Broad anti-SARS-CoV-2 antibody immunity induced by heterologous ChAdOx1/mRNA-1273 vaccination

Abstract

Heterologous prime-boost immunization strategies have the potential to augment COVID-19 vaccine efficacy. We longitudinally profiled severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) spike (S)-specific serological and memory B cell (MBC) responses in individuals who received either homologous (ChAdOx1:ChAdOx1) or heterologous (ChAdOx1:mRNA-1273) prime-boost vaccination. Heterologous messenger RNA (mRNA) booster immunization induced higher serum neutralizing antibody and MBC responses against SARS-CoV-2 variants of concern (VOCs) compared with that of homologous ChAdOx1 boosting. Specificity mapping of circulating B cells revealed that mRNA-1273 boost immunofocused ChAdOx1-primed responses onto epitopes expressed on prefusion-stabilized S. Monoclonal antibodies isolated from mRNA-1273-boosted participants displayed overall higher binding affinities and increased breadth of reactivity against VOCs relative to those isolated from ChAdOx1-boosted individuals. Overall, the results provide molecular insight into the enhanced quality of the B cell response induced after heterologous mRNA booster vaccination.

Fig. 1.. Serum binding and neutralizing activity following homologous and heterologous prime-boost vaccination.

(A) Immunization and blood draw schedule. (B to G) Serum IgG binding to recombinant SARS-CoV-2 S (1:1 mixture of WT S and S-2P antigens) (B), NTD (C), RBD (D), prefusion-stabilized S2 (E), HKU1 S (F), and OC43 S (G), as assessed by ELISA. Binding of pre-pandemic donor sera (n = 17) is shown for comparison in F-G. Geometric mean AUC values are shown below data points. (H to J) Serum neutralizing activity against authentic SARS-CoV-2 D614G (H) and Beta/B.1.351 (I) measured by plaque reduction assay (n = 27-28 donors from each cohort), Delta/B.1.617.2 (J) measured by cytopathic effect (CPE)-based colorimetric microneutralization assay (n = 10 donors from each cohort). Plotted values represent 50% serum neutralizing titers. Values below the dotted line indicate the percentage of samples with serum neutralizing titers below the limit of detection. (K) Serum neutralizing activity against authentic Omicron/B.1.1.529/BA.1 measured by 100% CPE inhibition (n = 10 donors from each cohort). Values below the dotted line indicate the percentage of samples with serum neutralizing titers below the limit of detection. Statistical comparisons between prime and boost were determined by Wilcoxon pair-matched rank sum test. Statistical comparisons across groups were determined by two-tailed Mann Whitney U test with Bonferroni correction [(B) to (E)] and (H) to (K)] or two-sided Kruskal Wallis test by ranks with subsequent Dunn's multiple comparisons [(F) and (G)]. *P < 0.05, **P < 0.01, ***P < 0.001, ****P<0.0001. P, Prime; B, Boost; AUC; area under the curve; LOD, limit of detection; ns, non-significant. All data are representative of at least two independent experiments.

Fig. 2.. SARS-CoV-2 S-specific B cell responses induced by homologous and heterologous prime-boost vaccination.

(A) Frequencies of plasmablasts (defined as CD19⁺CD20^—/loCD71⁺ cells) among circulating CD19⁺ B cells following prime and boost vaccination, as determined by flow cytometry. Pre-pandemic donor samples (n = 9) were included for comparison. Median values are shown above data points. (B) Representative fluorescence-activated cell sorting (FACS) gating strategy used for identifying WT S-specific, S-2P-specific, and WT/S-2P cross-reactive IgG⁺ B cells. (C) Frequencies of total (WT + S-2P) SARS-CoV-2 S-reactive B cells among circulating IgG⁺ B cells, as determined by flow cytometry. Median frequencies are shown above data points. (D) Frequency of circulating IgG⁺ B cells reactive with RBDs encoding mutations present in Beta, Delta, Kappa, and Lambda variants. The height of each bar indicates median frequency. (E) Proportion of donors with detectable B cell reactivity with the indicated number of variant RBDs. The total number of donors analyzed is indicated in the center of the pies. Statistical significance was determined by Fisher's exact test and calculated based on the proportion of donors with B cells displaying reactivity to ≥3 variant RBDs. (F) Proportions of WT S-specific, WT S/S-2P cross-reactive, and S-2P-specific B cells among total S-specific B cells following homologous (left) or heterologous (right) prime-boost immunization. Donors with S-specific B cell frequencies <1% of total IgG⁺ B cells at either time point were excluded from this analysis. Donor IDs are denoted above each bar. (G) Mean proportions of WT S-specific, WT S/S-2P cross-reactive, and S-2P-specific B cells across all donors within each cohort. Error bars indicate 95% confidence intervals. (H) Apparent binding affinities (K_D^App) of WT S-specific monoclonal antibodies for WT S, S-2P, prefusion S subdomains (NTD, RBD, prefusion-stabilized S2), HKU1 S, and OC43 S, as determined by biolayer interferometry (BLI). Competitive binding with an anti-S2 antibody (ADI-69962), as determined by a BLI competitive sandwich assay, is indicated below the heatmap. (I) Neutralizing activity of WT S-specific antibodies against MLV-SARS-CoV-2 Wuhan-1 at a concentration of 1 μg/ml. A previously described anti-RBD neutralizing antibody (ADG-2) was included as a positive control (29). Statistical comparisons between paired prime and boost samples were determined by [(A) and (C)] Wilcoxon pair-matched rank sum test. Statistical comparisons between vaccination cohorts were determined by two-sided Mann-Whitney U tests (D) and two-sided Kruskal-Wallis test by ranks with subsequent Dunn's multiple comparisons [(C) and (G)]. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. P, Prime; B, Boost; K_D^App; apparent equilibrium constant.

Fig. 3.. Binding and neutralization properties of monoclonal antibodies isolated from ChAdOx1 and mRNA-1273 boosted donors.

(A) Fab binding affinities for S-2P, as determined by BLI. Antibodies with no detectable monovalent binding activity are excluded and those with weak binding affinities that could not be fit to a 1:1 binding model are plotted as K_D >100 nM. Black bars indicate medians. Values in parentheses indicate the percentage of antibodies with K_D >100 nM. (B) Proportion of antibodies exhibiting <50%, 50-80%, and >80% neutralization activity against MLV-SARS-CoV-2 Wuhan-1 at a concentration of 1 μg/ml. (C) Proportion of S-2P-reactive antibodies directed to each of the indicated subdomains within prefusion S. Competitive hACE2 binding was determined using a BLI-based competition assay. (D) Heatmaps displaying fold change in binding affinity of anti-RBD antibodies to variant RBDs containing the indicated single point mutations. Competitive hACE2 binding activity and percentage neutralization against MLV-SARS-CoV-2 Wuhan-1 at a concentration of 1 μg/ml are shown in the bars on the right. (E) Summary of the distribution of anti-RBD antibodies belonging to each of the indicated classes. The numbers in the center of the pies indicate the total number of antibodies analyzed. Statistical comparisons were made by (A) two-sided Mann-Whitney U tests. K_D, equilibrium dissociation constant. **P < 0.01.

Fig. 4.. Breadth of antibody binding to SARS-CoV-2 variants.

(A) Proportion of anti-RBD antibodies with the indicated fold reduction in Fab binding affinity to RBDs incorporating mutations present in the Beta, Gamma, Delta, Kappa, Lambda, and Omicron variants relative to the Wuhan-1 RBD, as determined by a bead-based flow cytometric assay. (B) Fab binding affinities of anti-RBD antibodies to Wuhan-1 and variant RBDs. Antibodies that did not reach half-maximal saturation at the highest concentration tested (100 nM) are shown as K_D >100 nM. Black bars denote medians. (C) Heatmap showing the fold reduction in affinity to variant RBDs compared to the Wuhan-1 RBD among neutralizing and non-neutralizing anti-RBD antibodies. The top bar indicates the neutralization IC₅₀ of each mAb against MLV-SARS-CoV-2 Wuhan-1. Antibodies that did not reach 50% neutralization at 10 μg/ml were classified as non-neutralizing. The bottom bar denotes the class of each anti-RBD antibody. (D) Proportion of NTD-directed antibodies with the indicated fold reductions in binding activity to variant NTDs relative to the Wuhan-1 NTD. (E) Fab binding affinities of anti-NTD antibodies to Wuhan-1 and variant NTDs. Antibodies that did not reach half-maximal saturation at 100 nM are shown as K_D >100 nM. Black bars denote medians. (F) Combined proportions of anti-RBD and anti-NTD antibodies that bound the indicated number of variants of concern (Beta, Gamma, Delta, and Omicron) with K_D ≤50 nM. The number in the center of the pie indicates the total number of antibodies tested. Statistical significance was determined by [(B) and (E)] two-sided Mann-Whitney U tests or (F) Fisher's exact test. N.B., non-binding; K_D, equilibrium dissociation constant; IC₅₀, 50% inhibitory concentration. *P < 0.05.