Protein palmitoylation and sphingolipid metabolism control regulated exocytosis in cytotoxic lymphocytes

Regulated exocytosis controls key cellular functions ranging from neurotransmitter release to the secretion of immune mediators and its disruption is associated with numerous pathologies. The cytotoxic activity of lymphocytes is particularly dependent on regulated and polarized lytic granule delivery towards infected or malignant cells. Although genetic and mechanistic studies have identified factors regulating exocytosis in cytotoxic lymphocytes, a systematic mapping of the relevant factors and their relationships is lacking. Through a genome-scale CRISPR knockout screen in a human natural killer cell line, we characterized a complex genetic network regulating cytotoxic granule exocytosis, with lipid metabolism and protein lipidation amongst the most prominent pathways. By combining global protein lipidation and membrane lipid composition studies, we found that ZDHHC17 drives palmitoylation of the core SNARE complex protein SNAP23 to target cytotoxic granules to GM1-rich lipid rafts whose assembly is controlled by serine palmitoyltransferase. Brunello pooled degranulation screen: 377.6 million NK-92 Cas9-BFP clone1 cells were transduced with Brunello lentivirus at an infection percentage of 19.637, yielding a total of 74.15 million infected cells and a calculated coverage of 925.89 cells per sgRNA determined as described above upon puromycin selection. NK-92 Cas9-BFP clone1 cells were transduced with the lentiGuide.Puro vector. In example, 0.8 million cells were distributed per well in 12-well plates containing 200µL of unconcentrated Brunello viral supernatant (1:1 mix of 48h and 72h viral harvest) in total 1mL growth medium supplemented with 6µM BX795 and 8µg/mL Protamine sulfate. Cell were spinfected at 1000rcf, 60min, 37°C and then incubated for 6 hours at 37°C, 5% CO2. Thereafter, plates were centrifuged at 300rcf 5min and supernatant was gently removed and replaced by adding 2ml fresh growth medium per well. Next day cells were pooled and evenly distributed to T175 cell culture flasks. On day 3 cells were selected for a total of 6 days with 2.5 µg/ml puromycin, changing and splitting cells every 2 days. After complete selection puromycin was removed and cells were maintained at always at pre-determined coverage for another 8 days. Finally, 400 million Brunello transduced NK-92 Cas9-BFP clone1 cells were used for a degranulation assay with PMA/Ionomycin stimulation. Fixed cells were pre-gated on CD56 positive cells and sorted using BD FACS Aria into CD107a negative (LAMP1 neg) and CD107a high (LAMP1 hi) populations. NK-92 RAB27a knockout cells deficient in degranulation were used as gating controls. Sorted cells were pelleted, snap-frozen in LN2 and stored at -80°C for further usage. An aliquot of unstimulated and unstained cells after puromycin selection were taken along as control to assess sgRNA presentation in Brunello transduced cells.