Abstract
In living organisms, modified fatty acids are crucial for the functions of the cellular membranes and storage lipids where the fatty acids are esterified. Some bacteria produce a typical methyl-branched fatty acid, i.e., 10-methyl stearic acid (19:0Me10). The biosynthetic pathway of 19:0Me10 in vivo has not been demonstrated clearly yet. It had been speculated that 19:0Me10 is synthesized from oleic acid (18:1Δ9) by S-adenosyl-L-methionine-dependent methyltransfer and NADPH-dependent reduction via a methylenated intermediate, 10-methyelene octadecanoic acid. Although the recombinant methyltransferases UmaA and UfaA1 from Mycobacterium tuberculosis H37Rv synthesize 19:0Me10 from 18:1Δ9 and NADPH in vitro, these methyltransferases do not possess any domains functioning in the redox reaction. These findings may contradict the two-step biosynthetic pathway. We focused on novel S-adenosyl-L-methionine-dependent methyltransferases from Mycobacterium chlorophenolicum that are involved in 19:0Me10 synthesis and selected two candidate proteins, WP_048471942 and WP_048472121, by a comparative genomic analysis. However, the heterologous expression of these candidate genes in Escherichia coli cells did not produce 19:0Me10. We found that one of the candidate genes, WP_048472121, was collocated with another gene, WP_048472120, that encodes a protein containing a domain associated with flavin adenine dinucleotide-binding oxidoreductase activity. The co-expression of these proteins (hereafter called BfaA and BfaB, respectively) led to the biosynthesis of 19:0Me10 in E. coli cells via the methylenated intermediate.