Treg cells expressing CD44^act show an enhanced suppressive effect in alloreactive and GVH models in vivo. (A) C57BL/6-Thy1.1 congenic hosts were injected intraperitoneally with 1 × 10⁶ CFSE-labeled Balb/c CD4CD25^– T cells along with 2 × 10⁵ unlabeled unactivated or 1 × 10⁵ activated Balb/c CD4CD25⁺ cells sorted based on CD44^act expression. After 3 days, cells in the peritoneum were collected by lavage and analyzed by flow cytometry. Expression of CD25 as a reflection of alloactivation of CFSE-gated CD4CD25^– responder T cells is shown. Data are from individual mice and are representative of 4 independent experiments. (B) Average in vivo suppression of CD25 expression on T effectors in the presence of Treg cells. Transfers were conducted as in panel A. Experiments were done with 4 mice/group. Statistical comparison between relevant groups is shown. *P < .001; **P ≤ .002. Data are the mean ± SD of 4 experiments. (C) Lethally irradiated C57BL/6 mice were reconstituted with C57Bl/6 bone marrow cells (3 × 10⁶) and simultaneously given 2 × 10⁶ Balb/c CD4CD25^– effector T cells together with 1.5 × 10⁶ freshly isolated or CD44^act fractionated CD4CD25⁺ regulatory cells as in panel A. On day 5, serum was collected for IFNγ measurement and 2 × 10⁴ splenocytes were plated in Methocult M3234 plus stem-cell growth factors. Five to 7 days later, colonies were counted. The number of CFUs per plate is shown as a percentage of BM alone and is an average for at least 5 mice per group (bottom, ▪). The ▦ reflects the values when CFUs for fractionated cells are indexed for homing efficiency relative to freshly isolated, unactivated CD4CD25⁺ cells. IFNγ serum levels for each group are shown in the top panel. Statistical comparison between indicated groups is shown. *P ≤ .002; **P ≤ .01. Data shown are the mean ± SD of 5 experiments.