l-Arabinose/d-galactose 1-dehydrogenase of Rhizobium leguminosarum bv. trifolii characterised and applied for bioconversion of l-arabinose to l-arabonate with Saccharomyces cerevisiae

Four potential dehydrogenases identified through literature and bioinformatic searches were tested for l-arabonate production from l-arabinose in the yeast Saccharomyces cerevisiae. The most efficient enzyme, annotated as a d-galactose 1-dehydrogenase from the pea root nodule bacterium Rhizobium leguminosarum bv. trifolii, was purified from S. cerevisiae as a homodimeric protein and characterised. We named the enzyme as a l-arabinose/d-galactose 1-dehydrogenase (EC 1.1.1.-), Rl AraDH. It belongs to the Gfo/Idh/MocA protein family, prefers NADP⁺ but uses also NAD⁺ as a cofactor, and showed highest catalytic efficiency (k _cat/K _m) towards l-arabinose, d-galactose and d-fucose. Based on nuclear magnetic resonance (NMR) and modelling studies, the enzyme prefers the α-pyranose form of l-arabinose, and the stable oxidation product detected is l-arabino-1,4-lactone which can, however, open slowly at neutral pH to a linear l-arabonate form. The pH optimum for the enzyme was pH 9, but use of a yeast-in-vivo-like buffer at pH 6.8 indicated that good catalytic efficiency could still be expected in vivo. Expression of the Rl AraDH dehydrogenase in S. cerevisiae, together with the galactose permease Gal2 for l-arabinose uptake, resulted in production of 18 g of l-arabonate per litre, at a rate of 248 mg of l-arabonate per litre per hour, with 86 % of the provided l-arabinose converted to l-arabonate. Expression of a lactonase-encoding gene from Caulobacter crescentus was not necessary for l-arabonate production in yeast.     

Characterization of a recombinant l-fucose isomerase from Caldicellulosiruptor saccharolyticus that isomerizes l-fucose, d-arabinose, d-altrose, and l-galactose

A recombinant l-fucose isomerase from Caldicellulosiruptor saccharolyticus was purified as a single 68 kDa band with an activity of 76 U mg^?1. The molecular mass of the native enzyme was 204 kDa as a trimer. The maximum activity for l-fucose isomerization was at pH 7 and 75°C in the presence of 1 mM Mn²⁺. Its half-life at 70°C was 6.1 h. For aldose substrates, the enzyme displayed activity in decreasing order for l-fucose, with a k _cat of 11,910 min^?1 and a K _m of 140 mM, d-arabinose, d-altrose, and l-galactose. These aldoses were converted to the ketoses l-fuculose, d-ribulose, d-psicose, and l-tagatose, respectively, with 24, 24, 85, 55% conversion yields after 3 h.     

Enantioselective N-acetylation of 2-phenylglycine by an unusual N-acetyltransferase from Chryseobacterium sp.

The demand for d-2-phenylglycine used to synthesize semisynthetic antibiotics and pesticides is increasing. We have isolated a Chryseobacterium sp. that selectively transformed the l-form of racemic d,l-2-phenylglycine to (2S)-2-acetylamide-2-phenylacetic acid with a molar yield of 50 % and an enantiomer excess of >99.5 % under optimal culture conditions, consequently resulting in 99 % pure d-2-phenylglycine remaining in the culture. The enantioselective N-acetylation was catalyzed by an acetyl-CoA-dependent N-acetyltransferase whose synthesis was induced by l-2-phenylglycine. The enzyme differed from previously reported bacterial arylamine N-acetyltransferases in molecular mass and substrate specificity. The relative activity ratio of the enzyme with the substrates l-2-phenylglycine, d-2-phenylglycine, 2-(2-chlorophenyl)glycine, and 5-aminosalicylic acid (a good substrate of arylamine N-acetyltransferase) was 100:0:56.9:5.49, respectively. The biotransformation by the N-acetyltransferase-producing bacterium reported here could constitute a new preparative route for the enzymatic resolution of d,l-2-phenylglycine.     

NAD+-dependent xylitol dehydrogenase (xdhA) and l-arabitol-4-dehydrogenase (ladA) deletion mutants of Aspergillus oryzae for improved xylitol production

Xylitol dehydrogenase (XDHA) and l-arabitol dehydrogenase (LADA) are two key enzymes in xylan metabolism catalyzing the oxidation of xylitol to d-xylulose and arabitol to l-xylulose, respectively. In Aspergillus oryzae, XDHA and LADA are encoded by xdhA and ladA. We deleted xdhA and ladA and xdhA–ladA to generate mutants with decreased dehydrogenase activities and increased xylitol production. The mutants were constructed by homologous transformation into A. oryzae P4 (?pyrG) using pyrG as a selectable marker. The xylitol productivity of the mutants was measured using d-xylose as the sole carbohydrate source. xdhA, ladA, and the double-deletion mutant produced, respectively, 12.4 g xylitol/l with a yield of 0.24 g/g d-xylose, 12.4 g/l with a yield of 0.33 g/g d-xylose, and 8.6 g/l at a yield of 0.26 g/g d-xylose.     

Backbone and side-chain 1H, 15N and 13C assignment of apo- and imipenem-acylated l,d-transpeptidase from Bacillus subtilis

The d,d-transpeptidase activity of Penicillin Binding Proteins (PBPs) is essential to maintain cell wall integrity. PBPs catalyze the final step of the peptidoglycan synthesis by forming 4 → 3 cross-links between two peptide stems. Recently, a novel β-lactam resistance mechanism involving l,d-transpeptidases has been identified in Enterococcus faecium and Mycobacterium tuberculosis. In this resistance pathway, the classical 4 → 3 cross-links are replaced by 3 → 3 cross-links, whose formation are catalyzed by the l,d-transpeptidases. To date, only one class of the entire β-lactam family, the carbapenems, is able to inhibit the l,d-transpeptidase activity. Nevertheless, the specificity of this inactivation is still not understood. Hence, the study of this new transpeptidase family is of considerable interest in order to understand the mechanism of the l,d-transpeptidases inhibition by carbapenems. In this context, we present herein the backbone and side-chain ¹H, ¹⁵N and ¹³C NMR assignment of the l,d-transpeptidase from Bacillus subtilis (Ldt_Bs) in the apo and in the acylated form with a carbapenem, the imipenem.     

Function of aspartic acid residues in optimum pH control of l-arabinose isomerase from Lactobacillus fermentum

l-Arabinose isomerase (l-AI) catalyzes the isomerization of l-arabinose to l-ribulose and d-galactose to d-tagatose. Most reported l-AIs exhibit neutral or alkaline optimum pH, which is less beneficial than acidophilic ones in industrial d-tagatose production. Lactobacillus fermentum l-AI (LFAI) is a thermostable enzyme that can achieve a high conversion rate for d-galactose isomerization. However, its biocatalytic activity at acidic conditions can still be further improved. In this study, we report the single- and multiple-site mutagenesis on LFAI targeting three aspartic acid residues (D268, D269, and D299). Some of the lysine mutants, especially D268K/D269K/D299K, exhibited significant optimum pH shifts (from 6.5 to 5.0) and enhancement of pH stability (half-life time increased from 30 to 62 h at pH 6.0), which are more favorable for industrial applications. With the addition of borate, d-galactose was isomerized into d-tagatose by D268K/D269K/D299K at pH 5.0, resulting in a high conversion rate of 62 %. Based on the obtained 3.2-Å crystal structure of LFAI, the three aspartic acid residues were found to be distant from the active site and possibly did not participate in substrate catalysis. However, they were proven to possess similar optimum pH control ability in other l-AI, such as that derived from Escherichia coli. This study sheds light on the essential residues of l-AIs that can be modified for desired optimum pH and better pH stability, which are useful in d-tagatose bioproduction.     

Characterization of a recombinant l-rhamnose isomerase from Bacillus subtilis and its application on production of l-lyxose and l-mannose

The gene of an l-rhamnose isomerase (RhaA) from Bacillus subtilis was cloned to the pET28a(+) and then expressed in the E. coli ER2566. The expressed enzyme was purified with a specific activity of 3.58 U/mg by His-Trap affinity chromatography. The recombinant enzyme existed as a 194 kDa tetramer and the maximal activity was observed at pH 8.0 and 60°C. The RhaA displayed activity for l-rhamnose, l-lyxose, l-mannose, d-allose, d-gulose, d-ribose, and l-talose, among all aldopentoses and aldohexoses and it showed enzyme activity for l-form monosaccharides such as l-rhamnose, l-lyxose, l-mannose, and l-talose. The catalytic efficiency (k _cat/K _m) of the recombinant enzyme for l-rhamnose, l-lyxose, and l-mannose were 7,460, 1,013, and 258 M/sec. When l-xylulose 100 g/L and l-fructose 100 g/L were used as substrates, the optimum concentrations of RpiB were determined with 6 and 15 U/mL, respectively. The l-lyxose 40 g/L was produced from l-xylulose 100 g/L by the enzyme during 60 min, while l-mannose 25 g/L was produced from l-fructose 100 g/L for 80 min. The results suggest that RhaA from B. subtilis is a potential producer of l-form monosaccharides.     

Detection of d-amino acids in purified proteins synthesized in Escherichia coli

It has long been believed that amino acids comprising proteins of all living organisms are only of the l-configuration, except for Gly. However, peptidyl d-amino acids were observed in hydrolysates of soluble high molecular weight fractions extracted from cells or tissues of various organisms. This strongly suggests that significant amounts of d-amino acids are naturally present in usual proteins. Thus we analyzed the d-amino acid contents of His-tag-purified β-galactosidase and human urocortin, which were synthesized by Escherichia coli grown in controlled synthetic media. After acidic hydrolysis for various times at 110°C, samples were derivatized with 4-fluoro-7-nitro-2, 1, 3-benzoxadiazole (NBD-F) and separated on a reverse-phase column followed by a chiral column into d- and l-enantiomers. The contents of d-enantiomers of Ala, Leu, Phe, Val, Asp, and Glu were determined by plotting index d/(d + l) against the incubation time for hydrolysis and extrapolating the linear regression line to 0 h to eliminate the effect of racemization of amino acids during the incubation. Significant contents of d-amino acids were reproducibly detected, the d-amino acid profile being specific to an individual protein. This finding indicated the likelihood that d-amino acids are in fact present in the purified proteins. On the other hand, the d-amino acid contents of proteins were hardly influenced by the addition of d- or l-amino acids to the cultivation medium, whereas intracellular free d-amino acids sensitively varied according to the extracellular conditions. The origin of these d-amino acids detected in proteins was discussed.     

Protodioscin-glycosidase-1 hydrolyzing 26-O-β-d-glucoside and 3-O-(1 → 4)-α-l-rhamnoside of steroidal saponins from Aspergillus oryzae

A novel protodioscin-(steroidal saponin)-glycoside hydrolase, named protodioscin-glycosidase-1 (PGase-1), was purified and characterized from the Aspergillus oryzae strain. The molecular mass of this enzyme was determined to be about 55 kDa based on SDS-polyacrylamide gel electrophoresis. PGase-1 was able to hydrolyze the terminal 26-O-β-d-glucopyranoside of protodioscin (furostanoside) to produce dioscin (spirostanoside), and then further hydrolyze the terminal 3-O-(1?→?4)-α-l-rhamnopyranoside of dioscin to form progenin III. However, PGase-1 could hardly hydrolyze the 3-O-(1?→?2)-α-l-rhamnopyranoside of progenin III, 3-O-β-d-glucoside of trillin, and the 1-O-glycosides of ophiopogonin D (steroidal saponin). In addition, PGase-1 also could hydrolyze the α-d-galactopyranoside, β-d-glucopyranoside, and β-d-galactopyranoside of p-nitrophenyl-glycosides, but the enzyme could not hydrolyze the α-d-mannopyranoside, α-l-arabinopyranoside, α-d-glucopyranoside, β-d-xylopyranoside, and α-l-rhamnopyranoside of p-nitrophenyl-glycosides. These new properties of PGase-1 were significantly different from those of previously described steroidal saponin-glycosidases and the glycosidases currently described in Enzyme Nomenclature by the NC-IUBMB. The gene (termed as pgase-1) encoding PGase-1 was cloned, sequenced, and expressed in Pichia pastoris GS115. The complete nucleotide sequence of pgase-1 consists of 1,725 bp. The recombinant PGase-1 from recombinant P. pastoris GS115 strain also showed the activity hydrolyzing glycosides of steroidal saponins which was similar to that of the wild-type PGase-1 from A. oryzae. The PGase-1 gene is highly similar to Aspergilli α-amylase (EC 3.2.1.1), and PGase-1 should be classified as glycoside hydrolase family 13 by the method of gene sequence-based classification. But the enzyme properties of PGase-1 are different from those of α-amylase in this family.     

d-Tagatose production in the presence of borate by resting Lactococcus lactis cells harboring Bifidobacterium longum l-arabinose isomerase

Bifidobacterium longum NRRL B-41409 l-arabinose isomerase (l-AI) was overexpressed in Lactococcus lactis using a phosphate depletion inducible expression system. The resting L. lactis cells harboring the B. longum l-AI were used for production of d-tagatose from d-galactose in the presence of borate buffer. Multivariable analysis suggested that high pH, temperature and borate concentration favoured the conversion of d-galactose to d-tagatose. Almost quantitative conversion (92 %) was achieved at 20 g L^?1 substrate and at 37.5 °C after 5 days. The d-tagatose production rate of 185 g L^?1 day^?1 was obtained at 300 g L^?1 galactose, at 1.15 M borate, and at 41 °C during 10 days when the production medium was changed every 24 h. There was no significant loss in productivity during ten sequential 24 h batches. The initial d-tagatose production rate was 290 g L^?1 day^?1 under these conditions.     

Improved Protease Stability of the Antimicrobial Peptide Pin2 Substituted with d-Amino Acids

Cationic antimicrobial peptides (AMPs) have attracted a great interest as novel class of antibiotics that might help in the treatment of infectious diseases caused by pathogenic bacteria. However, some AMPs with high antimicrobial activities are also highly hemolytic and subject to proteolytic degradation from human and bacterial proteases that limit their pharmaceutical uses. In this work a d-diastereomer of Pandinin 2, d-Pin2, was constructed to observe if it maintained antimicrobial activity in the same range as the parental one, but with the purpose of reducing its hemolytic activity to human erythrocytes and improving its ability to resist proteolytic cleavage. Although, the hydrophobic and secondary structure characteristics of l- and d-Pin2 were to some extent similar, an important reduction in d-Pin2 hemolytic activity (30–40 %) was achieved compared to that of l-Pin2 over human erythrocytes. Furthermore, d-Pin2 had an antimicrobial activity with a MIC value of 12.5 μM towards Staphylococcus aureus, Escherichia coli, Streptococcus agalactiae and two strains of Pseudomonas aeruginosa in agar diffusion assays, but it was half less potent than that of l-Pin2. Nevertheless, the antimicrobial activity of d-Pin2 was equally effective as that of l-Pin2 in microdilution assays. Yet, when d- and l-Pin2 were incubated with trypsin, elastase and whole human serum, only d-Pin2 kept its antimicrobial activity towards all bacteria, but in diluted human serum, l- and d-Pin2 maintained similar peptide stability. Finally, when l- and d-Pin2 were incubated with proteases from P. aeruginosa DFU3 culture, a clinical isolated strain, d-Pin2 kept its antibiotic activity while l-Pin2 was not effective.     

Biosynthesis of ethylene glycol in Escherichia coli

Ethylene glycol (EG) is an important platform chemical with steadily expanding global demand. Its commercial production is currently limited to fossil resources; no biosynthesis route has been delineated. Herein, a biosynthesis route for EG production from d-xylose is reported. This route consists of four steps: d-xylose?→?d-xylonate?→?2-dehydro-3-deoxy-d-pentonate?→?glycoaldehyde?→?EG. Respective enzymes, d-xylose dehydrogenase, d-xylonate dehydratase, 2-dehydro-3-deoxy-d-pentonate aldolase, and glycoaldehyde reductase, were assembled. The route was implemented in a metabolically engineered Escherichia coli, in which the d-xylose?→?d-xylulose reaction was prevented by disrupting the d-xylose isomerase gene. The most efficient construct produced 11.7 g?L^?1 of EG from 40.0 g?L^?1 of d-xylose. Glycolate is a carbon-competing by-product during EG production in E. coli; blockage of glycoaldehyde?→?glycolate reaction was also performed by disrupting the gene encoding aldehyde dehydrogenase, but from this approach, EG productivity was not improved but rather led to d-xylonate accumulation. To channel more carbon flux towards EG than the glycolate pathway, further systematic metabolic engineering and fermentation optimization studies are still required to improve EG productivity.     

Characterization of a d-psicose-producing enzyme, d-psicose 3-epimerase, from Clostridium sp.

The gene coding for d-psicose 3-epimerase (DPEase) from Clostridium sp. BNL1100 was cloned and expressed in Escherichia coli. The recombinant enzyme was purified by Ni-affinity chromatography. It was a metal-dependent enzyme and required Co²⁺ as optimum cofactor. It displayed catalytic activity maximally at pH 8.0 and 65 °C (as measured over 5 min). The optimum substrate was d-psicose, and the K _m, turnover number (k _cat), and catalytic efficiency (k _cat/K _m) for d-psicose were 227 mM, 32,185 min^?1, and 141 min^?1 mM^?1, respectively. At pH 8.0 and 55 °C, 120 g d-psicose l^?1 was produced from 500 g d-fructose l^?1 after 5 h.     

Transportable and Non-transportable Inhibitors of L-glutamate Uptake Produce Astrocytic Stellation and Increase EAAT2 Cell Surface Expression

Astrocytic excitatory amino acid transporters (EAATs) regulate excitatory transmission and limit excitotoxicity. Evidence for a functional interface between EAATs and glial fibrillary acidic protein (GFAP) relevant to astrocytic morphology led to investigations of actions of transportable (d-Aspartate (d-Asp) and (2S,3S,4R)-2-(carboxycyclopropyl)glycine (l-CCG-III)) and non-transportable (dl-threo-β-benzyloxyaspartate (dl-TBOA)) inhibitors of Glu uptake in murine astrocytes. d-Asp (1 mM), l-CCG-III (0.5 mM) and dl-TBOA (0.5 mM) produced time-dependent (24–72 h) reductions in ³[H]d-Asp uptake (approximately 30–70%) with little or no gliotoxicity. All drugs induced a profound change in phenotype from cobblestone to stellate morphology and image analysis revealed increases in the intensity of GFAP immunolabelling for l-CCG-III and dl-TBOA. Cytochemistry indicated localized changes in F-actin distribution. Cell surface expression of EAAT2, but not EAAT1, was elevated at 72 h. Blockade of Glu uptake by both types of EAAT inhibitor exerts longer-term effects on astrocytic morphology and a compensatory homeostatic rise in EAAT2 abundance.     

Metabolic engineering of Escherichia coli for biosynthesis of d-galactonate

d-galactose is an attractive substrate for bioconversion. Herein, Escherichia coli was metabolically engineered to convert d-galactose into d-galactonate, a valuable compound in the polymer and cosmetic industries. d-galactonate productions by engineered E. coli strains were observed in shake flask cultivations containing 2 g L^?1 d-galactose. Engineered E. coli expressing gld coding for galactose dehydrogenase from Pseudomonas syringae was able to produce 0.17 g L^?1 d-galactonate. Inherent metabolic pathways for assimilating both d-galactose and d-galactonate were blocked to enhance the production of d-galactonate. This approach finally led to a 7.3-fold increase with d-galactonate concentration of 1.24 g L^?1 and yield of 62.0 %. Batch fermentation in 20 g L^?1 d-galactose of E. coli ?galK?dgoK mutant expressing the gld resulted in 17.6 g L^?1 of d-galactonate accumulation and highest yield of 88.1 %. Metabolic engineering strategy developed in this study could be useful for industrial production of d-galactonate.     

Principal component analysis of the relationship between the d-amino acid concentrations and the taste of the sake

We performed sensory evaluations on 141 bottles of sake and analyzed the relationship between the d-amino acid concentrations, and the taste of the sake using principal component analysis, which yielded seven principal components (PC1–7) that explained 100 % of the total variance in the data. PC1, which explains 33.6 % of the total variance, correlates most positively with strong taste and most negatively with balanced tastes. PC2, which explains 54.4 % of the total variance, correlates most positively with a sweet taste and most negatively with bitter and sour tastes. Sakes brewed with “Kimoto yeast starter” and “Yamahaimoto” had high scores for PC1 and PC2, and had strong taste in comparison with sakes brewed with “Sokujo-moto”. When present at concentrations below 50 μM, d-Ala did not affect the PC1 score, but all the sakes showed a high PC1 score, when the d-Ala was above 100 μM. Similar observations were found for the d-Asp and d-Glu concentrations with regard to PC1, and the threshold concentrations of d-Asp and d-Glu that affected the taste were 33.8 and 33.3 μM, respectively. Certain bacteria present in sake, especially lactic acid bacteria, produce d-Ala, d-Asp and d-Glu during storage, and these d-amino acids increased the PC1 score and produced a strong taste (Nojun). When d- and l-Ala were added to the sakes, the value for the umami taste in the sensory evaluation increased, with the effect of d-Ala being much stronger than that of l-Ala. The addition of 50–5,000 μM dl-Ala did not effect on the aroma of the sakes at all.     

A d-psicose 3-epimerase with neutral pH optimum from Clostridium bolteae for d-psicose production: cloning, expression, purification, and characterization

d-Tagatose 3-epimerase family enzymes can efficiently catalyze the epimerization of free keto-sugars, which could be used for d-psicose production from d-fructose. In previous studies, all optimum pH values of these enzymes were found to be alkaline. In this study, a d-psicose 3-epimerase (DPEase) with neutral pH optimum from Clostridium bolteae (ATCC BAA-613) was identified and characterized. The gene encoding the recombinant DPEase was cloned and expressed in Escherichia coli. In order to characterize the catalytic properties, the recombinant DPEase was purified to electrophoretic homogeneity using nickel-affinity chromatography. Ethylenediaminetetraacetic acid was shown to inhibit the enzyme activity completely; therefore, the enzyme was identified as a metalloprotein that exhibited the highest activity in the presence of Co²⁺. Although the DPEase demonstrated the most activity at a pH ranging from 6.5 to 7.5, it exhibited optimal activity at pH 7.0. The optimal temperature for the recombinant DPEase was 55 °C, and the half-life was 156 min at 55 °C. Using d-psicose as the substrate, the apparent K _m, k _cat, and catalytic efficiency (k _cat/K _m) were 27.4 mM, 49 s^?1, and 1.78 s^?1 mM^?1, respectively. Under the optimal conditions, the equilibrium ratio of d-fructose to d-psicose was 69:31. For high production of d-psicose, 216 g/L d-psicose could be produced with 28.8 % turnover yield at pH 6.5 and 55 °C. The recombinant DPEase exhibited weak-acid stability and thermostability and had a high affinity and turnover for the substrate d-fructose, indicating that the enzyme was a potential d-psicose producer for industrial production.     

Characterization of ergothionase from Burkholderia sp. HME13 and its application to enzymatic quantification of ergothioneine

We identified ergothionase, which catalyzes conversion of ergothioneine to thiolurocanic acid and trimethylamine, in a newly isolated ergothioneine-utilizing strain, Burkholderia sp. HME13. The enzyme was purified and its N-terminal amino acid sequence was determined. Based on the amino acid sequence, the gene encoding the enzyme was cloned and expressed in Escherichia coli. The recombinant enzyme was purified to homogeneity and characterized. The enzyme consisted of four identical 55-kDa subunits. The enzyme showed maximum activity at pH 8.0 and 65 °C and was stable between pH 7.0 and pH 10.0 and up to 60 °C. The enzyme acted on ergothioneine (K _m: 19 μM, V _max: 270 μmol/min/mg), but not d-histidine, l-histidine, d-tyrosine, l-tyrosine, d-phenylalanine, or l-phenylalanine. The enzyme was activated by BaCl₂ and strongly inhibited by CuSO₄, ZnSO₄, and HgCl₂. The amino acid sequence of ergothionase showed 23 % similarity to histidine ammonia-lyase (HAL) from Pseudomonas putida and 17 % similarity to phenylalanine ammonia-lyase (PAL) from parsley. However, the tripeptide sequence, Ala-Ser-Gly, which is important for catalysis in both HAL and PAL, was not conserved in ergothionase. The application of ergothionase for the quantification of ergothioneine contained in practical food and blood samples was investigated by performing a recovery test. Satisfactory recovery data (98.7–104 %) were obtained when ergothioneine was added to extract of tamogitake and hemolysis blood.     

Metabolic engineering Corynebacterium glutamicum for the l-lysine production by increasing the flux into l-lysine biosynthetic pathway

The experiments presented here were based on the conclusions of our previous results. In order to avoid introduction of expression plasmid and to balance the NADH/NAD ratio, the NADH biosynthetic enzyme, i.e., NAD-dependent glyceraldehyde-3-phosphate dehydrogenase (GADPH), was replaced by NADP-dependent GADPH, which was used to biosynthesize NADPH rather than NADH. The results indicated that the NADH/NAD ratio significantly decreased, and glucose consumption and l-lysine production drastically improved. Moreover, increasing the flux through l-lysine biosynthetic pathway and disruption of ilvN and hom, which involve in the branched amino acid and l-methionine biosynthesis, further improved l-lysine production by Corynebacterium glutamicum. Compared to the original strain C. glutamicum Lys5, the l-lysine production and glucose conversion efficiency (α) were enhanced to 81.0 ± 6.59 mM and 36.45 % by the resulting strain C. glutamicum Lys5-8 in shake flask. In addition, the by-products (i.e., l-threonine, l-methionine and l-valine) were significantly decreased as results of genetic modification in homoserine dehydrogenase (HSD) and acetohydroxyacid synthase (AHAS). In fed-batch fermentation, C. glutamicum Lys5-8 began to produce l-lysine at post-exponential growth phase and continuously increased over 36 h to a final titer of 896 ± 33.41 mM. The l-lysine productivity was 2.73 g l^?1 h^?1 and the α was 47.06 % after 48 h. However, the attenuation of MurE was not beneficial to increase the l-lysine production because of decreasing the cell growth. Based on the above-mentioned results, we get the following conclusions: cofactor NADPH, precursor, the flux through l-lysine biosynthetic pathway and DCW are beneficial to improve l-lysine production in C. glutamicum.