Partial Purification and Properties of l-Glutamine d-Fructose 6-Phosphate Amidotransferase from Phaseolus aureus

l-Glutamine d-fructose 6-phosphate amidotransferase (EC 2.6.1.16) was extracted and purified 600-fold by acetone fractionation and diethylaminoethyl cellulose column chromatography from mung bean seeds (Phaseolus aureus). The partially purified enzyme was highly specific for l-glutamine as an amide nitrogen donor, and l-asparagine could not replace it. The enzyme showed a pH optimum in the range of 6.2 to 6.7 in phosphate buffer. Km values of 3.8 mm and 0.5 mm were obtained for d-fructose 6-phosphate and l-glutamine, respectively. The enzyme was competitively inhibited with respect to d-fructose 6-phosphate by uridine diphosphate-N-acetyl-d-glucosamine which had a Ki value of 13 μm. Upon removal of l-glutamine and its replacement by d-fructose 6-phosphate and storage over liquid nitrogen, the enzyme was completely desensitized to inhibition by uridine diphosphate-N-acetyl-d-glucosamine. This indicates that the inhibitor site is distinct from the catalytic site and that uridine diphosphate-N-acetyl-d-glucosamine acts as a feedback inhibitor of the enzyme.     

An l-glucose Catabolic Pathway in Paracoccus Species 43P

An l-glucose-utilizing bacterium, Paracoccus sp. 43P, was isolated from soil by enrichment cultivation in a minimal medium containing l-glucose as the sole carbon source. In cell-free extracts from this bacterium, NAD⁺-dependent l-glucose dehydrogenase was detected as having sole activity toward l-glucose. This enzyme, LgdA, was purified, and the lgdA gene was found to be located in a cluster of putative inositol catabolic genes. LgdA showed similar dehydrogenase activity toward scyllo- and myo-inositols. l-Gluconate dehydrogenase activity was also detected in cell-free extracts, which represents the reaction product of LgdA activity toward l-glucose. Enzyme purification and gene cloning revealed that the corresponding gene resides in a nine-gene cluster, the lgn cluster, which may participate in aldonate incorporation and assimilation. Kinetic and reaction product analysis of each gene product in the cluster indicated that they sequentially metabolize l-gluconate to glycolytic intermediates, d-glyceraldehyde-3-phosphate, and pyruvate through reactions of C-5 epimerization by dehydrogenase/reductase, dehydration, phosphorylation, and aldolase reaction, using a pathway similar to l-galactonate catabolism in Escherichia coli. Gene disruption studies indicated that the identified genes are responsible for l-glucose catabolism.     

Accumulation of d-Glucose from Pentoses by Metabolically Engineered Escherichia coli

Escherichia coli that is unable to metabolize d-glucose (with knockouts in ptsG, manZ, and glk) accumulates a small amount of d-glucose (yield of about 0.01 g/g) during growth on the pentoses d-xylose or l-arabinose as a sole carbon source. Additional knockouts in the zwf and pfkA genes, encoding, respectively, d-glucose-6-phosphate 1-dehydrogenase and 6-phosphofructokinase I (E. coli MEC143), increased accumulation to greater than 1 g/liter d-glucose and 100 mg/liter d-mannose from 5 g/liter d-xylose or l-arabinose. Knockouts of other genes associated with interconversions of d-glucose-phosphates demonstrate that d-glucose is formed primarily by the dephosphorylation of d-glucose-6-phosphate. Under controlled batch conditions with 20 g/liter d-xylose, MEC143 generated 4.4 g/liter d-glucose and 0.6 g/liter d-mannose. The results establish a direct link between pentoses and hexoses and provide a novel strategy to increase carbon backbone length from five to six carbons by directing flux through the pentose phosphate pathway.     

myo-Inositol-1-Phosphatase from the Pollen of Lilium longiflorum Thunb

A Mg²⁺-dependent, alkaline phosphatase has been isolated from mature pollen of Lilium longiflorum Thunb., cv. Ace and partially purified. It hydrolyzes 1l- and 1d-myo-inositol 1-phosphate, myo-inositol 2-phosphate, and β-glycerophosphate at rates decreasing in the order named. The affinity of the enzyme for 1l- and 1d-myo-inositol 1-phosphate is approximately 10-fold greater than its affinity for myo-inositol 2-phosphate. Little or no activity is found with phytate, d-glucose 6-phosphate, d-glucose 1-phosphate, d-fructose 1-phosphate, d-fructose 6-phosphate, d-mannose 6-phosphate, or p-nitrophenyl phosphate. 3-Phosphosphoglycerate is a weak competitive inhibitor. myo-Inositol does not inhibit the reaction. Optimal activity is obtained at pH 8.5 and requires the presence of Mg²⁺. At 4 millimolar, Co²⁺, Fe²⁺ or Mn²⁺ are less effective. Substantial inhibition is obtained with 0.25 molar Li⁺. With β-glycerophosphate as substrate the K_m is 0.06 millimolar and the reaction remains linear at least 2 hours. In 0.1 molar Tris, β-glycerophosphate yields equivalent amounts of glycerol and inorganic phosphate, evidence that transphosphorylation does not occur.     

Nonreversible d-Glyceraldehyde 3-Phosphate Dehydrogenase of Plant Tissues

Preparations of TPN-linked nonreversible d-glyceraldehyde 3-phosphate dehydrogenase (EC 1.2.1.9), free of TPN-linked reversible d-glyceraldehyde 3-phosphate dehydrogenase, have been obtained from green shoots, etiolated shoots, and cotyledons of pea (Pisum sativum), cotyledons of peanut (Arachis hypogea), and leaves of maize (Zea mays). The properties of the enzyme were similar from each of these sources: the Km values for d-glyceraldehyde 3-phosphate and TPN were about 20 μm and 3 μm, respectively. The enzyme activity was inhibited by l-glyceraldehyde 3-phosphate, d-erythrose 4-phosphate, and phosphohydroxypyruvate. Activity was found predominantly in photosynthetic and gluconeogenic tissues of higher plants. A light-induced, phytochrome-mediated increase of enzyme activity in a photosynthetic tissue (pea shoots) was demonstrated. Appearance of enzyme activity in a gluconeogenic tissue (endosperm of castor bean, Ricinus communis) coincided with the conversion of fat to carbohydrate during germination. In photosynthetic tissue, the enzyme is located outside the chloroplast, and at in vivo levels of triose-phosphates and pyridine nucleotides, the activity is probably greater than that of DPN-linked reversible d-glyceraldehyde 3-phosphate dehydrogenase. Several possible roles for the enzyme in plant carbohydrate metabolism are considered.     

Structural Features of Sugars That Trigger or Support Conidial Germination in the Filamentous Fungus Aspergillus niger

The asexual spores (conidia) of Aspergillus niger germinate to produce hyphae under appropriate conditions. Germination is initiated by conidial swelling and mobilization of internal carbon and energy stores, followed by polarization and emergence of a hyphal germ tube. The effects of different pyranose sugars, all analogues of d-glucose, on the germination of A. niger conidia were explored, and we define germination as the transition from a dormant conidium into a germling. Within germination, we distinguish two distinct stages, the initial swelling of the conidium and subsequent polarized growth. The stage of conidial swelling requires a germination trigger, which we define as a compound that is sensed by the conidium and which leads to catabolism of d-trehalose and isotropic growth. Sugars that triggered germination and outgrowth included d-glucose, d-mannose, and d-xylose. Sugars that triggered germination but did not support subsequent outgrowth included d-tagatose, d-lyxose, and 2-deoxy-d-glucose. Nontriggering sugars included d-galactose, l-glucose, and d-arabinose. Certain nontriggering sugars, including d-galactose, supported outgrowth if added in the presence of a complementary triggering sugar. This division of functions indicates that sugars are involved in two separate events in germination, triggering and subsequent outgrowth, and the structural features of sugars that support each, both, or none of these events are discussed. We also present data on the uptake of sugars during the germination process and discuss possible mechanisms of triggering in the absence of apparent sugar uptake during the initial swelling of conidia.     

A Novel Bacterial Enzyme with d-Glucuronyl C5-epimerase Activity

Glycosaminoglycans are biologically active polysaccharides that are found ubiquitously in the animal kingdom. The biosynthesis of these complex polysaccharides involves complicated reactions that turn the simple glycosaminoglycan backbone into highly heterogeneous structures. One of the modification reactions is the epimerization of d-glucuronic acid to its C5-epimer l-iduronic acid, which is essential for the function of heparan sulfate. Although l-iduronic acid residues have been shown to exist in polysaccharides of some prokaryotes, there has been no experimental evidence for the existence of a prokaryotic d-glucuronyl C5-epimerase. This work for the first time reports on the identification of a bacterial enzyme with d-glucuronyl C5-epimerase activity. A gene of the marine bacterium Bermanella marisrubri sp. RED65 encodes a protein (RED65_08024) of 448 amino acids that has an overall 37% homology to the human d-glucuronic acid C5-epimerase. Alignment of this peptide with the human and mouse sequences revealed a 60% similarity at the carboxyl terminus. The recombinant protein expressed in Escherichia coli showed epimerization activity toward substrates generated from heparin and the E. coli K5 capsular polysaccharide, thereby providing the first evidence for bacterial d-glucuronyl C5-epimerase activity. These findings may eventually be used for modification of mammalian glycosaminoglycans.     

Substrate Specificity of a Mannose-6-Phosphate Isomerase from Bacillus subtilis and Its Application in the Production of l-Ribose

The uncharacterized gene previously proposed as a mannose-6-phosphate isomerase from Bacillus subtilis was cloned and expressed in Escherichia coli. The maximal activity of the recombinant enzyme was observed at pH 7.5 and 40°C in the presence of 0.5 mM Co²⁺. The isomerization activity was specific for aldose substrates possessing hydroxyl groups oriented in the same direction at the C-2 and C-3 positions, such as the d and l forms of ribose, lyxose, talose, mannose, and allose. The enzyme exhibited the highest activity for l-ribulose among all pentoses and hexoses. Thus, l-ribose, as a potential starting material for many l-nucleoside-based pharmaceutical compounds, was produced at 213 g/liter from 300-g/liter l-ribulose by mannose-6-phosphate isomerase at 40°C for 3 h, with a conversion yield of 71% and a volumetric productivity of 71 g liter⁻¹ h⁻¹.l-Ribose is a potential starting material for the synthesis of many l-nucleoside-based pharmaceutical compounds, and it is not abundant in nature (5, 19). l-Ribose has been produced mainly by chemical synthesis from l-arabinose, l-xylose, d-glucose, d-galactose, d-ribose, or d-mannono-1,4-lactone (2, 17, 23). Biological l-ribose manufacture has been investigated using ribitol or l-ribulose. Recently, l-ribose was produced from ribitol by a recombinant Escherichia coli containing an NAD-dependent mannitol-1-dehydrogenase (MDH) with a 55% conversion yield when 100 g/liter ribitol was used in a 72-h fermentation (18). However, the volumetric productivity of l-ribose in the fermentation is 28-fold lower than that of the chemical method synthesized from l-arabinose (8). l-Ribulose has been biochemically converted from l-ribose using an l-ribose isomerase from an Acinetobacter sp. (9), an l-arabinose isomerase mutant from Escherichia coli (4), a d-xylose isomerase mutant from Actinoplanes missouriensis (14), and a d-lyxose isomerase from Cohnella laeviribosi (3), indicating that l-ribose can be produced from l-ribulose by these enzymes. However, the enzymatic production of l-ribulose is slow, and the enzymatic production of l-ribose from l-ribulose has been not reported.Sugar phosphate isomerases, such as ribose-5-phosphate isomerase, glucose-6-phosphate isomerase, and galactose-6-phosphate isomerase, work as general aldose-ketose isomerases and are useful tools for producing rare sugars, because they convert the substrate sugar phosphates and the substrate sugars without phosphate to have a similar configuration (11, 12, 21, 22). l-Ribose isomerase from an Acinetobacter sp. (9) and d-lyxose isomerase from C. laeviribosi (3) had activity with l-ribose, d-lyxose, and d-mannose. Thus, we can apply mannose-6-phosphate (EC 5.3.1.8) isomerase to the production of l-ribose, because there are no sugar phosphate isomerases relating to l-ribose and d-lyxose. The production of the expensive sugar l-ribose (bulk price, $1,000/kg) from the rare sugar l-ribulose by mannose-6-phosphate isomerase may prove to be a valuable industrial process, because we have produced l-ribulose from the cheap sugar l-arabinose (bulk price, $50/kg) using the l-arabinose isomerase from Geobacillus thermodenitrificans (20) (Fig. ​(Fig.11).Open in a separate windowFIG. 1.Schematic representation for the production of l-ribulose from l-arabinose by G. thermodenitrificans l-arabinose isomerase and the production of l-ribose from l-ribulose by B. subtilis mannose-6-phosphate isomerase.In this study, the gene encoding mannose-6-phosphate isomerase from Bacillus subtilis was cloned and expressed in E. coli. The substrate specificity of the recombinant enzyme for various aldoses and ketoses was investigated, and l-ribulose exhibited the highest activity among all pentoses and hexoses. Therefore, mannose-6-phosphate isomerase was applied to the production of l-ribose from l-ribulose.     

Utilization of d-Ribitol by Lactobacillus casei BL23 Requires a Mannose-Type Phosphotransferase System and Three Catabolic Enzymes

Lactobacillus casei strains 64H and BL23, but not ATCC 334, are able to ferment d-ribitol (also called d-adonitol). However, a BL23-derived ptsI mutant lacking enzyme I of the phosphoenolpyruvate:carbohydrate phosphotransferase system (PTS) was not able to utilize this pentitol, suggesting that strain BL23 transports and phosphorylates d-ribitol via a PTS. We identified an 11-kb region in the genome sequence of L. casei strain BL23 (LCABL_29160 to LCABL_29270) which is absent from strain ATCC 334 and which contains the genes for a GlpR/IolR-like repressor, the four components of a mannose-type PTS, and six metabolic enzymes potentially involved in d-ribitol metabolism. Deletion of the gene encoding the EIIB component of the presumed ribitol PTS indeed prevented d-ribitol fermentation. In addition, we overexpressed the six catabolic genes, purified the encoded enzymes, and determined the activities of four of them. They encode a d-ribitol-5-phosphate (d-ribitol-5-P) 2-dehydrogenase, a d-ribulose-5-P 3-epimerase, a d-ribose-5-P isomerase, and a d-xylulose-5-P phosphoketolase. In the first catabolic step, the protein d-ribitol-5-P 2-dehydrogenase uses NAD⁺ to oxidize d-ribitol-5-P formed during PTS-catalyzed transport to d-ribulose-5-P, which, in turn, is converted to d-xylulose-5-P by the enzyme d-ribulose-5-P 3-epimerase. Finally, the resulting d-xylulose-5-P is split by d-xylulose-5-P phosphoketolase in an inorganic phosphate-requiring reaction into acetylphosphate and the glycolytic intermediate d-glyceraldehyde-3-P. The three remaining enzymes, one of which was identified as d-ribose-5-P-isomerase, probably catalyze an alternative ribitol degradation pathway, which might be functional in L. casei strain 64H but not in BL23, because one of the BL23 genes carries a frameshift mutation.     

Production of l-allose and d-talose from l-psicose and d-tagatose by l-ribose isomerase

l-ribose isomerase (L-RI) from Cellulomonas parahominis MB426 can convert l-psicose and d-tagatose to l-allose and d-talose, respectively. Partially purified recombinant L-RI from Escherichia coli JM109 was immobilized on DIAION HPA25L resin and then utilized to produce l-allose and d-talose. Conversion reaction was performed with the reaction mixture containing 10% l-psicose or d-tagatose and immobilized L-RI at 40 °C. At equilibrium state, the yield of l-allose and d-talose was 35.0% and 13.0%, respectively. Immobilized enzyme could convert l-psicose to l-allose without remarkable decrease in the enzyme activity over 7 times use and d-tagatose to d-talose over 37 times use. After separation and concentration, the mixture solution of l-allose and d-talose was concentrated up to 70% and crystallized by keeping at 4 °C. l-Allose and d-talose crystals were collected from the syrup by filtration. The final yield was 23.0% l-allose and 7.30% d-talose that were obtained from l-psicose and d-tagatose, respectively.     

myo-Inositol 1-Phosphate Synthase Inhibition and Control of Uridine Diphosphate-d-glucuronic Acid Biosynthesis in Plants

Of the eight intermediates associated with the two pathways of UDP-d-glucuronic acid biosynthesis found in plants, only d-glucuronic acid inhibited myo-inositol 1-phosphate synthase (EC 5.5.1.4), formerly referred to as d-glucose 6-phosphate cycloaldolase. Inhibition was competitive. An attempt to demonstrate over-all reversibility of the synthase indicated that it was less than 5% reversible, if at all.     

Ectopic expression of sorbitol-6-phosphate 2-dehydrogenase gene from <Emphasis Type="Italic">Haloarcula marismortui</Emphasis> enhances salt tolerance in transgenic <Emphasis Type="Italic">Arabidopsis thaliana</Emphasis>

d-Sorbitol-6-phosphate 2-dehydrogenase (S6PDH, E.C. 1.1.1.140) catalyzes the NADH-dependent conversion of d-fructose 6-phosphate (F6P) to d-sorbitol 6-phosphate (S6P). In this work, recombination and characterization of Haloarcula marismortui d-sorbitol-6-phosphate 2-dehydrogenase are reported. Haloarcula marismortui d-sorbitol-6-phosphate 2-dehydrogenase was expressed in P. pastoris and Arabidopsis thaliana. Enzyme assay indicated that HmS6PDH catalyzes the reduction of d-fructose 6-phosphate to d-sorbitol 6-phosphate and HmS6PDH activity was enhanced by NaCl. Furthermore, transgenic A. thaliana ectopic expressing HmS6PDH accumulate more sorbitol under salt stress. These results suggest that the ectopic expression of HmS6PDH in plants can facilitate future studies regarding the engineering and breeding of salt-tolerant crops.     

Crystal structure of a zinc-dependent D-serine dehydratase from chicken kidney

d-Serine is a physiological co-agonist of the N-methyl-d-aspartate receptor. It regulates excitatory neurotransmission, which is important for higher brain functions in vertebrates. In mammalian brains, d-amino acid oxidase degrades d-serine. However, we have found recently that in chicken brains the oxidase is not expressed and instead a d-serine dehydratase degrades d-serine. The primary structure of the enzyme shows significant similarities to those of metal-activated d-threonine aldolases, which are fold-type III pyridoxal 5′-phosphate (PLP)-dependent enzymes, suggesting that it is a novel class of d-serine dehydratase. In the present study, we characterized the chicken enzyme biochemically and also by x-ray crystallography. The enzyme activity on d-serine decreased 20-fold by EDTA treatment and recovered nearly completely by the addition of Zn²⁺. None of the reaction products that would be expected from side reactions of the PLP-d-serine Schiff base were detected during the >6000 catalytic cycles of dehydration, indicating high reaction specificity. We have determined the first crystal structure of the d-serine dehydratase at 1.9 Å resolution. In the active site pocket, a zinc ion that coordinates His³⁴⁷ and Cys³⁴⁹ is located near the PLP-Lys⁴⁵ Schiff base. A theoretical model of the enzyme-d-serine complex suggested that the hydroxyl group of d-serine directly coordinates the zinc ion, and that the ϵ-NH₂ group of Lys⁴⁵ is a short distance from the substrate Cα atom. The α-proton abstraction from d-serine by Lys⁴⁵ and the elimination of the hydroxyl group seem to occur with the assistance of the zinc ion, resulting in the strict reaction specificity.     

Structural and Functional Characterization of VanG d-Ala:d-Ser Ligase Associated with Vancomycin Resistance in Enterococcus faecalis

d-Alanyl:d-lactate (d-Ala:d-Lac) and d-alanyl:d-serine ligases are key enzymes in vancomycin resistance of Gram-positive cocci. They catalyze a critical step in the synthesis of modified peptidoglycan precursors that are low binding affinity targets for vancomycin. The structure of the d-Ala:d-Lac ligase VanA led to the understanding of the molecular basis for its specificity, but that of d-Ala:d-Ser ligases had not been determined. We have investigated the enzymatic kinetics of the d-Ala:d-Ser ligase VanG from Enterococcus faecalis and solved its crystal structure in complex with ADP. The overall structure of VanG is similar to that of VanA but has significant differences mainly in the N-terminal and central domains. Based on reported mutagenesis data and comparison of the VanG and VanA structures, we show that residues Asp-243, Phe-252, and Arg-324 are molecular determinants for d-Ser selectivity. These residues are conserved in both enzymes and explain why VanA also displays d-Ala:d-Ser ligase activity, albeit with low catalytic efficiency in comparison with VanG. These observations suggest that d-Ala:d-Lac and d-Ala:d-Ser enzymes have evolved from a common ancestral d-Ala:d-X ligase. The crystal structure of VanG showed an unusual interaction between two dimers involving residues of the omega loop that are deeply anchored in the active site. We constructed an octapeptide mimicking the omega loop and found that it selectively inhibits VanG and VanA but not Staphylococcus aureus d-Ala:d-Ala ligase. This study provides additional insight into the molecular evolution of d-Ala:d-X ligases and could contribute to the development of new structure-based inhibitors of vancomycin resistance enzymes.     

An Assessment of the Rubisco Inhibitor: 2-Carboxyarabinitol-1-Phosphate and d-Hamamelonic Acid 2-Phosphate Are Identical Compounds

2-Carboxyarabinitol-1-phosphate, the nocturnal inhibitor of ribulose-1,5-bisphosphate carboxylase/oxygenase is identical with d-hamamelonic acid-2¹-phosphate. Reasoning is based on theoretical considerations as well as on mass spectra and ¹H- and ¹³C-NMR spectra of the phosphate-free compounds. d-Hamamelonic acid-2¹-phosphate is interpreted as a metabolic derivative of d-hamamelose-2¹,5-bisphosphate which originates in the chloroplast from fructose-1,6-bisphosphate. A simple method for the synthesis of the inhibitor is suggested.     

Growth Inhibition and Metabolite Pool Levels in Plant Tissues Fed d-Glucosamine and d-Galactose

The growth of corn (Zea mays) roots and barley (Hordeum vulgare) coleoptiles is sensitive to the presence of external d-glucosamine and d-galactose. In order to investigate this effect, tissues were fed the radioactive monosaccharides at concentrations that ranged from those that were strongly inhibitory to those that had little influence on growth. At low concentrations, d-glucosamine is converted to uridine diphosphate-N-acetyl-d-glucosamine, phosphate esters of N-acetylglucosamine, and free N-acetylglucosamine. As the external concentrations were increased, the pool levels of each of these metabolites rose several fold; and, in corn roots, two unidentified compounds, which had not been detected previously, began to accumulate in the tissues. The major products of d-galactose metabolism were uridine diphosphate-d-galactose and d-galactose 1-phosphate at all the concentrations tested. Both these compounds showed a marked increase as the external galactose concentrations were raised to inhibitory levels. The experiments indicate that efficient pathways exist in plants for the metabolism of d-glucosamine and d-galactose. These pathways, however, do not appear to be under strict control, so that metabolites accumulate in unusually high amounts and presumably interfere competitively with normal carbohydrate metabolism.     

Further Studies on myo-Inositol-1-phosphatase from the Pollen of Lilium longiflorum Thunb

myo-Inositol-1-phosphatase has been purified to homogeneity from Lilium longiflorum pollen using an alternative procedure which includes pH change and phenyl Sepharose column chromatography. Sodium dodecyl sulfate-polyacrylamide gel electrophoretic analysis shows that the enzyme is a dimer (subunit molecular weight, 29,000 daltons). The enzyme is stable at low pH values and is inactivated only below pH 3.0. In addition to 1l-and 1d-myo-inositol-1-phosphate, it shows high specificity for 1l-chiro-inositol-3-phosphate. As observed earlier with other primary phosphate esters, d-glucitol-6-phosphate and d-mannitol-6-phosphate are hydrolyzed very slowly. No activity is observed with inorganic pyrophosphate or myo-inositol pentaphosphate as substrate. The enzyme is inhibited by fluoride, sulfate, molybdate, and thiol-directed reagents. Partial protection against N-ethylmaleimide inhibition by substrate and Mg²⁺ together suggests sulfhydryl involvement at the active site.     

Uptake of monosaccharides by guinea-pig cerebral-cortex slices

By the use of 1mm-iodoacetate to inhibit glycolysis in guinea-pig cerebral tissue slices, the kinetics of the uptake of monosaccharides on transfer of tissue from 0° to 37° were studied. d-Ribose, d-galactose, d-mannose, l-sorbose, and d-fructose showed diffusion kinetics, whereas 2-deoxy-d-glucose, d-glucose, d-arabinose and d-xylose showed saturation kinetics.     

Identification,Purification, and Characterization of a Novel Amino Acid Racemase,Isoleucine 2-Epimerase,from Lactobacillus Species

Accumulation of d-leucine, d-allo-isoleucine, and d-valine was observed in the growth medium of a lactic acid bacterium, Lactobacillus otakiensis JCM 15040, and the racemase responsible was purified from the cells and identified. The N-terminal amino acid sequence of the purified enzyme was GKLDKASKLI, which is consistent with that of a putative γ-aminobutyrate aminotransferase from Lactobacillus buchneri. The putative γ-aminobutyrate aminotransferase gene from L. buchneri JCM 1115 was expressed in recombinant Escherichia coli and then purified to homogeneity. The enzyme catalyzed the racemization of a broad spectrum of nonpolar amino acids. In particular, it catalyzed at high rates the epimerization of l-isoleucine to d-allo-isoleucine and d-allo-isoleucine to l-isoleucine. In contrast, the enzyme showed no γ-aminobutyrate aminotransferase activity. The relative molecular masses of the subunit and native enzyme were estimated to be about 49 kDa and 200 kDa, respectively, indicating that the enzyme was composed of four subunits of equal molecular masses. The K_m and V_max values of the enzyme for l-isoleucine were 5.00 mM and 153 μmol·min⁻¹·mg⁻¹, respectively, and those for d-allo-isoleucine were 13.2 mM and 286 μmol·min⁻¹·mg⁻¹, respectively. Hydroxylamine and other inhibitors of pyridoxal 5′-phosphate-dependent enzymes completely blocked the enzyme activity, indicating the enzyme requires pyridoxal 5′-phosphate as a coenzyme. This is the first evidence of an amino acid racemase that specifically catalyzes racemization of nonpolar amino acids at the C-2 position.