Exploring the molecular evolution of proteins with deep mutational scanning

"In this thesis, deep mutational scanning is expanded and applied to better understand the molecular evolution of proteins. Protein evolution is a complex process where subtle changes in molecular architecture can have massive impacts on biophysical properties, altering how well-adapted a protein is to a specific task or environment. Deep mutational scanning provides a finer level of understanding of molecular evolution by assessing the effect of every possible single-mutation on a protein's function. The technique combines site saturation mutant libraries, high throughput selections, and deep sequencing to tabulate the changes in mutant frequencies. From these changes the impacts of the mutations on protein function are characterized. This technology allows for efficient exploration of the local evolutionary landscape of a protein, making it a powerful tool for understanding evolution. Here, I use deep mutational scanning to study how the initial likelihood of obtaining the native folded state of an enzyme in vivo constrains its evolution. We designed two unique single-point mutants of AmiE, an aliphatic amidase from Pseudomonas aeruginosa. These mutant enzymes are significantly less likely to reach the native folded state in vivo than the unmutated precursor and have catalytic efficiencies that are statistically indistinguishable from the initial unmutated enzyme. I tested the impacts of nearly all single-point mutations for the two impaired enzymes using high-throughput growth selections and compared them to the precursor enzyme. These comparisons provided insights into how evolutionary outcomes are changed following decreases in the likelihood of native folding, and on how the impacts of single mutations combine to influence function. The other primary goal of this thesis is the development of a new method that expands the utility of deep mutational scanning studies. This method assembles comprehensive single-site saturation, and large multi-point, mutant genome libraries of the bacteriophage phi-X174. To assemble the mutant genome libraries we combine nicking scanning mutagenesis and Golden Gate cloning. With these viral genome libraries, deep mutational scanning experiments can be performed in situ. These libraries are a valuable tool for studying the molecular determinants of viral host switching, the combination of inter- and intra-subunit mutations, and other aspects of the molecular evolution of viruses."--Pages ii-iii.