Time-dependent effect of ATLG and role of complement in the presence of ATLG on NK cell viability. (A-B) Isolated NK cells were either incubated with 2 μg/mL ATLG (A) or 1000 μg/mL ATLG (B) for 1, 2, 4, 6, and 24 hours, respectively. Induction of cell death was determined by FACS after annexin V/PI staining at these time points. A minimum of 50 000 events was analyzed per sample. Dead cells include early and late apoptotic (annexin V⁺/annexin V⁺ and PI⁺) and necrotic (PI⁺) cells. Shown are means and standard deviation of duplicates of 2 different donors. (C) Isolated NK cells were incubated with various ATLG concentrations, as indicated, for 24 hours in the presence (i) of native human serum in a ratio of 1:4; (ii) of heat-inactivated human serum; or (iii) without human serum. Induction of cell death was determined by FACS after annexin V/PI staining. A minimum of 50 000 events was analyzed per sample. No significant differences within 1 ATLG dose (0-100 μg/mL) could be shown independent of the presence/absence of (active) complement (Kruskal-Wallis test; Dunn multiple comparisons). Only at 1000 μg/mL ATLG, cell death rates differed significantly between samples with functional and with inactivated sera (P = .03). Shown are means of dead cells of 3 independent experiments, and bars indicate standard deviation. (D) Cell death rates of isolated NK cells (sum of annexin V⁺/annexin V⁺ and PI⁺/PI⁺ stained CD56⁺ NK cells) preincubated with native human serum (1:4) and ATLG, concentrations as indicated, for 24 hours. Same assay as in panel C, analyzed by Dunnett T3 test.