Open Collections

Bayly, Carmen

University of British Columbia

Polyketides are a natural chemical class of complex, frequently bioactive molecules, from which several therapeutics have been derived. Large enzyme complexes, polyketide synthases, are responsible for constructing the carbon backbone of polyketides from simple metabolic precursors. For modular polyketide synthases (mPKSs), the chemical tailoring of each precursor incorporated is controlled by a distinct cluster of domains. These modules are organized as an assembly line. Because the chemistry of the final polyketide chain reflects the organization of modules on the assembly line, mPKSs have become targets for rational protein engineering. If module types could be functionally rearranged in any order, this would expand control over polyketide biosynthetic pathways, enabling access to libraries of novel polyketides from which new antibiotics or therapeutics may be derived. However, combinatorial use of individual mPKS modules has not surpassed 3-module assembly lines. Interactions between the acyl carrier protein (ACP) and ketosynthase (KS) module domains have been particularly sensitive to mismatched interfaces, and these largely constrain use of engineered modules to their specific locations in their natural mPKS assembly lines. ACP-KS interactions remain challenging to model and engineer. In this thesis, I propose an alternative mPKS platform using modules from Mycobacterium tuberculosis PKS12. This bimodular PKS uses repeating, identical KS-ACP interfaces to form 10-module multimers which build long, saturated carbon chains. This makes PKS12 modules attractive templates for creating more interchangeable modules. A prerequisite to developing a PKS12-derived combinatorial platform is the affixing of chain-onloading and chain-offloading domains onto PKS12 modules, enabling it to produce fatty acids. I show that a fatty-acid-producing mPKS can be created from PKS12 parts, and identified an unexpected fusion point for the offloading domain. I identified a plasmid system and culture conditions enabling detectable and reproducible output from PKS12-derived assembly lines in E. coli. I then used this system to demonstrate that antiparallel SYNZIP domains can functionally reattach saturating module halves, providing a new principle for combinatorial mPKS engineering. Finally, I investigate the structural basis for the unexpected C-terminal fusion point, and identify a particular ACP region, helix 2, as a likely contributor.

Text

2021-11-10

Attribution-NonCommercial-NoDerivatives 4.0 International

http://hdl.handle.net/2429/80180

Genome Science and Technology

University of British Columbia

UBCV

http://creativecommons.org/licenses/by-nc-nd/4.0/