Abstract
Functional tRNA molecules always contain a wide variety of post-transcriptionally modified nucleosides. These modifications stabilize tRNA structure, allow for proper interaction with other macromolecules and fine-tune the decoding of mRNAs during translation. Their presence in functionally important regions of tRNA is conserved in all domains of life. However, the identities of many of these modified residues depend much on the phylogeny of organisms the tRNAs are found in, attesting for domain-specific strategies of tRNA maturation. In this work we present a new tool, tRNAmodviz web server (http://genesilico.pl/trnamodviz) for easy comparative analysis and visualization of modification patterns in individual tRNAs, as well as in groups of selected tRNA sequences. We also present results of comparative analysis of tRNA sequences derived from 7 phylogenetically distinct groups of organisms: Gram-negative bacteria, Gram-positive bacteria, cytosol of eukaryotic single cell organisms, Fungi and Metazoa, cytosol of Viridiplantae, mitochondria, plastids and Euryarchaeota. These data update the study conducted 20 y ago with the tRNA sequences available at that time.
Introduction
Many nucleotide residues in tRNA transcripts are chemically modified by a wide variety of modification enzymes. These post-transcriptional modifications vary much in kind, degree of complexity and frequency of their occurrence. Some of them are present in the majority of tRNA molecules from most organisms, while others are specific to particular isoacceptor tRNA species or present in a limited number of tRNAs of a few evolutionarily related organisms. Some modifications are crucial for correct folding and stability of tRNA molecules.Citation1 They can also modulate (facilitate or prevent) interactions with other cellular macromolecules such as amino-acyl tRNA synthetases and translation factorsCitation2 and, most importantly, they allow for accurate decoding of mRNAs on the ribosome.Citation3-5 tRNA modification can be used by bacteria to suppress activation of the host immune response.Citation6 Recent works point out clear correlations between the absence of a normally modified nucleoside in certain tRNAs and the cellular response to stresses conditions, attesting for the vital importance and possibly a regulatory role of hypomodified tRNAs in cell metabolism (reviewed in refs.Citation7,8). Links between defects in tRNA modifications and human diseases have also been reported (reviewed in ref.Citation9).
The great variety of modified residues found in tRNAs can be subdivided into 3 main categories. The first category groups modified nucleosides for which both position and identity are conserved in the majority of tRNA species, such as dihydrouridine (D) and 2′-O-methylguanosine (Gm) in the D-loop or 5-methyluridine (m5U) and pseudouridine (Ψ) in the TΨ-loop. They are introduced by modification enzymes that are usually also evolutionary conserved (orthologs). The second category comprises modified residues for which only the position but not necessarily the identity of modification is conserved, such as the many modified, often hypermodified uridines and purines found at positions 34 and 37 of the anticodon loop, respectively. They are introduced by enzymes that are not necessarily conserved, attesting for the functional convergence (evolutionary analogy). The last group corresponds to all the other modified nucleosides, usually with simple chemical alterations, such as those methylated at base or ribose (mN or Nm, respectively, where N is any of the 4 canonical bases) or isomeric derivatives (Ψ). Modified residues of the third class are found at unique positions of only a limited group of tRNA species and the corresponding modification enzymes are, as expected, very diverse and often species-specific.Citation10
The distribution of modified residues in tRNAs of different groups of organisms was examined several decades ago.Citation11-13 The previous compilations demonstrated notable differences in the modification patterns of tRNA originating from organisms distantly related in evolution, the diversity of modifications was noted especially at position 34 and conserved purine at position 37 of anticodon (reviewed in ref.Citation14). For only a few organisms the complete repertoire of post-transcriptional modifications (position and identity) is known. These are Escherichia coli, Saccharomyces cerevisiae (see in ref.Citation15 and in ref.Citation16 and in ref.Citation17), Haloferax volcanii ( in ref.Citation18 Fig. 8.2 in ref.Citation19), Mycoplasma capricolum (Table 2 in ref.Citation20 and Supplementary Material in ref.Citation21) and Lactococcus lactis (published during the preparation of this manuscript).Citation22 Also, the complete landscape of mammalian mitochondrial tRNAs has been recently updated and the consequence of their defects on mitochondrial dysfunctions ultimately causing pathological changes were discussed.Citation23
Table 1. Minimal, average and maximal percentage of modified residues per tRNA molecule in 7 groups of organisms analyzed in this work
Figure 1. Consensus tRNA secondary structure presented in the “cloverleaf” form with the universal numbering system.Citation26
Figure 2. (A) Distribution of modified residues in tRNA sequences from E. coli as presented by the tRNAmodviz server. Percent of modified residues within each position is mapped onto the consensus tRNA secondary structure (small black and white pie charts within the cloverleaf). (B) Frequency of A, U, C or G bases occurrence and their derivatives in tRNA sequences from E. coli. (C) Color pie charts which present detailed information for chosen position in tRNA sorted by the originating base, sorted by the presence of modification or covering modifications only – these options the user can choose when working with tRNAmodviz.
Figure 3. Modification profile for tRNA sequences from Gram-negative bacteria (62 sequences from 8 species). The pie charts within each position in the cloverleaf correspond to the percentage of all modified nucleosides (modified being drawn in black). In the tables the series of numbers next to the series of symbols indicate the frequency of occurrence of listed nucleosides at the particular position. The last number corresponds to the frequency of occurrence of the encoded unmodified nucleoside. The numbering of the residues is presented in . The list of species from which the analyzed tRNA sequences originate is provided in Supplementary Table.1.
Figure 4. Modification profile for tRNA sequences from Gram-positive bacteria (72 sequences from 9 species). For the description of the tables and cloverleaf content see the legend for . The numbering of the residues is presented in . Among sequences considered for this group there are 2 tRNAGly isoacceptors from Staphylococcus epidermidis which probably do not function in protein synthesis.Citation59 These 2 lack the conserved pseudouridine in position 55. The list of species from which the analyzed tRNA sequences originate is provided in Supplementary Table.1.
To date nearly 600 tRNA sequences have been investigated with respect to the presence of modified nucleosides. With this amount of data answering the questions such as: “What patterns of modification do we observe in tRNA from different organisms?” or “What is the frequency of modifications at each position in the tRNA molecule?” would give valuable insight into the rules governing tRNA modification processes. Since doing the analysis by hand would be laborious and time-consuming, we provide an online software tool (tRNAmodviz) for easy and interactive visualization of the pattern (identity and frequency) of post-transcriptional modifications on the consensus tRNA cloverleaf structure (shown in a simplified secondary structure representation in ). Our tool facilitates comparative studies of tRNA modifications and identification of peculiarities of individual or any subset of tRNA isoacceptor species among a large pool of available fully sequenced tRNAs. The dataset comprises tRNA sequences present in the MODOMICS databaseCitation24 which currently contains all sequences available in the tRNAdbCitation25 (some of which were subsequently curated manually) and additional sequences retrieved from literature. In this article we provide an example analysis for tRNAs from organisms belonging to different phyla of the 3 domains of life.
Results
A new tool to visualize post-transcriptional modifications in tRNAs
tRNAmodviz (http://genesilico.pl/trnamodviz) is a web service that calculates the frequency of modified nucleosides at each position for a set of tRNA sequences available from the MODOMICS database (see Materials and Methods for details) or for its selected subsets and visualizes the results. tRNAmodviz counts the occurrences of each type of modified and non-modified residues in each column of the sequence alignment and calculates their frequencies. A modification profile is then displayed in the framework of the cloverleaf tRNA secondary structure diagram. tRNA sequences can be selected according to kingdom, species, cellular localization (organelle) and/or amino acid specificities or individually from the provided list. A percentage of modified nucleosides at each position is displayed as a pie chart at a corresponding position within the diagram as exemplified in . Detailed data for a single position can be obtained by clicking at the given residue. The modification profile is then visualized as a pie chart in a separate panel, next to the cloverleaf graph. A user can switch between 3 different ways of ordering the modification data within the pie chart. By default, all types of modifications of one of the original unmodified nucleosides (A, U, C or G) follow that nucleoside in a pie chart (so modified ‘As’ are next to the unmodified ‘A’ etc.). Alternatively, all modified residues can be displayed next to each other. A chart with modified residues only can also be obtained (). A summary of data for each position of the selected set of tRNA sequences can be also viewed as a table. Positions in the tRNA are listed in rows and numbers of sequences that contain the particular nucleoside (modified or not) are presented in columns. The frequencies of A, U, C or G base occurrence (including both nascent and modified versions) in tRNA sequences can also be visualized (). In such a case, results are presented as 4-color pie charts within the cloverleaf diagram. For the user convenience, a comprehensive video tutorial on the homepage of the tRNAmodviz server is provided.
Locations and frequencies of post-transcriptionally modified nucleosides in tRNA
To illustrate the applicability of the tRNAmodviz web server, we analyzed the occurrence of modified nucleosides in a selection of 584 tRNA sequences known as of May 2014. The results are presented in the form of cloverleaf diagrams accompanied by tables that collect information about modification profiles for each position in the tRNA molecule. In all tables and figures, universal numbering system for tRNA position corresponds to the one used in ref.Citation26 and is presented in . tRNAs were classified in 7 categories according to their origin: Gram-negative bacteria (62 sequences from 8 species, mostly E. coli, ), Gram-positive bacteria (72 sequences from 9 species, mostly Bacillus and Mycoplasma, ), cytosol of eukaryotic single cell organisms, Fungi and Metazoa (173 sequences from 22 species, mostly S. cerevisiae, ), cytosol of Viridiplantae (55 sequences from 14 species, mostly Lupinus luteus, ), mitochondria (124 sequences from 18 species, mostly Bos taurus, ), plastids (38 sequences from 12 species, mostly Spinacia oleacera,) and Euryarchaeota (60 sequences, 6 species, majority from Halobacteria, ). Euryarchaeota are the only phylum of Archaea taken into account in this study, since in the case of Crenoarchaeota there is only one fully sequenced tRNA available (Sulfolobus acidocaldarius tRNAIniCitation27), for Nanoarchaeota only partial or incomplete sequences are available, and for other Archaea groups there is no official data available, except for limited information about modified nucleotide content in bulk tRNAs, but with no information about their exact location in sequenced tRNAs.Citation18,19 The exact list of species with numbers of tRNA sequences considered in each subgroup is provided in Supplementary Table 1. They all correspond to cellular, naturally occurring and fully sequenced tRNAs. tRNAs of phages and viruses were not taken into account, since they are known to adopt the pattern of modification characteristic for their hosts (for examples see refs:Citation28,29). Likewise, information about mutants of wild type tRNA species was considered as redundant and was omitted from this analysis.
Figure 5. Modification profile for tRNA sequences from cytosol of eukaryotic single cell organisms, Fungi and Metazoa (173 sequences from 22 species). For the description of the tables and cloverleaf content see the legend for . The numbering of the residues is presented in . According to the original work xG37 in 2 R. norvegicus tRNALeu was converted to m1G by treatment with alkali.Citation60 The list of species from which the analyzed tRNA sequences originate is provided in Supplementary Table.1.
Figure 6. Modification profile for tRNA sequences from cytosol of Viridiplantae (55 sequences from 14 species). For the description of the tables and cloverleaf content see the legend for . The numbering of the residues is presented in . xG64 in Scenedesmus obliquus and Triticum aestivum is most probably Gr(p), like in cytoplasmic tRNAIni from S. cerevisiae. The list of species from which the analyzed tRNA sequences originate is provided in Supplementary Table.1.
Figure 7. Modification profile for tRNA sequences from mitochondria (124 sequences from 18 species). For the description of the tables and cloverleaf content see the legend for . The numbering of the residues is presented in . Note that sequences from Ascaris suum have a lot of deletions and as a result their alignment is problematic. In the majority of cases xA37 is i6A37 or ms2i6A. In Neurospora crassa tRNATyr m5C is present in position e21 or 48.Citation61 In this compilation m5C in e21 and xC48 were kept. The list of species from which the analyzed tRNA sequences originate is provided in Supplementary Table.1.
Figure 8. Modification profile for tRNA sequences from plastids (38 sequences from 12 species). For the description of the tables and cloverleaf content see the legend for . The numbering of the residues is presented in . The list of species from which the analyzed tRNA sequences originate is provided in Supplementary Table.1.
Figure 9. Modification profile for tRNA sequences from Euryarchaeota (60 sequences, mostly Haloferax volcanii). For the description of the tables and cloverleaf content see the legend for . The numbering of the residues is presented in . xA57 present in initiator tRNA from Thermoplasma acidophilum is not m1I.Citation27 The list of species, from which the analyzed tRNA sequences originate, is provided in Supplementary Table.1.
Standard symbols for modified residues were used (detailed information on each modified nucleoside can be found in http://modomics.genesilico.pl/modifications/). The series of numbers next to the series of symbols indicate the frequency of occurrence of listed nucleosides at the particular position in the group of tRNAs considered. The last number corresponds to the frequency of occurrence of the unmodified nucleoside in the mature tRNA molecules. The pie charts within each position of the cloverleaf diagram illustrate the percentage of all modified nucleosides (with modified residues shown in black and unmodified residues in white).
Prevalence of modified nucleosides in tRNAs
Depending on the organism considered, the average fraction of modified nucleosides ranges from about 6.5% of the total residues in tRNAs from Gram-positive bacteria up to about 16.5 % of the total residues in tRNAs from eukaryotic single cell organisms, Fungi and Metazoa (). However, inspection of individual tRNA sequences reveals that occurrence of modified nucleotides can be as low as 1.7% (1 modified residue in mitochondrial tRNASer of Mesocricetus auratus) up to 23.7% (18 modified residues in the cytoplasmic tRNATrp of Triticum aestivum), see Supplementary Table 2.
Among the 5293 modified positions examined in the 584 tRNAs analyzed in this work, the most conserved modified nucleosides in all sequenced tRNAs are Ψ at position 55 in the TΨ-loop (found in 492 tRNAs over a total of 584, only tRNA from mitochondria, Mycoplasma and several cytoplasmic initiator tRNAs from eukaryotes generally lack Ψ55), m5U or its derivatives: m5s2U (5-methyl-2-thiouridine, found exclusively in tRNAs of thermophilic bacteria) and m5Um (5,2′-O-dimethyluridine, found 335 times) at position 54 and D at position 20, 20a and/or 20b (443 times), see Supplementary Table 3 for the detailed list. Beside these almost universally conserved modifications of the TΨ-loop, other modified residues are located in single stranded regions of the D-loop and the variable loop. When viewed within the 3D L-shape architecture (), they are clearly concentrated in the core of the tRNA molecule, mainly at the interface of the D- and TΨ-loops and the hinge between the D-stem and the anticodon stem. They constitute a network of modified residues that control the tRNA folding and conformational dynamics of the molecule core. Such interplay within the network of modified residues of the tRNA molecule is conserved in tRNA molecules from all 3 domains of life (reviewed in refs:Citation5,30).
Figure 10. Frequency of modification marked on the structure of yeast tRNAPhe (PDB ID: 1ehz). The color scale from blue via white to red indicates the percentage in which each position is modified when all sequenced tRNAs are considered. The arrow indicates the interface of interactions between the D- and TΨ-loops.
The first nucleoside of the anticodon (position 34), most often occupied by various modified uridines, and the conserved purine in position 37, 3′-adjacent to the anticodon, are also on the list of most frequently modified tRNA residues (255 and 426 times, respectively). In contrast with positions 20, 54 and 55 that present a limited number of distinct conserved modifications, positions 34 and 37 present the largest variety of modified (often hypermodified) nucleoside derivatives: 29 different residues in position 34 and 13 different residues in position 37, which corresponds to about 70% of all modifications that are found in fully sequenced tRNA molecules (42 out of 60 different modified residues present in the data set of fully sequenced tRNA molecules included in this study). Despite their very different chemical structures, the functions of these modifications are likely the same. Irrespective of the organism or tRNA species, they fine-tune mRNA decoding by reducing conformational dynamics of the loop and ordering the anticodon branch structure (reviewed in ref.Citation3). The modifications of residue 34 in tRNA allow for or prevent certain codon-anticodon interactions (reviewed ref.Citation4). Some of these modifications are often specific for a very small subset of tRNAs, which gives them a potential to play a role in regulating the rate of translation elongation (for examples see refs:Citation8,31).
Other positions in tRNA are much less frequently modified and the functions of these modifications are still largely unknown. Only a few positions have never been found modified in any tRNA sequenced so far (these are positions 5, 11, 23, 24, 33, 42, 43, 45, 53, 59, 62, 63, 73–76 and the majority of positions from the variable loop: 13e, 15–17e, 1e, 3–5e, 21–27e).
Finally, the prevalence of basic modifications (methylation, isomerization, reduction, thiolation and deamination) of nucleotide residues in fully sequenced tRNAs was also examined. The most abundant modification types in tRNAs are Ψ, introduced by isomerization of U (and its derivatives: 1-methylpseudouridine, m1Ψ or 2′-O-methylpseudouridine, Ψm), D (introduced by reduction of U) and 2′-O-methyl derivatives of U, C, G and A. They are found in 1379, 901 and 424 out of 5293 modified positions in all tRNAs considered in this work, respectively (Supplementary Table 4).
Characteristic post-transcriptional modification patterns in different phyla
Although the most conserved modifications are present in all 3 domains of life, certain patterns of modifications in less conserved positions and/or the presence of certain derivatives of basic types of modifications are correlated with particular phyla. A number of features distinguish prokaryotic and eukaryotic tRNAs. Prokaryotes typically have acceptor stem and variable loop unmodified and they have 4-thiouridine (s4U) in the single-stranded linker between the acceptor stem and the D-loop. Five-methoxyuridine (mo5U) and uridine 5-oxyacetic acid (cmo5U) in the position 34 of the anticodon loop are also characteristic prokaryotic features.
Among tRNAs from prokaryotes, Euryarchaeota (represented in this study mostly by Halobacteria) have special features (). In the D-loop these tRNAs lack the conserved Gm in position 18. Euryarchaeota do not have modified residues in tRNA positions 17a, 20a and 20b and they do not have D in position 20. Instead, they have archaeosine (G+) in position 15, a modification characteristic for Euryarchaeota only. Besides, Euryarchaeota lack 2 relatively conserved modifications located 5′ to the T-arm: 7-methylguanosine (m7G) in position 46 and a modified variant of U47. While both Euryarchaeota and all Eukaryota have 5-methylcytidine (m5C) residue(s) in the 5′ region of the T-arm, only tRNAs from Euryarchaeota contain m5C not only in positions 48–50 but also 51 and 52. The TΨ-loop of tRNAs from Euryarchaeota is heavily modified and its characteristic feature is the presence of m1Ψ in position 54, which is otherwise usually occupied by m5U and its derivatives. The presence of 2′-O-methylcytidine (Cm) in position 56 and 1-methylinosine (m1I) in position 57 are also characteristic for this group of organisms.
Common features of all Bacteria are the lack of base methylations of G in positions 9 and/or 10, and 26, and the lack of m5C methylation of cytidine residues 5′ to the T-arm. Gram-positive bacteria have in general the lowest number of modified positions in their tRNAs and tend to have only the most conserved ones (). A characteristic modification present in Gram-positive bacteria only is 1-methyladenosine (m1A) in position 22. Gram-negative bacteria (), in contrast to Gram-positive bacteria and single cell organisms and Fungi and Metazoa, have 3-(3-amino-3-carboxypropyl)uridine (acp3U) in position 47; the other 2 groups have D47, while the remaining 4 phylogenetic groups analyzed in this study have a mixture of both. Some species from the Gram-negative bacteria group, all belonging to thermophilic species, possess a characteristic m5s2U in position 54 (a derivative of m5U often found in that position). The synergistic effect of both C5-methylation and 2-thiolation of uridine help to stabilize the 3D-structure of tRNAs at high temperature.Citation32
Eukaryotic tRNAs have in general a higher number of less conserved modified positions, compared to prokaryotic ones. Their specific feature is the presence of N4-acetylcytidine (ac4C) in position 12 and Ψ in position 27 (which occurs especially in the group of cytosol of eukaryotic single cell organisms, Fungi and Metazoa, ). In tRNAs from cytosol of eukaryotic single cell organisms, Fungi and Metazoa, and cytosol of Viridiplantae () also a variety of residues methylated at the 2′-hydroxyl group of the ribose are present in position 4. Only in the cytosolic tRNAs from single cell organisms, Fungi and Metazoa, the position 34 of the anticodon was found to be hypermodified to mannosyl- or galactosyl-queuosine. These modifications were not found in other groups of organisms. Two′-O-ribosyladenosine (phosphate) (Ar(p)) and 2′-O-ribosylguanosine (phosphate) (Gr(p)) in position 64 are also characteristic for this group of organisms only. A large fraction of these tRNAs have also Ψ13, which on the other hand is totally absent in plastids and Gram-positive bacteria.
Plastids have tRNAs with a relatively low number of modifications, compared to other groups of eukaryotic tRNAs analyzed in this work (). Their acceptor stem is not modified and their TΨ-loop has only the most conserved m5U54 and Ψ55. They do not have acp3U in the D-loop and 2′-O-methyl uridine (Um) in position 44 – 2 modifications present in all other groups of eukaryotic tRNAs. They also lack m1A58, which is otherwise present in all 3 domains of life.
Mitochondria have the highest number of modified positions that are not conserved (). Especially their acceptor stem is rich in Ψ residues. However, at the same time, some of the mitochondrial tRNAs lack the conserved Ψ55. The number of modifications located in positions 46–50, 5′ to the T-arm, is also relatively low in mitochondrial tRNAs. Taurine-containing residues in the first position of the anticodon are found only in mitochondria.Citation33
Discussion
Comparative analysis using tRNAmodviz suggests identity of unknown modifications
tRNAmodviz facilitates the identification of conserved modified residues in user-defined subsets of tRNA molecules. The presented comparative analysis of tRNA modification patterns, conducted using tRNAmodviz, allowed us to make predictions about the chemical structure of residues known to be modified, whose identity has not been determined experimentally.
In Euryarchaeota () an unknown modified G (xG) in position 15 is most probably G+, as in many other archaeal tRNAs. For the same reason, xA in position 57 in initiator tRNAMet from Thermoplasma acidophilum could be m1I. However, based on the original workCitation27 and the analysis of modifications of tRNAs from thermophilic archaeaCitation34 there is a possibility that the hypermodified m1Im is present in this position in this tRNAMet.
In Gram-positive bacteria (), xA at position 37 in tRNAThr from Mycoplasma mycoides is most probably the N6-threonylcarbamoyladenosine (t6A), like in all tRNAThr from this group of organisms (M. capricolum and B. subtilis). However, in 2 tRNATrp from Spiroplasma citri, xA37 is most probably N6-methyladenosine (m6A), like in M. capricolum tRNATrp and not i6A like in B. subtilis, since Spiroplasma are more closely related to Mycoplasma than to Bacilli.Citation21
In the cytoplasmic tRNAs from eukaryotic single cell organisms, Fungi and Metazoa (), xG most likely indicates different modifications in different positions. It is most probably m1G in position 9, m2G or m2,2G in position 26, and Gr(p) in position 64. xA in position 14 and 58 is most probably m1A. xU in position 20a in tRNAIle from Bombyx mori is most probably D, since all tRNAIle from this group have D20a. However, in tRNAVal from Drosophila melanogaster xU20 can be either D or acp3U. xU in positions 44 and 47 are most probably Um and D, respectively. xC34 in 2 tRNALeu from Rattus norvegicus are most probably 5-formyl-2′-O-methylcytidine (f5Cm), like in tRNALeu from Bos taurus. xC in positions 38, 48 and 49 are most probably m5C.
In cytosolic tRNAs from Viridiplantae (), xU in position 4 is probably Um, since in cytosolic tRNAs from eukaryotes residues in this position tend to be methylated at the 2′-hydroxyl group of the ribose. xC49, present in initiator tRNA from Lupinus luteus, could be m5C, but this prediction is not very confident because xC49 has been reported along with m5C in the neighboring position 48 (T. Zwierzynski, A. J. Rafalski, K. Gulewicz, W. Krzyzosiak (1983) personal communication in tRNAdbCitation25). xG64, present in initiator tRNA from Scenedesmus obliquus, is most probably Gr(p), like in initiator tRNA in yeast; in the original work it was reported to be “a hypermodified Gm.”Citation35
In mitochondria (), xU29 and xU40 are most probably Ψ, and xU39 is either Ψ or m1Ψ. xU34 in tRNALys from Mesocricetus auratus and R. norvegicus are most probably 5-taurinomethyl-2-thiouridine (τm5s2U) like in other mammals. On the other hand, xU34 in tRNATrp from R. norvegicus is likely 5-taurinomethyluridine (τm5U). xU34 in tRNAArg and tRNAGly from S. cerevisiae is likely 5-carboxymethylaminomethyluridine (cmnm5U). According to the original publications, xA37 in Oenothera sp. tRNAPhe, Phaseolus vulgaris tRNATrp and tRNATyr and S. cerevisiae tRNATrp should be either N6-isopentenyladenosine (i6A) or 2-methylthio-N6-isopentenyladenosine (ms2i6A).Citation36-38 Since in all known mitochondrial tRNATrp sequences A in position 37 is modified to ms2i6A, it is the most probable modification expected in the 2 tRNATrp from P. vulgaris and yeast, as well as in tRNATrp from Neurospora crassa. For the remaining 5 tRNAs, for which the identity of modified A37 is not known (tRNATyr from N. crassa, P. vulgaris and Tetrahymena pyriformis, and tRNAPhe from Oenothera sp and T. pyriformis) and which have different amino acid specificity, the identity of xA37 cannot be so easily predicted. Finally, in plastids () xU17 is most probably D and xC34 is most probably Cm. For a few additional unknown modifications, especially the less conserved ones, it remains difficult to propose the most probable identity based solely on currently available data.
The presented comparative analysis of tRNA modification patterns and further use of tRNAmodviz could also help in correlating the modification patterns with the conservation of enzymes that are known or predicted to be responsible for particular modifications, which is however beyond of the scope of this article.
Changes in the tRNA sequencing landscape during the last 2 decades
The current dataset of sequenced tRNAs has grown by around one fifth during the past 20 y (497 versus 584 sequences) and the overall picture of modification patterns has not changed dramatically compared to the previous overview.Citation13 Though new types of modified residues have been discovered in bulk tRNAs of a few archaea (reviewed in refs:Citation18,19), these discoveries were unfortunately not followed by tRNA sequencing and hence the locations of these residues have not been included in this study. From separate figures for Gram-positive and Gram-negative bacteria groups, differences in tRNA modification patterns between these organisms can be distinguished. It is clearly visible that Gram-positive bacteria tend to have a minimal set of modified residues. This observation was expected in parasitic bacteria with reduced genome and RNA modification apparatus like the Mollicutes class derived from Gram-positive bacteria.Citation20,21 In the presented compilation tRNAs from cytosol of eukaryotic single cell organisms, Fungi and Metazoa, were combined into one group. The number of known tRNA sequences of this origin has grown slightly since 1995 (from 160 to 173), new modifications of the first position in the anticodon loop have been discovered (f5C and f5Cm), but the overall pattern of modification presented in this work is not significantly different from the one presented in 1995.
On the other hand, the number of known tRNA sequences from Viridiplantae cytoplasm has increased by more than a half. Some new modified positions were identified, e.g. position 40, e11, 60 and 61, and for some positions that were known to be modified, new modifications were discovered such as N6-(cis-hydroxyisopentenyl)adenosine (io6A) in position 37. For the plastids, however, only 6 more sequences are known (32 vs 38). Though the newly discovered sequences contain modifications that could be considered as “unusual” in the context of the whole group, e.g., Gm in position 19 or queuosine in position 34, the picture of modification of these tRNAs has not changed much.
The most significant growth in the amount of data available can be observed for mitochondrial tRNAs. The number of known tRNA sequences of this origin has grown by half and several new positions are now known to contain modified bases (e.g. positions 2, 3 and 6). In particular mitochondria-specific base modifications τm5U and τm5s2U were discovered,Citation33 and linked with 2 major classes of mitochondrial diseases, MELAS (mitochondrial myopathy, encephalopathy, lactic acidosis, and stroke-like episodes) and MERRF (myoclonus epilepsy associated with ragged red fibers).Citation39 Taken together, these observations suggest that the general patterns of tRNA modifications as depicted above are not likely to change greatly as more tRNA sequences become available.
Conclusions and future perspectives
The focus of the scientific community on high-throughput DNA sequencing results in a deluge of tRNA gene sequence data (i.e. tDNA sequences), but the efforts aiming to provide a complete repertoire of tRNA sequences for any particular organism remain limited to only a few model organisms (E. coli, S. cerevisiae, H. volcanii, M. capricolum and most recent addition to this set: L. lactis). The hope is that studies of other organisms are underway, and the complete sequences will soon become available.
New, powerful, experimental techniques have allowed for identification of novel types of hypermodified nucleosides that were extremely difficult to identify in the past (reviewed in ref.Citation10). Among them are those found at position 34 of tRNAs of various organisms: 5-formylcytidine (f5C), 2′-O-methyl f5C (f5Cm), glutamyl-queuosine (GluQ), 5-taurinomethyluridine (τm5U) and its 2-thio-derivative (τm5s2U), agmatidine (C+), 5-cyanomethyluridine (cnm5U), 2-geranylthiouridine (ges2U) and its derivatives 5-aminomethyl-2-geranylthiouridine (nm5ges2U), 5-carboxymethylaminomethyl-2-geranylthiouridine (cmnm5ges2U) and 5-methylaminomethyl-2-geranylthiouridine (mnm5ges2U), raising the total number of post-transcriptional modified nucleotides found in naturally occurring tRNAs to up to about 100 out of a total of about 150 distinct chemical structures found in all types of cellular RNAs (including the intermediates of the multi-step enzymatic pathways). This is probably not an upper limit and more of these ‘tRNA decorations’ will still be found after sequencing tRNAs from yet poorly studied organisms.
For some of the newly discovered modifications the exact location of the modified residue within tRNA sequence have not been determined yet, and as a result they were not included in the compilation presented in this work. Also, one of the recent discoveries shows that most probably the widely conserved and well documented modification of position 37, t6A, is a hydrolyzed (artifactual) derivative of hypermodified cyclic N6-threonylcarbamoyladenosine (ct6A). Thus t6A may not necessarily occur in vivo, but is generated only during the experimental preparation of tRNA molecules.Citation40 Though in the data presented here we keep t6A, as it is in the original databases, it should be emphasized that, especially in Gram-negative bacteria, the naturally occurring derivative in position 37 of anticodon loop is most probably ct6A.
One big challenge in the analysis of tRNA modifications concerns the “degree” of modification one can expect for the hypermodified residues (understood as the fraction of experimentally characterized tRNA sequences that contain this modified residue). Partial character of certain modified bases has been fully documented in some cases,Citation41-45 but the information can usually be retrieved only from the original publications describing the sequencing data, and it is rarely quantitative. Partial modification was often pointed out when the tRNA to be sequenced originated from cells cultivated in different physiological conditions (e.g., aerobic vs anaerobic, different temperature of growth, availability of intermediate metabolites or cofactors, stress conditions, tumor malignancy etc.).Citation42,46-49 Currently, new, more sensitive methodologies used to determine whether a given nucleotide is modified or not, allow for more quantitative detection of such cases, which are now reported to be more frequent than initially anticipated.Citation8,31,50 It was proposed that such situation may reflect the linkage between tRNA modification processes and regulation of metabolism, e.g. via the influence on the translation of certain codons in mRNA. It should be kept in mind that hypermodifications (especially of U34) often require participation of many enzymes and different intermediates of the final product may accumulate depending on cell growth conditions and thus the availability of specific metabolites. For these reasons it may be difficult to elucidate the exact identity of the modification for any particular tRNA.
Not every type of post-transcriptional modification of the tRNA molecule can be detected by studying the presence of non-canonical modified residues. In addition to phylogenetically dispersed A-to-I editing of the first anticodon base, C-to-U editing was described in organellar tRNAs,Citation51 in cytoplasmic tRNAs from trypanosomesCitation52 and in some ArchaeaCitation53 (reviewed in ref.Citation54). In this case, the deamination results in the replacement of one canonical base, C, by another, U, which is not considered as “modified” in the presented compilation.
Moreover, modified nucleosides are not exclusive to tRNA, they are also found in rRNA, mRNA and small RNA,Citation14 yet their frequency and diversity in these RNA molecules are far lower than in tRNA. They probably influence mainly tertiary interactions, increase or decrease local flexibility of the RNA structure or fine-tune a specific function by influencing interactions with other molecules (from ions to small chemical molecules to large macromolecules including proteins and other RNAs). However, their exact functions are far from being fully understood. A classic example of rRNA modification function is the prevention of binding of antibiotics that would otherwise block the function of the ribosome.Citation55,56 Unfortunately, for non-tRNA molecules, usually only limited sets of homologous RNAs were sequenced and had their modified nucleoside content fully characterized. It would be very desirable to obtain more data on modified RNA sequences, to enable comparative analyses to be performed (such as one for tRNAs in this work), from which the influence of frequently modified residues on the cellular functions performed by those RNAs could be inferred.
Finally, the understanding of the whole process of RNA modification, not only tRNA modification, within an evolutionary framework remains an interesting challenge: do certain modified nucleotides correspond to relics of a prebiotic RNA World or do they correspond to a more elaborate program allowing progressive acquisition of discrete new functions? It is possible that both alternatives are true to some extent, which may reflect the special place of tRNA modifications in the biology of organisms from all domains of life.
Materials and Methods
Data collection
Only sequences obtained via tRNA sequencing (full length) were included in the data set – tRNA sequences predicted from DNA sequencing data were not considered. tRNA sequences and modification information were obtained from the MODOMICS databaseCitation24 (data available on 18 June 2014 were included). MODOMICS contains all sequences available in tRNAdbCitation25 as of June 2012 (some of which were subsequently curated manually based on data in the literature) and additional sequences retrieved from the literature. Both tRNAdb and MODOMICS databases derive the sequences and modification information from scientific literature and personal communications with users. The MODOMICS database is periodically updated based on newly published data.Citation24,57,58 For calculations, unknown nucleosides were treated as the fifth unmodified nucleoside type. When a given position of tRNA was reported to be incompletely modified, it was considered as fully modified. In the case of modified nucleotides resulting from sequential multienzymatic reactions, only the most abundant species was considered. tRNA sequences were divided into groups according to the phylogenetic grouping of their hosts, regardless of phylogenetic relationships between tRNA molecules themselves. The exact list of species assigned to each group with numbers of tRNA sequences considered is provided as Supplementary Table 1.
Server implementation
Statistics and data analysis were implemented in Python and JavaScript. Web Service backend was created using Django 1.6 framework with SQLite database engine. Frontend was created using jQuery and jQuery UI. JavaScript InfoVis Toolkit (http://thejit.org/) - [cite: Author: Nicolas Garcia Belmonte (http://philogb.github.com/)] - a JavaScript visualization library was used to prepare tRNA visualization module, amcharts.js library (http://www.amcharts.com/) was used to create interactive plots.
Disclosure of Potential Conflicts of Interest
No potential conflicts of interest were disclosed.
Tables.zip
Download Zip (65 KB)Acknowledgments
We thank Dr Elzbieta Purta, Dr Grzegorz Lach, Dr Lukasz Kozlowski, and Martyna Hogendorf for critical reading of the manuscript and useful comments.
Funding
This analysis was supported mainly by the Foundation for Polish Science (FNP, grant TEAM/2009–4/2 to J.M.B.). M.A.M. was additionally supported by funds from the European Union (European Social Fund - stipend for Mazovian PhD students Human Capital Operational Program 2007–2013). A.O. was additionally supported by the FNP (grant MPD/2009–3/2, project co-financed from the European Union - Regional Development Fund). H.G. is an emeritus scientist at the CNRS in France. J.M.B. was additionally supported by the Polish National Science Centre (NCN, grant 2012/04/A/NZ2/00455). Research on structural aspects of RNA in the Bujnicki lab has been also supported by grants from the European Union's Seventh Framework Program (REGPOT grant FishMed; grant agreement no. 535 316125), the Polish Ministry of Science and Higher Education (MNiSW, grant POIG.02.03.00-00-003/09) and by the infrastructure financed by the European Union—the European Regional Development Fund within the Operational Program “Innovative economy” for 2007–2013 (CePT, grant 540 POIG.02.02.00-14-024/08-00).
References
- Motorin Y, Helm M. tRNA stabilization by modified nucleotides. Biochemistry 2010; 49:4934-44; PMID:20459084; http://dx.doi.org/10.1021/bi100408z
- Giegé R, Lapointe J. Transfer RNA aminoacylation and modified nucleosides. In: Grosjean H, ed. DNA and RNA Modification Enzymes: Structure, Mechanism, Function and Evolution. Austin, Texas, USA: Landes Bioscience, 2009:476-92
- Agris PF. Bringing order to translation: the contributions of transfer RNA anticodon-domain modifications. EMBO Rep 2008; 9:629-35; PMID:18552770; http://dx.doi.org/10.1038/embor.2008.104
- Grosjean H, de Crecy-Lagard V, Marck C. Deciphering synonymous codons in the three domains of life: co-evolution with specific tRNA modification enzymes. FEBS letters 2010; 584:252-64; PMID:19931533; http://dx.doi.org/10.1016/j.febslet.2009.11.052
- Helm M, Alfonzo JD. Posttranscriptional RNA Modifications: playing metabolic games in a cell's chemical Legoland. Chem Biol 2014; 21:174-85; PMID:24315934; http://dx.doi.org/10.1016/j.chembiol.2013.10.015
- Gehrig S, Eberle ME, Botschen F, Rimbach K, Eberle F, Eigenbrod T, Kaiser S, Holmes WM, Erdmann VA, Sprinzl M, et al. Identification of modifications in microbial, native tRNA that suppress immunostimulatory activity. J Exp Med 2012; 209:225-33; PMID:22312113; http://dx.doi.org/10.1084/jem.20111044
- Yi C, Pan T. Cellular dynamics of RNA modification. Acc Chem Res 2011; 44:1380-8; PMID:21615108; http://dx.doi.org/10.1021/ar200057m
- Dedon PC, Begley TJ. A system of RNA modifications and biased codon use controls cellular stress response at the level of translation. Chem Res Toxicol 2014; 27:330-7; PMID:24422464; http://dx.doi.org/10.1021/tx400438d
- Torres AG, Batlle E, Ribas de Pouplana L. Role of tRNA modifications in human diseases. Trends Mol Med 2014; 20:306-14; PMID:24581449; http://dx.doi.org/10.1016/j.molmed.2014.01.008
- Björk G, Hagervall T. 2014. Transfer RNA Modification: Presence, Synthesis, and Function. EcoSal Plus 2014; doi:10.1128/ecosalplus.ESP-0007-2013
- Björk GR. Transfer RNA modification in different organisms. In: Baltscheffsky H, Jörnvall H, Rigler R, eds. Molecular Evolution of Life. Cambridge: Cambridge University Press, 1986:91-5.
- McCloskey JA. Nucleoside modification in archaebacterial transfer RNA. Syst Appl Microbiol 1986; 7:246-52; http://dx.doi.org/10.1016/S0723-2020(86)80014-5
- Grosjean H, Sprinzl M, Steinberg S. Posttranscriptionally modified nucleosides in transfer RNA: their locations and frequencies. Biochimie 1995; 77:139-41; PMID:7541252; http://dx.doi.org/10.1016/0300-9084(96)88117-X
- Grosjean H. Nucleic acids are not boring long polymers of only four types of nucleotides: a Guided tour. In: Grosjean H, ed. DNA and RNA Modification Enzymes: Structure, Mechanism, Function and Evolution. Austin, Texas, USA: Landes Bioscience, 2009:1-18.
- El Yacoubi B, Bailly M, de Crecy-Lagard V. Biosynthesis and function of posttranscriptional modifications of transfer RNAs. Annu Rev Genet 2012; 46:69-95; PMID:22905870; http://dx.doi.org/10.1146/annurev-genet-110711-155641
- Hopper AK. Transfer RNA post-transcriptional processing, turnover, and subcellular dynamics in the yeast saccharomyces cerevisiae. Genetics 2013; 194:43-67; PMID:23633143; http://dx.doi.org/10.1534/genetics.112.147470
- Phizicky EM, Hopper AK. tRNA biology charges to the front. Genes Dev 2010; 24:1832-60; PMID:20810645; http://dx.doi.org/10.1101/gad.1956510
- Phillips G, de Crecy-Lagard V. Biosynthesis and function of tRNA modifications in archaea. Curr Opin Microbiol 2011; 14:335-41; PMID:21470902; http://dx.doi.org/10.1016/j.mib.2011.03.001
- Grosjean H, Gupta R, Maxwell S. Modified nucleotides in archaeal RNAs. In: Blum P, ed. Archaea: New Models for Prokaryotic Biology. Norfolk, UK: Caister Academic Press, 2008:171-96.
- de Crecy-Lagard V, Marck C, Brochier-Armanet C, Grosjean H. Comparative RNomics and modomics in mollicutes: prediction of gene function and evolutionary implications. IUBMB Life 2007; 59:634-58; PMID:17852564; http://dx.doi.org/10.1080/15216540701604632
- Grosjean H, Breton M, Sirand-Pugnet P, Tardy F, Thiaucourt F, Citti C, Barre A, Yoshizawa S, Fourmy D, de Crecy-Lagard V, et al. Predicting the minimal translation apparatus: lessons from the reductive evolution of mollicutes. PLoS genetics 2014; 10:e1004363; PMID:24809820; http://dx.doi.org/10.1371/journal.pgen.1004363
- Puri P, Wetzel C, Saffert P, Gaston KW, Russell SP, Cordero Varela JA, van der Vlies P, Zhang G, Limbach PA, Ignatova Z, et al. Systematic identification of tRNAome and its dynamics in lactococcus lactis. Mol Microbiol 2014; 93:944-56; PMID:25040919; http://dx.doi.org/10.1111/mmi.12710
- Suzuki T, Suzuki T. A complete landscape of post-transcriptional modifications in mammalian mitochondrial tRNAs. Nucleic Acids Res 2014; 42:7346-57; PMID:24831542; http://dx.doi.org/10.1093/nar/gku390
- Machnicka MA, Milanowska K, Osman Oglou O, Purta E, Kurkowska M, Olchowik A, Januszewski W, Kalinowski S, Dunin-Horkawicz S, Rother KM, et al. MODOMICS: a database of RNA modification pathways–2013 update. Nucleic Acids Res 2013; 41:D262-7; PMID:23118484; http://dx.doi.org/10.1093/nar/gks1007
- Juhling F, Morl M, Hartmann RK, Sprinzl M, Stadler PF, Putz J. tRNAdb 2009: compilation of tRNA sequences and tRNA genes. Nucleic Acids Res 2009; 37:D159-62; PMID:18957446; http://dx.doi.org/10.1093/nar/gkn772
- Steinberg S, Misch A, Sprinzl M. Compilation of tRNA sequences and sequences of tRNA genes. Nucleic Acids Res 1993; 21:3011-5; PMID:7687348; http://dx.doi.org/10.1093/nar/21.13.3011
- Kuchino Y, Ihara M, Yabusaki Y, Nishimura S. Initiator tRNAs from archaebacteria show common unique sequence characteristics. Nature 1982; 298:684-5; PMID:6178978; http://dx.doi.org/10.1038/298684a0
- Guthrie C, Scholla CA, Yesian H, Abelson J. The nucleotide sequence of threonine transfer RNA coded by bacteriophage T4. Nucleic Acids Res 1978; 5:1833-44; PMID:353731; http://dx.doi.org/10.1093/nar/5.6.1833
- Becker HF, Motorin Y, Florentz C, Giege R, Grosjean H. Pseudouridine and ribothymidine formation in the tRNA-like domain of turnip yellow mosaic virus RNA. Nucleic Acids Res 1998; 26:3991-7; PMID:9705510; http://dx.doi.org/10.1093/nar/26.17.3991
- Jackman JE, Alfonzo JD. Transfer RNA modifications: nature's combinatorial chemistry playground. Wiley Interdiscip Rev RNA 2013; 4:35-48; PMID:23139145; http://dx.doi.org/10.1002/wrna.1144
- Chan CT, Pang YL, Deng W, Babu IR, Dyavaiah M, Begley TJ, Dedon PC. Reprogramming of tRNA modifications controls the oxidative stress response by codon-biased translation of proteins. Nat Commun 2012; 3:937; PMID:22760636; http://dx.doi.org/10.1038/ncomms1938
- Shigi N, Sakaguchi Y, Suzuki T, Watanabe K. Identification of two tRNA thiolation genes required for cell growth at extremely high temperatures. J Biol Chem 2006; 281:14296-306; PMID:16547008; http://dx.doi.org/10.1074/jbc.M511675200
- Suzuki T, Miyauchi K, Suzuki T, Yokobori S, Shigi N, Kondow A, Takeuchi N, Yamagishi A, Watanabe K. Taurine-containing uridine modifications in tRNA anticodons are required to decipher non-universal genetic codes in ascidian mitochondria. J Biol Chem 2011; 286:35494-8; PMID:21873425; http://dx.doi.org/10.1074/jbc.M111.279810
- Edmonds CG, Crain PF, Gupta R, Hashizume T, Hocart CH, Kowalak JA, Pomerantz SC, Stetter KO, McCloskey JA. Posttranscriptional modification of tRNA in thermophilic archaea (archaebacteria). J Bacteriol 1991; 173:3138-48; PMID:1708763
- Olins PO, Jones DS. Nucleotide sequence of scenedesmus obliquus cytoplasmic initiator tRNA. Nucleic Acids Res 1980; 8:715-29; PMID:6903883
- Marechal-Drouard L, Ramamonjisoa D, Cosset A, Weil JH, Dietrich A. Editing corrects mispairing in the acceptor stem of bean and potato mitochondrial phenylalanine transfer RNAs. Nucleic Acids Res 1993; 21:4909-14; PMID:8177739; http://dx.doi.org/10.1093/nar/21.21.4909
- Marechal L, Guillemaut P, Weil JH. Sequences of two bean mitochondria tRNAs(tyr) which differ in the level of post-transcriptional modification and have a prokaryotic-like large extra-loop. Plant Mol Biol 1985; 5:347-51; PMID:24306988; http://dx.doi.org/10.1007/BF00037555
- Marechal L, Guillemaut P, Grienenberger JM, Jeannin G, Weil JH. Sequence and codon recognition of bean mitochondria and chloroplast tRNAsTrp: evidence for a high degree of homology. Nucleic Acids Res 1985; 13:4411-6; PMID:3847869; http://dx.doi.org/10.1093/nar/13.12.4411
- Suzuki T, Suzuki T, Wada T, Saigo K, Watanabe K. Taurine as a constituent of mitochondrial tRNAs: new insights into the functions of taurine and human mitochondrial diseases. Embo J 2002; 21:6581-9; PMID:12456664; http://dx.doi.org/10.1093/emboj/cdf656
- Miyauchi K, Kimura S, Suzuki T. A cyclic form of N6-threonylcarbamoyladenosine as a widely distributed tRNA hypermodification. Nat Chem Biol 2013; 9:105-11; PMID:23242255; http://dx.doi.org/10.1038/nchembio.1137
- Dumelin CE, Chen Y, Leconte AM, Chen YG, Liu DR. Discovery and biological characterization of geranylated RNA in bacteria. Nat Chem Biol 2012; 8:913-9; PMID:22983156; http://dx.doi.org/10.1038/nchembio.1070
- Roe BA, Stankiewicz AF, Rizi HL, Weisz C, DiLauro MN, Pike D, Chen CY, Chen EY. Comparison of rat liver and walker 256 carcinosarcoma tRNAs. Nucleic Acids Res 1979; 6:673-88; PMID:424309; http://dx.doi.org/10.1093/nar/6.2.673
- Winkler ME. Genetics and regulation of base modification in the tRNA and rRNA of prokaryotes and eukaryotes. In: Grosjean H, Benne R, eds. Modification and Editing of RNA. Washington DC: ASM Press, 1998:441-69.
- Persson BC. Modification of tRNA as a regulatory device. Mol Microbiol 1993; 8:1011-6; PMID:7689685; http://dx.doi.org/10.1111/j.1365-2958.1993.tb01645.x
- Bjork GR, Rasmuson T. Links between tRNA modification and metabolism and modified nucleosides as tumor markers. In: Grosjean H, Benne R, eds. Modification and Editing of RNA. Washington DC: ASM Press, 1998:471-91.
- Kuchino Y, Borek E, Grunberger D, Mushinski JF, Nishimura S. Changes of post-transcriptional modification of wye base in tumor-specific tRNAPhe. Nucleic Acids Res 1982; 10:6421-32; PMID:6924749; http://dx.doi.org/10.1093/nar/10.20.6421
- Kowalak JA, Dalluge JJ, McCloskey JA, Stetter KO. The role of posttranscriptional modification in stabilization of transfer RNA from hyperthermophiles. Biochemistry 1994; 33:7869-76; PMID:7516708; http://dx.doi.org/10.1021/bi00191a014
- Moukadiri I, Garzon MJ, Bjork GR, Armengod ME. The output of the tRNA modification pathways controlled by the escherichia coli MnmEG and MnmC enzymes depends on the growth conditions and the tRNA species. Nucleic Acids Res 2014; 42:2602-23; PMID:24293650; http://dx.doi.org/10.1093/nar/gkt1228
- Buck M, Ames BN. A modified nucleotide in tRNA as a possible regulator of aerobiosis: synthesis of cis-2-methyl-thioribosylzeatin in the tRNA of Salmonella. Cell 1984; 36:523-31; PMID:6362893; http://dx.doi.org/10.1016/0092-8674(84)90245-9
- Chan CT, Dyavaiah M, DeMott MS, Taghizadeh K, Dedon PC, Begley TJ. A quantitative systems approach reveals dynamic control of tRNA modifications during cellular stress. PLoS Genet 2010; 6:e1001247; PMID:21187895; http://dx.doi.org/10.1371/journal.pgen.1001247
- Alfonzo JD, Blanc V, Estevez AM, Rubio MA, Simpson L. C to U editing of the anticodon of imported mitochondrial tRNA(Trp) allows decoding of the UGA stop codon in leishmania tarentolae. Embo J 1999; 18:7056-62; PMID:10601027; http://dx.doi.org/10.1093/emboj/18.24.7056
- Rubio MA, Ragone FL, Gaston KW, Ibba M, Alfonzo JD. C to U editing stimulates a to I editing in the anticodon loop of a cytoplasmic threonyl tRNA in trypanosoma brucei. J Biol Chem 2006; 281:115-20; PMID:16269406; http://dx.doi.org/10.1074/jbc.M510136200
- Randau L, Stanley BJ, Kohlway A, Mechta S, Xiong Y, Soll D. A cytidine deaminase edits C to U in transfer RNAs in archaea. Science 2009; 324:657-9; PMID:19407206; http://dx.doi.org/10.1126/science.1170123
- Su AA, Randau L. A-to-I and C-to-U editing within transfer RNAs. Biochemistry (Mosc) 2011; 76:932-7; PMID:22022967; http://dx.doi.org/10.1134/S0006297911080098
- Vester B, Long KS. Antibiotic Resistance in Bacteria Caused by Modified Nucleosides in 23S Ribosomal RNA In: Grosjean H, ed. DNA and RNA Modification Enzymes: Structure, Mechanism, Function and Evolution. Austin, Texas, USA: Landes Bioscience, 2009:537-49.
- Conn GL, Savic M, Macmaster R. Antibiotic resistance in bacteria through modification of nucleosides in 16S ribosomal RNA. In: Grosjean H, ed. DNA and RNA Modification Enzymes: Structure, Mechanism, Function and Evolution. Austin, Texas, USA: Landes Bioscience, 2009:525-36.
- Dunin-Horkawicz S, Czerwoniec A, Gajda MJ, Feder M, Grosjean H, Bujnicki JM. MODOMICS: a database of RNA modification pathways. Nucleic Acids Res 2006; 34:D145-9; PMID:16381833; http://dx.doi.org/10.1093/nar/gkj084
- Czerwoniec A, Dunin-Horkawicz S, Purta E, Kaminska KH, Kasprzak JM, Bujnicki JM, Grosjean H, Rother K. MODOMICS: a database of RNA modification pathways. 2008 update. Nucleic Acids Res 2009; 37:D118-21; PMID:18854352; http://dx.doi.org/10.1093/nar/gkn710
- Roberts RJ. Staphylococcal transfer ribonucleic acids. II. sequence analysis of isoaccepting glycine transfer ribonucleic acids IA and IB from staphylococcus epidermidis texas 26. J Biol Chem 1974; 249:4787-96; PMID:4367807
- Randerath E, Gupta RC, Morris HP, Randerath K. Isolation and sequence analysis of two major leucine transfer ribonucleic acids (anticodon mm-A-A) from a rat tumor, morris hepatoma 5123D. Biochemistry 1980; 19:3476-83; PMID:6773543; http://dx.doi.org/10.1021/bi00556a011
- Heckman JE, Alzner-Deweerd B, RajBhandary UL. Interesting and unusual features in the sequence of neurospora crassa mitochondrial tyrosine transfer RNA. Pro Nat Acad Sci U S A 1979; 76:717-21; PMID:154673; http://dx.doi.org/10.1073/pnas.76.2.717