A SARS-CoV-2 neutralizing antibody with extensive Spike binding coverage and modified for optimal therapeutic outcomes

Abstract

COVID-19 pandemic caused by SARS-CoV-2 constitutes a global public health crisis with enormous economic consequences. Monoclonal antibodies against SARS-CoV-2 can provide an important treatment option to fight COVID-19, especially for the most vulnerable populations. In this work, potent antibodies binding to SARS-CoV-2 Spike protein were identified from COVID-19 convalescent patients. Among them, P4A1 interacts directly with and covers majority of the Receptor Binding Motif of the Spike Receptor-Binding Domain, shown by high-resolution complex structure analysis. We further demonstrate the binding and neutralizing activities of P4A1 against wild type and mutant Spike proteins or pseudoviruses. P4A1 was subsequently engineered to reduce the potential risk for Antibody-Dependent Enhancement of infection and to extend its half-life. The engineered antibody exhibits an optimized pharmacokinetic and safety profile, and it results in complete viral clearance in a rhesus monkey model of COVID-19 following a single injection. These data suggest its potential against SARS-CoV-2 related diseases.

Fig. 1. Characterization of neutralizing antibodies from convalescent patients.

a Characterization of SARS-CoV-2 S protein-specific antibodies. Upper panels: binding of antibodies to the full-length S protein, S1 protein, and S2 protein was evaluated by ELISA (in duplicates with symbols show each of the replicates). Lower left panel: blockage of the binding of SARS-CoV-2 Spike S1 protein to Vero E6 cells by antibodies evaluated by flow cytometry (data in singleton). Lower middle panel: pseudovirus neutralization assay in Huh-7 cells (data in singleton). Lower right panel: in triplicates with symbols show each of the triplicates and SARS-CoV-2 live virus neutralization assay. All experiments were repeated at least two more times (except S2 binding that was repeated one more time) with similar results. b Images of Vero E6 cell-infected SARS-CoV-2 treated with antibodies of different concentrations. Green (stained with SARS-CoV-2 nucleocapsid protein (NP) antibody) indicates viral infected cells and blue (Hoechst 33258) represents cell nuclei. Experiment was performed in triplicates and repeated two more times with similar results.

Fig. 2. Structural analysis of P4A1 Fab and SARS-CoV-2 RBD complex.

a The overall P4A1-Fab-RBD complex structure superimposed with the hACE2-RBD complex. The P4A1 heavy chain (colored slate blue), light chain (colored salmon red), and hACE (colored pale green) are displayed in cartoon representation. The SARS-CoV-2 RBD is colored in gray and displayed in surface representation. b The epitope of P4A1 shown in surface representation. The CDR loops of heavy chain (HCDR) and light chain (LCDR) are colored in purple and magenta, respectively. The epitopes from the heavy chain and light chain are colored in slate blue and salmon red, respectively. The only residue K417, which contacts with both heavy chain and light chain, is colored in pink. The light-chain frame region 3 (LFR3) is colored in orange. The identical residues on RBD shared in P4A1 and hACE2 binding are labeled in red. The residues are numbered according to SARS-CoV-2 RBD. c The detailed interactions between SARS-CoV-2 RBD with HCDR, LCDR, and LFR3. The residues are shown in sticks with identical colors to (b).

Fig. 3. The activities of IgG4 antibody P4A1–2A to different SARS-CoV-2 S protein mutants, FcRs, and C1q.

a Binding of antibody P4A1 to SARS-CoV-2 S protein N354D/D364Y, R408I, W436R, V367F, or D614G mutants determined by surface plasmon resonance (SPR). b Pseudovirus neutralization assay in hACE2-overexpressing HEK293 cells. Experiment performed in triplicates with symbols represent each of the triplicates. c The binding affinity of P4A1 and P4A1–2A for different human FcRs and complement C1q.

Fig. 4. Therapeutic efficacy of in the rhesus macaque model of SARS-CoV-2 infection.

a Experimental design for therapeutic testing of P4A1–2A in the rhesus macaque (n = 3/group). b Viral load in oropharyngeal swabs tested by RT-qPCR was monitored for 7 days. c Viral load in the respiratory tissues (including trachea, left and right bronchus, and all six lung lobes) collected at necropsy on 7 days post infection (d.p.i., n = 1/group) was tested by RT-qPCR. d Representative images of histopathology in lung tissue from isotype control or P4A1–2A 50 mg/kg treated animals (collected at 7 d.p.i., n = 1/group).