Erythrocytes infected with a knobby variant of Plasmodium falciparum selectively bind IgG autoantibodies in normal human serum. Quantification of membrane-bound IgG, by use of '251-labeled protein A, revealed that erythrocytes infected with the knobby variant bound 30 times more protein A than did noninfected erythrocytes; infection with a knobless variant resulted in less than a 2-fold difference compared with noninfected erythrocytes. IgG binding to knobby erythrocytes appeared to be related to parasite development, since binding of 1251-labeled protein A to cells bearing young trophozoites (less than 20 hr after parasite invasion) was similar to binding to uninfected erythrocytes. By immunoelectron microscopy, the membrane-bound IgG on erythrocytes infected with the knobby variant was found to be preferentially associated with the protuberances (knobs) of the plasma membrane. The removal of aged or senescent erythrocytes from the peripheral circulation is reported to involve the binding of specific antibodies to an antigen (senescent antigen) related to the major erythrocyte membrane protein band 3. Since affinity-purified autoantibodies against band 3 specifically bound to the plasma membrane of erythrocytes infected with the knobby variant of P. fakiparum, it is clear that the malaria parasite induces expression of senescent antigen.Erythrocytes infected with the human malaria parasite Plasmodium falciparum become immunologically altered as a result ofparasite development. These changes, which occur primarily toward the end of the parasite's 48-hr developmental cycle, have been identified by techniques capable of detecting membrane-bound immunoglobulins (1-3). It has been proposed (4) that the newly exposed antigens on the surface of infected erythrocytes are the result of parasitesynthesized proteins that are inserted into the host cell membrane; however, in spite of intense efforts by several laboratories such parasite proteins have been neither fully characterized nor specifically localized, and the manner of transport from the parasite to the erythrocyte surface is yet to be convincingly demonstrated.It is entirely possible that some of the antigenic changes observed for malaria-infected erythrocytes result from parasiteprovoked exposure or modification of host cell antigens. For example, Kay (5) showed that immunoglobulins were specifically bound to the surface of old erythrocytes, and as a result in vitro phagocytosis by macrophages occurred. These autoantibodies, found in the sera ofnormal individuals, bound to the major erythrocyte integral membrane protein, band 3 (6, 7 Elution of membrane-bound IgG was performed according to the procedure of Rekvig and Hannesta (13), except that the incubation time in isotonic glycine buffer was increased to 5 min at 0WC. Cells were layered over a 10% (wt/vol) dextran cushion and centrifuged for 2 min in a Beckman Microfuge. The pelleted cells were washed five to seven times in PBS and then incubated with antibody (70 ,ug/ml) purified by band 3-Sepharose ...