In vitro expansion of T_regs. (A) Representative flow cytometry plots of CD25 and CD62L expression on CD4 cells from NOD (left), BDC2.5 (middle), and GAD286 (right) mice. FACS^®-purified T_regs (•) and CD4⁺ CD62L⁺ CD25⁻ cells (○) from NOD (B), BDC2.5 TCR Tg (C), or GAD286 TCR Tg (D) mice were stimulated in vitro with anti-CD3– and anti-CD28–coated beads along with IL-2. (E) T cells from BDC2.5 TCR Tg mice were expanded as described above with p31-linked IA^g7-mIgG2a immobilized on latex beads. All cultures were quantitated by viable cell counting.

Phenotype of in vitro–expanded T_regs. (A) Expression of CD25 and CD62L on expanded T_regs and CD4⁺ CD62L⁺ CD25⁻ cells was determined by flow cytometry on day 8 after the culture initiation. Results are representative of more than 20 independent experiments. (B) Levels of mRNA for the indicated genes in expanded NOD (filled symbols) or BDC2.5 TCR Tg T cells (open symbols) were determined by real time PCR analysis on day 10 after the initiation of the cultures. The relative expression ratio (T_reg/T_CD25⁻) for each pair of cultures was calculated from Ct values as described in Materials and Methods. The dashed line represents the ratio of 1 (i.e., identical level of gene expression in T_reg and CD4⁺ CD62L⁺ CD25⁻ cultures). (C) Western blot analysis of FoxP3 protein expression in fresh and expanded T cells. The level of tubulin expression was included as a loading control. Results are representative of three independent experiments. (D) Cytokine secretion by expanded BDC2.5 T cells 48 h after restimulation with antigenic peptide and splenic APC. Results are representative of two independent experiments.

In vitro suppression by expanded T_regs. (A) Fresh and expanded T_regs were compared for their ability to suppress the proliferation of CD4⁺ responder T cells stimulated with anti-CD3 and T cell–depleted splenocytes. (B) Suppression by T_regs expanded from BDC2.5 TCR Tg or GAD286 TCR Tg mice was assayed as described in A. (C) Suppressive activity of BALB/c-expanded T_regs on CD4⁺ responder T cells from DO11.10 TCR Tg mice stimulated with anti-CD3 or an OVA peptide. Results are representative of three independent experiments.

In vivo survival and activation of expanded T_regs. (A) Freshly isolated and expanded BALB/c T_regs were labeled with CFSE and then injected into normal BALB/c mice (10⁶/mouse). All recipient mice were killed on day 30 after injection and the numbers of CFSE⁺ cells in the peripheral LN and spleen (not shown) were determined by flow cytometry. The data are presented as the number of CD4⁺ CFSE⁺ cells/10⁶ endogenous CD4⁺ T cells. Each circle represents the value from one mouse and the black bar represents the mean of the group. (B) Expanded T_regs from NOD.Thy1.2 (top) and GAD286 TCR Tg Thy1.2 (middle) were labeled with CFSE and 3 × 10⁶ were transferred to normal 8–12-wk-old NOD.Thy1.1 recipients. Expanded BDC2.5 TCR Tg Thy1.1 were labeled with CFSE and 3 × 10⁶ were transferred to normal 8–12-wk-old NOD.Thy1.2 recipients. The presence of transferred cells and their activation status in spleens, LNs, and pancreatic LNs were determined by flow cytometry on day 7 after cell transfer. The dot plots shown are gated on the Thy1.2 for NOD and GAD286 cells and Thy1.1 for BDC2.5 cells. The percentages of cells with CFSE dilution are shown on the plots. Results are representative of at least three recipient mice in two separate experiments.

Prevention of diabetes transfer by expanded T_regs. (A) Activated diabetogenic BDC2.5 CD4⁺ CD62L⁺ CD25⁻ cells (3.5 × 10⁵) were cotransferred with BDC2.5-expanded T_regs to 8-wk-old NOD. RAG^−/− recipients at the indicated ratio. The blood glucose for individual recipient mouse was monitored and plotted to access diabetes. n = 3 for no T_regs and 1:9 groups; n = 4 for 1:1 and 1:3 groups. Results are representative of three independent experiments. (B) Diabetes was induced in 6-wk-old NOD.RAG^−/− mice in the same manner as described in A, except that the number of transferred expanded T_regs from GAD286 TCR Tg mice and BDC2.5 TCR Tg mice equaled the number of transferred T_effs (n = 3 mice/group). (C) Diabetes was induced in 5–8-wk-old NOD.RAG^−/− or NOD.TCR-α^−/− recipients by injection of 25 × 10⁶ pooled spleen and LN cells from diabetic donors (n = 8). Some recipient mice were coinjected with expanded T_regs from NOD (2 × 10⁶, n = 3; 5 × 10⁶, n = 3; 8 × 10⁶, n = 4) or BDC2.5 TCR Tg mice (2 × 10⁶, n = 4). Results represent two independent experiments.

Prevention of autoimmune diabetes in NOD.CD28^−/− mice with BDC2.5-expanded T_regs. 5-wk-old prediabetic NOD.CD28^−/− mice were injected with 5 × 10⁵ BDC2.5-expanded T_regs (n = 3) or left untreated (n = 4). The development of diabetes was monitored and blood glucose levels of individual mice were plotted. Results are representative of at least five independent experiments.

Reversal of diabetes with expanded T_regs. (A) NOD mice with chronic diabetes were transplanted with syngeneic islets under the kidney capsule. On the day of transplantation, some recipient mice received 5 × 10⁶ NOD-expanded T_regs (n = 4) or 2 × 10⁶ BDC2.5-expanded T_regs (n = 5), and the remaining mice (n = 3) were left untreated. Blood glucose level was monitored. All islet recipients normalized blood glucose within the first day after transplantation. Results are representative of two independent experiments. (B) NOD mice with new onset diabetes (blood glucose > 300 mg/dL, n = 7) were injected with 10⁷ BDC2.5-expanded T_regs and blood glucose was monitored. Two consecutive readings of blood glucose of <250 mg/dL was considered remission of diabetes.