Early 2022 breakthrough infection sera from India target the conserved cryptic class 5 epitope to counteract immune escape by SARS-CoV-2 variants

Abstract

During the coronavirus disease 2019 (COVID-19) pandemic, the vast majority of epitope mapping studies have focused on sera from mRNA-vaccinated populations from high-income countries. In contrast, here, we report an analysis of 164 serum samples isolated from patients with breakthrough infection in India during early 2022 who received two doses of the ChAdOx viral vector vaccine. Sera were screened for neutralization breadth against wild-type (WT), Kappa, Delta, and Omicron BA.1 viruses. Three sera with the highest neutralization breadth and potency were selected for epitope mapping, using charged scanning mutagenesis coupled with yeast surface display and next-generation sequencing. The mapped sera primarily targeted the recently identified class 5 cryptic epitope and, to a lesser extent, the class 1 and class 4 epitopes. The class 5 epitope is completely conserved across all severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) variants and for most sarbecoviruses. Based on these observations, an additional 26 sera were characterized, and all showed a broad neutralizing activity, including against XBB.1.5. This is in contrast with the results obtained with the sera from individuals receiving multiple doses of original and updated mRNA vaccines, where impaired neutralization of XBB and later variants of concern (VOCs) were observed. Our study demonstrates that two doses of the ChAdOx vaccine in a highly exposed population were sufficient to drive substantial neutralization breadth against emerging and upcoming variants of concern. These data highlight the important role of hybrid immunity in conferring broad protection and inform future vaccine strategies to protect against rapidly mutating viruses.

Importance: Worldwide implementation of coronavirus disease 2019 (COVID-19) vaccines and the parallel emergence of newer severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) variants have shaped the humoral immune response in a population-specific manner. While characterizing this immune response is important for monitoring disease progression at the population level, it is also imperative for developing effective countermeasures in the form of novel vaccines and therapeutics. India has implemented the world's second largest COVID-19 vaccination drive and encountered a large number of post-vaccination "breakthrough" infections. From a cohort of patients with breakthrough infection, we identified individuals whose sera showed broadly neutralizing immunity against different SARS-CoV-2 variants. Interestingly, these sera primarily target a novel cryptic epitope, which was not identified in previous population-level studies conducted in Western countries. This rare cryptic epitope remains conserved across all SARS-CoV-2 variants, including recently emerged ones and for the SARS-like coronaviruses that may cause future outbreaks, thus representing a potential target for future vaccines.

Keywords: Omicron subvariant XBB.1.5; SARS-CoV-2; breakthrough infection; class 5/RBD8 epitopes; convalescent serum; epitope mapping; immune escape; neutralizing antibodies; spike protein RBD; yeast display library.

Fig 1

Study cohort and timeline. (A) Schematic representation of study design. (B) Distribution of the study cohort as a function of the timeline of vaccination and breakthrough infection in the context of the second and third waves of pandemics caused by different VOCs of SARS-CoV-2. Red bars indicate the time points of serum sample collection for the entire cohort (early) and subsequently for the selected individuals (late). (C) Age distribution of male and female volunteers included in the study cohort.

Fig 2

Neutralizing activity of the polyclonal serum samples isolated from the study cohort. (A) Heatmap of neutralizing activity of the 164 serum samples (at dilutions; 0.01, 0.001, 0.001) against the parental strain and different VOCs of SARS-CoV-2, with green to red representing the lowest to highest extent of neutralization (%). (B) Comparison of the relative neuralization efficacies of the serum samples (at 0.01 dilution) against the parental strain and different VOCs. Results were compared by performing two-way ANOVA, and significance is defined as P < 0.0001 and denoted as asterisks (****). (C) Venn diagram showing a cross-reactivity of the selected serum samples displaying more than 65% neutralization (at 0.01 dilution) toward at least one of the pseudoviral variants tested in this study. A total of 29 serum samples with >65% neutralization toward all the VOCs tested in this study were designated as broadly neutralizing sera (red diamonds in A). Each image shown here is a representative of three independent experiments performed in triplicate.

Fig 3

Comparative neutralizing activities of the selected serum samples collected at early and late time points. (A) Comparative analysis of the neutralization titers (IC50) of serum samples isolated from nine selected volunteers during the first (early) and second (late) batches of sample collection, against WA.1, B.1.617.2, and BA.1. (B) Heatmap showing the fold change in the IC50 values for B.1.617.2 and BA.1 w.r.t. the parental WA.1 strain, with red to white indicating higher to lower neutralization titers. (C) Comparative analysis of the pseudoviral neutralization titers of the serum samples isolated at early and late time points. (D) Heatmap representing the variant-specific fold change in IC50 values. All data shown are averages of the results of at least three independent experiments.

Fig 4

The serum samples show high binding affinity toward spike RBD. (A) ELISA was performed to determine the binding affinity of the polyclonal serum samples toward spike RBD derived from the WA.1 and Omicron subvariants BA.1 and BA.5. (B) Comparison of the binding affinity of different serum samples (presented in the form of IC50) toward different RBD variants. (C) Fold change in IC50 values for the BA.1 and BA.5 variants w.r.t. the WA.1. Each data shown is a representative of at least three experiments performed independently.

Fig 5

Identification and validation of RBD epitope residues through yeast display analysis. (A) Schematic representation of the yeast display screening. (B) The heatmap represents the ratio of MFI expression to MFI binding extracted from analysis of deep sequencing reads. The higher to lower ratios are represented in a red to blue gradient. The higher the ratio (red), the higher the extent of serum escape at that position. The epitope mapping study was performed in duplicate. (C) The MFI expression/MFI binding ratios are mapped on the RBD structure (PDB ID: 6M0J) following the same color scheme. The RBM is demarcated on the structure using a black line. (D) Validation was carried out using FACS by quantifying the binding of WT and the mutant RBDs, harboring mutations at the epitope and non-epitope residues, to BCRTH-01, INK09, and INK36 serum samples. The MFI binding of WT and mutant RBDs w.r.t. the three serum samples is presented. Each image is representative of two independent experiments performed in triplicate. Results were compared by performing a one-way ANOVA, and significance is defined as P < 0.01 and denoted as asterisk (*).

Fig 6

Epitope classification and characterization. (A) Individual epitope residues, as identified for the BCRTH-01, INK-09, and INK-36 serum samples, are mapped on the surface of the RBD (PDB ID: 6M0J) and are color coded based on the previously published epitope classes. (B) Class 5/RBD8 epitopes identified in this study are mapped on the pre-fusion trimeric structure of the spike in its closed (PDB ID: 6VXX) and open (PDB ID: 6VYB) conformations. (C) Matrix representing the conservation of identified epitope residues across all major SARS-CoV-2 variants. Mutations of a specific residue in a particular variant are represented in blue squares. Epitope classes are designated using colored diamonds.

Fig 7

Neutralization titers (IC50) for WT and mutant pseudoviruses w.r.t. selected serum samples. (A) Neutralization curves for pseudoviruses with WT (WA.1) or mutant spike proteins, harboring specific mutations at the epitope residues, with serial dilutions of the BCRTH-01, INK-09, and INK-36 sera as indicated. (B) Neutralization titers (IC50) for WT and mutant pseudoviruses for three sera samples as calculated from (A). The neutralization experiments are performed in three independent biological replicates.

Fig 8

A considerable fraction of the study cohort shows neutralization activity against Omicron XBB.1.5. (A) Neutralization curves for WA.1, BA.1, and XBB.1.5 pseudoviruses with serial dilutions of BCRTH-01, INK-09, and INK-36 sera. The data shown are averages of the results of at least three independent experiments ± SD. (B) Neutralization titers (IC50) of 29 serum samples (from Fig. 2C) w.r.t. WA.1 and XBB.1.5 pseudoviruses as calculated from the neutralization curves shown in Fig. S6. Fold change in the IC50 values for XBB.1.5 w.r.t. the WA.1 are marked in salmon red.