PROJECT SUMMARY
Seasonal influenza epidemic causes 3-5 million infections and 250,000 to 500,000 deaths every year. While
seasonal influenza vaccine is available and being constantly updated, its effectiveness is often hampered by
the rapid antigenic drift of circulating strains. As a result, a major goal of influenza research is to develop a
more effective vaccine. Nevertheless, the poor ability to forecast the evolution of influenza virus poses a huge
challenge in influenza vaccine development. Consequently, understanding how the evolutionary trajectories of
influenza virus are being shaped can significantly benefit public health. Influenza virus has two surface
antigens, namely hemagglutinin (HA) and neuraminidase (NA). While influenza vaccine development has
traditionally focused on targeting the HA, NA has received increasing attention as an effective vaccine target in
recent years. Evolution of NA is under several biophysical constraints including protein stability, surface
expression, and enzymatic activity. These biophysical constraints determine not only the fitness effects of
individual mutations, but also how these fitness effects vary in the presence of other mutations (i.e. epistasis).
In fact, epistasis has been a main obstacle in evolution forecast since epistasis can lead to opposite fitness
effects of the same mutation in different influenza strains. This proposed study will use innovative high-
throughput experiments to systematically probe the fitness effects of all possible amino-acid mutations on NA
and map epistatic interactions that are involved in the natural evolution of NA. In addition, the molecular
mechanisms of epistasis will be characterized by biochemical and structural biology approaches. Statistical
modeling will further be applied to quantify the relationships between biophysical constraints of NA and viral
fitness. The results will comprehensively reveal the biophysical principles that govern the mutational fitness
effects and epistatic interactions in influenza NA, and hence its evolutionary trajectories in natural evolution.
This proposed study will therefore promote the construction of a unifying biophysical model to accurately
forecast the evolution of influenza virus, which will in turn facilitate the development of next-generation
influenza vaccines.