| Nucleic Acids Research | Pages |
The tRNA(guanine-26,N2-N2) methyltransferase (Trm1) from the hyperthermophilic archaeon Pyrococcus furiosus: cloning, sequencing of the gene and its expression in Escherichia coli
Introduction
Materials And Methods
Reagents, enzymes and plasmids
Strains
Preparation of P.furiosus genomic DNA
Preparation of a radioactive DNA probe for library screening
Mini-library construction and screening
Nucleotide sequence determination
Cloning and expression of P.furiosus tRNA(guanine-26,N2-N2) methyltransferase in E.coli
Preparation of E.coli cell-free extract
Preparation of total E.coli tRNA, post-labelling with 32P and analysis of modified nucleotides
Preparation of tRNA transcripts
In vitro enzymatic assay
Results
Cloning and sequencing of the pfTRM1 gene encoding tRNA(guanine-26,N2-N2) methyltransferase
Functional expression of the recombinant pfTRM1 gene product in E.coli
Identification of the activity of the recombinant pfTrm1 protein
Identification of the guanine residue modified in tRNA substrates
Discussion
m2G and m22G occur at positions 10 and 26 in fully mature tRNAs, formed by distinct site-specific methyltransferases
A single gene product pfTrm1p has both mono- and dimethylating activity at G26
Sequence analysis of the pfTRM1 ORF
Sequence alignment of several Trm1-like proteins and localization of putative SAM binding motifs
Acknowledgements
References
The tRNA(guanine-26,N2-N2) methyltransferase (Trm1) from the hyperthermophilic archaeon Pyrococcus furiosus: cloning, sequencing of the gene and its expression in Escherichia coli
DDBJ/EMBL/GenBank accession no. AF051912
ABSTRACT
INTRODUCTION
All living organisms can be grouped into three evolutionarily distinct domains: Bacteria, Eukarya and Archaea (1). Recent achievements in genome sequencing projects from organisms of all three domains allowed systematic comparative sequence analysis of archaeal proteins with known or putative functions with their corresponding bacterial and eukaryotic counterparts. It appears that almost all archaeal proteins and enzymes of the central metabolism and of the metabolite uptake systems, as well as enzymes involved in cell wall biosynthesis, show remarkable similarity to bacterial homologues. On the other hand, proteins and enzymes involved in DNA replication and transcription, DNA recombination and repair, RNA translation and also protein secretion are often more closely related to respective eukaryotic homologues (2; reviewed in 3-7).
Because genes that specify functions of genetic information processing are in the main most similar to their eukaryotic counterparts, one could expect that proteins and enzymes involved in RNA processing and modification are also of the eukaryotic type. Indeed, components involved in rRNA processing and tRNA intron excision share much similarity with those of eukaryotes (8,9; reviewed in 10). However, the recently studied archaeal CCA-adding enzyme does not exhibit strong similarity with either the bacterial or the eukaryotic CCA-adding enzymes (11).
A great variety of modified nucleosides is found in tRNAs of all three domains (12). Of 80 different modified nucleosides reported in tRNAs, 36 are found in Archaea (of which nine are unique to this domain), 43 in Bacteria (16 unique) and 47 in Eukarya (21 unique) (reviewed in 13,14). Among those modified nucleosides that are present in archaeal tRNAs, a few are also found in eukaryotic tRNAs but not in any bacterial tRNAs sequenced so far [such as m2G, m22G, m1I, m5C, s2U, Am; symbols and common names of modified nucleotides are those of McCloskey and Crain (12) and the universal numbering system for tRNA positions corresponds to that adopted in the tRNA sequence databank (15)]. On the other hand, a few modified nucleotides of bacterial type which have never been detected in eukaryotic tRNAs (such as s2T, s4U, m6A) exist in archaeal tRNAs. Only a few tRNA modifications are exclusively present in archaeal tRNAs (such as m1[Psi], mimG, archaeosine and several 2[prime]-O-ribose-methylated derivatives of base-modified nucleosides). The formation of all of these modified nucleosides is catalysed by specific enzymes, for which it would be interesting to know whether they exhibit common features within the three domains of life. From the phylogenetic point of view, this knowledge could help to root the universal phylogenetic tree (3,16; see also 17) and to speculate on a stepwise chronology for the early origin of the RNA maturation machinery and the maintenance of modified nucleosides in tRNAs (18).
Recently, the complete genomic sequences of three moderately thermophilic Archaea have been determined (Methanococcus jannaschii, Methanobacterium thermoautotrophicum and Archaeoglobus fulgidus; 19-21). Comparison of the protein sequences deduced from the open reading frames (ORFs) with the known identified proteins of Bacteria and Eukarya (of which the functions have been experimentally tested) led to the identification of numerous putative homologues in Archaea. Among them are the putative modification enzymes responsible for formation of pseudouridine, N2,N2-dimethylguanosine (m22G) and archaeosine in tRNA. While these assignments are plausible, they nevertheless remain hypothetical until experimental evidence of the enzyme activity is obtained.
Pyrococcus furiosus is a sulfur metabolizing hyperthermophile of the order Thermococcales which grows optimally at 100°C under strictly anaerobic conditions, as in the deep sea, near hydrothermal vents and in geothermally heated marine sediments (22). Several genes of this microorganism were sequenced and the corresponding protein products studied in vitro (23,24), but none of the enzymes involved in tRNA modification have yet been studied. Here we report the identification, cloning and sequencing of the gene corresponding to the tRNA(guanine-26,N2-N2) methyltransferase from P.furiosus. The recombinant pfTrm1 protein overexpressed in Escherichia coli catalyses the in vivo formation of m22G at position 26 in E.coli tRNAs or in vitro using yeast tRNAs as substrates and S-adenosyl-methionine as the methyl group donor.
MATERIALS AND METHODS
Reagents, enzymes and plasmids
32P-Radiolabelled nucleotide triphosphates (400 Ci/mmol) were from Amersham (UK). Reagents and restriction enzymes were from Sigma (St Louis, MO), Boehringer-Mannheim (Mannheim, Germany), MBI Fermentas (Vilnius, Lithuania) or Merck(Germany). Thin layer cellulose plates were from Schleicher & Schuell (Germany). Plasmids p67YF0, p67YF33 and p67YF59 bearing wild-type yeast tRNAPhe and its variants (G26-A44->A26-G44) and (G10-C24->C10-G24) have been described (25); they were a generous gift of Dr O.Uhlenbeck (Boulder, CO).
Strains
Pyrococcus furiosus strain DSM 3638 cells (22) grown under strictly anaerobic conditions were a gift from Prof. P. Forterre (Université d'Orsay, France). Escherichia coli strains JM103, supE thi-1 endA1 hsdR4 sbcB15 strA [Delta](lac-proAB) F[prime][traD36 proAB+lacI lacZ[Delta]M15], and BL21(DE3), hsdS gal ([lambda]cIts857 ind1 Sam7 nin5 lacUV5-T7 gene1), were used for library construction and expression of the cloned recombinant protein.
Preparation of P.furiosus genomic DNA
Cells of P.furiosus (1 g) were resuspended in 15 ml TNE (100 mM Tris-HCl, pH 7.5, 100 mM NaCl, 50 mM EDTA), 2 ml 10% N-lauryl sarcosyl, 2 ml 10% SDS and 1 ml proteinase K (1 mg/ml) were added, and the mixture was incubated for 3 h at 50°C. Proteins were extracted three times with phenol and once with chloroform/isoamyl alcohol. Nucleic acids were precipitated with ethanol and dissolved in 1 ml TE (100 mM Tris-HCl, pH 7.5, 10 mM EDTA). The remaining RNA was digested with 0.1 mg/ml RNase A for 1 h at 37°C.
Preparation of a radioactive DNA probe for library screening
The DNA fragment for hybridization was prepared by PCR amplification using the deoxyoligonucleotides primer A, 5[prime]-AAAACTCATAGGGAGACTTGTTC, and primer B, 5[prime]-CTGCAGGAGTCTCTAAGGCAAAC, corresponding respectively to the sequence of Pyrococcus woesei putative ribosomal protein L35A (GenBank accession no. M83987) and to the 5[prime]-sequence of ORFz (the first 73 codons of a putative tRNA methyltransferase gene; 26,27). Amplification was performed for 30 cycles: 1 min at 95°C, 2 min at 65°C, 2 min at 72°C. Amplified DNA was extracted with chloroform, ethanol precipitated and purified from 1% low melting point agarose after electrophoresis. The purified fragment was 32P-radiolabelled by random priming using the Megaprime kit (Amersham, Aylesbury, UK) according to the manufacturer's instructions.
Mini-library construction and screening
Pyrococcus furiosus genomic DNA (3 µg) was digested overnight with HindIII, HincII, XbaI, PstI, NciI or MvaI. The digests were loaded onto a 1% agarose gel and run at 80 V for several hours. After electrophoresis the gel was soaked in 0.2 M HCl, followed by a treatment in denaturing solution (1.5 M NaCl, 0.5 M NaOH) and neutralization with 1 M Tris-HCl, pH 7.5, 1.5 M NaCl. Separated DNA fragments were transferred to a GeneScreen hybridization transfer membrane (NEN, Boston, MA) and fixed by a 5 min irradiation under a 254 nm UV light. The membrane was incubated in the presence of 0.2 µg/ml salmon sperm DNA and then hybridized at 65°C with the 32P-radiolabelled probe according to Southern (28).
Cloning of the P.furiosus TRM1-like gene was performed as follows. Genomic DNA (100 µg) was digested with HindIII, loaded on 1% agarose and electrophoresed as described above. After separation, the region of the gel containing fragments of ~3 kb in size was excised and DNA fragments were recovered by electroelution. The extracted fragments were ligated in the cloning vector pUC8 cut with HindIII. Ligated DNA was introduced into E.coli JM103 by electroporation using a BioRad Gene Pulser II (0.2 cm cuvettes, 2500 V, 25 µF). Transformed cells were plated on isopropyl-[beta]-d-thiogalactoside (IPTG)/X-Gal-containing medium and grown overnight at 37°C. The transformants containing recombinant plasmid (white colonies) were selected and screened by Southern hybridization with the 32P-radiolabelled PCR fragment. The resulting plasmid containing the TRM1-like gene in a 3 kb insert (pNB5) was used for subcloning.
The HindIII-HincII fragment (1.9 kb) of pNB5 containing the P.furiosus TRM1-like ORF was purified in a 1% low melting point agarose gel and ligated into the HindIII and HincII sites of pUC8. This construct (pNB6) was used for sequencing the insert.
Nucleotide sequence determination
DNA sequencing was performed using the Taq DNA polymerase-based fmol[ DNA cycle sequencing system (Promega, Madison, WI) according to the manufacturer's recommendations. First, the 5[prime]-part of the insert (~800 nt altogether) was sequenced using three specific `walking' primers complementary to regions 182-201, 413-434 and 703-722 of the sequence. In parallel, the 1.9 kb insert in pNB6 was amplified by PCR using pUC universal forward and reverse primers. The amplified DNA fragments were then cut with BsuRI (HaeIII) or RsaI and the resulting short fragments were inserted into the SmaI site of pBluescript II SK(+) vector. As the digestion pattern for both BsuRI and RsaI gave rather long DNA fragments (600-1000 nt), only a part of the entire sequence was determined in this way. To cover the rest of the insert, the sequence was amplified by PCR using a specific `walking' primer for the 5[prime]-end and a pUC reverse primer for the 3[prime]-end and nested bi-directional deletions using ExoIII exonuclease/mung bean nuclease were obtained. Fragments (400-1000 nt) were inserted into the SmaI site of pBluescript II SK(+) and sequenced. The whole DNA insert (1.9 kb) was thereby sequenced at least three times using independent clones. The sequence described in this paper was deposited in GenBank under accession no. AF051912.
Cloning and expression of P.furiosus tRNA(guanine-26,N2-N2) methyltransferase in E.coli
Cloning of the P.furiosus TRM1-like gene in a T7 RNA polymerase promoter-based E.coli expression vector (pET28b) was done as follows. The coding sequence together with the 3[prime]-non-coding region (~800 nt) was amplified by PCR using pNB6 as template. The forward primer (5[prime]-AAAACCATGGGATCTATGGAATTATTTGAAG) contained a NcoI (underlined) restriction site at its 5[prime]-end. The NcoI site was introduced to provide the translation start (ATG), however, it adds an extra codon (GGA), coding for glycine, following the initiation codon. The reverse primer (5[prime]-TGTTGTGTGGAATTGTGAG) was complementary to the pUC8 sequence in pNB6. The amplified DNA was digested with NcoI and BamHI and the 1.7 kb fragment was inserted into the NcoI and BamHI sites of plasmid pET28b. The resulting recombinant plasmid (pET28b-pfTRM1), recovered from the E.coli JM103 transformant, was transferred to E.coli BL21(DE3). The bacteria were grown to mid-logarithmic phase (D600 ~ 0.7) at 37°C in MM9 minimal medium supplemented with 1 mM MgSO4, 0.3 mM CaCl2, 0.4% casamino acids and 50 µg/ml kanamycin. Expression of the recombinant protein was induced by addition of IPTG to a final concentration of 0.1 mM. After 4 h induction with IPTG at 37°C, the cells were harvested by centrifugation.
Preparation of E.coli cell-free extract
The cell-free extract of E.coli cells (BL21/DE3) was prepared by sonication in 5 vol 100 mM Tris-HCl buffer, pH 7.5, containing 10 mM MgCl2, 1 mM EDTA, 300 mM KCl, 1 mM phenylmethylsulphonyl fluoride, 1 mM DFP and 2 mM dithioerythreitol. The homogenate was centrifuged at 10 000 g for 15 min at 4°C. Glycerol was added to a final concentration of 20% and S10 extract was stored at -20°C.
Preparation of total E.coli tRNA, post-labelling with 32P and analysis of modified nucleotides
Total tRNA from E.coli BL21/DE3 cells was prepared by phenol extraction and high molecular mass RNA was removed by precipitation with 2 M NaCl. Further purification was performed by electrophoresis in a 6% denaturing polyacrylamide gel. The location of the large tRNA band in the gel was revealed by UV shadowing. The purified tRNA fraction was then analysed for its modified nucleotide content as described earlier (29). Separation of the 32P-labelled nucleotides was performed by two-dimensional chromatography on thin layer cellulose plates (30). The chromatographic systems and reference maps for identification of the nucleotide spots have been described previously (31).
Preparation of tRNA transcripts
[[alpha]-32P]GTP-labelled tRNA transcripts were obtained by in vitro transcription with T7 RNA polymerase of linearized plasmids, as previously described (32).
In vitro enzymatic assay
The standard reaction mixture for detection of m22G formation in tRNA transcripts was 100 mM Tris-HCl (pH 7.5 at 20°C), 5 mM MgSO4, 2 mM DTT, 0.1 mM EDTA, 100 mM ammonium acetate, 75 µM S-adenosyl L-methionine (AdoMet) and 50-100 fmol [32P]GTP-labelled tRNA substrate. The reaction was initiated by addition of S10 E.coli extract to a final protein concentration of 1 mg/ml in a final volume of 50 µl, overlayed by 10 µl paraffin oil. After incubation at 50°C, proteins were phenol extracted and the nucleic acids ethanol precipitated, collected by centrifugation and hydrolyzed overnight at 37°C with 2 µg nuclease P1 in 50 mM sodium acetate, pH 5.3. The resulting hydrolysate was then analysed by two-dimensional TLC as described above. Radioactivity in the spots of modified nucleoside 5[prime]-phosphates was quantified by scanning in a PhosphorImager (Molecular Dynamics, Sunnyvale, CA) using ImageQuant software.
RESULTS
Cloning and sequencing of the pfTRM1 gene encoding tRNA(guanine-26,N2-N2) methyltransferase
Sequencing of a 3.1 kb PstI fragment of the hyperthermophilic archaeon P.woesei genomic DNA, containing the glyceraldehyde 3-phosphate dehydrogenase gene (ORFgapdh), allowed mapping of the positions of three additional complete neighbouring ORFs (ORFa, ORFb and ORFy) and two truncated ORFs (ORFx and ORFz; see upper part of Fig.
Figure 1. Strategy for identification and cloning of the P.furiosus TRM1-like gene. The schematic representation of the 3 kb PstI fragment of P.woesei genomic DNA bearing the gene coding for glyceraldehyde 3-phosphate dehydrogenase (gapdh-gene; 26) is given at the top. The orientations of ORFs are indicated by arrows. All details are as described in Materials and Methods. To this end, a DNA sequence coding for the putative N-terminal part of P.woesei Trm1-like protein (73 amino acids) and the C-terminal part of ribosomal protein L35a protein (51 amino acids) was used to design oligonucleotides for PCR amplification from genomic DNA from P.furiosus. A DNA fragment of the expected size (~380 bp) was obtained, radiolabelled and used as a probe for Southern hybridization with the digested P.furiosus genomic DNA (Fig. Figure 2. The sequence of the TRM1-like ORF from P.furiosus and deduced protein sequence using the conventional one letter code for each amino acid (GenBank accession no. AF051912). Non-coding nucleotides are represented by lower case letters. Only 79 of the 170 nt sequenced from the HindIII restriction site are shown. The putative ribosomal binding site (region +9 to +14) and pyrimidine-rich sequence downstream of the stop codon (putative transcription stop signal) are boxed. Underlined letters correspond to the putative AdoMet bindings motifs (see text). A schematic map of the HindIII-HincII fragment of P.furiosus genomic DNA is given at the bottom. In order to remove the non-coding upstream DNA sequence adjacent to the pfTRM1 gene and to place the sequence under the control of an efficient promoter, a new plasmid, pET-pfTRM1, was constructed by inserting the coding sequence into the expression vector pET28b between the NcoI and BamHI sites, in-frame with the initiation codon within the NcoI site (Materials and Methods). The sequence of this insert was identical to that of the pfTRM1 ORF in pNB6 (Figs Figure 3. Expression of the Trm1-like protein of P.furiosus in E.coli BL21(DE3). IPTG-induced cells were harvested by centrifugation, lysed in the presence of 6% SDS and the extract electrophoresed on a 10% denaturing polyacrylamide gel. Control cells bearing empty pET28b vector and pET28b-pfTRM1-containing cells were induced for 4 h. Mid-range molecular mass markers (Promega, Madison, WI) were loaded in lane M. Bacterial tRNAs, including E.coli tRNAs, do not contain the modified nucleotide m22G26 (15) due to the absence of the corresponding tRNA(guanine-26,N2-N2) methyltransferase activity. Only the recently sequenced genome of the hyperthermophilic eubacteria Aquifex aeolicus reveals the presence of a TRM1-like gene (36). However, when bulk or individual E.coli tRNAs are incubated with protein extracts from eukaryotic cells, such as wheat embryo, yeast, Tetrahymena pyriformis or rat liver, m22G26 is easily formed in an AdoMet-dependent manner (see for example 37-40). Likewise, bulk tRNA isolated from E.coli cells transformed with pUC19 bearing the S.cerevisiae TRM1 gene contains m22G (41). Thus, despite the absence of a naturally modified G26 in bacterial tRNAs, their tRNAs can serve as perfect substrates for the eukaryotic N2,N2-dimethylguanine transferase. In this study that peculiarity was used to detect the enzyme activity of recombinant pfTrm1p in vivo. Qualitative analysis of the content of modified nucleotides in bulk tRNA isolated from E.coli cells transformed either with pET28b containing the insert pfTRM1 or pET-28b lacking the insert (control experiment) was performed by the post-labelling procedure on tRNA hydrolysate, followed by TLC of the resulting 5[prime]-32P-labelled nucleotide monophosphates. Figure Figure 4. Analysis of modified nucleotides present in the control E.coli tRNA sample (A) and in tRNA extracted from pET28b-pfTRM1-transformed E.coli BL21(DE3) (B). Analysis was performed by the post-labelling technique as described in Materials and Methods. Two-dimensional separation of radioactive 5[prime]-mononucleotides was achieved with the chromatographic system described in Materials and Methods. Assignment of modified nucleotides on the autoradiogram was by comparison with reference maps (31). The major spots correspond to pA, pC, pU and pG, as indicated. The spot corresponding to pm22G in the tRNA from the induced strain is circled (see B). The other spots on the autoradiograms represent the modified nucleotides normally present in mature E.coli tRNA: 1, pAm; 2, pm6A; 3, pGm; 4 and 5, pm2G and pt6A (visible only on the original autoradiogram); 6, pm1G; 7, probably pQ; 8, pT; 9, pm7G; 10, pCm; 11, pD; 12, p[Psi]; 13, most probably ribosyl phosphate; 14, pI (inosine). To formally identify the site of modification, additional experiments were performed in vitro with a protein extract of the induced pET28-pfTRM1-containing E.coli cells, using synthetic tRNA transcripts as substrates. We used the transcript of wild-type yeast tRNAPhe, which was shown to be an excellent substrate for yeast tRNA(guanine,N2-N2) methyltransferase when tested in an S100 yeast extract supplemented with AdoMet (42) as well as after microinjection into the cytoplasm of Xenopus laevis oocytes (32,43). Our results obtained with cell-free extracts of P.furiosus also demonstrate that the transcript of yeast tRNAPhe becomes modified to m22G26 and m2G10 upon incubation. The two other substrates correspond to variants of yeast tRNAPhe bearing mutations at positions 26-44 and 10-25 (Fig. Figure 5. Cloverleaf representation of the nucleotide sequence of wild-type yeast tRNAPhe and two yeast tRNAPhe mutants. mut1 corresponds to reversion of the G10C24 base pair to C10G24 and mut2 reversion of the G26A44 base pair to A26G44. T7 transcripts corresponding to these three tRNAs were radiolabelled with [32P]GTP and incubated for 30 min at 50°C with S10 extract from IPTG-induced E.coli cells BL21 carrying the plasmid pET28b-pfTRM1. The presence of methylated guanines was evaluated by TLC analysis of the P1 nuclease digest of radiolabelled tRNA as described in Materials and Methods. (A) Autoradiogram of the TLC plate when yeast tRNAPhe-mut1 was used as substrate; (B) wild-type yeast tRNAPhe; (C) tRNAPhe-mut2. Open circles indicate migration of markers as visualized by UV shading: AMP (pA); CMP (pC); UMP (pU). (D) Time courses for the biosynthesis of m2G (open symbols) and m22G (solid symbols) in in vitro transcripts of yeast wild-type tRNAPhe (triangles) and the two yeast tRNAPhe mutants, mut1 (circles) and mut2 (squares). Quantification of the radioactivity in each spot of the TLC plates made at different times of incubation allows determination of the molar ratio of m2G and m22G present in the modified tRNA transcript. Each transcript was [32P]GTP-radiolabelled and individually tested with S10 extract of induced transformed E.coli in the presence of added AdoMet. The incubation temperature 50°C was chosen, taking into account the results of preliminary experiments demonstrating that maximal enzymatic activity is achieved under these conditions. A further increase in incubation temperature (to 60°C) results in a decreased level of m22G modification in tRNAPhe transcript (data not shown). This temperature optimum (50°C) is related to a limited resistance of the macromolecular RNA substrate (unmodified transcript of yeast tRNAPhe) to thermal unfolding. The melting profiles obtained with yeast tRNAPhe and other tRNA transcripts demonstrate that unfolding of the proper three-dimensional structure of tRNA begins between 55 and 60°C, depending on the Mg2+ concentration and the nature of the transcript (44). The results of TLC analysis (Fig. In the present paper, we demonstrate that heterologous expression of the TRM1 gene of P.furiosus in E.coli, normally devoid of an m22G-forming enzyme, confers on the host the ability to express a polypeptide having G26 methyltransferase activity. Moreover, the protein extract from the induced transformed E.coli reveals an enzymatic activity that is able to catalyse the formation of the N2,N2-dimethyl derivative of G26 but not of G10. Therefore, the pfTRM1 ORF that we have cloned is the structural gene for the AdoMet-dependent tRNA(guanine-26,N2-N2) methyltransferase (pfTrm1p). However, considering that m22G is also found at position 10 of several naturally occurring archaeal tRNAs (see below), an unidentified distinct m22G-forming enzyme in P.furiosus has still to be discovered. These observations fit well with what is already known concerning the occurrence and biosynthesis of m22G in tRNAs. Indeed, analysis of the nucleotide sequences of 59 mature tRNAs from several Archaea (mostly from Haloferax volcanii, but also from Halobacterium cutirubrum and Halococcus morrhuae, and five tRNA sequences from the thermophilic Archaea M.thermoautotrophicum, Thermoplasma acidophilum and Sulfolobus acidocaldarius) has revealed that methylation of the N2-amino group of guanine occurs at two distinct sites: at position 10 in the proximal dihydrouridine stem and at position 26 in the bend between the dihydrouridine (D) stem and the anticodon (AC) stem of tRNAs. Depending on the tRNA considered, these guanine modifications involve a single or a double methylation to form m2G or m22G, dimethylation being most frequently found at those sites (15). In tRNAs from certain hyperthermophilic Archaea, a 2[prime]-O-methyl derivative of m22G has also been detected (m22Gm; 13), however, its exact location, probably at position(s) 10 and/or 26 has not yet been elucidated. Dimethylation of guanine residues is frequently found in eukaryotic tRNAs (cytoplasmic, mitochondrial and chloroplast), mainly at position 26 and rarely at position 27, while at position 10 m2G is exclusively present. In higher eukaryotic tRNAs, but not in single cell eukaryotes, m2G is also frequently found at position 6 (and exceptionally at position 7) of the amino acid stem (15). Previous studies with eukaryotic cell extracts (from rat liver, chicken embryo, T.pyriformis and yeast) demonstrated that these two sites are methylated by chromatographically distinct tRNA(guanine) methyltransferases: one specific for monomethylation of N2-G10 with a molecular mass of ~75-85 kDa (45-48) and the other for dimethylation of N2,N2-G26 with a molecular mass ranging between 180 and 260 kDa (37,38,40,49). In both cases, the molecular masses refer to the native proteins, depending on the origin of the enzymes and on the method used for their evaluation. Eukaryotic tRNA(guanine-26) methyltransferases were shown to be rather unstable under the purification procedure. Also, depending on the nucleotides surrounding G26, they may catalyse only formation of m2G26 (reviewed in 50). Thus, despite the close spatial proximity of G10 and G26 within the three-dimensional architecture of tRNA (see 51), distinct position-specific tRNA(guanine) methyltransferases exist in archaeal and eukaryotic cells. In this work, we have shown that bulk tRNA extracted from an E.coli strain transformed with plasmid pET28b-pfTRM1 contains m22G. Likewise, transfection of E.coli with the TRM1 gene of S.cerevisiae leads to the appearence of m22G in the tRNAs of the transformed E.coli cells (41). Altogether, these results demonstrate that a single gene product is able to catalyse the transfer of both methyl groups to G26 in tRNAs. This conclusion is in accord with our observation that a small amount of m2G26 was also formed as an intermediate when a T7 transcript of yeast tRNAPhe was incubated with the cell extract of transformed E.coli BL21(pET-pfTRM1). It also agrees with the results obtained for the native tRNA(guanine-26,N2-N2) methyltransferase from the protozoan T.pyriformis (40). This enzyme was capable of forming in vitro both m2G26 and m22G26 on a single tRNA isoacceptor species, the relative amount of m2G over m22G being dependent on the ratio of enzyme and tRNA substrate. Thus, G26 dimethylation is a two step process which might require dissociation of the enzyme-monomethylated tRNA intermediate (40). However, taking into account the subunit structure of the enzyme, other alternative scenarios can be envisaged (discussed in 50). As indicated in Figure Figure 6. Sequence alignment of TRM1-like proteins from C.elegans (C_ELEG), S.cerevisiae (S_CEREV), A.fulgidus (A_FULG), P.furiosus (P_FUR), M.thermoautotrophicum (M_THER), M.jannaschii (M_JAN) and A.aeolicus (A_AEOL). The scale in amino acid residues is presented at the top. Identical and conserved residues are shaded. Two putative AdoMet binding motifs and their consensus amino acid sequences are shown at the bottom. The one letter code for each amino acid is used, except for the hydrophobic amino acids I, L, V, M, F and A, which are designated by U. The upper line of the motifs corresponds to the ones defined in this paper, while the second line corresponds to the motifs defined by Kagan and Clarke (57). Asterisks indicate the sequence of proteins for which the activity of the enzymatic reaction was established experimentally. In both cases they correspond to tRNA(guanine-26,N2-N2) methyltransferases (EC 2.1.1.32). B, E and A designate Bacteria, Eukarya and Archaea respectively. Analysis of the 3[prime]-UTR of the sequenced HindIII-HincII fragment reveals the presence of another ORF in the opposite orientation. A BLAST search using the deduced protein sequence indicates that it is similar to the C-terminal end of a protein of unknown function found in all three archaeal genomes sequenced so far (AE000840, AE000989 and MJ0969). The G+C content of the pfTRM1 gene (39.3%) is rather close to the value reported for the total genome of P.furiosus (38%) (22). As expected for a gene with a low G+C content, A and T are the preferred bases in the third position of codons (reviewed in 54). In particular, codons NCG (ending with G) for Ser, Pro, Thr and Ala are not or rarely used, whereas for Arg, AGG and AGA codons are strongly preferred (16 out of a total of 20 Arg codons). The same trends have been noticed for many other hyperthermophilic archaeal coding genes (see for example 55,56). pfTrm1p and scTrm1p are the only two tRNA(guanine-26,N2-N2) methyltransferases for which the activity of the corresponding gene products have been experimentally verified. However, based on the similarity of amino acid sequences with scTrm1p, the genes coding for five tentatively assigned m22G-forming enzymes can be detected (GenBank entries Z77136, U67538, AE000887, AE001048 and AE000710). One of them is from a eukaryon Caenorhabditis elegans, three are from Archaea (M.jannaschii, M.thermoautotrophicum and A.fulgidus) and the last is from the thermophilic eubacterium A.aeolicus. Most probably these genes do indeed encode for tRNA(guanine-26,N2-N2) methyltransferases, yet this plausible hypothesis should be verified experimentally. Alignment of the amino acid sequences of all seven Trm1-like proteins is schematically illustrated in Figure In conclusion, pfTRM1 is the first gene of an archaeon coding for a tRNA modification enzyme that has been cloned and expressed in an active form in E.coli. The amino acid sequence of this newly identified tRNA(guanine-26,N2-N2) methyltransferase bears great similarity to those of the corresponding eukaryotic enzymes. The fact that the Trm1p of a hyperthermophilic organism would be functional under the physiological conditions of E.coli cells was not a priori evident. Indeed, there are cases where recombinant thermophilic enzymes were shown to be inactive at low temperatures or even required a heat treatment after cell lysis in order to recover enzymatic activity in vitro (see for example 56,62). Therefore, the possibility of functional expression of the pfTRM1 gene in mesophilic E.coli provides the opportunity for further biochemical, biophysical and possible crystallization studies on this tRNA modification enzyme. This work was performed within the programme `Bactocean', sponsored by the CNRS-IFREMER. It was supported by grants from the CNRS (ACC-5 and PCV) and Actions de Recherche sur le Cancer (ARC). F.C. is a pre-doctoral fellow supported by a fellowship from the Ministère de la Recherche et de l'Enseignement Superieurs. Y.M. has benefitted from a one year Associated Researcher position at the CNRS (Poste Rouge). We thank Prof. P. Forterre (Université d'Orsay, France), Dr J. DiRuggiero (Center of Marine Biotechnology, University of Maryland, MD), Dr O. Guerassimenko (Moscow University, Moscow, Russia) and Dr J.-P. Waller (CNRS, Gif-sur-Yvette, France) for discussions, encouragement and critical reading of the manuscript.
Functional expression of the recombinant pfTRM1 gene product in E.coli
Identification of the activity of the recombinant pfTrm1 protein
Identification of the guanine residue modified in tRNA substrates
DISCUSSION
m2G and m22G occur at positions 10 and 26 in fully mature tRNAs, formed by distinct site-specific methyltransferases
A single gene product pfTrm1p has both mono- and dimethylating activity at G26
Sequence analysis of the pfTRM1 ORF
Sequence alignment of several Trm1-like proteins and localization of putative SAM binding motifs
ACKNOWLEDGEMENTS
REFERENCES
This article has been cited by other articles:
This page is run by Oxford University Press, Great Clarendon Street, Oxford OX2 6DP, as part of the OUP Journals
Comments and feedback: www-admin{at}oup.co.uk
Last modification: 30 Jul 1998
Copyright©Oxford University Press, 1998.
CiteULike 
Connotea 
Del.icio.us What's this?
T. Awai, S. Kimura, C. Tomikawa, A. Ochi, Ihsanawati, Y. Bessho, S. Yokoyama, S. Ohno, K. Nishikawa, T. Yokogawa, et al.
Aquifex aeolicus tRNA (N2,N2-Guanine)-dimethyltransferase (Trm1) Catalyzes Transfer of Methyl Groups Not Only to Guanine 26 but Also to Guanine 27 in tRNA
J. Biol. Chem.,
July 31, 2009;
284(31):
20467 - 20478.
[Abstract]
[Full Text]
[PDF]
H. Walbott, S. Auxilien, H. Grosjean, and B. Golinelli-Pimpaneau
The Carboxyl-terminal Extension of Yeast tRNA m5C Methyltransferase Enhances the Catalytic Efficiency of the Amino-terminal Domain
J. Biol. Chem.,
August 10, 2007;
282(32):
23663 - 23671.
[Abstract]
[Full Text]
[PDF]
J. Armengaud, J. Urbonavicius, B. Fernandez, G. Chaussinand, J. M. Bujnicki, and H. Grosjean
N2-Methylation of Guanosine at Position 10 in tRNA Is Catalyzed by a THUMP Domain-containing, S-Adenosylmethionine-dependent Methyltransferase, Conserved in Archaea and Eukaryota
J. Biol. Chem.,
August 27, 2004;
279(35):
37142 - 37152.
[Abstract]
[Full Text]
[PDF]
F. Constantinesco, P. Forterre, E. V. Koonin, L. Aravind, and C. Elie
A bipolar DNA helicase gene, herA, clusters with rad50, mre11 and nurA genes in thermophilic archaea
Nucleic Acids Res.,
February 27, 2004;
32(4):
1439 - 1447.
[Abstract]
[Full Text]
[PDF]
J. Liu and K. B. Straby
The human tRNA(m22G26)dimethyltransferase: functional expression and characterization of a cloned hTRM1 gene
Nucleic Acids Res.,
September 15, 2000;
28(18):
3445 - 3451.
[Abstract]
[Full Text]
[PDF]
D. R. Stanford, N. C. Martin, and A. K. Hopper
SURVEY AND SUMMARY: ADEPTs: information necessary for subcellular distribution of eukaryotic sorting isozymes resides in domains missing from eubacterial and archaeal counterparts
Nucleic Acids Res.,
January 15, 2000;
28(2):
383 - 392.
[Abstract]
[Full Text]
[PDF]
This Article 
Abstract
Print PDF (238K)
Alert me when this article is cited
Alert me if a correction is posted
Services
Email this article to a friend
Similar articles in this journal
Similar articles in ISI Web of Science
Similar articles in PubMed
Alert me to new issues of the journal
Add to My Personal Archive
Download to citation manager
Search for citing articles in:
ISI Web of Science (14)
Request Permissions
Commercial Re-use Guidelines
for Open Access NAR Content
Google Scholar
Articles by Constantinesco, F.
Articles by Grosjean, H.
Search for Related Content
PubMed
PubMed Citation
Articles by Constantinesco, F.
Articles by Grosjean, H.
Social Bookmarking
What's this?