Codon usage determines translation rate in Escherichia coli

Pseudouridylation of helix 69 of 23S rRNA is necessary for an effective translation termination

Escherichia coli strains with inactivated rluD genes were previously found to lack the conserved pseudouridines in helix 69 of 23S ribosomal RNA and to grow slowly. A suppressor mutant was isolated with a near normal growth rate that had changed the conserved Glu-172 codon to a Lys codon in prfB, encoding translation termination factor RF2. When nonsense suppression in strains with all combinations of prfB(+)/prfB(E172K) and rluD(+)/rluD::cat was analyzed, misreading of all three stop codons as sense codons was found to be increased by rluD inactivation: Nonsense suppression was increased 2-fold at UAG codons, 9-fold at UAA, and 14-fold at UGA. The increased read-through at UGA corresponds to reading UGA as a sense codon in 30% of the cases. In contrast, the accuracy of reading sense codons appeared unaffected by loss of rluD. When the inactivated rluD gene was combined with the altered prfB, wild-type levels of termination were restored at UAA codons and termination was more efficient than wild type at UGA. These results strongly suggest that at least one of the helix 69 pseudouridines has a function in translation termination. To our knowledge, this is the first described function for a ribosomal RNA pseudouridine modification.

ERCC1 Genotype and Phenotype in Epithelial Ovarian Cancer Identify Patients Likely to Benefit From Paclitaxel Treatment in Addition to Platinum-Based Therapy

Purpose

To investigate the effect of excision repair cross-complementation group 1 (ERCC1) on treatment response and survival of patients treated with platinum chemotherapy with or without paclitaxel.

Patients and Methods

Tumor samples from epithelial ovarian cancer patients were evaluated for ERCC1 mRNA expression and a single nucleotide polymorphism at codon 118 (C>T). Of 178 patients treated with postoperative platinum-based chemotherapy, 75 were also given paclitaxel. For all of these patients, ERCC1 expression and genotype were analyzed for associations with treatment response and survival.

Results

Among the 103 patients treated with platinum without paclitaxel, the C/C genotype, compared with C/T and T/T, was associated with greater risk of disease progression and death (hazard ratio [HR], 1.95, P = .051; HR, 2.01, P = .033, respectively); high levels of ERCC1 mRNA, compared with low levels, were associated with greater risk of disease progression (HR, 2.41; P = .014). Similarly, when the ERCC1 data were combined, patients with the C/C genotype and high ERCC1 expression had greater risk for disease progression (HR, 3.73; P = .003) compared with those with low expression and non-C/C genotype. However, for the 75 patients treated with platinum plus paclitaxel, the C/C genotype and high ERCC1 expression were not associated with poor prognosis, suggesting that paclitaxel may help to alleviate ERCC1-related platinum resistance.

Conclusion

Ovarian cancer patients with high ERCC1 expression or the C/C genotype at codon 118 may benefit from the combination of platinum and paclitaxel, while those with low ERCC1 expression or the C/T or T/T genotype may respond well to platinum without paclitaxel.

Inhomogeneous exclusion processes with extended objects: the effect of defect locations

We study the effects of local inhomogeneities, i.e., slow sites of hopping rate q<1, in a totally asymmetric simple exclusion process for particles of size l>or=1 (in units of the lattice spacing). We compare the simulation results of l=1 and l>1 and notice that the existence of local defects has qualitatively similar effects on the steady state. We focus on the stationary current as well as the density profiles. If there is only a single slow site in the system, we observe a significant dependence of the current on the location of the slow site for both l=1 and l>1 cases. When two slow sites are introduced, more intriguing phenomena emerge, e.g., dramatic decreases in the current when the two are close together. In addition, we study the asymptotic behavior when q-->0. We also explore the associated density profiles and compare our findings to an earlier study using a simple mean-field theory. We then outline the biological significance of these effects.

Halting a cellular production line: responses to ribosomal pausing during translation

Cellular protein synthesis is a complex polymerization process carried out by multiple ribosomes translating individual mRNAs. The process must be responsive to rapidly changing conditions in the cell that could cause ribosomal pausing and queuing. In some circumstances, pausing of a bacterial ribosome can trigger translational abandonment via the process of trans-translation, mediated by tmRNA (transfer-messenger RNA) and endonucleases. Together, these factors release the ribosome from the mRNA and target the incomplete polypeptide for destruction. In eukaryotes, ribosomal pausing can initiate an analogous process carried out by the Dom34p and Hbs1p proteins, which trigger endonucleolytic attack of the mRNA, a process termed mRNA no-go decay. However, ribosomal pausing can also be employed for regulatory purposes, and controlled translational delays are used to help co-translational folding of the nascent polypeptide on the ribosome, as well as a tactic to delay translation of a protein while its encoding mRNA is being localized within the cell. However, other responses to pausing trigger ribosomal frameshift events. Recent discoveries are thus revealing a wide variety of mechanisms used to respond to translational pausing and thus regulate the flow of ribosomal traffic on the mRNA population.

Integrative analysis of transcriptomic and proteomic data: challenges, solutions and applications

Recent advances in high-throughput technologies enable quantitative monitoring of the abundance of various biological molecules and allow determination of their variation between biological states on a genomic scale. Two popular platforms are DNA microarrays that measure messenger RNA transcript levels, and gel-free proteomic analyses that quantify protein abundance. Obviously, no single approach can fully unravel the complexities of fundamental biology and it is equally clear that integrative analysis of multiple levels of gene expression would be valuable in this endeavor. However, most integrative transcriptomic and proteomic studies have thus far either failed to find a correlation or only observed a weak correlation. In addition to various biological factors, it is suggested that the poor correlation could be quite possibly due to the inadequacy of available statistical tools to compensate for biases in the data collection methodologies. To address this issue, attempts have recently been made to systematically investigate the correlation patterns between transcriptomic and proteomic datasets, and to develop sophisticated statistical tools to improve the chances of capturing a relationship. The goal of these efforts is to enhance understanding of the relationship between transcriptomes and proteomes so that integrative analyses may be utilized to reveal new biological insights that are not accessible through one-dimensional datasets. In this review, we outline some of the challenges associated with integrative analyses and present some preliminary statistical solutions. In addition, some new applications of integrated transcriptomic and proteomic analysis to the investigation of post-transcriptional regulation are also discussed.

Ribosome kinetics and aa-tRNA competition determine rate and fidelity of peptide synthesis

It is generally accepted that the translation rate depends on the availability of cognate aa-tRNAs. In this study it is shown that the key factor that determines translation rate is the competition between near-cognate and cognate aa-tRNAs. The transport mechanism in the cytoplasm is diffusion, thus the competition between cognate, near-cognate and non-cognate aa-tRNAs to bind to the ribosome is a stochastic process. Two competition measures are introduced; C(i) and R(i) (i=1, 64) are quotients of the arrival frequencies of near-cognates vs. cognates and non-cognates vs. cognates, respectively. Furthermore, the reaction rates of bound cognates differ from those of bound near-cognates. If a near-cognate aa-tRNA binds to the A site of the ribosome, it may be rejected at the anti-codon recognition step or proofreading step or it may be accepted. Regardless of its fate, the near-cognates and non-cognates have caused delays of varying duration to the observed rate of translation. Rate constants have been measured at a temperature of 20 degrees C by (Gromadski, K.B., Rodnina, M.V., 2004. Kinetic determinants of high-fidelity tRNA discrimination on the ribosome. Mol. Cell 13, 191-200). These rate constants have been re-evaluated at 37 degrees C, using experimental data at 24.5 degrees C and 37 degrees C (Varenne, S., et al., 1984. Translation in a non-uniform process: effect of tRNA availability on the rate of elongation of nascent polypeptide chains. J. Mol. Biol. 180, 549-576). The key results of the study are: (i) the average time (at 37 degrees C) to add an amino acid, as defined by the ith codon, to the nascent peptide chain is: tau(i)=9.06+1.445x[10.48C(i)+0.5R(i)] (in ms); (ii) the misreading frequency is directly proportional to the near-cognate competition, E(i)=0.0009C(i); (iii) the competition from near-cognates, and not the availability of cognate aa-tRNAs, is the most important factor that determines the translation rate - the four codons with highest near-cognate competition (in the case of E. coli) are [GCC]>[CGG]>[AGG]>[GGA], which overlap only partially with the rarest codons: [AGG]<[CCA]<[GCC]<[CAC]; (iv) based on the kinetic rates at 37 degrees C, the average time to insert a cognate amino acid is 9.06ms and the average delay to process a near-cognate aa-tRNA is 10.45ms and (vii) the model also provides estimates of the vacancy times of the A site of the ribosome - an important factor in frameshifting.

Posttranscriptional expression regulation: what determines translation rates?

Recent analyses indicate that differences in protein concentrations are only 20%–40% attributable to variable mRNA levels, underlining the importance of posttranscriptional regulation. Generally, protein concentrations depend on the translation rate (which is proportional to the translational activity, TA) and the degradation rate. By integrating 12 publicly available large-scale datasets and additional database information of the yeast Saccharomyces cerevisiae, we systematically analyzed five factors contributing to TA: mRNA concentration, ribosome density, ribosome occupancy, the codon adaptation index, and a newly developed “tRNA adaptation index.” Our analysis of the functional relationship between the TA and measured protein concentrations suggests that the TA follows Michaelis–Menten kinetics. The calculated TA, together with measured protein concentrations, allowed us to estimate degradation rates for 4,125 proteins under standard conditions. A significant correlation to recently published degradation rates supports our approach. Moreover, based on a newly developed scoring system, we identified and analyzed genes subjected to the posttranscriptional regulation mechanism, translation on demand. Next we applied these findings to publicly available data of protein and mRNA concentrations under four stress conditions. The integration of these measurements allowed us to compare the condition-specific responses at the posttranscriptional level. Our analysis of all 62 proteins that have been measured under all four conditions revealed proteins with very specific posttranscriptional stress response, in contrast to more generic responders, which were nonspecifically regulated under several conditions. The concept of specific and generic responders is known for transcriptional regulation. Here we show that it also holds true at the posttranscriptional level.

Large-scale mRNA concentration measurements are a hallmark of our post-genomic era. Usually they are taken as a surrogate for the corresponding protein concentrations. For most genes, proteins are the actual cellular players, but up to now it has been much more difficult to measure protein concentrations than mRNA concentrations. However, due to numerous posttranscriptional regulation mechanisms, mRNA levels only partly correlate with protein concentrations. Based on thoroughly composed reference datasets for protein and mRNA concentrations in yeast under standard growth conditions, we report the best corresponding correlation so far. We took into account additional factors, beyond mRNA concentrations, that influence protein levels in order to improve protein level predictions. Extending our previous approach, where ribosome occupancy and ribosome density were considered, we now also consider ORF-specific translation elongation rates. Different measures for elongation velocity were examined, and the codon adaptation index was found to be most appropriate. Moreover, saturation kinetics were introduced to better describe the translation process. The general findings were also applied to four stress conditions. Three new concepts, translation on demand, just-in-time translation, and general and specific posttranscriptional stress responders, are discussed.

Codon engineering for improved antibody expression in mammalian cells

While well established in bacterial hosts, the effect of coding sequence variation on protein expression in mammalian systems is poorly characterized outside of viral proteins or proteins from distant phylogenetic families. The potential impact is substantial given the extensive use of mammalian expression systems in research and manufacturing of protein biotherapeutics. We are studying the effect of codon engineering on expression of recombinant antibodies with an emphasis on developing manufacturing cell lines. CNTO 888, a human mAb specific for human MCP-1, was obtained by antibody phage display in collaboration with MorphoSys AG. The isolated DNA sequence of the antibody was biased towards bacterial codons, reflecting the engineering of the Fab library for phage display expression in Escherichia coli. We compared the expression of CNTO 888 containing the parental V-region sequences with two engineered coding variants. In the native codon exchanged (NCE) variant, the V-region codons were replaced with those used in naturally derived human antibody genes. In the human codon optimized (HCO) variant the V-region codons were those used at the highest frequency based on a human codon usage table. The antibody expression levels from stable transfections in mammalian host cells were measured. The HCO codon variant of CNTO 888 yielded the highest expressing cell lines and the highest average expression for the screened populations. This had a significant positive effect on the process to generate a CNTO 888 production cell line and indicates the potential to improve antibody expression in mammalian expression systems by codon engineering.

Codon usage and replicative strategies of hepatitis A virus

Hepatitis A virus (HAV), the prototype of genus Hepatovirus, has many biological characteristics that distinguish it from other members of the Picornaviridae family. Among these it is worth of note the need for an intact eIF4G factor for the initiation of translation and thus the inability to shut down host protein synthesis by a similar mechanism as in other picornaviruses. Consequently, HAV must inefficiently compete for the cellular translational machinery and this may explain its poor growth in cell culture. In this context of virus/cell competition HAV has strategically adopted a naturally highly deoptimized codon usage. Accordingly, a low protein synthesis may be expected with those proteins involved in RNA replication existing at limiting concentrations. Thus, a very low translation rate and a very low RNA replication rate may play a role in escaping to host cell defenses, allowing the virus to grow in a quiescent way. This could explain the high specific infectivity of HAV in spite of its naturally deoptimized codon usage, which would indicate non-abortive infections due to the antiviral cell response. Additionally, the deoptimized codon usage conveys in the use of abundant and rare codons. Many clusters of such rare codons are present in the capsid surface playing a seminal role in the highly cohesive stability of the HAV virion. Thus, the slow translation rate, resulting from the accumulation of rare codons, is likely to contribute to the highly stable viral capsid necessary for a prolonged survival outside the host body.

Selection on synthesis cost affects interprotein amino acid usage in all three domains of life

Most investigations of the forces shaping protein evolution have focused on protein function. However, cells are typically 50%-75% protein by dry weight, with protein expression levels distributed over five orders of magnitude. Cells may, therefore, be under considerable selection pressure to incorporate amino acids that are cheap to synthesize into proteins that are highly expressed. Such selection pressure has been demonstrated to alter amino acid usage in a few organisms, but whether "cost selection" is a general phenomenon remains unknown. One reason for this is that reliable protein expression level data is not available for most organisms. Accordingly, I have developed a new method for detecting cost selection. This method depends solely on interprotein gradients in amino acid usage. Applying it to an analysis of 43 whole genomes from all three domains of life, I show that selection on the synthesis cost of amino acids is a pervasive force in shaping the composition of proteins. Moreover, some amino acids have different price tags for different organisms--the cost of amino acids is changed for organisms living in hydrothermal vents compared with those living at the sea surface or for organisms that have difficulty acquiring elements such as nitrogen compared with those that do not--so I also investigated whether differences between organisms in amino acid usage might reflect differences in synthesis or acquisition costs. The results suggest that organisms evolve to alter amino acid usage in response to environmental conditions.

Correlation between mechanical strength of messenger RNA pseudoknots and ribosomal frameshifting

Programmed ribosomal frameshifting is often used by viral pathogens including HIV. Slippery sequences present in some mRNAs cause the ribosome to shift reading frame. The resulting protein is thus encoded by one reading frame upstream from the slippery sequence and by another reading frame downstream from the slippery sequence. Although the mechanism is not well understood, frameshifting is known to be stimulated by an mRNA structure such as a pseudoknot. Here, we show that the efficiency of frameshifting relates to the mechanical strength of the pseudoknot. Two pseudoknots derived from the Infectious Bronchitis Virus were used, differing by one base pair in the first stem. In Escherichia coli, these two pseudoknots caused frameshifting frequencies that differed by a factor of two. We used optical tweezers to unfold the pseudoknots. The pseudoknot giving rise to the highest degree of frameshifting required a nearly 2-fold larger unfolding force than the other. The observed energy difference cannot be accounted for by any existing model. We propose that the degree of ribosomal frameshifting is related to the mechanical strength of RNA pseudoknots. Our observations support the "9 A model" that predicts some physical barrier is needed to force the ribosome into the -1 frame. Also, our findings support the recent observation made by cryoelectron microscopy that mechanical interaction between a ribosome and a pseudoknot causes a deformation of the A-site tRNA. The result has implications for the understanding of genetic regulation, reading frame maintenance, tRNA movement, and unwinding of mRNA secondary structures by ribosomes.

5-Aminolevulinate production with recombinant Escherichia coli using a rare codon optimizer host strain

The 5-aminolevulinate (ALA) synthase gene (hemA) containing several codons rarely used by Escherichia coli was cloned from the genome of Rhodobacter sphaeroides and optimized in two strains of Escherichia coli: BL21(DE3) and Rosetta(DE3), which is a rare codon optimizer strain. The effects of initial isopropyl-beta-D: -thiogalactopyranoside (IPTG) concentration, induction time, and temperature on enzyme activity were studied and compared for two strains. The results indicated that the ALA synthase expressed by Rosetta(DE3)/pET-28a(+)-hemA was higher than that by BL21(DE3)/pET-28a(+)-hemA. The initial precursors, glycine and succinate, and initial glucose, which is an inhibitor for both ALA synthase and dehydratase, were observed to be the key factors affecting ALA production. ALA synthase activity was generally higher with Rosetta(DE3) than with BL21(DE3), so was ALA biosynthesis. Based on the optimal culture system using Rosetta(DE3), the yield of ALA achieved 3.8 g/l (29 mM) under the appropriate conditions in fermenter.

Experimental confirmation of a key role for non-optimal codons in protein export

Non-optimal codons are defined by low usage and low abundance of corresponding tRNA, and have an established role in translational pausing to allow the correct folding of proteins. Our previous work reported a striking abundance of non-optimal codons in the signal sequences of secretory proteins exported via the sec-dependent pathway in Escherichia coli. In the current study the signal sequence of maltose-binding protein (MBP) was altered so that non-optimal codons were substituted with the most optimal codon from their synonymous codon family. The expression of MBP from the optimized allele (malE-opt) was significantly less than wild-type malE. Expression of MBP from malE-opt was partially restored in a range of cytoplasmic and periplasmic protease deficient strains, confirming that reduced expression of MBP in malE-opt was due to its preferential degradation by cytoplasmic and periplasmic proteases. These data confirm a novel role for non-optimal codon usage in secretion by slowing the rate of translation across the N-terminal signal sequence to facilitate proper folding of the secreted protein.

Combinatorial expression vector engineering for tuning of recombinant protein production in Escherichia coli

The complex and integrated nature of both genetic and protein level factors influencing recombinant protein production in Escherichia coli makes it difficult to predict the optimal expression strategy for a given protein. Here, two combinatorial library strategies were evaluated for their capability of tuning recombinant protein production in the cytoplasm of E. coli. Large expression vector libraries were constructed through either conservative (ExLib1) or free (ExLib2) randomization of a seven-amino-acid window strategically located between a degenerated start codon and a sequence encoding a fluorescently tagged target protein. Flow cytometric sorting and analyses of libraries, subpopulations or individual clones were followed by SDS-PAGE, western blotting, mass spectrometry and DNA sequencing analyses. For ExLib1, intracellular accumulation of soluble protein was shown to be affected by codon specific effects at some positions of the common N-terminal extension. Interestingly, for ExLib2 where the same sequence window was randomized via seven consecutive NN(G/T) tri-nucleotide repeats, high product levels (up to 24-fold higher than a reference clone) were associated with a preferential appearance of novel SD-like sequences. Possible mechanisms behind the observed effects are discussed.

Error minimization explains the codon usage of highly expressed genes in Escherichia coli

Different organisms use synonymous codons with different preferences. Several measures have been introduced to compute the extent of codon usage bias within a gene or genome, among which the codon adaptation index (CAI) has been shown to be well correlated with mRNA levels of Escherichia coli. In this work an error adaptation index (eAI) is introduced, which estimates the level at which a gene can tolerate the effects of mistranslations. It is shown that the eAI has a strong correlation with CAI, as well as with mRNA levels, which suggests that the codons of highly expressed genes are selected so that mistranslation would have the minimum possible effect on the structure and function of the related proteins.

Codon usage and selection on proteins

Selection pressures on proteins are usually measured by comparing homologous nucleotide sequences (Zuckerkandl and Pauling 1965). Recently we introduced a novel method, termed volatility, to estimate selection pressures on proteins on the basis of their synonymous codon usage (Plotkin and Dushoff 2003; Plotkin et al. 2004). Here we provide a theoretical foundation for this approach. Under the Fisher-Wright model, we derive the expected frequencies of synonymous codons as a function of the strength of selection on amino acids, the mutation rate, and the effective population size. We analyze the conditions under which we can expect to draw inferences from biased codon usage, and we estimate the time scales required to establish and maintain such a signal. We find that synonymous codon usage can reliably distinguish between negative selection and neutrality only for organisms, such as some microbes, that experience large effective population sizes or periods of elevated mutation rates. The power of volatility to detect positive selection is also modest--requiring approximately 100 selected sites--but it depends less strongly on population size. We show that phenomena such as transient hyper-mutators can improve the power of volatility to detect selection, even when the neutral site heterozygosity is low. We also discuss several confounding factors, neglected by the Fisher-Wright model, that may limit the applicability of volatility in practice.

Correlation of mRNA expression and protein abundance affected by multiple sequence features related to translational efficiency in Desulfovibrio vulgaris: a quantitative analysis

The modest correlation between mRNA expression and protein abundance in large-scale data sets is explained in part by experimental challenges, such as technological limitations, and in part by fundamental biological factors in the transcription and translation processes. Among various factors affecting the mRNA-protein correlation, the roles of biological factors related to translation are poorly understood. In this study, using experimental mRNA expression and protein abundance data collected from Desulfovibrio vulgaris by DNA microarray and liquid chromatography coupled with tandem mass spectrometry (LC-MS/MS) proteomic analysis, we quantitatively examined the effects of several translational-efficiency-related sequence features on mRNA-protein correlation. Three classes of sequence features were investigated according to different translational stages: (i) initiation, Shine-Dalgarno sequences, start codon identity, and start codon context; (ii) elongation, codon usage and amino acid usage; and (iii) termination, stop codon identity and stop codon context. Surprisingly, although it is widely accepted that translation initiation is the rate-limiting step for translation, our results showed that the mRNA-protein correlation was affected the most by the features at elongation stages, i.e., codon usage and amino acid composition (5.3-15.7% and 5.8-11.9% of the total variation of mRNA-protein correlation, respectively), followed by stop codon context and the Shine-Dalgarno sequence (3.7-5.1% and 1.9-3.8%, respectively). Taken together, all sequence features contributed to 15.2-26.2% of the total variation of mRNA-protein correlation. This study provides the first comprehensive quantitative analysis of the mRNA-protein correlation in bacterial D. vulgaris and adds new insights into the relative importance of various sequence features in prokaryotic protein translation.

Expression, purification, and characterization of the maltooligosyltrehalose trehalohydrolase from the thermophilic archaeon Sulfolobus solfataricus ATCC 35092

The maltooligosyltrehalose trehalohydrolase (MTHase) mainly cleaves the alpha-1,4-glucosidic linkage next to the alpha-1,1-linked terminal disaccharide of maltooligosyltrehalose to produce trehalose and the maltooligosaccharide with lower molecular mass. In this study, the treZ gene encoding MTHase was PCR-cloned from Sulfolobus solfataricus ATCC 35092 and then expressed in Escherichia coli. A high yield of the active wild-type MTHase, 13300 units/g of wet cells, was obtained in the absence of IPTG induction. Wild-type MTHase was purified sequentially using heat treatment, nucleic acid precipitation, and ion-exchange chromatography. The purified wild-type MTHase showed an apparent optimal pH of 5 and an optimal temperature at 85 degrees C. The enzyme was stable at pH values ranging from 3.5 to 11, and the activity was fully retained after a 2-h incubation at 45-85 degrees C. The k(cat) values of the enzyme for hydrolysis of maltooligosyltrehaloses with degree of polymerization (DP) 4-7 were 193, 1030, 1190, and 1230 s(-1), respectively, whereas the k(cat) values for glucose formation during hydrolysis of DP 4-7 maltooligosaccharides were 5.49, 17.7, 18.2, and 6.01 s(-1), respectively. The K(M) values of the enzyme for hydrolysis of DP 4-7 maltooligosyltrehaloses and those for maltooligosaccharides are similar at the same corresponding DPs. These results suggest that this MTHase could be used to produce trehalose at high temperatures.

Influences on gene expression in vivo by a Shine-Dalgarno sequence

The Shine-Dalgarno (SD+: 5'-AAGGAGG-3') sequence anchors the mRNA by base pairing to the 16S rRNA in the small ribosomal subunit during translation initiation. We have here compared how an SD+ sequence influences gene expression, if located upstream or downstream of an initiation codon. The positive effect of an upstream SD+ is confirmed. A downstream SD+ gives decreased gene expression. This effect is also valid for appropriately modified natural Escherichia coli genes. If an SD+ is placed between two potential initiation codons, initiation takes place predominantly at the second start site. The first start site is activated if the distance between this site and the downstream SD+ is enlarged and/or if the second start site is weakened. Upstream initiation is eliminated if a stable stem-loop structure is placed between this SD+ and the upstream start site. The results suggest that the two start sites compete for ribosomes that bind to an SD+ located between them. A minor positive contribution to upstream initiation resulting from 3' to 5' ribosomal diffusion along the mRNA is suggested. Analysis of the E. coli K12 genome suggests that the SD+ or SD-like sequences are systematically avoided in the early coding region suggesting an evolutionary significance.

Gene Designer: a synthetic biology tool for constructing artificial DNA segments

Background

Direct synthesis of genes is rapidly becoming the most efficient way to make functional genetic constructs and enables applications such as codon optimization, RNAi resistant genes and protein engineering. Here we introduce a software tool that drastically facilitates the design of synthetic genes.

Results

Gene Designer is a stand-alone software for fast and easy design of synthetic DNA segments. Users can easily add, edit and combine genetic elements such as promoters, open reading frames and tags through an intuitive drag-and-drop graphic interface and a hierarchical DNA/Protein object map. Using advanced optimization algorithms, open reading frames within the DNA construct can readily be codon optimized for protein expression in any host organism. Gene Designer also includes features such as a real-time sliding calculator of oligonucleotide annealing temperatures, sequencing primer generator, tools for avoidance or inclusion of restriction sites, and options to maximize or minimize sequence identity to a reference.

Conclusion

Gene Designer is an expandable Synthetic Biology workbench suitable for molecular biologists interested in the de novo creation of genetic constructs.

Optimization of a ribosomal structural domain by natural selection

A conserved, independently folding domain in the large ribosomal subunit consists of 58 nt of rRNA and a single protein, L11. The tertiary structure of an rRNA fragment carrying the Escherichia coli sequence is marginally stable in vitro but can be substantially stabilized by mutations found in other organisms. To distinguish between possible reasons why natural selection has not evolved a more stable rRNA structure in E. coli, mutations affecting the rRNA tertiary structure were assessed for their in vitro effects on rRNA stability and L11 affinity (in the context of an rRNA fragment) or in vivo effects on cell growth rate and L11 content of ribosomes. The rRNA fragment stabilities ranged from -4 to +9 kcal/mol relative to the wild-type sequence. Variants in the range of -4 to +5 kcal/mol had almost no observable effect in vivo, while more destabilizing mutations (>7 kcal/mol) were not tolerated. The data suggest that the in vivo stability of the complex is roughly -6 kcal/mol and that any single tertiary interaction is dispensable for function as long as a minimum stability of the complex is maintained. On the basis of these data, it seems that the evolution of this domain has not been constrained by inherent structural or functional limits on stability. The estimated stability corresponds to only a few ribosomes per bacterial cell dissociated from L11 at any time; thus the selective advantage for any further increase in stability may be so small as to be outweighed by other competing selective pressures.

A periodic pattern of mRNA secondary structure created by the genetic code

Single-stranded mRNA molecules form secondary structures through complementary self-interactions. Several hypotheses have been proposed on the relationship between the nucleotide sequence, encoded amino acid sequence and mRNA secondary structure. We performed the first transcriptome-wide in silico analysis of the human and mouse mRNA foldings and found a pronounced periodic pattern of nucleotide involvement in mRNA secondary structure. We show that this pattern is created by the structure of the genetic code, and the dinucleotide relative abundances are important for the maintenance of mRNA secondary structure. Although synonymous codon usage contributes to this pattern, it is intrinsic to the structure of the genetic code and manifests itself even in the absence of synonymous codon usage bias at the 4-fold degenerate sites. While all codon sites are important for the maintenance of mRNA secondary structure, degeneracy of the code allows regulation of stability and periodicity of mRNA secondary structure. We demonstrate that the third degenerate codon sites contribute most strongly to mRNA stability. These results convincingly support the hypothesis that redundancies in the genetic code allow transcripts to satisfy requirements for both protein structure and RNA structure. Our data show that selection may be operating on synonymous codons to maintain a more stable and ordered mRNA secondary structure, which is likely to be important for transcript stability and translation. We also demonstrate that functional domains of the mRNA [5′-untranslated region (5′-UTR), CDS and 3′-UTR] preferentially fold onto themselves, while the start codon and stop codon regions are characterized by relaxed secondary structures, which may facilitate initiation and termination of translation.

Endonucleolytic cleavage of eukaryotic mRNAs with stalls in translation elongation

A fundamental aspect of the biogenesis and function of eukaryotic messenger RNA is the quality control systems that recognize and degrade non-functional mRNAs. Eukaryotic mRNAs where translation termination occurs too soon (nonsense-mediated decay) or fails to occur (non-stop decay) are rapidly degraded. We show that yeast mRNAs with stalls in translation elongation are recognized and targeted for endonucleolytic cleavage, referred to as 'no-go decay'. The cleavage triggered by no-go decay is dependent on translation and involves Dom34p and Hbs1p. Dom34p and Hbs1p are similar to the translation termination factors eRF1 and eRF3 (refs 3, 4), indicating that these proteins might function in recognizing the stalled ribosome and triggering endonucleolytic cleavage. No-go decay provides a mechanism for clearing the cell of stalled translation elongation complexes, which could occur as a result of damaged mRNAs or ribosomes, or as a mechanism of post-transcriptional control.

A model of protein translation including codon bias, nonsense errors, and ribosome recycling

We present and analyse a model of protein translation at the scale of an individual messenger RNA (mRNA) transcript. The model we develop is unique in that it incorporates the phenomena of ribosome recycling and nonsense errors. The model conceptualizes translation as a probabilistic wave of ribosome occupancy traveling down a heterogeneous medium, the mRNA transcript. Our results show that the heterogeneity of the codon translation rates along the mRNA results in short-scale spikes and dips in the wave. Nonsense errors attenuate this wave on a longer scale while ribosome recycling reinforces it. We find that the combination of nonsense errors and codon usage bias can have a large effect on the probability that a ribosome will completely translate a transcript. We also elucidate how these forces interact with ribosome recycling to determine the overall translation rate of an mRNA transcript. We derive a simple cost function for nonsense errors using our model and apply this function to the yeast (Saccharomyces cervisiae) genome. Using this function we are able to detect position dependent selection on codon bias which correlates with gene expression levels as predicted a priori. These results indirectly validate our underlying model assumptions and confirm that nonsense errors can play an important role in shaping codon usage bias.

Experimental and Computational Analysis of Translation Products in Apomyoglobin Expression

This work focuses on the experimental analysis of the time-course of protein expression in a cell-free system, in conjunction with the development of a computational model, denoted as progressive chain buildup (PCB), able to simulate translation kinetics and product formation as a function of starting reactant concentrations. Translation of the gene encoding the apomyoglobin (apoMb) model protein was monitored in an Escherichia coli cell-free system under different experimental conditions. Experimentally observed protein expression yields, product accumulation time-course and expression completion times match with the predictions by the PCB model. This algorithm regards elementary single-residue elongations as apparent second-order events and it accounts for aminoacyl-tRNA regeneration during translation. We have used this computational approach to model full-length protein expression and to explore the kinetic behavior of incomplete chains generated during protein biosynthesis. Most of the observed incomplete chains are non-obligatory dead-end species, in that their formation is not mandatory for full-length protein expression, and that they are unable to convert to the expected final translation product. These truncated polypeptides do not arise from post-translational degradation of full-length protein, but from a distinct subpopulation of chains which expresses intrinsically more slowly than the population leading to full-length product. The PCB model is a valuable tool to predict full-length and incomplete chain populations and formulate experimentally testable hypotheses on their origin. PCB simulations are applicable to E.coli cell-free expression systems (both in batch and dialysis mode) under the control of T7 RNA polymerase and to other environments where transcription and translation can be regarded as kinetically decoupled.

Introns and the origin of nucleus-cytosol compartmentalization

The origin of the eukaryotic nucleus marked a seminal evolutionary transition. We propose that the nuclear envelope's incipient function was to allow mRNA splicing, which is slow, to go to completion so that translation, which is fast, would occur only on mRNA with intact reading frames. The rapid, fortuitous spread of introns following the origin of mitochondria is adduced as the selective pressure that forged nucleus-cytosol compartmentalization.

tRNA properties help shape codon pair preferences in open reading frames

Translation elongation is an accurate and rapid process, dependent upon efficient juxtaposition of tRNAs in the ribosomal A- and P-sites. Here, we sought evidence of A- and P-site tRNA interaction by examining bias in codon pair choice within open reading frames from a range of genomes. Three distinct and marked effects were revealed once codon and dipeptide biases had been subtracted. First, in the majority of genomes, codon pair preference is primarily determined by a tetranucleotide combination of the third nucleotide of the P-site codon, and all 3 nt of the A-site codon. Second, pairs of rare codons are generally under-used in eukaryotes, but over-used in prokaryotes. Third, the analysis revealed a highly significant effect of tRNA-mediated selection on codon pairing in unicellular eukaryotes, Bacillus subtilis, and the gamma proteobacteria. This was evident because in these organisms, synonymous codons decoded in the A-site by the same tRNA exhibit significantly similar P-site pairing preferences. Codon pair preference is thus influenced by the identity of A-site tRNAs, in combination with the P-site codon third nucleotide. Multivariate analysis identified conserved nucleotide positions within A-site tRNA sequences that modulate codon pair preferences. Structural features that regulate tRNA geometry within the ribosome may govern genomic codon pair patterns, driving enhanced translational fidelity and/or rate.

Overproduction in Escherichia coli and purification of Epstein-Barr virus EBNA-1

Epstein-Barr virus nuclear antigen 1 (EBNA-1) is a multi-functional protein of the Epstein-Barr virus (EBV). Due to its low abundance in EBV-transformed cells, overproduction in a foreign host is preferred to obtain purified EBNA-1 protein. The EBNA-1 gene possesses a large number of Escherichia coli rare codons (23%). By using E. coli BL21(DE3)Rosetta2 cells that augment the low-abundance tRNA genes, the expression level of EBNA-1 in E. coli was greatly enhanced. EBNA-1 was then purified by applying the whole cell extract soluble fraction to a Ni-NTA Superflow column and eluting with an imidazole gradient. The improved overexpression in E. coli followed by a one-step Ni-NTA purification resulted in a sufficient amount of pure EBNA-1 protein to test DNA binding activity, and prepare and test EBNA-1-specific monoclonal antibodies (mAbs).

Selection in favor of nucleotides G and C diversifies evolution rates and levels of polymorphism at mammalian synonymous sites

The impact of synonymous nucleotide substitutions on fitness in mammals remains controversial. Despite some indications of selective constraint, synonymous sites are often assumed to be neutral, and the rate of their evolution is used as a proxy for mutation rate. We subdivide all sites into four classes in terms of the mutable CpG context, nonCpG, postC, preG, and postCpreG, and compare four-fold synonymous sites and intron sites residing outside transposable elements. The distribution of the rate of evolution across all synonymous sites is trimodal. Rate of evolution at nonCpG synonymous sites, not preceded by C and not followed by G, is approximately 10% below that at such intron sites. In contrast, rate of evolution at postCpreG synonymous sites is approximately 30% above that at such intron sites. Finally, synonymous and intron postC and preG sites evolve at similar rates. The relationship between the levels of polymorphism at the corresponding synonymous and intron sites is very similar to that between their rates of evolution. Within every class, synonymous sites are occupied by G or C much more often than intron sites, whose nucleotide composition is consistent with neutral mutation-drift equilibrium. These patterns suggest that synonymous sites are under weak selection in favor of G and C, with the average coefficient s approximately 0.25/Ne approximately 10(-5), where Ne is the effective population size. Such selection decelerates evolution and reduces variability at sites with symmetric mutation, but has the opposite effects at sites where the favored nucleotides are more mutable. The amino-acid composition of proteins dictates that many synonymous sites are CpGprone, which causes them, on average, to evolve faster and to be more polymorphic than intron sites. An average genotype carries approximately 10(7) suboptimal nucleotides at synonymous sites, implying synergistic epistasis in selection against them.

Comparisons of NMR Spectral Quality and Success in Crystallization Demonstrate that NMR and X-ray Crystallography Are Complementary Methods for Small Protein Structure Determination

X-ray crystallography and NMR spectroscopy provide the only sources of experimental data from which protein structures can be analyzed at high or even atomic resolution. The degree to which these methods complement each other as sources of structural knowledge is a matter of debate; it is often proposed that small proteins yielding high quality, readily analyzed NMR spectra are a subset of those that readily yield strongly diffracting crystals. We have examined the correlation between NMR spectral quality and success in structure determination by X-ray crystallography for 159 prokaryotic and eukaryotic proteins, prescreened to avoid proteins providing polydisperse and/or aggregated samples. This study demonstrates that, across this protein sample set, the quality of a protein's [15N-1H]-heteronuclear correlation (HSQC) spectrum recorded under conditions generally suitable for 3D structure determination by NMR, a key predictor of the ability to determine a structure by NMR, is not correlated with successful crystallization and structure determination by X-ray crystallography. These results, together with similar results of an independent study presented in the accompanying paper (Yee, et al., J. Am. Chem. Soc., accompanying paper), demonstrate that X-ray crystallography and NMR often provide complementary sources of structural data and that both methods are required in order to optimize success for as many targets as possible in large-scale structural proteomics efforts.

Optimality of codon usage in Escherichia coli due to load minimization

The canonical genetic code is known to be highly efficient in minimizing the effects of mistranslational errors and point mutations, an ability which in term is designated "load minimization". One parameter involved in calculating the load minimizing property of the genetic code is codon usage. In most bacteria, synonymous codons are not used with equal frequencies. Different factors have been proposed to contribute to codon usage preference. It has been shown that the codon preference is correlated with the composition of the tRNA pool. Selection for translational efficiency and translational accuracy both result in such a correlation. In this work, it is shown that codon usage bias in Escherichia coli works so as to minimize the consequences of translational errors, i.e. optimized for load minimization.

Human antibody response to outer membrane protein G1a, a lipoprotein of Moraxella catarrhalis

Moraxella catarrhalis is an important cause of respiratory infections in adults with chronic obstructive pulmonary disease (COPD) and of otitis media in children. Outer membrane protein (OMP) G1a is an approximately 29-kDa surface lipoprotein and is a potential vaccine candidate. The gene that encodes OMP G1a was expressed and purified using a novel plasmid vector. [(3)H]palmitic acid labeling demonstrated that both native and recombinant OMP G1a contain covalently bound palmitic acid. To assess the expression of OMP G1a during human infection, paired sera and sputum supernatants from adults with COPD followed prospectively were studied by enzyme-linked immunosorbent assays with recombinant lipidated OMP G1a to detect antibodies made specifically during carriage of M. catarrhalis. Overall, 23% of patients developed either a serum immunoglobulin G (IgG) response (9%) or sputum IgA response (21%) to OMP G1a, following 100 episodes of acquisition and clearance of M. catarrhalis. Patients developed antibody responses at similar rates following episodes of clinical exacerbation compared to asymptomatic colonization. Serum IgG antibodies following natural infection were directed predominantly at OMP G1a epitopes that are not exposed on the bacterial surface. These data show that OMP G1a is expressed during infection of the human respiratory tract and is a target of systemic and mucosal antibodies. These observations indicate that OMP G1a, a highly conserved surface protein, should be evaluated further as a vaccine candidate.

Modelling of translation of human protein disulfide isomerase in Escherichia coli—A case study of gene optimisation

Recombinant human protein disulfide isomerase (PDI) was expressed in vivo in Escherichia coli using a non-optimised gene sequence and an optimised sequence with four 5' codons substituted by synonymous codons that take less time to translate. The optimisation resulted in a 2-fold increase of total PDI concentration and by successive optimisation with expression at low temperature in a 10-fold increase of the amount of soluble PDI in comparison with the original wild-type construct. The improvement can be due to a faster clearing of the ribosome binding site on the mRNA, elevating the translation initiation rate and resulting in higher ribosome loading and better ribosome protection of the PDI mRNA against endonucleolytic cleavage by RNase. This hypothesis was supported by a novel computer simulation model of E. coli translational ribosome traffic based upon the stochastic Gillespie algorithm. The study indicates the applicability of such models in optimisation of recombinant protein sequences.

Combined effect of protein fusion and signal sequence greatly enhances the production of recombinant human GM-CSF in Escherichia coli

Granulocyte-macrophage colony-stimulating factor (GM-CSF) is a hematopoietic growth factor, that has been used as a therapeutic agent in facilitating bone marrow and stem cell transplantation and in other clinical cases like neutropenia. Although biologically active recombinant GM-CSF has been successfully produced in Escherichia coli, the reported levels are extremely poor. In this study we looked into the possible reasons for poor expression and found that protein toxicity coupled with protease-based degradation was the principal reason for low productivity. To overcome this problem we attached a signal sequence, as well as an amino-terminal His-tag fusion to the GM-CSF gene. This combination had a dramatic effect on expression levels, which increased from 0.8 microg/mL in the control to 40 microg/mL. When a larger fusion partner, such as the Maltose-binding protein (MBP-tag), was used the expression levels increased further to 69.5 microg/mL, which along with the MBP-tag represented approx 12% of the total cellular protein.

In vivo effect of inactivation of ribosome recycling factor - fate of ribosomes after unscheduled translation downstream of open reading frame

The post-termination ribosomal complex is disassembled by ribosome recycling factor (RRF) and elongation factor G. Without RRF, the ribosome is not released from mRNA at the termination codon and reinitiates translation downstream. This is called unscheduled translation. Here, we show that at the non-permissive temperature of a temperature-sensitive RRF strain, RRF is lost quickly, and some ribosomes reach the 3' end of mRNA. However, instead of accumulating at the 3' end of mRNA, ribosomes are released as monosomes. Some ribosomes are transferred to transfer-messenger RNA from the 3' end of mRNA. The monosomes thus produced are able to translate synthetic homopolymer but not natural mRNA with leader and canonical initiation signal. The pellet containing ribosomes appears to be responsible for rapid but reversible inhibition of most but not all of protein synthesis in vivo closely followed by decrease of cellular RNA and DNA synthesis.

JCat: a novel tool to adapt codon usage of a target gene to its potential expression host

A novel method for the adaptation of target gene codon usage to most sequenced prokaryotes and selected eukaryotic gene expression hosts was developed to improve heterologous protein production. In contrast to existing tools, JCat (Java Codon Adaptation Tool) does not require the manual definition of highly expressed genes and is, therefore, a very rapid and easy method. Further options of JCat for codon adaptation include the avoidance of unwanted cleavage sites for restriction enzymes and Rho-independent transcription terminators. The output of JCat is both graphically and as Codon Adaptation Index (CAI) values given for the pasted sequence and the newly adapted sequence. Additionally, a list of genes in FASTA-format can be uploaded to calculate CAI values. In one example, all genes of the genome of Caenorhabditis elegans were adapted to Escherichia coli codon usage and further optimized to avoid commonly used restriction sites. In a second example, the Pseudomonas aeruginosa exbD gene codon usage was adapted to E.coli codon usage with parallel avoidance of the same restriction sites. For both, the degree of introduced changes was documented and evaluated. JCat is integrated into the PRODORIC database that hosts all required information on the various organisms to fulfill the requested calculations. JCat is freely accessible at .

What makes ribosome-mediated transcriptional attenuation sensitive to amino acid limitation?

Ribosome-mediated transcriptional attenuation mechanisms are commonly used to control amino acid biosynthetic operons in bacteria. The mRNA leader of such an operon contains an open reading frame with “regulatory” codons, cognate to the amino acid that is synthesized by the enzymes encoded by the operon. When the amino acid is in short supply, translation of the regulatory codons is slow, which allows transcription to continue into the structural genes of the operon. When amino acid supply is in excess, translation of regulatory codons is rapid, which leads to termination of transcription. We use a discrete master equation approach to formulate a probabilistic model for the positioning of the RNA polymerase and the ribosome in the attenuator leader sequence. The model describes how the current rate of amino acid supply compared to the demand in protein synthesis (signal) determines the expression of the amino acid biosynthetic operon (response). The focus of our analysis is on the sensitivity of operon expression to a change in the amino acid supply. We show that attenuation of transcription can be hyper-sensitive for two main reasons. The first is that its response depends on the outcome of a race between two multi-step mechanisms with synchronized starts: transcription of the leader of the operon, and translation of its regulatory codons. The relative change in the probability that transcription is aborted (attenuated) can therefore be much larger than the relative change in the time it takes for the ribosome to read a regulatory codon. The second is that the general usage frequencies of codons of the type used in attenuation control are small. A small percentage decrease in the rate of supply of the controlled amino acid can therefore lead to a much larger percentage decrease in the rate of reading a regulatory codon. We show that high sensitivity further requires a particular choice of regulatory codon among several synonymous codons for the same amino acid. We demonstrate the importance of a high fraction of regulatory codons in the control region. Finally, our integrated model explains how differences in leader sequence design of the trp and his operons of Escherichia coli and Salmonella typhimurium lead to high basal expression and low sensitivity in the former case, and to large dynamic range and high sensitivity in the latter. The model clarifies how mechanistic and systems biological aspects of the attenuation mechanism contribute to its overall sensitivity. It also explains structural differences between the leader sequences of the trp and his operons in terms of their different functions.

When cells grow and divide, they must continually construct new proteins from the 20 amino acid building blocks according to the instructions of the genetic code. Proteins are made by large macromolecular complexes, ribosomes, where information encoded as base triplets (codons) in messenger RNA sequences, transcribed from the DNA sequences of the genes, is translated into amino acid sequences that determine the functions of all proteins. Rapid growth of cells requires that the supply of each free amino acid is balanced to the demand for it in protein synthesis.

The present work mathematically models a common control mechanism in bacteria, which regulates synthesis of amino acids to eliminate deviations from balanced supply and demand. The mechanism “measures” the speed by which the ribosome translates the codons of a regulated amino acid. When supply is less than demand, translation of these “control” codons is slow, which is sensed by the mechanism and used to increase synthesis of the amino acid.

This paper explains why the mechanism is “hyper-sensitive” to relative changes in supply and demand, and why it is differently designed for control of the enzymes that synthesize the amino acids histidine and tryptophan.

Expression of enterovirus 70 capsid protein VP1 in Escherichia coli

The VP1 gene of enterovirus 70 (EV70) possesses a large number of Escherichia coli low-usage codons (11.0%) and a bacterial ribosome binding site complementary sequence (RBSCS) 5'-UGUCUCCUUUUC-3' flanking the codon 139. Plasmids containing EV70 cDNA encoding the full-length VP1 failed to express in E. coli (BL21(DE3), Rosetta 2(DE3) or Rosetta (DE3)pLysS). High expression (>8% of total protein) of recombinant VP1 (rVP1m) in E. coli required engineering of the encoding cDNA (conserved modification of the native cDNA) by simultaneous substitution of a rare-codon cluster located between codons 103 and 132, and replacement of the RBSCS-TCCTTT sequence. The rare-codon frequencies of the cDNAs encoding VP1 non-overlapping terminal fragments N138 (1-138 aa) and C170 (141-310 aa) are similar (10.9 and 11.2%, respectively). However, in E. coli, high expression of recombinant C170 (rC170) required no modification of the native cDNA whereas high expression of recombinant N138 (rN138m) required minimal synonymous substitution of the above rare-codon cluster. The rare-codon cluster of EV70 VP1 gene has five least-usage arginine codons (AGG/AGA) and three tandem rare-codon pairs (AGGAGG, CUAAGG, and AGACUA). Our results suggest that the rare-codon cluster (its rare codon arrangement per se and/or its related mRNA secondary structure(s)) and the RBSCS in EV70 VP1 gene, not the rare-codon frequency, constitute the key elements that suppress its expression in E. coli.

Visualization of differential gene expression by improved cyan fluorescent protein and yellow fluorescent protein production in Bacillus subtilis

The distinguishable cyan and yellow fluorescent proteins (CFP and YFP) enable the simultaneous in vivo visualization of different promoter activities. Here, we report new cloning vectors for the construction of cfp and yfp fusions in Bacillus subtilis. By extending the N-terminal portions of previously described CFP and YFP variants, 20- to 70-fold-improved fluorescent-protein production was achieved. Probably, the addition of sequences encoding the first eight amino acids of the N-terminal part of ComGA of B. subtilis overcomes the slow translation initiation that is provoked by the eukaryotic codon bias present in the original cfp and yfp genes. Using these new vectors, we demonstrate that, within an isogenic population of sporulating B. subtilis cells, expression of the abrB and spoIIA genes is distinct in individual cells.

Differences in codon bias cannot explain differences in translational power among microbes

Background

Translational power is the cellular rate of protein synthesis normalized to the biomass invested in translational machinery. Published data suggest a previously unrecognized pattern: translational power is higher among rapidly growing microbes, and lower among slowly growing microbes. One factor known to affect translational power is biased use of synonymous codons. The correlation within an organism between expression level and degree of codon bias among genes of Escherichia coli and other bacteria capable of rapid growth is commonly attributed to selection for high translational power. Conversely, the absence of such a correlation in some slowly growing microbes has been interpreted as the absence of selection for translational power. Because codon bias caused by translational selection varies between rapidly growing and slowly growing microbes, we investigated whether observed differences in translational power among microbes could be explained entirely by differences in the degree of codon bias. Although the data are not available to estimate the effect of codon bias in other species, we developed an empirically-based mathematical model to compare the translation rate of E. coli to the translation rate of a hypothetical strain which differs from E. coli only by lacking codon bias.

Results

Our reanalysis of data from the scientific literature suggests that translational power can differ by a factor of 5 or more between E. coli and slowly growing microbial species. Using empirical codon-specific in vivo translation rates for 29 codons, and several scenarios for extrapolating from these data to estimates over all codons, we find that codon bias cannot account for more than a doubling of the translation rate in E. coli, even with unrealistic simplifying assumptions that exaggerate the effect of codon bias. With more realistic assumptions, our best estimate is that codon bias accelerates translation in E. coli by no more than 60% in comparison to microbes with very little codon bias.

Conclusions

While codon bias confers a substantial benefit of faster translation and hence greater translational power, the magnitude of this effect is insufficient to explain observed differences in translational power among bacterial and archaeal species, particularly the differences between slowly growing and rapidly growing species. Hence, large differences in translational power suggest that the translational apparatus itself differs among microbes in ways that influence translational performance.

Robotic cloning and Protein Production Platform of the Northeast Structural Genomics Consortium

In this chapter we describe the core Protein Production Platform of the Northeast Structural Genomics Consortium (NESG) and outline the strategies used for producing high-quality protein samples using Escherichia coli host vectors. The platform is centered on 6X-His affinity-tagged protein constructs, allowing for a similar purification procedure for most targets, and the implementation of high-throughput parallel methods. In most cases, these affinity-purified proteins are sufficiently homogeneous that a single subsequent gel filtration chromatography step is adequate to produce protein preparations that are greater than 98% pure. Using this platform, over 1000 different proteins have been cloned, expressed, and purified in tens of milligram quantities over the last 36-month period (see Summary Statistics for All Targets, ). Our experience using a hierarchical multiplex expression and purification strategy, also described in this chapter, has allowed us to achieve success in producing not only protein samples but also many three-dimensional structures. As of December 2004, the NESG Consortium has deposited over 145 new protein structures to the Protein Data Bank (PDB); about two-thirds of these protein samples were produced by the NESG Protein Production Facility described here. The methods described here have proven effective in producing quality samples of both eukaryotic and prokaryotic proteins. These improved robotic and?or parallel cloning, expression, protein production, and biophysical screening technologies will be of broad value to the structural biology, functional proteomics, and structural genomics communities.

Clustered bottlenecks in mRNA translation and protein synthesis

Using a model based on the totally asymmetric exclusion process, we investigate the effects of slow codons along messenger RNA. Ribosome density profiles near neighboring clusters of slow codons interact, enhancing suppression of ribosome throughput when such bottlenecks are closely spaced. Increasing the slow codon cluster size beyond approximately 3-4 codons does not significantly reduce the ribosome current. Our results are verified by both extensive Monte Carlo simulations and numerical calculation, and provide a biologically motivated explanation for the experimentally observed clustering of low-usage codons.

Whole genome analysis reveals a high incidence of non-optimal codons in secretory signal sequences of Escherichia coli

Translational pausing may occur due to a number of mechanisms, including the presence of non-optimal codons, and it is thought to play a role in the folding of specific polypeptide domains during translation and in the facilitation of signal peptide recognition during sec-dependent protein targeting. In this whole genome analysis of Escherichia coli we have found that non-optimal codons in the signal peptide-encoding sequences of secretory genes are overrepresented relative to the "mature" portions of these genes; this is in addition to their overrepresentation in the 5'-regions of genes encoding non-secretory proteins. We also find increased non-optimal codon usage at the 3' ends of most E. coli genes, in both non-secretory and secretory sequences. Whereas presumptive translational pausing at the 5' and 3' ends of E. coli messenger RNAs may clearly have a general role in translation, we suggest that it also has a specific role in sec-dependent protein export, possibly in facilitating signal peptide recognition. This finding may have important implications for our understanding of how the majority of non-cytoplasmic proteins are targeted, a process that is essential to all biological cells.

Codon usage bias from tRNA's point of view: redundancy, specialization, and efficient decoding for translation optimization

The selection-mutation-drift theory of codon usage plays a major role in the theory of molecular evolution by explaining the co-evolution of codon usage bias and tRNA content in the framework of translation optimization. Because most studies have focused only on codon usage, we analyzed the tRNA gene pool of 102 bacterial species. We show that as minimal generation times get shorter, the genomes contain more tRNA genes, but fewer anticodon species. Surprisingly, despite the wide G+C variation of bacterial genomes these anticodons are the same in most genomes. This suggests an optimization of the translation machinery to use a small subset of optimal codons and anticodons in fast-growing bacteria and in highly expressed genes. As a result, the overrepresented codons in highly expressed genes tend to be the same in very different genomes to match the same most-frequent anticodons. This is particularly important in fast-growing bacteria, which have higher codon usage bias in these genes. Three models were tested to understand the choice of codons recognized by the same anticodons, all providing significant fit, but under different classes of genes and genomes. Thus, co-evolution of tRNA gene composition and codon usage bias in genomes seen from tRNA's point of view agrees with the selection-mutation-drift theory. However, it suggests a much more universal trend in the evolution of anticodon and codon choice than previously thought. It also provides new evidence that a selective force for the optimization of the translation machinery is the maximization of growth.

A codon window in mRNA downstream of the initiation codon where NGG codons give strongly reduced gene expression in Escherichia coli

The influences on gene expression by codons at positions +2, +3, +5 and +7 downstream of the initiation codon have been compared. Most of the +2 codons that are known to give low gene expression are associated with a higher expression if placed at the later positions. The NGG codons AGG, CGG, UGG and GGG, but not GGN or GNG (where N is non-G), are unique since they are associated with a very low gene expression also if located at positions +2, +3 and +5. All codons, including NGG, give a normal gene expression if placed at positions +7. The negative effect by the NGG codons is true for both the lacZ and 3A' model genes. The low expression is suggested to originate at the translational level, although it is not the result of mRNA secondary structure or a lowered intracellular mRNA pool.

Intragenic codon bias in a set of mouse and human genes

To better conceptualize the mechanism underlying the evolution of synonymous codons, we have analysed intragenic codon usage in chosen "regions" of some mouse and human genes. We divided a given gene into two regions: one consisting of a trinucleotide repeat (TNR) and the other consisting of the "rest of the coding region" (RCR). Usually, a TNR is composed of a repetitive single codon, which may reflect its frequency in a gene. In contrast, a non-random frequency of a codon in the RCR versus TNR (or vice versa) of a gene should indicate a bias for that codon within the TNR. We examined this scenario by comparing codon frequency between the RCR and the cognate TNR(s) for a set of human and mouse genes. A TNR length of six amino acids or more was used to identify genes from the Genbank database. Twenty nine human and twenty one mouse genes containing TNRs coding for nine different amino acid runs were identified. The ratio of codon frequency in a TNR versus the corresponding RCR was expressed as "fold change" which was also regarded as a measure of codon bias (defined as preferential use either in TNR or in RCR). Chi-square values were then determined from the distribution of codon frequency in a TNR vs. the cognate RCR. At p<0.001, 22% and 27%, respectively, of human and mouse TNRs showed codon bias. Greater than 40% of the TNRs (29 out of 69 in human, and 18 of 42 in mouse) showed codon bias at p<0.05. In addition, we identify eight single-codon TNRs in mouse and ten in human genes. Thus, our results show intragenic codon bias in both mouse and human genes expressed in diverse tissue types. Since our results are independent of the Codon Adaptation Index (CAI) and starvation CAI, and since the tRNA repertoire in a cell or in a tissue is constant, our data suggest that other constraints besides tRNA abundance played a role in creating intragenic codon bias in these genes.

Experimental reduction of codon bias in the Drosophila alcohol dehydrogenase gene results in decreased ethanol tolerance of adult flies

The ethanol tolerance of adult transgenic flies of Drosophila containing between zero and ten unpreferred synonymous mutations that reduced codon bias in the alcohol dehydrogenase (Adh) gene was assayed. As the amino acid sequences of the ADH protein were identical in the four genotypes assayed, differences in ethanol tolerance were due to differences in the abundance of ADH protein, presumably driven by the effects of codon bias on translational efficiency. The ethanol tolerance of genotypes decreased with the number of unpreferred synonymous mutations, and a positive correlation between ADH protein abundance and ethanol tolerance was observed. This work confirms that the fitness effects of unpreferred synonymous mutations that reduce codon bias in a highly expressed gene are experimentally measurable in Drosophila melanogaster.

Tissue-specific codon usage and the expression of human genes

A diverse array of mechanisms regulate tissue-specific protein levels. Most research, however, has focused on the role of transcriptional regulation. Here we report systematic differences in synonymous codon usage between genes selectively expressed in six adult human tissues. Furthermore, we show that the codon usage of brain-specific genes has been selectively preserved throughout the evolution of human and mouse from their common ancestor. Our findings suggest that codon-mediated translational control may play an important role in the differentiation and regulation of tissue-specific gene products in humans.