cells are susceptible to cytokine deprivation apoptosis in vivo. (a) Frequency of CD4+ Foxp3+ (upper panels) or CD25+Foxp3+ (lower panels) in splenocytes isolated from WT, Bcl-2 tg mice ex vivo. (b) Splenocytes of WT or Bim-/- mice showing the frequency of CD4+ Foxp3+ (upper panel) or CD25+Foxp3+ (lower panel) Treg cells ex vivo (c) Mean percentage of Treg cells in spleens of WT (grey bar) or Bim-/- mice (solid black bar) (n = 3–5, n = number of mice). Data represent two independent experiments. (d) Viability of Tcon or Treg cells from WT or Bim-/- mice stimulated for 3 days with soluble anti-CD3 and anti-CD28. Cytokines were added at indicated 20 ng/ml concentrations in cultures at the beginning of stimulation. Percentage of events in the PIneg and FSChigh live gates are shown. Histograms of CFSE dilution of live Tresp cells (e) and level of apoptosis in Tresp cells (f) from WT mice co-cultured with Tcon cells or Treg cells from WT (upper panels) or Bim-/- (lower panels) mice. (g) Dot plots showing Foxp3 expression and CFSE dilution in live Tresp cells co-cultured with WT (upper panels) or BIM-/- (lower panels) Tcon or Treg cells stimulated as in 'd' for 4 days in the presence or absence of IL-7 (20 ng/ml).