Aging and CMV Infection Affect Pre-existing SARS-CoV-2-Reactive CD8+ T Cells in Unexposed Individuals

Abstract

Age is a major risk factor for COVID-19 severity, and T cells play a central role in anti-SARS-CoV-2 immunity. Because SARS-CoV-2-cross-reactive T cells have been detected in unexposed individuals, we investigated the age-related differences in pre-existing SARS-CoV-2-reactive T cells. SARS-CoV-2-reactive CD4⁺ T cells from young and elderly individuals were mainly detected in the central memory fraction and exhibited similar functionalities and numbers. Naïve-phenotype SARS-CoV-2-reactive CD8⁺ T cell populations decreased markedly in the elderly, while those with terminally differentiated and senescent phenotypes increased. Furthermore, senescent SARS-CoV-2-reactive CD8⁺ T cell populations were higher in cytomegalovirus seropositive young individuals compared to seronegative ones. Our findings suggest that age-related differences in pre-existing SARS-CoV-2-reactive CD8⁺ T cells may explain the poor outcomes in elderly patients and that cytomegalovirus infection is a potential factor affecting CD8⁺ T cell immunity against SARS-CoV-2. Thus, this study provides insights for developing effective therapeutic and vaccination strategies for the elderly.

Keywords: COVID-19; SARS-CoV-2; T cell aging; T-cell immunity; cross-reactive T cells; cytomegalovirus; senescent T cells.

FIGURE 1

SARS-CoV-2-specific T-cell responses in unexposed young and elderly individuals. PBMCs isolated from 30 young and 26 elderly blood samples were stimulated for 20 h with DW or overlapping peptides containing the SARS-CoV-2 S, N, and M protein sequences. Antigen-reactive T cells were identified by flow cytometry (FCM) according to the gating strategy and staining presented in Supplementary Figure 2A and Supplementary Table 1. (A) Representative FCM plots displaying AIMs (CD137 and IRF4) on CD4⁺ and CD8⁺ T cells after stimulation with the negative control (DW) or SARS-CoV-2 peptide pool (SCoV-2). Numbers indicate the population percentages in the gates. (B) Percentages of AIM⁺ (CD137⁺IRF4⁺) T cells in CD4⁺ and CD8⁺ T cells between the negative control (DW) and SARS-CoV-2 peptide stimulated samples (SCoV-2). (C) Percentages of AIM⁺ (CD137⁺IRF4⁺) T cells in CD4⁺ and CD8⁺ T cells and their numbers (×10⁶/L). Data were background subtracted against the DW and are shown as the median ± interquartile range (IQR). (D) Stimulation index (SI) quantification of AIM⁺CD4⁺ or AIM⁺CD8⁺ T cells; the same samples as in Figure 1C were analyzed. The cutoff value was defined as >3, as indicated by the dashed lines. (B-D) Each dot represents one donor. Pairwise comparisons were performed using Wilcoxon’s test. Statistical comparisons across cohorts were performed using the Mann-Whitney test. **p < 0.01, ***p < 0.001. n.s., not significant.

FIGURE 2

opt-SNE mapping of SARS-CoV-2-reactive T cells from young and elderly individuals. The opt-SNE plots were generated using all the markers described in the middle and bottom rows (CCR7, CD45RA, CD28, CD57, PD-1, and IRF4). (Top) opt-SNE plots showing the clustering of AIM⁺CD4⁺ (A) or AIM⁺CD8⁺ (B) T cells from young (blue) (n = 30) and elderly (red) (n = 26) cohorts. Background plots shown in gray represent total AIM⁺CD4⁺ (A) and AIM⁺CD8⁺ (B) T cells from all 56 samples. Presumable naïve phenotype (NP), central memory (CM), effector memory (EM), and terminally differentiated effector memory cells re-expressing CD45RA (TEMRA) were manually gated based on the expression levels of CD45RA and CCR7. (Middle and Bottom) opt-SNE plots showing the expression of individual markers. See Supplementary Figure 3.

FIGURE 3

Immunophenotyping of SARS-CoV-2-reactive CD4⁺ T cells from young and elderly individuals. (A, C) Representative flow cytometry plots showing the expressions of CD45RA and CCR7 in total CD4⁺ (A) and CD4⁺ AIM⁺ (CD137⁺IRF4⁺) (C) CD3⁺ T cells after stimulation with the SARS-CoV-2 peptide pool in the AIM assay. The subset definitions and gating strategies are outlined in Supplementary Figures 2A, 4A. Numbers indicate percentages in the drawn gates. (B, D) Frequency (left panels) and calculated number in whole blood (right panels) of NP, CM, EM and TEMRA cells in total CD4⁺ (B) and SARS-CoV-2-specific AIM⁺CD4⁺ (D) T cells in the young (n = 30) and elderly (n = 26) cohorts. (Left panel of D) Samples that a percentage of AIM⁺ cells in CD4⁺ T cells of at least 0.001% are indicated (young, n = 27 and elderly, n = 26). Data are shown as the median ± IQR. Statistical comparisons across cohorts were performed with the Mann-Whitney test. (E) Correlation between the percentage of each fraction in the total CD4⁺ T cell population and of AIM⁺CD4⁺ T cells in the NP, CM, EM, and TEMRA fractions. Correlation coefficients (R) were calculated using the Spearman rank correlation test. Linear approximations are plotted in figures. (B, D, E) Each dot represents one donor. Young (blue circles) and elderly (red circles). *p < 0.05; **p < 0.01; ***p < 0.001. n.s., not significant.

FIGURE 4

SARS-CoV-2-specific cytokine production and proliferation from young and elderly CD4⁺ T cells. PBMCs from the young (n = 30) and elderly (n = 25) cohorts were stimulated and cultured with the SARS-CoV-2 peptide pool (SCoV-2) or a negative control (DW) and analyzed after 6 days. One sample from the elderly cohort was excluded because of the low number of PBMCs obtained. (A) Percentages of IL-2⁺, IFNγ⁺, TNFα⁺, and IL-17A⁺ cells in the CD4⁺ T cell population from the young (blue circles) and elderly (red circles) cohorts. Each dot represents one donor. Data were background subtracted against DW and are shown as the median ± IQR. (B) Representative flow cytometry plots showing CTV and FSC gating of total CD4⁺ T cells. The boxed gates define CTV^lowFSC^high cells (left). Percentages of CTV^lowFSC^high T cells in CD4⁺ T cells between DW and SARS-CoV-2 peptide pool stimulation (SCoV-2) in the young and elderly cohorts (middle). Frequency of CTV^lowFSC^high T cells in CD4⁺ T cells. Data were background subtracted against DW and are shown as the median ± IQR (right). Each dot represents one donor (left and middle). Pairwise comparisons were performed using Wilcoxon’s test. Statistical comparisons across cohorts were performed using the Mann-Whitney test. ***p < 0.001. n.s., not significant. See Supplementary Figures 5.

FIGURE 5

Immunophenotyping of SARS-CoV-2-reactive CD8⁺ T cells from young and elderly individuals. (A,C) Representative flow cytometry plots showing the expression of CD45RA and CCR7 among total CD8⁺ (A) and CD8⁺CD137⁺IRF4⁺ (C) T cells after stimulation with the SARS-CoV-2 peptide pool in the AIM assay. Subset definitions and gating strategies are outlined in Supplementary Figures 2A, 4A. Numbers indicate population percentages in the gates. (B,D) Frequency (left panels) and calculated total number in the whole blood sample (right panels) of NP, CM, EM, and TEMRA cells in total CD8⁺ (B) an SARS-CoV-2-reactive AIM⁺ CD8⁺ (D) T cells in the young (n = 30) and elderly (n = 26) cohorts. (Left panel of D) Samples that a percentage of AIM⁺ cells in CD8⁺ T cells of at least 0.001% are indicated (young, n = 22 and elderly, n = 24). Data are shown as the median ± IQR. Statistical comparisons across cohorts were performed using the Mann-Whitney test. (E) Correlation between percentages of each fraction in total CD8⁺ T cells and of AIM⁺CD8⁺ T cells in the NP, CM, EM, and TEMRA fractions. Correlation coefficients (R) were calculated with the Spearman rank correlation test. Linear approximations are plotted in figures. (B,D,E) Each dot represents one donor. Young (blue circles) and elderly (red circles). *p < 0.05; **p < 0.01; ***p < 0.001. n.s., not significant.

FIGURE 6

SARS-CoV-2-specific cytokine production and proliferation from young and elderly CD8⁺ T cells. PBMCs from the young (n = 30) and elderly (n = 25) cohorts were stimulated and cultured with the SARS-CoV-2 peptide pool (SCoV-2) or negative control (DW) and analyzed after 6 days. One sample from the elderly cohort was excluded because of the low number of PBMCs obtained. (A) Percentages of IFNγ⁺ and TNFα⁺ cells in CD8⁺ T cells from the young (blue circles) and elderly (red circles) cohorts. Data were background subtracted against DW and are shown as the median ± IQR. (B) Representative flow cytometry (FCM) plots showing CTV and FSC gating of total CD8⁺ T cells. Boxed gates define CTV^lowFSC^high cells (left). Percentages of CTV^lowFSC^high T cells in CD8⁺ T cells between the DW and SARS-CoV-2 peptide pool stimulation in young and elderly cohorts. (middle). Percentages of CTV^lowFSC^high T cells in CD8⁺ T cells. Data were background subtracted against DW and are shown as the median ± IQR (right). (C) Representative FCM plots showing the expression of CD57 and CD28 in AIM⁺ CD8⁺ T cells in the AIM assay. Numbers indicate percentages in the drawn gates (left). Percentages of CD28⁻ and CD57⁺ cells in SARS-CoV-2-specific AIM⁺ CD8⁺T cells from young and elderly cohorts. Data are shown as the median ± IQR (right). (A-C) Each red or blue dot represents one donor. Pairwise comparisons were performed using Wilcoxon’s test. Statistical comparisons across cohorts were performed using the Mann-Whitney test. *p < 0.05, ***p < 0.001. n.s., not significant. See Supplementary Figure 6.

FIGURE 7

Immunophenotyping of SARS-CoV-2-reactive CD8⁺ T cells from young CMV⁺ or CMV⁻ individuals. (A,B) Percentages of NP, CM, EM and TEMRA cells in total CD8⁺ (A) or in SARS-CoV-2-reactive AIM⁺CD8⁺ T cells (B) from CMV seronegative (CMV(−)) and CMV seropositive (CMV(+)) individuals. (C) Frequency of CD28⁻ and CD57⁺ cells among SARS-CoV-2-specific AIM⁺CD8⁺ T cells from CMV⁺ (n = 14) and CMV⁻ (n = 16) individuals. Each dot represents one donor. Data are shown as the median ± IQR. Statistical comparisons across cohorts were performed using the Mann-Whitney test.