Human memory T cells with a naive phenotype accumulate with aging and respond to persistent viruses

Abstract

The number of naive T cells decreases and susceptibility to new microbial infections increases with age. Here we describe a previously unknown subset of phenotypically naive human CD8(+) T cells that rapidly secreted multiple cytokines in response to persistent viral antigens but differed transcriptionally from memory and effector T cells. The frequency of these CD8(+) T cells, called 'memory T cells with a naive phenotype' (TMNP cells), increased with age and after severe acute infection and inversely correlated with the residual capacity of the immune system to respond to new infections with age. CD8(+) TMNP cells represent a potential new target for the immunotherapy of persistent infections and should be accounted for and subtracted from the naive pool if truly naive T cells are needed to respond to antigens.

Figure 1. Age-related increase in phenotypically naive T cells capable of rapid cytokine production

(a) Intracellular IFN-γ staining of human PMA/iono- activated PBMCs from representative healthy adult, 32 y old, and old, 76 y old, donors. Overlaid dot plots show % of naive CD45RA⁺CCR7⁺ (yellow gate) and subsequently CD45RA⁺CCR7⁺CD95^loCD28^int (red gate) CD8 T cells producing IFN-γ⁺ (in blue) overlaid over total CD8⁺ T cells (in green). b. Percentage of CD8⁺ naïve phenotype CD45⁺CCR7⁺CD95^loCD28^int cells producing IFN-γ across age. Data are from one of two experiments. t-test was used to show significance in the increase between the group of 21–40 years old (n=27); 40 – 65 years old (n=44) and 66–97 year old donors (>65 years old; n=21; *p<0.05; **p<0.001). c. Bar graph showing cytokine production by phenotypically naive CD45RA⁺ CCR7⁺CD95^loCD28^int CD8⁺ T cells from freshly isolated blood (age 32–76;n=7).

Figure 2. SPADE analysis of CD8⁺ T cells demonstrates clustering of T_MNP cells with T_N group

Two representative subjects (adult, age 27, left; and old, age 68, right) are shown. a. IFN-γ producing cells form a unique naive CD8⁺ T cell subset. Node size depicts the relative count and the color represents IFN-γ staining intensity. Four canonical populations are outlined in black, CD45RA⁺CCR7⁺CD95^loCD28^int- defined naive population in red, and IFN-γ⁺ naive population in bright green. b. Naive phenotype of IFN-γ producing T_MNP CD8⁺T cells. Colors represent staining intensity of phenotypic markers indicated in the legend. Data are representative of a single experiment (n=9, 6 males, 3 females, age 27–78).

Figure 3. T_MNP cells form a unique T cell subset with naive phenotype that expresses higher levels of CD49d and CXCR3

a. Flow cytometry analysis of human PMA/iono-activated PBMCs from a representative donor (age 54). Dot plots show gating strategy used to identify T_MNP CD8⁺ T cells based on the expression of CD45RA⁺ CCR7⁺CD95^loCD28^int and IFN-γ production (four canonical populations are outlined on CCR7 vs CD45RA plot). b,c. Overlaid histogram plots show expression level [mean geometric mean fluorescence intensity (gMFI) ± standard error of the mean (SEM)] of an indicated surface antigen in different CD8⁺ T cell subsets from a representative donor from a. d. Repeated measures one-way ANOVA was used to quantify differences in CD49d and CXCR3 expression between T_N and T_MNP CD8 T cells (*p<0.01,**P<0.001; data are representative of two experiments, n=13 and 12, age 27–85).

Figure 4. T_MNP cells show signs of prior activation

a. CD8⁺T_MNP cell % after 3h stimulation with indicated agonistic Ab (left) and as a fraction of PMA+iono response (±SD, n=10, 26–68y, repeated once). b. Intracellular cytokine expression by CD8⁺T cell subsets (n=36, 24–82y). c. Left, 5% contour plots show representative degranulation (CD107a⁺) and Perforin(Perf) production by IFN-γ⁺CD45RA⁺CCR7⁺CD95^loCD28^intT_MNPCD8⁺ T cells (black) overlaid over total CD8⁺ T cells (gray) (n=2, male, 52,63y of n=8, 31–77y, repeated twice). Right, cumulative frequencies of CD107a⁺ and Perf⁺CD8⁺T_MNP cells as percentage of IFNγ⁺GzmB⁺T_MNP (n=8, 31–77y; repeated twice). d. Day 7 proliferation of sorted CD8⁺ T cell subsets in response to 50 ng/mL of IL-7 and IL-15 (subjects as in C; n=4/8 had all memory subsets as controls). e. Phenotype changes 3d after cytokine culture in 2 of 3 analyzed subjects (male, 52&64y). f. Mean gMFI±SEM of T-bet expression in different CD8⁺ T cell subsets before or 2h after PMA+iono activation (significance by repeated measures one- way ANOVA between populations, ns= not significant, *p<0.05,**p<0.01, ***p<0.0001, n=9, 6 males, 3 females, age 27–78). g. Time course of pErk expression (right, gMFI shown) in T_MNP (CCR7⁺CD45RA⁺CD95^loCD28^intCD49d^hi) vs T_N (CCR7⁺CD45RA⁺CD95^loCD28^intCD49d^lo) CD8⁺ T cells following 3h PMA+iono stimulation in a representative 68y old female. Left, mean pErk gMFI ±SD (repeated measures one-way ANOVA, n=3, 2 females, 1 male, of n=7, 67–83y,*p<0.05; repeated once). h. Dose-dependent decrease in IFN-γ production by PMA+iono stimulated CD8⁺ T cells after pre-incubation with Erk inhibitor (n=14, 5f, 9m, 52–83y, repeated once). (Repeated measures one-way ANOVA; *p<0.05, ***p<0.0001).

Figure 5. Telomere lengths and RNA transcriptome profile of T_MNP cells

a. Representative FACS plot analysis showing telomere length of different CD8⁺ subsets (T_N (CCR7⁺CD45RA⁺CD95^loCD28^int CD49d^lo), T_MNP (CCR7⁺CD45RA⁺CD95^loCD28^int CD49d^hi)_, T_CM, T_E+EM, T_EMRA). Numbers indicate gMFI of Cy5 PNA telomere probe staining. Bar graph comparing telomere length of CD8⁺ subsets with the length of naive subset within each individual (n=9, 6 females, 3 males, ages 34–83). b. Plot of the samples projected onto the first two principal components with variables of all Consensus Coding Sequence (CCDS) listed genes that were measured. The color distinguished the cell type (indicated in the figure legend) and shape distinguishes the subject from which the sample was derived. c. Heat map of differentially expressed genes (adjP<0.01, FDR<0.05. The expression values for each gene are standardized to a z-score and then binned to a color.

Figure 6. T_MNP cells have skewed TCR Vβ repertoire and can be detected among T cells responding to stimulation with EBV, CMV, and HIV peptides

a. 3D line graphs showing TCR Vβ distribution in different CD8⁺ T cells subsets (T_N (CD45RA⁺ CCR7⁺CD95^loCD28^intCD49d^lo), T_MNP (CD45RA⁺CCR7⁺CD95^loCD28^int CD49d^hi), T_CM, T_E+EM and T_EMRA) from 3 representative donors (of n=11, 6 males, 5 females, ages 27–78) shown as percentages of each of the 24 tested TCR Vβ (marked on the x axis) are shown. b,c. Flow cytometry analysis of PBMCs at 3 h after stimulation with (b) EBV BLZF1peptide pool, CMV pp65 peptide pool or IAV matrix protein peptide pool and (c) HIV (Gag) peptide pool. Dot plots show gating strategy to evaluate the naive phenotype of the responding (IFN-γ producing) CD8⁺ T cells. Numbers represent percentages of the population within each gate after subsequent gating (IFN-γ⁺ CD8⁺T cells followed by CCR7⁺CD45RA⁺ gate and subsequently by CD95^lo CD28^int gate). Total CD8⁺ T cells are shown in gray. (b) Data are representative of one healthy donor with an n=8 (3 females, 5 males, ages 34–68) for each peptide stimulation; (c) Data show the response of three HIV+ ART donors with no detectable HIV in plasma.

Figure 7. T_MNP cells are found among CMV and EBV -specific T cell pools and are not specific for influenza, smallpox, or self-antigens

a,b. Flow cytometry analysis of IFN-γ production by naive (CD45RA⁺CCR7⁺CD95^loCD28^int) CMV (a) and EBV (b) specific CD8⁺ T at 3 h after stimulation with PMA+iono. Four representative donors (one CMV seropositive (male, age 72) and one seronegative (female, age 65) and two EBV seropositive (males, ages 67 and 75) (high and low titer) of n=26, ages 28–83 (a) or n=7, ages 24–75) (b) are shown. Dot plots show gating strategy and the numbers represent percentages of the population within each gate after subsequent gating c, d. Flow cytometry analysis of IFN-γ production by naive (CCR7⁺CD45RA⁺CD95^loCD28^int) vaccinia virus-specific CD8⁺ T cells from two representative donors (one CMV seropositive (female, age 64) and one seronegative (female, age 36) (of n=4, ages 36–74)) and naive Melan-A-specific CD8⁺ T cells from two representative donors, females, ages 58 and 76 (of n=6, ages 29–82) after brief stimulation with PMA+iono (same gating strategy was used as in a, b). Data are representatives of one (a,b,c) and two (d) experiments.

Figure 8. Individuals with symptomatic WNV infection show increased levels of T_MNP cells

a,b. Intracellular IFN-γ and GzmB staining of human PMA+iono- activated PBMCs from six representative WNV⁺ donors (three WNV asymptomatic (a) and three WNV symptomatic (b)). Dot plots show gating strategy to quantify T_MNP (IFNγ⁺CD45RA⁺ CCR7⁺CD95^loCD28^int) CD8⁺ T cells (shown as percentages of the population within each gate after subsequent gating and quantified as number (mean ±SEM) of T_MNP cells per million CD8⁺ T cells (c) (Mann-Whitney test, **p<0.01, n=26). Data are representative of one experiment). d. Quantification of T_MNP CD8⁺ T cells (number (mean ±SEM) of T_MNP (IFNγ⁺CCR7⁺CD45RA⁺CD95^loCD28^int) cells per million CD8⁺ T cells) from individuals with asymptomatic and symptomatic WNV infections early (day 2–7) and late (day 60–90) after index donation (Repeated measures ANOVA was used to compare symptomatic group to asymptomatic group at two time points. Prior to applying repeated measures, equal variance assumption was tested by Levene’s test. The heterogeneous Compound Symmetry covariance structure was used to not only account for the within subject correlation but also adjust for the variance heterogeneity. p=NS, *p<0.05, **p<0.01,(asymptomatic n=11, 39–92y old; symptomatic n=16, 43–84y). Data are representative of one experiment.