NKG2D KO CD8⁺ T cells display a survival disadvantage and reduced cytotoxicity in a cell-intrinsic manner. Irradiated Balb/c mice were transplanted with B6-derived TCD BM cells and FACS-sorted WT CD8⁺ T cells (CD45.1⁺), NKG2D KO CD8⁺ T cells (CD45.2⁺), or a mix of WT CD8⁺ T cells and NKG2D KO CD8⁺ T cells. (A) On day 7, spleen cells were harvested from the transplanted mice and stimulated with phorbol 12-myristate 13-acetate/ionomycin for 4 hours. Scatter plots of the fraction of IFN-γ and (B) TNF-α–producing WT and NKG2D KO CD8⁺ T cells in the individually transplanted setting and in the mixed setting are shown. (C) Transplanted mice were treated with BrdU on day 6 post-BMT and harvested on day 7. The fraction of BrdU-incorporated WT and NKG2D KO CD8⁺ T cells in the individually transplanted setting and in the mixed setting is shown. (D) Irradiated Balb/c mice were transplanted with B6-derived TCD BM cells and WT B6.SJL CD8⁺ T cells mixed with either WT B6 CD8⁺ T cells (CD45.2⁺) or NKG2D KO CD8⁺ T cells (CD45.2). The percentage of B6.SJL WT, WT B6, or NKG2D KO CD8⁺ T cells out of all donor-derived CD8⁺ T cells is represented as a scatter plot. (E) The percent of dead (live/dead stain⁺) CD8⁺ T cells of WT or NKG2D KO origin from mice receiving a mix of WT and NKG2D KO CD8⁺ T cells is plotted. (F) WT or NKG2D KO CD8⁺ T cells from individually transplanted or mixed transplanted mice were pooled, FACS-sorted, and cultured with A20 cells with (right bars) or without (left bars) anti-NKG2D antibody treatment in vitro at a 1:20 effector-to-target ratio. Cytotoxicity of the A20 cells was measured 4 hours later. ns, **, and *** indicate that the groups compared are not significantly different or display statistical significance of P < .01 and P < .001, respectively by Student t test.