Bioprospecting for Novel Natural Products in Ancient Non‐Actinobacteria

Abstract

Polyketides are a particularly important category of secondary metabolite, and alone comprise 20% of the top‐selling drugs around the world. 1 Polyaromatic (type II) polyketides incorporate aromatic rings into their structures and are synthesized in vivo by a cohort of enzymes called a synthase. Three enzymes remain constant in all type II polyketide synthases: ketosynthase (KS), chain length factor (CLF), and acyl carrier protein (ACP). These three enzymes are known as the minimal polyketide synthase (PKS) and form the basis for the production of all type II polyketides. The most well‐studied of these polyketides are produced by a phylum known as Actinobacteria. Some of the polyketides that these bacteria produce include tetracycline, erythromycin, and lovastatin. Unfortunately, Actinobacteria are difficult to work with in the lab and researchers have historically been unsuccessful in expressing their KS‐CLF complexes in robust heterologous hosts such as E. coli. In 2015, a study by the Charkoudian and Hillenmeyer labs showed that the gene clusters responsible for the transcription and translation of the minimal PKS are present in some anciently diverged strains of non‐Actinobacteria. 2 These strains are known as orphan strains; the sequences of the genes are known but the enzymes and the product they manufacture remain uncharacterized. Many of the uncharacterized products include those made by non‐Actinobacterial species. Additionally, many of the identified non‐Actinobacterial species originate in phyla such as the Firmicutes and Proteobacteria, which means that their gene products could be expressed in tractable, non‐Actinobacterial heterologous hosts, such as S. pyogenes (a Firmicute) or E. coli (a Proteobacteria). Thus, non‐Actinobacterial ancient strains seemed ripe for bioprospecting.One of the first non‐Actinobacterial species that this research considered was the strain Gloeocapsa sp. PCC 7428, a cyanobacteria whose genome was found to contain a putative PKS gene cluster in the original 2015 study. Herein we show that the Gloeocapsa KS‐CLF can be expressed and purified in E. coli in high titers. Further characterization has shown this KS‐CLF to have several valuable qualities. The successful expression of such a protein may open doors to solutions to the antibiotic resistance problem that were previously closed.Support or Funding Information1. This work was supported by the Arnold & Mabel Beckman Foundation Scholar Award to G.S.H. and NSF CAREER grant number CHE‐1652424 to L.K.C.