Engineering Yeast as a Host for Modular Polyketide Production

Natural products have demonstrated incredible value to human welfare through their pharmaceutical properties. Polyketides are a particularly interesting class of natural products, however, these compounds are often difficult to synthesize and their hosts are rarely amenable to genetic engineering. We are interested in polyketides produced by bacterial polyketide synthases (PKSs). These modular enzymes assemble their products in an assembly-line fashion, and module-prediction at the DNA level corresponds to the specific chemical modifications that the enzyme performs on its growing poly-acyl chain. We seek to express these proteins and their natural products in yeast, and ultimately to engineer novel polyketides by “swapping” modules at the DNA level, resulting in novel natural products.

            Unfortunately, the expression of these large, heterologous proteins remains challenging. Preliminary work has shown that codon optimization significantly improves protein expression. We have partnered with Gen9, a DNA synthesis company, to assemble fully codon-optimized bacterial PKSs from the disorazole, nystatin, and amphotericin biosynthetic clusters. We have also developed C-terminally fused selectable markers (URA3, HIS3) for the purposes of selecting for the production of our proteins of interest. By fusing a bacterial PKS gene to a selectable marker and introducing the construct into the yeast deletion library (pilot experiments resulted in 85% coverage), we can identify strains which improve heterologous protein expression. These hits will be verified in individual growth experiments, and eventually combined into a yeast “superhost,” which can then be used in future metabolic engineering projects.