Investigations into the Stucture and Function of Type I Polyketide Synthases

The polyketide (PK) class of natural products constitutes an abundant array of secondary metabolites produced in microorganisms, many of which possess potential medicinal value, especially in the area of oncology. Polyketides are assembled biosynthetically via the megaenzymes polyketide synthases (PKSs) through an assembly line process of stepwise condensations of simple malonic acid building blocks derived from primary metabolism. Despite the usefulness of natural products in medicine, the development of polyketide natural products into new drugs is often hindered by their suboptimal pharmacological properties, highlighting the need for their modification by medicinal chemistry. However, low natural abundance and high structural complexity often necessitates lengthy and expensive synthetic routes to natural product analogs, thus impeding their clinical development. A promising method for expanding the chemical diversity within polyketide natural products is PKS bioengineering, whereby natural product analogs are generated by engineering new functionality into the enzymes responsible for their production instead of through synthetic derivatization. While notable successes in PKS engineering have been achieved, many attempts result in decreased product yields or fail to produce the predicted molecules entirely. The studies in this thesis focus on investigating the structural and mechanistic parameters that govern PKS catalysis in order to increase the potential of harnessing these enzymes as biocatalysts for the production of new polyketide analogs. First, a series of engineered PKS modules was generated by combining modules from the pikromycin, erythromycin, and juvenimycin biosynthetic pathways with non-native TE domains and analyzed for substrate flexibility in vitro. The results from this study implicated the TE domain as the dominant catalytic bottleneck in the full-module processing of unnatural substrates. We next focused our investigations on probing the TE directly as an excised domain, subsequently confirming the previously observed catalytic bottleneck. Mutational analysis of the Pik TE domain resulted in an engineered variant (S148C) with improved substrate flexibility and catalytic efficiency, which eliminated the aforementioned bottleneck and allowed for the production of diastereomeric macrolactone analogs. Finally, we performed molecular dynamics (MD) simulations coupled with quantum mechanical (QM) calculations of the native and engineered TE domains to provide a mechanistic rational for our experimental observations. Taken together, the results herein provide further insight into the catalytic and mechanistic parameters that govern the productive functioning of engineered PKSs. Our identification of the thioesterase domain as a key catalytic bottleneck in the processing of unnatural substrates builds the groundwork for future engineering of PKS TE domains in order to generate more flexible catalysts for the production of novel natural product analogs.PhDCancer BiologyUniversity of Michigan, Horace H. Rackham School of Graduate Studieshttps://deepblue.lib.umich.edu/bitstream/2027.42/140831/1/aakoch_1.pd