Structural definition of a neutralization epitope on the N-terminal domain of MERS-CoV spike glycoprotein

Abstract

Most neutralizing antibodies against Middle East respiratory syndrome coronavirus (MERS-CoV) target the receptor-binding domain (RBD) of the spike glycoprotein and block its binding to the cellular receptor dipeptidyl peptidase 4 (DPP4). The epitopes and mechanisms of mAbs targeting non-RBD regions have not been well characterized yet. Here we report the monoclonal antibody 7D10 that binds to the N-terminal domain (NTD) of the spike glycoprotein and inhibits the cell entry of MERS-CoV with high potency. Structure determination and mutagenesis experiments reveal the epitope and critical residues on the NTD for 7D10 binding and neutralization. Further experiments indicate that the neutralization by 7D10 is not solely dependent on the inhibition of DPP4 binding, but also acts after viral cell attachment, inhibiting the pre-fusion to post-fusion conformational change of the spike. These properties give 7D10 a wide neutralization breadth and help explain its synergistic effects with several RBD-targeting antibodies.

Fig. 1

7D10 binding specificity and neutralization potency. Recombinant NTD or RBD of MERS-CoV S glycoprotein at 1 μg mL⁻¹ were used to coat plates overnight at 4 °C, and each of the mAbs including 7D10, an antibody anti-RBD (MERS-GD27) and an unrelated antibody (3C11) were serially diluted in PBS and assessed for binding affinity and specificity to the NTD (a) and RBD (b). c Neutralization of the pseudotyped MERS-CoV. DPP4-expressing Huh7 cells were cultured with 200 TCID₅₀ of pseudotyped MERS-CoV in the presence of serially diluted mAbs. The neutralization percentage was calculated by measuring luciferase expression compared to the pseudotyped virus-infected cell control. d Neutralization of live MERS-CoV. Different concentrations of the mAbs were pre-cultured with the live MERS-CoV (EMC strain) in Vero E6 cell monolayers. The neutralization percentage was evaluated by calculating the decrease in plaque number compared with the virus-infected control. Data are shown as mean ± SD. e PFU images of viral infection in the presence of the mAbs on day 3. The images correspond to the neutralizing percentages in (d). Approximately, 30–35 PFU virus stocks (EMC strain) were used to infect Vero E6 cells in a 12-well plate with or without mAbs. MERS-GD27, 3C11 mAbs, and PBS were used as the positive, unrelated, and blank controls, respectively. Source data are provided as a Source Data file

Fig. 2

Crystal structure of 7D10-scFv bound to NTD and the binding interface. a An overall structure of the NTD/7D10-scFv complex in which the NTD, N222-linked glycans on the NTD, 7D10 V_L, and 7D10 V_H are colored in blue, gray, magenta, and cyan, respectively. b Epitope on the NTD recognized by 7D10. The NTD is represented as blue surface, on which the protein region bound by 7D10 is displayed in orange and the N222-linked glycans are displayed as gray sticks. c 7D10 residues that are involved in the binding. The V_L and V_H are colored in magenta and cyan, respectively, and the residues interacting with 7D10 are displayed in orange. d Interactions between the 7D10 V_H residues and the corresponding residues of NTD. e Interactions between the 7D10 V_L residues and the corresponding residues of NTD. f Zoom-in view of interactions between N222-linked glycans and 7D10

Fig. 3

Breadth of 7D10-H neutralization. a Neutralizing analysis of 7D10-H against MERS-CoV wild-type (WT) and its variant mutants; site-directed mutations were introduced into the EMC strain to create 22 variant mutants according to natural mutations of MERS-CoV S. b Summary of 7D10-H mediated inhibition of infection by all pseudotyped viruses. IC₅₀ neutralization titers for mutant EMC S variants are presented relative to wild-type S (set to 1). Source data are provided as a Source Data file

Fig. 4

Combination effects for 7D10-H with RBD-specific mAbs. a Effects of 7D10-H combined with MERS-GD27 in neutralizing pseudotyped MERS-CoV. Percent neutralization was calculated for serial threefold dilutions of each antibody alone and in combination at constant ratios in a range of concentrations from 27 times to 1/81 of IC₅₀s. The constant ratios of the combined antibodies were their IC₅₀s. On the x-axis, a dose of 1 was at the IC₅₀ concentration. Fractional effect (FA) plots generated by the CompuSyn program for 7D10, MERS-4 and their combination showing dosage versus effect. Median effect plot of calculated CI values (logarithmic) versus FA values, in which a log CI of <0 indicates synergism, a log CI of >0 indicates antagonism and a log CI of =0 indicates additive action. The percent neutralization, fractional effect, and CI values for 7D10 combined with MERS-4 (b) and 7D10 combined with MERS-27 (c) were calculated and generated using the same method. Source data are provided as a Source Data file

Fig. 5

7D10-H partially inhibiting virus binding to its receptor DPP4. a Inhibition of the binding between recombinant soluble MERS-CoV spike trimer (S) and human DPP4 expressed on the Huh7 cell-surface. Soluble MERS-CoV spike trimer (S) with strep-tag (1 μg) was incubated with monoclonal antibodies (mAbs) in advance at a molar ratio 1:1, 1:3, 1:9, and 1:27 for 1 h. Huh7 cells were incubated with S or S and mAbs mixtures for 1 h. After washing the unbound S, Huh7 cells were stained with streptavidin APC and analyzed by fluorescence-activated cell sorting (FACS). The amounts of S-bound Huh-7 cells were measured and characterized as median fluorescence intensity. See also Supplementary Table 4 and Supplementary Fig. 6. b–e Top view and side view of the MERS-CoV spike trimer in receptor-binding activated states (PDB: 5 × 5c [10.2210/pdb5X5C/pdb]) with two RBD in the up positions, on which the RBD/DPP4 and NTD/antibody structures are superimposed. The spike trimer (RBD in green, NTD in blue, and S2 subunit in gray) was shown as a cartoon. The DPP4 was shown as a semitransparent surface and colored in yellow. 7D10-scFv (b), 7D10-Fab (c), and 7D10-IgG (d, e) were all colored pink. 7D10-Fab and 7D10-IgG were modeled using homologous modeling in SWISS-MODEL

Fig. 6

7D10-H retaining neutralizing activity after viral cell attachment. a Neutralization curves. 7D10-H IgG, 7D10-H Fab, 7D10 scFv were tested for neutralizing activities against pseudotyped MERS-CoV. VRC01 mAb was used as unrelated control. b Pre- and post-attachment neutralizing activities. 7D10-H IgG was tested for neutralizing activity against pseudotyped MERS-CoV before or after receptor binding. VRC01 mAb was used as unrelated control. c The effect of 7D10-H Fab on the conformational change of the MERS-CoV S trimer was probed by western blotting using an anti-MERS-CoV S2 polyclonal antibody. Refolding to the postfusion conformation was detected by the appearance of a proteinase-K resistant band. Trypsin was used at 5 μg mL⁻¹ and proteinase K at 10 μg mL⁻¹. Digestion experiments and western blots were performed in triplicates, and a representative result is shown for each of them. d A cartoon representation designed by us showing the neutralization mechanism by which 7D10 blocks MERS-CoV entry. On the one hand, some virus particles can not bind to DPP4 due to steric hindrance caused by 7D10 binding. On the other hand, 7D10 still recognizes the particles when the up receptor-binding domain (RBD) binds to DPP4, and may inhibit the prefusion to postfusion transition of the S2 subunit and the initiation of membrane fusion. Source data are provided as a Source Data file