The influence of the reading context upon the suppression of nonsense codons

Dissecting the Contribution of Release Factor Interactions to Amber Stop Codon Reassignment Efficiencies of the Methanocaldococcus jannaschii Orthogonal Pair

Non-canonical amino acids (ncAAs) are finding increasing use in basic biochemical studies and biomedical applications. The efficiency of ncAA incorporation is highly variable, as a result of competing system composition and codon context effects. The relative quantitative contribution of the multiple factors affecting incorporation efficiency are largely unknown. This manuscript describes the use of green fluorescent protein (GFP) reporters to quantify the efficiency of amber codon reassignment using the Methanocaldococcus jannaschii orthogonal pair system, commonly employed for ncAA incorporation, and quantify the contribution of release factor 1 (RF1) to the overall efficiency of amino acid incorporation. The efficiencies of amber codon reassignments were quantified at eight positions in GFP and evaluated in multiple combinations. The quantitative contribution of RF1 competition to reassignment efficiency was evaluated through comparisons of amber codon suppression efficiencies in normal and genomically recoded Escherichia coli strains. Measured amber stop codon reassignment efficiencies for eight single stop codon GFP variants ranged from 51 to 117% in E. coli DH10B and 76 to 104% in the RF1 deleted E. coli C321.ΔA.exp. Evaluation of efficiency changes in specific sequence contexts in the presence and absence of RF1 suggested that RF1 specifically interacts with +4 Cs and that the RF1 interactions contributed approximately half of the observed sequence context-dependent variation in measured reassignment efficiency. Evaluation of multisite suppression efficiencies suggests that increasing demand for translation system components limits multisite incorporation in cells with competing RF1.

'Stop' in protein synthesis is modulated with exquisite subtlety by an extended RNA translation signal

Translational stop codons, UAA, UAG, and UGA, form an integral part of the universal genetic code. They are of significant interest today for their underlying fundamental role in terminating protein synthesis, but also for their potential utilisation for programmed alternative translation events. In diverse organisms, UAA has wide usage, but it is puzzling that the high fidelity UAG is selected against and yet UGA, vulnerable to suppression, is widely used, particularly in those archaeal and bacterial genomes with a high GC content. In canonical protein synthesis, stop codons are interpreted by protein release factors that structurally and functionally mimic decoding tRNAs and occupy the decoding site on the ribosome. The release factors make close contact with the decoding complex through multiple interactions. Correct interactions cause conformational changes resulting in new and enhanced contacts with the ribosome, particularly between specific bases in the mRNA and rRNA. The base following the stop codon (fourth or +4 base) may strongly influence decoding efficiency, facilitating alternative non-canonical events like frameshifting or selenocysteine incorporation. The fourth base is drawn into the decoding site with a compacted stop codon in the eukaryotic termination complex. Surprisingly, mRNA sequences upstream and downstream of this core tetranucleotide signal have a significant influence on the strength of the signal. Since nine bases downstream of the stop codon are within the mRNA channel, their interactions with rRNA, and r-proteins may affect efficiency. With this understanding, it is now possible to design stop signals of desired strength for specific applied purposes.

Transfer RNA gene numbers may not be completely responsible for the codon usage bias in asparagine, isoleucine, phenylalanine, and tyrosine in the high expression genes in bacteria

It is generally believed that the effect of translational selection on codon usage bias is related to the number of transfer RNA genes in bacteria, which is more with respect to the high expression genes than the whole genome. Keeping this in the background, we analyzed codon usage bias with respect to asparagine, isoleucine, phenylalanine, and tyrosine amino acids. Analysis was done in seventeen bacteria with the available gene expression data and information about the tRNA gene number. In most of the bacteria, it was observed that codon usage bias and tRNA gene number were not in agreement, which was unexpected. We extended the study further to 199 bacteria, limiting to the codon usage bias in the two highly expressed genes rpoB and rpoC which encode the RNA polymerase subunits β and β', respectively. In concordance with the result in the high expression genes, codon usage bias in rpoB and rpoC genes was also found to not be in agreement with tRNA gene number in many of these bacteria. Our study indicates that tRNA gene numbers may not be the sole determining factor for translational selection of codon usage bias in bacterial genomes.

Fine-tuning of translation termination efficiency in Saccharomyces cerevisiae involves two factors in close proximity to the exit tunnel of the ribosome

In eukaryotes, release factors 1 and 3 (eRF1 and eRF3) are recruited to promote translation termination when a stop codon on the mRNA enters at the ribosomal A-site. However, their overexpression increases termination efficiency only moderately, suggesting that other factors might be involved in the termination process. To determine such unknown components, we performed a genetic screen in Saccharomyces cerevisiae that identified genes increasing termination efficiency when overexpressed. For this purpose, we constructed a dedicated reporter strain in which a leaky stop codon is inserted into the chromosomal copy of the ade2 gene. Twenty-five antisuppressor candidates were identified and characterized for their impact on readthrough. Among them, SSB1 and snR18, two factors close to the exit tunnel of the ribosome, directed the strongest antisuppression effects when overexpressed, showing that they may be involved in fine-tuning of the translation termination level.

A little gene with big effects: a serT mutant is defective in flgM gene translation

A conditional-lethal mutant was isolated as having a flagellar regulatory phenotype at 30 degrees C and being unable to grow at 42 degrees C. Chromosomal mapping localized the mutation to the serT gene, which encodes an essential serine tRNA species (tRNA((cmo)5UGA)(Ser)). DNA sequence analysis revealed the mutation to be a single base change in G:A at position 10 of the serT gene that lies within the D-stem of the essential tRNA((cmo)5)UGA(Ser) species. tRNA((cmo)5)UGA(Ser) recognizes UCA, UCG, and UCU codons, but UCU is also recognized by tRNA(GGA)(Ser) and UCG by tRNA(CGA)(Ser). No other tRNAs are known to read the UCA codon. Thus, the UCA codon is specifically recognized by tRNA((cmo)5)UGA(Ser). We show that the anti-sigma(28) activity of FlgM is defective in the serT mutant strain. The serT allele causes a 10-fold increase in sigma(28)-dependent fliC promoter transcription, indicating a defect in FlgM anti-sigma(28) activity in the presence of the serT mutation. The flgM gene contains only one UCA codon. Changing the UCA of flgM to ACG reversed the effect of the serT allele. Implications for context effects in regulation of gene expression are discussed.

A sequence in the Escherichia coli fdhF "selenocysteine insertion Sequence" (SECIS) operates in the absence of selenium

The UGA codon context of the Escherichia coli fdhF mRNA includes an element called the selenocysteine insertion sequence (SECIS) that is responsible for the UGA-directed incorporation of the amino acid selenocysteine into a protein. Here, we describe an extended fdhF SECIS that includes the information for an additional function: the prevention of UGA readthrough under conditions of selenium deficiency. This information is contained in a short mRNA region consisting of a single C residue adjacent to the UGA on its downstream side, and an additional segment consisting of the six nucleotides immediately upstream from it. These two regions act independently and additively, and probably through different mechanisms. The single C residue acts as itself; the upstream region acts at the level of the two amino acids, arginine and valine, for which it codes. These two codons at the 5' side of the UGA correspond to the ribosomal E and P sites. Here, we present a model for the E. coli fdhF SECIS as a multifunctional RNA structure containing three functional elements. Depending on the availability of selenium, the SECIS enables one of two alternatives for the translational machinery: either selenocysteine incorporation into a polypeptide or termination of the polypeptide chain.

Does disparate occurrence of autoregulatory programmed frameshifting in decoding the release factor 2 gene reflect an ancient origin with loss in independent lineages?

In Escherichia coli an autoregulatory mechanism of programmed ribosomal frameshifting governs the level of polypeptide chain release factor 2. From an analysis of 20 sequences of genes encoding release factor 2, we infer that this frameshift mechanism was present in a common ancestor of a large group of bacteria and has subsequently been lost in three independent lineages.

Translational termination in Escherichia coli: three bases following the stop codon crosslink to release factor 2 and affect the decoding efficiency of UGA-containing signals

The observations that the Escherichia coli release factor 2 (RF2) crosslinks with the base following the stop codon (+4 N), and that the identity of this base strongly influences the decoding efficiency of stop signals, stimulated us to determine whether there was a more extended termination signal for RF2 recognition. Analysis of the 3' contexts of the 1248 genes in the E.coli genome terminating with UGA showed a strong bias for U in the +4 position and a general bias for A and against C in most positions to +10, consistent with the concept of an extended sequence element. Site-directed crosslinking occurred to RF2 from a thio-U sited at the +4, +5 and +6 bases following the UGA stop codon but not beyond (+7 to +10). Varying the +4 to +6 bases modulated the strength of the crosslink from the +1 invariant U to RF2. A strong selection bias for particular bases in the +4 to +6 positions of certain E. coli UGANNN termination sites correlated in some cases with crosslinking efficiency to RF2 and in vivo termination signal strength. These data suggest that RF2 may recognise at least a hexanucleotide UGA-containing sequence and that particular base combinations within this sequence influence termination signal decoding efficiency.

Is the in-frame termination signal of the Escherichia coli release factor-2 frameshift site weakened by a particularly poor context?

The synthesis of release factor-2 (RF-2) in bacteria is regulated by a high efficiency +1 frameshifting event at an in-frame UGA stop codon. The stop codon does not specify the termination of synthesis efficiently because of several upstream stimulators for frameshifting. This study focusses on whether the particular context of the stop codon within the frameshift site of the Escherichia coli RF-2 mRNA contributes to the poor efficiency of termination. The context of UGA in this recoding site is rare at natural termination sites in E.coli genes. We have evaluated how the three nucleotides downstream from the stop codon (+4, +5 and +6 positions) in the native UGACUA sequence affect the competitiveness of the termination codon against the frameshifting event. Changing the C in the +4 position and, separately, the A in the +6 position significantly increase the termination signal strength at the frameshift site, whereas the nucleotide in the +5 position had little influence. The efficiency of particular termination signals as a function of the +4 or +6 nucleotides correlates with how often they occur at natural termination sites in E.coli; strong signals occur more frequently and weak signals are less common.

On the relationship between preferred termination codon contexts and nonsense suppression in human cells

The nucleotide sequences 3' to the translational termination codons in a collection of human genes have been analysed for evidence of a preferred 3' context for natural UAG codons. The aim was to see whether human UAG contexts can be related to the recent demonstration of the effects of 3' context on nonsense suppression in human cells. Since mammalian genomes are known to consist of a patchwork of blocks of sequences or 'isochores' with different G+C contents, the collection of genes was split into 5 classes containing genes with similar frequencies of G+C at the 3rd position of synonymous codons. This analysis revealed that the frequency of bases 3' to UAG varies with the G+C frequency of the gene, and that these changes were mirrored by changes in the patterns of bases in GN and AGN strings. The identity of the next 3' base appears therefore to be determined by genome wide changes in G+C composition, rather than selection to maintain a particular tetranucleotide stop signal. These findings argue strongly that the failure to find bias in the patterns of bases used in human coding sequences is an insensitive guide for the existence of codon usage or codon context effects during translation in human cells.

Codon context and protein synthesis: enhancements of the genetic code

The sequence around stop codons strongly affects termination efficiency and the probability of alternative events to termination such as frameshifting and stop codon readthrough. Where tRNA suppressors of nonsense codons are present, both the efficiency of suppression and of the termination process appear to be affected by stop codon context. Since context can affect suppressor tRNA function directly, an influence on sense codon translation or suppression might be expected, but has not yet been clearly demonstrated. Statistical analyses of coding sequences indicate non-random contexts for both stop and sense codons, and non-random occurrence of codon pairs. Highly expressed genes show clear preferences among stop codons and their contexts, whereas inefficient stop signals are exploited in a variety of recoding processes.

The signal for translational readthrough of a UGA codon in Sindbis virus RNA involves a single cytidine residue immediately downstream of the termination codon

The nucleotide sequences surrounding termination codons influence the efficiency of translational readthrough. In this report, we examined the sequence requirement for efficient readthrough of the UGA codon in the Sindbis virus genomic RNA which regulates production of the putative viral RNA polymerase, nsP4. The UGA codon and its neighboring nucleotide sequences were subcloned into a heterologous coding context, and readthrough efficiency was measured by cell-free translation of RNA transcripts in rabbit reticulocyte lysates. The CUA codon immediately downstream of the UGA codon was found to be sufficient for efficient translational readthrough. Further mutagenesis of residues in the CUA triplet demonstrated that mutations at the second or third residues following the UGA codon (U and A, respectively) had little effect on readthrough efficiency. In contrast, replacement of the cytidine residue immediately downstream of the UGA codon with any of the other three nucleotides (U, A, or G) dramatically reduced the readthrough efficiency from approximately 10% to less than 1%. These results show that a simple sequence context can allow efficient readthrough of UGA codons in a mammalian translation system. Interestingly, compilation studies of nucleotide sequences surrounding eukaryotic termination codons indicate a strong bias against cytidine residues immediately 3' to UGA termination codons. Taken together with our results, this bias may reflect a selective pressure for efficient translation termination for most eukaryotic gene products.

Termination of translation in bacteria may be modulated via specific interaction between peptide chain release factor 2 and the last peptidyl-tRNA(Ser/Phe)

The 5' context of 671 Escherichia coli stop codons UGA and UAA has been compared with the context of stop-like codons (UAC, UAU and CAA for UAA; UGG, UGC, UGU and CGA for UGA). We have observed highly significant deviations from the expected nucleotide distribution: adenine is over-represented whereas pyrimidines are under-represented in position -2 upstream from UAA. Uridine is over-represented in position -3 upstream from UGA. Lysine codons are preferable immediately prior to UAA. A complete set of codons for serine and the phenylalanine UUC codon are preferable immediately 5' to UGA. This non-random codon distribution before stop codons could be considered as a molecular device for modulation of translation termination. We have found that certain fragment of E. coli release factor 2 (RF2) (amino acids 93-114) is similar to the amino acid sequences of seryl-tRNA synthetase (positions 10-19 and 80-93) and of beta (small) subunit (positions 72-94) of phenylalanyl-tRNA synthetase from E. coli. Three-dimensional structure of E. coli seryl-tRNA synthetase is known [1]: Its N-terminus represents an antiparallel alpha-helical coiled-coil domain and contains a region homologous to RF2. On the basis of the above-mentioned results we assume that a specific interaction between RF2 and the last peptidyl-tRNA(Ser/Phe) occurs during polypeptide chain termination in prokaryotic ribosomes.

Influence of codon context on UGA suppression and readthrough

We studied the influence of the codon context on UGA suppression by a suppressor tRNA and on UGA readthrough by a normal tRNA in Escherichia coli. This was done by a series of constructs where only the immediate context of the TGA codon was varied by only one nucleotide at a time. For both UGA suppression and UGA readthrough the codon context had a similar influence according to the following rules. (1) The nature of the nucleotide immediately adjacent to the 3' side of the UGA is an important determinant; at that position the level of UGA translation is influenced by the nucleotides in the order A greater than G greater than C greater than U. (2) At extremely high or low levels of UGA translation this influence of the adjacent 3' nucleotide is not seen. (3) In all cases, the nature of both the nucleotide immediately adjacent to the 5' side of the codon and that following the base adjacent to the 3' side of the codon have little effect, if any, on UGA translation. The varying influence of the codon context effect on UGA translation is discussed in relation to its role in gene expression.

The signal for a leaky UAG stop codon in several plant viruses includes the two downstream codons

Expression of the RNA replicase domain of tobacco mosaic virus (TMV) and certain protein-coding regions in other plant viruses, is mediated by translational readthrough of a leaky UAG stop codon. It has been proposed that normal tobacco tyrosine tRNAs are able to read the UAG codon of TMV by non-conventional base-pairing but recent findings that stop codons can also be bypassed as a result of extended translocational shifts (tRNA hopping) have encouraged a re-examination. In light of the alternatives, we investigated the sequences flanking the leaky UAG codon using an in vivo assay in which bypass of the stop codon is coupled to the transient expression of beta-glucuronidase (GUS) reporter genes in tobacco protoplasts. Analysis of GUS constructions in which codons flanking the stop were altered allowed definition of the minimal sequence required for read through as UAG-CAA-UUA. The effects of all possible single-base mutations in the codons flanking the stop indicated that 3' contexts of the form CAR-YYA confer leakiness and that the 3' context permits read through of UAA and UGA stop codons as well as UAG. Our studies demonstrate a major role for the 3' context in the read through process and do not support a model in which teh UAG is bypassed exclusively as a result of anticodon-codon interactions. No evidence for tRNA hopping was obtained. The 3' context apparently represents a unique sequence element that affects translation termination.

Codon context

The analysis of coding sequences reveals nonrandomness in the context of both sense and stop codons. Part of this is related to nucleotide doublet preference, seen also in non-coding sequences and thought to arise from the dependence of mutational events on surrounding sequence. Another nonrandom context element, relating the wobble nucleotides of successive codons, is observed even when doublet preference, codon usage and bias in amino acid doublets are all allowed for. Several phenomena related to protein synthesis have been shown in vivo to be affected by the nucleotide sequence around codons. Thus, nonsense and missense suppression, elongation rate, precision of tRNA selection and polypeptide chain termination are all affected by codon context. At present, it remains unclear how these phenomena may influence the evolution of nonrandomness in the context of codons in natural sequences.

Sequence analysis suggests that tetra-nucleotides signal the termination of protein synthesis in eukaryotes

An increasing number of cases where tri-nucleotide stop codons do not signal the termination of protein synthesis are being reported. In order to identify what constitutes an efficient stop signal, we analysed the region around natural stop codons in genes from a wide variety of eukaryotic species and gene families. Certain stop codons and nucleotides following stop codons are over-represented, and this pattern is accentuated in highly expressed genes. For example, the preferred signal for Saccharomyces cerevisiae and Drosophila melanogaster highly expressed genes is UAAG, and generally the signals UAA(A/G) and UGA(A/G) are preferred in eukaryotes. The GC% of the organism or DNA region can affect whether there is A or G in the second or fourth positions. We suggest therefore, that the stop codon and the nucleotide following it comprise a tetra-nucleotide stop signal. A model is proposed in which the polypeptide chain release factor, a protein, recognises this sequence, but will tolerate some substitution, particularly A to G in the second or third positions.

Third position base changes in codons 5' and 3' adjacent UGA codons affect UGA suppression in vivo

The base sequence around nonsense codons affects the efficiency of nonsense codon suppression. Published data, comparing different nonsense sites in a mRNA, implicate the two bases downstream of the nonsense codon as major determinants of suppression efficiency. However, the results we report here indicate that the nature of the contiguous upstream codon can also affect nonsense suppression, as can the third (wobble) base of the contiguous downstream codon. These conclusions are drawn from experiments in which the two Ser codons UCU233 and UCG235 in a nonsense mutant form (UGA234) of the trpA gene in Escherichia coli have been replaced with other Ser codons by site-directed mutagenesis. Suppression of these trpA mutants has been studied in the presence of a UGA nonsense suppressor derived from glyT. We speculate that the non-site-specific effects of the two adjacent downstream bases may be largely at the level of the termination process, whereas more site-specific or codon-specific effects may operate primarily on the activity of the suppressor tRNA.

Construction of Escherichia coli amber suppressor tRNA genes. II. Synthesis of additional tRNA genes and improvement of suppressor efficiency

Using synthetic oligonucleotides, we have constructed 17 tRNA suppressor genes from Escherichia coli representing 13 species of tRNA. We have measured the levels of in vivo suppression resulting from introducing each tRNA gene into E. coli via a plasmid vector. The suppressors function at varying efficiencies. Some synthetic suppressors fail to yield detectable levels of suppression, whereas others insert amino acids with greater than 70% efficiency. Results reported in the accompanying paper demonstrate that some of these suppressors insert the original cognate amino acid, whereas others do not. We have altered some of the synthetic tRNA genes in order to improve the suppressor efficiency of the resulting tRNAs. Both tRNA(CUAHis) and tRNA(CUAGlu) were altered by single base changes, which generated -A-A- following the anticodon, resulting in a markedly improved efficiency of suppression. The tRNA(CUAPro) was inactive, but a hybrid suppressor tRNA consisting of the tRNA(CUAPhe) anticodon stem and loop together with the remainder of the tRNA(Pro) proved highly efficient at suppressing nonsense codons. Protein chemistry results reported in the accompanying paper show that the altered tRNA(CUAHis) and the hybrid tRNA(CUAPro) insert only histidine and proline, respectively, whereas the altered tRNA(CUAGlu) inserts principally glutamic acid but some glutamine. Also, a strain deficient in release factor I was employed to increase the efficiency of weak nonsense suppressors.

The signal for the termination of protein synthesis in procaryotes

The sequences around the stop codons of 862 Escherichia coli genes have been analysed to identify any additional features which contribute to the signal for the termination of protein synthesis. Highly significant deviations from the expected nucleotide distribution were observed, both before and after the stop codon. Immediately prior to UAA stop codons in E. coli there is a preference for codons of the form NAR (any base, adenine, purine), and in particular those that code for glutamine or the basic amino acids. In contrast, codons for threonine or branched nonpolar amino acids were under-represented. Uridine was over-represented in the nucleotide position immediately following all three stop codons, whereas adenine and cytosine were under-represented. This pattern is accentuated in highly expressed genes, but is not as marked in either lowly expressed genes or those that terminate in UAG, the codon specifically recognised by polypeptide chain release factor-1. These observations suggest that for the efficient termination of protein synthesis in E. coli, the 'stop signal' may be a tetranucleotide, rather than simply a tri-nucleotide codon, and that polypeptide chain release factor-2 recognises this extended signal. The sequence following stop codons was analysed in genes from several other procaryotes and bacteriophages. Salmonella typhimurium, Bacillus subtilis, bacteriophages and the methanogenic archaebacteria showed a similar bias to E. coli.

Aminoglycoside suppression at UAG, UAA and UGA codons in Escherichia coli and human tissue culture cells

We have compared the suppression of nonsense mutations by aminoglycoside antibiotics in Escherichia coli and in human 293 cells. Six nonsense alleles of the chloramphenicol acetyl transferase (cat) gene, in the vector pRSVcat, were suppressed by growth in G418 and paromomycin. Readthrough at UAG, UAA and UGA codons was monitored with enzyme assays for chloramphenicol acetyl transferase (CAT), in stably transformed bacteria and during transient expression from the same plasmid in human 293 tissue culture cells. We have found significant differences in the degree of suppression amongst three UAG codons and two UAA codons in different mRNA contexts. However, the pattern of these effects are not the same in the two organisms. Our data suggest that context effects of nonsense suppression may operate under different rules in E. coli and human cells.

Nonsense suppression context effects in Escherichia coli bacteriophage T4

Nonsense suppression by supE44 has been examined in a collection of 14 T4 gene 22 and gene 23 UAG mutants, for which the precise gene location is known. In concordance with previous studies, UAG followed by a pyrimidine was inefficiently suppressed. However, among positions with similar 3' nucleotides, there was considerable variation in suppression efficiency. The competition between supE44 and Release Factor 1 (RF 1) was also investigated following the introduction of a multicopy RF 1 plasmid. An inverse relationship between the efficiency of suppression and RF 1 competition was observed.

Normal yeast tRNA(CAGGln) can suppress amber codons and is encoded by an essential gene

We have isolated a gene that can encode yeast tRNA(CAGGln). When present on a multicopy plasmid, this gene suppresses the phenotype of a number of amber mutants, but has no effect on the ocher mutants tested. We therefore conclude that the anticodon CUG in tRNA(CAGGln) can decode the amber codon UAG by G-U mispairing, possibly by wobble base-pairing in the first codon position. This represents the second example we have observed in this laboratory of nonsense suppression in yeast by natural tRNA(Gln), involving G-U mispairing in the first codon position. Replacing the genomic copy of the cloned gene with a disrupted tRNA gene results in recessive lethality in heterozygous diploids and is lethal to haploid cells. This lethality can be rescued by transformation of cells with a single copy plasmid containing the tRNA(CAGGln) gene. Thus, the gene encoding tRNA(CAGGln) is apparently essential for viability in yeast, suggesting that it is normally present as a single copy gene.

Functional interactions in vivo between suppressor tRNA and mutationally altered ribosomal protein S4

Ribosomal mutants (rpsD) which are associated with a generally increased translational ambiguity were investigated for their effects in vivo on individual tRNA species using suppressor tRNAs as models. It was found that nonsense suppression is either increased, unaffected or decreased depending on the codon context and the rpsD allele involved as well as the nature of the suppressor tRNA. Missense suppression of AGA and AGG by glyT(SuAGA/G) tRNA as well as UGG by glyT(SuUGG-8) tRNA is unaffected whereas suppression of UGG by glyT(SuUGA/G) or glyV(SuUGA/G) tRNA is decreased in the presence of an rpsD mutation. The effects on suppressor tRNA are thus not correlated with the ribosomal ambiguity (Ram) phenotype of the rpsD mutants used in this study. It is suggested that the mutationally altered ribosomes are changed in functional interactions with the suppressor tRNA itself rather than with the competing translational release factor(s) or cognate aminoacyl tRNA. The structure of suppressor tRNA, particularly the anticodon loop, and the suppressed codon as well as the codon context determine the allele specific functional interactions with these ribosomal mutations.

Influence of modification next to the anticodon in tRNA on codon context sensitivity of translational suppression and accuracy

Effects on translation in vivo by modification deficiencies for 2-methylthio-N6-isopentenyladenosine (ms2i6A) (Escherichia coli) or 2-methylthio-N6-(4-hydroxyisopentenyl)adenosine (ms2io6A) (Salmonella typhimurium) in tRNA were studied in mutant strains. These hypermodified nucleosides are present on the 3' side of the anticodon (position 37) in tRNA reading codons starting with uridine. In E. coli, translational error caused by tRNA was strongly reduced in the case of third-position misreading of a tryptophan codon (UGG) in a particular codon context but was not affected in the case of first-position misreading of an arginine codon (CGU) in another codon context. Misreading of UGA nonsense codons at two different positions was codon context dependent. The efficiencies of some tRNA nonsense suppressors were decreased in a tRNA-dependent manner. Suppressor tRNA which lacks ms2i6A-ms2io6A becomes more sensitive to codon context. Our results therefore indicate that, besides improving translational efficiency, ms2i6A37 and ms2io6A37 modifications in tRNA are also involved in decreasing the intrinsic codon reading context sensitivity of tRNA. Possible consequences for regulation of gene expression are discussed.

Constraints on codon context in Escherichia coli genes. Their possible role in modulating the efficiency of translation

The constraints on nucleotide sequences of highly and weakly expressed genes from Escherichia coli have been analysed and compared. Differences in synonymous codon spectra in highly and weakly expressed genes lead to different frequencies of nucleotides (in the first and third codon positions) and dinucleotides in the two groups of genes. It has been found that the choice of synonymous codons in highly expressed genes depends on the nucleotides adjacent to the codon. For example, lysine is preferably encoded by the AAA codon if guanosine is 3' to the lysine codon (AAA-G, P less than 10(-9)). And, on the contrary, AAG is used more often than AAA (P less than 0.001) if cytidine is 3' adjacent to lysine. Guanosine occurs more frequently than adenosine 5' to all the lysine codons (AAR, P less than 10(-5), i.e. NNG codons are preferred over the synonymous NNA codons 5' to the positions of lysine in the genes. The context effect was observed in nonsense and missense suppression experiments. Therefore, a hypothesis has been suggested that the efficiency of translation of some codons (for which the constraints on the adjacent nucleotides were found) can be modulated by the codon context. The rules for preferable synonymous codon choice in highly expressed genes depending on the nucleotides surrounding the codon are presented. These rules can be used in the chemical synthesis of genes designed for expression in E. coli.

Sense codons are found in specific contexts

The sequence environment of codons in structural genes has been investigated statistically, using computer methods. A set of Escherichia coli genes with abundant products was compared with a set having low gene product levels, in order to detect potential differences associated with expression. The results show striking non-randomness in the nucleotides occurring near codons. These effects are, unexpectedly, very much larger and more homogeneous among the genes with rare products. The intensity of effects in weakly expressed genes suggests that such non-random sequence environments decrease expression. In the weakly expressed set of genes, the 5' neighbor of a codon, and all positions of the 3' neighbor codon are biased. In the highly expressed genes, the first nucleotide of the next codon is a uniquely affected site. The distribution of non-randomness in weakly expressed genes suggests that sequence bias is primarily due to a constraint acting directly on the secondary or tertiary structure of the codon/anticodon. In highly expressed genes, the observed bias suggests an interaction between the codon/anticodon and a site outside the codon/anticodon. Much of the tendency to non-random near-neighbor sequences in weakly expressed genes can be ascribed to a correlation between nearby nucleotides and the wobble nucleotide of the codon, despite the fact that selection of such correlations will alter the amino acid sequence. The favored pattern, in genes expressed at low level, is R YYR or Y RRY. R indicates purine, Y indicates pyrimidine; the space is the boundary between codons. It seems likely that this preference for nearby sequences is the physical basis of the genetic context effect. Under this assumption such sequence biases will affect expression. On this basis, we predict new sites for contextual mutations which decrease expression, and suggest strategy for the design of messages having optimal translational activity.

Undermodification in the first position of the anticodon of supG-tRNA reduces translational efficiency

Two mutants of Escherichia coli, trmC1 and trmC2, which are both defective in the synthesis of 5-methylaminomethyl-2-thiouridine (mnm5s2U) were utilized to study the function of this complex modified nucleoside. Transfer RNAs specific for glutamine, glutamic acid and lysine as well as a specific ochre suppressor derived from lysine tRNA (tRNAUAAlys encoded by the supG allele), contain this modified nucleoside at position 34 (the wobble position). It was found that two different undermodified derivatives of mnm5s2U were present in the two trmC mutants, which suggests that the two mutations affect two different enzymatic activities. Using the lacI-Z fusion system (Miller and Albertini 1983), we found that the efficiency of supG-mediated suppression was reduced to 30%-90% of the wild-type value in the trmC mutants. The modification-deficient supG-tRNA in the mutants showed a higher sensitivity to codon context than the normal tRNAUAAlys.

Codon context effects in missense suppression

After our first observation of codon context effects in missense suppression ( Murgola & Pagel , 1983), we measured the suppression of missense mutations at two positions in trpA in Escherichia coli. The suppressible codons in the trpA messenger RNA were the lysine codons, AAA and AAG, and the glutamic acid codons, GAA and GAG. The mRNA sites of the codons correspond to amino acids 211 and 234 of the trpA polypeptide, positions at which glycine is the wild-type amino acid. Our data demonstrated codon context effects with both pairs of codons. The results indicate that suppression of AAA and AAG by mutant lysine transfer RNAs was more efficient at 211 than at 234, whereas suppression of GAA and GAG by two different mutant glycine tRNAs was more efficient at 234 than at 211. In general, the context effects were more pronounced with GAG and AAG than with GAA and AAA. (In some instances it appeared that suppression of GAA or AAA at a given position was more effective than suppression of GAG or AAG.) By contrast, no context effects were observed with a glyT suppressor of AAA and AAG, a glyT GAA/G-suppressor, and a glyU suppressor of GAG. Our observation of this phenomenon in missense suppression demonstrates that codon context can affect polypeptide elongation and that the effects can be different depending on the codons and tRNAs examined. It is suggested that tRNA-tRNA interaction on the ribosome is involved in the observed context effects.

An effect of codon context on the mistranslation of UGU codons in vitro

Effects of codon context on nonsense codon suppression may act either through release factor recognition of termination codons or aminoacyl-tRNA selection by the ribosome. The latter hypothesis has been studied by comparing misreading by Escherichia coli UGA suppressor tryptophan tRNA of UGU (cysteine) codons in two synthetic polymers, poly(U-G) and poly( U5 , G), which differ in sequence around the UGU codons. In vitro translation of these polymers in a cell-free system from E. coli yielded selection errors of 4 X 10(-3) and 1.75 X 10(-2) for UGU codons in poly(U-G) and poly( U5 , G), respectively. This difference suggests that codon context may significantly affect misincorporation of amino acids into protein.

A temperature-sensitive mutant of Escherichia coli that shows enhanced misreading of UAG/A and increased efficiency for some tRNA nonsense suppressors

A spontaneous mutant was isolated that harbors a weak suppressing activity towards a UAG mutation, together with an inability to grow at 43 degrees C in rich medium. The mutation is shown to be associated with an increased misreading of UAG at certain codon contexts and UAA. UGA, missense or frameshift mutations do not appear to be misread to a similar extent. The mutation gives an increased efficiency to several amber tRNA suppressors without increasing their ambiguity towards UAA. The ochre suppressors SuB and Su5 are stimulated in their reading of both UAG and UAA with preference for UAG. An opal suppressor is not affected. The effect of the mutation on the efficiency of amber and ochre suppressors is dependent on the codon context of the nonsense codon. The mutated gene (uar) has been mapped and found to be recessive both with respect to suppressor-enhancing ability as well as for temperature sensitivity. The phenotype is partly suppressed by the ochre suppressor SuC. It is suggested that uar codes for a protein, which is involved in translational termination at UAG and UAA stop codons.

Context effects: translation of UAG codon by suppressor tRNA is affected by the sequence following UAG in the message

The efficiency of various suppressor tRNAs in reading the UAG amber codon has been measured at 42 sites in the lacI gene. Results indicate that: (1) for all suppressors, efficiency is not an a priori value; rather, it is determined at each site by the specific reading context of the suppressed codon; (2) the degree of sensitivity to context effects differs among suppressors. Most affected is amber suppressor supE (su2), whose activity varies over a 20-fold range depending on context; (3) context effects are produced by residues present at the 3' side of the UAG codon. The most important role appears to be played by the base that is immediately adjacent to the codon. When this base is a purine, the amber codon is suppressed more efficiently than when a pyrimidine is in the same position. Superimposed on this initial pattern, the influence of bases further downstream to the UAG triplet can be detected also. The possibility is discussed that context effects are produced by the whole codon following UAG in the message.

Effects of surrounding sequence on the suppression of nonsense codons

Using a lacI-Z fusion system, we have determined the efficiency of suppression of nonsense codons in the I gene of Escherichia coli by assaying beta-galactosidase activity. We examined the efficiency of four amber suppressors acting on 42 different amber (UAG) codons at known positions in the I gene, and the efficiency of a UAG suppressor at 14 different UGA codons. The largest effects were found with the amber suppressor supE (Su2), which displayed efficiencies that varied over a 35-fold range, and with the UGA suppressor, which displayed a 170-fold variation in efficiency. Certain UGA sites were so poorly suppressed (less than 0.2%) by the UGA suppressor that they were not originally detected as nonsense mutations. Suppression efficiency can be correlated with the sequence on the 3' side of the codon being suppressed, and in many cases with the first base on the 3' side. In general, codons followed by A or G are well suppressed, and codons followed by U or C are poorly suppressed. There are exceptions, however, since codons followed by CUG or CUC are well suppressed. Models explaining the effect of the surrounding sequence on suppression efficiency are considered in the Discussion and in the accompanying paper.

UGA suppression by normal tRNA Trp in Escherichia coli: codon context effects

The nucleotide sequences at the 3' side of in-phase UGA termination codons in mRNAs of various prokaryotic genes were re-examined. An adenine (A) residue is found to be adjacent to the 3' side of UGA in mRNAs which code for readthrough proteins by the suppression of UGA by normal Escherichia coli tRNA Trp. It is suggested that the nature of the nucleotide following a UGA codon determines whether the UGA signals inefficiently or efficiently the termination of polypeptide chain synthesis: an A residue at this position permits the UGA readthrough process.

The influence of codon context on genetic code translation

A class of mutations that increase the deficiency of a suppressor tRNA in translating a particular amber codon has been characterized. The increased efficiency is due to a mutation resulting in a change in the mRNA that affects the nucleotide adjacent to the 3' side of the UAG triplet. Thus the interaction of tRNA with mRNA is influenced by mRNA sequences outside the triplet codon.

Isolation and characterization of context mutations affecting the suppressibility of nonsense mutations

Secondary mutations which increase the efficiency of suppression of nonsense mutations in the rIIB cistron of bacteriophage T4 have been isolated. These secondary mutations, called context mutations, map at sites very close to the nonsense codon, possibly on the promotor distal side. In context-nonsense double mutants, the amount of suppressed gene product is increased approximately 10-fold. The context mutations examined can act on the UAA (ochre) nonsense allele as well as on the UAG (amber) nonsense allele at a given site. These context mutations affect all suppression mechanisms analyzed (genetic suppressors, 5-fluorouracil suppression and spontaneous suppression). We suggest that context mutations affect information which is significant to the termination of polypeptide chains. According to our view, context mutations change the immediate neighborhood of nonsense mutations and so reduce the degree of resemblance to the sequences normally used for the termination of translation.

The suppression of defective translation by ppGpp and its role in the stringent response

Amino acid starvation is shown to decrease the fidelity of translation in E. coli. When proteins are analyzed by two-dimensional gel electrophoresis, missense errors are detected as an unusual heterogeneity in their isoelectric points, while premature termination of protein synthesis can be recognized by a decreased relative rate of synthesis of higher molecular weight proteins and by the the accumulation of a complex group of new small polypeptides. The types of translational errors observed are amino acid-specific. For example, starvation of a rel- strain for histidine produces severe isoelectric point heterogeneity with little evidence of premature termination, while starvation for leucine has little effect on the isoelectric points, but produces a drastic decrease in the average molecular weight of the newly synthesized protein. These differences suggest codon-specific errors in reading the genetic code. In these rel- cells, the effect of amino acid starvation on the rates of synthesis of complete individual proteins is both protein- and amino acid-specific. For example, ribosomal protein L7/12, which lacks histidine, is made at a higher level during histidine starvation than during isoleucine or leucine starvation. This suggests that in rel- cells, the modulation of gene expression caused by the lack of a particular amino acid is, at least in part, a function of the abundance of that amino acid in particular proteins-that is, the response of rel- cells to starvation is consistent with the theory that the inhibition of protein synthesis and the accompanying increase in error frequency both result from low levels of the correct substrate. In marked contrast, virtually no starvation-induced translational errors are detected in a rel+ strain, and the response is not amino acid-specific. Varoius data strongly imply that in this rel+ strain, essentially all the changes caused by starvation are due to the accumulation of ppGpp, which independently reduces protein synthesis, thereby suppressing all the direct effects of amino acid limitation seen in rel- strains (where ppGpp does not accumulate upon starvation). A model is presented which describes how ppGpp might suppress the direct effects of starvation and avoid the loss of translational fidelity. In addition, the direct and specific effects of ppGpp on gene expression are examined independently of amino acid starvation.

Low activity of -galactosidase in frameshift mutants of Escherichia coli

16 lac frameshift mutants induced by an acridine derivative, ICR-191D, in E. coli are leaky for beta-galactosidase activity. Activities of all mutants differ from each other and from the wild type in their stability to thermal denaturation. The leakiness is under ribosomal control, since it is strongly reduced by strA restrictive mutations and is restored by ram mutations that reverse restriction. Addition of streptomycin during growth has an effect similar to the presence of the ram mutation. These ribosomal alterations do not modify the thermal stability of the enzyme.It is suggested that the leakiness is due to an infrequent 2- or 4-base reading close to the frameshift mutation site. The possibility that not only the ribosome, but also the reading context in the messenger, plays a role in securing code fidelity is discussed.