SA-FasL–engineered T cells are effectively eliminated in response to alloantigens in vivo. (A) Frequency, proliferation, and apoptosis of 4C cells in F1 recipients. 4C.SJL splenocytes were labeled with CTV, engineered with SA-FasL protein (100 ng/10⁶ cells), and injected IV (5 × 10⁶ cells/mouse) into F1 recipients (H2K^b/d). Cells engineered with an equimolar of SA (50 ng/10⁶ cells) were used as controls. Splenic T cells were analyzed 48 hours after injection for the frequency of total donor live (7AAD^-CD45.1⁺) cells, CD4⁺T (CD45.1⁺CD4⁺) cells, proliferating (CTV^low) cells, and apoptotic (CD45.1⁺AnnexinV⁺) cells. Data were pooled from 2 independent experiments, with 3 to 4 per group. (B) Tracking of C57BL/6^hCD2 donor cells in F1 recipients. Donor splenocytes were labeled with CTV and engineered with the indicated doses of SA-FasL protein (ng/10⁶ cells). Cells engineered with equimolar of SA (50 ng/10⁶ cells) at the highest dose of SA-FasL were used as controls. Cells were adoptively transferred into F1 recipients (10 × 10⁶ cells per mouse) that were euthanized 72 hours later to harvest the spleen. Splenocytes were analyzed in flow cytometry by gating on donor cells negative for H-2K^d for the frequency of total cells (CTV⁺H-2K^d-), CD4⁺ (CTV⁺H-2K^d-CD4⁺), CD8⁺ (CTV⁺H-2K^d-CD8⁺) cells, as well as proliferating donor cells (CTV^lowH-2K^d-). Data were pooled from 3 independent experiments with n = 3 to 4 per group. (C) SA-FasL–engineered alloreactive T cells undergo autocrine apoptosis in vivo. CFSE-labeled and SA-FasL–engineered C57BL/6^hCD2 (CD45.1⁺) splenocytes were comixed at 1:1 ratio with CTV-labeled and SA-engineered splenocytes and injected (IV) into F1 recipients 6 hours after irradiation (1000 cGy). Mice also received 10 × 10⁶ nonlabeled and nonengineered bone marrow cells. F1 mice that underwent transplantation with nonengineered cells served as controls. Mice were euthanized 72 hours after cell infusion, and spleen cells were analyzed for the indicated donor cell types using flow cytometry. (D) Turnover kinetics of SA-FasL on engineered CD3⁺ T cells. CTV-labeled SA-FasL plenocytes (CD45.1) were injected into irradiated F1 (1000 cGy) animals. Mice were euthanized at various time points after infusion and analyzed for the presence of SA-FasL on donor CD3⁺ T cells using an antibody to the streptavidin portion of the molecule in flow cytometry. For comparison of mean, Mann Whitney test and 1-way ANOVA with Tukey post hoc test was used in panels A-C, respectively. Nonlinear regression analysis was done for panel D. Data are represented as mean ± SEM. ANOVA, analysis of variance; SEM, standard error mean. ∗P < .05; ∗∗P < .01; ∗∗∗P < .001; ∗∗∗∗P < .0001.