Continuous volatile fatty acid production from lignocellulosic biomass by a novel rumen-mimetic bioprocess.

Lignocellulosic biomass is an attractive source of biofuels and biochemicals, being abundant in various plant sources. However, processing this type of biomass requires hydrolysis of cellulose. The proposed rumen-mimetic bioprocess consists of dry-pulverization of lignocellulosic biomass and pH-controlled continuous cultivation of ruminal bacteria using ammonium as a nitrogen source. In this study, ruminal bacteria were continuously cultivated for over 60 days and used to digest microcrystalline cellulose, rice straw, and Japanese cedar to produce volatile fatty acids (VFAs). The ruminal bacteria grew well in the chemically defined medium. The amounts of VFAs produced from 20 g of cellulose, rice straw, and Japanese cedar were 183 ± 29.7, 69.6 ± 12.2, and 21.8 ± 12.9 mmol, respectively. Each digestion completed within 24 h. The carbon yield was 60.6% when 180 mmol of VFAs was produced from 20 g of cellulose. During the cultivation, the bacteria were observed to form flocs that enfolded the feed particles. These flocs likely contain all of the bacterial species necessary to convert lignocellulosic biomass to VFAs and microbial protein symbiotically. Denaturing gradient gel electrophoresis (DGGE) analysis of PCR-amplified 16S rDNA fragments revealed that the bacterial community was relatively stable after 1 week in cultivation, though it was different from the original community structure. Furthermore, sequence analysis of the DGGE bands indicates that the microbial community includes a cellulolytic bacterium, a bacterium acting synergistically with cellulolytic bacteria, and a propionate-producing bacterium, as well as other anaerobic bacteria.

Heterologous production of kasugamycin, an aminoglycoside antibiotic from Streptomyces kasugaensis, in Streptomyces lividans and Rhodococcus erythropolis L-88 by constitutive expression of the biosynthetic gene cluster.

Kasugamycin (KSM), an aminoglycoside antibiotic isolated from Streptomyces kasugaensis cultures, has been used against rice blast disease for more than 50 years. We cloned the KSM biosynthetic gene (KBG) cluster from S. kasugaensis MB273-C4 and constructed three KBG cassettes (i.e., cassettes I-III) to enable heterologous production of KSM in many actinomycetes by constitutive expression of KBGs. Cassette I comprised all putative transcriptional units in the cluster, but it was placed under the control of the P neo promoter from Tn5. It was not maintained stably in Streptomyces lividans and did not transform Rhodococcus erythropolis. Cassette II retained the original arrangement of KBGs, except that the promoter of kasT, the specific activator gene for KBG, was replaced with P rpsJ , the constitutive promoter of rpsJ from Streptomyces avermitilis. To enhance the intracellular concentration of myo-inositol, an expression cassette of ino1 encoding the inositol-1-phosphate synthase from S. avermitilis was inserted into cassette II to generate cassette III. These two cassettes showed stable maintenance in S. lividans and R. erythropolis to produce KSM. Particularly, the transformants of S. lividans induced KSM production up to the same levels as those produced by S. kasugaensis. Furthermore, cassette III induced more KSM accumulation than cassette II in R. erythropolis, suggesting an exogenous supply of myo-inositol by the ino1 expression in the host. Cassettes II and III appear to be useful for heterologous KSM production in actinomycetes. Rhodococcus exhibiting a spherical form in liquid cultivation is also a promising heterologous host for antibiotic fermentation.

Massilia sp. BS-1, a novel violacein-producing bacterium isolated from soil.

A novel bacterium, Massilia sp. BS-1, producing violacein and deoxyviolacein was isolated from a soil sample collected from Akita Prefecture, Japan. The 16S ribosomal DNA of strain BS-1 displayed 93% homology with its nearest violacein-producing neighbor, Janthinobacterium lividum. Strain BS-1 grew well in a synthetic medium, but required both L-tryptophan and a small amount of L-histidine to produce violacein.

Hydroxylation of oleanolic acid to queretaroic acid by cytochrome P450 from Nonomuraea recticatena.

A gene for cytochrome P450 (moxA) from Nonomuraea recticatena, coexpressed with camAB for pseudomonad redox partners in Escherichia coli, hydroxylated oleanolic acid to produce queretaroic acid. When we used the P450-induced whole-cell as a catalyst, only a small amount of queretaroic acid was produced, probably due to poor permeability of oleanolic acid into the E. coli cell. In an alternative approach with the cell-free reaction system, the conversion ratio increased up to 17%.

Hydroxylation of testosterone by bacterial cytochromes P450 using the Escherichia coli expression system.

Two hundred thirteen cytochrome P450 (P450) genes were collected from bacteria and expressed based on an Escherichia coli expression system to test their hydroxylation ability to testosterone. Twenty-four P450s stereoselectively monohydroxylated testosterone at the 2alpha-, 2beta-, 6beta-, 7beta-, 11beta-, 12beta-, 15beta-, 16alpha-, and 17-positions (17-hydroxylation yields 17-ketoproduct). The hydroxylation site usage of the P450s is not the same as that of human P450s, while the 2alpha-, 2beta-, 6beta-, 11beta-, 15beta-, 16alpha-, and 17-hydroxylation are reactions common to both human and bacterial P450s. Most of the testosterone hydroxylation catalyzed by bacterial P450s is on the beta face.

Production of human metabolites of cyclosporin A, AM1, AM 4 N and AM 9, by microbial conversion.

In humans, cyclosporin A (CyA) is primarily metabolized to two hydroxylated (AM1 and AM 9) and one N-demethylated (AM 4 N) derivative. To produce these derivatives, 1237 actinomycetes were screened for their ability to convert CyA. Among them, 89 strains (7.2%) produced these derivatives from CyA. Finally, Dactylosporangium variesporum IFO 14104, Actinoplanes sp. ATCC 53771 and Streptosporangium sp. AF 935 were selected for the production of AM1, AM 4 N, and AM 9, respectively. Composition of the production medium and the incubation period with CyA were important for obtaining a high yield of CyA derivatives. Large-scale microbial conversion of CyA (1.5 g) using a 30-l jar fermentor yielded 288 mg of AM1 (19.2%), 147 mg of AM 4 N (9.8%) and 115 mg of AM 9 (7.7%).

Increase in the rate of L-pipecolic acid production using lat-expressing Escherichia coli by lysP and yeiE amplification.

Biotransformation of L-lysine (L-Lys) to L-pipecolic acid (L-PA) using lat-expressing Escherichia coli has been reported (Fujii et al., Biosci. Biotechnol. Biochem., 66, 622-627 (2002)). The rate-limiting step of this biotransformation seemes to be the transport of L-Lys into cells. To improve the L-PA production rate, we attempted to increase the rate of L-Lys uptake. E. coli BL21 carrying a plasmid with lat and lysP (pRH125) caused a 5-fold increase in the rate of L-PA production above the level of cells carrying a plasmid with lat (pRH124). Moreover, E. coli BL21 carrying a plasmid with lat, lysP, and yeiE (pRH127) caused a 6.4-fold increase in the rate of L-PA production above the level of cells carrying pRH124. Our results from RT-PCR experiments and the sequence similarity of YeiE to LysR transcriptional regulators suggest the possibility that yeiE expression induces lysP expression. The amplification of lysP, or rather both lysP and yeiE, increases the rate of L-PA production using lat-expressing E. coli.

Biotransformation of L-lysine to L-pipecolic acid catalyzed by L-lysine 6-aminotransferase and pyrroline-5-carboxylate reductase.

The enzyme involved in the reduction of delta1-piperideine-6-carboxylate (P6C) to L-pipecolic acid (L-PA) has never been identified. We found that Escherichia coli JM109 transformed with the lat gene encoding L-lysine 6-aminotransferase (LAT) converted L-lysine (L-Lys) to L-PA. This suggested that there is a gene encoding "P6C reductase" that catalyzes the reduction of P6C to L-PA in the genome of E. coli. The complementation experiment of proC32 in E. coli RK4904 for L-PA production clearly shows that the expression of both lat and proC is essential for the biotransformation of L-Lys to L-PA. Further, We showed that both LAT and pyrroline-5-carboxylate (P5C) reductase, the product of proC, were needed to convert L-Lys to L-PA in vitro. These results demonstrate that P5C reductase catalyzes the reduction of P6C to L-PA. Biotransformation of L-Lys to L-PA using lat-expressing E. coli BL21 was done and L-PA was accumulated in the medium to reach at an amount of 3.9 g/l after 159 h of cultivation. It is noteworthy that the ee-value of the produced pipecolic acid was 100%.