Analysis of Protein Adaptation from High Throughput Mutagenesis Studies

Abstract

Proteins are sophisticated molecular machines, yet they arise through a simple process of mutation and selection. Understanding how proteins adapt to their ever-changing environments is one of the central challenges in molecular biology. In this dissertation, I first discuss the state of the molecular evolution field, which has recently been pushed forward by advances in deep sequencing, and I highlight some broad consensuses that have risen out of recent deep mutational scans. In chapters 2-4, I investigate three systems that highlight different aspects of protein adaptation. First, I characterize the mutational neighborhood of a bacterial protein that plays a dual role as a nutrient transporter and the receptor for a phage. By performing a deep mutational scan with respect to both properties, I show that specific resistance mutations are common relative to destabilizing mutations, and that these specific resistance mutations are spatially clustered around a particular structural feature, Loop L6. Second, I characterize the tail fiber tip of phage, which mediates attachment to its host and is the key player in overcoming host resistance. I challenged a library of phage variants to infect four hosts, expressing the wild type receptor or one of three resistance mutations. By comparing infectivity of variants across these hosts, I characterize two properties that underlie adaptation: specificity and promiscuity. Third, I aggregated data from 20 published deep mutational scans, showing that many proteins have a strong signature of genetic robustness. Across these datasets, most genes have more favorable mutational neighborhoods than would be expected by chance. In most cases, this effect cannot be explained by codon bias alone – the positions in which codons are used is much more important than how frequently they are used. In the fifth and final chapter, I draw some connections between the different systems I have investigated, and I provide a perspective on the standing questions in the field. In particular, the role of stability in biasing evolutionary outcomes rises to prominence. As technology advances, further dissection of the links between genetic alterations and protein properties is likely to enhance our understanding of how proteins adapt.