Development of a multi-antigenic SARS-CoV-2 vaccine candidate using a synthetic poxvirus platform

Abstract

Modified Vaccinia Ankara (MVA) is a highly attenuated poxvirus vector that is widely used to develop vaccines for infectious diseases and cancer. We demonstrate the construction of a vaccine platform based on a unique three-plasmid system to efficiently generate recombinant MVA vectors from chemically synthesized DNA. In response to the ongoing global pandemic caused by SARS coronavirus-2 (SARS-CoV-2), we use this vaccine platform to rapidly produce fully synthetic MVA (sMVA) vectors co-expressing SARS-CoV-2 spike and nucleocapsid antigens, two immunodominant antigens implicated in protective immunity. We show that mice immunized with these sMVA vectors develop robust SARS-CoV-2 antigen-specific humoral and cellular immune responses, including potent neutralizing antibodies. These results demonstrate the potential of a vaccine platform based on synthetic DNA to efficiently generate recombinant MVA vectors and to rapidly develop a multi-antigenic poxvirus-based SARS-CoV-2 vaccine candidate.

Fig. 1. sMVA construction and characterization.

a Schematic of MVA genome. The MVA genome is ~178 kbp in length and contains an internal unique region (UR) flanked by ~9.6 kbp inverted terminal repeat (ITR) sequences. b sMVA fragments. The three sub-genomic sMVA fragments (F1–F3) comprise ~60 kbp of the left, central, and right part of the MVA genome as indicated. sMVA F1/F2 and F2/F3 share ~3 kbp overlapping homologous sequences for recombination (red dotted crossed lines). Approximate genome positions of commonly used MVA insertion sites (Del2, IGR69/70, Del3) are indicated. c Terminal CR/HL/CR sequences. Each of the sMVA fragments contains at both ends a sequence composition comprising a duplex copy of the MVA terminal hairpin loop (HL) flanked by concatemeric resolution (CR) sequences. BAC bacterial artificial chromosome vector. d sMVA reconstitution. The sMVA fragments are maintained in E. coli, isolated from the bacteria, and co-transfected into BHK cells, which are subsequently infected with FPV as a helper virus to initiate sMVA virus reconstitution. e PCR analysis. CEFs infected with sMVA, derived with either FPV HP1.441 (sMVA hp) or with FPV TROVAC from two independent virus reconstitutions (sMVA tv1 and sMVA tv2), were investigated by PCR for several genome positions, including the ITR sequences, the transition from the left or right ITR into the internal UR (left ITR/UR; UR/right ITR), the Del2, IGR69/70 and Del3 insertion sites, and the F1/F2 and F2/F3 recombination sites. The removal of the BAC vector sequences was also investigated. PCR with wtMVA-infected and uninfected cells, without sample (mock), or with MVA BAC was performed as controls. *Note the different sample order for the BAC-specific PCR as indicated by numbers. f Restriction pattern analysis. Viral DNA isolated from ultra-purified sMVA (sMVA tv1 and sMVA tv2) or wtMVA virus was compared by KpnI and XhoI restriction enzyme digestion. Experiments in e and f were performed twice with similar results.

Fig. 2. sMVA replication properties.

The replication properties of sMVA, derived either with FPV HP1.441 (sMVA hp) or with FPV TROVAC from two independent sMVA virus reconstitutions (sMVA tv1 and sMVA tv2), were compared with that of wtMVA. a Replication kinetics. BHK or CEF cells were infected in triplicates (n = 3) for each time point at 0.02 multiplicity of infection (MOI) with sMVA or wtMVA and viral titers of the inoculum and each triplicate infection were determined at 24 and 48 h post infection on CEF. Mixed-effects model with the Geisser-Greenhouse correction followed by Tukey’s multiple comparison test were applied; at 24 and 48 h post-infection differences between groups were not significant (p > 0.05). b, c Viral foci size analysis. BHK or CEF cell monolayers were infected at 0.002 MOI with sMVA or wtMVA and areas of viral foci (n = 20) were determined at 24 h post infection following immunostaining with anti-Vaccinia polyclonal antibody (αVAC). Panel c provides examples of sMVA and wtMVA viral foci following immunostaining of infected CEF. d Host cell range analysis. Various human cell lines (HEK293, A549, 143b, and HeLa), CEF or BHK cells were infected in duplicates (n = 2) at 0.01 MOI with sMVA or wtMVA and virus titers of each duplicate infection were determined in duplicates (n = 4 in total) at 48 h post infection on CEF. Dotted lines indicate the calculated virus titer of the inoculum based on 0.01 MOI. Differences between groups in b and d were calculated using one-way ANOVA followed by Tukey’s (b) or Dunnett’s (d) multiple comparison tests; ns = not significant (p > 0.05). Data in a and d are presented as mean values + SD. Lines in b represent median values. Source data are provided as a Source Data file.

Fig. 3. sMVA in vivo immunogenicity.

sMVA derived either with FPV HP1.441 (sMVA hp) or with FPV TROVAC from two independent virus reconstitutions (sMVA tv1 and sMVA tv2) was compared by in vivo analysis with wtMVA. C57BL/6 mice (n = 4) were immunized twice at a 3-week interval with low (1 × 10⁷ PFU) or high (5 × 10⁷ PFU) dose of sMVA or wtMVA. Mock-immunized mice were used as controls. a Binding antibodies. MVA-specific binding antibodies (IgG titer) stimulated by sMVA or wtMVA were measured after the first and second immunization by ELISA. b NAb responses. MVA-specific NAb titers induced by sMVA or wtMVA were measured after the booster immunization against recombinant wtMVA expressing a GFP marker. c, d T-cell responses. MVA-specific IFNγ, TNFα, IL-4, and IL-10-secreting CD8+ (c) and CD4+ (d) T-cell responses induced by sMVA or wtMVA after two immunizations were measured by flow cytometry following ex vivo antigen stimulation using B8R immunodominant peptides. Differences between groups were evaluated using one-way ANOVA with Tukey’s multiple comparison test; ns = not significant (p > 0.05). Data in a and b are presented as mean values + SD. Lines in c and d represent median values. Source data are provided as a Source Data file.

Fig. 4. Construction and characterization of sMVA-CoV2 vectors.

a Schematic representation of vector construction. S and N antigen sequences (red spheres and green triangles) were inserted into sMVA fragments F2 and F3 by bacterial recombination methods in E. coli. The modified sMVA fragments of F2 and F3 with inserted antigen sequences and the unmodified sMVA fragment F1 were isolated from E. coli and co-transfected into FPV-infected BHK cells to initiate virus reconstitution. b Schematics of single (sMVA-S, sMVA-N) and double (sMVA-N/S, sMVA-S/N) recombinant sMVA-CoV2 vectors with S and N antigen sequences inserted into commonly used MVA insertion sites (Del2, IGR69/70, Del3) as indicated. All antigens were expressed via the Vaccinia mH5 promoter. ITR inverted terminal repeat. c PCR analysis. CEFs infected with the single and double recombinant sMVA-CoV2 vectors derived with FPV HP1.441 (sMVA-S/N hp, sMVA-N/S hp) or TROVAC (sMVA-S/N tv, sMVA-N/S tv, sMVA-S tv, sMVA-N tv) were evaluated by PCR with primers specific for the Del2 and Del3 insertion sites harboring the N and S antigen sequences or primers specific for the F1/F2 and F2/F3 recombination sites. d Western Blot. BHK cells infected with the sMVA-CoV2 vectors were evaluated for antigen expression by Western Blot using anti-S1 and anti-N antibodies (αS1 and αN). Vaccinia B5R protein was verified as infection control. Higher and lower molecular weight bands may represent mature and immature protein species. e Flow cytometry staining. HeLa cells infected with the vaccine vectors were evaluated by cell surface and intracellular flow staining using anti-S1, S2, and N antibodies (αS1, αS2, and αN). Live cells were used to evaluate cell surface antigen expression. Fixed and permeabilized cells were used to evaluate intracellular antigen expression. Anti-Vaccinia virus antibody (αVAC) was used as staining control to verify MVA protein expression. Cells infected with sMVA or wtMVA or uninfected cells were used as controls for experiments in c, d, and e as indicated. The experiments in c, d, and e were performed twice with similar results.

Fig. 5. Humoral immune responses stimulated by sMVA-CoV2 vectors.

Balb/c mice (n = 5) immunized twice in a 3-week interval with 5 × 10⁷ PFU of the single and double recombinant sMVA-CoV2 vectors derived with FPV HP1.441 (sMVA-S/N hp and sMVA-N/S hp) or TROVAC (sMVA-S/N tv, sMVA-N/S tv, sMVA-S tv, sMVA-N tv) were evaluated for SARS-CoV-2-specific humoral immune responses. a, b Binding antibodies. S-, RBD-, and N-specific binding antibodies induced by the vaccine vectors were evaluated after the first (a) and second (b) immunization by ELISA. As a comparison, antigen-specific endpoint titers measured in n = 19 plasma samples from SARS-CoV-2 convalescent individuals (Fig. S4) were added in a and b. Data are presented as mean values + SD. One-way ANOVA with Tukey’s multiple comparison test was used to evaluate the differences between binding antibody end-point titers. c IgG2a/IgG1 isotype ratio. S-, RBD-, and N-specific binding antibodies of the IgG2a and IgG1 isotype were measured after the second immunization using 1:10,000 serum dilution, and absorbance reading was used to calculate IgG2a/IgG1 antibody ratio. Lines represent median values. One-way ANOVA with Dunnett’s multiple comparison test was used to compare each group mean IgG2a/IgG1 ratio to a ratio of 1 (balanced Th1/Th2 response). d–g NAb responses. SARS-CoV-2-specific NAb (NT90 titer) induced by the vaccine vectors were measured after the first (d, f) and second (e, g) immunization against SARS-CoV-2 pseudovirus (pv) (d, e) or infectious SARS-CoV-2 virus (f, g) in pooled sera of immunized mice. Shown is the average NT90 measured in duplicate (d, e) or triplicate (f, g) infection. N/A = failed quality control of the samples. Dotted lines indicate lowest antibody dilution included in the analysis. h SARS-CoV-2/SARS-CoV-2pv correlation analysis. Correlation analysis of NT90 measured in mouse sera after one and two immunizations using infectious SARS-CoV-2 virus and SARS-CoV-2pv. Pearson correlation coefficient (r) was calculated in h; *0.05 < p ≪0.01, **0.01 < p < 0.001, ***0.001 < p < 0.0001, ****p < 0.0001; ns = not significant. Source data are provided as a Source Data file.

Fig. 6. Cellular immune responses stimulated by sMVA-CoV2 vectors.

Balb/c mice (n = 5) immunized twice in a 3-week interval with 5 × 10⁷ PFU of the single or double recombinant sMVA-CoV2 vectors derived with FPV HP1.441 (sMVA-S/N hp and sMVA-N/S hp) or TROVAC (sMVA-S/N tv, sMVA-N/S tv, sMVA-S tv, sMVA-N tv) were evaluated for SARS-CoV-2-specific cellular immune responses. Antigen-specific CD8+ (a and b) and CD4+ (c and d) T-cell responses induced by the vaccine vectors after two immunizations were evaluated by flow cytometry for IFNγ, TNFα, IL-4, and IL-10 secretion following ex vivo antigen stimulation using SARS-CoV-2 S- and N-specific peptide libraries. Due to technical issues, 1–3 animals/group were not included in the CD4/TNFα analysis in c and d. One-way ANOVA with Tukey’s multiple comparison test was used to compare % of cytokine-specific T cells between immunized mice and mock controls. Lines represent median values; *0.05 < p < 0.01, **0.01 < p < 0.001, ***0.001 < p < 0.0001, ****p < 0.0001; ns = not significant. Source data are provided as a Source Data file.

Fig. 7. CD8+ T-cell responses induced by sMVA-N/S in HLA transgenic mice.

HLA-B*07:02 (B7) transgenic mice (n = 4) were immunized twice in a 3-week interval with 1 × 10⁷ PFU of sMVA-N/S or sMVA control vector (reconstituted with FPV TROVAC). B7 mice (n = 3) were mock-immunized as additional control. Development of SARS-CoV-2-specific CD8+ T cells was evaluated 1 week post booster immunization. a–d Intracellular cytokine staining. Nucleocapsid- (a, c) and Spike-specific (b, d) CD8+ T cells were evaluated by intracellular cytokine staining for IFNγ, TNFα, and IL-4 secretion following ex vivo antigen stimulation by N and S peptide libraries, respectively. Panels a and b show the percentage of CD3+/CD8+ T cells secreting IFNγ, TNFα, or IL-4 following peptide stimulation. Panels c and d show relative frequencies of CD8+ T cells secreting one or more cytokines after peptide stimulation. Total percentage of cytokine-secreting cells within CD3+/CD8+ population is indicated under each pie chart. e ELISPOT analysis of IFNγ-secreting cells following stimulation with S and N peptide libraries, S library sub-pools (1S1, 2S1, S2), and N26 peptide containing the HLA-B*07:02-restricted N-specific immunodominant epitope SPRWYFYYL. One-way ANOVA with Dunnett’s multiple comparison test was used in a and b. Two-way ANOVA with Dunnett’s multiple comparison test was used in e. Data in a, b, and e are presented as mean values ± SD; *0.05 < p < 0.01, **0.01 < p < 0.001, ***0.001 < p < 0.0001, ****p < 0.0001; ns = not significant. Source data are provided as a Source Data file.