Adaptive laboratory evolution of Escherichia coli W enhances gamma-aminobutyric acid production using glycerol as the carbon source.

The microbial conversion of glycerol into value-added commodity products has emerged as an attractive means to meet the demands of biosustainability. However, glycerol is a non-preferential carbon source for productive fermentation because of its low energy density. We employed evolutionary and metabolic engineering in tandem to construct an Escherichia coli strain with improved GABA production using glycerol as the feedstock carbon. Adaptive evolution of E. coli W under glycerol-limited conditions for 1300 generations harnessed an adapted strain with a metabolic system optimized for glycerol utilization. Mutation profiling, enzyme kinetic assays, and transcriptome analysis of the adapted strain allowed us to decipher the basis of glycerol adaptation at the molecular level. Importantly, increased substrate influx mediated by the mutant glpK and modulation of intracellular cAMP levels were the key drivers of improved fitness in the glycerol-limited condition. Leveraging the enhanced capability of glycerol utilization in the strain, we constructed a GABA-producing E. coli W-derivative with superior GABA production compared to the wild-type. Furthermore, rationally designed inactivation of the non-essential metabolic genes, including ackA, mgsA, and gabT, in the glycerol-adapted strain improved the final GABA titer and specific productivity by 3.9- and 4.3-fold, respectively, compared with the wild-type.

Detection and purification of backbone-cyclized proteins using a bacterially expressed anti-myc-tag single chain antibody.

A myc-tag and of which recognition by an antibody 9E10 has long been used for the detection and purification of recombinant proteins. We have previously expanded the application of the tag to the specific detection and purification of backbone-cyclized proteins. Here we sought a more practical way to using the 9E10 antibody by expressing its single chain antibody (scAb) form in Escherichia coli. The combined use of a strong T7 promoter and auto-induction strategy rather than early to mid-log induction of a Lac promoter resulted in the soluble over-expression of 9E10 scAb. However, the co-expression of a chaperone, Skp, was absolutely necessary for the activity even when the protein was expressed in a soluble manner. We could purify about 4 mg of 9E10 scAb from 1 l of culture, and the resulting scAb could be used to detect and purify the backbone-cyclized protein as the parental full-length 9E10. Moreover, the immunoaffinity resin prepared using 9E10 scAb could be regenerated several times after the elution of bound proteins using an acid, which added more value to the ready preparation of the active antibody in bacteria.

The architecture of ArgR-DNA complexes at the genome-scale in Escherichia coli.

DNA-binding motifs that are recognized by transcription factors (TFs) have been well studied; however, challenges remain in determining the in vivo architecture of TF-DNA complexes on a genome-scale. Here, we determined the in vivo architecture of Escherichia coli arginine repressor (ArgR)-DNA complexes using high-throughput sequencing of exonuclease-treated chromatin-immunoprecipitated DNA (ChIP-exo). The ChIP-exo has a unique peak-pair pattern indicating 5' and 3' ends of ArgR-binding region. We identified 62 ArgR-binding loci, which were classified into three groups, comprising single, double and triple peak-pairs. Each peak-pair has a unique 93 base pair (bp)-long (±2 bp) ArgR-binding sequence containing two ARG boxes (39 bp) and residual sequences. Moreover, the three ArgR-binding modes defined by the position of the two ARG boxes indicate that DNA bends centered between the pair of ARG boxes facilitate the non-specific contacts between ArgR subunits and the residual sequences. Additionally, our approach may also reveal other fundamental structural features of TF-DNA interactions that have implications for studying genome-scale transcriptional regulatory networks.

Genetic Code Expansion by Degeneracy Reprogramming of Arginyl Codons.

The genetic code in most organisms codes for 20 proteinogenic amino acids or translation stop. In order to encode more than 20 amino acids in the coding system, one of stop codons is usually reprogrammed to encode a non-proteinogenic amino acid. Although this approach works, usually only one amino acid is added to the amino acid repertoire. In this study, we incorporated non-proteinogenic amino acids into a protein by using a sense codon. As all the codons are allocated in the universal genetic code, we destroyed all the tRNA(Arg) in a cell-free protein synthesis system by using a tRNA(Arg) -specific tRNase, colicin D. Then by supplementing the system with tRNACCU , the translation system was partially restored. Through this creative destruction, reprogrammable codons were successfully created in the system to encode modified lysines along with the 20 proteinogenic amino acids.

Implementing bacterial acid resistance into cell-free protein synthesis for buffer-free expression and screening of enzymes.

Cell-free protein synthesis utilizes translational machinery isolated from the cells for in vitro expression of template genes. Because it produces proteins without gene cloning and cell cultivation steps, cell-free protein synthesis can be used as a versatile platform for high-throughput expression of enzyme libraries. Furthermore, the open nature of cell-free protein synthesis allows direct integration of enzyme synthesis with subsequent screening steps. However, the presence of high concentration of chemical buffers in the conventional reaction mixture makes it difficult to streamline cell-free protein synthesis with pH-based assay of the synthesized enzymes. In this study, we have implemented an enzyme-assisted bacterial acid resistance mechanism into an Escherichia coli (E.coli) extract-based cell-free protein synthesis system in place of chemical buffers. When deployed in the reaction mixture for cell-free synthesis of enzymes, through proton-consuming conversion of glutamate into γ-aminobutyric acid (GABA), an engineered glutamate decarboxylase (GADß) was able to maintain the pH of reaction mixture during enzyme synthesis. Because the reaction mixture becomes free of buffering capacity upon the depletion of glutamate, synthesized enzyme could be directly assayed without purification steps. The designed method was successfully applied to the screening of mutant library of sialyltransferase genes to identify mutants with improved enzymatic activity.

Enhanced production of gamma-aminobutyrate (GABA) in recombinant Corynebacterium glutamicum by expressing glutamate decarboxylase active in expanded pH range.

BACKGROUND: Gamma-aminobutylate (GABA) is an important chemical in pharmacetucal field and chemical industry. GABA has mostly been produced in lactic acid bacteria by adding L-glutamate to the culture medium since L-glutamate can be converted into GABA by inherent L-glutamate decarboxylase. Recently, GABA has gained much attention for the application as a major building block for the synthesis of 2-pyrrolidone and biodegradable polyamide nylon 4, which opens its application area in the industrial biotechnology. Therefore, Corynebacterium glutamicum, the major L-glutamate producing microorganism, has been engineered to achieve direct fermentative production of GABA from glucose, but their productivity was rather low. RESULTS: Recombinant C. glutamicum strains were developed for enhanced production of GABA from glucose by expressing Escherichia coli glutamate decarboxylase (GAD) mutant, which is active in expanded pH range. Synthetic PH36, PI16, and PL26 promoters, which have different promoter strengths in C. glutamicum, were examined for the expression of E. coli GAD mutant. C. glutamicum expressing E. coli GAD mutant under the strong PH36 promoter could produce GABA to the concentration of 5.89±0.35 g/L in GP1 medium at pH 7.0, which is 17-fold higher than that obtained by C. glutamicum expressing wild-type E. coli GAD in the same condition (0.34±0.26 g/L). Fed-bath culture of C. glutamicum expressing E. coli GAD mutant in GP1 medium containing 50 µg/L of biotin at pH 6, culture condition of which was optimized in flask cultures, resulted in the highest GABA concentration of 38.6±0.85 g/L with the productivity of 0.536 g/L/h. CONCLUSION: Recombinant C. glutamicum strains developed in this study should be useful for the direct fermentative production of GABA from glucose, which allows us to achieve enhanced production of GABA suitable for its application area in the industrial biotechnology.

Genetically engineered myoblast sheet for therapeutic angiogenesis.

Peripheral arterial disease is a common manifestation of systemic atherosclerosis, which results in more serious consequences of ischemic events in peripheral tissues such as the lower extremities. Cell therapy has been tested as a treatment for peripheral ischemia that functions by inducing angiogenesis in the ischemic region. However, the poor survival and engraftment of transplanted cells limit the efficacy of cell therapy. In order to overcome such challenges, we applied genetically engineered cell sheets using a cell-interactive and thermosensitive hydrogel and nonviral polymer nanoparticles. C2C12 myoblast sheets were formed on Tetronic-tyramine (Tet-TA)-RGD hydrogel prepared through a highly efficient and noncytotoxic enzymatic reaction. The myoblast sheets were then transfected with vascular endothelial growth factor (VEGF) plasmids using poly(ß-amino ester) nanoparticles to increase the angiogenic potential of the sheets. The transfection increased the VEGF expression and secretion from the C2C12 sheets. The enhanced angiogenic effect of the VEGF-transfected C2C12 sheets was confirmed using an in vitro capillary formation assay. More importantly, the transplantation of the VEGF-transfected C2C12 sheets promoted the formation of capillaries and arterioles in ischemic muscles, attenuated the muscle necrosis and fibrosis progressed by ischemia, and eventually prevented ischemic limb loss. In conclusion, the combination of cell sheet engineering and genetic modification can provide more effective treatment for therapeutic angiogenesis.

Cyclization tag for the detection and facile purification of backbone-cyclized proteins.

Backbone-cyclized proteins, with their characteristic stability toward denaturants such as heat and chemicals, are becoming increasingly significant in many applications. Intein-mediated protein cyclization is the most efficient and frequently used method of choice and has been successfully applied to various targets, achieving stable proteins. However, the detection and isolation of the cyclic protein from the linear one after cyclization is very difficult because the backbone-cyclized protein and the linear one (a by-product formed during the cyclization reaction), which originated from the same molecule, are almost identical in terms of their size. Thus, we first developed a split c-myc tag system; the active c-myc tag was formed only in the backbone-cyclized protein and not in the linear by-product from the inactive precursor, and this helps both the detection and purification of the backbone-cyclized proteins. This tag system, which we called a cyclization tag, was further engineered in its sequence to develop an engineered c-myc (e-myc) tag with enhanced efficiency in the backbone cyclization reaction while keeping its specificity toward the commercial antibody intact. Using two different proteins as models, we show that the cyclization tag developed here can be used as a specific tag for the backbone-cyclized protein, thereby facilitating detection and purification.

Multiple Gene Expression in Cell-Free Protein Synthesis Systems for Reconstructing Bacteriophages and Metabolic Pathways.

As a fast and reliable technology with applications in diverse biological studies, cell-free protein synthesis has become popular in recent decades. The cell-free protein synthesis system can be considered a complex chemical reaction system that is also open to exogenous manipulation, including that which could otherwise potentially harm the cell's viability. On the other hand, since the technology depends on the cell lysates by which genetic information is transformed into active proteins, the whole system resembles the cell to some extent. These features make cell-free protein synthesis a valuable addition to synthetic biology technologies, expediting the design-build-test-learn cycle of synthetic biology routines. While the system has traditionally been used to synthesize one protein product from one gene addition, recent studies have employed multiple gene products in order to, for example, develop novel bacteriophages, viral particles, or synthetic metabolisms. Thus, we would like to review recent advancements in applying cell-free protein synthesis technology to synthetic biology, with an emphasis on multiple gene expressions.

Functional Characterization of Melanin Decolorizing Extracellular Peroxidase of Bjerkandera adusta.

Melanin pigmentation in the human skin results from complicated cellular mechanisms that remain to be entirely understood. Uneven melanin pigmentation has been counteracted by inhibiting synthesis or transfer of melanin in the skin. Recently, an enzymatic approach has been proposed, wherein the melanin in the skin is decolorized using lignin peroxidase. However, not many enzymes are available for decolorizing melanin; the most studied one is lignin peroxidase derived from a lignin degrading fungus, Phanerochaete chrysosporium. Our current study reveals that versatile peroxidase from Bjerkandera adusta can decolorize synthetic melanin. Melanin decolorization was found to be dependent on veratryl alcohol and hydrogen peroxide, but not on Mn2+. The degree of decolorization reached over 40% in 10 min at 37 °C and a pH of 4.5. Optimized storage conditions were slightly different from those for the reaction; crude enzyme preparation was the most stable at 25 °C at pH 5.5. Since the enzyme rapidly lost its activity at 50 °C, stabilizers were screened. As a result, glycerol, a major component in several cosmetic formulations, was found to be a promising excipient. Our results suggest that B. adusta versatile peroxidase can be considered for future cosmetic applications aimed at melanin decolorization.

Combinatorial, selective and reversible control of gene expression using oligodeoxynucleotides in a cell-free protein synthesis system.

Herein we describe the methods for selective and reversible regulation of gene expression using antisense oligodeoxynucleotides (ODNs) in a cell-free protein synthesis system programmed with multiple DNAs. Either a complete shut down or controlled level of gene expression was attained through the antisense ODN-mediated regulation of mRNA stability in the reaction mixture. In addition to the primary control of gene expression, we also demonstrate that the inhibition of protein synthesis can be reversed by using an anti-antisense ODN sequence that strips the antisense ODN off the target sequence of mRNA. As a result, sequential additions of the antisense and anti-antisense ODNs enabled the stop-and-go expression of protein molecules. Through the on-demand regulation of gene expression, presented results will provide a versatile platform for the analysis and understanding of the complicated networks of biological components.

In vitro selection of a 5'-purine ribonucleotide transferase ribozyme.

Here we report in vitro selection of a novel ribozyme that catalyzes the 5'-nucleotidyl transfer reaction forming the 2'-5' phosphodiester bond. This ribozyme was retrieved as a sole sequence in the pool enriched for the 5'-triphosphate-dependent activities in incorporating ATP-gammaS. The originally selected ribozyme consisting of 109-nucleotide (nt) was miniaturized to 45-nt M4 ribozyme via a series of mutation studies, and based on this mini-ribozyme a trans-acting system was constructed. One of the most challenging tasks in our study was to determine the chemistry occurring at the 5'-ppp site. We utilized various analytical methods including MALDI-TOF analysis of the product generated by the trans-acting system and elucidated the chemistry to be 3'-->5' mononucleotide extension forming the 2'-5' phosphodiester bond. Interestingly, M4 ribozyme promiscuously accepts a variety of purine nucleotides bearing 5'-mono-, di- and triphosphates as substrates. This remarkable ability of M4 ribozyme would lead us to the development of a new tool for the 5'-modification of RNAs with unique chemical groups.

Artificial Caprolactam-Specific Riboswitch as an Intracellular Metabolite Sensor.

Caprolactam is a monomer used for the synthesis of nylon-6, and a recombinant microbial strain for biobased production of nylon-6 was recently developed. An intracellular biosensor for caprolactam can facilitate high-throughput metabolic engineering of recombinant microbial strains. Because of the mixed production of caprolactam and valerolactam in the recombinant strain, a caprolactam biosensor should be highly specific for caprolactam. However, a highly specific caprolactam sensor has not been reported. Here, we developed an artificial riboswitch that specifically responds to caprolactam. This riboswitch was prepared using a coupled in vitro- in vivo selection strategy with a heterogeneous pool of RNA aptamers obtained from in vitro selection to construct a riboswitch library used in in vivo selection. The caprolactam riboswitch successfully discriminated caprolactam from valerolactam. Moreover, the riboswitch was activated by 3.36-fold in the presence of 50 mM caprolactam. This riboswitch enabled caprolactam-dependent control of cell growth, which will be useful for improving caprolactam production and is a valuable tool for metabolic engineering.

Identification of peptide inhibitors of matrix metalloproteinase 1 using an in-house assay system for the enzyme.

Matrix metalloproteinases (MMPs) are zinc-dependent proteases involved in the degradation of extracellular matrix proteins. As one of the isoforms, MMP-1 breaks down collagen, and its activity is known to be important in wound healing. Its timely and adequate level of expression is pivotal because MMP-1 is also involved in the damage or aging of skins as well as in certain types of cancers. Thus, both assaying the MMP-1 activity and developing its inhibitors are of great importance. We here developed an in-house assay system that gave us the high degree of freedom in screening peptide inhibitors of MMP-1. The assay system utilized a circularly permutated fusion of ß-lactamase and its inhibitory protein through an MMP-1-sensitive linker so that the activity of MMP-1 could be translated into that of ß-lactamase. As a proof of concept, we applied the developed assay system to initial screens of MMP-1 inhibitors and successfully identified one lead peptide that inhibited the collagenase activity of the enzyme.

Thermostable Ebola virus vaccine formulations lyophilized in the presence of aluminum hydroxide.

No United States Food and Drug Administration-licensed vaccines protective against Ebola virus (EBOV) infections are currently available. EBOV vaccine candidates currently in development, as well as most currently licensed vaccines in general, require transport and storage under a continuous cold chain in order to prevent potential decreases in product efficacy. Cold chain requirements are particularly difficult to maintain in developing countries. To improve thermostability and reduce costly cold chain requirements, a subunit protein vaccine against EBOV was formulated as a glassy solid using lyophilization. Formulations of the key antigen, Ebola glycoprotein (EBOV-GP), adjuvanted with microparticulate aluminum hydroxide were prepared in liquid and lyophilized forms, and the vaccines were incubated at 40 °C for 12 weeks. Aggregation and degradation of EBOV-GP were observed in liquid formulations during the 12-week incubation period, whereas changes were minimal in lyophilized formulations. Antibody responses against EBOV-GP following three intramuscular immunizations in BALB/c mice were used to determine vaccine immunogenicity. EBOV-GP formulations were equally immunogenic in liquid and lyophilized forms. After lyophilization and reconstitution, adjuvanted vaccine formulations produced anti-EBOV-GP IgG antibody responses in mice similar to those generated against corresponding adjuvanted liquid vaccine formulations. More importantly, antibody responses in mice injected with reconstituted lyophilized vaccine formulations that had been incubated at 40 °C for 12 weeks prior to injection indicated that vaccine immunogenicity was fully retained after high-temperature storage, showing promise for future vaccine development efforts.

Tuning the expression level of recombinant proteins by modulating mRNA stability in a cell-free protein synthesis system.

In this work, it was discovered that the stability of mRNA in a cell-free extract could be controlled by using engineered T7 terminator sequences. Specifically, it was found that mRNA stability gradually decreased as the length of the stem structure of the T7 terminator was reduced sequentially. As a result of the controlled abundance of mRNA species, it was possible to manipulate the relative expression level of target proteins by employing the T7 terminator of adjusted stem lengths.

Redesigning the substrate specificity of omega-aminotransferase for the kinetic resolution of aliphatic chiral amines.

Substrate specificity of the omega-aminotransferase obtained from Vibrio fluvialis (omega-ATVf) was rationally redesigned for the kinetic resolution of aliphatic chiral amines. omega-ATVf showed unique substrate specificity toward aromatic amines with a high enantioselectivity (E > 100) for (S)-enantiomers. However, the substrate specificity of this enzyme was much narrower toward aliphatic amines. To overcome the narrow substrate specificity toward aliphatic amines, we redesigned the substrate specificity of omega-ATVf using homology modeling and the substrate structure- activity relationship. The homology model and the substrate structure-activity relationship showed that the active site of omega-ATVf consists of one large substrate-binding site and another small substrate-binding site. The key determinant in the small substrate-binding site was D25, whose role was expected to mask R415 and to generate the electrostatic repulsion with the substrate's alpha-carboxylate group. In the large substrate-binding site, R256 was predicted to recognize the alpha-carboxylate group of substrate thus obeying the dual substrate recognition mechanism of aminotransferase subgroup II enzymes. Among the several amino acid residues in the large substrate-binding site, W57 and W147, with their bulky side chains, were expected to restrict the recognition of aliphatic amines. Two mutant enzymes, W57G and W147G, showed significant changes in their substrate specificity such that they catalyzed transamination of a broad range of aliphatic amines without losing the original activities toward aromatic amines and enantioselectivity.

Soluble expression of horseradish peroxidase in Escherichia coli and its facile activation.

Horseradish peroxidase (HRP) is widely used as a marker enzyme in immunoassays and biosensors, and can possibly be used in industries such as waste water treatments or fine chemical synthesis. Cost-effective production of active HRP is thus very important in the related fields. Also, engineering of HRP for its better performance in the designated application is expected to make the enzyme even more important in several areas of research and industry. One of obstacles to this end and to the large scale production of the enzyme has been its facile expression in a bacterial host. Here we show that HRP could be overexpressed as a soluble form by fusing the enzyme with Escherichia coli phosphoglycerate kinase (PGK). After simple incubation with calcium ion, hemin, and oxidized glutathione, PGK-HRP could be fully activated showing a higher molar specific activity than plant-derived HRP. Our established procedure did not use tedious and inefficient refolding steps that have been used to activate HRP produced as inclusion bodies and thus is superior in its overall yield (>72 mg purified HRP conjugate per L culture) to existing methods. By co-expressing PGK-HRP with ferrochelatase in a special host that permitted the formation of disulfide bonds in the cytoplasm, the activation steps could be simplified even though the resulting specific activity was low.

Bacterial tRNase-Based Gene Therapy with Poly(ß-Amino Ester) Nanoparticles for Suppressing Melanoma Tumor Growth and Relapse.

Here, a novel anticancer gene therapy with a bacterial tRNase gene, colicin D or virulence associated protein C (VapC), is suggested using biodegradable polymeric nanoparticles, such as poly(ß-amino esters) (PBAEs) as carriers. These genes are meticulously selected, aiming at inhibiting translation in the recipients by hydrolyzing specific tRNA species. In terms of nanoparticles, out of 9 PBAE formulations, a leading polymer, (polyethylene oxide)4 -bis-amine end-capped poly(1,4-butanediol diacrylate-co-5-amino-1-pentanol) (B4S5E5), is identified that displays higher gene delivery efficacy to cancer cells compared with the leading commercial reagent Lipofectamine 2000. Interestingly, the B4S5E5 PBAE nanoparticles complexed with colicin D or VapC plasmid DNA induce significant toxicity highly specific to cancer cells by triggering apoptosis. In contrast, the PBAE nanoparticles do not induce these cytotoxic effects in noncancerous cells. In a mouse melanoma model of grafted murine B16-F10 cells, it is demonstrated that treatment with PBAE nanoparticles complexed with these tRNase genes significantly reduces tumor growth rate and delays tumor relapse. Moreover, increased stability of PBAE by PEGylation further enhances the therapeutic effect of tRNase gene treatment and improves survival of animals. This study highlights a nonviral gene therapy that is highly promising for the treatment of cancer.

Development of Artificial Riboswitches for Monitoring of Naringenin In Vivo.

Microbial strains are considered promising hosts for production of flavonoids because of their rapid growth rate and suitability for large-scale manufacturing. However, productivity and titer of current recombinant strains still do not meet the requirements of industrial processes. Genetically encoded biosensors have been applied for high-throughput screening or dynamic regulation of biosynthetic pathways for enhancing the performance of microbial strains that produce valuable chemicals. Currently, few protein sensor-regulators for flavonoids exist. Unlike the protein-based trans-regulating controllers, riboswitches can respond to their ligands faster and eliminate off-target effects. Here, we developed artificial riboswitches that activate gene expression in response to naringenin, an important flavonoid. RNA aptamers for naringenin were developed using SELEX and cloned upstream of a dual selectable marker gene to construct a riboswitch library. Two in vivo selection routes were applied separately to the library by supplementing naringenin at two different concentrations during enrichments to modulate the operational ranges of the riboswitches. The selected riboswitches were responsive to naringenin and activated gene expression up to 2.91-fold. Operational ranges of the riboswitches were distinguished on the basis of their selection route. Additionally, the specificity of the riboswitches was assessed, and their applicability as dynamic regulators was confirmed. Collectively, the naringenin riboswitches reported in this work will be valuable tools in metabolic engineering of microorganisms for the production of flavonoids.