Age-dependent signature of metallothionein expression in primary CD4 T cell responses is due to sustained zinc signaling

Abstract

The ability to mount adaptive immune responses to vaccinations and viral infections declines with increasing age. To identify mechanisms leading to immunosenescence, primary CD4 T cell responses were examined in 60- to 75-year-old individuals lacking overt functional defects. Transcriptome analysis indicated a selective defect in zinc homeostasis. CD4 T cell activation was associated with zinc influx via the zinc transporter Zip6, leading to increased free cytoplasmic zinc and activation of negative feedback loops, including the induction of zinc-binding metallothioneins. In young adults, activation-induced cytoplasmic zinc concentrations declined after 2 days to below prestimulation levels. In contrast, activated naïve CD4 T cells from older individuals failed to downregulate cytoplasmic zinc, resulting in excessive induction of metallothioneins. Activation-induced metallothioneins regulated the redox state in activated T cells and accounted for an increased proliferation of old CD4 T cells, suggesting that regulation of T cell zinc homeostasis functions as a compensatory mechanism to preserve the replicative potential of naïve CD4 T cells with age.

FIG. 1.

Age affects naïve CD4 T cell proliferation independent of early activation events. (A) Purified CD4⁺CD45RA⁺ T cells were stimulated with TSST-1-loaded mDC and the induction of early activation markers CD25 and CD69 on TCR Vβ2⁺-gated cells was determined by flow cytometry. Results shown as mean ± SD do not show a difference between the two age groups (20–35 [open circles] and 60–75 years [solid circles]). (B) CFSE-labeled CD4⁺CD45RA⁺ T cells were stimulated with TSST-1-loaded mDC for 4 days. The percentages of Vβ2⁺CD4⁺ T cells that had entered the cell cycle and started to proliferate are shown as box plots displaying medians, 25th and 75th percentiles as boxes, and 10th and 90th percentiles as whiskers. Again, no age-dependent difference is seen. (C) Naïve CD4 T cells from the 60- to 75-year-old individuals (n = 15) underwent a higher number of cell divisions within the 4-day culture than the young adults (n = 15, p = 0.03. Results are shown for proliferating Vβ2⁺ T cells on day 4 after stimulation as mean ± SD of fold increases.

FIG. 2.

Age causes a selective transcriptional fingerprint in activated naïve CD4⁺ T cells. CD4⁺CD45RA⁺ T cells from 6 individuals 20–35 years old (closed symbols) and 6 individuals 60 to 75 years old (open symbols) were stimulated as described in Fig. 1A. (A) Gene expression analysis was performed on days 0, 3, and 10. PCA projection did not show an age-dependent discrimination. (B) Activation-induced change in whole-transcriptome expression values correlated highly between the two age groups (r = 0.976, p < 10⁻¹¹). Results are shown as the mean log₂ ratio for 6 individuals in each age group. (C) Separate analysis of cytokine and cytokine receptor gene expression (r = 0.981, p < 10⁻¹¹) confirmed that activation-induced gene expression was intact. (D) Probes differentially expressed were analyzed using DAVID's Gene Ontology enrichment tools. Metal binding was the GO term most significantly associated with genes overexpressed on day 3 in the 60- to 75-year-old individuals. Metallothioneins (MT) were entirely responsible for the annotation clustering. The dynamics of the expression of MT family members is shown as the log₂ ratio of the mean expression values in the two age groups. MT3 and MT4, which were not expressed in T cells at any time point, were omitted.

FIG. 3.

Overexpression of MT in naive CD4 T cells. CD4⁺CD45RA⁺ T cells were stimulated with TSST-1-pulsed DC. (A) T cells were harvested at indicated time points. MT1X transcription was quantified by qPCR. Data are representative of five independent experiments. (B) MT1X and MT2A transcription was quantified on days 0 and 3 in twelve 20- to 35- (open boxes) and eleven 60- to 75-year-old individuals (shaded boxes). Data are shown as box plots of fold increase. (C) Intracellular MT protein expression was quantified in gated Vβ2⁺CD25⁺ cells on day 4 following stimulation by FACS with Alexa 488-labeled anti-MT antibody. (Left panel) One representative histogram (bold line, Alexa 488 labeled anti-MT; gray shaded, isotype control). Results for seven donors age 20–35 years and five donors age 60 to 75 years are shown as mean ± SD in the right panel.

FIG. 4.

Zinc-induced gene transcription. (A) CD4⁺CD45RA⁺ T cells were stimulated with TSST-1-pulsed DC and cultured in different zinc concentrations. MT expression in activated Vβ2⁺CD25⁺ and nonactivated CD25⁻ T cells were assessed by flow cytometry on day 4. Histograms representative of seven experiments show the zinc dependence of MT induction in activated T cells. (B) CD4⁺CD45RA⁺ T cells were activated and then cultured in 3 μM and 45 μM zinc-containing medium. Zinc-induced gene expression was assessed by Affymetrix gene arrays and compared to the genes preferentially expressed in the 60- to 75-year-old individuals on day 3 after stimulation. Of 22 genes induced by the higher zinc concentrations, 14 were found to be preferentially induced in 60- to 75-year-old individuals with a probability of >0.75 (p = 5 × 10⁻⁹).

FIG. 5.

Influence of age on activation-induced zinc flux. (A) CD4 T cells were labeled with 2 μM FluoZin3 and stimulated with TSST-1-loaded DC. FluoZin3 fluorescence intensity indicative of cytoplasmic labile zinc concentrations in Vβ2⁺CD25⁺ cells peaked 2 days after stimulation. Cytoplasmic labile zinc kinetics were compared in T cells in young and old adults; samples were always run in parallel. Results from a 35-year-old (open cjrcles) and a 62-year-old individual (black circles) are shown as mean ± SD of triplicate wells and are representative of three donor pairs. (B) CD4 T cells were stimulated with anti-CD3/anti-CD28-coated beads. After 24 h of stimulation, the cells were transfected with Zip6-specific or control siRNAs and cultured in IL-2-supplemented medium. MT expression was assessed by flow cytometry on day 4 after stimulation. Results are representative of three experiments.

FIG. 6.

MT induction decreases reactive oxygen species and dampens the NF-κB pathway. (A) CD4 T cells were cultured with DC and TSST-1 in 3 μM and 75 μM Zn²⁺ supplemented medium. Cells were harvested on day 4, and total superoxide production was determined by ESR. The culture in the higher zinc concentrations significantly reduced superoxide production (p = 0.006). (B) Cells were cultured as in A and transfected with MT2A-specific or control siRNA on day 2. A representative experiment depicting the ESR signal reflecting superoxide anion production is shown. (C) Superoxide production in CD4 T cells cultured for 4 days in 75 μM Zn²⁺ and transfected with MT2A-targeted or control siRNA is shown as mean ± SEM. The production was significantly increased after MT2A silencing (p = 0.01). (D) CD4 T cells were stimulated with anti-CD3/anti-CD28 beads for 2 days and transfected with MT2A-specific or control siRNA. At 24 h after transfection, cells were stimulated with 50 ng/mL of TNF-α. NF-κB phosphorylation was quantified after 10 min by PhosFlow. Data are shown as Δ MFI of stimulated and unstimulated cells and are representative of three independent experiments.

FIG. 7.

Induction of MT supports T cell proliferation. (A) Purified CD4 T cells were stimulated with anti-CD3/anti-CD28 beads for 2 days and transfected with MT2A-specific or control siRNA. After 24 hours, 10⁵ transfected cells/200 μL were expanded in IL-2-supplemented medium. Cell recovery was assessed after 48 h by flow-based cell counting after PI staining. Data are representative of four experiments. (B) CFSE-labeled CD4⁺CD45RA⁺ T cells were stimulated with TSST-1-loaded DC. The mean number of divisions was calculated from the CFSE dilution profile on day 4 and correlated with the transcription of MT2A and MT1X on day 3 as quantified by qPCR.