Natural product discovery and engineering at the protein, pathway, and genome scales

For centuries, mankind has turned to natural sources for cures, remedies, and ways to improve its quality of life. It was not until the discovery of the antibiotic penicillin in 1928, however, that natural products – small molecule secondary metabolites produced by a variety of organisms for a variety of purposes – were truly appreciated as the source of these beneficial properties. In the ensuing decades, research in the field of natural products boomed, and a number of discoveries were made that revolutionized global health. In the 1970s and 1980s, however, these discoveries started to become fewer and farther between, prompting pharmaceutical companies to turn away from natural products research in favor of synthetic chemistry approaches to drug discovery. Nevertheless, despite diminishing returns from traditional natural product discovery methods, modern genomics has in fact revealed that vast numbers of natural product gene clusters exist across all domains of life, far in excess of the number of known natural products and far surpassing any previous predictions. Thus, natural product discovery efforts to date have only scratched the surface of Nature’s true capabilities. In this work, we sought to leverage modern techniques for natural product discovery at the protein, pathway, and genome scales to tap into this biosynthetic potential. The past several years have seen the development of many tools for the manipulation of DNA with ease and precision far exceeding the standards set by traditional methods. These methods, combined with the ever-increasing number of putative natural product pathways revealed by genome sequencing efforts, open the door for new platforms of natural product discovery and engineering. At the protein level, we demonstrate the synthesis of novel derivatives of the antimalarial natural product FR-900098 by leveraging the substrate promiscuity of the native biosynthetic machinery. Through structural and biochemical characterization of the N-acetyltransferase FrbF and the corresponding target for FR-900098 inhibition, Dxr from Plasmodium falciparum, we show that the novel FR-900098 derivatives can serve as more potent inhibitors. A platform for their biosynthesis is also established in Escherichia coli. We additionally demonstrate a genome-mining platform for fungal polyketide synthases via the one-step assembly of expression-ready plasmids by homologous recombination in yeast. Through the evaluation of previously uncharacterized endogenous promoters from Saccharomyces cerevisiae, we demonstrate the heterologous production of polyketides from the dimorphic fungus Talaromyces marneffei. Extension of this approach to the pathway level is also demonstrated to assemble both a 6-gene resorcylic acid lactone cluster and a 13-gene phosphonic acid biosynthetic cluster. The latter case, in which all 13-genes from a phosphonic acid non-producing strain were placed under individual strong promoters, enabled the production of novel phosphonic acid compounds when heterologously expressed in Streptomyces lividans. At the genome scale, we developed a versatile system from genome engineering in a broad range of Streptomyces species. As the genus Streptomyces is by far the most important producer of pharmaceutical natural products to date, tools to facilitate their genetic engineering are of significant interest to aid discovery and improve production. Here, we show that the CRISPR/Cas9 system of S. pyogenes can be reconstituted in S. lividans for high-efficiency, multiplex editing of genomic loci. We show that deletions ranging from 20 bp to 31 kb can be successfully introduced in one step, and extend this approach to additional Streptomyces strains