Identification of the alloreactive subset at the population level. Case no. 5 was characterized by C1 KIR-ligand mismatch, because the patient expressed HLA-A*2402,*2601, -B*3502,*5301 (Bw6/Bw4), -C*0401,*0602 (C2/C2) while the donor expressed HLA-A*24,*26, -B*18,*53 (Bw6/Bw4), -C*1203,*0602 (C1/C2). (A) The alloreactive subset was evaluated in the donor and in the recipient (2 months after transplantation). NK cells were incubated with the mixture of GL-183-PE (anti-KIR2DL2/L3/S2), EB6B-PE (anti-KIR2DL1/S1), Z27-PE (anti-KIR3DL1/S1), DX9-FITC (anti-KIR3DL1), no.143211-FITC (anti-KIR2DL1) and Z199 (anti-NKG2A) followed by anti–IgG2b-FITC. The alloreactive subset resides in the top left quadrant: 36% in the donor and 34% in the recipient. (B) Donor (filled symbols)– and recipient (open symbols)–derived polyclonal NK-cell populations were tested in cytolytic assays against the patient's leukemia blasts (CD10⁺ ALL, squares) and the B-EBV BOB (circles) expressing the homozygous haplotype HLA-A*2402, -B*5101 (Bw4), -C*1502 (C2). Both effector cells efficiently lysed these target cells at different E:T ratios as shown. The results are representative of 2 independent experiments; the standard deviation of the mean of the triplicates was less than 5%.