- Email: [email protected]
Evaluation of serodiagnosis of toxoplasmosis by using the recombinant nucleoside triphosphate hydrolase isoforms expressed in Escherichia coli
Parasitology International 48 Ž2000. 215]222
Evaluation of serodiagnosis of toxoplasmosis by using the recombinant nucleoside triphosphate hydrolase isoforms expressed in Escherichia coli q Kyoko Nakajima-Nakano a,b, Asao Makiokac , Naoya Yamashitab, Nobutake Matsuo b, Takashi Asai a,U a
Department of Tropical Medicine and Parasitology, Keio Uni¨ ersity School of Medicine, 35 Shinanomachi Shinjuku-ku, Tokyo 160-8582, Japan b Department of Pediatrics, Keio Uni¨ ersity School of Medicine, Shinjuku-ku, Tokyo 160-8582, Japan c Department of Tropical Medicine, Jikei Uni¨ ersity School of Medicine, 3-25-8 Nishishinbashi, Minato-ku, Tokyo 105-8461, Japan Received 8 June 1999; accepted 19 July 1999
Abstract The nucleoside triphosphate hydrolase ŽNTPase. isoforms termed, NTPase-I and NTPase-II of Toxoplasma gondii, were expressed in Escherichia coli as inclusion bodies and purified under denaturing condition. Furthermore, NTPase-I was refolded as an active form and purified under non-denaturing condition. The purified NTPase isoforms, both denatured and refolded, were tested for their usefulness as antigens for the serodiagnosis of acute toxoplasmosis in immunocompetent humans. The test was conducted by using the recombinant NTPase isoforms and comparing the enzyme linked immunosorbent assay ŽELISA. absorbances with the Sabin]Feldman dye test titer. Seventy-three sera from dye test-positive patients, and 30 sera from subjects with no T. gondii infection were examined. The total positive rates in dye test positive sera were: 82% Ž60r73. for denatured NTPase-I; 78% Ž57r73. for denatured NTPase-II; and 63% Ž46r73. for refolded NTPase-I. For all three antigen types of recombinant NTPase, the positive rates of sera of acute toxoplasmosis suspected patients were 93% Ž13r14.. A moderate correlation between the ELISA absorbance using these antigens and the dye test titer was observed with the correlation coefficients, 0.583 Ž r 2 . for denatured NTPase-I, 0.472 Ž r 2 . for denatured NTPase-II, and 0.604 Ž r 2 . for
Abbre¨ iations: ELISA, enzyme-linked immunosorbent assay; NTPase, nucleoside triphosphate hydrolase; PCR, polymerase chain reaction; SDS]PAGE, sodium dodecyl sulfate]polyacrylamide gel electrophoresis. q This study forms part of the dissertation submitted to Keio University School of Medicine in partial fulfillment of the requirement for the Degree of Medical Sciences. U Corresponding author. Tel.: q81-3-3353-1211, ext. 62747; fax: q81-3-3353-5958. E-mail address: [email protected] ŽT. Asai. 1383-5769r00r$ - see front matter Q 2000 Published by Elsevier Science Ireland Ltd. All rights reserved. PII: S 1 3 8 3 - 5 7 6 9 Ž 9 9 . 0 0 0 1 9 - 7
216
K. Nakajima-Nakano et al. r Parasitology International 48 (2000) 215]222
refolded NTPase-I in the linear regression analysis. There was no significant difference observed in the antigenicity between refolded and denatured NTPase-I, nor between the isoforms. Q 2000 Published by Elsevier Science Ireland Ltd. All rights reserved. Keywords: Toxoplasma gondii; NTPase; Recombinant antigen; Dye test; Serodiagnosis
1. Introduction For acute acquired infection with Toxoplasma gondii in immunocompetent humans, the Sabin] Feldman dye test w1x has been the standard for the diagnosis of the serologic tests because of its sensitivity and specificity. However, the dye test is available only in limited numbers for laboratories because of its requirement of live tachyzoites. Therefore, the development of simple and low economical serologic test that is equivalent to the dye test is highly required. We previously reported that dithiothreitolactivated nucleoside triphosphate hydrolase ŽNTPase. is one of the major proteins in tachyzoites of T. gondii w2x. We also reported high correlation in antibody titers between the dye test and the enzyme-linked immunosorbent assay ŽELISA. for detecting immunoglobulin G antibodies against NTPase w3x. Although this ELISA is useful in diagnosing of acute toxoplasmosis, the method requires monoclonal antibody against NTPase and tachyzoite cell extracts. Furthermore, we recently found that there were two isoforms termed NTPase-I and NTPase-II, and the gene encoding NTPase-II was found in all strains of T. gondii while the NTPase-I was confined only to virulent strains w4x. These facts have raised the necessity for the improvement of the ELISA procedure. In particular, direct antigen absorption to a microtiter plate is helpful for saving time and it is necessary to determine which isoform is suitable for the ELISA. For the improvement of ELISA procedure, a production of recombinant antigen in bacteria or yeast cells is quite helpful. It saves time and the antigen supply is economical for the procedure. However, it is reported that only 7% of human sera containing antibodies to the T. gondii tachyzoite had reacted positively in an ELISA using incomplete length Ž1r3. of recombinant NTPase
w5x. Therefore, we produced full lengths of recombinant NTPase isoforms expressed in Escherichia coli as inclusion bodies and one of the two isoforms was refolded to study whether the refolding of isoform makes it a better antigen. In this paper, we also report the evaluation of antigenicities of these recombinant antigens by comparison with the Sabin]Feldman dye test titer.
2. Materials and methods 2.1. Expressions of recombinant NTPase isoforms The NTPase isoform genes ŽNTPase-I and NTPase-II. excluding the coding regions of N-terminal signal sequences were amplified from cDNA clones, pBluescript II SKŽq. plasmids carrying either Type-I or Type-II gene w4x, using polymerase chain reaction ŽPCR. Žforward oligo 59TACCATATGACAGACTCATCGTCACTCCG39; reverse oligo 59-TCGCTCGAGTCACAGATTGTGAGAATATCC-39.. Both isoform genes were amplified with the same primer set. The amplified fragments were cloned into the NdeI]XhoI sites of the vector pET22bq ŽNovagen, Madison, WI, USA., and transformed into BL21 Escherichia coli host cells. The clone was confirmed with sequencing some randomly selected transformants of the plasmids. Sequencing was done by Sanger dideoxy termination reactions w6x. Expressions of recombinant NTPase isoforms were induced with 0.2 mM isopropyl-b-thiogalactoside for 2.5 h. 2.2. Purification of recombinant NTPase isoforms Purification of two NTPase isoforms was both performed with the same procedure. Inclusion bodies of NTPase isoform were dissolved in 8 M
K. Nakajima-Nakano et al. r Parasitology International 48 (2000) 215]222
urea in 20 mM Tris]HCl ŽpH 8.5. and applied to a column of DEAE-TOYOPEAL 650 S ŽTOSOH Corp., Tokyo, Japan., previously equilibrated with 20 mM Tris]HCl ŽpH 8. containing 6 M urea. The NTPase isoform was eluted with a 500-ml linear gradient from 0 to 0.5 M KCl in 20 mM Tris]HCl ŽpH 8. containing 6 M urea. The NTPase isoform eluted as the symmetrical peak, which was analyzed by sodium dodecyl sulfate ] polyacrylamide gel electrophoresis ŽSDS]PAGE. as described w7x. The pooled fractions of the peak were concentrated by ultrafiltration and applied to a gel filtration column of TOYOPEAL HW 55 S ŽTOSOH. equilibrated with 20 mM Tris]HCl ŽpH 7.5. containing 0.2 M KCl and 6 M urea. Elution was carried out with the same solution, and fractions containing only the isoform judged by SDS]PAGE were pooled and concentrated by ultrafiltration. The concentrated samples were stored at y808C until use. 2.3. Refolding of NTPase-I isoform as an acti¨ e form Inclusion bodies of NTPase-I isoform from 500 ml of culture fluid were dissolved in 5 ml of 6 M guanidine HCl in 50 mM Tris-HCl ŽpH 7.5. and diluted six times immediately with 50 mM Tris]HCl ŽpH 7.5.. Then the diluted solution Ž30 ml. was kept at 48C overnight, and dialyzed at 48C for 24 h against 3 l 50 mM Tris]HCl ŽpH 7.5.. The dialyzed solution containing refolded NTPase-I, checked by NTPase assay as described w4x, was brought to pH 8.5 by adding 1 M Tris base and applied to a DEAE-TOYOPEAL 650 S column equilibrated with 20 mM Tris]HCl ŽpH 8.5., and eluted with a linear gradient from 0 to 0.5 M KCl. Peak active fractions were pooled, concentrated, and applied to TOYOPEAL HW55 S column equilibrated with 50 mM Tris]HCl ŽpH 7.5. containing 0.2 M KCl. Peak active fractions were checked for purity by SDS]PAGE and Coomassie Blue staining. Fractions containing no other proteins except NTPase-I were pooled, concentrated, and added glycerol Žfinal 20% wrv.. The purified enzyme was stored at y808C until use.
217
2.4. Enzyme linked immunosorbent assay (ELISA) The purified antigen was diluted with 0.1 M bicarbonate buffer ŽpH 9.6. to a concentration of 10 mgrml, and 100 ml was pipetted into wells of a polystyrene microtiter plate ŽNunk, Roskilde, Denmark.. After 1 h at 378C, the wells were washed once with phosphate-buffered saline ŽPBS. containing 0.05% Tween 20 ŽPBS]Tween.. To each well was added 150 ml of PBS containing 0.5% skim milk. After 1 h at 378C, the wells were again washed once with PBS]Tween. Diluted serum sample Ž100 ml. of each patient, 100-fold with PBS containing 0.1% skim milk ŽPBS]SM., was transferred to a microtiter plate. After 1 h at 378C, the wells were washed five times with PBS]Tween. Subsequently, 100 ml of diluted Ž1000-fold with PBS]SM. goat anti-human IgG antibody Žaffinity-purified. conjugated with horse-radish peroxidase ŽCappel Laboratories, Inc., Malvern, PA, USA. was added to each well, and the preparation was incubated at 378C for 1 h. The wells were washed five times with PBS]Tween. Then 100 ml of 50 mM citrate] phosphate buffer ŽpH 4.9. containing 0.06 mg of 2,29-azino-bis-Ž3-ethylbenzthiazoline-6-sulfonic acid. and 0.003% H 2 O 2 were added to each well. After 1 h at 258C, the developed color intensities were read at 405 nm with a microplate photometer Ž550 EIA READER; Bio-Rad, Hercules, CA, USA.. 2.5. Other methods The dye test was performed as described w1x. Protein was determined by the dye-binding procedure described by Bradford w8x using bovine serum albumin as a standard, unless otherwise stated.
3. Results 3.1. Purification of recombinant NTPase isoforms In order to evaluate antigenicities of recombinant NTPase isoforms, designated as dNTPase-I and dNTPase-II, respectively, the proteins have
218
K. Nakajima-Nakano et al. r Parasitology International 48 (2000) 215]222
been expressed in E. coli and purified under denaturing conditions, or in the presence of urea. Furthermore, NTPase-I was refolded as an active form; designated rNTPase-I, and purified under non-denaturing condition for the evaluation. The specific activity of purified rNTPase was the same as of the native NTPase-I purified from tachyzoite cells. Approximately 5% of NTPase-I proteins in inclusion bodies were refolded as the active form Ždata not shown.. A SDS]PAGE profile of the homogeneously purified proteins by using DEAE-TOYOPEAL 650 S and TOYOPEAL HW 55 S is shown in Fig. 1. Apparent molecular weights of both purified NTPase isoforms were estimated to be 67 kDa by comparison of positions with corresponding values for seven proteins from a protein maker kit ŽNovagen, Madison, WI, USA.. These results indicate that the antigens were correctly expressed in size. 3.2. E¨ aluation of antigenicities of the recombinant proteins We performed the ELISA using three kinds of antigens: dNTPase-I, dNTPase-II and rNTPase-I, and sera from the Sabin]Feldman dye test negative and positive humans. Scatter plots for the comparison between ELISA absorbances using these three antigens and the Sabin]Feldman dye test titer are shown in Figs. 2]4. Thirty sera were collected from healthy donors for the negative control, and 73 sera were used for the dye test positive evaluation group. The positive sera were collected within the past 5 years at Jikei University Hospital and Keio University Hospital, and stored in a deep freezer at y808C until use. The dye test titers of these sera were checked again at the same time when the ELISA was performed. The ELISA absorbances of the negative control sera using all three kinds of recombinant antigens were very insignificant and no significantly high absorbance was observed ŽFigs. 2]4.. On the basis of these results, we determined the ELISA absorbances of greater than the mean value q 3 S.D. of the negative control to be positive for each antigen, respectively. Under the condition ruled, all of the 30 sera from healthy control humans were tested negative for every
Fig. 1. SDS]PAGE of purified recombinant antigens expressed in E. coli. Electrophoresis was carried out on 8% of polyacrylamide gel with the same buffer system as described w7x. Lane A, recombinant protein makers ŽThe Perfect Protein Makers; Novagen, Madison, WI, USA.. Lane B, 3 mg refolded and purified NTPase-I under non-denaturing condition ŽrNTPase-I.. Lane C, 3 mg purified NTPase-I under denaturing condition ŽdNTPase-I.. Lane D, 3 mg purified NTPase-II under denaturing condition ŽdNTPase-II.. The proteins were detected by Coomassie Brilliant Blue R-250 staining. Molecular mass is expressed in kDa.
recombinant antigens ŽFigs. 2]4.. When dNTPase-I was used as the antigen for the dye test positive sera, 82% of the sera Ž60r73. were positive ŽFig. 2.. Similarly, 78% of sera Ž57r73. were positive for dNTPase-II ŽFig. 3. and 63% of sera Ž46r73. were positive for rNTPase-I ŽFig. 4.. For all three antigen types of recombinant NTPase, the positive rates of sera showing the dye test titer more than 1:1024, those suspected to be acute toxoplasmosis patient sera, were 93% Ž13r14. ŽFigs. 2]4.. A serum in this suspected acute toxoplasmosis group was tested negative for all three antigens. Furthermore, this serum was
K. Nakajima-Nakano et al. r Parasitology International 48 (2000) 215]222
219
Fig. 2. Scatter plot for comparison between ELISA absorbance using denatured recombinant NTPase-I as the antigen and the Sabin]Feldman dye test titer. The dotted line represents the cutoff value for positive Žmean q 3 S.D. of the healthy control group.. The solid lines indicate the mean absorbances.
Fig. 3. Scatter plot for comparison between ELISA absorbance using denatured recombinant NTPase-II as the antigen and the Sabin]Feldman dye test titer. The dotted line represents the cutoff value for positive Žmean q 3 S.D. of the healthy control group.. The solid lines indicate the mean absorbances.
also tested negative for the native NTPase purified from tachyzoite cells Ždata not shown.. Although some negative sera were present, high correlations between ELISA absorbance and the dye test titer were observed when these three antigen types of NTPase were used ŽFigs. 2]4.. The correlation coefficients between the ELISA absorbance and the dye test titer were 0.583 Ž r 2 . for dNTPase-I, 0.472 Ž r 2 . for dNTPase-II and 0.604 Ž r 2 . for rNTPase-I in the linear regression analysis. A scatter plot for the comparison of ELISA absorbances measured in using dNTPase-I and rNTPase-I as the antigens is shown in Fig. 5. A very high correlation between these ELISA tests was observed and the correlation coefficient between them was 0.964 Ž r 2 . in the linear regression analysis. A scatter plot for comparison of
ELISA absorbances measured in using dNTPase-I and dNTPase-II as the antigens is shown in Fig. 6. Similarly, a significantly high correlation between the ELISA absorbances in those two ELISA tests was observed, and the correlation coefficient between them was 0.867 Ž r 2 . in the linear regression analysis.
4. Discussion We have produced recombinant NTPases and tested whether they were useful as antigens for the serodiagnosis of acute toxoplasmosis in immunocompetent humans. To evaluate these recombinant antigens, we compared the ELISA absorbances using recombinant NTPases with the dye test titer. There are several advantages in
220
K. Nakajima-Nakano et al. r Parasitology International 48 (2000) 215]222
Fig. 6. Scatter plot for comparison of ELISA absorbances measured in using denatured NTPase-I ŽdNTPase-I. and denatured NTPase-II ŽdNTPase-II..
Fig. 4. Scatter plot for comparison between ELISA absorbance using refolded NTPase-I as the antigen and the Sabin] Feldman dye test titer. The dotted line represents the cutoff value for positive Žmean q 3 S.D. of the healthy control group.. The solid lines indicate the mean absorbances.
having the NTPases as the antigen: Ž1. dithiothreitol-activated NTPases are thought to be restricted to T. gondii and Neospora caninum w9,10x;
Fig. 5. Scatter plot for comparison of ELISA absorbances measured in using refolded NTPase-I ŽrNTPase-I. and denatured NTPase-I ŽdNTPase-I..
Ž2. NTPase genes are expressed in tachyzoite form, which is the causative form of acute toxoplasmosis, but down regulated in bradyzoite form, which is the chronic infection form w11x; Ž3. NTPases are the major proteins of tachyzoite cells w2x; Ž4. NTPases are secreted from tachyzoite cells after invading the host cell w4,12x, and detected as a circulating antigen in the blood w13x; and Ž5. NTPases are the soluble enzymes, and no hydrophobic cluster region is present in their molecules w2,4x. The results obtained in this study, with reflection of the advantages listed above, are discussed below. Dithiothreitol-activated NTPases were identified only in T. gondii and N. caninum and not in other organisms including the related apicomplexan parasites of Plasmodium berghei w9x and Eimeria tenella ŽT. Asai, unpublished data.. Neospora caninum infection is quite unlikely to occur because N. caninum has not been reported to infect human w14x. Therefore, we did not obtain any significance in the ELISA absorbance for the tests in the negative control group ŽFigs. 2]4.. Furthermore, we checked a possible double infections with T. gondii and N. caninum in the dye test positive group by using a recombinant NTPase of N. caninum produced as described w10x. Although N. caninum NTPase shares approximately 69% identity to T. gondii NTPase isoforms in the predicted amino acid sequence w10x, only two positive sera for N. caninum NTPase were found. It was confirmed that these two
K. Nakajima-Nakano et al. r Parasitology International 48 (2000) 215]222
patients were not infected with N. caninum since the antibodies in these sera reacted to T. gondii antigens but did not react to several antigens of N. caninum Žprovided by Dr L. David Sibley, Washington University, USA. except NTPase Ždata not shown.. The dye test has been used for half a century and still has an important place as a gold standard in serodiagnosis of acute toxoplasmosis in immunocompetent humans. The most valid reason for this is that antibodies in concern with the dye test are highly specific to the cell surface antigens of tachyzoite form w15x. On the other hand, most of commercially available serodiagnostic tests are made with a crude extract from tachyzoite cells containing several common antigens to bradyzoite due to the easy handling of antigens and low economical means. Accordingly, a major problem of commercially available tests is that many sera from chronic toxoplasmosis patients show a high serologic titer by the tests. Since NTPases are down regulated in their expression in bradyzoite form, antibody response to NTPase by chronic patient is assumed to be weak. In fact, there were no significantly high ELISA absorbance observed by the sera from chronic cases with the dye test titers less than 1:256 ŽFigs. 2]4.. This suggests the fact of the advantage of NTPase as the antigen for serodiagnosis. A cellular content of NTPases is quite high, 2]8% of total parasite protein w2,13,16x, indicating that a large quantity of NTPases is exposed to host immunosurveillance system. A high correlation in antibody titers between the dye test and the ELISA detecting IgG antibodies to NTPase reported previously w3x and the results obtained in this study ŽFigs. 2]4. indicate that NTPases were highly recognized by the host immunosurveillance system. The production of recombinant NTPases in E. coli is quite easy because NTPases are the soluble enzymes. This is another one of the advantages of NTPase as the antigen. It is reported that only the refolded recombinant surface antigen 1 ŽSAG1; previously named P30. expressed in eukaryotic cells had significant antigenicity w17x. This is one of the disadvantages of a recombinant antigen expressed in E. coli. In our case, however,
221
there was no significant difference observed in the antigenicity between refolded and denatured NTPase-I ŽFig. 5.. This indicates that the production of recombinant NTPase with high antigenicity would be easily performed. A refolding of NTPase-II activity with the same protocol for NTPase-I has been unsuccessful despite the high homology of amino acid sequences Žmore than 97%. between NTPase-I and NTPase-II. Nakaar et al. w18x reported a successful refolding of NTPase-II activity, however, the efficiency in refolding of NTPase-II activity is too low to perform the ELISA. There was no significant difference in the antigenicity between NTPase-I and NTPase-II ŽFig. 6.. The data suggest that either recombinant NTPase-I or NTPase-II purified under denaturing condition is enough to obtain high correlation between ELISA absorbance and the Sabin]Feldman dye test titer. Although high correlation between ELISA absorbance using recombinant NTPase as the antigen and dye test titer was observed, there are some weak points for recombinant NTPase explained as follows: Ž1. total positivity is incomplete Ž63]82%.; and Ž2. there was one negative serum, only one but definitely, among the suspected acute toxoplasmosis group. The former may suggest an unknown difference between native and recombinant NTPase. The latter suggests that a patient history with immunodisorder, i.e. lymphoma, cancer bearing, etc. should be considered in the future since we could not follow the histories of patients. The characterization of molecular feature of refolded NTPase, and the relation between the ELISA absorbance and history of patients are presently under investigation.
Acknowledgements We thank Dr L. David Sibley, Washington University School of Medicine, for helpful discussions in producing recombinant NTPases and refolding of NTPase-I. This work was supported in part by the National Grant-in-Aid from the Ministry of Education of Japan Ž10670236., the Japan Society for the Promotion of Science ŽJSPSRFTF97L00701., and Promotion of AIDS Re-
222
K. Nakajima-Nakano et al. r Parasitology International 48 (2000) 215]222
search from The Ministry of Health and Welfare of Japan.
w10x
References w1x Sabin AB, Feldman HA. Dyes as microchemical indicators of a new immunochemical phenomenon affecting a protozoan parasite ŽToxoplasma.. Science 1948;108: 660]3. w2x Asai T, O’Sullivan WJ, Tatibana M. A potent nucleoside triphosphate hydrolase from the parasitic protozoan Toxoplasma gondii. J Biol Chem 1983;258:6816]22. w3x Asai T, Mizuno F, Kojima S, Takeuchi T, Kobayashi A, Suzuki Y. High correlation in antibody titers between the Sabin]Feldman dye test and an enzyme-linked immunosorbent assay detecting immunoglobulin G antibodies to the nucleoside triphosphate hydrolase of Toxoplasma gondii. J Clin Microbiol 1992;30:1291]3. w4x Asai T, Miura S, Sibley LD, Okabayashi H, Takeuchi T. Biochemical and molecular characterization of nucleoside triphosphate hydrolase isozymes from the parasitic protozoan Toxoplasma gondii. J Biol Chem 1995;270: 11391]7. w5x Tenter AM, Johnson AM. Human antibody response to the nucleoside triphosphate hydrolase of Toxoplasma gondii. J Immunoassay 1990:579]90. w6x Sanger F, Nicklen S, Coulson AR. DNA sequencing with chain-terminating inhibitors. Proc Natl Acad Sci USA 1977;74:5463]7. w7x Laemmli UK. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 1970;227:680]5. w8x Bradford MM. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 1979;72:248]54. w9x Asai T, Kanazawa T, Kobayashi S, Takeuchi T, Kim T. Do protozoa conceal a high potency of nucleoside
w11x
w12x
w13x
w14x w15x
w16x
w17x
w18x
triphosphate hydrolysis present in Toxoplasma gondii. Comp Biochem Physiol 1986;85B:365]8. Asai T, Howe DK, Nakajima K, Nozaki T, Takeuchi T, Sibley LD. Neospora caninum: tachyzoites express a potent type-I nucleoside triphosphate hydrolase, but lack nucleoside diphosphate hydrolase activity. Exp Parasitol 1998;90:277]85. Ferguson DJ, Cesbron-Delauw MF, Dubremetz JF, Sibley LD, Joiner KA, Wright S. The expression and distribution of dense granule proteins in the enteric ŽCoccidian. forms of Toxoplasma gondii in the small intestine of the cat. Exp Parasitol 1999;91:203]11. Sibley LD, Niesman IR, Asai T, Takeuchi T. Toxoplasma gondii: secretion of a potent nucleoside triphosphate hydrolase into parasitophorous vacuole. Exp Parasitol 1994;79:301]11. Asai T, Kim T, Kobayashi M, Kojima S. Detection of nucleoside triphosphate hydrolase as a circulating antigen in sera of mice infected with Toxoplasma gondii. Infect Immun 1987;55:1332]5. Dubey JP, Lindsay DS. A review of Neospora caninum and neosporosis. Vet Parasitol 1996;67:1]59. Handman E, Remington JS. Serological and immunochemical characterization of monoclonal antibodies to Toxoplasma gondii. Immunology 1980;40: 579]88. Nakaar V, Bermudes D, Peck KR, Joiner KA. Upstream elements required for expression of nucleoside triphosphate hydrolase genes of Toxoplasma gondii. Mol Biochem Parasitol 1998;92:229]39. Xiong C, Grieve RB, Kim K, Boothroyd JC. Expression of Toxoplasma gondii P30 as fusions with glutathione S-transferase in animal cells by Sindbis recombinant virus. Mol Biochem Parasitol 1993;61:143]8. Nakaar V, Beckers CJM, Polotsky V, Joiner KA. Basis for substrate specificity of the Toxoplasma gondii nucleoside triphosphate hydrolase. Mol Biochem Parasitol 1998;97:209]20.
Evaluation of serodiagnosis of toxoplasmosis by using the recombinant nucleoside triphosphate hydrolase isoforms expressed in Escherichia coli q Kyoko Nakajima-Nakano a,b, Asao Makiokac , Naoya Yamashitab, Nobutake Matsuo b, Takashi Asai a,U a
Department of Tropical Medicine and Parasitology, Keio Uni¨ ersity School of Medicine, 35 Shinanomachi Shinjuku-ku, Tokyo 160-8582, Japan b Department of Pediatrics, Keio Uni¨ ersity School of Medicine, Shinjuku-ku, Tokyo 160-8582, Japan c Department of Tropical Medicine, Jikei Uni¨ ersity School of Medicine, 3-25-8 Nishishinbashi, Minato-ku, Tokyo 105-8461, Japan Received 8 June 1999; accepted 19 July 1999
Abstract The nucleoside triphosphate hydrolase ŽNTPase. isoforms termed, NTPase-I and NTPase-II of Toxoplasma gondii, were expressed in Escherichia coli as inclusion bodies and purified under denaturing condition. Furthermore, NTPase-I was refolded as an active form and purified under non-denaturing condition. The purified NTPase isoforms, both denatured and refolded, were tested for their usefulness as antigens for the serodiagnosis of acute toxoplasmosis in immunocompetent humans. The test was conducted by using the recombinant NTPase isoforms and comparing the enzyme linked immunosorbent assay ŽELISA. absorbances with the Sabin]Feldman dye test titer. Seventy-three sera from dye test-positive patients, and 30 sera from subjects with no T. gondii infection were examined. The total positive rates in dye test positive sera were: 82% Ž60r73. for denatured NTPase-I; 78% Ž57r73. for denatured NTPase-II; and 63% Ž46r73. for refolded NTPase-I. For all three antigen types of recombinant NTPase, the positive rates of sera of acute toxoplasmosis suspected patients were 93% Ž13r14.. A moderate correlation between the ELISA absorbance using these antigens and the dye test titer was observed with the correlation coefficients, 0.583 Ž r 2 . for denatured NTPase-I, 0.472 Ž r 2 . for denatured NTPase-II, and 0.604 Ž r 2 . for
Abbre¨ iations: ELISA, enzyme-linked immunosorbent assay; NTPase, nucleoside triphosphate hydrolase; PCR, polymerase chain reaction; SDS]PAGE, sodium dodecyl sulfate]polyacrylamide gel electrophoresis. q This study forms part of the dissertation submitted to Keio University School of Medicine in partial fulfillment of the requirement for the Degree of Medical Sciences. U Corresponding author. Tel.: q81-3-3353-1211, ext. 62747; fax: q81-3-3353-5958. E-mail address: [email protected] ŽT. Asai. 1383-5769r00r$ - see front matter Q 2000 Published by Elsevier Science Ireland Ltd. All rights reserved. PII: S 1 3 8 3 - 5 7 6 9 Ž 9 9 . 0 0 0 1 9 - 7
216
K. Nakajima-Nakano et al. r Parasitology International 48 (2000) 215]222
refolded NTPase-I in the linear regression analysis. There was no significant difference observed in the antigenicity between refolded and denatured NTPase-I, nor between the isoforms. Q 2000 Published by Elsevier Science Ireland Ltd. All rights reserved. Keywords: Toxoplasma gondii; NTPase; Recombinant antigen; Dye test; Serodiagnosis
1. Introduction For acute acquired infection with Toxoplasma gondii in immunocompetent humans, the Sabin] Feldman dye test w1x has been the standard for the diagnosis of the serologic tests because of its sensitivity and specificity. However, the dye test is available only in limited numbers for laboratories because of its requirement of live tachyzoites. Therefore, the development of simple and low economical serologic test that is equivalent to the dye test is highly required. We previously reported that dithiothreitolactivated nucleoside triphosphate hydrolase ŽNTPase. is one of the major proteins in tachyzoites of T. gondii w2x. We also reported high correlation in antibody titers between the dye test and the enzyme-linked immunosorbent assay ŽELISA. for detecting immunoglobulin G antibodies against NTPase w3x. Although this ELISA is useful in diagnosing of acute toxoplasmosis, the method requires monoclonal antibody against NTPase and tachyzoite cell extracts. Furthermore, we recently found that there were two isoforms termed NTPase-I and NTPase-II, and the gene encoding NTPase-II was found in all strains of T. gondii while the NTPase-I was confined only to virulent strains w4x. These facts have raised the necessity for the improvement of the ELISA procedure. In particular, direct antigen absorption to a microtiter plate is helpful for saving time and it is necessary to determine which isoform is suitable for the ELISA. For the improvement of ELISA procedure, a production of recombinant antigen in bacteria or yeast cells is quite helpful. It saves time and the antigen supply is economical for the procedure. However, it is reported that only 7% of human sera containing antibodies to the T. gondii tachyzoite had reacted positively in an ELISA using incomplete length Ž1r3. of recombinant NTPase
w5x. Therefore, we produced full lengths of recombinant NTPase isoforms expressed in Escherichia coli as inclusion bodies and one of the two isoforms was refolded to study whether the refolding of isoform makes it a better antigen. In this paper, we also report the evaluation of antigenicities of these recombinant antigens by comparison with the Sabin]Feldman dye test titer.
2. Materials and methods 2.1. Expressions of recombinant NTPase isoforms The NTPase isoform genes ŽNTPase-I and NTPase-II. excluding the coding regions of N-terminal signal sequences were amplified from cDNA clones, pBluescript II SKŽq. plasmids carrying either Type-I or Type-II gene w4x, using polymerase chain reaction ŽPCR. Žforward oligo 59TACCATATGACAGACTCATCGTCACTCCG39; reverse oligo 59-TCGCTCGAGTCACAGATTGTGAGAATATCC-39.. Both isoform genes were amplified with the same primer set. The amplified fragments were cloned into the NdeI]XhoI sites of the vector pET22bq ŽNovagen, Madison, WI, USA., and transformed into BL21 Escherichia coli host cells. The clone was confirmed with sequencing some randomly selected transformants of the plasmids. Sequencing was done by Sanger dideoxy termination reactions w6x. Expressions of recombinant NTPase isoforms were induced with 0.2 mM isopropyl-b-thiogalactoside for 2.5 h. 2.2. Purification of recombinant NTPase isoforms Purification of two NTPase isoforms was both performed with the same procedure. Inclusion bodies of NTPase isoform were dissolved in 8 M
K. Nakajima-Nakano et al. r Parasitology International 48 (2000) 215]222
urea in 20 mM Tris]HCl ŽpH 8.5. and applied to a column of DEAE-TOYOPEAL 650 S ŽTOSOH Corp., Tokyo, Japan., previously equilibrated with 20 mM Tris]HCl ŽpH 8. containing 6 M urea. The NTPase isoform was eluted with a 500-ml linear gradient from 0 to 0.5 M KCl in 20 mM Tris]HCl ŽpH 8. containing 6 M urea. The NTPase isoform eluted as the symmetrical peak, which was analyzed by sodium dodecyl sulfate ] polyacrylamide gel electrophoresis ŽSDS]PAGE. as described w7x. The pooled fractions of the peak were concentrated by ultrafiltration and applied to a gel filtration column of TOYOPEAL HW 55 S ŽTOSOH. equilibrated with 20 mM Tris]HCl ŽpH 7.5. containing 0.2 M KCl and 6 M urea. Elution was carried out with the same solution, and fractions containing only the isoform judged by SDS]PAGE were pooled and concentrated by ultrafiltration. The concentrated samples were stored at y808C until use. 2.3. Refolding of NTPase-I isoform as an acti¨ e form Inclusion bodies of NTPase-I isoform from 500 ml of culture fluid were dissolved in 5 ml of 6 M guanidine HCl in 50 mM Tris-HCl ŽpH 7.5. and diluted six times immediately with 50 mM Tris]HCl ŽpH 7.5.. Then the diluted solution Ž30 ml. was kept at 48C overnight, and dialyzed at 48C for 24 h against 3 l 50 mM Tris]HCl ŽpH 7.5.. The dialyzed solution containing refolded NTPase-I, checked by NTPase assay as described w4x, was brought to pH 8.5 by adding 1 M Tris base and applied to a DEAE-TOYOPEAL 650 S column equilibrated with 20 mM Tris]HCl ŽpH 8.5., and eluted with a linear gradient from 0 to 0.5 M KCl. Peak active fractions were pooled, concentrated, and applied to TOYOPEAL HW55 S column equilibrated with 50 mM Tris]HCl ŽpH 7.5. containing 0.2 M KCl. Peak active fractions were checked for purity by SDS]PAGE and Coomassie Blue staining. Fractions containing no other proteins except NTPase-I were pooled, concentrated, and added glycerol Žfinal 20% wrv.. The purified enzyme was stored at y808C until use.
217
2.4. Enzyme linked immunosorbent assay (ELISA) The purified antigen was diluted with 0.1 M bicarbonate buffer ŽpH 9.6. to a concentration of 10 mgrml, and 100 ml was pipetted into wells of a polystyrene microtiter plate ŽNunk, Roskilde, Denmark.. After 1 h at 378C, the wells were washed once with phosphate-buffered saline ŽPBS. containing 0.05% Tween 20 ŽPBS]Tween.. To each well was added 150 ml of PBS containing 0.5% skim milk. After 1 h at 378C, the wells were again washed once with PBS]Tween. Diluted serum sample Ž100 ml. of each patient, 100-fold with PBS containing 0.1% skim milk ŽPBS]SM., was transferred to a microtiter plate. After 1 h at 378C, the wells were washed five times with PBS]Tween. Subsequently, 100 ml of diluted Ž1000-fold with PBS]SM. goat anti-human IgG antibody Žaffinity-purified. conjugated with horse-radish peroxidase ŽCappel Laboratories, Inc., Malvern, PA, USA. was added to each well, and the preparation was incubated at 378C for 1 h. The wells were washed five times with PBS]Tween. Then 100 ml of 50 mM citrate] phosphate buffer ŽpH 4.9. containing 0.06 mg of 2,29-azino-bis-Ž3-ethylbenzthiazoline-6-sulfonic acid. and 0.003% H 2 O 2 were added to each well. After 1 h at 258C, the developed color intensities were read at 405 nm with a microplate photometer Ž550 EIA READER; Bio-Rad, Hercules, CA, USA.. 2.5. Other methods The dye test was performed as described w1x. Protein was determined by the dye-binding procedure described by Bradford w8x using bovine serum albumin as a standard, unless otherwise stated.
3. Results 3.1. Purification of recombinant NTPase isoforms In order to evaluate antigenicities of recombinant NTPase isoforms, designated as dNTPase-I and dNTPase-II, respectively, the proteins have
218
K. Nakajima-Nakano et al. r Parasitology International 48 (2000) 215]222
been expressed in E. coli and purified under denaturing conditions, or in the presence of urea. Furthermore, NTPase-I was refolded as an active form; designated rNTPase-I, and purified under non-denaturing condition for the evaluation. The specific activity of purified rNTPase was the same as of the native NTPase-I purified from tachyzoite cells. Approximately 5% of NTPase-I proteins in inclusion bodies were refolded as the active form Ždata not shown.. A SDS]PAGE profile of the homogeneously purified proteins by using DEAE-TOYOPEAL 650 S and TOYOPEAL HW 55 S is shown in Fig. 1. Apparent molecular weights of both purified NTPase isoforms were estimated to be 67 kDa by comparison of positions with corresponding values for seven proteins from a protein maker kit ŽNovagen, Madison, WI, USA.. These results indicate that the antigens were correctly expressed in size. 3.2. E¨ aluation of antigenicities of the recombinant proteins We performed the ELISA using three kinds of antigens: dNTPase-I, dNTPase-II and rNTPase-I, and sera from the Sabin]Feldman dye test negative and positive humans. Scatter plots for the comparison between ELISA absorbances using these three antigens and the Sabin]Feldman dye test titer are shown in Figs. 2]4. Thirty sera were collected from healthy donors for the negative control, and 73 sera were used for the dye test positive evaluation group. The positive sera were collected within the past 5 years at Jikei University Hospital and Keio University Hospital, and stored in a deep freezer at y808C until use. The dye test titers of these sera were checked again at the same time when the ELISA was performed. The ELISA absorbances of the negative control sera using all three kinds of recombinant antigens were very insignificant and no significantly high absorbance was observed ŽFigs. 2]4.. On the basis of these results, we determined the ELISA absorbances of greater than the mean value q 3 S.D. of the negative control to be positive for each antigen, respectively. Under the condition ruled, all of the 30 sera from healthy control humans were tested negative for every
Fig. 1. SDS]PAGE of purified recombinant antigens expressed in E. coli. Electrophoresis was carried out on 8% of polyacrylamide gel with the same buffer system as described w7x. Lane A, recombinant protein makers ŽThe Perfect Protein Makers; Novagen, Madison, WI, USA.. Lane B, 3 mg refolded and purified NTPase-I under non-denaturing condition ŽrNTPase-I.. Lane C, 3 mg purified NTPase-I under denaturing condition ŽdNTPase-I.. Lane D, 3 mg purified NTPase-II under denaturing condition ŽdNTPase-II.. The proteins were detected by Coomassie Brilliant Blue R-250 staining. Molecular mass is expressed in kDa.
recombinant antigens ŽFigs. 2]4.. When dNTPase-I was used as the antigen for the dye test positive sera, 82% of the sera Ž60r73. were positive ŽFig. 2.. Similarly, 78% of sera Ž57r73. were positive for dNTPase-II ŽFig. 3. and 63% of sera Ž46r73. were positive for rNTPase-I ŽFig. 4.. For all three antigen types of recombinant NTPase, the positive rates of sera showing the dye test titer more than 1:1024, those suspected to be acute toxoplasmosis patient sera, were 93% Ž13r14. ŽFigs. 2]4.. A serum in this suspected acute toxoplasmosis group was tested negative for all three antigens. Furthermore, this serum was
K. Nakajima-Nakano et al. r Parasitology International 48 (2000) 215]222
219
Fig. 2. Scatter plot for comparison between ELISA absorbance using denatured recombinant NTPase-I as the antigen and the Sabin]Feldman dye test titer. The dotted line represents the cutoff value for positive Žmean q 3 S.D. of the healthy control group.. The solid lines indicate the mean absorbances.
Fig. 3. Scatter plot for comparison between ELISA absorbance using denatured recombinant NTPase-II as the antigen and the Sabin]Feldman dye test titer. The dotted line represents the cutoff value for positive Žmean q 3 S.D. of the healthy control group.. The solid lines indicate the mean absorbances.
also tested negative for the native NTPase purified from tachyzoite cells Ždata not shown.. Although some negative sera were present, high correlations between ELISA absorbance and the dye test titer were observed when these three antigen types of NTPase were used ŽFigs. 2]4.. The correlation coefficients between the ELISA absorbance and the dye test titer were 0.583 Ž r 2 . for dNTPase-I, 0.472 Ž r 2 . for dNTPase-II and 0.604 Ž r 2 . for rNTPase-I in the linear regression analysis. A scatter plot for the comparison of ELISA absorbances measured in using dNTPase-I and rNTPase-I as the antigens is shown in Fig. 5. A very high correlation between these ELISA tests was observed and the correlation coefficient between them was 0.964 Ž r 2 . in the linear regression analysis. A scatter plot for comparison of
ELISA absorbances measured in using dNTPase-I and dNTPase-II as the antigens is shown in Fig. 6. Similarly, a significantly high correlation between the ELISA absorbances in those two ELISA tests was observed, and the correlation coefficient between them was 0.867 Ž r 2 . in the linear regression analysis.
4. Discussion We have produced recombinant NTPases and tested whether they were useful as antigens for the serodiagnosis of acute toxoplasmosis in immunocompetent humans. To evaluate these recombinant antigens, we compared the ELISA absorbances using recombinant NTPases with the dye test titer. There are several advantages in
220
K. Nakajima-Nakano et al. r Parasitology International 48 (2000) 215]222
Fig. 6. Scatter plot for comparison of ELISA absorbances measured in using denatured NTPase-I ŽdNTPase-I. and denatured NTPase-II ŽdNTPase-II..
Fig. 4. Scatter plot for comparison between ELISA absorbance using refolded NTPase-I as the antigen and the Sabin] Feldman dye test titer. The dotted line represents the cutoff value for positive Žmean q 3 S.D. of the healthy control group.. The solid lines indicate the mean absorbances.
having the NTPases as the antigen: Ž1. dithiothreitol-activated NTPases are thought to be restricted to T. gondii and Neospora caninum w9,10x;
Fig. 5. Scatter plot for comparison of ELISA absorbances measured in using refolded NTPase-I ŽrNTPase-I. and denatured NTPase-I ŽdNTPase-I..
Ž2. NTPase genes are expressed in tachyzoite form, which is the causative form of acute toxoplasmosis, but down regulated in bradyzoite form, which is the chronic infection form w11x; Ž3. NTPases are the major proteins of tachyzoite cells w2x; Ž4. NTPases are secreted from tachyzoite cells after invading the host cell w4,12x, and detected as a circulating antigen in the blood w13x; and Ž5. NTPases are the soluble enzymes, and no hydrophobic cluster region is present in their molecules w2,4x. The results obtained in this study, with reflection of the advantages listed above, are discussed below. Dithiothreitol-activated NTPases were identified only in T. gondii and N. caninum and not in other organisms including the related apicomplexan parasites of Plasmodium berghei w9x and Eimeria tenella ŽT. Asai, unpublished data.. Neospora caninum infection is quite unlikely to occur because N. caninum has not been reported to infect human w14x. Therefore, we did not obtain any significance in the ELISA absorbance for the tests in the negative control group ŽFigs. 2]4.. Furthermore, we checked a possible double infections with T. gondii and N. caninum in the dye test positive group by using a recombinant NTPase of N. caninum produced as described w10x. Although N. caninum NTPase shares approximately 69% identity to T. gondii NTPase isoforms in the predicted amino acid sequence w10x, only two positive sera for N. caninum NTPase were found. It was confirmed that these two
K. Nakajima-Nakano et al. r Parasitology International 48 (2000) 215]222
patients were not infected with N. caninum since the antibodies in these sera reacted to T. gondii antigens but did not react to several antigens of N. caninum Žprovided by Dr L. David Sibley, Washington University, USA. except NTPase Ždata not shown.. The dye test has been used for half a century and still has an important place as a gold standard in serodiagnosis of acute toxoplasmosis in immunocompetent humans. The most valid reason for this is that antibodies in concern with the dye test are highly specific to the cell surface antigens of tachyzoite form w15x. On the other hand, most of commercially available serodiagnostic tests are made with a crude extract from tachyzoite cells containing several common antigens to bradyzoite due to the easy handling of antigens and low economical means. Accordingly, a major problem of commercially available tests is that many sera from chronic toxoplasmosis patients show a high serologic titer by the tests. Since NTPases are down regulated in their expression in bradyzoite form, antibody response to NTPase by chronic patient is assumed to be weak. In fact, there were no significantly high ELISA absorbance observed by the sera from chronic cases with the dye test titers less than 1:256 ŽFigs. 2]4.. This suggests the fact of the advantage of NTPase as the antigen for serodiagnosis. A cellular content of NTPases is quite high, 2]8% of total parasite protein w2,13,16x, indicating that a large quantity of NTPases is exposed to host immunosurveillance system. A high correlation in antibody titers between the dye test and the ELISA detecting IgG antibodies to NTPase reported previously w3x and the results obtained in this study ŽFigs. 2]4. indicate that NTPases were highly recognized by the host immunosurveillance system. The production of recombinant NTPases in E. coli is quite easy because NTPases are the soluble enzymes. This is another one of the advantages of NTPase as the antigen. It is reported that only the refolded recombinant surface antigen 1 ŽSAG1; previously named P30. expressed in eukaryotic cells had significant antigenicity w17x. This is one of the disadvantages of a recombinant antigen expressed in E. coli. In our case, however,
221
there was no significant difference observed in the antigenicity between refolded and denatured NTPase-I ŽFig. 5.. This indicates that the production of recombinant NTPase with high antigenicity would be easily performed. A refolding of NTPase-II activity with the same protocol for NTPase-I has been unsuccessful despite the high homology of amino acid sequences Žmore than 97%. between NTPase-I and NTPase-II. Nakaar et al. w18x reported a successful refolding of NTPase-II activity, however, the efficiency in refolding of NTPase-II activity is too low to perform the ELISA. There was no significant difference in the antigenicity between NTPase-I and NTPase-II ŽFig. 6.. The data suggest that either recombinant NTPase-I or NTPase-II purified under denaturing condition is enough to obtain high correlation between ELISA absorbance and the Sabin]Feldman dye test titer. Although high correlation between ELISA absorbance using recombinant NTPase as the antigen and dye test titer was observed, there are some weak points for recombinant NTPase explained as follows: Ž1. total positivity is incomplete Ž63]82%.; and Ž2. there was one negative serum, only one but definitely, among the suspected acute toxoplasmosis group. The former may suggest an unknown difference between native and recombinant NTPase. The latter suggests that a patient history with immunodisorder, i.e. lymphoma, cancer bearing, etc. should be considered in the future since we could not follow the histories of patients. The characterization of molecular feature of refolded NTPase, and the relation between the ELISA absorbance and history of patients are presently under investigation.
Acknowledgements We thank Dr L. David Sibley, Washington University School of Medicine, for helpful discussions in producing recombinant NTPases and refolding of NTPase-I. This work was supported in part by the National Grant-in-Aid from the Ministry of Education of Japan Ž10670236., the Japan Society for the Promotion of Science ŽJSPSRFTF97L00701., and Promotion of AIDS Re-
222
K. Nakajima-Nakano et al. r Parasitology International 48 (2000) 215]222
search from The Ministry of Health and Welfare of Japan.
w10x
References w1x Sabin AB, Feldman HA. Dyes as microchemical indicators of a new immunochemical phenomenon affecting a protozoan parasite ŽToxoplasma.. Science 1948;108: 660]3. w2x Asai T, O’Sullivan WJ, Tatibana M. A potent nucleoside triphosphate hydrolase from the parasitic protozoan Toxoplasma gondii. J Biol Chem 1983;258:6816]22. w3x Asai T, Mizuno F, Kojima S, Takeuchi T, Kobayashi A, Suzuki Y. High correlation in antibody titers between the Sabin]Feldman dye test and an enzyme-linked immunosorbent assay detecting immunoglobulin G antibodies to the nucleoside triphosphate hydrolase of Toxoplasma gondii. J Clin Microbiol 1992;30:1291]3. w4x Asai T, Miura S, Sibley LD, Okabayashi H, Takeuchi T. Biochemical and molecular characterization of nucleoside triphosphate hydrolase isozymes from the parasitic protozoan Toxoplasma gondii. J Biol Chem 1995;270: 11391]7. w5x Tenter AM, Johnson AM. Human antibody response to the nucleoside triphosphate hydrolase of Toxoplasma gondii. J Immunoassay 1990:579]90. w6x Sanger F, Nicklen S, Coulson AR. DNA sequencing with chain-terminating inhibitors. Proc Natl Acad Sci USA 1977;74:5463]7. w7x Laemmli UK. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 1970;227:680]5. w8x Bradford MM. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 1979;72:248]54. w9x Asai T, Kanazawa T, Kobayashi S, Takeuchi T, Kim T. Do protozoa conceal a high potency of nucleoside
w11x
w12x
w13x
w14x w15x
w16x
w17x
w18x
triphosphate hydrolysis present in Toxoplasma gondii. Comp Biochem Physiol 1986;85B:365]8. Asai T, Howe DK, Nakajima K, Nozaki T, Takeuchi T, Sibley LD. Neospora caninum: tachyzoites express a potent type-I nucleoside triphosphate hydrolase, but lack nucleoside diphosphate hydrolase activity. Exp Parasitol 1998;90:277]85. Ferguson DJ, Cesbron-Delauw MF, Dubremetz JF, Sibley LD, Joiner KA, Wright S. The expression and distribution of dense granule proteins in the enteric ŽCoccidian. forms of Toxoplasma gondii in the small intestine of the cat. Exp Parasitol 1999;91:203]11. Sibley LD, Niesman IR, Asai T, Takeuchi T. Toxoplasma gondii: secretion of a potent nucleoside triphosphate hydrolase into parasitophorous vacuole. Exp Parasitol 1994;79:301]11. Asai T, Kim T, Kobayashi M, Kojima S. Detection of nucleoside triphosphate hydrolase as a circulating antigen in sera of mice infected with Toxoplasma gondii. Infect Immun 1987;55:1332]5. Dubey JP, Lindsay DS. A review of Neospora caninum and neosporosis. Vet Parasitol 1996;67:1]59. Handman E, Remington JS. Serological and immunochemical characterization of monoclonal antibodies to Toxoplasma gondii. Immunology 1980;40: 579]88. Nakaar V, Bermudes D, Peck KR, Joiner KA. Upstream elements required for expression of nucleoside triphosphate hydrolase genes of Toxoplasma gondii. Mol Biochem Parasitol 1998;92:229]39. Xiong C, Grieve RB, Kim K, Boothroyd JC. Expression of Toxoplasma gondii P30 as fusions with glutathione S-transferase in animal cells by Sindbis recombinant virus. Mol Biochem Parasitol 1993;61:143]8. Nakaar V, Beckers CJM, Polotsky V, Joiner KA. Basis for substrate specificity of the Toxoplasma gondii nucleoside triphosphate hydrolase. Mol Biochem Parasitol 1998;97:209]20.