Date of Award

Spring 1-1-2018

Document Type


Degree Name

Doctor of Philosophy (PhD)

First Advisor

Ryan T. Gill

Second Advisor

Maciej Walczak

Third Advisor

Geoff Cameron

Fourth Advisor

Joel Kaar

Fifth Advisor

Jerome Fox


The diversity in structure and biological activity of natural products has long captured the imagination of chemists and molecular biologists alike. Natural products are synthesized by elaborate secondary metabolic pathways or enzymes. For example, modular megasynthases such as Type I Polyketide Synthases (PKSs), Non-ribosomal Peptide Synthetases (NRPSs), and PKS-NRPS hybrids, produce an array of quintessential natural products. The apparent modularity, defined architecture, and predictable chemistry of modular megasynthases make them especially attractive for combinatorial biosynthesis, which has long been a focus of biotechnology. Combinatorial biosynthetic libraries are of particular interest due to the high value of many natural products and for the potential directed evolution of small molecules.

Prior efforts to rationally design modular megasynthases have demonstrated some success but combinatorial forward engineering remains unfulfilled. The rise of synthetic biology has presented new capabilities to resolve the underlying deficiencies of past approaches. Synthetic biology abstracts complex biological systems into modular parts with simplified rules governing functions and interactions, enabling forward engineering of novel biological systems.

Herein, we present a precise and scalable synthetic biology approach to combinatorial biosynthesis built on improved abstraction of modular megasynthases into ‘parts’. The approach begins with bottom-up design of a chimeric modular megasynthase chassis using conserved catalytic domain linkers as modular parts for connecting catalytic domains. This minimal modular megasynthase ‘chassis’ is then assembled to produce a target molecule and finally diversified at scale to produce a library of new molecules using paralogous catalytic domains.

To demonstrate this approach, we first targeted the production of a triketide lactone food and fragrance ingredient, d-hexalactone. A retrosynthetic analysis identified catalytic domains, linkers, and substrates required for d-hexalactone biosynthesis via a PKS. The chimeric megasynthase chassis was then designed using known catalytic domains and a custom computational pipeline that identified context-independent linker sequences, built, and tested. To demonstrate the extensibility of this approach we designed, built, and tested Non-Ribosomal Peptide Synthetase (NRPS) and NRPS-PKS chassis. Finally, we applied a massively multiplex genome engineering tool for further rational diversification, providing a scalable, high efficiency approach to combinatorial biosynthesis.