Codon usage determines translation rate in Escherichia coli

A survey of codon and amino acid frequency bias in microbial genomes focusing on translational efficiency

Unequal use of synonymous codons has been found in several prokaryotic and eukaryotic genomes. This bias has been associated with translational efficiency. The prevalence of this bias across lineages is currently unknown. Here, a new method (GCB) to measure codon usage bias is presented. It uses an iterative approach for the determination of codon scores and allows the computation of an index of codon bias suitable for interspecies comparison. A server to calculate GCB-values of individual genes as well as a list of compiled results are available at www.g21.bio.uni-goettingen.de. The method was applied to complete bacterial genomes. The relation of codon usage bias with amino acid composition and the choice of stop codons were determined and discussed.

Dynamics and bistability in a reduced model of the lac operon

It is known that the lac operon regulatory pathway is capable of showing bistable behavior. This is an important complex feature, arising from the nonlinearity of the involved mechanisms, which is essential to understand the dynamic behavior of this molecular regulatory system. To find which of the mechanisms involved in the regulation of the lac operon is the origin of bistability, we take a previously published model which accounts for the dynamics of mRNA, lactose, allolactose, permease and beta-galactosidase involvement and simplify it by ignoring permease dynamics (assuming a constant permease concentration). To test the behavior of the reduced model, three existing sets of data on beta-galactosidase levels as a function of time are simulated and we obtain a reasonable agreement between the data and the model predictions. The steady states of the reduced model were numerically and analytically analyzed and it was shown that it may indeed display bistability, depending on the extracellular lactose concentration and growth rate.

Optimization of codon pair use within the (GGGGS)3 linker sequence results in enhanced protein expression

Here, we report that a significant increase in recombinant fusion antibody expression can be accomplished by adjusting the nucleotide sequence to conform to certain codon pairing rules. We investigated the expression of a protein in which a single chain Fv specific for HER2/neu with VH and VL joined by a flexible (GGGGS)3 linker was linked to the CH3 of a human anti-rat transferrin receptor IgG3 heavy chain with the same flexible (GGGGS)3 linker. In initial experiments we failed to achieve significant expression of this protein. However, when we made a single nucleotide change in each (GGGGS)3 linker we were able to achieve expression The change of one nucleotide within each linker did not alter either the amino acid sequence or the frequency score of these codon triplets' usage in mammalian cells. Instead they removed two codon pairs predicted to be detrimental to expression. In a transient transfection assay we find that this change results in an over 30-fold increase in expression that is not the result of an increase in the level of accumulated mRNA. In addition, the changes made it possible to isolate stably transfected mammalian cell clones producing high levels of fusion protein, which had not been possible using the original gene.

Protein structure preference, tRNA copy number, and mRNA stem/loop content

From statistical analyses of protein sequences for humans and Escherichia coli we found that the messenger RNA segment of m-codons (for m=2 to 6) with average high tRNA copy number (TCN) (larger than approximately 10.5 for humans or approximately 1.95 for E. coli) preferably code for the alpha helix and that with low TCN (smaller than approximately 7.5 for humans or approximately 1.7 for E. coli) preferably code for coil. Between them there is an intermediate region without correlation to structure preference. For the beta strand the preference/ avoidance tendency is not obvious. All strong preference-modes of TCN for protein secondary structures have been deduced. The mutual interaction between two factors--protein secondary structural type and codon TCN--is tested by F distribution. A phenomenological model on the relation between structure preference and translational efficiency or accuracy is proposed. It is pointed out that the structure preference of codons is related to the distribution of mRNA stem/loop content in three TCN regions.

Translational selection and yeast proteome evolution

The primary structures of peptides may be adapted for efficient synthesis as well as proper function. Here, the Saccharomyces cerevisiae genome sequence, DNA microarray expression data, tRNA gene numbers, and functional categorizations of proteins are employed to determine whether the amino acid composition of peptides reflects natural selection to optimize the speed and accuracy of translation. Strong relationships between synonymous codon usage bias and estimates of transcript abundance suggest that DNA array data serve as adequate predictors of translation rates. Amino acid usage also shows striking relationships with expression levels. Stronger correlations between tRNA concentrations and amino acid abundances among highly expressed proteins than among less abundant proteins support adaptation of both tRNA abundances and amino acid usage to enhance the speed and accuracy of protein synthesis. Natural selection for efficient synthesis appears to also favor shorter proteins as a function of their expression levels. Comparisons restricted to proteins within functional classes are employed to control for differences in amino acid composition and protein size that reflect differences in the functional requirements of proteins expressed at different levels.

Protein folding

The problem of protein folding is that how proteins acquire their native unique three-dimensional structure in the physiological milieu. To solve the problem, the following key questions should be answered: do proteins fold co- or post-translationally, i.e. during or after biosynthesis, what is the mechanism of protein folding, and what is the explanation for fast folding of proteins? The two first questions are discussed in the current review. The general lines are to show that the opinion, that proteins fold after they are synthesized is hardly substantiated and suitable for solving the problem of protein folding and why proteins should fold cotranslationally. A possible tentative model for the mechanism of protein folding is also suggested. To this end, a thorough analysis is made of the biosynthesis, delivery to the folding compartments, and the rates of the biosynthesis, translocation and folding of proteins. A cursory attention is assigned to the role of GroEL/ES-like chaperonins in protein folding.

High-level expression of the mycobacterial porin MspA in Escherichia coli and purification of the recombinant protein

MspA is the prototype of a new family of tetrameric porins and provides the main general diffusion pathway for hydrophilic compounds through the outer membrane of Mycobacterium smegmatis. Structural analysis was hampered by the scarce amount of pure protein. After replacement of the GC-rich codons of the mspA gene by codons optimal for high-level expression in Escherichia coli, the mature MspA protein was overproduced in E. coli. The recombinant MspA (rMspA) monomer (M(r) 20000) was purified by anion exchange and hydrophobic interaction chromatography yielding 2.6 mg pure protein per liter of culture. This exceeded the yield of the native protein 10-fold. Circular dichroism revealed that rMspA is folded in a native-like structure. rMspA assembled partially to the channel-forming tetramer both during expression in E. coli and after purification in vitro. Thus, overexpression in E. coli and chromatographic purification are key steps towards a high resolution structure of MspA.

Feedback regulation in the lactose operon: a mathematical modeling study and comparison with experimental data

A mathematical model for the regulation of induction in the lac operon in Escherichia coli is presented. This model takes into account the dynamics of the permease facilitating the internalization of external lactose; internal lactose; beta-galactosidase, which is involved in the conversion of lactose to allolactose, glucose and galactose; the allolactose interactions with the lac repressor; and mRNA. The final model consists of five nonlinear differential delay equations with delays due to the transcription and translation process. We have paid particular attention to the estimation of the parameters in the model. We have tested our model against two sets of beta-galactosidase activity versus time data, as well as a set of data on beta-galactosidase activity during periodic phosphate feeding. In all three cases we find excellent agreement between the data and the model predictions. Analytical and numerical studies also indicate that for physiologically realistic values of the external lactose and the bacterial growth rate, a regime exists where there may be bistable steady-state behavior, and that this corresponds to a cusp bifurcation in the model dynamics.

Growth at low temperature suppresses readthrough of the UGA stop codon during the expression of Bacillus subtilis flgM gene in Escherichia coli

The efficient production of recombinant proteins in Escherichia coli requires a proper termination of translation to ensure the synthesis of only the desired product. During the recombinant production of Bacillus subtilis flgM in E. coli, we detected an additional polypeptide of molecular mass higher than the expected, corresponding to a product of a translational readthrough of the UGA stop codon. In this paper we show that the readthrough was abolished when the synthesis of the recombinant protein was carried out at 25 degrees C. The possible causes that contribute to reduce the proportion of readthrough protein species against the correct terminated product are discussed.

Hepatitis B virus core antigen: enhancement of its production in Escherichia coli, and interaction of the core particles with the viral surface antigen

The core antigen (HBcAg) of hepatitis B Virus (HBV) can be expressed in Escherichia coil where it assembles into icosahedral particles containing 240 or 180 subunits. Analysis of the two kinds of particles by SDS-polyacrylamide gel electrophoresis (SDS-PAGE) showed that a substantial proportion of their subunits were smaller than the full-length HBcAg monomer and of variable size, but all had the same N-terminal sequence showing that the smaller species were heterogeneous in their arginine-rich C-terminal regions. Around 50% of these arginine residues are encoded by the triplet AGA which is rare in E. coli. Supplementation of the level of AGA tRNA in the cell by transformation with plasmids expressing the T4 AGA tRNA gene significantly enhanced the yield of HBcAg. Fusion phage carrying a ligand specific for HBcAg showed no significant difference in the affinity for the two sizes of HBcAg particles, but in similar reactions in solution HBV surface antigen exhibited differential affinities for the same two HBcAg preparations.

Periplasmic secretion of native ovalbumin without signal cleavage in Escherichia coli

In Escherichia coli cells carrying wild-type ovalbumin cDNA, some of the recombinant protein was secreted into the periplasmic space. In contrast, a signal-region mutant form of ovalbumin (deletion, Gly1 to Ala39) was not detected in the periplasm despite being synthesized at the same level as the wild-type protein. Chemical and spectroscopic analyses showed that periplasmic ovalbumin assumes a conformation indistinguishable from that of native egg white ovalbumin. We concluded that a process resembling the secretion of ovalbumin process in the oviduct occurs also in bacteria.

Effects of codon usage versus putative 5'-mRNA structure on the expression of Fusarium solani cutinase in the Escherichia coli cytoplasm

Matching the codon usage of recombinant genes to that of the expression host is a common strategy for increasing the expression of heterologous proteins in bacteria. However, while developing a cytoplasmic expression system for Fusarium solani cutinase in Escherichia coli, we found that altering codons to those preferred by E. coli led to significantly lower expression compared to the wild-type fungal gene, despite the presence of several rare E. coli codons in the fungal sequence. On the other hand, expression in the E. coli periplasm using a bacterial PhoA leader sequence resulted in high levels of expression for both the E. coli optimized and wild-type constructs. Sequence swapping experiments as well as calculations of predicted mRNA secondary structure provided support for the hypothesis that differential cytoplasmic expression of the E. coli optimized versus wild-type cutinase genes is due to differences in 5(') mRNA secondary structures. In particular, our results indicate that increased stability of 5(') mRNA secondary structures in the E. coli optimized transcript prevents efficient translation initiation in the absence of the phoA leader sequence. These results underscore the idea that potential 5(') mRNA secondary structures should be considered along with codon usage when designing a synthetic gene for high level expression in E. coli.

In vivo introduction of unpreferred synonymous codons into the Drosophila Adh gene results in reduced levels of ADH protein

The evolution of codon bias, the unequal usage of synonymous codons, is thought to be due to natural selection for the use of preferred codons that match the most abundant species of isoaccepting tRNA, resulting in increased translational efficiency and accuracy. We examined this hypothesis by introducing 1, 6, and 10 unpreferred codons into the Drosophila alcohol dehydrogenase gene (Adh). We observed a significant decrease in ADH protein production with number of unpreferred codons, confirming the importance of natural selection as a mechanism leading to codon bias. We then used this empirical relationship to estimate the selection coefficient (s) against unpreferred synonymous mutations and found the value (s >or= 10(-5)) to be approximately one order of magnitude greater than previous estimates from population genetics theory. The observed differences in protein production appear to be too large to be consistent with current estimates of the strength of selection on synonymous sites in D. melanogaster.

Cis control of gene expression in E.coli by ribosome queuing at an inefficient translational stop signal

An UGA stop codon context which is inefficient because of the 3'-flanking context and the last two amino acids in the gene protein product has a negative effect on gene expression, as shown using a model protein A' gene. This is particularly true at low mRNA levels, corresponding to a high intracellular ribosome/mRNA ratio. The negative effect is smaller if this ratio is decreased, or if the distance between the initiation and termination signals is increased. The results suggest that an inefficient termination codon can cause ribosomal pausing and queuing along the upstream mRNA region, thus blocking translation initiation of short genes. This cis control effect is dependent on the stop codon context, including the C-terminal amino acids in the gene product, the translation initiation signal strength, the ribosome/mRNA ratio and the size of the mRNA coding region. A large proportion of poorly expressed natural Escherichia coli genes are small, and the weak termination codon UGA is under-represented in small, highly expressed E.coli genes as compared with the efficient stop codon UAA.

Expression and one-step purification of a developmentally regulated protein from Dictyostelium discoideum

To overexpress Dictyostelium 5NT, a 1506bp fragment of the cDNA encoding the gene was cloned into a pET32c+ vector and expressed in the Escherichia coli expression host BL21-CodonPlus(DE3)-RIL by Isopropyl-beta-D-thiogalactoside (IPTG) induction. Maximum induction of insoluble recombinant protein was reached after incubation of the culture for 3h with 1.0mM IPTG. High level of 5NT expression was confirmed by SDS-PAGE and immunoblotting analysis. The recombinant 5NT was purified to homogeneity by a one-step purification using continuous-elution electrophoresis. Ten mg recombinant 5NT was purified per liter of growth medium. To achieve one of the goals of this study, polyclonal antibody against the recombinant 5NT was produced in a rabbit. We have shown previously by Northern blot and reporter gene analyses that 5nt is developmentally regulated. In this report, we used polyclonal antibody against the recombinant protein in Western blot analysis of membrane protein extracts from different developmental stages of Dictyostelium. The 5NT protein levels were first detected at the tight aggregation stage of development. Thus, there is no significant delay between transcription and translation of 5nt.

An insight into the possible mechanism of working of two-cistronic gene expression systems and rational designing of newer systems

The initial attempts at hyper-expressing buffalo/goat growth hormone (GH)-ORFs in Escherichia coli directly under various strong promoters were not successful despite the presence of a functional gene. High level expression of GH was achieved as a fusion protein with glutathione-S-transferase (GST). To produce native GH in an unfused state, we adapted an established strategy of two-cistronic approach in our system. In this strategy, utilizing one of the highly efficient reported sequences as the first cistron led to a nearly 1000-fold enhancement in the level of expression under an E. coli promoter (trc). In search of a newer first-cistron sequence as well as to see the generality of the two-cistronic approach, we explored the ability of different lengths of a highly expressing natural gene to act as an efficient first cistron. Surprisingly, GST, which is naturally highly expressible in E. coli, could not be fitted into a successful two-cistronic construct. In addition, placement of the entire two-cistronic expression cassette (which had earlier given high-level GH expression under trc promoter) under the T7 promoter in E. coli failed to hyper-express GH. These results suggest that the successful exploitation of the two-cistron arrangement for hyper-expression of eukaryotic ORFs in bacteria is not as straightforward as was previously thought. It appears probable that factors such as the sequence context, together with the length and codons used in the first cistron are important as well.

Investigating the relationship between genome structure, composition, and ecology in prokaryotes

Our thesis is that the DNA composition and structure of genomes are selected in part by mutation bias (GC pressure) and in part by ecology. To illustrate this point, we compare and contrast the oligonucleotide composition and the mosaic structure in 36 complete genomes and in 27 long genomic sequences from archaea and eubacteria. We report the following findings (1) High-GC-content genomes show a large underrepresentation of short distances between G(n) and C(n) homopolymers with respect to distances between A(n) and T(n) homopolymers; we discuss selection versus mutation bias hypotheses. (2) The oligonucleotide compositions of the genomes of Neisseria (meningitidis and gonorrhoea), Helicobacter pylori and Rhodobacter capsulatus are more biased than the other sequenced genomes. (3) The genomes of free-living species or nonchronic pathogens show more mosaic-like structure than genomes of chronic pathogens or intracellular symbionts. (4) Genome mosaicity of intracellular parasites has a maximum corresponding to the average gene length; in the genomes of free-living and nonchronic pathogens the maximum occurs at larger length scales. This suggests that free-living species can incorporate large pieces of DNA from the environment, whereas for intracellular parasites there are recombination events between homologous genes. We discuss the consequences in terms of evolution of genome size. (5) Intracellular symbionts and obligate pathogens show small, but not zero, amount of chromosome mosaicity, suggesting that recombination events occur in these species.

Production of monoclonal antibodies against chicken Pop1 (BVES)

Pop1 (BVES) is a member of the Popeye gene family which was named for its high-level expression in the heart and other muscle lineages. However, these proteins have no sequence similarity to any other gene family and their function is unclear. Here we report the production of recombinant chicken Pop1/BVES protein and the generation of two Pop1/BVES specific monoclonal antibodies (MAbs) 3F11-D9-E8 and 1B3-G11-A8. These antibodies detect recombinant Pop1/BVES in ELISAs and endogenous chicken Pop1/BVES in chicken heart extracts by Western blotting. Further, both 3F11-D9-E8 and 1B3-G11-A8 detect Pop1/BVES specifically in the cardiomyocytes of 8-day-old embryonic chicken hearts by immunofluorescence. These MAbs will be useful in immunolocalization and immunoblotting experiments of different tissue types to determine the location and levels of Pop1/BVES expression throughout development, as well as further analysis of the biochemistry of this protein.

The positive activator of cephamycin C and clavulanic acid production in Streptomyces clavuligerus is mistranslated in a bldA mutant

In Streptomyces coelicolor bldA encodes the principal leucyl tRNA for translation of UUA codons and controls pigmented antibiotic production by the presence of TTA codons in the genes encoding the pathway-specific activators of actinorhodin and undecylprodigiosin biosynthesis. In Streptomyces clavuligerus the gene encoding the pathway-specific activator of both cephamycin C and clavulanic acid production, ccaR, also contains a TTA codon and was expected to exhibit bldA-dependence. A cloned S. clavuligerus DNA fragment containing a sequence showing 91% identity to the S. coelicolor bldA-encoded tRNA was able to restore antibiotic production and sporulation to bldA mutants of S. coelicolor and the closely related Streptomyces lividans. A null mutation of the bldA gene in S. clavuligerus resulted in the expected sporulation defective phenotype, but unexpectedly had no effect on antibiotic production. Transcript analysis showed no difference in the levels of ccaR transcripts in the wild-type and bldA mutant strains, ruling out any effect of elevated levels of the ccaR mRNA. Furthermore, when compared to the wild-type strain, the bldA mutant showed no differences in the levels of CcaR, suggesting that the single TTA codon in ccaR is mistranslated efficiently. The role of codon context in bldA dependence is discussed.

Single synonymous codon substitution eliminates pausing during chloramphenicol acetyl transferase synthesis on Escherichia coli ribosomes in vitro

The coding sequence for chloramphenicol acetyl transferase (CAT) contains several rare codons; three of them are ATA encoding isoleucine in positions 13, 84 and 119 of the amino acid sequence. Expression of CAT on Escherichia coli ribosomes in vitro results in mostly full-length product but also distinct smaller polypeptides from less than 3 kDa to over 20 kDa. As reported earlier, the smaller polypeptides are the predominant products, if translation is initiated with fluorophore-Met-tRNA(f). All this translational pausing is eliminated when the first ATA codon is mutated to ATC, a frequently used codon for isoleucine in E. coli. Addition of large amounts of E. coli tRNA to the coupled transcription/translation reaction does not reduce the number of pause-site peptides seen in the expression of wild-type CAT. Thus we hypothesize that the mRNA structure may be an important determinant for translational pausing.

Codon-usage based regulation of colicin K synthesis by the stress alarmone ppGpp

The molecular mechanism of the upregulation of Escherichia coli colicin K (Cka) synthesis during stress conditions was studied. Nutrient starvation experiments and the use of relA spoT mutant strains, IPTG-regulated overproduction of ppGpp and lacZ fusions revealed that the stringent response alarmone guanosine 3',5'-bispyrophosphate (ppGpp) is the main positive effector of Cka synthesis. Comparison of the amounts of protein produced (Western blotting) and specific mRNA (Northern blotting) before and after nutrient starvation demonstrated increases in Cka protein with unaltered specific mRNA levels, suggesting a post-transcriptional regulatory mechanism. Reporter (beta-galactosidase) assays using truncated cka of variable length fused to lacZ located the key regulatory region close to the 5' end of the cka mRNA. Closer analysis of this region indicated the presence of several rare codons, including the leucine-encoding codon CUA. Synonymous exchange of the rare codons with more frequently used ones abolished the regulatory effect of ppGpp. Supplementation of the strain with the plasmid CodonPlus carrying several rare tRNA genes yielded similar results, indicating that codon usage (in particular, the fifth codon for the amino acid leucine) and tRNA availability (i.e. tRNAleu) are the key elements of the regulatory function of ppGpp. We conclude that ppGpp regulates Cka synthesis via a novel post-transcriptional mechanism that is based on rare codon usage and variable cognate tRNA availability.

Does recombination improve selection on codon usage? Lessons from nematode and fly complete genomes

Understanding the factors responsible for variations in mutation patterns and selection efficacy along chromosomes is a prerequisite for deciphering genome sequences. Population genetics models predict a positive correlation between the efficacy of selection at a given locus and the local rate of recombination because of Hill-Robertson effects. Codon usage is considered one of the most striking examples that support this prediction at the molecular level. In a wide range of species including Caenorhabditis elegans and Drosophila melanogaster, codon usage is essentially shaped by selection acting for translational efficiency. Codon usage bias correlates positively with recombination rate in Drosophila, apparently supporting the hypothesis that selection on codon usage is improved by recombination. Here we present an exhaustive analysis of codon usage in C. elegans and D. melanogaster complete genomes. We show that in both genomes there is a positive correlation between recombination rate and the frequency of optimal codons. However, we demonstrate that in both species, this effect is due to a mutational bias toward G and C bases in regions of high recombination rate, possibly as a direct consequence of the recombination process. The correlation between codon usage bias and recombination rate in these species appears to be essentially determined by recombination-dependent mutational patterns, rather than selective effects. This result highlights that it is necessary to take into account the mutagenic effect of recombination to understand the evolutionary role and impact of recombination.

Codon bias at the 3'-side of the initiation codon is correlated with translation initiation efficiency in Escherichia coli

The codon that follows the AUG initiation triplet (+2 codon) affects gene expression in Escherichia coli. We have extended this analysis using two model genes lacking any apparent Shine-Dalgarno sequence. Depending on the identity of the +2 codon a difference in gene expression up to 20-fold could be obtained. The effects did not correlate with the levels of intracellular pools of cognate tRNA for the +2 codon, with putative secondary mRNA structures, or with mRNA stability. However, most +2 iso-codons that were decoded by the same species of tRNA gave pairwise similar effects, suggesting that the effect on gene expression was associated with the decoding tRNA. High adenine content of the +2 codon was associated with high gene expression. Of the fourteen +2 codons that mediated the highest efficiency, all except two had an adenine as the first base of the codon. Analysis of the 3540 E. coli genes from the TransTerm database revealed that codons associated with high gene expression in the two expression systems are over-represented at the +2 position in natural genes. Codons that are associated with low gene expression are under-represented. The data suggest that evolution has favored codons at the +2 position that give high translation initiation.

A test of translational selection at 'silent' sites in the human genome: base composition comparisons in alternatively spliced genes

Natural selection appears to discriminate among synonymous codons to enhance translational efficiency in a wide range of prokaryotes and eukaryotes. Codon bias is strongly related to gene expression levels in these species. In addition, between-gene variation in silent DNA divergence is inversely correlated with codon bias. However, in mammals, between-gene comparisons are complicated by distinctive nucleotide-content bias (isochores) throughout the genome. In this study, we attempted to identify translational selection by analyzing the DNA sequences of alternatively spliced genes in humans and in Drosophila melanogaster. Among codons in an alternatively spliced gene, those in constitutively expressed exons are translated more often than those in alternatively spliced exons. Thus, translational selection should act more strongly to bias codon usage and reduce silent divergence in constitutive than in alternative exons. By controlling for regional forces affecting base-composition evolution, this within-gene comparison makes it possible to detect codon selection at synonymous sites in mammals. We found that GC-ending codons are more abundant in constitutive than alternatively spliced exons in both Drosophila and humans. Contrary to our expectation, however, silent DNA divergence between mammalian species is higher in constitutive than in alternative exons.

High-level periplasmic expression in Escherichia coli using a eukaryotic signal peptide: importance of codon usage at the 5' end of the coding sequence

We investigated the ability of signal peptides of eukaryotic origin (human, mouse, and yeast) to efficiently direct model proteins to the Escherichia coli periplasm. These were compared against a well-characterized prokaryotic signal peptide-OmpA. Surprisingly, eukaryotic signal peptides can work very efficiently in E. coli, but require optimization of codon usage by codon-based mutagenesis of the signal peptide coding region. Analysis of the 5' of periplasmic and cytoplasmic E. coli genes shows some codon usage differences.

FlbT, the post-transcriptional regulator of flagellin synthesis in Caulobacter crescentus, interacts with the 5' untranslated region of flagellin mRNA

Flagellar gene expression' in Caulobacter crescentus is regulated by a complex trans-acting hierarchy, in which the assembly of early structural proteins is required for the expression of later structural proteins. The flagellins that comprise the filament are regulated at both the transcriptional and the post-transcriptional levels. Post-transcriptional regulation is sensitive to the assembly of the flagellar basal body and hook structures. In mutant strains lacking these structures, flagellin genes are transcribed, but not translated. Mutations in the flagellar regulatory gene, flbT, restore flagellin translation in the absence of flagellar assembly. In this report, we investigate the mechanism of FlbT-mediated post-transcriptional regulation. We show that FlbT is associated with the 5' untranslated region (UTR) of fljK (25 kDa flagellin) mRNA and that this association requires a predicted loop structure in the transcript. Mutations within this loop abolished FlbT association and resulted in increased mRNA stability, indicating that FlbT promotes the degradation of flagellin mRNA by associating with the 5' UTR. We also assayed the effects on gene expression using mutant transcripts fused to lacZ. Interestingly, the mutant transcript that failed to associate with FlbT in vitro was still repressed in mutants defective in flagellum assembly, suggesting that other factors in addition to FlbT couple assembly to translation.

Mitochondrial gene therapy: an arena for the biomedical use of inteins

Mitochondrial DNA (mtDNA) mutations underlie many rare diseases and might also contribute to human ageing. Gene therapy is a tempting future possibility for intervening in mitochondriopathies. Expression of the 13 mtDNA-encoded proteins from nuclear transgenes (allotopic expression) might be the most effective gene-therapy strategy. Its only confirmed difficulty is the extreme hydrophobicity of these proteins, which prevents their import into mitochondria from the cytosol. Inteins (self-splicing 'protein introns') might offer a solution to this problem: their insertion into such transgenes could greatly reduce the encoded proteins' hydrophobicity, enabling import, with post-import excision restoring the natural amino acid sequence.

Use of modified BL21(DE3) Escherichia coli cells for high-level expression of recombinant peanut allergens affected by poor codon usage

We previously cloned a panel of peanut allergens by phage display technology. Examination of the codons used in these sequences indicated that most of the cDNAs contain an excess of the least used codons in Escherichia coli, namely AGG/AGA, that correspond to a minor tRNA, the product of the dnaY gene. To achieve high-level expression of the peanut allergens, the cDNAs were subcloned into an expression vector of the pET series (Novagen) in order to produce (His)(10)-tagged fusion proteins in conventional E. coli BL21(DE3) cells. The peanut allergens Ara h 1, Ara h 2, and Ara h 6 with an AGG/AGA codon content of 8-10% were only marginally expressed, whereas the peanut profilin Ara h 5, with an AGG/AGA codon content of only 0.8%, was efficiently expressed in these cells. Hence, by using modified BL21(DE3) E. coli cells, namely BL21-CodonPlus(DE3)-RIL cells (Stratagene) with extra copies of E. coli argU, ileY, and leuW tRNA genes, it was possible to attain high-level expression of the proteins affected by rare codon usage. IPTG-induced expression of several recombinant peanut allergens, such as Ara h 1, Ara h 2, and Ara h 6, was greatly increased in these special cells compared to the expression yield achieved by conventional E. coli hosts. The purification of the soluble and the insoluble fraction of Ara h 2 was performed by metal-affinity chromatography and yielded a total of about 30 mg (His)(10)-tagged recombinant protein per liter of culture of transformed BL21(DE3)CodonPlus-RIL cells. This is over 100 times more than achieved by production of Ara h 2 in conventional BL21(DE3) cells.

Expression of the Staphylococcus hyicus lipase in Lactococcus lactis

The extracellular Staphylococcus hyicus lipase was expressed under the control of different promoters in Lactococcus lactis and Bacillus subtilis. Its expression at high and moderate levels is toxic for the former and the latter hosts, respectively. In L. lactis, the lipase was expressed at a high level, up to 30% of the total cellular proteins, under the control of the inducible promoter PnisA. About 80% of the lipase remained associated with the cells. Close to half of this amount remained associated with the inner side of the cytoplasmic membrane as unprocessed pre-pro-lipase. The other half was trapped by the cell wall and partially degraded at the N-terminal end. This result suggests that extracellular proteases degrade the lipase. Surprisingly, the kinetics and the pattern of lipase degradation were different in the two L. lactis subspecies, L. lactis subsp. cremoris and L. lactis subsp. lactis. The extracellular proteolytic systems that degrade lipase are thus different in these closely related subspecies. The incorrect export of the lipase is not due to an inappropriate leader peptide but may be due to an inefficiency of several steps of lipase secretion. We propose that (i) the S. hyicus lipase may require a special accessory system to be correctly exported or (ii) the kinetics of lipase synthesis may be a critical factor for proper folding.

Ribosome traffic in E. coli and regulation of gene expression

The ribosome traffic during translation of E. coli coding sequences was simulated, assuming that the rate of translation of individual codons is limited by the cognate tRNA availability. Actual translation rates were taken from Solomovici et al. (J. theor. Biol. 185, 511-521, 1997). The mean translation rates of the 4271 sequences cover a broad, two-fold range, whereas the local rate of translation along messengers varies three-fold on average. The simulation allows one to sketch the ribosome traffic on the polysome, in particular by providing the extent of mRNA sequences uncovered between consecutive ribosomes and the time during which these sequences are exposed. These parameters may participate in the control of mRNA stability and transcriptional polarity. By averaging the translation rates in a 17-codon window, assumed to be the sequence covered by a translating ribosome, and sliding this window along a given coding sequence, the addresses KMAX and KMIN, and the times TMAX and TMIN of respectively the slowest and the fastest translated window were determined. It is shown that under the assumptions made, TMAX sets the number of proteins translated from a given mRNA molecule per unit time, in case the delay between consecutive translation starts is below TMAX. Both windows display two strong biases, one as expected on the usage of codon frequencies, and the other surprisingly on the occurrence of amino acids.

RNA secondary structure and compensatory evolution

The classic concept of epistatic fitness interactions between genes has been extended to study interactions within gene regions, especially between nucleotides that are important in maintaining pre-mRNA/mRNA secondary structures. It is shown that the majority of linkage disequilibria found within the Drosophila Adh gene are likely to be caused by epistatic selection operating on RNA secondary structures. A recently proposed method of RNA secondary structure prediction based on DNA sequence comparisons is reviewed and applied to several types of RNAs, including tRNA, rRNA, and mRNA. The patterns of covariation in these RNAs are analyzed based on Kimura's compensatory evolution model. The results suggest that this model describes the substitution process in the pairing regions (helices) of RNA secondary structures well when the helices are evolutionarily conserved and thermodynamically stable, but fails in some other cases. Epistatic selection maintaining pre-mRNA/mRNA secondary structures is compared to weak selective forces that determine features such as base composition and synonymous codon usage. The relationships among these forces and their relative strengths are addressed. Finally, our mutagenesis experiments using the Drosophila Adh locus are reviewed. These experiments analyze long-range compensatory interactions between the 5' and 3' ends of Adh mRNA, the different constraints on secondary structures in introns and exons, and the possible role of secondary structures in RNA splicing.

Effect of the codon following the ATG start site on the expression of ovine growth hormone in Escherichia coli

For expression of ovine growth hormone (OGH) in inclusion bodies without an affinity histidine tag at either end of the protein, three clones, differing only in the second codon following the ATG start site, were constructed. Their expression was studied by SDS-PAGE followed by immunoblotting. Clone Ala.OGH (clone 1), beginning with Met.Ala.Phe.Pro ellipsis, did not show any expression. Clone Phe.OGH (clone 3), beginning with Met.Phe.Pro ellipsis, gave very high levels of OGH expression following IPTG induction. However, in clone Gly.OGH (clone 2), in which the Ala codon was replaced with a Gly codon at the second position after the start site, a lower level of expression was obtained. Northern hybridization analysis showed that upon IPTG induction, OGH mRNA was transcribed from all three clones. These results therefore, imply that lack of expression in clone 1 and a lower level of expression in clone 2 are not due to a failure of transcription; however, they may be due to inefficient initiation of translation. The secondary structure analysis of mRNA predicts inaccessibility of different elements of the RBS in the case of Ala.OGH (clone 1). The present study highly underscores the importance of mRNA secondary structure at the start site in regulation of expression of a cloned gene in Escherichia coli, a prokaryotic expression system.

Membrane insertion kinetics of a protein domain in vivo. The bacterioopsin n terminus inserts co-translationally

The pathway by which segments of a polytopic membrane protein are inserted into the membrane has not been resolved in vivo. We have developed an in vivo kinetic assay to examine the insertion pathway of the polytopic protein bacterioopsin, the apoprotein of Halobacterium salinarum bacteriorhodopsin. Strains were constructed that express the bacteriorhodopsin mutants I4C:H(6) and T5C:H(6), which carry a unique Cys in the N-terminal extracellular domain and a polyhistidine tag at the C terminus. Translocation of the N-terminal domain was detected using a membrane-impermeant gel shift reagent to derivatize the Cys residue of nascent radiolabeled molecules. Derivatization was assessed by gel electrophoresis of the fully elongated radiolabeled population. The time required to translocate and fully derivatize the Cys residues of I4C:H(6) and T5C:H(6) is 46 +/- 9 and 61 +/- 6 s, respectively. This is significantly shorter than the elongation times of the proteins, which are 114 +/- 26 and 169 +/- 16 s, respectively. These results establish that translocation of the bacterioopsin N terminus and insertion of the first transmembrane segment occur co-translationally and confirm the use of the assay to monitor the kinetics of polytopic membrane protein insertion in vivo.

Codon optimization effect on translational efficiency of DNA vaccine in mammalian cells: analysis of plasmid DNA encoding a CTL epitope derived from microorganisms

Interspecific difference of codon usage is one of the major obstacles for effective induction of specific immune responses against bacteria and protozoa by DNA immunization. Using genes encoding major histocompatibility complex class I-restricted cytotoxic T-lymphocyte (CTL) epitopes, derived from an intracellular bacterium, Listeria monocytogenes and a mouse malaria parasite, Plasmodium yoelii, we report here that the codon optimization level of the genes is not precisely proportional to, but does correlate well with the translational efficiency in mammalian cells, which is concomitantly associated with the induction level of specific CTL response in the mouse. These results suggest that DNA immunization using the gene codon-optimized to mammals through the entire region is very effective.

Initiation of translation in prokaryotes and eukaryotes

The mechanisms whereby ribosomes engage a messenger RNA and select the start site for translation differ between prokaryotes and eukaryotes. Initiation sites in polycistronic prokaryotic mRNAs are usually selected via base pairing with ribosomal RNA. That straightforward mechanism is made complicated and interesting by cis- and trans-acting elements employed to regulate translation. Initiation sites in eukaryotic mRNAs are reached via a scanning mechanism which predicts that translation should start at the AUG codon nearest the 5' end of the mRNA. Interest has focused on mechanisms that occasionally allow escape from this first-AUG rule. With natural mRNAs, three escape mechanisms - context-dependent leaky scanning, reinitiation, and possibly direct internal initiation - allow access to AUG codons which, although not first, are still close to the 5' end of the mRNA. This constraint on the initiation step of translation in eukaryotes dictates the location of transcriptional promoters and may have contributed to the evolution of splicing.The binding of Met-tRNA to ribosomes is mediated by a GTP-binding protein in both prokaryotes and eukaryotes, but the more complex structure of the eukaryotic factor (eIF-2) and its association with other proteins underlie some aspects of initiation unique to eukaryotes. Modulation of GTP hydrolysis by eIF-2 is important during the scanning phase of initiation, while modulating the release of GDP from eIF-2 is a key mechanism for regulating translation in eukaryotes. Our understanding of how some other protein factors participate in the initiation phase of translation is in flux. Genetic tests suggest that some proteins conventionally counted as eukaryotic initiation factors may not be required for translation, while other tests have uncovered interesting new candidates. Some popular ideas about the initiation pathway are predicated on static interactions between isolated factors and mRNA. The need for functional testing of these complexes is discussed. Interspersed with these theoretical topics are some practical points concerning the interpretation of cDNA sequences and the use of in vitro translation systems. Some human diseases resulting from defects in the initiation step of translation are also discussed.

The positive relationship between codon usage bias and translation initiation AUG context in Saccharomyces cerevisiae

The relationship between the codon usage bias and the sequence context surrounding the AUG translation initiation codon was examined in 211 Saccharomyces cerevisiae mRNA sequences. The codon usage bias and the number of matches to optimal AUG context, (A/U)A(A/C)AA(A/C)AUGUC(U/C), for translation initiation showed a positive relationship, indicating that these two factors are evolutionally under the similar natural selection constraint at the translation level. A new index (AUGCAI = AUG Context Adaptation Index) for the measure of optimal AUG context was devised, and the importance of each position of AUG context was also examined.

The effect of a hydrophobic N-terminal probe on translational pausing of chloramphenicol acetyl transferase and rhodanese

The effect on translational pausing of a hydrophobic probe, coumarin, at the N terminus of nascent peptides was investigated. Two different proteins, bacterial chloramphenicol acetyltransferase and bovine rhodanese, were synthesized by coupled transcription/translation in a cell-free system derived from Escherichia coli. Protein synthesis was initiated with N-formyl-Met-tRNAf or N-acetyl-S-coumarin-Met-tRNAf. Cotranslational incorporation of the coumarin derivative generated nascent polypeptides with a hydrophobic residue at their N termini. The effect of the two N-terminal groups on the size distribution and quantity of the peptides formed by translational pausing was investigated. The N-terminal coumarin caused an accumulation of nascent chloramphenicol acetyltransferase peptides in the mass range of 3.5-4.0 kDa that reflects a delay in translation at this point. No similar effect on rhodanese pause-site peptides was observed. This effect on translational pausing cannot be explained by either mRNA secondary structure or rare codons and tRNA abundance. It is suggested that the effect of N-terminal coumarin on translational pausing is the result of the interaction of the nascent peptide with components of the large ribosomal subunit along the path it follows between the peptidyl transferase center and the exit site on the distal surface.

BirA enzyme: production and application in the study of membrane receptor-ligand interactions by site-specific biotinylation

The enzyme BirA is a key reagent because of its ability to biotinylate proteins at a specific residue in a recognition sequence. We report a rapid, efficient, and economical method for the production, purification, and application of this enzyme. The method is easily scaled up and the protein produced is of high purity and can be stored for many months with retention of activity. We have used this enzyme to biotinylate the C termini of membrane proteins, allowing these proteins to be tetramerized by binding to streptavidin. Because of the specificity of the biotinylation at the C terminus, the orientation of the membrane proteins on the streptavidin is equivalent to that of the native protein on the cell surface. These tetrameric proteins can be used to study protein receptor-ligand interactions at the cell surface, and site-specific biotinylation can be used to study proteins in vitro using a defined orientation.

The modification of the wobble base of tRNAGlu modulates the translation rate of glutamic acid codons in vivo

In Escherichia coli, uridine in the wobble position of tRNAGlu and tRNALys is modified to mnm5s2U34. This modification is believed to restrict the base-pairing capability, i.e. to prevent misreading of near-cognate codons and reduce the efficiency of cognate codon reading, especially of codons ending in G. We have determined the influence of the 5-methylaminomethyl and the 2-thio modifications of mnm5s2U34 in tRNAGlu on the translation rate of the glutamate codons GAA and GAG in vivo. In wild-type cells, GAG is translated slower (7. 7 codons/second) and GAA faster (18 codons/second) than the average codon (13 codons/second). Surprisingly, tRNAGlu lacking the 5-methylaminomethyl group, thus containing s2U34, translated GAA twofold faster (47 codons/second) and GAG fourfold slower (1.9 codons/second) than fully modified tRNAGlu. In contrast, tRNAGlu that contains mnm5U34 instead of mnm5s2U34 translated GAA fourfold slower (4.5 codons/second) and GAG only 20% slower (6.2 codons/second). Clearly, the 5-methylaminomethyl group of mnm5s2U34 facilitates base-pairing with G while decreasing base-pairing with A, resulting in rates of translation of GAG and GAA that approach that of the average codon. The 2-thio group increases the recognition of GAA and has only a minor effect on the decoding of GAG. Furthermore, the 2-thio group is important for aminoacylation (see the accompanying paper). These data imply that the function of mnm5s2U34 may be different from what has been suggested previously.

Aminoacylation of hypomodified tRNAGlu in vivo

The highly specific interaction of each aminoacyl-tRNA synthetase and its substrate tRNAs constitutes an intriguing problem in protein-RNA recognition. All tRNAs have the same overall three-dimensional structure in order to fit interchangeably into the translational apparatus. Thus, the recognition by aminoacyl-tRNA synthetase must be more or less limited to discrimination between bases at specific positions within the tRNA. The hypermodified nucleotide 5-methylaminomethyl-2-thiouridine (mnm5s2U) present at the wobble position of bacterial tRNAs specific for glutamic acid, lysine and possibly glutamine has been shown to be important in the recognition of these tRNAs by their synthetases in vitro. Here, we have determined the aminoacylation level in vivo of tRNAGlu, tRNALys, and tRNA1GIn in Escherichia coli strains containing undermodified derivatives of mnm5s2U34. Lack of the 5-methylaminomethyl group did not reduce charging levels for any of the three tRNAs. Lack of the s2U34 modification caused a 40% reduction in the charging level of tRNAGlu. Charging of tRNALys and tRNA1Gln were less affected. There was no compensating regulation of expression of glutamyl-tRNA synthetase because the relative synthesis rate was the same in the wild-type and mutant strains. These results indicate that the mnm5U34 modification is not an important recognition element in vivo for the glutamyl-tRNA synthetase. In contrast, lack of the s2U34 modification reduced the efficiency of charging by at least 40%. This is the minimal estimate because the turn-over rate of Glu-tRNAGlu was also reduced in the absence of the 2-thio group. Lack of either modification did not affect mischarging or mistranslation.

Silent mutations in the Escherichia coli ompA leader peptide region strongly affect transcription and translation in vivo

In order to test the effect of silent mutations on the regulation of gene expression, we monitored several steps of transcription and translation of the ompA gene in vivo , in which some or all codons between codons 6 and 14, frequently used in Escherichia coli , had been exchanged for infrequent synonymous codons. Northern blot analysis revealed an up to 4-fold reduction in the half-life of the mutated messengers and a >10-fold reduction in their steady-state amounts. Western blot analysis showed a 10-fold reduction in the amount of OmpA protein. Use of a system expressing a Rho-specific anti-terminator allowed us to detect a strong transcription polarity effect in the silent mutants. These results demonstrate that silent mutations can severely inhibit several steps of gene expression in E. coli and that code degeneracy is efficiently exploited in this species for setting signals for gene control and regulation.

The relationship between synonymous codon usage and protein structure

The hypothesis that synonymous codon usage is related to protein three-dimensional structure is examined by investigating the correlation between synonymous codon usage and protein secondary structure. All except two codons in E. coli show the same secondary structural preference for alpha-helix, beta-strand or coil as that of amino acids to be encoded by the respective codons, while 17 codons show secondary structural bias in mammalian proteins. The results indicate that there is no significant correlation between synonymous codon usage and protein secondary structure in E. coli, but there is a correlation in mammals. It could be deduced that synonymous codons carry much less structural information in prokaryotes than in eukaryotes due to their divergent evolutionary mechanism.

Ribosomal protein S1 is required for translation of most, if not all, natural mRNAs in Escherichia coli in vivo

We have deleted the chromosomal rpsA gene, encoding ribosomal protein S1, from an Escherichia coli strain carrying a plasmid where rpsA was controlled by the lac promoter and operator. This exogenous source of protein S1 was essential for growth. Thus we have verified the absolute requirement for protein S1. To see if translation of individual mRNAs differed in the requirements for protein S1, we removed the inducer and followed the time-course of the synthesis of several individual proteins and of total RNA, DNA and protein. Growth immediately shifted from being exponential to being linear, with a rate of protein synthesis defined by the pre-existing amount of protein S1. The expression pattern of the individual proteins indicated that the translation of all mRNAs was dependent on protein S1. Unexpectedly, we found that depletion for protein S1 for extended periods introduced a starvation for amino acids. Such starvation was indicated by an increased synthesis of ppGpp and could be reversed by addition of a mixture of all 20 amino acids. Measurements of the peptide chain elongation rate in vivo showed that ribosomes without protein S1 were unable to interfere with the peptide chain elongation rate of the active ribosomes and that, therefore, protein S1 was unable to diffuse from one ribosome to another during translation. We conclude that protein S1-deficient ribosomes are totally inactive in peptide chain elongation on most, if not all, naturally occurring E. coli mRNAs.

How optimized is the translational machinery in Escherichia coli, Salmonella typhimurium and Saccharomyces cerevisiae?

The optimization of the translational machinery in cells requires the mutual adaptation of codon usage and tRNA concentration, and the adaptation of tRNA concentration to amino acid usage. Two predictions were derived based on a simple deterministic model of translation which assumes that elongation of the peptide chain is rate-limiting. The highest translational efficiency is achieved when the codon recognized by the most abundant tRNA reaches the maximum frequency. For each codon family, the tRNA concentration is optimally adapted to codon usage when the concentration of different tRNA species matches the square-root of the frequency of their corresponding synonymous codons. When tRNA concentration and codon usage are well adapted to each other, the optimal content of all tRNA species carrying the same amino acid should match the square-root of the frequency of the amino acid. These predictions are examined against empirical data from Escherichia coli, Salmonella typhimurium, and Saccharomyces cerevisiae.

Expression of catalytically active human cytochrome p450scc in Escherichia coli and mutagenesis of isoleucine-462

Cytochrome P450scc (P450scc) catalyzes the first step in steroid hormone synthesis, the conversion of cholesterol to pregnenolone. Human P450scc has been poorly studied due to the difficulty of purifying reasonable quantities of enzyme from human tissue. To provide a more convenient source of the human enzyme and to enable structure-function studies to be done using site-directed mutagenesis, we expressed the mature form of human P450scc in Escherichia coli. The expression system enabled us to produce larger quantities of active cytochrome than have previously been isolated from placental mitochondria. The expressed P450scc was purified to near homogeneity and shown to have catalytic properties comparable to the enzyme purified from the human placenta. The mature form of human adrenodoxin was also expressed in E. coli and supported cholesterol side chain cleavage activity with the same Vmax as that observed using bovine adrenodoxin but with a higher Km. Mutation of Ile-462 to Leu in human P450scc caused a decrease in the catalytic rate constant (kcat) with cholesterol as substrate, increased the Km for 22R-hydroxycholesterol, but did not affect the kinetic constants for 20 alpha-hydroxycholesterol. This suggests that Ile-462 lies close to the side chain binding site and that the side chains of cholesterol, 22R-hydroxycholesterol, and 20 alpha-hydroxycholesterol occupy slightly different positions in the active site.

Purification, characterization, and crystallization of an N-hydroxyarylamine O-acetyltransferase from Salmonella typhimurium

The N-hydroxyarylamine O-acetyltransferase from Salmonella typhimurium has been expressed as a histidine-tagged fusion protein in Escherichia coli and purified to apparent homogeneity using single-step immobilized metal ion chromatography. Sufficient quantities of the purified protein have been obtained to allow its characterization by physical methods including dynamic light scattering and electrospray mass spectrometry. The substrate specificity and temperature sensitivity of the enzymatic activity have also been assessed. The enzyme has been crystallized from sodium, potassium tartrate and X-ray diffraction data have been obtained to allow the identification of an orthorhombic unit cell, point group P21212, with dimensions a = 137 A, b = 223 A, and c = 105 A. These crystals will provide a route to a crystallographic determination of the structure of the protein.

Antisense inhibition of gene expression in bacteria by PNA targeted to mRNA

Peptide nucleic acid (PNA) is a DNA mimic with attractive properties for developing improved gene-targeted antisense agents. To test this potential of PNA in bacteria, PNAs were designed to target the start codon regions of the Escherichia coli beta-galactosidase and beta-lactamase genes. Dose-dependent and specific gene inhibition was observed in vitro using low nanomolar PNA concentrations and in vivo using low micromolar concentrations. Inhibition was more efficient for a permeable E. coli strain relative to wild-type K-12. The potency of the anti-beta-lactamase PNAs was abolished by a six base substitution, and inhibition could be re-established using a PNA with compensating base changes. Antisense inhibition of the beta-lactamase gene was sufficient to sensitize resistant cells to the antibiotic ampicillin. The results demonstrate gene- and sequence-specific antisense inhibition in E. coli and open possibilities for antisense antibacterial drugs and gene function analyses in bacteria.

Inhibition of translation and bacterial growth by peptide nucleic acid targeted to ribosomal RNA

Peptide nucleic acid (PNA) is a DNA mimic that has shown considerable promise as a lead compound for developing gene therapeutic drugs. We report that PNAs targeted to functional and accessible sites in ribosomal RNA can inhibit translation in an Escherichia coli cell-free transcription/translation system, with 50% reductions caused by nanomolar PNA concentrations. The effect in vitro is quantitatively similar to that of the known translation inhibitor and antibiotic tetracycline. Also, the targeted PNAs inhibited bacterial growth on agar plates and in liquid culture. A strain of E. coli (AS19) that is more permeable to antibiotics was approximately 10-fold more sensitive to the active PNAs, suggesting that the effect on growth indeed was caused by PNAs that entered cells. Inhibition was not observed when using control PNAs of similar composition but with an unrelated or mismatched sequence. The results demonstrate that ribosomal RNA is a possible target for sequence-designed novel antibiotics based on DNA analogues or mimics.

The efficiency of a cis-cleaving ribozyme in an mRNA coding region is influenced by the translating ribosome in vivo

A cis -cleaving hammerhead ribozyme (Rz) expression system (3A'-Rz) in Escherichia coli has been constructed that can be used to study the involvement of factors that affect ribozyme cleavage in vivo . The ribozyme sequence is placed in the coding region of 3A' mRNA, which is expressed from a semi-synthetic translation assay gene. The size and the 5'-end sequences of the 3' cleavage fragments were determined and the efficiencies of different Rz variants were measured by quantitative primer extension. It is shown that one of the semi-active constructs (3A'-RzIII) can be used as an indicator for ribosomes that read through or terminate at a stop codon upstream of the Rz hammerhead sequence in the mRNA. Readthrough of the stop codon in an uncleaved mRNA gives a full length 3A' protein. Termination at the stop codon upstream of the ribozyme sequence gives a shortened termination product. However, the mRNA fragment that should arise as a result of the auto-cleavage does not give rise to any detectable corresponding truncated protein. Besides studies on translating ribosomes, the 3A'-Rz system can be used to isolate mutant strains that are changed in ribozyme activity either from internal base alterations, or changed interacting host factors.