Nuclear magnetic resonance studies of codon-anticodon interaction in tRNAPhe. I. Effect of binding complementary tetra and pentanucleotides to the anticodon

The effect of codon-anticodon interaction on the structure of two tRNA^Phe species was investigated by means of nuclear magnetic resonance spectroscopy. To this end n.m.r.² spectra of yeast and Escherichia coli tRNA^Phe were recorded in the absence and the presence of the oligonucleotides U-U-C-A, U-U-C-G and U-U-C-A-G, which all contain the sequence UUC complementary to the anticodon sequence GAA. The spectra of the hydrogen-bonded protons, the methyl protons and the internucleotide phosphorous nuclei served to monitor the structure of the anticodon loop and of the tRNA in the tRNA-oligonucleotide complex. From the changes in the methyl proton spectra and in the phosphorous spectra it could be concluded that the oligonucleotides bind to the anticodon. Moreover it turned out that the binding constants obtained from these n.m.r. experiments were, within experimental error, equal to the values obtained with other techniques. Using the resonances of the protons hydrogen-bonded between the oligonucleotide and the anticodon loop the structure of the latter could be studied. In particular, binding of the pentanucleotide U-U-C-A-G, which is complementary to the five bases on the 5′ side of the anticodon loop, resulted in the resolution of four to five extra proton resonances indicating that four to five base-pairs are formed between the pentanucleotide and the anticodon loop. The formation of five base-pairs was confirmed by an independent fluorescence binding study. The resonance positions of the hydrogen-bonded protons indicate, that an RNA double helix is formed by the anticodon loop and U-U-C-A-G with the five base-pairs forming a continuous stack. This structure can be accomodated in the so-called 5′ stacked conformation of the anticodon loop, a structure that has been suggested earlier as an alternative to the familiar 3′ stacked conformation in the crystal structure models of yeast tRNA^Phe. It turned out that structural adjustments of the anticodon loop to the binding of the oligonucleotides are propagated into the anticodon stem. The relevance of these results with respect to the mechanism of protein synthesis is discussed.     

Cytokinin-Active Ribonucleosides in Phaseolus RNA: II. DISTRIBUTION IN tRNA SPECIES FROM ETIOLATED P. VULGARIS L. SEEDLINGS

The distribution of cytokinin-active ribonucleosides in tRNA species from etiolated Phaseolus vulgaris L. seedlings has been examined. Phaseolus tRNA was fractionated by benzoylated diethylaminoethyl-cellulose and RPC-5 chromatography, and the distribution of cytokinin activity was compared with the distribution of tRNA species expected to correspond to codons beginning with U. Phaseolus tRNA^Cys, tRNA^Trp, tRNA^Tyr, a major peak of tRNA^Phe, and a large fraction of tRNA^Leu were devoid of cytokinin activity in the tobacco bioassay. Cytokinin activity was associated with all fractions containing tRNA^Ser species and with minor tRNA^Leu species. In addition, several anomalous peaks of cytokinin activity that could not be directly attributed to U group tRNA species were detected.     

Identification of transfer RNA suppressors in Escherichia coli. II. Duplicate genes for tRNA2Gln.

The number of gene copies for tRNA₂^Gln in λpsu⁺2 was determined by genetic and biochemical studies. The transducing phage stimulates the production of the su⁺2 (amber suppressor) and su°2 glutamine tRNAs and methionine tRNA_m. When the su⁺2 amber suppressor was converted to an ochre suppressor by single-base mutation, the phage stimulated ochre-suppressing tRNA₂^Gln, instead of the amber-suppressing tRNA₂^Gln. From the transducing phage carrying the ochre-suppressing allele, strains carrying both ochre and amber suppressors were readily obtainable. These phages stimulated both ochre-suppressing and amber-suppressing tRNA₂^Gln, but not the non-suppressing form. We conclude that the original transducing phage carries two tRNA₂^Gln genes, one su⁺2 and one su°2. The transducing phage carrying two suppressors, ochre and amber, segregates one-gene derivatives that encode only one or the other type of suppressor tRNA. These derivatives apparently arise by unequal recombination involving the two glutamine tRNA genes in the parental phage. This segregation is not accompanied by the loss of the tRNA_m^Met gene. Based on these results, it is suggested that Escherichia coli normally carries in tandem two identical genes specifying tRNA₂^Gln at 15 minutes on the bacterial chromosome. su⁺2 mutants may arise by single-base mutations in the anticodon region of either of these two, leaving the other intact. By double mutations, tRNA₂^Gln genes could also become ochre suppressors. A tRNA_m^Met gene is located near, but not between, these two tRNA₂^Gln genes.     

Letter: Nuclear magnetic resonance of protein-nucleic acid interactions. II. The E. coli tRNA-Glu complex with glutamyl-tRNA synthetase

We have observed the 300 MHz high-resolution proton nuclear magnetic resonance spectrum of the Escherichia coli tRNA^Glu complex with the glutamyl-tRNA synthetase. The observations are fitted very well by a computer-simulated spectrum of the E. coli tRNA^Glu itself, with the individual resonances broadened from 45 Hz to 150 Hz because of the increased molecular weight. This indicates that no helical arms open upon complex formation, nor is there evidence for any additional Watson-Crick base-pairs in the complex.     

In vivo identification of essential nucleotides in tRNALeu to its functions by using a constructed yeast tRNALeu knockout strain

The fidelity of protein biosynthesis requires the aminoacylation of tRNA with its cognate amino acid catalyzed by aminoacyl-tRNA synthetase with high levels of accuracy and efficiency. Crucial bases in tRNA^Leu to aminoacylation or editing functions of leucyl-tRNA synthetase have been extensively studied mainly by in vitro methods. In the present study, we constructed two Saccharomyces cerevisiae tRNA^Leu knockout strains carrying deletions of the genes for tRNA^Leu(GAG) and tRNA^Leu(UAG). Disrupting the single gene encoding tRNA^Leu(GAG) had no phenotypic consequence when compared to the wild-type strain. While disrupting the three genes for tRNA^Leu(UAG) had a lethal effect on the yeast strain, indicating that tRNA^Leu(UAG) decoding capacity could not be compensated by another tRNA^Leu isoacceptor. Using the triple tRNA knockout strain and a randomly mutated library of tRNA^Leu(UAG), a selection to identify critical tRNA^Leu elements was performed. In this way, mutations inducing in vivo decreases of tRNA levels or aminoacylation or editing ability by leucyl-tRNA synthetase were identified. Overall, the data showed that the triple tRNA knockout strain is a suitable tool for in vivo studies and identification of essential nucleotides of the tRNA.     

Identification of transfer RNA suppressors in Escherichia coli. I. Amber suppressor su+2, an anticodon mutant of tRNA2Gln.

In order to isolate the gene for amber suppressor su⁺2 (SupE) in Escherichia coli, a non-defective su⁺2-transducing phage lambda was isolated in three steps: first, deletion derivatives of F′su⁺2 gal (λ) were selected, linking su⁺2 to the right-hand prophage attachment site, att^λPB′; second, these F′-factors were relysogenized by λ and defective transducing phages, λdsu⁺2, were produced by induction; and third, non-defective λpsu⁺2 transducing phages were produced by recombination of λdsu⁺2 isolates with λ. Upon infection by λpsu⁺2, the production of transferRNAs accepting glutamine and methionine was markedly stimulated. Fingerprint analysis of these tRNAs revealed that they consisted of normal tRNA₂^Gln, mutant tRNA₂^Gln and tRNA_m^Met. The mutant tRNA₂^Gln carried a singlebase alteration from G to A at the 3′-end of the anticodon. The production of tRNA₁^Gln was not stimulated by the infection of λpsu⁺2. We conclude that the wild-type allele of su⁺2 (SupE) is the structural gene for tRNA₂^Gln, and the su⁺2 amber suppressor was derived by a single base mutation, changing the anticodon from CUG to CUA, in one of the multi-copy genes for tRNA₂^Gln. The fact that λpsu⁺2 also induces the production of tRNA_m^Met suggests that this tRNA is encoded in the same chromosomal region of E. coli as is tRNA₂^Gln.     

Assignment of the low field proton nuclear magnetic resonance spectrum of yeast phenylalanine transfer RNA to specific base pairs

High-resolution proton nuclear magnetic resonance spectra at 220 and 300 MHz have been used to investigate the base-pairing structure of fragments of yeast tRNA^Phe, of chemically modified tRNA^Phe and of intact tRNA^Phe. To a very good approximation the positions of the fragment spectra are additive within 0·2 part per million, indicating that factors responsible for certain structural features in the intact molecule are already present in the smaller fragments (half molecules, hairpins and $\text{3}\text{4}$ molecules). A simple first-order ring-current shift theory taken in conjunction with the cloverleaf model for tRNA^Phe (RajBhandary et al., 1967) has been used to predict the low-field (? 15 to ?11 part per million) nuclear magnetic resonance spectra and make assignments of the resolved resonances to ring NH protons of specific base pairs. The general agreement between the predicted and observed spectra to within 0·2 part per million confirms in detail the cloverleaf model for the secondary structure of tRNA^Phe in solution. It is also established that ring-current shifts are the principal factor responsible for the wide range of shifts observed in the low-field spectra. As a result it is evident that the resonances are very sensitive to small changes in the secondary structure and in some cases changes in the interbase distance as small as 0·2 Å could easily be detected. It is also clear from the analysis that certain of the resonances are sensitive to the tertiary structure of the molecule and specific examples are discussed. As with our previous study, we find no evidence for any strong Watson-Crick type base pairs beyond those predicted by the cloverleaf structure.     

Mutation in MTO1 involved in tRNA modification impairs mitochondrial RNA metabolism in the yeast Saccharomyces cerevisiae

The yeast MTO1 gene encodes an evolutionarily conserved protein for the biosynthesis of the 5-carboxymethylaminomethyl group of cmnm⁵s²U in the wobble position of mitochondrial tRNA. However, mto1 null mutant expressed the respiratory deficient phenotype only when coupled with the C1409G mutation of mitochondrial 15S rRNA. To further understand the role of MTO1 in mitochondrial RNA metabolism, the yeast mto1 null mutants carrying either wild-type (P^S) or 15S rRNA C1409G allele (P^R) have been characterized by examining the steady-state levels, aminoacylation capacity of mitochondrial tRNA, mitochondrial gene expression and petite formation. The steady-state levels of tRNA^Lys, tRNA^Glu, tRNA^Gln, tRNA^Leu, tRNA^Gly, tRNA^Arg and tRNA^Phe were decreased significantly while those of tRNA^Met and tRNA^His were not affected in the mto1 strains carrying the P^S allele. Strikingly, the combination of the mto1 and C1409G mutations gave rise to the synthetic phenotype for some of the tRNAs, especially in tRNA^Lys, tRNA^Met and tRNA^Phe. Furthermore, the mto1 strains exhibited a marked reduction in the aminoacylation levels of mitochondrial tRNA^Lys, tRNA^Leu, tRNA^Arg but almost no effect in those of tRNA^His. In addition, the steady-state levels of mitochondrial COX1, COX2, COX3, ATP6 and ATP9 mRNA were markedly decreased in mto1 strains. These data strongly indicate that unmodified tRNA caused by the deletion of MTO1 gene caused the instability of mitochondrial tRNAs and mRNAs and an impairment of aminoacylation of mitochondrial tRNAs. Consequently, the deletion of MTO1 gene acts in synergy with the 15S rRNA C1409G mutation, leading to the loss of COX1 synthesis and subsequent respiratory deficient phenotype.     

The role of the catalytic domain of E. coli GluRS in tRNA discrimination

Discrimination of tRNA^Gln is an integral function of several bacterial glutamyl-tRNA synthetases (GluRS). The origin of the discrimination is thought to arise from unfavorable interactions between tRNA^Gln and the anticodon-binding domain of GluRS. From experiments on an anticodon-binding domain truncated Escherichia coli (E. coli) GluRS (catalytic domain) and a chimeric protein, constructed from the catalytic domain of E. coli GluRS and the anticodon-binding domain of E. coli glutaminyl-tRNA synthetase (GlnRS), we show that both proteins discriminate against E. coli tRNA^Gln. Our results demonstrate that in addition to the anticodon-binding domain, tRNA^Gln discriminatory elements may be present in the catalytic domain in E. coli GluRS as well.     

Cloning of the yeast tyrosine transfer RNA genes in bacteriophage lambda.

Lambda bacteriophage containing yeast tyrosine transfer RNA genes were prepared by molecular recombination. These phage were identified by hybridization of ¹²⁵I-labeled yeast tRNA^Tyr to plaques from lambda-yeast recombinant phage pools. The cloned yeast EcoRI fragments that hybridize to ¹²⁵I-labeled tRNA^Tyr were compared in size with the fragments in total yeast DNA that hybridize to the same probe. These comparisons indicate that seven of the eight different tRNA^Tyr genes have been isolated. Unambiguous evidence that these seven fragments contain tRNA^Tyr coding regions was obtained by showing that they hybridize to aminoacylated [^3H]Tyr-tRNA^Tyr. Only one of the fragments hybridizes to ³²P-labeled total yeast tRNA in the presence of competing unlabeled tRNA^Tyr; the tRNA^Tyr genes, therefore, are not predominantly organized into heteroclusters of tRNA genes.     

Recognition of tRNAGln by Helicobacter pylori GluRS2—a tRNAGln-specific glutamyl-tRNA synthetase

Accurate aminoacylation of tRNAs by the aminoacyl-tRNA synthetases (aaRSs) plays a critical role in protein translation. However, some of the aaRSs are missing in many microorganisms. Helicobacter pylori does not have a glutaminyl-tRNA synthetase (GlnRS) but has two divergent glutamyl-tRNA synthetases: GluRS1 and GluRS2. Like a canonical GluRS, GluRS1 aminoacylates tRNA^Glu1 and tRNA^Glu2. In contrast, GluRS2 only misacylates tRNA^Gln to form Glu-tRNA^Gln. It is not clear how GluRS2 achieves specific recognition of tRNA^Gln while rejecting the two H. pylori tRNA^Glu isoacceptors. Here, we show that GluRS2 recognizes major identity elements clustered in the tRNA^Gln acceptor stem. Mutations in the tRNA anticodon or at the discriminator base had little to no impact on enzyme specificity and activity.     

Variation in 16S-23S rRNA Intergenic Spacer Regions in Photobacterium damselae: a Mosaic-Like Structure

Phenotypically, Photobacterium damselae subsp. piscicida and P. damselae subsp. damselae are easily distinguished. However, their 16S rRNA gene sequences are identical, and attempts to discriminate these two subspecies by molecular tools are hampered by their high level of DNA-DNA similarity. The 16S-23S rRNA internal transcribed spacers (ITS) were sequenced in two strains of Photobacterium damselae subsp. piscicida and two strains of P. damselae subsp. damselae to determine the level of molecular diversity in this DNA region. A total of 17 different ITS variants, ranging from 803 to 296 bp were found, some of which were subspecies or strain specific. The largest ITS contained four tRNA genes (tDNAs) coding for tRNA^Glu(UUC), tRNA^Lys(UUU), tRNA^Val(UAC), and tRNA^Ala(GGC). Five amplicons contained tRNA^Glu(UUC) combined with two additional tRNA genes, including tRNA^Lys(UUU), tRNA^Val(UAC), or tRNA^Ala(UGC). Five amplicons contained tRNA^Ile(GAU) and tRNA^Ala(UGC). Two amplicons contained tRNA^Glu(UUC) and tRNA^Ala(UGC). Two different isoacceptor tRNA^Ala genes (GGC and UGC anticodons) were found. The five smallest amplicons contained no tRNA genes. The tRNA-gene combinations tRNA^Glu(UUC)-tRNA^Val(UAC)-tRNA^Ala(UGC) and tRNA^Glu(UUC)-tRNA^Ala(UGC) have not been previously reported in bacterial ITS regions. The number of copies of the ribosomal operon (rrn) in the P. damselae chromosome ranged from at least 9 to 12. For ITS variants coexisting in two strains of different subspecies or in strains of the same subspecies, nucleotide substitution percentages ranged from 0 to 2%. The main source of variation between ITS variants was due to different combinations of DNA sequence blocks, constituting a mosaic-like structure.     

Two tyrosine residues outside the editing active site in Giardia lamblia leucyl-tRNA synthetase are essential for the post-transfer editing

Leucyl-tRNA synthetase (LeuRS) is responsible for the Leu-tRNA^Leu synthesis. The connective peptide 1 (CP1) domain inserted into the Rossmann nucleotide binding fold possesses editing active site to hydrolyze the mischarged tRNA^Leu with noncognate amino acid, then to ensure high fidelity of protein synthesis. A few co-crystal structures of LeuRS with tRNA^Leu in different conformations revealed that tRNA^Leu 3′ end shuttled between synthetic and editing active sites dynamically with direct and specific interaction with the CP1 domain. Here, we reported that Y515 and Y520 outside the editing active site of CP1 domain of Giardia lamblia LeuRS (GlLeuRS) are crucial for post-transfer editing by influencing the binding affinity with mischarged tRNA^Leu. Mutations on Y515 and Y520 also decreased tRNA^Leu charging activity to various extents but had no effect on leucine activation. Our results gave some biochemical knowledge about interaction of tRNA^Leu 3′ end with the CP1 domain in archaeal/eukaryotic LeuRS.     

The fate of norleucine as a replacement for methionine in protein synthesis.

In vivo synthesised protein with norleucine occupying one half of the normal methionine loci was prepared using a methionine auxotroph of Escherichia coli K12. The extent of charging of the analogue onto both tRNA_met species and subsequent incorporation into soluble protein was monitored with a double-labelling system comprising [G-³H]norleucine and [³⁵S]methionine. Further experiments established that norleucine can be formylated in vivo once charged onto the initiator tRNA^f_met. An N-terminal analysis of the crude soluble protein revealed that formylnorleucyl-tRNA^f_met can initiate protein synthesis and that the formyl group is then removed from the nascent polypeptide. We were also led to conclude that the N-terminal methionine-amino peptidase does not recognise the analogue in this position. Slow growth rates on the methionine analogue have been partly attributed to limiting levels of charged tRNA^m_met, resulting in turn from the inefficiency of norleucine charging by methionyl-tRNA synthetase. Finally no evidence has been found for the production of aberrant protein as a result of norleucine incorporation, implying that limited growth on the analogue is due to its inability to replace methionine as the precursor of S-adenosyl methionine.     

Crucial contribution of the multiple copies of the initiator tRNA genes in the fidelity of tRNAfMet selection on the ribosomal P-site in Escherichia coli

The accuracy of the initiator tRNA (tRNA^fMet) selection in the ribosomal P-site is central to the fidelity of protein synthesis. A highly conserved occurrence of three consecutive G–C base pairs in the anticodon stem of tRNA^fMet contributes to its preferential selection in the P-site. In a genetic screen, using a plasmid borne copy of an inactive tRNA^fMet mutant wherein the three G–C base pairs were changed, we isolated Escherichia coli strains that allow efficient initiation with the tRNA^fMet mutant. Here, extensive characterization of two such strains revealed novel mutations in the metZWV promoter severely compromising tRNA^fMet levels. Low cellular abundance of the chromosomally encoded tRNA^fMet allows efficient initiation with the tRNA^fMet mutant and an elongator tRNA^Gln, revealing that a high abundance of the cellular tRNA^fMet is crucial for the fidelity of initiator tRNA selection on the ribosomal P-site in E. coli. We discuss possible implications of the changes in the cellular tRNA^fMet abundance in proteome remodeling.     

Activation of several tRNAs of Escherichia coli by the phenoxyacetyl derivative of N-hydroxysuccinimide

Escherichia coli 15T^? treated with chloramphenicol produces tRNA^phe which is deficient in minor nucleosides. Undermodified tRNA^phe chromatographs as two new peaks from a benzoylated diethylaminoethyl-cellulose column. Chloramphenicol tRNA^phe was purified by phenoxyacetylation of phenylalanyl-tRNA and subsequent chromatography on benzoylated diethylaminoethyl-cellulose. Purified tRNA^phe had an altered Chromatographie profile as a result of the purification procedure. Phenoxyacetylation of an unpurified tRNA preparation, which was either charged with phenylalanine or kept discharged, resulted in a permanent alteration of tRNA^phe which was similar to the alteration of the purified tRNA^phe. The altered tRNAs eluted with higher salt or ethanol concentrations from benzoylated diethylaminoethyl-cellulose. The alteration was also shown for tRNA^phe of phenoxyacetylated tRNA from late log phase E. coli 15T?. tRNA^glu and tRNA^Leu were not changed, but both tRNA^Arg and tRNA^Ile were altered. tRNA₂^Val and tRNA^Met shifted in the elution profile; tRNA₁^Val and tRNA_f^Met were not affected.Comparison of the primary structures of the alterable and nonalterable tRNA's revealed that all alterable tRNA's have the undefined nucleoside X in the extra loop. Phenoxyacetylation of nucleoside X probably was the cause of the altered profiles.tRNA^phe from E. coli 15T^? treated with chloramphenicol was less reactive towards phenoxyacetylation than normal tRNA, possibly because of a different conformation of the modification-deficient molecule relative to the normal tRNA^phe. tRNA^phe from E. coli 15T^?, starved for cysteine and methionine and treated with chloram-phenicol, is more deficient in minor nucleosides and showed even less reactivity.Acceptor capacities of the altered tRNA species were not changed significantly; only the acceptor capacity for tRNA^Ile decreased approximately 25%. The recognition site for the aminoacyl-tRNA synthetases probably is not affected.