Cytoskeleton remodeling in DCs is critical for synapse formation and NK cell activation. (A) F-actin relocalized at the mDC/NK cell interface. Phalloidin-PE was used to stain the F-actin cytoskeleton of either mDCs or resting NK cells, cultured alone or together (25 minutes). Whereas phalloidin staining was uniform on cell membranes of mDCs (left panel) or resting NK cells alone (upper cell of the middle panel), it clustered at the mDC/NK cell interface in cocultures (as indicated with an arrow at the DCNK-IS). (B) Fascin relocalized at the mDC/NK cell interface. Up to 89% ± 5% of mDC/NK cell conjugates observed after a 25-minute coculture relocalized fascin toward the synapse. (C) Inhibitors of cytoskeleton remodeling impaired accumulation of phosphotyrosines at the DCNK-IS. Prior to coculture with NK cells, mDCs were pretreated with cytochalasin B (CytB) or with nocodazole. The percentages ± SEM of mDC/NK cell conjugates associated with pTyr clustering at the DCNK-IS is reported. DMSO used to solubilize nocodazole and CytB was also used as a negative control. The mean percentage ± SEM of 3 independent experiments is depicted. (D-F) DCs from WASp-mutant mice failed to promote productive DC/NK cell conjugates leading to NK cell activation. Mouse WASp^-/- or wild-type BM-DCs were cocultured for 25 minutes with wild-type mouse NK cells on a slide coated with poly-L-lysine (D) or for 18 hours in 96-well plates (E-F). The percentages of conjugates associated with clustering of phosphotyrosines at the BM-DC/NK cell interface are reported (D). Confocal microscopy confirmed that WASp^-/- BM-DCs failed to form conjugates with NK cells (not shown). The levels of IFN-γ released in the coculture supernatants were measured using ELISA (E) and YAC-1 target cell lysis was determined by a chromium release assay (F). One representative experiment of 3 is depicted. E:T indicates effector-target ratio.