Omicron BA.1 breakthrough infection drives cross-variant neutralization and memory B cell formation against conserved epitopes

Abstract

Omicron is the evolutionarily most distinct severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) variant of concern (VOC) to date. We report that Omicron BA.1 breakthrough infection in BNT162b2-vaccinated individuals resulted in strong neutralizing activity against Omicron BA.1, BA.2, and previous SARS-CoV-2 VOCs but not against the Omicron sublineages BA.4 and BA.5. BA.1 breakthrough infection induced a robust recall response, primarily expanding memory B (B_MEM) cells against epitopes shared broadly among variants, rather than inducing BA.1-specific B cells. The vaccination-imprinted B_MEM cell pool had sufficient plasticity to be remodeled by heterologous SARS-CoV-2 spike glycoprotein exposure. Whereas selective amplification of B_MEM cells recognizing shared epitopes allows for effective neutralization of most variants that evade previously established immunity, susceptibility to escape by variants that acquire alterations at hitherto conserved sites may be heightened.

Trial registration: ClinicalTrials.gov NCT04380701 NCT04949490 NCT04380701.

Fig. 1.. Cohorts, sampling and experimental setup.

Blood samples were drawn from four cohorts: Omicron-naïve individuals double- or triple-vaccinated with BNT162b2 (light and dark green), and individuals double- or triple-vaccinated with BNT162b2 that subsequently had a breakthrough infection with Omicron (light and dark purple) at a time of BA.1 dominance. PBMCs (red) and sera (yellow) were isolated in the Omicron-naïve cohorts at the time-points indicated following their most recent vaccination; for convalescent cohorts, the time from their most recent vaccination to Omicron infection, and infection to PBMC and serum isolation are indicated in purple text. All values are specified as median [range]. The age/gender composition of cohorts is further detailed in Tables S4 and S10. Serum neutralizing capacity was assessed using pseudovirus and live virus neutralization tests. SARS-CoV-2 S glycoprotein-specific B_MEM cells were assessed via a flow cytometry-based B cell phenotyping assay using bulk PBMCs. N/A, not applicable. Schematic was created with BioRender.com.

Fig. 2.. Omicron BA.1 breakthrough infection of BNT162b2 double- and triple-vaccinated individuals induces broad neutralization of Omicron BA.1, BA.2 and other VOCs, but not BA.4 and BA.5.

Serum was drawn from double-vaccinated individuals (BNT162b2²) at 22 days after the second dose (green, open circles), from triple-vaccinated individuals (BNT162b2³) at 28 days after the third dose (green, closed circles), from double-vaccinated individuals with Omicron breakthrough infection (BNT162b2² + Omi) at 46 days post-infection (purple, open triangles), and from triple-vaccinated individuals and Omicron breakthrough infection (BNT162b2³ + Omi) at 44 days post-infection (purple, closed triangles). Serum was tested in duplicate; 50% pseudovirus neutralization (pVN₅₀) geometric mean titers (GMTs) (in a), the geometric mean ratio of SARS-CoV-2 variants of concern (VOCs) and SARS-CoV-1 pVN₅₀ GMTs normalized against Wuhan pVN₅₀ GMTs (in b), 50% virus neutralization (VN₅₀) GMTs (in c), and the geometric mean ratio of SARS-CoV-2 variants of concern (VOCs) VN₅₀ GMTs normalized against Wuhan VN₅₀ GMTs (in d) were plotted. For titer values below the limit of detection (LOD), LOD/2 values were plotted. Values above violin plots represent the group GMTs. The nonparametric Friedman test with Dunn’s multiple comparisons correction was used to compare Wuhan neutralizing group GMTs with titers against the indicated variants and SARS-CoV-1. Multiplicity-adjusted p values are shown. (a) pVN₅₀ GMTs against Wuhan, VOC and SARS-CoV-1 pseudovirus. (b) Group geometric mean ratios with 95% confidence intervals for all cohorts shown in a. (c) VN₅₀ GMTs against live SARS-CoV-2 Wuhan, Beta, Delta and Omicron BA.1. (d) Group geometric mean ratios with 95% confidence intervals for all cohorts shown in c.

Fig. 3.. B_MEM cells of individuals double- and triple-vaccinated with BNT162b2 broadly recognize VOCs and are further boosted by Omicron BA.1 breakthrough infection.

PBMC samples from double-vaccinated individuals (BNT162b2²) at 22 days after the second dose (green, open squares) and 5 months after the second dose (green, open circles), from triple-vaccinated individuals (BNT162b2³) at 84 days after the third dose (green, closed circles), from double-vaccinated individuals with Omicron breakthrough infection (BNT162b2² + Omi) at 46 days post-infection (purple, open triangles), and from triple-vaccinated individuals with Omicron breakthrough infection (BNT162b2³ + Omi) at 44 days post-infection (purple, closed triangles) were analyzed via flow cytometry for SARS-CoV-2-specific B_MEM cell (B_MEM – CD3^-CD19⁺CD20⁺IgD^-CD38^int/low) frequencies via B cell bait staining. (a) Schematic of one-dimensional staining of B_MEM cells with fluorochrome-labeled SARS-CoV-2 S glycoprotein tetramer bait allowing discrimination of variant recognition. Frequencies of Wuhan or VOC full-length S glycoprotein- (b) and RBD- (c) specific B_MEM cells for the four groups of individuals were analyzed. Variant-specific B_MEM cell frequencies were normalized to Wuhan frequencies for S glycoprotein (d) and RBD (e) binding. (f) The frequency ratios of RBD protein-specific B_MEM cells over full-length S glycoprotein-specific B_MEM cells are depicted. Mean values and standard error of the mean (SEM) are shown in (b)-(f). Statistical comparisons between individuals of each group presented in (b) and (c) were performed using the nonparametric Friedman test with Dunn’s multiple comparisons correction presented in fig S4 (a-j). n = number of individuals per group. Schematic in (a) was created with BioRender.com.

Fig. 4.. Omicron BA.1 breakthrough infection of BNT162b2 double- and triple-vaccinated individuals primarily boosts B_MEM against conserved epitopes shared broadly between S glycoproteins of Wuhan and other VOCs rather than strictly Omicron S-specific epitopes.

PBMC samples from double-vaccinated individuals (BNT162b2²) at 22 days after the second dose (green, open squares) and 5 months after the second dose (green, open circles), from triple-vaccinated individuals (BNT162b2³) at 84 days after the third dose (green, closed circles), from double-vaccinated individuals with Omicron breakthrough infection (BNT162b2² + Omi) at 46 days post-infection (purple, open triangles), and from triple-vaccinated individuals with Omicron breakthrough infection (BNT162b2³ + Omi) at 44 days post-infection (purple, closed triangle) were analyzed via flow cytometry for SARS-CoV-2-specific memory B cells (B_MEM – CD3^-CD19⁺CD20⁺IgD^-CD38^int/low) frequencies via B cell bait staining (a). (b) Representative flow plots of Omicron and Wuhan S glycoprotein- and RBD-binding for each of the four groups of individuals investigated. Frequencies of B_MEM binding Omicron, Wuhan, or both (shared) full-length S glycoprotein (c) or RBD (d) for Omicron-experienced and naïve BNT162b2 double and triple vaccinees are shown. (e) Venn diagrams visualizing the combinatorial (Boolean) gating strategy to identify cross-reactive B_MEM recognizing all four variants simultaneously (All 4 +ve) and B_MEM recognizing only Omicron (only Omi) or only Wuhan (only Wuhan) S glycoproteins. Frequencies for these three B_MEM sub-groups were compared for full-length S glycoprotein (f) and RBD (g) in the four different groups of individuals investigated. RBD variant recognition pattern by B_MEM was assessed through Boolean flow cytometric gating strategy and frequencies recognizing combinations of Wuhan and/or variant RBDs are displayed in (h), for all Omicron convalescent subjects (double and triple vaccinees pooled, n=13). (i) Conserved site within the RBD domain recognized by RBD-specific B_MEM after Omicron break-through infection. Mean values are indicated in c, d, f, g, and h. n = number of individuals per group. Results for the nonparametric Friedman test with Dunn’s multiple comparisons correction testing for significance within treatment groups against shared (c,d) and all 4+ve (f,g) are presented in Table S13. Schematic in (a) was created with BioRender.com.