PfEMP1 is not exposed on the surface of erythrocytes infected with CS2ΔSBP1. (A) Intact erythrocytes infected with parental CS2, CS2ΔSBP1, and uninfected erythrocytes were treated with (+) or without (−) trypsin, and extracts were analyzed by Western blotting with rabbit (left) or mouse (right) anti-ATS antibody. Surface-exposed PfEMP1 is cleaved by trypsin, resulting in bands at 83 kDa. The blot was reprobed with anti-HSP70 antibodies to confirm equal loading. (B) PfEMP1 is more abundant in CS2ΔSBP1 saponin supernatant. Saponin pellets and supernatants of trophozoite-infected erythrocytes were analyzed by Western blotting with anti-ATS antibody. (C) Triton X-100–soluble PfEMP1 is more abundant in CS2ΔSBP1. Triton X-100, SDS soluble, and insoluble fractions of trophozoite-infected erythrocytes were analyzed by Western blotting with anti-ATS antibody. (D) More soluble PfEMP1 can be detected in SLO-treated CS2ΔSBP1-infected parasites. Four hemolytic units (HUs) of SLO were used to permeabilize the plasma membrane of infected erythrocytes. The soluble content of the erythrocyte cytosol was then analyzed for the presence of PfEMP1 with anti-ATS antibody. As a comparison the same experiment was done at a noneffective SLO concentration (2 HU). The rhoptry-associated protein RAP1 was used as a loading control. (E) Comparison of reactivity of sera from nonexposed (N1-N5) versus exposed persons (A-J) from Papua, New Guinea, toward erythrocytes infected with parental CS2 and with CS2ΔSBP1 is consistent with the lack of PfEMP1 on the host cell in the absence of PfSBP1 function. IgG binding to the surface of infected erythrocytes among sera from malaria-exposed multigravid women residing in PNG was significantly higher for CS2 than for CS2ΔSBP1 (P < .004, Wilcoxon signed rank sum test). There was little IgG reactivity among sera from Australian residents not exposed to malaria. IgG binding was measured by flow cytometry, and values (arbitrary units) represent the mean + range of samples tested in duplicate.