Systematic alteration of the nucleotide sequence preceding the translation initiation codon and the effects on bacterial expression of the cloned SV40 small-t antigen gene

Bicistronic mRNAs to enhance membrane protein overexpression

Functional overexpression of membrane proteins is essential for their structural and functional characterization. However, functional overexpression is often difficult to achieve, and frequently either no expression or expression as misfolded aggregates is observed. We present an approach for improving the functional overexpression of membrane proteins in Escherichia coli using transcriptional fusions. The method involves the use of a small additional RNA sequence upstream to the RNA sequence of the target membrane protein and results in the production of a bicistronic mRNA. In contrast to the common approach of translational fusions to enhance protein expression, transcriptional fusions do not require protease treatment and subsequent removal of the fusion protein. Using this strategy, we observed improvements in the quantity and/or the quality of the produced material for several membrane proteins to levels compatible with structural studies. Our analysis revealed that translation of the upstream RNA sequence was not essential for increased expression. Rather, the sequence itself had a large impact on protein yields, suggesting that alternative folding of the transcript was responsible for the observed effect.

Cost minimization of ribosomal frameshifts

Properties of mRNA leading regions that modulate protein synthesis are little known (besides effects of their secondary structure). Here I explore how coding properties of leading regions may account for their disparate efficiencies. Trinucleotides that form off frame stop codons decrease costs of ribosomal slippages during protein synthesis: protein activity (as a proxy of gene expression, and as measured in experiments using artificial variants of 5' leading sequences of beta galactosidase in Escherichia coli) increases proportionally to the number of stop motifs in any frame in the 5' leading region. This suggests that stop codons in the 5' leading region, upstream of the recognized coding sequence, terminate eventual translations that sometimes start before ribosomes reach the mRNA's recognized start codon, increasing efficiency. This hypothesis is confirmed by further analyses: mRNAs with 5' leading regions containing in the same frame a start preceding a stop codon (in any frame) produce less enzymatic activity than those with the stop preceding the start. Hence coding properties, in addition to other properties, such as the secondary structure of the 5' leading region, regulate translation. This experimentally (a) confirms that within coding regions, off frame stops increase protein synthesis efficiency by early stopping frameshifted translation; (b) suggests that this occurs for all frames also in 5' leading regions and that (c) several alternative start codons that function at different probabilities should routinely be considered for all genes in the region of the recognized initiation codon. An unknown number of short peptides might be translated from coding and non-coding regions of RNAs.

The Ambush Hypothesis: Hidden Stop Codons Prevent Off-Frame Gene Reading

Coding sequences lack stop codons, but many stops appear off-frame. Off-frame stops (stops in -1 and +1 shifted reading frames, termed hidden stops) terminate frame-shifted translation, potentially decreasing energy, and resource waste on nonfunctional proteins. Benefits may include reduced waste elimination costs and avoidance of potentially cytotoxic frame-shifted products. Our "ambush" hypothesis suggests that hidden stops are sometimes selected for. Codons of many amino acids can contribute to hidden stops, depending on the synonymous position state and adjacent codons. In vertebrate mitochondria, 31.75% of all amino acid combinations can form hidden stops. Codons with more potential to form hidden stops have greater usage frequency and bias in their favor among synonymous codons. Among primates, predicted mitochondrial rRNA secondary structure stability correlates negatively with the number of hidden stops in the mitochondrial genome. The taxonomic distribution of genetic codes suggests that +1 frameshifts might be more frequent than -1 frameshifts. This is confirmed by analyses of primate mitochondrial genomes: species with unstable rRNAs have more +1 stops, but the correlation is weak for -1 stops. High hidden stop density seems to be an adaptation in species with slippage prone ribosomes (unstable rRNAs). Hidden stops may thus compensate for reduced efficiency of some parts of the biosynthetic machinery. Some experimental data confirm our hypothesis: gene expression increases with the experimentally manipulated number of stops in the promoter region of a gene, suggesting biotechnological applications.

Production of engineered human pancreatic ribonucleases, solving expression and purification problems, and enhancing thermostability

Human pancreatic ribonuclease, the homolog of bovine pancreatic ribonuclease, has a significant therapeutic potential. Its study has been hindered by the difficulty of obtaining the enzyme in a pure and homogeneous form, either from human source or using heterologous expression. Engineering of different variants of human pancreatic ribonuclease has allowed us to study and overcome some problems encountered during its heterologous production in an Escherichia coli system and its purification from inclusion bodies. The 5'-end region of the mRNA that encodes the enzyme is critical for obtaining high expression levels. The results also suggest the importance of the proline 50 residue in the recovery yields of human pancreatic ribonuclease. All the variants produced are pure and homogeneous. Their homogeneity has been demonstrated by cation-exchange and reversed-phase chromatography and by mass spectrometry analysis. Moreover, enhancement of human pancreatic ribonuclease thermal stability is observed when residues R4, K6, Q9, D16, and S17 are changed to the corresponding residues of bovine seminal ribonuclease.

Strategies for achieving high-level expression of genes in Escherichia coli

Progress in our understanding of several biological processes promises to broaden the usefulness of Escherichia coli as a tool for gene expression. There is an expanding choice of tightly regulated prokaryotic promoters suitable for achieving high-level gene expression. New host strains facilitate the formation of disulfide bonds in the reducing environment of the cytoplasm and offer higher protein yields by minimizing proteolytic degradation. Insights into the process of protein translocation across the bacterial membranes may eventually make it possible to achieve robust secretion of specific proteins into the culture medium. Studies involving molecular chaperones have shown that in specific cases, chaperones can be very effective for improved protein folding, solubility, and membrane transport. Negative results derived from such studies are also instructive in formulating different strategies. The remarkable increase in the availability of fusion partners offers a wide range of tools for improved protein folding, solubility, protection from proteases, yield, and secretion into the culture medium, as well as for detection and purification of recombinant proteins. Codon usage is known to present a potential impediment to high-level gene expression in E. coli. Although we still do not understand all the rules governing this phenomenon, it is apparent that "rare" codons, depending on their frequency and context, can have an adverse effect on protein levels. Usually, this problem can be alleviated by modification of the relevant codons or by coexpression of the cognate tRNA genes. Finally, the elucidation of specific determinants of protein degradation, a plethora of protease-deficient host strains, and methods to stabilize proteins afford new strategies to minimize proteolytic susceptibility of recombinant proteins in E. coli.

Cloning and complete sequence of the DNA polymerase-encoding gene (BstpolI) and characterisation of the Klenow-like fragment from Bacillus stearothermophilus

A fragment of the DNA polymerase I-encoding gene (polI) from Bacillus stearothermophilus (Bst) was obtained by PCR. This was used as a probe to obtain a full-length gene from a Bst genomic DNA (gDNA) plasmid library. Comparison of the sequence to B. caldotenax (Bca) showed about 93% homology at the amino acid (aa) level. A Klenow-like (BstpolIk) clone was developed and the recombinant protein displayed DNA polymerase activity similar to the wild-type BstPolI enzyme.

High gene expression in Escherichia coli of recombinant alginate lyase as a fused protein with beta-galactosidase alpha-peptide

Escherichia coli LE392 (pAL28) was previously isolated as a positive clone harboring the alginate lyase gene (aly) from an alginate-degrading strain, Pseudomonas sp. OS-ALG-9. The plasmid pAL205, one of the constructs obtained after successive subcloning of pAL28, gave the highest expression of aly in E. coli cells. A 8-fold increase in the alginate lyase (Aly) activity in E. coli JM109 (pAL205) was induced with isopropyl-beta-D-thiogalactoside, which was 210 times higher than that in E. coli LE392 (pAL28). The highly significant increase in the expression of the Aly enzyme with pAL205 was investigated through the nucleotide sequence around the 5' region of aly as well as the N-terminal sequence of the purified enzyme. It was found that the Aly expressed in E. coli (pAL205) was a fused protein containing 7 residues from the N-terminus of beta-galactosidase alpha-peptide and the mature protein found in the Pseudomonas sp. except for three residues in the N-terminal.

Determination of the optimal aligned spacing between the Shine-Dalgarno sequence and the translation initiation codon of Escherichia coli mRNAs

The prokaryotic mRNA ribosome binding site (RBS) usually contains part or all of a polypurine domain UAAGGAGGU known as the Shine-Dalgarno (SD) sequence found just 5' to the translation initiation codon. It is now clear that the SD sequence is important for identification of the translation initiation site on the mRNA by the ribosome, and that as a result, the spacing between the SD and the initiation codon strongly affects translational efficiency (1). It is not as clear, however, whether there is a unique optimal spacing. Complications involving the definition of the spacing as well as secondary structures have obscured matters. We thus undertook a systematic study by inserting two series of synthetic RBSs of varying spacing and SD sequence into a plasmid vector containing the chloramphenicol acetyltransferase gene. Care was taken not to introduce any secondary structure. Measurements of protein expression demonstrated an optimal aligned spacing of 5 nt for both series. Since aligned spacing corresponds naturally to the spacing between the 3'-end of the 16S rRNA and the P-site, we conclude that there is a unique optimal aligned SD-AUG spacing in the absence of other complicating issues.

Potential secondary structure at the translational start domain of eukaryotic and prokaryotic mRNAs

In order to identify conserved potential secondary structures within translational start sites, mRNA sequences derived from different species were studied with programs able to depict such features. The potential secondary structure of 71 bases around the initiator AUG or AUGs in the coding sequences of 290 eukaryotic mRNAs was first examined and compared to 290 similarly analyzed regions derived from prokaryotic mRNA sequences (Nucleic Acids Res (1987) 15, 345-360). In both sets of sequences the initiator codon was often found to be in an open potential structure whereas a denser region characterized by nearly-periodic spacings defined the coding regions. Randomization of the sequences obliterated the observed patterns suggesting that the structure of the mRNA may determine these differences. Three sets of eukaryotic and prokaryotic mRNAs of approximately equal length were analyzed and found to preserve an open unpaired non-coding region 5' to the start codon. The start codon was found free of potential secondary structure in over 80% of all the sequences analyzed. These data, and study of mutants that restrict the accessibility of the start codon to the ribosomal initiation complex, suggest that both the prokaryotic and eukaryotic mRNA start sites must occur free of potential secondary structure for efficient initiation. A striking difference of the eukaryotic mRNA sequences analyzed was the high propensity of the coding region vicinal to the start codon to form secondary structures. Certain translation-defective mutants exhibit impaired formation of these secondary structures suggesting that the structure of the coding regions adjacent to the start codons of eukaryotic mRNAs may be an important, thus far unexamined, determinant of initiation. We propose that, for all genes studied, the transition in secondary structure between the coding and non-coding regions may be an important determinant of initiation.

The influence of ribosome-binding-site elements on translational efficiency in Bacillus subtilis and Escherichia coli in vivo

A method is described to determine simultaneously the effect of any changes in the ribosome-binding site (RBS) of mRNA on translational efficiency in Bacillus subtilis and Escherichia coli in vivo. The approach was used to analyse systematically the influence of spacing between the Shine-Dalgarno sequence and the initiation codon, the three different initiation codons, and RBS secondary structure on translational yields in the two organisms. Both B. subtilis and E. coli exhibited similar spacing optima of 7-9 nucleotides. However, B. subtilis translated messages with spacings shorter than optimal much less efficiently than E. coli. In both organisms, AUG was the preferred initiation codon by two- to threefold. In E. coli GUG was slightly better than UUG while in B. subtilis UUG was better than GUG. The degree of emphasis placed on initiation codon type, as measured by translational yield, was dependent on the strength of the Shine-Dalgarno interaction in both organisms. B. subtilis was also much less able to tolerate secondary structure in the RBS than E. coli. While significant differences were found between the two organisms in the effect of specific RBS elements on translation, other mRNA components in addition to those elements tested appear to be responsible, in part, for translational species specificity. The approach described provides a rapid and systematic means of elucidating such additional determinants.

Lack of a 5' non-coding region in Tn1721 encoded tetR mRNA is associated with a low efficiency of translation and a short half-life in Escherichia coli

The repressor-encoding tetR gene from Tn1721 is expressed with a very low efficiency. Its mRNA lacks an untranslated leader sequence. We have constructed protein fusions with the lacZ gene which contain between 14 and 157 5' nucleotides from the tetR gene. Since they are all expressed with similar efficiencies we conclude that the sequence information for initiation of translation is contained within the first 14 bases of the tetR coding region. These fusion transcripts are about 20-fold less efficiently translated than the wild type lacZ transcript. A toeprint analysis confirms that the initiation complex is indistinguishable from those formed by regular transcripts with 5' untranslated regions but occurs in a very low amount in vitro. Thus, the absence of a 5' leader causes a poor rate of translation initiation. The half-lives of tetR and tetR-lacZ mRNAs are about 30 seconds, which is 3-times lower than that of the wt lacZ mRNA. Inactivation of the ams/rne locus in E. coli stabilizes the tetR transcript more than ten-fold. The influence of translation on the tetR half-life is discussed.

Functional expression of the dihydrofolate reductase and thymidylate synthetase activities of the human malaria parasite Plasmodium falciparum in Escherichia coli

We have developed a recombinant system that directs the functional expression from Escherichia coli of both dihydrofolate reductase-thymidylate synthetase (DHFR-TS) and the isolated DHFR domain from Plasmodium falciparum. Both products are inhibitory to a number of E. coli cell lines to the extent that cell growth ceases immediately upon induction. This dramatic inhibition is not seen in strain AB1899, in which amounts of plasmodial protein of up to 100 times the basal E. coli TS level can be accumulated. However, as well as the full-length DHFR-TS molecule, smaller proteins carrying an intact TS substrate-binding site are produced. These represent ca. 60-75% of the total plasmodial protein expressed and are observed in every E. coli strain examined. We show that they are not derived by degradation of the parent DHFR-TS molecule, but can be correlated with the sizes of proteins expected to be produced if erroneous initiation of translation were occurring at 3 internal methionine residues.

Effects of gene dosage on the expression of human growth hormone cDNA in Escherichia coli

Gene dosage effects on the foreign protein production in Escherichia coli have been analyzed with the expression vectors containing one, two and four tandemly arranged human growth hormone cDNAs under various culture conditions. In a limited synthetic culture medium, a proportional relationship was observed between the number of the integrated cDNA and the amount of human growth hormone produced. With a culture medium rich in the nutrients, however, no such relationship was observed. These results suggest that, under the limited culture conditions, high-level expression of a foreign gene can be achieved by tandemly joined multiple copies of cDNA.

An additional ribosome-binding site on mRNA of highly expressed genes and a bifunctional site on the colicin fragment of 16S rRNA from Escherichia coli: important determinants of the efficiency of translation-initiation

For various genes of E. coli, three regions (-55 to -1; -35 to -1; -21 to -1) 5' to AUG codon on mRNA were searched for sites of interaction with colicin fragment of 16S rRNA. The detailed sequence comparison points out that apart from Shine-Dalgarno base pairing, an additional ribosome-binding site, a subsequence of 5'-UGAUCC-3' invariably exists in mRNA for highly expressed genes. Poorly expressed genes appear to be controlled by only Shine-Dalgarno base pairing. The analysis indicates that in the initiator region, the -55 to -1 region contains the signal which decides the efficiency of the translation-initiation. The site on 16S rRNA, 5'-GGAUCA-3' at position 1529, that can base pair to the above site, has a recognition site on 23S rRNA at position 2390. In the light of the conserved nature and accessibility of these sites, it is proposed that the site on 16S rRNA plays a bifunctional role--initially it binds to mRNA from highly expressed genes to form a stable 30S initiation complex, and upon association with 50S subunit it exchanges base pairing with 23S rRNA, thus leaving the site on mRNA free.

A translational enhancer derived from tobacco mosaic virus is functionally equivalent to a Shine-Dalgarno sequence

When present at the 5' end of mRNAs, the untranslated leader sequence (omega) of tobacco mosaic virus RNA significantly enhances translation in eukaryotes and prokaryotes. We have tested a deletion derivative of the omega sequence, omega delta 3, for its enhancing ability on gene constructs in which the ribosomal binding site was either present or deleted, in several Gram-negative bacterial species including Escherichia coli, Agrobacterium tumefaciens, Xanthomonas campestris pv. vitians, Erwinia amylovora, and Salmonella typhimurium. In vivo production of chloramphenicol acetyltransferase from a gene construct lacking its native ribosomal binding site was enhanced 40- to 120-fold by the presence of omega delta 3. Similar levels of enhancement (30- to 240-fold) were observed when the gene encoding beta-glucuronidase was tested. With a chloramphenicol acetyltransferase construct containing a ribosomal binding site, enhancement was markedly less, between 1- and 3.8-fold. Omega delta 3 appeared to enhance translation independent of its position upstream of the AUG codon used for initiation.

Point mutations that affect translation initiation of the transposon Tn10 tetA gene

Single base-pair changes affecting the 5' untranslated leader of the tetracycline-resistance gene (tetA) mRNA resulted in increased or decreased levels in expression of tetA-lacZ protein fusions. The base changes affected the rate of initiation of translation of tetA mRNA because operon fusions revealed that the mutations had little or no effect on transcription. The translational efficiency of wild-type and mutant tetA mRNAs varied by more than a factor of 2000. The observed variations could be correlated with stabilization and destabilization of RNA secondary structures. These structures, located 5' from the translation start codon, sequester the Shine-Dalgarno sequence.

The T7 phage gene 10 leader RNA, a ribosome-binding site that dramatically enhances the expression of foreign genes in Escherichia coli

Expression of foreign genes in Escherichia coli requires the juxtaposition of prokaryotic transcription and translation elements with a coding region for the foreign gene. Commonly, this results in only modest expression of the foreign gene product. Here we describe a novel ribosome-binding site (RBS; phage T7 'gene 10 leader') which is able to drive the translation of several foreign genes. This RBS dramatically enhanced the translation efficiency of all the genes we have tested to date, and was particularly effective for foreign genes. The enhanced expression was often more than 40-fold greater than that obtained using a 'consensus' RBS. A general plasmid vector has been constructed, incorporating the T7 gene 10 leader sequence, which allows the facile expression of important gene products. In this report we demonstrate the application of this system for the high-level expression of plant, mammalian and bacterial proteins in E. coli.

Design, synthesis and expression of a human interleukin-2 gene incorporating the codon usage bias found in highly expressed Escherichia coli genes

A synthetic gene encoding human interleukin-2 (IL-2) was designed such that the codon usage bias resembled that found in highly expressed Escherichia coli genes. The percentage of preferred codons was increased from 43% in the native cDNA sequence to 85% in the synthetic sequence. The cDNA and synthetic IL-2 genes were placed under the control of the trc promoter and expressed in E. coli JM101. While Northern blot analysis of IL-2 mRNA from each genetic construct demonstrated equivalent message half-lives, immunoblot and bioactivity analyses showed the synthetic gene to direct the synthesis of up to 16 times more IL-2 than the native cDNA sequence.

Sequence changes preceding a Shine-Dalgarno region influence trpE mRNA translation and decay

In studies with a trpE promoter-strength measuring system we observed that constructs containing the Escherichia coli trp promoter and its adjacent transcribed region yielded lower levels of trpE protein than were expected. To analyze this observation we introduced mutational changes in the nucleotide sequence preceding the trpE Shine-Dalgarno region and examined their effects on trpE mRNA synthesis, translation and decay. We found that certain deletion, insertion and substitution mutations in the pre-Shine-Dalgarno region caused a two- to fivefold increase in trpE enzyme activity. These increases were accompanied by increases in steady-state levels of trpE mRNA. Pulse-chase analyses of trpE mRNA degradation revealed that the observed steady-state trpE mRNA levels correlated with changes in trpE mRNA stability. These findings are interpreted in terms of alternative models in which the primary effect of mutational changes that elevate trpE expression is to increase trpE mRNA translation, versus increasing trpE mRNA stability.

Search for the optimal sequence of the ribosome binding site by random oligonucleotide-directed mutagenesis

Synthetic DNA duplexes corresponding to the ribosome binding site (RBS) were synthesized through the phosphite method on solid support. The synthetic RBS DNA with partial random sequences was inserted into an appropriate site between the lpp-lac promoter and the beta-galactosidase structural gene in plasmid pMKT2. The level of beta-galactosidase expression was correlated with the color intensity of the recombinant colonies on X-gal plates. The bluest colonies were isolated and characterized with respect to beta-galactosidase enzyme activity and RBS sequence. There was good correlation between color intensity and the level of the enzyme activity, and this provided a reliable phenotypic screening method in the search for the optimal regulatory sequences. Novel RBS sequences obtained here show not only the unique nucleotide distribution, but also strong complemetarity to the 3' end region of 16S rRNA, from which could be deduced a generalized RBS sequence, the position of the SD region, and the 16S rRNA position mediated during translation initiation.

Chemical synthesis, cloning and expression in mammalian cells of a gene coding for human tissue-type plasminogen activator

A 1610-bp DNA duplex coding for human tissue-type plasminogen activator has been chemically synthesized using the phosphoramidite procedure, adapted for a custom-built gene synthesizer. The synthesizer, which was designed for both simplicity and speed, permits the rapid construction of relatively large genes and compares favorably in speed with alternative cDNA isolation procedures. The plasminogen activator gene has been expressed in mammalian cells and shown to produce authentic protein by an immuno-activity assay.

Minimal energy foldings of eukaryotic mRNAs form a separate leader domain

We have investigated the minimal energy foldings of 38 mature mRNAs, including the globin family, the insulins, the growth hormones and interleukin-2, and have compared these foldings with those of fully and partly randomised sequences. The mRNAs differ from the random sequences in that they form a separate leader hairpin of 40-60 nucleotides, with the initiation codon typically located downstream of this hairpin, followed by a main fold in which a region flanking the initiation codon is basepaired with the trailer: resulting in a close proximity of the 5' and 3' end of the mRNA. The formation of this conformation depends not only--or primarily--on the structure of the leader, but on both the leader and trailer sequence and their interaction with the coding sequence. Thus if, as the frequent occurrence of this pattern suggests, the secondary structure of the leader regions plays a role in the initiation of translation, possibly accounting for the specificity of initiation and the different translational efficiencies of various mRNAs, we expect that these features may be influenced both by leader and trailer mutants.

Overexpression and DNA-binding properties of the mer-encoded regulatory protein from plasmid NR1 (Tn21)

In plasmid NR1 the expression of genes involved in mercury resistance (Tn21) is regulated by the trans-acting product of the merR gene. An in vivo T7 RNA polymerase-promoter overexpression system was used to detect a protein of approximately 16,000 daltons encoded by the merR reading frame. Overexpressed MerR constituted about 5% of labeled proteins. An in vitro MerR-mer-op (mer-op is the mer operator and promoter region) gel electrophoresis binding assay established that the binding site for MerR was located between the putative -35 and -10 sequences of the promoter for the mer structural genes. A nonsense mutation in the carboxyl half of MerR resulted in the loss of biological function and the loss of in vitro mer-op binding properties.

Spacer alterations which increase the expression of porcine growth hormone in E. coli

Full-length porcine growth hormone (PGH) cDNA clones were isolated from a porcine pituitary cDNA library. When the coding portion of the PGH gene was cloned into an E. coli expression vector downstream from the powerful trc promoter, high levels of mRNA, but no protein were detected. Mutation directed by an oligodeoxynucleotide primer altered 5'-non-coding sequences and raised the level of PGH produced from undetectable to 15% of the total cellular protein. Alteration of four codons infrequently used by E. coli in the 5'-end of the gene produced no further increases.

Potential secondary structure at translation-initiation sites

Since translational start codons also occur internally, more-complex features within mRNA must determine initiation. We compare the potential secondary structure of 123 prokaryotic mRNA start regions to that of regions coding for internal methionines. The latter display an unexpectedly-uniform, almost-periodic pattern of pairing potential. In contrast, sequences 5' to start codons have little self-pairing, and do not pair extensively with the proximal coding region. Pairing potential surrounding start codons was found to be less than half of that found near internal AUGs. In groups of random sequences where the distribution of nucleotides at each position, or of trinucleotides at each in-frame codon position, matched the observed natural distribution, there was no periodicity in the pairing potential of the internal sequences. Randomized internal sequences had less pairing: the ratio of pairing intensity between internals and starts was reduced from 2.0 to 1.6 by randomization. We propose that the transition from the relatively-unstructured start domains to the highly-structured internal sequences may be an important determinant of translational start-site recognition.

Expression of Lassa virus nucleocapsid protein segments in bacteria: purification of high-level expression products and their application in antibody detection

The Lassa virus nucleocapsid protein gene and segments from it were expressed in Escherichia coli under the control of the lac promoter in pUC-based plasmids. Expression of the near full-length protein [amino acid (aa) residues 12-570] fused to an N-terminal sequence of vector-derived 6 aa was not particularly efficient, and neither was that of a smaller N-terminal segment (aa 6-201) which was also fused at its C terminus to the remainder of the lacZ gene product. By contrast, the C-terminal 370 aa could be expressed at levels approaching 10% of total cellular protein. All the recombinant proteins were associated with the insoluble fraction after sonication of the bacteria. The inefficiently expressed products did not appear to be any more susceptible to proteolytic degradation. The distribution of codons rarely used in E. coli genes was relatively uniform along the nucleocapsid gene sequence. These results are consistent with the regulation of transcriptional or translational efficiency by features of the sequence downstream from the promoter and ribosome-binding site. The C-terminal segment (aa 201-570 representing 65% of the authentic protein) was purified by ion exchange chromatography and shown to be active when used as antigen in enzyme-linked immunoassays for virus-specific antibodies.

A 'phase-shift' fusion system for the regulation of foreign gene expression by lambda repressor in gram-negative bacteria

A 'phase-shift' translation fusion vector was constructed in which mutually compatible restriction sites BamHI, BclI and BglII are positioned in such a manner that the cut point is in a different reading frame, immediately following the ATG start codon and ribosome-binding site of the lambda cro gene. The lambda cro gene is expressed from promoter pR and controlled by a thermosensitive (cI857) lambda repressor. The usefulness of the expression vector was demonstrated using a galK gene lacking the ATG start codon and fusing this to the pR promoter and ATG start codon of the lambda cro gene, resulting in cI857-regulated expression of galactokinase. The vector is of general use for foreign gene expression in Escherichia coli when the target gene has a compatible cohesive end (5'-GATC-3') at the N terminus (provided, for example, by a BamHI linker). The lambda cI857-pR-cro-galK cassette was cloned into pJRD215, a wide-host-range plasmid and transferred by conjugation to a variety of Gram-negative bacteria. In all cases, thermosensitive regulation of galactokinase could be demonstrated, though the levels of induction varied considerably. These results show that the powerful lambda pR promoter and the efficient lambda repressor can be used to regulate expression of foreign genes in Gram-negative organisms other than E. coli.

RNA structural elements for expression in Escherichia coli. Alpha 1-antitrypsin synthesis using translation control elements based on the cII ribosome-binding site of phage lambda

Analysis of a series of lambda cII::alpha 1-antitrypsin (alpha 1AT) gene fusions of different sizes showed that increased alpha 1AT expression correlated with the stabilisation of a particular computer-predicted RNA secondary structure. Moreover, significant synthesis of unfused alpha 1AT was achieved by reconstruction of this conformation to permit interaction between the upstream region of the ribosome-binding site and the first part of the alpha 1AT coding sequence. This high-level expression was dependent upon certain silent point mutations in the coding sequence, indicating that RNA primary and secondary structure determinants can operate in concert to dictate the efficiency of protein synthesis.

Translation of a synthetic two-cistron mRNA in Escherichia coli

A synthetic two-cistron expression system was constructed for the high-level expression of eukaryotic genes in Escherichia coli. This system was designed to overcome translational inhibition of mRNAs containing eukaryotic sequences. The first cistron in this system is a 31-base A + T-rich synthetic sequence that provides for efficient translation initiation. The second cistron contains the protein coding sequence for the eukaryotic gene. Insertion of the first cistron between the 5' untranslated region of the mRNA and the protein coding region separates the two and thereby potentially minimizes the formation of local secondary structures that might prevent ribosomes from binding and initiating translation. The 31-base cistron contains three nonsense codons (TAA), one in each of the three translational reading frames, and an 8-base Shine-Dalgarno sequence that is complementary to the 3' end of the 16S rRNA. The effects of translation of the first cistron in all three reading frames on the expression of the second cistron was examined. The most efficient expression of the second cistron seemed to occur when the stop codon that terminates translation of the first cistron is located 3' to the Shine-Dalgarno sequence and close to the AUG start codon for the second cistron. When the Shine-Dalgarno sequence was deleted from the first cistron, no detectable expression of the second cistron was observed. This two-cistron system has been used to express the gene encoding methionylalanyl bovine growth hormone with its native codons and the gene encoding methionyl human growth hormone at a level greater than 20% of total cell protein. In the case of human growth hormone, we show that the amount of gene product is not significantly diminished by placing a "functional" first cistron in front of a gene that can be expressed without a cistron.

A mutation allowing an mRNA secondary structure diminishes translation of Saccharomyces cerevisiae iso-1-cytochrome c

The CYC1-239-O mutation in the yeast Saccharomyces cerevisiae produces a -His-Leu- replacement of the normal -Ala-Gly- sequence at amino acid positions 5 and 6, which lie within a dispensable region of iso-1-cytochrome c; this mutation can accommodate the formation of a hairpin structure at the corresponding site in the mRNA. The amount of the altered protein was diminished to 20% of the wild-type level, whereas the amount of the mRNA remained normal. However, in contrast to the normal CYC1+ mRNA that is associated mainly with four to seven ribosomes, the bulk of the CYC1-239-O mRNA is associated with one to four ribosomes. These results suggest that the stable secondary structure within the translated region of the CYC1 mRNA diminishes translation by inhibiting elongation.

Bacillus thuringiensis var. israelensis delta-endotoxin. Nucleotide sequence and characterization of the transcripts in Bacillus thuringiensis and Escherichia coli

The nucleotide sequence of a 1408 base-pair DNA fragment encoding the insecticidal 27,340 Mr delta-endotoxin of Bacillus thuringiensis var. israelensis has been determined by analysis of a recombinant plasmid from Escherichia coli. The hydropathy plot of the protein shows it to be highly hydrophobic, consistent with a postulated cytolytic mechanism of action for the toxin. In addition, the delta-endotoxin transcriptional start points that are used in B. thuringiensis and an E. coli recombinant have been determined. In B. thuringiensis var. israelensis, transcription initiates from a single start point, and gene-specific transcripts are not observed before stage II of sporulation. This is the stage at which delta-endotoxin antigen is first detected, indicating that control of expression is primarily at the transcriptional level for this protein. Analysis of gene-specific transcription in E. coli indicates that at least three start points are utilized in this organism. Interestingly, the highest level of delta-endotoxin mRNA is seen during mid-exponential growth of E. coli and the level appears to decrease as the cells enter the stationary phase of growth.

Use of a portable ribosome-binding site for maximizing expression of a eukaryotic gene in Escherichia coli

To maximize expression of a eukaryotic gene in Escherichia coli, a series of plasmids were constructed containing various synthetic ribosome-binding sites (RBS). These sites consist of a Shine-Dalgarno (SD) region (with translation stop codons in all three reading frames) positioned at distances 5-9 nucleotides (nt) from the AUG initiator codon of the gene coding for human T-cell growth factor (TCGF or IL-2). The region encompassing the RBS through the TCGF structural gene from each of these plasmids was inserted as a 'cassette' into seven different E. coli expression vectors, and TCGF production was measured. Our results demonstrate a greater than 2000-fold range of TCGF synthesis dependent upon the promoter and the synthetic RBS used. The translational efficiency of the TCGF gene was found to be influenced by the quality of the RBS, which is in part determined by the external sequence context of this site. The synthetic RBS, containing the necessary information for the translation initiation process, readily accessible by restriction sites, should be of general usefulness in obtaining maximum expression of eukaryotic genes in E. coli.

Binding of pea cytochrome f to the inner membrane of Escherichia coli requires the bacterial secA gene product

Various sequences from the 5' end of the pea chloroplast gene for cytochrome f have been fused in the correct reading frame with lacZ, and the cellular location of the hybrid polypeptides in Escherichia coli has been examined. Hybrid polypeptides containing N-terminal parts of cytochrome f are located in the cytoplasmic membrane of E. coli. Membrane localization is most efficient when the intact signal sequence of cytochrome f is present at the N-terminal end of the fusion proteins. Fusion within the signal sequence, so that the processing site is absent, reduces the efficiency of membrane binding. Membrane insertion of fusion proteins containing signal sequences is prevented in a temperature-sensitive secA strain at the nonpermissive temperature and the hybrid proteins accumulate in the cytoplasm. This indicates that specific recognition of the chloroplast signal sequence occurs in the bacterial secretory pathway.

A mutation allowing an mRNA secondary structure diminishes translation of Saccharomyces cerevisiae iso-1-cytochrome c

The CYC1-239-O mutation in the yeast Saccharomyces cerevisiae produces a -His-Leu- replacement of the normal -Ala-Gly- sequence at amino acid positions 5 and 6, which lie within a dispensable region of iso-1-cytochrome c; this mutation can accommodate the formation of a hairpin structure at the corresponding site in the mRNA. The amount of the altered protein was diminished to 20% of the wild-type level, whereas the amount of the mRNA remained normal. However, in contrast to the normal CYC1+ mRNA that is associated mainly with four to seven ribosomes, the bulk of the CYC1-239-O mRNA is associated with one to four ribosomes. These results suggest that the stable secondary structure within the translated region of the CYC1 mRNA diminishes translation by inhibiting elongation.

High-level expression in Escherichia coli of biologically active bovine growth hormone

High-level synthesis of bovine growth hormone (bGH) in Escherichia coli was achieved by maximizing gene transcription and optimizing the translational efficiency of bGH mRNA. Nearly all of the recombinant hormone was found in the pellet fraction after bacterial cell lysis. This property allowed the purification of bGH nearly to homogeneity. Protein sequence analysis indicated that greater than 93% of the purified hormone had the amino-terminal methionine residue removed by E. coli, yielding mature bGH. In a hypophysectomized rat assay system, purified bacterial-produced bGH demonstrated growth-promoting activity equivalent to that of pituitary-derived bovine growth hormone.