Current influenza virus vaccines poorly display key neuraminidase (NA) epitopes and do not robustly induce NA-reactive antibodies; instead, they focus on the induction of hemagglutinin (HA)-reactive antibodies. Next-generation influenza vaccines should be optimized in order to activate NA-reactive B cells and to induce a broadly cross-reactive and protective antibody response. We aimed at enhancing the immunogenicity of the NA on vaccines by two strategies: (i) modifying the HA:NA ratio of the vaccine preparation and (ii) exposing epitopes on the lateral surface or beneath the head of the NA by extending the NA stalk. The H1N1 glycoproteins from the influenza virus A/California/04/2009 strain were displayed on human immunodeficiency virus 1 (HIV-1) gag-based virus-like particles (VLP). Using the baculovirus insect cell expression system, we biased the quantity of surface glycoproteins employing two different promoters, the very late baculovirus p10 promoter and the early and late gp64 promoter. This led to a 1:1 to 2:1 HA:NA ratio, which was approximately double or triple the amount of NA as present on the wild-type influenza A virus (HA:NA ratio 3:1 to 5:1). Furthermore, by insertion of 15 amino acids from the A-New York/61/2012 strain (NY12) which prolongates the NA stalk (NA long stalk; NA-LS), we intended to improve the accessibility of the NA. Six different types of VLPs were produced and purified using a platform downstream process based on Capto-Core 700™ followed by Capto-Heparin™ affinity chromatography combined with ultracentrifugation. These VLPs were then tested in a mouse model. Robust titers of antibodies that inhibit the neuraminidase activity were elicited even after vaccination with two low doses (0.3 μg) of the H1N1 VLPs without compromising the anti-HA responses. In conclusion, our results demonstrate the feasibility of the two developed strategies to retain HA immunogenicity and improve NA immunogenicity as a future influenza vaccine candidate.
Read full abstract