An intranasal subunit vaccine induces protective systemic and mucosal antibody immunity against respiratory viruses in mouse models

Abstract

Although vaccines are usually given intramuscularly, the intranasal delivery route may lead to better mucosal protection and limit the spread of respiratory virus while easing administration and improving vaccine acceptance. The challenge, however, is to achieve delivery across the selective epithelial cell barrier. Here we report on a subunit vaccine platform, in which the antigen is genetically fused to albumin to facilitate FcRn-mediated transport across the mucosal barrier in the presence of adjuvant. Intranasal delivery in conventional and transgenic mouse models induces both systemic and mucosal antigen-specific antibody responses that protect against challenge with SARS-CoV-2 or influenza A. When benchmarked against an intramuscularly administered mRNA vaccine or an intranasally administered antigen fused to an alternative carrier of similar size, only the albumin-based intranasal vaccine yields robust mucosal IgA antibody responses. Our results thus suggest that this needle-free, albumin-based vaccine platform may be suited for vaccination against respiratory pathogens.

Fig. 1. Intranasal vaccination with RBD-fused MSA in female BALB/c mice induces protective antigen-specific IgG and IgA antibodies.

a Top panel: Illustration of a fusion design with albumin genetically fused to RBD, and intranasal vaccination followed by FcRn-mediated transport of the vaccine across polarized mucosal epithelial cells, immune cell processing, and transport of generated IgG and IgA antibodies by FcRn and the pIgR, respectively, to the mucosal surface, and also blood for IgG. Bottom panel: Flow chart outlining the intranasal vaccination protocol with live virus challenge or endpoint sample harvest at week 5. Mice were vaccinated with equimolar amounts of RBD and RBD-MSA (prime dose: 6.2 µg and 19.9 µg, respectively, boost: 10% of prime) in combination with 20 μg CpG, or given PBS. b Illustration of ELISA and data for determination of RBD epitope availability on the RBD-albumin fusion compared to RBD alone, using the commercial monoclonal antibodies sotrovimab, cilgavimab and tixagevimab. c Illustration of the flow cytometric bead array (FCBA) for detection of antigen-specific antibodies in sera and BALF and data from analyses at endpoint. d Illustration of the FCBA for detection of antibodies able to inhibit human ACE2-RBD binding. Antibodies can inhibit human ACE2 binding to RBD-coupled beads (1a-1b), or not, leading to detection of ACE2-DIG bound to RBD-coupled beads by anti-DIG-PE (2), and results from sera and BALF at endpoint, detected using FCBA and ELISA (only sera). e Illustration of a pseudovirus neutralization cellular assay based on lentiviral infection of 293T-hACE2-TMPRSS2 cells where the presence of antibodies blocks cellular infection (1), while reduced blocking ability leads to pseudoviral entry and GFP expression (2), and results from sera and BALF at endpoint using SARS-CoV-2 pseudovirus of the ancestral strain (Wuhan). Curve plots in b are presented as each replicate of technical duplicates. Bars indicating c, d group mean ± SD with individual mice represented as a single datapoint (sera: n = 6, BALF: RBD n = 3 and RBD-MSA n = 4) and e technical duplicates of pooled biological samples per group (sera: n = 6, BALF: RBD n = 3 and RBD-MSA n = 5). c, d Two-tailed unpaired t-test. a Created or b–e partially created in BioRender.

Fig. 2. Intranasal vaccination with RBD-fused MSA protects female K18-hACE2 mice from lethal challenge with SARS-CoV-2.

a Illustration of the K18-hACE2 mouse model and RBD-specific antibodies in sera at day 28 detected using FCBA. b The ability of antibodies in sera of vaccinated K18-hACE2 mice at day 28 to inhibit ACE2-RBD binding and block cellular infection by SARS-CoV-2 pseudovirus of the ancestral strain (Wuhan). c Weight change after viral challenge of intranasally vaccinated K18-hACE2 mice by SARS-COV-2 WA1/2020. d Viral load in the lungs of intranasally vaccinated K18-ACE2 mice 7 days post infection e Hematoxylin and eosin staining of lungs of intranasally vaccinated K18-ACE2 mice 7 days post infection. One representative lung from each group is shown with magnification x2 (scale bar 1 mm) and x4 (scale bar 500 µm). a, b, d Bar indicating group mean ± SD with individual mice represented as a single datapoint (PBS n = 10, RBD and RBD-MSA n = 12, except d PBS n = 7)). b Curve plot presented as group mean ± SD of technical duplicates (PBS n = 10, RBD and RBD-MSA n = 12). e Bars indicating number of biological samples per indication within each treatment group (group sizes: PBS n = 10, RBD and RBD-MSA n = 12). Weight in c presented as percentage weight compared to the weight at day of infection, as biological group mean ± SEM (PBS n = 10, RBD and RBD-MSA n = 12). a and c One-way ANOVA with Tukey’s multiple comparison, b Two-tailed unpaired t-test (ACE2-RBD binding) or Kruskal-Wallis test with Dunn’s multiple comparison (pseudovirus neutralization), and d-e Two-tailed Mann-Whitney test. a–e Partially created in BioRender.

Fig. 3. Intranasal vaccination with RBD-fused MSA in female BALB/c mice induces robust antigen-specific antibodies beyond that of intramuscular vaccination.

a Illustration of intranasal vaccination with an RBD-albumin fusion compared with intramuscular vaccination with the mRNA vaccine from BioNTech-Pfizer (Comirnaty/BNT162b2), and the vaccination regimen in female BALB/c mice. b RBD-specific IgG and IgA in sera and BALF at endpoint after vaccination of BALB/c mice (prime dose: 6.2 µg RBD and 19.9 µg RBD-MSA in combination with 20 μg CpG, 3 μg Comirnaty, or PBS), and c the ability of the antibodies to inhibit human ACE2-RBD binding. d Percentage of B-cells that are antigen-specific in mediastinal lymph nodes detected by flow cytometry. All serology detected using FCBA. Bars indicating group mean ± SD with individual mice represented as a single datapoint (b PBS and BnT-Pfizer n = 5, and RBD and RBD-MSA n = 6, or c n = 5, except RBD-MSA n = 6). d All mediastinal lymph nodes per group (PBS and BnT-Pfizer n = 5, and RBD and RBD-MSA n = 6) were merged and samples prepared in technical triplicates for flow cytometry. b One-way ANOVA with Tukey’s multiple comparison and c Kruskal–Wallis test with Dunn’s multiple comparison. a Created or b and c partially created in BioRender.

Fig. 4. Intranasal vaccination with RBD-fused engineered HSA in Tg32-hFc mice induces robust antigen-specific antibodies beyond that of intramuscular vaccination and of an alternative carrier.

a Illustration of Tg32-hFc mice expressing human FcRn and chimeric IgG1 with human Fc (chIgG1), and RBD-specific IgG and IgA in sera, BALF, nose and saliva of female mice at endpoint (prime dose: 6.2 μg RBD and 20.0 μg RBD-QMP in combination with 20 μg CpG), and the ability of the antibodies in sera and BALF to b inhibit human ACE2-RBD binding and c inhibit cellular infection by SARS-CoV-2 pseudovirus of the ancestral strain (Wuhan). d Percentage of antigen-specific B-cells in mediastinal lymph nodes. e Illustration of Tg32-hFc mice that were intranasally or intramuscularly vaccinated with RBD-QMP, and RBD-specific IgG and IgA in sera and BALF 5 weeks after initial vaccination of male mice (prime dose: 20.0 μg RBD-QMP in combination with 20 μg CpG) f RBD-specific IgG and IgA in sera, BALF, nose and saliva of female and male Tg32-hFc mice at endpoint (prime dose: 20.0 μg RBD-QMP (2 F, 3 M) and 22.0 μg RBD-Tf (3 F, 2 M) in combination with 20 μg CpG), and g the ability of the antibodies in sera and BALF to inhibit human ACE2-RBD binding. All serology was detected using FCBA. Bars indicating group mean ± SD with individual mice represented as a single datapoint (a RBD n = 5 and RBD-QMP n = 6, except nose RBD n = 3, b, e, f, g n = 5, except e RBD n = 3 (intranasal) or n = 4 (intramuscular)). Pseudovirus neutralization c presented as technical duplicates of pooled biological samples per group (RBD n = 5 and RBD-QMP n = 6). d All mediastinal lymph nodes per group (PBS n = 3, RBD and RBD-QMP n = 5) were merged and samples prepared in technical triplicate (PBS as duplicate) for flow cytometry. a, b, f, g Two-tailed unpaired t-test, and e Two-tailed Mann–Whitney test. a–c, e–g Partially created in BioRender.

Fig. 5. Intranasal vaccination with RBD-fused engineered HSA induces antibodies solely toward the subunit vaccine antigen.

a Illustration of the RBD-fused albumin with CpG site-specifically conjugated to cysteine 34 (C34) in domain I of albumin, distal from the principal FcRn binding site. b RBD-specific IgG and IgA in sera, BALF, nose, and saliva at endpoint after intranasal vaccination of female Tg32-hFc mice (prime dose: 20.0 μg RBD-QMP with or without 20 μg CpG, 21.4 μg CpG-RBD-QMP, or PBS). c Percentage antigen-specific B-cells in mediastinal lymph nodes. d Illustration of ELISA for detection of antibodies against human albumin (HSA) after vaccination, and HSA-specific antibodies generated at endpoint after intranasal prime and boost with QMP-containing albumin detected using human IgG. e Illustration of the HSA/hFcRn mouse model expressing human FcRn and HSA. Immune responses at endpoint after intranasal vaccination of female mice (prime dose: 6.2 μg RBD, 20.0 μg RBD-WT and 20.0 μg RBD-QMP in combination with 20 μg CpG) displaying f RBD-specific IgG and IgA antibodies in sera and BALF, respectively, and g HSA-specific mouse IgG antibodies in sera. Serology was detected using FCBA (b and f) or ELISA (d and g). Bars indicating group mean ± SD with individual mice represented as a single datapoint (except d) and curve plots presented as biological mean per group ± SD. b, d, f, g n = 6, except RBD-QMP n = 5. c All mediastinal lymph nodes per group n = 6, except RBD-QMP n = 5 were merged and samples prepared in technical triplicate for flow cytometry. b and f One-way ANOVA with Tukey’s multiple comparison. a and e Created or b, d, f, g partially created in BioRender.