNA intron-containing precursor [29]. 2′-O-methylation of both nucleotides C34 and U39 in H. volcanii pre-tRNATrp also requires the box C/D ribonucleoprotein (RNP) machinery. Methylation occurs sequentially with the C/D-box small RNA (sRNA) guide in trans, which would also eliminate the obligatory refolding of the pre-tRNA before splicing [30]. Most euryarchaeal tRNATrp genes require a box C/D guide RNA within their intron for specifying methylation at these two sites [31].
The archaeal-specific modification Cm56 is formed by the SPOUT S-adenosylmethionine (AdoMet)-dependent MTase aTrm56, which is found only in Archaea [32,33]. aTrm56 is present in almost all of the archaeal genomes sequenced up to now, except in the crenarchaeon Pyrobaculum aerophilum in which the tRNA Cm56 2′-O-methylation is provided by a C/D sRNP [31].

1.3. Eukaryotic tRNA 2′-O-Methylation

In yeast and human, tRNAs 2′-O-methylations occurs at five rather conserved positions and two minor sites, the best conserved locations being Nm4, Gm18, Cm32/Nm34 and Um44 (Table 1).
The formation of the highly conserved Gm18 in Saccharomyces cerevisiae tRNAs is catalyzed by the enzyme encoded by ORF YDL112w (now renamed “Trm3”). Genetic depletion of YDL112w resulted in the complete absence of Gm18 in all tRNAs that naturally contain this modification. Other tRNA ribose methylations and the complex pattern of ribosomal RNA (rRNA) ribose methylations were not affected [34].
Anticodon 2′-O-modifications both at positions 32 (Cm32) and 34 (Nm34) in yeast tRNAs require a unique MTase Trm7 (encoded by YBR061C) [35], which also tightly interacts with ORF YMR259c (now named Trm732) for 2′-O-methylation of C32 and with Rtt10 (named Trm734) for 2′-O-methylation of N34 in substrate tRNAs [36]. Moreover, the presence of Trm7-catalyzed modifications in the tRNA anticodon loop is required for efficient modification of m1G at position 37, in the biosynthesis pathway leading to yW (wybutosine) [36,37]. Activities towards Cm32 and Nm34 seem to be independent, since a specific mutation in Trm7 selectively abolishes the formation of Gm34 only [38].
2′-O-methylated Um44 is frequent among eukaryotic cytoplasmic tRNAs bearing a long variable loop. Yeast ORF YPL030w (now renamed to Trm44) was identified by screening for the Um44:MTase activity of yeast genomic library of affinity-purified proteins. Consistent with the conservation of Um44 in eukaryotic tRNAs, Trm44 is found in metazoans and fungi, but surprisingly absent in plants [39].
The last 2′-O-methylation site in yeast tRNAs was found at position 4, and it is the only 2′-O-methylation that occurs in the middle of a duplex region in tRNA. It was shown that S. cerevisiae ORF YOL125w (renamed to Trm13) is required and sufficient for this modification [40].
The tRNA 2′-O-methylation profile in vertebrates is somehow similar to the one observed for lower eukaryotes, like yeasts, with conserved sites at position Cm/Um4, Gm18, Cm/Um/ψm32, Cm/Gm34 and Um44. One notable exception is the presence of Tm(m5Um)54 in the Tψ-loop; this nucleotide occasionally replaces the universally conserved T54 [41]. Another minor site is ψm39 (Figure 1 and Table 1).
In addition to common simple 2′-O-methylated residues, tRNAs from higher eukaryotes contain other more complex modified residues, where base modification is coupled with ribose methylation. For example, 2′-O-methyl-5-formyl modification of cytidine (f5Cm) was found at position 34 in bovine tRNALeu and confirmed for rabbit and lamb. Thus, f5Cm could be a general feature of cytoplasmic tRNAsLeu in mammals [42].
Studies of higher eukaryotic RNA 2′-O-methylation enzymes have a long-standing history [43]. Already in the 1980s, the enzymatic activity acting at Gm34 in tRNAPhe was detected using microinjection into Xenopus laevis oocytes [44]. However, systematic identification of the corresponding enzymes has started only recently.
The only higher eukaryotic tRNA-specific 2′-O-MTase characterized up to now is a human homolog of yeast Trm7. Mutations in human FTSJ1, the Trm7 homolog, are present in two genetically-independent cell lines of patients and cause non-syndromic X-linked intellectual disability (NSXLID). Cells with loss-of-function FTSJ1 mutations nearly completely lack Cm32, as well as Gm34; moreover, an NSXLID patient with a novel FTSJ1-p.A26P missense allele specifically lacked Gm34, while levels of Cm32 and o2yW37 remained normal [38].
A homology search using yeast tRNA:MTases as probes revealed the presence of predicted homologs for all four known yeast tRNA:2′-O-MTases [5] (see Table 1); however, the activity of most of these proteins was not experimentally confirmed up to now.

1.4. Functions of tRNA 2′-O-Methylation in tRNA Stability and Immunostimulation

Ribose 2′-O-methylation has been known for a long time to affect the stability and physico-chemical properties of the RNA chain. 2′-O-methylation of pyrimidines stabilizes the C3′-endo sugar conformation (and thus, the A-type RNA helix) primarily due to the steric repulsion among the 2-carbonyl group, the 2′-O-methyl group and the 3′-phosphate group in the C2′-endo form [45].
Archaea-specific hypermodified residues N4-acetyl-2′-O-methylcytidine (ac4Cm), 5-methyl-2′-O-methylcytidine (m5Cm) and N2-dimethyl-2′-O-methylguanosine (m22Gm) were found to stabilize the C3′-endo form and therefore cause “conformational rigidity” as attested by proton Nuclear Magnetic Resonance spectroscopy. In particular, the ac4Cm was found to be extremely rigid due to additive effects of the N4-acetylation and 2′-O-methylation [46].
In addition to purely structural roles and general protection of RNA molecules from alkaline hydrolysis, tRNA 2′-O-methylation has an important functional role in fine-tuning of bacterial co-habitations with human (and other mammalian) organisms. It was noticed a long time ago that preparations of Salmonella typhimurium tRNA and rRNA provide RNA-dependent immunogenic activity [47]. More recently, it was clearly demonstrated that immunostimulation by tRNA is inhibited by tRNA:Gm18-MTase activity and even by the single methyl group on the 2′-oxygen of Gm18 as a natural modification in native tRNA [48,49]. This might explain differential immunostimulatory effects of different bacterial strains, and it was discussed that tRNA Gm18 modification might be used as an immune evasion mechanism. Of note, Gm18-modified tRNA acted as a TLR7 antagonist and blocked activation by RNA from plasmacytoid dendritic cells [47,48].
In addition, Gm18 also potently inhibited TLR7-independent human monocyte activation by RNA derived from a variety of bacterial strains. Antagonizing effects of Gm18-modified RNA are due to competition with stimulatory RNA for receptor binding. The effect of immunostimulation/suppression mediated by Gm18 in E. coli tRNA was observed for human and murine innate immune receptors TLR7/8, whereas the mouse-specific receptor TLR13 was insensitive to tRNA Gm18 [50], but inhibited by N6-methylation of rRNA (A2085 in Staphylococcus aureus/A2058 E. coli) [51].

1.5. Analytical Methods for Detection of 2′-O-Methylation in RNA

Despite the growing interest in the detection and quantification of 2′-O-methylated residues, no rapid and high-throughput analytical technologies have been applied to tRNAs, mostly due to their small size, very stable secondary and tertiary structure and an important proportion of other modified RNA residues. Reverse-transcription (RT) at low desoxynucleotide triphosphates (dNTP) concentrations is not very suitable for tRNAs, since natural RT-pauses are frequent, and RT-arresting modified residues prevent the generation of long primer extension products [52,53]. High-pressure liquid chromatography (HPLC) and HPLC coupled with mass spectrometry (MS)-related techniques require purification of tRNA species to near homogeneity, thus limiting their application to rather global analysis of the RNA modification profile [54,55,56]. Other approaches were also proposed, but did not become popular in the field [57].
Recently, high-throughput techniques have been developed for the analysis of a small subset of RNA modifications, namely m6A, m1A, pseudouridine and m5C [58,59,60,61,62]. These techniques allow transcriptome-wide screening, but the quantification of modification is uncertain and, when possible, requires complex calibration procedures. Only one type of modification in RNA can be detected and precisely quantified by next-generation sequencing (NGS), due to the induced resistance of the phosphodiester bond to cleavage in the presence of a neighboring 2′-O-Me group [63,64,65]. Thus, the measurement of protection against nucleolytic cleavage provides an excellent and quantitative way to measure 2′-O-modification.
The aim of the present study was to adapt the previously-published RiboMethSeq protocol [65] to 2′-O-methylation analysis of tRNAs. The main difficulties are related to the short length of most tRNA species, their stable 2D and 3D RNA structures, the existence of very similar, closely-related tRNA isoacceptors and the presence of numerous other modified nucleotides affecting both the efficiency of RNA cleavage, as well as primer extension by RT.

2. Results and Discussion

We first evaluated the results from the application of the standard RiboMethSeq protocol (previously adapted to rRNA) for comparatively short tRNA sequences from E. coli and S. cerevisiae. The RiboMethSeq protocol [65] is based on high-throughput analysis of an alkaline RNA digest and is able to precisely map resistant phosphodiester bonds, with such sites mostly corresponding to natural 2′-O-methylated 5′-adjacent riboses. However, spontaneous structure- or sequence-dependent resistance to alkaline hydrolysis may generate false-positive hits in the analysis.
One essential step of the RiboMethSeq protocol consists of the generation of a suitable library of RNA fragments after alkaline treatment. Short and structured RNA species (like tRNAs) are comparatively resistant to alkaline hydrolysis, and their fragmentation conditions have to be optimized to get sufficient amounts of short 20–40-nt fragments. Preliminary tests indicated that hydrolysis times of 6–10 min were appropriate for E. coli tRNAs, while a somewhat longer treatment of 8–12 min was required for the S. cerevisiae tRNA fraction (Figure 2). These tRNA fragments were converted to libraries and subjected to single-read Illumina sequencing for 50 nt.
With standard treatment parameters, the RiboMethSeq procedure uses reads of a minimum of 17 nt in length, which are aligned to the reference sequence with the end-to-end option. Only 5′-end extremities are used for analysis with calculation interval of six neighboring nucleotides. The application of these parameters for tRNA 2′-methylation analysis demonstrated, in principle, the detection of modified nucleotides, but also showed the necessity for a careful optimization of all treatment steps of the analysis pipeline.
The necessity for optimization is related to short and quite similar tRNA genes (transfer DNA, tDNA) sequences used for alignment, to the very irregular alkaline cleavage profile of these small and strongly-structured RNA molecules and certainly to the presence of numerous other modified nucleotides with RT-arresting properties, decreasing even further the number of available sequencing reads at certain tDNA regions.
Taking into account all of these difficulties, we proceeded with the optimization of all treatment steps (Figure 3), the most important concerning the used read length, selection of only uniquely aligned reads, simultaneous usage of 5′- and 3′-end information, as well as the calculation of scores for shorter neighboring regions and, finally, a careful optimization of the reference tDNA dataset. All of these optimization steps are discussed below.

2.1. Construction of the Reference tDNA Dataset

Initial tDNA datasets for alignment were constructed from sequences of modified tRNA species deposited in tRNA database (tRNAdb) [66], converted to DNA sequences in Fasta format. Inspection of the sequencing reads alignment to this tDNA reference revealed that, even at a minimum length of 17 nt, a significant number of reads mapped to multiple locations in tRNAs (Figure 4). This was mostly due to almost identical tDNA species present in the dataset and also to the presence of highly repetitive regions. In order to obtain better certitude in read mapping, we first decided to reduce and simplify the tDNA reference datasets for analysis.
In order to identify closely-related tDNA sequences, the Levenshtein distance was calculated for the complete tDNA dataset for known modified tRNA species, obtained from tRNAdb.
Distances between tDNA sequences are given in Figure S1; the calculation was performed for the dataset before and after collapsing of closely-related sequences. tDNA sequences with Levenshtein distance <4 (corresponds to a maximum of three nucleotide replacements) were collapsed into one unique sequence, and an undefined nucleotide “N” was placed at the divergent positions. Other closely-related pairs were retained in the dataset. At the same time, observed sequencing non-conformities and errors were corrected, and instances of the A-derived, but C-base pairing modified nucleotide inosine (I) present in some tRNA species were replaced by a G in the reference tDNA dataset. The same strategy was applied to E. coli and S. cerevisiae tDNA datasets. Details of the adaptation of both tRNA datasets are given in Tables S1 and S2.

2.2. Alignment and Analysis

For alignment to optimize the tDNA dataset, raw read trimming was performed to keep all sequences >8 nt. Of note, this length, while unsuitable for large genomic alignments, is still appropriate for a limited size tDNA “genome”. In order to gain additional information on both the 5′- and the 3′-end positions for every aligned read, only trimmed (with adapter removed) sequences were retained for subsequent alignment. This reduced the number of usable reads, but still provided some gain in information, since both 5′- and 3′-end locations were used.
Bowtie2 alignment parameters were adjusted for short input sequences (L = 10) in end-to-end mode. In order to avoid contamination by ambiguously (multiply) mapped sequencing reads, only uniquely-aligned sequences were taken for further analysis. Details for the number of retained reads at every step are given in Table S3. The positions of the 5′- and 3′-ends of every read were extracted after conversion of the aligned *.sam file to bed format with bedtools.
The positions of the 5′- and 3′-ends were cumulated together to obtain the global end-coverage profile, and these data were used to calculate RiboMethSeq scores (ScoreMean (derived from ScoreMax) and ScoresA, B and MethScore(C)) as described previously [63,65]. Since the length of tRNA sequences is rather limited and the cleavage profile of tRNAs is highly irregular, only two neighboring nucleotides on each side were taken into account (instead of six in the standard RiboMethSeq approach).
As a result of this optimization, over 95% of trimmed reads were unambiguously mapped to tDNA reference sequences, and a better homogeneity of the coverage was obtained. All tRNAs (except one species in E. coli and one in S. cerevisiae) received sufficient coverage, and further analysis of 2′-O-methylation could be performed.

2.3. Profiles for tRNAs from WT and ΔTrmH E. coli Mutant Strains

Analysis of E. coli tRNA 2′-O-methylation was performed in parallel for the wild-type (WT) strain and for a strain with the deletion of the nonessential trmH gene responsible for Gm18 formation in bacterial tRNAs. E. coli tRNAs were reported to contain 2′-O-methylated residues at three different locations (Gm18, Cm32 and Cm/Um/cmnm5Um34; see the Introduction). As expected, RiboMethSeq efficiently detected alkali-resistant positions, which mostly correspond to 2′-O-methylations, to other modified RNA residues affecting cleavage or phosphorylation/ligation efficiency for the adapter or to naturally insensitive non-modified residues. The present goal was to provide a set of potentially modified positions to be verified by independent approaches and also to gain relative quantification of the methylation level for known 2′-O-methylated nucleotides.
Representative cleavage profiles for the selection of well-detectable 2′-O-methylated sites are shown in Figure 5, and the corresponding scores are listed in Table 2. All previously reported 2′-O-methylated sites gave rather high values of ScoreMean, while ScoreA [63] varied considerably. The comparison of both scores obtained for WT and ΔTrmH strains demonstrated that the value of the score considerably decreased for all TrmH-dependent 2′-O-methylations (Gm18), but remained very stable for all 2′-O-methylated residues at positions 32 and 34, catalyzed by TrmJ and TrmL (Figure 1 and Table 1). Inspection of the corresponding cleavage profiles confirmed that cleavage efficiency at all TrmH-catalyzed positions drastically increased in the ΔTrmH strain. This clearly demonstrated that tRNA 2′-O-methylation can be detected and even precisely quantified using RiboMethSeq, even when the RNA cleavage profile is highly irregular. However, multiple false-positive hits were also detected in various tRNAs (Figure S2), some of them in proximity to highly accessible tRNA nucleotides or to some of the numerous modified residues in tRNA, namely pseudouridine or m7G46.
E. coli tRNAMet(ac4CAU), tdbR00000276 (highlighted in grey see Table 2), was previously reported to contain at least partial Gm18 methylation. Analysis of the RiboMethSeq results for this tRNA demonstrated that rather low scores were obtained for potential Gm18, and in addition, no difference was detected between WT and ΔTrmH strains. In order to check the real existence of Gm modification in this tRNA, it was purified to near homogeneity by hybridization with biotinylated DNA oligonucleotide and subjected to nucleotide analysis by HPLC-MS/MS (Figure S3a,b). As clearly seen in the figure, only trace amounts of Gm were detected, confirming the absence of Gm18 in E. coli tRNAMet(ac4CAU). Following up on our previous studies on the immunostimulatory effect of a Gm18 modification in E. coli tRNA [48], we tested the native tRNA for its effect on TLR7-mediated stimulation of interferon emission from peripheral blood mononuclear cells (PBMCs). Indeed, the native tRNA induced the substantial release of α-interferon, which would be in contrast to the possible presence of Gm18, but is in excellent agreement with its absence (Figure S3c). These findings illustrate that the mapping of ribose methylated residues in tRNA is accurate to a point that reports from the literature can be efficiently corrected.

2.4. Analysis of 2′-O-Methylation in WT and ΔTRM3 Yeast Mutant Strains

The optimized treatment pipeline described above was also applied for the analysis of alkaline hydrolysis-resistant positions in yeast total tRNAs extracted from WT and ΔTrm3-strains. All 19 known positions of 2′-O-Me in yeast gave strong and consistent values of ScoreMean (median value 0.93, standard deviation (SD) 0.077), as well as MethScore (median 0.83, SD 0.15). All scores for Gm18 demonstrated a considerable decrease (average ScoreMean −0.037 and MethScore −0.1) in the ΔTrm3 yeast strain compared to the WT cells. In contrast, the calculated values of the scores for other known 2′-O-methylated positions (Nm5-6, Nm32–34, Nm45) were not at all affected in the ΔTrm3 strain (see Table 2). Representative cleavage profiles are given in Figure 6 and false-positive hits in Figure S4.

2.5. Limitations

The results obtained in this study demonstrate that RiboMethSeq can be successfully applied not only to rRNA sequences in different species, but also to tRNAs, including also relative quantification of 2′-O-methylation levels in this type of RNA. However, rather important sequencing coverage is required for reliable analysis, and the cleavage profiles for tRNAs remain extremely irregular and, thus, generate an important number of false-positive hits. Some of them correspond to other modified tRNA nucleotides, and independent approaches are required to distinguish real RNA modification and phosphodiester bonds that are alkaline resistant for other reasons, such as three-dimensional structure. The ScoreMean parameter provides rather good sensitivity (all known 2′-O-methylated positions have a high ScoreMean value), but poor selectivity, while the opposite is true for ScoreA. Other proposed scoring schemes are even less discriminative and were not used here.
The use of cell lines or strains devoid of selected RNA-modification enzymes provided considerable support in the distinction between truly ribose-methylated positions from false-positive hits, since only signals corresponding to real modified nucleotides are affected, while signals for false-positives and other unaffected modified nucleotides remain generally stable. For a more precise quantification of the RNA modification level, the calibration of RiboMethSeq signals with unmodified RNA species extracted from the deleted or knockout strain is strongly recommended.

3. Materials and Methods

3.1. tRNA Sources

The S. cerevisiae parental strain (Y00000, strain BY4741; S288C isogenic yeast strain: MATa; his3Δ1; leu2Δ0; met15Δ0; ura3Δ0) and MTase mutant (Y03809, MATa; ura3Δ0; leu2Δ0; his3Δ1; met15Δ0; YDL112w::kanMX4) deficient for Trm3 (Gm18-2′-O-MTase) were obtained from Euroscarf, Johann Wolfgang Goethe-University Frankfurt and grown at 30 °C in yeast extract-peptone-dextrose (YPD) medium (Carl Roth, Karlsruhe, Germany) to mid-log phase.
The E. coli parental strain (BW25113; CGSC#7636; F-, Δ(araD-araB)567, ΔlacZ4787(::rrnB-3), λ, rph-1, Δ(rhaD-rhaB)568, hsdR514) and trmH (Gm18-2′-O-MTase) deficient mutant (JW3626-1; CGSC#11693; F-, Δ(araD-araB)567, ΔlacZ4787(::rrnB-3), λ-, rph-1, ΔtrmH755::kan, Δ(rhaD-rhaB)568, hsdR514) were purchased from E. coli Genetic Stock Center CGSC, Yale University, and grown to mid-log phase at 37 °C in Luria-Bertani (LB) medium (AppliChem, Darmstadt, Germany). Knockout strains were grown with 200 μg/mL geneticin (G418) (Carl Roth) and 50 μg/mL kanamycin (AppliChem), respectively.

3.2. Isolation of tRNA

For optimal amounts of RNA, bacteria and yeast were pelleted and frozen at −80 °C overnight and thereafter lysed in 40 mg/mL lysozyme (Serva, Heidelberg, Germany) or 50 U/mL lyticase (Sigma Aldrich, St. Louis, MO, USA), respectively. Total yeast and bacterial RNA were isolated using TRIzol reagent (Thermo Fisher Scientific, Waltham, MA, USA) according to the manufacturer’s instructions. Total RNA preparations were separated by denaturating PAGE (10% acrylamide, 8 M urea); tRNA-containing bands were cut out and shaken in 0.5 M sodium acetate overnight. tRNA was precipitated with ice cold ethanol and resuspended in nuclease-free water. The concentration of RNA was controlled via UV absorbance.

3.3. tRNA Fragmentation and Analysis

tRNA (about 100 ng) hydrolysis was performed in 50 mM bicarbonate buffer pH 9.2 for different times (6–10 min for E. coli or 8–12 min for S. cerevisiae) at 95 °C. The reaction was stopped by ethanol precipitation using 3 M Na-acetate, pH 5.2 and glycoblue as a carrier in liquid nitrogen. After centrifugation, the pellet was washed with 80% ethanol and resuspended in nuclease-free water. The sizes of generated RNA fragments were assessed by capillary electrophoresis using a Small RNA chip on the Bioanalyzer 2100 (Agilent).

3.4. Library Preparation

RNA fragments without any gel-purification step were directly 3′-end dephosphorylated using 5 U of Antarctic phosphatase (New England Biolabs, Ipswich, MA, USA) for 30 min at 37 °C. After the inactivation of the phosphatase for 5 min at 70 °C, RNA fragments were phosphorylated at the 5′-end using T4 polynucleotide kinase (PNK) and 1 mM ATP for 1 h at 37 °C. End-repaired RNA fragments were then purified using the RNeasy MinElute Cleanup kit (QIAGEN, Hilden, Germany) according to the manufacturer’s recommendations, except that the volume of 96% ethanol was adjusted for RNA binding. Elution was performed in 10 µL of nuclease-free water. RNA fragments were converted to the library using the NEBNext Small RNA Library kit (ref#E7330S, New England Biolabs; or equivalent from Illumina, San Diego, CA, USA) following the manufacturer’s instructions. DNA library quality was assessed using a High Sensitivity DNA chip on a Bioanalyzer 2100. Library quantification was done using a fluorometer (Qubit 2.0 fluorometer, Invitrogen, Waltham, MA, USA).
The library mix for sequencing was adjusted to obtain ~10–15 millions of raw reads for each library; this number was found to be sufficient for a 5000–8000 nt length reference sequence (end-coverage ~1000×).

3.5. Sequencing

Library sequencing was performed using the Illumina HiSeq 1000 sequencer in a single read mode for 50 nt with SBS kit v3. Primary analysis of sequencing quality was done with Illumina RTA 2.12 software, to insure a >Q30 quality score for >95% of the obtained sequences. Following the SR50 sequencing run, demultiplexing was done with Illumina BclToFastq v2.4; reads not passing the quality filter were removed.

3.6. Data Analysis

Raw reads after demultiplexing were trimmed to remove the sequence of the Illumina 3′ adapter. Trimming was performed with Trimmomatic v0.32 [67] with the following parameters: -phred33 /adapters/TruSeq3-SE.fa:2:30:10 LEADING:30 TRAILING:30 SLIDINGWINDOW:4:15 MINLEN:8 AVGQUAL:30. Tests with a training dataset demonstrated that with stringency parameter = 10, Trimmomatic removes the adapter with a minimal size of 16 nts. Thus, only trimmed and adapter-free reads of <35 nt were taken for alignment. Alignment to the reference tDNA sequences was performed with Bowtie2 ver2.2.4 [68] in end-to-end mode using the following parameter set: --no-1mm-upfront -D 15 -R 2 -N 0 -L 10 -i S,1,1.15. Preliminary tests demonstrated that the use of soft trimming during alignment (-- local settings) is not recommended, since the 5′- and 3-end may be incorrectly determined. Selection of only uniquely-mapped reads (single reported alignment position in the reference sequence) in the resulting *.sam file was done using values of the NM and XS fields (grep -E “@|NM:” *.sam|grep -v “XS:”). Uniquely-mapped reads were extracted from the *.sam file by RNA ID and converted to *.bed format using bedtools v2.25.0 [69]. Positional counting of the 5′-and 3′-ends of each read was performed with the awk Unix command. Further treatment steps were done in the open-source R environment (v3.0.1) [70]. In brief, 5′-end and 3′-end counts were merged together by RNA position and used for the calculation of ScoreMean (derived from the ScoreMax described previously), as well as Scores A and B [63] and MethScore [65]. Scores were calculated for 2 neighboring nucleotides instead of 6 in the standard RiboMethSeq procedure. Profiles of RNA cleavage at selected (candidate and previously known) positions were extracted and visually inspected.

4. Conclusions

The alkaline hydrolysis-based RiboMethSeq method (Figure 3) can be successfully applied to the analysis of known tRNA 2′-O-methylations and their dynamics or variations under different conditions. Quantification of the modification level is possible with careful calibration, obtained either with unmodified tRNAs extracted from a relevant KO strain or, if not available, with a synthetic RNA transcript with an identical sequence. The candidates for novel modified positions revealed by this high-throughput technique should be further validated by orthogonal approaches, either by physico-chemical methods, like HPLC-MS/MS analysis or by primer extension at low dNTP concentrations, an approach specific for RNA 2′-O-methylations. In the case of Gm18 in E. coli tRNAMet(ac4CAU), where we have exercised such a validation, the RiboMethSeq method proved accurate and led to the correction of the data from the literature. As this example has demonstrated, RiboMethSeq may be used for the correct prediction of the immunosilencing properties of RNA molecules based on the functional effect of ribose methylation on the recognition by components of the innate immune system.

Supplementary Materials

The following are available online at https://www.mdpi.com/2218-273X/7/1/13/s1, Supplementary Materials and Methods, Figure S1: Levenshtein distances between tRNA gene (tDNA) sequences from E. coli and S. cerevisiae, Figure S2: False-positive hits observed in the E. coli transfer RNA (tRNA) cleavage profiles, Figure S3: Quantification of Gm in E. coli tRNAMet by liquid chromatography coupled to tandem mass-spectrometry (LC-MS/MS), Figure S4: False-positive hits observed in the S. cerevisiae tRNA cleavage profiles, Table S1: Initial and collapsed E. coli tDNA dataset, Table S2: Initial and collapsed S. cerevisiae tDNA dataset, Table S3 :Number of sequencing reads used for the RiboMethSeq analysis of different samples.

Acknowledgments

This work was supported by Agence Nationale de Recherche HTRNAMod ANR-13-ISV8-0001 grant to Y.M. Part of this work was supported by grants of the German Research Foundation: Da592/5 to A.D. and by the DFG SPP1784, MH3397/12-1, MH3397/14-1 to M.H.

Author Contributions

Y.M., A.D. and M.H. conceived of and designed the experiments. V.M., I.F. and L.A. performed the experiments. F.P. and Y.M. analyzed the data. Y.M., V.M., A.D. and M.H. wrote the paper.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Cheng, X.; Roberts, R.J. AdoMet-dependent methylation, DNA methyltransferases and base flipping. Nucleic Acids Res. 2001, 29, 3784–3795. [Google Scholar] [CrossRef] [PubMed]
  2. Boschi-Muller, S.; Motorin, Y. Chemistry enters nucleic acids biology: Enzymatic mechanisms of RNA modification. Biochem. Biokhimiia 2013, 78, 1392–1404. [Google Scholar] [CrossRef] [PubMed]
  3. Motorin, Y.; Seidu-Larry, S.; Helm, M. DNA and RNA Pyrimidine Nucleobase Alkylation at the Carbon-5 Position. Adv. Exp. Med. Biol. 2016, 945, 19–33. [Google Scholar] [PubMed]
  4. Rana, A.K.; Ankri, S. Reviving the RNA World: An Insight into the Appearance of RNA Methyltransferases. Front. Genet. 2016, 7, 99. [Google Scholar] [CrossRef] [PubMed]
  5. Towns, W.L.; Begley, T.J. Transfer RNA methytransferases and their corresponding modifications in budding yeast and humans: Activities, predications, and potential roles in human health. DNA Cell Biol. 2012, 31, 434–454. [Google Scholar] [CrossRef] [PubMed]
  6. Motorin, Y.; Helm, M. RNA nucleotide methylation. Wiley Interdiscip. Rev. RNA 2011, 2, 611–631. [Google Scholar] [CrossRef] [PubMed]
  7. Gilbert, W.V.; Bell, T.A.; Schaening, C. Messenger RNA modifications: Form, distribution, and function. Science 2016, 352, 1408–1412. [Google Scholar] [CrossRef] [PubMed]
  8. Yue, Y.; Liu, J.; He, C. RNA N6-methyladenosine methylation in post-transcriptional gene expression regulation. Genes Dev. 2015, 29, 1343–1355. [Google Scholar] [CrossRef] [PubMed]
  9. Sarin, L.P.; Leidel, S.A. Modify or die?--RNA modification defects in metazoans. RNA Biol. 2014, 11, 1555–1567. [Google Scholar] [CrossRef] [PubMed]
  10. Machnicka, M.A.; Milanowska, K.; Osman Oglou, O.; Purta, E.; Kurkowska, M.; Olchowik, A.; Januszewski, W.; Kalinowski, S.; Dunin-Horkawicz, S.; Rother, K.M.; et al. MODOMICS: A database of RNA modification pathways--2013 update. Nucleic Acids Res. 2013, 41, D262–D267. [Google Scholar] [CrossRef] [PubMed]
  11. Hori, H. Methylated nucleosides in tRNA and tRNA methyltransferases. Front. Genet. 2014, 5, 144. [Google Scholar] [CrossRef] [PubMed]
  12. Gillingham, D.; Shahid, R. Catalysts for RNA and DNA modification. Curr. Opin. Chem. Biol. 2015, 25, 110–114. [Google Scholar] [CrossRef] [PubMed]
  13. El Yacoubi, B.; Bailly, M.; de Crécy-Lagard, V. Biosynthesis and function of posttranscriptional modifications of transfer RNAs. Annu. Rev. Genet. 2012, 46, 69–95. [Google Scholar] [CrossRef] [PubMed]
  14. Kawamura, T.; Anraku, R.; Hasegawa, T.; Tomikawa, C.; Hori, H. Transfer RNA methyltransferases from Thermoplasma acidophilum, a thermoacidophilic archaeon. Int. J. Mol. Sci. 2014, 16, 91–113. [Google Scholar] [CrossRef] [PubMed]
  15. Phillips, G.; de Crécy-Lagard, V. Biosynthesis and function of tRNA modifications in Archaea. Curr. Opin. Microbiol. 2011, 14, 335–341. [Google Scholar] [CrossRef] [PubMed]
  16. Gustafsson, C.; Reid, R.; Greene, P.J.; Santi, D.V. Identification of new RNA modifying enzymes by iterative genome search using known modifying enzymes as probes. Nucleic Acids Res. 1996, 24, 3756–3762. [Google Scholar] [CrossRef] [PubMed]
  17. Anantharaman, V.; Koonin, E.V.; Aravind, L. SPOUT: A class of methyltransferases that includes spoU and trmD RNA methylase superfamilies, and novel superfamilies of predicted prokaryotic RNA methylases. J. Mol. Microbiol. Biotechnol. 2002, 4, 71–75. [Google Scholar] [PubMed]
  18. Koonin, E.V.; Rudd, K.E. SpoU protein of Escherichia coli belongs to a new family of putative rRNA methylases. Nucleic Acids Res. 1993, 21, 5519. [Google Scholar] [CrossRef] [PubMed]
  19. Persson, B.C.; Jäger, G.; Gustafsson, C. The spoU gene of Escherichia coli, the fourth gene of the spoT operon, is essential for tRNA (Gm18) 2′-O-methyltransferase activity. Nucleic Acids Res. 1997, 25, 4093–4097. [Google Scholar] [CrossRef] [PubMed]
  20. Hori, H.; Suzuki, T.; Sugawara, K.; Inoue, Y.; Shibata, T.; Kuramitsu, S.; Yokoyama, S.; Oshima, T.; Watanabe, K. Identification and characterization of tRNA (Gm18) methyltransferase from Thermus thermophilus HB8: Domain structure and conserved amino acid sequence motifs. Genes Cells Devoted Mol. Cell. Mech. 2002, 7, 259–272. [Google Scholar] [CrossRef]
  21. Watanabe, K.; Nureki, O.; Fukai, S.; Ishii, R.; Okamoto, H.; Yokoyama, S.; Endo, Y.; Hori, H. Roles of conserved amino acid sequence motifs in the SpoU (TrmH) RNA methyltransferase family. J. Biol. Chem. 2005, 280, 10368–10377. [Google Scholar] [CrossRef] [PubMed]
  22. Pleshe, E.; Truesdell, J.; Batey, R.T. Structure of a class II TrmH tRNA-modifying enzyme from Aquifex aeolicus. Acta Crystallogr. Sect. F Struct. Biol. Cryst. Commun. 2005, 61, 722–728. [Google Scholar] [CrossRef] [PubMed]
  23. Purta, E.; van Vliet, F.; Tkaczuk, K.L.; Dunin-Horkawicz, S.; Mori, H.; Droogmans, L.; Bujnicki, J.M. The yfhQ gene of Escherichia coli encodes a tRNA:Cm32/Um32 methyltransferase. BMC Mol. Biol. 2006, 7, 23. [Google Scholar] [CrossRef] [PubMed]
  24. Somme, J.; Van Laer, B.; Roovers, M.; Steyaert, J.; Versées, W.; Droogmans, L. Characterization of two homologous 2′-O-methyltransferases showing different specificities for their tRNA substrates. RNA 2014, 20, 1257–1271. [Google Scholar] [CrossRef] [PubMed]
  25. Armengod, M.-E.; Moukadiri, I.; Prado, S.; Ruiz-Partida, R.; Benítez-Páez, A.; Villarroya, M.; Lomas, R.; Garzón, M.J.; Martínez-Zamora, A.; Meseguer, S.; et al. Enzymology of tRNA modification in the bacterial MnmEG pathway. Biochimie 2012, 94, 1510–1520. [Google Scholar] [CrossRef] [PubMed]
  26. Benítez-Páez, A.; Villarroya, M.; Douthwaite, S.; Gabaldón, T.; Armengod, M.-E. YibK is the 2′-O-methyltransferase TrmL that modifies the wobble nucleotide in Escherichia coli tRNALeu isoacceptors. RNA 2010, 16, 2131–2143. [Google Scholar] [CrossRef] [PubMed]
  27. Jaroensuk, J.; Atichartpongkul, S.; Chionh, Y.H.; Wong, Y.H.; Liew, C.W.; McBee, M.E.; Thongdee, N.; Prestwich, E.G.; DeMott, M.S.; Mongkolsuk, S.; et al. Methylation at position 32 of tRNA catalyzed by TrmJ alters oxidative stress response in Pseudomonas aeruginosa. Nucleic Acids Res. 2016, 44, 10834–10848. [Google Scholar] [CrossRef] [PubMed]
  28. Grosjean, H.; Gaspin, C.; Marck, C.; Decatur, W.A.; de Crécy-Lagard, V. RNomics and Modomics in the halophilic archaea Haloferax volcanii: Identification of RNA modification genes. BMC Genom. 2008, 9, 470. [Google Scholar] [CrossRef] [PubMed]
  29. Joardar, A.; Malliahgari, S.R.; Skariah, G.; Gupta, R. 2′-O-methylation of the wobble residue of elongator pre-tRNAMet in Haloferax volcanii is guided by a box C/D RNA containing unique features. RNA Biol. 2011, 8, 782–791. [Google Scholar] [CrossRef] [PubMed]
  30. Singh, S.K.; Gurha, P.; Tran, E.J.; Maxwell, E.S.; Gupta, R. Sequential 2′-O-methylation of archaeal pre-tRNATrp nucleotides is guided by the intron-encoded but trans-acting box C/D ribonucleoprotein of pre-tRNA. J. Biol. Chem. 2004, 279, 47661–47671. [Google Scholar] [CrossRef] [PubMed]
  31. Clouet-d’Orval, B.; Gaspin, C.; Mougin, A. Two different mechanisms for tRNA ribose methylation in Archaea: A short survey. Biochimie 2005, 87, 889–895. [Google Scholar] [CrossRef] [PubMed]
  32. Renalier, M.-H.; Joseph, N.; Gaspin, C.; Thebault, P.; Mougin, A. The Cm56 tRNA modification in archaea is catalyzed either by a specific 2′-O-methylase, or a C/D sRNP. RNA 2005, 11, 1051–1063. [Google Scholar] [CrossRef] [PubMed]
  33. Kuratani, M.; Bessho, Y.; Nishimoto, M.; Grosjean, H.; Yokoyama, S. Crystal structure and mutational study of a unique SpoU family archaeal methylase that forms 2′-O-methylcytidine at position 56 of tRNA. J. Mol. Biol. 2008, 375, 1064–1075. [Google Scholar] [CrossRef] [PubMed]
  34. Cavaillé, J.; Chetouani, F.; Bachellerie, J.P. The yeast Saccharomyces cerevisiae YDL112w ORF encodes the putative 2′-O-ribose methyltransferase catalyzing the formation of Gm18 in tRNAs. RNA 1999, 5, 66–81. [Google Scholar] [CrossRef] [PubMed]
  35. Pintard, L.; Lecointe, F.; Bujnicki, J.M.; Bonnerot, C.; Grosjean, H.; Lapeyre, B. Trm7p catalyses the formation of two 2′-O-methylriboses in yeast tRNA anticodon loop. EMBO J. 2002, 21, 1811–1820. [Google Scholar] [CrossRef] [PubMed]
  36. Guy, M.P.; Podyma, B.M.; Preston, M.A.; Shaheen, H.H.; Krivos, K.L.; Limbach, P.A.; Hopper, A.K.; Phizicky, E.M. Yeast Trm7 interacts with distinct proteins for critical modifications of the tRNAPhe anticodon loop. RNA 2012, 18, 1921–1933. [Google Scholar] [CrossRef] [PubMed]
  37. Guy, M.P.; Phizicky, E.M. Conservation of an intricate circuit for crucial modifications of the tRNAPhe anticodon loop in eukaryotes. RNA 2015, 21, 61–74. [Google Scholar] [CrossRef] [PubMed]
  38. Guy, M.P.; Shaw, M.; Weiner, C.L.; Hobson, L.; Stark, Z.; Rose, K.; Kalscheuer, V.M.; Gecz, J.; Phizicky, E.M. Defects in tRNA Anticodon Loop 2′-O-Methylation Are Implicated in Nonsyndromic X-Linked Intellectual Disability due to Mutations in FTSJ1. Hum. Mutat. 2015, 36, 1176–1187. [Google Scholar] [CrossRef] [PubMed]
  39. Kotelawala, L.; Grayhack, E.J.; Phizicky, E.M. Identification of yeast tRNA Um44 2′-O-methyltransferase (Trm44) and demonstration of a Trm44 role in sustaining levels of specific tRNASer species. RNA 2008, 14, 158–169. [Google Scholar] [CrossRef] [PubMed]
  40. Wilkinson, M.L.; Crary, S.M.; Jackman, J.E.; Grayhack, E.J.; Phizicky, E.M. The 2′-O-methyltransferase responsible for modification of yeast tRNA at position 4. RNA 2007, 13, 404–413. [Google Scholar] [CrossRef] [PubMed]
  41. Gross, H.J.; Simsek, M.; Raba, M.; Limburg, K.; Heckman, J.; Raj Bhandary, U.L. 2′-O-methyl ribothymidine: a component of rabbit liver lysine transfer RNA. Nucleic Acids Res. 1974, 1, 35–43. [Google Scholar] [CrossRef] [PubMed]
  42. Païs de Barros, J.P.; Keith, G.; El Adlouni, C.; Glasser, A.L.; Mack, G.; Dirheimer, G.; Desgrès, J. 2′-O-methyl-5-formylcytidine (f5Cm), a new modified nucleotide at the “wobble” of two cytoplasmic tRNAsLeu(NAA) from bovine liver. Nucleic Acids Res. 1996, 24, 1489–1496. [Google Scholar] [CrossRef] [PubMed]
  43. Eichler, D.C. Characterization of a nucleolar 2′-O-methyltransferase and its involvement in the methylation of mouse precursor ribosomal RNA. Biochimie 1994, 76, 1115–1122. [Google Scholar] [CrossRef]
  44. Droogmans, L.; Haumont, E.; de Henau, S.; Grosjean, H. Enzymatic 2′-O-methylation of the wobble nucleoside of eukaryotic tRNAPhe: Specificity depends on structural elements outside the anticodon loop. EMBO J. 1986, 5, 1105–1109. [Google Scholar] [PubMed]
  45. Kawai, G.; Yamamoto, Y.; Kamimura, T.; Masegi, T.; Sekine, M.; Hata, T.; Iimori, T.; Watanabe, T.; Miyazawa, T.; Yokoyama, S. Conformational rigidity of specific pyrimidine residues in tRNA arises from posttranscriptional modifications that enhance steric interaction between the base and the 2′-hydroxyl group. Biochemistry 1992, 31, 1040–1046. [Google Scholar] [CrossRef] [PubMed]
  46. Kawai, G.; Ue, H.; Yasuda, M.; Sakamoto, K.; Hashizume, T.; McCloskey, J.A.; Miyazawa, T.; Yokoyama, S. Relation between functions and conformational characteristics of modified nucleosides found in tRNAs. Nucleic Acids Symp. Ser. 1991, 49–50. [Google Scholar]
  47. Kita, E.; Kashiba, S. Immunogenicity of transfer RNA isolated from a two-heptose rough mutant of Salmonella typhimurium LT2 in mouse typhoid infection. Immunology 1983, 50, 369–376. [Google Scholar] [PubMed]
  48. Gehrig, S.; Eberle, M.-E.; Botschen, F.; Rimbach, K.; Eberle, F.; Eigenbrod, T.; Kaiser, S.; Holmes, W.M.; Erdmann, V.A.; Sprinzl, M.; et al. Identification of modifications in microbial, native tRNA that suppress immunostimulatory activity. J. Exp. Med. 2012, 209, 225–233. [Google Scholar] [CrossRef] [PubMed]
  49. Jöckel, S.; Nees, G.; Sommer, R.; Zhao, Y.; Cherkasov, D.; Hori, H.; Ehm, G.; Schnare, M.; Nain, M.; Kaufmann, A.; Bauer, S. The 2′-O-methylation status of a single guanosine controls transfer RNA-mediated Toll-like receptor 7 activation or inhibition. J. Exp. Med. 2012, 209, 235–241. [Google Scholar] [CrossRef] [PubMed]
  50. Rimbach, K.; Kaiser, S.; Helm, M.; Dalpke, A.H.; Eigenbrod, T. 2′-O-Methylation within Bacterial RNA Acts as Suppressor of TLR7/TLR8 Activation in Human Innate Immune Cells. J. Innate Immun. 2015, 7, 482–493. [Google Scholar] [CrossRef] [PubMed]
  51. Oldenburg, M.; Krüger, A.; Ferstl, R.; Kaufmann, A.; Nees, G.; Sigmund, A.; Bathke, B.; Lauterbach, H.; Suter, M.; Dreher, S.; et al. TLR13 recognizes bacterial 23S rRNA devoid of erythromycin resistance-forming modification. Science 2012, 337, 1111–1115. [Google Scholar] [CrossRef] [PubMed]
  52. Maden, B.E.; Corbett, M.E.; Heeney, P.A.; Pugh, K.; Ajuh, P.M. Classical and novel approaches to the detection and localization of the numerous modified nucleotides in eukaryotic ribosomal RNA. Biochimie 1995, 77, 22–29. [Google Scholar] [CrossRef]
  53. Motorin, Y.; Muller, S.; Behm-Ansmant, I.; Branlant, C. Identification of modified residues in RNAs by reverse transcription-based methods. Methods Enzymol. 2007, 425, 21–53. [Google Scholar] [PubMed]
  54. Gehrke, C.W.; Kuo, K.C.; McCune, R.A.; Gerhardt, K.O.; Agris, P.F. Quantitative enzymatic hydrolysis of tRNAs: Reversed-phase high-performance liquid chromatography of tRNA nucleosides. J. Chromatogr. 1982, 230, 297–308. [Google Scholar] [CrossRef]
  55. Su, D.; Chan, C.T.Y.; Gu, C.; Lim, K.S.; Chionh, Y.H.; McBee, M.E.; Russell, B.S.; Babu, I.R.; Begley, T.J.; Dedon, P.C. Quantitative analysis of ribonucleoside modifications in tRNA by HPLC-coupled mass spectrometry. Nat. Protoc. 2014, 9, 828–841. [Google Scholar] [CrossRef] [PubMed]
  56. Cao, X.; Limbach, P.A. Enhanced detection of post-transcriptional modifications using a mass-exclusion list strategy for RNA modification mapping by LC-MS/MS. Anal. Chem. 2015, 87, 8433–8440. [Google Scholar] [CrossRef] [PubMed]
  57. Abbate, J.; Rottman, F. Gas chromatographic method for determination of 2′-O-methylation in RNA. Anal. Biochem. 1972, 47, 378–388. [Google Scholar] [CrossRef]
  58. Dominissini, D.; Moshitch-Moshkovitz, S.; Salmon-Divon, M.; Amariglio, N.; Rechavi, G. Transcriptome-wide mapping of N6-methyladenosine by m6A-seq based on immunocapturing and massively parallel sequencing. Nat. Protoc. 2013, 8, 176–189. [Google Scholar] [CrossRef] [PubMed]
  59. 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.; et al. The dynamic N1-methyladenosine methylome in eukaryotic messenger RNA. Nature 2016, 530, 441–446. [Google Scholar] [CrossRef] [PubMed]
  60. Schwartz, S.; Bernstein, D.A.; Mumbach, M.R.; Jovanovic, M.; Herbst, R.H.; León-Ricardo, B.X.; Engreitz, J.M.; Guttman, M.; Satija, R.; Lander, E.S.; et al. Transcriptome-wide mapping reveals widespread dynamic-regulated pseudouridylation of ncRNA and mRNA. Cell 2014, 159, 148–162. [Google Scholar] [CrossRef] [PubMed]
  61. Squires, J.E.; Patel, H.R.; Nousch, M.; Sibbritt, T.; Humphreys, D.T.; Parker, B.J.; Suter, C.M.; Preiss, T. Widespread occurrence of 5-methylcytosine in human coding and non-coding RNA. Nucleic Acids Res. 2012, 40, 5023–5033. [Google Scholar] [CrossRef] [PubMed]
  62. Tserovski, L.; Marchand, V.; Hauenschild, R.; Blanloeil-Oillo, F.; Helm, M.; Motorin, Y. High-throughput sequencing for 1-methyladenosine (m1A) mapping in RNA. Methods 2016, 107, 110–121. [Google Scholar] [CrossRef] [PubMed]
  63. Birkedal, U.; Christensen-Dalsgaard, M.; Krogh, N.; Sabarinathan, R.; Gorodkin, J.; Nielsen, H. Profiling of ribose methylations in RNA by high-throughput sequencing. Angew. Chem. 2015, 54, 451–455. [Google Scholar] [CrossRef]
  64. Krogh, N.; Jansson, M.D.; Häfner, S.J.; Tehler, D.; Birkedal, U.; Christensen-Dalsgaard, M.; Lund, A.H.; Nielsen, H. Profiling of 2′-O-Me in human rRNA reveals a subset of fractionally modified positions and provides evidence for ribosome heterogeneity. Nucleic Acids Res. 2016, 44, 7884–7895. [Google Scholar] [CrossRef] [PubMed]
  65. Marchand, V.; Blanloeil-Oillo, F.; Helm, M.; Motorin, Y. Illumina-based RiboMethSeq approach for mapping of 2′-O-Me residues in RNA. Nucleic Acids Res. 2016, 44, e135. [Google Scholar] [CrossRef] [PubMed]
  66. Transfer RNA database (tRNAdb). Available online: http://trna.bioinf.uni-leipzig.de (accessed on 1 December 2016).
  67. Bolger, A.M.; Lohse, M.; Usadel, B. Trimmomatic: A flexible trimmer for Illumina sequence data. Bioinformatics 2014, 30, 2114–2120. [Google Scholar] [CrossRef] [PubMed]
  68. Langmead, B.; Trapnell, C.; Pop, M.; Salzberg, S.L. Ultrafast and memory-efficient alignment of short DNA sequences to the human genome. Genome Biol. 2009, 10, R25. [Google Scholar] [CrossRef] [PubMed]
  69. Quinlan, A.R. BEDTools: The Swiss-Army Tool for Genome Feature Analysis. Curr. Protoc. Bioinform. 2014, 47, 1–34. [Google Scholar]
  70. The R Foundation. Available online: https://www.r-project.org/foundation/ (accessed on 1 December 2016).
Biomolecules 07 00013 g001 550
Figure 1. Known positions of tRNA 2′-O-methylations and corresponding enzymes in Eubacteria (E. coli/Bacillus subtilis), Archaea (data almost exclusively from H. volcanii) and Eukarya (S. cerevisiae and various vertebrates). tRNAs are represented in folded 3D structures. 2′-O-methylated nucleotides are shown as spheres, orange for Eubacteria, pink for Archaea. Light blue color corresponds to yeast modifications; additional vertebrate positions are shown in dark blue. Names of homologous human proteins are in italics.
Figure 1. Known positions of tRNA 2′-O-methylations and corresponding enzymes in Eubacteria (E. coli/Bacillus subtilis), Archaea (data almost exclusively from H. volcanii) and Eukarya (S. cerevisiae and various vertebrates). tRNAs are represented in folded 3D structures. 2′-O-methylated nucleotides are shown as spheres, orange for Eubacteria, pink for Archaea. Light blue color corresponds to yeast modifications; additional vertebrate positions are shown in dark blue. Names of homologous human proteins are in italics.
Biomolecules 07 00013 g001
Biomolecules 07 00013 g002 550
Figure 2. Fragmentation profiles for E. coli and S. cerevisiae total tRNA fractions. Alkaline hydrolysis was performed for 6–12 min (as indicated on the traces). Fragments size was analyzed by capillary electrophoresis on a Bioanalyzer 2100 (Agilent, Santa Clara, CA, USA) Small RNA Chip. FU: fluorescence unit.
Figure 2. Fragmentation profiles for E. coli and S. cerevisiae total tRNA fractions. Alkaline hydrolysis was performed for 6–12 min (as indicated on the traces). Fragments size was analyzed by capillary electrophoresis on a Bioanalyzer 2100 (Agilent, Santa Clara, CA, USA) Small RNA Chip. FU: fluorescence unit.
Biomolecules 07 00013 g002
Biomolecules 07 00013 g003 550
Figure 3. Optimized treatment pipeline for tRNA 2′-O-methylation analysis by high-throughput sequencing. SAM: sequence alignment/map.
Figure 3. Optimized treatment pipeline for tRNA 2′-O-methylation analysis by high-throughput sequencing. SAM: sequence alignment/map.
Biomolecules 07 00013 g003
Biomolecules 07 00013 g004 550
Figure 4. Proportion of non-uniquely mapped reads with initial and collapsed (optimized) transfer DNA (tDNA) datasets for each tRNA species. For both datasets, the number of uniquely and non-uniquely mapped reads was calculated for each tRNA species and plotted as stacked bars.
Figure 4. Proportion of non-uniquely mapped reads with initial and collapsed (optimized) transfer DNA (tDNA) datasets for each tRNA species. For both datasets, the number of uniquely and non-uniquely mapped reads was calculated for each tRNA species and plotted as stacked bars.
Biomolecules 07 00013 g004
Biomolecules 07 00013 g005 550
Figure 5. Representative cleavage profiles for six E. coli tRNA species for wild-type (WT) and ΔTrmH strains. The cumulated number of the 5′- and 3′-end extremities at every position is traced as vertical bars. All profiles are centered for the known 2′-O-methylated position shown by the black arrow; false-positive alkali-resistant positions are indicated by gray arrows.
Figure 5. Representative cleavage profiles for six E. coli tRNA species for wild-type (WT) and ΔTrmH strains. The cumulated number of the 5′- and 3′-end extremities at every position is traced as vertical bars. All profiles are centered for the known 2′-O-methylated position shown by the black arrow; false-positive alkali-resistant positions are indicated by gray arrows.
Biomolecules 07 00013 g005
Biomolecules 07 00013 g006 550
Figure 6. Representative cleavage profiles for six S. cerevisiae tRNA species for WT and ΔTRM3 strains. The cumulated number of the 5′- and 3′-end extremities at every position is traced as vertical bars. All profiles are centered for the known 2′-O-methylated position shown by the black arrow; false-positive alkali-resistant positions are indicated by gray arrows.
Figure 6. Representative cleavage profiles for six S. cerevisiae tRNA species for WT and ΔTRM3 strains. The cumulated number of the 5′- and 3′-end extremities at every position is traced as vertical bars. All profiles are centered for the known 2′-O-methylated position shown by the black arrow; false-positive alkali-resistant positions are indicated by gray arrows.
Biomolecules 07 00013 g006
Table
Table 1. Known transfer RNA (tRNA):2′-O-methyltransferases in model organisms (Escherichia coli, Haloferax volcanii, Saccharomyces cerevisiae, Homo sapiens) and their tRNA substrates.
Table 1. Known transfer RNA (tRNA):2′-O-methyltransferases in model organisms (Escherichia coli, Haloferax volcanii, Saccharomyces cerevisiae, Homo sapiens) and their tRNA substrates.
OrganismEnzymeEnzyme Alt. NamePositionModificationtRNAAnticodontRNAdb Reference
E. coliTrmHspoU18GmIlek2CAUtdbR00000160
18Leucmnm5UmAAtdbR00000220
18LeuCAGtdbR00000221
18LeuGAGtdbR00000222
18Metac4CAUtdbR00000276
18GlnCUGtdbR00000332
18GlnU*UGtdbR00000333
18SerCGAtdbR00000390
18SerGGAtdbR00000392-3
18Sercmo5UGAtdbR00000394
18TyrQUAtdbR00000545-46
18LeuCmAAtdbR00000595
TrmJYfhQ, TrMet(Xm32)32CmSercmo5UGAtdbR00000394
32TrpCCAtdbR00000484
32IniCAUtdbR00000510-11
32UmGlnCUGtdbR00000332
32GlnU*UGtdbR00000333
32ProCGGtdbR00000623
TrmLyibK34CmLeuCmAAtdbR00000595
34cmnm5UmLeucmnm5UmAAtdbR00000220
H. volcaniiaTrmJHVO_290632CmLysU*UUtdbR00000175
Lysac4CUUtdbR00000176
TrpCmCAtdbR00000478
TyrGUAtdbR00000541
sRNA/aFib34CmMetCmAUtdbR00000271
34CmTrpCmCAtdbR00000478
39UmTrpCmCAtdbR00000478
aTrm56HVO_117356CmAll known tRNAs
S. cerevisiaeTrm13YOL125w4CmGlyGCCtdbR00000129
4CmProU*GGtdbR00000323-4
4AmHisGUGtdbR00000145
Trm3YDL112w18GmLeuCAAtdbR00000249
18LeuUAGtdbR00000250
18LeuUAAtdbR00000251
18TrpCmCAtdbR00000494
18SerIGAtdbR00000407-8
18SerU*GAtdbR00000406
18HisGUGtdbR00000145
18TyrGψAtdbR00000555
Trm7/Trm732YBR061C/YMR259c32CmTrpCmCAtdbR00000494
32PheGmAAtdbR00000083-4
32LeuUAAtdbR00000251
Trm7/Trm734YBR061C/Rtt1034CmTrpCmCAtdbR00000494
34GmPheGmAAtdbR00000083-4
34cmnm5UmLyscmnm5UmUUtdbR00000193
Trm44YPL030w44UmSerIGAtdbR00000407-8
44SerU*GAtdbR00000406
H. sapienshTrm13CCDC764CmGlyCCCtdbR00000135
4UmGlyGCCtdbR00000136
hTrm3TARBP118GmAsnNUGtdbR00000306
18GlnU*UGtdbR00000345
18GlnU*UGtdbR00000346
18SerUGAtdbR00000428
hTrm7FTSJ132CmPheGmAAtdbR00000103
32AsnNUGtdbR00000306
32GlnU*UGtdbR00000345
32GlnU*UGtdbR00000346
32UmAlaIGCtdbR00000016
32AlaIGCtdbR00000017
34CmMetCmAUtdbR00000289
34GmPheGmAAtdbR00000103
unknown39ψmGlyCCCtdbR00000135
hTrm44METTL1944UmLeuNAAtdbR00000269
44SerUGAtdbR00000428
unknown54m5UmGluCUCtdbR00000062
tRNA reference numbers are from tRNA database (tRNAdb). U*: unknown modified uridine; Q: queuosine; I: inosine; ψ: pseudouridine; N: unknown residue; other modified nucleotides abbreviated according to their conventional names. Trm: tRNA methyltransferase; aTrm: archaeal tRNA methyltransferase; hTrm: human tRNA methyltransferase; sRNA: small RNA; aFib: archaeal fibrillarin.
Table
Table 2. Calculated ScoreMean and MethScore for known 2′-O-methylated positions in E. coli and S. cerevisiae tRNAs extracted from WT and the strain lacking tRNA:Gm18-MTase (TrmH or TRM3).
Table 2. Calculated ScoreMean and MethScore for known 2′-O-methylated positions in E. coli and S. cerevisiae tRNAs extracted from WT and the strain lacking tRNA:Gm18-MTase (TrmH or TRM3).
Escherichia coli
WT E. coliΔTrmH E. coliDifference
tdbR NumbertRNAModificationPosition *ScoreMeanMethScoreScoreMeanMethScoreScoreMeanMethScore
tdbR00000160tRNAIle(k2CAU)Gm18180.920.680.29−0.370.631.05
tdbR00000332tRNAGln(CUG)Gm18180.970.850.940.790.030.06
tdbR00000333tRNAGln(U*UG)Gm18180.810.70.760.670.050.04
tdbR00000390tRNASer(CGA)Gm18180.60.250.24−0.30.350.55
tdbR00000392-393tRNASer(GGA)Gm18180.860.760.260.320.60.44
tdbR00000394tRNASer(cmo5UGA)Gm18180.980.880.940.80.040.08
tdbR00000545-546tRNATyr(QUA)Gm18180.930.840.350.170.580.67
tdbR00000220tRNALeu(cmnm5UmAA)Gm18190.870.420.46−0.780.421.21
tdbR00000221tRNALeu(CAG)Gm18190.970.890.35−0.020.620.9
tdbR00000222tRNALeu(GAG)Gm18190.990.950.25−0.480.741.44
tdbR00000276tRNAMet(ac4CAU)Gm18190.580.420.750.44−0.17−0.02
tdbR00000595tRNALeu(CmAA)Gm18190.930.90.730.670.190.22
tdbR00000332tRNAGln(CUG)Um32320.970.850.940.790.030.06
tdbR00000333tRNAGln(U*UG)Um32320.810.70.760.670.050.04
tdbR00000484tRNATrp(CCA)Cm32330.990.920.990.9100.01
tdbR00000394tRNASer(cmo5UGA)Cm32340.980.880.940.80.040.08
tdbR00000510-511tRNAMeti(CAU)Cm32340.970.840.970.850−0.02
tdbR00000623tRNAPro(CGG)Um32340.980.930.970.9200.01
tdbR00000595tRNALeu(CmAA)Cm343610.940.990.880.010.05
tdbR00000220tRNALeu(cmnm5UmAA)cmnm5Um343610.940.980.810.020.13
Saccharomyces cerevisae
WT S. cerevisiaeΔTRM3 S. cerevisiaeDifference
tdbR NumbertRNAModificationPosition *ScoreMeanMethScoreScoreMeanMethScoreScoreMeanMethScore
tdbR00000129tRNAGly(GCC)Cm450.890.630.930.76−0.04−0.13
tdbR00000323-4tRNAPro(U*GG)Cm450.950.830.950.810.010.02
tdbR00000145tRNAHis(GUG)Am560.980.890.960.850.010.03
tdbR00000249tRNALeu(CAA)Gm18180.990.890.840.030.150.87
tdbR00000250tRNALeu(UAG)Gm18180.930.84−0.35−0.461.281.29
tdbR00000251tRNALeu(UAA)Gm18180.930.86−0.16−0.141.101.01
tdbR00000406tRNASer(U*GA)Gm18180.910.850.750.570.150.29
tdbR00000407-8tRNASer(IGA)Gm18180.950.81−0.35−0.801.301.61
tdbR00000494tRNATrp(CmCA)Gm18180.930.810.26−0.050.670.87
tdbR00000145tRNAHis(GUG)Gm18190.910.80−0.98−0.841.891.64
tdbR00000555tRNATyr(GψA)Gm18190.960.920.090.130.870.79
tdbR00000494tRNATrp(CmCA)Cm32320.990.860.990.880.00−0.02
tdbR00000083-4tRNAPhe(GmAA)Cm32330.980.790.980.810.00−0.03
tdbR00000251tRNALeu(UAA)Cm32340.920.550.950.69−0.03−0.14
tdbR00000494tRNATrp(CmCA)Cm34340.990.870.990.890.00−0.03
tdbR00000083-4tRNAPhe(GmAA)Gm34350.960.740.930.630.030.10
tdbR00000193tRNALys(cmnm5UmUU)cmnm5Um34350.640.300.580.260.060.04
tdbR00000406tRNASer(U*GA)Um44450.930.860.930.83−0.010.03
tdbR00000407-8tRNASer(IGA)Um44450.900.780.860.710.040.07
tdbR identifiers are taken from tRNAdb [66]; when the collapsed sequence is used, both numbers are indicated. * The position corresponds to the real position of the signal in the sequence of tRNA; thus it appears at position +1 to the 2′-O-methylated residue. U*: unknown modified uridine; Q: queuosine; I: inosine; ψ: pseudouridine. Important changes in ScoreMean and MethScore indicating disappearence of Gm18 are bolded.

Share and Cite

MDPI and ACS Style

Marchand, V.; Pichot, F.; Thüring, K.; Ayadi, L.; Freund, I.; Dalpke, A.; Helm, M.; Motorin, Y. Next‐Generation Sequencing‐Based RiboMethSeq Protocol for Analysis of tRNA 2′‐O‐Methylation. Biomolecules 2017, 7, 13. https://doi.org/10.3390/biom7010013

AMA Style

Marchand V, Pichot F, Thüring K, Ayadi L, Freund I, Dalpke A, Helm M, Motorin Y. Next‐Generation Sequencing‐Based RiboMethSeq Protocol for Analysis of tRNA 2′‐O‐Methylation. Biomolecules. 2017; 7(1):13. https://doi.org/10.3390/biom7010013

Chicago/Turabian Style

Marchand, Virginie, Florian Pichot, Kathrin Thüring, Lilia Ayadi, Isabel Freund, Alexander Dalpke, Mark Helm, and Yuri Motorin. 2017. "Next‐Generation Sequencing‐Based RiboMethSeq Protocol for Analysis of tRNA 2′‐O‐Methylation" Biomolecules 7, no. 1: 13. https://doi.org/10.3390/biom7010013

APA Style

Marchand, V., Pichot, F., Thüring, K., Ayadi, L., Freund, I., Dalpke, A., Helm, M., & Motorin, Y. (2017). Next‐Generation Sequencing‐Based RiboMethSeq Protocol for Analysis of tRNA 2′‐O‐Methylation. Biomolecules, 7(1), 13. https://doi.org/10.3390/biom7010013

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop