Neospora caninum:Tachyzoites Express a Potent Type-I Nucleoside Triphosphate Hydrolase,but Lack Nucleoside Diphosphate Hydrolase Activity

EXPERIMENTAL PARASITOLOGY ARTICLE NO.

90, 277–285 (1998)

PR984346

Neospora caninum: Tachyzoites Express a Potent Type-I Nucleoside Triphosphate Hydrolase,1 but Lack Nucleoside Diphosphate Hydrolase Activity

Takashi Asai,* Daniel K. Howe,† Kyoko Nakajima,* Tomoyoshi Nozaki,* Tsutomu Takeuchi,* and L. David Sibley†,2 *Department of Tropical Medicine and Parasitology, Keio University School of Medicine, Shinjuku, Tokyo 160, Japan; and †Department of Molecular Microbiology, Washington University School of Medicine, St. Louis, Missouri 63110, U.S.A.

Asai, T., Howe, D. K., Nakajima, K., Nozaki, T., Takeuchi, T., and Sibley, L. D. Neospora caninum: Tachyzoites Express Type-I Nucleoside Triphosphate Hydrolase1. But Lack Nucleoside Diphosphate Hydrolase Activity. Experimental Parasitology 90, 277–285. We have identified type I nucleoside triphosphate hydrolase (NTPase; EC 3.6.1.3) activity, previously thought to be restricted to the virulent strains of Toxoplasma gondii, in the cell extracts of Neospora caninum tachyzoites. Sequence analysis of a complete cDNA from Nc-1 strain indicated that N. caninum NTPases shared approximately 69% identity to the NTPases of T. gondii and are most similar to the NTPase-I isozyme. Southern blot analysis of genomic DNA and sequence analysis of two independent NTP clones from the Nc-1 strain revealed the presence of multiple genes, at least two of which are transcribed. Substrate specificity and Km values for MgATP22 and MgADP2 hydrolysis for recombinant or partially purified native NcNTPase were the same as those for the type I isozyme (NTPaseI). Significantly, no type II enzyme (NTPase-II) activity for NDP hydrolysis was detected in cell extracts of N. caninum, although it is universally present in all T. gondii strains that have been tested. This intriguing difference between these two closely related apicomplexan parasites may provide insight into the function of the NTPases during intracellular parasitism. q 1998 Academic Press Index Descriptors and Abbreviations: apicomplexan protozoa; neosporosis; Toxoplasma gondii; ATPase; purine salvage; adenosine triphosphate, ATP; adenosine diphosphate, ADP; dithiothreitol, DTT; electron microscopy, EM; ethylenediaminetetraacetic acid, EDTA; fetal bovine serum, FBS; high-performance liquid chromatography, HPLC;

1

Sequence data reported have been submitted to GenBank under Accession No. AB010444. 2 To whom correspondence should be addressed. Fax: 314-362-1232. E-mail: [email protected].

0014-4894/98 $25.00 Copyright q 1998 by Academic Press All rights of reproduction in any form reserved.

N-2-hydroxyethylene piperazine-N8-2-ethansulfonic acid, Hepes; human foreskin fibroblasts, HFF; isopropyl b-D-thiogalactopyranoside, IPTG; nucleoside diphosphate, NDP; nucleoside triphosphate, NTP; nucleoide triphosphate phosphatase, NTPase; open reading frame, ORF; phosphate-buffered saline, PBS; polyacrylamide gel electrophoresis, PAGE; polymerase chain reaction, PCR; polyvinylpyrrolidone, PVP; sodium dodecyl sulfate, SDS; standard sodium citrate, SSC.

INTRODUCTION

The tachyzoite form of Toxoplasma gondii, an apicomplexan protozoan infecting many warm-blooded animals, has an active nucleoside triphosphate hydrolase (NTPase; EC 3.6.1.3) with a number of unusual properties, including a requirement for activation by dithiols such as DTT (Asai et al. 1983). It has been found that type I strains of T. gondii, which are acutely virulent in mice, contain two isoforms of NTPase (Bermudes et al. 1994; Asai et al. 1995); the type I enzyme (NTPase-I in Asai et al. 1995; NTP3 in Bermudes et al. 1994) preferentially hydrolyzes triphosphate nucleosides, while the type II enzyme (NTPase-II in Asai et al. 1995, NTP1 in Bermudes et al. 1994) hydrolyzes tri- and diphosphate nucleosides at approximately equal rates. The NTPase-I isoform appears to be present only in the type I virulent strains, while NTPase-II is universally present in all

277

278 T. gondii strains (Asai et al. 1995). Similarly, DTT-activated NTPase enzymes were not identified in the related apicomplexan parasites Plasmodium berghei (Asai et al. 1986) and Eimeria tenella (T. Asai, unpublished data) and were previously thought to be restricted to T. gondii. Neospora caninum is a newly recognized apicomplexan parasite (Dubey et al. 1988) that is morphologically very similar to T. gondii (Speer and Dubey 1989; Lindsay et al. 1993) and phylogenetically close, as determined by ribosomal RNA sequence comparison (Marsh et al. 1995). Despite their structural and phylogenetic similarities, polyclonal antisera against T. gondii show limited crossreactivity against N. caninum (Dubey et al. 1988), and many monoclonal antibodies against T. gondii antigens do not cross-react with N. caninum tachyzoites (Howe and Sibley 1997). N. caninum, while an important pathogen in domestic and companion animals where it causes paralysis and abortion, is less pathogenic in mice and has not been reported to infect humans (Dubey and Lindsay 1996). Given the close phylogenetic relationship between T. gondii and N. caninum, we chose to investigate whether N. caninum tachyzoites might also contain NTPase activity. Herein, we report the identification and molecular characterization of a potent NTPase in N. caninum which is similar to the T. gondii NTPase-I enzyme described previously.

MATERIALS AND METHODS

Reagents. Nucleotides and chemicals were obtained from Sigma (St. Louis, MO). Cell culture reagents were purchased from Gibco BRL. Other reagents were commercial products of the highest purity available. Parasite culture. Strains of N. caninum and the RH strain of T. gondii were propagated as tachyzoites by serial passage in human foreskin fibroblast host cells maintained in Dulbecco’s modified Eagle’s medium supplemented with 10% FBS (Hyclone Inc., Logan, UT), 2 mM glutamine, and 20 mg/ml gentamicin. Tachyzoites from freshly lysed host cells were harvested by separation through 3.0-mm polycarbonate filters (Nucleopore Corp., Pleasanton, CA), as described previously for T. gondii (Sibley and Boothroyd 1992). Tachyzoites were collected by centrifugation at 1000g for 10 min, washed three times with PBS, and counted using a hemocytometer. Cell pellets was stored at 2808C until used. Western blot analysis. Parasites were lysed in sample

ASAI ET AL.

buffer and separated by SDS–PAGE (Laemmli 1970). Proteins were transferred to nitrocellulose membranes by semidry electroblotting and stained with India ink. The blot was blocked with PBS containing nonfat dry milk, incubated with affinity-purified rabbit polyclonal antiserum against the T. gondii NTPases (Sibley et al. 1994), and then incubated with 125I-conjugated protein A (Amersham). After washing, the blot was exposed to Kodak XAR film. Southern blot analysis. Genomic DNAs (2.5 mg each) were treated with restriction endonucleases, separated in 1.0% agarose gels by electrophoresis, and transferred to nitrocellulose by capillary blotting. A probe corresponding to the ORF was amplified from genomic DNA of the Nc-1 strain and labeled with 32P by random priming according to the manufacturer’s instructions (Takara Shuzo Co., Japan). Blots were hybridized at 658C in 63 SSC containing 53 Denhardt’s solution, 1% SDS, and 100 mg/ml of salmon sperm DNA. After overnight incubation, blots were washed at a final stringency of 0.23 SSC, 0.2% SDS at 688C for 30 min and exposed to Kodak XAR film. ImmunoEM localization. Extracellular T. gondii tachyzoites were fixed for 1 h at 48C with 2.5% Formalin, 0.5% glutaralydehyde in Tris-buffered NaCl, pH 7.2. Samples were then embedded in 10% gelatin, infiltrated with sucrose/ PVP, and frozen in liquid N2. Ultrathin cryosections were blocked in 0.12 mM glycine and PBS–10% FBS, then incubated in primary antibodies diluted in PBS–1% FBS, followed by the secondary antibodies goat anti-rabbit coupled to 18-nm colloidal gold (Jackson Laboratories) diluted in PBS–1% FBS. Sections were counterstained with 0.15 M oxalic acid/2% uranyl acetate, stained with a 1:12 mixture of 2% methylcellulose and 4% uranyl acetate, and examined using a Zeiss EM902 microscope. NTP gene isolation and characterization. A unidirectional cDNA expression library from Nc-1 strain tachyzoites was constructed in the Uni-ZAP Lambda vector according to the manufacturer’s protocol (Stratagene, La Jolla, CA). The library was screened with the rabbit anti-TgNTPase serum to identify recombinant phage expressing putative NTP genes. One of these clones was sequenced in its entirety and proved to be a partial cDNA sequence that was highly homologous to the T. gondii NTP family of genes. The 58 end of the gene was subsequently obtained by PCR amplification from the cDNA library using a reverse-orientation oligonucleotide (58-AGTCGTGGAAATCACGAAC-38) corresponding to the partial NTP clone in combination with the T3 oligonucleotide from the lambda vector. Sequencing was done by Sanger dideoxy termination reactions and sequence analyses were performed with GCG version 8 (Devereaux et al. 1984).

279

N. caninum NTPase EXPRESSION

Expression of recombinant NcNTPase. The NTP gene was amplified from genomic DNA of Nc-1 strain N. caninum using PCR (forward oligo 58-TACCATATGGCCGACGAGCCAGCGACACTT-38; reverse oligo 58-TCGCTCGAGTCACATGTTGTAGGTAAATCC-38). The amplified fragment was cloned into the NdeI–XhoI sites of the vector pET22b1 (Novagen, Madison, WI), and transformed into BL21 Escherichia coli host cells. Expression of recombinant NTPase was induced with 0.2 mM IPTG for 2.5 h. Inclusion bodies were dissolved in 6 M guanidium HCl in 50 mM Tris–HCl, pH7.5, diluted, and dialyzed against Tris–HCl, pH 7.5. The dialyzed solution containing refolded NTPase was brought to pH 8.5 by adding 1 M Tris base and applied to a DEAE-Toyo Peal 650s column equilibrated with 20 mM Tris–HCl, pH 8.5, and eluted with a linear gradient from 0 to 0.5 M KCl. Peak active fractions were pooled and concentrated and applied to a Toyo peal HW55s column equilibrated with 50 mM Tris–HCl, pH 7.5, containing 0.2 M KCl. Peak active fractions were checked for purity by SDS–PAGE and Coomassie blue staining, and fractions containing no other proteins except NTPase were pooled, concentrated, and adjusted with glycerol (final 20% w/v). The purified enzyme was stored at 2808C until use. Determination of NTPase activity. Frozen tachyzoites (,3 3 107) were resuspended in 0.5 ml of a solution containing 20 mM Hepes/KOH, pH 7.5, 0.9% NaCl. The suspension was sonicated at 20 kHz for 10 s in an ice bath and centrifuged at 10,000 rpm for 10 min at 48C. The supernatant was collected and glycerol was added to a final concentration of 20% (w/v). Partial purification of native NTPase from the Nc-1 strain was accomplished by precipitation with ammonium sulfate. The contents of the reaction mixture and the reaction conditions for determining enzyme activity were as described previously (Asai et al. 1995). Degradation of nucleotides was analyzed by HPLC as described previously (Asai and Suzuki 1990). For determination of substrate specificities, 1 mM NTPs and 4 mM NDPs were used for the NTPase-I, while both NTPs and NDPs were used at 1 mM for the NTPase-II.

RESULTS Identification and immunoEM localization of NTPase. To determine if N. caninum tachyzoites contained an immunologically cross-reactive NTPase, we analyzed the Nc-1 strain by Western blot using a rabbit polyclonal serum that was raised against purified T. gondii NTPases. This antiserum reacts primarily to the NTPases which comigrate at 67

FIG. 1. Western blot analysis of RH strain T. gondii and Nc-1 strain N. caninum lysates with a rabbit polyclonal anti-serum against the NTPase of T. gondii. A protein that comigrates with the 67-kDa T. gondii NTPase is recognized in the N. caninum lysate.

kDa (greater than 95% of the signal) in lysates of T. gondii; on long exposure, it also recognizes several bands of lower and higher molecular weight which are of unknown composition (Fig. 1). A cross-reactive protein that comigrated with the NTPases from the RH strain of T. gondii was identified in the Nc-1 lysate by the anti-TgNTPase antibodies (Fig. 1). While the reaction of this antiserum to N. caninum is weaker than that observed in T. gondii, the pattern is extremely specific for a band that comigrates with NTPases in T. gondii. To directly test for analogous enzyme activity in N. caninum, cell homogenates of five different isolates were compared to the NTPase activity in a cell homogenate of the RH strain of T. gondii. This assay identified a potent DTT-dependent, ATP hydrolytic activity characteristic of an NTPase in the cell extract from all five N. caninum isolates (Table I). The NTPase of N. caninum (NcNTPase) was stable, as freezing and thawing did not cause loss of enzyme

TABLE I Comparison of NTPase Activities ( Vmax ) in the Cell Homogenates of N. caninum strains and T. gondii Strain RH Strain Nc-1 Nc-2 Nc-LIV BPA-1 BPA-3 Rh (T. gondii) a

Mean 6 SD (N 5 3, 4).

NTPase activity (mmol/min/mg protein) a 15.9 29.7 21.97 16.13 13.87 31.0

6 6 6 6 6 6

5.29 20.76 14.51 5.36 10.19 7.24

280 activity, similar to the previously characterized T. gondii NTPase (Asai et al. 1983; Asai and Suzuki 1990). In the absence of DTT, nucleotide hydrolytic activity in the crude extract of N. caninum was almost nil under the assay conditions (data not shown). No activity was detected in crude lysates of N. caninum using ADP as a substrate (data not shown). To localize the NTPase homologue in N. caninum, immunogold labeling was used to stain extracellular tachyzoites incubated with the anti-TgNTPase antibodies. The majority of gold particles were located in the dense granules (Fig. 2B), although minor labeling of the cytoplasm was occasionally seen (data not shown). Occasional gold particles were also observed in the cytoplasm of parasites labeled with a control rabbit antiserum (anti-GST) (Fig. 2A). These data indicated that NcNTPase is a dense granule protein, as has been previously observed for the NTPases of T. gondii (Sibley et al. 1994). Cloning and characterization of NcNTPase. To characterize NTP genes in Nc-1, a cDNA expression library was screened with the anti-TgNTPase polyclonal serum. An immunoreactive clone was isolated and sequenced and was found to contain a NTP homologue that lacked all but approximately 500 bp at the 58 ORF (about 175 amino acids) (data not shown). The 58 end of the N. caninum NTP gene

FIG. 2. ImmunoEM localization of NcNTPase in extracellular tachyzoites of N. caninum. (A) Control rabbit anti-GST. (B) Affinitypurified rabbit anti-TgNTPase. Secondary antisera was goat anti-rabbit IgG conjugated to 18 nm gold. Bar equals 200 nm.

ASAI ET AL.

was obtained by PCR amplication from the Nc-1 cDNA library with the T3 oligonucleotide and an internal oligonucleotide from the partial cDNA sequence. The full-length cDNA clone, designated pcNcNTP, encoded a 626-aminoacid protein with a predicted 24-residue leader peptide. When the predicted amino acid sequence was directly compared to T. gondii, the NcNTPase was found to share approximately 69% identity to either the NTPase-I or NTPase-II isoforms, with a single gap of two residues inserted into the primary sequence of NcNTPase to optimize the alignment (Fig. 3). Of the 16 amino acid polymorphisms between NTPase-I and NTPase-II (Bermudes et al. 1994; Asai et al. 1995), the NcNTPase shares 6 amino acids in common with NTPase-I and 4 amino acids in common with NTPase-II (Fig. 3). To determine if the NcNTPase was encoded by a singlecopy gene, the ORF of the pcNcNTP clone was used as a probe for Southern hybridization to Nc-1 genomic DNA that was digested with enzymes predicted to cut at a single site (SacII or SmaI) or not predicted to cut (ApaI, EcoRV, XhoI) within the ORF. Three separate fragments were detected in the DNA digested with SacII, which is predicted to cut at one site within the probe, and multiple fragments were detected in the DNA digested with enzymes not predicted to cut within the probe (Fig. 4). Hybridization to three fragments of EcoRV-digested DNA indicates that there are likely three NTP genes (Fig. 4), consistent with what has previously been observed in T. gondii (Bermudes et al. 1994; Asai et al. 1995). Since the Southern blot analysis indicated the presence of more than one gene, we also sequenced a genomic DNA clone that had been amplified from Nc-1 strain and cloned into the expression vector pET22b. This analysis indicated that the expression construct (designated prNcNTP) contained four nucleotide changes that resulted in four amino acid polymorphisms relative to the cDNA described above. Like the NTPases of T. gondii, this genomic clone encoded the amino acid residues S and Y at positions 316 and 395, respectively. Additionally, it contains two unique differences, T and V at positions 411 and 415, respectively. The two alleles are shown in Fig. 3 and are designated NTPaseIA and NTPase-IB based on biochemical evidence (below) that they encode enzymes with extremely similar activities. To determine if both genes were expressed in tachyzoites, NTP-specific primers were used to amplify from either genomic DNA or the cDNA library of Nc-1, and the amplification products were digested with NruI, which is predicted to cut pcNcNTP (cDNA clone) but not prNcNTP (genomic clone). Mixed PCR products, some of which contained the NruI site while others did not, were obtained in amplifications

N. caninum NTPase EXPRESSION

281

FIG. 3. Comparison of amino acid sequences for the T. gondii type I and type II NTPase isozymes and the NTPase-IA and NTPase-IB in N. caninum. Dashes represent amino acids that are identical to Tg/NTPase-I. Periods represent sequence gaps. Boxes indicate the T. gondii isozyme polymorphisms where the N. caninum enzyme is identical to either the type I or the type II NTPase. Amino acids indicated by * are believed to be involved in substrate binding.

282

ASAI ET AL.

T. gondii. Significantly, neither the recombinant or the native enzymes from N. caninum had measurable hydrolytic activity for diphosphate nucleosides (Table II). The Km value for MgATP22 hydrolysis by the N. caninum enzymes was also quite similar to that of the recombinant or native T. gondii NTPase-I (Table III). If N. caninum tachyzoites contain a NTPase-II-like enzyme, its activity was undetectable; therefore, a Km value for MgADP2 hydrolysis could not be determined for either the recombinant or the native NTPases of N. caninum.

DISCUSSION

FIG. 4. Southern blot analysis of NcNTP in Nc-1 strain genomic DNA. Genomic DNA was treated with the indicated restriction endonucleases, electrophoresed in an agarose gel, transferred to nitrocellulose, and probed with a 32P-labeled probe corresponding to the open reading frame of the NTP gene. The presence of multiple bands in digests which are predicted not to cut within the gene (i.e., EcoRV) indicate the presence of three distinct copies.

from both genomic and cDNAs, thus confirming that both forms of the gene are transcribed in tachyzoites (data not shown). Characterization of enzyme activities. Analysis of NTPase activities in crude cell homogenates indicated that N. caninum isolates contained a potent nucleoside triphosphate hydrolase activity that had previously been seen only in type I strains of T. gondii (Table 1). To better characterize the enzyme activities in N. caninum, we examined the substrate specificities and kinetic properties of partially purified native enzyme from tachyzoites of the Nc-1 strain and recombinant NcNTPase expressed in E. coli. A comparison of substrate specificities for the N. caninum and T. gondii NTPases is shown in Table II. Like the T. gondii NTPase-I isozyme, a broad substrate specificity toward nucleoside triphosphates was observed for both the recombinant NcNTPase and the partially purified native enzyme. The N. caninum enzymes hydrolyzed all nucleoside triphosphates (1 mM) at nearly the same rate; thus, the N. caninum enzymes exhibited properties similar to those of the previously characterized NTPase-I of

We have identified a potent NTPase in N. caninum that is antigenically cross-reactive to the NTPases of T. gondii and immunolocalizes to the dense granules of tachyzoites. NTPase activity in N. caninum is encoded by multiple genes, and sequence analysis of two independent clones revealed them to be most similar to the NTPase-I isoform of T. gondii. Consistent with this finding, NTPase activity in N. caninum exhibited enzymatic characteristics that were nearly identical to those of T. gondii NTPase-I. Significantly, a NTPase-IIlike activity for hydrolysis of NDPs was not detected in lysates of N. caninum. NTPase activity was detected in cell homogenates of five different N. caninum isolates, although three of the five N. caninum isolates had activities that were only about 50% of that seen in the RH strain of T. gondii (Table I). The observed NTPase activities in the cell homogenates were also quite variable from different cell pellets of the same isolate, as indicated by the relatively high standard deviations (Table I), suggesting that this difference was due to variability in growth or harvest. ImmunoEM analysis indicated that the NcNTPase was localized to the dense granule organelles in N. caninum (Fig. 2), and it has been demonstrated for T. gondii that dense granule proteins, like the NTPases, are constitutively secreted by extracellular parasites (Carruthers and Sibley 1997). Thus, the enzymatic activities detected in the cell homogenates could be significantly affected by the constitutive secretion and the subsequent levels of enzyme remaining in the dense granules at the time of parasite harvest. It has previously been shown that heterologous expression of the T. gondii NTPase-I (NTP3 isoform of Bermudes et al. 1994) in N. caninum resulted in correct targeting to the dense granule and secretion into the parasitophorous vacuole (Beckers et al. 1997). While we have not examined

283

N. caninum NTPase EXPRESSION

TABLE II Comparison of Substrate Specificity of N. caninum and T. gondii (RH) NTPase N. caninum (%)

Nucleotide ATP GTP ITP UTP CTP TTP dATP dGTP dUTP dCTP ADP GDP UDP CDP a

T. gondii (%)

Recombinant NcNTPase

Native NcNTPase

Recombinant NTPase-I

100a 104 112 92 87 87 105 105 85 83 ,0.1 ,0.1 ,0.1 ,0.1

100 105 110 87 86 86 105 104 86 82 ,0.1 ,0.1 ,0.1 ,0.1

100 103 115 98 91 90 103 104 90 81 0.9 0.9 0.6 0.6

Native NTPase-I

NTPase-II

100 102 109 96 91 91 105 106 90 82 1.0 0.9 0.6 0.6

100 98 97 91 90 87 100 106 90 80 97 96 65 70

The rate of ATP hydrolysis was taken as 100%.

the secretion of NcNTPase, it is likely also secreted into the parasitophorous vacuole following host cell invasion. The amino acid sequences of NcNTPase encoded by either pcNcNTP or prNcNTP were more similar to NTPase-I than NTPase-II of T. gondii. The two T. gondii NTPase isoforms share about 97% identity to one another, with 16 polymorphisms distributed throughout the 628-amino-acid sequence of the proteins (Bermudes et al. 1994; Asai et al. 1995). Among these 16 polymorphisms, NcNTPase shares 6 amino acids in common with NTPase-I and 4 amino acids in common with NTPase-II (Fig. 3). Since the recombinant NcNTPase has NTPase-I-like activity (see Tables II and III),

these data provide insight into the amino acid residues that determine the substrate specificities of these enzymes. The sequence QLIGAGKR at residues 101–107 of NTPase-I most closely matches the P loop motif found in many proteins that bind ATP or GTP. In NTPase-II, the amino acids arginine and glutamic acid are found at positions 104 and 105, and it has previously been speculated that these differences may explain the lower efficiency of ATP hydrolysis by NTPase-II (Asai et al. 1995). This hypothesis is supported by the current study, since the amino acids at positions 104 and 105 of NcNTPase are identical to those of NTPase-I, with which it shares hydrolytic properties (Fig. 3). Similarly,

TABLE III Comparison of Km Values for MgATP22 and MgADP2 Hydrolysis by N. caninum NcNTPase and T. gondii (RH) NTPase N. caninum

Substrate MgATP MgADP a b

T. gondii Native

Recombinant NcNTPase

Native NcNTPase

Recombinant NTPase-I

NTPase-I

NTPase-II

0.18 6 0.03 NDb

0.17 6 0.03 ND

0.14 6 0.02 1.20 6 0.20

0.12 6 0.02 1.20 6 0.20

0.50 6 0.05 0.50 6 0.07

a

Activity was too low to determine Km value. Mean 6 SE, using six different concentrations of substrates.

284 the extremely inefficient hydrolysis of ADP by NcNTPase and NTPase-I in T. gondii is presumably due to sequence differences between these alleles and NTPase-II. The basis for these differences will require further testing by sitedirected mutagenesis, which should be readily accomplished using the recombinant enzymes. The NTPase-I enzyme was previously observed to be restricted to the type I virulent strains of T. gondii, and it was suggested that the highly active hydrolysis of nucleoside triphosphates might contribute to the virulence phenotype (Asai et al. 1995). N. caninum tachyzoites also contain a NTPase-I-like enzyme that efficiently hydrolyzes NTPs, yet this parasite is extremely avirulent in the mouse model of infection. It is interesting to note that the NTPase-II isoform, which efficiently hydrolyzes diphosphate nucleosides, is present in all strains of T. gondii that have been tested (Asai et al. 1995), yet this enzyme activity was not detected in N. caninum (Tables II and III). Protozoan parasites are purine auxotrophs, and it has been speculated that the NTPases of T. gondii are involved in the salvage of purine nucleosides from the host cell (Bermudes et al. 1994). The data presented here indicate that the primary role of the NTPase of N. caninum tachyzoites may not involve purine acquisition, since the parasite lacks enzymatic activities for progressively cleaving nucleoside bases to their monophosphate form. Thus, the NcNTPase may serve an alternative function. We have previously shown that following secretion in T. gondii, the NTPase changes from a soluble form in the parasitophorous vacuole into an insoluble form, presumably through the association with other components in the vacuole (Carruthers and Sibley 1997). This suggests the NTPase provides energy necessary for transport within the vacuole, either in obtaining nutrients from the host cell or in mediating targeted insertion of GRA proteins into the PV membrane and the intravacuolar network. Alternatively, if the NTPases are involved in purine salvage in T. gondii, N. caninum would appear to be partially defective in this pathway, and this deficiency could conceivably affect its capacity to infect the vertebrate host. For example, the inability of Nc-1 to successfully establish chronic infections in outbred mice may be related to an inability to grow under nutrient-limited conditions in vivo. Whether this difference in survival is related to a lack of NTPase-II activity is presently under investigation.

ACKNOWLEDGMENTS We thank Marilyn Levy for expert technical assistance with electron microscopy and Dr. Helen Profous-Juchelka, Merck Research Labs,

ASAI ET AL.

for providing extracts of E. tenella. This work was supported in part by funds from the USDA (97-35204-4770), a National Grant-in-Aid from the Ministry of Education of Japan (10670236), the Japan Society for the Promotion of Science (JSPS-RFTF97L00701), and the Promotion of AIDS Research from The Ministry of Health and Welfare of Japan.

REFERENCES

Asai, T., O’Sullivan, W. J., and Tatibana, M. 1983. A potent nucleoside triphosphate hydrolase from the parasitic protozoan Toxoplasma gondii. Journal of Biological Chemisty 258, 6816–6822. Asai, T., Kanazawa, T., Kobayashi, S., Takeuchi, T., and, Kim, T. 1986. Do protozoa conceal a high potency of nucleoside triphosphate hydrolysis present in Toxoplasma gondii. Comparative Biochemistry and Physiology 85B, 365–368. Asai, T., and Suzuki, Y. 1990. Remarkable activities of nucleoside triphosphate hydrolase in the tachyzoites of both virulent and avirulent strains of Toxoplasma gondii. FEMS Microbiology Letters 72, 89–92. Asai, T., Miura, S., Sibley, L. D., Okabayashi, H., and Takeuchi, T. 1995. Biochemical and molecular characterization of nucleoside triphosphate hydrolase isozymes from the parasitic protozoan Toxoplasma gondii. Journal of Biological Chemistry 270, 11391–11397. Beckers, C. J. M., Wakefield, T., and Joiner, K. A. 1997. The expression of Toxoplasma proteins in Neospora caninum and the identification of a gene encoding a novel rhoptry protein. Molecular and Biochemical Parasitology 89, 209–223. Bermudes, D., Peck, K. R., Afifi, M. A., Beckers, C. J. M., and Joiner, K. A. 1994. Tandemly repeated genes encode nucleoside triphosphate hydrolase isoforms secreted into the parasitophorous vacuole of Toxoplasma gondii. The Journal of Biological Chemistry 269, 29252–29260. Carruthers, V. B., and Sibley, L. D. 1997. Sequential protein secretion from three distinct organelles of Toxoplasma gondii accompanies invasion of human fibroblasts. European Journal of Cell Biology 73, 114–123. Devereaux, J., Haeberli, P., and Smithies, O. 1984. A comprehensive set of sequence analysis programs for the VAX. Nucleic Acids Research 12, 387–295. Dubey, J. P., and Lindsay, D. S. 1996. A review of Neospora caninum and neosporosis. Veterinary Parasitology 67, 1–59. Dubey, J. P., Hattel, A. L., Lindsay, D. S., and Topper, M. J. 1988. Neonatal Neospora caninum infection in dogs: Isolation of the causative agent and experimental transmission. Journal of the American Veterinary Medical Association 193, 1259–1263. Howe, D. K., and Sibley, L. D. 1997. Development of molecular genetics for Neospora caninum: A complementary system to Toxoplasma gondii. Methods: A Companion to Methods in Enzymology 13, 123–133.

285

N. caninum NTPase EXPRESSION

Lindsay, D. S., and Dubey, J. P. 1989. Neospora caninum (Protozoa: Apicomplexa) infections in mice. Journal of Parasitology 75, 772– 779.

Sibley L. D., and Boothroyd, J. C. 1992. Construction of a molecular karyotype for Toxoplasma gondii. Molecular and Biochemical Parasitology 51, 291–300.

Lindsay D. S., Speer C. A., Toivio-Kinnucan M. A., Dubey J. P., and Blagburn B. L. 1993. Use of infected cultured cells to compare ultrastructural features of Neospora caninum from dogs and Toxoplasma gondii. American Journal of Veterinary Research 54, 103–106.

Sibley, L. D., Niesman, I. R., Asai, T., and Takeuchi, T. 1994. Toxoplasma gondii: Secretion of a potent nucleoside triphosphate hydrolase into the parasitophorous vacuole. Experimental Parasitology 79, 301–311.

Marsh, A. E., Barr, B. C., Sverlow, K., Ho, M., Dubey, J. P., and Conrad, P. A. 1995. Sequence analysis and comparison of ribosomal DNA from bovine Neospora to similar coccidial parasites. Journal of Parasitology 81, 530–535.

Speer, C. A., and Dubey, J. P. 1989. Ultrastructure of tachyzoites, bradyzoites, and tissue cysts of Neospora caninum. Journal of Protozoology 36, 458–463. Received 4 May 1998; accepted with revision 24 July 1998