A nanobody recognizes a unique conserved epitope and potently neutralizes SARS-CoV-2 omicron variants

Abstract

The severe acute respiratory syndrome coronavirus 2 (SARS-CoV2) Omicron variant sub-lineages spread rapidly worldwide, mostly due to their immune-evasive properties. This has put a significant part of the population at risk for severe disease and underscores the need for effective anti-SARS-CoV-2 agents against emergent strains in vulnerable patients. Camelid nanobodies are attractive therapeutic candidates due to their high stability, ease of large-scale production, and potential for delivery via inhalation. Here, we characterize the receptor binding domain (RBD)-specific nanobody W25 and show superior neutralization activity toward Omicron sub-lineages in comparison to all other SARS-CoV2 variants. Structure analysis of W25 in complex with the SARS-CoV2 spike glycoprotein shows that W25 engages an RBD epitope not covered by any of the antibodies previously approved for emergency use. In vivo evaluation of W25 prophylactic and therapeutic treatments across multiple SARS-CoV-2 variant infection models, together with W25 biodistribution analysis in mice, demonstrates favorable pre-clinical properties. Together, these data endorse W25 for further clinical development.

Keywords: Decision science; Information system model; Public health.

Graphical abstract

Figure 1

Authentic virus neutralization of SARS-CoV-2 variants by W25, EUA, and other mAbs (A) Neutralization curves comparing the sensitivity of SARS-CoV-2 strains Wu (gray line) and Omicron BA.1 (red line) and Omicron BA.2 (blue line) to antibody as indicated including W25-Fc, C05, CB6, REGN10933, CT-P59, LY-CoV555, S309, REGN10987, and DH1047. The data were analyzed and plotted using nonlinear regression (curve fit, three parameters). (B and C) Comparison of IC₅₀ values across mAbs tested between Wu and Omicron. IC₅₀ was calculated from the neutralization curve by Graphpad Prism 8 software. Color represents different mAbs as indicated. Data are generated from two independent experiments, each performed in technical triplicate. Data are represented as mean ± SEM.

Figure 2

Structure analysis of the W25/spike interaction (A) Cryo-EM densities of Wu (upper panel) and Omicron (lower panel) spike/W25 complexes (gray semi-transparent surfaces) (map 1, Figure S1 and map 4, Figure S3). Fitted trimeric SARS-COV2 spike protein structures (PDB: 6ZXN (Wu) and PBD: 7WG6 (Omicron)^, are shown in cartoon representation, with protomer A colored purple, protomer B red, and protomer C cyan. W25 is shown as green cartoon. Densities corresponding to W25 are colored green. (B) Three views of a focused refinement of the regions illustrated in A. Proteins are shown in cartoon representation, and N-linked glycans as sticks, in the same colors as in A. Cryo-EM densities (map 3, Figure S1, and map 5, Figure S4) are shown as semi-transparent light-gray surfaces. (C) Detailed “open book” view of the W25/RBD interaction interface between RBD and W25. Amino acids involved in intermolecular interaction are shown as sticks. (D) Visualization of the RBD surface in contact with W25 (yellow). The RBD is shown as red cartoon. Residues mutated in the Omicron RBD are shown as spheres, and colored light blue when located outside of the W25 contact surface, or dark blue when inside. (E) Two views of a superposition of selected RBD-bound NABs with the RBD/W25 complex, in cartoon representation (CB6 – PDB: 7C01, LY-CoV555 – PDB: 7l3N, S309 – PDB: 7R6W, CR3022 – PDB: 6YLA). W25 is shown as semi-transparent green molecular surface. For clarity, only the W25-bound RBD is shown as red cartoon. The inset shows an enlarged view of the unique portion of the W25 epitope as yellow surface.

Figure 3

Broad reactivity of W25-Fc is conferred by conserved patch on SARS-CoV-2 RBD and potential inhibition mechanism of W25 (A) SPR sensograms of W25-Fc. W25-Fc was immobilized on SPR protein A chips. Various concentrations of SARS-CoV-2 spike variant proteins as indicated were injected for 30 s, at 30 μL/min followed by dissociation for 600 s. Dissociation constants (K_D) were determined on the basis of fits, applying a 1:1 interaction model. Similar experiments were conducted for S309, DH1047, and C05 (as control, Figure S6). (B) Summary table showing K_D, K_a,K_d of indicated Nb/mAbs. (C) Kd values from ELISA binding curves of W25-Fc, EUA mAbs, and other epitope-specific mAbs to SARS-CoV-2 spikes variants. (D) Neutralization curves comparing the sensitivity of live SARS-CoV-2 viruses including Alpha, Beta, Delta, Kappa, Lambda, and Gamma to W25-Fc, EUA, and other RBD-specific mAbs as indicated. The data were analyzed and plotted using nonlinear regression (curve fit, three parameters), and IC₅₀ value was calculated from neutralization curves by Graphpad Prism 8 software. (E) Summary of IC₅₀ values (nM) of neutralistion of SARS-CoV-2 variants performed in Vero E6 cells. Values that approached 50% neutralization were estimated from (D). Data are generated from two independent experiments, each performed in technical duplicate. (F) Molecular surface conservation of RBD VOCs. (G) W25 and its derivatives enhance spike-mediated cell fusion. Cell-cell fusion assay was performed. Stable cells expressing the rLuc-GFP components (effector cells) were transiently transfected SARS-CoV-2 spike proteins bearing spike mutations for D614G (top left), Beta (top right), Omicron (Bottom left), or control plasmid (no envelope control). W25 Nb, W25 Dimeric Fc (W25-Fc), or W25 monomeric Fc (W25-FcM) were incubated with effector cells for 1 h prior to co-culturing with target cells (huACE2-HEK293T cells). After 24–48 h, Renilla luciferase was read and analyzed by subtraction with control plasmid treatment. Data are generated from at least three biological replicates. Data are represented as mean ± SEM.

Figure 4

W25-Fc protects mice from lethal Beta and Omicron SARS-CoV-2 infection and W25-Fc pharmacodynamic analysis (A) Experimental schematic: eight- to twelve-week-old male and female K18 transgenic mice were inoculated via the intranasal route with 1 × 10³ PFU of SARS-CoV-2 (Isolate B.1.351). W25-Fc or GFP Nb-Fc were administered intraperitoneally 4 h prior to infection. (B) Survival and (C) weight change were monitored and scored. Two experiments were performed (n = 9–12, Log rank (Mantel-Cox) test ∗∗∗∗p < 0.0001). (D) Viral burdens were determined in lung tissues 4 day post infection via plaque forming assays for infectious virus and (E) RT-qPCR for viral genome copy number. Two experiments were performed (N = 10–11). (F) Experimental schematic: five- to six-week-old female K18 transgenic mice were inoculated via the intranasal route with 1 × 10⁴ FFU of SARS-CoV-2 (Beta, Isolate B.1.351) N = 8, and N = 4–5 each group for survival and tissue harvest experiment, respectively. For therapeutic treatment, W25-Fc or GFP-Nb-Fc were administered intraperitoneally 24 h after SARS-CoV-2 Beta infection at 5 × 10³ FFU/mouse. (G) Survival and (H) weight change were monitored. Viral loaded were determined in (I) nasal turbinate and (J) lung tissues 2 day post infection by immuoplaque assays for infectious virus. (K) Experimental schematic: for Omicron infection, five- to six-week-old female K18 transgenic mice were inoculated via the intranasal route with 1 × 10⁵ FFU of SARS-CoV-2 (Isolate B.1.1.529). W25-Fc or GFP-Nb-Fc were administered intraperitoneally or intranasally 24 h after infection. (L and P) Survival and (M and Q) weight changes were monitored and scored. Viral loaded were determined in (N and R) nasal turbinate and (O and S) lung tissues 2 day post infection by immunoplaque assays for infectious virus. Bars represent medians, dots are individual animals, and dotted horizontal lines indicate the limit of detection. Data are represented as mean ± SEM. (T) The pharmacodynamics of W25-Fc in blood was determined by conjugation of W25-Fc to radioactive ¹¹¹Indium. One mg/kg were injected intravenously through the tail to 6 groups of mice (group 1: 5 min, group 2: 20 min, group 3: 60 min, group 4: 3 h, group 5: 5 h and group 6: 24 h). The mice were dissected, and concentration of ¹¹¹In W25-Fc was measured in the blood using an Auto-Gamma Counter.