Introduction of two prolines and removal of the polybasic cleavage site leads to optimal efficacy of a recombinant spike based SARS-CoV-2 vaccine in the mouse model

Abstract

The spike protein of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has been identified as the prime target for vaccine development. The spike protein mediates both binding to host cells and membrane fusion and is also so far the only known viral target of neutralizing antibodies. Coronavirus spike proteins are large trimers that are relatively instable, a feature that might be enhanced by the presence of a polybasic cleavage site in the SARS-CoV-2 spike. Exchange of K986 and V987 to prolines has been shown to stabilize the trimers of SARS-CoV-1 and the Middle Eastern respiratory syndrome coronavirus spikes. Here, we test multiple versions of a soluble spike protein for their immunogenicity and protective effect against SARS-CoV-2 challenge in a mouse model that transiently expresses human angiotensin converting enzyme 2 via adenovirus transduction. Variants tested include spike protein with a deleted polybasic cleavage site, the proline mutations, a combination thereof, as well as the wild type protein. While all versions of the protein were able to induce neutralizing antibodies, only the antigen with both a deleted cleavage site and the PP mutations completely protected from challenge in this mouse model.

Importance: A vaccine for SARS-CoV-2 is urgently needed. A better understanding of antigen design and attributes that vaccine candidates need to have to induce protective immunity is of high importance. The data presented here validates the choice of antigens that contain the PP mutation and suggests that deletion of the polybasic cleavage site could lead to a further optimized design.

Figure 1.. Spike construct design and protein characterization.

(A-D) described the wild type, ΔCS, PP and ΔCS-PP constructs used in this study. (B) shows the four antigens on a SDS-PAGE stained with Coomassie blue, while (C) shows the same protein on a Western blot developed with an antibody to the C-terminal hexahistidine tag. While all four proteins are detected as clean, single bands on the SDS-PAGE, the Western blot reveals a small fraction of degradation products at approximately 80 kDa for the wild type and PP variants and of approximately 40 kDa for the PP and ΔCS-PP constructs. (D) shows binding of mAb CR3022 to the constructs in ELISA. Data for the negative control mAb and the blank were combined for the different substrates.

Figure 2.. Immunogenicity of different spike variants in the mouse model.

(A) shows the vaccination regimen used for the five groups of mice and (B) shows the timeline. Animals were bled 3-weeks post prime (C) and 4 weeks post-boost (D) and antibody levels to a mammalian-cell expressed RBD were measured. Post-boost sera were also tested in cell-based ELISAs on cells infected with authentic SARS-CoV-2. Finally, post-boost sera were tested in a microneutralization assay against SARS-CoV-2.

Figure 3.. Challenge of mice with SARS-CoV-2.

Animals sensitized by transient expression of hACE2 via adenovirus transduction were challenged with 10⁵ PFU or SARS-CoV-2 and weight loss was monitored over a period of 14 days (A). (B) and (C) shows day 2 and day 5 lung titers respectively, while (D) shows lung immunohistochemistry staining for SARS-CoV-2 nucleoprotein on days 2 and 4 post challenge. Representative images from two animals each are shown at 5-fold magnification. Scale bar = 500 um.

Figure 4:. Lung pathology.

(A) shows a histopathological composite score for animals on day 2 post infection, (B) shows representative H&E stained tissue images from 2 animals per group. (C) and (D) show the same but for day 4 post challenge. Scale bar = 500 um.