Production of 6-deoxy-13-cyclopropyl-erythromycin B by Saccharopolyspora erythraea NRRL 18643.

Cyclopropane carboxylic acid was fed to Saccharopolyspora erythraea NRRL 18643 (6-deoxyerythromycin producer), resulting in the production of 6-deoxy-13-cyclopropyl-erythromycin B. These studies provide further evidence that deoxyerythronolide B synthase has a relaxed specificity for the starter unit.

A second branched-chain alpha-keto acid dehydrogenase gene cluster (bkdFGH) from Streptomyces avermitilis: its relationship to avermectin biosynthesis and the construction of a bkdF mutant suitable for the production of novel antiparasitic avermectins.

A second cluster of genes encoding the E1 alpha, E1 beta, and E2 subunits of branched-chain alpha-keto acid dehydrogenase (BCDH), bkdFGH, has been cloned and characterized from Streptomyces avermitilis, the soil microorganism which produces anthelmintic avermectins. Open reading frame 1 (ORF1) (bkdF, encoding E1 alpha), would encode a polypeptide of 44,394 Da (406 amino acids). The putative start codon of the incompletely sequenced ORF2 (bkdG, encoding E1 beta) is located 83 bp downstream from the end of ORF1. The deduced amino acid sequence of bkdF resembled the corresponding E1 alpha subunit of several prokaryotic and eukaryotic BCDH complexes. An S. avermitilis bkd mutant constructed by deletion of a genomic region comprising the 5' end of bkdF is also described. The mutant exhibited a typical Bkd- phenotype: it lacked E1 BCDH activity and had lost the ability to grow on solid minimal medium containing isoleucine, leucine, and valine as sole carbon sources. Since BCDH provides an alpha-branched-chain fatty acid starter unit, either S(+)-alpha-methylbutyryl coenzyme A or isobutyryl coenzyme A, which is essential to initiate the synthesis of the avermectin polyketide backbone in S. avermitilis, the disrupted mutant cannot make the natural avermectins in a medium lacking both S(+)-alpha-methylbutyrate and isobutyrate. Supplementation with either one of these compounds restores production of the corresponding natural avermectins, while supplementation of the medium with alternative fatty acids results in the formation of novel avermectins. These results verify that the BCDH-catalyzed reaction of branched-chain amino acid catabolism constitutes a crucial step to provide fatty acid precursors for antibiotic biosynthesis in S. avermitilis.

Branched-chain fatty acid requirement for avermectin production by a mutant of Streptomyces avermitilis lacking branched-chain 2-oxo acid dehydrogenase activity.

The eight natural avermectins produced by Streptomyces avermitilis have the carbon skeleton of either isobutyric or S-2-methylbutyric acid incorporated into their structures. A mutant of S. avermitilis has been isolated that contains no functional branched-chain 2-oxo acid dehydrogenase activity. The mutant, in contrast to its parent, is unable to grow with isoleucine, valine and leucine as carbon sources. In medium lacking both S(+)-2-methylbutyric and isobutyric acid, the mutant is also incapable of making the natural avermectins, while supplementation with either one of these compounds restores production of the corresponding four natural avermectins. These facts indicate that in S. avermitilis the branched-chain 2-oxo acid dehydrogenase enzyme functions not only to catabolize the cellular branched-chain amino acids in order to meet energy and growth requirements but also to provide the small branched-chain organic acid precursor molecules necessary for avermectin biosynthesis. Supplementation of the mutant strain with R(-)-2-methylbutyric acid yields novel isomeric avermectins unseen in the (unsupplemented) wild-type strain. It was also concluded that acetate and propionate production by branched-chain 2-oxo acid degradation is not absolutely essential for avermectin production.