T cell memory revisited using single telomere length analysis

Abstract

The fundamental basis of T cell memory remains elusive. It is established that antigen stimulation drives clonal proliferation and differentiation, but the relationship between cellular phenotype, replicative history, and longevity, which is likely essential for durable memory, has proven difficult to elucidate. To address these issues, we used conventional markers of differentiation to identify and isolate various subsets of CD8⁺ memory T cells and measured telomere lengths in these phenotypically defined populations using the most sensitive technique developed to date, namely single telomere length analysis (STELA). Naive cells were excluded on the basis of dual expression of CCR7 and CD45RA. Memory subsets were sorted as CD27⁺CD45RA⁺, CD27^intCD45RA⁺, CD27^-CD45RA⁺, CD27⁺CD45RA^int, CD27^-CD45RA^int, CD27⁺CD45RA^-, and CD27^-CD45RA^- at >98% purity. The shortest median telomere lengths were detected among subsets that lacked expression of CD45RA, and the longest median telomere lengths were detected among subsets that expressed CD45RA. Longer median telomere lengths were also a feature of subsets that expressed CD27 in compartments defined by the absence or presence of CD45RA. Collectively, these data suggested a disconnect between replicative history and CD8⁺ memory T cell differentiation, which is classically thought to be a linear process that culminates with revertant expression of CD45RA.

Keywords: T cell differentiation; T cell memory; T cell senescence; replicative history; telomere length (TL).

Figure 1

Telomere length distributions among CD8⁺ memory T cell subsets revealed using STELA. (A) Flow cytometric gating strategy. Populations labeled J–P were flow-sorted at >98% purity for telomere length assessment via STELA. Data from donor 1. (B) Southern blot showing XpYp telomere length data for the subsets in (A). ^#DNA ladder; ^§fibroblasts (control). (C) Scatter plot depicting 17p telomere length distributions for the subsets in (A). Horizontal red lines indicate median values. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001 (Kruskal-Wallis test with Dunn’s post-hoc test for multiple comparisons). (D) Boxplot depicting pooled telomere length distributions (n = 5 donors). Significance was assessed using the Kruskal-Wallis test. Individual comparisons and corrected values are shown in Table 1 . (E) Flow cytometry plots showing the distribution of CD8⁺ memory T cells according to expression levels of CD27 and CD45RA in the presence of CMV. Data from donor 4.

Figure 2

Individual telomere length distribution patterns among CD8⁺ memory T cell subsets revealed using STELA. (A–D) Scatter plots depicting 17p telomere length distributions from donors 2 (A), 3 (B), 4 (C), and 5 (D). Horizontal red lines indicate median values. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001 (Kruskal-Wallis test with Dunn’s post-hoc test for multiple comparisons). Donors 4 and 5 were known to be seropositive for CMV.