Influenza Virus-like Particle (VLP) Vaccines Expressing the SARS-CoV-2 S Glycoprotein, S1, or S2 Domains

Abstract

The ongoing severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) pandemic had brought disastrous consequences throughout the entire world. While several manufactured vaccines have been approved for emergency use, continuous efforts to generate novel vaccines are needed. In this study, we developed SARS-CoV-2 virus-like particles (VLPs) containing the full length of spike (S) glycoprotein (S full), S1, or S2 together with the influenza matrix protein 1 (M1) as a core protein. Successfully constructed VLPs expressing the S full, S1, and S2 via Sf9 cell transfections were confirmed and characterized by Western blot and transmission electron microscopy (TEM). VLP immunization in mice induced higher levels of spike protein-specific IgG and its subclasses compared to naïve control, with IgG2a being the most predominant subclass. S full and S1 immune sera elicited virus-neutralizing activities, but these were not strong enough to fully inhibit receptor-ligand binding of the SARS-CoV-2. Neutralizing activities were not observed from the S2 VLP immune sera. Overall, our findings revealed that S full or S1 containing VLPs can be developed into effective vaccines.

Keywords: COVID-19; SARS-CoV-2; antibody; neutralization; vaccine; virus-like particle.

Figure 1

Transmembrane domain prediction and schematic diagram of the gene constructs. Original sequences of the S glycoprotein gene were analyzed by the TMHMM database and transmembrane protein topologies were predicted (A). Codon-optimized genes were constructed following the schematic illustrated above (B).

Figure 1

Transmembrane domain prediction and schematic diagram of the gene constructs. Original sequences of the S glycoprotein gene were analyzed by the TMHMM database and transmembrane protein topologies were predicted (A). Codon-optimized genes were constructed following the schematic illustrated above (B).

Figure 2

Successful cloning of S protein inserts. Codon-optimized genes were successful cloned into pFastBac vector (A). Vectors containing each insert were transformed into DH10Bac competent cells and bacmids were analyzed via colony PCR (B). Lane identifications for both panels are as follows: M, marker; 1, S full; 2, S1; 3, S2; 4, pFastBac vector.

Figure 3

Confirming recombinant baculovirus construction. Polyclonal antibodies were raised against the S RBD and S2 antigens. The two antibodies were used to detect successful transfection of bacmid DNAs into Sf9 cells via immunocytochemistry. The S RBD antibodies were used to detect protein expression for the S full and S1 rBVs, whereas S2 antibodies were specifically used for the S2 rBV.

Figure 4

Confirming baculovirus removal after sucrose gradient purification. S full VLPs were used as a representative to confirm baculovirus removal from the VLPs described in this study. Sf9 cells were inoculated with fractions of S full VLPs acquired post-sucrose purification, rBV control, and VLPs before purification (pre-sucrose). Non-inoculated Sf9 cells were used as a negative control. Purified VLPs were labeled as VLP1 (band 1) and VLP2 (band 2) (A). Sf9 cells were monitored daily for 4 days to assess baculovirus cell infectivity under the microscope (B). All images were acquired at 100× magnification.

Figure 5

VLPs were characterized by Western blot and TEM. Expressions of the S full, S1, S2, and M1 proteins in the VLPs were evaluated by Western blot. Lanes 1, 2, and 3 correspond to protein loading concentrations of 40 μg, 20 μg, and 10 μg, each respectively (A). VLP images were acquired under TEM (B).

Figure 6

Antigen-specific antibody responses induced by VLP immunization. Sera of mice were collected 1 week after each immunization and were used to assess the antigen-specific antibody responses via ELISA. VLP immunization and boosting effect of VLP immunization was confirmed by reacting the sera against S full VLP (A). All of the sera collected over the course of the animal studies were used to observe changes in IgG (B), IgG1 (C), IgG2a (D), IgG2b (E) responses against S1 and S2 antigens. Data are expressed as mean ± SEM (* p < 0.05, ** p < 0.01, *** p < 0.001).

Figure 7

Surrogate virus neutralization requires the presence of S1 domain. Confirmation of HRP conjugation to S RBD (A) and successful RBD-HRP interaction with the hACE2 (B) was assessed through ELISA. Surrogate virus neutralization assay was conducted using serially diluted sera of immunized mice after heat-inactivation (C). Data are expressed as mean ± SEM (* p < 0.05 vs. naïve control).