Localization and phosphorylation of HS1 during the formation of lytic synapses. Top rows, localization of primary NK cells expressing GFP-HS1 incubated for 5, 10 or 15 min (top right corners) at 37 1C with K562 target cells (a) or Dynabeads coated with ICAM-1 and ULBP (b). Scale bars, 3 mm. Below, immunoprecipitation (IP) of HS1 from cell lysates followed by immunoblot analysis (IB) with monoclonal antibody 4G10 to assess phosphorylated tyrosine (p-Tyr) or with antibody to HS1 (anti-HS1) to show equal loading of lanes. For all immunoblot analyses in all figures, an equal amount of protein was loaded into each well, as determined by the Bradford assay and Ponceau staining. Data are representative of three separate experiments (images) or at least three separate experiments (immunoblots).

HS1 in NK cell adhesion.
(a) Tyrosine phosphorylation of HS1 during the adhesion of primary NK cells expressing shRNA and/or ‘rescue’ constructs in six-well plates coated with ICAM-1 or fibronectin (Fn) in the presence of PMA (10 ng/ml); after 15 min, HS1 was immunoprecipitated and analyzed by immunoblot with antibody to phosphorylated tyrosine. WCL, immunoblot analysis of one tenth of the preimmunoprecipitation whole-cell lysate of NK cells on ICAM-1 with anti-HS1 to assess total HS1.
(b) Adhesion of primary NK cell populations to 96-well plates coated with ICAM-1 or fibronectin, in the presence of PMA, assessed as absorbance at 595 nm (A₅₉₅).
*, P < 0.01. (c) Morphology of primary NK cells (as described in a) allowed to adhere to ICAM-1 or fibronectin-coated six-well plates with glass coverslips on the bottom, then fixed and stained with Alexa Fluor 568–phalloidin. (d) TIRF microscopy of HS1 localization in NKL cells expressing GFP-HS1, allowed to adhere to ICAM-1 - or fibronectin-coated plates, then fixed, made permeable and stained with antibodies specific for various proteins (left margins). Original magnification, × 100 (c,d). Data are representative of at least three separate experiments (mean and s.e.m. of triplicate wells, b).

Function of HS1 in the formation of NK cell–target cell lytic synapses and conjugates, (a) Overlays of fluorescence and transmitted-light images of fixed control (HS1 WT) or HS1 -knockdown NKL cells stably expressing GFP-tagged proteins and incubated for 15 min at 37 1C with target K562 cells. Original magnification, × 100. (b) Two-color flow cytometry of the time course of cell-cell conjugate formation by K562 cells stained with the red dye PKH26 and incubated for 1–15 min (horizontal axes) at 37 1C with control or HS1-knockdown NKL cells expressing GFP-tagged actin or Vav1, presented as the percentage of red-green events relative to total events, (c) Quantification of cell-cell conjugate formation by K562 cells labeled with PKH26 and incubated for 15 min at 37 1C with NKL cells (horizontal axis) expressing various GFP-tagged proteins (key), presented as described in b. Data are representative of at least three separate experiments (error bars, s.e.m.).

Function of HS1 in chemotaxis. (a) Tyrosine phosphorylation of HS1 in primary NK cells allowed to adhere to fibronectin-coated six-well plates for 15 min at 37 1C in the presence of PMA (Adhesion), or treated as described above and then exposed for 15 min to SDF-1a (Chemotaxis), followed by immunoprecipitation of HS1 and immunoblot analysis with antibody to phosphorylated tyrosine. Bottom blot (IB: HS1), one tenth of the preimmunoprecipitation whole-cell lysate. (b) Transwell assay of the migration of primary NK cells (Supplementary Methods), (c) Activity of signaling proteins in primary NK cells adhering to fibronectin and treated with SDF-1a (as described in a), analyzed by immunoblot with monoclonal antibodies specific for the proteins along the left margin. WCL, immunoblot analysis of one tenth of the preimmunoprecipitation whole-cell lysate. PLC-b, phospholipase C-b. (d) Signaling by Rho-family GTPases in NKL cell populations treated as described in a; lysates were ‘CRIB precipitated’ on GST-PBD agarose (PAK-binding domain) or GST-RBD agarose (Rhotekin-binding domain) and analyzed by immunoblot. Lysate total, immunoblot analysis of one tenth of the preimmunoprecipitation whole-cell lysate of HS1-knockdown cells. Data are representative of at least three separate experiments (mean ± s.e.m. of triplicate wells, b).

Function of HS1 in cytolysis. (a) HS1 knockdown in NKL cells and primary NK cells expressing a tetracycline-inducible shRNA sequence targeting HS1 (HS1A or HS1 B) or a nontargeting control shRNA (HS1sc) and treated for 0–72 h (above lanes) with doxycycline; HS1 was immunoprecipitated and analyzed by immunoblot with anti-HS1. (b) Restoration of HS1 expression in HS1-knockdown cells treated with doxycycline and then transfected by nucleofection 72 h later with shRNA-resistant GFP-tagged wild-type or mutant HS1 constructs (above lanes), followed by immunoprecipitation of HS1 at 72 h later. For immunoblot analysis of phosphorylated tyrosine, cells were treated with 2 mM Na₃VO₄ for 15 min before lysis. HS1 was immunoprecipitated and analyzed by immunoblot with anti-HS1. WCL (a,b), immunoblot of one tenth of the preimmunoprecipitation whole-cell lysate with anti-actin to show equal loading of lanes, (c) Cytolysis assays of NK cell populations and K562 target cells combined for 4 h at 37 1C, presented as the killing index, based on the release of adenylate kinase into the media. WT, wild-type; KD, knockdown. *, P < 0.01. Data are representative of at least three separate experiments (mean and s.e.m. of triplicate wells, c).

Integrin-mediated signaling in HS1-deficient NK cells and NK cells expressing HS1 mutants, (a) HS1 in the activation of specific signaling proteins ‘downstream’ of LFA-1 in control primary NK cells or primary NK cells expressing shRNA targeting HS1; cells were allowed to adhere to ICAM-1 or fibronectin in the presence of PMA for 0, 5 or 10 min at 37 1C, then were analyzed by immunoblot of lysates with antibodies specific for phosphorylated molecules (phosphorylated residues in parentheses), except Syk, which was immunoprecipitated and then analyzed with antibody to phosphorylated tyrosine. (b) Association of specific signaling proteins with integrin β₂ in primary NK cells allowed to adhere to ICAM-1 in the presence of PMA for 0, 5 or 10 min at 37 1C; lysates immunoprecipitated with Dynabeads coated with anti-β₂ integrin were analyzed by immunoblot with monoclonal antibodies specific for the proteins along the left margin. WCL (a,b), immunoblot of one tenth of the preimmunoprecipitation whole-cell lysate. (c) Immunoblot analysis of GTP-bound Rac, Cdc42 and Rap1 precipitated from lysates of primary NK cells allowed to adhere to ICAM-1 in the presence of PMA for 0, 5 or 10 min at 37 1C. WCL, immunoblot analysis of one tenth of the preimmunoprecipitation whole-cell lysates of HS1 -knockdown cells to show that the effects produced were not due to loss of protein expression (such as Rac, Cdc42 or Rapl) with loss of HS1 expression. Data are representative of at least three separate experiments.

Selective functions for HS1 tyrosine residues in NK cell migration and cytolysis. Killing index of primary NK cell populations expressing shRNA and/or ‘rescue’ constructs (key) and incubated for 4 h at 37 1C with K562 target cells in 96-well flat-bottomed plates (a), 96-well round-bottomed plates (b) or 6-well flat-bottomed plates (c), assessed on the basis of release of adenylate kinase into the media. Data are the mean ± s.e.m. of triplicate wells from at least three separate experiments.

Function of HS1 in NK receptor signaling, (a) DAP10 localization in control and HS1-knockdown primary NK cells and target K562 cells pelleted together, incubated for 10 min at 37 1C, then fixed and stained with anti-DAP10. Original magnification, × 100. (b) Activation of signaling proteins in primary NK cells expressing shRNA and/or ‘rescue’ constructs and incubated for 0, 5 or 10 min at 37 1C with Dynabeads coated with ICAM-1 and ULBP. For DAP10, lysates were immunoprecipitated with anti-DAP10 and analyzed by immunoblot with antibody to phosphorylated tyrosine; for Lyn, Vav1 and PI(3)K, immunoblots of lysates were analyzed with antibodies specific for phosphorylated molecules, (c) Association of proteins with DAP10 in primary NK cells incubated with Dynabeads; lysates immunoprecipitated with anti-DAP10 were analyzed by immunoblot with monoclonal antibodies specific for proteins along left margin. WCL (b,c), immunoblot analysis of one tenth of the preimmunoprecipitation whole-cell lysate. (d) Concentration of the cytokines interferon-g (black) and tumor necrosis factor (gray) in the media of primary NK cell populations cultured with K562 cells, (e). Rho-family GTPase activity in NKL cell populations incubated with Dynabeads as describe in b, assessed by analysis of lysates for GTP-bound Rac, Cdc42 and Rap1. WCL, immunoblot analysis of one tenth of the preimmunoprecipitation whole-cell lysate from HS1 -knockdown cells. Data are representative of at least three separate experiments (mean and s.e.m. of triplicate wells, d).