Shark nanobodies with potent SARS-CoV-2 neutralizing activity and broad sarbecovirus reactivity

Abstract

Despite rapid and ongoing vaccine and therapeutic development, SARS-CoV-2 continues to evolve and evade, presenting a need for next-generation diverse therapeutic modalities. Here we show that nurse sharks immunized with SARS-CoV-2 recombinant receptor binding domain (RBD), RBD-ferritin (RFN), or spike protein ferritin nanoparticle (SpFN) immunogens elicit a set of new antigen receptor antibody (IgNAR) molecules that target two non-overlapping conserved epitopes on the spike RBD. Representative shark antibody variable NAR-Fc chimeras (ShAbs) targeting either of the two epitopes mediate cell-effector functions, with high affinity to all SARS-CoV-2 viral variants of concern, including the divergent Omicron strains. The ShAbs potently cross-neutralize SARS-CoV-2 WA-1, Alpha, Beta, Delta, Omicron BA.1 and BA.5, and SARS-CoV-1 pseudoviruses, and confer protection against SARS-CoV-2 challenge in the K18-hACE2 transgenic mouse model. Structural definition of the RBD-ShAb01-ShAb02 complex enabled design and production of multi-specific nanobodies with enhanced neutralization capacity, and picomolar affinity to divergent sarbecovirus clade 1a, 1b and 2 RBD molecules. These shark nanobodies represent potent immunotherapeutics both for current use, and future sarbecovirus pandemic preparation.

Fig. 1. Induction of shark IgNAR antibody responses and isolation of antigen-specific VNARs.

a Immunization schedule of six nurse sharks for VNAR isolation. Sharks “Pink” and “Red” were immunized with SARS-CoV-2 RBD protein, followed by panning and identification of ShAb01 and ShAb02. Nurse sharks ‘Green’ and ‘Yellow’ were immunized with SARS-CoV-2 RBD-ferritin nanoparticle, followed by panning and identification of ShAb17-18 & 21 VNAR molecules. Nurse sharks ‘Purple’ and ‘Blue’ were immunized with SARS-CoV-2 Spike-ferritin nanoparticle, followed by panning and identification of ShAb09 and ShAb22-25 & 27–29 VNAR molecules. b Shark IgNAR peak plasma titers from all animals tested against SARS-CoV-2 RBD by ELISA. Pre-bleed samples serve as the negative control for each animal. c Shark IgNAR peak plasma titers from SpFN immunized animals (‘Purple’ and ‘Blue’) tested against SARS-CoV-2 S-2P by ELISA. Pre-bleed samples serve as the negative control for each animal. d Flow-chart depicting the steps involved from initial shark peripheral blood lymphocyte (PBL) isolation to ShAb protein expression. e ELISA of ShAb-Fc molecules against SARS-CoV-2 RBD (top) and S-2P trimer (bottom). Background binding to blocking agent, bovine serum albumin, has been subtracted from each data point. Source data are provided as a Source Data file. The shark image was created with BioRender.com.

Fig. 2. Antigenic characterization of the ShAb molecules.

a ShAb RBD-competition assay assessed by Bio-Layer Interferometry (BLI). BLI measurements are performed with immobilized SARS-CoV-2 RBD and ShAb in solution. ShAb01 (top) or ShAb02 (bottom) are bound to the RBD molecule followed by incubation with other ShAbs. b Epitope binning of the ShAb molecules. Values represent the % residual binding of the indicated second molecule (ShAb01, ShAb02, ACE2, or CR3022) after saturation of the antigen (WA-1 RBD) with the indicated first antibody (left column). Shading from dark to light indicates competition strength ranging from strong competition (0–33%), to reduced competition (> 50%). Competition groups are indicated by black boxes. Human ACE2-Fc and CR3022 were used as controls. c RBD kinetic binding constants, and neutralization IC₅₀ titers of SARS-CoV-2 viral VoC and SARS-CoV-1 pseudoviruses by ShAb01 and ShAb02. n/d indicates no binding was detected. d ShAb01 and ShAb02 binding assays performed by BLI with immobilized ShAbs and SARS-CoV-1 RBD and SARS-CoV-2 RBD, or clade 1b bat sarbecovirus trimeric S molecules in solution. Source data are provided as a Source Data file.

Fig. 3. Passive immunization with ShAb01 and ShAb02 protect mice from SARS-CoV-2 challenge.

a Schematic of K18-hACE2 mice SARS-CoV-2 challenge study. Mice (n = 13/group, 7 female, 6 male; 15/group for the isotype control, 8 female, 7 male) received an intraperitoneal injection of ShAb01 (blue), or ShAb02 (green), or IgG1 isotype control mAb (black), one day prior to challenge with 1.25 × 10⁴ PFU of SARS-CoV-2 virus (WA-1/2020). A cohort of mice (n = 5/group) were sacrificed 2 days post-challenge with the BAL fluid analyzed for viral load. The remaining mice (n = 8/group or 10 for the isotype control group) were assessed daily for weight and clinical symptoms. b Body weight measurements for K18-hACE2 mice. Percentage of initial weight is plotted. Isotype control mAb (black diamond), or ShAb01 (blue triangle) or ShAb02 (green circle). c Clinical score measurements of the K18-hACE2 study groups. All animals in the isotype control group were euthanized by study day 8. d Survival of K18-hACE2 mice for the 3 study groups. Key is shown in panel. Survival curves were compared individually to the isotype control using a Mantel-Cox log-rank test, and the significant P-value (P < 0.0001) is shown. n.s. indicates not significant differences. e SARS-CoV-2 viral loads in BAL were measured 2 days post-challenge in a subset of animals (n = 5/group) by plaque assay. Asterisks indicate significance compared to the antibody isotype control group by one-way ANOVA with Dunn’s multiple comparisons test, **P < 0.01. Source data are provided as a Source Data file. The mouse image was created with BioRender.com.

Fig. 4. Structural analysis of the interaction of ShAb01 and ShAb02 with SARS-CoV-2 RBD.

a Crystal structure of ShAb01 and ShAb02 VNARs in complex with SARS-CoV-2 RBD. The critical residues of ShAb01-RBD interaction are shown in two zoom-in panels, with extensive contacts made by the CDR3 loop. b Crystal structure of the ShAb01-ShAb02-RBD complex rotated from a. The critical residues of ShAb02-RBD interaction are shown in two zoom-in panels, with extensive contacts made by the VNAR CDR loops 1–3. c Epitope mapping by alanine mutagenesis scanning analysis identified SARS-CoV-2 RBD residue Y369 as a critical contact residue for ShAb01 binding. d Epitope mapping by alanine mutagenesis scanning analysis identified SARS-CoV-2 RBD residues R346, N354, and K356 as critical contact residues for ShAb02 binding. e Footprint of ACE2, ShAb01, and ShAb02 shown in multiple orientations. Overlay of the footprints are shown as an enclosed dashed line. SARS CoV-2 RBD is shown as white surface presentation; ACE2, ShAb01, and ShAb02 footprints on RBD are colored with salmon, green and slate respectively. f Footprint of ACE2, ShAb01 and ShAb02 footprints on SARS-CoV-2 WA-1 RBD. SARS-CoV-2 Omicron BA.1 and SARS-CoV-1 Urbani residue differences are highlighted in orange and purple, respectively.

Fig. 5. Crystal structure of ShAb01 and ShAb02 in the context of prefusion S.

a SARS-CoV-2 RBD (light gray) is shown in surface representation, with ShAb01, ShAb02, and 10 other nanobodies overlaid and shown in smooth surface representation. b Structure alignment of the ShAb01-ShAb02-RBD complex (center) to a partially open S-2P structure (1-RBD-up conformation, PDB ID:6VYB). Both ShAb01 and ShAb02 have full access to their respective epitopes in this structure. ACE2 (left) and CR3022 (right) are shown for reference. c The SARS-CoV-2 spike is displayed as in b with the RBD in wheat color and ShAb02 (green) aligned with (left), RBD-up conformation and (right) RBD-down conformation. The clash of ShAb02 and NTD in the RBD-down conformation is indicated by the red circle. d Structure of ShAb01-ShAb02-RBD complex (surface representation) overlaid on previously reported antibody-RBD complex structures (representing frequently observed antibody recognition classes). mAbs CB6, P2B-2F6, S309, and CR3022 are shown in cartoon representation and colored sand, pink, cyan, and orange respectively.

Fig. 6. Structure-based design of multi-domain ShAb molecules.

a Left, SARS-CoV-2 RBD is shown in surface representation with ShAb01 and ShAb02 VNARs in ribbon representation. The distance from the ShAb02 VNAR C-terminus to either the N- or C-terminus of ShAb01 is shown. Right, design schematics of multi-specific molecules BiShAb0201, and ShAb01H02K ‘knob-in-hole’ constructs, and trivalent ShAb-Foldon constructs. b BLI measurement of ACE2-inhibition to SARS-CoV-2 WA-1 RBD (center) and S-2P (right). The measurements are performed with immobilized SARS-CoV-1 RBD, with ShAb molecules and ACE2 in solution. c Neutralization of SARS-CoV-2 WA-1, Beta, Delta, Omicron BA.1, Omicron BA.5, and SARS-CoV-1 Urbani pseudoviruses by ShAb01, ShAb02, ShAb01H02K, BiShAb0201, ShAb01-Foldon and Shab02-Foldon. P2B-2F6 was used as a control antibody for SARS-CoV-2 strains except for Delta and Omicron BA.1 where COV2-2196 was used, and CR3022 was used for SARS-CoV-1. IC₅₀ values for pseudovirus neutralization, are shown in ng ml⁻¹, for each virus and ShAb or control mAb, with SD indicated by vertical lines. Source data are provided as a Source Data file.

Fig. 7. Fc-mediated effector functions by ShAb molecules.

a ADCP assay using SARS-CoV-2 WA-1 or SARS-CoV-1 Urbani S-2P-labelled beads incubated with phagocytic cells. ADCP scores measure the relative uptake of the bead-labeled proteins. Dotted black line indicates positive control (CR3022), and gray shaded area indicates positivity cutoff as determined by an irrelevant anti-Zika virus antibody (MZ4). b NK cell degranulation in the presence of ShAbs and CoV S-2P. The dotted black line indicates positive control (1:1 mixture of WRAIR-2039 and WRAIR-2123 antibodies for SARS-CoV-2, and CR3022 for SARS-CoV-1). The gray shaded area indicates positivity cutoff as determined by an irrelevant anti-Zika virus antibody (MZ4). Source data are provided as a Source Data file.

Fig. 8. ShAb molecules show activity against diverse sarbecoviruses.

a Epitope conservation and cross-reactivity analysis. Highly conserved RBD residues are colored in white, similar residues are colored in pink, and differing residues are colored in red. ShAb01 and ShAb02 epitopes are shown as a light blue and light green surface respectively. The epitope residues are numbered according to the Wuhan reference strain; the strength of the interaction between the ShAb and RBD is indicated by the height of the histogram bars above the sequence alignment (right). Sequences are ordered by phylogenetic relationships based on a maximum likelihood phylogenetic tree derived from RBD amino acid sequences. b BLI measurement of ShAb molecules binding to various sarbecovirus RBDs, (clade 1b (black), clade 1a (gray), clade 2 (gold) and clade 3 (blue)). Heat map shows area under curve (AUC) values. RBD molecules were immobilized to the probe with ShAb molecules in solution. c ACE2-inhibition by ShAbs to SARS-CoV-2 Omicron BA.1 RBD, SARS-CoV-1 RBD, and two ACE2-utilizing sarbecovirus RBD molecules. The measurements are performed with immobilized RBD, and with ShAbs and ACE2-Fc in solution. CR3022 was used as a control. Source data are provided as a Source Data file.