The role of modifications in codon discrimination by tRNA(Lys)UUU

The natural modification of specific nucleosides in many tRNAs is essential during decoding of mRNA by the ribosome. For example, tRNA(Lys)(UUU) requires the modification N6-threonylcarbamoyladenosine at position 37 (t(6)A37), adjacent and 3' to the anticodon, to bind AAA in the A site of the ribosomal 30S subunit. Moreover, it can only bind both AAA and AAG lysine codons when doubly modified with t(6)A37 and either 5-methylaminomethyluridine or 2-thiouridine at the wobble position (mnm(5)U34 or s(2)U34). Here we report crystal structures of modified tRNA anticodon stem-loops bound to the 30S ribosomal subunit with lysine codons in the A site. These structures allow the rationalization of how modifications in the anticodon loop enable decoding of both lysine codons AAA and AAG.     

How are tRNAs and mRNA arranged in the ribosome? An attempt to correlate the stereochemistry of the tRNA-mRNA interaction with constraints imposed by the ribosomal topography.

Two tRNA molecules at the ribosomal A- and P-sites, with a relatively small angle between the planes of the L-shaped molecules, can be arranged in two mutually exclusive orientations. In one (the 'R'-configuration), the T-loop of the A-site tRNA faces the D-loop of the P-site tRNA, whereas in the other (the 'S'-configuration) the D-loop of the A-site tRNA faces the T-loop of the P-site tRNA. A number of stereochemical arguments, based on the crystal structure of 'free' tRNA, favour the R-configuration. In the ribosome, the CCA-ends of the tRNA molecules are 'fixed' at the base of the central protuberance (the peptidyl transferase centre) of the 50S subunit, and the anticodon loops lie in the neck region (the decoding site) of the 30S subunit. The translocation step is essentially a rotational movement of the tRNA from the A- to the P-site, and there is convincing evidence that the A-site must be located nearest to the L7/L12 protuberance of the 50S subunit. The mRNA in the two codon-anticodon duplexes lies on the 'inside' of the 'elbows' of the tRNA molecules (in both the S-type and R-type configurations), and runs up between the two molecules from the A- to the P-site in the 3' to 5'-direction. These considerations have the consequence that in the S-configuration the mRNA in the codon-anticodon duplexes is directed towards the 50S subunit, whereas in the R-configuration it is directed towards the 30S subunit. The results of site-directed cross-linking experiments, in particular cross-links to mRNA at positions within or very close to the codons interacting with A- or P-site tRNA, favour the latter situation. This conclusion is in direct contradiction to other current models for the arrangement of mRNA and tRNA on the ribosome.     

The distance between the anticodon loops of two tRNAs bound to the 70 S Escherichia coli ribosome.

Using singlet-singlet energy transfer, we have measured the distance between the anticodons of two transfer RNAs simultaneously bound to a messengerprogramed Escherichia coli 70 S ribosome. The fluorescent Y base adjacent to the anticodon of yeast tRNA_Y^Phe serves as a donor. A proflavine (Pf) chemically substituted for the Y base in tRNA_Pf^Phe serves as an acceptor. By exploiting the sequential binding properties of 70 S ribosomes for two deacylated tRNAs, we can fill the strong site with either tRNA_Y^Phe or tRNA_Pf^Phe and then the weak site with the other tRNA. In both cases donor quenching and sensitized emission of the acceptor are observed. Analysis of these results leads to an estimate for the Y-proflavine distance of 18 ± 2 Å. This distance is very short and suggests strongly that the two tRNAs are simultaneously in contact with adjacent codons of the message. Separate experiments show that binding of a tRNA to the weak site does not perturb the environment of the hypermodified base of a tRNA bound to the strong site. This supports the assignment of the strong site as the peptidyl site. It also indicates that binding of the second tRNA proceeds without a change in the anticodon structure of a pre-existing tRNA at the peptidyl site.     

Rapid kinetic analysis of EF-G-dependent mRNA translocation in the ribosome

Precise and coordinated movement of the tRNA-mRNA complex within the ribosome is a fundamental step during protein biosynthesis. The molecular mechanism for this process is still poorly understood. Here we describe a new sensitive method for monitoring elongation factor G-dependent translocation of the mRNA in the ribosome. In this method, the fluorescent probe pyrene is covalently attached to the 3' end of a short mRNA sequence at position +9. Translocation of the mRNA by one codon results in a significant decrease in the fluorescence emission of pyrene and can be used to directly monitor mRNA movement using rapid kinetic methods. Importantly, this method offers the flexibility of using any tRNA or tRNA analog in order to elucidate the molecular mechanism of translocation. Our results show that the mRNA is translocated at the same rate as the tRNAs, which is consistent with the view that the movement of the tRNAs and the mRNA are coupled in the ribosome. Furthermore, an anticodon stem-loop analog of tRNA is translocated from the ribosomal A site at a rate constant that is 350-fold lower than peptidyl tRNA, indicating that the D stem, T stem and acceptor stem of A site tRNA contribute significantly to the rate of translocation.     

Models for two tRNAs bound to successive codons on mRNA on the ribosome

We have investigated the structural changes necessary to build a model complex of two molecules of phenylalanine transfer RNA (tRNA(Phe) bound to successive codons in a short segment of a model messenger RNA (mRNA), consisting of U6. We keep the mRNA in an ideal helical conformation, deforming the tRNAs as necessary to eliminate steric overlaps while bringing the two 3' termini together. The resulting model has the two tRNAs oriented relative to one another in a manner that is very similar to a model developed by McDonald and Rein (1) in which the tRNAs maintain their ideal crystallographic conformations and all of the deformations are introduced into the mRNA. Consequently, regardless of how one divides the deformations between the tRNAs and the mRNA it is clear that, on the ribosome, the tRNA in the P site has its "front" side (that side with the variable loop) close to the "back" side of the tRNA in the A site (that side with the D loop). The space between the two molecules must be left free on the ribosome, in order to facilitate the transition from the A site to the P site. A detailed pathway is also proposed for changing the anticodon loop structure from that of the A site to that of the P site. The anticodon loop is always kept in a 3'-stacked conformation, since we find that the shift between the 3'-stacked and 5'-stacked structures proposed by Woese (2) is not feasible.     

Accommodating the bacterial decoding release factor as an alien protein among the RNAs at the active site of the ribosome

The decoding release factor （RF） triggers termination of protein synthesis by functionally mimicking a tRNA to span the decoding centre and the peptidyl transferase centre （PTC） of the ribosome. Structurally, it must fit into a site crafted for a tRNA and surrounded by five other RNAs, namely the adjacent peptidyl tRNA carrying the completed polypeptide, the mRNA and the three rRNAs. This is achieved by extending a structural domain from the body of the protein that results in a critical conformational change allowing it to contact the PTC. A structural model of the bacterial termination complex with the accommodated RF shows that it makes close contact with the first, second and third bases of the stop codon in the mRNA with two separate loops of structure＂ the anticodon loop and the loop at the tip of helix orS. The anticodon loop also makes contact with the base following the stop codon that is known to strongly influence termination efficiency. It confirms the close contact of domain 3 of the protein with the key RNA structures of the PTC. The mRNA signal for termination includes sequences upstream as well as downstream of the stop codon, and this may reflect structural restrictions for specific combinations of tRNA and RF to be bound onto the ribosome together. An unbiased SELEX approach has been investigated as a tool to identify potential rRNA-binding contacts of the bacterial RF in its different binding conformations within the active centre of the ribosome.     

Accurate translation of the genetic code depends on tRNA modified nucleosides

Transfer RNA molecules translate the genetic code by recognizing cognate mRNA codons during protein synthesis. The anticodon wobble at position 34 and the nucleotide immediately 3' to the anticodon triplet at position 37 display a large diversity of modified nucleosides in the tRNAs of all organisms. We show that tRNA species translating 2-fold degenerate codons require a modified U(34) to enable recognition of their cognate codons ending in A or G but restrict reading of noncognate or near-cognate codons ending in U and C that specify a different amino acid. In particular, the nucleoside modifications 2-thiouridine at position 34 (s(2)U(34)), 5-methylaminomethyluridine at position 34 (mnm(5)U(34)), and 6-threonylcarbamoyladenosine at position 37 (t(6)A(37)) were essential for Watson-Crick (AAA) and wobble (AAG) cognate codon recognition by tRNA(UUU)(Lys) at the ribosomal aminoacyl and peptidyl sites but did not enable the recognition of the asparagine codons (AAU and AAC). We conclude that modified nucleosides evolved to modulate an anticodon domain structure necessary for many tRNA species to accurately translate the genetic code.     

Recognition and positioning of mRNA in the ribosome by tRNAs with expanded anticodons

Mutant tRNAs containing an extra nucleotide in the anticodon loop are known to suppress +1 frameshift mutations, but in no case has the molecular mechanism been clarified. It has been proposed that the expanded anticodon pairs with a complementary mRNA sequence (the frameshift sequence) in the A site, and this quadruplet "codon-anticodon" helix is translocated to the P site to restore the correct reading frame. Here, we analyze the ability of tRNA analogs containing expanded anticodons to recognize and position mRNA in ribosomal complexes in vitro. In all cases tested, 8 nt anticodon loops position the 3' three-quarters of the frameshift sequence in the P site, indicating that the 5' bases of the expanded anticodon (nucleotides 33.5, 34, and 35) pair with mRNA in the P site. We also provide evidence that four base-pairs can form between the P-site tRNA and mRNA, and the fourth base-pair involves nucleotide 36 of the tRNA and lies toward (or in) the 30 S E site. In the A site, tRNA analogs with the expanded anticodon ACCG are able to recognize either CGG or GGU. These data imply a flexibility of the expanded anticodon in the A site. Recognition of the 5' three-quarters of the frameshift sequence in the A site and subsequent translocation of the expanded anticodon to the P site results in movement of mRNA by four nucleotides, explaining how these tRNAs can change the mRNA register in the ribosome to restore the correct reading frame.     

Translocation of a tRNA with an extended anticodon through the ribosome

Coordinated translocation of the tRNA-mRNA complex by the ribosome occurs in a precise, stepwise movement corresponding to a distance of three nucleotides along the mRNA. Frameshift suppressor tRNAs generally contain an extra nucleotide in the anticodon loop and they subvert the normal mechanisms used by the ribosome for frame maintenance. The mechanism by which suppressor tRNAs traverse the ribosome during translocation is poorly understood. Here, we demonstrate translocation of a tRNA by four nucleotides from the A site to the P site, and from the P site to the E site. We show that translocation of a punctuated mRNA is possible with an extra, unpaired nucleotide between codons. Interestingly, the NMR structure of the four nucleotide anticodon stem-loop reveals a conformation different from the canonical tRNA structure. Flexibility within the loop may allow conformational adjustment upon A site binding and for interacting with the four nucleotide codon in order to shift the mRNA reading frame.     

Interaction of tRNA with 23S rRNA in the ribosomal A, P, and E sites

Three sets of conserved nucleotides in 23 rRNA are protected from chemical probes by binding of tRNA to the ribosomal A, P, and E sites, respectively. They are located almost exclusively in domain V, primarily in or adjacent to the loop identified with the peptidyl transferase function. Some of these sites are also protected by antibiotics such as chloramphenicol, which could explain how these drugs interfere with protein synthesis. Certain tRNA-dependent protections are abolished when the 3'-terminal A or CA or 2',3'-linked acyl group is removed, providing direct evidence for the interaction of the conserved CCA terminus of tRNA with 23S rRNA. When the EF-Tu.GTP.aminoacyl-tRNA ternary complex is bound to the ribosome, no tRNA-dependent A site protections are detected in 23S rRNA until EF-Tu is released. Thus, EF-Tu prevents interaction of the 3' terminus of the incoming aminoacyl-tRNA with the peptidyl transferase region of the ribosome during anticodon selection, thereby permitting translational proofreading.     

Modified nucleotides in tRNA(Lys) and tRNA(Val) are important for translocation

Ribosomes translate genetic information encoded by messenger RNAs (mRNAs) into proteins. Accurate decoding by the ribosome depends on the proper interaction between the mRNA codon and the anticodon of transfer RNA (tRNA). tRNAs from all kingdoms of life are enzymatically modified at distinct sites, particularly in and near the anticodon. Yet, the role of these naturally occurring tRNA modifications in translation is not fully understood. Here we show that modified nucleosides at the first, or wobble, position of the anticodon and 3'-adjacent to the anticodon are important for translocation of tRNA from the ribosome's aminoacyl site (A site) to the peptidyl site (P site). Thus, naturally occurring modifications in tRNA contribute functional groups and conformational dynamics that are critical for accurate decoding of mRNA and for translocation to the P site during protein synthesis.     

Ribosome-initiator tRNA complex as an intermediate in translation initiation in Escherichia coli revealed by use of mutant initiator tRNAs and specialized ribosomes.

For functional studies of mutant Escherichia coli initiator tRNAs in vivo, we previously described a strategy based on the use of tRNA genes carrying an anticodon sequence change from CAU to CUA along with a mutant chloramphenicol acetyltransferase (CAT) gene carrying an initiation codon change from AUG to UAG. Surprisingly, under conditions where the mutant initiator tRNA is optimally active, the CAT gene with the UAG initiation codon produced more CAT protein (3- to 9-fold more depending on the conditions) than the wild-type CAT gene. Here we show that two new mutant CAT genes having GUC and AUC initiation codons also produce more of the CAT protein in the presence of the corresponding mutant initiator tRNAs. These results are most easily understood if assembly of the 30S ribosome-initiator tRNA-mRNA initiation complex in vivo proceeds with the 30S ribosome binding first to the initiator tRNA and then to the mRNA. In cells overproducing the mutant initiator tRNAs, most ribosomes would carry the mutant initiator tRNA and these ribosomes would select the mutant CAT mRNA over the other mRNAs.     

EF-G-catalyzed translocation of anticodon stem-loop analogs of transfer RNA in the ribosome.

Translocation, catalyzed by elongation factor EF-G, is the precise movement of the tRNA-mRNA complex within the ribosome following peptide bond formation. Here we examine the structural requirement for A- and P-site tRNAs in EF-G-catalyzed translocation by substituting anticodon stem-loop (ASL) analogs for the respective tRNAs. Translocation of mRNA and tRNA was monitored independently; mRNA movement was assayed by toeprinting, while tRNA and ASL movement was monitored by hydroxyl radical probing by Fe(II) tethered to the ASLs and by chemical footprinting. Translocation depends on occupancy of both A and P sites by tRNA bound in a mRNA-dependent fashion. The requirement for an A-site tRNA can be satisfied by a 15 nucleotide ASL analog comprising only a 4 base pair (bp) stem and a 7 nucleotide anticodon loop. Translocation of the ASL is both EF-G- and GTP-dependent, and is inhibited by the translocational inhibitor thiostrepton. These findings show that the D, T and acceptor stem regions of A-site tRNA are not essential for EF-G-dependent translocation. In contrast, no translocation occurs if the P-site tRNA is substituted with an ASL, indicating that other elements of P-site tRNA structure are required for translocation. We also tested the effect of increasing the A-site ASL stem length from 4 to 33 bp on translocation from A to P site. Translocation efficiency decreases as the ASL stem extends beyond 22 bp, corresponding approximately to the maximum dimension of tRNA along the anticodon-D arm axis. This result suggests that a structural feature of the ribosome between the A and P sites, interferes with movement of tRNA analogs that exceed the normal dimensions of the coaxial tRNA anticodon-D arm.     

ArfA recruits release factor 2 to rescue stalled ribosomes by peptidyl‐tRNA hydrolysis in Escherichia coli

The ribosomes stalled at the end of non‐stop mRNAs must be rescued for productive cycles of cellular protein synthesis. Escherichia coli possesses at least three independent mechanisms that resolve non‐productive translation complexes (NTCs). While tmRNA (SsrA) mediates trans‐translation to terminate translation, ArfA (YhdL) and ArfB (YaeJ) induce hydrolysis of ribosome‐tethered peptidyl‐tRNAs. ArfB is a paralogue of the release factors (RFs) and directly catalyses the peptidyl‐tRNA hydrolysis within NTCs. In contrast, the mechanism of the ArfA action had remained obscure beyond its ability to bind to the ribosome. Here, we characterized the ArfA pathway of NTC resolution in vitro and identified RF2 as a factor that cooperates with ArfA to hydrolyse peptidyl‐tRNAs located in the P‐site of the stalled ribosome. This reaction required the GGQ (Gly–Gly–Gln) hydrolysis motif, but not the SPF (Ser–Pro–Phe) codon–recognition sequence, of RF2 and was stimulated by tRNAs. From these results we suggest that ArfA binds to the vacant A‐site of the stalled ribosome with possible aid from association with a tRNA, and then recruits RF2, which hydrolyses peptidyl‐tRNA in a GGQ motif‐dependent but codon‐independent manner. In support of this model, the ArfA‐RF2 pathway did not act on the SecM‐arrested ribosome, which contains an aminoacyl‐tRNA in the A‐site.     

Orientations of transfer RNA in the ribosomal A and P sites.

In protein synthesis, peptide bond formation requires that the tRNA carrying the amino acid (A site tRNA) contact the tRNA carrying the growing peptide chain (P site tRNA) at their 3' termini. Two models have been proposed for the orientations of two tRNAs as they would be bound to the mRNA in the ribosome. Viewing the tRNA as an upside down L, anticodon loop pointing down, acceptor stem pointing right, and calling this the front view, the R (Rich) model would have the back of the P site tRNA facing the front of the A site tRNA. In the S (Sundaralingam) model the front of the P site tRNA faces the back of the A site tRNA. Models of two tRNAs bound to mRNA as they would be positioned in the ribosomal A and P sites have been created using MC-SYM, a constraint satisfaction search program designed to build nucleic acid structures. The models incorporate information from fluorescence energy transfer experiments and chemical crosslinks. The models that best answer the constraints are of the S variety, with no R conformations produced consistent with the constraints.     

A stereochemical model of the transpeptidation complex

Molecular models are proposed to describe the relative arrangement of aminoacyl and peptidyl tRNAs when bound to their respective A and P sites on the ribosome. The crystallographically determined structures of tRNAasp and tRNAphe have served as the models for these bound structures, while the imposed steric constraints for the model complexes were based on the results of published experimental data. The constructed models satisfy the stereochemical requirements needed for codon-anticodon interaction and for peptide bond formation. In this paper, the results of the complex containing tRNAphe as the A and P site bound transfer RNAs, is compared to a similarly constructed model which uses tRNAasp as the ribosome-bound transfer RNAs. The models have the following three major features: 1) the aminoacyl and peptidyl transfer RNAs assume an angle of approximately 45 degrees relative to each other; 2) in providing the proper stereochemistry for peptide bond condensation, a significant kink must be present in the messenger RNA between the A site and P site codons; and 3) a comparison of the two model complexes indicates that structural variations between the tRNAs or any allosteric transitions of the transfer RNAs associated with codon-anticodon recognition may be accommodated in the model by way of freedom of rotation about the phosphate backbone bonds in the mRNA between consecutive codons.     

Frameshift suppressor mutations outside the anticodon in yeast proline tRNAs containing an intervening sequence.

Extragenic suppressors of +1 frameshift mutations in proline codons map in genes encoding two major proline tRNA isoacceptors. We have shown previously that one isoacceptor encoded by the SUF2 gene (chromosome 3) contains no intervening sequence. SUF2 suppressor mutations result from the base insertion of a G within a 3'-GGA-5' anticodon, allowing the tRNA to read a 4-base code word. In this communication we describe suppressor mutations in genes encoding a second proline tRNA isoacceptor (wild-type anticodon 3'-GGU-5') that result in a novel mechanism for translation of a 4-base genetic code word. The genes that encode this isoacceptor include SUF7 (chromosome 13), SUF8 (chromosome 8), trn1 (chromosome 1), and at least two additional unmapped genes, all of which contain an intervening sequence. We show that suppressor mutations in the SUF7 and SUF8 genes result in G-to-U base substitutions at position 39 that disrupted the normal G . C base pairing in the last base pair of the anticodon stem adjacent to the anticodon loop. These anticodon stem mutations might alter the size of the anticodon loop and permit the use of a 3'-GGGU-5' sequence within the loop to read 4-base proline codons. Uncertainty regarding the exact structure of the mature suppressor tRNAs results from the possibility that anticodon stem mutations might affect sites of intervening sequence removal. The possible role of the intervening sequence in the generation of mature suppressor tRNA is discussed. Besides an analysis of suppressor tRNA genes, we have extended previous observations of the apparent relationship between tRNA genes and repetitive delta sequences found as solo elements or in association with the transposable element TY1. Hybridization studies and a computer analysis of the DNA sequence surrounding the SUF7 gene revealed two incomplete, inverted delta sequences that form a stem and loop structure located 165 base pairs from the 5' end of the tRNA gene. In addition, sequences beginning 164 base pairs from the 5' end of the trn1 gene also exhibit partial homology to delta. These observations provide further evidence for a nonrandom association between tRNA genes and delta sequences.     

Motion of transfer RNA from the A/T state into the A‐site using docking and simulations

The ribosome catalyzes peptidyl transfer reactions at the growing nascent polypeptide chain. Here, we present a structural mechanism for selecting cognate over near‐cognate A/T transfer RNA (tRNA). In part, the structural basis for the fidelity of translation relies on accommodation to filter cognate from near‐cognate tRNAs. To examine the assembly of tRNAs within the ribonucleic–riboprotein complex, we conducted a series of all‐atom molecular dynamics (MD) simulations of the entire solvated 70S Escherichia coli ribosome, along with its associated cofactors, proteins, and messenger RNA (mRNA). We measured the motion of the A/T state of tRNA between initial binding and full accommodation. The mechanism of rejection was investigated. Using novel in‐house algorithms, we determined trajectory pathways. Despite the large intersubunit cavity, the available space is limited by the presence of the tRNA, which is equally large. This article describes a “structural gate,” formed between helices 71 and 92 on the ribosomal large subunit, which restricts tRNA motion. The gate and the interacting protein, L14, of the 50S ribosome act as steric filters in two consecutive substeps during accommodation, each requiring: (1) sufficient energy contained in the hybrid tRNA kink and (2) sufficient energy in the Watson–Crick base pairing of the codon–anticodon. We show that these barriers act to filter out near‐cognate tRNA and promote proofreading of the codon–anticodon. Since proofreading is essential for understanding the fidelity of translation, our model for the dynamics of this process has substantial biomedical implications. Proteins 2012. © 2012 Wiley Periodicals, Inc.     

Analysis of the roles of tRNA structure, ribosomal protein L9, and the bacteriophage T4 gene 60 bypassing signals during ribosome slippage on mRNA

A 50-nucleotide coding gap divides bacteriophage T4 gene 60 into two open reading frames. In response to cis-acting stimulatory signals encrypted in the mRNA, the anticodon of the ribosome-bound peptidyl tRNA dissociates from a GGA codon at the end of the first open reading frame and pairs with a GGA codon 47 nucleotides downstream just before the second open reading frame. Mutations affecting ribosomal protein L9 or tRNA(Gly)(2), the tRNA that decodes GGA, alter the efficiency of bypassing. To understand the mechanism of ribosome slippage, this work analyzes the influence of these bypassing signals and mutant translational components on -1 frameshifting at G GGA and hopping over a stop codon immediately flanked by two GGA glycine codons (stop-hopping). Mutant variants of tRNA(Gly)(2) that impair bypassing mediate stop-hopping with unexpected landing specificities, suggesting that these variants are defective in ribosomal P-site codon-anticodon pairing. In a direct competition between -1 frameshifting and stop-hopping, the absence of L9 promotes stop-hopping at the expense of -1 frameshifting without substantially impairing the ability of mutant tRNA(Gly)(2) variants to re-pair with the mRNA by sub-optimal pairing. These observations suggest that L9 defects may stimulate ribosome slippage by enhancing mRNA movement through the ribosome rather than by inducing an extended pause in translation or by destabilizing P-site pairing.Two of the bypassing signals, a cis-acting nascent peptide encoded by the first open reading frame and a stemloop signal located in the 5' portion of the coding gap, stimulate peptidyl-tRNA slippage independently of the rest of the gene 60 context. Evidence is presented suggesting that the nascent peptide signal may stimulate bypassing by destabilizing P-site pairing.     

Novel anticodon composition of transfer RNAs in Micrococcus luteus, a bacterium with a high genomic G + C content. Correlation with codon usage.

The number and relative amount of isoacceptor tRNAs for each amino acid in Micrococcus luteus, a Gram-positive bacterium with high genomic G + C content, have been determined by sequencing their anticodon loop and its adjacent regions and by selective labelling of tRNAs. Thirty-one tRNA species with 29 different anticodon sequences have been detected. All the tRNAs have G or C at the anticodon first position except for tRNA(ICGArg) and tRNA(NGASer), in response to the abundant usage of NNC and NNG codons. No tRNA with the anticodon UNN capable of translating codon NNA has been detected, in accordance with a very low or zero usage of NNA codons. The relative amount of isoacceptor tRNAs for an amino acid determined by selective labelling strongly correlates with usage of the corresponding codons. On the basis of these and other observations in this and other eubacterial species, we conclude that the relative amount and anticodon composition of isoacceptor tRNA species are flexible, and their changes are mainly adaptive phenomena that have been primarily affected by codon usage, which in turn is affected by directional mutation pressure.