Heterologous Sarbecovirus Receptor Binding Domains as Scaffolds for SARS-CoV-2 Receptor Binding Motif Presentation

Figures

• 

  Figure 1. Selection of scaffolds and design. (A) Phylogenetic tree of 20 selected receptor binding domains used in this study; SARS-CoV and SARS-CoV-2 are included for reference. Design schematic of versions 1 (B), 2 (C), and 3 (D) resurfaced constructs, depicting RBM grafts (gray) on heterologous coronavirus scaffolds (indigo). Red dashed boxes indicate differences relative to version 1. (PDB 6M0J).

• 

  Figure 2. Resurfaced monomeric immunogen expression. (A) SDS-PAGE analysis of v2 monomers under nonreducing and reducing conditions. (B) Representative size exclusion trace with (*) marking the v2 monomeric constructs. Fractions in this peak were pooled for further characterization. Construct-specific protein yield for a 25 mL Expi293 transfection is shown in parentheses.

• 

  Figure 3. Resurfaced dimeric immunogen design and expression. (A) RBM-directed Fab B38 was used to confirm the conformational integrity of the SARS-2 RBM grafted onto rsYNLF-34C. FAB2G sensors were used with immobilized Fab to measure binding via BLI to rsYNLF-34C at 10, 5, 1, and 0.5 μM. Binding of CR3022 and S309 Fabs to rsYNLF-34C at 10 μM was also assayed, but no binding was detected. (B) SDS-PAGE analysis of dimers under nonreducing and reducing conditions. (C) Representative size exclusion trace with (*) marking the dimeric constructs. Fractions in this peak were pooled for use as immunogens. Expression volume of Expi293 cells and construct-specific protein yield are shown in parentheses. (D) Pairwise comparisons of amino acid sequence identity in the non-RBM portions of selected sarbecovirus RBDs. (E) Pairwise comparisons of amino acid sequence identity in the RBM portions of selected sarbecovirus RBDs. (F) Tm values were calculated from a Two-State Boltzmann Fit for the unfolding state of each construct using nanoDSF technique.

• 

  Figure 4. Antigenic distance investigation immunizations. Schematic of immunization regimens. Two immunization cohorts (n = 5 mice each) were primed with SARS-2 two-proline-stabilized (2P) spike protein on day -21 and then boosted with a cocktail of either the rsSARS-1 and rsWIV1 dimers (“AgDistlow” cohort) or the rsWIV1 and rsYNLF-34C dimers (“AgDisthigh” cohort) at days 0 and 21.

• 

  Figure 5. Antigenic distance investigation immunization serum reactivity. (A) Serum collected at day 35 was assayed in ELISA against different coronavirus antigens. (B) Human SARS-2 convalescent sera from 8 subjects were assayed in ELISA against the rsYNLF-34C homodimeric immunogen. (C) Pseudovirus neutralization was assayed against a panel of sarbecoviruses, ordered here based on spike amino acid sequence similarity to SARS-2. Statistical significance for (A, C) was determined using Kruskal–Wallis test with post hoc analysis using Dunn’s test corrected for multiple comparisons (*p < 0.05, **p < 0.01, ***p < 0.001); nonsignificant comparisons not marked.