The Integration of Genome Mining, Comparative Genomics, and Functional Genetics for Biosynthetic Gene Cluster Identification.

Antimicrobial resistance is a worldwide health crisis for which new antibiotics are needed. One strategy for antibiotic discovery is identifying unique antibiotic biosynthetic gene clusters that may produce novel compounds. The aim of this study was to demonstrate how an integrated approach that combines genome mining, comparative genomics, and functional genetics can be used to successfully identify novel biosynthetic gene clusters that produce antimicrobial natural products. Secondary metabolite clusters of an antibiotic producer are first predicted using genome mining tools, generating a list of candidates. Comparative genomic approaches are then used to identify gene suites present in the antibiotic producer that are absent in closely related non-producers. Gene sets that are common to the two lists represent leading candidates, which can then be confirmed using functional genetics approaches. To validate this strategy, we identified the genes responsible for antibiotic production in Pantoea agglomerans B025670, a strain identified in a large-scale bioactivity survey. The genome of B025670 was first mined with antiSMASH, which identified 24 candidate regions. We then used the comparative genomics platform, EDGAR, to identify genes unique to B025670 that were not present in closely related strains with contrasting antibiotic production profiles. The candidate lists generated by antiSMASH and EDGAR were compared with standalone BLAST. Among the common regions was a 14 kb cluster consisting of 14 genes with predicted enzymatic, transport, and unknown functions. Site-directed mutagenesis of the gene cluster resulted in a reduction in antimicrobial activity, suggesting involvement in antibiotic production. An integrated approach that combines genome mining, comparative genomics, and functional genetics yields a powerful, yet simple strategy for identifying potentially novel antibiotics.

A broad-spectrum antibacterial natural product from the cystic fibrosis isolate, Pantoea agglomerans Tx10.

The prevalence of antibiotic-resistant Gram-positive and Gram-negative pathogens has prompted considerable efforts to identify new antibacterials. Here we show that Pantoea agglomerans Tx10-an isolate from the sputum sample of a cystic fibrosis patient-is a strong competitor that inhibits the growth of a wide range of Gram-positive and Gram-negative bacteria through the production of a secreted compound. A genetic screen to identify the genes involved in the production of this compound resulted in the delineation of a 6-gene biosynthetic cluster. We called this compound Pantoea Natural Product 2 (PNP-2). Assays with mutants deficient in PNP-2 production revealed they were still able to inhibit Erwinia amylovora, suggesting the production of a second antibiotic, which we identified as Pantocin A. We generated Pantocin A knockouts, and a PNP-2/Pantocin A double knockout and used these to evaluate the spectrum of activity of both natural products. We show that strains of Enterobacter, E. coli, Klebsiella, Kosakonia, Pseudocitrobacter, Salmonella, Staphylococcus, and Streptococcus as well as the majority of Pantoea strains assayed are susceptible to PNP-2, indicating a broad spectrum of activity, and potential for therapeutic development.