Chimeric spike mRNA vaccines protect against Sarbecovirus challenge in mice

Abstract

The emergence of severe acute respiratory syndrome coronavirus (SARS-CoV) in 2003 and SARS-CoV-2 in 2019 highlights the need to develop universal vaccination strategies against the broader Sarbecovirus subgenus. Using chimeric spike designs, we demonstrate protection against challenge from SARS-CoV, SARS-CoV-2, SARS-CoV-2 B.1.351, bat CoV (Bt-CoV) RsSHC014, and a heterologous Bt-CoV WIV-1 in vulnerable aged mice. Chimeric spike messenger RNAs (mRNAs) induced high levels of broadly protective neutralizing antibodies against high-risk Sarbecoviruses. By contrast, SARS-CoV-2 mRNA vaccination not only showed a marked reduction in neutralizing titers against heterologous Sarbecoviruses, but SARS-CoV and WIV-1 challenge in mice resulted in breakthrough infections. Chimeric spike mRNA vaccines efficiently neutralized D614G, mink cluster five, and the UK B.1.1.7 and South African B.1.351 variants of concern. Thus, multiplexed-chimeric spikes can prevent SARS-like zoonotic coronavirus infections with pandemic potential.

Fig. 1.. Genetic design of chimeric Sarbecovirus spike vaccines.

(A) Genetic diversity of pandemic and bat zoonotic coronaviruses. HKU3-1 is shown in yellow, SARS-CoV is shown in light blue, RsSHC014 is shown in orange, and SARS-CoV-2 is shown in purple. (B) Spike chimera 1 includes the NTD from HKU3-1, the RBD from SARS-CoV, and the rest of the spike from SARS-CoV-2. Spike chimera 2 includes the RBD from SARS-CoV-2 and the NTD and S2 from SARS-CoV. Spike chimera 3 includes the RBD from SARS-CoV and the NTD and S2 SARS-CoV-2. Spike chimera 4 includes the RBD from RsSHC014 and the rest of the spike from SARS-CoV-2. SARS-CoV-2 furin KO spike vaccine and is the norovirus capsid vaccine. (C) Table summary of chimeric spike constructs.

Fig. 2.. Human and animal coronavirus spike binding and hACE2-blocking responses in chimeric and monovalent SARS-CoV-2 spike-vaccinated mice.

Serum antibody ELISA binding responses were measured in the five different vaccination groups. Before immunization, after prime, and after boost binding responses were evaluated against Sarbecoviruses, MERS-CoV, and common-cold CoV antigens including (A) SARS-CoV Toronto Canada (Tor2) S2P, (B) SARS-CoV-2 S2P D614G, (C) SARS-CoV-2 RBD, (D) SARS-CoV-2 NTD, (E) Pangolin GXP4L spike, (F) RaTG13 spike, (G) RsSHC014 S2P spike, (H) HKU3-1 spike, (I) MERS-CoV spike, and (J) hACE2 blocking responses against SARS-CoV-2 spike in the distinct immunization groups. Blue squares indicate mice from group 1, orange triangles indicate mice from group 2, green triangles indicate mice from group 3, red rhombuses indicate mice from group 4, and upside-down triangles indicate mice from group 5. Statistical significance for the binding and blocking responses is reported from a Kruskal-Wallis test after Dunnett’s multiple comparison correction. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001.

Fig. 3.. Live Sarbecovirus neutralizing antibody responses in vaccinated mice.

Neutralizing antibody responses in mice from the five different vaccination groups were measured by using nanoluciferase-expressing recombinant viruses. (A) SARS-CoV neutralizing antibody responses from baseline and after boost in the distinct vaccine groups. (B) SARS-CoV-2 neutralizing antibody responses from baseline and after boost. (C) RsSHC014 neutralizing antibody responses from baseline and after boost. (D) WIV-1 neutralizing antibody responses from baseline and after boost. (E) The neutralization activity in groups 1 and 4 against SARS-CoV-2 D614G, South African B.1.351, UK B.1.1.7, and mink cluster 5 variant. (F) Neutralization comparison of SARS-CoV-2 D614G versus South African B.1.351, versus UK B.1.1.7, and versus mink cluster 5 variant. Statistical significance for the live-virus neutralizing antibody responses is reported from a Kruskal-Wallis test after Dunnett’s multiple comparison correction. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001.

Fig. 4.. In vivo protection against Sarbecovirus challenge after mRNA-LNP vaccination.

(A) Percent starting weight from the different vaccine groups of mice challenged with SARS-CoV MA15. (B) SARS-CoV MA15 lung viral titers in mice from the distinct vaccine groups. (C) SARS-CoV MA15 nasal turbinate titers. (D) Percent starting weight from the different vaccine groups of mice challenged with SARS-CoV-2 MA10. (E) SARS-CoV-2 MA10 lung viral titers in mice from the distinct vaccine groups. (F) SARS-CoV-2 MA10 nasal turbinate titers. (G) Percent starting weight from the different vaccine groups of mice challenged with WIV-1. (H) WIV-1 lung viral titers in mice from the distinct vaccine groups. (I) WIV-1 nasal turbinate titers. (J) Percent starting weight from the different vaccine groups of mice challenged with SARS-CoV-2 B.1.351. (K) SARS-CoV-2 B.1.351 lung viral titers in mice from the distinct vaccine groups. (L) SARS-CoV-2 B.1.351 nasal turbinate titers. The vaccines used in the different groups are denoted at bottom. Statistical significance for weight loss is reported from a two-way analysis of variance (ANOVA) after Dunnett’s multiple comparison correction. For lung and nasal turbinate titers, statistical significance is reported from a one-way ANOVA after Tukey’s multiple comparison correction. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001.

Fig. 5.. Lung pathology in vaccinated mice after SARS-CoV and SARS-CoV-2 challenge.

(A) Hematoxylin and eosin 4 days after infection lung analysis of SARS-CoV MA15–challenged mice from the different groups: group 1, chimeras 1 to 4 prime and boost; group 2, chimeras 1 and 2 prime and 3 and 4; group 3, chimera 4 prime and boost, SARS-CoV-2 furin KO prime and boost, and norovirus capsid prime and boost. (B) Lung pathology quantitation in SARS-CoV MA15–challenged mice from the different groups. Macroscopic lung discoloration score, microscopic ALI score, and DAD in day 4 after infection lung tissues are shown. (C) Hematoxylin and eosin 4 days after infection lung analysis of SARS-CoV-2 MA10–challenged mice from the different groups. (D) Lung pathology measurements in SARS-CoV-2 MA10–challenged mice from the different groups. Macroscopic lung discoloration score, microscopic ALI score, and DAD in day 4 after infection lung tissues are shown. Statistical significance is reported from a one-way ANOVA after Dunnet’s multiple comparison correction. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001.