A homologue of N-ethylmaleimide-sensitive factor in the malaria parasite Plasmodium falciparum is exported and localized in vesicular structures in the cytoplasm of infected erythrocytes in the brefeldin A-sensitive pathway.

N-Ethylmaleimide-sensitive factor (NSF) and its homologues play a central role in vesicular trafficking in eukaryotic cells. We have identified a NSF homologue in Plasmodium falciparum (PfNSF). The reported PfNSF gene sequence (GenBank accession number CAB10575) indicated that PfNSF comprises 783 amino acids with a calculated molecular weight of 89,133. The overall identities of its gene and amino acid sequences with those of rat NSF are 50.9 and 48.8%, respectively. Reverse transcription-polymerase chain reaction analysis and Northern blotting with total P. falciparum RNA indicated expression of the PfNSF gene. Polyclonal antibodies against a conserved region of NSF specifically recognized an 89-kDa polypeptide in the parasite cells. After homogenization of the parasite cells, approximately 90% of an 89-kDa polypeptide is associated with particulate fraction, suggesting membrane-bound nature of PfNSF. PfNSF was present within both the parasite cells and the vesicular structure outside of the parasite cells. The export of PfNSF outside of the parasite cells appears to occur at the early trophozoite stage and to terminate at the merozoite stage. The export of PfNSF is inhibited by brefeldin A, with 9 microM causing 50% inhibition. Immunoelectromicroscopy indicated that intracellular PfNSF was associated with organelles such as food vacuoles and that extracellular PfNSF was associated with vesicular structures in the erythrocyte cytoplasm. These results indicate that PfNSF expressed in the malaria parasite is exported to the extracellular space and then localized in intraerythrocytic vesicles in a brefeldin A-sensitive manner. It is suggested that a vesicular transport mechanism is involved in protein export targeted to erythrocyte membranes during intraerythrocytic development of the malaria parasite.

Plasmodium falciparum, a human malaria parasite, invades an erythrocyte during one stage of its life cycle. In an infected erythrocyte, the P. falciparum organism develops a membrane structure called the parasitophorus vacuolar membrane. The parasitophorus vacuolar membrane extends into the host cell cytoplasm and forms a complex membrane structure, thus called the tubovesicular membrane network (reviewed in Refs. [1][2][3]. These membrane systems outside the P. falciparum cell are important for the transport of various nutrients such as glucose, phospholipids, and amino acids and for extrusion of antimalarial agents so as to maintain suitable circumstances for them (3)(4)(5)(6). In addition to the formation of such intraerythrocytic membrane systems, P. falciparum cells also transport some proteins such as erythrocyte membrane protein-1 of P. falciparum (PfEMP1) 1 and PfEMP3 to the erythrocyte plasma membrane, which results in the formation of a knob-like structure on the surfaces of the infected erythrocytes. These proteins are responsible for protection against immunological attack and attachment of infected erythrocyte to endothelial cells, one of the crucial steps for cerebral malaria (1)(2)(3)(7)(8)(9)(10)(11). Importantly, the extraparasite protein transport process can not rely upon the endogenous transport machinery in the host cells, because mature erythrocytes are completely devoid of machinery for protein trafficking. Thus, the malaria parasite must transport proteins through the plasma membrane and the membrane structure in the cytoplasm of the host cells by means of their own mechanism, although the molecular pathway for the transport of proteins through the parasite plasma membrane is less understood.
proteins to the plasma membranes of parasitized erythrocytes. Consistently, a homologue of the ADP-ribosylation factor and a variety of small GTP-binding proteins, including a Sar1 homologue, which are components of the common machinery for membrane traffic, have been identified in P. falciparum (18 -22). However, in mature intraerythrocytic malaria parasites, a morphologically distinguishable Golgi apparatus has not yet been identified, although the cis-Golgi marker, PfERD2, is present separately with the trans-Golgi marker, PfRab6. Sphingomyelin synthase, a marker for the Golgi apparatus in other eukaryotes, is present at least partially in tubovesicular membrane networks (3,23,24). Upon treatment with BFA, Plasmodium proteins exported and localized to erythrocyte membrane such as merozoite surface protein-1 are accumulated in a novel compartment similar to but distinct from the endoplasmic reticulum within the malaria parasite (25). Thus, the mechanism of protein secretion through the malaria parasite plasma membrane seems to be unusual, and it may be more complex than that in other eukaryotes.
In eukaryotic cells, protein transport along the secretory pathways is mediated by vesicles that move between the organelles (26). Transport vesicles are formed from the donor compartment and are targeted to acceptor organelles, where they deliver cargo molecules through membrane fusion. The docking and/or fusion of transport vesicles with the target membranes is mediated by the supramolecular protein complex consisting of NSF, soluble NSF-attachment protein (SNAP), and receptors for soluble NSF attachment protein (SNARE) (27)(28)(29). During docking and/or fusion of transport vesicles, NSF may form a 20 S complex together with receptors for soluble NSF attachment protein at the target membrane (tSNARE) and vesicular receptors for soluble NSF-attachment protein (vSNARE) to trigger membrane fusion with the plasma membrane (30,31). It would be interesting to determine whether proteins involved in the above mentioned docking/ fusion of vesicles are present in malaria parasites.
Very recently, Bowman et al. (32) reported the complete nucleotide sequence of chromosome 3 of P. falciparum. In the sequence, they found a gene homologous to the NSF gene and called the protein MP03103 (GenBank TM accession number CAB10575). In the present study, we found that this protein is actually expressed and present in P. falciparum cells. To our surprise, part of P. falciparum NSF (PfNSF) appears to be exported from the parasite cells and localized in vesicular structures in the erythrocyte cytoplasm in a BFA-sensitive manner. These results are consistent with the idea that a supramolecular protein complex consisting of NSF, SNAP, and receptors for soluble NSF attachment protein is involved in the targeting of P. falciparum proteins to the plasma membranes of host erythrocytes.

EXPERIMENTAL PROCEDURES
P. falciparum Culture and Drug Treatment-P. falciparum strain FCR-3 cells were cultured in RPMI 1640 medium (Life Technologies, Inc.) containing 50 mg/liter gentamycin and 10% A ϩ human serum at a hematocrit of 5%, according to the method of Trager and Jensen (33). Erythrocytes exhibiting 0.3% parasitemia were added to each well of plates in 990 l of culture medium to give a final hematocrit of 3%. Then the plates were incubated at 37°C under 5% CO 2 , 5% O 2 , and 90% N 2 gas for 72 h. In some experiments, the cultures were synchronized by hemolysis of mature, trophozoite stage-parasitized erythrocytes by suspension in a 5% (w/v) sorbitol solution (34). Then BFA dissolved in ethanol was added to the cultures at the concentrations shown in Fig. 7. After incubation for 2 h, the parasitized erythrocytes were washed and used for further experiments.
Cell Fractionation-P. falciparum cells were obtained by saponin treatment as follows (34). Parasitized erythrocytes exhibiting 3.5-7% parasitemia (about 1 ϫ 10 8 cells) were washed three times with PBS containing 10 g/ml pepstatin A, 10 g/ml leupeptin, and 1 mM phen-ylmethylsulfonyl fluoride and suspended in 40 ml of the same buffer. Then saponin was added to a final concentration of 0.075%. The solution was incubated for 5 min at 4°C and centrifuged at 10,000 ϫ g for 10 min. The pellet (malaria parasites) was washed three times with the same buffer, suspended in 100 l of 20 mM MOPS-Tris buffer (pH. 7.0) containing 5 mM EDTA, 0.25 M sucrose, 10 g/ml pepstatin A, 10 g/ml leupeptin, and 1 mM phenylmethylsulfonyl fluoride, and kept in an ice bath until use. The supernatant, which contained the particulate fraction in the extraparasitic space, was centrifuged at 105,000 ϫ g for 1 h, and the pellet and supernatant were obtained. The pellet was suspended in the same buffer and designated as the extracellular particulate fraction.
Reverse Transcription-Polymerase Chain Reaction (RT-PCR) Analysis-Total RNA extracted from parasitized erythrocytes (1 g each) was transcribed into cDNA in a final volume of 20 l of reaction buffer containing 0.5 mM each dNTP, 10 mM dithiothreitol, 100 pmol of random octamers, and 200 units of Molony murine leukemia virus reverse transcriptase (Amersham Pharmacia Biotech). After a 1-h incubation at 42°C, the reaction was terminated by heating at 90°C for 5 min. For PCR amplification, the 100-fold diluted synthesized cDNA solution was added to the reaction buffer containing 0.12 mM dNTPs (30 M each dNTP), 25 pmol of primers, and 1.5 units of AmpliTaq-Gold polymerase (PerkinElmer Life Sciences). 35 temperature cycles were conducted as follows: denaturation at 94°C for 30 s, annealing at the temperature specific for each set of primers for 30 s, and extension at 72°C for 30 s. The amplification products were finally analyzed by polyacrylamide gel electrophoresis. The sequences of the oligonucleotides used as primers were based on published sequences (32). For amplification of the PfNSF gene, the specific sense primer was 5Ј-GGGAATATAGGGAGAGAA-GAA-3Ј (bases 163-183), and the antisense primer was 5Ј-AGCCACTA-AATCCCGAAGAAT-3Ј (bases 539 -559). The antisense primer (5Ј-CTCATCAAAATCGGCTGTCTT-3Ј (bases 475-495) was also used for 2 second amplification. For amplification of the parasite actin-1 gene (GenBank TM accession number M22719), the specific sense primer was 5Ј-GCAGCAGGAATCCACACAAC-3Ј (bases 1119 -1138), and the antisense primer was 5Ј-GTGGACAATACTTGGTCCTG-3Ј (bases 1402-1421).
Northern Blotting-Total RNA (25 g) isolated from parasitized eryhtrocytes were separated on a formaldehyde-agarose gel (1%) and transferred to a nylon membrane (Amersham Pharmacia Biotech). The immobilized RNA was probes with a cDNA fragment of PfNSF, amplified as described above, labeled with [ 32 P]dATP by random priming. After extensive washing, the membrane was subjected to autoradiography using BAS 1000 film (Fuji Film Co.).
Antibodies-Site-specific antibodies against PfNSF were raised in rabbits by injecting the following peptide conjugated to thyroglobulin with glutaraldehyde: 412 DLIDEALLRPGR 423 (which is conserved among mammalian NSFs; see Fig. 1). The antibodies recognized NSFs of mammalian origin and plant sources (data not shown). Site-specific polyclonal antibodies against vacuolar H ϩ -ATPase subunit A were prepared as described previously (35). Polyclonal antibodies against the serine repeat antigen protein and H ϩ -pumping pyrophosphatase from mung bean were kindly provided by Dr. Mitamura (Institute of Microbial Diseases, Osaka University, Japan) and Dr. Maeshima (Nagoya University, Japan), respectively.
Immunoblotting-Samples were denatured with SDS-sample buffer containing 1% SDS and 10% ␤-mercaptoethanol and then electrophoresed on a 12% polyacrylamide gel in the presence of SDS (36). Following electrotransfer at 0.3 amperes for 2 h, the nitrocellulose filters were blocked in a buffer consisting of 20 mM Tris-HCl (pH 7.6), 5 mM EDTA, 0.1 M NaCl, and 0.5% bovine serum albumin for 4 h and then probed with 1000-fold diluted antiserum for 2 h. The filters were washed with 20 mM Tris-HCl buffer (pH 7.6) containing 5 mM EDTA, 0.1 M NaCl, and 0.05% Tween 20, treated with peroxidase-labeled anti-rabbit IgG at a dilution of 1:2000 for 30 min, washed further with the same buffer, and then subjected to ECL amplification according to the manufacturer's manual (Amersham Pharmacia Biotech).
Indirect Immunofluorescence Microscopy-Parasitized erythrocytes on polylysine-coated glass coverslips were fixed in PBS containing 4% paraformaldehyde for 30 min, washed three times with PBS, then incubated with PBS containing 0.1% Triton X-100 for 20 min and further with 2% goat serum and 0.5% bovine serum albumin in PBS, and finally reacted with antibodies at 10 g/ml for 1 h. The samples were washed three times with PBS and reacted with the second antibodies conjugated with fluorescein, and then the immunoreactivity (green color) was observed under an Olympus FLUOVIEW confocal laser microscope or Olympus BX60 fluorescence microscope.
Immunoelectromicroscopy-The pre-embedding silver enhancement immunogold method described by Burry et al. (36) was used with a slight modification. The parasitized erythrocytes on polylysine-coated plastic coverslips were fixed in 4% paraformaldehyde plus 0.1% glutaraldehyde dissolved in 0.1 M sodium phosphate buffer (pH 7.4) for 30 min and then washed three times with sodium phosphate buffer. Then the cells were incubated in a blocking buffer containing 0.25% saponin and 5% bovine serum albumin for 30 min and reacted with anti-NSF antiserum (200ϫ dilution) in blocking buffer containing 0.005% saponin, 10% bovine serum albumin, 10% goat serum, and 0.1% cold water fish gelatin at 4°C overnight. Then the cells were washed in sodium phosphate buffer containing 0.005% saponin and incubated with goat antirabbit IgG conjugated with colloidal gold (1.4-nm diameter, Nanogold, Nanoprobes) in blocking buffer for 2 h at room temperature. Cells were washed five times with sodium phosphate buffer containing 0.005% saponin for 10 min, washed with sodium phosphate buffer for 5 min, and fixed with 1% glutaraldehyde for 10 min. After washing, the gold particles were intensified using a silver enhancement kit (HQ silver, Nanoprobes) for 5 min at 20°C in the dark. After washing in distilled water, the cells were post-fixed with 0.5% OsO 4 for 90 min at 4°C, washed with distilled water, dehydrated with a graded series of ethanol, and embedded in epoxy resin. Ultrathin sections were doubly stained with uranyl acetate and lead citrate and observed under a Hitachi H7000 electron microscope.
Other Procedures, Preparations, and Chemicals-ATPase activity was measured as described previously (37). Synaptic vesicles were prepared from rat brains as described previously (38). A Golgi membrane-rich fraction was prepared from CHO cells as described previously (38). BFA and C5-ceramide were purchased from Molecular Probes. Other chemicals used in the study were the highest grade available commercially.

RESULTS
Expression of PfNSF-The complete nucleotide sequence of chromosome 3 of P. falciparum enabled us to identify the gene encoding NSF or its homologue. A gene that is homologous to the mammalian NSF gene consisting of 2352 base pairs without any intron structures (GenBank TM accession number CAB10575) encodes 783 amino acids with a calculated molecular weight of 89,133 (Fig. 1). The overall identities of its gene sequence and amino acid sequence with those of rat NSF are 50.9 and 48.8%, respectively. This gene product is abbreviated as PfNSF in this study. Like other NSFs, PfNSF possesses two conserved ATP-binding motifs (Walker motifs), which are GXXXXGKT at positions 294 -301 and 575-582 (Fig. 1, boxed).
We examined whether or not the PfNSF gene is actually expressed in P. falciparum cells. We prepared a total RNA fraction from parasitized erythrocytes, and then RT-PCR reaction was performed using specific primers for a PfNSF gene as described under "Experimental Procedures." When the reverse transcripts were used, an amplified product with the expected size was obtained (Fig. 2A, lane 3), whereas no amplified product was obtained when samples without reverse transcriptase reaction were used (Fig. 2A, lane 4). The nucleotide and deduced amino acid sequences of the amplified DNA product (333 base pairs corresponding to 55-165 amino acid sequence positions) exactly matched those of the PfNSF gene (Fig. 1). Northern blotting with the amplified product as a probe demonstrated the presence of a single species of mRNA of PfNSF (ϳ3.6 kilobases) (Fig. 2B). These results indicate that the PfNSF gene is actually expressed in P. falciparum cells.
To identify PfNSF at the protein level, we raised polyclonal antibodies specific to the conserved region of NSF (see Fig. 1, dashed line). The anti-NSF antibodies recognized single 83and 89-kDa polypeptides in rat brain synaptic vesicles and parasitized erythrocytes, respectively (Fig. 3A, lanes 1 and 2). The apparent molecular masses of these polypeptides are those expected from their cDNA sequences (Fig. 1). No immunological cross-reactivity was detected when noninfected control erythrocytes were used (data not shown). The presence of the antigenic peptide, which was used for preparation of anti-NSF antibodies, during immunodecoration, prevented the immuno-logical cross-reactivities (Fig. 3A, lanes 3 and 4). These results strongly suggest that PfNSF was present in the parasitized erythrocytes.
Isolated parasites (about 10 8 cells) were homogenized vigorously and then centrifuged again at 105,000 ϫ g for 1 h. More than 90% of the anti-NSF immunoreactivity was recovered in the pellet, suggesting that most PfNSF is present as a membrane-bound form.
Properties of PfNSF-Mammalian NSFs have been shown to be N-ethylmaleimide-sensitive ATPases; upon the addition of MgATP, the enzymes may hydrolyze ATP, thereby forming ADP and inorganic phosphate, although the rate of hydrolysis is quite slow (39). We isolated PfNSF by solubilization with polyoxyethylene lauryether followed by immunoprecipitation with anti-NSF antibodies (Fig. 3B). The isolated PfNSF showed MgATP hydrolytic activity (0.11 mol of P i liberated/h/mg of protein), which was inhibited completely by N-ethylmaleimide at 1 mM. The N-ethylmaleimide-sensitive ATPase activity was weak but comparable with those in mammalian NSFs (39). These results suggested that PfNSF is a N-ethylmaleimidesensitive ATPase as in the case of mammalian NSFs.
NSF in the CHO Golgi fraction is known to be released from the membrane upon treatment with MgATP (26,27), although NSF in neuronal synaptic vesicles or endocrine synaptic-like microvesicles does not have such an effect (38,40). We examined whether or not PfNSF is released from the membrane upon the addition of ATP. As shown in Fig. 4, neither PfNSF nor NSF in synaptic vesicles was released from parasite membranes upon treatment with MgATP, whereas the same treatment released NSF from CHO Golgi membranes. These results indicated that PfNSF shares properties with NSFs of synaptic vesicles and synaptic-like microvesicles.
Presence of Extracellular PfNSF-During isolation of malaria parasites from parasitized erythrocytes with saponin (see "Experimental Procedures"), we noticed that an appreciable level of anti-NSF immunoreactivity remained in the supernatant after isolation of the parasite cells (Fig. 5A). This fraction contained the erythrocyte cytoplasm, erythrocyte plasma membranes, and extraparasitized membrane structures including tubovesicular membrane networks. This fraction was then centrifuged at 105,000 ϫ g for 1 h, a pellet (the extracellular particulate fraction) and a supernatant being obtained. Western blotting experiments indicated that PfNSF was present in the extracellular particulate fraction as well as in the parasites but not in the supernatant (Fig. 5A). Vacuolar H ϩ -ATPase and H ϩ -pumping pyrophosphatase, which are known to be present in the parasite (35,41), were detected in parasite cells but not detected in either the extracellular particulate fraction or the supernatant (Fig. 5A). These results strongly suggest the presence of extraparasitized membrane-bound PfNSF.
Indirect immunofluorescence microscopy with anti-NSF antibodies further demonstrated the presence of extraparasitized PfNSF in erythrocytes. The anti-NSF antibodies immunostained the parasite cells in parasitized erythrocytes (Fig. 5B), whereas no immunoreactivity was found in noninfected erythrocytes (Fig. 5C). Significantly, the PfNSF immunoreactivity  lanes 1 and 2) and PfNSF (lanes 3 and 4) was examined by RT-PCR, and the amplified products were visualized on an acrylamide gel stained with ethidium bromide. The amplified products were not detected when samples without reverse transcriptase reaction were used in lanes 2 and 4, respectively. The position of the PfNSF transcript is indicated by an arrow. The positions of molecular markers are also indicated (from top to bottom, 1412, 517, 396, and 221 base pairs (bp)). B, Northern blotting. Total RNA (25 g) obtained from asynchoronous parasites was hybridized to the amplified product for PfNSF shown above, washed, and visualized using BAS 1000 film after overnight exposure.

FIG. 3. Immunological detection of PfNSF.
A, rat brain synaptic vesicles (50 g protein) (lanes 1 and 3) and malarial cells isolated by saponin treatment (10 l) (lanes 2 and 4) were dissolved in SDS-sample buffer. After electrophoresis, the proteins were transferred to a nitrocellulose sheet followed by decoration with anti-NSF antibodies (1000 diluted antiserum), and the immunoreactivity was visualized with ECL. For lanes 3 and 4, the nitrocellulose sheet was incubated with 2 mg of antigenic peptide during the antibody treatment. B, isolated parasites (about 10 8 cells) were solubilized with 1% polyoxyethylene lauryether and then centrifuged at 105,000 ϫ g for 1 h. The supernatant was carefully removed, anti-NSF antiserum (10 l) or control serum (10 l) was added, and then successively 30 l of protein A-labeled Sephadex (Bio-Rad) was added. The solution was then gently shaken at 4°C overnight. Then protein A-labeled particles were washed several times with PBS, dissolved in the sample buffer as described above, and finally electrophoresed. Immunoblotting with anti-NSF antibodies was performed as described above. was present within the vesicular structures outside the parasite cells (Fig. 5B). Consistent with the distribution observed on the immunoblotting shown in Fig. 5A, no such extraparasitized vesicular structures were observed in the immunoreactivities against antibodies for vacuolar H ϩ -ATPase (Fig. 5D), H ϩpumping pyrophosphatase (Fig. 5E), serine repeat antigen pro-tein, markers for the peripheral space between the parasitophorus vacuolar membranes, and the plasma membrane of the malaria parasite (Fig. 5F). Vital staining with C5-ceramide revealed tubovesicular membrane networks (6) (Fig. 5G). From these results, we concluded that PfNSF is present in both parasite cells and vesicular structures outside of parasite cells.
During development, a similar degree of anti-PfNSF immunoreactivity was observed in all stages of intraerythrocyte parasites, indicating that PfNSF is expressed in all cell stages. At the early trophozoite stage, PfNSF appeared in the extraparasite space and seemed to be associated with several apparent vesicular structures outside the parasites in the erythrocyte cytoplasm. At the trophozoite stage, the PfNSF-positive vesicular structure seems to be more discrete and distributed throughout erythrocyte cytoplasm, which becomes weak at the schizont stage and disappears at the merozoite stage (Fig. 6).
These results indicate that export of PfNSF occurs at the early trophozoite stage.
Export of PfNSF Is BFA-sensitive-To obtain information on the mechanism by which PfNSF is transported outside the parasites, we next examined the effect of BFA (Fig. 7). It was found that BFA effectively blocked the export of PfNSF from malarial parasites; the concentration required for 50% inhibition was 9 M. Almost all immunoreactivity against anti-NSF antibodies outside of parasite cells disappeared upon treatment with 50 M BFA for 2 h. The effect of BFA was reversible, because the immunoreactivity outside the parasite cells appeared again when the erythrocytes were washed several times and resuspended in culture medium. Under similar assay conditions, BFA did not affect the distribution of either serine repeat antigen protein or C5-ceramide (data not shown). These results indicated that the export of PfNSF is sensitive to BFA.
Prolonged exposure to 50 M BFA for 24 h inhibited maturation of the parasite; BFA-treated cells did not progress to the trophozoite stage, whereas control cells matured normally. This arrest was also reversible. These results were consistent with previous observations (12,22). Subcellular Localization of PfNSF-Finally, the localization of PfNSF in parasitized erythrocytes at the subcellular level was investigated by immunoelectromiscroscopy. Consistent with the immunohistochemistry described in Fig. 5A, immunogold particles for PfNSF were selectively and intensely labeled in infected P. falciparum cells, whereas few immunogold particles were observed in erythrocyte cytoplasm (Fig. 8A). The immunogold particles seem to be associated with intracellular organelles such as food vacuoles (Fig. 8A) and plastid-like organelle (Fig. 8B), supporting the membrane-bound nature of PfNSF (Fig. 4). As shown in Fig. 8, C and D, immunogold particles were also associated with vesicular structures with a diameter of 30 -70 nm and electron translucent contents in erythrocytes. The immunogold particles associated with vesicles outside the parasite cells disappeared when the cells were treated with BFA, although the BFA treatment did not change the number or morphology of the vesicles (data not shown). Immunogold particles were also associated with larger membrane structure in erythrocyte cytoplasm, which may correspond to part of the tubovesicular membrane network (Fig. 8A,  arrowhead). DISCUSSION NSF and its homologue are key proteins that comprise a supramolecular complex with SNAP and its receptors (which contain synaptotagmin, vesicular-associated membrane protein (VAMP), and syntaxin), catalyze the docking and fusion of vesicles, and facilitate protein transport during the biogenesis of organelles in eukaryotes (26 -29). Because the gene encoding NSF or its homologue was identified in P. falciparum cells, one can expect that NSF or its homologues is expressed and functions in P. falciparum cells. The identification and characterization of the PfNSF protein may provide a clue as to the vesicular transport systems in the malaria parasites and the mechanism underlying protein transport to the erythrocyte membrane. The present study was, therefore, undertaken to obtain the direct evidence of PfNSF in P. falciparum cells.
We detected the mRNA of PfNSF by RT-PCR and Northern blotting (Fig. 2), and identified PfNSF by Western blotting with site-directed polyclonal antibodies against a conserved region of NSF (Fig. 3). The antibodies immunostained the whole body of malaria parasites infecting erythrocytes (Fig. 5). Furthermore, the immunoprecipitated polypeptide showed a weak Nethylmaleimide-sensitive ATPase activity. Taken together, these results constitute evidence for the functional occurrence of the PfNSF protein in the malaria parasite.
The presence of PfNSF suggests that a vesicular transport mechanism is operating in the malaria parasite. PfNSF is associated with organelles such as food vacuoles, suggesting that PfNSF plays a role in the biogenesis of organelles through vesicular transport. It is noteworthy that PfNSF is also associated with plastid-like organelles (Fig. 8B). The plastid of P. falciparum is an evolutionary homologue of the plant chloroplast. Very recently, signal and plant-like transit peptides were found to be involved in the protein trafficking to plastids in P. falciparum (42). The association of PfNSF with plastid-like organelles suggests that PfNSF plays some role in protein trafficking to plastids. Consistently, a NSF homologue was shown to be important for vesicle fusion and/or membrane protein translocation in plastids of the higher plant Capsicum annuum (43). Another important finding of the present study is that PfNSF is exported from parasite cells and localized in vesicular structures in the erythrocyte cytoplasm. This suggests that PfNSF plays some role in protein transport from the parasite to the erythrocyte plasma membrane. Immunoelectromicroscopy clearly revealed that extraparasitized PfNSF is associated with vesicles with diameters of 30 -70 nm. In this respect, it is noteworthy that PfEMP1 and PfEMP3, which are targeted to the plasma membrane of infected erythrocytes, are also known to be associated with vesicles (11). PfNSF containing vesicles and PfEMP1-and PfEMP3-containing vesicles have electron translucent contents and are morphologically similar to each other. It is possible that PfNSF is involved in the targeting of PfEMP1 and pfEMP3 into erythrocyte plasma membrane.
Moreover, very recently, it was reported that P. falciparum cells have the ability to transport erythrocyte membrane proteins to internal organelles of the parasite cells and that cholesterol and sphingomyelin are important for this process (44). Because the vacuolar uptake of host components seems to correspond to the endocytotic process in other eukaryotes, it appears that a vesicular transport mechanism is involved in the endocytotic process as well as the targeting of the parasite membrane to the erythrocyte membrane.
PfNSF is the first example of the presence of the SNAP receptor complex in the malaria parasite. Since other constituents of the SNAP receptor complex have not yet been detected in the malaria parasite, the identification and characterization of such proteins will be important in revealing all of the features of the putative vesicular transport mechanism in the malaria parasite. Phylogenetically, the vesicular machinery participating in membrane biogenesis, such as docking and fusion of vesicles, is broadly conserved across the species barrier in higher eukaryotes. Consistent with this idea, Toxoplasma uses trafficking mechanisms, that is the NSF/SNAPs/ SNAP receptor/Rabs, suggesting a role in exocytotic and endocytotic pathways (45,46). It is possible that a vesicular transport mechanism operates in pathogenic protozoa in general.
In conclusion, we obtained evidence of NSF or its homologue in P. falciparum cells. Like the Sar1 protein, parts of PfNSF are associated with vesicles in the erythrocyte cytoplasm. It is possible that these vesicles are involved in protein targeting to the erythrocyte plasma membrane and that the SNAP receptor complex is involved in this transport process. P. falciparum cells may constitute a unique experimental system for studies on vesicular trafficking.