An infectivity-enhancing site on the SARS-CoV-2 spike protein targeted by antibodies

Abstract

Antibodies against the receptor-binding domain (RBD) of the SARS-CoV-2 spike protein prevent SARS-CoV-2 infection. However, the effects of antibodies against other spike protein domains are largely unknown. Here, we screened a series of anti-spike monoclonal antibodies from coronavirus disease 2019 (COVID-19) patients and found that some of antibodies against the N-terminal domain (NTD) induced the open conformation of RBD and thus enhanced the binding capacity of the spike protein to ACE2 and infectivity of SARS-CoV-2. Mutational analysis revealed that all of the infectivity-enhancing antibodies recognized a specific site on the NTD. Structural analysis demonstrated that all infectivity-enhancing antibodies bound to NTD in a similar manner. The antibodies against this infectivity-enhancing site were detected at high levels in severe patients. Moreover, we identified antibodies against the infectivity-enhancing site in uninfected donors, albeit at a lower frequency. These findings demonstrate that not only neutralizing antibodies but also enhancing antibodies are produced during SARS-CoV-2 infection.

Keywords: ADE; angiotensin converting enzyme 2; antibody-dependent enhancement; cryo-EM; cryo-electron microscopy; docking model.

Graphical abstract

Figure 1

Effect of anti-spike antibodies on ACE2 binding (A) HEK293T cells transfected with vectors expressing NTD-TM, RBD-TM, and S2-TM were stained with anti-spike antibodies. The effect of antibodies on ACE2-binding to spike transfectants was analyzed with ACE2-Fc-fusion protein. (B) Mean fluorescence intensities (MFIs) of the stained cells are shown (top three columns). The binding of ACE2-Fc-fusion protein to full-length spike transfectants was analyzed in the presence of the indicated antibodies at 1 μg/mL (bottom column). Antibodies that enhanced binding of ACE2-Fc to the spike transfectants by more than 1.9 times are indicated in red. See also Figures S1 and S2.

Figure S1

Specificities of the antibodies used in this study, related to Figures 1 and 3 (A) The expression constructs used to analyze the specificity of the anti-spike antibodies. (B) The plasmids encoding the full-length spike, Flag-NTD-TM, Flag-RBD-TM, and Flag-S2-TM were cotransfected separately with a GFP vector into HEK293T cells. The transfectants were then stained with the indicated antibodies. The antibodies bound to the transfectants were detected with APC-labeled secondary antibodies. The fluoresce intensities of APC on the GFP-expressing cells are shown (red line). Control stainings were shown as shaded histogram. (C) The plasmids expressing the wild-type spike protein and the D614G mutant were cotransfected separately with a GFP vector into HEK293T cells. The transfectants were stained with the anti-NTD infectivity-enhancing antibody COV2-2490 and anti-NTD non-enhancing antibody 4A8. The fluorescence intensities of APC on the GFP-expressing cells are shown (red line). Control stainings were shown as shaded histogram. (D) Parental HEK293T cells and ACE2-transfected HEK293T cells were stained with anti-ACE2 mAb (red line). Control stainings were shown as shaded histogram.

Figure S2

Enhanced binding of ACE2 to the spike protein by specific anti-NTD antibodies, related to Figures 1 and 2 (A) The plasmids expressing the full-length spike, NTD-TM, RBD-TM, and mock were transfected into HEK293T cells with the GFP vector, and the transfectants were mixed with the indicated anti-NTD antibodies at 10 μg/ml. 4A8 is a non-enhancing antibody and the remaining is enhancing antibodies. Transfectants not mixed with antibodies were used as a control (shaded histogram). Afterward, the cells were stained with biotin-labeled ACE2-Fc fusion protein, followed by APC-labeled streptavidin. The fluorescence intensities of APC on the GFP-expressing cells are shown (red line). Mean fluorescent intensities (MFI) of red lines were shown in the figure. (B) HEK293T cells transfected with spike protein were mixed with the indicated concentrations of enhancing antibodies. Subsequently, APC-conjugated anti-human IgG Fc antibody (blue) or biotin-ACE2-Fc fusion protein, followed by APC-labeled streptavidin (red), were mixed with the transfectants. The mean fluorescence intensities of cells stained in the absence of the enhancing antibody were set to 0, and the maximum fluorescence intensities in the presence of the enhancing antibody were set to 100. EC₅₀ was calculated from the concentrations of the enhancing antibody that induced half the maximum fluorescence intensities. (C) EC₅₀s of antibody binding to the spike transfectants and ACE2 binding enhancement by the antibodies were shown. (D) ACE2-Fc binding to wild-type spike protein in the presence of COV2-2490 antibodies at 3 μg/ml and various concentrations of anti-RBD neutralizing antibodies indicated in the figure (red line). ACE2-Fc binding in the absence of the enhancing antibodies was shown as the control (black line). (E) Neutralizing C144 Ab was added to the spike transfectants first and then an enhancing antibody COV2-2490 was added. Concentrations of antibodies were indicated in the figure. ACE2-Fc binding to wild-type spike protein was shown. The data are presented as mean ± SD. The representative data from three independent experiments are shown.

Figure 2

Enhanced ACE2 binding to spike protein by anti-NTD antibodies (A) ACE2-Fc binding to the spike transfectants in the various concentrations of antibodies. (B) ACE2-Fc binding to the wild-type or D614G spike protein in the presence of 3 μg/mL COV2-2490 monoclonal antibody (mAb). The statistical significance derived from an unpaired t test is indicated. (C) ACE2-Fc binding to wild-type spike protein in the presence of the indicated antibodies at 3 μg/mL and various concentrations of anti-RBD neutralizing antibody C144 (red line). ACE2-Fc binding in the absence of the enhancing antibodies was shown as the control (black line). The data from triplicates are presented as mean ± SD. Representative data from three independent experiments are shown. See also Figure S2.

Figure 3

Enhanced SARS-CoV-2 infectivity by specific anti-NTD antibodies. (A) ACE2-expressing HEK293T cells (MOI: 0.3) were infected by a SARS-CoV-2 spike pseudovirus carrying a GFP reporter gene in the presence of various concentrations of indicated antibodies. The proportion of GFP-positive infected cells is shown. (B) ACE2-expressing HEK293T cells were infected with a SARS-CoV-2 spike pseudovirus carrying a GFP reporter gene at different MOIs with (red line, 3 μg/mL) or without (black line) the indicated antibodies. (C) HEK293T cells, ACE2-expressing HEK293T cells, and Huh7 cells were infected with authentic SARS-CoV-2 virus in the presence (+) or absence (−) of enhancing antibody COV2-2490 at 1 μg/mL. The amounts of SARS-CoV-2 virus produced in the cell culture supernatants were analyzed 48 h after infection. The statistical significance derived from an unpaired t test is indicated. NS, not significant. The data are presented as mean ± SD. Representative data from three independent experiments are shown. See also Figure S1.

Figure 4

Epitope mapping of SARS-CoV-2 infectivity-enhancing antibodies (A) Binding of enhancing antibodies (binder) to full-length spike transfectants was analyzed in the presence of the indicated antibodies (competitor). The effect of competitors on ACE2-Fc binding to the spike transfectants was also analyzed. The non-enhancing anti-S2 antibody, COV2-2147, was used as a control. Relative antibody- or ACE2-binding levels observed in the presence of competitor are shown. (B) Relative antibody-binding levels to a series of NTD mutants compared to wild-type NTD are shown. Non-enhancing anti-NTD antibody 4A8 was used as a control. The most affected residues are shown as red. (C) Full-length mutant spike proteins were stained with the indicated enhancing antibodies (red line). Staining of wild-type spike is shown as a shaded histogram. See also Figure S3.

Figure S3

Binding of anti-spike antibodies against spike mutants with mutated antibody epitopes in the SARS-CoV-2 infectivity site, related to Figure 4 (A) The plasmids expressing the full-length spike proteins with alanine mutations at the indicated amino acid residues were transfected separately with the GFP vector into HEK293T cells, and the binding of 4A8 (anti-NTD non-enhancing antibody), C144 (anti-RBD antibody), and COV2-2454 (anti-S2 antibody) against the GFP-positive cells were analyzed (red line). Control stainings were shown as shaded histogram. (B) The relative binding of each enhancing antibody to the B.1.1.7 spike transfectants was compared to the wild-type spike transfectants. (C) The relative bindings of ACE2-Fc to the transfectants of wild-type or B.1.1.7 spike protein were analyzed in the presence or absence of 10 μg/ml of enhancing antibodies. The data are presented as mean ± SD.

Figure 5

Structures of SARS-CoV-2 infectivity-enhancing antibodies bound to spike protein (A) Amino acid residues that affected the binding of each enhancing antibody are shown as a heatmap based on their percent reduction of the MFIs in Figure 4B, with higher reduction indicated by darker shades, as shown in bar in the figure. NTD, dark gray; RBD, medium gray; other regions, light gray. (B) MFI reductions of the affected residues are averaged across the six antibodies and shown as a heatmap, with higher reduction indicated by darker shades, as shown in bar in the figure. (C) Each SARS-CoV-2 infectivity-enhancing antibody was docked onto the spike protein as described in STAR Methods. (D) A spike model built from PDB: 7KEB was found to fit the apo density best, while another model built from PDB: 7K8W fit the two other antibody-bound densities best. The spike subunit in the one RBD-“up” form is colored green. Antibodies are colored pink (heavy chain) and blue (light chain). The scale bar represents 30 Å. See also Figure S4 and Table S1.

Figure S4

Cryo-EM density map of spike of SARS-CoV-2 with antibody, related to Figure 5 (A-C) A representative micrographs (left), CTF estimation of a micrograph on left panel (middle), and typical 2D class averages. (A) Apo spike protein of SARS-CoV-2. (B) Spike protein with 8D2 antibody. (C) Spike protein with COV2-2490 antibody. The final density map from single particle cryo-EM is colored by local resolution (A-H). (D) Apo spike protein (EMDBID: 30915). (E-I) Spike protein with 8D2 antibody (EMDBID: 30916, 30917, 30918, 30919, and 30920). (J-K) Spike protein with 2490 antibody (EMDBID: 30921 and 30922). Asterisks indicate the rise form of RBD domains. Triangles indicate the binding positions of antibody. (L-N) The GS-FSC curves for each corresponding map are shown. Red line indicates FSC = 0.143 criteria. Scale bars are 30 Å.

Figure 6

Induction of the open conformation of the RBD by divalent enhancing antibodies (A) The binding of antibodies specific to the open RBD (H014) or the open and closed RBD (C009, C144, and C135) against spike transfectants was analyzed in the presence of the indicated enhancing antibodies (3 μg/mL). ACE2 binding to spike protein was also analyzed. The non-enhancing anti-NTD antibodies 4A8 and 4A2 were used as controls. Relative antibody- or ACE2-binding levels observed in the presence of anti-NTD antibodies are shown. (B) H014 antibody binding and ACE2 binding to spike protein in the presence of indicated antibodies are shown. The correlation coefficient between them is indicated. (C) H014 antibody binding and ACE2 binding to spike protein in the presence of various concentrations of Fab and F(ab′)₂ and fragments of the enhancing COV2-2490 antibody. (D) A model of spike protein bound with divalent enhancing antibodies (8D2 and COV2-2490). Spike protein and Fab complexes demonstrated in Figure 5D were used as a template for the models. See also Figure S5.

Figure S5

Induction of the open conformation of the RBD by divalent enhancing antibodies, related to Figure 6 (A) The binding of antibodies specific to open RBD (H014) or open and closed RBD (C009, C144, C135) against spike transfectants was analyzed in the presence or absence of enhancing (COV2-2490) or non-enhancing (4A8) antibodies (3 ug/ml). (B) SDS-PAGE analysis of whole IgG antibody, F(ab’)₂ and Fab fragments of enhancing COV2-2490 antibody. (C) Whole spike transfectants were stained with whole IgG antibody, F(ab’)₂ or Fab fragment of COV2-2490 antibody. Bound antibodies were detected by APC-labeled anti-human IgG Fc or anti-human IgG Fab specific antibodies.

Figure 7

SARS-CoV-2 infectivity-enhancing antibodies in COVID-19 patients and uninfected individuals (A) A method of detecting the enhancing or neutralizing antibodies using a competitive binding assay. DyLight-650-labeled 8D2 and C144 were used to detect the enhancing and neutralizing antibodies, respectively. (B) The binding of the enhancing antibody, 8D2, or the neutralizing antibody, C144, to beads coated with the spike protein was analyzed in the presence of the serially diluted serum of a representative COVID-19 patient or an uninfected donor. (C) The levels of SARS-CoV-2 infectivity-enhancing antibodies and neutralizing antibodies in noninfected individuals and non-severe and severe COVID-19 patients. (D) Enhancing antibody titers subtracted with neutralizing antibody titers were compared between non-severe and severe COVID-19 patients. Statistical significance derived from Mann-Whitney U test is indicated. (E) Enhancing antibodies were detected by comparing antibody binding to the wild-type NTD-TM (WT) and the mutant NTD-TM lacking the enhancing antibody epitopes (Mut). (F) Serum levels of antibodies in uninfected individuals against the wild-type NTD (blue bar) and mutant NTDs whose epitopes for the enhancing antibodies were mutated (red bar). (G) SARS-CoV-2 infectivity-enhancing antibody titers were calculated by subtracting antibody levels against the mutant NTD from those against the wild-type NTD in uninfected individuals (red bar). Anti-RBD antibody titers were analyzed using RBD-TM transfectants (blue bar). See also Figure S6.

Figure S6

Specificity of competitive binding assay to detect enhancing and neutralizing antibodies, related to Figure 7 (A) Specificity of 8D2 and C144 antibody competitive binding assay. Each antibody derived from COVID-19 patients used in Figure 1B was mixed with spike protein transfectants, and DyLight 647-labeled 8D2 or C144 antibody binding as well as ACE2-Fc binding was analyzed. Anti-NTD and S2 antibodies were plotted as black dots and anti-RBD antibodies were plotted as blue dots. Approximate line and correlation coefficient between ACE2 binding and 8D2 binding by anti-NTD and S2 antibodies were shown in the figure. (B) Comparison of competitive binding assay using C144 antibody with that using C009 and C135 antibody. Spike protein transfectants were mixed with serially diluted COVID-19 patient sera followed by the staining with DyLight 647-labeled C009, C144 and C135 anti-RBD antibodies recognizing different epitopes. Relative fluorescent intensities of the stained cells were shown. The data are presented as mean ± SD.