TFEB overexpression resolved autophagic arrest but not cytotoxic defect in Npc1-KO CTLs. (A) Subcellular localization of TFEB was determined by western immunoblotting in naïve and day-3 activated CD8⁺ T cells from Npc1^−/−.CL4 and Npc1^+/+.CL4 mice. (B) Confocal immunofluorescence microscopy shows morphological restoration of CGs after TFEB overexpression in Bl/6.OTI CTLs that had the Npc1 gene knocked out using CRISPR/Cas9 (Npc1 KO). The cells were also retrovirally transduced with an empty murine stem cell virus (MSCV) vector or with TFEB-flag and stained for intracellular GzmB (green) and Prf (magenta). (C) Diameter of CGs was estimated by computational 3-dimensional re-creation (Imaris vesicles) based on GzmB staining. Isolated Bl/6.OTI CD8⁺ T cells were either transfected with NT or guide RNAs that target the Npc1 gene and retrovirally transduced with an empty MSCV vector or TFEB-flag complementary DNA cloned into MSCV vector. (D) Area of LC3-B⁺ organelles per cell was assessed after Npc1 KO and TFEB overexpression, as shown in panel B. Individual experiments are shown in supplemental Figure 4D. (E) TEM of control and Npc1-deficient cells transduced with an empy vector or TFEB-flag (n = 30 cells from each group). Shown is mean ± standard error of the mean (SEM). Npc1^+/+ cells were compared using Mann-Whitney test, and Npc1^−/− cells were compared using 1-tailed unpaired t test (this was due to nonnormal and normal distributions, respectively). (F) TFEB overexpression had no effect on the colocalization of GzmB and cholera toxin B (CTxB) in Npc1-deficient cells. (G) TFEB overexpression had no effect on CTxB accumulation (per cell) in Npc1-deficient cells. Individual experiments are shown in panel F. (H) Chromium-51 release assay using (SIINFEKL) EL-4 target cells and genetically modified Bl/6.OTI CTLs (as per panel B) showed no restoration of CTL function after overexpression of TFEB in Npc1-KO cells; Npc1-KO-TFEB-flag-MSCV and Npc1-KO-MSCV CTLs showed near-identical cytotoxic activity. Values plotted are standardized to maximum killing observed in NT-empty MSCV cells at 10:1 effector-to-target (E:T) ratio and set at 100% (average cytotoxicity at 10:1 E:T ratio was 75.9% ± 18.9% [mean ± standard deviation of n = 3 independent experiments]). Each value shown represents mean ± SEM (n = 3). Detailed description of microscopy and analysis is provided in data supplement. *P < .05, ***P < .001, ****P < .0001. C, cytosolic; DAPI, 4′,6-diamidino-2-phenylindole; N, nuclear fractions; ns, not significant.