Membrane-bound sugar alcohol dehydrogenase in acetic acid bacteria catalyzes L-ribulose formation and NAD-dependent ribitol dehydrogenase is independent of the oxidative fermentation.

To identify the enzyme responsible for pentitol oxidation by acetic acid bacteria, two different ribitol oxidizing enzymes, one in the cytosolic fraction of NAD(P)-dependent and the other in the membrane fraction of NAD(P)-independent enzymes, were examined with respect to oxidative fermentation. The cytoplasmic NAD-dependent ribitol dehydrogenase (EC 1.1.1.56) was crystallized from Gluconobacter suboxydans IFO 12528 and found to be an enzyme having 100 kDa of molecular mass and 5 s as the sedimentation constant, composed of four identical subunits of 25 kDa. The enzyme catalyzed a shuttle reversible oxidoreduction between ribitol and D-ribulose in the presence of NAD and NADH, respectively. Xylitol and L-arabitol were well oxidized by the enzyme with reaction rates comparable to ribitol oxidation. D-Ribulose, L-ribulose, and L-xylulose were well reduced by the enzyme in the presence of NADH as cosubstrates. The optimum pH of pentitol oxidation was found at alkaline pH such as 9.5-10.5 and ketopentose reduction was found at pH 6.0. NAD-Dependent ribitol dehydrogenase seemed to be specific to oxidoreduction between pentitols and ketopentoses and D-sorbitol and D-mannitol were not oxidized by this enzyme. However, no D-ribulose accumulation was observed outside the cells during the growth of the organism on ribitol. L-Ribulose was accumulated in the culture medium instead, as the direct oxidation product catalyzed by a membrane-bound NAD(P)-independent ribitol dehydrogenase. Thus, the physiological role of NAD-dependent ribitol dehydrogenase was accounted to catalyze ribitol oxidation to D-ribulose in cytoplasm, taking D-ribulose to the pentose phosphate pathway after being phosphorylated. L-Ribulose outside the cells would be incorporated into the cytoplasm in several ways when need for carbon and energy sources made it necessary to use L-ribulose for their survival. From a series of simple experiments, membrane-bound sugar alcohol dehydrogenase was concluded to be the enzyme responsible for L-ribulose production in oxidative fermentation by acetic acid bacteria.

Purification and characterization of membrane-bound quinoprotein cyclic alcohol dehydrogenase from Gluconobacter frateurii CHM 9.

A quinoprotein catalyzing oxidation of cyclic alcohols was found in the membrane fraction for the first time, after extensive screening among aerobic bacteria. Gluconobacter frateurii CHM 9 was finally selected in this study. The enzyme tentatively named membrane-bound cyclic alcohol dehydrogenase (MCAD) was found to occur specifically in the membrane fraction, and pyrroloquinoline quinone (PQQ) was functional as the primary coenzyme in the enzyme activity. MCAD catalyzed only oxidation reaction of cyclic alcohols irreversibly to corresponding ketones. Unlike already known cytosolic NAD(P)H-dependent alcohol-aldehyde or alcohol-ketone oxidoreductases, MCAD was unable to catalyze the reverse reaction of cyclic ketones or aldehydes to cyclic alcohols. MCAD was solubilized and purified from the membrane fraction of the organism to homogeneity. Differential solubilization to eliminate the predominant quinoprotein alcohol dehydrogenase (ADH), and the subsequent two steps of column chromatographies, brought MCAD to homogeneity. Purified MCAD had a molecular mass of 83 kDa by SDS-PAGE. Substrate specificity showed that MCAD was an enzyme oxidizing a wide variety of cyclic alcohols. Some minor enzyme activity was found with aliphatic secondary alcohols and sugar alcohols, but not primary alcohols, differentiating MCAD from quinoprotein ADH. NAD-dependent cytosolic cyclic alcohol dehydrogenase (CCAD) in the same organism was crystallized and its catalytic and physicochemical properties were characterized. Judging from the catalytic properties of CCAD, it was apparent that CCAD was distinct from MCAD in many respects and seemed to make no contributions to cyclic alcohol oxidation.

Isolation and characterization of thermotolerant Gluconobacter strains catalyzing oxidative fermentation at higher temperatures.

Thermotolerant acetic acid bacteria belonging to the genus Gluconobacter were isolated from various kinds of fruits and flowers from Thailand and Japan. The screening strategy was built up to exclude Acetobacter strains by adding gluconic acid to a culture medium in the presence of 1% D-sorbitol or 1% D-mannitol. Eight strains of thermotolerant Gluconobacter were isolated and screened for D-fructose and L-sorbose production. They grew at wide range of temperatures from 10 degrees C to 37 degrees C and had average optimum growth temperature between 30-33 degrees C. All strains were able to produce L-sorbose and D-fructose at higher temperatures such as 37 degrees C. The 16S rRNA sequences analysis showed that the isolated strains were almost identical to G. frateurii with scores of 99.36-99.79%. Among these eight strains, especially strains CHM16 and CHM54 had high oxidase activity for D-mannitol and D-sorbitol, converting it to D-fructose and L-sorbose at 37 degrees C, respectively. Sugar alcohols oxidation proceeded without a lag time, but Gluconobacter frateurii IFO 3264T was unable to do such fermentation at 37 degrees C. Fermentation efficiency and fermentation rate of the strains CHM16 and CHM54 were quite high and they rapidly oxidized D-mannitol and D-sorbitol to D-fructose and L-sorbose at almost 100% within 24 h at 30 degrees C. Even oxidative fermentation of D-fructose done at 37 degrees C, the strain CHM16 still accumulated D-fructose at 80% within 24 h. The efficiency of L-sorbose fermentation by the strain CHM54 at 37 degrees C was superior to that observed at 30 degrees C. Thus, the eight strains were finally classified as thermotolerant members of G. frateurii.

Enzymes Responsible for Acetate Oxidation by Acetic Acid Bacteria.

Several acetic acid bacteria of the genus Acetobacter oxidize much acetate oxidation, which is not desired in vinegar manufacturing. Acetobacter rancens SKU 1111, a strong acetate oxidant, grew rapidly with a biphasic growth curve while consuming acetate in the second growth phase. Acetobacter aceti IFO 3284 did not show extensive acetate oxidation. Addition of glycerol to the culture medium of Acetobacter rancens SKU 1111 increased acetate oxidation and resulted in more biomass in the second growth phase than when glycerol was not added. Enzyme activities of acetyl-CoA synthetase and phosphotransacetylase in the organism were high during acetate oxidation. The activity of phosphoenolpyruvate carboxylase was most stimulated by a trace amount of acetyl-CoA among the enzymes of glycerol catabolism. Phosphoenolpyruvate carboxylase in A. rancens SKU 1111 showed a sigmoidal saturation curve with acetyl-CoA. This finding suggested that strong acetate oxidation caused by acetyl-CoA synthetase or phosphotransacetylase activity, together with phosphoenolpyruvate carboxylase, increased the biomass.

Crystallization and Properties of NADPH-Dependent L-Sorbose Reductase from Gluconobacter melanogenus IFO 3294.

NADPH-Dependent L-sorbose reductase (SORD, synonimously NADP-dependent D-srobitol dehydrogenase) was purified and crystallized for the first time from the cytosolic fraction of Gluconobacter melanogenus IFO 3294. The enzyme catalyzed oxidoreduction between D-sorbitol and L-sorbose in the presence of NADP or NADPH. Affinity chromatography by a Blue-dextran Sepharose 4B column was effective for purifying the enzyme giving about 770-fold purification with an overall yield of more than 50%. The crystalline enzyme showed a single sedimentation peak in analytical ultracentrifugation, giving an apparent sedimentation constant of 3.8 s. Gel filtration on a Sephadex G-75 column gave the molecular mass of 60 kDa to the enzyme, which dissociated into 30 kDa subunit on SDS-PAGE, indicating that the enzyme is composed of 2 identical subunits. Reduction of L-sorbose to D-sorbitol predominated in the presence of NADPH with the optimum pH of 5.0-7.0. Oxidation of D-sorbitol to L-sorbose was observed in the presence of NADP at the optimum pH of 7.0-9.0. The relative rate of L-sorbose reduction was more than seven times higher to that of D-sorbitol oxidation. NAD and NADH were inert for both reactions. D-Fructose reduction in the presence of NADPH did not occur with SORD. Since the reaction rate in L-sorbose reduction highly predominated over D-sorbitol oxidation over a wide pH range, the enzyme could be available for direct enzymatic measurement of L-sorbose. Even in the presence of a large excess of D-glucose and other substances, oxidation of NADPH to NADP was highly specific and stoichiometric to the L-sorbose reduced. Judging from the enzymatic properties, SORD would contribute to the intracellular assimilation of L-sorbose incorporated from outside the cells where L-sorbose is accumulated in huge amounts in the culture medium.

Crystallization and Properties of NAD-Dependent D-Sorbitol Dehydrogenase from Gluconobacter suboxydans IFO 3257.

NAD-dependent D-sorbitol dehydrogenase (EC 1.1.1.14) was crystallized from the cytosolic fraction of Gluconobacter suboxydans IFO 3257. This is the first example of the enzyme crystallized from acetic acid bacteria. The enzyme catalyzed oxidoreduction between D-sorbitol and D-fructose in the presence of NAD or NADH. The crystalline enzyme showed a single sedimentation peak in analytical ultracentrifugation, giving an apparent sedimentation constant of 5.1s. Gel filtration on a Sephadex G-200 column gave the molecular mass of 98 kDa for the enzyme, which dissociated into 26-kDa subunits on SDS-PAGE, indicating that the enzyme is composed of four identical subunits. Oxidation of D-sorbitol to D-fructose and xylitol to D-xylulose predominated in the presence of NAD at the optimum pH of 9.5-10.0. Reductions of D-fructose to D-sorbitol and D-xylulose to xylitol were also observed in the presence of NADH with the optimum pH around 6.0. The relative rate of D-fructose reduction was about one-fourth of that of D-sorbitol oxidation. NADP and NADPH were inert for the both reactions. Since the reation rate in D-sorbitol oxidation predominated over D-fructose reduction at some alkaline pH, the enzyme could be available for direct enzymatic measurement of D-sorbitol. Even in the presence of a large excess of D-glucose and other substances, reduction of NAD to NADH was highly specific and stoichiometric to the D-sorbitol oxidized.

In vivo expression of the Bacillus subtilis spoVE gene.

In vivo expression of the Bacillus subtilis spoVE gene was studied by S1 nuclease mapping and spoVE gene fusion analysis. Transcription of spoVE is induced at about the second hour of sporulation from two closely spaced promoters designated P1 and P2. Examination of the precise transcription initiation site by high-resolution primer extension mapping indicated that the nucleotide sequences of the -10 and -35 regions of both P1 and P2 were similar to those of promoters recognized by E sigma E. Moreover, S1 nuclease mapping and translational spoVE-lacZ fusion studies with various spo mutants suggest that the expression of spoVE P2 requires the spoIIG gene product, sigma E. The sporulation of a wild-type strain was inhibited severely in the presence of a multicopy plasmid, pKBVE, carrying the spoVE promoter, indicating the possible titration of a transcriptional regulatory element(s).

Bacillus subtilis spoVE gene is transcribed by sigma E-associated RNA polymerase.

Expression of the Bacillus subtilis sporulation gene spoVE was examined by runoff transcription assay with an RNA polymerase preparation obtained from vegetative and sporulating cells. Transcripts from tandem promoters (P1 and P2 promoters) located just upstream of the spoVE structure gene were detected. The transcription of spoVE initiated within an hour after the onset of sporulation and coincided with the presence of RNA polymerase associated with a 33-kDa protein. Amino acid sequence analysis of the 33-kDa protein revealed that it is a sigma factor, sigma E. Reconstitution analysis of sigma E purified from the sporulating cell extracts and vegetative core RNA polymerase showed that sigma E recognizes the P2 promoter. SpoVE protein could not be synthesized in the transcription-translation coupled system prepared from vegetative cells (M. Okamoto, S. Fukui, and Y. Kobayashi, Agric. Biol. Chem. 49:1077-1082, 1985). However, addition of sigma E-associated RNA polymerase to the coupled system restored SpoVE protein synthesis. These results indicate that spoVE expression in sporulating cells is controlled essentially by sigma E-associated RNA polymerase.

Sequence of the Bacillus subtilis homolog of the Escherichia coli cell-division gene murG.

The Bacillus subtilis homology of the Escherichia coli murG gene [encoding UDP-N-acetylglucosamine:N-acetylmuramyl-(pentapeptide) pyrophosphoryl-undecaprenol N-acetylglucosamine transferase] was cloned in E. coli K-12 and sequenced. The murG homolog encodes a protein of M(r) 39,936 [363 amino acid (aa) residues] of which 108 aa residues (29.8%) are identical with the E. coli murG product.