Targeted Reduction of Nucleoside Triphosphate Hydrolase by Antisense RNA Inhibits Toxoplasma gondii Proliferation
1999; Elsevier BV; Volume: 274; Issue: 8 Linguagem: Inglês
10.1074/jbc.274.8.5083
ISSN1083-351X
AutoresValerian Nakaar, Benjamin U. Samuel, Emily O. Ngo, Keith A. Joiner,
Tópico(s)Cytomegalovirus and herpesvirus research
ResumoNucleoside triphosphate hydrolase (NTPase) is a very abundant protein secreted by the obligate intracellular parasiteToxoplasma gondii shortly after invasion of the host cell. When activated by dithiols, NTPase is one of the most potent apyrases known to date, but its physiological function remains unknown. The genes encoding NTPase have been cloned (Bermudes, D., Peck, K. R., Afifi-Afifi, M., Beckers, C. J. M., and Joiner, K. A. (1994) J. Biol. Chem. 269, 29252–29260). We have recently shown that the enzyme is tightly controlled within the vacuolar space and may influence parasite exit from the host cell (Silverman, J. A., Qi, H., Riehl, A., Beckers, C., Nakaar, V., and Joiner, K. A (1998) J. Biol. Chem. 273, 12352–12359). In the present study, we have generated an antisense NTP RNA construct in which the 3′-untranslated region is replaced by a hammerhead ribozyme. The constitutive synthesis of the chimeric antisense RNA-ribozyme construct in parasites that were stably transfected with this construct resulted in a dramatic reduction in the steady-state levels of NTPase. This inhibition was accompanied by a decrease in the capacity of the parasites to replicate. The reduction in parasite proliferation was due to a specific effect of antisense NTP RNA, since a drastic inhibition of hypoxanthine-xanthine-guanine phosphoribosyl transferase (HXGPRT) expression by a chimeric antisense HXGPRT RNA-ribozyme construct did not alter NTPase expression nor compromise parasite replication. These data implicate NTPase in an essential parasite function and suggest that NTPase may have more than one function in vivo. These results also establish that it is possible to study gene function in apicomplexan parasites using antisense RNA coupled to ribozymes. Nucleoside triphosphate hydrolase (NTPase) is a very abundant protein secreted by the obligate intracellular parasiteToxoplasma gondii shortly after invasion of the host cell. When activated by dithiols, NTPase is one of the most potent apyrases known to date, but its physiological function remains unknown. The genes encoding NTPase have been cloned (Bermudes, D., Peck, K. R., Afifi-Afifi, M., Beckers, C. J. M., and Joiner, K. A. (1994) J. Biol. Chem. 269, 29252–29260). We have recently shown that the enzyme is tightly controlled within the vacuolar space and may influence parasite exit from the host cell (Silverman, J. A., Qi, H., Riehl, A., Beckers, C., Nakaar, V., and Joiner, K. A (1998) J. Biol. Chem. 273, 12352–12359). In the present study, we have generated an antisense NTP RNA construct in which the 3′-untranslated region is replaced by a hammerhead ribozyme. The constitutive synthesis of the chimeric antisense RNA-ribozyme construct in parasites that were stably transfected with this construct resulted in a dramatic reduction in the steady-state levels of NTPase. This inhibition was accompanied by a decrease in the capacity of the parasites to replicate. The reduction in parasite proliferation was due to a specific effect of antisense NTP RNA, since a drastic inhibition of hypoxanthine-xanthine-guanine phosphoribosyl transferase (HXGPRT) expression by a chimeric antisense HXGPRT RNA-ribozyme construct did not alter NTPase expression nor compromise parasite replication. These data implicate NTPase in an essential parasite function and suggest that NTPase may have more than one function in vivo. These results also establish that it is possible to study gene function in apicomplexan parasites using antisense RNA coupled to ribozymes. Apicomplexan parasites (e.g. Toxoplasma,Plasmodium, Cryptosporidium, and Eimeria) are prevalent worldwide and cause significant diseases in humans and animals. In particular, the obligate intracellular parasiteToxoplasma gondii is an opportunistic infection associated with AIDS and congenital neurological birth defects (1Luft B.J. Remington J.S. Clin. Infect. Dis. 1992; 15: 211-222Crossref PubMed Scopus (1055) Google Scholar). Treatment regimens for this infection are not always effective, and this has mandated the search for novel drug targets. As reported previously, we and others have identified the gene encoding nucleoside triphosphate (NTPase), 1The abbreviations NTPasenucleoside triphosphate hydrolaseHXGPRThypoxanthine-xanthine-guanine phosphoribosyl transferaseGRA3dense granule antigen 3SAG1surface antigen 1HFFhuman foreskin fibroblastsAS-antisenseWTwild type 1The abbreviations NTPasenucleoside triphosphate hydrolaseHXGPRThypoxanthine-xanthine-guanine phosphoribosyl transferaseGRA3dense granule antigen 3SAG1surface antigen 1HFFhuman foreskin fibroblastsAS-antisenseWTwild typea very abundant protein secreted by T. gondii shortly after invasion of host cells (2Bermudes D. Peck K.R. Afifi-Afifi M. Beckers C.J.M. Joiner K.A. J. Biol. Chem. 1994; 269: 29252-29260Abstract Full Text PDF PubMed Google Scholar, 3Asai T. Miura S. Sibley D.L. Okabayashi H. Tsutomu T. J. Biol. Chem. 1995; 270: 11391-11397Crossref PubMed Scopus (169) Google Scholar, 4Asai T. O'Sullivan W.J. Tatibana M. J. Biol. Chem. 1983; 258: 6816-6822Abstract Full Text PDF PubMed Google Scholar, 5Carruthers V. Sibley D. Eur. J. Cell Biol. 1997; 73: 114-123PubMed Google Scholar). The 63-kDa NTPase isoforms are encoded by two genes,NTP1 and NTP3, which share more than 97% sequence identity (2Bermudes D. Peck K.R. Afifi-Afifi M. Beckers C.J.M. Joiner K.A. J. Biol. Chem. 1994; 269: 29252-29260Abstract Full Text PDF PubMed Google Scholar). When activated in vitro by dithiols, NTPase is one of the most potent apyrases known to date, capable of sequentially degrading ATP to ADP and AMP (4Asai T. O'Sullivan W.J. Tatibana M. J. Biol. Chem. 1983; 258: 6816-6822Abstract Full Text PDF PubMed Google Scholar). Because T. gondii is a purine auxotroph, it is thought that NTPase may participate in purine salvage (2Bermudes D. Peck K.R. Afifi-Afifi M. Beckers C.J.M. Joiner K.A. J. Biol. Chem. 1994; 269: 29252-29260Abstract Full Text PDF PubMed Google Scholar, 6Schwab J.C. Beckers C.J.M. Joiner K.A. Proc. Natl. Acad. Sci. U. S. A. 1994; 91: 509-513Crossref PubMed Scopus (322) Google Scholar, 7Sibley L.D. Niesman I.R. Asai T. Takeuchi T. Exp. Parasitol. 1994; 79: 301-311Crossref PubMed Scopus (84) Google Scholar). Recently we have shown that the enzyme may facilitate parasite exit from infected host cells (8Silverman J.A. Qi H. Riehl A. Beckers C. Nakaar V. Joiner K.A J. Biol. Chem. 1998; 273: 12352-12359Abstract Full Text Full Text PDF PubMed Scopus (75) Google Scholar). However, the biological function of NTPase in vivo is yet to be defined. nucleoside triphosphate hydrolase hypoxanthine-xanthine-guanine phosphoribosyl transferase dense granule antigen 3 surface antigen 1 human foreskin fibroblasts antisense wild type nucleoside triphosphate hydrolase hypoxanthine-xanthine-guanine phosphoribosyl transferase dense granule antigen 3 surface antigen 1 human foreskin fibroblasts antisense wild type It is possible to generate null mutants by gene knock out in the apicomplexan parasites (9Kim K.D. Soldati D. Boothroyd J.C. Science. 1993; 262: 911-914Crossref PubMed Scopus (231) Google Scholar, 10Donald R.G.K. Roos D. Mol. Biochem. Parasitol. 1998; 91: 295-305Crossref PubMed Scopus (117) Google Scholar, 11Menard R. Sultan A.A. Cortes C. Altszuler R. van Dijk M.R. Janse C.J. Waters A.P. Nussenzweig R.S. Nussenzweig V. Nature. 1997; 385: 336-340Crossref PubMed Scopus (254) Google Scholar). This strategy enables the study of gene function that can lead to the identification of potential drug targets. However, the presence in the genome of multiple copies of a gene (such as NTP) combined with low frequency of homologous recombination in T. gondii, hamper gene targeting efforts in these organisms. Moreover, it is not feasible to investigate the biological function of essential genes using this approach because of the lethality of the phenotype. To circumvent these difficulties, we sought to develop an antisense strategy to inhibit gene expression inT. gondii. Antisense RNA has been shown to regulate gene expression in bacteria (12Green P.J. Pines O. Inouye M. Annu. Rev. Biochem. 1986; 55: 569-597Crossref PubMed Scopus (315) Google Scholar), Dictyostelium (13Knecht D.A. Loomis W.F. Science. 1987; 236: 1081-1086Crossref PubMed Scopus (500) Google Scholar), Leishmania (14Zhang W.-W. Matlashewski G. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 8807-8811Crossref PubMed Scopus (131) Google Scholar),Drosophila (15Cabrera C.V. Alonso M.C. Johnston P. Phillips R.G. Lawrence P.A. Cell. 1987; 50: 659-663Abstract Full Text PDF PubMed Scopus (157) Google Scholar), Xenopus oocytes (16Melton D.A. Proc. Natl. Acad. Sci. U. S. A. 1985; 82: 144-148Crossref PubMed Scopus (263) Google Scholar), mammalian cells, (17Bevilacqua A. Erickson R.P. Hieber V. Proc. Natl. Acad. Sci. U. S. A. 1988; 85: 831-835Crossref PubMed Scopus (32) Google Scholar) and plants (18Ecker J.R. Davis R.W. Proc. Natl. Acad. Sc. U. S. A. 1986; 83: 5372-5376Crossref PubMed Scopus (181) Google Scholar, 19Rothstein S.J. DeMaio J. Strand M. Rice D. Proc. Natl. Acad. Sci. U. S. A. 1987; 84: 8439-8443Crossref PubMed Google Scholar, 20van der Krol A.R. Lenting P.E. Veenstra J. van der Meer I.M. Koes R.E. Gerats A.G.M. Mol J.N.M. Stuitje A.R. Nature. 1988; 333: 866-869Crossref Scopus (328) Google Scholar) but has not yet been demonstrated in the apicomplexan parasites. This strategy is at present still by trial and error. This may be partly ascribed to the finding that in eukaryotes, 3′-end formation of mRNA promotes the export of the mature transcript to the cytoplasm (21Eckner R. Ellmeier W. Birnstiel M. EMBO J. 1991; 10: 3513-3522Crossref PubMed Scopus (170) Google Scholar). Since antisense-induced mRNA degradation in eukaryotes occurs in the nucleus (22Cornelissen M. Nucleic Acids Res. 1989; 17: 7203-7209Crossref PubMed Scopus (30) Google Scholar), the use of a polyadenylated RNA may reduce the overall efficiency of antisense RNA. Eckner et al. (21Eckner R. Ellmeier W. Birnstiel M. EMBO J. 1991; 10: 3513-3522Crossref PubMed Scopus (170) Google Scholar) have shown that substituting the normal polyadenylation signal with cis-acting ribozymes leads to the nuclear retention of the product RNAs by generating export-deficient transcripts (21Eckner R. Ellmeier W. Birnstiel M. EMBO J. 1991; 10: 3513-3522Crossref PubMed Scopus (170) Google Scholar). Ribozymes are catalytic RNA molecules that are active in vitro, both in cell-free systems and in living cells (23Haseloff J. Gerlach W.L. Nature. 1988; 334: 585-591Crossref PubMed Scopus (957) Google Scholar, 24Gu J.-L. Veerapanane D. Rossi J. Natarajan R. Thomas L. Nadler J. Circ. Res. 1995; 77: 14-20Crossref PubMed Scopus (39) Google Scholar, 25Taylor N.R. Kaplan B.E. Swidersk P. Lin H. Rossi J.J. Nucleic Acids Res. 1992; 20: 4559-4564Crossref PubMed Scopus (115) Google Scholar, 26Montgomery R.L. Dietz H.C. Hum. Mol. Genet. 1997; 6: 519-525Crossref PubMed Scopus (59) Google Scholar) and are currently being developed for gene therapy applications as inhibitors of gene expression and viral replication (27Cech T.R. J. Am. Med. Assoc. 1988; 260: 3030-3034Crossref PubMed Scopus (92) Google Scholar, 28Sarver M. Cantin E. Chang P. Ladne P. Stephens D. Zaia J. Rossi J. Science. 1990; 247: 1222-1225Crossref PubMed Scopus (579) Google Scholar). In this report we show that the expression of NTPase can be substantially inhibited by stably introducing into the parasite genome an antisense gene in which its 3′ end is modified by a hammerhead ribozyme. The attenuation of NTPase activity inhibits parasite proliferation, suggesting a role for NTPase in parasite metabolism. The inhibition of parasite replication was due to specific reduction in NTPase by antisense RNA, since a drastic reduction in hypoxanthine-xanthine-guanine phosphoribosyl transferase (HXGPRT) expression did not alter NTPase expression nor compromise parasite replication. The chimeric antisense RNA-ribozyme strategy should be useful for the study of gene function in other protozoan parasites. RH-strain of T. gondii parasites were cultivated in Vero cells or in primary human foreskin fibroblasts (HFF), and the tachyzoites were transfected with plasmid constructs pαNTP-RZ, pASHXRZ, or p5′NTP-RZ (see Fig. 1) subcloned into a vector containing the dihydrofolate reductase-thymidylate-synthase (DHFR-TS) gene, which confers pyrimethamine resistance for selection (29Donald R.G.K. Roos D.S. Proc. Natl. Acad. Sci. 1993; 90: 11703-11707Crossref PubMed Scopus (287) Google Scholar). Stable transformants were selected in 1 μm pyrimethamine, and individual clones were isolated. Confluent cultures of HFF cells in 10-cm plates or 25-cm2T-flasks were inoculated with 106 freshly lysed-out parasite tachyzoites and incubated at 37 °C for 4 h, and the cultures were replaced with fresh medium so as to remove parasites that had not yet invaded. The doubling of intracellular parasites was followed by counting the number of tachyzoites per parasitophorous vacuole at the various times (at least 50 vacuoles were scored at each time point). The number of parasite divisions since infection was determined using log2 (parasite number) formula (30Fichera M.E. Bhopale M.K. Roos D.S. Antimicrob. Agent Chemother. 1995; 39: 1530-1537Crossref PubMed Scopus (164) Google Scholar). Statistical significance was assessed by measures of analysis of variance and two-tailed t test. NTPase enzymatic activity was done as described previously using equivalent amounts of parasite extracts (2Bermudes D. Peck K.R. Afifi-Afifi M. Beckers C.J.M. Joiner K.A. J. Biol. Chem. 1994; 269: 29252-29260Abstract Full Text PDF PubMed Google Scholar). Briefly, 1–10 μg of cell lysate was incubated with [2,8-3H]ATP (Amersham Pharmacia Biotech) in the presence of 100 mm HEPES-KOH, pH 7.2, 30 mmmagnesium acetate, 10 mm ATP, 0.1 mg/ml soybean trypsin inhibitor, and 1 mm dithiothreitol at 37 °C for 10 min. The reaction products were separated by TLC, spots corresponding to the nucleobase and 5′ products were excised, and the radioactivity was counted under liquid scintillation. HXGPRT activity and parasite replication were assayed by measuring the incorporation of [8-3H]xanthine (Moravek Biochemicals) and [5,6-3H]uracil (Amersham Pharmacia Biotech), respectively. 2–5 × 105 parasites/well were used to infect monolayers of HFF in a 24-well plate at 37 °C. Unattached parasites were removed 4 h later, and after 24 h post-infection, 1 μCi of radiolabel was added to each well. Incubation was continued for 2–6 h before the monolayers were fixed with trichloroacetic acid, rinsed, and counted as described (31Pfefferkorn E.R. Nothnagel R.F. Borotz S.E. Antimicrob. Agents Chemother. 1992; 36: 1091-1096Crossref PubMed Scopus (72) Google Scholar,32Donald R.G.K. Carter D. Ullman B. Roos D.S. J. Biol. Chem. 1996; 271: 14010-14019Abstract Full Text Full Text PDF PubMed Scopus (351) Google Scholar). Total RNAs were isolated from parasites with TRIzol reagent (Life Technologies, Inc.), and Northern blot analyses were performed with 10 μg of total RNA using gel-purified NTP cDNA probe (corresponding to the 1.98-kilobaseXhoI-EcoO109IA fragment of NTP1; see Fig. 1) labeled by random priming with [α32P]dCTP (Amersham Pharmacia Biotech). Hybridizations were done at 65 °C in a solution containing 5× saline/sodium phosphate/EDTA, 5× Denhardt and 0.5% SDS, and 100 μg/ml tRNA. The filters were washed at 65 °C with several changes of 1× saline, sodium phosphate, EDTA, 0.1% SDS followed by 0.1× saline, sodium phosphate, EDTA, 0.1% SDS and exposed for autoradiography. Between 106 and 107parasites were separated on SDS-polyacrylamide electrophoresis gels, and the material was transferred onto polyvinylidene difluoride membranes (Millipore). The filters were probed with anti-HXGPRT antiserum (generously given by Buddy Ullman, Portland, Oregon) or with anti-NTPase polyclonal antibody, GRA3, and SAG1 monoclonal antibodies as described previously (2Bermudes D. Peck K.R. Afifi-Afifi M. Beckers C.J.M. Joiner K.A. J. Biol. Chem. 1994; 269: 29252-29260Abstract Full Text PDF PubMed Google Scholar, 33Ossorio P.N. Dubremetz J.-F. Joiner K.A. J. Biol. Chem. 1994; 269: 15350-15357Abstract Full Text PDF PubMed Google Scholar). The primary antibodies were generally used at 1:1000 dilution. Secondary anti-mouse or anti-rabbit were horseradish peroxidase conjugates (Boehringer Mannheim) used at 1:2000 dilution followed by detection with ECL kit (Amersham Pharmacia Biotech). We targeted the gene encoding the abundant NTPase in T. gondii for inhibition by antisense RNA. Previous experiments by others suggest that a high ratio of transfected antisense RNA to the endogenous mRNA is required for effective reduction of expression of the target gene (34Kim S.K. Wold B.J. Cell. 1985; 42: 129-138Abstract Full Text PDF PubMed Scopus (208) Google Scholar, 35Izant J.G. Weintraub H. Science. 1985; 229: 345-352Crossref PubMed Scopus (223) Google Scholar). To increase the effectiveness of the antisense expressing vector (Fig. 1), we designed a construct that incorporated potentially enhancing features: a potent NTP promoter (36Nakaar V. Bermudes D. Peck K.R. Joiner K.A. Mol. Biochem. Parasitol. 1998; 92: 229-239Crossref PubMed Scopus (56) Google Scholar), the inclusion of an autocatalytic ribozyme structure and RNA structural elements that may act to stabilize the antisense RNA, as well as the possible enrichment for antisense RNA within the nuclear compartment (21Eckner R. Ellmeier W. Birnstiel M. EMBO J. 1991; 10: 3513-3522Crossref PubMed Scopus (170) Google Scholar, 37Liu Z. Batt D.B. Carmichael G.G. Proc. Natl. Acad. Sci. U. S. A. 1994; 91: 4258-4262Crossref PubMed Scopus (40) Google Scholar). Substituting the polyadenylation signal with a cis-acting ribozyme that cleaves in a site-specified manner leads to nuclear retention of the product RNAs by generating export-deficient transcripts (21Eckner R. Ellmeier W. Birnstiel M. EMBO J. 1991; 10: 3513-3522Crossref PubMed Scopus (170) Google Scholar, 37Liu Z. Batt D.B. Carmichael G.G. Proc. Natl. Acad. Sci. U. S. A. 1994; 91: 4258-4262Crossref PubMed Scopus (40) Google Scholar). We inserted a hammerhead ribozyme (23Haseloff J. Gerlach W.L. Nature. 1988; 334: 585-591Crossref PubMed Scopus (957) Google Scholar) immediately downstream of the antisense DNA, thus replacing the 3′-untranslated region of the gene. This operation potentially creates a pool of antisense RNA in the nucleus comparable with the endogenous mRNA. Since 3′-end formation promotes the export of mature polyadenylated RNA from the nuclear compartment where antisense effects most likely occur in eukaryotes, we did not include in these experiments a plasmid containing NTP antisense sequences with a polyadenylation signal (21Eckner R. Ellmeier W. Birnstiel M. EMBO J. 1991; 10: 3513-3522Crossref PubMed Scopus (170) Google Scholar, 22Cornelissen M. Nucleic Acids Res. 1989; 17: 7203-7209Crossref PubMed Scopus (30) Google Scholar, 34Kim S.K. Wold B.J. Cell. 1985; 42: 129-138Abstract Full Text PDF PubMed Scopus (208) Google Scholar, 35Izant J.G. Weintraub H. Science. 1985; 229: 345-352Crossref PubMed Scopus (223) Google Scholar, 37Liu Z. Batt D.B. Carmichael G.G. Proc. Natl. Acad. Sci. U. S. A. 1994; 91: 4258-4262Crossref PubMed Scopus (40) Google Scholar). Clones expressing antisense NTP RNA (and hereafter referred to as AS-NTP) displayed a dramatic reduction in NTPase enzymatic activity, ranging from ∼4 to >19-fold (Fig.2 A). The variation in the degree of inhibition as revealed by the enzymatic assay is considered a general feature of antisense inhibition (20van der Krol A.R. Lenting P.E. Veenstra J. van der Meer I.M. Koes R.E. Gerats A.G.M. Mol J.N.M. Stuitje A.R. Nature. 1988; 333: 866-869Crossref Scopus (328) Google Scholar, 38Smith C.J.S. Watson C.F. Ray J. Bird C.R. Morris P.C. Schuch W. Grierson D. Nature. 1988; 334: 724-726Crossref Scopus (499) Google Scholar). This variation may be related to the difference in the steady-state levels of antisense RNA generated due to chromosomal location of the gene or in copy number of the integrated gene (38Smith C.J.S. Watson C.F. Ray J. Bird C.R. Morris P.C. Schuch W. Grierson D. Nature. 1988; 334: 724-726Crossref Scopus (499) Google Scholar, 39Rodermel S.R. Abbott M.S. Bogorad L. Cell. 1988; 55: 673-681Abstract Full Text PDF PubMed Scopus (194) Google Scholar). To ascertain that reduction in enzyme activity was due to reduced synthesis of the protein, the amount of NTPase protein was assessed by immunoblot. NTPase levels were reduced between 86 to 88% when compared with WT in selected clones expressing antisense RNA and correlated with the decrease in enzymatic activity (Fig. 2 B). To control for the possible nonspecific effects of antisense RNA, we measured the levels of another dense granule protein, GRA3, and of the surface antigen, SAG1. The level of each of these proteins was comparable in all the samples analyzed. By immunofluorescence, NTPase signal was reduced in the transformants, whereas SAG1 expression was not affected (data not shown). The controls in these experiments indicated that the expression of antisense NTP RNA did not globally impair parasite metabolism. We determined the steady-state levels of NTP transcripts by Northern blot analysis (Fig. 2 C). NTP RNA levels were depressed at least by 2- to 6-fold in clones expressing antisense RNA. There appeared to be a good correlation between the steady-state levels of NTP RNA and the synthesis of the protein as assessed by Northern blot and immunoblot, respectively. Using a double-stranded DNA probe in Northern hybridization, we did not detect antisense RNA, although dot blot analysis revealed the presence of the gene in the genomic DNA of stably transfected clones (data not shown). One possible reason for this is that the antisense transcripts or their cleavage products are highly unstable and are degraded extremely rapidly in the parasite. However, the reduced steady-state levels of NTP mRNA coupled with the lack of detection of antisense RNA in the transfected clones is consistent with observations in other systems (38Smith C.J.S. Watson C.F. Ray J. Bird C.R. Morris P.C. Schuch W. Grierson D. Nature. 1988; 334: 724-726Crossref Scopus (499) Google Scholar, 39Rodermel S.R. Abbott M.S. Bogorad L. Cell. 1988; 55: 673-681Abstract Full Text PDF PubMed Scopus (194) Google Scholar, 40Kuipers A.G.J. Soppe W.J.J. Jacobsen E. Visser R.G.F. Mol. Gen. Genet. 1995; 246: 745-755Crossref PubMed Scopus (35) Google Scholar). This is generally assumed to be caused by duplex formation between antisense RNA and the target mRNA in the nucleus, thereby interfering with normal nuclear processing, cytoplasmic export, and translation of mRNA (34Kim S.K. Wold B.J. Cell. 1985; 42: 129-138Abstract Full Text PDF PubMed Scopus (208) Google Scholar, 35Izant J.G. Weintraub H. Science. 1985; 229: 345-352Crossref PubMed Scopus (223) Google Scholar). A direct inhibition of the production of NTPase through disruption of mRNA processing and transport seems plausible since ribozyme-processed RNA has been shown to be retained in the nuclear compartment (21Eckner R. Ellmeier W. Birnstiel M. EMBO J. 1991; 10: 3513-3522Crossref PubMed Scopus (170) Google Scholar, 37Liu Z. Batt D.B. Carmichael G.G. Proc. Natl. Acad. Sci. U. S. A. 1994; 91: 4258-4262Crossref PubMed Scopus (40) Google Scholar). These results indicate that antisense RNA can block mRNA and protein expression in T. gondii. Parasites expressing antisense RNA were impaired in intracellular replication but not invasion. At 8 h after infection, equal numbers of parasitophorous vacuoles were established by both control (108 ± 4.9) and antisense-expressing clones (114 ± 7.2, 111 ± 14.3; Table I), indicating that transgenic parasites that had a normal morphology invaded host cells normally. However, once inside the cell, the transgenic parasites divided more slowly than wild type, as evidenced by two different assays for parasite replication: intracellular doubling (Table I; Fig.3, A, B, andC) and nucleobase uptake (Fig. 3 D). For example, at 24 h after inoculation, the average doubling for the control was 2.89, which differed from 2.17 and 1.67 (p < 0.01), displayed by the antisense-expressing parasites (Table I). For the control parasites, this doubling corresponds to a replication time of about 8 h, similar to that previously reported (30Fichera M.E. Bhopale M.K. Roos D.S. Antimicrob. Agent Chemother. 1995; 39: 1530-1537Crossref PubMed Scopus (164) Google Scholar). The reduced replication of the antisense parasites is coincident with reduced [3H]uracil incorporated into parasite nucleic acid pools as shown by the lower values (37 and 48%) in transformants relative to wild type (Fig. 3 D). The lag in replication of the transformants in the first 24–32 h after infection may reflect the deficit in initial NTPase levels and represents the time needed to overcome this deficit by synthesis, packaging into dense granules, and exocytosis of a threshold amount of NTPase into the vacuolar space.Table IDistribution of parasitophorous vacuole sizes in antisense-expressing and control parasitesSampleTime after entry into vacuole% vacuoles with the following No. of parasite doublings (No. of parasites/vacuole)aMean ± S.E. from experiments involving the examination of at least 50 independent vacuoles at each time point in four different experiments is shown. At 8 h, the mean total vacuoles counted for each group were 108 ± 4.9, 114 ± 7.2, and 111 ± 14.3 for control, AS-NTP 12, and AS-NTP 14, respectively. There was no significant difference (P < 0.5) between control and the experimental groups. Antisense-expressing parasites continued to replicate but at a reduced rate. At the 24-h time point, 66% of parasitophorous vacuoles of the control parasites were at ≥8-cell stage, whereas for the antisense clones, only 25 and 36% were at ≥8-cell stage. Similarly, at 32 h, 48% of the parasitophorous vacuoles of the control were at ≥16-cell stage, and for the transformants, only 18 and 24% were at ≥16-cell stage.Average No. of doublings per vacuole (S.E.)P bNS, not significant; P was determined by analysis of variance against control.0 (1Luft B.J. Remington J.S. Clin. Infect. Dis. 1992; 15: 211-222Crossref PubMed Scopus (1055) Google Scholar)1 (2Bermudes D. Peck K.R. Afifi-Afifi M. Beckers C.J.M. Joiner K.A. J. Biol. Chem. 1994; 269: 29252-29260Abstract Full Text PDF PubMed Google Scholar)2 (4Asai T. O'Sullivan W.J. Tatibana M. J. Biol. Chem. 1983; 258: 6816-6822Abstract Full Text PDF PubMed Google Scholar)3 (8Silverman J.A. Qi H. Riehl A. Beckers C. Nakaar V. Joiner K.A J. Biol. Chem. 1998; 273: 12352-12359Abstract Full Text Full Text PDF PubMed Scopus (75) Google Scholar)4 (16Melton D.A. Proc. Natl. Acad. Sci. U. S. A. 1985; 82: 144-148Crossref PubMed Scopus (263) Google Scholar)5 (32Donald R.G.K. Carter D. Ullman B. Roos D.S. J. Biol. Chem. 1996; 271: 14010-14019Abstract Full Text Full Text PDF PubMed Scopus (351) Google Scholar)hControl89730.03 (.028)AS-NTP 1289730.03 (.023)NSAS-NTP 1481000.02 (.013)NSControl2443043232.89 (.24)AS-NTP 12245203924122.17 (.30)P < 0.01AS-NTP 1424141331961.67 (.35)P < 0.005Control32173538103.56 (.12)AS-NTP 123253833242.67 (.25)P < 0.05AS-NTP 143284925182.53 (.04)P< 0.01Stably transfected transgenic clones and control RH parasites were inoculated into fresh HFF cells. After 4 h, parasites that did not infect were washed off, and the number of parasites per vacuole was scored at the indicated times. The average number of vacuoles is expressed as % and plotted in Fig. 3. Parasite doublings were determined by log2(parasite number).a Mean ± S.E. from experiments involving the examination of at least 50 independent vacuoles at each time point in four different experiments is shown. At 8 h, the mean total vacuoles counted for each group were 108 ± 4.9, 114 ± 7.2, and 111 ± 14.3 for control, AS-NTP 12, and AS-NTP 14, respectively. There was no significant difference (P < 0.5) between control and the experimental groups. Antisense-expressing parasites continued to replicate but at a reduced rate. At the 24-h time point, 66% of parasitophorous vacuoles of the control parasites were at ≥8-cell stage, whereas for the antisense clones, only 25 and 36% were at ≥8-cell stage. Similarly, at 32 h, 48% of the parasitophorous vacuoles of the control were at ≥16-cell stage, and for the transformants, only 18 and 24% were at ≥16-cell stage.b NS, not significant; P was determined by analysis of variance against control. Open table in a new tab Stably transfected transgenic clones and control RH parasites were inoculated into fresh HFF cells. After 4 h, parasites that did not infect were washed off, and the number of parasites per vacuole was scored at the indicated times. The average number of vacuoles is expressed as % and plotted in Fig. 3. Parasite doublings were determined by log2(parasite number). To further control for nonspecific effects of antisense RNA, we employed two additional constructs; the histone-ribozyme vector, p5′NTPRZ (Fig. 1 A) and pAS-HXRZ, containing the HXGPRTgene in the antisense orientation (Fig. 1 B). Stable, pyrimethamine-resistant lines expressing these constructs showed no significant reduction in NTPase levels (Fig.4 A) nor in NTPase enzymatic activity (clones AS-HXRZ-26, 32, andRZ; Fig. 4 C). In contrast, parasites harboring pAS-HXRZ displayed a profound reduction in HXGPRT levels (Fig.4 A), which was accompanied by a significant reduction in HXGPRT activity (Fig. 4 B). These transgenic parasites did not reveal any morphological changes nor any growth retardation (Fig.4 D) and were comparable with the control HXGPRT knock out strain of RH (KO; Fig. 4, C and D). Similarly, parasite replication was not significantly reduced in parasites harboring the p5′NTPRZ construct (RZ; Fig.4 D). The reduction in HXGPRT activity with no adverse effects on parasite replication is consistent with the finding thatHXGPRTis not an essential gene (32Donald R.G.K. Carter D. Ullman B. Roos D.S. J. Biol. Chem. 1996; 271: 14010-14019Abstract Full Text Full Text PDF PubMed Scopus (351) Google Scholar). These results also suggest that promoter interference or squelching is not responsible for the inhibition of endogenous NTP expression. Although we have not directly tested these constructs (Fig. 1) in host cells, it is unlikely that they affect host cell growth or viability because they are driven by the developmentally regulated NTP promoter (32Donald R.G.K. Carter D. Ullman B. Roos D.S. J. Biol. Chem. 1996; 271: 14010-14019Abstract Full Text Full Text PDF PubMed Scopus (351) Google Scholar). Therefore, we conclude that the inhibition of parasite replication is a consequence of a deficit in NTPase production brought about by antisense NTP RNA. We propose that NTPase is functionally required for proliferation ofT. gondii. The molecular mechanism by which the reduction in NTPase levels leads to inhibition of parasite replication is unknown. Nevertheless, this study implicates NTPase for the first time in an essential parasite function. Data from other studies support the notion that NTPase may have a regulatory function in T. gondiigrowth and replication. First, NTPase activity is tightly regulated invivo, as unfettered activation of the enzyme leads to rapid exit of parasites from the host cell (8Silverman J.A. Qi H. Riehl A. Beckers C. Nakaar V. Joiner K.A J. Biol. Chem. 1998; 273: 12352-12359Abstract Full Text Full Text PDF PubMed Scopus (75) Google Scholar). Second, it has been shown that quercetin (P-type ATPase inhibitor) effectively prevents growth ofT. gondii in culture at concentrations that inhibit NTPase by >90% in vitro (41.Asai, T. & Sibley, L. D. (1996) Fourth International Biennial Toxoplasma Conference, July 22–26, 1996 (abstract), Drymen, Scotland.Google Scholar). Third, the secretion of NTPase into the parasitophorous vacuoles occurs shortly before replication ofToxoplasma sporozoites (infective forms within oocysts generated after the sexual cycle in the definitive feline host) (42Tilley M. Fichera M.E. Jerome M.E. Roos D.S. White M.W. Infect. Immun. 1997; 65: 4598-4605Crossref PubMed Google Scholar). Finally, NTP promoter activity is developmentally down-regulated in bradyzoites, the slowly dividing form of T. gondii that persists in the host as a chronic infection (36Nakaar V. Bermudes D. Peck K.R. Joiner K.A. Mol. Biochem. Parasitol. 1998; 92: 229-239Crossref PubMed Scopus (56) Google Scholar). Of interest, in gene knock out experiments we have never recovered any clones with a disrupted NTP1 or NTP3 gene locus, although we have employed targeting vectors with large contiguous genomic sequences. This suggests indirectly that NTPase is an essential gene and may thus provide a potential target for chemotherapy in T. gondii. The experiments were undertaken in this study to demonstrate the biological role of NTPase, one of the most potent ATP degrading enzymes, in the metabolism of the parasite. To this end, we employed antisense RNA coupled to a hammerhead ribozyme to reveal the role of NTPase in parasite proliferation. Although it remains to be formally proven, a complete abrogation of NTPase expression may be incompatible with parasite growth and survival. Since NTPase can also influence parasite exit from the host cell (8Silverman J.A. Qi H. Riehl A. Beckers C. Nakaar V. Joiner K.A J. Biol. Chem. 1998; 273: 12352-12359Abstract Full Text Full Text PDF PubMed Scopus (75) Google Scholar), these findings suggest that the physiological role of NTPase may be multifunctional. In this study, we have also shown that even a very abundant protein such as NTPase or HXGPRT is titratable by antisense RNA. Because it is effective and simple, antisense RNA strategy should find widespread application among apicomplexan parasites for the rapid analysis of gene function. For example, genes that are refractory to targeted gene disruption or essential genes that generate a lethal phenotype when they are disrupted, are especially amenable to this strategy. We thank G. Carmichael, Storrs, CT for generously donating pBS-RZ and Buddy Ullman, Oregon Health Sciences University for HXGPRT antibody. We also thank Heinreich Hoppe, Huân Ngô, Christian Tschudi, and Elisabetta Ullu for critically reading the manuscript.
Referência(s)