Co-opting the cell machinery for directed evolution

Over millions of years, Nature has generated an array of biomolecules such as DNA, RNA, proteins and small molecules. In Chapter 1 we present tools to study biomolecule interactions inside the cell. We cover protein interactions with RNA, DNA and small molecules. Emphasis is put on high-throughput methods to detect these interactions, such as the hybrid-technology and protein complementation assays.;Mimicking natural evolution, directed evolution has generated designer enzymes. However, directed evolution---mutagenesis of a target gene followed by selection for a protein variant with the desired properties---is limited to enzymatic reactions that are screenable or selectable. To address this bottleneck the Cornish laboratory developed chemical complementation, a general high-throughput assay for enzyme catalysis. Previously, chemical complementation has been adapted to select for glycosynthase activity. In Chapter 2 , we describe the synthesis optimization of the small molecule substrates used to detect glycosynthase activity, as well as the characterization of the products obtained with a newly evolved glycosynthase.;In Chapter 3 we adapt chemical complementation to detect bond cleavage reactions as a growth selection using a cellulase enzyme. Previously, chemical complementation was able to detect bond cleavage reactions only as a screen. This growth selection should allow the search of libraries four orders of magnitude larger than with our previous screen. In Chapter 4, the chemical complementation selection for bond cleavage reactions is used to carry out the directed evolution cellulases. Cellulases are an economically important target in the conversion of biomass to fermentable sugars. The cellulase selection was able to isolate cellulases with improved activity after only five days of selection. Due to the large number of enzyme variants selections can test compared to existing medium-throughput screens for cellulases, this assay has the potential to impact the discovery of improved cellulases and other glycosyl hydrolases for biomass conversion. Appendix 1 and 2 describe alternative libraries and selection strategies explored to carry out the directed evolution of cellulases.;Chapter 5 presents a novel in vivo targeted mutagenesis method exploiting the high efficiency homologous recombination in S. cerevisiae. This method should allow for the rapid and easy generation of targeted protein variant libraries when compared to current methods targeted mutagenesis methods. In Chapter 6, we take the in vivo targeted mutagenesis a step further and generate libraries of linear DNA cassettes inside the cell. By taking advantage of homologous recombination, the HO endonuclease, and sexual reproduction, we use this methodology to carry out the directed evolution of a gene in the tryptophan biosynthesis pathway. As sexual reproduction allows for the swapping of DNA cassette libraries without re-transformation, we anticipate searching larger libraries of protein variants than currently possible.