A two-adjuvant multiantigen candidate vaccine induces superior protective immune responses against SARS-CoV-2 challenge

Abstract

An ideal vaccine against SARS-CoV-2 is expected to elicit broad immunity to prevent viral infection and disease, with efficient viral clearance in the upper respiratory tract (URT). Here, the N protein and prefusion-full S protein (SFL_mut) are combined with flagellin (KF) and cyclic GMP-AMP (cGAMP) to generate a candidate vaccine, and this vaccine elicits stronger systemic and mucosal humoral immunity than vaccines containing other forms of the S protein. Furthermore, the candidate vaccine administered via intranasal route can enhance local immune responses in the respiratory tract. Importantly, human ACE2 transgenic mice given the candidate vaccine are protected against lethal SARS-CoV-2 challenge, with superior protection in the URT compared with that in mice immunized with an inactivated vaccine. In summary, the developed vaccine can elicit a multifaceted immune response and induce robust viral clearance in the URT, which makes it a potential vaccine for preventing disease and infection of SARS-CoV-2.

Keywords: SARS-CoV-2; SFL(mut); cGAMP; flagellin; mucosal immunity; nucleocapsid; protective efficacy.

Graphical abstract

Figure 1

KF and cGAMP synergistically promoted the maturation of BMDCs and enhanced mucosal immune responses (A and B) BMDCs were stimulated with KF (4 μg/mL), cGAMP (4 μg/mL), or KF (4 μg/mL) plus cGAMP (4 μg/mL) in vitro for 24 h. The cells were treated with lipopolysaccharide (LPS, 1 μg/mL) or sterile PBS as positive and negative controls, respectively. BMDCs were harvested and analyzed via flow cytometry to determine the surface expression of MHC II, CD86, CD80, or CD40. The data are expressed as the mean ± the SEM of four independent experiments. (C) Mice were intranasally administered OVA alone or adjuvanted with KF, cGAMP, or KF plus cGAMP three times at 2-week intervals and euthanized on day 14 after the last immunization. PBS-immunized mice served as negative controls. Serum was collected for the determination of IgG, IgG1, and IgG2a responses (n = 6 mice per group). (D) Nasal washes (NW) were collected for the detection of OVA-specific IgA antibodies (n = 6 mice per group). (E) Lung and spleen tissues were harvested for the detection of OVA-specific IgA-secreting cells by ELISpot assay (n = 6 mice per group). (F) Lung tissues were assayed by ELISpot to assess specific IFN-γ and IL-4 production. Data are expressed as the mean ± the SEM (n = 6 mice per group). The dotted line indicates the limit of detection (LOD), and values that fell below the detection limit are represented by the limit of detection value for statistical analysis. “ND” indicates that no individuals in this group had detectable levels. Significance was determined via one-way ANOVA with a Tukey multiple comparison test. ^∗p < 0.05, ^∗∗p < 0.01, ^∗∗∗p < 0.001, ns, no significance. See also Figure S1.

Figure 2

Induction of antibody responses by experimental vaccines (A) Diagram of C57BL/6 mouse immunization. Six- to 8-week-old C57BL/6 mice were immunized with three doses of experimental vaccines (the N protein and one of the following S protein-associated antigens: (1) the RBD (RBD), (2) the S1 subunit (S1), (3) the full-length S protein (SFL), or (4) the prefusion S protein structure generated by mutating eight amino acid sites (SFL_mut)) via the intranasal route at 14-day intervals. Two weeks after the third immunization, the mice were sacrificed, and antibody responses in the serum, nasal wash (NW), and bronchoalveolar lavage fluid (BALF) were evaluated. (B and C) Humoral immune responses in the serum were evaluated using S protein-, RBD-, N protein-, or inactivated SARS-CoV-2- based IgG ELISA on day 14 after the third immunization (n = 6 mice per group). (D) Neutralizing antibodies (NAbs) in the serum were evaluated by a SARS-CoV-2 neutralization assay in a BSL-3 laboratory (n = 6 mice per group). (E) Mucosal immune responses in the NW were evaluated using S protein- or RBD protein-based IgA ELISA on day 14 after the third immunization (n = 6 mice per group). (F) Mucosal immune responses in the BALF were evaluated using S protein- or RBD-based IgA ELISA on day 14 after the third immunization (n = 6 mice per group). Data are expressed as the mean ± the SEM. The dotted line indicates the limit of detection (LOD), and values that fell below the detection limit are represented by the limit of detection value for statistical analysis. Significance was determined via one-way ANOVA with a Tukey multiple comparison test. ^∗p < 0.05, ^∗∗p < 0.01, ^∗∗∗p < 0.001.

Figure 3

Intranasal administration of the candidate vaccine drives strong systemic and mucosal immune responses C57BL/6 mice were given the SFL_mut and N proteins supplemented with or without combined adjuvant via the intranasal route (i.n.) or immunized with the SFL_mut and N proteins combined with KF and cGAMP via the i.p. route three times at 2-week intervals. These groups were compared to a group that received a reference inactivated vaccine via the intraperitoneal (i.p.) route twice at 4-week intervals. These mice were euthanized 14 days after the last immunization for sample collection. (A) Vaccination scheme. (B) Specific anti-S and anti-RBD IgG titers in serum (n = 6 mice per group). (C) Specific anti-N IgG titers in serum (n = 6 mice per group). (D) Splenocytes were assayed for IFN-γ and IL-4 cells production after re-stimulation with N and S protein by ELISpot assay (n = 4 mice per group). (E) Lung tissues were assayed for IFN-γ and IL-4 cells production after re-stimulation with N and S protein by ELISpot assay (n = 4 mice per group). (F and G) Nasal-associated lymphoid tissue (NALT) was collected from animals immunized with the candidate vaccine via the i.n. route, the inactivated vaccine via the i.p. route, and PBS via the i.n. route, and flow-cytometric analysis was performed. In addition, four mice in these three groups were intravenously (i.v.) injected with an anti-CD45 antibody 8 min prior to euthanasia. The lungs were collected, and lung mononuclear cells were stained with antibodies specific for CD3, CD4, CD44, CD62L, and CD69 for flow-cytometric analysis. The results are expressed as CD4⁺ TRM cells (CD3⁺CD4⁺CD44⁺CD62L^–CD45^–CD69⁺) and CD8⁺ TRM cells (CD3⁺CD4^–CD44⁺CD62L^–CD45^–CD69⁺). (F) NALT was assayed to determine the levels CD3⁺CD8⁺ T cells and CD3^–CD19⁺IgA⁺ B cells by flow cytometry (n = 4 mice per group). (G) CD4⁺ TRM and CD8⁺ TRM cells in the lungs. Representative flow cytometry gating strategies for CD4⁺ TRM cells and CD8⁺ TRM cells in the lungs are shown in Figure S3 (n = 4 mice per group). Data are expressed as the mean ± the SEM. The dotted line indicates the limit of detection (LOD), and values that fell below the detection limit are represented by the limit of detection value for statistical analysis. “ND” indicates that no individuals in this group had detectable levels. Significance was determined via one-way ANOVA with a Tukey multiple comparison test. ^∗p < 0.05, ^∗∗p < 0.01, ^∗∗∗p < 0.001, ns, no significance. See also Figures S2 and S3.

Figure 4

mRNA expression profiles in mouse nasal mucosal samples collected on day 14 after the last immunization (A) A Venn diagram of RNA-seq data was used to illustrate the differentially expressed genes (DEGs) in the nasal mucosa among mice immunized with the candidate vaccine (n = 6 mice), inactivated vaccine (n = 6 mice), or placebo (n = 6 mice). (B) Gene-enrichment analyses of the 470 DEGs. GO terms are labeled with the name and ID and were sorted by the –log10 (P) value. (C) Heatmap of the 22 DEGs in the response to interferon-gamma term (GO: 0034341) among the three groups (n = 6 mice per group). (D) Heatmap of the 13 DEGs in the cellular response to interferon-beta term (GO: 0035458) among the three groups (n = 6 mice per group). (E) Heatmap of the 18 DEGs in the cell chemotaxis term (GO: 0060326) among the three groups (n = 6 mice per group). High-expression values are colored red, and low-expression values are colored blue. See also Figure S4.

Figure 5

Immunogenicity and protective efficacy of the candidate vaccine (A) Diagram of human ACE2 transgenic (hACE2 Tg) mouse immunization and challenge. Six- to 8-week-old hACE2 Tg mice were immunized with three doses of the candidate vaccine or placebo (control) via the intranasal route at 14-day intervals or primed via the intraperitoneal route with 100 U of inactivated SARS-CoV-2 vaccine and then boosted at 28 days with the same dose of the inactivated vaccine. Seven days after the last immunization, the mice were bled, and S-, RBD-, and N-specific IgG antibodies, S- and RBD- IgA antibodies, and NAbs in the serum were evaluated. On day 10 after the last immunization, mice were challenged intranasally with 10³ PFU of SARS-CoV-2. The challenged mice were monitored for mortality and weight loss for 7 days after infection. (B–D) The antibody responses in the serum of vaccinated mice at 7 days after the last immunization were evaluated. An ELISA method was used to measure SARS-CoV-2 S-, RBD-, and N- specific IgG antibodies (B), NAbs (C), and IgA levels (D). n = 4 mice per group. (E) Survival: placebo-immunized mice were humanely euthanized at 4 dpi (n = 4) and 5 dpi (n = 6) due to IMBCAMS-defined endpoints (n = 6–14 mice per group per time point). (F) Weight loss (n = 6–14 mice per group per time point). Data are expressed as the mean ± the SEM. The dotted line indicates the limit of detection (LOD), and values that fell below the detection limit are represented by the limit of detection value for statistical analysis. Significance was determined via one-way ANOVA with a Tukey multiple comparison test. ^∗p < 0.05, ^∗∗p < 0.01, ^∗∗∗p < 0.001. Asterisks in (F) indicate statistical significance compared with weight of before infection by the unpaired two-sided Student’s t test (^∗∗p < 0.01).

Figure 6

Intranasal-immunized candidate vaccine offers superior protection at URT against challenge with SARS-CoV-2 (A–D) Nasal wash (A) and throat swab (B) samples were collected for 7 days in the vaccinated groups and until animals succumbed to infection in the placebo group (by 5 dpi) (n = 4–10 mice per group per time point). RNA was extracted, and viral RNA was assessed as copies per 100 μL. Four animals were euthanized at 2, 5, and 7 dpi in the vaccinated group. For the placebo group, four animals were euthanized at 2 dpi, four animals were euthanized based on humane endpoints at 4 dpi, and six animals were humanly euthanized at 5 dpi. Tissue samples were collected, and viral RNA was assessed as copies per mg. Viral RNA and infectious virus loads in nasal turbinates (C), lungs (D), and brains were determined (n = 4–10 mice per group per time point). (F) Images showing H&E staining and IHC staining for the SARS-CoV-2 N protein following infection with 10³ PFU/mouse. The images shown are from 2 and 5 dpi for all groups. Scale bar, 50 or 20 μm (insets). Each image is representative of a group of 4 mice for 2 dpi and 4 or 6 mice for 5 dpi. The dotted line indicates the limit of detection (LOD), and values below the detection limit are represented by the value of the detection limit for statistical analysis. Data are expressed as the mean ± the SEM. The dotted line indicates the limit of detection (LOD), and values that fell below the detection limit are represented by the limit of detection value for statistical analysis. Significance was determined via one-way ANOVA with a Tukey multiple comparison test. ^∗p < 0.05, ^∗∗p < 0.01, ^∗∗∗p < 0.001. See also Figure S5.

Figure 7

Neutralization of SARS-CoV-2 pseudovirus in sera and detection of N- and S-specific T cell responses against the B.1351 and B.1.617.2 variants SARS-CoV-2 wild-type (WT), B.1.351 variant, or B.1.617 variant pseudoviruses were incubated with different serum sample dilutions for 1 h at 37°C before the mixtures were added to ACE2-overexpressing 293T cells. Transduction efficiency was quantified by measuring virus-encoded luciferase activity in cell lysates 48 h after transduction and used to calculate the serum dilution factor that resulted in a 50% reduction in pseudovirus particles that were associated with different degrees of S protein-mediated cell entry. (A and B) The 50% pseudovirus neutralization (pVNT₅₀) in serum from mice immunized with the candidate vaccine (A) and mice immunized with the inactivated vaccine (B) against the B.1.351 and B.1.617.2 variants compared with that against the wild-type (WT) virus (n = 4 mice per group). (C–F) Antigen-specific activation of T cells by the N and S proteins of the B.1.351 and B.1.617.2 variants compared with the homologous WT proteins. N-specific (C) and S-specific (D) activation of T cells in splenocytes from mice immunized with the candidate vaccine or the inactivated vaccine at 14 days after the third immunization. N-specific (E) and S-specific (F) activation of T cells in lung tissues from mice immunized with the candidate vaccine or the inactivated vaccine at 14 days after the third immunization (n = 4 mice per group). Significance was determined via one-way ANOVA with a Tukey multiple comparison test. ^∗p < 0.05, ^∗∗p < 0.01, ^∗∗∗p < 0.001, ns, no significance.