Intranasal Delivery of MVA Vector Vaccine Induces Effective Pulmonary Immunity Against SARS-CoV-2 in Rodents

Abstract

Antigen-specific tissue-resident memory T cells (Trms) and neutralizing IgA antibodies provide the most effective protection of the lungs from viral infections. To induce those essential components of lung immunity against SARS-CoV-2, we tested various immunization protocols involving intranasal delivery of a novel Modified Vaccinia virus Ankara (MVA)-SARS-2-spike vaccine candidate. We show that a single intranasal MVA-SARS-CoV-2-S application in mice strongly induced pulmonary spike-specific CD8⁺ T cells, albeit restricted production of neutralizing antibodies. In prime-boost protocols, intranasal booster vaccine delivery proved to be crucial for a massive expansion of systemic and lung tissue-resident spike-specific CD8⁺ T cells and the development of Th1 - but not Th2 - CD4⁺ T cells. Likewise, very high titers of IgG and IgA anti-spike antibodies were present in serum and broncho-alveolar lavages that possessed high virus neutralization capacities to all current SARS-CoV-2 variants of concern. Importantly, the MVA-SARS-2-spike vaccine applied in intramuscular priming and intranasal boosting treatment regimen completely protected hamsters from developing SARS-CoV-2 lung infection and pathology. Together, these results identify intramuscular priming followed by respiratory tract boosting with MVA-SARS-2-S as a promising approach for the induction of local, respiratory as well as systemic immune responses suited to protect from SARS-CoV-2 infections.

Keywords: bronchus-associated lymphoid tissue (BALT); lungs; modified vaccinia virus Ankara (MVA); respiratory tract; severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2); spike (S) protein; vaccination; vaccine.

Figure 1

Intranasal priming with MVA-SARS-2-S induces BALT and transient recruitment of T cells and macrophages in the lung. (A) Immunization protocol scheme. (B) Representative photomicrographs of lung sections reveal induction of BALT peaking at d11 after vaccine application. (C) Quantification of cumulative BALT size per lung section averaged on 3-4 lung sections per mouse. (D–G) Cellular composition of broncho-alveolar lavage (BAL; (D, E) and lung (F, G) analyzed by spectral flow cytometry and depicted as tSNE plot. Cluster identities were revealed by manual gating as shown in Supplementary Figures 1A, B ; antibodies used are listed in Supplementary Table S1 (panel 1). Cells inside the area delineated by the black line were negative for anti-CD45-FITC antibodies i.v injected 3 -5 min before mice were sacrificed and are thus considered to be from the lung parenchyma. Cells outside that area were CD45-FITC⁺ and considered to be within or in close vicinity to blood vessels. (D, F) Representative tSNE plot of concatenated samples from one mouse each sacrificed before (control) or 11 or 40 days after vaccine application; colors refer to indicated cell populations; antibodies used are listed in Supplementary Table S1 (panel 2). (E, G) Upper left, concatenated data as in (D, F), colors indicate different mice analyzed. Upper right and bottom plots: de-concatenated data, colors indicate cell populations. (H) Absolute cell numbers of interstitial macrophages (int MF), type 1 conventional dendritic cells (cDC1), NK, T and B cells in lung. (C, H) Pooled data from 3-4 experiments with n = 10 per group. (C, H) Individual values (symbols) and mean group value (lines). Statistical analysis was done on log-transformed values using ordinary or Welch’s ANOVA followed by Dunnett’s T3 multiple comparisons test. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

Figure 2

A single intranasal vaccine application induces local and systemic spike-specific cellular and humoral responses. (A, B) Spike-specific CD8⁺ T cells accumulate in in the organs at different time points after vaccine application. (A) Representative dot plots and percentage of tetramer⁺ lung CD8⁺CD44^hi T cells in groups indicated. (B) Absolute cell counts of spike-specific CD8⁺CD44^hi T cells in different organs. (C, D) Spike-specific CD4⁺ T cells transiently accumulate in lungs and spleens up to day 35 after immunization. (C) Representative dot plots and percentages of IFN-γ-expressing lung CD8⁺CD44^hi T cells of groups indicated after ex vivo re-stimulation with the pool of spike-specific and immnodominant peptides ( Supplementary Table S2 ) for 6hr. (D) Frequencies of IFN-γ-expressing CD4⁺CD44^hi cells in different organs isolated from mice at indicated time points after i.n. vaccine administration. (E–G) Spike-specific antibodies in serum (E, F) and broncho-alveolar lavage fluid (BAL; (G) measured by ELISA. (E) Mean OD group values of serial serum dilutions and (F) and individual EC₅₀ values for Spike-specific serum IgG for the groups indicated. (G) Mean group OD values in serial BAL dilutions for Spike-specific IgG (left) and IgA (right). (B, D, F) Pooled data from 3-4 experiments with n = 4-10 mice per group. Individual values (symbols) and mean group value (line). Statistical analysis was done on log-transformed values using ordinary or Welch’s ANOVA followed by Dunnett’s T3 multiple comparisons test. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

Figure 3

Intranasal boost with MVA-SARS-2-S induces strong local and systemic cellular immune responses irrespectively of the route of priming. (A) Immunization protocol scheme. (B) Cumulative size of BALT structures averaged on 3-4 central lung sections per mouse. (C, D) Composition of lung CD3⁺ T cells analyzed by spectral flow cytometry and depicted as tSNE plot. Cluster identities were revealed by manual gating as shown in Supplementary Figure 2 ; antibodies used are listed in Supplementary Table S1 (panel 3). Cells inside the area delineated by the black line were negative for anti-CD45-FITC antibodies i.v. injected 3 -5 min. before mice were sacrificed and are thus considered to be from the lung parenchyma. Cells outside that area were CD45-FITC⁺ and considered to be within or in close vicinity to blood vessels. (C) tSNE plot of concatenated FCS files from one representative mouse of each prime-boost group immunized with 10⁷ PFU vaccine and one non-immunized control mouse. (D) Individual tSNE plots of de-concatenated FCS files from (D). (E) Absolute counts of CD8⁺ T cells specific for S-V8L evaluated by tetramer staining in tissues and groups indicated. (F) Frequencies of IFN-γ⁺ CD4 T cells in organs indicated determined by intracellular cytokine staining after re-stimulation with the pool of S_1-129 together with immnodominant peptides and brefeldin A ( Supplementary Table S2 ) for 6hr. (G) Spike-specific antibodies in serum and (H, I) BAL determined by ELISA. Mean OD group values of serial serum and BAL dilutions and individual EC₅₀ values for Spike-specific IgG in serum (G) and BAL (H) and Spike-specific IgA in BAL (I) are shown for the groups indicated analyzed at day 40. Pooled data from 3-4 experiments with n = 10 per group. (B,E–I) Individual values (signs) and mean group value (line). Statistical analysis was done on log-transformed values using ordinary or Welch’s ANOVA followed by Dunnett’s T3 multiple comparisons test. Green stars - difference to the mice immunized with PO^7na-d24; Black stars - difference between different treatment groups as indicated with line. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. (B, E–I) Data from control mice and mice immunized with 10⁷ IU MVA-SARS-2-S are identical to that shown in Figure 1C ; (B, D–F).

Figure 4

Intramuscular-intranasal prime-boost immunization protocol generates S-specific T cells with tissue resident memory phenotype. (A) Immunization protocol scheme. (B) Relative change of serum cytokine concentrations at times indicated after i.n. boost determined by Ayoxxa’s LUNARIS™ multiplex biomarker platform; cytokine concentrations are given Supplementary Table S4 . (C) Cumulative size of BALT structures averaged from 3-4 central lung sections per mouse. (D) Absolute counts of S-V8L-T⁺CD44^hi CD8 T cells in different organs of vaccination groups indicated. (E) tSNE plot illustrating CD3⁺CD8⁺CD44^hi S-V8L-tet⁺ T cells from various organs. Shown are concatenated FCS file from each organ indicated of 2 representative mice immunized by the sPB^8mu-7na protocol and analyzed by spectral flow cytometry. Colors refer to cells originating from indicated tissues (left, big plot) or to expression levels with red indicating high and blue low expression (6 smaller plots). (F) Frequencies of CD44^hiIFN-γ⁺ CD4 T cells analyzed by intracellular cytokine staining after stimulation with the pool of S_1-129 together with immnodominant peptides and brefeldin A ( Supplementary Table S2 ) for 6hr. Pooled data from 2 experiments with n = 6-10 per group. Individual values (signs) and mean group value (line). Statistical analysis was done on log-transformed values using ordinary or Welch’s ANOVA followed by Dunnett’s T3 multiple comparisons test. **p < 0.01, ***p < 0.001, ****p < 0.0001.

Figure 5

Intranasal boosting with MVA-SARS-2-S induces high titer of neutralizing antibodies in lung and serum. (A–C) Spike-specific antibodies in serum (A) and BAL (B, C) measured by ELISA. Mean group OD values for S-specific IgG (A, B) and IgA (C) determined on serially diluted serum (A) and BAL (B, C) of indicated groups 14 days post boost (d40). (D–G) Neutralizing antibodies in serum (D, E) and BAL (F, G) measured by surrogate virus neutralization test (sVNT). (D, F) Inhibition of binding interaction of SARS-CoV-2 spike-RBD from different SARS-CoV-2 variants of concern (VoC) with ACE2 by addition of sera (D) or BAL (F) of immunized or control mice. Assay performed in triplicate; mean percentages of neutralization. (E, G) Reciprocal titers of neutralizing antibodies against SARS-CoV-2 VoCs from serum (E) or BAL (G) determined as the dilution retaining binding reduction > mean+2SD of non-immunized control mice. (A–G) Pooled data from 2-4 experiments with n = 10 per group. (E, G) Individual values (signs) and mean group value (line). Statistical analysis was done using Chi-square test for trend, ns, not significant, ****p < 0.001.

Figure 6

Intranasal immunization with MVA-SARS-2-S protects hamsters from SARS-CoV-2 infection. (A) Immunization and infection protocol scheme. (B, C) Relative body weight (B) and clinical scores (C) of hamsters after infection with SARS-CoV-2. (D) SARS-CoV-2 virus titers in the lungs measured by quantitative RT-PCR. (E) Representative stitched photomicrographs showing an overview of the entire left lung lobe stained for SARS-CoV-2 nuceloprotein (brown staining). Original magnification 40x, inset 400x. (F) Semiquantitative scoring of SARS-CoV-2 nuceloprotein per lung section averaged on 2 lung sections per hamster. (G) Representative stitched photomicrographs showing an overview of the entire left lung lobe stained with hematoxylin and eosin (H, E) indicate severe inflammation in the lungs of hamsters immunized with control, MVA-WT virus, which is almost not existent in the lungs of hamsters immunized with MVA-SARS-CoV-2. Original magnification 40x. (H) Lung inflammation scores calculated as described supplemental methods. (B–H) Pooled data from one experiment with n = 7 per group. Individual values (signs) and mean group value (line). Statistical analysis was done using Mann-Whitney t test. **p < 0.01, ***p < 0.001.

Figure 7

High titer of neutralizing antibodies in serum of immunized hamsters. (A, B) Levels of neutralizing Spike-specific antibodies in serum of hamsters immunized with MVA-WT or MVA-SARS-CoV-2 and measured using SARS-CoV-2 plaque reduction neutralization test (PRNT). Reciprocal titer values that result in 50% (A) or 90% (B) reduction in virus infectivity determined on serially diluted serum of indicated groups 20 days post boost (d46) and 6 days post SARS-CoV-2 infection. (C) Neutralizing antibodies against different SARS-CoV-2-S variants measured by surrogate virus neutralization test (sVNT). Inhibition of binding interaction of SARS-CoV-2 spike-RBD with ACE2 by addition of sera of non-immunized and uninfected hamsters (Invitrogen HS), MVA-WT, and MVA-SARS-CoV-2-S immunized and SARS-CoV-2 infected hamsters. Assay performed in duplicate for each sample; mean percentages of neutralization per group ± SD are shown. (A–C) Pooled data from 2-4 experiments with n = 7 per group. Individual values (signs) and mean group value (line). Statistical analysis was done using Mann-Whitney t test. *p < 0.05, **p < 0.01.