Diverse array of neutralizing antibodies elicited upon Spike Ferritin Nanoparticle vaccination in rhesus macaques

Abstract

The repeat emergence of SARS-CoV-2 variants of concern (VoC) with decreased susceptibility to vaccine-elicited antibodies highlights the need to develop next-generation vaccine candidates that confer broad protection. Here we describe the antibody response induced by the SARS-CoV-2 Spike Ferritin Nanoparticle (SpFN) vaccine candidate adjuvanted with the Army Liposomal Formulation including QS21 (ALFQ) in non-human primates. By isolating and characterizing several monoclonal antibodies directed against the Spike Receptor Binding Domain (RBD), N-Terminal Domain (NTD), or the S2 Domain, we define the molecular recognition of vaccine-elicited cross-reactive monoclonal antibodies (mAbs) elicited by SpFN. We identify six neutralizing antibodies with broad sarbecovirus cross-reactivity that recapitulate serum polyclonal antibody responses. In particular, RBD mAb WRAIR-5001 binds to the conserved cryptic region with high affinity to sarbecovirus clades 1 and 2, including Omicron variants, while mAb WRAIR-5021 offers complete protection from B.1.617.2 (Delta) in a murine challenge study. Our data further highlight the ability of SpFN vaccination to stimulate cross-reactive B cells targeting conserved regions of the Spike with activity against SARS CoV-1 and SARS-CoV-2 variants.

Fig. 1. Isolation of SARS-CoV-2 neutralizing antibodies elicited by SpFN-vaccinated rhesus macaques.

a SpFN vaccination timeline. Animals received a 50 μg dose at week 0 and 4. PBMCs from week 6 (2 weeks post-boost) were used for B cell sorting and mAb isolation. b Reactivity of isolated WRAIR mAbs towards SpFN¹ (expressing SARS-CoV-1 S protein) and SpFN (expressing SARS-CoV-2 S protein) through flow cytometry as measured during sorting. c The proportion of isolated mAbs binding to H. pylori ferritin, subdomains of S protein or the stabilized S protein (HexaPro). d Binding to SARS-CoV-2 antigens or the stabilized S protein (HexaPro) in a multiplexed bead-based assay by twenty purified monoclonal antibodies. e Neutralization potency of isolated WRAIR mAbs segregated by S protein subdomain binding specificity. Shown are the IC₅₀ values (µg ml⁻¹) from the SARS-CoV-2 (IL1/2020) pseudotyped assay, calculated from three independent experiments. f Neutralization curves of the (left) RBD-directed and (right) NTD-directed neutralizing mAbs against SARS-CoV-2 (WA-1). Shown in parentheses are the IC₅₀ values for each mAb. g Neutralization of SARS-CoV-1 as measured in pseudotyped viral inhibition assays, with the IC₅₀ value for WRAIR-5001 shown in parentheses. Plotted are the mean ± s.e.m. from two independent experiments. h Assessment of RBD (n = 6), NTD (n = 5) and S2 (n = 3) -directed mAbs in Fc-mediated effector functions including antibody-dependent cell surface S binding (opsonization), cellular phagocytosis (ADCP) (from top to bottom: *P = 0.02, **P = 0.009, *P = 0.04), cell membrane transfer (trogocytosis), and cellular cytotoxicity (ADCC) (from top to bottom: *P = 0.02, *P = 0.04). Black horizontal lines indicate the mean value and asterisks represent significance by two-tailed Mann–Whitney t-test. The dotted line indicates the positivity threshold as determined by non-SARS-CoV-2 (negative) monoclonal antibody control (Zika mAb rhMZ134). Monoclonal antibody CR3022 is used as a positive control. Each data point is the mean of duplicate data from a single experiment. Source data are provided as a Source Data file. The rhesus macaque image was created with BioRender.com.

Fig. 2. Epitope binning of WRAIR NTD and RBD-targeted mAbs against variants of concern.

Epitope binning of a NTD-directed or b RBD-directed mAbs as measured through a BLI-based competition assay. Values are the percentage of residual binding of the indicated WRAIR second antibody after saturation of the antigen (NTD or RBD subdomain) with a representative first antibody (NTD: Group A: WRAIR-2025, Group B: WRAIR-2137, Group C: WRAIR-2054) (RBD: Group A: WRAIR-2125, Group B: WRAIR-2063, Group C: WRAIR-2151). Competition groups are indicated by boxes in shades of brown (NTD) or green/purple (RBD). (Right) Closed S trimer (PDB 6ZGE) with epitopes of the NTD-targeted mAb competition groups indicated in tan, light brown, and dark brown. Each protomer of the S trimer is colored cyan, light grey, or dark grey. (Inset a, center) The NTD is shown in surface representation, with residues identified for competition shown in respective shades of brown. Residue deletions observed in Omicron subvariants are highlighted in red. c Mapping of RBD A and B mAbs using alanine mutagenesis across the spike glycoprotein. d Residues identified by e viral escape assays are highlighted to show the targeted epitope of Group A and Group B RBD WRAIR mAbs. Residues of each group are shown in sphere representation on the SARS-CoV-2 RBD. Residues targeted by more than one antibody are highlighted. Source data are provided as a Source Data file.

Fig. 3. Select RBD mAbs demonstrate ACE2 binding inhibition, neutralize across SARS-CoV-2 variants of concern, and demonstrate protection in vivo.

ACE2 inhibition by WRAIR RBD mAbs in a BLI-based assay. RBD mAbs were assessed for their ability to block hACE2 binding to a SARS-CoV-2 RBD or b S-2P. The half-maximal effective concentration (EC₅₀) in µg ml⁻¹ is indicated in parentheses. Cross-reactivity of mAbs were assessed for activity against SARS-CoV-2 VoCs and SARS-CoV-1, using c BLI, or d IC₅₀ values measured against pseudotyped virus. e Binding kinetics of WRAIR-5001 with SARS-CoV-2 WA-1, Delta, Omicron BA.1, Omicron BA.4/5, and SARS-CoV-1 RBDs, as measured by BLI. Additional kinetics values may be found in Supplementary Table 2. f In vivo protection study plan and design of prophylactic administration of RBD mAbs. RBD mAbs WRAIR-5001 and WRAIR-5021 were given at single dose of 10 mg/kg intravenously in K18-hACE2 transgenic mice, followed by challenge with SARS-CoV-2 Delta (B.1.617.2) intranasally 24 hours later (n = 13 animals/group). 5 mice from each group were sacrificed on day 2 for plaque reduction neutralization test (PRNT) assay, and surviving mice were measured for body weight changes and survival out to day 14. RBD mAb WRAIR-2125 was used a positive control, and the negative control was an IgG isotype Zika-specific mAb, MZ4. g Particle forming units (PFU) measured in the lungs or bronchoalveolar lavage (BAL) of mice on study day 2 (n = 5 animals/group). Error bars indicate the standard deviation. (****P < 0.0001 using ordinary one-way ANOVA compared to IgG isotype control, MZ4) h Percent loss of body weight out to study day 10 and i survival curves out to study day 14 (study end date). (****P < 0.0001 using ordinary one-way ANOVA compared to IgG isotype control, MZ4). Source data are provided as a Source Data file. The mouse image was created with BioRender.com.

Fig. 4. Structure and epitope analysis of WRAIR-5021.

a (Left) Crystal structure of WRAIR-5021, in complex with SARS-CoV-2 RBD (white) shown in cartoon representation. WRAIR-5021 heavy and light chains are colored dark and light green, respectively (color scheme applies to all panels in Fig. 4). The ACE2 binding ridge is indicated by Ⓣ. (Right) Structure is shown at 90° rotation. b (Left) Overlay of SARS-CoV-2 RBD bound ACE2 structure (PDB: 6M0J) onto the WRAIR-5021-RBD complex structure. ACE2 is shown in light blue/grey surface. (Right) Epitope of WRAIR-5021 shown on the surface of the RBD. The ACE2 epitope is outlined in cyan. c Buried surface area (BSA) for the CDR loops is shown as a bar diagram. d (Left) Key antibody contacting residues of RBD are shown as sticks, with residues reported in VoCs in red. (Right) Important heavy and light chain contacting residues shown as thin sticks. RBD residues reported in VoCs are represented in red sticks. e Structure of the WRAIR-5001-RBD complex overlaid onto previously reported antibodies in complex with SARS-CoV-2 RBD (representing frequently observed SARS-CoV-2 epitopes). f (Left) Omicron mutations highlighted as red spheres on the surface of SARS-CoV-2 RBD. The WRAIR-5021 epitope is shown in tubular representation and colored dark green. Omicron mutations that fall within the mAb epitope are shown as green spheres and labeled. (Right) RBD sequence alignment with WRAIR-5021 epitope indicated. Mutated residues in VoCs are highlighted in red. BSA for epitope residues are shown in the bar graph at the bottom. g Structural superimposition of the WRAIR-5021-RBD complex with closed (all RBD down conformation, PDB code: 6ZGE) and open (1-RBD-up, PDB: 6X2B) conformations of SARS-CoV-2 S-2P. WRAIR-5021-RBD is overlaid onto the RBD (dark gray surface) from one protomer. Side and top views are shown. h Sequence alignment of WRAIR-5021 with its precursor germline gene. CDRs are shaded grey, with residue numbering and CDR loops designated using the Kabat system. Residues interacting with the RBD are colored green. Symbols *,:, and. denote identical, similar, and less similar residues, respectively. Source data are provided as a Source Data file.

Fig. 5. Structure and epitope analysis of WRAIR-5001.

a (Left) Crystal structure of WRAIR-5001, in complex with SARS-CoV-2 RBD (white) shown in cartoon representation. Heavy and light chains of WRAIR-5001 are colored dark and light pink, respectively (color scheme applies to all panels in Fig. 5). The ACE2 binding ridge is indicated by Ⓣ. (Right) Structure is shown at 180° rotation. b SARS-CoV-2 RBD shown in surface representation. c Overlay of the SARS-CoV-2 RBD bound ACE2 structure (PDB: 6M0J) onto the WRAIR-5001-RBD complex structure. ACE2 is shown in cartoon representation and colored light blue/grey. d Buried surface area (BSA) for the WRAIR-5001 heavy and light chain CDR loops shown as a bar diagram. e (Left) Key antibody contacting residues of RBD shown as sticks. (Right) Important heavy and light chain contacting residues of contributing CDRs shown as thin sticks and labeled as per antibody coloring scheme. f Structure of WRAIR-5001-RBD complex overlaid onto previously reported antibodies in complex with SARS-CoV-2 RBD (representing frequently observed SARS-CoV-2 epitopes). g (Left-Middle) Crystal structures of S309-RBD and WRAIR-2057-RBD complexes overlaid onto the WRAIR-5001-RBD structure. Antibodies S309 and WRAIR-2057 are shown in ribbon representation and colored yellow and red, respectively. (Right) The WRAIR-5001 epitope is colored light and dark pink on the surface of SARS-CoV-2 RBD while S309 and WRAIR-2057 epitopes are outlined and labeled accordingly. h Omicron mutations highlighted as red spheres on the surface of SARS-CoV-2 RBD. The WRAIR-5001 epitope is shown in tubular representation and colored dark pink. i, j Structural superimposition of the WRAIR-5001-RBD complex with closed (all RBD down conformation, PDB code: 6ZGE) and open (1-RBD-up, PDB: 6X2B) conformations of SARS-CoV-2 S-2P. WRAIR-5001-RBD is overlaid onto the RBD (dark gray surface) from one protomer. Side and top views are shown. k Sequence alignment of WRAIR-5001 with its precursor germline gene. CDRs are shaded grey, with residue numbering and CDR loops designated using the Kabat system. Residues interacting with the RBD are colored purple. Symbols *,:, and. denote identical, similar, and less similar residues, respectively. Source data are provided as a Source Data file.

Fig. 6. Epitope conservation and mAb cross-reactivity.

a Structural and sequence analysis of the WRAIR-5021 and WRAIR-5001 footprints across sarbecoviruses. (Right) The epitope residues are numbered according to the Wuhan reference sequence; the strength of the interaction between the mAb and the RBD is indicated by the height and color of the histogram bars above the sequence alignment. Sequences are ordered based on their phylogenetic relationships based on a maximum likelihood phylogenetic tree derived from the RBD protein sequences. (Left) The RBD structure is shown in surface representation and depicts mutations between SARS-CoV-1 and SARS-CoV-2 in red; the WRAIR-5021 and WRAIR-5001 epitopes are outlined and labeled. (Right) Sequence alignment of mAb epitopes across sarbecoviruses. b The binding of mAbs was measured with a set of sarbecovirus RBDs using BLI to assess cross-reactivity. Heat-map represents the relative binding strengths for WRAIR-mAbs. Coloring legend indicating the relative binding strength is shown on the right. c ACE2 blocking activity of WRAIR RBD mAbs in a BLI-based assay. mAbs were assessed for their ability to block human ACE2 binding to selected sarbecovirus clade 1b or 1a RBDs. Source data are provided as a Source Data file.