Artificial genetic selection for an efficient translation initiation site for expression of human RACK1 gene in Escherichia coli

Importance of the 5' regulatory region to bacterial synthetic biology applications

The field of synthetic biology is evolving at a fast pace. It is advancing beyond single-gene alterations in single hosts to the logical design of complex circuits and the development of integrated synthetic genomes. Recent breakthroughs in deep learning, which is increasingly used in de novo assembly of DNA components with predictable effects, are also aiding the discipline. Despite advances in computing, the field is still reliant on the availability of pre-characterized DNA parts, whether natural or synthetic, to regulate gene expression in bacteria and make valuable compounds. In this review, we discuss the different bacterial synthetic biology methodologies employed in the creation of 5' regulatory regions - promoters, untranslated regions and 5'-end of coding sequences. We summarize methodologies and discuss their significance for each of the functional DNA components, and highlight the key advances made in bacterial engineering by concentrating on their flaws and strengths. We end the review by outlining the issues that the discipline may face in the near future.

Exploring the 5'-UTR DNA region as a target for optimizing recombinant gene expression from the strong and inducible Pm promoter in Escherichia coli

By using the strong and inducible Pm promoter as a model, we recently reported that the β-lactamase production (encoded by bla) can be stimulated up to 20-fold in Escherichia coli by mutating the DNA region corresponding to the 5'-untranslated region of mRNA (UTR). One striking observation was the unexpected large stimulatory effect some of these UTR variants had on the bla transcript production level. We here demonstrate that such UTR variants can also be used to improve the expression level of the alternative genes celB (encoding phosphoglucomutase) and inf-α2b (encoding human cytokine interferon α2b), which both can be expressed to high levels even with the wild-type Pm UTR DNA sequence. Our data indicated some degree of context dependency between the UTR DNA and concomitant recombinant gene sequences. By constructing and using a synthetic operon, we demonstrated that UTR variants optimized for high-level expression of probably any recombinant gene can be efficiently selected from large UTR mutant libraries. The stimulation affected both the transcript production and translational level, and such modified UTR sequences therefore clearly have a significant applied potential for improvement of recombinant gene expression processes.

The carboxyl terminal of the archaeal nuclease NurA is involved in the interaction with single-stranded DNA-binding protein and dimer formation

The nuclease NurA is present in all known thermophilic archaea and has been implicated to facilitate efficient DNA double-strand break end processing in Mre11/Rad50-mediated homologous recombinational repair. To understand the structural and functional relationship of this enzyme, we constructed five site-directed mutants of NurA from Sulfolobus tokodaii (StoNurA), D56A, E114A, D131A, Y291A, and H299A, at the conserved motifs, and four terminal deletion mutants, StoNurAΔN (19-331), StoNurAΔNΔC (19-303), StoNurAΔC (1-281), and StoNurAΔC (1-303), and characterized the proteins biochemically. We found that mutation at the acidic residue, D56, E114, D131, or at the basic residue, H299, abolishes the nuclease activity, while mutation at the aromatic residue Y291 only impairs the activity. Interestingly, by chemical cross-linking assay, we found that the mutant Y291A is unable to form stable dimer. Additionally, we demonstrated that deletion of the C-terminal amino acid residues 304-331 of StoNurA results in loss of the physical and functional interaction with the single-stranded DNA-binding protein (StoSSB). These results established that the C-terminal conserved aromatic residue Y291 is involved in dimer formation and the C-terminal residues 304-331 of NurA are involved in the interaction with single-stranded DNA-binding protein.

Introduction of amino acid residues at the N-terminus of the zeocin-resistance protein increases its expression in Saccharomyces cerevisiae

Nucleotide sequences proximal to the initiation codon of a gene are known to affect the expression efficiency of that gene. We screened 10-bp random sequences upstream of the initiation codon of the zeocin-resistance gene to identify sequences that could enhance its expression in Saccharomyces cerevisiae. Of the isolated sequences, 20 sequences exhibited a common feature, i.e. ATG at the position -9 through -7, which resulted in the incorporation of three amino acids at the N-terminus of the protein. The introduction of these sequences upstream of the initiation codon increased the expression levels of zeocin-resistance protein by 2.2-6.5-fold. One of these sequences increased the expression levels of three out of four human proteins, thereby suggesting that this sequence may also enhance the expression efficiency of mammalian proteins in yeast.

Mathematical modeling of translation initiation for the estimation of its efficiency to computationally design mRNA sequences with desired expression levels in prokaryotes

Background

Within the emerging field of synthetic biology, engineering paradigms have recently been used to design biological systems with novel functionalities. One of the essential challenges hampering the construction of such systems is the need to precisely optimize protein expression levels for robust operation. However, it is difficult to design mRNA sequences for expression at targeted protein levels, since even a few nucleotide modifications around the start codon may alter translational efficiency and dramatically (up to 250-fold) change protein expression. Previous studies have used ad hoc approaches (e.g., random mutagenesis) to obtain the desired translational efficiencies for mRNA sequences. Hence, the development of a mathematical methodology capable of estimating translational efficiency would greatly facilitate the future design of mRNA sequences aimed at yielding desired protein expression levels.

Results

We herein propose a mathematical model that focuses on translation initiation, which is the rate-limiting step in translation. The model uses mRNA-folding dynamics and ribosome-binding dynamics to estimate translational efficiencies solely from mRNA sequence information. We confirmed the feasibility of our model using previously reported expression data on the MS2 coat protein. For further confirmation, we used our model to design 22 luxR mRNA sequences predicted to have diverse translation efficiencies ranging from 10^-5 to 1. The expression levels of these sequences were measured in Escherichia coli and found to be highly correlated (R² = 0.87) with their estimated translational efficiencies. Moreover, we used our computational method to successfully transform a low-expressing DsRed2 mRNA sequence into a high-expressing mRNA sequence by maximizing its translational efficiency through the modification of only eight nucleotides upstream of the start codon.

Conclusions

We herein describe a mathematical model that uses mRNA sequence information to estimate translational efficiency. This model could be used to design best-fit mRNA sequences having a desired protein expression level, thereby facilitating protein over-production in biotechnology or the protein expression-level optimization necessary for the construction of robust networks in synthetic biology.

The translation of recombinant proteins in E. coli can be improved by in silico generating and screening random libraries of a -70/+96 mRNA region with respect to the translation initiation codon

Recombinant protein translation in Escherichia coli may be limited by stable (i.e. low free energy) secondary structures in the mRNA translation initiation region. To circumvent this issue, we have set-up a computer tool called ‘ExEnSo’ (Expression Enhancer Software) that generates a random library of 8192 sequences, calculates the free energy of secondary structures of each sequence in the −70/+96 region (base 1 is the translation initiation codon), and then selects the sequence having the highest free energy. The software uses this ‘optimized’ sequence to create a 5′ primer that can be used in PCR experiments to amplify the coding sequence of interest prior to sub-cloning into a prokaryotic expression vector. In this article, we report how ExEnSo was set-up and the results obtained with nine coding sequences with low expression levels in E. coli. The free energy of the −70/+96 region of all these coding sequences was increased compared to the non-optimized sequences. Moreover, the protein expression of eight out of nine of these coding sequences was increased in E. coli, indicating a good correlation between in silico and in vivo results. ExEnSo is available as a free online tool.

Development of expression vectors for Escherichia coli based on the pCR2 replicon

Background

Recent developments in metabolic engineering and the need for expanded compatibility required for co-expression studies, underscore the importance of developing new plasmid vectors with properties such as stability and compatibility.

Results

We utilized the pCR2 replicon of Corynebacterium renale, which harbours multiple plasmids, for constructing a range of expression vectors. Different antibiotic-resistance markers were introduced and the vectors were found to be 100% stable over a large number of generations in the absence of selection pressure. Compatibility of this plasmid was studied with different Escherichia coli plasmid replicons viz. pMB1 and p15A. It was observed that pCR2 was able to coexist with these E.coli plasmids for 60 generations in the absence of selection pressure. Soluble intracellular production was checked by expressing GFP under the lac promoter in an expression plasmid pCR2GFP. Also high level production of human IFNγ was obtained by cloning the h-IFNγ under a T7 promoter in the expression plasmid pCR2-IFNγ and using a dual plasmid heat shock system for expression. Repeated sub-culturing in the absence of selection pressure for six days did not lead to any fall in the production levels post induction, for both GFP and h-IFNγ, demonstrating that pCR2 is a useful plasmid in terms of stability and compatibility.

Conclusion

We have constructed a series of expression vectors based on the pCR2 replicon and demonstrated its high stability and sustained expression capacity, in the absence of selection pressure which will make it an efficient tool for metabolic engineering and co-expression studies, as well as for scale up of expression.

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.

Directed evolution of the 5'-untranslated region of the phoA gene in Escherichia coli simultaneously yields a stronger promoter and a stronger Shine-Dalgarno sequence

Directed evolution has been widely applied for gene improvement through random mutagenesis of coding sequences. Through error-prone PCR both in the coding sequence and the regulatory sequence of E. coli alkaline phosphatase, the cellular enzyme activity has been efficiently enhanced. Sequence analysis revealed that the resultant mutant 34-B12, which showed a sevenfold increased enzyme activity at the cellular level, contains three mutations in the regulatory sequence and another three mutations in the coding sequence. Activity assays of the enzyme containing the corresponding amino acid substitutions proved that the amino acid mutations contribute only to a small portion to the increased cellular enzyme activity. So the mutations in the 5'-untranslated region were analyzed separately and combinationally. The results suggested that one mutation yielded a stronger promoter and the other two mutations both elevated the E. coli alkaline phosphatase expression at the translational level; moreover, a stronger Shine-Dalgarno sequence was generated.

Regulation of translation via mRNA structure in prokaryotes and eukaryotes

The mechanism of initiation of translation differs between prokaryotes and eukaryotes, and the strategies used for regulation differ accordingly. Translation in prokaryotes is usually regulated by blocking access to the initiation site. This is accomplished via base-paired structures (within the mRNA itself, or between the mRNA and a small trans-acting RNA) or via mRNA-binding proteins. Classic examples of each mechanism are described. The polycistronic structure of mRNAs is an important aspect of translational control in prokaryotes, but polycistronic mRNAs are not usable (and usually not produced) in eukaryotes. Four structural elements in eukaryotic mRNAs are important for regulating translation: (i) the m7G cap; (ii) sequences flanking the AUG start codon; (iii) the position of the AUG codon relative to the 5' end of the mRNA; and (iv) secondary structure within the mRNA leader sequence. The scanning model provides a framework for understanding these effects. The scanning mechanism also explains how small open reading frames near the 5' end of the mRNA can down-regulate translation. This constraint is sometimes abrogated by changing the structure of the mRNA, sometimes with clinical consequences. Examples are described. Some mistaken ideas about regulation of translation that have found their way into textbooks are pointed out and corrected.

Improved solubility of TEV protease by directed evolution

The efficiency and high specificity of tobacco etch virus (TEV) protease has made it widely used for cleavage of recombinant fusion proteins. However, the production of TEV protease in E. coli is hampered by low solubility. We have subjected the gene encoding TEV protease to directed evolution to improve the yield of soluble protein. Libraries of mutated genes obtained by error-prone PCR and gene shuffling were introduced into the Gateway cloning system for facilitated transfer between vectors for screening, purification, or other applications. Fluorescence based in vivo solubility screening was carried out by cloning the libraries into a plasmid encoding a C-terminal GFP fusion. Mutant genes giving rise to high GFP fluorescence intensity indicating high levels of soluble TEV-GFP were subsequently transferred to a vector providing a C-terminal histidine tag for expression, purification, and activity tests of mutated TEV. We identified a mutant, TEV(SH), in which three amino acid substitutions result in a five-fold increase in the yield of purified protease with retained activity.