Binding of fluid phase C3b to nonsensitized bystander human red cells. A model for in vivo effects of complement activation on blood cells.

There is evidence that activated complement fragments (C3b) can bind to human red cells (RBCs) serving as adsorbing surfaces, but not as antigens. This evidence prompted the present in vitro experiments. Using the standard antiglobulin test (AGT), the radioimmune antiglobulin test (RIAT), and the immune adherence test, we found that C3 can indeed be attached to "bystander" human RBCs if complement is activated either through the classical pathway (anti-A hemolysins plus blood group A1 RBCs) or the alternative pathway (activator surfaces, i.e., Escherichia coli bacteria, rabbit RBCs, or inulin). On Scatchard plot analysis, between 24,000 and 44,000 radiolabeled anti-C3d molecules were bound per one adjacent "bystander" RBC, while untreated control RBCs, or RBCs preincubated with fresh compatible serum, bound only 200 to 300, and 600 to 800 molecules, respectively. Despite strong coating with C3 fragments, only "bystander" paroxysmal nocturnal hemoglobinuria, but not normal, RBCs were hemolyzed by complement activation, i.e., through E. coli at pH 8.0. RBCs were not coated with C3 when complement was activated in the fluid phase by heat-aggregated IgG or a staphylococcal "decomplementation antigen." We conclude that the findings of our in vitro experiments accurately mimic some old, but as yet insufficiently understood, in vivo effects of complement activation on circulating blood cells.