Toxoplasma gondii Homologue of Plasmodium Apical Membrane ...

INFECTION AND IMMUNITY, Dec. 2000, p. 7078–7086 0019-9567/00/$04.00⫹0 Copyright © 2000, American Society for Microbiology. All Rights Reserved.

Vol. 68, No. 12

Toxoplasma gondii Homologue of Plasmodium Apical Membrane Antigen 1 Is Involved in Invasion of Host Cells ADRIAN B. HEHL,1,2 CHRISTINE LEKUTIS,1 MICHAEL E. GRIGG,1 PETER J. BRADLEY,1 JEAN-FRANC ¸ OIS DUBREMETZ,3 EDUARDO ORTEGA-BARRIA,1,4 1 AND JOHN C. BOOTHROYD * Department of Microbiology and Immunology, Stanford University School of Medicine, Stanford, California 94305-51241; Institute of Parasitology, University of Zu ¨rich, CH-8057 Zu ¨rich, Switzerland2; EP CNRS 525, Institut de Biologie de Lille, 59021 Lille Cedex, France3; and Department of Parasitology, Gorgas Memorial Health Research and Information Center, Panama 5, Panama4 Received 1 August 2000/Returned for modification 21 August 2000/Accepted 9 September 2000

Proteins with constitutive or transient localization on the surface of Apicomplexa parasites are of particular interest for their potential role in the invasion of host cells. We describe the identification and characterization of TgAMA1, the Toxoplasma gondii homolog of the Plasmodium apical membrane antigen 1 (AMA1), which has been shown to elicit a protective immune response against merozoites dependent on the correct pairing of its numerous disulfide bonds. TgAMA1 shows between 19% (Plasmodium berghei) and 26% (Plasmodium yoelii) overall identity to the different Plasmodium AMA1 homologs and has a conserved arrangement of 16 cysteine residues and a putative transmembrane domain, indicating a similar architecture. The single-copy TgAMA1 gene is interrupted by seven introns and is transcribed into an mRNA of ⬃3.3 kb. The TgAMA1 protein is produced during intracellular tachyzoite replication and initially localizes to the micronemes, as determined by immunofluorescence assay and immunoelectron microscopy. Upon release of mature tachyzoites, TgAMA1 is found distributed predominantly on the apical end of the parasite surface. A ⬃54-kDa cleavage product of the large ectodomain is continuously released into the medium by extracellular parasites. Mouse antiserum against recombinant TgAMA1 blocked invasion of new host cells by approximately 40%. This and our inability to produce a viable TgAMA1 knock-out mutant indicate that this phylogenetically conserved protein fulfills a key function in the invasion of host cells by extracellular T. gondii tachyzoites. Toxoplasma gondii is an obligate, intracellular parasite of warm-blooded animals. In humans, it is best known as a pathogen in the developing fetus and in immunocompromised (e.g., AIDS) patients. It is related to other members of the phylum Apicomplexa, such as Plasmodium (the cause of malaria) and Eimeria (the cause of coccidiosis). Over the last few years, T. gondii has been actively developed as a model organism with which to study the biology of apicomplexan parasites (6). As part of this effort, a large database of expressed sequence tags (ESTs) has been generated, which led to the tentative identification of numerous Toxoplasma genes based on homology to coding regions from other organisms (1, 19, 25). Parallel efforts to sequence the genomes of Plasmodium (31) and Cryptosporidium (24, 38, 40) are also contributing to the number of apicomplexan sequences deposited in the databases. Of particular interest is a class of homologs representing genes that are unique and conserved among apicomplexan parasites (1). This group of phylogenetically restricted sequences code for proteins closely linked to the particular biology common to apicomplexans, which opens avenues for functional studies in T. gondii which would be difficult to do in a less tractable system. Invasion of host cells by the asexual stages of apicomplexan parasites is a complex, receptor-mediated event, which is still not well understood. It involves structures of the apical complex (15), specialized surface antigens (17, 32), and products released by secretory organelles (i.e., micronemes, rhoptries,

and dense granules) (7, 14, 30). While some members of this phylum, such as T. gondii, are extremely promiscuous with respect to the cell types they are able to infect, others, such as the asexual stages of Plasmodium and Eimeria, are able to selectively invade only certain specialized cells or tissues. Despite these differences in host and tissue specificity, there appears to be a significant conservation of the invasion apparatus, on the level of both ultrastructure and proteins associated with apical organelles (15, 16, 39, 41, 42). These common elements are of particular interest biologically, as they constitute the phylogenetically conserved, basic machinery for host cell invasion essential for the survival of these obligate intracellular parasites. Presumably, additional features and/or adaptations of this basic invasion apparatus have provided each species with the ability to infect its respective host(s) or host cell(s) with various degrees of specificity. T. gondii molecules involved in the interaction with host cell receptors have been elusive, mainly because of the wide range of cells that this parasite is able to invade. Some of the relevant molecules are most likely concentrated on or at the apical surface membrane at the time of invasion and control temporally discrete events. These events include initial attachment, reorientation, triggered secretion of vesicle contents, building and translocation of the moving junction, and finally the establishment of a functional parasitophorous vacuole (5, 7, 13). Some putative players in these events have been identified by antibodies raised to proteins of apical structures (27); others have been discovered as part of the excreted-secreted fraction stored in organelles and released in a controlled fashion upon contact with the host cell (4, 37). One of the Plasmodium proteins implicated in invasion is the apical membrane antigen (AMA1) expressed by merozoites.

* Corresponding author. Mailing address: Department of Microbiology and Immunology, Stanford University School of Medicine, Stanford, CA 94305-5124. Phone: (650) 723-7984. Fax: (650) 723-6853. E-mail: [email protected]. 7078

T. GONDII HOMOLOGUE OF PLASMODIUM AMA1

VOL. 68, 2000

This is a type Ia transmembrane protein with a conserved core structure determined by 16 cysteine residues in the mature extracellular domain (20). The extracellular portion can be subdivided into three structural domains (I to III), containing six, four, and six cysteines, respectively, which form intradomain disulfide bridges (20). The Plasmodium knowlesi and Plasmodium falciparum AMA1 (PkAMA1 and PfAMA1, respectively), for example, are synthesized as proproteins of 66 and 80 kDa, respectively, in mature trophozoites and segmenting schizonts and become concentrated at the apical end of the parasites. Upon merozoite release, AMA1 is proteolytically processed and secreted as a membrane-bound protein onto the surface of free merozoites, where it distributes over the entire parasite (9, 45). While the biological function of AMA1 is still unclear, the importance of this minor antigen in the invasion of red blood cells (RBCs) by free merozoites has been shown in several studies (8, 10, 43). Monoclonal antibodies (MAbs) raised against native PkAMA1 were able to prevent invasion of rhesus RBCs in vitro (43), and immunization of mice with recombinant and refolded AMA1 or passive transfer of specific AMA1 polyclonal antibodies into Plasmodium chabaudi-infected mice prevented lethal parasitemias (2). These data point to an important role of Plasmodium AMA1 in the invasion of erythrocytes. Here, we show that T. gondii has an AMA1 homologue that is also implicated in the invasion process but localizes to the micronemes rather than the rhoptry necks, as seen with Plasmodium AMA1. MATERIALS AND METHODS Parasites and cultivation. Tachyzoites of the representative T. gondii strains RH (36), ME49 (21), and CEP (34) were grown in monolayers of human foreskin fibroblasts (HFF) in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% Nu-serum, 2 mM glutamine and gentamicin (20 ␮g/ml) at 37°C in a humid 5% CO2 atmosphere. Library searches and nucleic acid techniques. Assembly of sequences and homology searches of the Toxoplasma EST databases have been described (1, 19, 25). Complete descriptions of the databases can be found at the Genome Web sites (http://www.ebi.ac.uk/parasites/toxo/toxopage.html and http://cbil.humgen .upenn.edu/toxodb/toxodb.html). TgAMA1 ESTs were identified by searching the GenBank database with Toxoplasma EST sequences using BLASTX at NCBI (http://www.ncbi.nlm.nih.gov) and assembled into a single nucleotide sequence contig with the assembly program of the Wisconsin Package version 9.0 (Genetics Computer Group, Madison, Wis.). The ␭ZapII insert TgESTzy99b08.r1, containing a full-length cDNA of TgAMA1 from strain ME49, was obtained from the Washington University–Merck Toxoplasma EST repository via David Sibley (St. Louis, Mo.). The insert was excised for further sequencing as pBluescript SKII using the Stratagene ExAssist helper phage according to the manufacturer’s protocol. Full-length cDNAs from three T. gondii strains (RH, ME49, and CEP) were obtained by reverse transcription-PCR amplification of total RNA prepared from tachyzoites with Ultraspec RNA (Biotecx Laboratories Inc., Houston, Tex.). Reverse transcription was performed with standard protocols using an oligo(dT) anchor primer (dTAP) and Superscript II (Gibco-Life Sciences). PCR amplification of coding regions was done with nested primers (CSEQ-S and CSEQ-AS) and the adapter primer (AP) using Taq polymerase (Sigma) and standard cycling conditions. Products were cloned into the pCR2.1-Topo vector (Invitrogen). Sequencing was performed on an ABI PRISM at the Stanford University sequencing facility using custom oligonucleotide primers. Sequence alignments were done using CLUSTALW with the blosum weight matrix and default settings via the world-wide web (http://ferrari.ibcp.fr/cgi-bin/Mail _clustalw.pl). Southern blot. Genomic DNA was obtained from RH strain tachyzoites lysed in a Tris-EDTA-LiCl-Triton buffer (26). The DNA was digested overnight with NheI, KpnI, HindIII, EcoRI, or BglII, separated through a 0.9% agarose gel, and transferred to a Nytran membrane through capillary action. The membrane was probed with a 600-bp AMA1 promoter fragment labeled by random priming with [␣-32P]dGTP. After autoradiography, the membrane was stripped and reprobed with a 1,600-bp AMA1 C-terminal open reading frame (ORF) fragment, also labeled by random priming with [␣-32P]dGTP. Northern blot. Total RNA was prepared from tachyzoites with Ultraspec as directed by the manufacturer (Biotecx Laboratories). Poly(A)-enriched RNA was derived from the total RNA preparation using the Oligotex mRNA kit (Qiagen, Valencia, Calif.). The RNA was then size fractionated in a 1.0% agarose gel and transferred through capillary action to a Zeta-Probe membrane

7079

(Bio-Rad Laboratories, Hercules, Calif.). The membrane was probed sequentially with the 5⬘ and 3⬘ AMA1 fragments described above. Oligonucleotide primers. The oligonucleotides used included LF-S (AGTGA ATTCGTCGACCTTGGACAAGACA), LF-AS (TGAGGATCCTTAGTCGG CCGTGCACTGAAGT), SF-S (ATCGAATTCTGCGCCGAGTTTGCCTTTA AGA), SF-AS (TGAGGATCCTTAACTCCCCGCTGCTGTATACGA), CF-S (CGTGAATTCGCGAAGAGGTTGGACAGA), CF-AS (CGTGGATCCTTA GTAATCCCCCTCGACCATAACA), dTAP (CCGGAATTCGGTACCTCTA GAT18VN), AP (CCGGAATTCGGTACCTCTAGA), CSEQ-S (ATGGGGCT CGTGGGCGTA), CSEQ-AS (GTAGTAATCCCCATCGACCA), AMAKOS (CGAAGCTTGGGACTCAGCTCAAGCACA), AMAKOAS (GCGGCCGCT ACGGAATCGCTGTTCT), AMAKOUS1 (GGGCGAGGTCAGCAGATGT), AMAKOUAS (GCGGACAGGCGTAGTAACT), 3DHFRS (GCCATTCATG CCAGTCAGT), KOUS3 (TCCGGCCAAATACATTAAATC), 5DHFRA (GA ACAGCAGCAAGATCGGAT), KODAS (ACATAATGTCAACAGCGTAA G), and AMAG (GCCCCATGTGCTTCGTCTCA). Generation of fusion proteins. Three constructs, long (LF), short (SF), and cytosolic (CF), for production of fusion proteins were generated in the pMal-P2 system (New England Biolabs). The fragments were amplified from TgAMA1 cDNA from strain RH using the primer pairs LF-S and LF-AS, SF-S and SF-AS, and CF-S and CF-AS. The primers were designed with EcoRI (sense) and BamHI (antisense) restriction sites for cloning into the respective sites of the expression vector, downstream of the maltose-binding protein (MBP) gene, giving rise to fusion genes MBP-LF, MBP-SF, and MBP-CF. MBP-LF contained a region corresponding to domains I and II of the ectodomain from Val-69 to Asp-410. MBP-SF covered domain I, starting at the second cysteine (Cys-166) and extending through domain II to Asp-379. MBP-CF contained the COOHterminal 62-amino-acid stretch corresponding to the presumed cytoplasmic domain from Ala-479 to Tyr-541. Bacterial overexpression was induced using 0.5 mM IPTG (isopropyl-␤-D-thiogalactopyranoside) for 2 h at 37°C, and fusion proteins were purified from bacterial cold shock lysates on amylose resin according to standard protocols (35) and lyophilized as previously described (18). Peptide synthesis. Two peptides corresponding to Ser-21 through Ser-36, starting at the predicted cleavage site of the signal sequence (N-pep: NH-S-GL-S-S-S-T-R-S-R-E-S-Q-T-L-S-C-COOH) and from the cytoplasmic tail region Gln-489 through lysine 505 (C-pep: NH-C-Y-Q-A-A-H-H-E-H-E-F-Q-S-D-RG-A-R-K-COOH) of TgAMA1 were synthesized and coupled to keyhole limpet hemocyanin (KLH). Generation of polyclonal antibodies. BALB/c mice were immunized intraperitoneally on days 0, 15, and 30 with approximately 50 ␮g of fusion protein or KLH-coupled peptide resuspended in 100 ␮l of phosphate-buffered saline (PBS) and emulsified with an equal volume of RIBI adjuvant (RIBI Immunochem Research Inc., Hamilton, Mont.). Blood was collected prior to initial immunization and after each boost from the tail vein, and the serum fraction was assayed for specific antibody content. MAb production. Splenocytes were harvested from an adult BALB/c mouse 3 days after boosting with the AMA1 cytoplasmic tail peptide conjugated to KLH. The splenocytes were fused to P3x63Ag8.653 myeloma cells at a 5:1 ratio, and hybridomas were selected in DMEM containing hypoxanthine-aminopterin-thymidine. Tissue culture supernatants were screened for AMA1 reactivity by Western blot using strips derived from a sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) gel of reduced parasite lysate. From this screen, one AMA1 cytoplasmic tail-specific hybridoma was obtained, CL22 (immunoglobulin G2A [IgG2A]). Inhibition of invasion. Briefly, tachyzoites of the RH strain were pretreated for 30 min at 37°C with heat-inactivated antiserum obtained from mice immunized with MBP-AMA1 fusion proteins or with normal mouse serum. Pretreated tachyzoites were added to HFF monolayers in 24-well plates in duplicate. After a 1-h incubation period at 37°C, unbound tachyzoites were washed off the monolayers. The number of cell-associated tachyzoites was assessed microscopically after fixation or by tritiated uracil incorporation as previously described (33). Protein analysis. Total lysates were prepared from extracellular tachyzoites in SDS-PAGE sample buffer under reducing conditions. Analytical SDS-PAGE and transfer of proteins to nitrocellulose membranes were performed according to standard protocols (3). Filters were blocked in 5% dry milk–0.5% Tween 20–PBS and incubated with antisera or MAbs diluted in blocking solution. Bound antibodies were detected with horseradish peroxidase-conjugated goat anti-mouse IgG and developed using enhanced chemiluminescence (ECL; Amersham). Secretion assays. Culture supernatants containing secreted protein were prepared by washing freshly lysed tachyzoites twice in RPMI without serum and incubating the parasites at a density of 108/ml in this solution for 5 min to 1 h at 37°C. Cells were cooled on ice, pelleted by centrifugation at 1,000 ⫻ g, and lysed in 200 ␮l of reducing SDS-PAGE sample buffer. The supernatant was recentrifuged under the same conditions and again at 14,000 ⫻ g for 30 min (4°C). The cleared supernatant was lyophilized in 200-␮l aliquots and dissolved in 100 ␮l of reducing SDS-PAGE sample buffer. Approximately 10 ␮l of each fraction was separated on SDS-PAGE and blotted as described above. Subcellular fractionation. Fractionation of T. gondii (RH strain) was carried out essentially as described (22). Briefly, parasites were lysed in a Stansted cell disruptor in SMDI (250 mM sucrose, 20 mM MOPS [morpholinepropanesulfo-

7080

HEHL ET AL.

INFECT. IMMUN.

nic acid, pH 7.2], 2 mM dithiothreitol [DTT], 5 ␮g of leupeptin per ml, and 1 mM phenylmethylsulfonyl fluoride [PMSF]). Unbroken cells and debris were pelleted at 500 ⫻ g for 15 min in a clinical centrifuge. The supernatant was spun for 25 min at 25,000 ⫻ g to obtain a high-speed pellet containing organelles, which were subsequently separated on a 30% Percoll gradient. Three fractions were collected: a top band previously described as containing tachyzoite “ghosts,” a middle fraction enriched in micronemes, and a bottom band containing rhoptries and dense granules. The three fractions were centrifuged at 100,000 ⫻ g for 90 min, and the organelles were collected from the top of the sedimented Percoll, resuspended in an equal volume of SMDI, and separated by SDS-PAGE. Proteins were transferred to nitrocellulose by a wet electrophoretic technique. Western blots were probed with MAbs using either a 1:50 dilution of tissue culture supernatant or a 1:1,000 to 1:2,000 dilution of ascites fluid in PBS. The T. gondii-specific MAbs used included MAbs against AMA1 (CL22), MIC2 (6D10) (44), ROP2, ROP3, and ROP4 (T34A7) (22), and GRA3 (2H11) (23). Immunofluorescence assay. Live tachyzoites were prepared for indirect surface immunofluorescence as described previously (18). Cell integrity was monitored by labeling with anti-ROP1 and anti-MIC2 antibodies. For analysis of intracellular parasites, infected fibroblasts grown on glass coverslips were washed in PBS, fixed in cold 2% formaldehyde, and permeabilized with 0.5% Triton X-100 in PBS (PBS-T) for 20 min. Blocking and incubations with diluted antisera or secondary antibodies were done according to standard methods in 2% bovine serum albumin in PBS-T, and coverslips were washed in PBS between incubations with antibody. Labeled cells were embedded with Vectashield (Vector Laboratories) solution for microscopy. Immunoelectron microscopy. Immunoelectron microscopy was performed on ultrathin cryosections. A tachyzoite-infected Vero cell monolayer was fixed with 4% paraformaldehyde–0.05% glutaraldehyde in 0.2 M sodium phosphate buffer for 90 min, then washed in PBS containing 10% fetal calf serum (PBS-FCS), and infused in 2.3 M sucrose containing 10% polyvinylpyrrolidone prior to freezing in liquid nitrogen. Sections were obtained on an FCS cryoattachment-equipped Leica Ultracut operating at ⫺100°C. Sections were floated successively on PBSFCS, mouse MBP-SF antiserum diluted 1:20 in PBS-FCS, anti-mouse IgG rabbit serum diluted 1:400 in PBS-FCS, 8-nm protein A-gold diluted in PBS to an optical density at 525 nm (OD525) of 0.05, with five 3-min washings in PBS between each step. Sections were then embedded in methylcellulose (2%)– uranyl acetate (0.4%) and observed with a Philips EM 420 electron microscope. AMA1 knockout. An AMA1 knockout vector was constructed by insertion of AMA1 flanking sequences on either side of a hypoxanthine-xanthine-guanine phosphoribosyltransferase (HXGPRT) expression cassette in pBluescript (12). Briefly, a 1-kb upstream AMA1 fragment and a 4.5-kb downstream AMA1 fragment were obtained by PCR from a genomic DNA template and cloned into polylinker sites adjacent to the dihydrofolate reductase (DHFR) 5⬘ untranslated region (UTR) and 3⬘ UTR so that the AMA1 and HXGPRT sequences sat in the reverse orientation. Specifically, the 4.5-kb AMA1 downstream fragment was amplified by PCR with the AMAKOS and AMAKOAS primers, whereas the 1.0-kb AMA1 upstream fragment was amplified by PCR with the AMAKOUS1 and AMAKOUAS primers. Both PCR fragments were shuttled into pCR2.1Topo (Invitrogen) prior to insertion adjacent to the HXGPRT expression cassette. Thus, the HXGPRT expression cassette replaces (in opposite orientation) the AMA1 promoter, start codon, and signal peptide. Approximately 50 ␮g of the AMA1 knock-out plasmid was linearized with NotI prior to electroporation into RH strain tachyzoites lacking HXGPRT. After 24 h, transformants were selected with 100 ␮g of mycophenolic acid and 50 ␮g of xanthine per ml of RPMI containing 10% FCS. On day 18 posttransfection, tachyzoites were cloned from the drug-resistant population. Genomic DNA was isolated from 80 distinct drug-resistant clones as well as the drug-resistant population to determine whether targeted disruption of AMA1 had occurred. Homologous recombination upstream of AMA1 was analyzed by PCR using a DHFR 3⬘ UTR sense primer (3DHFRS) and an AMA1 primer (KOUS3), which anneals to a region beyond that included in the knockout vector. Similarly, homologous recombination downstream of AMA1 was analyzed by PCR using a DHFR 5⬘ UTR antisense primer (5DHFRA) and an AMA1 primer which anneals to a region beyond that included in the knockout vector (KODAS). Interruption of the AMA1 ORF was monitored by PCR using AMA1-specific primers KOUS3 and AMAG. The latter primer anneals to a segment of the AMA1 ORF not present in the knockout vector. Southern blot analyses were also done on a subset of the drug-resistant clones as described above, in order to confirm the results obtained by PCR.

RESULTS AMA1 is a conserved protein in Plasmodium and Toxoplasma. BLASTX analysis of the Toxoplasma EST database revealed a clear homologue of Plasmodium AMA1 represented by three ESTs. The complete nucleotide sequence of this Toxoplasma homologue (TgAMA1) was determined from clone TgEST zy99b08.r1 containing an insert of 2,507 bp (GenBank accession number AF010264). The source of the cDNA was a library made from mRNA of strain ME49 tachyzoites

FIG. 1. (A) To-scale graphic depiction of the TgAMA1 gene locus, depicting exon (black boxes) and intron segments. Black line, noncoding genomic sequence. (B) Graphic representation of the predicted full-length TgAMA1 protein (GenBank accession number AF010264). Black boxes, hydrophobic sequences. Cysteines are represented by vertical lines. Domains I, II, and III are defined by analogy to those described by Hodder et al. (20) for the P. knowlesi AMA1 ectodomain. (C) Southern blot analysis. Genomic DNA from RH tachyzoites was isolated and digested overnight with NheI (N), KpnI (K), HindIII (H), EcoRI (R), or BglII (B). The DNA was resolved by agarose gel electrophoresis, and blots were probed sequentially with a 600-bp AMA1 5⬘ probe and a 1,200-bp AMA1 3⬘ probe. The migration of size markers is indicated (in kilobases).

(1). A single ORF from nucleotide (nt) 52 to nt 1677 codes for a 541-amino-acid (aa) protein with a predicted molecular mass of ⬃60 kDa and a theoretical pI of 5.47. Upstream and downstream AMA1 gene fragments were obtained by inverse PCR using AMA1-specific primers and an EcoRV-digested, circularized genomic DNA template. The complete sequence of the PCR-amplified AMA1 gene was acquired by primer walking. This revealed seven introns within the AMA1 coding region, ranging from 240 to 702 bp in length. The genome organization is shown schematically in Fig. 1A. A hydrophobicity plot shows an N-terminal hydrophobic region, which is identified by the program PSORT as a signal sequence with a predicted cleavage site between Ala-20 and Ser-21. After cleavage of the signal sequence, the calculated molecular mass of the mature protein is 57.9 kDa. A putative hydrophobic membrane-spanning region was identified between Ala-456 and Leu-472. The overall arrangement of these elements in TgAMA1 as well as in the Plasmodium homologues is indicative of a type Ia membrane protein (Fig. 1B), with the COOH-terminal 69 aa (Glu473 to Tyr-541) constituting a presumptive cytoplasmic tail. Southern blot analyses were done to determine how many copies of AMA1 exist within the Toxoplasma genome. As shown in Fig. 1C, AMA1 is a single-copy gene, since probes derived from the 5⬘ or 3⬘ end of AMA1 detected only a single fragment on sequentially probed blots. Likewise, a single 3.3-kb transcript was detected on Northern blots of tachyzoite poly(A)-enriched RNA (data not shown). The EST sequence predicts poly(A) addition ⬃900 nt downstream of the stop codon. This places the predicted 5⬘ end of the transcript about 700 nt upstream of the apparent start codon. Neither of these sites, however, has been experimentally confirmed. The ectodomain of mature Plasmodium AMA1 proteins contains 16 cysteine residues which form a secondary structure of three domains stabilized by disulfide bonds, with six, four,

FIG. 2. Multiple alignment of AMA1 amino acid sequences from P. berghei, P. yoelii, P. chabaudi, P. knowlesi, P. vivax, and P. falciparum, and TgAMA1. Identical residues in four or more of the seven sequences are boxed. Gaps inserted to optimize alignments are represented by dashes.

7081

7082

HEHL ET AL.

and six cysteines in domains I, II, and III, respectively (20). The correct folding of the ectodomain conformation is apparently important for proper function of Plasmodium AMA1, as only antibodies generated against epitopes in the native conformation are fully protective (2, 10, 43). A similar arrangement of 16 cysteines is preserved in the TgAMA1 ectodomain. The 10 cysteines corresponding to domains I and II can be easily aligned with those of the malaria AMA1 sequences. In contrast, the spacing of the six cysteines in domain III, which are predicted to form a “knot”-like structure in the P. chabaudi AMA1 (20), is not sufficiently conserved to allow sequence alignment without introduction of several gaps (Fig. 1B and 2). Specifically, the CPC and CXC motifs found in domain III of Plasmodium AMA1 are not present in TgAMA1. Multiple sequence alignment of the predicted AMA1 amino acid sequences from six Plasmodium species (P. berghei [U45969], P. yoelii [U45971], P. chabaudi [U49743], P. vivax [AF063138], P. knowlesi [P21303], and P. falciparum [320941]) with TgAMA1 (AF010264) was done to help identify phylogenetically conserved functional regions (Fig. 2). A strikingly high degree of conservation is found in a stretch of 34 aa, from residues His-63 to Ile-96 (numbering is for the T. gondii sequence). A minimum of 19 and a maximum of 22 identities were found in pairwise ungapped alignments of the Toxoplasma and each Plasmodium sequence in this region alone, corresponding to 56% (P. berghei, P. yoelii, and P. chabaudi) and 65% (P. falciparum) identity. This conserved region also includes the first cysteine residue of domain I (Fig. 1B), which predicts that 26 aa of this conserved block extend from the native protein, while 7 residues form part of a loop structure in domain I. On the NH-terminal side of His-63 in TgAMA1, the divergence is greater, with very few conserved positions in individual alignments and none in the multiple alignment of all seven sequences. In the putative 69-aa cytoplasmic domain, we find surprisingly little conservation in TgAMA1: only 12 identical positions (17%), including a conserved terminal tyrosine, are found in a multiple alignment with the Plasmodium homologues. Individual alignment with the P. vivax AMA1 shows the best conservation, with 19 identities and 21 similarities in this region. This dissimilarity may explain the differences in subcellular localization discussed below. TgAMA1 is expressed in tachyzoites of T. gondii. To derive specific antisera against TgAMA1, three TgAMA1 fragments were expressed as fusions with MBP in Escherichia coli. MBP-LF (long fragment) contained a region corresponding to domains I and II of the ectodomain, from Val-69 to Asp-410. MBP-SF (short fragment) covered domain I, starting at the second cysteine (Cys-166) but excluding the highly conserved region and extending through domain II to Asp-379. MBP-CF (cytosolic fragment) contained the COOH-terminal 62 aa, corresponding to the presumed cytoplasmic domain from Ala-479 to Tyr-541. Antiserum to MBP-LF detected a single strong band migrating at ⬃65 kDa in Western blots of total lysates or Triton X-100 extracts from RH tachyzoites separated on reducing SDS-PAGE (Fig. 3A). Identical results were obtained for the two other reference strains (ME49 and CEP) and when antibodies to MBP-SF and MBP-CF were used (data not shown). Antisera to peptides corresponding to the predicted mature NH terminus (N-pep: Ser-21 to Ser-36) and the central portion of the cytoplasmic tail (C-pep: Gln-489 to Lys-505) were generated in mice. Antibodies raised to C-pep strongly labeled the same 65-kDa band in lysates, while the antiserum to N-pep showed consistently weaker reactivity, with a band migrating slightly above the 65-kDa TgAMA1 signal (Fig. 3B). The apparent lack of reactivity of the N-pep antibodies with the major

INFECT. IMMUN.

FIG. 3. Western blot strips of parasite-associated TgAMA1 separated on SDS-PAGE. (A) Triton X-100 extracts of RH strain tachyzoites incubated with mouse polyclonal antibodies raised against MBP-LF (LF) or preimmune serum (PI). (B) Reactivity of C-pep (C) and N-pep (N) antibodies with tachyzoite extracts separated on SDS-PAGE gels. A putative precursor band, detected with anti-N-pep, is indicated by an arrow. The migration of size markers is indicated (in kilodaltons).

protein species, together with their detection of a minor and slightly larger species, suggests N-terminal posttranslational processing of TgAMA1 after signal sequence cleavage. This result is consistent with what has been observed for Plasmodium AMA1 proteins (9) but would require confirmation by direct N-terminal sequence analysis. The site at which this putative prepro cleavage occurs during synthesis and trafficking of TgAMA1 is not yet known. TgAMA1 is a microneme protein and relocalizes to the surface membrane upon egress of tachyzoites from the parasitophorous vacuole. To localize TgAMA1 in T. gondii, immunofluorescence studies using antisera against fusion proteins and peptides were performed on native extracellular or fixed and detergent-permeabilized intracellular and extracellular parasites. With fixed, permeabilized intracellular parasites, MBP-SF antiserum gave a crescent-shaped pattern at the anterior end of the cell (Fig. 4A and B). For fixed but unpermeabilized extracellular tachyzoites, the staining pattern was clearly distinct from that of intracellular parasites, showing a concentrated surface fluorescence of the apical half of the parasites, with a weak, circumferential staining posterior to this (Fig. 4C and D). Thus, the comparison of parasite surface and intracellular signals obtained with antiserum to the MBP-SF fusion protein shows an apparent redistribution of TgAMA1 subsequent to release of tachyzoites from the parasitophorous vacuole. Using N-pep antibodies on intracellular parasites, staining anterior to the nucleus was seen (Fig. 4E and F) that is clearly distinct from the staining pattern obtained with antisera to the MBP fusion proteins (Fig. 4A and B). Whether the anti-N-pep staining represents precursor proteins in transit through the secretory pathway could not be determined. Extracellular parasites which were not permeabilized with detergent showed predominantly staining of the very tip of the apical complex plus some peripheral staining, possibly representing a small amount of immature TgAMA1 just reaching the surface (Fig. 4G and H). The anterior fluorescence staining shown in Fig. 4A is most reminiscent of microneme proteins. Specifically, the staining pattern observed using the MBP-AMA1 polyclonal closely resembles the apical cap-like pattern observed for the microneme protein MIC2 (27). This is in contrast to the club-like staining pattern seen for the rhoptry protein ROP1 and the mottled staining pattern seen for the dense granule protein GRA3 (data not shown). To determine if TgAMA1 is indeed

VOL. 68, 2000

T. GONDII HOMOLOGUE OF PLASMODIUM AMA1

7083

FIG. 4. Immunofluorescence localization of TgAMA1 in T. gondii tachyzoites. (A, C, E, and G) Phase-contrast images; (B, D, F, and H) corresponding immunofluorescence images. (A and B) Formaldehyde-fixed and detergent-permeabilized intracellular parasites treated with MBP-SF antiserum. (C and D) Fixed and nonpermeabilized extracellular tachyzoites labeled with MBP-SF antiserum, showing apical localization of TgAMA1 on the surface of intact cells, with distinct circumferential staining. (E and F) Permeabilized intracellular tachyzoites incubated with N-pep antiserum, showing distinct perinuclear staining in addition to the subpellicular apical signal. (G and H) N-pep antiserum labeled mostly a small area at the conoid of extracellular parasites not treated with detergent.

localized to micronemes, dual immunofluorescence was performed with the mouse MAb CL22, raised to the C-pep of TgAMA1, and with rabbit antiserum for TgMIC2, a well-studied micronemal protein (44). The results (Fig. 5A to D) show colocalization of TgAMA1 and MIC2, strongly suggesting that the former protein is predominantly found in micronemes. To confirm this, immunoelectron microscopy on cryosections of intracellular tachyzoites was done using MBP-SF antiserum. Gold particles consistently decorated microneme vesicles but rarely other organelles or subcellular structures (Fig. 5E). A previous report had concluded that AMA1 of P. falciparum localizes to the anterior part of the rhoptry vesicles (9), the rhoptry necks. Virtually none of the organellar TgAMA1, however, is associated with rhoptries or rhoptry necks, suggesting that trafficking and secretion of AMA1 homologues to the parasite surface might follow a different pathway in these two genera. To determine if the occasional gold particle over rhoptries represents background or real labeling, the C-pep MAb CL22 was used to detect the protein in subcellularly fractionated T. gondii lysates. Both TgAMA1 and TgMIC2 are located in the ghost fraction and the middle fractions but are conspicuously absent from the rhoptry-dense granule band (Fig. 6). In contrast, ROP2, ROP3, ROP4, and GRA3 are enriched in the rhoptry-dense granule band compared to the middle fractions. Thus, biochemical analyses confirm that AMA1 is a microneme, not a rhoptry protein, in T. gondii. TgAMA1 is secreted as a lower-molecular-weight form. Shedding of lower-molecular-weight forms (Pk42/44) of PkAMA1(PK66) into the culture supernatant has been reported for free P. knowlesi merozoites prior to and during invasion of RBCs (11). Large amounts of a soluble lowermolecular-weight form of TgAMA1 were also found in the excreted-secreted fraction after incubation of extracellular tachyzoites in serum-free medium. In Western blots of this fraction collected at different time points, a single ⬃54-kDa soluble fragment, sTgAMA1, was detected with antibodies to MBP-LF (Fig. 7). The stable fragment was first detected after 10 to 20 min of incubation and rapidly increased in concentration over the course of the incubation. Detectable amounts of the 54-kDa soluble form could not be found associated with

free parasites (Fig. 7, pellet fractions). Western blots of secreted material and the corresponding parasite lysates were probed with the C-pep antibodies. The C-pep antiserum labeled the ⬃65-kDa forms in total lysates, but did not react with sTgAMA1 (data not shown). From these results and the observed size difference of approximately 10 kDa between the membrane-associated and soluble forms of TgAMA1, we propose that sTgAMA1 is the product of a proteolytic cleavage in domain III (i.e., in the ectodomain but relatively close to the transmembrane region). TgAMA1 is involved in host cell invasion and may be an essential protein. To investigate the function of TgAMA1, we tested the ability of mouse antibodies produced against fusion proteins to inhibit invasion of HFF by extracellular parasites. Tachyzoites were treated with the AMA1-reactive antiserum or pooled normal mouse serum prior to infection of an HFF monolayer. The number of parasites found associated with host cells was assessed microscopically. Coating parasites with the MBP-LF serum reduced the number of cell-associated tachyzoites by ⬃35 to 75% compared to control serum-coated parasites (data not shown). Precise quantitation of the inhibitory effect of the antiserum is difficult with this assay, so the more quantitative uracil incorporation assay was also done. Antisera obtained from two mice immunized with MBP-LF reproducibly inhibited invasion by ⬃40% (Fig. 8). This effect does not appear to be a result of antibodies to the MBP portion of the fusion protein, as antiserum from mice immunized with MBP-SF did not alter invasion by tachyzoites. As the MBP-LF protein differs from MBP-SF in containing most of the highly conserved region between His-63 and Ile-96, this segment may have an important biological role. To analyze AMA1 function in T. gondii more precisely, attempts were made to delete the AMA1 gene. Specifically, an AMA1 knockout vector designed to produce a gene disruption and containing HXGPRT as a selectable marker was constructed. RH⌬HXGPRT tachyzoites were transfected with a linearized AMA1-HXGPRT cassette and subsequently selected in MPA/XAN-containing medium. PCR analysis of genomic DNA isolated from the drug-resistant population indicated that upstream and downstream homologous recombination and integration of vector sequences had occurred. Even so, 80

7084

HEHL ET AL.

INFECT. IMMUN.

FIG. 5. Immunofluorescence localization of TgAMA1 in intracellular tachyzoites detected with the mouse MAb CL22 (A) and MIC2 detected with a rabbit polyclonal antiserum using intracellular tachyzoites that had been fixed and permeabilized (B). The merged image shows colocalization in large areas of overlap of the two signals (C). The corresponding phase-contrast image is depicted in panel D. (E) Localization of TgAMA1 by immunoelectron microscopy; intracellular tachyzoite section was labeled with MBP-SF antiserum. M, micronemes; R, rhoptries; D, dense granules. Immunogold staining of TgAMA1 in micronemes is marked with arrows. Bar, 0.2 ␮m.

MPA/XAN-resistant clones derived from this population contained an uninterrupted AMA1 gene, as determined by PCR. Several of these clones were analyzed further by Southern blot. As suggested by PCR analysis of the transfected population, homologous recombination in those selected clones had occurred on one but never on both sides of the HXGPRT cassette, leaving an intact TgAMA1 locus (data not shown).

FIG. 6. Western blot of intracellular organelles from T. gondii fractionated on a Percoll gradient. Lane 1, lysed cells; lane 2, top fraction containing tachyzoite ghosts; lane 3, middle fraction enriched in micronemes; lane 4, bottom fraction containing rhoptries and dense granules. Blots were probed with MAbs to organellar proteins AMA1 (CL22), MIC2 (6D10), ROP2, ROP3, ROP4 (T34A7), and GRA3 (2H11). The migration of size markers is indicated (in kilodaltons).

DISCUSSION In this report, we describe the cloning and characterization of TgAMA1, a Toxoplasma homologue of the Plasmodium AMA1. We show that the Toxoplasma and Plasmodium proteins are highly conserved in overall organization and, in many places, in their actual sequence. As with PfAMA1 and Plasmodium, antibodies to TgAMA1 significantly impair invasion by Toxoplasma, although to a lesser degree than seen in the Plasmodium experiments. This difference could reflect a greater redundancy in invasion mechanisms, less exposure of key epitopes, or a lower titer of the antibodies used in the Toxoplasma experiments.

FIG. 7. Secretion assay. Washed parasites were incubated in serum-free RPMI medium at 37°C. Excreted-secreted fractions (S) collected at 0, 5, 10, 20, 30, and 60 min of incubation of extracellular tachyzoites in serum-free medium and corresponding cell pellets (P) were separated on SDS–10% PAGE. Detection of membrane-bound TgAMA1 and secreted TgAMA1 was done with MBP-LF antibodies and subsequent incubation with peroxidase-conjugated goat anti-mouse Ig secondary antibodies.

VOL. 68, 2000

T. GONDII HOMOLOGUE OF PLASMODIUM AMA1

7085

ACKNOWLEDGMENTS We thank Gary Ward for exchange of information prior to publication and David Sibley for the rabbit anti-MIC2 serum and the MIC2specific MAb 6D10. This work was supported in part by grants from the Swiss National Science Foundation (31-45841.95 and 31-58912.99) to A.B.H., by a Bank of America-Giannini Foundation Postdoctoral Fellowship and an NRSA fellowship (AI10373) to C.L., a grant to J.-F.D. from the Ministe`re de la Recherche (PRFMMIP), and grants to J.C.B. from the National Institutes of Health (AI21423 and AI45057). A. Hehl and C. Lekutis contributed equally to this work. REFERENCES FIG. 8. Antibodies against recombinant TgAMA1 inhibit infection of HFF. Invasion in the presence of different concentrations of MBP-LF antiserum from mouse 1 (black column) and mouse 2 (stippled column) and MBP-SF antiserum from mouse 3 (white column) is shown as a percentage of invasion in the presence of pooled normal mouse serum as measured by uracil incorporation. Error bars represent the standard deviation for duplicate wells.

TgAMA1 is processed at least once and possibly twice during its transit to the extracellular milieu. The removal of a short segment at the N terminus apparently occurs at some stage during transit to the surface. In extracellular parasites, at least, the mature ectodomain is eventually shed into the supernatant, apparently as a result of cleavage off of the surface (i.e., just N-terminal to the membrane-anchoring domain). All of these properties are analogous to the situation with PfAMA1, although there are differences in detail, as expected from the very different modes of parasite replication (endodyogeny versus schizogony). A major difference between the Plasmodium AMA1s and TgAMA1 is in their subcellular localization in intracellular parasites: in Plasmodium, they are reported to be in the rhoptries (9, 28), whereas our data indicate a micronemal location for the Toxoplasma protein. Ward and colleagues have likewise colocalized TgAMA1 to the micronemes (11a). As the intracellular targeting of type I transmembrane proteins is typically encoded in the cytoplasmic tail (29), extensive differences in the sequence of this domain in AMA1 from the two parasites may explain the different localizations. It is also possible that the two compartments flow into one another and we are observing the protein at different stages in its trafficking from when the Plasmodium researchers localized it. There is no indication of such a flow in any other studies on rhoptry or microneme proteins, and so we consider this possibility unlikely. We were unable to fatally disrupt the AMA1 gene in Toxoplasma. Similarly negative results have also been obtained with Plasmodium and PfAMA1 (J. Adams, personal communication). While these techniques are not routine enough in either system to make a negative result conclusive, combined with the antibody studies, they strongly suggest a critical and potentially essential role for AMA1 in parasite growth. As yet, there are no clues to the presumptive ligands that are recognized by the extracellular and cytoplasmic domains of AMA1. The former may yield important information about host cell molecules that mediate attachment and/or invasion by the parasite. Molecules interacting with the cytoplasmic domain could include those necessary for correct targeting and/or for transducing a signal or kinetic energy from the outside to the inside of the parasite. Experiments that identify these ligands are now crucially needed.

1. Ajioka, J. W., J. C. Boothroyd, B. P. Brunk, A. Hehl, L. Hillier, I. D. Manger, M. Marra, G. C. Overton, D. S. Roos, K. L. Wan, R. Waterston, and L. D. Sibley. 1998. Gene discovery by EST sequencing in Toxoplasma gondii reveals sequences restricted to the Apicomplexa. Genome Res. 8:18–28. 2. Anders, R. F., P. E. Crewther, S. Edwards, M. Margetts, M. L. Matthew, B. Pollock, and D. Pye. 1998. Immunization with recombinant AMA-1 protects mice against infection with Plasmodium chabaudi. Vaccine 16:240–247. 3. Ausubel, F. M., R. Brent, R. E. Kingston, D. D. Moore, J. G. Seidman, J. A. Smith, and K. Struhl. 1994. Current protocols in molecular biology, p. 10.2.1–10.2.35, vol. 2. J. Wiley and Sons, Inc., New York, N.Y. 4. Beckers, C. J., J. F. Dubremetz, O. Mercereau-Puijalon, and K. A. Joiner. 1994. The Toxoplasma gondii rhoptry protein ROP 2 is inserted into the parasitophorous vacuole membrane, surrounding the intracellular parasite, and is exposed to the host cell cytoplasm. J. Cell Biol. 127:947–961. 5. Bonhomme, A., L. Pingret, and J. M. Pinon. 1992. Review: Toxoplasma gondii cellular invasion. Parassitologia 34:31–43. 6. Boothroyd, J. C., M. Black, S. Bonnefoy, A. Hehl, L. J. Knoll, I. D. Manger, E. Ortega-Barria, and S. Tomavo. 1997. Genetic and biochemical analysis of development in Toxoplasma gondii. Phil. Trans. R. Soc. Lond. B Biol. Sci. 352:1347–1354. 7. Carruthers, V. B., and L. D. Sibley. 1997. Sequential protein secretion from three distinct organelles of Toxoplasma gondii accompanies invasion of human fibroblasts. Eur. J. Cell Biol. 73:114–123. 8. Collins, W. E., D. Pye, P. E. Crewther, K. L. Vandenberg, G. G. Galland, A. J. Sulzer, D. J. Kemp, S. J. Edwards, R. L. Coppel, J. S. Sullivan, et al. 1994. Protective immunity induced in squirrel monkeys with recombinant apical membrane antigen-1 of Plasmodium fragile. Am. J. Trop. Med. Hyg. 51:711– 719. 9. Crewther, P. E., J. G. Culvenor, A. Silva, J. A. Cooper, and R. F. Anders. 1990. Plasmodium falciparum: two antigens of similar size are located in different compartments of the rhoptry. Exp. Parasitol. 70:193–206. 10. Deans, J. A., A. M. Knight, W. C. Jean, A. P. Waters, S. Cohen, and G. H. Mitchell. 1988. Vaccination trials in rhesus monkeys with a minor, invariant, Plasmodium knowlesi 66 kD merozoite antigen. Parasite Immunol. 10:535– 552. 11. Deans, J. A., A. W. Thomas, T. Alderson, and S. Cohen. 1984. Biosynthesis of a putative protective Plasmodium knowlesi merozoite antigen. Mol. Biochem. Parasitol. 11:189–204. 11a.Donahue, C. G., V. B. Carruthers, S. D. Gilk, and G. E. Ward. The Toxoplasma homolog of the Plasmodium apical membrane antigen-1 (AMA-1) is a microneme protein secreted in response to elevated intracellular calcium levels. Mol. Biochem. Parasitol., in press. 12. Donald, R. G. K., D. Carter, B. Ullman, and D. S. Roos. 1996. Insertional tagging, cloning, and expression of the Toxoplasma gondii hypoxanthinexanthine-guanine phosphoribosyltransferase gene: use as a selectable marker for stable transformation. J. Biol. Chem. 271:14010–14019. 13. Dubremetz, J. F. 1998. Host cell invasion by Toxoplasma gondii. Trends Microbiol. 6:27–30. 14. Dubremetz, J. F., A. Achbarou, D. Bermudes, and K. A. Joiner. 1993. Kinetics and pattern of organelle exocytosis during Toxoplasma gondii/hostcell interaction. Parasitol. Res. 79:402–408. 15. Dubremetz, J. F., N. Garcia-Reguet, V. Conseil, and M. N. Fourmaux. 1998. Apical organelles and host-cell invasion by Apicomplexa. Int. J. Parasitol. 28:1007–1013. 16. Fourmaux, M. N., A. Achbarou, O. Mercereau-Puijalon, C. Biderre, I. Briche, A. Loyens, C. Odberg-Ferragut, D. Camus, and J. F. Dubremetz. 1996. The MIC1 microneme protein of Toxoplasma gondii contains a duplicated receptor-like domain and binds to host cell surface. Mol. Biochem. Parasitol. 83:201–210. 17. Grimwood, J., and J. E. Smith. 1996. Toxoplasma gondii: the role of parasite surface and secreted proteins in host cell invasion. Int. J. Parasitol. 26:169– 173. 18. Hehl, A., T. Krieger, and J. C. Boothroyd. 1997. Identification and characterization of SRS1, a Toxoplasma gondii surface antigen upstream of and related to SAG1. Mol. Biochem. Parasitol. 89:271–282. 19. Hehl, A., I. D. Manger, and J. C. Boothroyd. 1997. Genetic analysis in Toxoplasma: gene discovery with expressed sequence tags and rapid map-

7086

HEHL ET AL.

ping of natural polymorphisms. Methods 13:89–102. 20. Hodder, A. N., P. E. Crewther, M. L. Matthew, G. E. Reid, R. L. Moritz, R. J. Simpson, and R. F. Anders. 1996. The disulfide bond structure of Plasmodium apical membrane antigen-1. J. Biol. Chem. 271:29446–29452. 21. Kasper, L. H., and P. L. Ware. 1985. Recognition and characterization of stage-specific oocyst/sporozoite antigens of Toxoplasma gondii by human antisera. J. Clin. Investig. 75:1570–1577. 22. Leriche, M. A., and J. F. Dubremetz. 1991. Characterization of the protein contents of rhoptries and dense granules of Toxoplasma gondii tachyzoites by subcellular fractionation and monoclonal antibodies. Mol. Biochem. Parasitol. 45:249–259. 23. Leriche, M. A., and J. F. Dubremetz. 1990. Exocytosis of Toxoplasma gondii dense granules into the parasitophorous. Parasitol. Res. 76:559–562. 24. Liu, C., V. Vigdorovich, V. Kapur, and M. S. Abrahamsen. 1999. A random survey of the Cryptosporidium parvum genome. Infect. Immun. 67:3960– 3969. 25. Manger, I. D., A. Hehl, S. Parmley, L. D. Sibley, M. Marra, L. Hillier, R. Waterston, and J. C. Boothroyd. 1998. Expressed sequence tag analysis of the bradyzoite stage of Toxoplasma gondii: identification of developmentally regulated genes. Infect. Immun. 66:1632–1637. 26. Medina-Acosta, E., and G. A. Cross. 1993. Rapid isolation of DNA from trypanosomatid protozoa using a simple “mini-prep” procedure. Mol. Biochem. Parasitol. 59:327–329. 27. Morrissette, N. S., V. Bedian, P. Webster, and D. S. Roos. 1994. Characterization of extreme apical antigens from Toxoplasma gondii. Exp. Parasitol. 79:445–459. 28. Narum, D. L., and A. W. Thomas. 1994. Differential localization of fulllength and processed forms of PF83/AMA-1 an apical membrane antigen of Plasmodium falciparum merozoites. Mol. Biochem. Parasitol. 67:59–68. 29. Ngo, H. M., H. C. Hoppe, and K. A. Joiner. 2000. Differential sorting and post-secretory targeting of proteins in parasitic invasion. Trends Cell Biol. 10:67–72. 30. Nichols, B. A., M. L. Chiappino, and G. R. O’Connor. 1983. Secretion from the rhoptries of Toxoplasma gondii during host-cell invasion. J. Ultrastruct. Res. 83:85–98. 31. Pennisi, E. 1999. Malarial genome comes into view. Science 286:1263–1265. 32. Perkins, M. E. 1984. Surface proteins of Plasmodium falciparum merozoites binding to the erythrocyte receptor, glycophorin. J. Exp. Med. 160:788–798. 33. Pfefferkorn, E. R., and L. C. Pfefferkorn. 1977. Toxoplasma gondii: specific labeling of nucleic acids of intracellular parasites in Lesch-Nyhan cells. Exp. Parasitol. 41:95–104.

Editor: W. A. Petri, Jr.

INFECT. IMMUN. 34. Pfefferkorn, E. R., L. C. Pfefferkorn, and E. D. Colby. 1977. Development of gametes and oocysts in cats fed cysts derived from cloned trophozoites of Toxoplasma gondii. J. Parasitol. 63:158–159. 35. Riggs, P. 1994. Expression and purification of maltose-binding protein fusions, p. 16.6.1–16.6.14. In F. M. Ausubel, R. Brent, R. E. Kingston, D. D. Moore, J. G. Seidman, J. A. Smith, and K. Struhl (ed.), Current protocols in molecular biology, vol. 2. John Wiley and Sons, Inc., New York, N.Y. 36. Sabin, A. B. 1941. Toxoplasmic encephalitis in children. J. Am. Med. Assoc. 116:801–807. 37. Sam-Yellowe, T. Y., H. Shio, and M. E. Perkins. 1988. Secretion of Plasmodium falciparum rhoptry protein into the plasma membrane of host erythrocytes. J. Cell Biol. 106:1507–1513. 38. Spano, F., and A. Crisanti. 2000. Cryptosporidium parvum: the many secrets of a small genome. Int. J. Parasitol. 30:553–565. 39. Spano, F., L. Putignani, S. Naitza, C. Puri, S. Wright, and A. Crisanti. 1998. Molecular cloning and expression analysis of a Cryptosporidium parvum gene encoding a new member of the thrombospondin family. Mol. Biochem. Parasitol. 92:147–162. 40. Strong, W. B., and R. G. Nelson. 2000. Preliminary profile of the cryptosporidium parvum genome: an expressed sequence tag and genome survey sequence analysis. Mol. Biochem. Parasitol. 107:1–32. 41. Suarez, C. E., T. F. McElwain, E. B. Stephens, V. S. Mishra, and G. H. Palmer. 1991. Sequence conservation among merozoite apical complex proteins of Babesia bovis, Babesia bigemina and other apicomplexa. Mol. Biochem. Parasitol. 49:329–332. 42. Taylor, D. W., C. B. Evans, S. B. Aley, J. R. Barta, and H. D. Danforth. 1990. Identification of an apically-located antigen that is conserved in sporozoan parasites. J. Protozool. 37:540–545. 43. Thomas, A. W., J. A. Deans, G. H. Mitchell, T. Alderson, and S. Cohen. 1984. The Fab fragments of monoclonal IgG to a merozoite surface antigen inhibit Plasmodium knowlesi invasion of erythrocytes. Mol. Biochem. Parasitol. 13:187–199. 44. Wan, K. L., V. B. Carruthers, L. D. Sibley, and J. W. Ajioka. 1997. Molecular characterisation of an expressed sequence tag locus of Toxoplasma gondii encoding the micronemal protein MIC2. Mol. Biochem. Parasitol. 84:203– 214. 45. Waters, A. P., A. W. Thomas, J. A. Deans, G. H. Mitchell, D. E. Hudson, L. H. Miller, T. F. McCutchan, and S. Cohen. 1990. A merozoite receptor protein from Plasmodium knowlesi is highly conserved and distributed throughout Plasmodium. J. Biol. Chem. 265:17974–17979.