Bacteriophage Φ21s receptor-binding protein evolves new functions through destabilizing mutations that generate non-genetic phenotypic heterogeneity

How viruses evolve to expand their host range is a major question with implications for predicting the next pandemic. Gain-of-function experiments have revealed that host-range expansions can occur through relatively few mutations in viral receptor-binding proteins, and the search for molecular mechanisms that explain such expansions are underway. Previous research on expansions of receptor use in bacteriophage λ has shown that mutations that destabilize λs receptor-binding protein cause the receptor-binding protein to fold into new conformations that can utilize novel receptors but have weakened thermostability. These observations led us to hypothesize that other viruses may take similar paths to expand their host range. Here, we find support for our hypothesis by studying another virus, bacteriophage 21 (Φ21), which evolves to use two new host receptors within two weeks of laboratory evolution. By measuring the thermodynamic stability of Φ21 and its descendants, we show that, as Φ21 evolves to use new receptors and expands its host range, it becomes less stable and produces viral particles that are genetically identical but vary in their thermostabilities. Next, we show that this non-genetic heterogeneity between particles is directly associated with receptor use innovation, as phage particles with more derived receptor use capabilities are more unstable and decay faster. Lastly, by manipulating the expression of protein chaperones during Φ21 infection, we demonstrate that heterogeneity in thermostability and receptor use breadth of phage particles is directly related to the folding of phage proteins into different conformers. Altogether, our results provide support for the hypothesis that viruses can evolve new receptor-use tropisms through mutations that destabilize the receptor-binding protein and produce multiple protein conformers.