Mapping immunodominant sites on the MERS-CoV spike glycoprotein targeted by infection-elicited antibodies in humans

Abstract

Middle East respiratory syndrome coronavirus (MERS-CoV) first emerged in 2012 and causes human infections in endemic regions. Vaccines and therapeutics in development against MERS-CoV focus on the spike (S) glycoprotein to prevent viral entry into target cells. These efforts are limited by a poor understanding of antibody responses elicited by infection. Here, we analyze S-directed antibody responses in plasma collected from MERS-CoV-infected individuals. We observe that binding and neutralizing antibodies peak 1-6 weeks after symptom onset/hospitalization, persist for at least 6 months, and neutralize human and camel MERS-CoV strains. We show that the MERS-CoV S₁ subunit is immunodominant and that antibodies targeting S₁, particularly the receptor-binding domain (RBD), account for most plasma neutralizing activity. Antigenic site mapping reveals that plasma antibodies frequently target RBD epitopes, whereas targeting of S₂ subunit epitopes is rare. Our data reveal the humoral immune responses elicited by MERS-CoV infection, which will guide vaccine and therapeutic design.

Keywords: CP: Immunology; MERS-CoV; antibody; binding; epitopes; merbecovirus; neutralization; plasma; spike; therapeutic; vaccine.

Figure 1.. MERS-CoV infection induces robust S binding and neutralizing antibody responses.

A) IgG binding titers (half-maximal effective dilution [ED₅₀]) against the prefusion MERS-CoV EMC/2012 S ectodomain trimer and B) neutralizing antibody titers (half-maximal inhibitory dilution [ID₅₀]) against VSV pseudotyped with the MERS-CoV EMC/2012 S measured for 98 plasma samples collected from 30 individuals hospitalized with MERS-CoV infection between 2017–2019 in Saudi Arabia. The GMT ± GSD is represented by the black bar and displayed above the plot and the limit of detection (ED₅₀ or ID₅₀: 10) is represented by the dashed line. Data reflect results obtained from one biological replicate and are representative of data obtained from at least two biological replicates conducted with unique batches of S protein or pseudovirus. C) Correlation analysis between neutralizing antibody and S IgG binding titers in the 97 plasma samples with detectable S binding titers. Kinetics of the D) S IgG binding titers and E) neutralizing antibody titers for 90 plasma samples collected from 22 individuals who contributed two or more samples. Samples collected from the same individual are connected with gray lines. F) Neutralization potency (ID₅₀) of plasma samples against VSV pseudotyped with the spike protein of the indicated MERS-CoV variant or the related merbecovirus MjHKU4r-CoV-1. Only the sample with highest S binding titer per individual was included in the analysis. GMTs ± GSD are represented by the bars and displayed above the plot. The fold-change in GMT compared to MERS-CoV EMC/2012 pseudovirus is indicated below the GMT for each pseudovirus. Data reflect results obtained from one biological replicate and are representative of data obtained from at least two biological replicates conducted with unique batches of pseudoviruses. See also Figures S1, S2, S3, and S4 and Tables S1 and S2.

Figure 2.. Antibodies directed against the S₁ subunit of the MERS-CoV spike protein are responsible for nearly all neutralizing activity of plasma.

A) IgG binding titers (ED₅₀) against the MERS-CoV EMC/2012 S₁ subunit measured for plasma samples. The sample with the highest S binding titer per individual was included in the analysis with 29 samples being included in total. The GMT ± GSD is represented by the black bar and displayed above the plot. The limit of detection (ED₅₀: 10) of the assay is represented by the dashed line. Data reflect results obtained from one biological replicate and are representative of data obtained from at least two biological replicates conducted with unique batches S₁ protein. B) Correlation analysis between S₁ IgG binding titers and neutralizing antibody titers for samples with detectable S₁ binding titers. C) S₁ and D) S IgG binding titers and E) neutralizing antibody titers in mock-depleted and S₁-depleted plasma samples. One sample per individual was included in the analysis with 27 samples being included in total. Mock- and S₁-depleted samples from the same individual are connected with a gray line. GMTs are displayed above the plot and the limit of detection (ED₅₀ or ID₅₀: 10) is represented by the dashed line. Data reflect results obtained from one biological replicate and are representative of data obtained from at least two biological replicates conducted with unique batches of S₁ and S proteins and pseudovirus. Comparisons between mock- and S1-depleted groups were made using the Wilcoxon matched-pairs signed rank test. ****p < 0.0001. See also Figures S5 and S6 and Table S2.

Figure 3.. RBD-directed antibodies account for the majority of neutralizing activity in plasma.

A) IgG binding titers (ED₅₀) against the MERS-CoV EMC/2012 RBD measured for the plasma samples. B) Correlation analysis between RBD IgG binding titers and neutralization potency of the plasma samples with detectable RBD binding titers. C) IgG binding titers (ED₅₀) against the MERS-CoV EMC/2012 NTD determined for the plasma samples. D) Correlation analysis between NTD IgG binding titers and neutralizing antibody titers in the plasma samples with detectable NTD binding titers. For the RBD and NTD ELISAs, the sample with the highest S binding titer per individual was included in the analysis. The GMT ± GSD is represented by the black bar and presented above the plot. The limit of detection (ED₅₀: 10) of the assay is indicated by the dashed line. Data reflect results obtained from one biological replicate and are representative of data obtained from at least two biological replicates conducted with unique batches of RBD or NTD protein. E) RBD IgG binding titers for mock- and RBD-depleted plasma samples. F) NTD IgG binding titers for mock- and NTD-depleted plasma samples. G) S IgG binding titers and H) neutralizing antibody titers for mock-, RBD-, and NTD-depleted plasma. Six of the 28 RBD-depleted plasma and 1 of the 28 NTD-depleted plasma samples exhibited neutralizing titers below the limit of detection. The one sample with the highest S binding titer per individual was included in the analysis. Data presented are from one biological replicate and reflective of two biological replicates completed with unique batches of RBD, NTD, and S proteins as well as distinct batches of pseudovirus. GMTs are displayed above the plots and the limit of detection (ED₅₀ or ID₅₀: 10) is indicated by the dashed line. Comparisons between RBD or NTD IgG binding titers of the mock- and RBD or NTD-depleted groups, respectively, were made using the Wilcoxon matched-pairs signed. Comparison between the spike IgG bindings titers and neutralizing antibody titers of mock-, RBD- and NTD-depleted groups were made using Dunn’s multiple comparison test. *p<0.05; ****p < 0.0001. See also Figures S7 and S8 and Table S2.

Figure 4.. Plasma antibodies target S₁ neutralizing epitopes, but rarely S₂ epitopes.

A) Epitopes targeted by S₁-directed antibodies included for the competition ELISAs. The RBM is outlined in black and glycans are represented in dark blue. B) Blocking titers (half-maximal blocking dilution [BD₅₀]) measured for plasma samples against the indicated S₁-directed antibodies. C) Epitopes targeted by S₂-directed antibodies included in the competition ELISAs. Glycans are displayed in dark blue. D) Blocking titers (BD₅₀) in plasma samples against the specified S₂-directed antibodies. The sample with the highest S binding titer per individual was analyzed. GMTs ± GSD are indicated by black bar and displayed above the plot. The portion of individuals exhibiting detectable blocking titers is displayed below the GMT for each antibody. Data presented are from one biological replicate and representative of data collected from at least two biological replicates conducted with distinct batches of antibodies and biotinylated S protein. See also Figure S9 and Table S2.