Convalescent-Phase Sera and Vaccine-Elicited Antibodies Largely Maintain Neutralizing Titer against Global SARS-CoV-2 Variant Spikes

Abstract

The increasing prevalence of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) variants with spike protein mutations raises concerns that antibodies elicited by natural infection or vaccination and therapeutic monoclonal antibodies will become less effective. We show that convalescent-phase sera neutralize pseudotyped viruses with the B.1.1.7, B.1.351, B.1.1.248, COH.20G/677H, 20A.EU2, and mink cluster 5 spike proteins with only a minor loss in titer. Similarly, antibodies elicited by Pfizer BNT162b2 vaccination neutralized B.1.351 and B.1.1.248 with only a 3-fold decrease in titer, an effect attributable to E484K. Analysis of the Regeneron monoclonal antibodies REGN10933 and REGN10987 showed that REGN10933 has lost neutralizing activity against the B.1.351 and B.1.1.248 pseudotyped viruses, and the cocktail is 9- to 15-fold decreased in titer. These findings suggest that antibodies elicited by natural infection and by the Pfizer vaccine will maintain protection against the B.1.1.7, B.1.351, and B.1.1.248 variants but that monoclonal antibody therapy may be less effective for patients infected with B.1.351 or B.1.1.248 SARS-CoV-2. IMPORTANCE The rapid evolution of SARS-CoV-2 variants has raised concerns with regard to their potential to escape from vaccine-elicited antibodies and anti-spike protein monoclonal antibodies. We report here on an analysis of sera from recovered patients and vaccinated individuals and on neutralization by Regeneron therapeutic monoclonal antibodies. Overall, the variants were neutralized nearly as well as the wild-type pseudotyped virus. The B.1.351 variant was somewhat resistant to vaccine-elicited antibodies but was still readily neutralized. One of the two Regeneron therapeutic monoclonal antibodies seems to have lost most of its activity against the B.1.351 variant, raising concerns that the combination therapy might be less effective for some patients. The findings should alleviate concerns that vaccines will become ineffective but suggest the importance of continued surveillance for potential new variants.

Keywords: 20A.EU2; B.1.1.248; B.1.1.7; B.1.351; COH.20G/677H; Pfizer BNT162b2; REGN10933; REGN10987; SARS-CoV-2; mink cluster 5; neutralization; spike protein.

FIG 1

Infectivity of viruses pseudotyped by variant spike proteins. (A, top) Diagram showing the domain structure of the SARS-CoV-2 spike monomer. NTD, N-terminal domain; RBD, receptor binding domain; RBM, receptor binding motif; SD1 subdomain 1; SD2, subdomain 2; FP, fusion peptide; HR1, heptad repeat 1; HR2, heptad repeat 2; TM, transmembrane region; IC, intracellular domain. (Bottom) Diagrams indicating the locations of the mutations of the United Kingdom B.1.1.7; South Africa B.1.351; Brazilian B.1.1.248; Columbus, OH, COH.20G/677H; European 20A.EU2; and mink cluster 5 variant spikes. Expression vectors for the variant spike proteins or with the individual mutations were generated, each with the D614G mutation and deleted for the carboxy-terminal 19 amino acids. The vectors were used to produce pseudotyped lentiviral virions. The infectivity of the virions normalized for RT activity was tested on ACE2.293T cells. Luciferase activity was measured at 2 days postinfection. (B) Locations of key mutated amino acid residues in the spike protein. Side (top) and top (bottom) views of the SARS-CoV-2 spike protein prefusion structure with a single RBD (gray) are shown. Arrows indicate RBD amino acid residues 417, 453, 484, 486, and 501 and the Δ69–70 and D614G mutations. (C) Infectivity of viruses pseudotyped with the individual mutations of the B.1.1.7 spike protein or combinations thereof. RLU, relative luminescence units. (D) Infectivity of D614G; European 20A.EU2; Columbus, OH, COH.20G/677H; and Brazilian B.1.1.248 spike protein-pseudotyped viruses. (E) Infectivity of viruses pseudotyped with the South Africa B.1.351 variant spike protein and its individual mutations. (F) Infectivity of virus with the mink-associated variant spike protein individual mutations and combinations thereof. The experiments were repeated three times, with similar results.

FIG 2

Neutralization of spike protein variants by convalescent-phase sera. (A) Neutralization of spike protein-pseudotyped viruses by serum samples from 18 (donors 1 to 18) (left) and 10 (donors 19 to 28) (right) convalescent individuals. The COH.20G/677H variant was tested in a separate experiment, as shown on the right. Each dot represents the IC₅₀ for a single donor. Correlations were calculated with GraphPad Prism software using Pearson’s correlation coefficients, and error bars indicate standard deviations. The analyses were repeated twice, with similar results. n.s., not significant. (B) Neutralization by convalescent-phase donor serum of viruses pseudotyped by N501Y, Δ69–70, B.1.1.7 (Δ69–70/N501Y/P681H), COH.20G/677H, 20A.EU2, and mink cluster 5 spike proteins compared to D614G. The IC₅₀s are based on serum dilution. (C, left) Neutralization of virus pseudotyped by the D614G, B.1.351, or E484K spike proteins by convalescent-phase sera. (Right) Comparison of neutralization of B.1.351 spike protein-pseudotyped virus with neutralization of the E484K pseudotype. (D) Neutralization of virus pseudotyped by the D614G and B.1.1.248 spike proteins by convalescent-phase sera.

FIG 3

Neutralization of variant spike protein pseudotypes by the sera of BNT162b2-vaccinated individuals. (A) Serum samples from vaccinated individuals (n = 5) were serially diluted and incubated with the indicated variant spike protein-pseudotyped viruses normalized for infectivity. The data are relative to the infectivity of unneutralized virus. (B) Neutralization IC₅₀s of D614G, N501Y, S982A, B.1.1.7 (Δ69–70/N501Y/P681H), COH.20G/677H, 20A.EU2, B.1.351, and B.1.1.248 pseudotypes. The experiments were repeated twice, with similar results.

FIG 4

Neutralization of variant spike protein pseudotypes by REGN10933 and REGN10987 monoclonal antibodies. The neutralization of D614G, B.1.1.7, B.1.351, B.1.1.248, and mink cluster 5 pseudotyped viruses by REGN10933 and REGN10987 was measured. (A) Neutralization of B.1.1.7 (Δ69–70/N501Y/P681H), B.1.351, B.1.1.248, mink cluster 5, and COH.20G/677H pseudotypes (top left); individual B.1.1.7 mutations (top right); individual B.1.351 mutations (bottom left); and mink cluster 5 mutation pseudotyped viruses (bottom right) by REGN10987. (B) Neutralization curves of D614G, B.1.1.7 (Δ69–70/N501Y/P681H), B.1.351, B.1.1.248, mink cluster 5, and COH.20G/677H (top left); individual B.1.1.7 mutations (top right); individual B.1.351 mutations (bottom left); and mink cluster 5 mutation pseudotyped viruses (bottom right) by REGN10933. (C) Neutralization of D614G, B.1.1.7 (Δ69–70/N501Y/P681H), B.1.351, B.1.1.248, COH.20G/677H, and mink cluster 5 pseudotyped viruses by a 1:1 mixture of REGN10933 and REGN10987. The analyses were repeated three times, with similar results.

FIG 5

Relative affinity of viruses pseudotyped by B.1.1.7 and B.1.351 spike proteins for ACE2. The relative affinity of the variant spike proteins for ACE2 was measured by a soluble ACE2 (sACE2) neutralization assay (A to D) and a virion binding assay (E). For the sACE2 neutralization assay, viruses pseudotyped with variant spike proteins were incubated with a serial dilution of recombinant sACE2, and their infectivity was then measured on ACE2.293T cells. The data represent percent infectivities, at each concentration of sACE2, of the spike variants plotted against sACE2 neutralization of D614G pseudotyped virus. The histograms at the bottom right summarize the IC₅₀s for each of the curves. For the virion binding assay, pseudotyped virions (30 ng p24) were incubated with Ni-NTA beads coated with the indicated amounts of His-tagged recombinant sACE2 protein. Unbound virions were removed by centrifugation, and bound virions were analyzed on an immunoblot probed with anti-p24 antibody. (A) sACE2 neutralization of viruses pseudotyped by spike proteins with the individual B.1.1.7 mutations. (B) sACE2 neutralization of viruses pseudotyped by spike proteins with the individual B.1.351 mutations. (C) sACE2 neutralization of COH.20G/677H (left) and B.1.1.248 (right) pseudotyped viruses. The histograms to the right of the curves show the calculated IC₅₀s. (D) Virion binding assay of viruses pseudotyped with the variant spike proteins. (E) Immunoblot analysis of the input sACE2 (left) and pseudotyped viruses (right) used in the virus-sACE2 binding assay. The mass of bead-bound p24 (nanograms) as determined using a standard curve with recombinant protein is indicated below each lane.