Portal de Periódicos da CAPES

Inhibition of Ca2+/Calmodulin-dependent Protein Kinase Blocks Morphological Differentiation of Plasmodium gallinaceum Zygotes to Ookinetes

2002; Elsevier BV; Volume: 277; Issue: 16 Linguagem: Inglês

10.1074/jbc.m107903200

1083-351X

Mário A.C. Silva-Neto, Geórgia C. Atella, Mohammed Shahabuddin,

Trypanosoma species research and implications

Once ingested by mosquitoes, malaria parasites undergo complex cellular changes. These include zygote formation, transformation of zygote to ookinete, and differentiation from ookinete to oocyst. Within the oocyst, the parasite multiplies into numerous sporozoites. Modulators of intracellular calcium homeostasis A23187, MAPTAM, and TMB-8 blocked ookinete development as did the calmodulin (CaM) antagonists W-7 and calmidazolium. Ca2+/CaM-dependent protein kinase inhibitor KN-93 also blocked zygote elongation, while its ineffective analog KN-92 did not have such effect. In vitro both zygote and ookinete extracts efficiently phosphorylated autocamtide-2, a classic CaM kinase substrate, which could be blocked by calmodulin antagonists W-7 and calmidazolium and CaM kinase inhibitor KN-93. These results demonstrated the presence of calmodulin-dependent CaM kinase activity in the parasite. KN-93-treated parasites, however, expressed the ookinete-specific enzyme chitinase and the ookinete surface antigen Pgs28 normally, suggesting that the morphologically untransformed parasites are biochemically mature ookinetes. In mosquitoes, KN-93-treated parasites did not develop as oocysts, while KN-92-treated parasites produced similar numbers of oocysts as controls. These data suggested that inPlasmodium gallinaceum morphological development of zygote to ookinete, but not its biochemical maturation, relies on Ca2+/CaM-dependent protein kinase activity and demonstrated that the morphological differentiation is essential for the further development of the parasite in infected blood-fed mosquitoes. Once ingested by mosquitoes, malaria parasites undergo complex cellular changes. These include zygote formation, transformation of zygote to ookinete, and differentiation from ookinete to oocyst. Within the oocyst, the parasite multiplies into numerous sporozoites. Modulators of intracellular calcium homeostasis A23187, MAPTAM, and TMB-8 blocked ookinete development as did the calmodulin (CaM) antagonists W-7 and calmidazolium. Ca2+/CaM-dependent protein kinase inhibitor KN-93 also blocked zygote elongation, while its ineffective analog KN-92 did not have such effect. In vitro both zygote and ookinete extracts efficiently phosphorylated autocamtide-2, a classic CaM kinase substrate, which could be blocked by calmodulin antagonists W-7 and calmidazolium and CaM kinase inhibitor KN-93. These results demonstrated the presence of calmodulin-dependent CaM kinase activity in the parasite. KN-93-treated parasites, however, expressed the ookinete-specific enzyme chitinase and the ookinete surface antigen Pgs28 normally, suggesting that the morphologically untransformed parasites are biochemically mature ookinetes. In mosquitoes, KN-93-treated parasites did not develop as oocysts, while KN-92-treated parasites produced similar numbers of oocysts as controls. These data suggested that inPlasmodium gallinaceum morphological development of zygote to ookinete, but not its biochemical maturation, relies on Ca2+/CaM-dependent protein kinase activity and demonstrated that the morphological differentiation is essential for the further development of the parasite in infected blood-fed mosquitoes. Despite decades of effort to control malaria, it is still the most devastating parasitic disease of humans, killing millions of children each year. This is due in part to continuous transmission of the malaria parasite by vector mosquitoes. Malaria transmission is a result of complex multistep development of the parasite in the insect. Upon feeding on an infected host, mosquitoes acquire gametocytes, which in turn produce female (macrogamete) and male (microgamete) gametes. These gametes mate to form zygotes, which transform into ookinetes (1.Shahabuddin M. Parasitology. 1998; 116: S83-S93Crossref PubMed Google Scholar). The transition from zygotes to ookinetes is a complex cellular and biochemical process. Stage-specific proteins such as P28 and chitinase are expressed as the parasites transform from round zygotes to slender motile ookinetes (1.Shahabuddin M. Parasitology. 1998; 116: S83-S93Crossref PubMed Google Scholar). Development to motile ookinetes is crucial for malaria transmission, giving the malaria parasites the ability to egress from the ingested blood meal and cross the midgut wall and develop as oocysts (1.Shahabuddin M. Parasitology. 1998; 116: S83-S93Crossref PubMed Google Scholar). Within oocysts, the parasites multiply into thousands of sporozoites. A week or two later, the oocysts burst and release the sporozoites. The sporozoites then invade the salivary glands from which they can be transmitted to the vertebrate host during the next blood meal. This extensive developmental program, including repeated shape changes, implies the activation of specific signaling pathways in each stage. Shape change is required for the precise functions of various cells such as lymphocytes (2.Pettit E.J. Fay F.S. Physiol. Rev. 1998; 78: 949-967Crossref PubMed Scopus (98) Google Scholar), neutrophils (3.Coates T.D. Watts R.G. Hartman R. Howard T.H. J. Cell Biol. 1992; 117: 765-774Crossref PubMed Scopus (100) Google Scholar), macrophages (4.Yin H.L. Hartwig J.H. J. Cell Sci. Suppl. 1988; 9: 169-184Crossref PubMed Google Scholar), and platelets (5.Blockmans D. Deckmyn H. Vermylen J. Blood Rev. 1995; 9: 143-156Crossref PubMed Scopus (265) Google Scholar) in mammals as well as in protozoans such asDictyostelium, Euglena, and Entamoeba(6.Newell P.C. Biosci. Rep. 1995; 15: 445-462Crossref PubMed Scopus (19) Google Scholar, 7.Lonergan T.A. Williamson L.C. J. Cell Sci. 1988; 89: 365-371PubMed Google Scholar, 8.Carbajal M.E. Manning-Cela R. Pina A. Franco E. Meza I. Exp. Parasitol. 1996; 82: 11-20Crossref PubMed Scopus (45) Google Scholar). The mosquito stages of malaria parasite comprise a unique example of the biology of cell shape because zygotes are round cells that develop into elongated ookinetes, which then round up again when they become oocysts. These transitions in cell morphology define the functional properties of each stage of the parasite: invasive stages are polarized and elongated cells such as ookinetes and sporozoites, whereas noninvasive or vegetative stages are round and nonpolarized cells such as zygotes and oocysts. In most cells, calcium-dependent signaling pathways drive shape changes (9.Berridge M.J. Lipp P. Bootman M.D. Nat. Rev. Mol. Cell. Biol. 2000; 1: 11-21Crossref PubMed Scopus (4491) Google Scholar). In Plasmodium, calcium-triggered changes have been described earlier. Examples include invasion of erythrocytes and intraerythrocytic development of merozoites (10.McCallum-Deighton N. Holder A.A. Mol. Biochem. Parasitol. 1992; 50: 317-323Crossref PubMed Scopus (45) Google Scholar, 11.Garcia C.R. Parasitol. Today. 1999; 15: 488-491Abstract Full Text Full Text PDF PubMed Scopus (54) Google Scholar, 12.Hotta C.T. Gazarini M.L. Beraldo F.H. Varotti F.P. Lopes C. Markus R.P. Pozzan T. Garcia C.R. Nat. Cell Biol. 2000; 2: 466-468Crossref PubMed Scopus (163) Google Scholar). Also the presence of specific calcium storage compartments and calcium transport mechanisms in infected erythrocytes has been reported (13.Desai S.A. McCleskey E.W. Schlesinger P.H. Krogstad D.J. Am. J. Trop. Med. Hyg. 1996; 54: 464-470Crossref PubMed Scopus (46) Google Scholar,14.Garcia C.R. Ann S.E. Tavares E.S. Dluzewski A.R. Mason W.T. Paiva F.B. Eur. J. Cell Biol. 1998; 76: 133-138Crossref PubMed Scopus (51) Google Scholar). In addition, production of inositol 1,4,5-trisphosphate and diacylglycerol, which are able to elevate intracellular calcium concentration, has been demonstrated during gametogenesis (15.Kawamoto F. Fujioka H. Murakami R. Syafruddin Hagiwara M. Ishikawa T. Hidaka H. Eur. J. Cell Biol. 1993; 60: 101-107PubMed Google Scholar, 16.Martin S.K. Jett M. Schneider I. J. Parasitol. 1994; 80: 371-378Crossref PubMed Scopus (51) Google Scholar). Different components of calcium signaling systems such as calcium-dependent protein kinases, Ca2+-ATPases, and noncytosolic EF-hand Ca2+-binding proteins from Plasmodium falciparumhave been purified and cloned, but the role of these pathways in parasite development or infection is not fully understood (11.Garcia C.R. Parasitol. Today. 1999; 15: 488-491Abstract Full Text Full Text PDF PubMed Scopus (54) Google Scholar). In the vector mosquitoes, although cyclic nucleotides and calcium were shown to affect gametogenesis, no involvement of a signal transduction pathway has been attributed to any of the subsequent stages (13.Desai S.A. McCleskey E.W. Schlesinger P.H. Krogstad D.J. Am. J. Trop. Med. Hyg. 1996; 54: 464-470Crossref PubMed Scopus (46) Google Scholar,14.Garcia C.R. Ann S.E. Tavares E.S. Dluzewski A.R. Mason W.T. Paiva F.B. Eur. J. Cell Biol. 1998; 76: 133-138Crossref PubMed Scopus (51) Google Scholar). In this study we investigated the role of the calcium/calmodulin (CaM) 1The abbreviations used are: CaMcalmodulinAIPautocamtide-2-related inhibitory peptideCaMKCa2+/calmodulin-dependent protein kinaseKN-92(2-[N-(4-methoxybenzenesulfonyl)]amino-N-(4-chlorocinnamyl)-N-methylbenzylamine)KN-932-[N-(2-hydroxyethyl)-N-(4-methoxybenzenesulfonyl)]amino-N-(4-chlorocinnamyl)-N-methylbenzylamineMyrmyristoylMAPTAM1,2-bis-(2-amino-5-methylphenoxy)ethane tetraacetic acid tetraacetoxymethyl estersignaling pathway in the control of zygote to ookinete differentiation and the role of Ca2+/CaM-dependent protein kinase (CaMK) in this process. Our data suggest that the calcium/CaM pathway controls morphological changes during zygote to ookinete differentiation. calmodulin autocamtide-2-related inhibitory peptide Ca2+/calmodulin-dependent protein kinase (2-[N-(4-methoxybenzenesulfonyl)]amino-N-(4-chlorocinnamyl)-N-methylbenzylamine) 2-[N-(2-hydroxyethyl)-N-(4-methoxybenzenesulfonyl)]amino-N-(4-chlorocinnamyl)-N-methylbenzylamine myristoyl 1,2-bis-(2-amino-5-methylphenoxy)ethane tetraacetic acid tetraacetoxymethyl ester P. gallinaceum strain 8A was maintained in White Leghorn chickens by serial passage. Processing of infected chicken blood for zygote preparation was performed as described previously (17.Kaushal D.C. Carter R. Rener J. Grotendorst C.A. Miller L.H. Howard R.J. J. Immunol. 1983; 131: 2557-2562PubMed Google Scholar). Briefly, infected blood was obtained by heart puncture of chicken and immediately diluted in suspended animation buffer (9 mmglucose, 8 mm Tris-HCl, pH 7.3, 138 mm NaCl) and centrifuged to separate cells from serum. Packed cells were resuspended in exflagellation medium (75% suspended animation buffer, 8% heat-inactivated chicken serum, 26 mmNaHCO3, 0.1 mm xanthurenic acid) for 45 min. The fertilized zygotes, macrogametes, and blood cells were separated from red blood cells on a cushion of Ficoll-PaqueTM Plus (Amersham Biosciences). Chicken white blood cells were removed by agglutinating with wheat germ agglutinin (0.1 mg/ml). The parasites were finally resuspended in ookinete medium (M199 medium with glutamine (Invitrogen) supplemented with glucose and penicillin/streptomycin) to a concentration of 4 × 106 cells/ml. Preparations routinely yielded 1–10 × 107 parasites with ookinete differentiation efficiency of 50–90%. Unless otherwise stated, Me2SO was used as the control in all experiments because it was the solvent for all drugs used in this study. The final concentration of Me2SO was never more than 0.1%; this concentration does not influence the rate of zygote to ookinete differentiation (data not shown). In experiments with protein kinase inhibitors, these drugs were initially used at the following final concentrations, which correspond to their publishedKi values for mammalian cells: bistyrphostin (inhibitor of epidermal growth factor receptor tyrosine kinase activity, Ki 0.4 μm), LY294002 (phosphatidylinositol 3-kinase inhibitor, Ki 1.6 μm), roscovitine (inhibitor of cyclin-dependent kinases; the IC50 of 0.7 μm was used as a reference because at this concentration roscovitine inhibits most p34cdc2, p33cdk2, and p33cdk5-cyclin complexes), H-89 (inhibitor of cyclic AMP-dependent protein kinase,Ki 0.048 μm), H-9 (inhibitor of cyclic GMP-dependent protein kinase, Ki 0.87 μm), bisindolylmaleimide I (inhibitor of protein kinase C, Ki 0.01 μm), ML-7 (inhibitor of myosin light chain kinase, Ki 0.3 μm), KN-93 (inhibitor of CaM kinases, Ki 0.37 μm) (18.Sumi M. Kiuchi K. Ishikawa T. Ishii A. Hagiwara M. Nagatsu T. Hidaka H. Biochem. Biophys. Res. Commun. 1991; 181: 968-975Crossref PubMed Scopus (461) Google Scholar), and KN-92 (an ineffective KN-93 analog). Autocamtide-2-related inhibitory peptide (AIP) (KKALRRQEAVDAL) and its myristoylated form Myr-AIP are peptides derived from the CaM kinase II substrate autocamtide-2, but alanine is substituted for threonine at the 9-position, which turns them into highly specific and potent inhibitors of CaM kinase II (19.Ishida A. Kameshita I. Okuno S. Kitani T. Fujisawa H. Biochem. Biophys. Res. Commun. 1995; 212: 806-812Crossref PubMed Scopus (248) Google Scholar). All the above drugs as well as calcium and CaM antagonists EGTA, A23187, MAPTAM, TMB-8, W-7, and calmidazolium were obtained from CalBiochem-NovaBiochem Corp. Once zygotes were obtained, drugs were immediately added to the differentiation medium. The following day, samples were centrifuged for 5 min at 700 × g to concentrate the cells. The number of zygotes and ookinetes were estimated in each sample by direct counting in a Neubauer chamber. The values obtained in controls (only Me2SO-treated cells), referred to as 100% differentiation, were used as a reference for estimating the effect of each drug. Statistical analyses were conducted with GraphPad Prism 3 software (GraphPad Software, Inc., San Diego, CA). Parasites were photographed in suspension with a Zeiss Axiolphot microscope with 40× objectives linked to a digital camera and using an illumination ring and differential interference contrast objectives (Carl Zeiss, Inc., Thornwood, NJ). Images were acquired and processed in Adobe Photoshop (Adobe Systems Inc., San José, CA). CaMK activity was assayed as suggested by the manufacturer (Upstate Biotechnologies, Piscataway, NJ) using autocamtide-2 and [32P]ATP (Amersham Biosciences, 6000 Ci/mmol) as substrates. Typically parasite extracts (4 × 106 cells) were prepared by addition of 0.2 ml of 20 mm Tris-HCl, pH 7.5, 200 mm NaCl, 10 mm NaF, 1 mm sodium vanadate, 0.1 mm EDTA, 1 mm EGTA, 0.2 mg/ml sodium azide, 0.1% Triton X-100, 1 mm phenylmethylsulfonyl fluoride. Cells were then immediately frozen at −70 °C. The following day extracts were thawed and centrifuged at 15,000 × g for 15 min, and the supernatants were used for kinase assays. Reactions (50 μl) were incubated with parasite extracts (0.1–0.4 mg/ml) for 10 min in the presence of 1 mm CaCl2, 100 μm autocamtide-2, 8 μg/ml calmodulin, 2 μm cAMP-dependent protein kinase peptide inhibitor, 2 μm protein kinase C peptide inhibitor, and 100 μm [32P]ATP (500–1000 cpm/pmol) at 30 °C. Aliquots were then removed, spotted in phosphocellulose discs, washed three times in 0.75% phosphoric acid and one time in acetone, dried, and counted by liquid scintillation. Chitinase activity was assayed essentially as in Huber et al. (20.Huber M. Cabib E. Miller L.H. Proc. Natl. Acad. Sci. U. S. A. 1991; 88: 2807-2810Crossref PubMed Scopus (180) Google Scholar) in both culture supernatants and parasite extracts. Briefly, supernatants were concentrated by centrifugal ultrafiltration (Centripep 10, Amicon, Inc., Beverly, MA) and resuspended in 20 mm sodium phosphate, pH 6.8, and then used in enzyme assays. Parasite extracts were prepared by addition of 20 mm sodium phosphate, pH 6.8, to the cell pellet followed by vigorous vortexing, three cycles of freeze-thawing (dry ice to room temperature), and sonication (six cycles for 20 s on ice). Enzyme sources (0.01 ml) were added to 0.160 ml of 20 mm Tris-HCl, pH 8.0, to which 0.03 ml of 4-methylumbelliferyl-N′,N′,N“-β-d-triacetylchitotrioside was added. Enzyme reactions were incubated at room temperature. A Dynatek Fluorite 1000 (filters, excitation 365 nm, and emission 450 nm) was used for kinetic fluorescence detection for 60 min. Protein concentration was estimated using the MicroBCA kit (Pierce). Final protein concentration in the assays was 0.05 mg/ml for cell supernatant samples and 0.025 mg/ml for pellets; the results are expressed as the normalized fluorescence corresponding to 0.001 mg of protein in each condition. Cell extracts obtained as above for the chitinase assay were used for Western blotting experiments. Either Me2SO-, KN-92-, KN-93-treated, or untreated parasite extracts from the same preparation were used. Protein content was estimated as above, and samples were adjusted for the same final concentration. For immunoblotting, proteins were separated on 10% SDS-PAGE gels (Novex Experimental Technology, San Diego, CA) and electroblotted to nitrocellulose using the Novex X Cell Blot II module. After overnight blocking with 3% albumin in Tris-buffered saline, 0.05% Tween 20 (TBS-T), blots were incubated with primary polyclonal antibody at 1:500 dilution in TBS-T for 1 h at room temperature. After three washes with TBS-T, blots were incubated with mouse secondary antibody linked to horseradish peroxidase (ECL System, Amersham Biosciences) at 1:10,000 dilution in TBS-T for 1 h. After three washes with Tris-buffered saline, blots were developed in a chemiluminescence assay using the ECL System. Monoclonal antibodies against Pgs28 were described earlier (21.Grotendorst C.A. Kumar N. Carter R. Kaushal D.C. Infect. Immun. 1984; 45: 775-777Crossref PubMed Google Scholar). The Liverpool/blackeye strain ofAedes aegypti were raised at 26 °C and 80% relative humidity and fed on sugar ad libitum. The effects of KN-93 and KN-92 on oocyst formation were tested as follows. Zygotes were treated for 3 h with 2.0 μm of each drug, a dose of KN-93 that blocks 80% of zygote elongation with no detectable effects on KN-92-treated parasites. These cells were then washed three times with M199 medium supplemented as described above. The cells were then mixed with blood and administered to the insects in a density of 0.5–1.0 × 107 cells/ml. Mosquitoes, starved for 6–18 h, were fed on blood through a Parafilm membrane. Eight days after blood feeding, mosquito midguts were dissected and stained with 1.0% mercurochrome in water, and oocysts were counted by light microscopy. As a control, a group of zygotes was treated only with 0.1% Me2SO, the solvent used for KN-92 and KN-93. All the mosquitoes in the control groups that received a blood meal were included in the analysis. Photographs of infected mosquito midguts were taken with a brightfield Zeiss Axiolphot microscope with 10× objectives attached to a digital camera. Results are described as prevalence of infection, which is the proportion of mosquitoes with oocysts in each experiment (infected mosquitoes/total mosquitoes fed), intensity of infection, which is the mean (geometric mean) number of oocysts in infected mosquitoes only, mean oocysts per midgut in each batch of mosquitoes (geometric mean of oocysts in all infected blood-fed mosquitoes including those without oocysts), and probability of occurring the differences between the control and inhibitor fed groups determined by nonparametric Wilcoxon rank sum analysis on raw data using GraphPad Prism software. Because cell shape change is triggered by calcium signaling in many cell types, we tested the effect of known calcium and CaM antagonists in zygote to ookinete differentiation. Agents that affect both intracellular and extracellular calcium homeostasis were added to incubation media containing zygotes. The following day, cells were examined for the presence of ookinetes. Incubation of zygotes with the extracellular calcium chelator EGTA did not affect ookinete formation (Fig. 1A), suggesting that extracellular calcium does not play a role during zygote to ookinete transition; however, the calcium ionophore A23187, which affects intracellular calcium homeostasis, completely blocked ookinete differentiation. A similar blockage of differentiation was observed when the intracellular calcium chelator MAPTAM was added to zygote media. Inositol trisphosphate receptors are involved in the release of stored intracellular calcium (9.Berridge M.J. Lipp P. Bootman M.D. Nat. Rev. Mol. Cell. Biol. 2000; 1: 11-21Crossref PubMed Scopus (4491) Google Scholar). The addition of TMB-8, a known antagonist of inositol trisphosphate receptors, also blocked zygote differentiation. This set of results suggests that zygote to ookinete differentiation relies on the release of intracellular stored calcium pools. Because calcium often acts through its binding to CaM, we added either W-7 or calmidazolium, two commonly used CaM antagonists. Both drugs blocked zygote to ookinete differentiation in a dose-dependent manner (Fig. 1B). Together these results suggested that both intracellular calcium and CaM are required for the differentiation of zygote to ookinete. Protein kinases are the effectors of most known signal transduction pathways (22.Hanks S.K. Hunter T. FASEB J. 1995; 9: 576-596Crossref PubMed Scopus (2295) Google Scholar). Therefore a set of inhibitors of these enzymes was tested for the ability to block zygote to ookinete differentiation (Fig. 1C). The final concentration of these drugs in these initial experiments is within the Kidetermined for their effects on mammalian cells. Fig. 1Cshows that only KN-93, a specific inhibitor of CaM kinase (18.Sumi M. Kiuchi K. Ishikawa T. Ishii A. Hagiwara M. Nagatsu T. Hidaka H. Biochem. Biophys. Res. Commun. 1991; 181: 968-975Crossref PubMed Scopus (461) Google Scholar, 23.Means A.R. Mol. Endocrinol. 2000; 14: 4-13Crossref PubMed Scopus (159) Google Scholar), significantly blocked the formation of ookinetes. CaMKs are enzymes usually kept inactive due to the presence of an autoinhibitory domain (23.Means A.R. Mol. Endocrinol. 2000; 14: 4-13Crossref PubMed Scopus (159) Google Scholar, 24.Soderling T.R. Curr. Opin. Neurobiol. 2000; 10: 375-380Crossref PubMed Scopus (203) Google Scholar). Upon an increase in intracellular calcium, the binding of calcium-CaM complex to the enzyme disrupts the interaction of autoinhibitory domains. The enzyme is then fully activated following autophosphorylation. KN-93 interacts with the calcium/CaM-binding domain of CaMK and blocks the binding of calcium-CaM complex. A hydroxyethyl domain on KN-93 is responsible for this interaction. This region is missing in KN-92, an inactive analog of KN-93 (18.Sumi M. Kiuchi K. Ishikawa T. Ishii A. Hagiwara M. Nagatsu T. Hidaka H. Biochem. Biophys. Res. Commun. 1991; 181: 968-975Crossref PubMed Scopus (461) Google Scholar). Thus KN-92 was used to test the specificity of KN-93 effects. Fig. 1D shows a dose-dependent effect of both KN-93 and KN-92 on ookinete formation. Unlike KN-93, despite its similar structure, KN-92 did not block the parasite elongation. The IC50 for KN-93 inhibition of zygote development is 0.7 μm, and few ookinetes are detected when zygotes are treated with concentrations greater than 2 μm KN-93 (Fig. 1D). KN-93 affects zygotes irreversibly when added within 4 h after their formation as removal of the inhibitor did not reverse the effect (data not shown). This suggests that the commitment between this calcium signaling pathway and the change in the parasite shape might occur in the early hours following zygote formation. Fig. 2 shows the morphological features of drug-treated parasites. KN-92-treated samples show fully differentiated ookinetes (Fig. 2D) similar to control ookinete cells (Fig. 2B). In contrast, the CaMK inhibitor KN-93 almost completely blocked the ookinete formation (Fig. 2C), and the shape of the parasite remained similar to the zygote (Fig. 2A). In some KN-93 samples, partially elongated parasites were seen. These data together suggest that a calcium/CaM signaling pathway through the activation of CaMK is required for differentiation of zygotes to ookinetes. To determine whether CaMK activity is present in P. gallinaceumthe following experiments were conducted. Autocamtide-2 is a peptide whose sequence is based on the autophosphorylation site of CaMK and thus is a specific substrate for these enzymes (25.Hashimoto Y. Schworer C.M. Colbran R.J. Soderling T.R. J. Biol. Chem. 1987; 262: 8051-8055Abstract Full Text PDF PubMed Google Scholar). This peptide was used to assay P. gallinaceum extracts in the presence of [32P]ATP. As shown in Fig. 3 autocamtide-2 is phosphorylated by zygote extract (Fig. 3A) and ookinete extract (Fig. 3B). The activity could be blocked by classical calmodulin antagonists such as W-7, calmidazolium, and KN-93 but not by KN-92. These results demonstrated that both cell types contain CaMK activity. To test the activity of CaMK in live cells we used a cell-permeable inhibitor of the enzyme. AIP is a nonphosphorylatable analog of the peptide substrate autocamtide-2 (19.Ishida A. Kameshita I. Okuno S. Kitani T. Fujisawa H. Biochem. Biophys. Res. Commun. 1995; 212: 806-812Crossref PubMed Scopus (248) Google Scholar) and therefore a highly specific and potent inhibitor of CaMK II. Both AIP and its myristoylated cell-permeable form Myr-AIP were then added to parasite cultures to test for their ability to block zygote to ookinete differentiation. Fig. 3C shows that Myr-AIP but not its nonpermeable analog, AIP, is able to block zygote to ookinete differentiation. The apparent low efficiency in the blocking ability of the peptide when compared with KN-93 can be attributed to the higher permeability of the latter molecule. This set of results reinforces that P. gallinaceumbears a protein kinase activity with overall characteristics of the CaMK described in the mammalian system (23.Means A.R. Mol. Endocrinol. 2000; 14: 4-13Crossref PubMed Scopus (159) Google Scholar, 24.Soderling T.R. Curr. Opin. Neurobiol. 2000; 10: 375-380Crossref PubMed Scopus (203) Google Scholar, 26.Hook S.S. Means A.R. Annu. Rev. Pharmacol. Toxicol. 2001; 41: 471-505Crossref PubMed Scopus (418) Google Scholar). The process of differentiation of zygote to ookinete is not well understood. As the parasite elongates, some stage-specific genes are expressed by the ookinetes. A few hours after fertilization, a 28-kDa surface antigen is expressed in ookinetes (27.Kumar N. Carter R. Mol. Biochem. Parasitol. 1985; 14: 127-139Crossref PubMed Scopus (65) Google Scholar). Also ∼12–15 h after fertilization, when P. gallinaceum ookinetes are almost transformed, a chitinase activity is secreted by the parasite (20.Huber M. Cabib E. Miller L.H. Proc. Natl. Acad. Sci. U. S. A. 1991; 88: 2807-2810Crossref PubMed Scopus (180) Google Scholar). The chitinase is required by the parasite to cross the chitinous peritrophic matrix that surrounds the blood meal in mosquito midgut (20.Huber M. Cabib E. Miller L.H. Proc. Natl. Acad. Sci. U. S. A. 1991; 88: 2807-2810Crossref PubMed Scopus (180) Google Scholar). To examine whether KN-93-treated parasites express these ookinete-specific proteins, we assayed chitinase secretion by treated and untreated parasites. Fig. 4A shows the levels of chitinase activity in KN-92- and KN-93-treated parasite extracts and cell supernatants. As can be seen, both groups exhibit high levels of enzyme activity when compared with nontransformed zygotes. This result indicates that although KN-93-treated parasites do not undergo the characteristic shape change of fully transformed ookinetes they are able both to produce and to secrete chitinase. We also tested whether expression of Pgs28 is affected by KN-93. Fig. 4B shows that the levels of this protein are similar in both KN-93- and KN-92-treated as well as in nontreated parasites. These results show that zygote treatment with an inhibitor of CaMK produced a biochemically mature but morphologically impaired ookinete. Expression of chitinase activity and ookinete surface antigen Pgs28 in KN-93-treated parasites suggested that some properties of mature ookinetes remain normal in untransformed parasites. We then examined whether the morphologically impaired parasites are able to develop as oocysts. Zygotes were treated for 3 h with either KN-93 or KN-92. These parasites were then washed to remove the drugs, mixed with blood, and fed to mosquitoes. Eight days later, mosquitoes were dissected, and their midguts were examined for oocysts. Typically oocysts are distributed toward the posterior end of posterior midgut. Fig. 5 shows that both untreated (Fig. 5A) and KN-92-treated (Fig. 5C) parasites infected the mosquitoes with high intensity. However, in agreement with previous results, only a few or no oocysts developed from the KN-93-treated parasites (Fig. 5B). The few parasites that developed as oocysts probably are those that could avoid the effect of the drug and developed as fully elongated ookinetes. The overall blocking of oocyst implantation is within 70–100% in three different experiments (Table I). The data suggested that expression of ookinete-specific genes is not sufficient, and morphologic elongation is required for the further development as oocysts.Table IBlocking of sporogonic development of P. gallinaceum by KN-93ExperimentGroupPrevalence1-aPrevalence of infection in parasite-fed mosquitoes (infected mosquitoes/total mosquitoes fed).Intensity1-bIntensity of infection in the mosquitoes with oocysts (geometric mean (range of number of oocysts in infected midguts)).Mean Oocysts1-cMean oocysts per gut in each batch of mosquitoes (geometric mean of all infected blood fed mosquitoes including those without oocysts).Blockingp1-dProbability of occurring the differences between the control and inhibitor-fed groups determined by nonparametric rank sum analysis on raw data using GraphPad Prism software.%1Control20 /4015.38 (1–82)2.921KN-9221 /4010.06 (1–95)2.360KN-932 /401.00 (1)0100<0.00012Control32 /4062.37 (5–216)26.29KN-9237 /4043.31 (3–142)31.65KN-9326 /4010.91 (1–73)3.72788.22<0.00013Control40 /4019.87 (2–91)18.87KN-9233 /4016.52 (3–93)9.11KN-9327 /406.816 (1–31)2.65371.070.00161-a Prevalence of infection in parasite-fed mosquitoes (infected mosquitoes/total mosquitoes fed).1-b Intensity of infection in the mosquitoes with oocysts (geometric mean (range of number of oocysts in infected midguts)).1-c Mean oocysts per gut in each batch of mosquitoes (geometric mean of all infected blood fed mosquitoes including those without oocysts).1-d Probability of occurring the differences between the control and inhibitor-fed groups determined by nonparametric rank sum analysis on raw data using GraphPad Prism software. Open table in a new tab Here we report evidence that suggests elongation of the zygote to ookinete and expression of ookinete-specific genes in P. gallinaceum are regulated by separate pathways. We show that CaMK is involved in the elongation step of the parasite, but the blocking of this enzyme activity did not interfere with expression of ookinete-specific genes Pgs28 and chitinase. Furthermore the untransformed parasites failed to develop as oocysts in mosquitoes. Several members of the eukaryotic protein kinase superfamily (22.Hanks S.K. Hunter T. FASEB J. 1995; 9: 576-596Crossref PubMed Scopus (2295) Google Scholar) have been reported in Plasmodium spp. (28.Kappes B. Doerig C.D. Graeser R. Parasitol. Today. 1999; 15: 449-454Abstract Full Text Full Text PDF PubMed Scopus (61) Google Scholar). These include members of three subgroups of the Ser/Thr protein kinase family such as PfPK5 (a member of the family of cyclin-dependent protein kinases), casein kinase 1 (28.Kappes B. Doerig C.D. Graeser R. Parasitol. Today. 1999; 15: 449-454Abstract Full Text Full Text PDF PubMed Scopus (61) Google Scholar), Pfmap-2 (a member of the mitogen-activated protein kinase family) and its upstream regulator Pfnek-1 (29.Dorin D. Alano P. Boccaccio I. Ciceron L. Doerig C. Sulpice R. Parzy D. J. Biol. Chem. 1999; 274: 29912-29920Abstract Full Text Full Text PDF PubMed Scopus (99) Google Scholar, 30.Dorin D. Le Roch K. Sallicandro P. Alano P. Parzy D. Poullet P. Meijer L. Doerig C. Eur. J. Biochem. 2001; 268: 2600-2608Crossref PubMed Scopus (106) Google Scholar), PfPKAc (a cAMP-dependent protein kinase) (31.Li J. Cox L.S. Mol. Biochem. Parasitol. 2000; 109: 157-163Crossref PubMed Scopus (36) Google Scholar), and calcium-dependent protein kinases (32.Zhao Y. Franklin R.M. Kappes B. Mol. Biochem. Parasitol. 1994; 66: 329-343Crossref PubMed Scopus (46) Google Scholar). Recently a CaM-like domain protein kinase enzyme has been described in the apicomplexan parasite Toxoplasma gondii (33.Kieschnick H. Wakefield T. Narducci C.A. Beckers C. J. Biol. Chem. 2001; 276: 12369-12377Abstract Full Text Full Text PDF PubMed Scopus (131) Google Scholar). However, by using classical CaM antagonists both in vivo (Fig. 1B) and in vitro (Fig. 3, A andB) as well as the CaM kinase-specific inhibitor KN-93 we demonstrate that CaMK activity plays a role in the zygote to ookinete differentiation process. There are two main mechanisms by which calcium can perform a regulatory function in cell shape change: first through direct binding to proteins involved in the process of polymerization/depolymerization of the cytoskeleton and second by the formation of a complex with CaM, which then further interacts with different target proteins (34.Acuto O. Cantrell D. Annu. Rev. Immunol. 2000; 18: 165-184Crossref PubMed Scopus (225) Google Scholar). One of these proteins is CaMK (23.Means A.R. Mol. Endocrinol. 2000; 14: 4-13Crossref PubMed Scopus (159) Google Scholar, 24.Soderling T.R. Curr. Opin. Neurobiol. 2000; 10: 375-380Crossref PubMed Scopus (203) Google Scholar). The multifunctional CaM kinase family, consisting of CaM kinase I, II, and IV, can mediate the action of signals that elevate intracellular free calcium (23.Means A.R. Mol. Endocrinol. 2000; 14: 4-13Crossref PubMed Scopus (159) Google Scholar, 24.Soderling T.R. Curr. Opin. Neurobiol. 2000; 10: 375-380Crossref PubMed Scopus (203) Google Scholar). This enzyme family is widely distributed in mammalian tissues and in all eukaryotic systems examined (23.Means A.R. Mol. Endocrinol. 2000; 14: 4-13Crossref PubMed Scopus (159) Google Scholar, 24.Soderling T.R. Curr. Opin. Neurobiol. 2000; 10: 375-380Crossref PubMed Scopus (203) Google Scholar). Once activated, CaM kinases play an important role in cell morphology and cytoskeletal assembly in various cell models through phosphorylation of several cytoskeleton components (35.le Gouvello S. Manceau V. Sobel A. J. Immunol. 1998; 161: 1113-1122PubMed Google Scholar, 36.Sanchez C. Diaz-Nido J. Avila J. Prog. Neurobiol. 2000; 61: 133-168Crossref PubMed Scopus (410) Google Scholar). CaM kinases are also involved with the control of cell growth and elongation (23.Means A.R. Mol. Endocrinol. 2000; 14: 4-13Crossref PubMed Scopus (159) Google Scholar, 36.Sanchez C. Diaz-Nido J. Avila J. Prog. Neurobiol. 2000; 61: 133-168Crossref PubMed Scopus (410) Google Scholar, 37.Oberstar J.V. Challacombe J.F. Roche F.K. Letourneau P.C. J. Neurobiol. 1997; 33: 161-171Crossref PubMed Scopus (14) Google Scholar, 38.Yamauchi T. Yoshimura Y. Nomura T. Fujii M. Sugiura H. Brain Res. Brain Res. Protoc. 1998; 2: 250-258Crossref PubMed Scopus (14) Google Scholar). In Plasmodium, involvement of the CaM kinase family has been suspected due to sensitivity of the parasite to some CaM antagonists (11.Garcia C.R. Parasitol. Today. 1999; 15: 488-491Abstract Full Text Full Text PDF PubMed Scopus (54) Google Scholar, 39.Matsumoto Y. Perry G. Scheibel L.W. Aikawa M. Eur. J. Cell Biol. 1987; 45: 36-43PubMed Google Scholar, 40.Scheibel L.W. Int. Rev. Cytol. 1992; 134: 165-242Crossref PubMed Scopus (14) Google Scholar). Several CaM antagonists block merozoite invasion and further intraerythrocytic development byP. falciparum (39.Matsumoto Y. Perry G. Scheibel L.W. Aikawa M. Eur. J. Cell Biol. 1987; 45: 36-43PubMed Google Scholar, 40.Scheibel L.W. Int. Rev. Cytol. 1992; 134: 165-242Crossref PubMed Scopus (14) Google Scholar) as well as gametogenesis ofPlasmodium berghei and P. falciparum (15.Kawamoto F. Fujioka H. Murakami R. Syafruddin Hagiwara M. Ishikawa T. Hidaka H. Eur. J. Cell Biol. 1993; 60: 101-107PubMed Google Scholar, 41.Kawamoto F. Alejo-Blanco R. Fleck S.L. Kawamoto Y. Sinden R.E. Mol. Biochem. Parasitol. 1990; 42: 101-108Crossref PubMed Scopus (96) Google Scholar). Curiously, classical antimalarial drugs such as primaquine and chloroquine, besides their overall effect as weak bases at high concentrations, also may act as CaM antagonists at lower concentrations and block the traffic of endosomal vacuoles in vivo and CaM kinase II activity in vitro (42.van Weert A.W. Geuze H.J. Groothuis B. Stoorvogel W. Eur. J. Cell Biol. 2000; 79: 394-399Crossref PubMed Scopus (88) Google Scholar). Some studies have addressed CaMK activity in protozoan cells such as inDictyostelium discoideum (43.Dunbar A.J. Wheldrake J.F. FEMS Microbiol. Lett. 1994; 115: 113-118Crossref PubMed Scopus (5) Google Scholar), Tetrahymena (44.Hirano-Ohnishi J. Watanabe Y. J. Biochem. (Tokyo). 1989; 105: 858-860Crossref PubMed Scopus (22) Google Scholar), and Trypanosoma cruzi (45.Ogueta S.B. Macintosh G.C. Tellez-Inon M.T. J. Eukaryot. Microbiol. 1998; 45: 392-396Crossref PubMed Scopus (14) Google Scholar). In Tetrahymena, this enzyme has been shown to phosphorylate cytoskeletal components such as ciliary β-tubulin (44.Hirano-Ohnishi J. Watanabe Y. J. Biochem. (Tokyo). 1989; 105: 858-860Crossref PubMed Scopus (22) Google Scholar). In this protozoan it was further demonstrated that elongation factor 1α F-actin bundling activity is also regulated by calcium and CaM (46.Kurasawa Y. Hanyu K. Watanabe Y. Numata O. J. Biochem. (Tokyo). 1996; 119: 791-798Crossref PubMed Scopus (54) Google Scholar). In epimastigotes of T. cruzi, CaMK is tightly bound to parasite cytoskeleton and flagella, and its autophosphorylation can regulate its distribution in the cell (45.Ogueta S.B. Macintosh G.C. Tellez-Inon M.T. J. Eukaryot. Microbiol. 1998; 45: 392-396Crossref PubMed Scopus (14) Google Scholar,47.Ogueta S. Intosh G.M. Tellez-Inon M.T. Mol. Biochem. Parasitol. 1996; 78: 171-183Crossref PubMed Google Scholar). The elongate shape of ookinetes allows these cells to egress from the blood bolus, cross the peritrophic matrix, and invade the mosquito midgut epithelium. Development from round, sedentary zygote to slender, motile ookinetes is therefore crucial for malaria transmission. This study is the first demonstration that zygote to ookinete differentiation during the mosquito stage of the malaria parasite cycle is a signal transduction-mediated process. Further study of this process may identify novel mechanisms involved in the complex sporogonic development of the malaria parasite and may contribute to the design of novel strategies to control malaria transmission. We are thankful to Drs. José Ribeiro and Sanjay Desai for comments on the manuscript. We also thank Dr. Adriana Costero for help in parasite preparation. We acknowledge the assistance of André Laughinghouse and Kevin Lee in rearing mosquitoes and cultivating parasites and Brenda Rae Marshall for preparing the manuscript.