A protease inhibitor associated with the surface of Toxoplasma gondii

Molecular & Biochemical Parasitology 116 (2001) 137– 145 www.parasitology-online.com.

A protease inhibitor associated with the surface of Toxoplasma gondii  Johan G. Lindh a,b,1, Silvia Botero-Kleiven a,b, Juan-Ignacio Arboleda a,b, Mats Wahlgren a,b,* a

Microbiology and Tumor Biology Center (MTC), Karolinska Institutet, Box 280, S-171 77 Stockholm, Sweden b Swedish Institute for Infectious Disease Control S-171 82 Solna, Sweden Received 15 November 2000; received in revised form 14 May 2001; accepted 31 May 2001

Abstract Toxoplasma gondii has a broad host-range including man and a variety of warm-blooded animals. The ability to infect and survive in this wide spectrum of hosts suggests highly evolved mechanisms to handle the harsh environments encountered. Here we show that extracellular tachyzoites are resistant to milligram levels of trypsin and describe the presence of an inhibitor of trypsin associated with the surface of T. gondii, TgTI. TgTI has an estimated molecular mass of 37 000 dalton and is encoded by the TgTI-gene which is found at low abundance as an expressed sequence tag (EST) in both the bradyzoite and tachyzoite stages. The inhibitory binding region was found to be in the N-terminus of TgTI where aminoacid-alignment to earlier described protease inhibitors demonstrates 75% similarity. In functional analysis, recombinant TgTI-protein inhibits the activity of trypsin approximately 10 times more efficiently than an inhibitor isolated from soybean. In contrast to other known trypsin inhibitors, TgTI also possesses a predicted membrane-binding region. Polyclonal antibodies raised against recombinant TgTI bind to the surface of the tachyzoite stage as seen both by immunofluorescence and immunoprecipitation of surface labelled parasite proteins. The high survival rate of the parasite in the upper gastrointestinal tract may be enhanced by the presence of the TgTI-molecule. © 2001 Elsevier Science B.V. All rights reserved. Keywords: Toxoplasma gondii; Trypsin inhibitor; Surface located; EST

1. Introduction Infections of humans by the parasitic protozoan Toxoplasma gondii normally occurs orally (reviewed in [1,2]). The parasite enters the host through the upper digestive tract and eventually invades the epithelial cells of the small intestine, a process during which it is exposed to a protease rich, hostile environment [3,4]. Some of the enzymatic activity found in the small Abbre6iations: aa, aminoacid; EST, expressed sequence tag; GPI, glycosylphosphatidylinositol; GST, glutathione-S-transferase; nt, nucleotide; ORF, open reading frame; SAG, surface antigen.  Note: Nucleotide sequence data reported in this paper are available in the GenbankTM database under the accession number AY007253. * Corresponding author. Tel.: + 46-8457-2510; fax: + 46-8310-525. E-mail address: [email protected] (M. Wahlgren). 1 Present address: Institute of Bioscience and Process Technology, Vaxjo University, SE-352 00 Vaxjo, Sweden.

intestine is due to serine proteases such as trypsin and chymotrypsin, which cleave peptide substrates on the carboxyl side of lysine and arginine and the lateral chains of aromatic aminoacids, respectively [5]. In an attempt to understand the resistance of T. gondii to protease-rich environments such as that encountered in the small intestine, we have screened an EST database for sequences presenting similarity to trypsin inhibitors (Toxoplasma Database of Clustered ESTs) [6–9]. An EST with a high score (1165401) and a rather long open-reading frame (ORF) was identified. The predicted protease inhibitor was suggested by computer analysis to be cleaved by a signal peptidase at aminoacid (aa) 22 in order to be located either at the surface or integrated into a membrane [10,11]. In opposition to other known trypsin inhibitors, which are secreted, the suggested inhibitor from T. gondii further possesses a predicted membrane-spanning region towards the carboxyl end. We have here identified the

0166-6851/01/$ - see front matter © 2001 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 6 - 6 8 5 1 ( 0 1 ) 0 0 3 1 4 - 0

138

J.G. Lindh et al. / Molecular & Biochemical Parasitology 116 (2001) 137–145

gene encoding this protease inhibitor and studied the function as well as the location of the encoded polypeptide within the parasite. We propose that EST 1165401 from T. gondii encodes a gene-product that acts as a trypsin inhibitor in association with the surface of the parasite. The polypeptide has been named T. gondii Trypsin Inhibitor (TgTI) and we think that it could be an important molecule for the survival of the parasite at extracellular locations.

2. Materials and methods

2.1. Identification of TgTI The Toxoplasma Database of Clustered ESTs [6] was used in a word search for trypsin and inhibitor. Further analysis was performed in MacVector™ 7.0 analysis software (Oxford Molecular, England) and AnalyzeSignalase, a free ware Macintosh program that utilizes the algorithm of von Heijne, 1986 [10,11].

2.2. Parasites T. gondii RH strain tachyzoites, kindly given by L.D. Sibley (Washington University, St. Louis, MO, USA) and the lacZ-expressing clone RH2b1, generously provided by D. Soldati (Universita¨ t Heidelberg, Heidelberg, Germany) [12] were routinely maintained in human foreskin fibroblasts (ATCC CRL-1634) cultured in RPMI-5% fetal calf serum.

2.3. Molecular cloning and sequencing of the TgTI gene PCR primers for the amplification of EST 1165401 were designed based on the nucleotide (nt) sequence corresponding to the ORF identified within EST 1165401. The EST 1165401 primer set (numbers correspond to nt position, bold nucleotides correspond to added restriction sites) was For.254 (TTTCAATTGATGAAGAGCAAATGTAC) and Rev.455 (GGTCGCACGGGCCTGTTG). PCR amplification was performed under standard conditions [13]. A RHtachyzoite specific lZAPII library was used as template (AIDS Research and Reference Reagent Program, NIAID, NIH, USA). Cycle conditions were 25×(94 °C 30¦, 52 °C 30¦, 72 °C 90¦) ending with 10 min extension at 72 °C. In order to get a full-length cDNA clone, the PCR amplification product was labelled with aCTP-P32 as described previously [13] and used as a probe in a cDNA library screen according to the protocols of the manufacturer (Stratagene, USA). The plasmid from the positive cDNA lphage was excised in vivo according to the description of the manufacturer (Stratagene).

Sequencing reactions were performed with Pharmacia dyekit (Pharmacia Biotech, Sweden). The sequence was analyzed with the MacVector™ 7.0 analysis software (Oxford Molecular).

2.4. RNA extraction and Northern blot analysis Parasites from freshly lysed host cells were harvested, washed twice in phosphate buffered saline (PBS) and counted before lysis with solution D for RNA isolation as described [14]. RNA blotting and hybridisation was performed as described earlier [13,15]. The TgTI specific probe was a 977 nt PCR amplification product from the coding region (see Section 2.5).

2.5. Expression of TgTI in E. coli PCR primers for the amplification of the full length ORF were designed based on the nt sequence corresponding to the ORF identified within the EST 1165401-gene. The ORF primer set (numbers correspond to nt position, bold nucleotides correspond to added restriction sites) was For.254 (TTTCAATTGATGAAGAGCAAATGTAC) and Rev.1231 (TTTCTCGAGTCATTCCAAAGGAATATAT). The SK-Bluescript plasmid (Stratagene) containing the positive cDNA insert was used as a template in a standard PCR (see above). The amplified product was ligated into the EcoR1 and Xho1 sites of a PGEX4t1 plasmid (Pharmacia Biotech). The GST-fusion protein was expressed and purified as recommended by the manufacturer (Pharmacia Biotech, Sweden). The malaria GST-fusion protein used as a negative control corresponds to the Duffy-Binding-Like-1-domain of PfEMP-1 [16].

2.6. Antibody production and Western blot analysis The recombinant TgTI GST-fusion protein was injected into a female New Zealand rabbit together with Freund’s complete adjuvant at day 0 after which the animal was further boosted at days 20, 40, 60 and 80 with recombinant protein and Freund’s incomplete adjuvant as described previously [17]. The total volume of serum was collected at day 90 and antibodies were batch-purified against pure GST protein coupled to glutathione coated beads in order to eliminate crossreactivity as described by the manufacturer (Pharmacia Biotech). Total toxoplasma lysate was obtained from 107 freshly harvested washed tachyzoites, which were sonicated twice for 30 s each at 4 °C with a Soniprep 150 (MSE Scientific Instruments, England) at 10 m in 100 ml PBS. Approximately 10 mg of the proteins in the lysate was mixed with an equal volume of 2×SDSPAGE loading buffer and boiled for 5 min prior to size-separation by SDS-PAGE using a mini- or maxi-

J.G. Lindh et al. / Molecular & Biochemical Parasitology 116 (2001) 137–145

gel system (BioRad, USA), fixed and stained with Coomassie Brilliant Blue [17]. Alternatively, the proteins were blotted onto PVDF membranes (BioRad) after separation using a semi-wet method. The membranes were blocked in 3% (w/v) bovine serum albumin (BSA) in PBS for 1 h. The primary antibodies, either anti-TgTI or anti-SAG1 monoclonal antibody (generous gift from E. Petersen, Laboratory of Parasitology, Statens Serum Institut, Copenhagen, Denmark) were added in PBS/3% BSA at room temperature for 1 h followed by 3×15 min washes with PBS. Incubation with horseradish peroxidase-conjugated goat anti-rabbit IgG (Dako, Denmark, diluted 1:20 000 in PBS/3% BSA) proceeded for 1 h at room temperature, followed by 3×15 min washes with Tris– buffered – saline (TBS, 10 mM Tris–base, pH8, 150 mM NaCl), and enzymatic reaction developed with ECL Plus as described (Pharmacia Biotech). Intensity measurements of the enzymatic reaction on the membrane were performed using Image Gauge (Fujifilm Science Lab) software.

2.7. Localisation by immunofluorescence For immunolocalisation in live tachyzoites, 106 parasites were washed 3× 5 min with PBS and incubated with blocking buffer (PBS/3% BSA) at 37 °C for 1 h. Rabbit anti-TgTI serum was applied at a dilution of 1:100 in blocking buffer for 1 h, followed by 3× 5 min washes with PBS. Fluorescein (DTAF)-conjugated donkey anti-rabbit IgG (H+ L), (Jackson ImmunoResearch Laboratories, Inc.) was used at a dilution of 1:100 in blocking buffer for 1 h. Parasites were then washed 3× 5 min in PBS, mounted and analyzed under incident UV-light in a Nikon optiphot-2 UV microscope. For immunolocalisation in fixed tachyzoites, 106 parasites were washed with PBS and fixed in 0.1 M phosphate buffer, pH 7.4 containing 3% paraformaldehyde for 2 h at room temperature, washed two times in PBS after which they were applied onto poly-L-lysine coated glass slides, and left to dry. Samples were then incubated in blocking buffer (PBS/1% BSA/50 mM glycine) at 37 °C for 1 h, and the rabbit anti-TgTI serum was applied at a dilution of 1:100 in blocking buffer at 37 °C for 1 h followed by PBS washes. The Texas Red-conjugated donkey anti-rabbit IgG (H+L) (Jackson ImmunoResearch Laboratories, Inc) was used at a dilution of 1:1000 in blocking buffer for 1 h. Parasites were subsequently washed in PBS, mounted and analysed under incident UV-light in a Nikon optiphot-2 UV microscope.

2.8. Phospholipase C digestion Freshly harvested RH tachyzoites (5× 106) were washed three times in PBS and incubated in 100 ml

139

PBS/10 mM dithiothreitol for 2 h at 37 °C with 1 unit of phosphatidylinositol-specific phospholipase C (PIPLC, Sigma, USA) [18,19]. The enzymatic treatment did not cause any significant loss of parasite viability as assessed by microscopic examination. For immunoblot analysis of possible TgTI glycosylphosphatidylinositol (GPI)-anchoring after PI-PLC treatment, tachyzoites were lysed and treated as in Section 2.6.

2.9. Iodination and immunoprecipitation Freshly harvested tachyzoites were washed three times with PBS and purified from host cell debris and dead parasites by separation on a Percoll gradient as in [20]. Purified tachyzoites were evaluated for viability with trypan blue. After counting, they were adjusted to 2× 108 ml − 1 and surface labelled with 125Iodine in a lactoperoxidase catalyzed reaction [21]. Labelled parasites were lysed (1 h on ice with periodic vortex) in 0.5% Triton-X100 with 300 mM NaCl, 50 mM Tris–HCl pH 7.6, and protease inhibitors: 10 mg ml − 1 Aprotinin, 10 mg ml − 1 Leupeptin (Sigma), 10 mM Pefabloc (Roche Diagnostics) and 1 mM EDTA. The parasite lysate was separated by centrifugation at 10 000× g for 15 min into a soluble and an insoluble fraction and the former was immunoprecipitated as described earlier [17] with either a non-immune or immune serum raised against the recombinant fusion protein using protein A-agarose beads (Sigma). All fractions were evaluated in a gamma counter and equal counts of each sample were loaded and size separated with SDS-PAGE. The dried gels were scanned with a Phosphorimager and analyzed with an ImageQuant analysis software (Molecular Dynamics, Ann Arbor, USA).

2.10. Trypsin functional assays Freshly harvested tachyzoites of the lacZ-expressing clone RH2b1 were washed three times in PBS and 0.5× 106 parasites were incubated separately together with 0, 0.1, 1, 10 and 100 mg ml − 1 of bovine trypsin (Sigma) for 1 h at 37 °C. The parasites were washed three times in PBS prior to infection of a 25 cm2 flask containing a confluent monolayer of human foreskin fibroblasts. After 24 h infection debris and non-infective parasites were washed away and infected fibroblasts lysed in 4 ml lysis buffer (60 mM Na2HPO4, 40 mM NaH2PO4, 10 mM KCl, 1 mM MgSO4, 50 mM 2 Mercaptoethanol, 2.5 mM EDTA and 0.125% NP-40) at 4 °C with periodic vortexing every 20 min for 2 h. For the enzymatic reaction 300 ml of lysed cells/parasites were added to 700 ml of chlorophenolred-b-Dgalactopyranoside (CPRG) mixture as described [12,22] and incubated at 37 °C for 0.5 h, after which absorbance was measured at 570 nm.

140

J.G. Lindh et al. / Molecular & Biochemical Parasitology 116 (2001) 137–145

2.11. Trypsin inhibitor assay Trypsin inhibiting activity was measured as described earlier [23]. Briefly, the substrate Bz– Ile– Glu – Gly – Arg –4-NitroAniline (Sigma) was incubated together with 17 pmol of bovine trypsin (Sigma). The activity of trypsin could be monitored as the amount of 4-nitroaniline released per unit. The amount of 4-nitroaniline was measured as an increase in absorbance with time at 405 nm. Recombinant TgTI fusion protein or Soybean trypsin inhibitor (Sigma) were added into the system at different concentrations and the effects were monitored and analysed in the same way. The TgTI concentration was determined using the BioRad Protein Assay (Bio-Rad, USA).

3. Results

3.1. Identification and organisation of a potential secreted/membrane-bound protein Our preliminary observations and those of others [24] suggested that tachyzoites are tolerant to physiological concentrations of trypsin (about 0.1 mg ml − 1 in the small intestine) [25,26]. In order to further address this question, freshly harvested tachyzoites containing the lacZ gene from E. coli as a marker were incubated for 1 h at 37 °C with 0– 100 mg ml − 1 of trypsin, washed and introduced to a monolayer of human fibroblasts for 24 h [12]. The rate of infection was determined by measuring b-galactosidase activity in the cell lysate. Infecting tachyzoites were shown to be completely resistant to up to 1 mg ml − 1 trypsin treatment, while higher concentrations gave a progressive inhibitory effect (Fig.

Fig. 1. Resistance of T. gondii tachyzoites to trypsin. T. gondii RH2b1 strain tachyzoites stably transfected with the b-galactosidase gene were incubated together with different concentrations of bovine trypsin for 1 h at 37 °C and returned to a confluent monolayer of human fibroblasts. After a 24 h infection, debris and non-infective parasites were washed away and infected fibroblasts lysed in lysis buffer and b-galactosidase activity measured. Control parasites were incubated without trypsin prior to infection of fibroblasts. Uninfected fibroblasts did not have any detectable b-galactosidase activity. There is a marked difference between 10 and 1 mg ml − 1. The results are derived from duplicates of two independent experiments.

Fig. 2. Northern blot of TgTI-gene using T. gondii RNA. Total RNA (3 mg) isolated from the tachyzoite stage was size separated, blotted onto a nylon membrane and hybridized to the cDNA insert obtained from the library screening (see text for details). The positions of RNA size markers are shown on the left hand side of the blot.

1). In order to pursue this finding, a word search of the T. gondii EST database using ‘trypsin and inhibitor’ was performed. One single EST-sequence was identified as having similarity to a trypsin inhibitor, EST 1165401. A primer was designed starting at the first potential ATG codon (nt 254), which was surrounded by the consensus cystidines at positions -4 and -1 and adenines at position -2 and -3 [27], and used together with a primer going upstream (nt 455). The PCR product was labeled and used to probe a lZAPII cDNA library specific for the tachyzoite stage of T. gondii, from which a 1.4 kb cDNA insert was cloned. In order to confirm that a full length cDNA had been obtained, a blot of total RNA isolated from the tachyzoite stage and separated after size was probed with the cDNA insert. The cDNA insert hybridized to a RNA molecule of approximately 1.4 kb, suggesting that the cDNA insert obtained was full length (Fig. 2). The entire cDNA-insert was sequenced and one ORF was identified. Starting codon located at nt 254 (Fig. 3A) and the first termination codon at nt 1228, this predicted ORF would be translated into a polypeptide of a predicted molecular weight of 36 900 Da. The AnalyzeSignalase 2.03 prediction program for identification of signalpeptides identified a potential cleavage site for a signal peptide at aminoacid 20 (Fig. 3A). Using MacVector software and Dense Alignment Surface method (DAS), further aminoacid analysis revealed at least two potential transmembrane regions towards the carboxyl end of the predicted ORF [28]. It could be argued that the parasite uses another start codon, e.g. at nt 281–283 or at nt 317–319, which would result in a protein without a signal peptide. These start codons are not surrounded by any favourable contexts (nt 250–252) while the suggested start codon at position nt 264–266 is, hence this is most likely the start codon used by the parasite [27].

J.G. Lindh et al. / Molecular & Biochemical Parasitology 116 (2001) 137–145

The sequence data did reveal an area between aa 32 and 51 with 70% similarity to other trypsin inhibitors (Fig. 3B).

3.2. Localisation studies Polyclonal antibodies were raised towards the TgTI fusion protein and depleted of GST reactivity by absorption on glutathione beads bound to a malarial GST fusion protein (Fig. 4A). The remaining negatively affinity purified antibodies were named PabTgTI. PabTgTI recognised an antigen of 34 kDa in whole tachyzoite lysate while no remaining cross-reactivity was seen towards the control GST fusion protein (Fig. 4B). The small size difference between the recognised protein band (34 kDa) and the predicted size (37 kDa) of the corresponding translated ORF could be due to cleavage of the signalpeptide from the endogenous protein. We therefore think that these antibodies are specific for the protein deriving from EST 1165401, TgTI. Previously described surface proteins of T. gondii have been demonstrated to be GPI-anchored, e.g. the major surface antigen, SAG1 [18]. In order to investigate if TgTI is GPI-anchored to the surface, whole lysates of PI-PLC treated parasites were size-separated by SDS-PAGE and blotted. PabTgTI recognised a protein band of 34 kDa at an intensity similar to a band of equivalent molecular weight in the untreated lysate, while antibodies against SAG1 recognised a band which was reduced by 61% in intensity when compared to one in untreated lysate (Fig. 4C). These

141

results suggest that unlike SAG proteins, the TgTI molecule is not affected by PI-PLC and thus not GPIanchored. Live and fixed parasites were used in immunofluorescence localization studies with similar results. The PabTgTI was found to bind to the surface of the parasite (Fig. 5), although fluorescence was of lower intensity than that of parasites probed with antibodies against the main surface antigen, SAG1 (data not shown). In order to get confirmation of surface localization, immunoprecipitation of iodinated surface proteins was performed. PabTgTI gave a weak precipitation of two different molecules, the expected protein at 34 kDa and another molecule at 80 kDa. Since reactivity to the 80 kDa protein was also seen with the control non-immune antibodies at an identical concentration, this was probably due to background reactivity found in the non-immune rabbit (Fig. 6).

3.3. Functional studies When the aminoacid sequence of the ORF was aligned in a BLAST search (NCBI), the highest scoring sequences were found to belong to a family of protease inhibitors, including different types of trypsin- and tryptase-inhibitors. In order to determine if the TgTI protein could have an inhibitory effect on trypsin, a trypsin reactivity assay was set up. In this method, the reaction products Bz–Ile–Glu –Gly –Arg and 4-nitroaniline are obtained by trypsin cleavage of the substrate Bz–Ile–Glu –Gly –Arg –4-NA. The amount of

Fig. 3. Nucleotide map of the gene corresponding to EST 1165401, the TgTI gene. (A) The nucleotide sequence is numbered at the top left. The deduced ORF, starting at nt 254 is shown below the DNA sequence, in single letter code. Bold letters indicate conserved nucleotide sequence upstream of initiation codon; the open arrow indicates suggested site of signal peptide cleavage; the filled arrows shows location of primers used. The grey box indicates region with similarity to other trypsin inhibitors. The black box corresponds to the suggested trans-membrane region. Scale: 1 cm approx. corresponds to 200 nt; (B) sequence alignment showing the similarity between TgTI (top) and other known trypsin/tryptase inhibitors at the protease binding region (swine trypsin inhibitor, fourth line; rat pancreatic trypsin inhibitor, third line; leech derived tryptase inhibitor, second line) as determined by a Macvector software program. Dark gray shaded boxes, show identical residues; light gray shaded boxes, conserved residues. Numbering of the TgTI aa sequence seen above TgTI sequence.

142

J.G. Lindh et al. / Molecular & Biochemical Parasitology 116 (2001) 137–145

[6]. Characterisation of the full gene product demonstrates the presence of a trypsin inhibitor associated with the surface of T. gondii and suggests that it holds a role in the biology of the parasite. The immunofluorescence pattern seen with antibodies against TgTI suggests that it is associated with the surface of tachyzoites, and that the abundance is much lower than the main surface antigen described, SAG1 [29,30]. This is furthermore indicated by the low frequency of EST-sequences with similarity to TgTI compared with ESTs identified as SAG1 (1 versus 100 EST sequences, respectively, from the tachyzoite stage) [6]. The low abundance of the anti-TgTI antibodies binding to the surface of the tachyzoites could also be due to Fig. 4. Western blot of T. gondii showing the expression of the TgTI protein. (A) TgTI/GST recombinant fusion protein and a malarial/ GST recombinant fusion protein (GST) were loaded onto the gel in equal amounts. The PabTgTI recognises a protein of approximately 65 kDa which corresponds to the TgTI/GST recombinant protein (37+ 28 kDa). No crossreaction could be demonstrated between the PabTgTI and the malarial/GST recombinant protein; (B, C) total protein extracts from T. gondii tachyzoites were size separated, blotted and immunoprobed using either PabTgTI, diluted 1:500 (B) or MabSAG-1, diluted 1:1000 (C). Prior to lysis parasites were treated (+ p) or not treated ( − p) with PI-PLC. PabTgTI recognises one major product at 34 kDa with the same intensity in treated and non-treated samples while MabSAG-1 recognises one major product at 32 kDa with 61% less band intensity in lane + p. Protein size markers are shown on the left hand side.

4-nitroaniline released is measured colorimetrically by measuring absorbance at 405 nm (Fig. 7A). When a soybean trypsin inhibitor with known effect was added at different concentrations into the system, a decrease in absorbance could be demonstrated as a result of the reduction in 4-nitroaniline released (Fig. 7B). The same effect was also seen when the recombinant TgTI-fusion protein was added at different concentrations, while no effect was observed with a control malarial GST-fusion protein (data not shown). In order to obtain complete inhibition of trypsin activity by the soybean inhibitor in our system, a 1:0.6 (trypsin:soybean inhibitor) molar ratio was needed, while only a 1:0.06 (trypsin:TgTI-fusion protein) molar ratio was needed when the TgTI-fusion protein was used.

4. Discussion Our data demonstrate that the tachyzoite stage of T. gondii is resistant to trypsin at concentrations higher than the physiological concentrations found in the intestinal fluid [26]. In order to understand the molecular mechanism underlying the tolerance to trypsin, we searched a T. gondii EST database and identified a sequence with similarity to eukaryotic trypsin inhibitors

Fig. 5. Indirect immunoflourescence of T. gondii tachyzoites showing the localisation of TgTI. (A, B) Live parasites were incubated with PabTgTI (diluted 1:100); (C, D) control with non-immune serum (diluted 1:100). DTAF-labelled anti-rabbit antibodies were used to visualize binding of primary antibodies; (A, C) corresponding light microscopy pictures; (E, F) paraformaldehyde fixed parasites were incubated with PabTgTI (diluted 1:100); (G, H) control with non-immune serum (diluted 1:100). Texas-Red-labelled anti-rabbit antibodies were used to visualize binding of primary antibodies; (E, G) corresponding phase-contrast pictures. Scale bar, 10 mm.

J.G. Lindh et al. / Molecular & Biochemical Parasitology 116 (2001) 137–145

Fig. 6. Immunoprecipitation of the TgTI. Live T. gondii tachyzoites were surface labelled with 125Iodine, then lysed with Triton-X100. (A) Triton soluble proteins were immunoprecipitated with non-immune serum; (B) Triton soluble proteins were immunoprecipitated with PabTgTI antibody. Arrow indicates specific PabTgTI immunoprecipitated protein at 34 kDa. Protein size markers are shown on the left hand side.

sterical hindrance from SAG-1, which corresponds to 5% of the whole protein production in the tachyzoite [30]. Nevertheless, in order to investigate the abundance of the TgTI protein, a more detailed analysis needs to be performed. Western blot analysis of PI-PLC treated tachyzoites suggests that TgTI does not use a GPI-anchor in order to be associated with the surface. The low efficiency of the PI-PLC enzymatic digestion seen (61% reduced intensity of SAG1) could also be due to sterical hindrance by the high abundance of SAG1 proteins at the surface [18,19]. Previously described trypsin/tryptase inhibitors are often of molecular masses less then 10 kDa, while the predicted molecular weight for TgTI is 37 kDa. The difference in molecular weight could be due to the fact that TgTI is found associated with the parasite surface and contains suggested transmembrane regions while most previously described inhibitors are secreted

143

[5,31,32]. A 3-D structure of a leech-derived tryptase inhibitor (LDTI) has previously been resolved, allowing us to predict the active region with high accuracy [33]. Aminoacids 3–10 of the LDTI are the most important aa in order to allow an interaction between the inhibitor and the protease (corresponding to aa 29–36 in TgTI) [31,33]. In this region, the similarity between TgTI and other inhibitors is above 62% [5,31–34], and lysine 8, corresponding to lysine 34 in TgTI, is suggested to be the aa with the strongest binding to the protease (Fig. 2B). This domain of similarity is located downstream of a sequence suggested to be a signal-peptide. While previously described trypsin/tryptase inhibitors terminate after the domain carrying the inhibitory effect, TgTI continues with approximately another 250 aminoacids. We believe that the membrane-spanning domain of TgTI might be localised here, since several distinct prediction programs found within the MacVector software and DAS identify membrane-spanning domains towards the carboxyl end of the protein. It is likely, therefore, that the molecular weight difference between the TgTI protein and other trypsin/tryptase inhibitors is at least in part due to membrane spanning regions located towards the carboxyl end of the protein. The localisation and the suggested function of this molecule raise several interesting hypothesis about its function in vivo. Its role could be to participate as a mechanism of defence when the parasites are released in the upper small intestine, either as bradyzoites or tachyzoites, where they will be attacked by proteases, specially trypsin and chymotrypsin. Alternatively, it might be important for survival when extracellular parasites are provoking an inflammatory response in the tissues. Here, another family of trypsin-like serine proteases, the tryptases, which are secreted from stimulated mast cells, could act as target molecules for TgTI [35]. The low abundance of TgTI EST-sequences identified within both tachyzoites and bradyzoites suggest that the inhibitor protein is expressed in both stages but at

Fig. 7. Trypsin inhibition by the recombinant TgTI protein. (A) The Bz – Ile– Glu– Gly– Arg– 4-NA substrate was incubated together with 17 pmol of trypsin 10 min prior to the addition of different concentrations of soybean inhibitor. The absorbance of released 4-nitroaniline product was measured at 405 nm every 10 min. Trypsin 17 pmol, solid line; soybean inhibitor 10 pmol (STi), ----; soybean inhibitor 1 pmol, ….; soybean inhibitor 0.1 pmol, -..-..; (B) The Bz –Ile –Glu– Gly– Arg–4-NA substrate was incubated together with 17 pmol of trypsin 10 min prior to the addition of different concentrations of recombinant TgTI. Trypsin 17 pmol, solid line; TgTI 1 pmol, ----; TgTI 0.5 pmol, ….; TgTI 0.1 pmol, -..-.. Y-axes show absorbance at 405 nm, X-axes show time of incubation in minutes.

144

J.G. Lindh et al. / Molecular & Biochemical Parasitology 116 (2001) 137–145

a low abundance compared to other known surface molecules [36–38]. Recently, another proposed inhibitor of serine proteases has been identified in T. gondii, TgPI [39]. In contrast to TgTI, TgPI seems to be secreted out from the parasite, although its original cellular location is not known. There is no sequence similarity between TgTI and TgPI and taken together with the above, this make us believe that they might have similar functions, but act independently. This is the first description of a surface associated trypsin inhibitor in a protozoan, and could probably be one of the multiple ways this parasite has developed in order to survive the harsh environments of the vertebrate host.

Acknowledgements We thank the AIDS Research and Reference Reagent Program, NIAID, NIH, USA, for the gift of the tachyzoite specific lZAPII cDNA-library; L.D. Sibley, Washington University, St. Louis, MO, USA, and D. Soldati, Universita¨ t Heidelberg, Heidelberg, Germany for parasite strains and E. Petersen, Laboratory of Parasitology, Statens Serum Institut, Copenhagen, Denmark, for the anti SAG1 monoclonal antibody. This study was supported by the Swedish Medical Society and ‘Stiftelsen Clas Groschinskys Minnesfond’.

References [1] Dubey JP, editor. Parasitic Protozoa. Toxoplasma, Neuspora, Sarcosystis and Other Tissue Cyst-Forming Coccidia of Humans and Animals, vol. 6. New York: Academic Press, 1993:1 – 158. [2] Aikawa M, Sterling CR. Intracellular Parasitic Protozoa. New York: Academic Press, 1974. [3] Boothroyd JC, Kim K, Sibley D, Soldati D. Toxoplasma as a paradigm for the use of genetics in the study of protozoan parasites. In: Boothroyd JC, Komuniecki R, editors. MBL Lectures in Biology, vol. 12. New York: Wiley-Liss, Inc., 1995:209 – 25. [4] Dubey JP. Toxoplasma, Neospora, Sarcocystis, and other tissue cyst-forming coccidia of humans and animals. In: Kreier JP, editor. Parasitic Protozoa. San Diego: Academic Press, Inc., 1993:1 – 57. [5] Czapinska H, Otlewski J. Structural and energetic determinants of the S1-site specificity in serine proteases. Eur J Biochem 1999;260:571 – 95. [6] http://www.cbil.upenn.edu/ParaDBs/Toxoplasma/toxodb.html. [7] Ajioka JW, Boothroyd JC, Brunk BP, Hehl A, Hillier L, Manger ID, Marra M, Overton GC, Roos DS, Wan KL, Waterston R, Sibley LD. Gene discovery by EST sequencing in Toxoplasma gondii reveals sequences restricted to the Apicomplexa. Genome Res 1998;8:18 – 28. [8] Ajioka JW. Toxoplasma gondii: ESTs and gene discovery. Int J Parasitol 1998;28:1025 –31. [9] Wan KL, Blackwell JM, Ajioka JW. Toxoplasma gondii expressed sequence tags: insight into tachyzoite gene expression. Mol Biochem Parasitol 1996;75:179 –86.

[10] von Heijne G. Signal sequences the limits of variation. J Mol Biol 1985;184:99 – 105. [11] von Heijne G. A new method for predicting signal sequence cleavage sites. Nucleic Acid Res 1986;14:4683 – 90. [12] Seeber F, Boothroyd JC. Escherichia coli b-galactosidase as an in vitro and in vivo reporter enzyme and stable transfection marker in the intracellular protozoan parasite Toxoplasma gondii. Gene 1996;169:39 – 45. [13] Sambrook J, Fritsch EF, Maniatis T. Molecular Cloning: A laboratory Manual, 2nd. New York: Cold Spring Harbour Laboratory Press, 1989. [14] Chomczynski P, Sacchi N. Single-step method of RNA isolation by acid guanidinium thiocyanate – phenol – chloroform extraction. Anal Biochem 1987;162:156 – 9. [15] Church GM, Gilbert W. Genomic sequencing. Proc Natl Acad Sci USA 1984;81:1991 – 5. [16] Chen Q, Barragan A, Fernandez V, Sundstrom A, Schlichtherle M, Sahlen A, Carlson J, Datta S, Wahlgren M. Identification of Plasmodium falciparum erythrocyte membrane protein (PfEMP1) as the rosetting ligand of the malaria parasite P. falciparum. J Exp Med 1998;187:15 – 23. [17] Harlow E, Lane D. Antibodies: A Laboratory Manual. New York: Cold Spring Harbour Laboratory Press, 1988. [18] Nagel SD, Botthroyd JC. The major surface antigen, P30, of Toxoplasma gondii is anchored by a glycolipid. J Biol Chem 1989;264:5569 – 74. [19] Tomavo S, Dubremetz JF, Schwarz RT. Structural analysis of glycosyl-phosphatidylinositol membrane anchor of the Toxoplasma gondii tachyzoite surface glycoprotein gp23. Biol Cell 1993;78:155 – 62. [20] Cornelissen AWCA, Overdulve JP, Hoenderboom JM. Separation of Isopora (Toxoplasma) gondii cysts and cystozoites from mouse brain tissue by continuous density-gradient centrifugation. Parasitology 1981;83:103 – 8. [21] Fernandez V, Treutiger CJ, Nash GB, Wahlgren M. Multiple adhesive phenotypes linked to rosetting binding of erythrocytes in Plasmodium falciparum malaria. Infect Immun 1998;66:2969 – 75. [22] McFadden DC, Seeber F, Boothroyd JC. Use of Toxoplasma gondii expressing b-galactosidase for colorimetric assessment of drug activity in vitro. Antimicrob Agents Chemother 1997;41:1849 – 53. [23] Bergmeyer HU, Bergmeyer J, Grassl M. Enzymes 3: peptidases, proteinases and their inhibitors. In: Geiger R, Fritz H, editors. Methods of Enzymatic Analysis, 3rd. Weinheim, Deerfield Beach, Florida, Basel: Verlag chemie, 1984:119 – 29. [24] Sharma SP, Dubey JP. Quantitative survival of Toxoplasma gondii tachyzoites and bradyzoites in pepsin and trypsin solutions. Am J Vet Res 1981;42:128 – 30. [25] Barrett JD, Eggena P. Inactivation of renin substrate by soybean trypsin inhibitors: implications for measurement of circulating inactive renin. Clin Exp Hypertension 1993;15(5):761 –80 New York. [26] Metheny NAS, Smith L, Yan H, Diebold M, Clouse RE. pH and concentrations of pepsin and trypsin in feeding tube aspirates as predictors of tube placement. J Parenter Enteral Nutr 1997;21(5):279 – 85. [27] Seeber F. Consensus sequence of translational initiation sites from Toxoplasma gondii genes. Parasitol Res 1997;83(3):309 –11. [28] Cserzo M, Wallin E, Simon I, von Heijne G, Elofsson A. Prediction of transmembrane alpha-helices in procariotic membrane proteins: the Dense Alignment Surface method, Prot Eng 1997;10(6):673 – 676. [29] Burg JL, Perelman D, Kasper LH, Ware PL, Boothroyd JC. Molecular analysis of the gene encoding the major surface antigen of Toxoplasma gondii. J Immun 1988;141:3584 –91.

J.G. Lindh et al. / Molecular & Biochemical Parasitology 116 (2001) 137–145 [30] Kasper LH, Crabb JH, Pfefferkorn ER. Purification of a major membrane protein of Toxoplasma gondii by immunoabsorption with a monoclonal antibody. J Immun 1983;130(5):2407 – 12. [31] Di Marco S, Priestle JP. Structure of the complex of leechderived tryptase inhibitor(LDTI) with trypsin and modeling of the LDTI-tryptase system. Structure 1997;5(11):1464 –74. [32] Sommerhoff CP, Sollner C, Mentele R, Piechottka GP, Auerswald EA, Fritz H. A Kazal-type inhibitor of human mast cell tryptase: isolatin from the medical leech Hirudo medicinalis characterization, and sequence analysis. J Biol Chem 1994;375:685 – 94. [33] Stubbs MT, Morenweiser R, Sturzebecher J, Bauer M, Bode W, Huber R, Piechottka GP, Matschiners G, Sommerhoff CP, Fritz H, Auerswald EA. The three-dimensional structure of recombinant leech-derived tryptase inhibitor in complex with trypsin. J Biol Chem 1997;272:19 931 –9 937. [34] Auerswald EA, Morenweisser R, Sommerhoff CP, Piechottka GP, Eckerskorn C, Gurtler LG, Fritz H. Recombinant leechderived tryptase inhibitor: construction, production, protein chemical characterization and inhibition of HIV-1 replication. J Biol Chem 1994;375:695 –703.

.

145

[35] Henderson WRJ, Chi EY. The importance of leukotrienes in mast cell-mediated Toxoplasma gondii cytotoxicity. J Infect Dis 1998;177(5):1437 – 43. [36] Boothroyd JC, Hehl A, Knoll LJ, Manger IA. The surface of Toxoplasma: more and less. Internat J Parasitol 1997;28:3 – 9. [37] Prince JB, Auer KL, Huskinson J, Parmley SF, Araujo FG, Remington JS. Cloning expression and cDNA sequence of surface antigen p22 from Toxoplasma gondii. Mol Biochem Parasitol 1990;43:97 – 106. [38] Cesbron-Delauw MF, Tomavo S, Beauchamps P, Fourmaux MP, Camus D, Capron A, Dubremetz JF. Similarities between the primary structures of two distinct major surface proteins of Toxoplasma gondii. J Biol Chem 1994;269:16 217 – 6 222. [39] Pszenny V, Angel SO, Duschak VG, Paulino M, Ledesma B, Yabo MI, Guarnera E, Ruiz AM, Bontempi EJ. Molecular cloning, sequencing and expression of a serine proteinase inhibitor gene from Toxoplasma gondii. Mol Biochem Parasitol 2000;15(107):241 – 9.