Evolution of SARS-CoV-2 T cell responses as a function of multiple COVID-19 boosters

Abstract

The long-term effects of repeated COVID-19 vaccinations on adaptive immunity remain incompletely understood. Here, we conducted a comprehensive three-year longitudinal study examining T cell and antibody responses in 78 vaccinated individuals without reported symptomatic infections. We observed distinct dynamics in Spike-specific humoral and cellular immune responses across multiple vaccine doses. While antibody titers incrementally increased and stabilized with each booster, T cell responses rapidly plateaued, maintaining remarkable stability across CD4+ and CD8+ subsets. Notably, approximately 30% of participants showed CD4+ T cell reactivity to non-Spike antigens, consistent with asymptomatic infections. Single-cell RNA sequencing revealed a diverse landscape of Spike-specific T cell phenotypes, with no evidence of increased exhaustion or significant functional impairment. However, qualitative changes were observed in individuals with evidence of asymptomatic infection, exhibiting unique immunological characteristics, including increased frequencies of Th17-like CD4+ T cells and GZMKhi/IFNR CD8+ T cell subsets. Remarkably, repeated vaccinations in this group were associated with a progressive increase in regulatory T cells, potentially indicating a balanced immune response that may mitigate immunopathology. By regularly stimulating T cell memory, boosters contribute to a stable and enhanced immune response, which may provide better protection against symptomatic infections.

Keywords: COVID-19; SARS-CoV-2; T cell responses; antibody responses; asymptomatic infection; booster vaccination; exhaustion; mRNA vaccines; scRNA-seq.

Figure 1.. Study design and prevalence of asymptomatic infections.

(A) Overall cohort surveillance for selection of 78 participants without COVID-19 symptoms or previous evidence of infection. (B) Participants categorization by the number of COVID-19 vaccine doses administered (color-coded). The number of donors included in each group is indicated and the groups are not mutually exclusive, allowing individual donors to be represented in multiple groups. (C) Multiple blood donations were collected from each subject at various timepoints (T) post-vaccination. The y axis represents each individual participant, and the x axis represents different time points at which their blood was collected in a longitudinal follow-up throughout the study timeline (Jan 2021-Jan 2024). Samples within the period of each vaccine dose are color-coded. Sampling timepoints include pre-vaccination (T1), 30 days after full vaccination (Dose 2) (T2), 30 days after the first booster (Dose 3) (T5), and 30 days after the second booster (Dose 4) (T8). Time intervals between vaccinations encompass T3, T6 and T9 (1–6 months after Dose 2, 3 and 4, respectively) and T4, T7 and T10 (6–12 months after Dose 2, 3 and 4, respectively). (D) CD4+ T cell responses to Spike and CD4RE MPs were measured as percentage of AIM+ (OX40+CD137+) CD4+ T cells and plotted in 2 dimensions as stimulation index (SI). Each dot represents a sample of a donor for a given timepoint. Percentages denote the frequency of samples that are either negative (CD4RE^NEG) or positive (CD4RE^POS) for CD4RE reactivity using a cut-off of SI=10 (dotted line).

Figure 2.. Evolution of antibody responses over repeated vaccinations and exposure.

(A) Longitudinal monitoring of RDB IgG levels across multiple vaccinations and timepoints. Each donor is represented by a colored dot and longitudinal samples for each individual donor connected by a gray line. Black bold line represents the geometric mean. Geometric mean value and % of donors with positive response for each timepoint is indicated. (B) Representation of paired RDB IgG levels before and after vaccination for each dose cohort for all samples (upper panel), or CD4RE^NEG (middle panel) and CD4RE^POS (lower panel) donors. Fold change (FC) is indicated in each graphic. (C) Cross-sectional RBD IgG responses from time of vaccination (days) for the different dose cohorts in donors associated with CD4RE^NEG (upper panel) or CD4RE^POS (lower panel) reactivity. t_1/2 is shown as the median half-life calculated based on linear mixed effects model. (A-C) Dotted lines indicate threshold of positivity. Data were analyzed for statistical significance using the paired Wilcoxon’s (B) or Spearman’s (C) test. p values are shown.

Figure 3.. Evolution of memory Spike-specific CD4+ T cell responses over repeated exposures and vaccinations.

(A) Longitudinal monitoring of AIM+ (OX40+CD137+) CD4+ T cell responses across multiple vaccinations and timepoints. Each donor is represented by a colored dot and longitudinal samples for each individual donor connected by a gray line. Black bold line represents the geometric mean. Geometric mean value and % of donors with positive response for each timepoint is indicated. (B) Representation of paired AIM+ (OX40+CD137+) CD4+ T cell responses before and after vaccination for each dose cohort for all samples (upper panel), or CD4RE^NEG (middle panel) and CD4RE^POS (lower panel) donors. Fold change (FC) is indicated in each graphic. (C) Cross-sectional AIM+ (OX40+CD137+) CD4+ T cell responses from time of vaccination (days) for the different dose cohorts in donors associated with CD4RE^NEG (upper panel) or CD4RE^POS (lower panel) reactivity. t_1/2 is shown as the median half-life calculated based on linear mixed effects model. (A-C) Dotted lines indicate threshold of positivity. Data were analyzed for statistical significance using the paired Wilcoxon’s (B) or Spearman’s (C) test. p values are shown.

Figure 4.. Evolution of memory Spike-specific CD8+ T cell responses over repeated exposures and vaccinations.

(A) Longitudinal monitoring of AIM+ (CD69+CD137+) CD8+ T cell responses across multiple vaccinations and timepoints. Each donor is represented by a colored dot and longitudinal samples for each individual donor connected by a gray line. Black bold line represents the geometric mean. Geometric mean value and % of donors with positive response for each timepoint is indicated. (B) Representation of paired AIM+ (CD69+CD137+) CD8+ T cell responses before and after vaccination for each dose cohort for all samples (upper panel), or CD4RE^NEG (middle panel) and CD4RE^POS (lower panel) donors. Fold change (FC) is indicated in each graphic. (C) Cross-sectional AIM+ (CD69+CD137+) CD8+ T cell responses from time of vaccination (days) for the different dose cohorts in donors associated with CD4RE^NEG (upper panel) and CD4RE^POS (lower panel) reactivity. t_1/2 is shown as the median half-life calculated based on linear mixed effects model. (A-C) Dotted lines indicate threshold of positivity. Data were analyzed for statistical significance using the paired Wilcoxon’s (B) or Spearman’s (C) test. p values are shown.

Figure 5.. Spike-Specific CD4+ and CD8+ T Cell subsets are highly diverse and phenotypically stable upon multiple vaccinations.

Single-cell transcriptomes of sorted AIM+ (A) CD4+ or (B) CD8+ T cells displayed by uniform manifold approximation and projection (UMAP). Cells are colored according to cluster identity as defined by unbiased clustering of 608,720 CD4+ T cells and 460,937 CD8+ T cells. Annotations of cell populations are shown for each cluster. (C) Plots show normalized proportion of spike-specific CD4+ T cells in each cluster as function of different vaccine doses. Proportions were normalized according to minimum cell totals across vaccine doses to ensure comparability. (D) Pie charts represent the relative frequency of spike-specific CD4+ T cell clusters within each dose cohort in CD4RE^NEG (upper panel) or CD4RE^POS (lower panel) samples. (E) Proportion of spike-specific CD8+ T cells in each cluster as function of different vaccine doses. Proportions were normalized as in (C). (F) Pie charts represent the relative frequency of spike-specific CD8+ T cell clusters within each dose cohort in CD4RE^NEG (upper panel) or CD4RE^POS (lower panel) samples.

Figure 6.. T cell functionality and exhaustion profiles remain stable after booster vaccination.

Graphs show relative expression of relevant genes cross-sectionally across different vaccine doses for (A) CD4+ Th1 cells (cluster 2; top panel) and CD4+ Tfh cells (cluster 3; lower panel), (B) CD8+ GZMKhi cells (cluster 0; upper panel), and CD8+ TEFF cells (cluster 3; lower panel) or (C) CD8+ TEX cells (cluster 1). Each color-coded gene is indicated. (A-C) Data were analyzed for statistical significance using the Kruskal-Wallis test followed by Dunn’s post-hoc test with FDR correction. (D) GSEA for exhaustion signature genes (Exhaustion consensus) across vaccine doses in spike-specific TEX CD8+ T cells. P-value and normalized enrichment score (NES) are indicated. (E-F) Graphs represent the relative frequency of CD8+ TEX (cluster 1) cells across timepoints for the indicated vaccine doses in all samples (E) or in samples associated with negative or positive CD4RE reactivity (F). Data were analyzed for statistical significance using the unpaired Mann-Whitney test. p values are shown.

Figure 7.. Repeated vaccinations limit peripheral Tfh cell expansion

(A) Graphs represent the relative frequency of Tfh cells in multiple vaccinations and timepoints. (B, C) Graphs show relative gene expression of (B) CXCL13 and (C) CXCR5 cross-sectionally across different vaccine doses and timepoints (months from vaccination). Color-coded dots represent the median response for each vaccine dose and timepoint. Interquartile range is represented. Data were analyzed for statistical significance comparing timepoints before and after each vaccination, using the unpaired Mann-Whitney test. * p = 0.0056; ** p = 0.0071; *** p = 0.0076. (D) Volcano plot shows differentially expressed genes in Tfh cells before (T1) and 30 days after (T2) Dose 2 vaccination. Vertical dotted lines indicate log2 fold change (LFC) cutoff, to call differentially expressed genes. Number of significant genes up- or down-regulated are indicated and dots are color-coded accordingly.

Figure 8.. Asymptomatic infections are associated with increased Th17-like and GZMKhi/IFNR cell phenotypes concurrent with Tregs expansion over repeated vaccinations.

(A) Graphs represent the relative frequency of CD4+ Th17 cells in CD4RE negative and positive samples across doses and timepoints. (B) Graphs represent the relative frequency of CD8+ GZMKhi, LTBhi and IFNR cells in CD4RE negative and positive samples across doses and timepoints. (C, D) Graph represents the relative frequency of CD4+ Tregs in different vaccine doses across samples associated with (C) positive or (D) negative CD4RE reactivity. Data were analyzed for statistical significance using the unpaired Mann-Whitney test. p values are shown.