The authors regret that two of the flow cytometry histograms in Fig. 3 C were inadvertently duplicated. The corrected figure and its legend appear below.

View larger version (43K): [in this window] [in a new window] [Download PPT slide]   Figure 3. Tumor-infiltrating T reg cells express functional OX40. (A) TILs were purified from a pool of CT26 tumor nodules, and the CD4+ cell population was analyzed by flow cytometry (percentages are shown). As controls, CD4+ T cells were purified from the spleens of tumor-free mice (STF). (top) On gated CD4+ T cells, the expression of Foxp3 versus CD25 was evaluated. (bottom) Among CD4+ CD25+ T cells, Foxp3 versus OX40 expression is shown. (B) OX40 expression was evaluated on gated CD4+ Foxp3+ T cells from TILs of tumor-bearing mice. As controls, isotype staining on Foxp3+ TILs and OX40 staining on Foxp3+ CD4+ splenocytes from tumor-free mice are shown. (C) CD4+ CD25– T cells (Teff), obtained from the spleens of tumor-free mice, were CFSE labeled and seeded either alone or combined at 1:1, 1:2, or 1:4 ratios with unlabeled CD4+ T cells purified from TILs (T reg TIL) of tumor-bearing mice, either untreated or 6 h after intratumor OX86 injection (anti-OX40 in vivo). No major differences in the composition of tumor-infiltrating CD4+ T cells or in Foxp3 expression were evident 6 h after treatment. As controls, CD4+ CD25+ T cells from spleens of tumor-free mice were used (T reg STF). T reg cells were either pretreated with anti-OX40 mAb in vitro or with rat IgG as mock control. After 72 h, effector T cells were tested for CFSE dilution as a marker of proliferation. Representative plots of one out of three independent experiments are shown and indicate percentages of proliferating (CFSElow) versus resting (CFSEhigh) effector T cells.