Broadly Protective CD8+ T Cell Immunity to Highly Conserved Epitopes Elicited by Heat Shock Protein gp96-Adjuvanted Influenza Monovalent Split Vaccine

Abstract

Currently, immunization with inactivated influenza virus vaccines is the most prevalent method to prevent infections. However, licensed influenza vaccines provide only strain-specific protection and need to be updated and administered yearly; thus, new vaccines that provide broad protection against multiple influenza virus subtypes are required. In this study, we demonstrated that intradermal immunization with gp96-adjuvanted seasonal influenza monovalent H1N1 split vaccine could induce cross-protection against both group 1 and group 2 influenza A viruses in BALB/c mouse models. Vaccination in the presence of gp96 induced an apparently stronger antigen-specific T cell response than split vaccine alone. Immunization with the gp96-adjuvanted vaccine also elicited an apparent cross-reactive CD8⁺ T cell response that targeted the conserved epitopes across different influenza virus strains. These cross-reactive CD8⁺ T cells might be recalled from a pool of memory cells established after vaccination and recruited from extrapulmonary sites to facilitate viral clearance. Of note, six highly conserved CD8⁺ T epitopes from the viral structural proteins hemagglutinin (HA), M1, nucleoprotein (NP), and PB1 were identified to play a synergistic role in gp96-mediated cross-protection. Comparative analysis showed that most of conservative epitope-specific cytotoxic T lymphocytes (CTLs) apparently induced by heterologous virus infection were also activated by gp96-adjuvanted vaccine, thus resulting in broader protective CD8⁺ T cell responses. Our results demonstrated the advantage of adding gp96 to an existing seasonal influenza vaccine to improve its ability to provide better cross-protection.IMPORTANCE Owing to continuous mutations in hemagglutinin (HA) or neuraminidase (NA) or recombination of the gene segments between different strains, influenza viruses can escape the immune responses developed by vaccination. Thus, new strategies aimed to efficiently activate immune response that targets to conserved regions among different influenza viruses are urgently needed in designing broad-spectrum influenza vaccine. Heat shock protein gp96 is currently the only natural T cell adjuvant with special ability to cross-present coupled antigen to major histocompatibility complex class I (MHC-I) molecule and activate the downstream antigen-specific CTL response. In this study, we demonstrated the advantages of adding gp96 to monovalent split influenza virus vaccine to improve its ability to provide cross-protection in the BALB/c mouse model and proved that a gp96-activated cross-reactive CTL response is indispensable in our vaccine strategy. Due to its unique adjuvant properties, gp96 might be a promising adjuvant for designing new broad-spectrum influenza vaccines.

Keywords: CD8+ T cells; conserved epitopes; cross-protection; gp96; influenza vaccine.

FIG 1

gp96-adjuvanted H1N1 split vaccine provides cross-protection against heterologous or heterosubtypic influenza virus infection. Shown is a schematic diagram of the vaccination and challenge protocol. BALB/c mice were intradermally immunized with two doses of H1N1 split-virus vaccine alone or formulated with gp96. Mice receiving the same dose of split vaccine or gp96 alone were used as adjuvant control group (A). Two weeks after the last immunization, mice were challenged with lethal dose of mouse-adapted A/PR/8/34 H1N1 (B), mouse-adapted A/Aichi/2/68 H3N2 (C), or A/Anhui/1/2013 H7N9 (D) influenza virus. Body weight changes and survival rates were monitored for 14 days following challenge (n = 10). Data are means ± SD for 10 mice. *, P < 0.05; ***, P < 0.001 (compared to vaccine control group). The statistical significances of survival data were determined using log rank tests. Experiments were repeated twice, with similar results.

FIG 2

Humoral immune responses induced by gp96 adjuvant. BALB/c mice were immunized twice with H1N1 split-virus vaccine with or without gp96. Two weeks after the last immunization, serum anti-H1N1 IgG titers were detected by ELISA (A), or titers of serum hemagglutination inhibition antibody against homologous virus, PR8, or H3N2 virus were determined by HAI assay or cross-reactive antibodies that bound to β-propiolactone (BPL)-inactivated PR8 virus were detected by ELISA (B and C). The cross-reactive antibodies titers were from a 1:2,000 dilution of serum. Sera from vaccinated mice or PR8 infected mice were administered to recipient mice. Each mouse was administered with 250 μl of serum intraperitoneally on 4 separate days as indicated. Recipients were then challenged with a lethal dose (3× the LD₅₀) of PR8 influenza virus (D). Body weight changes and survival rates were monitored for 14 days following challenge (E and F). Data are means ± SD for five mice. *, P < 0.05; **, P < 0.01 (compared to vaccine control group). The P value was calculated using the Student t test. Data are representative of those from three independent experiments.

FIG 3

Cellular immune responses induced by gp96 adjuvant. BALB/c mice were immunized twice with gp96-adjuvanted H1N1 split vaccine. Splenocytes from immunized mice were stimulated with H1N1 split-virus vaccine, inactivated PR8, or H3N2. Antigen-specific T cells were detected by IFN-γ ELISpot assays (A). Splenocytes from immunized mice were cocultured with PR8-infected syngeneic splenocytes for 20 h. Cross-reactive IFN-γ-producing CD8⁺ T cells were detected by flow cytometry. Data are means ± SD for five mice (B). Immunized mice were challenged on day 29 after the last immunization with 3× the LD₅₀ of PR8, and NP_147–155-specific or virus-specific IFN-γ⁺ CD8⁺ T cells in the lungs were analyzed by flow cytometry 5 days later. *, P < 0.05; **, P < 0.01; ***, P < 0.001 (compared to vaccine control groups) (C). Zebra picture depicting the percentage of responsive IFN-γ⁺ CD8⁺ T cells in the lung following stimulation with PR8-infected APCs (D). Mice were immunized and challenged according to the aforementioned procedure and depleted of CD4⁺ or CD8⁺ T cells or both with anti-CD4 or anti-CD8 antibodies at the indicated time (E). Viral titers in lung homogenates were determined by plaque formation assay at day 5 postinfection (n = 5) (F). Survival rates were also monitored for 14 days following challenge (n = 6) (G). *, P < 0.05; **, P < 0.01 (compared to isotype control group). The P value of viral titers was calculated using the Student t test. The statistical significances of mouse survival data were determined using log rank tests. Data are representative of those from three independent experiments.

FIG 4

CD8⁺ T cell response generated by vaccination or viral infection. Peptide-specific T cell responses were detected by ex vivo IFN-γ ELISpot assays. Mice were immunized with each single peptide formulated with gp96 at day 0, day 7, or day 21. Seven days after the last immunization, peptide-specific CTL responses were evaluated using ex vivo IFN-γ ELISpot assay. HBc_87–95 and HBc_82–90 were used as positive and negative controls, respectively (n = 3). Responses were considered positive if results were at least three times the mean of the negative-control wells and>25 SFCs/10⁶ splenocytes (A). BALB/c mice were immunized twice with gp96-adjuvanted H1N1 split vaccine or vaccine alone. Peptide-specific T cells were detected by ex vivo IFN-γ ELISpot assays. HBc_87–95 was used as a negative control. *, P < 0.05; **, P < 0.01 (compared to vaccine control group) (B). Peptide-specific T cells of natural infection were detected by ex vivo IFN-γ ELISpot assays. HBc_87–95 was used as a negative control (C). Each individual peptide was summed, and the fractional response of gp96-adjuvanted vaccine (D) or natural infection (E) elicited by each peptide was calculated. Results are represented as the average percentage of the response with the SEM indicated. *, peptide with a fractional response higher than >5%. Data are representative of those from three independent experiments.

FIG 5

Identification of potential protective K^d-restricted epitopes elicited by gp96 adjuvant. Mice were immunized with a mixture of the 6 protective peptides (HA_437–445, M_31–39, NP_39–47, NP_147–155, NP_218–226, and PB1_81–89) and challenged with 25 μl of the viral suspension containing 3× the LD₅₀ of PR8. Body weight changes and survival rates were monitored for 14 days (A). Splenocytes from the mice immunized with the mixture of six protective peptides or from control peptide-immunized mice were collected. CD8⁺ T cells were then purified and injected intravenously into recipient mice (6 × 10⁶ cells/per mouse) in a volume of 250 μl. After 8 h, recipient mice were challenged intravenously with 3× the LD₅₀ of PR8 virus. Body weight changes and survival rates for 14 days postinfection were monitored. Data are shown as means ± SD for five mice. The statistical significances of mouse survival data were determined using log rank tests. Data are representative of those from two independent experiments (B). Sequence similarity analysis of six identified protective peptides across different influenza viruses. The plot shows the sequence conservation level in bits (sequence logo) and percentage (histogram). Protein location and starting amino acid positions were indicated on top of the logo (C).