Bioprocess development for universal influenza vaccines based on inactivated split chimeric and mosaic hemagglutinin viruses

Abstract

Seasonal influenza viruses account for 1 billion infections worldwide every year, including 3-5 million cases of severe illness and up to 650,000 deaths. The effectiveness of current influenza virus vaccines is variable and relies on the immunodominant hemagglutinin (HA) and to a lesser extent on the neuraminidase (NA), the viral surface glycoproteins. Efficient vaccines that refocus the immune response to conserved epitopes on the HA are needed to tackle infections by influenza virus variants. Sequential vaccination with chimeric HA (cHA) and mosaic HA (mHA) constructs has proven to induce immune responses to the HA stalk domain and conserved epitopes on the HA head. In this study, we developed a bioprocess to manufacture cHA and mHA inactivated split vaccines and a method to quantify HA with a prefusion stalk based on a sandwich enzyme-linked immunosorbent assay. Virus inactivation with beta-propiolactone (βPL) and splitting with Triton X-100 yielded the highest amount of prefusion HA and enzymatically active NA. In addition, the quantity of residual Triton X-100 and ovalbumin (OVA) was reduced to very low levels in the final vaccine preparations. The bioprocess shown here provides the basis to manufacture inactivated split cHA and mHA vaccines for pre-clinical research and future clinical trials in humans, and can also be applied to produce vaccines based on other influenza viruses.