Deep mutational scanning gives new insights into flu virus evolution

Hopefully, you have already received your seasonal flu vaccine this year. These shots are highly effective at preventing illness, ultimately lowering the spread of the virus throughout the population. Panels of experts meet biannually to choose which flu strains to vaccinate the general population against for the coming flu season. These decisions are well-informed – experts factor which strains are getting people sick throughout the year, how fast those strains are spreading, and how well the previous year’s vaccine worked to make the final recommendation for the vaccine. Still, some strains of flu may break through the vaccine to infect people. If you have ever been vaccinated and still caught the flu, it was most likely due to one of these breakthrough strains. Understanding how the flu virus evolves to escape vaccination could help guide researchers’ vaccine recommendations to prevent fewer flu breakthrough cases in the future.

Dr. Jesse Bloom and his laboratory are working to characterize flu virus evolution to tackle these problems. A new study in his lab, led by recent graduate Dr. Tim Yu, used deep mutational scanning to analyze how mutations in the hemagglutinin protein impacted flu virus fitness. Hemagglutinin is expressed on the surface of viral particles where it binds to receptors on human cells and facilitates viral entry. It is also a target of neutralizing antibodies produced by human cells in response to a flu infection or vaccination. “What we see during evolution is that these viruses are constantly getting mutations that escape our antibodies, which is why we have vaccines to keep up with that evolution and also why we continually get infected over time,” explains Yu.

While hemagglutinin mutations may help the virus evade the immune system, they are not necessarily good for the virus’ lifecycle. Mutations could make viruses less likely to interact with host cells, unable to propagate more viruses once in the host cell, or otherwise impaired. “During evolution, you would expect that most mutations to a protein are deleterious, so how does the virus keep getting antibody escape mutations without it affecting viral fitness?” Yu continues.

To begin answering this question, the team used a technique called deep mutational scanning to generate a library of viruses with every possible hemagglutinin mutation. The hemagglutinin ectodomain is 504 amino acids long, and for every position, there are 19 other amino acids that could possibly be substituted in for a grand total of 9,576 mutations. With these tools in hand, the group quantified how well the mutated viruses were able to get into cells and how stable the mutant hemagglutinin proteins were.

To infect a cell, hemagglutinin first binds a host receptor. Next, the viruses are brought into cells within small membrane-bound compartments called endosomes. Inside the acidic endosome, hemagglutinin triggers fusion between the viral and endosomal membranes, helping the viruses escape from endosomes and infect the rest of the cell. The team analyzed the impact of mutations in three key regions of hemagglutinin on viral cell entry. The receptor binding pocket interacts with host cell receptors to facilitate viral entry. They found that mutations in the receptor binding pocket of hemagglutinin generally impaired virus entry into host cells. Mutations in the regions of hemagglutinin targeted by antibodies generally did not impact viral entry unless these regions overlapped with the receptor binding pocket. The fusion loop allows the flu virus membrane to fuse with the endosome. Mutations in the fusion loop were strongly deleterious for viral entry.

Because viral fusion in the endosome is triggered when it reaches a certain level of acidity, mutations that change the acid stability of hemagglutinin could impair viral fitness. Yu and his colleagues treated their mutant viruses with acidic buffers and tested their ability to infect cells. Using this approach, they identified several destabilizing mutations in hemagglutinin, many of which were in parts of the protein that conformationally change during fusion, highlighting the importance of these regions for viral fitness.

When a new mutation arises in hemagglutinin, that amino acid mutation should be tolerated as well as the ancestral amino acid that it replaces. However, as time passes and the viruses acquire other random mutations, the ancestral hemagglutinin amino acid can become intolerable for the newer virus. This form of evolution is called entrenchment. Yu and his team analyzed hemagglutinin mutations in current and ancestral flu viruses to identify entrenched mutations. They found several mutations in the receptor binding pocket that impaired cell entry in their deep mutational scan but had been part of past flu viruses, indicating that these mutations became entrenched in specific genetic flu variants. The group did not find entrenchment for hemagglutinin mutations in other regions that impacted cell entry or hemagglutinin stability.

Finally, the group analyzed how each hemagglutinin mutation impacts the ability of the viruses to escape antibody neutralization. They found one mutation that strongly promoted antibody escape in their screen and became part of all flu strains by 2022, highlighting how natural selection favored its ability to escape antibodies. However, Yu was puzzled why the mutation did not arise earlier. Further work revealed that the exact same mutation strongly impaired cell entry in older flu strains. Basically, the virus needed to evolve in other ways so the antibody escape mutation stopped impacting cell entry. The authors also found some mutations that escape antibodies but at the cost of destabilizing hemagglutinin in their screen. Those mutations have never arisen in flu strains, suggesting the virus has not figured out a way to evolve around the stability constraint. Together, their results highlight the power of deep mutational scanning to infer the effects of specific mutations on viral fitness. “One of the general consensuses is that the virus is always going to be able to mutate and figure out a solution [to keep infecting hosts],” says Yu. Accounting for how constraints on mutations change across genetic backgrounds is an exciting frontier for predicting viral evolution and, hopefully, creating better vaccines to counter this evolution.

Fred Hutch/University of Washington/Seattle Children’s Cancer Consortium Members Drs. Jesse Bloom and Janet Englund contributed to this research.

This work was supported by funding from the National Institutes of Health, the National Science Foundation, and Howard Hughes Medical Institute.

Yu TC, Kikawa C, Dadonaite B, Loes AN, Englund JA, Bloom JD. 2025. Pleiotropic mutational effects on function and stability constrain the antigenic evolution of influenza haemagglutinin. Nat Ecol Evol. doi: 10.1038/s41559-025-02895-1.