Novel aceQ and aceR genes involved in the acetan biosynthesis of Acetobacter xylinum were newly isolated. The homology search with DNA Data Bank of Japan indicated that aceQ and aceR were glycosyltransferases. Their gene-disrupted mutants were obtained by homologous recombination using the tetracycline resistance gene and the electroporation method. By NMR and ESI-MS analyses, aceQ-disrupted mutant DQ was found to secrete a water-soluble polysaccharide harboring the -Man-GlcUA side chain and the aceR-disrupted mutant DR was found to secrete an acetan analog, lacking the terminal Rha residue. These results suggested that aceQ and aceR encode a glucosyltransferase and a rhamnosyltransferase, respectively. It was indicated that acetan analogs harboring various side chains can be generated easily by genetic engineering.

A gene fragment encoding a putative pyrroloquinoline quinone glucose dehydrogenase (PQQ GDH) was cloned from a bacterial cellulose (BC)-forming acetic acid bacterium, Gluconacetobacter xylinus (=Acetobacter xylinum) strain BPR 2001, which was isolated as a high BC producer when using fructose as the carbon source. A GDH-deficient mutant of strain BPR 2001, namely GD-I, was then generated via gene disruption using the cloned gene fragment. Strain GD-I produced no gluconic acid but produced 4.1 g.l(-1) of BC aerobically in medium containing glucose as the carbon source. The ability of strain GD-I to convert glucose to BC was approximately 1.7-fold higher than that of the wild type. Strain GD-I was also able to produce 5.0 g.l(-1) of BC from a saccharified solution, which was derived from sweet potato pulp by enzymatic saccharification. Supplementation of ethanol during aerobic cultivation further increased the concentration of BC produced by strain GD-I to 7.0 g.l(-1). The rate of conversion from glucose to BC under these cultivation conditions was equivalent to that of strain BPR 2001 cultivated with fructose as the carbon source.

An open reading frame was found 214 bp downstream of the cellulose synthase operon of Acetobacter. The encoded amino acid sequence was found to be similar to some beta-glucosidases (G3ases). We detected G3ase activity in the culture medium and analysis of the N-terminal amino acid sequence showed that this gene encodes the enzyme. Therefore, it is possible that this region is a gene cluster for cellulose synthesis.

The 5' upstream region (about 3.1kb) of the cellulose synthase operon (bcs operon) has been isolated by cloning from Acetobacter xylinum strain BPR 2001. The expression level of the upstream region was determined using sucrose synthase cDNA as a reporter gene in the shuttle vector pSA19. The expression occurred with the 1.1-kb upstream sequence from the ATG start codon of the bcs operon but not with the 241-bp upstream sequence in A. xylinum, although neither the 1.1-kb nor the 241-bp upstream sequence caused any expression as a promoter in Escherichia coli. The level of expression with the 1. 1-kb upstream sequence in A. aceti was 75% of that in A. xylinum. These results suggest that the upstream region functions as a specific promoter for the Acetobacter genus. The expression was reduced by the introduction of the 241-bp upstream region between the lac promoter and the reporter gene in E. coli and was not detected in A. xylinum. This suggests that the short upstream region composed of 241bp contains the site(s) which causes a negative regulation on the transcription for bcs operon. The production of recombinant protein with the ribosome-binding site (RBS) of A. xylinum obtained from the bcs operon, was reduced to about half in E. coli, and that with the site of the lac promoter was also reduced to about half in A. xylinum. This shows that a species-specific predominance occurs during interaction between mRNA and 16S rRNA in the RBS between A. xylinum and E. coli.