Sarcocystis neurona Merozoites Express a Family of Immunogenic Surface Antigens That Are Orthologues of the Toxoplasma gondii Surface Antige
http://www.100md.com
感染与免疫杂志 2005年第2期
M. H. Gluck Equine Research Center, Department of Veterinary Science, University of Kentucky, Lexington, Kentucky
Center for Tropical and Emerging Global Diseases and Department of Cellular Biology, University of Georgia, Athens, Georgia
ABSTRACT
Sarcocystis neurona is a member of the Apicomplexa that causes myelitis and encephalitis in horses but normally cycles between the opossum and small mammals. Analysis of an S. neurona expressed sequence tag (EST) database revealed four paralogous proteins that exhibit clear homology to the family of surface antigens (SAGs) and SAG-related sequences of Toxoplasma gondii. The primary peptide sequences of the S. neurona proteins are consistent with the two-domain structure that has been described for the T. gondii SAGs, and each was predicted to have an amino-terminal signal peptide and a carboxyl-terminal glycolipid anchor addition site, suggesting surface localization. All four proteins were confirmed to be membrane associated and displayed on the surface of S. neurona merozoites. Due to their surface localization and homology to T. gondii surface antigens, these S. neurona proteins were designated SnSAG1, SnSAG2, SnSAG3, and SnSAG4. Consistent with their homology, the SnSAGs elicited a robust immune response in infected and immunized animals, and their conserved structure further suggests that the SnSAGs similarly serve as adhesins for attachment to host cells. Whether the S. neurona SAG family is as extensive as the T. gondii SAG family remains unresolved, but it is probable that additional SnSAGs will be revealed as more S. neurona ESTs are generated. The existence of an SnSAG family in S. neurona indicates that expression of multiple related surface antigens is not unique to the ubiquitous organism T. gondii. Instead, the SAG gene family is a common trait that presumably has an essential, conserved function(s).
INTRODUCTION
Sarcocystis neurona is an apicomplexan parasite and the primary cause of equine protozoal myeloencephalitis (EPM). EPM is a serious debilitating disease and is the most commonly diagnosed neurologic disorder of horses. Although seroprevalence studies have indicated that approximately 30 to 50% of horses have been exposed to S. neurona, the incidence of EPM is estimated to be less than 1% (15, 46). Thus, S. neurona infection clearly does not equate directly with clinical illness, and it is not clear what factors influence the progression from simple infection to severe neurologic disease. The normal life cycle of S. neurona alternates between the definitive host, the opossum (20), and various small mammal intermediate hosts, including skunks (9), raccoons (17), armadillos (8), and cats (16). Although S. neurona readily infects equids, these animals are currently believed to be aberrant hosts for this parasite species since latent forms (sarcocysts) have not been found in infected horses.
Like other members of the Apicomplexa, S. neurona is an obligate intracellular parasite that requires a number of unique molecules (i.e., virulence factors) to support its parasitic lifestyle. Apicomplexan surface molecules are important virulence factors that are responsible for the pathogen's initial interactions with the host cell surface and components of the host immune response. A broad family of more than 20 related surface antigens has been found to be expressed by Toxoplasma gondii (42). A recent bioinformatic search of the Toxoplasma genome database (www.toxodb.org) indicated that the full assemblage of surface antigen genes is even more extensive, and 161 related sequences have been identified in the genome of the ME49 strain (38). Although many of these sequences may be pseudogenes that are not expressed, it is apparent that T. gondii has the capacity to produce a complex array of surface antigens. These paralogous molecules, which have been designated SAGs and SAG-related sequences (SRSs), are developmentally regulated and exhibit various levels of sequence similarity to one of the major T. gondii surface antigens, TgSAG1 or TgSAG2. The T. gondii SAGs appear to be involved in receptor-ligand interactions with the host cell surface, most likely through binding of sulfated proteoglycans (27, 37), and there is increasing evidence that some of the SAGs can modulate host immune responses (42). The evolutionary advantage provided by expansion of the SAG/SRS gene family is unknown, but it has been speculated that the comprehensive array of surface antigens allows the very wide host range of T. gondii (4, 42). Individual SAG homologues have been described in the genus Sarcocystis (18, 19). However, it was not clear whether Sarcocystis spp. also express a complex family of related surface antigens.
In an effort to identify and characterize virulence factors of S. neurona, we have conducted a sequencing project that has generated approximately 8,500 expressed sequence tags (ESTs) from this organism (31, 43). Examination of this sequence database revealed a family of S. neurona surface antigens that are orthologues of the SAG/SRS family of surface proteins in T. gondii. Here, we present the results of characterization of the four S. neurona surface antigens identified thus far. Based on their clear homology to T. gondii SAGs, the S. neurona surface antigens have been designated SnSAG1, SnSAG2, SnSAG3, and SnSAG4.
MATERIALS AND METHODS
Parasite cultures. S. neurona strain SN3 (21) merozoites were propagated by serial passage in bovine turbinate cells and were maintained in RPMI 1640 medium supplemented with 10% fetal bovine serum, 2 mM sodium pyruvate, and Pen/Strep Fungizone (BioWhittaker, Inc.). Extracellular merozoites were harvested and purified from disrupted host cell monolayers by filtration through 3.0-μm-pore-size membranes, as described previously for Neospora caninum (33).
Immunoscreening of S. neurona cDNA library. Construction and analyses of the cSn.1 S. neurona merozoite cDNA library have been described previously (31). Phage plaques were allowed to form for 3 h at 42°C on Escherichia coli XL1-Blue MRF' host cells (Stratagene) grown on 150-mm NZY agar plates. When plaques became visible, the agar was overlaid with nitrocellulose filters previously soaked in 10 mM isopropyl--D-thiogalactopyranoside (IPTG), and the plates were incubated for an additional 3 h at 37°C. The filters were lifted from the plates, washed with TNT buffer (10 mM Tris-HCl [pH 7.5], 150 mM NaCl, 0.05% Tween 20), and blocked in phosphate-buffered saline (PBS)-5% dry milk-5% normal goat serum-0.05% Tween 20.
Antigenic cDNA clones were identified by screening the filters with cerebrospinal fluid (CSF) from a horse that had been naturally infected with S. neurona and exhibited a high titer of intrathecal antibodies against S. neurona in a Western blot analysis. Prior to screening of the S. neurona cDNA library, the CSF was depleted of antibodies that were reactive with E. coli and phage proteins. Briefly, the CSF was diluted 1:20 in PBS-0.1% dry milk-0.1% normal goat serum-0.05% Tween 20 and incubated for 30 min with filters carrying plaque lifts of a previously described N. caninum cDNA library (34). After adsorption of potential cross-reactive antibodies, the diluted CSF solution was incubated for 1 h with the cSn.1 filters. The filters were washed and then incubated for 1 h with goat anti-equine immunoglobulin G (IgG) conjugated to horseradish peroxidase (Jackson Immunoresearch Labs, Inc.) diluted 1:10,000. Immunoreactive phage plaques were picked with sterile pipette tips and suspended in 40 μl of SM buffer (50 mM Tris-HCl [pH 7.5], 100 mM NaCl, 8 mM MgSO4, 0.01% gelatin). The cDNA inserts of the selected phage plaques were PCR amplified by using the T3 and T7 oligonucleotide primers, and the resulting products were analyzed by agarose gel electrophoresis. Sequencing reactions with the T3 primer were conducted for the amplified cDNAs to provide preliminary identification of the immunoreactive clones. Phagemid excision was performed with selected cDNA clones, and plasmids were rescued in SOLR cells by using the manufacturer's protocol (Stratagene).
Immunoaffinity purification of antibodies. An immunoreactive phage clone was plated at a high density on two NZY agar plates and overlaid with IPTG-soaked nitrocellulose membranes. After 4 h of incubation at 37°C, the membranes were lifted from the plates and washed twice with TNT buffer. The membranes were blocked for 15 min and then incubated overnight at 4°C in CSF from an S. neurona-infected horse. After three washes, the two membranes were incubated sequentially for 15 min in 10 ml of 0.2 M glycine (pH 2.8) to elute the bound antibodies. One milliliter of 1 M Tris (pH 8.0) was added to neutralize the solution, followed by 1 ml of 10x PBS. The eluted antibodies were used for Western blot analysis, as described below.
S. neurona EST database searches and sequence analyses. Genes for surface antigen homologues were identified in the S. neurona clustered EST database (http://paradb.cis.upenn.edu/sarco/index.html) by using the BLAST (basic local alignment search tool) set of programs (1). At the time that the searches were conducted, the database contained 686 consensus sequences that had been generated from 1,883 S. neurona ESTs. Selected cDNAs were obtained from the archived collection of EST clones and sequenced by using a ABI Prism BigDye terminator cycle sequencing reaction mixture (Perkin-Elmer Applied Biosystems). The reaction mixtures were purified by using Centri-Sep spin columns (Princeton Separations), and the eluted extension products were resolved and analyzed with an ABI 310 genetic analyzer. Sequence analyses were conducted with Genetics Computer Group software (14) and programs available at the National Center for Biotechnology Information website (http://www.ncbi.nlm.nih.gov/) and the Expert Protein Analysis System (ExPASy) server of the Swiss Institute of Bioinformatics (http://www.expasy.ch/). Multiple-sequence alignments of the putative mature forms of the proteins (without the signal peptide and carboxyl-terminal tails) were constructed by using Multalin software (10). The individual SAG domains D1 and D2, as defined previously (27) and demarcated by the conserved cysteine residues within each domain, were aligned, and a phylogenetic dendrogram was constructed by using ClustalW (59).
Recombinant antigen expression and generation of polyclonal antisera. The predicted SnSAG open reading frames, without their predicted amino-terminal signal peptides and carboxyl-terminal hydrophobic tails, were amplified by PCR by using primers that incorporate a 5' end NdeI restriction site and a 3' end XhoI restriction site. The amplification products were digested with NdeI and XhoI and ligated into NdeI/XhoI-digested expression vector pET22b (Novagen), and the constructs were transformed into E. coli INVF' for plasmid propagation. The resulting expression plasmids were transformed into E. coli BL21-CodonPlus (Stratagene), and clones that expressed high levels of recombinant protein were selected for use. The histidine-tagged recombinant SnSAGs (rSnSAG1, rSnSAG2, rSnSAG3, and rSnSAG4) were purified by nickel-column chromatography by using the manufacturer's protocol (Novagen), and monospecific polyclonal antisera against the purified proteins were produced by immunization of rabbits and rats (Cocalico Biologicals, Inc.).
Western blot analysis. Parasites were lysed in sodium dodecyl sulfate sample buffer supplemented with a protease inhibitor cocktail (Sigma) and 2% 2-mercaptoethanol (2-ME), and the lysates were separated in 10 or 12% polyacrylamide gels (40). Dithiothreitol-reduced low-range molecular mass standards (Bio-Rad Laboratories, Inc.) were included in all Western blot analyses. Proteins were transferred to nitrocellulose membranes by semidry electrophoretic transfer in Tris-glycine buffer (pH 8.3). The membranes were blocked with PBS containing 5% nonfat dry milk, 5% goat serum, and 0.05% Tween 20 and then incubated for 1 h with primary antibody. After washing, the membranes were incubated with horseradish peroxidase-conjugated immunoglobulin G secondary antibody (Jackson Immunoresearch Labs, Inc.). The blots were washed, processed for chemiluminescence with the Supersignal substrate (Pierce Chemical Company), and exposed to film or visualized with a FluorChem 8800 imaging system (Alpha Innotech, Inc.).
Biotinylation of surface proteins and precipitation with immobilized streptavidin. Approximately 3 x 107 freshly harvested merozoites were resuspended in 1 ml of cold PBS (pH 7.8). Sulfo-N-hydroxy-succinimide-biotin (Pierce) was added to a concentration of 0.5 mg/ml and incubated at room temperature for 30 min. The labeled parasites were washed twice with 5 ml of PBS and stored at –20°C. The labeled parasite pellet was lysed with 1 ml of radioimmunoprecipitation assay buffer (50 mM Tris [pH 7.5], 1% Triton X-100, 0.5% sodium deoxycholate, 0.2% sodium dodecyl sulfate, 100 mM NaCl, 5 mM EDTA) supplemented with RNase, DNase, and a protease inhibitor cocktail, and the sample was centrifuged at 16,000 x g to remove the insoluble fraction. Biotin-labeled proteins were precipitated from the soluble fraction with UltraLink immobilized streptavidin (Pierce), and the precipitated proteins were analyzed by Western blotting, as described above.
Indirect immunofluorescent labeling of intracellular parasites. For detection of the SnSAGs on intracellular parasites, merozoites were inoculated onto bovine turbinate cells grown in LabTek chamber slides (Nunc Nalgene) or onto 12-mm round coverslips in 24-well plates. At 24, 48, or 72 h postinoculation, the cells were fixed either in –20°C methanol for 10 min or in 2.5% formalin-PBS-0.01% glutaraldehyde for 30 min on ice. The formalin-fixed cells were permeabilized with 0.2% Triton X-100. The slides were blocked by 30 min of incubation with 10% normal goat serum in PBS. The primary antibodies were rabbit antisera against SnSAG1, SnSAG2, SnSAG3, or SnSAG4 or rat antisera against SnSAG1, SnSAG2, or SnSAG3. A rabbit antiserum against T. gondii IMC3 (24) was used to visualize the inner membrane complex with the SnSAGs in methanol-fixed cells. After incubation with the primary antibodies, the slides were rinsed and then incubated with goat anti-rabbit IgG or goat anti-rat IgG conjugated to either fluorescein isothiocyanate (FITC) or Texas Red (Jackson Immunoresearch Labs, Inc.). The slides were mounted in Vectashield with 4',6'-diamidino-2-phenylindole (DAPI) (Vector Laboratories, Inc.) and examined with a Zeiss axioscope or a Leica DM IRBE equipped for epifluorescence microscopy.
Nucleotide sequence accession numbers. The sequences reported here have been deposited in the GenBank database under accession numbers AY032845, AY191006, AY191007, and AY191008.
RESULTS
Isolation and sequence analysis of immunoreactive cDNA clones. Primary screening of the cSn.1 cDNA library resulted in identification of multiple immunoreactive phage plaques, and 25 plaques were selected and resuspended in SM buffer. Amplification of the cDNA inserts with the T3 and T7 oligonucleotides revealed that 22 of the phage clones had similar lengths (approximately 1,500 bp), and sequence analysis with the T3 primer indicated that these 22 cDNAs represented the same gene. A secondary immunoscreening was performed for five of the selected cDNAs, and two highly reactive phage clones, designated SnAgI.8 and SnAgI.9, were chosen for further analyses.
To obtain preliminary identification of the parasite protein encoded by the selected cDNAs, the SnAgI.9 clone was used to affinity purify antibodies that bind the antigen expressed by this clone, and the eluted antibodies were used to probe a Western blot of S. neurona merozoite lysate. The antigen revealed by the phage-purified antibodies migrated at approximately 31 kDa under reducing conditions, and it comigrated with a protein that is recognized by equine or rabbit antisera against S. neurona as a major immunodominant antigen of this parasite (data not shown). This result implied that the 22 matching cDNA clones isolated during the library screening and represented by SnAgI.8 and SnAgI.9 encode an immunodominant antigen of S. neurona.
Full-length sequence analysis of SnAgI.8 revealed a 1,493-nucleotide cDNA insert with an open reading frame (ORF) that encodes a 276-amino-acid protein. Sequence analysis of SnAgI.9 indicated that this clone was virtually identical to SnAgI.8, although its 3' untranslated region extended an additional 160 nucleotides. Although this may have been due to alternative polyadenylation, mispriming by oligo(dT) during cDNA synthesis cannot be ruled out. A hydrophobicity plot of the encoded protein showed that there were hydrophobic domains at both termini, which corresponded to a predicted signal peptide at the amino terminus and a glycosylphosphatidylinositol (GPI) anchor addition sequence at the carboxyl terminus (data not shown). The signal peptide cleavage was predicted to occur at Ala15-Arg16 (SignalP), and the most likely GPI transamidase cleavage site was predicted to be at Ala247-Asn248 (DGPI; Swiss Institute of Bioinformatics). A single N-glycosylation site was predicted at residues 140 to 143. Removal of the N-terminal and C-terminal signal sequences resulted in a mature protein consisting of 232 amino acids that had a predicted molecular mass of 24.2 kDa, not including any potential posttranslational modifications (e.g., glycosylation).
A blastx search (1) of the nonredundant GenBank databases was conducted with the SnAgI.8 coding sequence as the query. This search revealed statistically significant similarity to the 31-kDa major surface antigen of Sarcocystis muris SmMSA (19) and a less significant but recognizable similarity to several SAG2-related surface antigens of T. gondii (41). In conjunction with the Western blot analysis and the predictions of a signal peptide and a GPI anchor addition, these results collectively suggested that the SnAgI.8 and SnAgI.9 cDNAs represent a gene that encodes an immunodominant surface antigen of S. neurona. Consequently, this protein was designated SnSAG1 by using the genetic nomenclature that is utilized for the related apicomplexan parasites T. gondii and N. caninum (34, 57).
A blastn search of the S. neurona ESTs (conducted on 4 October 2002) with the SnAgI.8 nucleotide sequence as the query resulted in identification of 654 overlapping ESTs (7.7% of the 8,482 S. neurona ESTs) that correspond to the SnSAG1 gene. This result indicated that the SnSAG1 sequence is extensively represented in the cDNA library, and it further suggested that the SnSAG1 surface antigen is expressed at high levels by S. neurona merozoites.
Identification of additional S. neurona surface antigens. To determine if other SnSAG surface antigens are expressed by S. neurona merozoites, the database of S. neurona EST consensus sequences (http://paradb.cis.upenn.edu/sarco/index.html) was searched with the BLAST tool tblastn (1) by using SnSAG1 and the various available T. gondii SAGs and SRSs as query sequences. At the time that the searches were conducted, the dbEST database contained 1,883 S. neurona ESTs, which had been clustered into 686 consensus gene sequences. These searches resulted in identification of gene sequences that appeared to encode three additional SnSAG paralogues, which were tentatively designated SnSAG2, SnSAG3, and SnSAG4. A representative cDNA clone for each of the three novel SnSAGs was recovered from the cryopreserved stocks of S. neurona EST clones and used for further analysis.
SnSAG2 was identified from the S. neurona clustered EST database by using T. gondii SAG1 as the query sequence. The SnSAG2 gene was initially represented by a cluster of 52 ESTs, which provided a 961-nucleotide consensus sequence that spanned the entire ORF of the gene. For further analysis and manipulation of SnSAG2, cDNA SnEST4a02h07 was recovered from the archived collection of S. neurona EST clones. The SnSAG2 gene encodes a 168-amino-acid protein that has a predicted amino-terminal signal peptide and a carboxyl-terminal signal for addition of a GPI anchor; the sequence does not contain any potential N-linked glycosylation sites. After removal of the signal peptide (28 residues) and the carboxyl-terminal tail (approximately 23 residues), a molecular mass of 11.8 kDa was predicted for the mature SnSAG2 protein. A blastn search of the current database of 8,482 S. neurona ESTs revealed 148 entries (approximately 1.7%) that correspond to the SnSAG2 sequence, indicating that this gene is well represented in the cSn.1 cDNA library and is likely expressed at high levels by S. neurona merozoites.
Both SnSAG3 and SnSAG4 were discovered by searching the S. neurona ESTs with the SnSAG1 protein sequence as the query. The SnSAG3 gene was initially identified as a cluster of three overlapping ESTs, from which clone SnEST4a18g06 was selected and recovered for further study. Full-length sequence analysis of this cDNA clone revealed a 1,496-nucleotide insert with an ORF that encodes a 281-amino-acid peptide. The derived SnSAG3 protein sequence was predicted to contain a 21-residue signal peptide and a 34-amino-acid carboxyl-terminal tail for GPI anchor addition. Removal of the signal peptide and carboxyl-terminal tail resulted in a mature protein consisting of 226 residues with a molecular mass of 23.1 kDa. A blastn search of the current database with the full-length nucleotide sequence of clone SnEST4a18g06 as the query resulted in identification of 19 S. neurona ESTs that correspond to SnSAG3.
The SnSAG4 gene was initially identified as a singleton cDNA clone, SnEST4a07g08. Full-length sequence analysis of this clone revealed a 1,093-nucleotide cDNA insert with an ORF encoding 287 amino acids. The SnSAG4 protein has a predicted 28-residue signal peptide and a 27-residue carboxyl-terminal tail for GPI anchor addition. Removal of these signals resulted in a mature protein that is 232 amino acids long and has a molecular mass of 24.4 kDa. A blastn search of the S. neurona EST database resulted in identification of four ESTs that correspond to the SnSAG4 gene, suggesting that this locus is transcribed at a relatively low level.
Comparison of SnSAGs with surface antigen orthologues. To assess the extent of sequence conservation in the S. neurona proteins, the SnSAGs were aligned with surface antigens that have been described in other members of the Coccidia. These analyses demonstrated that SnSAG1, SnSAG3, and SnSAG4 are most similar to one another, but they also exhibited clear homology to the SmMSA major surface antigen of S. muris and the TgSAG2 family of surface antigens from T. gondii (Fig. 1A). The SnSAG2 sequence was determined to be most similar to the T. gondii SAG1 family of surface antigens (Fig. 1B). Pairwise alignments indicated that SnSAG1 exhibits approximately 36% sequence identity with either SnSAG3 or SnSAG4, whereas SnSAG3 and SnSAG4 exhibit about 33% sequence identity. SnSAG1, SnSAG3, and SnSAG4 were found to exhibit approximately 30% sequence identity with SmMSA, and they generally exhibited less than 25% sequence identity with the TgSAG2 family members. Similar to SnSAG1, SnSAG3, and SnSAG4, the overall sequence identity exhibited by SnSAG2 with any one of the previously described TgSAG/SRS proteins was found to be quite modest (<25%). Despite the moderate sequence similarities that were observed, the SnSAGs were clearly homologues of the T. gondii surface antigens. This was primarily revealed by the presence of multiple conserved cysteine residues (Fig. 1), which have been described previously in the TgSAGs (7) and have been shown to participate in the formation of intramolecular disulfide bonds (27).
To further define the similarity between the SnSAG sequences and the T. gondii SAG families, the individual domains (D1 and D2) of the SnSAGs and the surface antigen prototypes TgSAG1 and TgSAG2A were aligned by using ClustalW, and a dendrogram was constructed from the resulting analysis. In the derived tree, the single domain of TgSAG2A clustered with the D2 and D1 domains of SnSAG1, SnSAG3, and SnSAG4 (Fig. 2). The D1 and D2 domains of TgSAG1 formed a second clade with the single domain of SnSAG2. TgSAG2A appeared to be most similar to the D2 domain of the SnSAGs, whereas SnSAG2 was most closely allied with the D1 domain of TgSAG1. These sequence comparisons suggest that the SAG1 and SAG2 gene family dichotomy that has been described in T. gondii is also present in the SnSAGs of S. neurona merozoites.
Although the S. neurona surface antigen genes were readily identified as orthologues of the SAG/SRS gene family in T. gondii, a clear line of phylogenetic descent could not be established for each SnSAG gene. Consequently, the basis for naming SnSAG1 through SnSAG4 was provided by their chronology of identification and their representation (i.e., abundance) in the S. neurona EST database. Thus, the numerical designations assigned to the SnSAGs do not imply direct homology to the TgSAGs with the corresponding numbers.
Characterization of the SnSAG gene products. In order to produce specific reagents that could be utilized for investigating the S. neurona surface antigens, each of the SnSAG genes was expressed as a His-tagged recombinant protein in E. coli. The expressed proteins were purified by nickel-column chromatography and used to immunize rabbits and rats for the production of monospecific polyclonal antibodies. Western blot analysis of 2-ME-reduced S. neurona merozoite lysate with the resulting SnSAG antisera revealed proteins that migrated at approximately 32, 20, 28, and 32 kDa for SnSAG1, SnSAG2, SnSAG3, and SnSAG4, respectively (Fig. 3A). When nonreduced lysate was analyzed, SnSAG1 to SnSAG4 migrated at approximately 26, 16, 25, and 26 kDa (Fig. 3B); these findings are reasonably consistent with the molecular mass predicted for each protein. Notably, the comigrating SnSAG1 and SnSAG4 proteins and the lower-molecular-mass protein SnSAG2 corresponded to the two most dominant bands typically recognized by antisera from infected animals (Fig. 3).
Homology to the TgSAGs indicated that the SnSAGs were likely surface antigens of S. neurona. To establish that the SnSAGs are membrane associated, we performed Triton X-114 phase partitioning with S. neurona merozoites, which separates proteins based on their amphiphilicity (5). Western blot analysis of the partitioned proteins demonstrated that all four SnSAGs were separated exclusively into the detergent phase (Fig. 4), signifying that these proteins are associated with membranes of S. neurona merozoites. To verify that the SnSAGs are displayed on the merozoite surface, extracellular parasites were labeled with sulfo-N-hydroxysuccinimide-biotin, which does not permeate membranes and selectively labels surface-exposed proteins. The biotin-labeled proteins were specifically precipitated with immobilized streptavidin, and Western blot analysis with the anti-SnSAG antisera revealed that all four SnSAGs were in the fraction of biotinylated proteins (Fig. 5), thus confirming that they are displayed on the surface of merozoites.
Presence and distribution during intracellular development. Intracellular S. neurona undergoes a complex process of cell division termed endopolygeny. This process uncouples genome replication from cytokinesis, thereby producing a polyploid nucleus in a large mother cell called a schizont. The nuclear material eventually divides concurrent with cytokinesis, resulting in the formation of multiple daughter merozoites. To assess the expression of the SnSAGs during endopolygeny and to determine the distribution of the surface antigens in the maturing schizonts, intracellular parasites were analyzed by indirect immunofluorescence. This analysis revealed that the four SnSAGs are present at all stages of intracellular development, although the pattern of labeling was somewhat different for each surface antigen (Fig. 6). In general, methanol fixation of cells yielded a uniform distribution of label for the four SnSAGs throughout intracellular development, with punctations of label appearing in the late stages of endopolygeny. For SnSAG2 (Fig. 6C) and SnSAG3 (data not shown), these punctations of label appeared to define the newly forming daughter merozoites that were randomly dispersed throughout the schizont. In the mature schizonts, SnSAG1 (Fig. 6A), SnSAG2 (data not shown), and SnSAG4 (data not shown) were fairly evenly distributed on the individual merozoites, as expected. The exception was SnSAG3, which exhibited distinct labeling at the apical end of the merozoites (Fig. 6D). The distributions of the SnSAGs were similar for extracellular merozoites (data not shown).
Notably, when the cells had been fixed with formaldehyde and permeabilized with Triton X-100, antibodies to SnSAG1 (Fig. 6B) and SnSAG4 (Fig. 6E) highlighted a reticulum that was most apparent during the intermediate phases of endopolygeny. This pattern of fluorescent labeling was not observed for SnSAG2 and SnSAG3. Occasionally, a vague network of SnSAG1 or SnSAG4 could be perceived in some methanol-fixed cells (data not shown). Collectively, the immunofluorescence results demonstrated that the four surface antigens are present throughout intracellular growth, but they may be concentrated in dissimilar regions in the developing and mature schizonts.
To further clarify the localization of the SnSAGs during parasite development, the inner membrane complex protein IMC3 was labeled as a marker for daughter merozoite formation (24, 36). Throughout much of the intracellular development of the schizonts, the IMC3 label was weak and nondescript (Fig. 7A). However, strong and distinct labeling of IMC3 became apparent in late-stage schizonts, coinciding with nuclear division and preceding the completion of cytokinesis. At this stage, IMC3 delineated the early daughter merozoites (Fig. 7A), similar to what was observed for SnSAG2 and SnSAG3. Indeed, covisualization of IMC3 with SnSAG3 demonstrated that there is a tight association between these two antigens during the formation of the merozoites (Fig. 7B). IMC3 and SnSAG1 also exhibited partial colocalization, although a considerable amount of additional SnSAG1 label was observed (data not shown). Further development of the parasites resulted in a strong, circumferential fluorescence of IMC3 on individual merozoites of the mature schizont (Fig. 7A). Unfortunately, immunofluorescence analysis of IMC3 was possible only with cells fixed with methanol. Therefore, the inner membrane complex protein could not be visualized in formaldehyde-fixed, Triton X-100-permeabilized cells. However, the labeling pattern seen for IMC3 (Fig. 7) was clearly distinct from the reticulum of label that was observed for SnSAG1 and SnSAG4 (Fig. 6B and E).
DISCUSSION
Sarcocystis is a very broad genus that encompasses some of the most prevalent parasites of vertebrate animals. Despite their abundance and importance as pathogens, the Sarcocystis spp. have been understudied relative to other members of the Apicomplexa. Consequently, there is a significant information gap for a large segment of this phylum. In an effort to define the genetic and molecular composition of S. neurona, we have conducted an EST sequencing project (31, 43) and explored the sequence database to identify S. neurona virulence factors. These searches have revealed a gene family that encodes at least four paralogous proteins with obvious homology and probable structural similarity to the SAG/SRS family of surface antigens in T. gondii (42). The four SnSAGs were all found to be membrane associated and displayed on the merozoite surface, consistent with their homology to surface antigens in related parasites. Sequence predictions suggested the presence of glycolipid anchors, as observed for the SAGs of T. gondii (49, 60) and N. caninum (32, 55). Although the SnSAGs could not be liberated from the merozoite surface by treatment with phosphatidylinositol-specific phospholipase C (data not shown), itis likely that this was due to inositol modification in the S. neurona GPI anchors, as observed previously for N. caninum tachyzoites (54, 55). It is apparent that the SnSAGs are highly immunogenic, similar to the major surface antigens of T. gondii (56) and N. caninum (30, 32). Indeed, SnSAG1 and SnSAG2 corresponded to two of the most immunodominant antigens (Fig. 4), which were previously designated Sn30 and Sn16 and were speculated to be surface proteins of S. neurona merozoites (44). Precedence from studies of T. gondii suggests that these SnSAGs may serve as multitasking virulence factors that are important for S. neurona's ability to invade cells and evade the host immune response. Further in-depth investigation of this gene family in S. neurona should allow comparative studies that could provide insight into the evolution and function of the SAGs in the tissue-forming coccidia.
The SnSAG1 surface antigen, in particular, appeared to be exceedingly immunogenic. Our cDNA library immunoscreening resulted in identification of the SnSAG1 gene in 22 of 25 reactive phage clones, which is consistent with this antigen's immunodominance. Additionally, the SnSAG1 gene was represented by the largest assembly of S. neurona ESTs (43), thus implying that the SnSAG1 mRNA is abundant and expressed at high levels in S. neurona merozoites. In view of its abundance and immunodominance, it was predictable that SnSAG1 would be the initial S. neurona antigen identified by us, as well as by other workers (18). Identical to the data presented here, SnSAG1 was described as an immunodominant antigen of S. neurona that localized to the surface of extracellular merozoites in immunofluorescence and immunoelectron microscope analyses (18). Notably, Ellison et al. found that the SnSAG1 genomic locus contains a 128-bp intron that is located in the hinge region separating the D1 and D2 domains (18), identical to what was described in the TgSAG2 family of surface antigens (27, 41). Consistent with our predicted gene phylogeny (Fig. 2), these data confirm that the S. neurona major surface antigen SnSAG1 gene is most similar to the TgSAG2 gene family rather than the gene encoding the predominant surface antigen of T. gondii, TgSAG1. Amplification of the SnSAG3 and SnSAG4 genomic loci also indicated the presence of an intron(s) in both genes, although the specific location of the intron within each locus was not determined (data not shown).
Fluorescence labeling demonstrated that all four SnSAGs are present throughout intracellular development (Fig. 6). Additionally, these analyses revealed a reticulate structure that was most evident in cells that were formaldehyde fixed and permeabilized with Triton X-100 (Fig. 6B). Permeabilization with strong detergents such as Triton X-100 can cause partial extraction of membrane proteins, especially from the periphery of cells, along with an accompanying increased immunoreactivity in intracellular membranes and compartments (26). Thus, the network of SnSAGs likely represents the deep invaginations of the schizont pellicle that have been observed in electron micrographs (58). Although the inner membrane complex appears to be present during all stages of intracellular development (58), the inner membrane complex protein IMC3 was not patently detectable with the infolded pellicle network throughout much of schizont maturation (Fig. 7A). Instead, IMC3 became evident at the latest stages of development when daughter cell formation had initiated, which is consistent with what has been observed during T. gondii endodyogeny (24). At this point of development, the SnSAGs became associated with the inner membrane complex, and SnSAG2 and SnSAG3 were almost exclusively linked to the newly forming merozoites (Fig. 6C and 7B).
Although the SnSAGs were found to be present throughout intracellular development (Fig. 6), it is probable that these proteins function primarily during the extracellular merozoite stage. By virtue of their location on the parasite surface, the SnSAGs are likely central to the parasite's initial interactions with the host cell and components of the immune response. Indeed, a role in the invasion of host cells has been previously proposed for the Sn16 surface antigen (identified here as SnSAG2) of S. neurona merozoites (44). Likewise, the surface antigens of T. gondii (22, 23, 47, 48) and N. caninum (28) have been implicated in attachment to and invasion of host cells. It was apparent from these prior studies, however, that the participation of the SAGs in this essential process is relatively complex. In fact, the individual surface antigens may each perform a discrete function during host cell invasion, which is consistent with the differential localization observed for the SnSAGs (Fig. 6).
In addition to their potential role as adhesins, the SnSAGs may act as immunomodulatory molecules that influence the infection and the progression of disease. In the present study, two of the most immunodominant protein species seen in the Western blot analysis of S. neurona correspond to SnSAG1, SnSAG2, and SnSAG4 (Fig. 4). Similarly, many of the SAGs from T. gondii (12, 25, 56) and N. caninum (29, 30, 32) have been described as immunodominant antigens. As suggested for toxoplasmosis (4), the robust immune response collectively elicited by the multiple immunogenic SAGs may be critical for controlling the acute stage of disease, thereby preventing a rampant infection that results in death of the host. Additionally, most of the TgSAGs exhibit stage-specific expression during the T. gondii life cycle (42). As a consequence, the immunodominant TgSAGs present on the tachyzoite surface are absent from the latent cysts that form in the infected host animal. In effect, this provides a mechanism whereby T. gondii bradyzoites become incognito to effectors of the host immune response. It remains to be determined whether there is stage-specific expression of the SnSAGs during the life cycle of S. neurona, but given the prominent immunogenicity exhibited by the SnSAGs, it is plausible that these surface antigens are important for immune evasion by the parasite.
Due to their surface display, immunogenicity, and apparent omnipresence, the SAGs collectively represent a promising general target for intervention in coccidian diseases (e.g., as vaccine antigens). It is well established that TgSAG1 from T. gondii can protect against acute toxoplasmosis in mice (2, 3, 6, 11, 39). A number of studies have since investigated the development of protective immune responses with assorted other SAG family members from T. gondii (45) and N. caninum (50, 51). Logically, the SnSAGs may similarly prove to be useful as vaccine antigens that protect against EPM in horses. Furthermore, the immunogenicity displayed by many of the SAG family members makes these antigens appealing markers for use in serodiagnostic assays. In individuals infected with T. gondii, anti-TgSAG1 antibodies can be detected early in the infection, and these antibodies include IgM, IgA, and various IgG subisotypes (13, 52, 53). Likewise, the two major surface antigens of N. caninum, NcSAG1 and NcSRS2, are consistently recognized by antisera from Neospora-infected animals (32), and a recombinant form of NcSAG1 has been implemented in an enzyme-linked immunosorbent assay (35). These prior studies provide considerable precedence for the use of the SnSAGs as diagnostic molecules.
In summary, the existence of a family of paralogous SAGs in S. neurona merozoites indicates that expression of numerous, semiredundant surface antigens is a general theme among the heteroxenous (two-host) Coccidia. Although the purpose served by these surface antigen families remains to be elucidated, their conserved presence is presumably important for an essential function(s) that is common to this group of parasitic organisms. Further investigation of Sarcocystis spp. and their complements of surface molecules should allow comparisons with the much-better-characterized organism T. gondii, thereby providing greater insight into the importance of the SAG/SRS gene families as virulence factors.
ACKNOWLEDGMENTS
This research was supported by grants from the Amerman Family Foundation and Fort Dodge Animal Health and by gifts from various donors.
We gratefully acknowledge Anthony Sinai for helpful discussions and a critical review of the manuscript.
Kentucky Agricultural Experiment Station article no. 04-14-017.
REFERENCES
1. Altschul, S. F., W. Gish, W. Miller, E. W. Myers, and D. J. Lipman. 1990. Basic local alignment search tool. J. Mol. Biol. 215:403-410.
2. Angus, C. W., D. Klivington-Evans, J. P. Dubey, and J. A. Kovacs. 2000. Immunization with a DNA plasmid encoding the SAG1 (P30) protein of Toxoplasma gondii is immunogenic and protective in rodents. J. Infect. Dis. 181:317-324.
3. Bonenfant, C., I. Dimier-Poisson, F. Velge-Roussel, D. Buzoni-Gatel, G. Del Giudice, R. Rappuoli, and D. Bout. 2001. Intranasal immunization with SAG1 and nontoxic mutant heat-labile enterotoxins protects mice against Toxoplasma gondii. Infect. Immun. 69:1605-1612.
4. Boothroyd, J. C., A. Hehl, L. J. Knoll, and I. D. Manger. 1998. The surface of Toxoplasma: more and less. Int. J. Parasitol. 28:3-9.
5. Brusca, J. S., and J. D. Radolf. 1994. Isolation of integral membrane proteins by phase partitioning with Triton X-114. Methods Enzymol. 228:182-193.
6. Bulow, R., and J. C. Boothroyd. 1991. Protection of mice from fatal Toxoplasma infection by immunization with p30 antigen in liposomes. J. Immunol. 147:3496-3500.
7. Cesbron-Delauw, M. F., S. Tomavo, P. Beauchamps, M. P. Fourmaux, D. Camus, A. Capron, and J. F. Dubremetz. 1994. Similarities between the primary structures of two distinct major surface proteins of Toxoplasma gondii. J. Biol. Chem. 269:16217-16222.
8. Cheadle, M. A., S. M. Tanhauser, J. B. Dame, D. C. Sellon, M. Hines, P. E. Ginn, R. J. MacKay, and E. C. Greiner. 2001. The nine-banded armadillo (Dasypus novemcinctus) is an intermediate host for Sarcocystis neurona. Int. J. Parasitol. 31:330-335.
9. Cheadle, M. A., C. A. Yowell, D. C. Sellon, M. Hines, P. E. Ginn, A. E. Marsh, J. B. Dame, and E. C. Greiner. 2001. The striped skunk (Mephitis mephitis) is an intermediate host for Sarcocystis neurona. Int. J. Parasitol. 31:843-849.
10. Corpet, F. 1988. Multiple sequence alignment with hierarchical clustering. Nucleic Acids Res. 16:10881-10890.
11. Couper, K. N., H. V. Nielsen, E. Peterson, F. Roberts, C. W. Roberts, and J. Alexander. 2003. DNA vaccination with the immunodominant tachyzoite surface antigen (SAG-1) protects against adult acquired Toxoplasma gondii infection but does not prevent maternofoetal transmission. Vaccine 21:2813-2820.
12. Couvreur, G., A. Sadak, B. Fortier, and J. F. Dubremetz. 1988. Surface antigens of Toxoplasma gondii. Parasitology 97:1-10.
13. Decoster, A., A. Caron, F. Darcy, and A. Capron. 1988. IgA antibodies against p30 as markers of congenital and acute toxoplasmosis. Lancet ii:1104-1106.
14. Devereux, J., P. Haeberli, and O. Smithies. 1984. A comprehensive set of sequence analysis programs for the VAX. Nucleic Acids Res. 12:387-395.
15. Dubey, J. P., D. S. Lindsay, W. J. Saville, S. M. Reed, D. E. Granstrom, and C. A. Speer. 2001. A review of Sarcocystis neurona and equine protozoal myeloencephalitis (EPM). Vet. Parasitol. 95:89-131.
16. Dubey, J. P., W. J. Saville, D. S. Lindsay, R. W. Stich, J. F. Stanek, C. A. Speer, B. M. Rosenthal, C. J. Njoku, O. C. Kwok, S. K. Shen, and S. M. Reed. 2000. Completion of the life cycle of Sarcocystis neurona. J. Parasitol. 86:1276-1280.
17. Dubey, J. P., W. J. Saville, J. F. Stanek, D. S. Lindsay, B. M. Rosenthal, M. J. Oglesbee, A. C. Rosypal, C. J. Njoku, R. W. Stich, O. C. Kwok, S. K. Shen, A. N. Hamir, and S. M. Reed. 2001. Sarcocystis neurona infections in raccoons (Procyon lotor): evidence for natural infection with sarcocysts, transmission of infection to opossums (Didelphis virginiana), and experimental induction of neurologic disease in raccoons. Vet. Parasitol. 100:117-129.
18. Ellison, S. P., A. L. Omara-Opyene, C. A. Yowell, A. E. Marsh, and J. B. Dame. 2002. Molecular characterisation of a major 29 kDa surface antigen of Sarcocystis neurona. Int. J. Parasitol. 32:217-225.
19. Eschenbacher, K. H., I. Sommer, H. E. Meyer, H. Mehlhorn, and W. Ruger. 1992. Cloning and expression in Escherichia coli of cDNAs encoding a 31-kilodalton surface antigen of Sarcocystis muris. Mol. Biochem. Parasitol. 53:159-167.
20. Fenger, C. K., D. E. Granstrom, J. L. Langemeier, S. Stamper, J. M. Donahue, J. S. Patterson, A. A. Gajadhar, J. V. Marteniuk, Z. Xiaomin, and J. P. Dubey. 1995. Identification of opossums (Didelphis virginiana) as the putative definitive host of Sarcocystis neurona. J. Parasitol. 81:916-919.
21. Granstrom, D. E., O. Alvarez, Jr., J. P. Dubey, P. F. Comer, and N. M. Williams. 1992. Equine protozoal myelitis in Panamanian horses and isolation of Sarcocystis neurona. J. Parasitol. 78:909-912.
22. Grimwood, J., and J. E. Smith. 1992. Toxoplasma gondii: the role of a 30-kDa surface protein in host cell invasion. Exp. Parasitol. 74:106-111.
23. 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.
24. Gubbels, M. J., M. Wieffer, and B. Striepen. 2004. Fluorescent protein tagging in Toxoplasma gondii: identification of a novel inner membrane complex component conserved among Apicomplexa. Mol. Biochem. Parasitol. 137:99-110.
25. Handman, E., J. W. Goding, and J. S. Remington. 1980. Detection and characterization of membrane antigens of Toxoplasma gondii. J. Immunol. 124:2578-2583.
26. Hannah, M. J., U. Weiss, and W. B. Huttner. 1998. Differential extraction of proteins from paraformaldehyde-fixed cells: lessons from synatophysin and other membrane proteins. Methods Companion Methods Enzymol. 16:170-181.
27. He, X. L., M. E. Grigg, J. C. Boothroyd, and K. C. Garcia. 2002. Structure of the immunodominant surface antigen from the Toxoplasma gondii SRS superfamily. Nat. Struct. Biol. 9:606-611.
28. Hemphill, A. 1996. Subcellular localization and functional characterization of Nc-p43. Infect. Immun. 64:4279-4287.
29. Hemphill, A., N. Fuchs, S. Sonda, B. Gottstein, and B. Hentrich. 1997. Identification and partial characterization of a 36 kDa surface protein on Neospora caninum tachyzoites. Parasitology 115:371-380.
30. Hemphill, A., and B. Gottstein. 1996. Identification of a major surface protein on Neospora caninum tachyzoites. Parasitol. Res. 82:497-504.
31. Howe, D. K. 2001. Initiation of a Sarcocystis neurona expressed sequence tag (EST) sequencing project: a preliminary report. Vet. Parasitol. 95:233-239.
32. Howe, D. K., A. C. Crawford, D. Lindsay, and L. D. Sibley. 1998. The p29 and p35 immunodominant antigens of Neospora caninum tachyzoites are homologous to the family of surface antigens of Toxoplasma gondii. Infect. Immun. 66:5322-5328.
33. Howe, D. K., and L. D. Sibley. 1997. Development of molecular genetics for Neospora caninum: a complementary system to Toxoplasma gondii. Methods Companion Methods Enzymol. 13:123-133.
34. Howe, D. K., and L. D. Sibley. 1999. Comparison of the major antigens of Neospora caninum and Toxoplasma gondii. Int. J. Parasitol. 29:1489-1496.
35. Howe, D. K., K. Tang, P. A. Conrad, K. Sverlow, J. P. Dubey, and L. D. Sibley. 2002. Sensitive and specific identification of Neospora caninum infection of cattle based on detection of serum antibodies to recombinant Ncp29. Clin. Diagn. Lab. Immunol. 9:611-615.
36. Hu, K., T. Mann, B. Striepen, C. J. Beckers, D. S. Roos, and J. M. Murray. 2002. Daughter cell assembly in the protozoan parasite Toxoplasma gondii. Mol. Biol. Cell 13:593-606.
37. Jacquet, A., L. Coulon, J. De Neve, V. Daminet, M. Haumont, L. Garcia, A. Bollen, M. Jurado, and R. Biemans. 2001. The surface antigen SAG3 mediates the attachment of Toxoplasma gondii to cell-surface proteoglycans. Mol. Biochem. Parasitol. 116:35-44.
38. Jung, C., C. Y. F. Lee, and M. E. Grigg. 2004. The SRS superfamily of Toxoplasma surface proteins. Int. J. Parasitol. 34:285-296.
39. Khan, I. A., K. H. Ely, and L. H. Kasper. 1991. A purified parasite antigen (p30) mediates CD8+ T cell immunity against fatal Toxoplasma gondii infection in mice. J. Immunol. 147:3501-3506.
40. Laemmli, U. K. 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature (London) 227:680-685.
41. Lekutis, C., D. J. Ferguson, and J. C. Boothroyd. 2000. Toxoplasma gondii: identification of a developmentally regulated family of genes related to SAG2. Exp. Parasitol. 96:89-96.
42. Lekutis, C., D. J. Ferguson, M. E. Grigg, M. Camps, and J. C. Boothroyd. 2001. Surface antigens of Toxoplasma gondii: variations on a theme. Int. J. Parasitol. 31:1285-1292.
43. Li, L., B. P. Brunk, J. C. Kissinger, D. Pape, K. Tang, R. H. Cole, J. Martin, T. Wylie, M. Dante, S. J. Fogarty, D. K. Howe, P. Liberator, C. Diaz, J. Anderson, M. White, M. E. Jerome, E. A. Johnson, J. A. Radke, C. J. Stoeckert, Jr., R. H. Waterston, S. W. Clifton, D. S. Roos, and L. D. Sibley. 2003. Gene discovery in the Apicomplexa as revealed by EST sequencing and assembly of a comparative gene database. Genome Res. 13:443-454.
44. Liang, F. T., D. E. Granstrom, X. M. Zhao, and J. F. Timoney. 1998. Evidence that surface proteins Sn14 and Sn16 of Sarcocystis neurona merozoites are involved in infection and immunity. Infect. Immun. 66:1834-1838.
45. Lunden, A., S. F. Parmley, K. L. Bengtsson, and F. G. Araujo. 1997. Use of a recombinant antigen, SAG2, expressed as a glutathione-S-transferase fusion protein to immunize mice against Toxoplasma gondii. Parasitol. Res. 83:6-9.
46. MacKay, R. J., D. E. Granstrom, W. J. Saville, and S. M. Reed. 2000. Equine protozoal myeloencephalitis. Vet. Clin. N. Am. Equine Pract. 16:405-425.
47. Mineo, J. R., and L. H. Kasper. 1994. Attachment of Toxoplasma gondii to host cells involves major surface protein SAG-1 (P30). Exp. Parasitol. 79:11-20.
48. Mineo, J. R., R. McLeod, D. Mack, J. Smith, I. A. Khan, K. H. Ely, and L. H. Kasper. 1993. Antibodies to Toxoplasma gondii major surface protein (SAG-1, P30) inhibit infection of host cells and are produced in murine intestine after peroral infection. J. Immunol. 150:3951-3964.
49. Nagel, S. D., and J. C. Boothroyd. 1989. The major surface antigen, p30, of Toxoplasma gondii is anchored by a glycolipid. J. Biol. Chem. 264:5569-5574.
50. Nishikawa, Y., N. Inoue, X. Xuan, H. Nagasawa, I. Igarashi, K. Fujisaki, H. Otsuka, and T. Mikami. 2001. Protective efficacy of vaccination by recombinant vaccinia virus against Neospora caninum infection. Vaccine 19:1381-1390.
51. Nishikawa, Y., X. Xuan, H. Nagasawa, I. Igarashi, K. Fujisaki, H. Otsuka, and T. Mikami. 2001. Prevention of vertical transmission of Neospora caninum in BALB/c mice by recombinant vaccinia virus carrying NcSRS2 gene. Vaccine 19:1710-1716.
52. Potasman, I., F. G. Araujo, G. Desmonts, and J. S. Remington. 1986. Analysis of Toxoplasma gondii antigens recognized by human sera obtained before and after acute infection. J. Infect. Dis. 154:650-657.
53. Santoro, F., D. Afchain, R. Pierce, J. Y. Cesbron, G. Ovalaque, and A. Capron. 1985. Serodiagnosis of Toxoplasma infection using a purified parasite protein (p30). Clin. Exp. Immunol. 62:262-269.
54. Schares, G., J. F. Dubremetz, J. P. Dubey, A. Barwald, A. Loyens, and F. J. Conraths. 1999. Neospora caninum: identification of 19-, 38-, and 40-kDa surface antigens and a 33-kDa dense granule antigen using monoclonal antibodies. Exp. Parasitol. 92:109-119.
55. Schares, G., C. F. Zinecker, J. Schmidt, N. Azzouz, F. J. Conraths, P. Gerold, and R. T. Schwarz. 2000. Structural analysis of free and protein-bound glycosyl-phosphatidylinositols of Neospora caninum. Mol. Biochem. Parasitol. 105:155-161.
56. Sharma, S. D., J. Mullenax, F. G. Araujo, H. A. Ehrlich, and J. S. Remington. 1983. Western blot analysis of the antigens of Toxoplasma gondii recognized by human IgM and IgG antibodies. J. Immunol. 131:977-983.
57. Sibley, L. D., E. R. Pfefferkorn, and J. C. Boothroyd. 1991. Proposal for a uniform genetic nomenclature in Toxoplasma gondii. Parasitol. Today 7:327-328.
58. Speer, C. A., and J. P. Dubey. 2001. Ultrastructure of schizonts and merozoites of Sarcocystis neurona. Vet. Parasitol. 95:263-271.
59. Thompson, J. D., D. G. Higgins, and T. J. Gibson. 1994. CLUSTAL W: improving the sensitivity of progressive multiple sequence alignments through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res. 22:4673-4680.
60. Tomavo, S., R. T. Scwartz, and J. F. Dubremetz. 1989. Evidence for glycosyl-phosphatidylinositol anchoring of Toxoplasma gondii major surface antigens. Mol. Cell. Biol. 9:4576-4580.(Daniel K. Howe, Rajshekha)
Center for Tropical and Emerging Global Diseases and Department of Cellular Biology, University of Georgia, Athens, Georgia
ABSTRACT
Sarcocystis neurona is a member of the Apicomplexa that causes myelitis and encephalitis in horses but normally cycles between the opossum and small mammals. Analysis of an S. neurona expressed sequence tag (EST) database revealed four paralogous proteins that exhibit clear homology to the family of surface antigens (SAGs) and SAG-related sequences of Toxoplasma gondii. The primary peptide sequences of the S. neurona proteins are consistent with the two-domain structure that has been described for the T. gondii SAGs, and each was predicted to have an amino-terminal signal peptide and a carboxyl-terminal glycolipid anchor addition site, suggesting surface localization. All four proteins were confirmed to be membrane associated and displayed on the surface of S. neurona merozoites. Due to their surface localization and homology to T. gondii surface antigens, these S. neurona proteins were designated SnSAG1, SnSAG2, SnSAG3, and SnSAG4. Consistent with their homology, the SnSAGs elicited a robust immune response in infected and immunized animals, and their conserved structure further suggests that the SnSAGs similarly serve as adhesins for attachment to host cells. Whether the S. neurona SAG family is as extensive as the T. gondii SAG family remains unresolved, but it is probable that additional SnSAGs will be revealed as more S. neurona ESTs are generated. The existence of an SnSAG family in S. neurona indicates that expression of multiple related surface antigens is not unique to the ubiquitous organism T. gondii. Instead, the SAG gene family is a common trait that presumably has an essential, conserved function(s).
INTRODUCTION
Sarcocystis neurona is an apicomplexan parasite and the primary cause of equine protozoal myeloencephalitis (EPM). EPM is a serious debilitating disease and is the most commonly diagnosed neurologic disorder of horses. Although seroprevalence studies have indicated that approximately 30 to 50% of horses have been exposed to S. neurona, the incidence of EPM is estimated to be less than 1% (15, 46). Thus, S. neurona infection clearly does not equate directly with clinical illness, and it is not clear what factors influence the progression from simple infection to severe neurologic disease. The normal life cycle of S. neurona alternates between the definitive host, the opossum (20), and various small mammal intermediate hosts, including skunks (9), raccoons (17), armadillos (8), and cats (16). Although S. neurona readily infects equids, these animals are currently believed to be aberrant hosts for this parasite species since latent forms (sarcocysts) have not been found in infected horses.
Like other members of the Apicomplexa, S. neurona is an obligate intracellular parasite that requires a number of unique molecules (i.e., virulence factors) to support its parasitic lifestyle. Apicomplexan surface molecules are important virulence factors that are responsible for the pathogen's initial interactions with the host cell surface and components of the host immune response. A broad family of more than 20 related surface antigens has been found to be expressed by Toxoplasma gondii (42). A recent bioinformatic search of the Toxoplasma genome database (www.toxodb.org) indicated that the full assemblage of surface antigen genes is even more extensive, and 161 related sequences have been identified in the genome of the ME49 strain (38). Although many of these sequences may be pseudogenes that are not expressed, it is apparent that T. gondii has the capacity to produce a complex array of surface antigens. These paralogous molecules, which have been designated SAGs and SAG-related sequences (SRSs), are developmentally regulated and exhibit various levels of sequence similarity to one of the major T. gondii surface antigens, TgSAG1 or TgSAG2. The T. gondii SAGs appear to be involved in receptor-ligand interactions with the host cell surface, most likely through binding of sulfated proteoglycans (27, 37), and there is increasing evidence that some of the SAGs can modulate host immune responses (42). The evolutionary advantage provided by expansion of the SAG/SRS gene family is unknown, but it has been speculated that the comprehensive array of surface antigens allows the very wide host range of T. gondii (4, 42). Individual SAG homologues have been described in the genus Sarcocystis (18, 19). However, it was not clear whether Sarcocystis spp. also express a complex family of related surface antigens.
In an effort to identify and characterize virulence factors of S. neurona, we have conducted a sequencing project that has generated approximately 8,500 expressed sequence tags (ESTs) from this organism (31, 43). Examination of this sequence database revealed a family of S. neurona surface antigens that are orthologues of the SAG/SRS family of surface proteins in T. gondii. Here, we present the results of characterization of the four S. neurona surface antigens identified thus far. Based on their clear homology to T. gondii SAGs, the S. neurona surface antigens have been designated SnSAG1, SnSAG2, SnSAG3, and SnSAG4.
MATERIALS AND METHODS
Parasite cultures. S. neurona strain SN3 (21) merozoites were propagated by serial passage in bovine turbinate cells and were maintained in RPMI 1640 medium supplemented with 10% fetal bovine serum, 2 mM sodium pyruvate, and Pen/Strep Fungizone (BioWhittaker, Inc.). Extracellular merozoites were harvested and purified from disrupted host cell monolayers by filtration through 3.0-μm-pore-size membranes, as described previously for Neospora caninum (33).
Immunoscreening of S. neurona cDNA library. Construction and analyses of the cSn.1 S. neurona merozoite cDNA library have been described previously (31). Phage plaques were allowed to form for 3 h at 42°C on Escherichia coli XL1-Blue MRF' host cells (Stratagene) grown on 150-mm NZY agar plates. When plaques became visible, the agar was overlaid with nitrocellulose filters previously soaked in 10 mM isopropyl--D-thiogalactopyranoside (IPTG), and the plates were incubated for an additional 3 h at 37°C. The filters were lifted from the plates, washed with TNT buffer (10 mM Tris-HCl [pH 7.5], 150 mM NaCl, 0.05% Tween 20), and blocked in phosphate-buffered saline (PBS)-5% dry milk-5% normal goat serum-0.05% Tween 20.
Antigenic cDNA clones were identified by screening the filters with cerebrospinal fluid (CSF) from a horse that had been naturally infected with S. neurona and exhibited a high titer of intrathecal antibodies against S. neurona in a Western blot analysis. Prior to screening of the S. neurona cDNA library, the CSF was depleted of antibodies that were reactive with E. coli and phage proteins. Briefly, the CSF was diluted 1:20 in PBS-0.1% dry milk-0.1% normal goat serum-0.05% Tween 20 and incubated for 30 min with filters carrying plaque lifts of a previously described N. caninum cDNA library (34). After adsorption of potential cross-reactive antibodies, the diluted CSF solution was incubated for 1 h with the cSn.1 filters. The filters were washed and then incubated for 1 h with goat anti-equine immunoglobulin G (IgG) conjugated to horseradish peroxidase (Jackson Immunoresearch Labs, Inc.) diluted 1:10,000. Immunoreactive phage plaques were picked with sterile pipette tips and suspended in 40 μl of SM buffer (50 mM Tris-HCl [pH 7.5], 100 mM NaCl, 8 mM MgSO4, 0.01% gelatin). The cDNA inserts of the selected phage plaques were PCR amplified by using the T3 and T7 oligonucleotide primers, and the resulting products were analyzed by agarose gel electrophoresis. Sequencing reactions with the T3 primer were conducted for the amplified cDNAs to provide preliminary identification of the immunoreactive clones. Phagemid excision was performed with selected cDNA clones, and plasmids were rescued in SOLR cells by using the manufacturer's protocol (Stratagene).
Immunoaffinity purification of antibodies. An immunoreactive phage clone was plated at a high density on two NZY agar plates and overlaid with IPTG-soaked nitrocellulose membranes. After 4 h of incubation at 37°C, the membranes were lifted from the plates and washed twice with TNT buffer. The membranes were blocked for 15 min and then incubated overnight at 4°C in CSF from an S. neurona-infected horse. After three washes, the two membranes were incubated sequentially for 15 min in 10 ml of 0.2 M glycine (pH 2.8) to elute the bound antibodies. One milliliter of 1 M Tris (pH 8.0) was added to neutralize the solution, followed by 1 ml of 10x PBS. The eluted antibodies were used for Western blot analysis, as described below.
S. neurona EST database searches and sequence analyses. Genes for surface antigen homologues were identified in the S. neurona clustered EST database (http://paradb.cis.upenn.edu/sarco/index.html) by using the BLAST (basic local alignment search tool) set of programs (1). At the time that the searches were conducted, the database contained 686 consensus sequences that had been generated from 1,883 S. neurona ESTs. Selected cDNAs were obtained from the archived collection of EST clones and sequenced by using a ABI Prism BigDye terminator cycle sequencing reaction mixture (Perkin-Elmer Applied Biosystems). The reaction mixtures were purified by using Centri-Sep spin columns (Princeton Separations), and the eluted extension products were resolved and analyzed with an ABI 310 genetic analyzer. Sequence analyses were conducted with Genetics Computer Group software (14) and programs available at the National Center for Biotechnology Information website (http://www.ncbi.nlm.nih.gov/) and the Expert Protein Analysis System (ExPASy) server of the Swiss Institute of Bioinformatics (http://www.expasy.ch/). Multiple-sequence alignments of the putative mature forms of the proteins (without the signal peptide and carboxyl-terminal tails) were constructed by using Multalin software (10). The individual SAG domains D1 and D2, as defined previously (27) and demarcated by the conserved cysteine residues within each domain, were aligned, and a phylogenetic dendrogram was constructed by using ClustalW (59).
Recombinant antigen expression and generation of polyclonal antisera. The predicted SnSAG open reading frames, without their predicted amino-terminal signal peptides and carboxyl-terminal hydrophobic tails, were amplified by PCR by using primers that incorporate a 5' end NdeI restriction site and a 3' end XhoI restriction site. The amplification products were digested with NdeI and XhoI and ligated into NdeI/XhoI-digested expression vector pET22b (Novagen), and the constructs were transformed into E. coli INVF' for plasmid propagation. The resulting expression plasmids were transformed into E. coli BL21-CodonPlus (Stratagene), and clones that expressed high levels of recombinant protein were selected for use. The histidine-tagged recombinant SnSAGs (rSnSAG1, rSnSAG2, rSnSAG3, and rSnSAG4) were purified by nickel-column chromatography by using the manufacturer's protocol (Novagen), and monospecific polyclonal antisera against the purified proteins were produced by immunization of rabbits and rats (Cocalico Biologicals, Inc.).
Western blot analysis. Parasites were lysed in sodium dodecyl sulfate sample buffer supplemented with a protease inhibitor cocktail (Sigma) and 2% 2-mercaptoethanol (2-ME), and the lysates were separated in 10 or 12% polyacrylamide gels (40). Dithiothreitol-reduced low-range molecular mass standards (Bio-Rad Laboratories, Inc.) were included in all Western blot analyses. Proteins were transferred to nitrocellulose membranes by semidry electrophoretic transfer in Tris-glycine buffer (pH 8.3). The membranes were blocked with PBS containing 5% nonfat dry milk, 5% goat serum, and 0.05% Tween 20 and then incubated for 1 h with primary antibody. After washing, the membranes were incubated with horseradish peroxidase-conjugated immunoglobulin G secondary antibody (Jackson Immunoresearch Labs, Inc.). The blots were washed, processed for chemiluminescence with the Supersignal substrate (Pierce Chemical Company), and exposed to film or visualized with a FluorChem 8800 imaging system (Alpha Innotech, Inc.).
Biotinylation of surface proteins and precipitation with immobilized streptavidin. Approximately 3 x 107 freshly harvested merozoites were resuspended in 1 ml of cold PBS (pH 7.8). Sulfo-N-hydroxy-succinimide-biotin (Pierce) was added to a concentration of 0.5 mg/ml and incubated at room temperature for 30 min. The labeled parasites were washed twice with 5 ml of PBS and stored at –20°C. The labeled parasite pellet was lysed with 1 ml of radioimmunoprecipitation assay buffer (50 mM Tris [pH 7.5], 1% Triton X-100, 0.5% sodium deoxycholate, 0.2% sodium dodecyl sulfate, 100 mM NaCl, 5 mM EDTA) supplemented with RNase, DNase, and a protease inhibitor cocktail, and the sample was centrifuged at 16,000 x g to remove the insoluble fraction. Biotin-labeled proteins were precipitated from the soluble fraction with UltraLink immobilized streptavidin (Pierce), and the precipitated proteins were analyzed by Western blotting, as described above.
Indirect immunofluorescent labeling of intracellular parasites. For detection of the SnSAGs on intracellular parasites, merozoites were inoculated onto bovine turbinate cells grown in LabTek chamber slides (Nunc Nalgene) or onto 12-mm round coverslips in 24-well plates. At 24, 48, or 72 h postinoculation, the cells were fixed either in –20°C methanol for 10 min or in 2.5% formalin-PBS-0.01% glutaraldehyde for 30 min on ice. The formalin-fixed cells were permeabilized with 0.2% Triton X-100. The slides were blocked by 30 min of incubation with 10% normal goat serum in PBS. The primary antibodies were rabbit antisera against SnSAG1, SnSAG2, SnSAG3, or SnSAG4 or rat antisera against SnSAG1, SnSAG2, or SnSAG3. A rabbit antiserum against T. gondii IMC3 (24) was used to visualize the inner membrane complex with the SnSAGs in methanol-fixed cells. After incubation with the primary antibodies, the slides were rinsed and then incubated with goat anti-rabbit IgG or goat anti-rat IgG conjugated to either fluorescein isothiocyanate (FITC) or Texas Red (Jackson Immunoresearch Labs, Inc.). The slides were mounted in Vectashield with 4',6'-diamidino-2-phenylindole (DAPI) (Vector Laboratories, Inc.) and examined with a Zeiss axioscope or a Leica DM IRBE equipped for epifluorescence microscopy.
Nucleotide sequence accession numbers. The sequences reported here have been deposited in the GenBank database under accession numbers AY032845, AY191006, AY191007, and AY191008.
RESULTS
Isolation and sequence analysis of immunoreactive cDNA clones. Primary screening of the cSn.1 cDNA library resulted in identification of multiple immunoreactive phage plaques, and 25 plaques were selected and resuspended in SM buffer. Amplification of the cDNA inserts with the T3 and T7 oligonucleotides revealed that 22 of the phage clones had similar lengths (approximately 1,500 bp), and sequence analysis with the T3 primer indicated that these 22 cDNAs represented the same gene. A secondary immunoscreening was performed for five of the selected cDNAs, and two highly reactive phage clones, designated SnAgI.8 and SnAgI.9, were chosen for further analyses.
To obtain preliminary identification of the parasite protein encoded by the selected cDNAs, the SnAgI.9 clone was used to affinity purify antibodies that bind the antigen expressed by this clone, and the eluted antibodies were used to probe a Western blot of S. neurona merozoite lysate. The antigen revealed by the phage-purified antibodies migrated at approximately 31 kDa under reducing conditions, and it comigrated with a protein that is recognized by equine or rabbit antisera against S. neurona as a major immunodominant antigen of this parasite (data not shown). This result implied that the 22 matching cDNA clones isolated during the library screening and represented by SnAgI.8 and SnAgI.9 encode an immunodominant antigen of S. neurona.
Full-length sequence analysis of SnAgI.8 revealed a 1,493-nucleotide cDNA insert with an open reading frame (ORF) that encodes a 276-amino-acid protein. Sequence analysis of SnAgI.9 indicated that this clone was virtually identical to SnAgI.8, although its 3' untranslated region extended an additional 160 nucleotides. Although this may have been due to alternative polyadenylation, mispriming by oligo(dT) during cDNA synthesis cannot be ruled out. A hydrophobicity plot of the encoded protein showed that there were hydrophobic domains at both termini, which corresponded to a predicted signal peptide at the amino terminus and a glycosylphosphatidylinositol (GPI) anchor addition sequence at the carboxyl terminus (data not shown). The signal peptide cleavage was predicted to occur at Ala15-Arg16 (SignalP), and the most likely GPI transamidase cleavage site was predicted to be at Ala247-Asn248 (DGPI; Swiss Institute of Bioinformatics). A single N-glycosylation site was predicted at residues 140 to 143. Removal of the N-terminal and C-terminal signal sequences resulted in a mature protein consisting of 232 amino acids that had a predicted molecular mass of 24.2 kDa, not including any potential posttranslational modifications (e.g., glycosylation).
A blastx search (1) of the nonredundant GenBank databases was conducted with the SnAgI.8 coding sequence as the query. This search revealed statistically significant similarity to the 31-kDa major surface antigen of Sarcocystis muris SmMSA (19) and a less significant but recognizable similarity to several SAG2-related surface antigens of T. gondii (41). In conjunction with the Western blot analysis and the predictions of a signal peptide and a GPI anchor addition, these results collectively suggested that the SnAgI.8 and SnAgI.9 cDNAs represent a gene that encodes an immunodominant surface antigen of S. neurona. Consequently, this protein was designated SnSAG1 by using the genetic nomenclature that is utilized for the related apicomplexan parasites T. gondii and N. caninum (34, 57).
A blastn search of the S. neurona ESTs (conducted on 4 October 2002) with the SnAgI.8 nucleotide sequence as the query resulted in identification of 654 overlapping ESTs (7.7% of the 8,482 S. neurona ESTs) that correspond to the SnSAG1 gene. This result indicated that the SnSAG1 sequence is extensively represented in the cDNA library, and it further suggested that the SnSAG1 surface antigen is expressed at high levels by S. neurona merozoites.
Identification of additional S. neurona surface antigens. To determine if other SnSAG surface antigens are expressed by S. neurona merozoites, the database of S. neurona EST consensus sequences (http://paradb.cis.upenn.edu/sarco/index.html) was searched with the BLAST tool tblastn (1) by using SnSAG1 and the various available T. gondii SAGs and SRSs as query sequences. At the time that the searches were conducted, the dbEST database contained 1,883 S. neurona ESTs, which had been clustered into 686 consensus gene sequences. These searches resulted in identification of gene sequences that appeared to encode three additional SnSAG paralogues, which were tentatively designated SnSAG2, SnSAG3, and SnSAG4. A representative cDNA clone for each of the three novel SnSAGs was recovered from the cryopreserved stocks of S. neurona EST clones and used for further analysis.
SnSAG2 was identified from the S. neurona clustered EST database by using T. gondii SAG1 as the query sequence. The SnSAG2 gene was initially represented by a cluster of 52 ESTs, which provided a 961-nucleotide consensus sequence that spanned the entire ORF of the gene. For further analysis and manipulation of SnSAG2, cDNA SnEST4a02h07 was recovered from the archived collection of S. neurona EST clones. The SnSAG2 gene encodes a 168-amino-acid protein that has a predicted amino-terminal signal peptide and a carboxyl-terminal signal for addition of a GPI anchor; the sequence does not contain any potential N-linked glycosylation sites. After removal of the signal peptide (28 residues) and the carboxyl-terminal tail (approximately 23 residues), a molecular mass of 11.8 kDa was predicted for the mature SnSAG2 protein. A blastn search of the current database of 8,482 S. neurona ESTs revealed 148 entries (approximately 1.7%) that correspond to the SnSAG2 sequence, indicating that this gene is well represented in the cSn.1 cDNA library and is likely expressed at high levels by S. neurona merozoites.
Both SnSAG3 and SnSAG4 were discovered by searching the S. neurona ESTs with the SnSAG1 protein sequence as the query. The SnSAG3 gene was initially identified as a cluster of three overlapping ESTs, from which clone SnEST4a18g06 was selected and recovered for further study. Full-length sequence analysis of this cDNA clone revealed a 1,496-nucleotide insert with an ORF that encodes a 281-amino-acid peptide. The derived SnSAG3 protein sequence was predicted to contain a 21-residue signal peptide and a 34-amino-acid carboxyl-terminal tail for GPI anchor addition. Removal of the signal peptide and carboxyl-terminal tail resulted in a mature protein consisting of 226 residues with a molecular mass of 23.1 kDa. A blastn search of the current database with the full-length nucleotide sequence of clone SnEST4a18g06 as the query resulted in identification of 19 S. neurona ESTs that correspond to SnSAG3.
The SnSAG4 gene was initially identified as a singleton cDNA clone, SnEST4a07g08. Full-length sequence analysis of this clone revealed a 1,093-nucleotide cDNA insert with an ORF encoding 287 amino acids. The SnSAG4 protein has a predicted 28-residue signal peptide and a 27-residue carboxyl-terminal tail for GPI anchor addition. Removal of these signals resulted in a mature protein that is 232 amino acids long and has a molecular mass of 24.4 kDa. A blastn search of the S. neurona EST database resulted in identification of four ESTs that correspond to the SnSAG4 gene, suggesting that this locus is transcribed at a relatively low level.
Comparison of SnSAGs with surface antigen orthologues. To assess the extent of sequence conservation in the S. neurona proteins, the SnSAGs were aligned with surface antigens that have been described in other members of the Coccidia. These analyses demonstrated that SnSAG1, SnSAG3, and SnSAG4 are most similar to one another, but they also exhibited clear homology to the SmMSA major surface antigen of S. muris and the TgSAG2 family of surface antigens from T. gondii (Fig. 1A). The SnSAG2 sequence was determined to be most similar to the T. gondii SAG1 family of surface antigens (Fig. 1B). Pairwise alignments indicated that SnSAG1 exhibits approximately 36% sequence identity with either SnSAG3 or SnSAG4, whereas SnSAG3 and SnSAG4 exhibit about 33% sequence identity. SnSAG1, SnSAG3, and SnSAG4 were found to exhibit approximately 30% sequence identity with SmMSA, and they generally exhibited less than 25% sequence identity with the TgSAG2 family members. Similar to SnSAG1, SnSAG3, and SnSAG4, the overall sequence identity exhibited by SnSAG2 with any one of the previously described TgSAG/SRS proteins was found to be quite modest (<25%). Despite the moderate sequence similarities that were observed, the SnSAGs were clearly homologues of the T. gondii surface antigens. This was primarily revealed by the presence of multiple conserved cysteine residues (Fig. 1), which have been described previously in the TgSAGs (7) and have been shown to participate in the formation of intramolecular disulfide bonds (27).
To further define the similarity between the SnSAG sequences and the T. gondii SAG families, the individual domains (D1 and D2) of the SnSAGs and the surface antigen prototypes TgSAG1 and TgSAG2A were aligned by using ClustalW, and a dendrogram was constructed from the resulting analysis. In the derived tree, the single domain of TgSAG2A clustered with the D2 and D1 domains of SnSAG1, SnSAG3, and SnSAG4 (Fig. 2). The D1 and D2 domains of TgSAG1 formed a second clade with the single domain of SnSAG2. TgSAG2A appeared to be most similar to the D2 domain of the SnSAGs, whereas SnSAG2 was most closely allied with the D1 domain of TgSAG1. These sequence comparisons suggest that the SAG1 and SAG2 gene family dichotomy that has been described in T. gondii is also present in the SnSAGs of S. neurona merozoites.
Although the S. neurona surface antigen genes were readily identified as orthologues of the SAG/SRS gene family in T. gondii, a clear line of phylogenetic descent could not be established for each SnSAG gene. Consequently, the basis for naming SnSAG1 through SnSAG4 was provided by their chronology of identification and their representation (i.e., abundance) in the S. neurona EST database. Thus, the numerical designations assigned to the SnSAGs do not imply direct homology to the TgSAGs with the corresponding numbers.
Characterization of the SnSAG gene products. In order to produce specific reagents that could be utilized for investigating the S. neurona surface antigens, each of the SnSAG genes was expressed as a His-tagged recombinant protein in E. coli. The expressed proteins were purified by nickel-column chromatography and used to immunize rabbits and rats for the production of monospecific polyclonal antibodies. Western blot analysis of 2-ME-reduced S. neurona merozoite lysate with the resulting SnSAG antisera revealed proteins that migrated at approximately 32, 20, 28, and 32 kDa for SnSAG1, SnSAG2, SnSAG3, and SnSAG4, respectively (Fig. 3A). When nonreduced lysate was analyzed, SnSAG1 to SnSAG4 migrated at approximately 26, 16, 25, and 26 kDa (Fig. 3B); these findings are reasonably consistent with the molecular mass predicted for each protein. Notably, the comigrating SnSAG1 and SnSAG4 proteins and the lower-molecular-mass protein SnSAG2 corresponded to the two most dominant bands typically recognized by antisera from infected animals (Fig. 3).
Homology to the TgSAGs indicated that the SnSAGs were likely surface antigens of S. neurona. To establish that the SnSAGs are membrane associated, we performed Triton X-114 phase partitioning with S. neurona merozoites, which separates proteins based on their amphiphilicity (5). Western blot analysis of the partitioned proteins demonstrated that all four SnSAGs were separated exclusively into the detergent phase (Fig. 4), signifying that these proteins are associated with membranes of S. neurona merozoites. To verify that the SnSAGs are displayed on the merozoite surface, extracellular parasites were labeled with sulfo-N-hydroxysuccinimide-biotin, which does not permeate membranes and selectively labels surface-exposed proteins. The biotin-labeled proteins were specifically precipitated with immobilized streptavidin, and Western blot analysis with the anti-SnSAG antisera revealed that all four SnSAGs were in the fraction of biotinylated proteins (Fig. 5), thus confirming that they are displayed on the surface of merozoites.
Presence and distribution during intracellular development. Intracellular S. neurona undergoes a complex process of cell division termed endopolygeny. This process uncouples genome replication from cytokinesis, thereby producing a polyploid nucleus in a large mother cell called a schizont. The nuclear material eventually divides concurrent with cytokinesis, resulting in the formation of multiple daughter merozoites. To assess the expression of the SnSAGs during endopolygeny and to determine the distribution of the surface antigens in the maturing schizonts, intracellular parasites were analyzed by indirect immunofluorescence. This analysis revealed that the four SnSAGs are present at all stages of intracellular development, although the pattern of labeling was somewhat different for each surface antigen (Fig. 6). In general, methanol fixation of cells yielded a uniform distribution of label for the four SnSAGs throughout intracellular development, with punctations of label appearing in the late stages of endopolygeny. For SnSAG2 (Fig. 6C) and SnSAG3 (data not shown), these punctations of label appeared to define the newly forming daughter merozoites that were randomly dispersed throughout the schizont. In the mature schizonts, SnSAG1 (Fig. 6A), SnSAG2 (data not shown), and SnSAG4 (data not shown) were fairly evenly distributed on the individual merozoites, as expected. The exception was SnSAG3, which exhibited distinct labeling at the apical end of the merozoites (Fig. 6D). The distributions of the SnSAGs were similar for extracellular merozoites (data not shown).
Notably, when the cells had been fixed with formaldehyde and permeabilized with Triton X-100, antibodies to SnSAG1 (Fig. 6B) and SnSAG4 (Fig. 6E) highlighted a reticulum that was most apparent during the intermediate phases of endopolygeny. This pattern of fluorescent labeling was not observed for SnSAG2 and SnSAG3. Occasionally, a vague network of SnSAG1 or SnSAG4 could be perceived in some methanol-fixed cells (data not shown). Collectively, the immunofluorescence results demonstrated that the four surface antigens are present throughout intracellular growth, but they may be concentrated in dissimilar regions in the developing and mature schizonts.
To further clarify the localization of the SnSAGs during parasite development, the inner membrane complex protein IMC3 was labeled as a marker for daughter merozoite formation (24, 36). Throughout much of the intracellular development of the schizonts, the IMC3 label was weak and nondescript (Fig. 7A). However, strong and distinct labeling of IMC3 became apparent in late-stage schizonts, coinciding with nuclear division and preceding the completion of cytokinesis. At this stage, IMC3 delineated the early daughter merozoites (Fig. 7A), similar to what was observed for SnSAG2 and SnSAG3. Indeed, covisualization of IMC3 with SnSAG3 demonstrated that there is a tight association between these two antigens during the formation of the merozoites (Fig. 7B). IMC3 and SnSAG1 also exhibited partial colocalization, although a considerable amount of additional SnSAG1 label was observed (data not shown). Further development of the parasites resulted in a strong, circumferential fluorescence of IMC3 on individual merozoites of the mature schizont (Fig. 7A). Unfortunately, immunofluorescence analysis of IMC3 was possible only with cells fixed with methanol. Therefore, the inner membrane complex protein could not be visualized in formaldehyde-fixed, Triton X-100-permeabilized cells. However, the labeling pattern seen for IMC3 (Fig. 7) was clearly distinct from the reticulum of label that was observed for SnSAG1 and SnSAG4 (Fig. 6B and E).
DISCUSSION
Sarcocystis is a very broad genus that encompasses some of the most prevalent parasites of vertebrate animals. Despite their abundance and importance as pathogens, the Sarcocystis spp. have been understudied relative to other members of the Apicomplexa. Consequently, there is a significant information gap for a large segment of this phylum. In an effort to define the genetic and molecular composition of S. neurona, we have conducted an EST sequencing project (31, 43) and explored the sequence database to identify S. neurona virulence factors. These searches have revealed a gene family that encodes at least four paralogous proteins with obvious homology and probable structural similarity to the SAG/SRS family of surface antigens in T. gondii (42). The four SnSAGs were all found to be membrane associated and displayed on the merozoite surface, consistent with their homology to surface antigens in related parasites. Sequence predictions suggested the presence of glycolipid anchors, as observed for the SAGs of T. gondii (49, 60) and N. caninum (32, 55). Although the SnSAGs could not be liberated from the merozoite surface by treatment with phosphatidylinositol-specific phospholipase C (data not shown), itis likely that this was due to inositol modification in the S. neurona GPI anchors, as observed previously for N. caninum tachyzoites (54, 55). It is apparent that the SnSAGs are highly immunogenic, similar to the major surface antigens of T. gondii (56) and N. caninum (30, 32). Indeed, SnSAG1 and SnSAG2 corresponded to two of the most immunodominant antigens (Fig. 4), which were previously designated Sn30 and Sn16 and were speculated to be surface proteins of S. neurona merozoites (44). Precedence from studies of T. gondii suggests that these SnSAGs may serve as multitasking virulence factors that are important for S. neurona's ability to invade cells and evade the host immune response. Further in-depth investigation of this gene family in S. neurona should allow comparative studies that could provide insight into the evolution and function of the SAGs in the tissue-forming coccidia.
The SnSAG1 surface antigen, in particular, appeared to be exceedingly immunogenic. Our cDNA library immunoscreening resulted in identification of the SnSAG1 gene in 22 of 25 reactive phage clones, which is consistent with this antigen's immunodominance. Additionally, the SnSAG1 gene was represented by the largest assembly of S. neurona ESTs (43), thus implying that the SnSAG1 mRNA is abundant and expressed at high levels in S. neurona merozoites. In view of its abundance and immunodominance, it was predictable that SnSAG1 would be the initial S. neurona antigen identified by us, as well as by other workers (18). Identical to the data presented here, SnSAG1 was described as an immunodominant antigen of S. neurona that localized to the surface of extracellular merozoites in immunofluorescence and immunoelectron microscope analyses (18). Notably, Ellison et al. found that the SnSAG1 genomic locus contains a 128-bp intron that is located in the hinge region separating the D1 and D2 domains (18), identical to what was described in the TgSAG2 family of surface antigens (27, 41). Consistent with our predicted gene phylogeny (Fig. 2), these data confirm that the S. neurona major surface antigen SnSAG1 gene is most similar to the TgSAG2 gene family rather than the gene encoding the predominant surface antigen of T. gondii, TgSAG1. Amplification of the SnSAG3 and SnSAG4 genomic loci also indicated the presence of an intron(s) in both genes, although the specific location of the intron within each locus was not determined (data not shown).
Fluorescence labeling demonstrated that all four SnSAGs are present throughout intracellular development (Fig. 6). Additionally, these analyses revealed a reticulate structure that was most evident in cells that were formaldehyde fixed and permeabilized with Triton X-100 (Fig. 6B). Permeabilization with strong detergents such as Triton X-100 can cause partial extraction of membrane proteins, especially from the periphery of cells, along with an accompanying increased immunoreactivity in intracellular membranes and compartments (26). Thus, the network of SnSAGs likely represents the deep invaginations of the schizont pellicle that have been observed in electron micrographs (58). Although the inner membrane complex appears to be present during all stages of intracellular development (58), the inner membrane complex protein IMC3 was not patently detectable with the infolded pellicle network throughout much of schizont maturation (Fig. 7A). Instead, IMC3 became evident at the latest stages of development when daughter cell formation had initiated, which is consistent with what has been observed during T. gondii endodyogeny (24). At this point of development, the SnSAGs became associated with the inner membrane complex, and SnSAG2 and SnSAG3 were almost exclusively linked to the newly forming merozoites (Fig. 6C and 7B).
Although the SnSAGs were found to be present throughout intracellular development (Fig. 6), it is probable that these proteins function primarily during the extracellular merozoite stage. By virtue of their location on the parasite surface, the SnSAGs are likely central to the parasite's initial interactions with the host cell and components of the immune response. Indeed, a role in the invasion of host cells has been previously proposed for the Sn16 surface antigen (identified here as SnSAG2) of S. neurona merozoites (44). Likewise, the surface antigens of T. gondii (22, 23, 47, 48) and N. caninum (28) have been implicated in attachment to and invasion of host cells. It was apparent from these prior studies, however, that the participation of the SAGs in this essential process is relatively complex. In fact, the individual surface antigens may each perform a discrete function during host cell invasion, which is consistent with the differential localization observed for the SnSAGs (Fig. 6).
In addition to their potential role as adhesins, the SnSAGs may act as immunomodulatory molecules that influence the infection and the progression of disease. In the present study, two of the most immunodominant protein species seen in the Western blot analysis of S. neurona correspond to SnSAG1, SnSAG2, and SnSAG4 (Fig. 4). Similarly, many of the SAGs from T. gondii (12, 25, 56) and N. caninum (29, 30, 32) have been described as immunodominant antigens. As suggested for toxoplasmosis (4), the robust immune response collectively elicited by the multiple immunogenic SAGs may be critical for controlling the acute stage of disease, thereby preventing a rampant infection that results in death of the host. Additionally, most of the TgSAGs exhibit stage-specific expression during the T. gondii life cycle (42). As a consequence, the immunodominant TgSAGs present on the tachyzoite surface are absent from the latent cysts that form in the infected host animal. In effect, this provides a mechanism whereby T. gondii bradyzoites become incognito to effectors of the host immune response. It remains to be determined whether there is stage-specific expression of the SnSAGs during the life cycle of S. neurona, but given the prominent immunogenicity exhibited by the SnSAGs, it is plausible that these surface antigens are important for immune evasion by the parasite.
Due to their surface display, immunogenicity, and apparent omnipresence, the SAGs collectively represent a promising general target for intervention in coccidian diseases (e.g., as vaccine antigens). It is well established that TgSAG1 from T. gondii can protect against acute toxoplasmosis in mice (2, 3, 6, 11, 39). A number of studies have since investigated the development of protective immune responses with assorted other SAG family members from T. gondii (45) and N. caninum (50, 51). Logically, the SnSAGs may similarly prove to be useful as vaccine antigens that protect against EPM in horses. Furthermore, the immunogenicity displayed by many of the SAG family members makes these antigens appealing markers for use in serodiagnostic assays. In individuals infected with T. gondii, anti-TgSAG1 antibodies can be detected early in the infection, and these antibodies include IgM, IgA, and various IgG subisotypes (13, 52, 53). Likewise, the two major surface antigens of N. caninum, NcSAG1 and NcSRS2, are consistently recognized by antisera from Neospora-infected animals (32), and a recombinant form of NcSAG1 has been implemented in an enzyme-linked immunosorbent assay (35). These prior studies provide considerable precedence for the use of the SnSAGs as diagnostic molecules.
In summary, the existence of a family of paralogous SAGs in S. neurona merozoites indicates that expression of numerous, semiredundant surface antigens is a general theme among the heteroxenous (two-host) Coccidia. Although the purpose served by these surface antigen families remains to be elucidated, their conserved presence is presumably important for an essential function(s) that is common to this group of parasitic organisms. Further investigation of Sarcocystis spp. and their complements of surface molecules should allow comparisons with the much-better-characterized organism T. gondii, thereby providing greater insight into the importance of the SAG/SRS gene families as virulence factors.
ACKNOWLEDGMENTS
This research was supported by grants from the Amerman Family Foundation and Fort Dodge Animal Health and by gifts from various donors.
We gratefully acknowledge Anthony Sinai for helpful discussions and a critical review of the manuscript.
Kentucky Agricultural Experiment Station article no. 04-14-017.
REFERENCES
1. Altschul, S. F., W. Gish, W. Miller, E. W. Myers, and D. J. Lipman. 1990. Basic local alignment search tool. J. Mol. Biol. 215:403-410.
2. Angus, C. W., D. Klivington-Evans, J. P. Dubey, and J. A. Kovacs. 2000. Immunization with a DNA plasmid encoding the SAG1 (P30) protein of Toxoplasma gondii is immunogenic and protective in rodents. J. Infect. Dis. 181:317-324.
3. Bonenfant, C., I. Dimier-Poisson, F. Velge-Roussel, D. Buzoni-Gatel, G. Del Giudice, R. Rappuoli, and D. Bout. 2001. Intranasal immunization with SAG1 and nontoxic mutant heat-labile enterotoxins protects mice against Toxoplasma gondii. Infect. Immun. 69:1605-1612.
4. Boothroyd, J. C., A. Hehl, L. J. Knoll, and I. D. Manger. 1998. The surface of Toxoplasma: more and less. Int. J. Parasitol. 28:3-9.
5. Brusca, J. S., and J. D. Radolf. 1994. Isolation of integral membrane proteins by phase partitioning with Triton X-114. Methods Enzymol. 228:182-193.
6. Bulow, R., and J. C. Boothroyd. 1991. Protection of mice from fatal Toxoplasma infection by immunization with p30 antigen in liposomes. J. Immunol. 147:3496-3500.
7. Cesbron-Delauw, M. F., S. Tomavo, P. Beauchamps, M. P. Fourmaux, D. Camus, A. Capron, and J. F. Dubremetz. 1994. Similarities between the primary structures of two distinct major surface proteins of Toxoplasma gondii. J. Biol. Chem. 269:16217-16222.
8. Cheadle, M. A., S. M. Tanhauser, J. B. Dame, D. C. Sellon, M. Hines, P. E. Ginn, R. J. MacKay, and E. C. Greiner. 2001. The nine-banded armadillo (Dasypus novemcinctus) is an intermediate host for Sarcocystis neurona. Int. J. Parasitol. 31:330-335.
9. Cheadle, M. A., C. A. Yowell, D. C. Sellon, M. Hines, P. E. Ginn, A. E. Marsh, J. B. Dame, and E. C. Greiner. 2001. The striped skunk (Mephitis mephitis) is an intermediate host for Sarcocystis neurona. Int. J. Parasitol. 31:843-849.
10. Corpet, F. 1988. Multiple sequence alignment with hierarchical clustering. Nucleic Acids Res. 16:10881-10890.
11. Couper, K. N., H. V. Nielsen, E. Peterson, F. Roberts, C. W. Roberts, and J. Alexander. 2003. DNA vaccination with the immunodominant tachyzoite surface antigen (SAG-1) protects against adult acquired Toxoplasma gondii infection but does not prevent maternofoetal transmission. Vaccine 21:2813-2820.
12. Couvreur, G., A. Sadak, B. Fortier, and J. F. Dubremetz. 1988. Surface antigens of Toxoplasma gondii. Parasitology 97:1-10.
13. Decoster, A., A. Caron, F. Darcy, and A. Capron. 1988. IgA antibodies against p30 as markers of congenital and acute toxoplasmosis. Lancet ii:1104-1106.
14. Devereux, J., P. Haeberli, and O. Smithies. 1984. A comprehensive set of sequence analysis programs for the VAX. Nucleic Acids Res. 12:387-395.
15. Dubey, J. P., D. S. Lindsay, W. J. Saville, S. M. Reed, D. E. Granstrom, and C. A. Speer. 2001. A review of Sarcocystis neurona and equine protozoal myeloencephalitis (EPM). Vet. Parasitol. 95:89-131.
16. Dubey, J. P., W. J. Saville, D. S. Lindsay, R. W. Stich, J. F. Stanek, C. A. Speer, B. M. Rosenthal, C. J. Njoku, O. C. Kwok, S. K. Shen, and S. M. Reed. 2000. Completion of the life cycle of Sarcocystis neurona. J. Parasitol. 86:1276-1280.
17. Dubey, J. P., W. J. Saville, J. F. Stanek, D. S. Lindsay, B. M. Rosenthal, M. J. Oglesbee, A. C. Rosypal, C. J. Njoku, R. W. Stich, O. C. Kwok, S. K. Shen, A. N. Hamir, and S. M. Reed. 2001. Sarcocystis neurona infections in raccoons (Procyon lotor): evidence for natural infection with sarcocysts, transmission of infection to opossums (Didelphis virginiana), and experimental induction of neurologic disease in raccoons. Vet. Parasitol. 100:117-129.
18. Ellison, S. P., A. L. Omara-Opyene, C. A. Yowell, A. E. Marsh, and J. B. Dame. 2002. Molecular characterisation of a major 29 kDa surface antigen of Sarcocystis neurona. Int. J. Parasitol. 32:217-225.
19. Eschenbacher, K. H., I. Sommer, H. E. Meyer, H. Mehlhorn, and W. Ruger. 1992. Cloning and expression in Escherichia coli of cDNAs encoding a 31-kilodalton surface antigen of Sarcocystis muris. Mol. Biochem. Parasitol. 53:159-167.
20. Fenger, C. K., D. E. Granstrom, J. L. Langemeier, S. Stamper, J. M. Donahue, J. S. Patterson, A. A. Gajadhar, J. V. Marteniuk, Z. Xiaomin, and J. P. Dubey. 1995. Identification of opossums (Didelphis virginiana) as the putative definitive host of Sarcocystis neurona. J. Parasitol. 81:916-919.
21. Granstrom, D. E., O. Alvarez, Jr., J. P. Dubey, P. F. Comer, and N. M. Williams. 1992. Equine protozoal myelitis in Panamanian horses and isolation of Sarcocystis neurona. J. Parasitol. 78:909-912.
22. Grimwood, J., and J. E. Smith. 1992. Toxoplasma gondii: the role of a 30-kDa surface protein in host cell invasion. Exp. Parasitol. 74:106-111.
23. 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.
24. Gubbels, M. J., M. Wieffer, and B. Striepen. 2004. Fluorescent protein tagging in Toxoplasma gondii: identification of a novel inner membrane complex component conserved among Apicomplexa. Mol. Biochem. Parasitol. 137:99-110.
25. Handman, E., J. W. Goding, and J. S. Remington. 1980. Detection and characterization of membrane antigens of Toxoplasma gondii. J. Immunol. 124:2578-2583.
26. Hannah, M. J., U. Weiss, and W. B. Huttner. 1998. Differential extraction of proteins from paraformaldehyde-fixed cells: lessons from synatophysin and other membrane proteins. Methods Companion Methods Enzymol. 16:170-181.
27. He, X. L., M. E. Grigg, J. C. Boothroyd, and K. C. Garcia. 2002. Structure of the immunodominant surface antigen from the Toxoplasma gondii SRS superfamily. Nat. Struct. Biol. 9:606-611.
28. Hemphill, A. 1996. Subcellular localization and functional characterization of Nc-p43. Infect. Immun. 64:4279-4287.
29. Hemphill, A., N. Fuchs, S. Sonda, B. Gottstein, and B. Hentrich. 1997. Identification and partial characterization of a 36 kDa surface protein on Neospora caninum tachyzoites. Parasitology 115:371-380.
30. Hemphill, A., and B. Gottstein. 1996. Identification of a major surface protein on Neospora caninum tachyzoites. Parasitol. Res. 82:497-504.
31. Howe, D. K. 2001. Initiation of a Sarcocystis neurona expressed sequence tag (EST) sequencing project: a preliminary report. Vet. Parasitol. 95:233-239.
32. Howe, D. K., A. C. Crawford, D. Lindsay, and L. D. Sibley. 1998. The p29 and p35 immunodominant antigens of Neospora caninum tachyzoites are homologous to the family of surface antigens of Toxoplasma gondii. Infect. Immun. 66:5322-5328.
33. Howe, D. K., and L. D. Sibley. 1997. Development of molecular genetics for Neospora caninum: a complementary system to Toxoplasma gondii. Methods Companion Methods Enzymol. 13:123-133.
34. Howe, D. K., and L. D. Sibley. 1999. Comparison of the major antigens of Neospora caninum and Toxoplasma gondii. Int. J. Parasitol. 29:1489-1496.
35. Howe, D. K., K. Tang, P. A. Conrad, K. Sverlow, J. P. Dubey, and L. D. Sibley. 2002. Sensitive and specific identification of Neospora caninum infection of cattle based on detection of serum antibodies to recombinant Ncp29. Clin. Diagn. Lab. Immunol. 9:611-615.
36. Hu, K., T. Mann, B. Striepen, C. J. Beckers, D. S. Roos, and J. M. Murray. 2002. Daughter cell assembly in the protozoan parasite Toxoplasma gondii. Mol. Biol. Cell 13:593-606.
37. Jacquet, A., L. Coulon, J. De Neve, V. Daminet, M. Haumont, L. Garcia, A. Bollen, M. Jurado, and R. Biemans. 2001. The surface antigen SAG3 mediates the attachment of Toxoplasma gondii to cell-surface proteoglycans. Mol. Biochem. Parasitol. 116:35-44.
38. Jung, C., C. Y. F. Lee, and M. E. Grigg. 2004. The SRS superfamily of Toxoplasma surface proteins. Int. J. Parasitol. 34:285-296.
39. Khan, I. A., K. H. Ely, and L. H. Kasper. 1991. A purified parasite antigen (p30) mediates CD8+ T cell immunity against fatal Toxoplasma gondii infection in mice. J. Immunol. 147:3501-3506.
40. Laemmli, U. K. 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature (London) 227:680-685.
41. Lekutis, C., D. J. Ferguson, and J. C. Boothroyd. 2000. Toxoplasma gondii: identification of a developmentally regulated family of genes related to SAG2. Exp. Parasitol. 96:89-96.
42. Lekutis, C., D. J. Ferguson, M. E. Grigg, M. Camps, and J. C. Boothroyd. 2001. Surface antigens of Toxoplasma gondii: variations on a theme. Int. J. Parasitol. 31:1285-1292.
43. Li, L., B. P. Brunk, J. C. Kissinger, D. Pape, K. Tang, R. H. Cole, J. Martin, T. Wylie, M. Dante, S. J. Fogarty, D. K. Howe, P. Liberator, C. Diaz, J. Anderson, M. White, M. E. Jerome, E. A. Johnson, J. A. Radke, C. J. Stoeckert, Jr., R. H. Waterston, S. W. Clifton, D. S. Roos, and L. D. Sibley. 2003. Gene discovery in the Apicomplexa as revealed by EST sequencing and assembly of a comparative gene database. Genome Res. 13:443-454.
44. Liang, F. T., D. E. Granstrom, X. M. Zhao, and J. F. Timoney. 1998. Evidence that surface proteins Sn14 and Sn16 of Sarcocystis neurona merozoites are involved in infection and immunity. Infect. Immun. 66:1834-1838.
45. Lunden, A., S. F. Parmley, K. L. Bengtsson, and F. G. Araujo. 1997. Use of a recombinant antigen, SAG2, expressed as a glutathione-S-transferase fusion protein to immunize mice against Toxoplasma gondii. Parasitol. Res. 83:6-9.
46. MacKay, R. J., D. E. Granstrom, W. J. Saville, and S. M. Reed. 2000. Equine protozoal myeloencephalitis. Vet. Clin. N. Am. Equine Pract. 16:405-425.
47. Mineo, J. R., and L. H. Kasper. 1994. Attachment of Toxoplasma gondii to host cells involves major surface protein SAG-1 (P30). Exp. Parasitol. 79:11-20.
48. Mineo, J. R., R. McLeod, D. Mack, J. Smith, I. A. Khan, K. H. Ely, and L. H. Kasper. 1993. Antibodies to Toxoplasma gondii major surface protein (SAG-1, P30) inhibit infection of host cells and are produced in murine intestine after peroral infection. J. Immunol. 150:3951-3964.
49. Nagel, S. D., and J. C. Boothroyd. 1989. The major surface antigen, p30, of Toxoplasma gondii is anchored by a glycolipid. J. Biol. Chem. 264:5569-5574.
50. Nishikawa, Y., N. Inoue, X. Xuan, H. Nagasawa, I. Igarashi, K. Fujisaki, H. Otsuka, and T. Mikami. 2001. Protective efficacy of vaccination by recombinant vaccinia virus against Neospora caninum infection. Vaccine 19:1381-1390.
51. Nishikawa, Y., X. Xuan, H. Nagasawa, I. Igarashi, K. Fujisaki, H. Otsuka, and T. Mikami. 2001. Prevention of vertical transmission of Neospora caninum in BALB/c mice by recombinant vaccinia virus carrying NcSRS2 gene. Vaccine 19:1710-1716.
52. Potasman, I., F. G. Araujo, G. Desmonts, and J. S. Remington. 1986. Analysis of Toxoplasma gondii antigens recognized by human sera obtained before and after acute infection. J. Infect. Dis. 154:650-657.
53. Santoro, F., D. Afchain, R. Pierce, J. Y. Cesbron, G. Ovalaque, and A. Capron. 1985. Serodiagnosis of Toxoplasma infection using a purified parasite protein (p30). Clin. Exp. Immunol. 62:262-269.
54. Schares, G., J. F. Dubremetz, J. P. Dubey, A. Barwald, A. Loyens, and F. J. Conraths. 1999. Neospora caninum: identification of 19-, 38-, and 40-kDa surface antigens and a 33-kDa dense granule antigen using monoclonal antibodies. Exp. Parasitol. 92:109-119.
55. Schares, G., C. F. Zinecker, J. Schmidt, N. Azzouz, F. J. Conraths, P. Gerold, and R. T. Schwarz. 2000. Structural analysis of free and protein-bound glycosyl-phosphatidylinositols of Neospora caninum. Mol. Biochem. Parasitol. 105:155-161.
56. Sharma, S. D., J. Mullenax, F. G. Araujo, H. A. Ehrlich, and J. S. Remington. 1983. Western blot analysis of the antigens of Toxoplasma gondii recognized by human IgM and IgG antibodies. J. Immunol. 131:977-983.
57. Sibley, L. D., E. R. Pfefferkorn, and J. C. Boothroyd. 1991. Proposal for a uniform genetic nomenclature in Toxoplasma gondii. Parasitol. Today 7:327-328.
58. Speer, C. A., and J. P. Dubey. 2001. Ultrastructure of schizonts and merozoites of Sarcocystis neurona. Vet. Parasitol. 95:263-271.
59. Thompson, J. D., D. G. Higgins, and T. J. Gibson. 1994. CLUSTAL W: improving the sensitivity of progressive multiple sequence alignments through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res. 22:4673-4680.
60. Tomavo, S., R. T. Scwartz, and J. F. Dubremetz. 1989. Evidence for glycosyl-phosphatidylinositol anchoring of Toxoplasma gondii major surface antigens. Mol. Cell. Biol. 9:4576-4580.(Daniel K. Howe, Rajshekha)