D-stem mutation in an essential tRNA increases translation speed at the cost of fidelity

Abstract

The efficiency with which aminoacyl-tRNA and GTP-bound translation elongation factor EF-Tu recognizes the A-site codon of the ribosome is dependent on codons and tRNA species present in the polypeptide (P) and exit (E) codon sites. To understand how codon context affects the efficiency of codon recognition by tRNA-bound EF-Tu, a genetic system was developed to select for fast translation through slow-translating codon combinations. Selection for fast translation through the slow-translated UCA-UAC pair, flanked by histidine codons, resulted in the isolation of an A25G base substitution mutant in the D-stem of an essential tRNA LeuZ, which recognizes the UUA and UUG leucine codons. The LeuZ(A25G) substitution allowed for faster translation through all codon pairs tested that included the UCA codon. Insertion of leucine at the UCA serine codon was enhanced in the presence of LeuZ(A25G) tRNA. This work, taken in context with the Hirsh UGA nonsense suppressor G24A mutation in TrpT tRNA, provides genetic evidence that the post-GTP hydrolysis proofreading step by elongation factor Tu may be controlled by structural interactions in the hinge region of tRNA species. Our results support a model in which the tRNA bending component of the accommodation step in mRNA translation allows EF Tu time to enhance its ability to differentiate tRNA interactions between cognate and near-cognate mRNA codons.

Citation:Schrock MN, Parsawar K, Hughes KT, Chevance FFV (2025) D-stem mutation in an essential tRNA increases translation speed at the cost of fidelity. PLoS Genet 21(2): e1011569. https://doi.org/10.1371/journal.pgen.1011569

Editor:Jue D. Wang,, University of Wisconsin-Madison, UNITED STATES OF AMERICA

Received:October 8, 2024; Accepted:January 8, 2025; Published: February 4, 2025

Copyright: © 2025 Schrock et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability:All relevant data are within the manuscript and its Supporting information files.

Funding:This work was supported by PHS grants GM056141 and AI184395 from the National Institutes of Health to K.T.H. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. MNS, FFVC and KTH received salary support from the grants.

Competing interests: The authors have declared that no competing interests exist.

Introduction

DNA is transcribed by RNA polymerase into messenger RNA (mRNA), which is then translated into protein by the ribosome. In prokaryotes, the ribosome is composed of 50s (large) and 30s (small) subunits. During translation elongation, charged, aminoacyl transfer RNAs (tRNAs) interact with mRNA codon positions within the ribosome decoding center: the aminoacyl (A-), peptidyl (P-), or exit (E-) sites. Translation elongation involves four principal steps: 1) recognition of the incoming aminoacyl-tRNA, 2) accommodation of the aminoacyl-tRNA into the A-site, 3) peptide bond formation between adjacent amino acid residues, and 4) translocation of the ribosome down the mRNA (Fig 1).

[Figure omitted. See PDF.]

Fig 1. Overview of translation elongation.

tRNA’s are randomly brought into the A-site by GTP-bound elongation factor Tu (EF-Tu). If a non-cognate tRNA is brought in, it is immediately rejected. If a near-cognate or cognate tRNA is brought in for potential recognition, the ribosomal subunits start assessing full complementarity between the anticodon and codon, allowing for GTPase activation. The large ribosomal subunit will change conformation causing the tRNA in the A-site to bend, allowing EF-Tu bound to GTP to interact with the catalytic residues responsible for GTP hydrolysis. As the tRNA in the A-site relaxes, EF-Tu dissociates from the ribosome and accommodation of the cognate tRNA occurs. Near- cognate tRNA’s are rejected prior to accommodation. Accommodation of the tRNA into the A-site positions the 3’ CCA end of the tRNA for peptidyl transfer. Elongation factor G (EF-G) promotes peptide bond formation and translocation of the ribosome.

https://doi.org/10.1371/journal.pgen.1011569.g001

Selection of the correct aminoacyl-tRNA whose anticodon base pairs with the mRNA codon in the A-site of the ribosome during elongation is essential for translation fidelity. GTP-bound elongation factor-Tu (EF-Tu) randomly brings aminoacylated tRNAs to the A-site for potential recognition at a rate of about 1,000 times per second [1–5]. The 30s ribosomal subunit then assesses complementarity of the aminoacyl-tRNAs anticodon and the mRNA codon in the ribosomal A-site [6,7]. When the two bases of the codon (usually the first two bases) are complementary to the tRNA’s anticodon, the aminoacyl-tRNA remains in the A-site of the ribosome to assess pairing of all three bases. This is especially important at the wobble base, which has less strict base-pairing rules [8,9]. If a near-cognate or non-cognate aminoacyl-tRNA is located in the A-site, the ribosome senses an incorrect match and the aminoacyl-tRNA is rejected [10]. As the anticodon/codon complementary is being assessed, the ribosome undergoes a conformational change, which induces a bending in the tRNA. This in turn allows the GTP-binding domain of EF-Tu to interact with the L7/L12 stalk on the 50s ribosomal subunit and catalyze the hydrolysis of GTP [11–20]. Following GTP hydrolysis, the aminoacyl-tRNA is accommodated into the A-site [18,21], EF-Tu and GDP dissociate from the ribosome, and elongation factor-G facilitates translocation of the ribosome by three bases [22].

Translation is fine-tuned to maximize fitness of an organism by balancing translation speed with accuracy of amino acid incorporation into the growing polypeptide chain. Failure of translation to proceed at an appropriate rate can result in changes in mRNA stability, lowered protein expression, and even the synthesis of non-functional proteins [23–26]. There are two fidelity checks that an aminoacyl-tRNA in the A-site must undergo to maintain accuracy of translation. The first is anticodon-codon complementarity [9,27]. The second is sufficient energy supplied following a conformational change in the ribosome structure that results in a bending of the tRNA that is coupled to GTP hydrolysis [18,27,28]. The rate at which the correct aminoacyl-tRNA is incorporated into the A-site depends on its relative abundance in the cellular tRNA pool, which controls translation speed [29,30]. Other factors that contribute to translation efficiency include codon context and tRNA modifications [31–33]. Codon context can influence the speed of elongation to ensure appropriate expression of individual genes [34]. Codon context effects on translation occur when neighboring codons and tRNAs in the E-site and P-site influence A-site aminoacyl-tRNA recognition [31,33].

To gain a better understanding of the effects of codon context on translation, we report here a genetic selection to isolate fast translation mutants for codon pairs that are translated slowly. It utilizes the Salmonella histidine (his) biosynthetic operon control region, which acts as a translation speedometer through His codons. The region between the his operon promoter and the biosynthetic structural genes includes a small leader peptide gene sixteen codons long that includes seven consecutive His codons (Fig 2A–C). The leader peptide gene is followed by an RNA stem-loop Rho-independent transcriptional attenuator. Formation of this attenuator stem-loop prevents the continuation of RNA polymerase transcription into the his structural genes [35–37]. An alternative RNA secondary structure can also form that disrupts the attenuator stem-loop and allows transcription of the his operon structural genes. Attenuator formation depends on the speed that ribosomes move through the consecutive His codons. This translation speedometer allows the bacterium to assay the cellular concentration of charged histidyl-tRNA and produce the required amount of His biosynthetic enzymes. Transcription into the his structural genes demands stalling of the ribosome in the consecutive His codons. If the level of charged histidyl-tRNA is high, ribosome stalling does not occur, the ribosome continues behind RNA polymerase to the stop codon of the leader peptide sequence. This results in attenuator formation, removal of RNA polymerase from the DNA and no further his operon transcription. Importantly, this translation speedometer assay is independent of protein and mRNA stability [38]. We have shown that the his leader peptide system can be used to measure translation rates through individual codons or through multiple codon combinations [39]. Measurement of β-galactosidase expression of a lac operon reporter fused to the hisD gene allows for a quantitative readout through measurement of β-galactosidase activity. These assays provide an indirect measurement of translation speed through any codon that is placed in the run of consecutive His codons in the leader peptide gene. Furthermore, the use of MacConkey-lactose (Mac-Lac) indicator medium allows for a simple visual qualitative readout as colonies vary in color from white (Lac-) to pink to red (Lac+) with increasing β-galactosidase (Fig 2D). The highest levels of β-galactosidase activity result in the precipitation of bile salts present in the indicator medium.

[Figure omitted. See PDF.]

Fig 2. The his operon leader peptide control region.

(A) The his operon leader peptide region contains a 16 amino acid open reading frame with seven consecutive histidine (His) codons and RNA sequences that can form alternate secondary structures. At high levels of charged histidyl-tRNA, fast translation through the seven histidine codons results in the formation of a strong Rho-independent stem-loop structure (E-F) that attenuates further transcription by RNA polymerase into the downstream structural genes encoding the histidine biosynthetic enzymes. (B) Low levels of charged histidyl-tRNA results in stalled translation through the seven histidine codons of the leader peptide coding region, which results in alternative RNA secondary structures that lack the attenuator. This allows RNA polymerase to continue transcription through the his operon structural genes. (C) When transcription of the his leader peptide region is not coupled to translation, the default RNA secondary structures include the formation of the strong Rho-independent terminator. (D) Fusion of the lac operon to the hisD structural gene (hisD::Mud-lac) allows for a simple phenotypic determination on MacConkey lactose (Mac-Lac) indicator medium where attenuator formation in the his leader region results in low levels of his-lac transcription and a Lac- or white colony phenotype. Loss of attenuation in the his leader region results in high levels of his-lac transcription and a Lac+ or red colony phenotype and the precipitation of bile salts present in the medium.

https://doi.org/10.1371/journal.pgen.1011569.g002

The his leader peptide system has been used to determine the effect of codon context on translation speed through codon pairs [36,37,39]. This work revealed the importance of codon orientation (context) on the efficiency of translation through codon pairs [39]. For example, translation through the UAC-UCA (Tyr-Ser) codon pair was fast in the his leader peptide system while in the reverse order, translation of UCA-UAC was slow, resembling what was observed in assays with stop codons. In the present study, we modified the translation speedometer assay in a way that allows selection for faster translation through translation-slow codons. A selection for faster translation through the translation-slow UCA-UAC codon pair resulted in the isolation of a surprising mutation in the D-stem of tRNALeuZ (there are no leucine codons present in the his leader peptide sequence). Further analysis revealed that this mutant tRNALeuZ causes increased translation speeds in the his leader through many different codon pairs where UCA was one of the two codons in the pair being assayed. We describe the isolation of this tRNA mutation and its properties and discuss its implications for understanding control of translation speed and fidelity.

Results

The slow-translating phenotype of a UCA-UAC codon pair at positions His4 and His5 of the his leader peptide region is suppressed in cis by changing adjacent codons

In the his leader peptide translation speedometer system, apparent translation rates can be measured using a hisD::Mud-lac transcriptional reporter. Visualization of colony color intensity – due to the expression of the lac operon – on lactose-utilization indicator plates can be used to estimate speed of translation through the his leader gene, and expression can be quantified by measuring β-galactosidase (β-gal) activity. On MacConkey-lactose (Mac-Lac) indicator plates, fermentation of lactose lowers the pH of the colony and depending on the level of lac expression, colony color varies from a white (Lac-) to dark red (Lac+). As mentioned above, a UCA-UAC codon pair placed at the His4-His5 positions in the his leader peptide region exhibited a translation-slow phenotype yet the reverse orientation with UAC-UCA at His4-His5 was translated fast [39]. We tested for potential cis-acting effects on translation through the UCA-UAC slow-translating pair by changing the upstream codons at positions His2 and His3. All 64 codons were incorporated at either the His2 or His3 codon positions upstream of UCA-UAC at His4-His5 in the his leader peptide sequence and their effect on the expression of a hisD-lac operon reporter was determined on lactose indicator medium. Mac-Lac indicator plates were used to provide a qualitative screen for suppression of the translation-slow phenotype. The variation in degrees of translation stalling translates to variation of hisD-lac operon expression, which was visualized on lactose indicator medium. Fast translation through the his leader sequence occurs with all seven His codons present and results in a Mac-Lac white Lac- colony phenotype in this assay (Fig 2D). The translation-stalled phenotype of UCA-UAC codons at His4-His5 of the leader peptide exhibits a strong Lac+ phenotype (red colony with bile salts precipitation). With the UCA-UAC codons at His4-His5 in our assay, the addition of substitutions at His3 resulted in only four coding codons, UAU (Tyr), UAC (Tyr), UGG (Trp) and CCG (Pro), that retained the translation-slow phenotype of UCA-UAC at His4-His5 (Fig 3). The remaining fifty-seven coding codons introduced at His3 resulted in faster translation phenotypes (Fig 3). Similarly, for substitutions at His2, only four coding codons, UGG (Trp), CGA (Arg), AGA (Arg) and AGG (Arg), retained the translation-slow phenotype of UCA-UAC at His4-His5. Most substitutions at either His2 or His3 resulted in a faster-translation Lac- phenotype as compared to the translation-stalled Lac+ parent strain phenotype (Fig 3). It is worth noting that the slow translation phenotype of the UCA-UAC at His4-His5 is in the context of flanking histidine codons. These results are consistent with earlier observations suggesting that codon context can have a profound effect on the efficiency of mRNA translation [34].

[Figure omitted. See PDF.]

Fig 3. Effect of adjacent codons on the slow translation phenotype through the UCA-UAC codon pair placed at the His4-His5 codon positions in the his leader peptide.

The presence of UCA-UAC at the His4-His5 positions in the leader peptide sequence resulted in a translation-slow phenotype similar to what was observed with a stop codon placed at either his4 or His5 [39]. The effect of codons at positions His2 and His3 on the slow translation phenotype of UCA-UAC at His4-His5 was determined. The His2 and His3 codons were changed to all 64 codons and the effect on translation through the his leader peptide was assayed on Mac-Lac indicator medium using a his-lac reporter (hisD9953::Mud-lac) to measure his operon transcription levels. The Mac-Lac plate phenotypes are shown as ranging from white (Lac-) to dark red (Lac+). Intermediate Mac-Lac plate phenotypes are indicated by the gradation of darkening pink colors. The Mac-Lac color phenotypes represent an indirect readout of the speed of translation through the his leader peptide region.

https://doi.org/10.1371/journal.pgen.1011569.g003

A hisD-sacB reporter provides a positive selection against slow translation speed in the his leader peptide translation speedometer system

Using the his leader peptide translation speedometer system we were able to distinguish seven Mac-Lac colony color phenotypes that we have named a (white), b through f (increasing dark shades of red) to g (dark red with bile salts precipitation visible under the colony) (Fig 4B). For example, when one of the seven consecutive His codons in the leader peptide was replaced by a stop codon, a g phenotype on Mac-Lac plates was observed. However, when translation is fastest through the his leader peptide (attenuation is occurring) we observed the a phenotype (Fig 4B).

[Figure omitted. See PDF.]

Fig 4. Use of sacB and lac operon reporters to select and screen for mutants resulting in faster translation in the his leader peptide region.

(A) Growth phenotypes on MacConkey lactose (Mac-Lac) and sucrose-containing indicator plates for strains containing either lac operon or sacB reporters expressed from the his operon control region that have different speeds of translation through the his leader peptide region. A fast-translation speed (labeled A) results in a Lac-, white colony phenotype and growth on sucrose-containing media for his-lac or his-sacB reporter constructs, respectively. A slow-translation speed (labeled G) results in a Lac+ with precipitation of bile salts red colony phenotype and no growth on sucrose-containing media for his-lac or his-sacB reporter constructs, respectively. Intermediate translation speeds are labeled B through F. Classification of lactose phenotype are indicated by a letter from A-G. ΔhisD::sacB, in response to different translation rates results in growth or no growth on 6% sucrose media, +: robust growth, +/-: smaller single colonies and mucoid phenotype, +/--: no single colonies and mucoid phenotype, and -: no growth. Phenotypes for Mac-Lac were determined using strains TH2141, TH28870, TH28871, TH28872, TH28873, TH28874, TH28875. A 6% sucrose plate with the range of life/death seen with A-G. Phenotypes for sucrose growth phenotypes were determined using strains TH27455, TH28679, TH28680, TH28781, TH28782, TH28783, TH28784. (B) The approximate β-galactosidase (β-gal) activities and growth phenotypes on sucrose (Suc) medium associated with the different his operon expression levels (A through G) are indicated. (W = white, P = pink, R = red, CR = red colony center, RPPT = red colony with precipitated bile salts). (C) A strain that is diploid for the his biosynthetic operon was constructed to select for mutants resulting in a faster-translation speed through the his leader peptide region. One of the duplicated his operon regions contains a replacement of the hisD structural gene with the sacB coding sequence; the second his operon contains a lac operon fusion to the hisD structural gene. The resulting diploid strain was used to isolate mutants that were simultaneously defective in the expression of both his operons so that both the sacB and lac operon reporters were not expressed. Stalled translation through the his leader peptide region results in a G phenotype (Suc- Lac+). Fast translation is selected for on 6% sucrose plates, and then screened on MacConkey- lactose indicator plates for mutants with a Suc+ Lac- phenotype. In this study TH27508 was used to isolate fast-translation mutants.

https://doi.org/10.1371/journal.pgen.1011569.g004

In order to isolate mutants that result in faster translation through translation-slow codons in the his leader gene, the selectable marker sacB was employed. SacB is levan sucrase, which converts sucrose to levan polymers. In the periplasm of Gram-negative bacteria, levan accumulation results in cell death [40]. A selection strain was constructed in which the hisD gene was replaced by a sacB gene and a chloramphenicol resistance cassette (FCF) (ΔhisD::sacB-FCF) so that sacB expression is dependent on the his operon promoter and on failure of attenuation (Fig 4A). Slow translation of the his leader, which produces a g (Lac+) phenotype on Mac-Lac plates, results in cell death on 6% sucrose media (Suc-), whereas fast translation produces an A (Lac-) Mac-Lac phenotype and growth on 6% sucrose media (Suc+) (Fig 4A). The transition from an a through c-f to g white to pink to darkening red color phenotypes corresponds to increasing lac operon expression (Fig 4B). Intermediate c-f colony color translation phenotypes often have mucoid colonies on 6% glucose. A strain that is diploid for the region of the chromosome that includes the his operon was constructed that carries the hisD::Mud-lac indicator in one of the duplicated regions and ΔhisD::sacB-FCF construct in the other (Fig 4C). This construct allows for the selection and screening of fast-translation mutants that have a Suc+ Lac- phenotype; the diploidy eliminates survival of recessive mutants that alter the leader gene or attenuator.

Isolation of trans-acting mutations that increase translation rates

A strain merodiploid for the his operon carrying the hisD::Mud-lac reporter in one his operon and the ΔhisD::sacB-FCF construct in the other his operon (above) with the slowly translated UCA-UAC codon pair replacing His4-His5 in both his leader genes was constructed. This strain has a Lac+ Suc- growth phenotype due to failure of attenuation and high levels of transcription into both his operons. Selection for Suc+ and screening for Lac- should isolate mutants that allow attenuation to occur in this strain (see Fig 3A). Four mutant types were expected to have a Suc+ Lac- phenotype: (i) global regulatory mutants defective in transcription of the his operon promoter(ii) global regulatory mutants defective in translation of both sacB and the lac operon(iii) mutants that result in the inability to translate through the his leader peptide mRNA upstream of the seven consecutive His codons and result in attenuator formation, and (iv) mutants that result in an increase in translation speed through the UCA-UAC codon pair.

Selection for spontaneous Suc+ cells was performed with 15 independent cultures. A total of ~6 x 108 cells from each culture was subject to Suc+ selection conditions yielding mutants at the relatively high frequency of ~1 in 105 cells. This is a frequency typical for the loss-of-function of a single Salmonella gene. We presumed the majority of these did not affect attenuation and were defective in the ΔhisD::sacB-FCF locus such that SacB function was decreased or eliminated. Only 1/103 of the Sac+ mutants had a Lac- phenotype indicating that the mutation affected expression of both his operon constructs. These Suc+ Lac- mutants were found to reside in four chromosomal linkage classes and DNA sequence analysis identified mutations in two genes encoding subunits of RNA polymerase (rpoA and rpoD) and two essential tRNA genes (thrU and leuZ).

Global regulatory mutants in rpoA and rpoD, defective in transcription of the his operon promoter

Global regulatory mutants were expected that decrease transcription enough to lower sacB expression to a non-toxic level. Transcription is carried out by RNA polymerase, which has five subunits α2ββ’ω. RpoA is the α-subunit of RNA polymerase. The C-terminal domain of Salmonella RpoA (residues 249-329) utilizes the UP element of promoters to enhance recognition [41]. This region includes the Suc+ Lac- rpoA (E261K) mutant suppressor that came out of the selection. We presume that the RpoA E261K substitution interferes with promoter recognition, thus resulting in decreased transcription of both his operons. The other transcription-affected mutant was isolated in rpoD (R605L). RpoD is sigma70, the primary promoter recognition factor in Salmonella Typhimurium. RpoD is commonly divided into four contiguous peptide regions, where Region 4 binds the -35 promoter region to allow recognition by RNA polymerase [42]. R506L in RpoD is in region 4 of the protein, which is expected to affect the ability of RpoD to recognize the -35 promoter regions of the two copies of the his operon promoter. Thus, both mutations likely decrease promoter recognition by RNA polymerase which leads to decreased transcription and translation, resulting in the observed Suc+ Lac- phenotype.

ThrU(C40A) results in a general defect in mRNA translation that is suppressed by gene amplification

ThrU is the only tRNA that recognizes the ACA(Thr) codon, and it is an essential tRNA in Salmonella. The thrU(C40A) allele has a base substitution in the anticodon stem (Fig 5A). The ThrU(C40A) substitution results in the suppression of the slow translation phenotype with the UCA-UAC (Ser-Tyr) codon pair at His4-His5 in the his leader peptide. The effect of the ThrU(C40A) substitution is an approximately three-fold decrease in hisD::Mud-lac expression from 140 ± 13 β-Gal units in a thrU+ background to 44 ± 15 units in the thrU(C40A) background (Fig 5B). The ThrU(C40A) substitution represented a potential global regulatory mutant defective in translation of both sacB and the lac operon to produce the Suc+ Lac- phenotype.

[Figure omitted. See PDF.]

Fig 5. The ThrU(C40A) substitution results in his leader attenuation.

(A) The RNA secondary structure of tRNAThrU showing known RNA base modifications. ThrU is an essential threonine tRNA, as it is the only tRNA to decode ACA codons (red). The C40A substitution in ThrU is indicated by an arrow. (B) Quantification of the effect of ThrU(C40A) on hisD-lac expression by β-galactosidase assay in strains with different codon pair substitutions at positions His4-His5 in the His leader peptide. The effects of the ThrU(C40A) allele on his leader translation speed compared to ThrU+ with different codon pairs at His4-His5 positions in the his leader peptide region was measured and quantified. Various his leader constructs were tested, WT hisL (TH2141, TH28797), hisL(His4-His5 = UCA-UAC) (TH27505, TH27798), hisL(His4-His5 = CCC-CCC) (TH17685, TH28798) and hisL(His4-His5 = CGA-CGA) (TH28084, TH28799). One-tailed heteroscedastic t-test showed that all data sets were statistically different (P = <0.001) (C) Quantification of the effect of ThrU(C40A) on hisD-lac expression by β-galactosidase assay in strains with different codon substitutions at codon 2 in the His leader peptide. The effects of the ThrU(C40A) allele on his leader translation speed compared to ThrU+ with different codon pairs at the codon 2 position in the his leader peptide region was measured and quantified. Various his leader constructs were tested, WT hisL(codon 2 = ACA) (TH27505, TH27698), hisL(codon 2 = ACT) (TH29795, TH29800), hisL(codon 2 = ACG) (TH29794, TH29799), hisL(codon 2 = TGT) (TH29792, TH29797), hisL(codon 2 = GAC) (TH29793, TH29798), and hisL(codon 2 = GTT) (TH29796, TH29801). One-tailed heteroscedastic t-test showed that all data sets were statistically different (P = <0.001) (D) The MudF transposon contains an intact lac operon with its native lac promoter-operator region. The effect of the ThrU(C40A) allele on lac operon expression independent of his operon expression was determined by measuring β-gal levels expressed from a MudF chromosomal insertion unlinked to the his operon (zdx-3729::MudF) (TH24401, TH28796). β-Galactosidase activities are listed in S3 Table. One-tailed heteroscedastic t-test showed that all data sets were statistically different (P = <0.001).

https://doi.org/10.1371/journal.pgen.1011569.g005

We tested if the ThrU(C40A) effect was specific to the his attenuation mechanism. If this were true, we expected that β-galactosidase expression in hisD::Mud-lac strains with different slow translating codon pairs in the his leader peptide should be reduced. We tested the effect of the ThrU(C40A) substitution on hisD::Mud-lac expression with leader peptides carrying either a CCC-CCC (Pro-Pro) or CGA-CGA (Arg-Arg) at His4-His5 in the his leader peptide. Proline is the only N-alkylamino acid (imino acid) among the 20 natural amino acids. It has been shown that translation of proline codons occurs at a slower rate because the cyclic proline residue constrains the secondary structure of the translating protein as compared to insertion of the other 19 amino acids [43]. Arginine codons are also translated slowly due to the positive charge interfering with transit through the exit channel on the ribosome [44]. The presence of the ThrU(C40A) substitution resulted in decreased β-Gal activities for both constructs. With the CCC-CCC codon pair at His4-His5, the levels of hisD::Mud-lac expression decreased from 290 ± 9 to 49 ± 3 β-Gal units. With the CGA-CGA codon pair at His4-His5, the levels of hisD::Mud-lac expression decreased from 270 ± 20 to 49 ± 3 β-Gal units (Fig 5B).

The sequence of the his leader peptide has a threonine codon, ACA, at amino acid 2. Previous work on the his leader peptide has shown that early termination of translation in the his leader peptide results in attenuation [37]. We hypothesized that either ThrU(C40A) decodes ACA poorly resulting in premature stalling and termination of His leader peptide translation preventing hisD::sacB-FCF and hisD::Mud-lac from being expressed or the ThrU(C40A) had an indirect effect on leader peptide translation independent of the ACA codon at amino acid 2. To test these possibilities, we replaced codon 2 of the His leader peptide with different codons that had a similar hisD-lac expression phenotype (Mac-Lac Red) as the original strain TH27505 (hisL amino acid 2 = ACA, His4-His5 = UCA(Ser)-UAC(Tyr)). Five codon replacements at amino acid 2 were tested: UCU (Cys), GAC (Asp), ACU (Thr), ACG (Thr) and GUU (Val). Unlike the UCA codon, which is only recognized by ThrU, the ACU (Thr) codon is recognized buy three tRNA species: ThrT, ThrU and ThrV, and the ACG (Thr) codon is recognized by ThrU and ThrX. The effects of the amino acid 2 substitutions on hisD-lac expression in the presence or absence of the ThrU(C40A) allele was determined by measuring β-gal levels and the results are shown in Fig 5C. For all five strains the presence of the ThrU(C40A) allele resulted in a similar decrease in β-gal levels (3- to 5-fold) as the parent strain. These results indicates that the effect of the ThrU(C40A) substitution is not due to a defect in translation at the ACA(Thr) codon at amino acid 2 of the His leader and likely results in a general defect in mRNA translation.

If ThrU(C40A) does decrease general translation of the lac operon in the hisD::Mud-lac fusion, we would expect reduced expression of a lac operon to not be under the control of the his operon attenuation control mechanism. Salmonella has no native lac operon, but we inserted a MudF element [45], which carries an intact E. coli lac operon, into a chromosomal location that is not linked to the his operon (zdx-3729::MudF). We tested the effect of the ThrU(C40A) substitution on lacZ expression from the MudF element. The β-Gal activity was reduced ~3-fold in the thrU(C40A) background (68 ± 11 β-Gal units) compared to β-Gal levels in the thrU+ strain background (240 ± 2 β-Gal units) (Fig 5B). This result demonstrates that the ThrU(C40A) effect was due to an overall reduction in cellular translation.

In working with the various hisD-lac fusion constructs in the thrU(C40A) mutant background, we observed a high reversion rate to a Lac+ and a larger-colony phenotype when streaked on Mac-Lac indicator plates. Strain TH27698 (hisL10592(His4-5 = TCA-TAC) hisD993::MudJ thrU(C40A)) was streaked onto two Mac-Lac plates and incubated at 30°C and 37°C. The frequency of Lac+, larger-colony phenotype mutants appeared to be greater at 37°C. Five independent cultures were started from small, Lac- (Mac-Lac white) colonies and grown overnight at either 30°C or 37°C. The cultures were serially diluted, plated on Mac-Lac indicator plates and incubated at 30°C. Plates from cultures grown at 30°C yielded larger-colony, Lac+ mutants at frequencies of 10-3 to 10-4, while plates from cultures grown at 37°C yielded larger-colony, Lac+ mutants at frequencies of ~10-2. These mutant frequencies are those expected from chromosomal duplication events rather than chromosomal mutations. The genomes of four independent larger-colony, Lac+ mutants were sequenced and all four were found to be duplicated for the region of the chromosome covering the thrU locus (S1 Fig). It had been previously reported that this region of the chromosome experiences duplications at a frequency of ~10-2 when grown in LB medium at 37°C [46], which is consistent with our findings. This suggests that the ThrU(C40A) allele results in a temperature-sensitive, slow-growth phenotype due to reduced aminoacyl-ThrU in the cell, which can be compensated by increasing the gene dosage of the thrU locus in the cell.

tRNALeuZ (A25G) does not affect translation of lacZ or sacB

As mentioned above, the Suc+ Lac- mutants included a leuZ mutant allele LeuZ(A25G). Genome sequencing revealed the presence of a second mutation in the structural gene of polynucleotide phosphorylase, pnp, resulting in an I645T substitution. When the leuZ(A25G) allele was separated from the pnp (I654T) allele, the cells grew poorly at 37°C. Thus, all further work was carried out in a leuZ(A25G) pnp(I654T) double mutant background. The leuZ(A25G) pnp(I654T) double mutant also resulted in a faster translation phenotype with the UCA-UAC codon pair at His4-His5 in the his leader peptide at 37°C (all substitutions discussed below are at this position). tRNALeuZ is an essential leucine tRNA that recognizes UUA and UUG codons. Polynucleotide phosphorylase (Pnp) is a 3’ to 5’ exoribonuclease. Pnp has been shown to degrade unstable uncharged tRNAs. Loss of Pnp increases the intracellular levels of precursor tRNAs [47–49]. The leuZ(A25G) allele was separated from the pnp allele and the resulting leuZ(A25G) pnp+ strain showed a poor-growth phenotype at 37°C (Fig 6B). We conclude that the pnp(I654T) substitution likely stabilizes the intracellular concentration of tRNALeuZ(A25G), thus allowing a normal growth phenotype at 37°C. Expression of the hisD::Mud-lac reporter with the UCA-UAC codon pair decreased from 130 β-gal units in a leuZ+ control to 85 in the presence of the leuZ(A25G) pnp(I654T) alleles (Fig 6C). We then tested the effect of the leuZ(A25G) pnp(I654T) mutations on hisD::Mud-lac expression with leader peptides carrying either a CCC-CCC (Pro-Pro) or CGA-CGA (Arg-Arg). Contrary to the results obtained with tRNAThrU(C40A) (Fig 5), the presence of the leuZ(A25G) pnp(I654T) mutations had no effect on hisD::Mud-lac expression in both constructs. With CCC-CCC the levels of hisD::Mud-lac expression was 250 β-gal units in both the leuZ+ pnp+ and leuZ(A25G) pnp(I654T) strains. With CGA-CGA the levels of hisD::Mud-lac expression was 180 β-gal units in both leuZ+ pnp+ and leuZ(A25G) pnp(I654T) backgrounds (Fig 6B). These results suggest that the effect of leuZ(A25G) pnp(I654T) on translation through UCA-UAC codon pair was not due to a general defect in mRNA translation in the cell or an overall effect on the his attenuation mechanism as observed with the thrU(C40A) allele.

[Figure omitted. See PDF.]

Fig 6. The tRNALeuZ(A25G) substitution results in faster translation speeds through the his leader peptide region containing UCA (serine) codons.

(A) The secondary structure of tRNALeuZ showing known RNA base modifications. tRNALeuZ is an essential leucine tRNA. tRNALeuZ is the only tRNA that decodes UUA codons (Red). The leuZ(A25G) substitution is depicted by an arrow. (B) The effect of a pnp(I654T) allele on the growth rate of a strain with the tRNALeuZ(A25G) allele at 37°C. Growth rates were performed by diluting overnight cultures in fresh medium and measuring OD600 at 37°C until the OD600 was greater than 1. Wild-type Salmonella Typhimurium (Blue, TH437), leuZ(A25G) (Orange, TH27694), leuZ(A25G) pnp(I654T) (Gray, TH28496), and Pnp(I654T) (Yellow, TH28581) were all assayed for growth in LB medium at 37°C. The strain carrying the leuZ(A25G) allele exhibited a slower growth phenotype at 37°C in a pnp+ strain background. This slow-growth phenotype was suppressed by the pnp(I654T) allele. (C) β-Galactosidase assays were used to quantify the effect of tRNALeuZ(A25G) on hisD::Mud-lac expression. The leuZ(A25G) mutation suppressed the slow-translation phenotype of the his leader with His4-His5 replaced by UCA(Ser)-UAC(Tyr) (TH27505, TH27689), but did not suppress the slow-translation phenotype with His4-His5 replaced with either the CCC(Pro)-CCC(Pro) (TH17685, TH28497) of CGA(Arg)-CGA(Arg) (TH28084, TH28498) codon pairs. One-tailed heteroscedastic t-test showed that the WT and UCA(Ser)-UAC(Tyr) in leuZ+ and leuZ(A25G) backgrounds were statistically different (P = 0.01 and 0.001, respectively). The CCC(Pro)-CCC(Pro) and CGA(Arg)-CGA(Arg) results were not significantly different (D) The leuZ(A25G) mutation suppressed the stalled translation phenotypes of codon pairs containing the UCA codon in the his leader peptide system. In order to determine if the effect of tRNALeuZ(A25G) on the suppression of slow-translation was specific to translation through UCA codons, β-galactosidase levels were determined in strains containing the hisD::Mud-lac reporter and several His4-His5 leader peptide constructs where the UCA codon was present at either His4 or His5. The his leader alleles that were assayed were hisL10634(His4-His5 = UCA-UUU; TH28865 (leuZ+) and TH28503 (leuZ(A25G)), hisL10633(His4-His5 = UCA-UUC; TH28866 (leuZ+) and TH28502 (leuZ(A25G)), hisL10632(His4-His5 = UCA-AAU; TH28867 (leuZ+) and TH28501 (leuZ(A25G)), hisL10635(His4-His5 = UCA-GCC; TH28868 (leuZ+) and TH28504 (leuZ(A25G)), and hisL10636(His4-His5 = GGC-UCA; TH28869 (leuZ+) and TH28505 (leuZ(A25G)). Cells were grown at 37°C. β-Galactosidase activities are listed in S3 Table. One-tailed heteroscedastic t-test showed that all data sets were statistically different (P = <0.001).

https://doi.org/10.1371/journal.pgen.1011569.g006

The possibility that tRNALeuZ(A25G) was defective in sacB translation was also tested to confirm that the Lac- Suc+ phenotype of the leuZ(A25G) allele in our assay was not due to an overall defect in translation. There are 10 LeuZ-dependent UUA Leu codons within the coding region of sacB at positions 13, 90, 109, 177, 259, 277, 308, 384, 398 and 469. A defect in translating UUA codons by tRNALeuZ(A25G) could result in the Suc+ selection phenotype. We tested this hypothesis by replacing the 10 UUA codons in sacB with different leucine codons (CTN) that are not decoded by tRNALeuZ. Sucrose phenotypes were assayed on 6% sucrose plates at 37°C in either a leuZ+ or a leuZ(A25G) pnp(I654T) background with this modified sacB(CTN) gene. The effect of leuZ+ and leuZ(A25G) alleles on sacB expression (low expression = Suc+ versus high expression = Suc-) was determined in the strain background with the UCA-UAC (Ser-Tyr) codon pair at His4-His5 in the his leader peptide and the hisD coding sequence replaced with sacB containing the Leu codon substitutions (ΔhisD::sacB(CTN)-FCF). The leuZ+ background strain remained Suc-; the leuZ(A25G) pnp(I654T) mutant background remained Suc+. Had the leuZ(A25G) pnp(I654T) mutant background become Suc- we would have concluded that the Suc+ phenotype in our original selection was due to a defect in translating sacB and not faster translation through UCA-UAC in the leader sequence. We conclude that the leuZ(A25G) mutation does not result in a general decrease in translation of UUA codons in either the ΔhisD::sacB-FCF or ΔhisD::sacB(CTN)-FCF strain and that the presence of tRNALeuZ(A25G) results in faster translation through the UCA-UAC codon pair in the his leader sequence.

The effect of tRNALeuZ(A25G) is specific to UCA codons

The above results suggest that the leuZ(A25G) allele does not have a direct effect on his leader attenuation, and that it results in an increased translation speed through the UCA-UAC codon pair in our speedometer assay. Previously, we reported the construction of 128 codon pair combinations at His4-His5 with UCA(Ser) held constant at either His4 or His5 in the his leader peptide sequence [39]. We tested whether the leuZ(A25G) allele could affect the apparent translation speed through other slow-translated codon pairs with UCA in the His4 or His5 position. In our previous measurements, apparent translation rates through UCA-UUU(Phe), UCA-UUC(Phe), UCA-AAU(Asn), UCA-GCC(Gly), GGC(Ala)-UCA replacements resulted in stalled translation when compared to a strain with the WT his leader sequence [39]. Rates were quantified using β-gal assays for these codon pair combinations in the presence of leuZ(A25G) pnp(I654T). For each codon pair that included UCA, stalled translation was rescued in the presence of tRNALeuZ(A25G) (Fig 6D), while all pairs without a UCA were not affected by tRNALeuZ(A25G) (Fig 6C). We conclude that the tRNALeuZ(A25G) rate enhancement effect on slow-translating UCA-UAC is specific to the UCA codon.

tRNALeuZ(A25G) mediates leucine incorporation at UCA codons

If tRNALeuZ(A25G) is decoding UCA codons, it should cause misincorporation of leucine at serine codons. To test this hypothesis, we took advantage of the finding that a change in amino acid 201 of eGFP to serine (UCU) resulted in a nonfluorescent, unstable phenotype in yeast [50]. We found that a UCA serine substitution at amino acid 201 of eGFP had the same phenotype in Salmonella as the UCU substitution reported in yeast. The eGFP- and eGFP(L201S(UCA))-expressing genes were cloned into an IPTG inducible pTrc99a-10xHis-FLAG (ApR) vector plasmid and introduced into Salmonella strains to characterize the effect of tRNALeuZ(A25G) on translation of the eGFP(L201S) gene. Under inducing conditions, the strains expressing eGFP were fluorescent while those expressing eGFP(L201S), were not fluorescent. The eGFP(L201S) substitution resulted in reduced expression levels, which is consistent with reduced stability (Fig 7A). The eGFP(L201S) also exhibited a slightly slower gel migration rate than the wild-type protein (Fig 7A). The eGFP(L201S) protein was purified from leuZ+ and leuZ(A25G) Salmonella strains, and mass spectrometric sequencing of the peptide region that includes AA201, showed that there was only serine present (Fig 7B and 7C). We reasoned that the inability to detect leucine incorporation in eGFP(L201S) was because the misincorporation event might be rare due to failure of the mutant tRNA to compete successfully with the native tRNASerT that normally translates the UCA codon. To increase the chances of leucine incorporation, the experiment was repeated in a tRNASerT mutant strain that is defective in translating the UCA codon. tRNASerT is the only tRNA that decodes the UCA serine codon. A tRNASerT(G10A) substitution was previously isolated in E. coli as a temperature-sensitive lethal mutant and was found to be important in cell cycle regulation [51]. In Salmonella Typhimurium, the serT(G10A) allele was isolated as a temperature-sensitive lethal mutant defective in translation of the UCA codon at permissive temperatures in the flagellar regulatory gene, flgM [52]. The presence of the serT(G10A) allele should increase the frequency of misincorporation of leucine into UCA codons by tRNALeuZ(A25G). The presence of the serT(G10A) allele resulted in a significant reduction of eGFP(L201S) levels in both the leuZ+ and leuZ(A25G) backgrounds (Fig 7A). His-FLAG-tagged eGFP(L201S) was purified from leuZ+ and leuZ(A25G) Salmonella strains containing the serT(G10A) allele. Analysis of their relevant peptides revealed that misincorporation of leucine at the UCA 201 codon in eGFP(L201S) increased significantly to 20% in the serT(G10A) leuZ(A25G) background relative to the leuZ+ control (the latter remained at <1% leucine) (Fig 7D and E). This result clearly shows that tRNALeuZ(A25G) can incorporate leucine at a UCA serine codon.

[Figure omitted. See PDF.]

Fig 7. The effect of tRNALeuZ(A25G) on the incorporation of leucine at UCA codon 201 in eGFP(L201S).

eGFP(L201S) was expressed and purified from leuZ+ serT(G10A), leuZ(A25G) serT(G10A), leuZ(A25G) serT+, and leuZ+ serT+ strain backgrounds. (A) eGFP expression patterns (western blots using anti-GFP antibody) without and with IPTG-induction conditions in different strain backgrounds indicated. (B–E) Isolated eGFP protein was digested with trypsin and peptides were analyzed by LC-MS/MS. (B–C) Only serine was present at amino acid position 201 of eGFP(L201S) in the wild-type (leuZ+ (B) and leuZ(A25G) (C)) mutant backgrounds. (D) There were two peptides isolated from the eGFP(L201S)/ serT(G10A) strain (TH28590). The major peak had serine incorporated at AA201, while the minor peak had leucine incorporated at AA201. The leucine peak is 1% relative to the serine peak. The two peptides containing Serine or Leucine were eluted at time 73.66 and 80.79 respectively. (E) There were two peptides isolated from eGFP(L201S)/ serT(G10A) leuZ(A25G) (TH28592). The major peak had serine incorporated at AA201, while the minor peak had leucine incorporated at AA201. The leucine peak is 20% relative to the serine peak. The two peptides containing Serine or Leucine were eluted at time 66.03 and 71.11 respectively.

https://doi.org/10.1371/journal.pgen.1011569.g007

Stalled translation results in eGFP (L201S) fluorescence in the presence of tRNALeuZ(A25G)

Even though the mass spectrometry data demonstrated that tRNALeuZ(A25G) can incorporate leucine at UCA codon 201 of eGFP (L201S) in the above construct, not enough was produced to allow us to detect fluorescence. Given the strong effect of the tRNALeuZ(A25G) substitution on translation of the UCA codon in the his leader peptide system, we were surprised at the difficulty in detecting misincorporation of leucine in place of serine for eGFP(L201S) and the fact that the serT(G10A) allele is required to facilitate misincorporation. We have noted that the his leader peptide system is extremely sensitive to translation defects. For example, successive proline codons at His4-His5 in the his leader peptide gives the same loss of attenuation as a stop codon at either one of those positions 34. In other systems proline pauses are only detected with four consecutive prolines with an additional polyproline-specific elongation factor, EF-P, removed [39,53].

The his leader peptide system is unusual in that we also detect effects of codon changes on translation rates when the ribosome is after RNA polymerase during transcription-translation coupling. In the eGFP(L201S) suppression assay described above the majority of ribosomes translating the UCA codon at position 201 would not be the first ribosome that is coupled to RNA polymerase. Recent work from the Bustamante lab supports the assertion that the RNA polymerase coupled ribosome is unique [54]. In an in vitro transcription-translation coupled system, they found that the coupled, initial ribosome exerts mechanical force on the RNAP, effectively pushing it from behind, which resulted in an increased transcription speed at the expense of fidelity. The initial ribosome is unique in that it is physically connected to RNA polymerase through NusG protein [55]. Ribosome stalling in the his leader peptide’s consecutive His codons should result in a physical separation between the first ribosome and RNA polymerase and allow RNA polymerase to continue on into the attenuator stem-loop region. We hypothesized that failure to detect fluorescence of eGFP(L201S) with tRNALeuZ(A25G) and a much higher frequency of misincorporation of leucine at codon 201 of eGFP(L201S) was because we were not accounting for separation of RNA polymerase from the first ribosome as in the his leader peptide. It is possible that stalling of the first ribosome at UCA codon 201 in eGFP(L201S) might be mimicked by the presence of translation-elongation inhibitors, which might allow tRNALeuZ(A25G) time to be recognized by translating ribosomes not associated with RNA polymerase. Thus, slowing of translating ribosomes would phenocopy the effect of directly following RNA polymerase and provide time to recognize UCA by tRNALeuZ(A25G) bound to EF-Tu.

There are numerous antibiotics that target the ribosome at different steps during translation [56]. We tested one translation-initiation inhibitor (kasugamycin) and three translation-elongation inhibitors (tetracycline, chloramphenicol and spectinomycin) on the ability of tRNALeuZ(A25G) to misincorporate leucine into codon UCA 201 in the eGFP(L201S) to produce detectable fluorescence. Intermediate concentrations of elongation inhibitors should slow ribosome progression through the cell. Exponentially growing cells were applied to a microscope slide with an agarose pad and antibiotic was applied to one end of the agarose covered slide. This allowed the antibiotic to form a concentration gradient by diffusion through the agar pad. A zone of cell growth inhibition was observed close to where the antibiotic was applied, but growth was observed at the lower subinhibitory antibiotic levels and throughout the rest of the slide. With each elongation inhibitor, but not with the initiation inhibitor, we observed a zone containing fluorescent cells (Fig 8). We suspect that the concentration of antibiotic in this zone was sufficient to cause an effect on translation of codons that was similar to the effect of separation of the leading ribosome from RNA polymerase resulting in high enough levels of mis-incorporation of leucine at UCA codon 201 of eGFP(L201S) by tRNALeuZ(A25G).

[Figure omitted. See PDF.]

Fig 8. Stalling of translation-elongation resulted in increased fluorescence for strains expressing eGFP(L201S) in the leuZ(A25G) background.

Incorporation of leucine at codon 201 in eGFP(L201S), resulted in a non-fluorescent phenotype. On a microscope slide, eGFP(L201S)/ leuZ(A25G) pnp(I654T) (TH29056) and eGFP(L201S)/ leuZ+ pnp(I654T) (TH29056 or TH29134) were plated onto an agarose pad. Antibiotics that inhibit either translation elongation (tetracycline, chloramphenicol, and spectinomycin) or translation initiation (kasugamycin) were spotted onto the far edge of the agarose slide. Images were taken every 5-10 steps starting at the edge of zone of inhibition, where each step is 0.89μm. Each assay was done in triplicate. Average eGFP fluorescence was quantified per cell for each replicate. The average between each replicate was calculated and plotted on the graph (solid line). The effect of each antibiotic on fluorescence was determined for both tRNALeuZ (Blue; ●,▲,◆) and tRNALeuZ(A25G) (Orange; ●,▲,◆) expressing strains.

https://doi.org/10.1371/journal.pgen.1011569.g008

The results presented in Fig 8 showed a significant increase in fluorescent eGFP(L201S)-expressing cells in the leuZ(A25G) background as compared to leuZ+ for translation-elongation inhibitors tetracycline, chloramphenicol and spectinomycin. The effect of spectinomycin was seen close to the edge of the zone of inhibition while the effects of tetracycline and chloramphenicol were further from the edge. This may be due to differences in sensitivity of Salmonella to the different antibiotics. The translation-initiation inhibitor kasugamycin had no effect on leucine mis-incorporation (Fig 8).

Discussion

We showed that an A25G substitution in the D-stem of the essential leucine tRNALeuZ is able to suppress the slow translation speed through tandem UCA-UAC codons at His4-His5 in the his operon leader sequence. The leuZ(A25G) allele is also able to increase the apparent translation speed through different codon pairs at His4-His5 in the his leader peptide system as long as one of the codons is UCA, but it does not affect the apparent translation speed through codon pairs that did not include a UCA codon. The his leader peptide does not contain leucine codons leading us to hypothesize that tRNALeuZ(A25G) is able to recognize UCA in the ribosome decoding center. Consistent with this hypothesis, we showed that leucine can be incorporated into eGFP with UCA at codon 201 in a strain expressing tRNALeuZ(A25G). This result suggests that the mutant tRNALeuZ(A25G) has an increased decoding capacity beyond the UUA and UUG codons recognized by wild-type tRNALeuZ.

How might the A25G substitution in the D-stem of tRNALeuZ lead to extended decoding capacity and increased translation speed through UCA codons? tRNA species are often presented in two dimensions as “cloverleaf” 2° structures depicting the acceptor stem, the D-stem, and loop, the anticodon stem and loop, and the TΨC-stem and loop (Fig 9A). A tRNA’s three-dimensional structure forms an inverted L-shape (Fig 9B). The D-loop and TΨC-loop interact at the elbow region of the tRNA 3° structure. During tRNA recognition and prior to its acceptance as the correct tRNA for a specific codon, the ribosome undergoes a conformational change that leads to a bending of the tRNA at the elbow (Fig 9C). The bending of the cognate tRNA is coupled to hydrolysis of EF-Tu bound GTP, which is followed by accommodation of the correct aminoacyl-tRNA or rejection of an incorrect one [18,27,28]. This GTP hydrolysis step in A-site recognition provides a final fidelity check to ensure that the correct tRNA is chosen for that specific codon. If a non-cognate tRNA is present in the A-site during the GTP hydrolysis step, the tRNA is rejected due to the mismatch in the codon-anticodon interaction [10,18,21]. Near cognate tRNAs that differ by only one base mismatch in the anticodon-codon interaction can remain in the A-site longer than a non-cognate interaction until the GTP hydrolysis step in codon recognition. In order to discriminate near-cognate from cognate tRNAs, the ribosome relies on the final recognition step that includes the bending of the tRNA in the D+TΨC-stem interacting (hinge) region near the elbow of the tRNA structure [28,57] (Fig 9).

[Figure omitted. See PDF.]

Fig 9. Model depicting straight versus bent tRNA structures.

(A) A classical color-coded “cloverleaf” tRNA 2° structure is shown depicting the acceptor stem (red), the D-stem and loop (blue), the anticodon stem and loop (yellow), the variable loop (green) and the TΨC-stem and loop (purple). (B) A three-dimensional color-coded tRNA inverted L-shaped tRNA structure. C. Straight and bent three-dimensional tRNA structures presented at a 90° angle from the structure shown in B.

https://doi.org/10.1371/journal.pgen.1011569.g009

One possible way to suppress the need for a complementary anticodon/ codon interaction is to have a pre-bent or more flexible tRNA [13]. If a near-cognate tRNA that is more flexible enters the A-site, the ability of the ribosome to hydrolyze GTP and accommodate the tRNA is faster than its ability to sense an incorrect near-cognate tRNA in the A-site [58] (Fig 10). There exists a single example of this in the literature: the “Hirsh suppressor” in E. coli tRNATrpT. In 1971 Hirsh isolated a G24A D-stem substitution in tRNATrpT as a UGA nonsense suppressor that allowed recognition of both the UGG(Trp) and the UGA nonsense codons [15,59]. It was later shown that translation with tRNATrpT(G24A) exhibits an accelerated forward reaction rate and structural data revealed that tRNATrpT(G24A) has a stable bent conformation [15,58]. The genetic data presented here predicts that tRNALeuZ(A25G) results in a structural change in the tRNA hinge region similar to that in tRNATrpT(G24A) that allows recognition of the near-cognate UCA(Ser) codon resulting in an increase in translation speed in the his translation speedometer assay. Our findings suggest that this may be a general phenomenon and that base substitution mutants in the hinge region of other tRNA species might also increase translation speed through near-cognate codons and result in amino acid misincorporation.

[Figure omitted. See PDF.]

Fig 10. Model for near-cognate codon recognition by pre-bent/flexible tRNAs.

(A) During translation, EF-Tu bound to GTP and aminoacylated-tRNA enters the A- site of the decoding center straight. As the codon- anticodon interaction is being assessed, the ribosome changes conformation resulting in a bent tRNA structure. After recognition of cognate tRNA, GTP is hydrolyzed to GDP and EF-Tu dissociates from the ribosome, which allows for accommodation of the tRNA into the A-site fully. (B) A mutation in the hinge region of a tRNA that results in a bent tRNA structure bypasses the final fidelity check in codon recognition. When the mutant, bent tRNA enters the ribosome A-site, GTP is hydrolyzed faster than the ribosome allowing the tRNA to insert an incorrect amino acid with a near-cognate codon. Thus, evolution has provided a mechanism to increase fidelity for codon recognition, but at a cost of translation speed. Structures adapted from [13,14]. Structures were taken from the Protein Data Bank. More specifically, PDB files: 4V68, 4V6F, 4V5S, and 4V5R. tRNA selection and tRNA recognition from [14], tRNA accommodation from structures in [70], and bent tRNA and straight tRNA came from structures in [15].

https://doi.org/10.1371/journal.pgen.1011569.g010

Translation is fine tuned to balance speed and accuracy. Amino acid misincorporation during translation in E. coli has been shown to occur at a frequency of ~10-3-10-4 [60,61]. Hyper-accurate ribosomal mutants, initially isolated in the 1960s, were found to result in a reduction of growth rates due to a reduction in translation speed that was concomitant with increased translational accuracy [62,63]. The translation machinery has evolved to balance the relative costs of translation speed and fidelity. The study presented here demonstrates that it is possible to select tRNA mutants that change this balance and allow for faster translation speed at the cost of fidelity.

One striking result obtained in this study was that only the thrU(C40A) mutant was obtained that was defective in overall mRNA translation out of ~1010 cells were obtained. The UCA-UAC codon pair used for selection has a translation-slow phenotype similar to a stop codon in our his leader peptide system (a “g” Mac-Lac and sucrose-sensitive phenotype in Fig 4B) [39]. It is possible that selection for faster translation using codon pairs with less stringent selection phenotypes would allow for the isolation of overall translation-defective mutants. Otherwise, the leuZ(A25G) allele would likely not have been obtained by our genetic selection method. It is noteworthy that the leuZ(A25G) was isolated together with the pnp(I654T) allele. Transduction of the leuZ(A25G) into a pnp+ background resulted in poor growth at 37°C and no growth at 42°C. It is possible that inclusion of a pnp-defective allele into the original selection strain would have allowed for the isolation of other general mRNA translation tRNA mutants besides the thrU(C40A) allele. Another important result was the requirement of the serT(G10A) allele with the leuZ(A25G) to detect leucine incorporation at UCA codon 201 of eGFP. We hypothesize that this represents the difference in mRNA translation in the his leader peptide system as compared to translation of eGFP mRNA. Translation in the his leader peptide system only occurs where the translating ribosome is coupled to transcription by RNA polymerase. For eGFP gene translation, only the initial translating ribosome is coupled to transcription by RNA polymerase. All but the initial translating ribosomes are not directly behind RNA polymerase during translation of eGFP mRNA. A recent study reported that in bacteria where transcription and translation are coupled, the initial ribosome trailing RNA polymerase speeds up RNA polymerase at the expense of transcript fidelity via force and allostery [54]. If the reciprocal is true (RNA polymerase impedes trailing ribosomes), the speed of the translating ribosome in the his speedometer assay may be slowed by RNA polymerase and may explain the apparent amplification effect of tRNALeuZ(A25G) on the speed through UCA codons and the misincorporation of leucine relative to translation effects on ribosomes that follow the initial translating ribosome.

Provided that the ribosome can move faster through codons than RNA polymerase can transcribe DNA, it would be advantageous to have a mechanism to simultaneously enhance fidelity and reduce translation speed that would prevent RNA polymerase from incorporating incorrect bases and reduce the frequency of ribosome collisions during translation. We propose that the final fidelity check coupled to GTP hydrolysis at EF-Tu provides such a mechanism.

In addition, data presented in Fig 8 data showed that even in the leuZ+ background significant fluorescence was observed in sublethal concentrations of tetracycline and chloramphenicol. This result raises the possibility of mechanisms where, under conditions of stress, a cell could express a translation-elongation inhibitor to a level that would allow a high degree of misincorporation resulting in transient production of a plethora of altered proteins without mutation at the DNA level. Such a mechanism might provide an evolutionary survival advantage to the organism.

Materials and Methods

Bacterial strains, plasmids, and media

All strains used in this study not designated as E. coli were derived from Salmonella enterica serovar Typhimurium wild-type strain LT2 and are listed in S1 Table. The list of DNA oligonucleotides used for strain constructions are listed in S2 Table. Cells were cultured in lysis broth (LB: Bacto-tryptone 10g/l, Bacto-yeast extract 5g/l, NaCl 5 g/l). Antibiotics were added to LB at the following final concentrations: 100 µg/ml sodium ampicillin (Ap), 12.5 µg/ml chloramphenicol (Cm), 15 µg/ml tetracycline-HCl (Tc), 50µg/ml kanamycin sulfate (Km), 100 µg/ml carbenicillin. Apex agar (12g/l) was used for preparation of solid medium. Selection of tetracycline-sensitive (TcS) recombinants was done on zinc-fusaric acid selection media [64]. LB-arabinose was supplemented to 0.2% (w/v) as needed. Sucrose plates used the same recipe as above with a couple of minor changes, 60g/l of LB-sucrose and no NaCl added. Minimal E salts medium supplemented with 0.2% glucose was used as a minimal medium. supplements were added to minimal medium to a final concentration of the following compounds: 2 µg/ml X-Gal and 1.55 µg/ml histidine. Phenotypic lactose activity was observed using MacConkey lactose indicator medium (Mac-Lac), and 3,4,5-triphenyltetrazolium chloride-lactose (TTC-Lac) indicator medium supplemented with 0.2% arabinose if required.

Transduction and λ-red recombination for strain construction

The generalized transducing phage of S. Typhimurium P22 HT105/1 int-201 was using in all transductional crosses [65]. P22 phage lysates were prepared on donor strains. Briefly, 100ul of a P22 transducing lysate, diluted to 10-6 to 10-7 plaque forming units (PFUs) per ml, was mixed with 100ul of an overnight culture (2 x 108 cells/ml) of the recipient strains. The mixture was allowed to incubate at either 37°C for 30 minutes or 30°C for 30 minutes for temperature sensitive mutants. The mixture was then plated onto selective media. The MudJ and MudA lac operon fusion elements were mobilized for strain construction by P22-mediated transduction selecting for kanamycin (KmR) or ampicillin resistance (ApR) respectively. λ-red recombination was used for allele construction using insertion and replacement of the tetRA cassette from transposon Tn10 following the protocol provided in [64]. Replacement of a tetRA cassette either by λ-red recombination or P22-mediated transduction was done by selecting for fusaric acid resistance as described [64]. To avoid background spontaneous tetracycline sensitive (TcS) mutants in P22 transductional crosses to fusaric acid resistance, the P22 transducing lysates are diluted to ~108 pfu/ml, let sit at either room temperature for one hour or incubated at 37°C for 30 minutes, and diluted 1000-fold prior to selection. This typically yields 100- to 1000-fold more TcS colonies than a no phage control cross. TcS colonies are then purified on green indicator plates [65] and genetic markers are verified by PCR or DNA sequence analysis. Gene replacements with the chloramphenicol acetyl transferase (cat) chloramphenicol resistance (CmR) cassette from plasmid pKD3 [66] were also done following the protocol provided in [64]. Insertions of the FRT-CmR-FRT (FCF) cassette from plasmid pKD3 [66] were also made following the protocol provided in [64] and used as linked selectable markers for strain construction purposes. The ΔSTM1941::tetRA deletion/insertion and the ΔSTM1941::cat deletion/insertion were constructed in a leuZ(A25G) background and used to move the leuZ(A25G) allele into various strain backgrounds by P22-mediated transduction selecting for tetracycline (TcR) or chloramphenicol resistance (CmR), respectively. The argG::Tn10dCm allele was used as a linked marker to move pnp alleles various strain backgrounds by P22-mediated transduction selecting for CmR. The hpaB::Tn10dTc allele used as a linked marker to move serT alleles various strain backgrounds by P22-mediated transduction selecting for TcR. The ΔSTM4140::tetRA deletion/insertion was constructed in a thrU(C40A) background and used to move the leuZ(A25G) allele into various strain backgrounds by P22-mediated transduction selecting for TcR. If the plasmid pSim-5 or pSim-6 was used for λ-red recombination, the only alterations to the protocol were as follows. Cells were grown in LB-Cm (pSim-5) or LB-Ap (pSim-6) at 30°C until OD600= 0.2. The culture was placed on ice for 5 minutes, and then heat shocked at 42°C for 15 minutes to induce expression of the λ-red genes.

Randomization of the His2 and His3 codons upstream of UCA-UAC at His4-His5

A DNA fragment containing NNN His3 upstream of His4-His5 was generated using 4 cycles fill-in reaction between primers 8125 (his345-NNN-Ser-tyr) and 8000 (hisleader-rv), followed by 9 cycle amplification of the fill-in DNA fragment using primers 7999 (hisleader-fw) and 8000 (hisleader-rv). The DNA fragment containing NNN His2 upstream of His4-His5 was produced similarly using primers 8484 (His2NNNSerTyr)/8000 (hisleader-rv)/ 8484 (His2NNNSerTyr). Phusion DNA polymerase was used for all fill-in and DNA amplification reactions. The resulting DNA fragments were purified by ethanol precipitation and electroporated into electrocompetant strain TH15671 (pKD46/hisO10508::tetRA hisD9953::MudJ), previously grown with arabinose at 30°C. Following electroporation, the cells were resupended into 1mL LB media and incubated at room temperature for 1h, before being plated onto TcS plates at 37°C. TcS recombinants were patched onto TcS and MacConkey-Lactose plates (to assess expression of hisD9953::MudJ). The Lac- phenotype recombinants were purified and pooled together. The Lac- culture was diluted to approximate an overnight OD and genomic DNA was extracted using the Qiagen DNeasy Blood and Tissue Kit. A 20 cycle PCR reaction was performed using the genomic DNA template using primers 8256/8257 and Phusion polymerase. The purified PCR was submitted for amplicon sequencing (GeneWiz) to identify at once the identity of the Lac- phenotype recombinants. Lac+ phenotype recombinants were purified and sequenced individually.

Randomization of amino acid 2 (Thr) of the his leader peptide

Amino acid 2 of the his leader peptide was randomized using a similar strategy as described above. A tetRA element was first inserted at amino acid 2 of the his leader in strain TH27505 (hisL10592(His4-5 = TCA-TAC) hisD9953::MudJ) by λ-red recombination (with primers 10331/10858). A DNA fragment containing NNN instead of the amino acid 2 of the his leader was produced using a fill-in reaction between primers 9827 and 10859, followed by 9-cycles amplification of the fill-in reaction using primers 10864 and 8094. The DNA product was electroporated into the strain constructed above (containing the tetRA element at amino acid 2 and the λ-red plasmid pSim6). TcS recombinants were streaked directly onto Mac-Lac plates and the Lac+ colonies similar color phenotypes to the parent containing His4-5 = TCA-TAC (TH27505; red on Mac-Lac) were further purified and sent to DNA sequencing analysis.

Mud-generated duplication of his operon

The duplication of the his operon was made using a strategy described in [67]. P22 phage lysates was grown on hisD9953::MudA (TH653) and on fliN5103::MudB (TH2094). Equal volumes of each phage lysates were incubated with an LT2 (TH437) overnight culture at 37°C for 30 minutes. Mud-containing transductants were selecting for on LB-Ap medium and ApR transductants were screened for prototrophic growth on minimal-glucose (E-dex) medium. ApR E-dex+ colonies were single colony purified and then streaked on minimal-Xgal medium and screened for blue-white segregation. Blue-white color segregation in the absence of ApR selection indicated that the region of the chromosome between the sites of insertion of the hisD9953::MudA and fliN5103::MudB elements had been duplicated yielding strain TH27503. When selection is not held for the duplication, the duplication is spontaneously lost resulting in a mixture of blue and white single colonies on minimal medium with X-gal [67]. The duplication from TH27503, was moved by P22-mediated transduction into Δhis-3050 (TH85) selecting for ApR, resulting in strain TH27504. The Δhis-3050 allele is a non-transducible deletion of the his operon. Within the duplication of TH27504, the his leader construct was changed with His4-His5= UCA-UAC (TH27505) via P22-mediated transduction by selecting for KmR and screening for blue-white color segregation E-dex-Xgal medium lacking Ap resulting in strain TH27506. The duplication in TH27506 was moved by P22-mediated transduction into strain TH27507 (hisL10592 ΔhisD10591::sacB-FCF) selecting for KmR and screening for blue-white color segregation E-dex-Xgal medium lacking Ap to produce strain TH27508.

Selection for fast-translation mutants

Strain TH27508 has the His4-His5 codons of the his leader peptide region replaced with codons UCA-UAC in each duplicated copy. One duplicated region carries a hisD9953::MudJ allele, which results in Lac+ phenotype on Mac-Lac medium. The other duplicated region of the his operon contains the ΔhisD10591::sacB-FCF allele, which results in a Suc- phenotype on 6% sucrose medium. Ten independent cultures of TH27508 were grown overnight in LB-Km at 37°C. A 0.1ml portion from each overnight culture was plated onto 6% sucrose-Km media and incubated overnight at 37°C. The following day each 6% sucrose-Km plate yielded ~1500 Suc+ colonies. Suc+ colonies were replica printed to Mac-Lac-Km medium and screened for those with a Lac- phenotype. Suc+ Lac- colonies were purified on Km-containing media and then either mapped or subject to next generation sequencing.

Next generation sequencing

Genomic DNA from overnight cultures was isolated using the Qiagen DNA Blood and Tissue Kit. Next Generation Sequencing analysis was performed either using Nanopore Sequencing (Plasmidsaurus) or Illumina (150x150 bp Novaseq Sequencing) using Ilumina DNA prep for small genomes at the University of Utah Core Facility. Sequences were aligned to Salmonella Typhimurium LT2 genomic sequence using the program Geneious. SNP alleles were identified in strains TH27509 and TH27511 as thrU(C40A) and rpoA(E261K), respectively. Strains TH27509 and TH27511 were subject to NexGen sequencing and found to harbor the thrU(C40A) and rpoA(E261K) alleles, respectively. For the four independent larger-colony, Lac+ mutants from strain TH27698 (hisL10592(His4-5 = TCA-TAC) hisD993::MudJ thrU(C40A)), one of the four genomes (suppressor 1) was subject to Nanopore Sequencing (Plasmidsaurus) and the other three were sequenced using Illumina Sequencing (University of Utah Core Facility) for comparative purposes (S1 Fig).

Isolation of Tn10dTc insertions near the chromosomal rpoD and leuZ loci

Mutations in TH27510 and TH27512 were mapped using a linked Tn10dTc transposon. P22 was grown on strain TT10423, which carries the Tn10dTc insertions in the F’ plasmid and used to transduce strain TT10427 to Tn10dTc-encoded TcR on LB-Tc medium. Strain TT10427 carries plasmid pNK972, which constitutively expresses the Tn10 transposase gene. Because there is no F’ plasmid DNA in strain TT10427, it is not possible to inherit the Tn10dTc transposons by homologous recombination. The TcR transductants are inherited by transposition into the recipient chromosome. More than 50,000 TcR transductants were pooled together, and P22 lysates were prepared on the pooled cells that carried insertions located at random positions through the chromosome. To isolate linked Tn10dTc insertions to alleles later shown to be in the rpoD and leuZ genes, the pooled Tn10dTc lysate was used to transduce strains TH27510 and TH27512, respectively, to TcR by plating on LB-Tc plates. Following overnight incubation at 37°C, the TcR colonies were screened for Lac+ on Mac-Lac medium. Co-transductional analysis of several linked insertions identified one Tn10dTc allele in the STM3216 locus that was 68% linked by P22-mediated transduction to the Lac phenotype associated with the rpoD mutant allele. Three Tn10dTc insertions were found to be in the STM1939, STM1941, and uvrC genes and were 84%, 94% and 84% linked, respectively, to the Lac phenotype associated with the leuZ mutant allele. PCR amplification of the strain carrying the closely linked insertions with a primer that would recognize the ends to Tn10dTc gave a product which after DNA sequence analysis revealed the Tn10dTc insertion to be in the loci mentioned above.

β-Galactosidase assays

Thirty µl of overnight cultures were each sub-cultured into 3 ml of fresh LB medium. Sub-cultures were incubated under aeration conditions at 37°C and grown to mid-log cell growth density (OD600 = 0.4). For strains that carry the thrU(C40A) allele, cells were incubated with aeration at 30°C and grown to mid-log (OD600 = 0.4). Following incubation on ice, cultures were pelleted by centrifugation, and cell pellets were resuspended in 3 ml of iced cold buffered saline. 0.5 ml culture samples (diluted if necessary) were added to 0.55 ml of Z-buffer (8.54 mM Na2HPO4, 5.5mM NaH2PO4•H2O, 0.75mM KCl, 0.25mM MgSO4•7H2O, pH 7.0) plus 5 ul of 10% sodium dodecyl sulfate and 100μl of chloroform. 0.2 ml of ONPG (4mg/ml) was added to each tube periodically. Reactions were stopped after 10 mins with the addition of 0.5 ml of 1M NaCO3. OD at wavelengths 420 and 550 were measured. Cell cultures were warmed to 30°C and their OD650 was measured. For each strain, the β-galactosidase activities reported are the average from at least three independent cell cultures started from different single colony isolates. β-galactosidase activity was expressed in nmol/min/OD650 and determined as described [68]. The one-tailed heteroscedastict-test was used for the statistical analysis shown in Figs 5 and 6.

Identification of a secondary pnp(I654T) mutation in the leuZ(A25G) mutant strain

The leuZ(A25G) allele linked to a ΔSTM1941::tetRA (TcR) marker was introduced via P22-mediated transduction into strain TH27505 (hisL10592(His4-His5=TCA-TAC) hisD9953::MudJ) selecting TcR. TcR transductants were sequenced for the leuZ region and one containing the leuZ(A25G) allele was kept as strain TH27691. Strain TH27691 had a noticeable growth defect at 37°C, indicating that a second mutation was present in the original background that suppressed the temperature-sensitive growth phenotype. A 2ml overnight culture of the parent strain TH28496 was prepared, genomic DNA was extracted using the kit and protocol from Qiagen DNeasy Blood and Tissue kit. Genomic DNA was then sent to the University of Utah’s core facilities for whole genome sequencing. The sequencing results were then aligned to the LT2 genome using the program Geneious. A SNP in the polynucleotide phosphorylase gene was identified as the pnp(I654T) allele, which was shown to be responsible for suppressing the temperature-sensitive growth defect caused by the LeuZ(A25G) substitution.

The effect of pnp(I654T) on strain growth phenotypes

Growth curves were determined for the strains carrying the following mutant alleles: none (wildtype strain LT2), leuZ(A25G), leuZ(A25G) pnp(I654T), and pnp(I654T). Three independent cultures of strains LT2, TH27694 (ΔSTM1941::tetRA leuZ(A25G)), TH28496 (leuZ(A25G) pnp(I654T)), and TH28581 argG::Tn10dCm pnp(I654T)), were grown overnight at 30°C. The overnight cultures were sub-cultured into 3 ml of fresh LB medium and incubated under aerobic growth conditions at 37°C. The OD600 for each culture was measured every 30 min until cultures reached OD 0.4 and every 15 minutes after that until the OD600 measurement reached >1.0. Measurements obtained for each replicate were averaged and plotted using Excel.

TTA to CTN codon replacement in SacB

The 10 UUA leucine codons (at positions 13, 90, 109, 177, 259, 277, 308, 384, 398 and 469) were replaced by other leucine codons (CTN) using λ-Red recombination in two steps. First, codons 13 through 177 of sacB were replaced by a tetRA element using primers 9496 and 9497 into a strain containing the sacB gene integrated into the chromosome of Salmonella (ΔhisD10591::sacB-FCF). Two PCR amplifications of sacB DNA with primer pairs 9479/9480 and 9481/9482 were produced. The resulting DNA fragments have a 24 bp overlap and were mixed together without primers followed by a 10-cycle PCR reaction, allowing the fragments to stitch together. A final 15-cycle PCR reaction on the stitched product, using external primers 9609 and 9610 was used for the λ-Red substrate product to replace the sacB::tetRA (ΔAA13-77) insertion sequence. The resulting strain had UUA leucine codons at positions 13, 90, 109 and 177 of sacB changed to CTC, CTG, CTG and CTG, respectively.

The second step consisted of inserting a tetRA element (made with primers 9483/9484) deleting amino acid codons 259 through 469 of sacB in the strain constructed above. First, Leu UUA codons at positions 259 and 277 of sacB were changed to CTG and CTC using primers 9485/9486 and sacB DNA as template. This PCR reaction was used as a template for another PCR reaction using the external primer pair 9487/9488. Second, changes of Leu UUA codons at positions 308 and 384 of SacB to CTA and CTC were generated by PCR using the primer pair 9489/9490 and sacB DNA as the template. Third, changes of Leu UUA codons at positions 398 and 469 of sacB to CTC and CTC were generated by PCR reaction using the primer pair 9491/9492 and sacB DNA as the template. The resulting three PCR products above were mixed in equimolar ratio, stitched together, and amplified (15 cycles reaction) with external primers 8487 and 9519, yielding a 711 bp DNA fragment encoding changes of Leu UUA codons at positions 259, 277, 308, 384, 398 and 469. This DNA fragment was used to replace the tetRA element deleting codons 259 through 469 of sacB.

e GFP plasmid construction

Chromosomal insertion of the eGFP and eGFP(L201S) coding sequences were previously constructed at the hisD locus of S. Typhimurium by λ-red recombination. These strains were available in the lab’s strain collection. Primers 10149/10150 were used to PCR amplify the eGFP DNA region from strain TH28176 and the eGFP(L201S) DNA region from strain TH28397. Both PCR DNA products were purified using the Qiagen QIAquick PCR purification kit following the provided protocol. Strain TH27367 contains a ptrc99A-based plasmid derived from plasmid pMM1702 [69] that was modified to express fusion proteins containing both 10xHis- and FLAG-tags at their N-termini. Plasmid DNA was extracted from strain TH27367 using Zymo Research’s Plasmid Miniprep-Classic Kit following the provided protocol. The extracted plasmid DNA and purified PCR-generated eGFP DNA samples were digested with restriction enzymes NdeI and BamHI. Digested PCR-generated and plasmid DNA samples were then ligated together using T4 DNA ligase. The ligation reactions were electroporated into strain Ecoli XL1. The corresponding strains containing plasmids ptrc-99A-10His-FLAG-eGFP and ptrc-99A-10His-FLAG-eGFP(L201S) were isolated following ApR selection. The plasmid DNA sequences were verified, and purified plasmid DNA was introduced into Salmonella strains by electroporation.

eGFP protein purification

The eGFP and eGFP(L201S)/ proteins were expressed from strain TH28567 (ptrc-His-Flag-eGFP/LT2) and strain TH28592 (ptrc-His-Flag-eGFP(L201S)/hpaB::Tn10dTc fla-5398 ΔSTM1941::cat leuZ(A25G)). One liter of cells was grown at 37°C in LB supplemented with carbenicillin. At an OD600 of 0.6, IPTG was added to the growing cell culture at 0.25 mM. Cells were grown overnight and harvested the next day. Cells were centrifuged and the cell pellet was resuspended into 16 ml of lysis buffer (50 mM NaH2PO4; 300 mM NaCl; 10 mM imidazole). Lysozyme was added at a concentration of 1 mg/ml and the cells were sonicated (3 x 10 sec bursts with 15 sec cooling). The lysate was then centrifuged at 10,000 g for 30 mins at 4°C to pellet cellular debris. A 1ml portion of Ni-NTA resin (Qiagen) was added to the supernatant and the mixture was gently mixed at 4°C for 1 to 2 hrs. The slurry was then transferred onto a gravity column and the resin was washed with 4 ml of Wash Buffer (50 mM NaH2PO4; 300 mM NaCl; 20 mM imidazole). Bound protein was eluted in 0.5ml fractions of Elution buffer (50 mM NaH2PO4; 300 mM NaCl; 250 mM imidazole). Each fraction was analyzed by SDS-PAGE and the most concentrated eGFP fraction was sent to University of Arizona’s Analytical and Biological Mass Spectrometry Core for analysis.

Intact protein analysis

LC-MS analyses of intact protein samples were analyzed by LC-MS using a Vanquish Horizon UHPLC Duo system (Thermo Scientific, San Jose, CA) coupled to an Orbitrap Exploris 480 mass spectrometer (Thermo Scientific, Bremen, Germany) fitted with an H-ESI source. Proteins were desalted and separated on a BioResolve RP mAb Polyphenyl Column (450Å, 2.7 µm, 2.1 mm X 50 mm, Waters Corp.) heated at 50°C with a gradient of 0.1% FA in ACN at 300 µl/min: from 10 to 90% in 15 min. For all experiments spray voltage was set to 3.8 kV, the sheath gas setting was 50, the auxiliary gas setting was 10, the sweep Gas 1 vaporizer temperature was 225°C, and the ion transfer tube temperature was 325°C. MS spectra were acquired at 7,500 resolving power (at m/z 200) with a scan range set to m/z 500- 4,000 Da, 10 microscans per MS scan, a normalized automatic gain control (AGC) target value of 300%, and a maximum injection time mode set to 200 ms. RF value was set to 60%. Source fragmentation energy on the Orbitrap Exploris 480 was 15V. MS spectra were deconvolved with BioPharma Finder 5.0 software.

Nano-LC-MS/MS method and data searching

Nano-LC-MS/MS analysis was performed on a Q Exactive Plus mass spectrometer (Thermo Fisher Scientific, San Jose, CA) equipped with an EASY-Spray nanoESI source. The protein sample were digested with trypsin using micro spin columns using the manufacturer provided protocol. The resulted peptides (300 ng) were eluted from an Acclaim Pepmap 100 trap column (75 micron ID x 2 cm, Thermo Scientific) onto an Acclaim PepMap RSLC analytical column (75 micron ID × 25 cm, Thermo Scientific using a 3-25% gradient of solvent B (acetonitrile, 0.1% formic acid) over 150 min, 25-50% solvent B over 20 min, 50-70% of solvent B over 10 min, 70-95% B over 10 min then a hold of solvent 95% B for 20 min, and finally a return to 3% solvent B for 10 min. Solvent A consisted of water and 0.1% formic acid. Flow rates were 300 nL/min using a Vanquish Neo UHPLC System (Thermo Fisher Scientific, San Jose, CA). Data dependent scanning was performed by the Xcalibur v 4.3.73.11 software using a survey scan at 70,000 resolution scanning mass/charge (m/z) 300-1600 at an automatic gain control (AGC) target of 1e6 and a maximum injection time (IT) of 65 msec, followed by higher-energy collisional dissociation (HCD) tandem mass spectrometry (MS/MS) at 27 (N)CE (normalized collision energy), of the 11 most intense ions at a resolution of 17,500, an isolation width of 1.5 m/z, an AGC of 5e4 and a maximum IT of 65 msec. Dynamic exclusion was set to place any selected m/z on an exclusion list for 30 seconds after a single MS/MS. Ions of charge state +1, 6-8, >8, unassigned, and isotopes were excluded from MS/MS acquisition.

Protein identification

MS and MS/MS data were searched against the amino acid sequence of the customized database using Thermo Proteome Discoverer v 2.4.1.15 (Thermo Fisher Scientific). MS/MS spectra matches considered fully tryptic peptides with up to 2 missed cleavage sites. Variable modifications considered were methylthio on Cys (+45.988Da) and Met (M) oxidation (+15.995Da). Proteins were identified at 95% confidence with XCorr score cut-offs as determined by a reversed database search.

Translation inhibition to detect eGFP(L201S) fluorescence

Overnight cultures of strain TH29056 (ptrc-His-Flag-eGFP(L201S)/ leuZ(A25G) pnp(I654T)), strain TH29058 (ptrc-His-Flag-eGFP(L201S)/argG::Tn10dCm pnp(I654T)), and strain TH29134 (ptrc-His-Flag-eGFP(L201S)/ pnp(I654T)) were grown at 30°C in LB supplemented with 100 μg/ml Ap and 250 μM IPTG. The overnight culture was diluted, 1:100, into fresh medium and grown until it reached an OD600 of 0.4. A 1 ml portion of cell culture was spun down and resuspended into 100 μl of buffered saline. Agarose pads were prepared on microscope slides for the detection of functional eGFP protein, which results in cell fluorescence. The agar pads contained 1% agarose supplemented with minimal E-salts medium containing 50μg/ml Ap, 250 μM IPTG, and 0.6 mM arginine. A 10μl portion of cells were spread onto the entire surface of each agarose pad. Liquid antibiotic solutions were applied onto each microscope slide at one edge yielding a final applied amount of antibiotic of 3μg for Tc, 10μg for Kasugamycin, 20μg for Spectinomycin, and 2.5μg for Cm. A coverslip was placed onto each agarose pad. The microscope slide was incubated at 37°C for 6 hours, then left at room temperature until the following day. Fluorescent microscopy analysis was performed using a Zeiss Axio Observer microscope equipped with a 100x oil immersion objective, a Solid-State Light Source Colibri 7 and an Axiocam 705 mono camera. Images were acquired and generated using the Zeiss Zen 3.7 software every 5-10 steps starting at the zone of inhibition. One step is equal to 0.89 μm. The average fluorescence per cell was determined by dividing the total fluorescence by the number of cells in the captured image. The average fluorescence/cell was then plotted on a graph against the distance from the Zone of Inhibition (expressed in steps). Each antibiotic was tested in triplicate for each strain. The average between each replicate was calculated using Microsoft Excel and plotted on the graph.

Supporting information

S1 Table. List of strains.

https://doi.org/10.1371/journal.pgen.1011569.s001

(PDF)

S2 Table. Lists of oligos.

https://doi.org/10.1371/journal.pgen.1011569.s002

(PDF)

S3 Table. β-galatosidase activity values for samples analyzed in Figs 5 and 6.

*All strains carry the hisD9953::MudJ insertion. **These codons correspond to the His4 and His5 codon positions in the His leader peptide. ***Assays were performed 3 to 6 times from independent cultures. Activity is expressed in nmol/min/OD650.

https://doi.org/10.1371/journal.pgen.1011569.s003

(PDF)

S1 Fig. Next generation sequence results for four, independent spontaneous Mac-Lac red colony revertants from strain TH27698 (hisL10592(His4-5 = TCA-TAC) hisD993::MudJ thrU(C40A)) as described in Methods.

Suppressor 1 was subject to Nanopore Sequencing (Plasmidsaurus) and suppressors 2 through 4 were sequenced using Illumina Sequencing (University of Utah Core Facility) for comparative purposes.

https://doi.org/10.1371/journal.pgen.1011569.s004

(PDF)

Acknowledgments

We acknowledge Max Gottesman, Rachel Green, Diarmaid Hughes, and Venki Ramakrishnan for helpful discussions and Sherwood Casjens for critically reading this manuscript prior to submission. We thank Ning Zhang of the Analytical and Biological Mass Spectrometry Core at the University of Arizona for technical expertise and River Hoerer for technical assistance.

References

1. 1. Mustafi M, Weisshaar JC. Simultaneous binding of multiple EF-Tu copies to translating ribosomes in live Escherichia coli. mBio. 2018;9(1):e02143-17. pmid:29339430

* View Article

* PubMed/NCBI

* Google Scholar

2. 2. Rudorf S, Thommen M, Rodnina MV, Lipowsky R. Deducing the kinetics of protein synthesis in vivo from the transition rates measured in vitro. PLoS Comput Biol. 2014;10(10):e1003909. pmid:25358034

* View Article

* PubMed/NCBI

* Google Scholar

3. 3. Morse JC, Girodat D, Burnett BJ, Holm M, Altman RB, Sanbonmatsu KY, et al. Elongation factor-Tu can repetitively engage aminoacyl-tRNA within the ribosome during the proofreading stage of tRNA selection. Proc Natl Acad Sci USA. 2020;117(7):3610–20. pmid:32024753

* View Article

* PubMed/NCBI

* Google Scholar

4. 4. Diaconu M, Kothe U, Schlünzen F, Fischer N, Harms JM, Tonevitsky AG, et al. Structural basis for the function of the ribosomal L7/12 stalk in factor binding and GTPase activation. Cell. 2005;121(7):991–1004. pmid:15989950

* View Article

* PubMed/NCBI

* Google Scholar

5. 5. Kothe U, Wieden H-J, Mohr D, Rodnina MV. Interaction of helix D of elongation factor Tu with helices 4 and 5 of protein L7/12 on the ribosome. J Mol Biol. 2004;336(5):1011–21. pmid:15037065

* View Article

* PubMed/NCBI

* Google Scholar

6. 6. Fischer N, Neumann P, Bock LV, Maracci C, Wang Z, Paleskava A, et al. The pathway to GTPase activation of elongation factor SelB on the ribosome. Nature. 2016;540(7631):80–5. pmid:27842381

* View Article

* PubMed/NCBI

* Google Scholar

7. 7. Loveland AB, Demo G, Grigorieff N, Korostelev AA. Ensemble cryo-EM elucidates the mechanism of translation fidelity. Nature. 2017;546(7656):113–7. pmid:28538735

* View Article

* PubMed/NCBI

* Google Scholar

8. 8. Demeshkina N, Jenner L, Westhof E, Yusupov M, Yusupova G. A new understanding of the decoding principle on the ribosome. Nature. 2012;484(7393):256–9. pmid:22437501

* View Article

* PubMed/NCBI

* Google Scholar

9. 9. Ogle JM, Brodersen DE, Clemons WM Jr, Tarry MJ, Carter AP, Ramakrishnan V. Recognition of cognate transfer RNA by the 30S ribosomal subunit. Science. 2001;292(5518):897–902. pmid:11340196

* View Article

* PubMed/NCBI

* Google Scholar

10. 10. Dale T, Fahlman RP, Olejniczak M, Uhlenbeck OC. Specificity of the ribosomal A site for aminoacyl-tRNAs. Nucleic Acids Res. 2009;37(4):1202–10. pmid:19129224

* View Article

* PubMed/NCBI

* Google Scholar

11. 11. Åqvist J, Kamerlin SCL. The conformation of a catalytic loop is central to GTPase activity on the ribosome. Biochemistry. 2015;54(2):546–56. pmid:25515218

* View Article

* PubMed/NCBI

* Google Scholar

12. 12. Wool IG, Glück A, Endo Y. Ribotoxin recognition of ribosomal RNA and a proposal for the mechanism of translocation. Trends Biochem Sci. 1992;17(7):266–9. pmid:1502728

* View Article

* PubMed/NCBI

* Google Scholar

13. 13. Schmeing TM, Voorhees RM, Kelley AC, Gao Y-G, Murphy FV 4th, Weir JR, et al. The crystal structure of the ribosome bound to EF-Tu and aminoacyl-tRNA. Science. 2009;326(5953):688–94. pmid:19833920

* View Article

* PubMed/NCBI

* Google Scholar

14. 14. Schuette J-C, Murphy FV 4th, Kelley AC, Weir JR, Giesebrecht J, Connell SR, et al. GTPase activation of elongation factor EF-Tu by the ribosome during decoding. EMBO J. 2009;28(6):755–65. pmid:19229291

* View Article

* PubMed/NCBI

* Google Scholar

15. 15. Schmeing TM, Voorhees RM, Kelley AC, Ramakrishnan V. How mutations in tRNA distant from the anticodon affect the fidelity of decoding. Nat Struct Mol Biol. 2011;18(4):432–6. pmid:21378964

* View Article

* PubMed/NCBI

* Google Scholar

16. 16. Voorhees RM, Schmeing TM, Kelley AC, Ramakrishnan V. The mechanism for activation of GTP hydrolysis on the ribosome. Science. 2010;330(6005):835–8. pmid:21051640

* View Article

* PubMed/NCBI

* Google Scholar

17. 17. Maracci C, Peske F, Dannies E, Pohl C, Rodnina MV. Ribosome-induced tuning of GTP hydrolysis by a translational GTPase. Proc Natl Acad Sci USA. 2014;111(40):14418–23. pmid:25246550

* View Article

* PubMed/NCBI

* Google Scholar

18. 18. Caulfield T, Devkota B. Motion of transfer RNA from the A/T state into the A-site using docking and simulations. Proteins. 2012;80(11):2489–500. pmid:22730134

* View Article

* PubMed/NCBI

* Google Scholar

19. 19. Valle M, Zavialov A, Li W, Stagg SM, Sengupta J, Nielsen RC, et al. Incorporation of aminoacyl-tRNA into the ribosome as seen by cryo-electron microscopy. Nat Struct Biol. 2003;10(11):899–906. pmid:14566331

* View Article

* PubMed/NCBI

* Google Scholar

20. 20. Mohr D, Wintermeyer W, Rodnina MV. GTPase activation of elongation factors Tu and G on the ribosome. Biochemistry. 2002;41(41):12520–8. pmid:12369843

* View Article

* PubMed/NCBI

* Google Scholar

21. 21. Girodat D, Blanchard SC, Wieden H-J, Sanbonmatsu KY. Elongation factor Tu switch I element is a gate for aminoacyl-tRNA selection. J Mol Biol. 2020;432(9):3064–77. pmid:32061931

* View Article

* PubMed/NCBI

* Google Scholar

22. 22. Frank J, Agrawal RK. A ratchet-like inter-subunit reorganization of the ribosome during translocation. Nature. 2000;406(6793):318–22. pmid:10917535

* View Article

* PubMed/NCBI

* Google Scholar

23. 23. Presnyak V, Alhusaini N, Chen Y-H, Martin S, Morris N, Kline N, et al. Codon optimality is a major determinant of mRNA stability. Cell. 2015;160(6):1111–24. pmid:25768907

* View Article

* PubMed/NCBI

* Google Scholar

24. 24. Komar AA, Lesnik T, Reiss C. Synonymous codon substitutions affect ribosome traffic and protein folding during in vitro translation. FEBS Lett. 1999;462(3):387–91. pmid:10622731

* View Article

* PubMed/NCBI

* Google Scholar

25. 25. Kudla G, Murray AW, Tollervey D, Plotkin JB. Coding-sequence determinants of gene expression in Escherichia coli. Science. 2009;324(5924):255–8. pmid:19359587

* View Article

* PubMed/NCBI

* Google Scholar

26. 26. Zhou M, Guo J, Cha J, Chae M, Chen S, Barral JM, et al. Non-optimal codon usage affects expression, structure and function of clock protein FRQ. Nature. 2013;495(7439):111–5. pmid:23417067

* View Article

* PubMed/NCBI

* Google Scholar

27. 27. Ogle JM, Carter AP, Ramakrishnan V. Insights into the decoding mechanism from recent ribosome structures. Trends Biochem Sci. 2003;28(5):259–66. pmid:12765838

* View Article

* PubMed/NCBI

* Google Scholar

28. 28. Gromadski KB, Daviter T, Rodnina MV. A uniform response to mismatches in codon-anticodon complexes ensures ribosomal fidelity. Mol Cell. 2006;21(3):369–77. pmid:16455492

* View Article

* PubMed/NCBI

* Google Scholar

29. 29. Rojiani MV, Jakubowski H, Goldman E. Relationship between protein synthesis and concentrations of charged and uncharged tRNATrp in Escherichia coli. Proc Natl Acad Sci USA. 1990;87(4):1511–5. pmid:2106136

* View Article

* PubMed/NCBI

* Google Scholar

30. 30. Elf J, Nilsson D, Tenson T, Ehrenberg M. Selective charging of tRNA isoacceptors explains patterns of codon usage. Science. 2003;300(5626):1718–22. pmid:12805541

* View Article

* PubMed/NCBI

* Google Scholar

31. 31. Bossi L, Roth JR. The influence of codon context on genetic code translation. Nature. 1980;286(5769):123–7. pmid:7402305

* View Article

* PubMed/NCBI

* Google Scholar

32. 32. Bouadloun F, Srichaiyo T, Isaksson LA, Björk GR. Influence of modification next to the anticodon in tRNA on codon context sensitivity of translational suppression and accuracy. J Bacteriol. 1986;166(3):1022–7. pmid:3086285

* View Article

* PubMed/NCBI

* Google Scholar

33. 33. Björnsson A, Isaksson LA. UGA codon context which spans three codons. Reversal by ms2i6A37 in tRNA, mutation in rpsD(S4) or streptomycin. J Mol Biol. 1993;232(4):1017–29. pmid:8371264

* View Article

* PubMed/NCBI

* Google Scholar

34. 34. Chevance FFV, Hughes KT. Case for the genetic code as a triplet of triplets. Proc Natl Acad Sci USA. 2017;114(18):4745–50. pmid:28416671

* View Article

* PubMed/NCBI

* Google Scholar

35. 35. Johnston HM, Barnes WM, Chumley FG, Bossi L, Roth JR. Model for regulation of the histidine operon of Salmonella. Proc Natl Acad Sci USA. 1980;77(1):508–12. pmid:6987654

* View Article

* PubMed/NCBI

* Google Scholar

36. 36. Johnston HM, Roth JR. Genetic analysis of the histidine operon control region of Salmonella typhimurium. J Mol Biol. 1981;145(4):713–34. pmid:7021855

* View Article

* PubMed/NCBI

* Google Scholar

37. 37. Johnston HM, Roth JR. DNA sequence changes of mutations altering attenuation control of the histidine operon of Salmonella typhimurium. J Mol Biol. 1981;145(4):735–56. pmid:6167727

* View Article

* PubMed/NCBI

* Google Scholar

38. 38. Spöring I, M V, Hotz C, Schwarz-Linek J, Grady KL, Nava-Sedeño JM, et al. Hlotbfiofmsotf bundle. PBS. 16(9):e2006989.

* View Article

* Google Scholar

39. 39. Chevance FFV, Le Guyon S, Hughes KT. The effects of codon context on in vivo translation speed. PLoS Genet. 2014;10(6):e1004392. pmid:24901308

* View Article

* PubMed/NCBI

* Google Scholar

40. 40. Li X-T, Thomason LC, Sawitzke JA, Costantino N, Court DL. Positive and negative selection using the tetA-sacB cassette: recombineering and P1 transduction in Escherichia coli. Nucleic Acids Res. 2013;41(22):e204. pmid:24203710

* View Article

* PubMed/NCBI

* Google Scholar

41. 41. Busby S, Ebright RH. Transcription activation by catabolite activator protein (CAP). J Mol Biol. 1999;293(2):199–213. pmid:10550204

* View Article

* PubMed/NCBI

* Google Scholar

42. 42. Gross CA, Chan C, Dombroski A, Gruber T, Sharp M, Tupy J, et al. The functional and regulatory roles of sigma factors in transcription. Cold Spring Harb Symp Quant Biol. 1998;63:141–55. pmid:10384278

* View Article

* PubMed/NCBI

* Google Scholar

43. 43. Pavlov MY, Watts RE, Tan Z, Cornish VW, Ehrenberg M, Forster AC. Slow peptide bond formation by proline and other N-alkylamino acids in translation. Proc Natl Acad Sci USA. 2009;106(1):50–4. pmid:19104062

* View Article

* PubMed/NCBI

* Google Scholar

44. 44. Petrone PM, Snow CD, Lucent D, Pande VS. Side-chain recognition and gating in the ribosome exit tunnel. Proc Natl Acad Sci USA. 2008;105(43):16549–54. pmid:18946046

* View Article

* PubMed/NCBI

* Google Scholar

45. 45. Chaconas G, de Bruijn FJ, Casadaban MJ, Lupski JR, Kwoh TJ, Harshey RM, et al. In vitro and in vivo manipulations of bacteriophage Mu DNA: cloning of Mu ends and construction of mini-Mu’s carrying selectable markers. Gene. 1981;13(1):37–46. pmid:6263754

* View Article

* PubMed/NCBI

* Google Scholar

46. 46. Anderson P, Roth J. Spontaneous tandem genetic duplications in Salmonella typhimurium arise by unequal recombination between rRNA (rrn) cistrons. Proc Natl Acad Sci USA. 1981;78(5):3113–7. pmid:6789329

* View Article

* PubMed/NCBI

* Google Scholar

47. 47. Aiso T, Ohki R. An rne-1 pnp-7 double mutation suppresses the temperature-sensitive defect of lacZ gene expression in a divE mutant. J Bacteriol. 1998;180(6):1389–95. pmid:9515904

* View Article

* PubMed/NCBI

* Google Scholar

48. 48. Li Z, Reimers S, Pandit S, Deutscher MP. RNA quality control: degradation of defective transfer RNA. EMBO J. 2002;21(5):1132–8. pmid:11867541

* View Article

* PubMed/NCBI

* Google Scholar

49. 49. Shi Z, Yang W-Z, Lin-Chao S, Chak K-F, Yuan HS. Crystal structure of Escherichia coli PNPase: central channel residues are involved in processive RNA degradation. RNA. 2008;14(11):2361–71. pmid:18812438

* View Article

* PubMed/NCBI

* Google Scholar

50. 50. Bezerra AR, Simões J, Lee W, Rung J, Weil T, Gut IG, et al. Reversion of a fungal genetic code alteration links proteome instability with genomic and phenotypic diversification. Proc Natl Acad Sci USA. 2013;110(27):11079–84. pmid:23776239

* View Article

* PubMed/NCBI

* Google Scholar

51. 51. Tamura F, Nishimura S, Ohki M. The E. coli divE mutation, which differentially inhibits synthesis of certain proteins, is in tRNASer1. EMBO J. 1984;3(5):1103–7. pmid:6376117

* View Article

* PubMed/NCBI

* Google Scholar

52. 52. Chevance FFV, Karlinsey JE, Wozniak CE, Hughes KT. A little gene with big effects: a serT mutant is defective in flgM gene translation. J Bacteriol. 2006;188(1):297–304. pmid:16352846

* View Article

* PubMed/NCBI

* Google Scholar

53. 53. Ude S, Lassak J, Starosta AL, Kraxenberger T, Wilson DN, Jung K. Translation elongation factor EF-P alleviates ribosome stalling at polyproline stretches. Science. 2013;339(6115):82–5. pmid:23239623

* View Article

* PubMed/NCBI

* Google Scholar

54. 54. Wee LM, Tong AB, Florez Ariza AJ, Cañari-Chumpitaz C, Grob P, Nogales E, et al. A trailing ribosome speeds up RNA polymerase at the expense of transcript fidelity via force and allostery. Cell. 2023;186(6):1244–62.e34. pmid:36931247

* View Article

* PubMed/NCBI

* Google Scholar

55. 55. Burmann BM, Schweimer K, Luo X, Wahl MC, Stitt BL, Gottesman ME, et al. A NusE:NusG complex links transcription and translation. Science. 2010;328(5977):501–4. pmid:20413501

* View Article

* PubMed/NCBI

* Google Scholar

56. 56. Arenz S, Wilson DN. Bacterial protein synthesis as a target for antibiotic inhibition. Cold Spring Harb Perspect Med. 2016;6(9):a025361. pmid:27481773

* View Article

* PubMed/NCBI

* Google Scholar

57. 57. Gromadski KB, Rodnina MV. Kinetic determinants of high-fidelity tRNA discrimination on the ribosome. Mol Cell. 2004;13(2):191–200. pmid:14759365

* View Article

* PubMed/NCBI

* Google Scholar

58. 58. Cochella L, Green R. An active role for tRNA in decoding beyond codon:anticodon pairing. Science. 2005;308(5725):1178–80. pmid:15905403

* View Article

* PubMed/NCBI

* Google Scholar

59. 59. Hirsh D. Tryptophan transfer RNA as the UGA suppressor. J Mol Biol. 1971;58(2):439–58. pmid:4933412

* View Article

* PubMed/NCBI

* Google Scholar

60. 60. Bouadloun F, Donner D, Kurland CG. Codon-specific missense errors in vivo. EMBO J. 1983;2(8):1351–6. pmid:10872330

* View Article

* PubMed/NCBI

* Google Scholar

61. 61. Edelmann P, Gallant J. Mistranslation in E. coli. Cell. 1977;10(1):131–7. pmid:138485

* View Article

* PubMed/NCBI

* Google Scholar

62. 62. Ruusala T, Andersson D, Ehrenberg M, Kurland CG. Hyper-accurate ribosomes inhibit growth. EMBO J. 1984;3(11):2575–80. pmid:6391914

* View Article

* PubMed/NCBI

* Google Scholar

63. 63. Andersson DI, van Verseveld HW, Stouthamer AH, Kurland CG. Suboptimal growth with hyper-accurate ribosomes. Arch Microbiol. 1986;144(1):96–101. pmid:3963992

* View Article

* PubMed/NCBI

* Google Scholar

64. 64. Karlinsey JE. lambda-red genetic engineering in Salmonella enterica serovar Typhimurium. Methods Enzymol. 2007;421:199–209. pmid:17352924

* View Article

* PubMed/NCBI

* Google Scholar

65. 65. Davis RW, Botstein D, Roth JR. Advanced bacterial genetics. Cold Spring Harbor, N.Y.: Cold Spring Harbor Laboratory; 1980.

66. 66. Datsenko KA, Wanner BL. One-step inactivation of chromosomal genes in Escherichia coli K-12 using PCR products. Proc Natl Acad Sci USA. 2000;97(12):6640–5. pmid:10829079

* View Article

* PubMed/NCBI

* Google Scholar

67. 67. Hughes KT, Roth JR. Directed formation of deletions and duplications using Mud(Ap, lac). Genetics. 1985;109(2):263–82. pmid:3156064

* View Article

* PubMed/NCBI

* Google Scholar

68. 68. Maloy SR. Experimental techniques in bacterial genetics. Boston: Jones and Bartlett; 1990.

69. 69. Minamino T, Macnab RM. Components of the Salmonella flagellar export apparatus and classification of export substrates. J Bacteriol. 1999;181(5):1388–94. pmid:10049367

* View Article

* PubMed/NCBI

* Google Scholar

70. 70. Jenner LB, Demeshkina N, Yusupova G, Yusupov M. Structural aspects of messenger RNA reading frame maintenance by the ribosome. Nat Struct Mol Biol. 2010;17(5):555–60. pmid:20400952

* View Article

* PubMed/NCBI

* Google Scholar

Citation: Schrock MN, Parsawar K, Hughes KT, Chevance FFV (2025) D-stem mutation in an essential tRNA increases translation speed at the cost of fidelity. PLoS Genet 21(2): e1011569. https://doi.org/10.1371/journal.pgen.1011569

References

1. Mustafi M, Weisshaar JC. Simultaneous binding of multiple EF-Tu copies to translating ribosomes in live Escherichia coli. mBio. 2018;9(1):e02143-17. pmid:29339430

2. Rudorf S, Thommen M, Rodnina MV, Lipowsky R. Deducing the kinetics of protein synthesis in vivo from the transition rates measured in vitro. PLoS Comput Biol. 2014;10(10):e1003909. pmid:25358034

3. Morse JC, Girodat D, Burnett BJ, Holm M, Altman RB, Sanbonmatsu KY, et al. Elongation factor-Tu can repetitively engage aminoacyl-tRNA within the ribosome during the proofreading stage of tRNA selection. Proc Natl Acad Sci USA. 2020;117(7):3610–20. pmid:32024753

4. Diaconu M, Kothe U, Schlünzen F, Fischer N, Harms JM, Tonevitsky AG, et al. Structural basis for the function of the ribosomal L7/12 stalk in factor binding and GTPase activation. Cell. 2005;121(7):991–1004. pmid:15989950

5. Kothe U, Wieden H-J, Mohr D, Rodnina MV. Interaction of helix D of elongation factor Tu with helices 4 and 5 of protein L7/12 on the ribosome. J Mol Biol. 2004;336(5):1011–21. pmid:15037065

6. Fischer N, Neumann P, Bock LV, Maracci C, Wang Z, Paleskava A, et al. The pathway to GTPase activation of elongation factor SelB on the ribosome. Nature. 2016;540(7631):80–5. pmid:27842381

7. Loveland AB, Demo G, Grigorieff N, Korostelev AA. Ensemble cryo-EM elucidates the mechanism of translation fidelity. Nature. 2017;546(7656):113–7. pmid:28538735

8. Demeshkina N, Jenner L, Westhof E, Yusupov M, Yusupova G. A new understanding of the decoding principle on the ribosome. Nature. 2012;484(7393):256–9. pmid:22437501

9. Ogle JM, Brodersen DE, Clemons WM Jr, Tarry MJ, Carter AP, Ramakrishnan V. Recognition of cognate transfer RNA by the 30S ribosomal subunit. Science. 2001;292(5518):897–902. pmid:11340196

10. Dale T, Fahlman RP, Olejniczak M, Uhlenbeck OC. Specificity of the ribosomal A site for aminoacyl-tRNAs. Nucleic Acids Res. 2009;37(4):1202–10. pmid:19129224

11. Åqvist J, Kamerlin SCL. The conformation of a catalytic loop is central to GTPase activity on the ribosome. Biochemistry. 2015;54(2):546–56. pmid:25515218

12. Wool IG, Glück A, Endo Y. Ribotoxin recognition of ribosomal RNA and a proposal for the mechanism of translocation. Trends Biochem Sci. 1992;17(7):266–9. pmid:1502728

13. Schmeing TM, Voorhees RM, Kelley AC, Gao Y-G, Murphy FV 4th, Weir JR, et al. The crystal structure of the ribosome bound to EF-Tu and aminoacyl-tRNA. Science. 2009;326(5953):688–94. pmid:19833920

14. Schuette J-C, Murphy FV 4th, Kelley AC, Weir JR, Giesebrecht J, Connell SR, et al. GTPase activation of elongation factor EF-Tu by the ribosome during decoding. EMBO J. 2009;28(6):755–65. pmid:19229291

15. Schmeing TM, Voorhees RM, Kelley AC, Ramakrishnan V. How mutations in tRNA distant from the anticodon affect the fidelity of decoding. Nat Struct Mol Biol. 2011;18(4):432–6. pmid:21378964

16. Voorhees RM, Schmeing TM, Kelley AC, Ramakrishnan V. The mechanism for activation of GTP hydrolysis on the ribosome. Science. 2010;330(6005):835–8. pmid:21051640

17. Maracci C, Peske F, Dannies E, Pohl C, Rodnina MV. Ribosome-induced tuning of GTP hydrolysis by a translational GTPase. Proc Natl Acad Sci USA. 2014;111(40):14418–23. pmid:25246550

18. Caulfield T, Devkota B. Motion of transfer RNA from the A/T state into the A-site using docking and simulations. Proteins. 2012;80(11):2489–500. pmid:22730134

19. Valle M, Zavialov A, Li W, Stagg SM, Sengupta J, Nielsen RC, et al. Incorporation of aminoacyl-tRNA into the ribosome as seen by cryo-electron microscopy. Nat Struct Biol. 2003;10(11):899–906. pmid:14566331

20. Mohr D, Wintermeyer W, Rodnina MV. GTPase activation of elongation factors Tu and G on the ribosome. Biochemistry. 2002;41(41):12520–8. pmid:12369843

21. Girodat D, Blanchard SC, Wieden H-J, Sanbonmatsu KY. Elongation factor Tu switch I element is a gate for aminoacyl-tRNA selection. J Mol Biol. 2020;432(9):3064–77. pmid:32061931

22. Frank J, Agrawal RK. A ratchet-like inter-subunit reorganization of the ribosome during translocation. Nature. 2000;406(6793):318–22. pmid:10917535

23. Presnyak V, Alhusaini N, Chen Y-H, Martin S, Morris N, Kline N, et al. Codon optimality is a major determinant of mRNA stability. Cell. 2015;160(6):1111–24. pmid:25768907

24. Komar AA, Lesnik T, Reiss C. Synonymous codon substitutions affect ribosome traffic and protein folding during in vitro translation. FEBS Lett. 1999;462(3):387–91. pmid:10622731

25. Kudla G, Murray AW, Tollervey D, Plotkin JB. Coding-sequence determinants of gene expression in Escherichia coli. Science. 2009;324(5924):255–8. pmid:19359587

26. Zhou M, Guo J, Cha J, Chae M, Chen S, Barral JM, et al. Non-optimal codon usage affects expression, structure and function of clock protein FRQ. Nature. 2013;495(7439):111–5. pmid:23417067

27. Ogle JM, Carter AP, Ramakrishnan V. Insights into the decoding mechanism from recent ribosome structures. Trends Biochem Sci. 2003;28(5):259–66. pmid:12765838

28. Gromadski KB, Daviter T, Rodnina MV. A uniform response to mismatches in codon-anticodon complexes ensures ribosomal fidelity. Mol Cell. 2006;21(3):369–77. pmid:16455492

29. Rojiani MV, Jakubowski H, Goldman E. Relationship between protein synthesis and concentrations of charged and uncharged tRNATrp in Escherichia coli. Proc Natl Acad Sci USA. 1990;87(4):1511–5. pmid:2106136

30. Elf J, Nilsson D, Tenson T, Ehrenberg M. Selective charging of tRNA isoacceptors explains patterns of codon usage. Science. 2003;300(5626):1718–22. pmid:12805541

31. Bossi L, Roth JR. The influence of codon context on genetic code translation. Nature. 1980;286(5769):123–7. pmid:7402305

32. Bouadloun F, Srichaiyo T, Isaksson LA, Björk GR. Influence of modification next to the anticodon in tRNA on codon context sensitivity of translational suppression and accuracy. J Bacteriol. 1986;166(3):1022–7. pmid:3086285

33. Björnsson A, Isaksson LA. UGA codon context which spans three codons. Reversal by ms2i6A37 in tRNA, mutation in rpsD(S4) or streptomycin. J Mol Biol. 1993;232(4):1017–29. pmid:8371264

34. Chevance FFV, Hughes KT. Case for the genetic code as a triplet of triplets. Proc Natl Acad Sci USA. 2017;114(18):4745–50. pmid:28416671

35. Johnston HM, Barnes WM, Chumley FG, Bossi L, Roth JR. Model for regulation of the histidine operon of Salmonella. Proc Natl Acad Sci USA. 1980;77(1):508–12. pmid:6987654

36. Johnston HM, Roth JR. Genetic analysis of the histidine operon control region of Salmonella typhimurium. J Mol Biol. 1981;145(4):713–34. pmid:7021855

37. Johnston HM, Roth JR. DNA sequence changes of mutations altering attenuation control of the histidine operon of Salmonella typhimurium. J Mol Biol. 1981;145(4):735–56. pmid:6167727

38. Spöring I, M V, Hotz C, Schwarz-Linek J, Grady KL, Nava-Sedeño JM, et al. Hlotbfiofmsotf bundle. PBS. 16(9):e2006989.

39. Chevance FFV, Le Guyon S, Hughes KT. The effects of codon context on in vivo translation speed. PLoS Genet. 2014;10(6):e1004392. pmid:24901308

40. Li X-T, Thomason LC, Sawitzke JA, Costantino N, Court DL. Positive and negative selection using the tetA-sacB cassette: recombineering and P1 transduction in Escherichia coli. Nucleic Acids Res. 2013;41(22):e204. pmid:24203710

41. Busby S, Ebright RH. Transcription activation by catabolite activator protein (CAP). J Mol Biol. 1999;293(2):199–213. pmid:10550204

42. Gross CA, Chan C, Dombroski A, Gruber T, Sharp M, Tupy J, et al. The functional and regulatory roles of sigma factors in transcription. Cold Spring Harb Symp Quant Biol. 1998;63:141–55. pmid:10384278

43. Pavlov MY, Watts RE, Tan Z, Cornish VW, Ehrenberg M, Forster AC. Slow peptide bond formation by proline and other N-alkylamino acids in translation. Proc Natl Acad Sci USA. 2009;106(1):50–4. pmid:19104062

44. Petrone PM, Snow CD, Lucent D, Pande VS. Side-chain recognition and gating in the ribosome exit tunnel. Proc Natl Acad Sci USA. 2008;105(43):16549–54. pmid:18946046

45. Chaconas G, de Bruijn FJ, Casadaban MJ, Lupski JR, Kwoh TJ, Harshey RM, et al. In vitro and in vivo manipulations of bacteriophage Mu DNA: cloning of Mu ends and construction of mini-Mu’s carrying selectable markers. Gene. 1981;13(1):37–46. pmid:6263754

46. Anderson P, Roth J. Spontaneous tandem genetic duplications in Salmonella typhimurium arise by unequal recombination between rRNA (rrn) cistrons. Proc Natl Acad Sci USA. 1981;78(5):3113–7. pmid:6789329

47. Aiso T, Ohki R. An rne-1 pnp-7 double mutation suppresses the temperature-sensitive defect of lacZ gene expression in a divE mutant. J Bacteriol. 1998;180(6):1389–95. pmid:9515904

48. Li Z, Reimers S, Pandit S, Deutscher MP. RNA quality control: degradation of defective transfer RNA. EMBO J. 2002;21(5):1132–8. pmid:11867541

49. Shi Z, Yang W-Z, Lin-Chao S, Chak K-F, Yuan HS. Crystal structure of Escherichia coli PNPase: central channel residues are involved in processive RNA degradation. RNA. 2008;14(11):2361–71. pmid:18812438

50. Bezerra AR, Simões J, Lee W, Rung J, Weil T, Gut IG, et al. Reversion of a fungal genetic code alteration links proteome instability with genomic and phenotypic diversification. Proc Natl Acad Sci USA. 2013;110(27):11079–84. pmid:23776239

51. Tamura F, Nishimura S, Ohki M. The E. coli divE mutation, which differentially inhibits synthesis of certain proteins, is in tRNASer1. EMBO J. 1984;3(5):1103–7. pmid:6376117

52. Chevance FFV, Karlinsey JE, Wozniak CE, Hughes KT. A little gene with big effects: a serT mutant is defective in flgM gene translation. J Bacteriol. 2006;188(1):297–304. pmid:16352846

53. Ude S, Lassak J, Starosta AL, Kraxenberger T, Wilson DN, Jung K. Translation elongation factor EF-P alleviates ribosome stalling at polyproline stretches. Science. 2013;339(6115):82–5. pmid:23239623

54. Wee LM, Tong AB, Florez Ariza AJ, Cañari-Chumpitaz C, Grob P, Nogales E, et al. A trailing ribosome speeds up RNA polymerase at the expense of transcript fidelity via force and allostery. Cell. 2023;186(6):1244–62.e34. pmid:36931247

55. Burmann BM, Schweimer K, Luo X, Wahl MC, Stitt BL, Gottesman ME, et al. A NusE:NusG complex links transcription and translation. Science. 2010;328(5977):501–4. pmid:20413501

56. Arenz S, Wilson DN. Bacterial protein synthesis as a target for antibiotic inhibition. Cold Spring Harb Perspect Med. 2016;6(9):a025361. pmid:27481773

57. Gromadski KB, Rodnina MV. Kinetic determinants of high-fidelity tRNA discrimination on the ribosome. Mol Cell. 2004;13(2):191–200. pmid:14759365

58. Cochella L, Green R. An active role for tRNA in decoding beyond codon:anticodon pairing. Science. 2005;308(5725):1178–80. pmid:15905403

59. Hirsh D. Tryptophan transfer RNA as the UGA suppressor. J Mol Biol. 1971;58(2):439–58. pmid:4933412

60. Bouadloun F, Donner D, Kurland CG. Codon-specific missense errors in vivo. EMBO J. 1983;2(8):1351–6. pmid:10872330

61. Edelmann P, Gallant J. Mistranslation in E. coli. Cell. 1977;10(1):131–7. pmid:138485

62. Ruusala T, Andersson D, Ehrenberg M, Kurland CG. Hyper-accurate ribosomes inhibit growth. EMBO J. 1984;3(11):2575–80. pmid:6391914

63. Andersson DI, van Verseveld HW, Stouthamer AH, Kurland CG. Suboptimal growth with hyper-accurate ribosomes. Arch Microbiol. 1986;144(1):96–101. pmid:3963992

64. Karlinsey JE. lambda-red genetic engineering in Salmonella enterica serovar Typhimurium. Methods Enzymol. 2007;421:199–209. pmid:17352924

65. Davis RW, Botstein D, Roth JR. Advanced bacterial genetics. Cold Spring Harbor, N.Y.: Cold Spring Harbor Laboratory; 1980.

66. Datsenko KA, Wanner BL. One-step inactivation of chromosomal genes in Escherichia coli K-12 using PCR products. Proc Natl Acad Sci USA. 2000;97(12):6640–5. pmid:10829079

67. Hughes KT, Roth JR. Directed formation of deletions and duplications using Mud(Ap, lac). Genetics. 1985;109(2):263–82. pmid:3156064

68. Maloy SR. Experimental techniques in bacterial genetics. Boston: Jones and Bartlett; 1990.

69. Minamino T, Macnab RM. Components of the Salmonella flagellar export apparatus and classification of export substrates. J Bacteriol. 1999;181(5):1388–94. pmid:10049367

70. Jenner LB, Demeshkina N, Yusupova G, Yusupov M. Structural aspects of messenger RNA reading frame maintenance by the ribosome. Nat Struct Mol Biol. 2010;17(5):555–60. pmid:20400952