Evolutionary Constraints and Potential of the Influenza A Virus RNA-dependent RNA Polymerase

Abstract

Due to rapid evolution and adaptation, influenza viruses remain a major health concern despite co-existing with human beings for centuries. The influenza virus polymerase is a major driver of influenza virus evolution. Mutations within the viral polymerase can change replication efficiency, affecting the replicative fitness of the virus. The polymerase also controls the rate at which influenza virus acquires mutations, opening up possibilities for new phenotypes such as host range expansion, drug resistance, and antigenic drift. Despite its importance to viral evolution, our understanding of the mutational effects on the influenza virus polymerase is relatively limited. The influenza virus’s segmented genome and the multi-unit structure of its polymerase add further complexity to the polymerase’s evolutionary constraints and potential.

My dissertation focuses on characterizing the mutational effects of the core subunit of the influenza virus polymerase complex, the RNA-dependent RNA polymerase (RdRp) subunit, and reveals key constraints on the RdRp that shape influenza virus evolution in nature. The second chapter of my thesis evaluated the fitness effects of mutations and the mutational tolerance of influenza virus RdRp. I performed deep mutational scanning of the influenza A virus PB1 protein and measured the replicative fitness of nearly all variants with single amino acid substitutions. Deep mutational scanning measured replicative fitness with high accuracy and precision and revealed purifying selection against mutations with more dramatic changes. While most missense and nonsense mutations were highly detrimental, some near-neutral and beneficial mutations did exist. I calculated mutational tolerance as the Shannon entropy of the enrichment of all amino acid variants at a site. The mutational tolerance of residues on the influenza virus RdRp was highly constrained by specific functions and site interactions and was not well characterized by the global protein structure. Many beneficial mutations revealed by deep mutational scanning were seen in the natural evolution history of PB1 or shown important to adaptation experimentally. Accessibility by single nucleotide mutations was a crucial factor in determining whether a beneficial mutation would arise in nature. My third chapter established a foundation to study the key mutations that would influence the virus’s replicative fidelity using the variant library created by deep mutational scanning. I examined the growth of the influenza A virus under different concentrations of five mutagenic drugs in different cell lines and determined the proper drug concentrations to induce a moderate selective pressure. 5-Azacytidine and molnupiravir exhibited similar inhibition curves when the infections happened in MDCK or A549 cells, while the inhibition curves of ribavirin, favipiravir, and 5-fluorouracil were vastly different in different cells. These results highlight the complexity of the mechanisms by which mutagenic drugs inhibit influenza virus replication and the varying cell responses to mutagens. Overall, my dissertation provided a comprehensive map of mutational effects on a viral RdRp and revealed the evolutionary constraints and potential of influenza virus polymerase, which would be a valuable resource for future studies on influenza and RNA virus evolution.