Loss of naive T cells and repertoire constriction predict poor response to vaccination in old primates

Abstract

Aging is usually accompanied by diminished immune protection upon infection or vaccination. Although aging results in well-characterized changes in the T cell compartment of long-lived, outbred, and pathogen-exposed organisms, their relevance for primary Ag responses remain unclear. Therefore, it remains unclear whether and to what extent the loss of naive T cells, their partial replacement by oligoclonal memory populations, and the consequent constriction of TCR repertoire limit the Ag responses in aging primates. We show in this study that aging rhesus monkeys (Macaca mulatta) exhibit poor CD8 T cell and B cell responses in the blood and poor CD8 responses in the lungs upon vaccination with the modified vaccinia strain Ankara. The function of APCs appeared to be maintained in aging monkeys, suggesting that the poor response was likely intrinsic to lymphocytes. We found that the loss of naive CD4 and CD8 T cells, and the appearance of persisting T cell clonal expansions predicted poor CD8 responses in individual monkeys. There was strong correlation between early CD8 responses in the transitory CD28+ CD62L- CD8+ T cell compartment and the peak Ab titers upon boost in individual animals, as well as a correlation of both parameters of immune response to the frequency of naive CD8+ T cells in old but not in adult monkeys. Therefore, our results argue that T cell repertoire constriction and naive cell loss have prognostic value for global immune function in aging primates.

Figure 1. Age-related decline in RM primary CD8 response to vaccination

8 experimental cohorts of Old (n=28) and adult (n=21) RM were vaccinated with MVA as described in methods. MVA-specific antibody responses were calculated and displayed as fold increase of ELISA titers over baseline (panelA). Cohorts of young and old animals were compared at indicated time points post prime (Upper panel) or boost (lower panel) by repeated measures ANOVA, and assessed for significance (n.s.-p>0.05; *-p<0.05). Histograms indicate group means, error bars show Standard errors of the mean (SEM). (B) Kinetic of IFNγ CD8+ cell responses in blood on days 7, 14, 28 and 42. Cells were in vitro stimulated with VACV WR at MOI=1 for 15h, followed by additional 2h incubation with BrefeldinA. Upon surface staining with αCD4 and αCD8, cells were fixed, permeablized and stained for cytokine expression. Frequency of IFNγ⁺ cells in uninfected controls were subtracted from vaccinia-infected samples to eliminate background IFNγ expression. Results are shown as means±SEM from three pooled experiments. p-value from repeated measures ANOVA is indicated. (C) Kinetic of IFNγ responding CD8+ cells in BAL on days 14, 28 and 42 of adult and old monkeys. Note that age-related differences were more pronounced in BAL than in blood (see panel B). Pooled means±SEM from three independent experiments and p-value from repeated measures ANOVA are shown. (D) Three representative contour plots of IFNγ (y axes) and TNFα (x axes) expression in control CD8 T-cells (upper panels – CTRL) or vaccinia stimulated cells (lower panels – VACV)

Fig. 2. No age-related differences in number and migration efficacy of DC

(A) Skin bioptic samples were stained with HLA-DR antibodies to define DC, and DC per visual field were counted in duplicate, with 10 fields/replicate. Median values and interquartile ranges for adult (n=11) and old (n=9) monkeys are shown. (B) Dendritic cells were enriched from blood CD14+ cells by a 5-day cultivation in GM-CSF and IL-4 media, and stimulated with 1µg/ml CD40L for 48h. IL-12 release in supernatants was measured by ELISA and normalized to cell numbers. Median values and interquartile ranges for adult (n=13) and old (n=22) monkeys are shown. (C) DC obtained as in panel B were used to process and present MHC1b- restricted peptides to the indicated CD8 cell clones. Functional CD8 responses to DC presentation were measured by ELISPOT, and normalized to reflect fold increases over background. Geometric means and 95% confidence intervals are shown for old (n=36) and adult (n=18) cohorts.

Fig. 3. Age-related differences in CD8 naïve pool size and stable TCE occurrence correlate inversely

(A, B). Three weeks before vaccination, blood lymphocytes were analyzed by FCM for the frequency and absolute count of naïve CD4 and CD8 T-cells Naive CD4 T-cells (left panels), and naïve CD8 T-cells (right panels) defined by restrictive progressive gating (via CD4 or CD8 , then through the CD28hiCD95lo gate) were quantified in individual monkeys in terms of their frequency with CD4 or CD8 pools (A), or in terms of their number per ml of blood (B). Symbols indicate the naïve T-cell percentage in individual old or adult monkeys, horizontal lines show means. p-values reflect Wilcoxon-Mann-Whitney test results. (C) cDNA from blood lymphocytes was analyzed yearly by PCR for TCR length polymorphism in each of the 24 V regions of the β TCR chain for 4 consecutive years, and Vβ families exhibiting consistently a single PCR band for at least the last two time points were defined as TCE⁺. Symbols indicate percentages of TCE⁺ Vβ families in individual monkeys, horizontal lines show means. (D) Naïve cell frequencies (x-axis) were correlated to the percentage of TCE⁺ Vβ families (y-axis) in individual adult (black diamonds) or old (white triangles) monkeys. A semilogarithmic correlation index for combined groups is indicated.

Fig. 4. Age-related difference in TCE and naïve CD8 frequency predicts antigen specific responses

(A) Frequency of naïve blood CD8 lymphocytes prior to MVA immunization (x-axis) was correlated to peak (day28) CD8 response in BAL (y-axis) in individual adult (●) or old monkeys (△).(B) Peak tissue CD8 responses from monkeys with at least one Vβ family with a clonal expansion (TCE+) were compared to those without any TCE (TCE−). Adult (●) and old (△) monkeys belonging to either TCE group are indicated by respective symbols, horizontal lines indicate means.

Fig. 5. Early CD8 responses to MVA are attenuated in old monkeys

(A) The frequency of IFNγ responders in CD28⁺ and CD28⁻ subsets of individual adult monkeys at day 7 post vaccination was pairwise compared by Wilcoxon signed rank test and p value is shown. Similar results (p=0.0039) were observed in old animals (not shown). (B) In vitro VACV stimulation was followed by FCM as in figure 1B. The frequency of CD28⁺ and CD28⁻ cells in CD62L⁻ IFNγ responders was calculated for each individual adult monkey and time point (see supportive figure 3 for gating). Means and SEM for each time point are shown. (C) The frequency of IFNγ responders in CD28⁺ subsets of adult or old animals was compared by Mann-Whitney test and p value is shown. Symbols indicate values relative to individual monkeys, horizontal bars indicate means. (D) Frequency of naïve CD8 lymphocytes prior to MVA immunization (x-axis) was correlated to IFNγ responses in CD62L⁻CD28⁺ cells at day 7 following infection (y-axis) in animals with less than 20% of naïve CD8 cells prior to vaccination. Symbols show values in individual adult (●) or old monkeys (△), the line shows the linear regression curve. (E) Naïve CD8 counts per ml blood prior to MVA immunization (x axis) was correlated to IFNγ responses in CD62L⁻CD28⁺ cells at day 7 following infection (y-axis) in animals old animals. Symbols show values in individual animals, the line shows the linear regression curve. (F) Immediate CD28⁺CD62L⁻ CD8⁺ T-cell IFNγ responses from Figure 2, panel C (x-axis) were correlated to antibody titers at day 14 post boost (y-axis). Values for individual adult (●) or old (△) animals, and the linear regression curve for combined datasets are shown.