Fig. 2

NBAS is required for cytotoxic granulation in the NK-cell line. a Plot showing the HLH time course, therapeutic approaches and response status in patient P007 with presumed primary HLH. b Functional investigations of cytotoxic lymphocytes. NK-cell cytotoxicity (upper row) and NK (middle row) and T-cell degranulation (bottom row) were defective in P007 compared with her healthy parents but gradually recovered after HSCT. NBAS mRNA (c) and protein (d) levels in an NK-cell line (IMC-1) after NBAS knockdown (n = 3). e Histograms showing the cytotoxic activity of scramble or shNBAS-targeting IMC-1 cells analyzed by FACS after coculture with K562-GFP target cells (n = 3). K562-GFP was used as the negative control, and K562-GFP cocultured with wide-type IMC-1 (without transfection of any shRNA) was used as the positive control. f Histograms show surface CD107a expression, indicating the degranulation ability of scramble or shNBAS-targeting IMC-1 cells in the presence of K562-GFP target cells. g Representative electron microscopic images of sorted scramble or shNBAS IMC-1 cells stimulated with K562-GFP cells for 4 h. LG, lytic (cytotoxic) granule; N, nucleus; M, mitochondria; GA, Golgi apparatus. Scale bars = 500 nm. h The relative number of lytic granules per field was quantified (n = 3). i Representative images showing decreased expression of AP3B1 (magenta; indicated by white arrows) and CD107a (red; indicated by yellow arrows) stimulated by K562-GFP (green) cells for 4 h. Nuclei were stained with DAPI (blue). Scale bar: 10 μm. n = 3. j Schematic depicting the process by which cytotoxic cells kill target cells through the granule-mediated degranulation pathway (left panel). The pop-up panel shows a cartoon of NBAS, along with RINT and ZW10, as part of the syntaxin 18 complex between the endoplasmic reticulum (ER) and Golgi. Data were presented as means ± SEM. *P < 0.05; **P < 0.01; ***P < 0.001; ns, not significant