Escape from recognition of SARS-CoV-2 variant spike epitopes but overall preservation of T cell immunity

Abstract

SARS-CoV-2 variants that escape neutralization and potentially affect vaccine efficacy have emerged. T cell responses play a role in protection from reinfection and severe disease, but the potential for spike mutations to affect T cell immunity is incompletely understood. We assessed neutralizing antibody and T cell responses in 44 South African COVID-19 patients either infected with the Beta variant (dominant from November 2020 to May 2021) or infected before its emergence (first wave, Wuhan strain) to provide an overall measure of immune evasion. We show that robust spike-specific CD4 and CD8 T cell responses were detectable in Beta-infected patients, similar to first-wave patients. Using peptides spanning the Beta-mutated regions, we identified CD4 T cell responses targeting the wild-type peptides in 12 of 22 first-wave patients, all of whom failed to recognize corresponding Beta-mutated peptides. However, responses to mutated regions formed only a small proportion (15.7%) of the overall CD4 response, and few patients (3 of 44) mounted CD8 responses that targeted the mutated regions. Among the spike epitopes tested, we identified three epitopes containing the D215, L18, or D80 residues that were specifically recognized by CD4 T cells, and their mutated versions were associated with a loss of response. This study shows that despite loss of recognition of immunogenic CD4 epitopes, CD4 and CD8 T cell responses to Beta are preserved overall. These observations may explain why several vaccines have retained the ability to protect against severe COVID-19 even with substantial loss of neutralizing antibody activity against Beta.

Fig. 1.. T cell recognition of SARS-CoV-2 spike in first and second wave COVID-19 patients.

(A) Clinical characteristics of acute COVID-19 patients recruited during the first and second wave of the COVID-19 pandemic in South Africa. *: median and interquartile range.^$: Disease severity was defined based on oxygen therapy requirement according to the WHO ordinal scale scoring system Moderate (no O₂ or O₂ via nasal prongs) or severe (O₂ via high flow to ECMO).^#: SARS-CoV-2 polymerase chain reaction (PCR) was performed using the Allplex 2019-nCoV Assay (Seegene). The cycle threshold (CT) value for the N-gene is reported. (B) SARS-CoV-2 epidemiological dynamics in the Western Cape (South Africa). Prevalence of SARS-CoV-2 strains is on the left y-axis (based on sequencing 4549 samples). Ancestral strains are depicted in blue, Beta in red, Alpha in grey and Delta in green. Monthly COVID-19 cases are on the right y-axis. Bars above the graph indicate when samples were collected. (C) Representative flow cytometry plots of IFN-γ, TNF-α and IL-2 production by CD4 T cells in response to ancestral full spike peptide pool (Full Spike) in one first wave (blue) and one second wave (red) COVID-19 patient. Frequencies of cytokine-producing cells are indicated. (D) Frequency of SARS-CoV-2-specific CD4 or CD8 T cells producing IFN-γ, TNF-α or IL-2, in first wave (n = 22, blue) and second wave (n = 22, red) COVID-19 patients. Bars represent medians of responders. Statistical analyses were performed using the Mann-Whitney test between T cell responders from the first and second wave and the Wilcoxon test between CD4 and CD8 responders. (E) Comparison of polyfunctional profiles of SARS-CoV-2-specific CD4 T cells in first and second wave patients. (F) Comparison of polyfunctional profiles of SARS-CoV-2-specific CD8 T cells in first and second wave patients. The medians are shown. Each response pattern is color-coded, and data summarized in the pie charts. Statistical differences between pies were defined using a permutation test.

Fig. 2.. T cell recognition of WT SARS-CoV-2 spike (S). nucleocapsid (N) and membrane (M) proteins in first and second wave convalescent COVID-19 patients.

(A) Clinical characteristics of convalescent COVID-19 patients recruited during the first and second wave of the COVID-19 pandemic. *: median and interquartile range. ^$: Disease severity was defined based on oxygen therapy requirement according to the WHO ordinal scale scoring system. (B) Summary graph of the frequency of ancestral SARS-CoV-2 S-, N- or M-specific CD4 or CD8 T cells, producing IFN-γ, TNF-α or IL-2, in first wave (n = 10, light blue) and second wave (n = 14, orange) convalescent COVID-19 patients. Due to limited cell availability, T cell responses to M were tested in 10 first wave participants and 7 s wave participants. The proportion of participants exhibiting a detectable CD8 response is indicated. Bars represent medians of responders. Statistical analyses were performed using the Mann-Whitney test. (C&D) Polyfunctional profiles of ancestral Spike-specific CD4 and CD8 T cells in first and second wave convalescent COVID-19 patients. Medians and IQR are shown. Each response pattern is color-coded, and data are summarized in pie charts. Statistical differences between pies were defined using a permutation test.

Fig. 3.. Loss of recognition of SARS-CoV-2 Beta variant epitopes and neutralizing antibody responses.

(A) Representative flow cytometry plots of IFN-γ production by CD4 T cells in response to ancestral full spike peptide pool (Full spike), and smaller pools spanning the mutated regions of ancestral (WT pool) or Beta spike (Beta pool) in two first wave (blue) and two second wave (red) COVID-19 patients. Frequencies (%) of IFN-γ positive cells are indicated. (B) The frequency of IFN-γ-producing SARS-CoV-2-specific CD4 T cells in first wave (n = 22, left) and second wave (n = 22, right) COVID-19 patients. The proportion of patients exhibiting a detectable response to the different peptide pools (i.e., responders) is indicated at the bottom of each graph. (C) Plasma samples from COVID-19 patients recruited during the first (n = 18) or the second wave (n = 19) were tested for neutralization cross-reactivity against ancestral or Beta pseudoviruses. The threshold of detection was a 50% inhibitory dilution (ID₅₀) of 20. Gray dots indicate patients who displayed a detectable CD4 T cell response to WT pool, selectively covering the variable regions of spike, and lost recognition to the Beta pool. Neutralization data on the second wave cohort are from (25). (D) Fold-change in neutralization titers is shown for data in c. Bars represent medians. Statistical analyses were performed using the Wilcoxon test and the Fisher’s exact-squared test.

Fig. 4.. Infrequent recognition of SARS-CoV-2 ancestral or Beta variant spike epitopes by CD8 T cells.

(A) Representative flow cytometry plots of IFN-γ production by CD8 T cells in response to ancestral full spike peptide pool (Full Spike), and pools covering the mutated regions of ancestral spike (WT pool) or Beta spike (Beta pool) in two first wave (blue) and two second wave (red) COVID-19 patients. Frequencies (%) of IFN-γ positive cells are indicated. (B) Frequency of IFN-γ-producing SARS-CoV-2-specific CD8 T cells in first wave (n = 22, left) and second wave (n = 22, right) patients. The proportion of responders is indicated. Bars represent medians. Statistical analyses were performed using the Wilcoxon test.

Fig. 5.. Identification of SARS-CoV-2 spike epitopes targeted by CD4 T cells.

(A) Representative flow plots of IFN-γ production by CD8 and CD4 T cells in response to the Beta pool, WT pool and peptide pairs containing the spike 6-25 sequence (containing L18), the 73-92 sequence (containing D80) and the 206-225 sequence (containing D215) in three first wave patients. HLA Class-II alleles of each participant are listed on the right. (B) Number of tested first wave participants (n = 6) exhibiting a response to Beta pool, WT pool and each of the peptide pairs tested individually. (C) Frequency of IFN-γ+ CD4 T cells in response to indicated stimuli. Each participant is depicted by a different color.