CTLs from Flower-deficient mice show a specific block of CG endocytosis. (A) Time-lapse maximal-intensity projections of syb2-mRFP (red) transfected CTLs conjugated to target cells in the presence of anti-RFP488 antibody (green) in the medium. The endocytosis of CGs (yellow) is shown in WT (top), Flower-deficient (middle), and Flower-deficient CTLs overexpressing Flower-pHluorin (magenta; bottom). Bars, 10 µm. (B) Quantitative analysis of the accumulation of endocytic CGs at the IS (one third of CTL volume projecting toward the target cell) over time from panel A. Time zero is defined as the appearance of the first endocytic CG at the IS. Data are given as mean ± SEM; unpaired Student’s t test: ***, P < 0.001. (C) CTLs from synaptobrevin2-mRFP knock-in (SybKI; top) and Flower-deficient/SybKI (Flower KO/SybKI; bottom) mice were incubated with target cells (P815) in the presence of Alexa Fluor 488–conjugated anti-RFP antibody (anti-RFP488) in the medium to visualize endocytic granules. Cells were then fixed after 40 min and imaged by SIM. Bars, 5 µm. (D) CTLs from Flower KO/SybKI (bottom) mice were electroporated with Flower-pHluorin and incubated with target cells (P815) in the presence of Alexa Fluor 647–conjugated anti-RFP antibody (anti-RFP647) in the medium to visualize endocytic granules. Cells were then fixed after 40 min and imaged by SIM. The CTL expressing Flower-pHluorin (top cell) shows a clear rescue in endocytosis, whereas the untransfected one (bottom cell) does not. Bar, 5 µm. (E and F) Real-time calcein release-based killing assay. WT and Flower-deficient CTLs were co-cultured with P815 cells at the indicated CTL/target cell ratios, and killing was measured every 10 min for 4 h. Data are given as mean ± SEM from five independent experiments performed in duplicates (n = 10; N = 5). One-tailed unpaired Student´s t test: *, P < 0.05; ***, P < 0.001.