Earticle

A large number of natural and unnatural products from microorganisms such as biologically active compounds and commodity chemicals have attracted huge attention in the various industries including pharmaceutical and chemical industries. However, functional improvement of microorganisms meeting with commercial need is limited by the biological constraints developed during natural evolutionary process. “Metabolic Engineering” aims the purposeful redesign of the biological systems and requires the accurate information of the cellular metabolic networks and proper tools for the reconstruction of the biological systems such as the regulatory elements and engineered proteins. The goal of “Synthetic Biology” is to synthesize whole biological system or its subsystem intentionally and designing parts such as the regulatory elements and functional gene should be necessarily predictable. Numerous regulatory elements such as promoter libraries and tunable intergenic regions (TIGRs) can be applied for the modulation of gene expression. However, without considering 5’-unstranslated region (5’-UTR) sequence, it is insufficient to precisely design and modulate the gene expression level. To address this issue, in this study, we obtained 5’-UTR variants for gene expression in Escherichia coli, followed by GFP as a reporting system. Even if using same promoter sequence, 5’-UTR variants showed expression levels in a range of 100-fold. This indicates that additional information besides promoter strength, such as 5’-UTR, should be considered for precise gene expression.

The main goal of this research is to achieve development of recombinant E. coli for fatty acid synthesis(FAS). Genes for pyruvate dehydrogenase (aceE) and acetyl-CoA carboxylase (accA), which are the enzymes that catalyze the first step in the synthesis of fatty acids in Escherichia coli MG1655, were cloned and characterized. That is, the E. coli strains were midterm recombinant strains to produce fatty acid. The enzyme of aceE gene converts pyruvate to acetyl-CoA and that of accA gene catalyzes the addition of CO2 to acetyl-CoA to generate malonyl-CoA. The genes were identified as homologous gene of E. coli through a metabolic pathway. The recombinant E. coli MG1655 containing the aceE or accA gene-inserted expression vector(pTrc99A) were constructed. As a result, the intermediate products were analyzed from in vivo metabolites and in vitro metabolites of recombinant E. coli. The two strains produced more malonic acid than its wild type E. coli. However, there was insignificant the difference between E. coli harboring a aceE gene-inserted pTrc99A vector and E. coli harboring a accA gene-inserted pTrc99A vector.

Polylactic acid (PLA) has been considered as a good alternative to petroleum-based plastic. However, the current process for PLA synthesis is not simple: fermentative production of lactic acid followed by chemical polymerization. Differently from PLA which requires chemical polymerization of lactic acid or lactide, polyhydroxyalkanoates (PHAs) are synthesized in vivo by PHA synthase which polymerizes various (D)-hydroxyacyl-CoAs generated through diverse metabolic pathways in the cell1. In this study, we were able to show that poly(3-hydroxybutyrate-co-lactate) could be synthesized in recombinant Cupriavidus necator H16 (formerly, Ralstonia eutropha H16) expressing the Clostridium propionicum propionate CoA-transferase gene2, which allows generation of (D)-lactyl-CoA in the cell3. Also, four representative PHA synthases (type I-IV) were examined for their ability to synthesize lactate-containing polymers in Escherichia coli. This work was supported by LG Chem and by the Korean Systems Biology Research Project (20090065571) of the Ministry of Education, Science and Technology. Further supports by the LG Chem Chair Professorship, Microsoft, and IBM SUR, WCU (World Class University) program through the National Research Foundation of Korea funded by the Ministry of Education, Science and Technology (R32-2008-000-10142-0) program are appreciated.

Polylactic acid (PLA) is a promising biomass-derived polymer, but is currently synthesized by a two-step process: fermentative production of lactic acid followed by chemical polymerization. Here we report production of PLA homopolymer and its copolymer, poly(3-hydroxybutyrate-co-lactate), P(3HB-co-LA), by direct fermentation of metabolically engineered Escherichia coli. Introduction of the heterologous metabolic pathways involving engineered propionate CoA-transferase and polyhydroxyalkanoate (PHA) synthase for the efficient generation of lactyl-CoA and incorporation of lactyl-CoA into the polymer, respectively, allowed synthesis of PLA and P(3HB-co-LA) in E. coli, but at relatively low efficiency1. In this study, the metabolic pathways of E. coli were further engineered based on in silico genome-scale metabolic flux analysis in addition to rational approach2. Using this engineered strain, PLA homopolymer and P(3HB-co-LA) copolymers containing up to 70 mol% lactate could be produced up to 11 wt% and 46 wt% from glucose, respectively. Thus, the combined approaches of systems-level metabolic engineering and enzyme engineering allowed efficient bio-based one-step production of PLA and its copolymers. This work was supported by LG Chem and by the Korean Systems Biology Research Project (20090065571) of the Ministry of Education, Science and Technology. Further supports by the LG Chem Chair Professorship, Microsoft, and IBM SUR, WCU (World Class University) program through the National Research Foundation of Korea funded by the Ministry of Education, Science and Technology (R32-2008-000-10142-0) program are appreciated. References 1. Yang, T.H., Kim, T.W., Kang, H.O., Lee, S.-H., Lee, E.J., Lim, S.-C., Oh, S.O., Song, A.-J., Park, S.J., and Lee, S.Y. Biosynthesis of polylactic acid and its copolymers using evolved propionate CoA transferase and PHA synthase (2009). Biotechnol. Bioeng. In press. 2. Jung, Y.K., Kim, T.Y., Park, S.J., and Lee, S.Y. Metabolic engineering of Escherichia coli for the production of polylactic acid and its copolymers (2009). Biotechnol. Bioeng. In press.

The sphingolipid, representatively ceramides, and sphingosine, recently is being researched as signal transfer materials and induction apoptosis in cell and studied about cell-to-cell signaling including acceptor on the cell membrane representatively sphingosine 1-phosphae and sphingosylphosphorylcholine [1, 2]. And ceramides has been known that have roles of intermediate material of sphingolipids metabolism and importantly participate including apoptosis, cell cycle arrest, senescence, differentiation, and stress responses [1, 2]. The goals of this study are comparison of production of ceramides between Saccharomyces cerevisiae and transformants containing genes on sphingolipid pathway. The vector system was constructed with YCplac22 and YCplac33 that are under the control of lac promoter and include TRP1 and URA3 selectable marker, respectively. Ceramide were prepared by solvent extraction and quantitative analysis was analyzed by HPLC system with ELSD [3].

The retinoids are represented essential compounds in many biological systems. β-Carotene is cleaved by β-carotene monooxygenase to yield two molecules of retinal. β-Carotene is synthesized by using IPP (isopentenyl diphosphate) and DMAPP (dimethylallyl diphosphate) as building blocks. IPP and DMAPP can be formed in E. coli via endogenous MEP pathway and exogenous Mevalonate pathway. The rate-limiting enzyme of DXP synthase (Dxs) of MEP pathway is overexpressed and an optimized artificial Mevalonate pathway is introduced in E. coli. In order to make E. coli produce β-carotene, it is also introduced the β-carotene pathway composed of 4 foreign enzymes; GGPP synthase (CrtE), phytoene synthase (CrtB), phytoene desaturase (CrtI), and lycopene β-cyclases (CrtY). Artificial E. coli codon optimized gene was synthesized, based on the amino acid sequence of blh gene of uncultured marine bacterium 66A03. The artificial b-Carotene 15,15'-monooxygenase gene was combined with b-carotene pathway genes to make retinal plasmid, pTDHBSR. Recombinant E. coli harboring pTDHBSR produced retinol (25 mg/L), retinal (19 mg/L), and retinyl acetate (9 mg/L). This work was supported by the Basic Research Program(Grant No.2009-0084490) of MEST, the EB-NCRC (Grant No. R15-2003-012-02001-0) and BK21 program of Korea.

In diatom Cylindrotheca fusiformis, modified peptides called silaffin polypeptides are responsible for silica deposition in vivo at ambient condition1. Recently, it was discovered that the synthetic R5 peptide, the repeat unit of silaffin polypeptide without posttranslational modification, was capable of precipitating silica in vitro at ambient condition2. Herein, chemiric proteins were generated by incorporating synthetic silaffin domains (R1-R7) from Cylindrotheca fusiformis onto green fluorescent protein (GFP) by genetic engineering3. His tagged chimeric proteins were accommodated using selfassembled monolayer of disulfide-NTA after chelating with Ni2+ ions on Au surface. Biosilicification catalyzed by silafifn chimeric protein was monitored in situ using quartz crystal microbalance (QCM) after injection of prehydrolyzed TMOS. It was found out that the kinetics of silaffin polypeptide catalyzed biosilicfication is the first order. And the biosilicification is composed of two different stages. At initial stage, the process of biosilicificaiton is dominant by catalysis of silaffin polypeptide and terminates in only 2 minutes. And then, spontaneous silica polymerization occurs in the second step. Silica nanoparticles deposited on Au gold were analyzed using SEM, STEM, AFM, and confocal microscope. The morphology of silica nanoparticles is heterogeneous with the size distribution ranging from 10 nm to 100 nm. This is the first demonstration for biosilicification kinetics catalyzed by silaffin polypeptide.

Budding yeast Saccharomyces cerevisiae plays an important role in producing bioethanol from lignocellulosic biomass. Pretreatment of lignocellulosic biomass releases a number of inhibitors reducing cell growth and ethanol production, which suggests an elaborate research should be performed to increase resistance of S. cerevisiae against fermentation inhibitors. Sucrose non-fermenting 1 (Snf1) is an AMPactivated serine/threonine protein kinase in yeast and is an important kinase for metabolic and developmental response in S. cerevisiae. Phenotypes of snf1 null mutant of S. cerevisiae on inhibitor-containing medium clearly indicated that Snf1 kinase plays crucial roles in tolerance of S. cerevisiae against fermentation inhibitors, which can be used as useful milestone for development of economically feasible bioethanol production process using yeast, S. cerevisiae.

Previous studies have focused on the AP-1 and CRE motifs located in the near upstream region of the tyrosine hydroxylase (TH) promoter due to their importance in regulating TH gene expression. The composition of both TH AP-1 and CRE DNA-protein complexes in olfactory bulb (OB) were consistent with immunohistochemical results showing that, in odor deprived mice, FosB, but not CRE-binding protein (CREB) and CRE modulator (CREM), binding activities and protein expression were reduced in parallel with TH levels. In addition, transgenic mice expressing a 9kb TH/lacZ construct with 2 base substitutions in the AP-1 site showed reduced expression of the reporter gene in the OB and not other brain regions. Interestingly, TH activity and immunoreactivity were not reduced in the OB of FosB knockout animals suggesting that another immediate early gene (IEG) may substitute for FosB as a Jun binding partner.

The plasmepsins involved in the degradation of host cell hemoglobin during malaria infection. Plasmepsin II initiate the degradative process, and have been suggested as attractive targets for treatment of malaria. Thirty compounds were already identified by post-processing the results of a large docking screen of commercially available compounds using an automated procedure based on molecular dynamics refinement and binding free-energy estimation using MM-PBSA and MM-GBSA1). In this work, these were experimentally validated using an inhibition assay based on FRET and Hemoglobin substrate degradation. Remarkably, 26 of the 30 tested compounds were proved to be active as plasmepsin II inhibitors, with IC50 values (FRET) ranging from 4.3 nM to 1.8 μM. Also, hemoglobin degradation by recombinant plasmepsin II was completely inhibited by 100 μM compound 7. These results show that new structural class was not only inhibiting plasmepsin II in vitro but was also active in an human hemoglobin. These experiments suggest the overall approach in the design of powerful and selective plasmepsin II inhibitor.

Biofuels are produced from living organisms by metabolic products. We are in the process of developing a new green technology that can produce a biodiesel from gram-positive bacteria, Bacillus subtilis 168. Naturally, most bacteria produce fatty acids as a cell envelope precursor, and multiple steps are needed to synthesize fatty acids. Among them, acetyl-CoA carboxylase (encoded by accA, accB, accC) is a multi-subunit key enzyme for the biosynthesis of fatty acids. The main function of an acetyl-CoA carboxylase is catalyzing the biotindependent carboxylation of acetyl-CoA to provide malonyl-CoA. This process is the most essential step in the biosynthesis of long-chain fatty acids. Therefore, overexpression of the acetyl-CoA carboxylase subunits increases the productivity of fatty acids. We have engineered a competent producer of fatty acids via introducing three different genes into the Bacillus subtilis genome. In this study, we constructed two recombinant plasmids, pET22bA and pET22bBC. These two plasmids were transformed into Bacillus subtilis 168 and tried to express the target genes consecutively. The compounds produced by two distinct recombinant strains have been analyzing by GC. This study implies that these genes encode the key enzyme for fatty acid biosynthesis in Bacillus subtilis, and ultimately indicates that a large amount of biodiesel can be produced by metabolic engineering.

A number of studies have shown that neurotrophic factors such as glial cell line-derived neurotrophic factor (GDNF) and brain-derived neurotrophic factor (BDNF) are required for either induction or maintenance of dopaminergic neurons in vivo. Transgenic mice expressing GDNF driven by the 9kb tyrosine hydroxylase (TH) promoter showed reduced numbers of TH neurons in the substantia nigra and locus ceruleus. In contrast, these same mice displayed many ectopic TH expressing granule-like cells in the mitral cell layer suggesting altered periglomerular (PG) neuron migration. Fibers containing olfactory marker protein (OMP) immunoreactivity were often found in the external plexiform and mitral cell layers. These data suggest that GDNF and other growth factors should be examined for a role in migration and even differentiation of dopaminergic PG neurons in olfactory bulb.

Mycothiol that is comprised of N-acetylcysteine amide-linked to GlcN- α(1-1)-Ins has been identified as a major thiol in a number of actinomycetes. It plays to be its resistance to autoxidation and a significant role in the detoxification of thiol-reactive substances, such as formaldehyde, various electrophiles and antibiotics. Homology searches from the genome of Streptomyces peucetius ATCC 27952 have found a complete set of biosynthetic genes putatively involved in the biosynthesis of mycothiol. The four genes were designated as mshA, mshB, mshC and mshD. The four genes were colned into pIBR25 to make pMSH25. It is for expression of MSH biosynthetic genes under ermE* promoter. The pMSH25 has been transformed into Streptomyces venezuelae wild type, Streptomyces venezuelae YJ028, Streptomyces lividans TK24 and Streptomyces peucetius DM07. The mycothiol is being extracted from these Streptomyces transformed with recombinant plasmid and analyzing.

There have been several systematic approaches to identify novel synthetic pathways in the metabolic network. However, the prediction was too general or too specified. To achieve both of generality and specificity of the prediction at a reliable level, we propose a systematic approach and related framework. To allow applications, five factors that can affect the novel pathways were considered. Three types of abilities, such as similarity, thermodynamics, and co-expression, are evaluated. The five factors are binding site covalence, chemical similarity, thermodynamic favorability, pathway distance, and organism specificity. We suggest a method to evaluate these factors and calculate priority scores which rank the detected candidates. The factors are evolutionary optimized, and thus the system evolves itself. Our approach will enhance the production of desired biochemicals. [This work was supported by the Korean Systems Biology Research Program (M10309020000-03B5002- 00000) of the Ministry of Education, Science and Technology (MEST). Further supports by LG Chem Chair Professorship, Microsoft, World Class University Program of MEST and IBM-SUR program are greatly appreciated.]

Escherichia coli B and K12 derivatives are commonly used in laboratory and bioindustry due to their fast growing and easy cultivation. After sequencing the complete genome of parent B and K12 strains, we have the demanding task of investigating the proteomes of these strains and obtaining valuable physiological information through comparative proteomics. Here, using two-dimensional gel electrophoresis coupled with MS/MS, we profiled the proteomes of an E. coli B derivative and three K12 derivatives grown with a defined medium at exponential phase and stationary phase, respectively. Over 100 proteins were differentially expressed, including those involved in cellular processes and stress responses, and those in the main metabolic pathways (glycolysis, TCA cycle, gluconeogenesis and fermentative pathway) and the synthesis pathways (amino acid biosynthesis, fatty acid biosynthesis, nucleotide synthesis and energy synthesis). A correlation between the proteome profiling and metabolic pathway analysis revealed significant metabolic differences among the four strains, which provided valuable information for the production of various primary metabolites. Also, analysis of amino acid synthesis and protein processing shed light on rational host selection for recombinant protein production. In conclusion, comparative proteomic analysis revealed systematic metabolic diversity among the four strains studied, and provided valuable information for rational host strain. [This work was supported by the Korean Systems Biology Project of the Ministry of Education, Science and Technology. Further supports by the LG Chem Chair Professorship and KOSEF through the CUPS are appreciated].

Putrescine (also known as 1,4-diaminobutane), is an important platform chemical with a wide range of applications in chemical industry. In particular, putrescine is currently polycondensed with adipic acid to synthesize nylon-4,6 (Stanyl®, DSM), a superior engineering plastic because of its high melting point and mechanical strength as well as excellent solvent resistance. Current production of putrescine on industrial scale relies mainly on chemical synthesis from petrochemicals under environmentally harsh conditions. The chemical synthesis route is thus undesirable from an environmental point of view and human health standpoint. Therefore, there has been increasing need for biotechnological production of putrescine from renewable feedstock.Here we report a sustainable bio-based process for putrescine production using metabolically engineered strain of Escherichia coli. First, a base strain was constructed by inactivating the putrescine degradation and utilization pathways, and deleting the ornithine carbamoyltransferase chain I gene argI to make more precursors available for putrescine synthesis. Next, ornithine decarboxylase, which converts ornithine to putrescine, was amplified by a combination of plasmid-based and chromosome-based overexpression of the coding genes. Furthermore, the ornithine biosynthetic genes (argC-E) were overexpressed from the trc promoter, which replaced the native promoter in the genome, to increase the ornithine pool. Finally, strain performance was further improved by the deletion of the stress responsive RNA polymerase sigma factor RpoS, a well-known global transcription regulator that controls the expression of ca. 10% of the E. coli genes. The final engineered E. coli strain was able to produce 1.68 g L-1 of putrescine with a yield of 0.168 g per g glucose. Furthermore, high cell density cultivation allowed production of 24.2 g L-1 of putrescine with a productivity of 0.75 g L-1 h-1. The strategy reported here should be useful for the bio-based production of putrescine from renewable resources, and also for the development of strains capable of producing other diamines, which are important as nitrogen-containing platform chemicals. [This work was supported by the Korean Systems Biology Research Project of the Ministry of Education, cience and Technology through Korea Science and Engineering Foundation (M10309020000-03B5002-00000). Further supports by LG Chem Chair Professorship, Microsoft, World Class University Program of the Ministry of Education, Science and Technology, and KAIST Institute for the BioCentury are appreciated].

B-type natriuretic peptide is a cardiac neurohormone secreted from the ventricles in volume expansion and pressure overload. B-type natriuretic peptide levels are elevated in patients with left ventricular dysfunction and correlated to the severity as well as prognosis. Therefore, B-type natriuretic peptide is a potential biomarker as well as a therapeutic target. Herein, we report the expression of anti-BNP scFv in the cytoplasm of Escherichia coli for the detection of B-type natriuretic peptide. In this study, we investigated optimal vector, temperature and IPTG concentration for expression of anti-BNP scFv and compare affinity for BNP with ANP. The genetic codes of anti-BNP scFv were chemically synthesized and cloned into both pET22b(+) and pColdⅣ vector, respectively. The recombinant scFv was successfully expressed as a functional form in cytoplasm of E. coli and detected through Western blot and ELISA. The highest level of functional expression of anti-BNP scFv was achieved using pET22b(+) vector at 15℃ by addition of 0.1 mM IPTG. And we confirm expressed anti-BNP scFv specifically captured only BNP. Additionally, we also examined the effect of molecular chaperones on the expression level of functional anti-BNP scFv by coexpressing chaperone gene with the scFv. It was found out that co-expression of groES-groEL chaperon plasmid increased the scFv soluble expression level and activity compared to the scFv expressed without chaperone coexpression.

Since soil is very complex and variable environment, soil-dwelling microorganisms are exposed to many kinds of external compounds. Such compounds may accumulate to toxic levels unless they are metabolized to water- soluble products that can be readily excreted. Cytochrome P450 enzymes often catalyze the initial step in such detoxification pathways. There are multiple forms of cytochrome P450 in a single strain. They are notable for their broad and overlapping substrate specificities. These enzymes catalyze industrially useful reaction such as hydroxylation, epoxidation etc. A number of studies have been reported that these are differentially expressed by the nducers. In many cases, inducers are also substrates for the induced enzymes. Thus, cytochrome P450 activities remain elevated only as needed. Streptomycetes are one of the most numerous soil bacteria. Among them, we have previously demonstrated that Streptomyces avermitilis is suited for cytochrome P450 reactions of phytochemicals such as isoflavonoid. There are 33 numbers of cytochrome P450 in Sterptomyces avermitilis. We could observe differential induction of cytochrome P450 by phytochemicals in Streptomyces avermitilis through RT-PCR. These knowledges can provide insights into screening of cytochrome P450 mainly participated in the reaction.

Recently, approximately 98% of the theoretical maximum xylitol yield has been reached using XYL2-disrupted Candida tropicalis(BSXDH-3) in batch fermentation. As a follow-up study, this study aims at enhancing xylitol production rate using glucose as a co-substrate. First, we introduced Neurospora crassa xylose reductase gene under the control by C. tropicalis GAPDH promoter for high level expression in the presence of glucose. After this, xylitol production increased in glucose-limited fed-batch fermentation remarkably. Second, glycolytic enzymes which consist of EMP pathway were blocked. As well known, NADPH is mainly generated through the oxidative part of the pentose phosphate pathway. By this knockout strategy, resulting extensive flux through the PP pathway can produce NADPH. Phosphoglucose isomerase(pgi) and phosphofructokinase (pfk1,2) are key enzymes in glucose metabolism. Two strands of pgi and each pfk1,2 gene in the diploid yeast C. tropicalis were sequentially disrupted using the Urablasting method. Interestingly, inactivation of pgi and pfk1,2 didn't led to great reduction of maximum specific growth rate. As a consequence, only deleting pfk from strain BSXDH-3 improved the xylitol production rate during batch culture and deleting either or both pgi and pfk1,2 showed increased xylitol production rate in glucose-limited fed-batch fermentations, which was about 1.2-fold higher than that obtained from the control strain.

Isoprene is a volatile C5 terpenoid that is released mainly from the leaves of many deciduous broad-leaved trees and some bacteria. Isoprene is present under standard conditions as a colorless liquid and it is the monomer of natural rubber and is a precursor to an immense variety of other naturally occurring compounds. Isoprene is an important chemical feedstock used in the synthetic rubber industry. About 800,000 tons per year of cis-polyisoprene are produced from the polymerization of isoprene, most of this polyisoprene is used in the tire and rubber industry. Isoprene is made from dimethylallyl diphosphate (DMAPP) by isoprene synthase via methyl erythritol 1-phosphate pathway (MEP). In this study, we constructed plasmid containing isoprene synthase (ispS) from three kinds of plant, Poplar alba, Poplar trichocarpa,and kudzu and introduced into E. coli to produce isoprene in E. coli. We confirmed isoprene production in E. coli and expecting that isoprene can be made in E. coli much more than now by our advanced metabolic engineering system. Isoprene formation is observed in the recombinant E. coli, which suggests E. coli as potential host strain for isoprene production. This work was supported by the EB-NCRC (Grant No. R15-2003-012-02001-0), KRIBB Research Initiative Program, and BK21 program of Korea.

Bioethanol is the most common renewable fuel today. During pretreatment processes for cellulosic biomass, furan-derived inhibitors such as furfural and hydroxymethylfurfural(HMF) which decrease ethanol productivity are generated. The conversion rate of furfural and HMF is a key factor for obtaining a high fermentation performance. Enzymes which supply cofactors, or those which convert furfural and HMF to less toxic compounds were chosen as candidate targets. Aldehyde dehydrogenase 6 (ALD6) is involved in supplying a cofactor, NADPH, which is needed for reduction of furfural and HMF. In addition, the specific rate of furfural oxidation was increased by 2.2-fold higher, and that of HMF oxidation by 2.5-fold higher in the ALD6 overexpressed strain compared to the control strain. Furthermore, ethanol productivity of the ALD6 overexpressed strain in fermentations with media containing both furfural and HMF increased by 33.2 % compared to the control strain. Consequently, the tolerant properties of S. cerevisiae against furan-derived inhibitors were improved by overproduction of ALD6.

Ethanol has a toxic effect on the cell that inhibits the cellular growth and metabolism. Therefore, ethanol resistance of the cell is a considerable factor in the fermentation process and for this reason the goal of this study was to develop genetically engineered mutants having improved ethanol tolerance. In the previous study, an ethanol producing Escherichia coli BL21 was developed, by introducing pyruvate decarboxylase (pdc) and alcohol dehydrogenase (adhB) into the cell and heat shock genes, BEM1 and SOD2 from Saccharomyces cerevisiae, were inserted into the mutated Eschericia coli BL21 in this study. Finally, we could have three different Escherichia coli strains; the ethanol producing strain, BEM1 gene inserted strain, and SOD2 gene inserted strain. By comparing the three different strains, we were able to measure the functions of heat shock genes, BEM1 and SOD2, and also their functions in ethanol tolerance. The three strains were tested by measuring the cellular growth in various ethanol concentrations, 0%, 1%, 3%, 5% (w/v) ethanol concentration medium. As a result, ethanol resistance mutants showed about two times higher cell growth rate than the ethanol producing mutant in the 5% ethanol concentration medium. In addition, ethanol production yields of the ethanol resistance mutants remained at similar level to the ethanol producing mutant; 2.02g/l for SOD2 inserted mutant, 1.57g/l for BEM1 inserted mutant and 2.03g/l for the ethanol producing mutant.

Many models describing glucose homeostasis in blood stream have been developed to predict the glucose levels after meals for diabetic patients. The minimal model of glucose kinetics was developed using only three differential equations [1] and has been widely employed in clinical studies in spite of its low degree of mechanistic detail. On the other hand, mechanistic model of such a complex system often causes over-parameterization and predicts the metabolic dynamics only under limited conditions. In this study, cybernetic principle condensing sophisticated regulations in metabolic networks was applied in modeling glucose homeostasis in blood stream in the presence of periodical supply of glucose and fat as diet. The mechanistic role of glucokinase was revealed as a metabolic glucose sensor [2], which was simulated by employing the matching law and proportional law in a cybernetic model composed of major biochemical reactions including glycogen formation/hydrolysis and lipid synthesis/degradation. The objective function was designed to maximize the rate of glucose metabolism in the model and stoichiometric parameters were obtained from the elementary mode analysis of the liver metabolic network. The model fits well with the profiles of blood glucose, fatty acid and triglyceride during glucose tolerance test. It also showed a robust glucose homeostasis under randomized supply of either high-carbohydrate or high-fat diet over a long period of time. An extended model including other organs such as adipose tissue and muscle is under development to simulate the metabolic dynamics caused by environmental perturbations, for example, drug, disease or exercise.

From the genomic sequence of Cyanothece ATCC 51148 [1], a central metabolic pathway was re-constructed by comparing to the previously known metabolic pathways in KEGG and by BLAST searching. Glycerate-3-phosphate dehydrogenase [EC1.2.1.13] utilizing NADPH as a cofactor and sedoheptulosebisphosphatase [EC:3.1.3.37] were not found thus the carbon dioxide fixation pathway was proposed to consist of pentose phosphate pathway and glycerate-3- phosphate dehydrogenase [EC1.2.1.12] in glycolysis. An elementary mode analysis for the metabolic network was carried out to identify the optimized flux distribution for hydrogen production. A metatool analysis generated 675 modes of pathways for chemoheterotrophic growth using glycerol and nitrate. Hydrogen production and simultaneous cell growth was stoichiometically and thermodynamically possible in 126 modes. The mode of pathway was visualized as each symbol in a plot of inverse cell yield per glucose and oxygen. At the optimized flux distribution in the central metabolic pathway, the theoretical yields of hydrogen production were calculated to be 700 mmol H2/g cell and 5.3 mol H2/g glycerol. Glycerol flows back to the glycolytic pathway and is metabolized in TCA cycle via the oxidative PPP and RuBP carboxylase because glyceraldehyde 3-phosphate dehydrogenase is inactive in the optimized flux distribution. Hydrogen is produced from the NAD(P)H generated by pyruvate dehydrogenase, malate dehydrogenase, gycerol-3-phosphate dehydrogenase, glucose-6-phosphate dehydrogenase, and 6-phosphogluconate dehydrogenase.

Tylosin is a 16-membered macrolide antibiotics produced by Streptomyces fradiae, composed of a branched lactone ring, tylactone and three deoxysugars, mycinose, mycaminose and mycarose. Of these different deoxysugars, mycinose is attached to C-23 position of tylactone. In this study, the mycinose biosynthetic pathway was identified through the heterologous expression of several gene sets in an engineered strain of Streptomyces venezuelae. A plasmid harboring TDPdeoxyallose biosynthetic gene set (tylD-tylJ-tylN-desIII-desIV) was constructed and introduced into the S. venezuelae YJ028 bearing a deletion of native pikromycin polyketide synthase and desosamine gene cluster. As a result, deoxyallosyl-OMT and deoxyallosyl-DMT were produced from 5-O-mycaminosyl tylonolide (OMT) and demycinosyl tylosin (DMT), respectively. In addition, expression of TDP-deoxyallose biosynthetic genes plus three different sets of methyltransferase genes tylE and tylF (tylE, tylF and tylE-tylF) in the S. venezuelae YJ028 supplemented with OMT and DMT, led to the production of 2'''-or/and 3'''-methyldeoxylallosyl OMT and 2'''-or/and 3'''-methyldeoxylallosyl DMT, respectively. These results describe more precise mycinose biosynthetic pathway and biosynthetic routes to tylosin of S. fradiae.

Butanol, which is a primary alcohol with a C4-structure (C4H10O), as a liquid biofuel has very attractive performances including high energy density, phase stability on blending with other fuels, and noncorrosiveness. Clostridium acetobutylicum ATCC 824 is one of the major butanol producers. To improve butanol formation of C. acetobutylicum ATCC 824, a gene knockout system for metabolic engineering was constructed by using an L1.LtrB group II intron from Lactococcus lactis. The gene knockout plasmid pCACYS3 was constructed by cloning the L1.LtrB group II intron into the pIMPH, which was generated by removing both HindIII restriction sites from C. acetobutylicum-Escherichia coli shuttle vector pIMP1. The transcription of the L1.LtrB group II intron on pCACYS3 was controlled by a thl promoter. To verify the knockout system, pCACYS3, buk gene encoding the butyrate kinase was disrupted in ATCC 824 strain. [This work was supported by the Ministry of Knowledge Economy grant funded by the Korea government (#10030795). Further supports by the GS-Caltex and the BioFuelChem].

Sirolimus (rapamycin) is a 31-memberbed ring macrolide produced by Streptomyces hygroscopicus, and has immunosuppressive, antifungal and anticancer activity. A combined approach using classical mutagenesis and metabolic engineering was tried to improve of production level of rapamycin in S. hygroscopicus ATCC29253. Rapamycin is synthesized from building blocks derived from coenzyme A thioesters (CoA) such as malonyl-CoA and methylmalonyl-CoA. However, the productivity of rapamycin during fermentation processes in S. hygroscopicus is limited by the availability of these precursors in vivo. This study describes the development of a novel high-throughput method for rapid screening of rapamycin-producing strains by Ultraviolet (UV)-induced metagenesis and metabolic engineering of propionyl-CoA carboxylase (PCC) pathway. Therefore, we selected two distinct strains exhibiting approximately 3.2-fold enhanced rapamycin production compared to that of wild-type strain by UV mutagenesis. Moreover, PCC pathway participating in the biosynthesis of methylmalonyl-CoA from propionyl-CoA was introduced into UV-mutated strains, resulting in additional 1.4-fold improved production of rapamycin.

Farnesol is an acyclic sesquiterpene in many essential oils. It can be used to emphasize the odors of sweet floral perfumes and also applied to a natural pesticide for mites. Properties of hydrophobicity and high energy content also present farnesol as an ideal biofuel molecule. Farnesol is formed through FPP (farnesyl diphosphate) depyrophosphorylation either by terpene synthase in plants or promiscuous phosphatase in microoganisms. FPP is generated by FPP synthase (IspA) condensing the universal five-carbon building blocks, IPP (isopentenyl diphosphate) and DMAPP (dimethylallyl diphosphate). The introduction of foreign mevalonate (MVA) pathway in E.coli can lead to supply of IPP and DMAPP in plenty amount. This study showed recombinant E. coli harboring MVA pathway was capable of producing farnesol, since fransol production was observed only in the recombinant E. coli strain. Approximately 80mg/L of farnesol was achieved in test-tube culture by additional ispA overexpression in E.coli with MVA pathway. Two phase culture of decane overlay to culture broth was applied for entrapment of volatile farnesol. As farnesol production was previously reported to 0.37mg/L in E.coli by acid phosphatse hydrolysis in vitro, and less than 30 mg/L in the yeast at aerobic and long time incubation, this work demonstrates the feasibility of farnesol production using E. coli as a host and its metabolic engineering. This work was supported by the 21C Frontier Microbial Genomics and Applications Center Program, EB-NCRC (Grant No. R15-2003-012-02001-0), and BK21 program of Korea.

We have previously developed an Escherichia coli strain which is 100% genetically defined, producing high-yield L-valine in batch culture by targeted genome engineering combined with transcriptome analysis and gene knock-out simulation of the in silico genome-scale metabolic network. This engineered strain was able to produce 7.55 g/L L-valine with 20 g/L glucose, resulting in an impressively high yield of 0.378 g L-valine per g glucose. In this study, the development of a fed-batch fermentation process based on in silico simulation was carried out. The results obtained here clearly demonstrate that the trade-off between L-valine and biomass formation should be optimized specifically towards L-valine production. Finally, the fedbatch culture of the further engineered strain allowed production of high concentration of 35.0 g/L L-valine. To our knowledge, the Lvaline concentration obtained here is the highest concentration ever reported in E. coli. [This work was supported by the Korean Systems Biology Research Project (20090065571) of the Ministry of Education, Science and Technology (MEST).]

Type III secretion systems (T3SS) are important virulence factors that Gram-negative pathogen bacteria use to inject virulence related proteins into eukaryotic host cells. The T3SS have been found in many animal pathogens, such as Salmonella spp., Shigella spp., Pseudomonas aeruginosa and Chlamydia spp.. The T3SS apparatus spans the bacterial inner and outer membranes and exports proteins past the cell wall. Salmonella subspecies are able to invade, survive, and replicate in multiple cell types from a wide range of species. Salmonella encode two virulence-associated T3SS, Salmonella pathogenicity islands 1 and 2 (SPI-1 and SPI-2). SPI-1 is expressed on the contact with host cells, and translocate effector proteins into host. In this study, we fused Salmonella SPI-1 effectors with model protein (lipase) to secret it into media. [This work was supported by National Research Foundation of Korea Grant funded by the Korean Government(KRF-2008-313- D00304)]