- Email: [email protected]
Molecular cloning of the Toxoplasma gondii sag4 gene encoding an 18 kDa bradyzoite specific surface protein
MOLECULAR BI-CAL PARAslToLoGy
ELSEVIER
Molecular
and Biochemical
Parasitology
82 (1996) 237-244
Molecular cloning of the Toxoplasma gondii sag4 gene encoding an 18 kDa bradyzoite specific surface protein’ Carmen odberg-FerraguP**‘, Martine Soete”,‘, Anne Engels”*‘, Bart Samynb, Anne Loyens”, Jozef Van Beeumenb, Daniel Camus”,‘, Jean-Fraqois Dubremetz” ‘Fuulteif
“INSERM 1142. 369, rue Jules Gursde, BP. 39, 59651 Villmeuvr d’Ascq cPdes. France Ct’ermsc~huppm.lhiversiteit Gent. Department of’ Biochemistry, Ph_vsiolog~~ and Miuobiology. K. L. Lcdeg~mckstruur 35. 9000 Gent, Belgium ‘FucultP de Medecine. Place dc Verdun. 59850 Lille. France
Received
7 May
1996: revised
5 August
1996: accepted
I5 August
1996
Abstract An 18 kDa bradyzoite specific surface protein of Toxopplasmu gondii (T. gondii) has been purified by affinity chromatography with a specific monoclonal antibody using parasites grown in vitro under conditions inducing the biosynthesis of bradyzoite specific proteins. N-terminal and internal amino acid sequences obtained by microsequencing enabled us to design degenerate oligonucleotides. A fragment of 187 bp was amplified by polymerase chain reaction (PCR). It was used as a probe to clone a 4 kb-BanrHI fragment encompassing the gene encoding the 18 kDa protein. Nucleotide sequence analysis revealed a single open reading frame of 516 nucleotides encoding a 172 amino acid protein. The deduced amino acid sequence matched perfectly the peptides microsequenced from the native protein. The N-terminal hydrophobic region was found to possess the characteristics of a signal peptide of 27 amino acids. The hydrophobic C-terminal part could represent a signal for a glycan-phosphoinositide anchor. The full-length cDNA was also isolated and both the 5’ and 3’ untranslated regions were determined. Reverse transcriptase-PCR (RT-PCR) using pl8-specific primers showed a stage-specific expression of this gene. Comparison of the nucleic acid sequence and the predicted amino acid sequence with databases did not reveal significant homology with known genes or proteins. This gene is proposed to be named sug4, according to the existing T. gondii nomenclature. Keywords:
Toxoplusrrzu gondii;
Bradyzoite
surface
antigen;
Gene cloning
AbhretWions: GPI, Glycan-phosphoinositide: PCR, polymerase chain reaction: ORF. open reading frame; mab. monoclonal antibody. * Corresponding author. Tel.: + 33 20 472397; fax: + 33 20 059172; e-mail [email protected] ’ Note: Nucleotide sequence data reported in this paper are available in the EMBL databases under the accession number 269373. z These two authors contributed equally to the work. 0166-6851/96]$15.00
CD 1996 Elsevier
PIf SO166-6851(96)02740-5
Science
B.V. All rights
reserved
238
C. odberg-Ferragut
et al. / Molecular and Biochemical Parasitology 82 (1996) 237-244
1. Introduction Toxoplasma gondii (T. gondii) is the intracellular protozoan parasite causing toxoplasmosis. This disease is responsible for transplacental infections resulting eventually in malformations or the death of congenitally infected infants [1,2]. More recently, T. gondii has been shown to be one of the major opportunistic pathogens in immunocompromised patients [3,4]. Indeed, this pathogen persists in immunocompetent hosts in cyst form within brain and muscle tissues. In a state of immune deficiency these quiescent cysts may rupture and cause a disseminated and potentially fatal disease [5,6]. A better understanding of conversion between the acute (tachyzoite) and dormant (bradyzoite) stages would therefore be a major step in controlling the disease. As this conversion is mainly a differentiation process characterized by the differential expression of stage specific molecules, the identification of these molecules is a prerequisite to study this process. Then the cloning of the stage specifically expressed genes would allow the study of their regulation, and therefore of the differentiation mechanisms. Because the tachyzoite stage is the most readily available for study, many proteins and the corresponding genes have been characterized from tachyzoites. Most of these have been shown to be common to tachyzoites and bradyzoites. Only a few have proven to be specific to the tachyzoite stage, and these are glycan-phosphoinositide (GPI) anchored surface molecules [7,8]. More recently, because of the rising interest in stage conversion, several groups have identified bradyzoite or cyst specific proteins [9-121 and the genes coding for three of them have been cloned and sequenced [13- 161. These genes code for cyst wall or bradyzoite cytoplasmic proteins. Because of the importance of surface molecules in the parasite’s interaction with its host cell and with the immune response, we have been interested in the characterization of stage specific surface proteins, as they may also be directly related to stage specific biological properties of the invasive stages. We had previously identified four bradyzoite specific proteins located at the parasite
plasmalemma [lo]. One of them, P18, a protein of 18 kDa, has been further analyzed and shown to possess characteristics of a GPI anchored surface protein (unpublished observations). We have purified this protein and obtained amino acid sequence data that have been used to clone the corresponding gene. We report here the structure of this gene and of a full length cDNA. We propose to name this gene sag4, and the gene product SAG4, according to the nomenclature in use for T. gondii genes [17].
2. Materials and methods 2.1. Cells and parasites
The parasites used in the study were of strain RH [18], of strain PLK [19] and of strain 76 K [20]. They were grown either in the peritoneal cavity of Swiss mice, or in cell cultures, on Vero cells. Induction of bradyzoite development in cell cultures was obtained by raising the pH of the culture medium to 8 [21]. 2.2. Microsequencing of SAG4 SAG4 was purified by immunoaffinity on monoclonal antibody (mab) T83Bl [lo] coupled to Sepharose 4B. A T. gondii PLK culture that had been subjected to high pH for 2 days was lysed in phosphate buffered saline (PBS) containing 10 pgg/ml leupeptin, 100 pg/ml aprotinin, 1 mM PMSF, 2 mM ehthelenediaminetetraacetate (EDTA) and 1% Nonidet P40. The purified protein was separated by sodium dodecyl sulfatepolyacrylamide gel electrophoresis (SDS-PAGE) and electroblotted on polyvinylidene fluoride membrane (Applied Biosystems). The membrane was stained with Ponceau and the 18 kDa band was excised for microsequencing. In a first experiment, the N-terminal sequence of the 18 kDa protein was obtained using the 476A protein sequencer (Applied Biosystems) with on-line PTH analysis. A second round of purification was performed. A fraction of the blot was used for amino acid analysis. The remaining fraction was submitted to an endoproteinase Lys-C digest. The pep-
C. tjdberg-Ferragut
et al. / Molecular and Biochemical Parasitology 82 (1996) 237-244
239
B
_
c
c-c__
_
_
o 1kb
Fig. I, Schematic representation of genomic locus encompassing the bradyzoite sag4 gene. (A) Relevant restriction enzyme cleavage sites. (B) Cloned genomic DNA fragment of about 4 kb. The shaded area represents the coding region and the solid bar indicates the hybridization site of the probe. Sequenced fragments are enclosed between two vertical lines and the primers used shown by the arrows.
tides were separated on a C8 RPLC (SMART System, Pharmacia, Uppsala, and sequenced as described above.
column Sweden)
spectively) were cut with Hind111 or PstI enzymes. T. gondii PLK DNA was also digested by either EcoRI, PstI or BamHI enzymes alone and by EcoRI and each of the other enzymes. DNAs
2.3. Polymerase chain reaction (PCR) Sense and antisense degenerate oligonucleotides based on the amino acid sequences of PeptIlPept-3 (Fig. 2) were synthetized by Eurogentec, Belgium. These were designated as oligola: TGGACITAYGAYTTYAARAARGC (nucleotides (nt) 88 - 110) and oligo-lb: GARATHATHACICCIGG (nt 136 152) for the sense sequences and oligo-2*: GGIGGDATRTAYTCIARIGG (nt 187-206) and oligo-3*: GCIGCIARYTCNACNGGYTCRTC (nt 253-275) for the anti-sense sequences (Fig. 2). PCR amplifications were performed on genomic DNA using Goldstar polymerase (Eurogentec, Belgium). Conditions were: 95°C for 5 min, 95°C for 1 min, 50°C for 1 min, 72°C for 1 min (5 cycles), 95°C for 1 min, 45°C for 1 min, 72°C for 1 min (5 cycles), 95°C for 1 min, 40°C for 1 min, 72°C for 1 min (25 cycles), 72°C for 5 min. PCR products were analyzed by agarose gel electrophoresis. The largest amplified product, a 187 bp fragment, was cloned into PCR-Script vector (Stratagene) and sequenced. 2.4. Southern blot and hybridization Twenty micrograms mouse, Vero cells and
of genomic
DNA
(from re-
T. gondii PLK strain
Fig. 2. Sequence of the gene encoding the SAG4 bradyzoite antigen. The nt sequence contains an ORF encoding a polypeptide of 172 amino acids. Boxed regions reflect the homologous amino acid sequences of Pept-1 (28852), Pept-2 (46-76) and Pept-3 (85-95) obtained from the microsequencing of the purified SAG4. The PCR primers are highlighted and italicised; they are designated oligo-I-oligo-7: asteriks indicate that reverse primers have been chosen. The double arrow shows the putative cleavage of the signal sequence. The unique EcoRI restriction site is underlined. Amino acid residues are numbered on the right, and the nt on the left of these sequences.
240
C. odberg-Ferragut
et al. / Molecular and Biochemical Parasitology 82 (1996) 237-244
were electrophoretically separated and transferred to nylon membranes (Immobilon-P, Millipore) by capillary blotting in alkaline conditions. Prehybridizations and hybridizations were carried out in 50% formamide, 5 x Denhart’s solution, 5 x SSPE, 0.2% SDS, 0.5% dextran sulfate and 200 pg/ml of sheared salmon sperm DNA. The 187 bp fragment was used as a probe and labelled by random priming (Boehringer kit) using [u-~*P] dATP. Radiolabeled probe was incubated with filter overnight at 42°C and unbound probe removed by washing filters three times in 2 x SSC at 42°C. Autoradiographic exposure was for a week at - 80°C.
2.5. Cloning and screening Fifty micrograms of genomic 7’. gondii DNA were digested with BamHI and electrophoresed in a 1% agarose gel. The 3.5-4.5 kb fraction was ligated to predigested ZAP Express BamHI/CIAP vector (Stratagene), packaged in vitro into /iphage particles with the Gigapack II Gold extract (Stratagene) and titered on Escherichia coli XLlBlue MRF’ strain. This size-selected DNA library was screened by plaque hybridization using the 187 bp fragment amplified as digoxigenin-labeled probe (Boehringer kit). Detection was performed by enzyme immunoassay with luminescence. Positive clones were excised in vivo to obtain directly the recombinant pBK-CMV phagemid.
2.7. Analysis of the 5’ and 3’ ends Total RNA was isolated by the procedure described by Chomczynski and Sachi [22] and modified by Clontech (total RNA separator kit). One micrgram of total T. gondii RNA was reverse transcribed using the Marathon system of Clontech. PCR amplifications of the 5’ and 3’ ends of the sag4 transcript were performed using universal adaptor primers and specific oligonucleotides of the sag4 coding sequence. The 5’ end was PCR oligo-4*: amplified the anti-sense using GCAGTGGTCTCCCTTTTCAAC (nt 229-249, Fig. 2). Similarly, the 3’ end was PCR amplified using the sense oligo-5: GGAGATTATCACACCGGGCG (nt 135-154, Fig. 2). Nested PCRs were performed using the anti-sense oligo-6*: CTCGCCAAGCCGAGCACCC (nt 55-73, Fig. 2) and the sense oligo-7: GAGAAGGAAACCATACCC (nt 361-378, Fig. 2). The largest PCR products generated were cloned into pGEMT plasmid (Promega), transfected into E. coli JM109 and partially sequenced. 2.8. Reverse transcriptase-PCR
(RT-PCR)
Total RNA was reverse transcribed by using Superscript II enzyme (BRL) and specific 3’ primers for either sag4 (oligo-5, Fig. 2) or a T. gondii microneme protein mic 1 gene [23]. Tfl polymerase (Epicentre) was used in PCR experiments with additional 5’ primers specific for both genes (oligo-4* for sag4, Fig. 2). Annealing steps were performed at 50°C.
2.6. Sequencing The 187 bp PCR insert was entirely sequenced using primers flanking the cloning sites. The 4 kb BarnHI fragment was partially sequenced using gene-specific primers. Plasmid and phagemid were purified on Qiagen columns and sequenced with the Sequenase kit (Amersham) by the double strand DNA sequencing method, using [u-~~P] dCTP. Sequence data were analyzed using the PC gene program. Homology searches of protein (Swissprot and PIR) and nucleic acid (GenBank and EMBL) databases were performed using Blitz and FASTA programs.
3. Results 3. I. Native SAG4 sequence
Direct N-terminal sequencing of SAG4 purified from in vitro grown bradyzoites resulted in the sequence designated as Pept- 1: KSWTYDFKKALDDDETKKEIITPGD (Fig. 2). Further, analysis of amino acid composition revealed that this protein was Lys-rich. The sequencing results of the Lys-C peptides obtained after in situ cleavage [24] confirmed the N-terminus and defined
C. odberg-Ferragut
et al. / Molecular and Biochemical Parasitology 82 (1996) 237~m244
two new sequences designated as Pept-2: EIITPGDSVSIENSGSRPLEYIPPNPSQVLK (Fig. 2) and Pept-3: DEPVELAALFK (Fig. 2). 3.2. Characterization
of sag4
PCR experiments were carried out on PLK, RH and 76 K T. gondii genomic DNA. Using was amoligos la and 3*, a 187 bp fragment plified. Nested PCR using oligos lb and 2* validated the first amplification product as being a fragment of the sug4 gene. No amplification was found with either mouse DNA or Vero cell DNA. Also, only T. gorzdii DNA hybridized to this 187 bp fragment. It was sequenced and translated. The deduced amino-acid sequence showed complete sequence identity with those found in Pept1 -Pept-3 (Fig. 2) and the nt sequence revealed an EcoRI site in its central area. Thus, T. gondii genomic DNA was digested with several single restriction enzymes or in combination with EcoRI and probed using the 187 bp amplified fragment. The resulting restriction map is shown in Fig. 1. The entire sug4 was cloned in ZAP Express vector from a BcrnzHI size selected library. About 3.6 x lo3 recombinant clones were screened, three clones were found positive and further characterized by PCR, one of them was sequenced. The genomic sequence encoding SAG4 from the PLK T. gondii strain consists of an open reading frame (ORF) of 5 16 bp long. The ATG codon marking the beginning of the ORF had the favourable environment for initiation of translation according to the Kozak criteria [25]. Also, an in-frame termination codon preceded this start codon. The coding region had a codon usage similar to that of other T. gondii coding regions, [26] excepted for the Ala codon. In the sag4, T was the major preference at the third position for this codon. Comparison of the sag4 and several T. gondii genes in the region immediately upstream from the ATG codon revealed putative consensus translation sequences (Fig. 3) [7,27-291. The strongest similarity was found with B-tub genes. On the contrary, the T-streches found in many 5’ untranslated sequences of T. gondii genes [29] are not obvious in sag4.
m-tub
~~~-----______~~__~~
E-tub
TTTTTC-TG----CAm--TG
sag4
-~TTTCGTC-~~~TGTCTTCAACCATG
gra5
-TTTTC%TG---PAGT-MC-AaAATG
RraC
TTmTTGGGAGmCGi-G-TG
241
Fig. 3. Comparison of 5 T. go:o12diisequences immediately upstream from the ATG. Bold letters indicate identity with the canonical sequence described by Kozak [25].Underlined bases show similarity of sequences upstream from the ATG between sag4 and various T. go&ii genes.
Three different 5’ ends were mapped by sequencing the subcloned 5’ Marathon products. They matched perfectly with the genomic sequence and ranged at 257, 229 and 198 bp respectively upstream from the ATG initiation codon. These end-points should be confirmed by Northern blot experiments. There were no typical eukaryotic TATA nor CCAAT consensus promoter sequences. Nevertheless, the hexameric sequence ACGCCG was found three times in this part 01 the gene but was not present in other sequences specifically expressed at the bradyzoite stage (1316). The polyadenylation site was at 709 bp downstream of the stop codon. This site was identified by analysis of the amplification product primed with a modified oligo dT. The higher eucaryotic polyadenylation signal AAUAAA normally found about 30 nt upstream of the poly A tail was not present. PCR amplification of both genomic and cDNA showed no significant size difference, suggesting that no intron was present in sag4.
3.3. Regulation
of sag4 expression
RT-PCR was used to study gene expression at bradyzoite and tachyzoite stages. The micl gene was used as a control for expression in both stages. Using sag4-specific primers a transcript of the expected size was amplified only in bradyzoites while the rnicl gene was expressed in both stages (Fig. 4). Parallel amplification of genomic DNA with micl-specific primers resulted in a band of larger size, due to the presence of one intron in this part of the gene (unpublished results). This band was not found in RNA amplification and demonstrated that no genomic
C. tidberg-Ferragut
242
contamination tion.
et al. / Molecular and Biochemical Parasitology 82 (1996) 237-244
was present in the RNA prepara-
3.4. Analysis of the deduced protein sequence
The gene translation product was composed of 172 amino acids and had a theoretical molecular mass of 18.5 kDa. This deduced protein showed two hydrophobic domains at both the amino and the carboxy terminus. Both regions framed a polypeptide rich in charged amino acids (14% polar negative and 15% polar positive) (Fig. 2). The sequence in the N-terminal part of the polypeptide was typical for an eukaryotic signal
1 23
45
peptide showing three positively charged aminoterminal residues and a central hydrophobic area. Two putative cleavage sites were inferred by PC gene analysis. One was between Ala 22 and Trp 23; the other was located between Ala 27 and Lys 28. This last prediction coincided with the first amino acid obtained by N- terminal sequencing of the protein and therefore assigned a secretory signal sequence of 27 amino acids. The hydrophobic signal sequence was followed by a strongly hydrophilic region (Fig. 2). This region is rich in lysine and contains only two cysteines. The C-terminal hydrophobic region consists of 21 residues. The protein exhibits the N-glycosylation amino acid consensus sequence Asn-X-Ser/Thr at residues 70 and 137 respectively. The calculated isoelectric point of the protein is 8.97.
4. Discussion 622 404 242+238 123
<
Fig. 4. RT-PCR with gene-specific primers. Total RNA was isolated either from tachyzoites (T) grown intraperitoneally in mice or from bradyzoites (B) grown in vitro under switching conditions. Tachyzoite genomic DNA was prepared from in vitro Vero cell cultures. Lane 1 represents the amplification obtained from genomic DNA with mic 1-specific primers; lanes 2 and 4 show the RT-PCR results using total RNA as template and gene-specific primer pairs for micl; lanes 3 and 5 show the RT-PCR results utilizing total RNA as template and gene-specific primer pairs for sag4. The apparent molecular mass of DNA standards is indicated on the left. The arrowhead indicates the sug4-specific signal detected in bradyzoites.
The present study reports the cloning of the gene encoding an 18 kDa protein which had been previously identified by mab as a bradyzoite surface molecule ([lo] and unpublished observations). According to the nomenclature proposed by Sibley et al. [17], the gene encoding P18 should therefore be named sag4 as surface antigen gene 4 and P18 named SAG4. Several convincing evidences suggest that this clone represents the bradyzoite sag4 gene. First, there is perfect matching of the sequences obtained from the microsequencing analysis on the native protein compared to the sequences derived from the cloned gene. Second, the size of the coding sequence is as expected and this latter contains a signal peptide. In addition, a C-terminal sequence likely to be removed and replaced by a GPI anchor has been found. Further evidence for this C-terminal removal is given by the fact that the mature protein is not metabolically labeled with [35S] methionine, indicating that this amino acid position 172 is removed during processing. Third, the RT-PCR experiment using sag4-specific primers shows a stage-specific expression of this gene. This type of regulation has also been found for bag 1 [ 141.
C. odberg-Ferragut
et al. 1 Molecular and Biochemical Parasitology 82 (1996) 237-244
Comparison of the predicted sequence with databases does not reveal significant homology with any known protein. Results of our Southern blot analysis (data not shown) are consistent with the sag4 gene being a single copy gene. It is noteworthy that the putative promoter region of the sag4 gene showed perfect identity between strains PLK and RH. As these two strains show a significant difference in susceptibility to bradyzoite differentiation triggers, i.e. in the initiation of SAG4 expression, this result suggests that this difference is likely to be due to factors which are independent of the sag4 promoter. This hypothesis is further supported by the fact that expression of P36 and P34 bradyzoite antigens follows similar kinetics as SAG4 antigen [21]. Nevertheless, it will be necessary to compare the sequence of the 3’ untranslated region of the sag4 gene from both strains and to study their influence on the .expression level. The influence of this region has been demonstrated for other protozoan genes [30]. We cannot draw any conclusion concerning a bradyzoite specific promoter. The consensus sequence TGCTGTGTC detected in two bradyzoites genes [13,14] was not found in the sequenced part of sag4. Other genes specifically expressed at the bradyzoite stage encode proteins found in the cyst wall [13] or in the parasite cytoplasm [l&16]. The surface localization of SAG4 makes the cloning of this gene a unique tool to study the functional aspect of differential surface protein expression. As bradyzoites and tachyzoites develop in different locations, their interaction with the environment is primarily dependent on their surface properties. Differential characteristics such as digestive enzyme susceptibility, or enterocyte invasion, could be related to differences in surface protein expression. A powerful way of studying these functions will be the transfection of these genes under the control of heterologous promoters in order to modify the stage specificity of expression and analyse the effects on the biology of the parasite. These experiments are presently under way.
243
Acknowledgements
The authors are indebted to M.N. Fourmaux for excellent advice in using digoxigenin-labelled probes and to J.P. Touzel for his kindness and help in using database programs. This work was funded by the Agence National de Recherche sur le SIDA (ANRS), by EC (program BMHl-CT921535) and by a Concerted Research Action of the Flemish Government (contract 12052293). Martine Soete was supported by ANRS.
References McCabe, R.E. and Remington, J.S. (1983) The diagnosis and treatment of toxoplasmosis Eur. J. Clin. Microbial. 2. 15. VI Remington, J.S. and Desmonts, G. (1990)Toxoplasmosis. In: Infectious Diseases of the Fetus and Newborn Infant (Remington, J.S. and Klein, J.O., eds.). pp. 899195, W.B. Saunders, Philadelphia. 131 Luft, B.J., Brooks, R.G., Conley, F.K.. McCabe, R.E. and Remington, J.S. (1984) Toxoplasmic encephalitis in patients with acquired immune response deficiency syndrome. JAMA 252, 913-917. [41 Wong, J., Gold, W.M., Brown, A.E., Lange, M., Fried, R.. Grieco, M., Mildvan, D., Giron, J., Tapper, M.L., Lerner, C.W. and Armstrong, D. (1984) Central-nervoussystem toxoplasmosis in homosexual men and parenteral drug abusers. Ann. Intern. Med. 100. 36642. J.S. (1989) VI Suzuki. Y., Conley. F.K. and Remington. Differences in virulence and development of encephalitis during chronic infection vary with the strain of Toroplasma gondii. J. Infect. Dis. 159. 790-794. WI Gazinelli, R.T., Eltoum, I., Wynn. T.A. and Sher, A. (1993) Acute cerebral toxoplasmosis is induced by in vivo neutralization of TNF-alpha and correlates with the down-regulated expression of inducible nitric oxide synthase and other markers of macrophage activation. J. Immunol. I5 1, 3672-368 1. [71 Burg. J.L., Perelman, D., Kasper, L.H.. Ware, P.L. and Boothroyd. J.C. (1988) Molecular analysis of the gene encoding the major surface antigen of To.wplasma gondii. J. Immunol. 141, 358443591. J., Parmley. S.F., PI Prince. J.B.. Auer. K.L.. Huskinson. Ardujo, F.G. and Remington. J.S. (1990) Cloning, expression, and cDNA sequence of surface antigen P22 from Toxoplasma gondii. Mol. Biochem. Parasitol. 43. 97- 106. [91 Omata. Y., Igarashi, M., Ramos, M.I. and Nakabayashi, T. (1989) Toxoplasma gondii: antigenic differences between endozoites and cystozoites defined by monoclonal antibodies. Parasitol. Res. 75. 189- 193.
[‘I
244
C. iidberg-Ferragut
et al.
I Molecular and Biochemical Parasitology 82 (1996) 237-244
[lo] Tomavo, S., Fortier, B., Soete, M., Ansel, C., Camus, D. and Dubremetz, J.F. (1991) Characterization of bradyzoite-specific antigens of Toxoplasma gondii major surface antigens. Infect. Immun. 59, 3750-3753. [Ill Weiss, L.M., LaPlace, D.. Tanowitz, H.B. and Wittner, M. (1992) Identification of Toxoplasma gondii bradyzoitespecific monoclonal antibodies. J. Infect. Dis. 166, 213215. [12] Bohne, W., Heesemann, J. and Gross, U. (1993) Induction of bradyzoite-specific Toxoplasma gondii antigens in gamma interferon-treated mouse macrophages. Infect, Immun. 61. 1141-1145. [I31 Parmley, S.F., Yang, SM., Harth, G., Sibley. L.D., Sucharczuk. A. and Remington, J.S. (1994) Molecular characterization of a 65 kilodalton Toxoplasma gondii antigen expressed abundantly in the matrix of tissue cysts. Mol. Biochem. Parasitol. 66. 2833296. [14] Bohne, W., Gross. U., Ferguson, D.J.P. and Heesemann, J. (1995) Cloning and characterization of a bradyzoitespecifically expressed gene (hsp /bag 1) of Toxoplasma gondii, related to genes encoding small heat-shock proteins of plants. Mol. Microbial. 16, 1221-1230. [15] Parmley, S.F.. Weiss, L.M. and Yang, S. (1995) Cloning of a bradyzoite-specific gene of Toxoplasma gondii encoding a cytoplasmic antigen. Mol. Biochem. Parasitol. 73, 2533257. [16] Yang, S. and Parmley, S.F. (1995) A bradyzoite stagespecifically expressed gene of Toxoplasma gondii encodes a polypeptide homologous to lactate dehydrogenase. Mol. Biochem. Parasitol. 73, 291-294. [17] Sibley, L.D., Pfefferkorn, E.R. and Boothroyd, J.C. (1991) Proposal for a uniform genetic nomenclature in Toxoplasma gondii. Parasitol. Today 7, 327-328. [18] Sabin. A. (1941) Toxoplasmic encephalitis in children. JAMA 116. 801-814. [19] Kasper, L.H. and Ware, P.L. (1985) Recognition and characterization of stage-specific oocyst sporozoite antigens of T. gondii by human anti-serum. J. Clin. Invest. 75, 1570-1577. [20] Laugier, M.N. and Quilici, M. (1970) Inter&t experimental d’une souche de toxoplasme peu pathogene pour la souris. Ann. Parasitol. Hum. Comp. 45, 3899403.
[21] Soete, M.. Camus, D. and Dubremetz, J.F.D. (1994) Experimental induction of bradyzoite-specific antigen expression and cyst formation by the RH strain of TO.XOplasma gondii in vitro. Exp. Parasitol. 78, 361-370. [22] Chomczynski, P. and Sacchi, N. (1987) Single-step method of RNA isolation by acid guanidium thiocyanatephenol-choroform extraction. Anal. Biochem. 162, 156159. [23] Mercereau-Puijalon, 0.. Fourmaux. M.N. and Dubremetz, J.F.D. (1993) Expression of apical organelles antigens by a Tosoplasma gondii genomic library. In: Toxoplasmosis (Smith, J.E., ed.), pp. 19931, SpringerVerlag. Berlin. [24] Fernandez, J., Andrews, L. and Mische, S.M. (1994) An improved procedure for enzymatic digestion of polyvinylidene difluoride-bound proteins for internal sequence analysis. Anal. Biochem. 218, 112-117. [25] Kozak, M. (1986) Point mutations define a sequence flanking the AUG initiator codon that modulates translation by eukaryotic ribosomes. Cell, 44, 283-292. [26] Ellis. J., Griffin, H., Morrison. D. and Johnson, A.M. (1993) Analysis of dinucleotide frequency and codon usage in the phylum Apicomplexa. Gene, 126, 163- 170. [27] Nagel, S.D. and Boothroyd, J.C. (1988) The cc-and Rtubulins of To.xoplasma gondii are encoded by single copy genes containing multiple introns. Mol. Biochem. Parasitol. 29, 261-273. [28] Lecordier, L., Mercier, C., Torpier, G., Tourvieille. B., Darcy. F., Liu, J.L.. Maes, P., Tartar, A., Capron, A. and Cesbron-Delauw, M.F. (1993) Molecular structure of a Toxoplasma gondii dense granule antigen (GRAS) associated with the parasitophorous vacuole membrane. Mol. Biochem. Parasitol. 59, 1433154. [29] Lecordier, L., Moleon-Borodowsky, I., Dubremetz, J.F., Tourvieille, B., Mercier. C., Deslee, D., Capron, A. and Cesbron-Delauw, M.F. (1995) Characterization of a dense granule antigen of To.yoplasma gondii (GRA6) associated to the network of the parasitophorous vacuole. Mol. Biochem. Parasitol. 70, 85-94. [30] Hug, M.. Carruthers, V.B., Hartmann, C., Sherman, D.S.. Cross, G.A.M. and Clayton, C. (1993) Mol. Biochem. Parasitol. 61, 87796.
ELSEVIER
Molecular
and Biochemical
Parasitology
82 (1996) 237-244
Molecular cloning of the Toxoplasma gondii sag4 gene encoding an 18 kDa bradyzoite specific surface protein’ Carmen odberg-FerraguP**‘, Martine Soete”,‘, Anne Engels”*‘, Bart Samynb, Anne Loyens”, Jozef Van Beeumenb, Daniel Camus”,‘, Jean-Fraqois Dubremetz” ‘Fuulteif
“INSERM 1142. 369, rue Jules Gursde, BP. 39, 59651 Villmeuvr d’Ascq cPdes. France Ct’ermsc~huppm.lhiversiteit Gent. Department of’ Biochemistry, Ph_vsiolog~~ and Miuobiology. K. L. Lcdeg~mckstruur 35. 9000 Gent, Belgium ‘FucultP de Medecine. Place dc Verdun. 59850 Lille. France
Received
7 May
1996: revised
5 August
1996: accepted
I5 August
1996
Abstract An 18 kDa bradyzoite specific surface protein of Toxopplasmu gondii (T. gondii) has been purified by affinity chromatography with a specific monoclonal antibody using parasites grown in vitro under conditions inducing the biosynthesis of bradyzoite specific proteins. N-terminal and internal amino acid sequences obtained by microsequencing enabled us to design degenerate oligonucleotides. A fragment of 187 bp was amplified by polymerase chain reaction (PCR). It was used as a probe to clone a 4 kb-BanrHI fragment encompassing the gene encoding the 18 kDa protein. Nucleotide sequence analysis revealed a single open reading frame of 516 nucleotides encoding a 172 amino acid protein. The deduced amino acid sequence matched perfectly the peptides microsequenced from the native protein. The N-terminal hydrophobic region was found to possess the characteristics of a signal peptide of 27 amino acids. The hydrophobic C-terminal part could represent a signal for a glycan-phosphoinositide anchor. The full-length cDNA was also isolated and both the 5’ and 3’ untranslated regions were determined. Reverse transcriptase-PCR (RT-PCR) using pl8-specific primers showed a stage-specific expression of this gene. Comparison of the nucleic acid sequence and the predicted amino acid sequence with databases did not reveal significant homology with known genes or proteins. This gene is proposed to be named sug4, according to the existing T. gondii nomenclature. Keywords:
Toxoplusrrzu gondii;
Bradyzoite
surface
antigen;
Gene cloning
AbhretWions: GPI, Glycan-phosphoinositide: PCR, polymerase chain reaction: ORF. open reading frame; mab. monoclonal antibody. * Corresponding author. Tel.: + 33 20 472397; fax: + 33 20 059172; e-mail [email protected] ’ Note: Nucleotide sequence data reported in this paper are available in the EMBL databases under the accession number 269373. z These two authors contributed equally to the work. 0166-6851/96]$15.00
CD 1996 Elsevier
PIf SO166-6851(96)02740-5
Science
B.V. All rights
reserved
238
C. odberg-Ferragut
et al. / Molecular and Biochemical Parasitology 82 (1996) 237-244
1. Introduction Toxoplasma gondii (T. gondii) is the intracellular protozoan parasite causing toxoplasmosis. This disease is responsible for transplacental infections resulting eventually in malformations or the death of congenitally infected infants [1,2]. More recently, T. gondii has been shown to be one of the major opportunistic pathogens in immunocompromised patients [3,4]. Indeed, this pathogen persists in immunocompetent hosts in cyst form within brain and muscle tissues. In a state of immune deficiency these quiescent cysts may rupture and cause a disseminated and potentially fatal disease [5,6]. A better understanding of conversion between the acute (tachyzoite) and dormant (bradyzoite) stages would therefore be a major step in controlling the disease. As this conversion is mainly a differentiation process characterized by the differential expression of stage specific molecules, the identification of these molecules is a prerequisite to study this process. Then the cloning of the stage specifically expressed genes would allow the study of their regulation, and therefore of the differentiation mechanisms. Because the tachyzoite stage is the most readily available for study, many proteins and the corresponding genes have been characterized from tachyzoites. Most of these have been shown to be common to tachyzoites and bradyzoites. Only a few have proven to be specific to the tachyzoite stage, and these are glycan-phosphoinositide (GPI) anchored surface molecules [7,8]. More recently, because of the rising interest in stage conversion, several groups have identified bradyzoite or cyst specific proteins [9-121 and the genes coding for three of them have been cloned and sequenced [13- 161. These genes code for cyst wall or bradyzoite cytoplasmic proteins. Because of the importance of surface molecules in the parasite’s interaction with its host cell and with the immune response, we have been interested in the characterization of stage specific surface proteins, as they may also be directly related to stage specific biological properties of the invasive stages. We had previously identified four bradyzoite specific proteins located at the parasite
plasmalemma [lo]. One of them, P18, a protein of 18 kDa, has been further analyzed and shown to possess characteristics of a GPI anchored surface protein (unpublished observations). We have purified this protein and obtained amino acid sequence data that have been used to clone the corresponding gene. We report here the structure of this gene and of a full length cDNA. We propose to name this gene sag4, and the gene product SAG4, according to the nomenclature in use for T. gondii genes [17].
2. Materials and methods 2.1. Cells and parasites
The parasites used in the study were of strain RH [18], of strain PLK [19] and of strain 76 K [20]. They were grown either in the peritoneal cavity of Swiss mice, or in cell cultures, on Vero cells. Induction of bradyzoite development in cell cultures was obtained by raising the pH of the culture medium to 8 [21]. 2.2. Microsequencing of SAG4 SAG4 was purified by immunoaffinity on monoclonal antibody (mab) T83Bl [lo] coupled to Sepharose 4B. A T. gondii PLK culture that had been subjected to high pH for 2 days was lysed in phosphate buffered saline (PBS) containing 10 pgg/ml leupeptin, 100 pg/ml aprotinin, 1 mM PMSF, 2 mM ehthelenediaminetetraacetate (EDTA) and 1% Nonidet P40. The purified protein was separated by sodium dodecyl sulfatepolyacrylamide gel electrophoresis (SDS-PAGE) and electroblotted on polyvinylidene fluoride membrane (Applied Biosystems). The membrane was stained with Ponceau and the 18 kDa band was excised for microsequencing. In a first experiment, the N-terminal sequence of the 18 kDa protein was obtained using the 476A protein sequencer (Applied Biosystems) with on-line PTH analysis. A second round of purification was performed. A fraction of the blot was used for amino acid analysis. The remaining fraction was submitted to an endoproteinase Lys-C digest. The pep-
C. tjdberg-Ferragut
et al. / Molecular and Biochemical Parasitology 82 (1996) 237-244
239
B
_
c
c-c__
_
_
o 1kb
Fig. I, Schematic representation of genomic locus encompassing the bradyzoite sag4 gene. (A) Relevant restriction enzyme cleavage sites. (B) Cloned genomic DNA fragment of about 4 kb. The shaded area represents the coding region and the solid bar indicates the hybridization site of the probe. Sequenced fragments are enclosed between two vertical lines and the primers used shown by the arrows.
tides were separated on a C8 RPLC (SMART System, Pharmacia, Uppsala, and sequenced as described above.
column Sweden)
spectively) were cut with Hind111 or PstI enzymes. T. gondii PLK DNA was also digested by either EcoRI, PstI or BamHI enzymes alone and by EcoRI and each of the other enzymes. DNAs
2.3. Polymerase chain reaction (PCR) Sense and antisense degenerate oligonucleotides based on the amino acid sequences of PeptIlPept-3 (Fig. 2) were synthetized by Eurogentec, Belgium. These were designated as oligola: TGGACITAYGAYTTYAARAARGC (nucleotides (nt) 88 - 110) and oligo-lb: GARATHATHACICCIGG (nt 136 152) for the sense sequences and oligo-2*: GGIGGDATRTAYTCIARIGG (nt 187-206) and oligo-3*: GCIGCIARYTCNACNGGYTCRTC (nt 253-275) for the anti-sense sequences (Fig. 2). PCR amplifications were performed on genomic DNA using Goldstar polymerase (Eurogentec, Belgium). Conditions were: 95°C for 5 min, 95°C for 1 min, 50°C for 1 min, 72°C for 1 min (5 cycles), 95°C for 1 min, 45°C for 1 min, 72°C for 1 min (5 cycles), 95°C for 1 min, 40°C for 1 min, 72°C for 1 min (25 cycles), 72°C for 5 min. PCR products were analyzed by agarose gel electrophoresis. The largest amplified product, a 187 bp fragment, was cloned into PCR-Script vector (Stratagene) and sequenced. 2.4. Southern blot and hybridization Twenty micrograms mouse, Vero cells and
of genomic
DNA
(from re-
T. gondii PLK strain
Fig. 2. Sequence of the gene encoding the SAG4 bradyzoite antigen. The nt sequence contains an ORF encoding a polypeptide of 172 amino acids. Boxed regions reflect the homologous amino acid sequences of Pept-1 (28852), Pept-2 (46-76) and Pept-3 (85-95) obtained from the microsequencing of the purified SAG4. The PCR primers are highlighted and italicised; they are designated oligo-I-oligo-7: asteriks indicate that reverse primers have been chosen. The double arrow shows the putative cleavage of the signal sequence. The unique EcoRI restriction site is underlined. Amino acid residues are numbered on the right, and the nt on the left of these sequences.
240
C. odberg-Ferragut
et al. / Molecular and Biochemical Parasitology 82 (1996) 237-244
were electrophoretically separated and transferred to nylon membranes (Immobilon-P, Millipore) by capillary blotting in alkaline conditions. Prehybridizations and hybridizations were carried out in 50% formamide, 5 x Denhart’s solution, 5 x SSPE, 0.2% SDS, 0.5% dextran sulfate and 200 pg/ml of sheared salmon sperm DNA. The 187 bp fragment was used as a probe and labelled by random priming (Boehringer kit) using [u-~*P] dATP. Radiolabeled probe was incubated with filter overnight at 42°C and unbound probe removed by washing filters three times in 2 x SSC at 42°C. Autoradiographic exposure was for a week at - 80°C.
2.5. Cloning and screening Fifty micrograms of genomic 7’. gondii DNA were digested with BamHI and electrophoresed in a 1% agarose gel. The 3.5-4.5 kb fraction was ligated to predigested ZAP Express BamHI/CIAP vector (Stratagene), packaged in vitro into /iphage particles with the Gigapack II Gold extract (Stratagene) and titered on Escherichia coli XLlBlue MRF’ strain. This size-selected DNA library was screened by plaque hybridization using the 187 bp fragment amplified as digoxigenin-labeled probe (Boehringer kit). Detection was performed by enzyme immunoassay with luminescence. Positive clones were excised in vivo to obtain directly the recombinant pBK-CMV phagemid.
2.7. Analysis of the 5’ and 3’ ends Total RNA was isolated by the procedure described by Chomczynski and Sachi [22] and modified by Clontech (total RNA separator kit). One micrgram of total T. gondii RNA was reverse transcribed using the Marathon system of Clontech. PCR amplifications of the 5’ and 3’ ends of the sag4 transcript were performed using universal adaptor primers and specific oligonucleotides of the sag4 coding sequence. The 5’ end was PCR oligo-4*: amplified the anti-sense using GCAGTGGTCTCCCTTTTCAAC (nt 229-249, Fig. 2). Similarly, the 3’ end was PCR amplified using the sense oligo-5: GGAGATTATCACACCGGGCG (nt 135-154, Fig. 2). Nested PCRs were performed using the anti-sense oligo-6*: CTCGCCAAGCCGAGCACCC (nt 55-73, Fig. 2) and the sense oligo-7: GAGAAGGAAACCATACCC (nt 361-378, Fig. 2). The largest PCR products generated were cloned into pGEMT plasmid (Promega), transfected into E. coli JM109 and partially sequenced. 2.8. Reverse transcriptase-PCR
(RT-PCR)
Total RNA was reverse transcribed by using Superscript II enzyme (BRL) and specific 3’ primers for either sag4 (oligo-5, Fig. 2) or a T. gondii microneme protein mic 1 gene [23]. Tfl polymerase (Epicentre) was used in PCR experiments with additional 5’ primers specific for both genes (oligo-4* for sag4, Fig. 2). Annealing steps were performed at 50°C.
2.6. Sequencing The 187 bp PCR insert was entirely sequenced using primers flanking the cloning sites. The 4 kb BarnHI fragment was partially sequenced using gene-specific primers. Plasmid and phagemid were purified on Qiagen columns and sequenced with the Sequenase kit (Amersham) by the double strand DNA sequencing method, using [u-~~P] dCTP. Sequence data were analyzed using the PC gene program. Homology searches of protein (Swissprot and PIR) and nucleic acid (GenBank and EMBL) databases were performed using Blitz and FASTA programs.
3. Results 3. I. Native SAG4 sequence
Direct N-terminal sequencing of SAG4 purified from in vitro grown bradyzoites resulted in the sequence designated as Pept- 1: KSWTYDFKKALDDDETKKEIITPGD (Fig. 2). Further, analysis of amino acid composition revealed that this protein was Lys-rich. The sequencing results of the Lys-C peptides obtained after in situ cleavage [24] confirmed the N-terminus and defined
C. odberg-Ferragut
et al. / Molecular and Biochemical Parasitology 82 (1996) 237~m244
two new sequences designated as Pept-2: EIITPGDSVSIENSGSRPLEYIPPNPSQVLK (Fig. 2) and Pept-3: DEPVELAALFK (Fig. 2). 3.2. Characterization
of sag4
PCR experiments were carried out on PLK, RH and 76 K T. gondii genomic DNA. Using was amoligos la and 3*, a 187 bp fragment plified. Nested PCR using oligos lb and 2* validated the first amplification product as being a fragment of the sug4 gene. No amplification was found with either mouse DNA or Vero cell DNA. Also, only T. gorzdii DNA hybridized to this 187 bp fragment. It was sequenced and translated. The deduced amino-acid sequence showed complete sequence identity with those found in Pept1 -Pept-3 (Fig. 2) and the nt sequence revealed an EcoRI site in its central area. Thus, T. gondii genomic DNA was digested with several single restriction enzymes or in combination with EcoRI and probed using the 187 bp amplified fragment. The resulting restriction map is shown in Fig. 1. The entire sug4 was cloned in ZAP Express vector from a BcrnzHI size selected library. About 3.6 x lo3 recombinant clones were screened, three clones were found positive and further characterized by PCR, one of them was sequenced. The genomic sequence encoding SAG4 from the PLK T. gondii strain consists of an open reading frame (ORF) of 5 16 bp long. The ATG codon marking the beginning of the ORF had the favourable environment for initiation of translation according to the Kozak criteria [25]. Also, an in-frame termination codon preceded this start codon. The coding region had a codon usage similar to that of other T. gondii coding regions, [26] excepted for the Ala codon. In the sag4, T was the major preference at the third position for this codon. Comparison of the sag4 and several T. gondii genes in the region immediately upstream from the ATG codon revealed putative consensus translation sequences (Fig. 3) [7,27-291. The strongest similarity was found with B-tub genes. On the contrary, the T-streches found in many 5’ untranslated sequences of T. gondii genes [29] are not obvious in sag4.
m-tub
~~~-----______~~__~~
E-tub
TTTTTC-TG----CAm--TG
sag4
-~TTTCGTC-~~~TGTCTTCAACCATG
gra5
-TTTTC%TG---PAGT-MC-AaAATG
RraC
TTmTTGGGAGmCGi-G-TG
241
Fig. 3. Comparison of 5 T. go:o12diisequences immediately upstream from the ATG. Bold letters indicate identity with the canonical sequence described by Kozak [25].Underlined bases show similarity of sequences upstream from the ATG between sag4 and various T. go&ii genes.
Three different 5’ ends were mapped by sequencing the subcloned 5’ Marathon products. They matched perfectly with the genomic sequence and ranged at 257, 229 and 198 bp respectively upstream from the ATG initiation codon. These end-points should be confirmed by Northern blot experiments. There were no typical eukaryotic TATA nor CCAAT consensus promoter sequences. Nevertheless, the hexameric sequence ACGCCG was found three times in this part 01 the gene but was not present in other sequences specifically expressed at the bradyzoite stage (1316). The polyadenylation site was at 709 bp downstream of the stop codon. This site was identified by analysis of the amplification product primed with a modified oligo dT. The higher eucaryotic polyadenylation signal AAUAAA normally found about 30 nt upstream of the poly A tail was not present. PCR amplification of both genomic and cDNA showed no significant size difference, suggesting that no intron was present in sag4.
3.3. Regulation
of sag4 expression
RT-PCR was used to study gene expression at bradyzoite and tachyzoite stages. The micl gene was used as a control for expression in both stages. Using sag4-specific primers a transcript of the expected size was amplified only in bradyzoites while the rnicl gene was expressed in both stages (Fig. 4). Parallel amplification of genomic DNA with micl-specific primers resulted in a band of larger size, due to the presence of one intron in this part of the gene (unpublished results). This band was not found in RNA amplification and demonstrated that no genomic
C. tidberg-Ferragut
242
contamination tion.
et al. / Molecular and Biochemical Parasitology 82 (1996) 237-244
was present in the RNA prepara-
3.4. Analysis of the deduced protein sequence
The gene translation product was composed of 172 amino acids and had a theoretical molecular mass of 18.5 kDa. This deduced protein showed two hydrophobic domains at both the amino and the carboxy terminus. Both regions framed a polypeptide rich in charged amino acids (14% polar negative and 15% polar positive) (Fig. 2). The sequence in the N-terminal part of the polypeptide was typical for an eukaryotic signal
1 23
45
peptide showing three positively charged aminoterminal residues and a central hydrophobic area. Two putative cleavage sites were inferred by PC gene analysis. One was between Ala 22 and Trp 23; the other was located between Ala 27 and Lys 28. This last prediction coincided with the first amino acid obtained by N- terminal sequencing of the protein and therefore assigned a secretory signal sequence of 27 amino acids. The hydrophobic signal sequence was followed by a strongly hydrophilic region (Fig. 2). This region is rich in lysine and contains only two cysteines. The C-terminal hydrophobic region consists of 21 residues. The protein exhibits the N-glycosylation amino acid consensus sequence Asn-X-Ser/Thr at residues 70 and 137 respectively. The calculated isoelectric point of the protein is 8.97.
4. Discussion 622 404 242+238 123
<
Fig. 4. RT-PCR with gene-specific primers. Total RNA was isolated either from tachyzoites (T) grown intraperitoneally in mice or from bradyzoites (B) grown in vitro under switching conditions. Tachyzoite genomic DNA was prepared from in vitro Vero cell cultures. Lane 1 represents the amplification obtained from genomic DNA with mic 1-specific primers; lanes 2 and 4 show the RT-PCR results using total RNA as template and gene-specific primer pairs for micl; lanes 3 and 5 show the RT-PCR results utilizing total RNA as template and gene-specific primer pairs for sag4. The apparent molecular mass of DNA standards is indicated on the left. The arrowhead indicates the sug4-specific signal detected in bradyzoites.
The present study reports the cloning of the gene encoding an 18 kDa protein which had been previously identified by mab as a bradyzoite surface molecule ([lo] and unpublished observations). According to the nomenclature proposed by Sibley et al. [17], the gene encoding P18 should therefore be named sag4 as surface antigen gene 4 and P18 named SAG4. Several convincing evidences suggest that this clone represents the bradyzoite sag4 gene. First, there is perfect matching of the sequences obtained from the microsequencing analysis on the native protein compared to the sequences derived from the cloned gene. Second, the size of the coding sequence is as expected and this latter contains a signal peptide. In addition, a C-terminal sequence likely to be removed and replaced by a GPI anchor has been found. Further evidence for this C-terminal removal is given by the fact that the mature protein is not metabolically labeled with [35S] methionine, indicating that this amino acid position 172 is removed during processing. Third, the RT-PCR experiment using sag4-specific primers shows a stage-specific expression of this gene. This type of regulation has also been found for bag 1 [ 141.
C. odberg-Ferragut
et al. 1 Molecular and Biochemical Parasitology 82 (1996) 237-244
Comparison of the predicted sequence with databases does not reveal significant homology with any known protein. Results of our Southern blot analysis (data not shown) are consistent with the sag4 gene being a single copy gene. It is noteworthy that the putative promoter region of the sag4 gene showed perfect identity between strains PLK and RH. As these two strains show a significant difference in susceptibility to bradyzoite differentiation triggers, i.e. in the initiation of SAG4 expression, this result suggests that this difference is likely to be due to factors which are independent of the sag4 promoter. This hypothesis is further supported by the fact that expression of P36 and P34 bradyzoite antigens follows similar kinetics as SAG4 antigen [21]. Nevertheless, it will be necessary to compare the sequence of the 3’ untranslated region of the sag4 gene from both strains and to study their influence on the .expression level. The influence of this region has been demonstrated for other protozoan genes [30]. We cannot draw any conclusion concerning a bradyzoite specific promoter. The consensus sequence TGCTGTGTC detected in two bradyzoites genes [13,14] was not found in the sequenced part of sag4. Other genes specifically expressed at the bradyzoite stage encode proteins found in the cyst wall [13] or in the parasite cytoplasm [l&16]. The surface localization of SAG4 makes the cloning of this gene a unique tool to study the functional aspect of differential surface protein expression. As bradyzoites and tachyzoites develop in different locations, their interaction with the environment is primarily dependent on their surface properties. Differential characteristics such as digestive enzyme susceptibility, or enterocyte invasion, could be related to differences in surface protein expression. A powerful way of studying these functions will be the transfection of these genes under the control of heterologous promoters in order to modify the stage specificity of expression and analyse the effects on the biology of the parasite. These experiments are presently under way.
243
Acknowledgements
The authors are indebted to M.N. Fourmaux for excellent advice in using digoxigenin-labelled probes and to J.P. Touzel for his kindness and help in using database programs. This work was funded by the Agence National de Recherche sur le SIDA (ANRS), by EC (program BMHl-CT921535) and by a Concerted Research Action of the Flemish Government (contract 12052293). Martine Soete was supported by ANRS.
References McCabe, R.E. and Remington, J.S. (1983) The diagnosis and treatment of toxoplasmosis Eur. J. Clin. Microbial. 2. 15. VI Remington, J.S. and Desmonts, G. (1990)Toxoplasmosis. In: Infectious Diseases of the Fetus and Newborn Infant (Remington, J.S. and Klein, J.O., eds.). pp. 899195, W.B. Saunders, Philadelphia. 131 Luft, B.J., Brooks, R.G., Conley, F.K.. McCabe, R.E. and Remington, J.S. (1984) Toxoplasmic encephalitis in patients with acquired immune response deficiency syndrome. JAMA 252, 913-917. [41 Wong, J., Gold, W.M., Brown, A.E., Lange, M., Fried, R.. Grieco, M., Mildvan, D., Giron, J., Tapper, M.L., Lerner, C.W. and Armstrong, D. (1984) Central-nervoussystem toxoplasmosis in homosexual men and parenteral drug abusers. Ann. Intern. Med. 100. 36642. J.S. (1989) VI Suzuki. Y., Conley. F.K. and Remington. Differences in virulence and development of encephalitis during chronic infection vary with the strain of Toroplasma gondii. J. Infect. Dis. 159. 790-794. WI Gazinelli, R.T., Eltoum, I., Wynn. T.A. and Sher, A. (1993) Acute cerebral toxoplasmosis is induced by in vivo neutralization of TNF-alpha and correlates with the down-regulated expression of inducible nitric oxide synthase and other markers of macrophage activation. J. Immunol. I5 1, 3672-368 1. [71 Burg. J.L., Perelman, D., Kasper, L.H.. Ware, P.L. and Boothroyd. J.C. (1988) Molecular analysis of the gene encoding the major surface antigen of To.wplasma gondii. J. Immunol. 141, 358443591. J., Parmley. S.F., PI Prince. J.B.. Auer. K.L.. Huskinson. Ardujo, F.G. and Remington. J.S. (1990) Cloning, expression, and cDNA sequence of surface antigen P22 from Toxoplasma gondii. Mol. Biochem. Parasitol. 43. 97- 106. [91 Omata. Y., Igarashi, M., Ramos, M.I. and Nakabayashi, T. (1989) Toxoplasma gondii: antigenic differences between endozoites and cystozoites defined by monoclonal antibodies. Parasitol. Res. 75. 189- 193.
[‘I
244
C. iidberg-Ferragut
et al.
I Molecular and Biochemical Parasitology 82 (1996) 237-244
[lo] Tomavo, S., Fortier, B., Soete, M., Ansel, C., Camus, D. and Dubremetz, J.F. (1991) Characterization of bradyzoite-specific antigens of Toxoplasma gondii major surface antigens. Infect. Immun. 59, 3750-3753. [Ill Weiss, L.M., LaPlace, D.. Tanowitz, H.B. and Wittner, M. (1992) Identification of Toxoplasma gondii bradyzoitespecific monoclonal antibodies. J. Infect. Dis. 166, 213215. [12] Bohne, W., Heesemann, J. and Gross, U. (1993) Induction of bradyzoite-specific Toxoplasma gondii antigens in gamma interferon-treated mouse macrophages. Infect, Immun. 61. 1141-1145. [I31 Parmley, S.F., Yang, SM., Harth, G., Sibley. L.D., Sucharczuk. A. and Remington, J.S. (1994) Molecular characterization of a 65 kilodalton Toxoplasma gondii antigen expressed abundantly in the matrix of tissue cysts. Mol. Biochem. Parasitol. 66. 2833296. [14] Bohne, W., Gross. U., Ferguson, D.J.P. and Heesemann, J. (1995) Cloning and characterization of a bradyzoitespecifically expressed gene (hsp /bag 1) of Toxoplasma gondii, related to genes encoding small heat-shock proteins of plants. Mol. Microbial. 16, 1221-1230. [15] Parmley, S.F.. Weiss, L.M. and Yang, S. (1995) Cloning of a bradyzoite-specific gene of Toxoplasma gondii encoding a cytoplasmic antigen. Mol. Biochem. Parasitol. 73, 2533257. [16] Yang, S. and Parmley, S.F. (1995) A bradyzoite stagespecifically expressed gene of Toxoplasma gondii encodes a polypeptide homologous to lactate dehydrogenase. Mol. Biochem. Parasitol. 73, 291-294. [17] Sibley, L.D., Pfefferkorn, E.R. and Boothroyd, J.C. (1991) Proposal for a uniform genetic nomenclature in Toxoplasma gondii. Parasitol. Today 7, 327-328. [18] Sabin. A. (1941) Toxoplasmic encephalitis in children. JAMA 116. 801-814. [19] Kasper, L.H. and Ware, P.L. (1985) Recognition and characterization of stage-specific oocyst sporozoite antigens of T. gondii by human anti-serum. J. Clin. Invest. 75, 1570-1577. [20] Laugier, M.N. and Quilici, M. (1970) Inter&t experimental d’une souche de toxoplasme peu pathogene pour la souris. Ann. Parasitol. Hum. Comp. 45, 3899403.
[21] Soete, M.. Camus, D. and Dubremetz, J.F.D. (1994) Experimental induction of bradyzoite-specific antigen expression and cyst formation by the RH strain of TO.XOplasma gondii in vitro. Exp. Parasitol. 78, 361-370. [22] Chomczynski, P. and Sacchi, N. (1987) Single-step method of RNA isolation by acid guanidium thiocyanatephenol-choroform extraction. Anal. Biochem. 162, 156159. [23] Mercereau-Puijalon, 0.. Fourmaux. M.N. and Dubremetz, J.F.D. (1993) Expression of apical organelles antigens by a Tosoplasma gondii genomic library. In: Toxoplasmosis (Smith, J.E., ed.), pp. 19931, SpringerVerlag. Berlin. [24] Fernandez, J., Andrews, L. and Mische, S.M. (1994) An improved procedure for enzymatic digestion of polyvinylidene difluoride-bound proteins for internal sequence analysis. Anal. Biochem. 218, 112-117. [25] Kozak, M. (1986) Point mutations define a sequence flanking the AUG initiator codon that modulates translation by eukaryotic ribosomes. Cell, 44, 283-292. [26] Ellis. J., Griffin, H., Morrison. D. and Johnson, A.M. (1993) Analysis of dinucleotide frequency and codon usage in the phylum Apicomplexa. Gene, 126, 163- 170. [27] Nagel, S.D. and Boothroyd, J.C. (1988) The cc-and Rtubulins of To.xoplasma gondii are encoded by single copy genes containing multiple introns. Mol. Biochem. Parasitol. 29, 261-273. [28] Lecordier, L., Mercier, C., Torpier, G., Tourvieille. B., Darcy. F., Liu, J.L.. Maes, P., Tartar, A., Capron, A. and Cesbron-Delauw, M.F. (1993) Molecular structure of a Toxoplasma gondii dense granule antigen (GRAS) associated with the parasitophorous vacuole membrane. Mol. Biochem. Parasitol. 59, 1433154. [29] Lecordier, L., Moleon-Borodowsky, I., Dubremetz, J.F., Tourvieille, B., Mercier. C., Deslee, D., Capron, A. and Cesbron-Delauw, M.F. (1995) Characterization of a dense granule antigen of To.yoplasma gondii (GRA6) associated to the network of the parasitophorous vacuole. Mol. Biochem. Parasitol. 70, 85-94. [30] Hug, M.. Carruthers, V.B., Hartmann, C., Sherman, D.S.. Cross, G.A.M. and Clayton, C. (1993) Mol. Biochem. Parasitol. 61, 87796.