Targeted Discovery of β‑Branched Conjugated Polyketides from Bacteria Based on Genomic and Metabolomic Hallmarks

A logical discovery method based on metabologenomic targeting was designed to search for new β-branched conjugated polyketides. PCR screening of a bacterial DNA library (1,638 strains) targeting the key enzyme hydroxymethylglutaryl-CoA synthase (HCS) homolog-encoding gene as a genomic hallmark identified 67 hit strains. Phylogenetic analysis of HCS homolog-encoding gene amplicons prioritized strains potentially producing new β-branched conjugated polyketides. Vibronic fine structures in ultraviolet spectra as a metabolomic hallmark clearly identified β-branched conjugated polyketides in bacterial extracts, logically leading to the discovery of paenillaene (1), paeniformicins A-D (2-5) from two strains belonging to the relatively unexplored genus Paenibacillus, and pulvomycins E-F (6-7) from a Streptomyces strain. Combined analysis of spectroscopic data, chemical reactions, quantum mechanics-based calculations, and biosynthetic gene clusters led to the structure elucidations of 1-7 as new β-branched conjugated polyketides. Paenillaene (1) dissociated amyloid-β aggregates, paeniformicin D (5) showed significant anti-tubercular activity against Mycobacterium tuberculosis, and pulvomycin E (6) displayed remarkable anti-proliferative activity against several human cancer cells, including a docetaxel-resistant breast cancer cell line. This metabologenomic targeting method provides an efficient tool for systematic discovery of new bioactive β-branched conjugated polyketides.