Toxoplasma gondii Attachment to Host Cells Is Regulated by a Calmodulin-like Domain Protein Kinase*

The role of calcium-dependent protein kinases in the invasion of Toxoplasma gondii into its animal host cells was analyzed. KT5926, an inhibitor of calcium-dependent protein kinases in other systems, is known to block the motility of Toxoplasma tachyzoites and their attachment to host cells. In vivo, KT5926 blocks the phosphorylation of only three parasite proteins, and in parasite extracts only a single KT5926-sensitive protein kinase activity was detected. This activity was calcium-dependent but did not require calmodulin. In a search for calcium-dependent protein kinases in Toxoplasma, two members of the class of calmodulin-like domain protein kinases (CDPKs) were detected. TgCDPK2 was only expressed at the mRNA level in tachyzoites, but no protein was detected. TgCDPK1 protein was expressed in Toxoplasmatachyzoites and cofractionated precisely with the peak of KT5926-sensitive protein kinase activity. TgCDPK1 kinase activity was calcium-dependent but did not require calmodulin or phospholipids. TgCDPK1 was found to be inhibited effectively by KT5926 at concentrations that block parasite attachment to host cells.In vitro, TgCDPK1 phosphorylated three parasite proteins that migrated identical to the three KT5926-sensitive phosphoproteins detected in vivo. Based on these observations, a central role is suggested for TgCDPK1 in regulatingToxoplasma motility and host cell invasion.

The role of calcium-dependent protein kinases in the invasion of Toxoplasma gondii into its animal host cells was analyzed. KT5926, an inhibitor of calcium-dependent protein kinases in other systems, is known to block the motility of Toxoplasma tachyzoites and their attachment to host cells. In vivo, KT5926 blocks the phosphorylation of only three parasite proteins, and in parasite extracts only a single KT5926-sensitive protein kinase activity was detected. This activity was calcium-dependentbutdidnotrequirecalmodulin.Inasearchforcalciumdependent protein kinases in Toxoplasma, two members of the class of calmodulin-like domain protein kinases (CDPKs) were detected. TgCDPK2 was only expressed at the mRNA level in tachyzoites, but no protein was detected. TgCDPK1 protein was expressed in Toxoplasma tachyzoites and cofractionated precisely with the peak of KT5926-sensitive protein kinase activity. TgCDPK1 kinase activity was calcium-dependent but did not require calmodulin or phospholipids. TgCDPK1 was found to be inhibited effectively by KT5926 at concentrations that block parasite attachment to host cells. In vitro, TgCDPK1 phosphorylated three parasite proteins that migrated identical to the three KT5926-sensitive phosphoproteins detected in vivo. Based on these observations, a central role is suggested for TgCDPK1 in regulating Toxoplasma motility and host cell invasion.
Toxoplasma gondii is an obligate intracellular parasite of the phylum Apicomplexa. Infection with this parasite is typically asymptomatic, although acute toxoplasmosis can be fatal in immunocompromised individuals and result in severe birth defects or abortion during the first trimester of pregnancy (1). Three developmental stages of Toxoplasma have been described. The highly infective sporozoite stage is shed in the feces of infected felines and can infect all warm-blooded animals, where it differentiates into the rapidly replicating tachyzoite stage. This form of the parasite rapidly spreads throughout the infected animal and eventually differentiates into the slow growing, encysted bradyzoite stage. The latter can infect another animal upon ingestion of infected tissues. The three stages are immunologically and biochemically distinct, due to the expression of many stage-specific proteins (2). All three developmental stages are obligate intracellular parasites, suggesting that infected host cells supply nutrients that supplement T. gondii biosynthetic deficiencies, such as its inability to make purines de novo (3). As a result of these biosynthetic requirements, invasion of a host cell is critical for the growth and reproduction of T. gondii.
The biochemical pathways involved in host cell invasion have not yet been identified, although it has been shown that cytoplasmic calcium in the parasite is essential for this process (4,5). The involvement of calcium in regulating parasite interaction with host cells was further strengthened by the observation of Sibley et al. that both the attachment of Toxoplasma tachyzoites to its host cells as well as parasite motility are sensitive to an inhibitor of calcium-dependent protein kinases, KT5926 (6,7).
In eukaryotic cells, protein kinases often mediate the cellular responses to external stimuli. Three major classes of protein kinases are often involved in this: cyclic nucleotide-dependent protein kinases, calcium/calmodulin-dependent protein kinases, and calcium/phospholipid-dependent protein kinases. In a number of organisms, an unusual class of calcium-dependent protein kinases has been described. These enzymes, the calmodulin-like domain protein kinases (CDPKs), 1 are activated by calcium in the absence of calmodulin or phospholipids. Initially identified, characterized, and cloned in plants (8 -10), CDPKs have also been identified in algae (11,12), Paramecium tetraurelia (13)(14)(15), and the apicomplexan parasites Plasmodium falciparum (16,17), Eimeria maxima, and Eimeria tenella (18).
The typical CDPK domain structure consists of an Nterminal serine/threonine kinase domain homologous to that of calcium/calmodulin-dependent protein kinases, followed by a highly conserved junction domain, which joins the kinase region to a C-terminal calmodulin-like domain (10). This calmodulin-like domain, which is 30 -40% homologous to calmodulin, imparts calcium sensitivity to the CDPKs (19,20). Activation of CDPKs has been shown to be dependent on calcium and independent of calmodulin (8). The elucidation of the function of specific CDPKs has been complicated, in part, by the fact that often multiple isoforms are expressed at the same time (19). Three soybean isoforms were found to differ in sensitivity to calcium and substrate specificity (19), suggesting that they are involved in the regulation of different phenomena. In plants, CDPKs have been found to regulate Ca 2ϩpumps (21), K ϩ -transport (22), and a Cl Ϫ channel (23) and have also been implicated in the response to environmental stress * This work was supported by National Institutes of Health Grant AI41765 and a New Investigator Award from the Burroughs Wellcome Fund. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
The nucleotide sequence(s) reported in this paper has been submitted to the GenBank TM  and infections (24,25). The role of CDPKs in protists and apicomplexan parasites is unclear at this time.
In light of the importance of calcium-dependent regulation in host cell invasion by Toxoplasma tachyzoites and the suggestion that this might be affected in the presence of KT5926, we performed an analysis of calcium-dependent protein kinase activities in tachyzoites and their sensitivity to KT5926. We describe here the identification of two CDPKs, TgCDPK1 and TgCDPK2, in T. gondii. Of these, only TgCDPK1 is expressed in the tachyzoite stage of the parasite. This enzyme appears to be the target of KT5926 in tachyzoites and is therefore likely to play a role in their invasion of host cells.

EXPERIMENTAL PROCEDURES
Growth and Isolation of Parasites-T. gondii strain RH(EP) was obtained from D. S. Roos (University of Pennsylvania) and was maintained by serial passage in human foreskin fibroblasts or in the peritoneal cavity of Swiss/Webster mice as has been described (26). Parasites were harvested from the culture supernatant or the peritoneal fluid, washed by centrifugation, and filtered through a 3-m filter before use.
Attachment and Motility Assays-These were essentially performed as described by Dobrowolski et al. (7). For attachment and invasion assays, parasites were added to confluent coverslips of human foreskin fibroblast cells at a concentration of 10 6 /ml in Hanks' balanced salt solution containing 0.1% bovine serum albumin in the presence of increasing concentrations of KT5926 (Calbiochem) or Me 2 SO. Attachment and invasion were allowed to occur for 15 min at 37°C. All subsequent operations were performed on ice. The coverslips were gently washed three times in cold phosphate-buffered saline to remove any unattached parasites. Extracellular parasites were detected by incubation with an anti-Toxoplasma rabbit antiserum in phosphatebuffered saline containing 3% bovine serum albumin. After a 30-min incubation, the coverslips were washed as before. Cells were fixed in 3% paraformaldehyde in phosphate-buffered saline, washed, and permeabilized for 5 min in ice-cold methanol. Both extracellular and intracellular parasites were detected using the mouse monoclonal antibody T41E5 (obtained from Dr. J. F. Dubremetz (Institut Pasteur, Lille, France)) to the parasite surface marker SAG1. Cells were washed and incubated in the secondary antibody solution containing fluoresceinconjugated goat anti-rabbit antibody and rhodamine-conjugated goat anti-mouse antibody and 5 g/ml 4Ј,6-diamidino-2-phenylindole to label nuclei. After washing and mounting, the number of attached parasites, total parasites, and host cell nuclei were counted in eight individual fields at ϫ 1000 magnification. All incubations were performed in triplicate. Attachment and invasion were expressed as the number of events observed per host cell nucleus.
Preparation and Fractionation of Toxoplasma Extracts-All manipulations were performed on ice or at 4°C. Parasites, prepared as described above, were resuspended to a final concentration of 10 9 parasites/ml in 25 mM HEPES-KOH, 50 mM KCl, 1 mM phenylmethylsulfonyl fluoride, 50 g/ml antipain, 10 g/ml leupeptin, and 10 g/ml aprotinin at pH 7.5. Parasites were disrupted either by two rapid freeze-thaw cycles, by brief sonication, or by the addition of Triton X-100 to a final concentration of 1%. Lysates were clarified by centrifugation for 10 min at 13,000 ϫ g. Heattreated extracts were prepared by incubation at 70°C for 10 min, and denatured proteins were removed by centrifugation.
A 2-ml aliquot of a parasite lysate prepared by sonication was loaded onto a 2 ϫ 1-cm DEAE-Sepharose (Amersham Pharmacia Biotech) column equilibrated in 10 mM Tris-HCl, pH 8, 50 mM KCl. The column was washed with 10 ml of equilibration buffer, and bound protein was eluted with a 20-ml gradient of 50 -500 mM KCl in 10 mM Tris-HCl, pH 8. Each 1-ml fraction was analyzed for the presence of protein kinase activity as described below. The three fractions containing the peak of calcium-dependent, KT5926-sensitive protein kinase activity were pooled, and KCl was added to a final concentration of 2 M. This was loaded onto a 2 ϫ 1-cm phenyl-Sepharose (Amersham Pharmacia Biotech) column equilibrated in 10 mM Tris-HCl, pH 8, 2 M KCl. The column was washed with 10 ml of equilibration buffer, and bound proteins were eluted with a 20-ml gradient of 2 to 0 M KCl in 10 mM Tris-HCl, pH 8, and protein kinase activity in each fraction was determined as described below.
Protein Kinase Assay-Kinase assays were performed in a total volume of 20 l containing 20 mM HEPES-KOH (pH 7.5), 10 mM MgCl 2 , 2 mM EGTA, 1.93 mM CaCl 2 (2.5 M free calcium), 1 mg/ml bovine serum albumin, and 0.3 g/ml (ϳ5 nM) purified TgCDPK1. The purified enzyme was diluted in buffer containing 20 mM HEPES-KOH (pH 7.5), 10 mM MgCl 2 , and 1 mg/ml bovine serum albumin. For assays during enzyme purification, 0.2 mg/ml histones III-S and 90 M [␥-32 P]ATP (ϳ200 Ci/mol) (Amersham Pharmacia Biotech) were used. To assess calcium-dependent activity of the enzyme, the reactions were performed in the presence of 1 mM EGTA, or the concentration of free calcium reactions was controlled by a series of buffers containing a ratio of EGTA/CaCl 2 according to Fabiato (27). Where indicated, KT5926, calphostin C, and fluphenazine-N-2-chloroethane (all from Calbiochem) were included in the kinase reaction at the indicated concentrations. For kinetic analysis, the concentrations of peptide/protein substrates and [␥-32 P]ATP were varied according to Table I. Control reactions for ATP titrations contained 25 mM EGTA, while controls for peptide titrations lacked peptide substrate. Reactions were incubated at 30°C for 10 min, transferred immediately to an ice-water bath, and sampled by spotting 10 l on a 2.5-cm P-81 filter paper (Whatman). Filters were air-dried and washed with constant shaking for 30 min in five changes of 75 mM o-phosphoric acid followed by 5 min in 100% acetone. Filters were air-dried, and Cerenkov radiation was quantitated using a liquid scintillation counter.
Values for K m and V max were obtained by analyzing data according to Eadie-Hofstee and Lineweaver-Burk.
Amplification of Protein Kinase-related Sequences from Toxoplasma cDNA-Total RNA was isolated from parasites with the Trizol reagent (Life Technologies) using the conditions suggested by the manufacturer. First strand cDNA was synthesized using SuperScript reverse transcriptase (Stratagene, La Jolla, CA) and either random hexamers or oligo(dT) according to the manufacturer's directions. The resulting cDNA was analyzed for the presence of protein kinase-related sequences using PCR and two primers derived from conserved domains present in a wide variety of serine/threonine protein kinases: TgPK1 (5Ј-CGGATCCAYMGIGAYYT-3Ј) and TgPK2 (5Ј-GGAATTCCRWAR-GACCAIACRTC-3Ј). A typical 100-l PCR contained 2.5 units of Taq polymerase, 100 pmol of primers TgPK1 and -2, 50 ng of cDNA, 200 M dNTPs, 1.5 mM MgCl 2 , 50 mM KCl, 10 mM Tris-HCl, pH 8.3, and 0.001% gelatin. The following PCR conditions were used: 30 cycles of 1 min at 95°C, 1 min at 37°C, and 1 min at 72°C, followed by a 10-min incubation at 72°C. PCR fragments of the expected size were isolated from agarose gels and cloned directly into pGEM-T vector (Promega, Madison, WI) according to the manufacturer's instructions. Clones containing inserts were identified, and the inserts were sequenced manually with Sequenase (Amersham Pharmacia Biotech).
A Toxoplasma cDNA library in ZAPII (AIDS Research and Reference Reagent Program, McKesson Biosciences, Rockville, MD) was screened with TgCDPK1-or TgCDPK2-specific probes. Positive phage clones were converted to plasmids using in vivo excision according to the manufacturer's instructions (Stratagene). The resulting clones were analyzed by restriction enzyme digestion and automated DNA sequence analysis (Keck Biotechnology Resource Laboratory, Yale University, New Haven, CT).
Phylogenetic Analysis of TgCDPK Sequences-Sequences were aligned employing the ClustalW algorithm, available through the Baylor University site on the World Wide Web. The phylogeny was developed from the aligned sequence data employing the branch and bound algorithm available in the PAUP program package (28). In undertaking the analysis, gaps introduced during the alignment process were not considered as character states, all characters were equally weighted, and the phylogenies were not rooted. A single most parsimonious tree was identified in the branch and bound algorithm, which is shown here. Of the 632 characters analyzed, 363 were informative. The statistics for the tree are as follows: length, 2015 (minimum 1752, maximum 2672); consistency index, 0.869; rescaled consistency index, 0.621. Numbers at the nodes indicate the percentage the grouping distal to the labeled node was identified in a bootstrap analysis of 1000 replicate data sets.
Production of Antisera to TgCDPK1 and TgCDPK2-The entire open reading frames of TgCDPK1 and -2 were amplified using the following primers: 5Ј-GAAGATCTGATGGGGCAGCAGGAAAGCAC-3Ј and 5Ј-C-CAAGCTTTAGTTTCCGCAGAGCTTC-3Ј for TgCDPK1; 5Ј-CGGGATC-CCGCATCACCAGTGCAGCACC-3Ј and 5Ј-CCAAGCTTTTACCCCGT-AGCGCGAGG-3Ј for TgCDPK2 with PfuTurbo DNA polymerase (Stratagene) as described by the manufacturer. Amplified products were digested with BglII (TgCDPK1) or BamHI (TgCDPK2) and Hin-dIII and cloned into the bacterial expression vector pRSETB digested with BamHI and HindIII, resulting in the plasmids pRSETB-CDPK1 and pRESTB-CDPK2. These plasmids encode the open reading frame of the protein kinases fused at their N terminus to a polyhistidine tag and Xpress tag. For the expression of recombinant protein, these plasmids were transfected into JM109 (DE3). Overnight cultures were diluted 1:100 in 200 ml of fresh LB medium containing 50 g/ml ampicillin and grown at 37°C until A 600 was 0.4 -0.6. Isopropyl-1-thio-␤-D-galactopyranoside was added to a final concentration of 1 mM, and growth was continued for an additional 4 h at 37°C. Bacteria were harvested, lysed in 6 M guanidine HCl in 10 mM Tris-HCl, pH 8.0, and loaded onto 2 ml of Ni 2ϩ -nitrilotriacetic acid resin (Qiagen). The resin was washed with 20 ml of lysis buffer and 20 ml of 8 M urea, 10 mM Tris-HCl, pH 8.0. The polyhistidine-tagged fusion proteins were eluted using a linear 0 -250 mM imidazole gradient in 8 M urea, 10 mM Tris-HCl, pH 8.0. Fractions containing pure fusion proteins were dialyzed overnight against phosphate-buffered saline and used as antigen.
Expression and Purification of Enzymatically Active TgCDPK1-The entire open reading frame of TgCDPK1 was amplified using the primers 5Ј-GGAATTCCATATGGGGCAGCAGGAAAGCAC-3Ј and 5Ј-CCAAGC-TTTAGTTTCCGCAGAGCTTC-3Ј and PfuTurbo as described above. The amplified product was purified by agarose electrophoresis, digested with the restriction enzymes NdeI and HindIII, and repurified by agarose gel electrophoresis. The expression plasmid pRSET-B (Invitrogen, Carlsbad, CA) was digested with the restriction enzymes NdeI and HindIII and purified by agarose gel electrophoresis. The amplification product was ligated into pRSETB, resulting in a construct (pRSETB-TgCDPK1) that encoded only TgCDPK1 without extraneous sequences. A 20-ml overnight culture of Escherichia coli JM109(DE3) containing the plasmid pRSETB-TgCDPK1 was used to inoculate 2 liters of LB medium containing 50 g/ml ampicillin and grown at 37°C until an A 600 of 0.8 was reached. Isopropyl-1-thio-␤-D-galactopyranoside was added to a final concentration of 1 mM, and the culture was incubated for an additional 5 h at 37°C. All subsequent manipulations were performed at 0 -4°C. The pellet (8.1 g) was resuspended in 40 ml of ice-cold lysis buffer (50 mM Tris-HCl (pH 7.6), 25 mM NaCl, 1 mM EDTA, and 1 mM phenylmethylsulfonyl fluoride). Lysozyme was added to a final concentration of 1 mg/ml, and after a 10-min incubation the lysate was sonicated on ice. After centrifugation, the clarified supernatant was frozen in liquid N 2 and stored at Ϫ80°C. One-fourth of the frozen supernatant was thawed, and (NH 4 ) 2 SO 4 was added to 40% saturation. After 1 h, the precipitate was removed by centrifugation at 10,000 ϫ g for 15 min. Additional (NH 4 ) 2 SO 4 was added to the supernatant to 60% saturation. After 1 h, the precipitate was collected by centrifugation at 10,000 ϫ g for 15 min. The pellet was resuspended in 3 ml of lysis buffer and dialyzed overnight against two changes of 2 liters of lysis buffer. The dialyzed material was clarified by centrifugation and applied to a DEAE-Sepharose anion exchange column (3 ϫ 2 cm) equilibrated with 10 mM Tris-HCl, pH 7.6, 50 mM NaCl. The column was washed with 30 ml of this buffer, and bound proteins were eluted with a 150-ml linear gradient of NaCl (50 -500 mM) in 10 mM Tris-HCl, pH 7.6. Fractions (5 ml) were analyzed for kinase activity as described. The fractions containing the peak of kinase activity were pooled and frozen in liquid nitrogen. Calcium-dependent hydrophobic interaction chromatography was used as the final purification step. The peak fractions of kinase activity were pooled, and CaCl 2 was added to a final concentration of 2 mM. This was applied to a (2.5 ϫ 2 cm) phenyl-Sepharose column equilibrated with loading buffer containing 10 mM Tris-HCl pH 7.6, 50 mM NaCl, and 2 mM CaCl 2 . The column was washed with 45 ml of loading buffer followed by 45 ml of wash buffer (10 mM Tris-HCl, pH 7.6, 50 mM NaCl, and 0.1 mM CaCl 2 ). The enzyme was eluted with buffer containing 10 mM Tris-HCl, pH 7.6, 50 mM NaCl, and 2 mM EGTA. Fractions (2.5 ml) were assayed for kinase activity as described. Purity of the recombinant protein was assessed by SDS-PAGE.
SDS-PAGE and Immunoblotting-Proteins were separated by SDS-PAGE on 10 or 12% polyacrylamide gels. Following electrophoresis, proteins were transferred to nitrocellulose membrane for 60 min at 100 V. The nitrocellulose membrane was blocked in Tris-buffered saline containing 5% nonfat dry milk and 0.1% Tween 20. Rabbit or mouse antisera raised against purified TgCDPK1 and TgCDPK2 was used at a 1:2000 dilution in the blocking buffer. Bound antibodies were detected using horseradish peroxide-conjugated secondary antisera and the Su-perSignal detection system (Pierce).
Protein Quantitation-Protein concentrations were measured by the method of Bradford (29) using bovine serum albumin as a standard.

KT5926 Blocks Parasite Motility and Cell Attachment-It
was reported by Dobrowolski et al. (7) that the protein kinase inhibitor KT5926 blocked the ability of T. gondii to attach to their host cell and to move over substrates. We confirmed these findings, as can be seen in Fig. 1, and established that the IC 50 of KT5926 for Toxoplasma attachment is ϳ100 nM. KT5926 is a selective inhibitor of calmodulin-dependent and myosin light chain kinases in animal cells.
KT5926 Is a Selective Inhibitor of Toxoplasma Protein Phosphorylation in Vivo-To determine the extent to which KT5926 blocks protein phosphorylation in T. gondii, tachyzoites were incubated with inorganic radioactive phosphate in the presence or absence of KT5926. As a control, we also performed an in vivo phosphorylation experiment in the presence of staurosporine, a general inhibitor of serine/threonine protein kinases. To obtain the highest possible sensitivity, the effect of the inhibitor on incorporation of [ 32 P]PO 4 3Ϫ in proteins was assessed by two-dimensional gel electrophoresis. As expected, the addition of staurosporine blocks incorporation of 32 P into most proteins (data not shown). The addition of KT5926 to parasites, as seen in Fig. 2, selectively blocks the incorporation of phosphate in three major proteins: PP1 (M r (app) ϭ 67,000, pI ϭ 6.7), PP2 (M r (app) ϭ 31,000, pI ϭ 5.6), and PP3 (M r (app) ϭ 6500, pI ϭ 4.7). These data suggest that the effect of KT5926 on protein phosphorylation in T. gondii is quite selective and might be due to an inhibition of one or only a few protein kinases. Although the IC 50 of KT5926 for parasite attachment is only 100 nM, the in vivo phosphorylation experiment in Fig. 2 was performed in the presence of 10 M inhibitor to exacerbate its effects on protein phosphorylation. A similar effect of KT5926 on protein phosphorylation was observed at a concentration of 1 M.
A Calcium-dependent Protein Kinase Activity Is Inhibited by KT5926 -To characterize the effects of KT5926 on protein kinase activity in Toxoplasma tachyzoites further, we analyzed parasite extracts for the presence of protein kinase activities. As can be seen in Fig. 3, protein kinase activity is readily detected in Toxoplasma extracts using histones as substrates. Since KT5926 is thought to be specific for calcium-dependent protein kinase (6), we determined whether the presence or absence of calcium affects protein kinase activity in Toxoplasma extracts. In the presence of calcium, overall protein kinase activity in parasite extracts was stimulated 2-fold when compared with reactions performed in the presence of EGTA. These data suggest that Toxoplasma extracts do indeed contain one or more potent calcium-dependent protein kinases. The calcium-dependent increase in protein kinase activity was completely blocked in the presence of 500 nM KT5926, confirming the hypothesis that this inhibitor targets one or more calciumdependent protein kinases in Toxoplasma. The addition of recombinant Toxoplasma calmodulin, a calmodulin antagonist, or calphostin C had no effect on the KT5926-sensitive protein kinase activity. These observations suggest that Toxoplasma does not possess major calmodulin-dependent protein kinase or protein kinase C activities and that therefore the target of KT5926 may be a different class of calcium-dependent protein kinase.
To characterize the calcium-dependent protein kinase activity in Toxoplasma extracts further, we fractionated parasite extracts by ion exchange chromatography and analyzed each fraction for calcium-dependent and KT5926-sensitive protein kinase activity and the effect of KT5926 on that activity. As can be seen in Fig. 4A, soluble extracts of Toxoplasma contain a number of protein kinase activities that can phosphorylate histones, but only a single peak of calcium-dependent protein kinase activity was detected, and this activity was effectively inhibited in the presence of 500 nM KT5926. Further fractionation of this activity by hydrophobic interaction chromatography revealed only a single sharp peak of calcium-dependent, KT5926-sensitive protein kinase activity (Fig. 4B). As before, the addition of neither recombinant Toxoplasma calmodulin nor calmodulin antagonists affected any of the protein kinase activities detected in Fig. 4 (data not shown). Taken together, these observations suggest that Toxoplasma tachyzoites contain a single KT5926-sensitive protein kinase activity that is calcium-dependent but does not require calmodulin.
Identification of Two CDPKs in Toxoplasma-To identify possible candidates for the calcium-dependent, KT5926-sensitive protein kinase activity in Fig. 4, we performed PCR on Toxoplasma cDNA using primers that were known to allow amplification of the catalytic domain of a wide variety of serine/threonine protein kinases. The only candidates for calcium-dependent protein kinases identified in an exhaustive analysis of protein kinase sequences were two members of the family of CDPKs. No candidates for calmodulin-dependent protein kinases or protein kinase C were identified. Full-length clones were isolated from a cDNA library using both PCR products and characterized by DNA sequencing. Since the predicted amino acid sequences of both clones are highly homologous to known CDPKs, these are henceforth referred to a TgCDPK1 and TgCDPK2. Low stringency Southern blot analysis of Toxoplasma genomic DNA did not reveal any additional CDPK-related sequences (data not shown). The open reading frame of TgCDPK1 encodes a 507amino acid protein with a predicted molecular mass of 57.3 kDa. TgCDPK2 is predicted to be a 520-amino acid protein with a predicted molecular mass of 58.3 kDa.
The predicted sequences of both proteins contain the domains typically encountered in CDPKs: an N-terminal catalytic domain, a short autoinhibitory domain, and a C-terminal calcium-binding domain (Fig. 5A). The two CDPK isoforms of Toxoplasma are 36% identical to each other. TgCDPK1 is most homologous to the CDPKs described in the related apicomplexan parasites E. maxima and E. tenella and CDPK1 of P. falciparum. TgCDPK2 is more homologous to CDPK2 and -3 of P. falciparum and Arabidopsis thaliana CDPK4 and soybean CDPK-␥. Homologies extend along the entire length of the predicted protein sequences but are most pronounced in the catalytic domains (residues 1-298 in TgCDPK1 and 1-293 in TgCDPK2), where they range from 44 to 93%.
Motifs commonly found in protein kinases (30)  The evolutionary relationship of the Toxoplasma CDPKs to known CDPKs is shown in Fig. 5B. The CDPKs from higher plants and Paramecium form two clades that are distinct from each other and from the CDPKs from apicomplexans. The latter form two distinct clades. One contains TgCDPK1, the Eimeria CDPKs, and P. falciparum CDPK1; the second clade comprises TgCDPK2 and P. falciparum CDPK2 and CDPK3.
Expression of TgCDPK1 and -2 in Toxoplasma Tachyzoites-Monospecific antisera to both TgCDPK1 and -2 were prepared using recombinant proteins and were used to determine their expression in the parasite. The resulting antisera are highly specific for each of the TgCDPKs, and no cross-reactivity is observed (Fig. 6A). As can be seen in Fig. 6A, TgCDPK1 can be   FIG. 3. Toxoplasma extracts contain calcium-dependent and KT5926sensitive protein kinase activities but no calmodulin-dependent protein kinase activity. Parasite extracts were prepared, and the endogenous protein kinase activities were analyzed using histones III-S as substrate as described under "Experimental Procedures." Each 20-l reaction contained an amount of parasite extract equivalent to 5 ϫ 10 6 organisms and was performed in the presence of 2.5 M free calcium, 1 mM EGTA, . This resulted in the detection of a single peak of calcium-dependent, KT5926-sensitive protein kinase activity. Using immunoblot analysis, TgCDPK1 was shown to cofractionate precisely with the KT5926-sensitive protein kinase activity during chromatography on DEAE-Sepharose (C) and phenyl-Sepharose (D).

FIG. 5. Alignment of phylogenetic analysis of TgCDPK sequences.
A, the predicted protein sequences of TgCDPK1 and TgCDPK2 were aligned with the sequence of Glycine max CDPK␣ and human calmodulin. Boxes I and II mark two conserved sequence elements conserved between protein kinases and thought to be required for binding of ATP. Box III indicates a sequence element conserved between serine/threonine protein kinases and thought to be involved in catalysis. Box IV indicates the four EF-hand calcium-binding domains found in all CDPKs and calmodulin. The sequences of TgCDPK1 and TgCDPK2 have been deposited at GenBank TM under accession numbers AF333958 and AF333959. Sequences of Toxoplasma cDNA clones that are homologous to TgCDPK1 and TgCDPK2 were identified in the GenBank TM data base (accession numbers AAC02532 and AAD17247, respectively). B, TgCDPK1 and TgCDPK2 are members of two distinct classes of apicomplexan CDPKs. CDPK sequences most homologous to TgCDPK1 and -2 were identified using BLAST and FASTA, and a phylogenetic tree was generated as described under "Experimental Procedures." AtCDPK2, A. thaliana CDPK2 (BAA04830); AtCDPK4, A. thaliana CDPK4 (CAB82124); EmCDPK, E. maxima CDPK (CAA96438); EtCDPK, E. tenella CDPK (CAA96439); GmCDPK␥, G. max CDPK-␥ (AAB80693); PfCDPK1, P. falciparum CDPK1 (CAA47704); PfCDPK2, P. falciparum CDPK2 (CAA68090); PfCDPK3, P. falciparum CDPK3 (AAF63154); PtCDPKA, P. tetraurelia CDPK-A (AAC13356); PtCDPKB, P. tetraurelia CDPK-B (AAC13355).
readily detected in parasite extracts as a protein with an apparent molecular mass of 58 kDa, close to the predicted value of 57.3 kDa. The antiserum to TgCDPK2 failed to detect any protein in tachyzoite extracts, although it reacted very strongly with recombinant protein (Fig. 6A). To address the possibility that the TgCDPK2 open reading frame might be part of a pseudogene, we analyzed TgCDPK2 expression in tachyzoites using reverse transcription-PCR. As can be seen in Fig. 6B, the TgCDPK2 transcript can be readily detected by reverse transcription-PCR, suggesting that the TgCDPK2 gene is indeed transcribed but that expression of the encoded protein may be regulated at the post-transcriptional level. Since the TgCDPK2 protein does not appear to be expressed in Toxoplasma tachyzoites, it was not analyzed further.
TgCDPK1 Is a KT5926-sensitive, Calcium-dependent Protein Kinase-PurifiedrecombinantTgCDPK1demonstratedcalciumdependent protein kinase activity. To analyze activation of TgCDPK1 by calcium, in vitro protein kinase reactions were performed at different concentrations of free calcium (Fig. 7). Half-maximal activation of TgCDPK1 occurred at ϳ600 nM free calcium, and enzyme activity was maximal at calcium concentrations between 1 and 5 M. A slight inhibition of activity was observed at higher calcium concentrations. For all subsequent experiments, reactions were performed in the presence of 2.5 M free calcium. Activation of TgCDPK1 was solely dependent on the presence of calcium. The addition of 200 g/ml phosphatidyl serine to in vitro protein kinase reactions failed to stimulate the activity of TgCDPK1 beyond that seen with calcium alone (data not shown). The addition of recombinant Toxoplasma calmodulin also failed to enhance calcium-stimulated TgCDPK1 activity (data not shown).
The results of kinetic analysis of TgCDPK1 are shown in Table I. Several protein and peptide substrates commonly used in the characterization of protein kinases were investigated as possible TgCDPK1 substrates. Protamine sulfate and neurogranin peptide were very poor substrates. Histone III-S, which has been used as a substrate for plant CDPKs, was a poor substrate for TgCDPK1 (K m ϭ 3 mg/ml, V max ϭ 1 mol min Ϫ1 mg Ϫ1 ). While histone III-S initially proved useful in the identification and purification of a number of CDPKs, it has been noted that potential variation in peptide composition among different preparations of the substrate limits the reliability of kinetic comparisons (36). We found that the synthetic peptides, syntide-2 and the glycogen synthase peptide, provided more consistent kinetic measurements than histone III-S. We obtained values of 107 M and 5.3 mol min Ϫ1 mg Ϫ1 for K m and V max for the glycogen synthase peptide and found the K m to be 155 M and the V max to be 5.2 mol min Ϫ1 mg Ϫ1 with respect to ATP.
To characterize the enzymatic activity of TgCDPK1 further, we investigated the effects of a variety of inhibitors on the enzyme. Several inhibitors of serine/threonine kinases had no effect on the activity of TgCDPK1. Calphostin, an inhibitor of protein kinase C had an IC 50 greater than 10 M.  sin light chain kinase that are inhibited with K i values of 4 and 18 nM, respectively (6). As can be seen in Fig. 1, TgCDPK1 is less sensitive to KT5926 than either enzyme with half-maximal inhibition of protein kinase activity occurring at about 100 nM and complete inhibition at 1 M. Strikingly, however, TgC-DPK1 protein kinase activity and Toxoplasma attachment to host cells display a virtually identical sensitivity to KT5926 (Fig. 1), suggesting that TgCDPK1 is the main target of KT5926 in T. gondii and is required for the attachment of the parasite host cells.
To determine the relationship between TgCDPK1 and the major KT5926-sensitive protein kinase activity detected during fractionation of Toxoplasma tachyzoite extracts, we analyzed all chromatographic fractions by immunoblot analysis. As can be seen in Fig. 4, TgCDPK1 cofractionates exactly with the calcium-dependent, KT5926-sensitive protein kinase, suggesting that they are identical.
Proteins Phosphorylated in Vitro by TgCDPK1 Are Similar to KT5926-sensitive Phosphoproteins Identified in Vivo-To identify soluble endogenous substrates for TgCDPK1, in vitro protein phosphorylation reactions were performed and analyzed by two-dimensional gel electrophoresis and autoradiography. We first identified proteins within T. gondii cytosol that were phosphorylated as a result of endogenous calcium-dependent protein kinase activities. To this end, protein kinase reactions containing parasite cytosol were performed in the presence and absence of calcium, and the corresponding autoradiographs were compared (Fig. 8, A and B). Three major substrates of endogenous calcium-dependent protein kinase activity were identified as well as several minor ones.
To identify specifically the substrates for TgCDPK1, T. gondii cytosol was heated at 70°C for 10 min to inactivate endogenous protein kinases. As is evident from Fig. 8C, this treatment is very effective at inactivating the endogenous protein kinases. The addition of recombinant TgCDPK1 to in vitro protein kinase reactions containing heat-treated cytosol results in the calciumdependent phosphorylation of a number of proteins. The three major TgCDPK1 substrates appear identical to the major substrates of calcium-dependent protein kinase activity in parasite extracts. More importantly, with respect to molecular weight and isoelectric point, they appear to be identical to the three KT5926-sensitive phosphoproteins identified in Toxoplasma tachyzoites in vivo, PP1-3 (Fig. 2). In addition to the major substrates, heat-treated cytosol also contains a number of minor substrates for TgCDPK1. Since these are absent in untreated cytosol, they most likely represent proteins that were denatured during the heat treatment of cytosol, exposing cryptic TgCDPK1 phosphorylation sites in the process.
Although TgCDPK1 is a major protein kinase activity in parasite extracts (Fig. 4), comparison of Figs. 2 and 8 suggests that this enzyme actually represent only a small fraction of the total protein kinase activity found in Toxoplasma tachyzoites. This discrepancy suggests that other protein kinase activities are lost during preparation and fractionation of extracts or that they are inactive in the incubation conditions used to detect TgCDPK1. In addition, the removal or inactivation of one or more TgCDPK1specific protein phosphatase activities during the preparation and subsequent fractionation of the extracts would also result in an apparent increase in its relative activity. DISCUSSION Cytoplasmic calcium fluctuations have been implicated in the regulation of host cell invasion by the apicomplexan parasites Theileria parva (31), P. falciparum (32), and T. gondii tachyzoites (4,7,33), suggesting that this process is regulated by a common mechanism in apicomplexan parasites. Although the specific calcium-dependent regulatory mechanism involved has not been identified directly, inhibitor studies have implicated calmodulin and a myosin light chain kinase-like activity (5,7).
Here, we describe the identification of one calcium-dependent regulatory system required for Toxoplasma attachment to host cells as a member of the CDPK family. Two members of this family, TgCDPK1 and TgCDPK2, were identified in T. gondii. Whereas TgCDPK1 protein is expressed in tachyzoites, TgCDPK2 protein is not detectable in this stage of the Toxoplasma life cycle. The predicted amino acid sequences of these protein kinases are highly homologous to those of CDPKs previously described in plants (8 -10) and in the apicomplexan parasites P. falciparum (16,17) and E. maxima and E. tenella (18). Like these enzymes, the Toxoplasma CDPKs are composed of three domains: an N-terminal catalytic domain, a short junction domain, and a C-terminal calmodulin-like do- Soluble parasite extracts were prepared as described under "Experiments Procedures." To detect endogenous calcium-dependent protein kinase activity in parasite extracts, aliquots corresponding to 10 7 parasites were incubated directly with ␥-[ 32 P]ATP in the presence of 2 mM EGTA (A) or 2.5 M free calcium (B). The presence of specific TgCDPK1 substrates in parasite extracts was determined using heattreated extracts (10 min at 75°C) corresponding to 10 7 parasites that were incubated with [␥-32 P]ATP in the absence (C) or presence of 200 ng of purified recombinant TgCDPK1 (D). Phosphorylation reactions were analyzed by two-dimensional chromatography and autoradiography. Phosphoproteins that are phosphorylated specifically in the presence of TgCDPK1 are boxed. The asterisk indicates a minor contaminant of recombinant TgCDPK1 that is phosphorylated effectively by this enzyme. main with four EF-hand calcium-binding sites. Further characterization of TgCDPK1 revealed that, in addition to structural homology, the Toxoplasma enzyme also shares functional properties with other CDPKs in that its activity is absolutely dependent on the presence of calcium yet is insensitive to calmodulin antagonists. Despite the high degree of homology, the catalytic properties of TgCDPK1 are different from other CDPKs with respect to the observed K m (app) values for peptide substrates and ATP (9,19). These differences most likely reflect the different substrate specificities of these enzymes in vivo. Another indication of diverse cellular activities among CDPKs lies in their activation by calcium (19). While the activity of TgCDPK1 is half-maximal at 600 nM free calcium, a concentration often encountered in activated cells, the activity of the P. falciparum CDPK1 is half-maximal at 15 M free calcium (20). This difference could reflect the need for different CDPKs to be activated at different levels of cellular stimulation, although it could also reflect the absence of essential cofactors in the kinase assays for P. falciparum CDPK1.
A possible function for TgCDPK1 in T. gondii is suggested by its sensitivity to KT5926, an inhibitor of myosin light chain kinases in animal cells. This compound was previously shown to be an effective inhibitor of Toxoplasma motility as well as its attachment to host cells. The IC 50 of KT5926 for in vitro kinase activity of TgCDPK1 and parasite attachment to host cells is almost identical. In addition, in vivo phosphorylation experiments reveal that the effect of KT5926 on protein phosphorylation in Toxoplasma is quite specific, since it blocks the phosphorylation of only three proteins. These three proteins are, with respect to their apparent molecular weights and isoelectric points, indistinguishable from the three major substrates of TgC-DPK1 identified in Toxoplasma extracts in vitro. Finally, during fractionation of Toxoplasma cytosol, a single peak of KT5926sensitive protein kinase activity was detected, which cofractionated consistently with TgCDPK1. Taken together, these data demonstrate that TgCDPK1 is the most likely target for KT5926 in Toxoplasma and is therefore likely to be involved in bringing about parasite motility and attachment to host cells.
The physiological effects of TgCDPK1 activation are most likely brought about by the phosphorylation of the three substrates we detected in Toxoplasma extracts. The identity of these substrates and the manner in which their phosphorylation could contribute to parasite motility and host attachment is unclear at this time. One possible mechanism is suggested by the observation that treatment of Toxoplasma with KT5926 inhibits the secretion of the adhesin MIC2 (and probably other proteins as well) from the parasites' micronemes (7). It is therefore possible that phosphorylation of one or more of the TgCDPK1 substrates is required to bring about an aspect of microneme secretion, such as movement of micronemes to their site of secretion or the actual fusion of micronemes with the parasite plasma membrane. Since the secretion of MIC2 and other adhesins by Toxoplasma appears to be an essential prerequisite for parasite motility, it is possible that all three TgCDPK1 substrates are involved in microneme secretion. It is equally likely, however, that microneme secretion and parasite motility are regulated independently by TgCDPK1 phosphorylation of different substrates. Identification of the TgC-DPK1 substrates will probably shed light on these issues.
The possible role of TgCDPK2 is unclear at this time. Although its transcript is readily detected in the tachyzoite stage of the Toxoplasma life cycle, the protein product is not. A number of examples have been described where the transcripts of bradyzoite-specific antigens are actually detectable in both tachyzoites and bradyzoites, but the proteins themselves are only detectable in either the tachyzoites (34) or the bradyzoites (35). Therefore, the possibility remains that TgCDPK2 protein will be expressed in bradyzoites or sporozoites. The recent observation that P. falciparum CDPK3 is only expressed in the sexual erythrocytic stage of P. falciparum (17) strengthens this hypothesis.
The role of CDPKs in other apicomplexan parasites is unclear at this time. But considering that all apicomplexan parasites are obligate intracellular parasites and that cytoplasmic calcium appears to be essential for host cell invasion, they are likely to possess similar calcium-dependent regulatory mechanisms to bring about host cell attachment and invasion. It is therefore likely that one or more of the CDPKs identified thus far in other apicomplexan parasites will perform a role similar to that of TgCDPK1 in Toxoplasma gondii. If true, this would put apicomplexan CDPKs in control of one of the most essential processes in parasite survival, the ability to invade animal host cells. Since no members of the CDPK family have been described in the animal hosts of apicomplexan parasites, these enzymes may prove to be suitable targets for the development of novel antiparasitic chemotherapeutics.