Development and experimental verification of a genome-scale metabolic model for Corynebacterium glutamicum
- Yohei Shinfuku†1,
- Natee Sorpitiporn†1,
- Masahiro Sono,
- Chikara Furusawa1, 2Email author,
- Takashi Hirasawa1 and
- Hiroshi Shimizu1Email author
© Shinfuku et al; licensee BioMed Central Ltd. 2009
Received: 22 April 2009
Accepted: 3 August 2009
Published: 3 August 2009
In silico genome-scale metabolic models enable the analysis of the characteristics of metabolic systems of organisms. In this study, we reconstructed a genome-scale metabolic model of Corynebacterium glutamicum on the basis of genome sequence annotation and physiological data. The metabolic characteristics were analyzed using flux balance analysis (FBA), and the results of FBA were validated using data from culture experiments performed at different oxygen uptake rates.
The reconstructed genome-scale metabolic model of C. glutamicum contains 502 reactions and 423 metabolites. We collected the reactions and biomass components from the database and literatures, and made the model available for the flux balance analysis by filling gaps in the reaction networks and removing inadequate loop reactions. Using the framework of FBA and our genome-scale metabolic model, we first simulated the changes in the metabolic flux profiles that occur on changing the oxygen uptake rate. The predicted production yields of carbon dioxide and organic acids agreed well with the experimental data. The metabolic profiles of amino acid production phases were also investigated. A comprehensive gene deletion study was performed in which the effects of gene deletions on metabolic fluxes were simulated; this helped in the identification of several genes whose deletion resulted in an improvement in organic acid production.
The genome-scale metabolic model provides useful information for the evaluation of the metabolic capabilities and prediction of the metabolic characteristics of C. glutamicum. This can form a basis for the in silico design of C. glutamicum metabolic networks for improved bioproduction of desirable metabolites.
A coryneform bacterium, Corynebacterium glutamicum is a facultatively aerobic, gram-positive bacterium that can grow on various sugars or organic acids [1, 2]. This organism can produce various amino acids such as glutamate [1, 2] and lysine  with high efficiency, and is thus widely used for the large-scale production of amino acids [4, 5]. Furthermore, the production of ethanol and organic acids such as lactate and succinate by growing C. glutamicum under oxygen deprivation conditions has recently been proposed [6, 7]. Owing to its importance for bioproduction, C. glutamicum has been chosen as one of the effective hosts for metabolic engineering purposes [8–10]. Thus, the construction and exploration of appropriate in silico metabolic models, which help predict the cellular behavior and production of useful chemicals, are highly desired.
Recently, on the basis of whole-genome information, the genome-scale metabolic networks of cells have been reconstructed and applied to metabolic flux balance analysis (FBA) [11, 12] for many organisms, including representatives of each of the 3 major domains of life, namely, archaea , bacteria [14–17], and eukarya [18–20]. FBA is an analysis of metabolic flux profiles, in which a steady state of metabolic flux is assumed and the profile of metabolic fluxes is calculated by optimizing an objective function using linear programming. Although genome-scale metabolic models cannot compute the detailed kinetic dynamics of metabolic reactions in a cell, they enable the description of the range of possible metabolic states on the basis of the constraints defined by the stoichiometry of metabolic reactions and transport steps at a steady state. Furthermore, we can obtain a solution, i.e., a set of all metabolic fluxes, which maximizes an objective function using linear programming. The biomass production rate is generally adopted as the objective function. It has been shown that the metabolic profiles calculated by the maximization of biomass production can describe those obtained experimentally in many organisms and environmental conditions, suggesting that organisms can maximize their growth rate by adaptation and evolution [21, 22]. Using the appropriate genome-scale metabolic network and an objective function to be maximized, FBA can be used to predict the relationship among the genotype, environmental conditions, and product yields at the steady state; this data can be utilized for the improvement of the microbial production [23, 24].
In this study, we present the reconstruction of a genome-scale metabolic model of C. glutamicum. Metabolic reactions and other parameters for biomass were collected using databases and literatures. After reconstruction of the genome-scale metabolic model, we performed FBA simulations with the maximization of biomass production and evaluated the results of the simulations by using experimental data of C. glutamicum cultures grown at different oxygen uptake rates (OURs). Our results revealed that the production rates of biomass and organic acids predicted using our model agreed well with the experimental rates; this result suggests that this model well represents the intracellular metabolic profiles of C. glutamicum. It should be noted that the representation of such changes in the metabolic profiles that occur on changing the oxygen uptake is difficult for another genome-scale model of C. glutamicum reported recently . Furthermore, by using the model proposed in this paper, we performed comprehensive simulations of gene deletions in order to identify candidate genes for genetic modification(s) to improve the productivity of organic acids by C. glutamicum.
Materials and methods
Modeling and Simulations
The known metabolic reactions in the C. glutamicum metabolic network were collected by a search of public databases and scientific publications. The genome-scale metabolic network was based on the pathways in the BioCyc database collection http://www.biocyc.org for C. glutamicum. We also referred to the information on C. glutamicum in the Kyoto Encyclopedia of Genes and Genomes database (KEGG; http://www.kegg.jp). In general, the genome-scale model constructed using only public databases contains incorrect and insufficient metabolic pathways owing to the incompleteness of database information. The most frequently observed incorrectness was missing enzymes in metabolic pathways. Thus, the resulting network was then subjected to the gap-filling process to allow biomass formation. For gap filling, we referred to published cell-specific data available in the literatures, such as Ref. [27, 28] and references threin. For example, the mycolic acid and arabinogalactan synthetic pathways are constructed on the basis of Refs. [29, 30].
Determination of biomass composition
The reaction "biomass synthesis" is a hypothetical metabolite in the metabolic network, which represents the requirement of precursors and coenzymes for the biomass formation. Biomass synthesis consists of a linear combination of 43 components, including amino acids, DNA, RNA, lipids, and cell envelope components. In our model, the biomass composition was determined from various reported data. The demands of precursors such as pyruvate, acetyl-CoA, and oxaloacetate for biomass production were based on the data in Ref. . For the macromolecular composition, we referred to the data in Ref. ; the total biomass is composed of 52% protein, 5% RNA, 1% DNA, 13% lipid, 19% cell wall components, and 10% other components. The C. glutamicum cells have a characteristic cell membrane termed as MAPc , which consists of the polysaccharides peptidoglycan and arabinogalactan as well as mycolic acids. Since the MAPc biosynthesis pathway is quite complex and some metabolic reactions in MAPc synthesis have not been characterized in detail, we described this pathway by some lumped reactions. The biomass composition of MAPc was determined to satisfy the macromolecular composition and the precursor demand described above. The biomass compositions of nucleic acids were calculated on the basis of the genomic sequence ; the composition ratios of DNA and RNA to the total biomass are 5% and 1%, respectively . The biomass synthesis reactions and its composition are shown in additional file 1. For all simulations presented here, the composition of every component was fixed independent of the environmental conditions. The energy requirement for biomass production was set to 41.26 mmol of ATP per 1 g biomass on the basis of the data in Ref..
In silico computation: FBA
where S is the stoichiometric matrix representing the stoichiometry of metabolic reactions in the network and v is a vector of all metabolic fluxes. vmin and vmax indicate the minimum and maximum constraints on the fluxes and are used to define the constraints for maximal enzymatic rate, irreversibility of reaction, or constant uptake from the environment. cT is a vector representing the objective function to be maximized, as a linear combination of metabolic fluxes. In general, the biomass production rate mentioned above is used as the objective function to be maximized. In this study, we followed this method to calculate the metabolic flux profile under the assumption that organisms have been evolved toward growth maximization. For all simulations in this paper, glucose was chosen as the sole carbon source and the following external metabolites were allowed to freely transport through the cell membrane: CO2, H2O, SO3, NH3, and PO4. All calculations, including the linear programming problems, were performed using the commercially available software Lindo (Lindo Systems, Inc.) and Matlab (Mathworks, Inc.).
Strain and medium
C. glutamicum strain ATCC 13032 was used in the culture experiments. The composition of the synthetic medium used for the preculture of the microorganism was the same as that employed in our previous study  (per liter of deionized water): 40 g of glucose, 30 g of (NH4)2SO4, 3.0 g of Na2HPO4, 6.0 g of KH2PO4, 2.0 g of NaCl, 84 mg of CaCl2, 3.9 mg of FeCl3, 0.9 mg of ZnSO4.7H2O, 0.1 mg of (NH4)6Mo7O21.4H2O, 0.3 mg of Na2B4O7.10H2O, 0.4 mg of MgSO4.7H2O, 40 mg of FeSO4.7H2O, 500 μg of vitamin B1·HCl, 0.1 g of EDTA, and 10 μg of biotin. The medium composition for the main culture was the same as that for the preculture except the initial glucose concentration was changed to 80 g/L for batch cultivation and 20 g/L for continuous cultivation.
All culture experiments were performed at 31.5°C, and the pH was maintained at 7.2 by the automatic addition of 25% ammonium solution using an autoclavable pH probe (Fermprobe F-635; Broadley-James Corporation, Irvine, CA) and a pH controller (DT-1023; Able Corporation, Japan).
The culture supernatant was used for measuring the concentration of glucose and organic acids. Glucose was measured using an enzymatic assay kit, the Glucose CII test Wako (Wako Pure Chemicals, Inc., Japan) according to the manufacturer's protocol. Organic acids were quantified using a HPLC system, Hitachi L-6200 equipped with a L-4000H UV detector (210 nm; HITACHI, Japan). Samples were eluted with 0.75 mM H2SO4 in an ion-exclusion column TSKgel Oapak-P (Tosho, Japan) at 40°C. The flow rate was set to 0.8 ml/min. The concentration of dissolved oxygen (DO) in the culture was monitored by using an oxygen electrode (DKK-Toa Corporation, Japan), and the concentrations of exhaust oxygen and carbon dioxide were monitored by using an exhaust gas analyzer (Model DEX-1562-1; Able Corporation, Japan).
Results and discussions
Reconstruction of the metabolic network
Genomic features and characteristics of a reconstructed metabolic model of C. glutamicum ATCC 13032.
No. of open reading frames (ORFs)
Total coding sequences (CDS)
CDS encoding annotated proteins
In silico metabolic networks
No. of genes included
No. of associated reactions
No. of other reactions
No. of metabolites
No. of internal fluxes
No. of exchange fluxes
Functional classification of metabolic reactions in a C. glutamicum genome-scale model.
Metabolism of complex carbohydrate
pentose phosphate pathway
glycerol and glycerophosphodiester degradation
fatty acid biosynthesis
isopentenyl diphosphate biosynthesis
glutathione redox reactions
nucleotide salvage pathway
Amino acid metabolism
Metabolism of complex lipid
Metabolism of cofactors and vitamins
coenzyme A biosynthesis
riboflavin and FMN and FAD biosynthesis
Metabolism of other amino acids
Metabolism of sugars
interconversion of arginine, ornithine and proline
The reconstructed metabolic network of C. glutamicum has several characteristics that distinguish it from the networks of other microorganisms. The cell envelopes of coryneform bacteria and mycobacteria have a unique structure consisting of a covalently linked complex comprising mycolic acid, arabinogalactan, and peptidoglycan (MAPc) . In order to represent the characteristics of cell envelope biosynthesis, we introduced metabolic reactions for MAPc biosynthesis into the model. The synthetic pathways and composition of MAPc were decided on the basis of previous studies [29, 30] and the demand of MAPc for biomass production was considered to be consistent with the cell wall composition of C. glutamicum. In the central metabolic pathway, the reconstructed model of C. glutamicum has pyruvate carboxylase in the anaplerotic pathway and lacks pyruvate formate lyase. In addition, C. glutamicum has 2 pathways for the biosynthesis of diaminopimelate, which is a precursor for lysine biosynthesis. One pathway involves the direct conversion of Δ1-piperideine-2,6-dicarboxylate to diaminopimelate, which is catalyzed by diaminopimelate dehydrogenase; the other pathway involves an indirect conversion, which is catalyzed by 4 independent enzymes . Additionally, C. glutamicum does not have glycine C-acetyltransferase, which is involved in the conversion of threonine to glycine. These distinguishing characteristics of the metabolic pathways are necessary to represent the flux profile of C. glutamicum.
Verification of the C. glutamicum genome-scale model: Measuring metabolic profiles under various oxygen supply conditions
In order to verify the results of the FBA using our genome-scale model of C. glutamicum, we compared the growth and metabolic profiles obtained by FBA simulations with those obtained by experiments performed at various oxygen levels. Here, we focused on the OUR as a parameter to change the metabolic profiles of C. glutamicum; this is because OUR is known to alter the metabolic profile drastically and is a key factor for controlling the productivity of several materials by this microorganism, such as organic acids. We performed a series of experiments involving the culture of C. glutamicum ATCC 13032 at various oxygen levels. The results of these experiments are summarized in Fig. 1(a). For aerobic conditions, i.e., conditions with relatively high OURs, we used continuous cultures with a dilution rate of 0.2 h-1; in these cultures, the oxygen supply was changed by altering the agitation speed to 800, 1000, and 1100 rpm (experiments 3, 4, and 5 in Fig. 1(a)). Here, the steady state was defined as that CV of cell concentration measured by optical density (OD660) is less than 10% over 10 hours. The production rates of CO2 and organic acids were measured in the steady states of the continuous cultures. We also confirmed that, for GUR, OUR, and production rates of organic acids, the ranges of deviation from the means were less than 5% during the steady state. In microaerobic and oxygen anaerobic conditions, C. glutamicum exhibited low specific growth rate (e.g., less than 0.05 h-1), and maintaining a steady state in the continuous cultures was difficult. Thus, we used batch cultures in this study and altered the oxygen supply conditions during the culture. In the batch cultures, cells were first grown under aerobic conditions with aeration and a high agitation speed; during the exponential growth phase, nitrogen gas started to be supplied instead of air for maintaining the anaerobic condition (experiment 1) and the agitation speed was controlled to maintain a constant OUR for the microaerobic condition (experiment 2). After changing the oxygen supply conditions, the production rates of CO2 and organic acids were measured (the time series of cell concentration and organic acid concentrations in experiments 1 and 2 are presented in additional files 2 and 3). Here, we obtained GUR, OUR, and the production rate of organic acids by linear regression of the time series. To check the error of this regression, we also calculated the confidence intervals of the regression and found that the 95% confidence intervals are less than 10% from the mean. In experiment 1, data of OUR was set to zero since nitrogen gas was supplied to the fermentor instead of air. As shown in Fig. 1(a), the production rates of biomass, CO2, and organic acids depended on the oxygen supply conditions. The experimental data presented in Fig. 1(a) indicated that under the anaerobic and micro-aerobic conditions (experiment 1 and 2), the cells converted most of the glucose to organic acids as lactate and succinate. With an increase in OUR/GUR ratio, cells changed their metabolism to produce acetate (experiment 3), and a further increase in OUR/GUR ratio resulted metabolic shift to CO2 production phase in which the tricarboxylic acid (TCA) cycle was activated (experiment 4 and 5).
Using the experimental data summarized in Fig. 1(a), we evaluated the results of the FBA simulations of our genome-scale model. In these simulations, we used biomass maximization as the objective function of FBA. We calculated the production yields of organic acids, CO2, and biomass using experimentally observed OUR and GUR values. The predicted yields are presented in Fig. 1(a). Also, in Fig. 1(b), we show a scatter plot of carbon yields, in which the carbon yields in the 5 sets of experimental and simulation results are presented. In Fig. 1(b), x-axis shows the predicted yield by FBA simulation and y-axis represents experimentally observed ones. As shown in the Fig. 1(a) and 1(b), the predictions of our genome-scale model with maximization of the biomass yield agreed well with the experimentally obtained yields. For example, the FBA simulation predicted that 10% of carbon is secreted as acetate when OUR/GUR ~1.5, which is consistent with the experimental result. The most significant discrepancy between the experimental results and FBA simulation was in the succinate production yield under the micro-aerobic and anaerobic conditions. The FBA simulation with biomass production maximization predicted that 10~20% of carbon is secreted as succinate in that condition, and the experimental results revealed that around 5% of carbon was secreted as succinate. The possible cause of this discrepancy will be discussed in the next section.
FBA of the metabolic profiles under various oxygen supply conditions
As shown in Fig. 1, the changes in the metabolic profiles in the FBA simulations mentioned above agreed well with the changes observed experimentally, except for the succinate production yield in the micro-aerobic and anaerobic conditions (phase IV and phase V). This discrepancy in the succinate production might be due to the differences between the simulation and experiment in the ATP demand for biomass production or in the P/O ratio. As mentioned above, succinate production from pyruvate requires ATP consumption in the anaplerotic pathway; thus, the oxidation of NADH by such succinate production tends to be preferred when the ATP demand for biomass production is small or the P/O ratio is large. In fact, when we changed the ATP demand for biomass production to 70.1 mmol per 1 g biomass, which was 1.7 fold higher than the original coefficient based on the previous report, the predicted production yield of succinate in the micro-aerobic condition showed good agreement with the experimental one. However, this change of coefficient resulted in the decrease of the biomass yield, and a significant discrepancy in the biomass yield between simulation and experimental results arose, for example, it was more than 30% in aerobic condition (OUR/GUR > 2). This result might suggest that the coefficient for ATP in the genome-scale model should be changed depending on OUR.
Comparison with previously proposed genome-scale model of C. glutamicum
The difference between the previously proposed genome-scale model of C. glutamicum  and our model is summarized in additional file 5. It should be noted that the previous model could not represent the changes that occur in the metabolic profile on altering the oxygen supply. In the FBA simulation using that model, the metabolic profile was not altered by changing the oxygen supply condition. One reason for this discrepancy was the inclusion of inadequate reaction loops in the genome-scale model. For example, that model consists of 2 transport reactions for urea; 1 reaction involves diffusion through the membrane without coupling to any other molecule and the other reaction is the urea-proton symport reaction. In FBA simulations, the combination of 2 such transport reactions results in an arbitrary proton flux from/to inside the cell; for example, the efflux of urea into the medium through the urea-proton symport and the intake of the same amount of urea by diffusion result in proton efflux without any changes in other metabolic balances except that of protons. Of course, the proton efflux can be balanced by proton-ATPase in the cell membrane with the generation of ATP. The genome-scale model reported in Ref.  consists several such inadequate reaction loops that allow the arbitrary generation of metabolites such as ATP and NADH; thus, representing the changes in the metabolic profiles by changing the oxygen supply condition is difficult. Furthermore, the model in Ref.  lacks the production pathways for lactate and succinate, which also make the representation of the metabolic state in the micro-aerobic condition difficult.
Analysis of the metabolic profile in the amino acid production phase
C. glutamicum is also widely used for glutamate production. However, the flux profile of the glutamate production phase cannot be represented by using the genome-scale model so far. Glutamate production by C. glutamicum can be induced by several triggers such as biotin depletion , Tween 40 addition , and penicillin addition . After receiving such a trigger, the cells cease growing and then start producing glutamate. Flux analysis using 13C-tracer experiments has revealed the changes in the metabolic flux profiles, indicating that a large fraction of carbon derived from glucose is converted to glutamate with the activation of the phosphoenol-pyruvate carboxylase pathway, which provides carbon flux to the TCA cycle . However, the identification of a metabolic state in which glutamate is produced and cell growth is arrested is difficult using analysis by a genome-scale model; this is because the possible range of metabolic states is fairly large and no appropriate objective function exists to predict such a metabolic state. In order to represent the metabolic profile with glutamate production by simulating the genome-scale model, further improvements in the simulation scheme are necessary. For example, it is well known that a decrease in the 2-oxoglutarate dehydrogenase complex activity plays an essential role in the metabolic change in the glutamate production phase [43, 44]. Additionally, a recent study suggested the existence of a membrane exporter of glutamate, whose conformation change can be involved in the transition to the glutamate production phase . The inclusion of such regulations of enzymatic activity might be important for the prediction of flux profiles of the glutamate production phase, and should be considered in future studies.
Analysis of gene deletion for improvement in the production of organic acids
Candidate reactions whose disruption increases the lactate or succinate production flux predicted by FBA simulations.
Reaction disabled by gene deletion
Lactate production flux (mmol/gDW/h)
Growth rate (1/h)
9.54 × 10-2 (1.00)
ADP + Pi + 4H [e] → ATP + H2O + H
8.50 × 10-2 (0.89)
R5P + Xu5P ↔ S7P + GAP
9.06 × 10-2 (0.95)
MAL ↔ FUM + H2O
8.97 × 10-2 (0.94)
G6P + NADP → 6PGL + NADPH + H
9.06 × 10-2 (0.95)
Reaction disabled by gene deletion
Succinate production flux (mmol/gDW/h)
Growth rate (1/h)
9.54 × 10-2 (1.00)
NADH + PYR + H ↔ LAC + NAD
8.11 × 10-2 (0.85)
In this study, we developed a genome-scale metabolic model of C. glutamicum, which is commercially important for the production of amino acids and useful chemicals. This model includes hundreds of metabolites and reactions among them, and also includes that the hypothetical reaction representing synthesis of biomass to calculate the requirement of metabolites for cell growth. It should be stressed that, this model is not just a collection of all the public available data about the metabolic reaction of C. glutamicum. Instead, we constructed the model by adding and deleting metabolic reactions to/from those on the databases, to make this model available for the flux balance analysis. For example, based on some literatures, we added some reactions to the metabolic model to fill gaps in the network to allow biomass formation. Also, we removed some reactions to avoid the formation of inadequate loop reactions, which makes impossible to calculate balance of metabolic fluxes due to the arbitrary generation of metabolites by this loop reaction as discussed above. Only after fixing these problems, we could use the metabolic model for FBA and predict the metabolic profiles. Using the genome-scale model, we performed FBA to understand the characteristics of the metabolic network and to identify candidates of the target metabolic pathways that can be manipulated to improve organic acid production by C. glutamicum. The results revealed that the FBA agreed well with the experimental results; this suggests that when the cells grow exponentially, the metabolic profiles can be predicted by our genome-scale model with maximization of the biomass production rate. We also performed simulations to predict the metabolic profiles in amino acid production phases, and succeeded in representing the metabolic profiles in the lysine production phase. However, the glutamate production phase, in which the cells stop growing, could not be represented by the genome-scale model. Further improvements in the model, such as inclusion of some gene regulatory machinery, should be considered. Furthermore, we performed in silico screening to identify candidates of the target metabolic pathways that can be manipulated to improve organic acid production by C. glutamicum. Our results revealed that the disruption of H+-ATPase activity is the most effective in improving lactate production under oxygen deprivation conditions. In fact, recent studies showed that the disruption of H+-ATPase results in significant changes of metabolic flux profiles in C. glutamicum , although the flux profiles under oxygen deprivation have not yet been investigated. The experimental verification of this in silico screening remains as future works. We expect that further extensive studies using our genome-scale model with experimental verifications will enable us to understand in detail the characteristics of the metabolic networks of C. glutamicum.
We thank to Mr. Satoshi Ohno for help of some simulations. This work was supported by Grants-in-Aid for Young Scientists (B) to CF and TH from the Ministry of Education, Culture, Sports, Science and Technology of Japan. This work was also supported in part by a grant from the Global COE (Center of Excellence) of the Ministry of Education, Culture, Sports, Science and Technology, Japan.
- Kinoshita S, Udaka S, Shimono M: Studies on the amino acid fermentation. Appl Microbiol Jpn. 1957, 3: 193-205. 10.2323/jgam.3.193.Google Scholar
- Udaka S: Screening method for microorganisms accumulating metabolites and its use in the isolation of micrococcus glutamicus. Jour Bacteriol. 1960, 79 (5): 754-755.Google Scholar
- Nakayama K, Kitada S, Kinoshita S: Studies on lysinefermentation. I. The control mechanism on lysine accumulation by homoserine and threonine. J Gen Appl Microbiol. 1961, 7: 145-154. 10.2323/jgam.7.145.View ArticleGoogle Scholar
- Kumagai H: Microbial Production of Amino Acids in Japan. Adv Biochem Eng Biotechnol. 2000, 69: 71-85.Google Scholar
- Leuchtenberger W, Huthmacher K, Drauz K: Biotechnological production of amino acids and derivatives: current status and prospects. Appl Microbiol Biotechnol. 2005, 69 (1): 1-8. 10.1007/s00253-005-0155-y.View ArticleGoogle Scholar
- Okino S, Inui M, Yukawa H: Production of organic acids by Corynebacterium glutamicum under oxygen deprivation. Appl Microbiol Biotechnol. 2005, 68 (4): 475-480. 10.1007/s00253-005-1900-y.View ArticleGoogle Scholar
- Inui M, Murakami S, Okino S, Kawaguchi H, Vertes AA, Yukawa H: Metabolic analysis of Corynebacterium glutamicum during lactate and succinate productions under oxygen deprivation conditions. J Mol Microbiol Biotechnol. 2004, 7 (4): 182-196. 10.1159/000079827.View ArticleGoogle Scholar
- Stephanopoulos G: Metabolic fluxes and metabolic engineering. Metab Eng. 1999, 1 (1): 1-11. 10.1006/mben.1998.0101.View ArticleGoogle Scholar
- de Graaf AA, Eggeling L, Sahm H: Metabolic engineering for L-lysine production by Corynebacterium glutamicum. Adv Biochem Eng Biotechnol. 2001, 73: 9-29.Google Scholar
- Shirai T, Fujimura K, Furusawa C, Nagahisa K, Shioya S, Shimizu H: Study on roles of anaplerotic pathways in glutamate overproduction of Corynebacterium glutamicum by metabolic flux analysis. Microb Cell Fact. 2007, 6: 19- 10.1186/1475-2859-6-19.View ArticleGoogle Scholar
- Edwards JS, Palsson BO: The Escherichia coli MG1655 in silico metabolic genotype: its definition, characteristics, and capabilities. Proc Natl Acad Sci USA. 2000, 97 (10): 5528-5533. 10.1073/pnas.97.10.5528.View ArticleGoogle Scholar
- Edwards JS, Ibarra RU, Palsson BO: In silico predictions of Escherichia coli metabolic capabilities are consistent with experimental data. Nat Biotechnol. 2001, 19 (2): 125-130. 10.1038/84379.View ArticleGoogle Scholar
- Feist AM, Scholten JC, Palsson BO, Brockman FJ, Ideker T: Modeling methanogenesis with a genome-scale metabolic reconstruction of Methanosarcina barkeri. Mol Syst Biol. 2006, 2:Google Scholar
- Feist AM, Henry CS, Reed JL, Krummenacker M, Joyce AR, Karp PD, Broadbelt LJ, Hatzimanikatis V, Palsson BO: A genome-scale metabolic reconstruction for Escherichia coli K-12 MG1655 that accounts for 1260 ORFs and thermodynamic information. Mol Syst Biol. 2007, 3: 121- 10.1038/msb4100155.View ArticleGoogle Scholar
- Oh YK, Palsson BO, Park SM, Schilling CH, Mahadevan R: Genome-scale reconstruction of metabolic network in Bacillus subtilis based on high-throughput phenotyping and gene essentiality data. J Biol Chem. 2007, 282 (39): 28791-28799. 10.1074/jbc.M703759200.View ArticleGoogle Scholar
- Thiele I, Vo TD, Price ND, Palsson BO: Expanded metabolic reconstruction of Helicobacter pylori (iIT341 GSM/GPR): an in silico genome-scale characterization of single- and double-deletion mutants. J Bacteriol. 2005, 187 (16): 5818-5830. 10.1128/JB.187.16.5818-5830.2005.View ArticleGoogle Scholar
- Oliveira AP, Nielsen J, Forster J: Modeling Lactococcus lactis using a genome-scale flux model. BMC Microbiol. 2005, 5: 39- 10.1186/1471-2180-5-39.View ArticleGoogle Scholar
- Duarte NC, Herrgard MJ, Palsson BO: Reconstruction and validation of Saccharomyces cerevisiae iND750, a fully compartmentalized genome-scale metabolic model. Genome Res. 2004, 14 (7): 1298-1309. 10.1101/gr.2250904.View ArticleGoogle Scholar
- Sheikh K, Forster J, Nielsen LK: Modeling hybridoma cell metabolism using a generic genome-scale metabolic model of Mus musculus. Biotechnol Prog. 2005, 21 (1): 112-121. 10.1021/bp0498138.View ArticleGoogle Scholar
- Duarte NC, Becker SA, Jamshidi N, Thiele I, Mo ML, Vo TD, Srivas R, Palsson BO: Global reconstruction of the human metabolic network based on genomic and bibliomic data. Proc Natl Acad Sci USA. 2007, 104 (6): 1777-1782. 10.1073/pnas.0610772104.View ArticleGoogle Scholar
- Ibarra RU, Edwards JS, Palsson BO: Escherichia coli K-12 undergoes adaptive evolution to achieve in silico predicted optimal growth. Nature. 2002, 420 (6912): 186-189. 10.1038/nature01149.View ArticleGoogle Scholar
- Fong SS, Palsson BO: Metabolic gene-deletion strains of Escherichia coli evolve to computationally predicted growth phenotypes. Nat Genet. 2004, 36 (10): 1056-1058. 10.1038/ng1432.View ArticleGoogle Scholar
- Lee SJ, Lee DY, Kim TY, Kim BH, Lee J, Lee SY: Metabolic engineering of Escherichia coli for enhanced production of succinic acid, based on genome comparison and in silico gene knockout simulation. Appl Environ Microbiol. 2005, 71 (12): 7880-7887. 10.1128/AEM.71.12.7880-7887.2005.View ArticleGoogle Scholar
- Alper H, Miyaoku K, Stephanopoulos G: Construction of lycopene-overproducing E. coli strains by combining systematic and combinatorial gene knockout targets. Nat Biotechnol. 2005, 23 (5): 612-616. 10.1038/nbt1083.View ArticleGoogle Scholar
- Kjeldsen KR, Nielsen J: In silico genome-scale reconstruction and validation of the Corynebacterium glutamicum metabolic network. Biotechnol Bioeng. 2009, 102 (2): 583-597. 10.1002/bit.22067.View ArticleGoogle Scholar
- Karp PD, Ouzounis CA, Moore-Kochlacs C, Goldovsky L, Kaipa P, Ahren D, Tsoka S, Darzentas N, Kunin V, Lopez-Bigas N: Expansion of the BioCyc collection of pathway/genome databases to 160 genomes. Nucleic Acids Res. 2005, 33 (19): 6083-6089. 10.1093/nar/gki892.View ArticleGoogle Scholar
- Eggeling L, Bott M: Handbook of corynebacterium glutamicum. 2005, Boca Raton: CRC PressView ArticleGoogle Scholar
- Burkovski A: Corynebacteria: Genomics and Molecular Biology. 2008, Norforlk: Caister Academic PressGoogle Scholar
- Birch HL, Alderwick LJ, Bhatt A, Rittmann D, Krumbach K, Sing A, Bai Y, Lowary TL, Eggeling L, Besra GS: Biosynthesis of mycobacterial arabinogalactan: identification of a novel α (1 → 3) arabinofuranosyltransferase. Molecular Microbiology. 2008, 69 (5): 1191-1206.Google Scholar
- Kacem R, De Sousa-D'Auria C, Tropis M, Chami M, Gounon P, Leblon G, Houssin C, Daffe M: Importance of mycoloyltransferases on the physiology of Corynebacterium glutamicum. Microbiology. 2004, 150 (Pt 1): 73-84. 10.1099/mic.0.26583-0.View ArticleGoogle Scholar
- Cocaign-Bousquet M, Guyonvarch A, Lindley ND: Growth Rate-Dependent Modulation of Carbon Flux through Central Metabolism and the Kinetic Consequences for Glucose-Limited Chemostat Cultures of Corynebacterium glutamicum. Appl Environ Microbiol. 1996, 62 (2): 429-436.Google Scholar
- Crick DC, Mahapatra S, Brennan PJ: Biosynthesis of the arabinogalactan-peptidoglycan complex of Mycobacterium tuberculosis. Glycobiology. 2001, 11 (9): 107R-118R. 10.1093/glycob/11.9.107R.View ArticleGoogle Scholar
- Kalinowski J, Bathe B, Bartels D, Bischoff N, Bott M, Burkovski A, Dusch N, Eggeling L, Eikmanns BJ, Gaigalat L, et al: The complete Corynebacterium glutamicum ATCC 13032 genome sequence and its impact on the production of L-aspartate-derived amino acids and vitamins. J Biotechnol. 2003, 104 (1–3): 5-25. 10.1016/S0168-1656(03)00154-8.View ArticleGoogle Scholar
- Cocaign-Bousquet M, Lindley ND: Pyruvate overflow and carbon flux within the central metabolic pathways of Corynebacterium glutamicum during growth on lactate. Enzyme and Microbial Technology. 1995, 17 (3): 260-267. 10.1016/0141-0229(94)00023-K.View ArticleGoogle Scholar
- Shirai T, Nakato A, Izutani N, Nagahisa K, Shioya S, Kimura E, Kawarabayasi Y, Yamagishi A, Gojobori T, Shimizu H: Comparative study of flux redistribution of metabolic pathway in glutamate production by two coryneform bacteria. Metab Eng. 2005, 7 (2): 59-69. 10.1016/j.ymben.2004.10.001.View ArticleGoogle Scholar
- Ikeda M, Nakagawa S: The Corynebacterium glutamicum genome: features and impacts on biotechnological processes. Appl Microbiol Biotechnol. 2003, 62 (2–3): 99-109. 10.1007/s00253-003-1328-1.View ArticleGoogle Scholar
- Schrumpf B, Schwarzer A, Kalinowski J, Puhler A, Eggeling L, Sahm H: A functionally split pathway for lysine synthesis in Corynebacterium glutamicium. J Bacteriol. 1991, 173 (14): 4510-4516.Google Scholar
- Sonntag K, Eggeling L, De Graaf AA, Sahm H: Flux partitioning in the split pathway of lysine synthesis in Corynebacterium glutamicum. Quantification by 13C- and 1H-NMR spectroscopy. Eur J Biochem. 1993, 213 (3): 1325-1331. 10.1111/j.1432-1033.1993.tb17884.x.View ArticleGoogle Scholar
- Vallino JJ, Stephanopoulos G: Metabolic flux distributions in Corynebacterium glutamicum during growth and lysine overproduction. Reprinted from Biotechnology and Bioengineering, Vol. 41, Pp 633–646 (1993). Biotechnol Bioeng. 2000, 67 (6): 872-885. 10.1002/(SICI)1097-0290(20000320)67:6<872::AID-BIT21>3.0.CO;2-X.View ArticleGoogle Scholar
- Shiio I, Otsuka SI, Takahashi M: Effect of biotin on the bacterial formation of glutamic acid. I. Glutamate formation and cellular premeability of amino acids. J Biochem. 1962, 51: 56-62.Google Scholar
- Takinami K, Yoshida H, Tsuri H, Okada H: Biochemical effects of fatty acid and its derivatives on L-glutamic acid fermentation. Part III. Biotin-Tween 60 relationship in the accumulation of L-glutamic acid and the growth of Brevibacterium lactofermentum. Agric Biol Chem. 1965, 29: 351-359.View ArticleGoogle Scholar
- Nunheimer TD, Birnbaum J, Ihnen ED, Demain AL: Product inhibition of the fermentative formation of glutamic acid. Appl Microbiol. 1970, 20 (2): 215-217.Google Scholar
- Shimizu H, Tanaka H, Nakato A, Nagahisa K, Kimura E, Shioya S: Effects of the changes in enzyme activities on metabolic flux redistribution around the 2-oxoglutarate branch in glutamate production by Corynebacterium glutamicum. Bioprocess Biosyst Eng. 2003, 25 (5): 291-298.View ArticleGoogle Scholar
- Bott M: Offering surprises: TCA cycle regulation in Corynebacterium glutamicum. Trends Microbiol. 2007, 15 (9): 417-425. 10.1016/j.tim.2007.08.004.View ArticleGoogle Scholar
- Nakamura J, Hirano S, Ito H, Wachi M: Mutations of the Corynebacterium glutamicum NCgl1221 gene, encoding a mechanosensitive channel homolog, induce L-glutamic acid production. Appl Environ Microbiol. 2007, 73 (14): 4491-4498. 10.1128/AEM.02446-06.View ArticleGoogle Scholar
- Aoki R, Wada M, Takesue N, Tanaka K, Yokota A: Enhanced glutamic acid production by a H+-ATPase-defective mutant of Corynebacterium glutamicum. Biosci Biotechnol Biochem. 2005, 69 (8): 1466-1472. 10.1271/bbb.69.1466.View ArticleGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.