In Vivo Roles of Fatty Acid Biosynthesis Enzymes in Biosynthesis of Biotin and α-Lipoic Acid in Corynebacterium glutamicum

Abstract

For fatty acid biosynthesis, Corynebacterium glutamicum uses two type I fatty acid synthases (FAS-I), FasA and FasB, in addition to acetyl-coenzyme A (CoA) carboxylase (ACC) consisting of AccBC, AccD1, and AccE. The in vivo roles of the enzymes in supplying precursors for biotin and α-lipoic acid remain unclear. Here, we report genetic evidence demonstrating that the biosynthesis of these cofactors is linked to fatty acid biosynthesis through the FAS-I pathway. For this study, we used wild-type C. glutamicum and its derived biotin vitamer producer BFI-5, which was engineered to express Escherichia coli bioBF and Bacillus subtilis bioI Disruption of either fasA or fasB in strain BFI-5 led to decreased production of biotin vitamers, whereas its amplification contributed to increased production, with a larger impact of fasA in both cases. Double disruptions of fasA and fasB resulted in no biotin vitamer production. The acc genes showed a positive effect on production when amplified simultaneously. Augmented fatty acid biosynthesis was also reflected in pimelic acid production when carbon flow was blocked at the BioF reaction. These results indicate that carbon flow down the FAS-I pathway is destined for channeling into the biotin biosynthesis pathway, and that FasA in particular has a significant impact on precursor supply. In contrast, fasB disruption resulted in auxotrophy for lipoic acid or its precursor octanoic acid in both wild-type and BFI-5 strains. The phenotypes were fully complemented by plasmid-mediated expression of fasB but not fasA These results reveal that FasB plays a specific physiological role in lipoic acid biosynthesis in C. glutamicumIMPORTANCE For the de novo biosynthesis of fatty acids, C. glutamicum exceptionally uses a eukaryotic multifunctional type I fatty acid synthase (FAS-I) system comprising FasA and FasB, in contrast to most bacteria, such as E. coli and B. subtilis, which use an individual nonaggregating type II fatty acid synthase (FAS-II) system. In this study, we reported genetic evidence demonstrating that the FAS-I system is the source of the biotin precursor in vivo in the engineered biotin-prototrophic C. glutamicum strain. This study also uncovered the important physiological role of FasB in lipoic acid biosynthesis. Here, we present an FAS-I enzyme that functions in supplying the lipoic acid precursor, although its biosynthesis has been believed to exclusively depend on FAS-II in organisms. The findings obtained here provide new insights into the metabolic engineering of this industrially important microorganism to produce these compounds effectively.

Keywords: Corynebacterium glutamicum; biotin; fatty acid biosynthesis; lipoic acid; metabolic engineering.

FIG 1

Proposed de novo biosynthetic pathways and the relevant genes of biotin and lipoic acid in C. glutamicum. Fatty acid biosynthesis in this organism begins with the reaction of acetyl-CoA carboxylase consisting of three subunits, AccBC, AccD1, and AccE, and then proceeds to the FAS-I pathway consisting of FasA and FasB. The biotin biosynthesis pathway of C. glutamicum is incomplete due to the lack of bioF and the gene for the de novo synthesis of pimeloyl-CoA (or pimeloyl-ACP). In the previous study, we demonstrated that E. coli bioBF and B. subtilis bioI could bridge the gaps (18). The origin of pimeloyl-CoA (or pimeloyl-ACP) in vivo could be the fatty acid biosynthesis pathway, but this remains speculative. In contrast, lipoic acid is assumed to be synthesized from octanoyl-CoA (or octanoyl-ACP) in a manner similar to that of E. coli (3). The octanoyl moiety is first transferred to the apoprotein (E2) by LipB and is then converted to lipoic acid by LipA to form lipoyl-E2. Also in this case, the origin of octanoyl-CoA (or octanoyl-ACP) remains an enigma. TCA, tricarboxylic acid.

FIG 2

Schematic diagram of the creation of strain C. glutamicum BFI-5 carrying E. coli bioBF and B. subtilis bioI on its genome. We previously constructed C. glutamicum BF-3, in which the E. coli genomic region comprising the bioBF gene cluster and its promoter/operator sequence (P/O) was inserted into the wild-type genome (18). Likewise, the B. subtilis bioI gene was inserted in the vicinity of the bioBF genes so as to be constitutively expressed under the promoter (P_gapA) of the endogenous gapA gene.

FIG 3

Biotin vitamer production by strain BFI-5 with disrupted fatty acid biosynthesis genes (A) and amplified fatty acid biosynthesis genes (B). Cultivations were carried out in 30 ml of biotin-free MM (1% glucose) supplemented with 10 μg of lipoic acid per liter in 300-ml baffled Erlenmeyer flasks. For cultures of strains BFI ΔfasA and BFI ΔfasAB, 1 g of Tween 80 per liter was added to satisfy the oleic auxotrophy. The control strain BFI-5 and strain BFI ΔfasB were cultivated under the conditions both with (+) and without (−) Tween 80 (1 g · liter⁻¹). Plasmid carriers were cultivated in the presence of 10 mg of kanamycin per liter. Under these conditions, the plasmid maintenance rate at the end of cultivation was more than 97.0% in all cultures. Titers of biotin vitamers are shown as the means and standard deviations of the results from three independent cultures. Growth values (■) are means of the results from three independent cultures, which showed <5% differences among them. Data for comparison between groups of the control vector carriers and the pCfasB carriers (*) were analyzed by Student's t test using JMP statistical software version 8.0.1 (SAS Institute, Cary, NC), and the differences were considered statistically significant at P values of <0.03. OD₆₀₀, OD at 600 nm.

FIG 4

Growth responses of wild-type strain ATCC 13032 and its fasA- and fasB-disrupted strains, WT ΔfasA and WT ΔfasB, respectively, to oleate, lipoic acid, and octanoic acid. (A) After appropriate dilutions of the cultures, an aliquot (approximately 10³ cells) was spread onto biotin (100 μg · liter⁻¹)-supplemented MM agar plates with and without 100 mg of sodium oleate, 10 μg of lipoic acid, or 1 mg of octanoic acid per liter and cultured at 30°C for 2 days. The pictures show one representative result from three independent experiments. (B) Cultivations were carried out at 30°C in biotin (100 μg · liter⁻¹)-supplemented MM liquid culture with no additions (○), 50 mg of sodium oleate (▲), 50 mg of sodium oleate plus 10 μg of lipoic acid (△), 100 mg of sodium oleate (◆), 100 mg of sodium oleate plus 10 μg of lipoic acid (♢), 10 μg of lipoic acid (■), or 1 mg of octanoic acid (□) per liter. The inoculum size from the seed culture to the main culture corresponds to 0.01%, as indicated in Materials and Methods. Values are means of the results from three independent cultures, which showed <5% differences among them.

FIG 5

Growth of strains WT ΔfasB carrying the vector plasmids pCS299P, pCfasB, and pCfasA. Cultivations were carried out at 30°C in biotin (100 μg · liter⁻¹)-supplemented MM liquid culture with no additions (○), 10 μg of lipoic acid (■), or 1 mg of octanoic acid (□) per liter. The inoculum size from the seed culture to the main culture corresponds to 0.01%. The plasmid maintenance rate at the end of cultivation was more than 97.0% in all cultures. Values are means of the results from three independent cultures, which showed <5% differences among them.