The enduracidin biosynthetic gene cluster from Streptomyces fungicidicus

Abstract

The biosynthetic gene cluster for the 17 aa peptide antibiotic enduracidin has been cloned and sequenced from Streptomyces fungicidicus ATCC 21013. The 84 kb gene cluster contains 25 ORFs and is located within a 116 kb genetic locus that was fully sequenced. Targeted disruption of non-ribosomal peptide synthetase (NRPS) genes in the cluster abolished enduracidin production and confirmed function. The cluster includes four genes, endA-D, encoding two-, seven-, eight- and one-module NRPSs, respectively, and includes unique modules for the incorporation of citrulline and enduracididine. The NRPS organization generally follows the collinearity principle, and starts with a condensation domain (C domain) similar to those found in other lipopeptide systems for the coupling of an acyl group to the starting amino acid. The sixth module of EndB, corresponding to Thr8, is missing an adenylation domain (A domain) and this module is presumed to be loaded in trans by the single module protein EndD. The most striking feature of the NRPS organization is the lack of epimerization domains (E domains) in light of the fact that the product has seven d-amino acid residues. Sequence analysis reveals that C domains following modules corresponding to d-amino acids belong to a unique subset of C domains able to catalyse both epimerization and condensation reactions. Other genes directing lipid modification and activation, and formation of the non-proteinogenic amino acids 4-hydroxyphenylglycine and enduracididine are readily identified, as are genes possibly involved in regulation of antibiotic biosynthesis and export. These findings provide the basis to further genetically manipulate and improve lipodepsipeptide antibiotics via combinatorial and chemical methods.