Binding of C3 fragments to PNH erythrocytes incubated in acidified C5-deficient NHS. PNH erythrocytes from patient 2 were incubated in acidified C5-deficient human serum (C5def NHS; A-C), in acidified C5-deficient human serum chelated with EDTA (D), or in aNHS (E). After washing, the cells were stained with a combination of fluorescently labeled mAbs specific for activation and degradation products of C3 (Table 1) and subsequently analyzed by flow cytometry. In panels A and B, the cells were stained with a cocktail that contained 3 mAbs: Al488 mAb 3E7, PE mAb 1H8, and Al647 mAb 7C12. In panels C through E, only 2 mAbs, PE mAb 1H8 and Al647 mAb 7C12, were used to stain the cells. Cells that are doubly positive for both anti-C3 antibodies are contained in the area circumscribed by the rectangular area labeled R1. The readouts are similar in both panels A (all 3 mAbs used to stain, including 3E7) and C (only 2 mAbs used to stain), indicating that mAb 3E7 does not interfere with binding of mAbs 7C12 and 1H8. Dot plots show both erythrocytes (red) and ghosts (green). In the case of the cells incubated in acidified C5-deficient serum, the fluorescently labeled anti-C3b antibodies bound to intact erythrocytes because lysis did not occur, whereas in the case of the cells incubated in aNHS, the fluorescently labeled anti-C3b antibodies bound to ghost because the GPI-AP–deficient cells are hemolyzed. The low level of binding of PE mAb 1H8 to the erythrocytes observed along the horizontal axis of panels A and C through E is likely due to the presence of C3dg that was deposited on these cells in vivo, as the patient whose cells were used in this experiment was being treated with eculizumab. The results of 1 experiment, representative of 2, are illustrated.