NKG2D-mediated cytolysis in CTL is cPLA₂ dependent. (A) cPLA₂ inhibition by the pharmacological inhibitor AACOCF3 (CF3) blocks NKG2D-mediated cytotoxicity. Fresh IE-CTL prestimulated with IL-15 for 48 h, celiac IE-CTL clones, PB-CTL clones, and TALL-104 leukemia CTLs were pretreated for 30 min with vehicle control or CF3 before the cytolysis assay. Depending on the nature of the CTLs studied, distinct E/T ratios were used to achieve similar levels of specific lysis (50:1 for fresh IE-CTLs and PB-CTLs, 15:1 for celiac IE-CTLs, and 5:1 for TALL-104 leukemia CTLs). Data are means ± SD of three independent experiments for each cell line. P values reported are from a Tukey-adjusted pairwise comparison. (left) Cytolysis of FcγR⁺ P815 cells was significantly blocked by 20 µM CF3 in the presence of anti-NKG2D, but not anti-CD3, antibody. (right) Cytolysis of C1R/MIC transfectants was blocked upon pretreatment of CTLs with 20 µM CF3 and restored when 100 µM AA was added 1 h after the beginning of the cytolysis assay. Importantly, CF3 did not affect the level of intracellular granzyme and perforin expression (Fig. S2). (B) Knockdown of cPLA₂ in TALL-104 CTLs significantly impairs NKG2D-mediated cytolysis, but not TCR-mediated cytolysis under optimal conditions of stimulation. Untransfected cPLA₂ and control siRNA-transfected TALL-104 CTLs were studied 72 h after transfection. (left) cPLA₂ siRNA significantly decreases cPLA₂ expression without affecting JNK and ERK activation. Untransfected and transfected TALL-104 were cultured for 72 h, serum-starved overnight, and stimulated for 5 min (p-ERK) or 15 min (p-JNK), respectively, with anti-NKG2D. Cell lysates were analyzed by Western blot with anti-cPLA₂, then stripped and blotted with anti-phospho-ERK (p-ERK) or -JNK (p-JNK) mAbs. Right panel: cPLA₂ siRNA inhibited significantly NKG2D-mediated cytolysis in a redirected cytolytic assay using ⁵¹Cr-labeled FcγR⁺ P815 targets. (C) cPLA₂ knocked down in PB-CTLs blocked the release of cytolytic granule. An esterase assay was performed 36 h after transfection. Data are means ± SD of six experimental points from two independent experiments. P values are from a Tukey-adjusted pairwise comparison. (D) AA reconstitutes NKG2D cytolytic function against C1R/MIC⁺ target cells in TALL-104 transfected with cPLA₂ siRNA. Rescue experiments were performed by adding 50 µM of AA or a control polyunsaturated fatty acid LA after 1 h co-culture with target cells. (left) Knock-down of cPLA₂ did not affect 5-LO expression, whereas 5-LO siRNA decreased significantly 5-LO expression without affecting cPLA₂ expression. Cells were sorted into transfected (GFP⁺) and untransfected cells (GFP^-) cell populations 24 h after transfection with cPLA₂ siRNA, control siRNA or 5-LO siRNA in addition to GFP plasmids. Cell lysates were analyzed 12 h later by Western blot with anti-cPLA₂ antibody, then stripped and immunoblotted with anti-5-LO antibody. (middle) Cytolysis assays were performed 36 h after transfection at the indicated E:T ratios in transfected (GFP⁺) and untransfected (GFP^-) cells. One representative experiment out of two independent experiments is shown. (right) Cells were cultured for 72 h after transfection with cPLA₂ or control siRNA and cytolysis assays were performed in absence of cell sorting on the total cell population. The transfection efficiency, determined by GFP expression, was ∼60%. Data are means ± SD of six experimental points from two independent experiments. P values are from a Tukey-adjusted pairwise comparison.