NKG2D expression in CD4+ T lymphocytes as a marker of senescence in the aged immune system

Abstract

Human aging is characterized by changes in the immune system which have a profound impact on the T-cell compartment. These changes are more frequently found in CD8+ T cells, and there are not well-defined markers of differentiation in the CD4+ subset. Typical features of cell immunosenescence are characteristics of pathologies in which the aberrant expression of NKG2D in CD4+ T cells has been described. To evaluate a possible age-related expression of NKG2D in CD4+ T cells, we compared their percentage in peripheral blood from 100 elderly and 50 young adults. The median percentage of CD4+ NKG2D+ in elders was 5.3% (interquartile range (IR): 8.74%) versus 1.4% (IR: 1.7%) in young subjects (p < 0.3 × 10(-10)). CD28 expression distinguished two subsets of CD4+ NKG2D+ cells with distinct functional properties and differentiation status. CD28+ cells showed an immature phenotype associated with high frequencies of CD45RA and CD31. However, most of the NKG2D+ cells belonged to the CD28(null) compartment and shared their phenotypical properties. NKG2D+ cells represented a more advanced stage of maturation and exhibited greater response to CMV (5.3 ± 3.1% versus 3.4 ± 2%, p = 0.037), higher production of IFN-γ (40.56 ± 13.7% versus 24 ± 8.8%, p = 0.015), lower activation threshold and reduced TREC content. Moreover, the frequency of the CD4+ NKG2D+ subset was clearly related to the status of the T cells. Higher frequencies of the NKG2D+ subset were accompanied with a gradual decrease of NAIVE and central memory cells, but also with a higher level of more differentiated subsets of CD4+ T cells. In conclusion, CD4+ NKG2D+ represent a subset of highly differentiated T cells which characterizes the senescence of the immune system.

Fig. 1

CD4+ NKG2D+ T cells in peripheral blood from young and elderly individuals. (a) Representative dot plots showing the frequency of NKG2D expression in CD4+ T cells in elderly and young individuals. (b) Percentages of CD4+ T cells expressing NKG2D in a group of 100 elderly subjects compared with a group of 50 young subjects. Whole blood was stained with CD45-FITC/NKG2D-PE/CD3-PerCP/CD4-APC and 10⁵ cells were acquired in each experiment. Frequencies of NKG2D+ cells in gated CD45+ CD3+ CD4+ lymphocytes were analyzed. Outlier values are represented by circles and extreme values by stars, and were calculated by adding 1.5 and 3 times the IR to the 75th percentile, respectively. The horizontal dotted line illustrates the 75th percentile in young donors (2.3%). The non-parametric Mann-Whitney U method was used to compare frequencies between groups

Fig. 2

Distribution of CD4+ and CD8+ T cells into naïve, central memory, effector memory, and effector memory RA. CD45RA and CCR7 expression was analysed by flow cytometry in isolated CD4+ T cells from young people (n = 20) and elderly subjects with different frequencies of NKG2D (<5%, 5–20%, and >20%; n = 20 in each group). a Schematic model of the T cell differentiation subsets according to CD45RA and CCR7 expression. b Representative dot plots of the subsets defined by CD45RA and CCR7 expression for individuals in each group. c Histograms depict cell percentage in each subset (mean ± SEM) in young (white bars) and elderly subjects according to NKG2D expression (light grey bars <5%, dark grey bars 5–20%, black bars >20%). Significant differences between subsets are indicated (Student’s t test or non-parametric Mann–Whitney method). Asterisks represent significant differences with all the groups (p < 0.05)

Fig. 3

Distribution of EM and EMRA CD4+ T cells into subsets defined by CD28 and CD27 expression. a Schematic model of the EM and EMRA differentiation subsets according to CD28 and CD27 expression. b Representative dot plots of the subsets defined by CD27 and CD28 expression for individuals in each group. c Individual segments of the pie charts represent the proportions of cells with each combination of CD28 and CD27 in the EM and EMRA CD4 T cell subsets in young donors (n = 20) and in the three NKG2D groups (<5%, 5–20%, and >20% NKG2D, n = 20 in each group) of elderly individuals. P values of differences between young and elderly people in each NKG2D group are shown in the rectangles below the pie charts (Student’s t test or non-parametric Mann–Whitney method). No significant differences were represented by white; p < 0.05 by light grey; and p > 0.001 by dark grey rectangles. d Quantification of telomerase activity. CD4+ T cells were isolated from eight individuals in <5%, seven in 5–20%, and five in >20% NKG2D groups. The telomerase repeat sequence product added to the oligo substrate was determined in triplicate in the cellular extracts and the results were corrected by the total protein content. Outlier values were represented by circles and extreme values by stars, and calculated by adding 1.5 and 3 times the IR to the 75th percentile, respectively. The non-parametric Kruskal–Wallis test was used to compare frequencies between the three groups

Fig. 4

Phenotypic characterization of CD4+ NKG2D+ T cells. a Expression of CD28 was analyzed in CD4+ NKG2D+ T cells from 20 elderly and 20 young individuals. Whole blood was stained with CD28-FITC/NKG2D-PE/CD3-PerCP/CD4-APC and 10⁵ cells were acquired in each experiment. Histograms depict the percentage of CD28+ NKG2D+ (black bar) and CD28^nullNKG2D+ (white bar) cells in gated CD3+ CD4+ lymphocytes. Percentages of CD28^null and CD28+ cells within NKG2D+ subset in elderly and young individuals are summarized. b (a) Characterization of CD4+ NKG2D+ T cells related to expression of the CD28 marker. (b) CD4+ T cells from five aged donors were isolated and the expression of CD45RA-FITC, CD45RO-PercP, CD25-FITC, and HLA-DR-PerCP was analyzed in CD28^nullNKG2D+ (Bb) and in CD28+ NKGD+ cells (c). Intracellular staining of granzyme B and perforin expression. The percentages of positive cells in the indicated cell populations in this representative experiment were expressed in each histogram plot. (d) Histograms summarize the percentage of positive cells for each marker in CD28^nullNKG2D+ and in CD28+ NKG2D+ cells (mean ± SEM). Asterisks significant differences between the groups (p < 0.05). (Student’s t test or non-parametric Mann–Whitney method). (c) CD4+ T cells from the same donors were stained with CD45RA-FITC/NKG2D-PE/CD28-PercP/CD31-APC. The frequencies of CD45RA+ and CD31+ expression in the CD45RA subset were analyzed in CD28+ NKG2D+ and in CD28+ NKG2D⁻ subsets. The percentage of positive cells in the indicated cell populations in this representative experiment were depicted in each histogram plot. Histograms summarize the percentage of positive cells to each marker (mean ± SEM) in CD28+ NKG2D+ (black bars) and in CD28+ NKG2D⁻ cells (white bars) from five elderly donors. Student’s t test was used to compare means between groups

Fig. 5

CD69 expression in response to CMV antigens and to the influenza vaccine. PBMCs from elderly individuals were stimulated for 18 h and CD69 expression was evaluated by flow cytometry. a Representative dot plots of the frequency of CMV- and influenza vaccine-specific CD4 T cells with and without NKG2D expression. Percentages of positive cells in this representative experiment are expressed in dot plots. b Histograms summarize the percentage of CD69-cells positive for CMV and influenza vaccine in the CD4+ CD28^nullNKG2D+ (black bars) and CD4+ CD28^nullNKG2D- subsets (white bars) obtained from 11 elderly donors. Paired t test was used to compare paired frequencies

Fig. 6

Differentiated status of CD4+ subsets defined by IL-2/IFN-γ production and TREC content. a PBMCs were stimulated for 6 h with soluble anti-CD3 (1 μg/mL). The responder cells were analyzed for intracellular IL-2-APC staining in the indicated populations. b Cells were treated as described previously and IFN-γ expression was analyzed by intracellular staining. Percentages of positive cells in the indicated populations in this representative experiment are expressed on the histogram-plots. Histograms summarize the percentage of IFN-γ-positive cells in CD4+ CD28^null and in CD4+ CD28+ lymphocytes NKG2D− and NKG2D⁺ (mean ± SEM) from the six elderly donors tested. A paired t test was used to compare paired means. c Dose-response curves of the IFN-γ production by CD4 subsets defined by CD28 and NKG2D expression in response to anti-CD3. The cells were cultured for 6 h in medium alone or with increasing concentrations of anti-CD3 (1–1,000 ng/mL). A fluorescence analysis was carried out as previously described. One representative experiment of three is shown. d Quantification of TREC copy number. CD28+, CD28^nullNKG2D⁻, and CD28^nullNKG2D+ populations were isolated by sorting (magnetic bead separation) and the TREC copy number was determined by real-time PCR. Experiments were conducted in duplicate and bars represented results from three aging donors (mean ± SEM)