Preferential Binding of Plasmodium falciparum SERA and Rhoptry Proteins to Erythrocyte Membrane

Transcription

1 INFECTiON AND IMMUNITY, Apr. 1994, p /94/$ Copyright 1994, American Society for Microbiology Vol. 62, No. 4 Preferential Binding of Plasmodium falciparum SERA and Rhoptry Proteins to Erythrocyte Membrane Inner Leaflet Phospholipids MARGARET E. PERKINS* AND ANNEMARIE ZIEFER Laboratory of Biochemical Parasitology, The Rockefeller University, New York, New York Received 7 October 1993/Returned for modification 20 December 1993/Accepted 11 January 1994 Proteins of an apical organelle, the rhoptry, of Plasmodium falciparum are secreted into the host erythrocyte membrane during merozoite invasion. To identify the membrane-binding site for rhoptry proteins, we examined the binding of parasite proteins to phospholipid vesicles. A specific interaction between the rhoptry proteins of 140, 130, and 110 kda to vesicles containing phosphatidylserine and phosphatidylinositol was observed. Both phospholipids are preferentially localized on the inner leaflet of the bilayer. Binding to other phospholipids, including sphingomyelin, was considerably less. In addition, the 120-kDa serine repeat antigen known as SERA, which was determined to be present on the merozoite, bound to phosphatidylserine vesicles and much less to vesicles of other phospholipids. Both the rhoptry and SERA proteins exhibited a preference for phosphatidylserine with short acyl side chains. Specific binding of SERA and the rhoptry proteins to phospholipids of the inner leaflet of membranes suggests a possible mechanism by which the proteins facilitate invasion into host cells. Pathogenic microorganisms have evolved various strategies to invade mammalian cells and survive intracellularly. Certain features, however, are common to many pathogens. Recognition of the correct host cell is usually regulated by receptors. Invasion occurs within an endocytotic vacuole. Entry into nonphagocytic host cells may involve subversion of transmembrane signalling mechanisms that result in reorganization of the cytoskeleton around the endocytotic vacuole much in the same way that growth factors and cytokines are endocytosed (30). Once inside the host cell, the pathogens appear to thwart lysosomal destruction by secreting factors that block endosome-lysosome fusion or inactivate lysosomal enzymes. Many of these processes, which have been demonstrated for certain bacterial pathogens, also hold true for intracellular protozoan parasites although the detailed knowledge of the mechanism of invasion for many parasites is not yet known (21). One feature that differentiates plasmodium and other parasites of the phylum Apicomplexa, such as Toxoplasma gondii and Babesia bovis, from other intracellular parasites is that invasion only proceeds when the apical end of the parasite is aligned with the erythrocyte. This so-called apical reorientation is taken to imply that contents of the apical organelles, the rhoptries and micronemes, have a role in invasion and must interact with the erythrocyte plasma membrane. Although many of the rhoptry proteins of Plasmodium falciparum have been identified and their primary structure is known (6, 29), we have little idea of the mechanism by which they interact with the erythrocyte and function to facilitate invasion. Electron microscopy shows that contents of the rhoptries are exocytosed, and immunofluorescence studies indicate that a 110- kda rhoptry protein is secreted into the host plasma membrane during invasion (1, 32). It has been proposed that the rhoptry proteins and possibly lipid are responsible for the * Corresponding author. Mailing address: Laboratory of Biochemical Parasitology, Rockefeller University, 1230 York Ave., New York, NY Phone: (212) Fax: (212) Electronic mail address: perkinm@rockvax.rockefeller.edu. formation of the endocytotic vacuole membrane known as the parasitophorous vacuole membrane (17, 26). The aim of the present study was to identify the membrane target of the rhoptry proteins in an effort to understand their function in invasion. To identify such sites initially, we examined binding of parasite proteins to intact erythrocytes and membranes. The major merozoite surface protein that binds to intact erythrocytes is MSP-1, the merozoite surface glycoprotein of 195 kda, as reported previously (27). Several other parasite proteins bound weakly to erythrocytes but more strongly to membranes. Binding of rhoptry proteins of 140, 130, and 110 kda, which form a complex known as RHOP-H (16), to membranes was not significantly affected by proteases, suggesting that the primary binding site of the proteins may be lipid. Thus, we examined binding to phospholipid vesicles. MATERIALS AND METHODS In vitro culture of P. falciparum and metabolic labeling. The following isolates of P. falciparum were cultured (34): FCR-3 (Gambia) and Kl (Thailand). For metabolic labeling, synchronous cultures were concentrated at the mid-schizont stage by centrifugation on a 60% Percoll gradient (11). Parasites were then labeled with [35S]methionine (100,uCi/ml) for 6 h. Typically, this was 8 h prior to the beginning of merozoite release. The radiolabeled medium was exchanged for RPMI medium-10% human serum. The spent culture supernatant was collected for 6 h, replaced with fresh medium, and collected 14 h later. The later culture supernatant was used for binding assays since we had observed that rhoptry proteins are more abundant in this fraction. Erythrocyte binding assay. For the binding assay, normal human erythrocytes, type A positive, were used. [35S]methionine-labeled culture supernatant (0.3 ml) was added to erythrocytes (0.1 ml) for 15 min at room temperature. The cell suspension was layered over silicon oil (Dow Corning Corp., Midland, Mich.) and centrifuged in a 1.5-ml Eppendorf tube for 2 min at 14,000 x g. The pellet was mixed with 0.5 M NaCl in PBS buffer (0.08 ml) (16) and centrifuged for 2 min at 14,

2 1208 PERKINS AND ZIEFER x g. The supernatant was removed and analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS- PAGE). Membranes were prepared by isotonic lysis of washed erythrocytes in cold 10 mm Tris buffer (ph 8.0). Membranes were collected by centrifugation at 14,000 x g in a microcentrifuge. For binding assays, membranes were incubated with [35S]methionine-labeled culture supernatant for 30 min. Unbound proteins were removed by washing membranes in cold RPMI medium twice by centrifugation at 14,000 x g. Bound proteins were eluted with 0.5 M NaCl. Preparation of liposomes. The phospholipids (Avanti Polar Lipids, Alabaster, Ala.) used were phosphatidylserine (PS; dilauroyl, dimyristoyl, dipalmitoyl, distearoyl, dioleoyl, dilinoleoyl), phosphatidylethanolamine (PE; dipalmitoyl), phosphatidylcholine (PC; dipalmitoyl), phosphatidylinositol (PI; from bovine liver, predominantly stearoyl-arachidonyl), and sphingomyelin (from egg). For the initial binding assays, all phospholipids used had dipalmitoyl side chains except for PI, which was available only with stearoyl-arachidonyl side chains. Cholesterol was from Sigma. Liposomes were prepared by the reverse-phase evaporation method as follows. Phospholipids (40 mg) were mixed in 1.2 ml of chloroform, and then 0.2 ml of phosphate-buffered saline (PBS) was added (ratio, 1:6). The mixture was sonicated for 10 min on ice until a homogeneous suspension was formed. The chloroform was removed by evaporation under reduced pressure at 100 to 400 mm Hg (1 mm Hg = Pa). The liposomes were mixed with 1.0 ml of ice-cold RPMI medium, vortexed, centrifuged for 4 min at 4 C, and used immediately in binding assays. Liposomes prepared by the reverse-phase evaporation method are large unilamellar vesicles and have a mean diameter of 0.16,um (33). Liposome binding assay. Liposomes (10 mg of lipid) were incubated with [35S]methionine-labeled culture supernatant (0.3 ml) for 30 min, centrifuged for 4 min at 4 C, washed in ice-cold RPMI medium twice, and then eluted with 0.08 ml of NaCl (0.5 M). The fraction eluted with NaCl was solubilized in sample buffer and analyzed by SDS-PAGE (27). For binding in the presence of Ca2, Ca2+ (2 mm) was added to culture supernatant prior to the addition to liposomes. The presence of proteins on liposomes after the NaCl wash was determined by boiling the liposomes in sample buffer and analyzing them by SDS-PAGE. Immunoprecipitation. For immunoprecipitation of parasite proteins from total supernatant (0.05 ml) and from NaCl eluates from erythrocytes or liposomes (0.08 ml), the samples were diluted with water (0.1 ml). The following monoclonal antibodies (MAbs) were used: MAb 1B9 directed against RHOP-H (32) and MAb 5E3 directed against SERA (8). MAb 5E3 was a generous gift from Jeff Lyon. MAbs were precoupled to goat anti-mouse immunoglobulin G (Cappel Laboratories, Cochranville, Pa.). Sepharose-coupled MAbs were incubated with total supernatant or NaCl eluate for 1 h and washed twice with buffer containing 1% Nonidet P-40 and 1% bovine serum albumin and once with PBS buffer containing 1% Nonidet P-40. Precipitates were analyzed by SDS-PAGE. Immunofluorescence. Smears of cultures of P. falciparum in the process of reinvading were fixed in cold acetone for S min, overlaid with MAb 5E3 for 30 min, washed in PBS, overlaid with fluorescein isothiocyanate-goat anti-mouse immunoglobulin G (Cappel Laboratories) for 30 min, washed in PBS, and mounted in 50% glycerol in PBS. Fluorescence was viewed with a Nikon epifluorescence microscope. a b c d " ma-l 25 -,,_ OP-H L i-sera -110 FIG. 1. Identification of P. falciparum proteins binding to erythrocyte membranes. [35S]methionine-labeled proteins from P. falciparum (Kl isolate) culture supernatant were added to erythrocytes and membranes. After being washed in PBS, bound proteins were eluted with 0.5 M NaCl and immunoprecipitated with MAb 1B9 against RHOP-H and MAb 5E3 against SERA. Lanes: a, NaCl eluate of proteins bound to normal erythrocytes; b, NaCl eluate of proteins bound to membranes; c, immunoprecipitate of proteins bound to membranes with MAb 1B9; d, immunoprecipitate with MAb 5E3. Lanes c and d were from a different SDS-polyacrylamide gel that had been electrophoresed for an extended time to clearly separate the components of RHOP-H from those of SERA. The numbers to the left of lane a and the right of lane d are molecular weights (in thousands). RESULTS INFECT. IMMUN. Binding of P.fakciparum proteins to erythrocyte membranes. The major P. falciparum protein binding to intact human erythrocytes is a protein of 195 kda (Fig. 1, lane a) which can be immunoprecipitated by monospecific antibodies to the major merozoite surface protein, MSP-1 (27), and the microneme protein EBA-175 (23). Several proteins in the molecular size range of 100 to 140 kda bind weakly to erythrocytes but significantly stronger to membranes (lane b). The major membrane-binding proteins were identified as the rhoptry complex of 140, 130, and 110 kda by immunoprecipitation with MAb 1B9 (lane c) and the SERA antigen by immunoprecipitation with MAb 5E3 (lane d). Lanes c and d of Fig. 1 were from an SDS-polyacrylamide gel electrophoresed for an extended time to separate clearly the 140-kDa protein from the 130-kDa one. In this gel, it appears that there is a doublet at 140 kda. MAb 5E3 immunoprecipitates several minor proteins of 140 and 60 kda. Proteins of 250, 50, and <10 kda also bound to membranes (lane b). Binding of the RHOP-H complex and SERA proteins to membranes was unaffected by treatment of membranes with trypsin, chymotrypsin, and neuraminidase (data not shown), prompting the idea that they bound to the membrane lipid. Binding of parasite proteins to phospholipids. To identify the possible lipid binding sites for parasite proteins, binding to phospholipids and cholesterol was examined (Fig. 2). The RHOP-H complex bound most strongly to vesicles of PS and PI (Fig. 2, lanes b, h, and i). Binding to other phospholipids, PC and PE, could be detected upon longer exposure of the radioactivity but was considerably less than the binding to PS

3 VOL. 62, 1994 MALARIAL INVASION PROTEINS BIND PHOSPHOLIPIDS 1209 a b c d e f g h i i k I m n o 200-., i;i 97- -I Ca+ + PC PE + ri ri C PS immunoprecipi tate FIG. 2. Binding of RHOP-H and SERA to phospholipid vesicles. Lanes: a, [35S]methionine-labeled culture supernatant (FCR-3 isolate) used as starting material; b to k, proteins bound to and eluted with NaCl from vesicles of PS (lane b), PC (lane c), PC and Ca2+ (2 mm; lane d), PE (lane e), PE and Ca2" (2 mm; lane f), sphingomyelin (lane g), PI (lane h), PI and Ca2+ (2 mm; lane i), cholesterol (lane j), and PS (lane k); 1 and m, immunoprecipitation from proteins bound to PS vesicles with MAb 1B9 (lane 1) and MAb 5E3 (lane m); n and o, immunoprecipitation from culture supernatant with MAb 1B9 (lane n) and MAb 5E3 (lane o). The acyl side chain of PS, PC, and PE was dipalmitoyl. The acyl side chain of PI was stearoyl-arachidonyl. The numbers to the left of lane a and the right of lanes i and o are molecular weights (in thousands). and PI (lanes c and e). Insignificant amounts bound to sphingomyelin (lane g) and cholesterol (lane j). RHOP-H bound to PS was identified by immunoprecipitation with MAb 1B9 (lane 1). A different pattern of binding was observed for the SERA protein. SERA bound to PS, considerably less to PC and PE, and insignificantly to PI (Fig. 2, lanes b to i). Binding of SERA to PS vesicles was confirmed by immunoprecipitation of bound proteins with MAb 5E3 (lane m). Ca2+ is known to cause the fusion of liposomes, resulting in the formation of large cochleate cylinders (33). Also, since PS and PI are negatively charged phospholipids, and Ca2+ often influences binding to this class of lipids, the effects of Ca2+ (2 mm) on binding were examined. Ca2+ moderately increased the binding of SERA to PC vesicles (lane d) but had little effect on binding to PS and PI. This would suggest that binding of RHOP-H to phospholipids was not influenced by the type or structure of the vesicles but by the inherent composition. This is important because PE vesicles can under some circumstances form cochleate cylinders. Binding of SERA to vesicles of sphingomyelin and cholesterol was not detected (lanes g and j). No proteins remaining bound to liposomes after the NaCl elution were detected by SDS-PAGE. Phospholipid binding specificity was further explored by examining binding to vesicles of PS with different fatty acid side chains. SERA exhibited a preference for short-chain fatty acids: binding to dilauroyl PS was greater than to PS with longer side chains (Fig. 3, lane a). The 130-kDa protein of the RHOP-H complex did not bind to dilauroyl PS (lane a). The 110-kDa protein of the RHOP-H complex exhibited a strong preference for dimyristoyl PS (lane b). Binding of RHOP-H to dimyristoyl (14:0) and dipalmitoyl (16:0) was greater than to distearoyl (18:0; lanes c to e) but increased for unsaturated acyl side chains (lanes g and h). Binding of additional proteins to vesicles of PS having unsaturated fatty acid side chains, dioleoyl (18:1) and dilinoleyl (18:2), was observed (lanes g and h). Ca2' had little effect on binding to PS (lanes d and f). The different binding specificities of SERA and RHOP-H suggest that they are not binding as a complex. Immunofluorescence localization of SERA during merozoite invasion. By immunofluorescence, SERA was localized on the extracellular merozoites (Fig. 4a and b) and intracellularly in the ring-infected erythrocyte (Fig. 4c and d). From the micrographs, it is not possible to determine whether SERA is localized on the merozoite surface or underneath the membrane; however, it does appear to have a polar distribution on both extracellular and intracellular parasites. Some SERA appears to be shed from the merozoite in the immediate vicinity of the released merozoite (Fig. 4b), suggesting that it is 120- a b c d e f g h :0 14:0 16:0 16:0 18:0 18:0 18:1 18:2 Ca FIG. 3. Binding of RHOP-H and SERA to vesicles of PS with different acyl side chains. Lanes contain proteins eluted from vesicles of dilauroyl PS (12:0; lane a), dimyristoyl PS (14:0; lane b), dipalmitoyl PS (16:0; lane c), dipalmitoyl PS and Ca2" (lane d), distearoyl PS (18:0; lane e), distearoyl PS and Ca2+ (lane f), dioleoyl PS (18:1; lane g), and dilinoleyl PS (18:2; lane h). The numbers to the left of lane a and the right of lane h are molecular weights (in thousands).

4 1210 PERKINS AND ZIEFER INFECT. IMMUN. Downloaded from FIG. 4. Immunofluorescence localization of SERA during merozoite invasion of erythrocytes. (a and b) Released merozoite; (c) newly invaded ring-infected erythrocyte; (d) ring-infected erythrocytes and schizont-infected erythrocyte (S). derived from the merozoite surface. In the ring-infected erythrocyte, it appears that SERA is associated with the ring or parasitophorous vacuole membrane. A detailed electron microscopic study will be required to clarify the exact localization of SERA at different stages. The results shown in Fig. 4 indicate that SERA is associated with the extracellular merozoite and is transferred with the parasite to the ring-infected stage. The SERA detected on ring stages is not due to de novo synthesis because SERA is not synthesized until the trophozoite stage. DISCUSSION Invasion of erythrocytes by the malarial merozoites is a two-step process. The first step is host cell recognition and attachment, determined by binding to an erythrocyte receptor. In the case of P. falciparum, the most lethal and widespread of the human malarial parasites, the major erythrocyte receptor has been identified as glycophorin (24, 25). The receptor for Plasmodium vivax, another species infecting humans, has been identified as the Duffy glycoprotein (3, 20). A major P. falciparum merozoite surface protein, MSP-1, and a microneme protein, EBA-175, bind to glycophorin and may be the parasite proteins that mediate host cell recognition and attachment (7, 27). The second step of invasion is internalization, which involves entry in the parasitophorous vacuole. Receptor binding alone does not appear to be sufficient to trigger endocytosis as it would in phagocytic entry. It is commonly assumed that the parasite factors that orchestrate parasite entry are localized in the apical organelles, the micronemes and rhoptries. The origin of the parasitophorous vacuole membrane has been debated (17). It has been proposed that it is derived from the host or the parasite. Although the vacuole membrane does not appear to contain host protein, it does contain erythrocytederived lipid (12). Other studies indicate that lipid derived from the apical end of the merozoite enters the host erythrocyte possibly in the parasitophorous vacuole (10, 19). These two models are not mutually exclusive, and it is possible that the vacuole contains lipid derived from both the parasite and host (12). We have suggested that the proteins of the rhoptry may be organized in a membrane-like structure in the rhoptry organelle since they are solubilized in the isolated organelle by phospholipases (11). Electron microscopy studies strongly suggest that they are secreted in a lamellar form (1). RHOP-H complex is secreted into the erythrocyte membrane during invasion (32), and it is possible that RHOP-H and other rhoptry proteins, associated with lipid, integrate into the host plasma membrane to initiate formation of the parasitophorous vacuole. How such an integration occurs is of considerable interest and the impetus for the present study. We have been able to demonstrate in this study that the rhoptry proteins of 140, 130, and 110 kda, which form a on December 7, 2018 by guest

5 VOL. 62? 1994 MALARIAL INVASION PROTEINS BIND PHOSPHOLIPIDS 1211 protein complex known as RHOP-H, specifically bind to PS and PI. Unexpectedly, we found that another well-characterized merozoite protein, known as SERA, also binds to phosphatidylserine. Both the RHOP-H complex and SERA bound only weakly to intact erythrocytes (Fig. 1). This may be due to the inaccessibility of the phospholipids in the bilayer since they are probably covered by surface glycoproteins, most notably glycophorin. Disruption of the surface proteins during the preparation of membranes would expose these sites. The weak binding of RHOP-H and SERA to intact cells made it difficult to test the inhibitory effects of the phospholipids. We found that high concentrations of Ca 2 (2 mm) do increase the amount of RHOP-H and SERA bound, and under these circumstances, PS inhibits the binding of RHOP-H and SERA to intact erythrocytes (data not shown). SERA was originally identified by an MAb that blocks merozoite invasion (2, 28). Furthermore, immunization of Aotus monkeys with recombinant fragments of SERA have been demonstrated recently to be protective against challenge with P. falciparum (15). The primary structure of the protein indicates that there is limited sequence homology with the cysteine protease-active site (13, 18). Because of its apparent localization in the parasitophorous vacuole space, SERA has been proposed to play a role primarily in merozoite release or at the merozoite surface (9). Binding to lipids of the parasitophorous vacuole membrane and erythrocyte plasma membrane could certainly facilitate membrane lysis. However, we observed that a considerable amount of SERA is still present on free merozoites and also on the newly invaded parasite although some of the protein is shed at the merozoite stage (Fig. 4). Detailed electron microscopic studies will be required to determine the exact localization of SERA at the merozoite and ring-infected stage. Localization of SERA on the merozoite surface is suggested by the studies of Chulay et al. (8), who demonstrated that merozoites could be agglutinated with antibody directed against SERA. Furthermore, we have observed in invasion assays that the presence of MAb 5E3 agglutinates free merozoites. Thus, it is possible that SERA interacts with the host erythrocyte membrane during invasion. This would explain the fact that MAbs to SERA are effective in blocking invasion (2, 28). In a previous study, we had shown that RHOP-H proteins bind to phospholipid vesicles (31). In that study, RHOP-H appeared to exhibit a similar preference for PS, PC, and PI. However, there are several differences in the preparation of vesicles between the two studies. In the previous study, the phospholipids used contained mixed fatty-acid side chains and were prepared as mixtures with cholesterol. In this study, we used phospholipids of greater purity and with identical acyl side chains to compare binding preference. Furthermore, cholesterol is known to modify packing of the acyl side chains in the vesicle, rendering the vesicles thicker and decreasing their permeability (5, 33). Both PS and PI are localized preferentially on the inner leaflet of the membrane (22). Sphingomyelin and PC are the major outer leaflet phospholipids in erythrocytes, and PE and PS are the major inner leaflet lipids. PI is a minor inner leaflet phospholipid, constituting less than 1% of the total (14). Thus, we could propose that selective binding and integration of RHOP-H and SERA to the inner leaflet could result in inward expansion of the cytoplasmic side of the bilayer, initiating the formation of the parasitophorous vacuole membrane. By not binding to the major lipids, the bilayer would not be destabilized to the point of lysis. The biological importance of the preference for short acyl side chains is not clear since these are a relatively minor fraction of the total lipids. The apparent differential binding of the 1 10-kDa protein for the short-chain phospholipids is not readily explained since we know that the three polypeptides of RHOP-H bind as a complex; the three can be coprecipitated from the NaCl-eluted fraction with an MAb against one of the proteins, MAb 11B9. It is possible, however, that, after binding, one of the proteins of the complex localizes inside the phospholipid vesicle and is less influenced by the washing treatment and thus appears to be enriched in the NaCl-eluted fraction. We intend to perform electron microscopy studies of the phospholipid vesicles after binding to determine the localization of the RHOP-H complex within the bilayer. The consequences of SERA and RHOP-H binding to phospholipids should be viewed in the context of the immunofluorescence localization of these proteins during and after merozoite invasion. In the present study, we show that SERA is present inside the infected erythrocyte after invasion, on the newly invaded parasite or vacuole membrane. Electron microscopic studies will be required to determine its exact localization. In contrast, RHOP-H is localized in both the plasma membrane of the erythrocyte and the parasitophorous vacuole membrane as shown by immunofluorescence and electron microscopic studies (32). Thus, RHOP-H spreads out along the bilayer from the initial point of secretion, whereas SERA appears only to be localized on the membrane at the site of formation of the vacuole. We could propose that SERA is the primary protein involved in expansion of the bilayer to form the vacuole and that RHOP-H has some other function in invasion, perhaps in resealing the plasma membrane. SERA, if present on the surface, would presumably make contact with the erythrocyte before rhoptry proteins in the normal sequence of events during invasion. The finding that SERA and RHOP-H have different binding specificities also indicates that they bind separately and have distinct functions. It is interesting to note that the YopH protein of Yersinia spp. has been proposed to be translocated across the lipid bilayer during the invasion of these bacteria into epithelial cells (4). The translocation may modify phosphorylation of cytoskeletal proteins, resulting in the reorganization necessary for invasion (4). It is possible that binding of SERA and RHOP-H or other rhoptry proteins either directly or indirectly results in the disruption of the interaction of the cytoskeleton with transmembrane proteins in a similar fashion as has been proposed for Yersinia spp. (30). ACKNOWLEDGMENTS We acknowledge the technical help of Catherine Rozario. We are grateful to J. Lyon for the gift of MAb 5E3. This work was supported by National Institutes of Health grant Al to M.E.P. REFERENCES 1. Bannister, L. H., and G. H. Mitchell The fine structure of secretion by Plasmodium knowlesi merozoites during red cell invasion. J. Protozool. 36: Banyal, H. S., and J. Inselburg Isolation and characterization of parasite-inhibitory Plasmodium falciparum monoclonal antibodies. Am. J. Trop. Med. Hyg. 34: Barnwell, J. W., M. E. Nichols, and P. Rubinstein In vitro evaluation of the role of the Duffy blood group in erythrocyte invasion by Plasmodium vivax. J. Exp. Med. 169: Bliska, J. B., J. E. Galan, and S. Falkow Signal transduction in the mammalian cell during bacterial attachment and entry. Cell 73: Bretscher, M. S., and S. Munro Cholesterol and the Golgi apparatus. Nature (London) 261: Brown, H. J., and R. L. Coppel Primary structure of a

6 1212 PERKINS AND ZIEFER Plasmodium falciparum rhoptry antigen. Mol. Biochem. Parasitol. 49: Camus, D., and T. J. Hadley A Plasmodium falciparum antigen that binds to host erythrocytes and merozoites. Science 230: Chulay, J. D., J. A. Lyon, J. D. Haynes, A. I. Meierovics, C. T. Atkinson, and M. Aikawa Monoclonal antibody characterization of Plasmodium falciparum antigens in immune complexes formed when schizonts rupture in the presence of immune serum. J. Immunol. 139: Delplace, P., B. Fortier, G. Tronchin, J.-F. Dubremetz, and A. Vernes Localization, biosynthesis, processing and isolation of a major 126 kda antigen of the parasitophorous vacuole of Plasmodium falciparum. Mol. Biochem. Parasitol. 23: Dluzewski, A. R., G. H. Mitchell, P. R. Fryer, S. Griffiths, R. J. M. Wilson, and W. B. Gratzer Origins of the parasitophorous vacuole membrane of the malaria parasite, Plasmodium falciparum, in human red blood cells. J. Cell. Sci. 102: Etzion, Z., M. C. Murray, and M. E. Perkins Isolation and characterization of rhoptries of Plasmodium falciparum. Mol. Biochem. Parasitol. 47: Haldar, K., and L. Uyetake The movement of fluorescent endocytic tracers in Plasmodium falciparum infected erythrocytes. Mol. Biochem. Parasitol. 50: Horii, T., D. J. Bzik, and J. Inselburg Characterization of antigen-expressing Plasmodium falciparum cdna clones that are reactive with parasite inhibitory antibodies. Mol. Biochem. Parasitol. 30: Hsiao, L. L., R. J. Howard, M. Aikawa, and T. F. Taraschi Modification of host cell membrane lipid composition by the intra-erythrocytic human malaria parasite Plasmodium falciparum. Biochem. J. 274: Inselburg, J., D. J. Bzik, W.-B. Li, K. M. Green, J. Kansopon, B. K. Hahm, I. C. Bathurst, P. J. Barr, and R. N. Rossan Protective immunity induced in Aotus monkeys by recombinant SERA proteins of Plasmodium falciparum. Infect. Immun. 59: Jaikaria, N. S., C. Rozario, R. G. Ridley, and M. E. Perkins Biogenesis of rhoptry organelles in Plasmodium falciparum. Mol. Biochem. Parasitol. 57: Joiner, K. A Rhoptry lipids and parasitophorous vacuole formation: a slippery issue. Parasitol. Today 7: Knapp, B., E. Hundt, U. Nau, and H. A. Kupper Molecular cloning, genomic structure and localization in a blood stage antigen of Plasmodium falciparum characterized by a serine stretch. Mol. Biochem. Parasitol. 32: Mikkelsen, R. B., M. Kamber, K. S. Wadwa, P.-S. Lin, and R. Schmidt-Ullrich The role of lipids in Plasmodium falcipa- INFECT. IMMUN. rum invasion of erythrocytes: a coordinated biochemical and microscopic analysis. Proc. Natl. Acad. Sci. USA 85: Miller, L. H., S. J. Mason, J. A. Dvorak, M. H. McGinniss, and I. K. Rothman Erythrocyte receptors for (Plasmodium knowlesi) malaria: Duffy blood group determinants. Science 189: Moulder, J. W Comparative biology of intracellular parasitism. Microbiol. Rev. 49: Op den Kamp, J. A. F Lipid asymmetry in membranes. Annu. Rev. Biochem. 48: Orlandi, P. A., B. K. L. Sim, J. D. Chulay, and J. D. Haynes Characterization of the 175-kilodalton erythrocyte binding antigen of Plasmodium falciparum. Mol. Biochem. Parasitol. 40: Pasvol, G., J. S. Wainscoat, and D. J. Weatherall Erythrocytes deficient in glycophorin resist invasion by the malarial parasite, Plasmodium falciparum. Nature (London) 297: Perkins, M. E Inhibitory effects of erythrocyte membrane proteins on the in vitro invasion of the human malarial parasite (Plasmodium falciparum) into its host cell. J. Cell Biol. 90: Perkins, M. E Rhoptry organelles of apicomplexan parasites. Parasitol. Today 8: Perkins, M. E., and L. J. Rocco Sialic acid-dependent binding of Plasmodium falciparum merozoite surface antigen, Pf200, to human erythrocytes. J. Immunol. 141: Perrin, L. H., B. Merkii, M. Loche, C. Chizzolini, J. Smart, and R. Richle Antimalarial immunity in Saimiri monkeys. Immunization with surface components of asexual blood stages. J. Exp. Med. 160: Ridley, R. G., B. Takacs, H.-W. Lahm, C. J. Delves, M. Goman, U. Certa, H. Matile, G. R. Woollett, and J. G. Scaife Characterisation and sequence of a protective rhoptry antigen from Plasmodium falciparum. Mol. Biochem. Parasitol. 41: Rosenshine, I., and B. B. Finlay Exploitation of host signal transduction pathways and cytoskeletal functions by invasive bacteria. BioEssays 15: Sam-Yellowe, T. Y., and M. E. Perkins Interaction of the 140/130/110 kd rhoptry protein complex of Plasmodium falciparum with the erythrocyte membrane and liposomes. Exp. Parasitol. 73: Sam-Yellowe, T. Y., H. Shio, and M. E. Perkins Secretion of Plasmodium falciparum rhoptry protein into the plasma membrane of host erythrocytes. J. Cell Biol. 106: Szoka, F., Jr., and D. Papahadjopoulos Comparative properties and methods of preparation of lipid vesicles (liposomes). Annu. Rev. Biophys. Bioeng. 9: Trager, W., and J. B. Jensen Human malaria parasites in continuous culture. Science 193: