A Fc-enhanced NTD-binding non-neutralizing antibody delays virus spread and synergizes with a nAb to protect mice from lethal SARS-CoV-2 infection

Abstract

Emerging evidence indicates that both neutralizing and Fc-mediated effector functions of antibodies contribute to protection against SARS-CoV-2. It is unclear whether Fc-effector functions alone can protect against SARS-CoV-2. Here, we isolated CV3-13, a non-neutralizing antibody, from a convalescent individual with potent Fc-mediated effector functions. The cryoelectron microscopy structure of CV3-13 in complex with the SARS-CoV-2 spike reveals that the antibody binds from a distinct angle of approach to an N-terminal domain (NTD) epitope that only partially overlaps with the NTD supersite recognized by neutralizing antibodies. CV3-13 does not alter the replication dynamics of SARS-CoV-2 in K18-hACE2 mice, but its Fc-enhanced version significantly delays virus spread, neuroinvasion, and death in prophylactic settings. Interestingly, the combination of Fc-enhanced non-neutralizing CV3-13 with Fc-compromised neutralizing CV3-25 completely protects mice from lethal SARS-CoV-2 infection. Altogether, our data demonstrate that efficient Fc-mediated effector functions can potently contribute to the in vivo efficacy of anti-SARS-CoV-2 antibodies.

Keywords: ADCC; COVID-19; Fc-effector functions; K18-hACE2 mice; SARS-CoV-2; coronavirus; non-neutralizing antibodies; spike; synergy.

Graphical abstract

Figure 1

Recognition of SARS-CoV-2 spikes by CV3-13 (A) Staining of CV3-13 (5 μg/mL) on the spike of the WT (Wuhan-Hu-1) or the B.1.1.7 (alpha variant) strain of SARS-CoV-2 expressed at the surface of 293T-transfected cells. (B) Staining of CV3-13 (5 μg/mL) on the different individual mutations of the spike of the B.1.1.7 strain of SARS-CoV-2 (D614G, Δ69-70, Δ144, N501Y, A570D, P681H, T716I, S982A, and D1118H). CV3-13 binding was further normalized to the binding obtained with the conformational-independent CV3-25 mAb (5 μg/mL). Statistical significance was evaluated using a non-parametric Mann-Whitney U test (^∗∗p < 0.01). Data are the average of the median of each experiment done at least two times. Mean values ± standard error of the mean (SEM). (C) Surface plasmon resonance-based epitope mapping reveals that CV3-13 explicitly binds to SARS-CoV-2 spike S1 subunit. Different viral antigens from SARS-CoV or SARS-CoV-2 were injected to the immobilized CV3-13 IgG (∼5,800 RU) at the indicated concentrations. (D) Kinetics measurement of CV3-13 Fab binding to the immobilized SARS-CoV-2 HexaPro spike (∼800 RU) with concentrations ranging from 3.125 to 200 nM (2-fold serial dilution). The experimental data (red) were fitted to a 1:1 Langmuir model (black) in BIA evaluation software.

Figure 2

CV3-13 is a non-nAb that has potent Fc-mediated effector functions (A) Neutralizing activity of CV3-13 WT, CV3-13 LALA, CV3-13 GASDALIE, and CV3-1 on SARS-CoV-2 spike D614G bearing pseudoviruses using 293T-ACE2 target cells. (B) Neutralizing activity of CV3-13 WT, CV3-13 LALA, CV3-13 GASDALIE, and CV3-1 on SARS-CoV-2 D614G authentic virus using Vero E6 target cells. The concentrations are the same as in (A). (C) Binding of CR3022, CV3-1, CV3-25, CV3-13 WT, CV3-13 GASDALIE, and CV3-13 LALA on the surface of Vero E6 cells infected with authentic SARS-CoV-2 virus 48 h post infection. Intracellular nucleocapsid (N) staining was done to separate the infected from the uninfected cells. (D) Binding of CV3-13 WT, CV3-13 LALA, CV3-13 GASDALIE. and CV3-1 on CEM.NKr.Spike cells. The non-specific staining obtained of CEM.NKr parental cells was subtracted from the staining on CEM.NKr.Spike cells. (E) % ADCC in the presence of CV3-13 WT, CV3-13 LALA, CV3-13 GASDALIE, and CV3-1 using a 1:1 ratio of parental CEM.NKr cells and CEM.NKr.Spike cells as target cells while PBMCs from uninfected donors were used as effector cells. (F) Percent ADCP mediated by CV3-13 WT, CV3-13 LALA and CV3-13 GASDALIE using CEM.NKr.Spike cells as target cells and THP-1 cells as phagocytic cells. (G) Indirect ELISA on SARS-CoV-2 spike 6P using CV3-13 WT, CV3-13 LALA, CV3-13 GASDALIE, CV3-1, CV3-25, and CR3022 (50 ng/mL). CR3022 was used as a positive control in each ELISA plate and for each experiment the data were further normalized on the signal obtained with this antibody. Data shown is the average of four independent experiments. (H) Staining with CV3-13 WT, CV3-13 LALA, CV3-13 GASDALIE, CV3-1, CV3-25, and CR3022 (5 μg/mL) of 293T cells transfected with the SARS-CoV-2 spike. CR3022 was used as a positive control in each experiment and for each experiment the data were further normalized on the signal obtained with this antibody. Data shown are the average of three independent experiments. Statistical significance was evaluated using a non-parametric Mann-Whitney U test (n.s., not significant). Mean FI, mean fluorescence intensity; Median FI, median fluorescence intensity; R.L.U., relative light units. The error bars indicate the SEM. Data are the average of at least two experiments for (A–F).

Figure 3

Cryo-EM structure of SARS-CoV-2 spike in complex with CV3-13 Fab (A) Side and top views of the NTD-targeting CV3-13 Fab and SARS-CoV-2 spike complex reveal a symmetrical Fab-spike assembly. In the left panel, protomer A of the spike colored in salmon is shown in ribbons while the other two protomers colored in green or gray and the variable regions (heavy chain, light yellow; light chain, cyan) of three CV3-13 Fabs binding to the lateral surface of the NTD are shown as cryo-EM density (C3 symmetry map). (B) The total buried surface area (BSA) at the Fab/NTD interface contributed by heavy/light chains of CV3-13 and 11 other structurally available NTD-directed mAbs, 4A8 (PDB: 7CL2), S2M28 (PDB: 7LY3), S2L28 (PDB: 7LXX), 5–24 (PDB: 7L2F), 4–18 (PDB: 7L2E), 4–8 (PDB: 7LQV), 1–87 (PDB: 7L2D), 2–51 (PDB: 7L2C), CM25 (PDB: 7M8J), FC05 (PDB: 7CWS), and DH1052 (PDB: 7LAB). The BSA values of all the equivalent biological assemblies were calculated and averaged by PISA (Krissinel and Henrick, 2007). (C) Expanded view of CV3-13 interactions with the protruding NTD loops. CV3-13 is shown as surface with the CDRs H1, H2, H3 L1, L2, and L3 colored in yellow, orange, gold, dark cyan, sky blue, and deep blue, respectively. The NTD regions are displayed as ribbons and the N1 to N5 loops as defined by Chi et al. (2020) are colored in green, purple, deep red, blue, and pink, respectively. Of note, the identified N2-3 hairpin is highlighted in light green. (D) CV3-13 epitope footprint on the electrostatic potential surface of NTD (colored red, blue, and white for negative, positive, and neutral electrostatic potential, respectively). The NTD residues interacting with CV3-13, which were defined as those with BSA > 0 Å² as calculated by PISA, are colored as in (C) in accordance with loop locations. The CV3-13 epitope footprints are outlined in green. The deleted Y¹⁴⁴ identified in the B.1.1.7 variant that disrupts CV3-13 binding is marked with a red box.

Figure 4

Structural basis of SARS-CoV-2 spike recognition by CV3-13 and NTD-directed nAbs (A) Two-view diagram of CV3-13 binding to the NTD lateral epitope as compared with 11 other NTD-targeting neutralizing mAbs and 3 other infectivity-enhancing mAbs. The NTD defined in this study is used here and shown as a light blue surface with the NTD supersite (residues 14–20, 140–158, and 245–264) highlighted in blue. On the left panel, the CV3-13 variable region is shown as a surface in light yellow for heavy chains and cyan for light chains. The approach of other NTD-binding mAbs are graphically represented by arrows (using the average Cα position of the variable region for individual NTD-targeting antibodies [Cα-_Fv] pointing toward the average Cα position of the NTD domains as a whole [Cα-_NTD] determined by individual NTD/spike-mAb complexes) in the indicated colors. On the right panel, the NTD is shown in a semi-transparent surface with NTD loops N1-N5 and the defined N2-3 hairpin shown as colored ribbons. The CV3-13 epitope boundary is shown with a green line. The Cα-_Fv of each NTD antibody is shown as colored spheres. (B) Angles of approach of CV3-13 and other NTD antibodies. The values for individual mAbs are calculated as the angle between Cα-_Fv and Cα-_NTD, with Cα_-NTD-Cter acting as the origin. (C) Diagram of NTD antibody epitopes reveal a distinct set of CV3-13 interacting NTD residues. The buried surface area (BSA) for the NTD residues contacting individual mAbs were calculated by PISA. (D) B factor representation of the CV3-13-bound NTD domain. (E) Orthogonal views of the NTD superimposition of reported EM/X-ray structures of NTD-binding mAbs in complex with SARS-CoV-2 spike or NTD. On the left panel, N1-N5 loops and the N2-3 hairpin of CV3-13-bound NTD are shown as colored ribbons with the NTD scaffold colored in yellow, while the other NTDs are uniformly depicted as gray ribbons. In the right panel, CV3-13-bound NTD is shown by yellow ribbons and the antigen NTD loops from other mAb-NTD structures are shown with colored ribbons with the scaffold colored in gray.

Figure 5

Prophylactic treatment with non-nAb CV3-13 does not protect K18-hACE2 mice from lethal SARS-CoV-2 infection (A) Experimental design for testing in vivo efficacy of non-nAb CV3-13 administered 1 day before challenging K18-hACE2 mice (i.n.) with SARS-CoV-2-nLuc followed by non-invasive BLI every 2 days. Human IgG1-treated (12.5 mg IgG/kg) mice were use as the isotype control (Iso). (B) Representative images from BLI of SARS-CoV-2-nLuc-infected mice in ventral (v) and dorsal (d) positions at the indicated dpi and after necropsy for experiment as in (A). (C) Ex vivo quantification of the nLuc signal as flux (photons/sec) after necropsy. (D and E) Temporal quantification of the nLuc signal as flux (photons/s) computed non-invasively in indicated areas of each animal. (F) Temporal changes in mouse body weight with initial body weight set to 100%. (G) Kaplan-Meier survival curves of mice statistically compared by log rank (Mantel-Cox) test for experiment as in (A). (H and I) Ex vivo imaging of organs and quantification of the nLuc signal as flux (photons/s) at the indicated dpi after necropsy. (J) Viral loads (nLuc activity/g) from the indicated organs using Vero E6 cells as targets. (K and L) Cytokine mRNA levels in lung and brain tissues after necropsy normalized to Gapdh in the same sample and that in uninfected mice. Viral loads (J) and inflammatory cytokine profile (K and L) were determined after necropsy for mice that succumbed to infection. Scale bars in (B) and (H) denote radiance (photons/s/cm²/steradian). Each curve in (D–F) and each data point in (C) and (I–L) represents an individual mouse. An equal number of female and male mice were used in each group. The data in (C) and (I–L) were analyzed by non-parametric Mann-Whitney test. Comparisons shown as “n.s.” or lacking ^∗ indicate non-significant statistical significance p > 0.05, Mean values ± SD are depicted.

Figure 6

Prophylactic treatment with CV3-13 GASDALIE delays SARS-CoV-2 neuroinvasion and dissemination in K18-hACE2 mice (A) Experimental design for testing virus dissemination in CV3-13 WT and CV3-13 GASDALIE-administered K18-hACE2 mice, 1 day before challenging (i.n.) with SARS-CoV-2-nLuc followed by non-invasive BLI every 2 days. Human IgG1-treated (12.5 mg IgG/kg) mice were use as the isotype control. (B) Representative images from BLI of SARS-CoV-2-nLuc-infected mice in ventral (v) as well as dorsal (d) positions and after necropsy at the indicated dpi for experiment as in (A). (C) Ex vivo quantification of the nLuc signal as flux (photons/s) at 4 dpi after necropsy. (D and E) Temporal quantification of the nLuc signal as flux (photons/s) computed non-invasively in indicated areas of each animal. (F) Temporal changes in mouse body weight with initial body weight set to 100%. (G and H) Ex vivo imaging of organs and quantification of the nLuc signal as flux (photons/s) at the indicated dpi after necropsy. (I) Viral loads (nLuc activity/g) from indicated organs using Vero E6 cells as targets. (J and K) Cytokine mRNA levels in lung and brain tissues after necropsy normalized to Gapdh in the same sample and that in uninfected mice. Scale bars in (B) and (G) denote radiance (photons/s/cm²/steradian). Each curve in (D–F) and each data point in (C) and (H–J) represents an individual mouse. An equal number of female and male mice were used in each group. The group data in (C) and (H–K) were analyzed by two-way ANOVA followed by Tukey's multiple comparison tests. Statistical significance for group comparisons to control are shown in black, CV3-13 WT in red and CV3-13 GASDALIE in green. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001. Comparisons shown as “ns” or lacking ^∗ indicate non-significant statistical significance p > 0.05; mean values ± SD are depicted.

Figure 7

Fc-enhanced CV3-13 in combination with Fc-compromised nAb CV3-25 protects K18-hACE2 mice from lethal SARS-CoV-2 infection (A) Experimental design for testing in vivo efficacy of indicated antibodies or combinations administered 1 day before challenging K18-hACE2 mice (i.n.) with SARS-CoV-2-nLuc followed by non-invasive BLI every 2 days. Human IgG (Iso) or HIV-1 co-receptor site binding 17b antibody carrying GASDALIE mutations (12.5 mg IgG/kg) were used as controls. (B) Representative images from BLI of SARS-CoV-2-nLuc-infected mice in ventral (v) and dorsal (d) positions at the indicated dpi and after necropsy for experiment as in (A). (C and D) Temporal quantification of the nLuc signal as flux (photons/s) computed non-invasively in indicated areas of each animal. (E) Temporal changes in mouse body weight with initial body weight set to 100%. “†” denote mice that succumbed to infection. (F) Kaplan-Meier survival curves of mice statistically compared by log rank (Mantel-Cox) test for experiment as in (A). (G) Ex vivo imaging of indicated organs after necropsy. (H) Viral loads (nLuc activity/g) from indicated organs using Vero E6 cells as targets. (I and J) Cytokine mRNA levels in lung and brain tissues after necropsy normalized to Gapdh in the same sample and that in uninfected mice. Viral loads (H) and inflammatory cytokine profile (I and J) were determined after necropsy for mice when they succumbed to infection. Scale bars in (B) and (G) denote radiance (photons/s/cm²/steradian). Each curve in (C–E) and each data point in (H–J) represents an individual mouse. An equal number of female and male mice were used in each group. The group data in (C–E) and (H–J) were analyzed by two-way ANOVA followed by Tukey's multiple comparison tests. Statistical significance for group comparisons to control are shown in black, 17b GASDALIE in blue, CV3-13 WT in green, CV3-13 GASDALIE in red, CV3-25 LALA in peach, and combination of CV3-25 LALA + CV3-13 GASDALIE in pink. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001. Comparisons shown as “ns” or lacking ^∗ indicate non-significant statistical comparisons p > 0.05; mean values ± SD are depicted.