Codon optimization of bacterial luciferase (lux) for expression in mammalian cells

Expression of the bacterial luciferase (lux) system in mammalian cells would culminate in a new generation of bioreporters for in vivo monitoring and diagnostics technology. Past efforts to express bacterial luciferase in mammalian cells have resulted in only modest gains due in part to low overall expression of the bacterial genes. To optimize expression, we have designed and synthesized codon-optimized versions of the luxA and luxB genes from Photorhabdus luminsecens. To evaluate these genes in vivo, stable HEK293 cell lines were created harboring wild type luxA and luxB (WTA/WTB), codon-optimized luxA and wild type luxB (COA/WTB), and codon-optimized versions of both luxA and luxB genes (COA/COB). Although mRNA levels within these clones remained approximately equal, LuxA protein levels increased significantly after codon optimization. On average, bioluminescence levels were increased by more than six-fold [5×10⁵ vs 2.9×10⁶ relative light units (RLU)/mg total protein] with the codon-optimized luxA and wild type luxB. Bioluminescence was further enhanced upon expression of both optimized genes (2.7×10⁷ RLU/mg total protein). These results show promise toward the potential development of an autonomous light generating lux reporter system in mammalian cells     

Enhanced stable expression of aVibrio luciferase under the control of the Ω-translational enhancer in transgenic plants

A fusion gene usingluxA andluxB genes ofVibrio species has been designed to express light autonomously in plants.LuxA:luxB was introduced into plants by a high-efficiency transformation system consisting of a high-copy virulence helper plasmid pUCD2614 and T-vector pUCD2715 containingluxA:luxB. The expression ofluxA:luxB fusion gene was optimized by adjusting the spacing between the genes and by placing the translational efficiency of its mRNA under the control of the -3 translational enhancer. The resulting transgenic plants synthesized luciferase at levels greater than 1% of the total leaf protein. These plants produced light autonomously and light intensity was enhanced by the addition of aldehyde. That theluxA:luxB fusion has been optimized enables its use as a reporter for gene activity in plants during development and under various stress-inducing conditions. These results show that a specific protein from an introduced foreign gene can be produced with high efficiency in cultivated plants and such a system is therefore amenable for production of desired proteins through conventional farming methods.     

Fusion of luxA and luxB and its expression in E. coli,S. cerevisiae and D. melanogaster

Luciferase from Vibrio harveyi is encoded by two adjacent genes, luxA and luxB. The two genes were fused by replacing a segment extending from near the end of luxA into the N-terminal end of luxB by a synthetic oligonucleotide. The construction removed the TAA stop codon at the end of luxA, the intervening region of 26 base pairs, and the initial methionine of luxB. A Smal site was included at the junction between the two genes and an Aatll site was created near the end of luxA without altering its amino acid sequence. In Escherichia coli the fused luxAB gene could be expressed to produce functional luciferase that gave about 20% of the activity in cells without the fusion. An out-of frame ATG exists close to and preceding the ATG of the luxA gene. This was removed and the entire fused gene bracketed by several restriction enzyme sites. The fused luxAB gene was successfully expressed in Saccharomyces cerevisiae and Drosophila melanogaster by transferring it to appropriate plasmid vectors.     

Mobilization of cloned luciferase genes into Vibrio harveyi luminescence mutants

A recombinant plasmid which carried a 5 kb fragment of Vibrio harveyi DNA containing the luxA and luxB genes was mobilized from Escherichia coli into luminescence-deficient mutants of V. harveyi. The cloned genes complemented a temperature sensitive luciferase mutation, but failed to complement lesions in two different aldehyde deficient mutants. Expression of the cloned genes was not subject to autoinduction in either E. coli or in V. harveyi.     

Experimental Evidence for the Physiological Role of Bacterial Luciferase in the Protection of Cells Against Oxidative Stress

The origin and function of bioluminescence was considered a problematic question of the Charles Darwin theory. Early evolution of bacterial luminescence and its current physiological importance seem to be especially mysterious. Recently, it was proposed that stimulation of DNA repair may be a physiological role for production of light by bacterial cells. On the other hand, it was also proposed that primary role of luminescent systems could be detoxification of the deleterious oxygen derivatives. Although some previous results might suggest that this hypothesis can be correct, until now experimental evidence for such a mechanism operating in bacterial cells and having physiological importance was generally lacking. Here we demonstrate that in the presence of various oxidants (hydrogen peroxide, cumene hydroperoxide, t-butyl hydroperoxide, and ferrous ions) at certain concentrations in the culture medium, growth of Vibrio harveyi mutants luxA and luxB, but not of the mutant luxD, is severely impaired relative to wild-type bacteria. This deleterious effect of oxidants on the mutants luxA and luxB could be significantly reduced by addition of the antioxidants A-TEMPO or 40H-TEMPO. We conclude that bacterial luciferase may indeed play a physiological role in the protection of cells against oxidative stress.     

High level expression and single-step purification of hexahistidine-tagged L-2-hydroxyisocaproate dehydrogenase making use of a versatile expression vector set

Affinity tags as fusions to the N- or C-terminal part of proteins are valuable tools to facilitate the production and purification of proteins. In many cases, there may be the necessity to remove the tag after protein preparation to regain activity. Removal of the tag is accomplished by insertion of a unique amino acid sequence that is recognized and cleaved by a site specific protease. Here, we report the construction of an expression vector set that combines N- or C-terminal fusion to either a hexahistidine tag or Streptag with the possibility of tag removal by factor Xa or recombinant tobacco etch virus protease (rTEV), respectively. The vector set offers the option to produce different variants of the protein of interest by cloning the corresponding gene into four different Escherichia coli expression vectors. Either immobilized metal affinity chromatography or streptactin affinity chromatography can be used for the one-step purification. Furthermore, we show the successful application of the expression vector for C-terminal hexahistidine tagging. The expression and purification of His-tagged L-2-hydroxyisocaproate dehydrogenase yields fully active enzyme. The tag removal is here accomplished by a derivative of rTEV.     

A site-directed mutant of Cu,Zn-superoxide dismutase modeled after native extracellular superoxide dismutase

The well-studied cytosolic Cu,Zn-superoxide dismutase (SOD) protects against reperfusion injury, although its short (6 min) plasma half-life and negative charge create undesirable pharmacokinetics. We have designed, cloned, and expressed a genetic variant of SOD with altered pharmacological properties. A fusion gene consisting of the entire coding region of human SOD followed by a positively charged carboxy-terminal (C-terminal) “tail” of eight glycine and six arginine residues was constructed. The tail was modeled after the extracellular SOD (EC-SOD) C-terminal 26-amino acid basic peptide. This EC-SOD tail binds to heparin-like proteoglycans on cell surfaces and contributes to the enzyme’s very long (30 h) plasma clearance time. After expression inEscherichia coli, the mutant enzyme was purified and characterized. No differences in specific activity or UV absorption spectrum between the mutant and the native enzyme were found. The thermal stability of the fusion protein was greater than that of native SOD. Although native SOD has no affinity for heparin, the modified enzyme bound to a heparin-agarose column. A “designer” SOD able to bind to cell surfaces may aid in the prevention of superoxide-mediated endothelial damage.     

New Unstable Variants of Green Fluorescent Protein for Studies of Transient Gene Expression in Bacteria

Use of the green fluorescent protein (Gfp) from the jellyfish Aequorea victoria is a powerful method for nondestructive in situ monitoring, since expression of green fluorescence does not require any substrate addition. To expand the use of Gfp as a reporter protein, new variants have been constructed by the addition of short peptide sequences to the C-terminal end of intact Gfp. This rendered the Gfp susceptible to the action of indigenous housekeeping proteases, resulting in protein variants with half-lives ranging from 40 min to a few hours when synthesized in Escherichia coli and Pseudomonas putida. The new Gfp variants should be useful for in situ studies of temporal gene expression.     

Relationship of the luminous bacterial symbiont of the Caribbean flashlight fish,Kryptophanaron alfredi (family Anomalopidae) to other luminous bacteria based on bacterial luciferase (luxA) genes

Flashlight fishes (family Anomalopidae) have light organs that contain luminous bacterial symbionts. Although the symbionts have not yet been successfully cultured, the luciferase genes have been cloned directly from the light organ of the Caribbean species, Kryptophanaron alfredi. The goal of this project was to evaluate the relationship of the symbiont to free-living luminous bacteria by comparison of genes coding for bacterial luciferase (lux genes). Hybridization of a luxAB probe from the Kryptophanaron alfredi symbiont to DNAs from 9 strains (8 species) of luminous bacteria showed that none of the strains tested had lux genes highly similar to the symbiont. The most similar were a group consisting of Vibrio harveyi, Vibrio splendidus and Vibrio orientalis. The nucleotide sequence of the luciferase subunit gene luxA of the Kryptophanaron alfredi symbiont was determined in order to do a more detailed comparison with published luxA sequences from Vibrio harveyi, Vibrio fischeri and Photobacterium leiognathi. The hybridization results, sequence comparisons and the mol% G+C of the Kryptophanaron alfredi symbiont luxA gene suggest that the symbiont may be considered as a new species of luminous Vibrio related to Vibrio harveyi.The nucleotide sequence reported in this article has been deposited in Genbank under accession number M36597     

Optimal conditions for levan formation by an overexpressed recombinant levansucrase

Summary A genetically modified levansucrase, which contained His-affinity tag in its C-terminal, was constructed by PCR reaction using two synthetic primers. This modified protein was produced up to 30 % in total cell protein of E. coli, and was purified by a one-step affinity chromatography. The optimum pH for levan production was pH 5 and the optimum temperature was 0 °C. The higher velocity of levan formation within shorter enzyme reaction times was achieved by increasing the levels of enzyme concentration. The optimal sucrose concentration for levan production was around 20 %. Under these conditions, more than 50 g levan/l was produced.     

Tailoring of recombinant FDH: effect of histidine tag location on solubility and catalytic properties of Chaetomium thermophilum formate dehydrogenase (CtFDH)

Abstract

Several protein expression systems can be used to get enzymes in required quantities and study their functions. Incorporating a polyhistidine tag is a beneficial way of getting various enzymes such as FDHs for industrial applications. The NAD⁺ dependent formate dehydrogenase from Chaetomium thermophilum (CtFDH) can be utilized for interconversion of formate to carbon dioxide coupled with the conversion of NAD⁺ to NADH. In this study, N-terminal His tagged CtFDH (N-CtFDH) and C-terminal His tagged CtFDH (C-CtFDH) was constructed to learn the effect of His tag location on the activity and kinetic parameters of the enzyme. The solubility of proteins was not affected by tag position, however, an interference on the N-terminal region caused a deterioration in specific activity and the kinetic ability of enzyme. The obtained results indicated that the C-terminus of the enzyme is an appropriate region for tag engineering. The C-CtFDH has an approximately three-fold larger specific activity and two-fold higher catalytic efficiency than N-CtFDH. The results suggest that insertion of a His-tag at the N-terminal or C-terminal end of CtFDH has different effects on the protein and the N-terminal fragment of the protein is crucial for the function of CtFDH.     

Functional expression of the lipase gene Lip2 of <Emphasis Type="Italic">Pleurotus</Emphasis> <Emphasis Type="Italic">sapidus</Emphasis> in <Emphasis Type="Italic">Escherichia</Emphasis> <Emphasis Type="Italic">coli</Emphasis>

The lipase Lip2 of the edible basidiomycete, Pleurotus sapidus, is an extracellular enzyme capable of hydrolysing xanthophyll esters with high efficiency. The gene encoding Lip2 was expressed in Escherichia coli TOP10 using the gene III signal sequence to accumulate proteins in the periplasmatic space. The heterologous expression under control of the araBAD promoter led to the high level production of recombinant protein, mainly as inclusion bodies, but partially in a soluble and active form. A fusion with a C-terminal His tag was used for purification and immunochemical detection of the target protein. This is the first example of a heterologous expression and periplasmatic accumulation of a catalytically active lipase from a basidiomycete fungus.     

The use of the luciferase reporter system forin planta gene expression studies

The properties of the firefly luciferase (LUC) make it a very good nondestructive reporter to quantify and image transgene promoter activity in plants. The short half-life of the LUC mRNA and protein, and the very limited regeneration of the LUC protein after reacting with luciferin, enables monitoring of changes in gene activity with a high time resolution. However, the ease at which luciferase activity is measuredin planta, using a light sensitive camera system (2D-luminometer), contrasts sharply with the complications that arise from interpreting the results. A variegated pattern of luciferase activity, that is often observed inin planta measurements, might either be caused by differences in influx, availability of the substrates (luciferin, oxygen, ATP) or by local differences in reporter gene activity. Here we tested the possible contribution of differences in the availability of each substrate to the variegatedin planta luciferase activity, and we show whenin planta luciferase activity is measured under substrate equilibrium conditions and can be related to the promoter activity of the reporter gene. Furthermore, we demonstrate the effects of protein stability, apparent half-life of luciferase activity, regeneration of luciferase and pH on thein vivo andin vitro luciferase measurements. The combined results give the prerequisites for the correct utilisation of the luciferase reporter system, especially forin vivo gene expression studies in plant research.     

Phylogenetic analysis of host–symbiont specificity and codivergence in bioluminescent symbioses

Several groups of marine fishes and squids form mutualistic bioluminescent symbioses with luminous bacteria. The dependence of the animal on its symbiont for light production, the animal's specialized anatomical adaptations for harboring bacteria and controlling light emission, and the host family bacterial species specificity characteristic of these associations suggest that bioluminescent symbioses are tightly coupled associations that might involve coevolutionary interactions. Consistent with this possibility, evidence of parallel cladogenesis has been reported for squid–bacterial associations. However, genetic adaptations in the bacteria necessary for and specific to symbiosis have not been identified, and unlike obligate endosymbiotic associations in which the bacteria are transferred vertically, bacterially bioluminescent hosts acquire their light‐organ symbionts from the environment with each new host generation. These contrasting observations led us to test the hypotheses of species specificity and codivergence in bioluminescent symbioses, using an extensive sampling of naturally formed associations. Thirty‐five species of fish in seven teleost families (Chlorophthalmidae, Macrouridae, Moridae, Trachichthyidae, Monocentridae, Acropomatidae, Leiognathidae) and their light‐organ bacteria were examined. Phylogenetic analysis of a taxonomically broad sampling of associations was based on mitochondrial 16S rRNA and cytochrome oxidase I gene sequences for the fish and on recA, gyrB and luxA sequences for bacteria isolated from the light organs of these specimens. In a fine‐scale test focused on Leiognathidae, phylogenetic analysis was based also on histone H3 subunit and 28S rRNA gene sequences for the fish and on gyrB, luxA, luxB, luxF and luxE sequences for the bacteria. Deep divergences were revealed among the fishes, and clear resolution was obtained between clades of the bacteria. In several associations, bacterial species identities contradicted strict host family bacterial species specificity. Furthermore, the fish and bacterial phylogenies exhibited no meaningful topological congruence; evolutionary divergence of host fishes was not matched by a similar pattern of diversification in the symbiotic bacteria. Re‐analysis of data reported for squids and their luminous bacteria also revealed no convincing evidence of codivergence. These results refute the hypothesis of strict host family bacterial species specificity and the hypothesis of codivergence in bioluminescent symbioses. © The Willi Hennig Society 2007.     

Single Molecule Assays Reveal Differences Between In Vitro and In Vivo Synthesized β-Galactosidase

Genomic DNA was extracted from wild-type Escherichia coli strains ATCC 35321 and 8677. The lac Z gene was amplified and used as a template for in vitro synthesis of β-galactosidase. In addition the enzyme was synthesized in vitro with a C-terminal His₆ tag. The enzyme expression was also induced in these strains using isopropyl-β-D-galactoside. Single enzyme molecule assays were performed using a capillary electrophoresis-based protocol on both the in vitro and in vivo synthesized enzyme. In vivo produced enzyme from strains 35321 and 8677 showed average combined turnover numbers for the 4 active sites of the individual enzyme molecules of 53,400 ± 18,400 (N = 139) and 34,300 ± 17,800 min⁻¹ (N = 181) respectively. Average combined turnover numbers of 35,800 ± 20,900 (N = 302) and 31,700 ± 17,700 min⁻¹ (N = 315) were obtained respectively for the in vitro synthesized enzyme from strain 35321 with the absence and presence of a C-terminal His₆ tag. For strain 8677, the average combined turnover numbers were 29,000 ± 17,900 (N = 288) and 25,200 ± 12,600 min⁻¹ (N = 240) respectively for the absence and presence of a C-terminal His₆ tag. The average combined turnover numbers of the enzyme from both strains synthesized in vivo and in vitro and with the presence and absence of a His₆ tag were found to differ significantly. This indicates that the in vivo and in vitro produced enzymes are not identical and the presence of a C-terminal His₆ tag alters the activity of β-galactosidase.     

Generating functional protein variants with variational autoencoders

The vast expansion of protein sequence databases provides an opportunity for new protein design approaches which seek to learn the sequence-function relationship directly from natural sequence variation. Deep generative models trained on protein sequence data have been shown to learn biologically meaningful representations helpful for a variety of downstream tasks, but their potential for direct use in the design of novel proteins remains largely unexplored. Here we show that variational autoencoders trained on a dataset of almost 70000 luciferase-like oxidoreductases can be used to generate novel, functional variants of the luxA bacterial luciferase. We propose separate VAE models to work with aligned sequence input (MSA VAE) and raw sequence input (AR-VAE), and offer evidence that while both are able to reproduce patterns of amino acid usage characteristic of the family, the MSA VAE is better able to capture long-distance dependencies reflecting the influence of 3D structure. To confirm the practical utility of the models, we used them to generate variants of luxA whose luminescence activity was validated experimentally. We further showed that conditional variants of both models could be used to increase the solubility of luxA without disrupting function. Altogether 6/12 of the variants generated using the unconditional AR-VAE and 9/11 generated using the unconditional MSA VAE retained measurable luminescence, together with all 23 of the less distant variants generated by conditional versions of the models; the most distant functional variant contained 35 differences relative to the nearest training set sequence. These results demonstrate the feasibility of using deep generative models to explore the space of possible protein sequences and generate useful variants, providing a method complementary to rational design and directed evolution approaches.