Egg-based influenza split virus vaccine with monoglycosylation induces cross-strain protection against influenza virus infections

Abstract

Each year influenza virus infections cause hundreds of thousands of deaths worldwide and a significant level of morbidity with major economic burden. At the present time, vaccination with inactivated virus vaccine produced from embryonated chicken eggs is the most prevalent method to prevent the infections. However, current influenza vaccines are only effective against closely matched circulating strains and must be updated and administered yearly. Therefore, generating a vaccine that can provide broad protection is greatly needed for influenza vaccine development. We have previously shown that vaccination of the major surface glycoprotein hemagglutinin (HA) of influenza virus with a single N-acetylglucosamine at each of the N-glycosylation sites [monoglycosylated HA (HA_mg)] can elicit better cross-protection compared with the fully glycosylated HA (HA_fg). In the current study, we produced monoglycosylated inactivated split H1N1 virus vaccine from chicken eggs by the N-glycosylation process inhibitor kifunensine and the endoglycosidase Endo H, and intramuscularly immunized mice to examine its efficacy. Compared with vaccination of the traditional influenza vaccine with complex glycosylations from eggs, the monoglycosylated split virus vaccine provided better cross-strain protection against a lethal dose of virus challenge in mice. The enhanced antibody responses induced by the monoglycosylated vaccine immunization include higher neutralization activity, higher hemagglutination inhibition, and more HA stem selectivity, as well as, interestingly, higher antibody-dependent cellular cytotoxicity. This study provides a simple and practical procedure to enhance the cross-strain protection of influenza vaccine by removing the outer part of glycans from the virus surface through modifications of the current egg-based process.

Keywords: N-glycosylation; hemagglutinin stem-specific antibody; influenza vaccines; monoglycosylated influenza split virus vaccine.

Fig. 1.

Preparation of monoglycosylated virus in the embryonated chicken egg. (A) Schematic overview of monoglycosylated split virus vaccine production and viral HA with different glycan states. Viral HA_fg, HA with the typical complex type N-glycans; viral HA_hm, HA with high mannose type N-glycans; viral HA_mg, HA with GlcNAc at its N-glycosylation sites. Models were created with Protein Data Bank (PDB) ID code 3LZG by adding glycan using GlyProt (www.glycosciences.de/modeling/glyprot/php/main.php) and PDB files of lipid bilayer from Lipid Bilayer Membranes for RasMol (https://www.umass.edu/microbio/rasmol/bilayers.htm). The images are displayed using the PyMOL (https://pymol.org/2/) program. (B) Western blot analysis of viral HA in allantoic fluids that were inoculated with different concentrations of kifunensine. (C) Western blot analysis of viral HA in kifunensine-treated allantoic fluids with different concentration of Endo H digestion. In B and C, viral HA was detected by rabbit anti-HA polysera in Western blot analysis.

Fig. 2.

Characterization of the fully glycosylated and monoglycosylated viruses and split vaccine products. (A) SDS/PAGE analysis of the purified fully glycosylated and monoglycosylated viruses. Arrowheads indicate the viral HA protein (black, HA1; red, HA2). NP, nucleoprotein. (B) Ratio of the amount of HA to total viral proteins of the purified viruses. Proteins deglycosylated by peptide-N-glycosidase F (PNGase F) of the purified viruses were quantified with SDS/PAGE analysis. (C) Transmission electron microscopic images of the fully glycosylated and monoglycosylated viruses. (D) Hemagglutination assay of X-181_fg and X-181_mg split vaccine. HAU, hemagglutination unit. (E) Binding of FI6v3, F10, and CR9114 to X-181_fg or X-181_mg split vaccine. *P < 0.05; **P < 0.01; ***P < 0.001. The P value was calculated with Prism software using the Student’s t test and two-way ANOVA. Values are mean ± SEM. conc., concentration.

Fig. 3.

Protection against Cal/09, NC/99, and WSN/33 viruses with monoglycosylated split virus vaccine. X-181 split virus vaccines (X-181_fg and X-181_mg) and commercial TIV were used as vaccines. The sera of vaccinated mice were assayed by HI and microneutralization (MN) against X-181 virus (A and B), NC/99 virus (D and E), and WSN/33 virus (G and H), respectively. For virus challenge, the immunized mice were inoculated with a lethal dose of X-181 (10-fold LD₅₀) (C), NC/99 (2.5-fold LD₅₀) (F) and WSN/33 (10-fold LD₅₀) (I) viruses, and the efficacy was evaluated by survival over 14 d. GMT, geometric mean titer. *P < 0.05; **P < 0.01; ***P < 0.001. The P value of HI and MN was calculated with Prism software using the Student’s t test and two-way ANOVA. The statistical significances of mice survival data were determined using log-rank tests.

Fig. 4.

More cross-strain immune responses elicited by vaccination with monoglycosylated split virus vaccine. (A) ELISA and ELISpot of antibody bindings to a cross-strain Brisbane (Bris/07) HA. (B) ELISA and ELISpot of antibody bindings to stem-HA. (C) Competition ELISA binding to Brisbane HA with the biotinylated F10 antibody. The percentage of competition showed the amount of anti-HA stem antibodies in immunized sera. (D) Alternative competition ELISA binding to Brisbane HA with or without F10 preblocking. The percentage of competition in the alternative competition ELISA showed the ratio of anti-HA stem antibodies (F10-competitive antibodies) to total HA-specific antibodies in immunized sera. *P < 0.05; **P < 0.01; ***P < 0.001. The P value was calculated with Prism software using the Student’s t test. Values are mean ± SEM.

Fig. 5.

ADCC assay of immunized sera against HA-expressed HEK cells. ADCC-mediated antibodies of X-181_fg–, X-181_mg–, and TIV-vaccinated sera were detected by ADCC assay using Bris/07 (A) or Cal/09 (B) HA protein expressing HEK293T cells as target cells. The cytotoxicity was measured and indicated the ADCC effect of vaccinated mice sera. *P < 0.05; **P < 0.01. The P value was calculated with Prism software using the Student’s t test. Values are mean ± SEM.