Fig. 5: Itgβ8^pos Tregs activate TGF-β1 produced by cancer cells to repress tumor-infiltrating CD8 T cells.

a–c Foxp3^Ctrl mice and Foxp3^ΔItgβ8 animals were injected i.d. with B16 cells. 15 days later tumors were harvested and analyzed by immunostaining for LAP and melanoma (GP100), 11–13 field were analyzed for each tumor. The graph illustrates the numbers of LAP aggregates observed per field. c Histogram illustrates the expression of Tgfb1 by RT-qPCR on isolated tumor cells after normalization with gadph expression. For b, c, data are presented as mean ± SD. d, e Foxp3^ΔItgβ8 animals were injected i.d. with either B16 shTgfb1 (TGFβ1^KO) or B16 cells having received empty vector control (TGFβ1^CTRL), and tumors were harvested 15 days later and analyzed by immunostaining for the presence of LAP aggregates. Graph illustrates the numbers of LAP aggregates observed per field, with 11–12 fields were analyzed for each tumor. f, g Foxp3^Ctrl mice were injected i.d. with either TGFβ1^CTRL B16 cells (black), 4 tumors, or TGFβ1^KO B16 cells (green), 8 tumors, and CD8^pos T cells infiltrating the tumors analyzed 15 days later flow cytometry. Representative counterplots illustrating the cytotoxic functions are shown as well as graphs demonstrating the percentage of CD8^pos T cells expressing either granzyme B (GzB) or and GzB and CD107. b P = 0.0005 (***), f P = 0.0061 (**). All experiments were conducted on four mice per group in three independent experiments. Means are shown ± SD. ***P < 0.001 was determined by unpaired two-tailed Student t test (b) and means are shown ± SD. **P < 0.01 was determined by a two-tailed Mann–Whitney test (c, e). ns statistically not significant. Scale bar (50 µm). Source data are provided as a Source Data file.