Proliferation and cytotoxicity. Sorted T-cell subsets were cultured with irradiated allogeneic spleen cells for 5 days before they were tested for proliferation and cytotoxicity. Results represent mean ± SD of triplicate wells. (A) Sorting strategy. Enriched T cells were stained with anti-CD4, anti-CD8, anti-CD44, anti-CD45RB, and anti-CD62L antibodies. We first gated on CD4/CD8⁺ T cells. CD45RB⁺CD62L⁺ T cells (region A) were further separated into 2 populations based on the expression of CD44 (A1, CD45RB⁺CD62L⁺CD44^−/low; A2, CD45RB⁺CD62L⁺CD44^high). Cells were sorted into naive (A1) and memory (all other T cells except naive [B+C+A2]) phenotype T cells. In some experiments (Figure 5), memory (total) T cells were further separated into T_EM (region B) and T_CM (C+A2) based on the expression of CD62L. Shown in the third column are cells from region A in the second column. Numbers represent percentages of parent gates. Donor mice for this sort were 4 to 6 months old. (B) Phenotypes of sorted T-cell subsets. Shown in the second column are cells from region A in the first column. Data presented in the last 2 columns were not from post-sort analyses. Logarithmic displays were used in all histograms or dot plots except those in the last column in which “logicle” displays (biexponential displays) were used. These data represent the results of at least 4 similar experiments. (C) Mixed lymphocyte reaction. Proliferation was determined by ³H-thymidine incorporation. The responder alone was always lower than 200 cpm. Data represent the results of 3 to 4 similar experiments. (D) Cytotoxic T-lymphocyte assay. Cytotoxicity was measured by 4-hour ⁵¹Cr release assay. *P < .01, memory versus other groups; #P < .01, naive versus bulk.