Conservation of Epidermal Growth Factor Receptor Function in the Human Parasitic Helminth Schistosoma mansoni*

The epidermal growth factor receptor (EGF-R) plays an important role in development and cell differentiation, and homologues of EGF-R have been identified in a broad range of vertebrate and invertebrate organisms. This work concerns the functional characterization of SER, the EGF-R-like molecule previously identified in the helminth parasite Schistosoma mansoni (Shoemaker, C. B., Ramachandran, H., Landa, A., dos Reis, M. G., and Stein, L. D. (1992) Mol. Biochem. Parasitol. 53, 17–32). Transactivation assays performed in epithelial Madin-Darby canine kidney cells co-transfected with SER and a Ras-responsive reporter vector indicated that SER was able to trigger a Ras/ERK pathway in response to human epidermal growth factor (EGF). These results were confirmed in Xenopus oocytes showing that human EGF induced meiosis reinitiation characterized by germinal vesicle breakdown in SER-expressing oocytes. Germinal vesicle breakdown induced by EGF was dependent on receptor kinase activity and shown to be associated with phosphorylation of SER and of downstream ERK proteins. 125I-EGF binding experiments performed on SER-expressing oocytes revealed high affinity (2.9 × 10–9 m) of the schistosome receptor for human EGF. Phosphorylation of the native SER protein present in S. mansoni membranes was also shown to occur upon binding of human EGF. These data demonstrate the ability of the SER schistosome receptor to be activated by vertebrate EGF ligands as well as to activate the classical ERK pathway downstream, indicating the conservation of EGF-R function in S. mansoni. Moreover, human EGF was shown to increase protein and DNA synthesis as well as protein phosphorylation in parasites, supporting the hypothesis that host EGF could regulate schistosome development. The possible role of SER as a receptor for host EGF peptides and its implication in host-parasite signaling and parasite development are discussed.

The epidermal growth factor receptor (EGF-R) 1 is a major key mediator of cell communication during animal develop-ment and homeostasis. EGF-R was the first receptor tyrosine kinase to be cloned (1), and its structure and activation pathways have been studied extensively. EGF-R represents the archetype of receptor tyrosine kinase with an extracellular ligand-binding part with two cysteine-rich repeats and an intracellular domain containing tyrosine kinase activity (2). In mammals, four isoforms of EGF-R have been characterized (EGF-R/ErbB-1, HER2/ErbB-2, HER3/ErbB-3, and HER4/ ErbB-4), and a number of different ligands, including epidermal growth factor (EGF)-like molecules, can selectively bind each isoform (3). Ligand binding activates the receptor by inducing the formation of homo-heterodimers. Dimerization triggers trans-phosphorylation and subsequent autophosphorylation of receptor molecules on tyrosine residues that provide docking sites for diverse effector and adaptor proteins. These partners (Grb2/Sos, p85-PI3K, PLC␥, and JAK) are active in different signal transduction cascades, such as the mitogenactivated protein kinase (MAPK), phosphoinositol 3-kinase, antiapoptotic kinase Akt, and several transcriptional regulatory pathways (reviewed in Ref. 4). Different homodimer-heterodimer combinations formed by EGF-R family members drive a complex signaling network within the MAPK pathway. The ERK pathway is the most recurrent and is mainly responsible for the mitogenic action of EGF receptors. Dysregulation of EGF-R signaling is therefore strongly oncogenic, and the direct implication of EGF-R isoforms in various cancers has been widely demonstrated. For this reason, EGF-R currently represents one of the major drug targets in human cancer therapy (5).
In invertebrates, EGF-R isoforms appeared to be expressed in more limited numbers. A single isoform has been characterized in Caenorhabditis elegans (LET-23) (6) as well as in Drosophila melanogaster (DER) (7,8). A single cognate ligand (LIN-3) would be present in the worm (9), and four distinct cognate ligands (Vein, Gurken, Spitz, and Argos) would be present in the fly (10). These observations indicated that the EGF-R signaling module has grown in complexity from invertebrates to mammals. However, except for C. elegans and D. melanogaster models, few data are available at present about the role of the EGF-R family in invertebrate development.
SER, the Schistosoma mansoni EGF-R homologue, is one of the three receptor tyrosine kinases that have been characterized in this trematode parasite (11,12). SER is present predominantly in schistosome muscles, suggesting that it could participate in muscle development and functions (13). Similar to other EGF-R family members, the intracellular tyrosine kinase domain is the most highly conserved region, whereas the extracellular domain is poorly conserved. Previous work has shown that recombinant SER expressed in insect cells possessed kinase activity and was capable of autophosphorylation, but no evidence was given regarding its function in the parasite (13). In this study, we have undertaken a functional study of SER to analyze the signaling pathways triggered upon its activation and its potential role as a receptor for host ligands. The expression of the SER recombinant protein has been examined in two different vertebrate cellular systems (epithelial MDCK cells and Xenopus oocytes) to analyze the affinity of the parasite receptor for human EGF and its ability to phosphorylate and activate an ERK-dependent signaling pathway. SER analysis was extended to the different schistosome life stages, and the effect of human EGF on parasite metabolism was studied in vitro on sporocysts, the only proliferative stage that develops in the mollusc intermediate host.
The possible role of SER as a receptor for host EGF peptides and its implication in host-parasite signaling and parasite development are discussed.

EXPERIMENTAL PROCEDURES
Parasites-A Puerto Rican strain of S. mansoni was maintained by passage through albino Biomphalaria glabrata snails and Mesocricetus auratus hamsters. Miracidia and cercariae were prepared as described previously (14). Miracidia were transformed in vitro into sporocysts by incubation in minimum salt medium at 28°C for 18 h (15). Adult schistosomes were collected by portal perfusion from infected hamsters.
Cloning of SER-Total RNA was isolated from S. mansoni adult worms by the method of Chirgwin et al. (16) and purified by centrifugation through a cesium chloride gradient. The first-strand cDNA was synthesized using the Thermoscript TM RT-PCR System (Invitrogen) and the oligo(dT) 20 primer. SERfwd (5Ј-AAAAGCTTTAAATGGATTTT-TTACTACCTACTCTG-3Ј) and SERrev (5Ј-GCGGCCGCATAATTCCA-GCGAGGATGACTCATTTTGG-3Ј) complementary to the N-and Cterminal sequences of the SER cDNA (GenBank TM accession no. M86396) were used as primers in RT-PCR to amplify the schistosome cDNA. PCR was carried out with Taq Platinum HiFi polymerase (Invitrogen). Final products were purified, cloned into TOPO 4.0 TA cloning vector (Invitrogen), and then inserted in-frame into the pcDNA3.1-V5/His expression vector (Invitrogen). SER-pcDNA3.1 constructs were sequenced using the Big-Dye Terminator cycle sequencing kit and analyzed on an ABI Prism 377 DNA sequencer (PerkinElmer Life Sciences). Dead kinase SER (SER TKϪ ) constructs were obtained by site-directed mutagenesis of the active D 1202 FG 1204 motif into a DNA inactive motif using the QuikChange® Site-Directed Mutagenesis Kit (Stratagene). The 5Ј-CATGTACAGATTACTGATaaTGcTTAGCTAAAA-TGTTAG-3Ј mutated sequence and its reverse complement were used as primers (mutated residues are lowercase italic).
Transactivation Assays in MDCK Cells-Transactivation assays were performed as described previously, with minor modifications (17). Briefly, 2 ϫ 10 4 MDCK cells were cultured in 12-well plates and transiently transfected using a lipofection method. Cells were rinsed and incubated in 500 l of OptiMEM (Invitrogen) with 1 g of SER-pcDNA3.1 or empty pcDNA3.1 plasmid DNA (as a control), 0.25 g of Py-Luc reporter vector (18), and 10 g of LipofectAMINE. Twentyfour hours later, cells were stimulated with increasing concentrations (up to 30 nM) of recombinant human EGF. After 24 h of incubation, cells were disrupted in lysis buffer (Promega) and centrifuged. Cell supernatants were processed in triplicates for luciferase assays. Results are expressed in relative luciferase units, considering the value obtained in control cells without EGF as a unit. Three experiments were performed with independent plasmid preparations to assess reproducibility.
cRNA in Vitro Synthesis-Capped mRNA (cRNA) was synthesized using the T7 mMessage mMachine Kit (Ambion). SER and SER TKϪ -pcDNA3.1 plasmids were linearized by the enzyme PmeI, and pOBER plasmid (19) containing the human EGF-R (HER) was digested by NotI. cRNAs transcribed from 1 g of each linearized plasmid were precipitated by 2.5 M LiCl, washed in 70% ethanol, resuspended in 20 l of diethyl pyrocarbonate-treated water, and then quantified by spectrophotometry. Finally, 1 g of cRNA was analyzed on a denaturating agarose gel. Gel staining with 10 g ml Ϫ1 ethidium bromide allowed confirmation of the size of cRNA and verification of the absence of abortive transcripts.
Expression of SER in Xenopus Oocytes-After anesthesia with MS 222 (1g l Ϫ1 ; Sandoz), Xenopus laevis ovarian fragments were surgically removed and placed in ND96 medium (96 mM NaCl, 2 mM KCl, 1 mM MgCl 2 , 1.8 mM CaCl 2 , and 5 mM Hepes adjusted to pH 7.4 with NaOH) supplemented with streptomycin/penicillin (50 g ml Ϫ1 each; Eurobio), sodium pyruvate (225 g ml Ϫ1 ; Sigma), and soybean trypsin inhibitor (30 g ml Ϫ1 ; Sigma). Stage VI oocytes were harvested after a 1-h treatment with 1 mg ml Ϫ1 collagenase A (Roche Applied Science). Complete defolliculation of oocytes was achieved by manual dissection. Oocytes were kept at 19°C in ND96 medium before microinjection in the equatorial region. In each assay, 20 -30 oocytes removed from at least two different animals were used. 60 nl (equivalent to 60 ng) of SER, SER TKϪ , and HER cRNA were injected in each oocyte 48 h before stimulation by recombinant human EGF (R&D Systems). GVBD was detected by the appearance of a white spot at the center of the animal pole after 15 h of human EGF treatment. Student's t test was used to assess the significance of the observed differences.
Immunoprecipitation and Western Blot Analyses of SER Expressed in Xenopus Oocytes-Expression of SER in oocytes was confirmed by immunoprecipitation of membrane extracts from [ 35 S]methionine-labeled oocytes according to the procedure described previously (20). Oocytes were injected with SER or SER TKϪ , incubated in ND96 medium containing [ 35 S]methionine (3000 Ci mmol Ϫ1 , 0.5 mCi ml Ϫ1 ) for 48 h, lysed in buffer A (50 mM Hepes, pH 7.4, 500 mM NaCl, 0.05% SDS, 5 mM MgCl 2 , 1 mg/ml bovine serum albumin, 10 g ml Ϫ1 leupeptin, 10 g ml Ϫ1 aprotinin, 10 g ml Ϫ1 soybean trypsin inhibitor, 10 g ml Ϫ1 benzamidine, 1 mM phenylmethylsulfonyl fluoride, and 1 mM sodium vanadate), and centrifuged at 4°C for 15 min at 10,000 ϫ g. Membrane pellets were resuspended and incubated for 15 min at 4°C in buffer A containing 1% Triton X-100 and then centrifuged under the same conditions. Supernatants were incubated with anti-V5 antibodies (1:100; Invitrogen) overnight at 4°C. Protein A-Sepharose beads (5 mg; Amersham Biosciences) were added for 1 h at 4°C. Immune complexes were collected by centrifugation, rinsed three times, resuspended in Laemmli sample buffer, and subjected to a 7.5% SDS-PAGE. Gels were dried and autoradiographed. The detection of tyrosine-phosphorylated SER proteins was performed in oocytes expressing SER or SER TKϪ for 48 h and stimulated or not for 5 min by EGF (49.8 nM, final concentration). Briefly, 30 oocytes were lysed in 300 l of buffer A, and membrane extracts were immunoprecipitated by anti-V5 antibodies under the same conditions as described above. Immune complexes were analyzed by Western blotting using anti-V5 (1:50,000) and PY20 (1:10,000; BD Biosciences) antibodies and the ECL advance detection kit (Amersham Biosciences).
The detection of ERK proteins in oocyte homogenates was performed by Western blotting using anti-ERK2 mouse antibodies (Santa Cruz Biotechnology) as described previously (21).
EGF Binding Analysis-Oocytes were injected with 60 nl of SER cRNA (equivalent to 60 ng) or with 60 nl of water 48 h before EGF binding assays. Sets of 10 oocytes were placed in 24-well plates and preincubated in ND96 medium and 1% bovine serum albumin for 1 h at 0°C. A 10 Ϫ7 M human EGF solution containing 125 I-labeled human EGF (250 Ci ml Ϫ1 ; specific activity, 1266 Ci mmol Ϫ1 ; Amersham Biosciences) was prepared, and serial dilutions of the radioactive mixture were made in ND96 medium/bovine serum albumin. Oocyte groups were incubated for 4 h (until equilibrium) at 0°C in a total volume of 200 l in the presence of 10 Ϫ8 , 6.6 ϫ 10 Ϫ9 , 3.3 ϫ 10 Ϫ9 , 10 Ϫ9 , and 5 ϫ 10 Ϫ10 m EGF final concentrations. At the end of the incubation, aliquots of media were removed and used to determine the amount of free radioactivity in each well. Oocytes were washed three times in cold ND96 medium/bovine serum albumin buffer, and the radioactivity bound to oocytes was determined. Nonspecific binding to water-injected oocytes represented Ͻ10% of total binding and was deducted from the values of bound radioactivity. Scatchard analysis was performed using EnzFitter software (Biosoft). The experiment was repeated twice using oocytes removed from two different animals. In each experiment, the efficacy of oocyte transfection and SER expression was confirmed after GVBD induction upon EGF stimulation.
Detection of Phosphorylated SER Protein in Schistosome Membrane Extracts-S. mansoni adult worms freshly perfused from infected hamsters were homogeneized in PY buffer (20 mM Tris-HCl, pH 7.4, 50 mM NaCl, 5 mM EDTA, 1% aprotinin, 1 mM phenylmethylsulfonyl fluoride, 1 M leupeptin, 1 mM sodium orthovanadate, and 20 mM ␤-glycerophosphate) and then centrifuged at 4°C for 5 min at 15,000 ϫ g. Membrane pellets equivalent to 100 parasites were resuspended in 1 ml of PY buffer with or without 1 M EGF and incubated at 37°C for 10 min.
After centrifugation, membrane pellets were homogeneized in 100 l of PY supplemented with 1% Triton X-100 and incubated for 30 min at 0°C. Membrane lysates were complemented with 4 volumes of kinase buffer (50 mM Hepes, pH 7.5, 150 mM Nacl, 12.5 mM MgCl 2, 1 mM dithiothreitol, 1 mM sodium orthovanadate, and 1 mM phenylmethylsulfonyl fluoride) containing 50 M ATP and incubated for 30 min at 30°C before immunoprecipitation by purified rabbit anti-SER antibodies directed against the ligand binding domain (residues 357-517) of the receptor (1 g ml Ϫ1 ; kindly provided by P. Skelly; Ref. 13) and protein A-Sepharose beads. Immune complexes were analyzed in 7.5% SDS-PAGE gel and revealed by Western blotting using anti-PY20 (1:10,000; BD Biosciences).
Quantitative RT-PCR-Total RNA was extracted from miracidia, sporocysts, cercariae, and adult male and female worms using the technique of Chirgwin et al. (16) and then reverse-transcribed using the Thermoscript TM RT-PCR System (Invitrogen). cDNAs were used as templates for PCR amplification using the SYBR ® Green PCR Master Mix and the ABI PRISM 7000 sequence detection system (Applied Biosystems). Primers specific for S. mansoni tubulin (GenBank TM accession no. 80214, positions 851-873 and 925-904) and SER (Gen-Bank TM accession no. M86396, positions 1002-1022 and 1102-1081) were designed by the Primer Express Program (Applied Biosystems) and used for amplification in triplicate assays. For graphical representation of quantitative PCR data, raw cycle threshold values (Ct values) obtained for male worms, female worms, miracidia, and cercariae were deducted from the Ct value obtained for sporocyst transcript levels using the ⌬⌬Ct method (22), with tubulin gene levels serving as the internal standard.
Sporocyst Labeling-200 transformed sporocysts were incubated for 24 h in 96-well plates in minimum salt medium at a concentration of 1000 sporocysts ml Ϫ1 in the presence of radioactive precursors (2 Ci of L-[ 35 S]methionine or 1 Ci of [ 3 H]thymidine (Amersham Biosciences)) and different doses (10 Ϫ6 to 10 Ϫ9 M, final concentration) of human EGF (R&D Systems). After incubation, parasites were collected, and 35 S radioactivity was measured in the trichloroacetic acid-precipitated material using a scintillation liquid (Beckman) and a Packard scintillation counter. [ 3 H]Thymidine incorporation was determined after filtration onto GF/C filters (Packard) using a cell harvester (Packard) and a Packard microplate scintillation counter. 32 P labeling was performed under similar conditions with the addition of 2 Ci of [␥-32 P]ATP (3000 Ci mmol Ϫ1 ). Labeling was stopped after 5 min of incubation by the addition of 0.2 volume of 10% SDS and boiling for 2 min. Proteins were precipitated with 5% trichloroacetic acid, and radioactivity was determined.

Human EGF Activates SER in MDCK Epithelial Cells-The
full-length SER sequence (GenBank TM accession no. M86396) was amplified by RT-PCR from adult schistosome RNA, cloned in-frame into the pcDNA3.1-V5/His expression vector and expressed in MDCK epithelial cells. In vitro kinase assays on immune complexes formed after precipitation of cell extracts with anti-V5 antibodies confirmed that the recombinant protein was an active kinase, able to phosphorylate MBP (bovine myelin basic protein) substrate and to autophosphorylate, as already shown by Ramachandran et al. (13) (data not shown).
Transactivation assays were further performed in SER-expressing MDCK cells co-transfected with the Py-Luc reporter vector. This vector contains Ras-responsive elements with three tandem copies of a polyoma virus enhancer-derived sequence with ETS/AP1 binding sites linked to the thymidine kinase promoter and the luciferase reporter gene (18). This co-transfection system has been demonstrated to be a sensitive assay to analyze functional activity of receptor tyrosine kinase (17). Results in Fig. 1 indicated that in SER-transfected cells, human EGF is able to increase the Ras-dependent transcriptional response in a dose-dependent manner, with maximum luciferase activity detected with EGF concentrations of Ն10 Ϫ9 M. In control cells transfected with the empty vector, a basal level of luciferase activity was detected, and its moderate increase in the presence of EGF was attributed to activation of the endogeneous canine EGF-R. These results already suggested that SER expressed in mammalian cells was able to trigger a Ras pathway in response to EGF, but the level of protein expression in MDCK cells was too low to pursue the analysis of receptor activation and phosphorylation in this expression system.
EGF Induces GVBD in SER-expressing Oocytes-Xenopus oocytes are frequently used to express proteins, and they have already been shown to express with efficacy S. mansoni proteins (23). Therefore, we considered these cells as an alternative model for the expression of the SER receptor. Results presented in Fig. 2 demonstrated that translation of SER cRNA, in vitro-transcribed from SER-containing plasmids, was efficient in oocytes. After 48 h of incubation, the presence of the [ 35 S]methionine-labeled SER protein was detected in oocyte membranes. The labeled band of 215 kDa exhibited an apparent molecular mass larger than that expected for the recombinant protein (192 kDa), which could be attributable to glycosylation, a processing required for translocation of the receptor to the cell surface and subsequent acquisition of function (24).
Besides their use in examining protein production, oocytes have been shown to represent a suitable model for investigating receptor tyrosine kinase activation and transduction cascades. They are physiologically arrested at the G 2 -M transition of the first meiotic prophase, and this blockade can be abolished upon activation of receptor tyrosine kinase-dependent pathways. Ras activation triggers the MAPK cascade and causes phosphorylation of the ERK2/MAPK, which is necessary and sufficient to induce meiosis reinitiation characterized by oocyte GVBD (25). The results presented in Fig 3A indicate that human EGF induced GVBD in a high percentage of oocytes expressing the human EGF-R (HER). GVBD resulted essentially from HER activation because no GVBD was detected in noninjected control oocytes or in HER-expressing oocytes in the absence of EGF. When similar experiments were performed with SER-expressing oocytes, we observed that human EGF was also able to induce GVBD in such oocytes, at levels comparable with that found for HER-expressing oocytes. The percentage of GVBD induced by 16.6 nM EGF in HER oocytes (61.6 Ϯ 35.4%) was not significantly different from that found in SER oocytes (40.8 Ϯ 27.5%) at the same ligand concentration. In these experiments, we could demonstrate that SER kinase activity was directly implicated in GVBD induction by analyzing oocytes injected with a SER sequence mutated in its thymidine kinase domain (SER TKϪ ). Results in Fig. 3A show that a single mutation in the Mg 2ϩ -binding site of the receptor kinase domain completely abolished GVBD in SER TKϪ -expressing oocytes. Results of immunoprecipitation confirmed that the mutated SER protein was expressed at a similar level as the native receptor (Fig. 2).
SER Activates ERK Phosphorylation in Oocytes-To determine the nature of the signaling pathway triggered by SER, we first demonstrated that the induction of GVBD in oocytes required receptor autophosphorylation. Results in Fig. 3B confirm the recognition of SER by anti-phosphotyrosine antibodies only when oocytes expressed the active receptor and were activated by EGF. Further analyses have concerned the effect of EGF on the phosphorylation of ERK proteins. Results in Fig.  3C indicate a shift of phosphorylated ERK2 proteins in SER-and HER-expressing oocytes after the addition of EGF. ERK phosphorylation appeared to be dependent on SER kinase activity because the dead mutation in SER TKϪ was sufficient to totally inhibit the shift of ERK proteins in EGF-treated oocytes.
Determination of the Affinity of SER for Human EGF-Results from experiments of transactivation in MDCK cells and maturation of oocytes already suggested that SER could bind EGF from vertebrates with high affinity. 125 I-EGF binding experiments were further performed on SER-expressing oocytes to determine the affinity of the schistosome receptor for human EGF. As shown in Fig. 4, Scatchard analysis generated linear plots and provided a K d value of 2.9 ϫ 10 Ϫ9 M (inset) for apparent affinity of SER for human EGF. This calculated equilibrium affinity was similar to that found by Wiley et al. (26) for the human fibroblast receptor (2.2 ϫ 10 Ϫ9 M).
EGF-dependent Phosphorylation of the Native SER Protein in Schistosome Membranes-Previous work has demonstrated the expression of the SER protein in adult schistosomes and its predominant localization in parasite muscular membranes (13). To confirm the ability of the parasite receptor to bind and be activated by human EGF, schistosome membrane preparations were incubated with EGF and subjected to kinase assay conditions before analysis by SDS-PAGE. Results in Fig. 5A show that a limited number of proteins in schistosome mem-  ϭ 2). B, receptor and phosphotyrosine Western blot analysis of EGF-activated SER. Oocytes expressing SER or SER TKϪ for 48 h were stimulated (ϩ) or not (Ϫ) by 49.8 nM human EGF for 5 min. SER and SER TKϪ proteins were immunoprecipitated by anti-V5 antibodies, and phosphorylation on tyrosine residues was detected by Western blot analysis using anti-phosphotyrosine PY20 antibodies. Immunoprecipitation of SER and SER TKϪ proteins was confirmed by the use of anti-V5 antibodies. C, Western blot analysis of EGF-induced ERK2 phosphorylation in oocytes. After 48 h of SER TKϪ , SER, or HER expression, oocytes were incubated without (Ϫ) or with (ϩ) 16.6 nM human EGF for 15 h. Oocyte lysates were analyzed by Western blotting using anti-ERK2 antibodies. The respective positions of ERK and phosphorylated ERK (ERK-P) proteins are indicated by arrows.
brane extracts react with anti-phosphotyrosine antibodies but indicate the presence in EGF-treated membranes of a major tyrosine-phosphorylated protein with an apparent molecular mass of 190 kDa, corresponding to the size of the activated EGF-R. Immunoprecipitation of EGF-treated membrane extracts by anti-SER antibodies conclusively identified the 190-kDa protein as SER (Fig. 5B) and confirmed phosphorylation of the native SER protein upon EGF binding. The apparent molecular mass of SER detected in these assays is slightly different from that found by previous authors using the same preparation of antibodies (170 kDa, Ref. 13), and this is probably due to different conditions of protein analysis. The disparity in size of native and recombinant SER (Fig. 2) probably results from differential glycosylation in adult schistosomes and Xenopus oocytes.
Potential Role of EGF in Host-Parasite Interactions-As a preliminary step toward a functional study of SER in schistosomes, we have analyzed the relative expression of SER transcripts in the different S. mansoni parasite stages living in both the definitive and intermediate hosts using the technique of real-time RT-PCR. Results presented in Fig. 6 confirm the high representation of SER in adult schistosomes, particularly in male worms, in which muscular structures are the most abundant. Data also indicate that SER is expressed throughout the parasite life cycle but that lower amounts of transcripts are present in larval stages as compared with adult forms. EGF-R has been demonstrated to be involved in diverse proliferative processes. The observation that SER was also expressed in sporocysts, the only larval stage responsible for asexual reproduction and proliferation of schistosomes in the mollusc host, prompted us to analyze a potential effect of EGF on sporocysts. Experiments shown in Fig. 7 indicate that in vitro incubation of sporocysts with human recombinant EGF stimulated both [ 35 S]methionine and [ 3 H]thymidine incorporation in a dose-dependent manner, with a maximal effect at a 10 Ϫ8 M concentration. Additionally, after a short period (5 min) of incubation with the optimal EGF dose, we observed a 2-fold increase in the level of 32 P incorporation as compared with control sporocysts, demonstrating that human EGF could activate a signaling pathway dependent on phosphorylation in the parasite. Further experiments confirmed that a similar dose of human EGF (10 Ϫ8 M) was also able to stimulate (2-fold) protein synthesis in in vitro-cultured adult schistosomes (data not shown). DISCUSSION Much knowledge has been accumulating in recent years on the importance of EGF signaling in embryonic development and physiology. Progress on the developmental role of this growth factor has come, in large part, from the two inverte- FIG. 4. Scatchard analysis of 125 I-EGF binding to oocytes. Oocytes injected with 60 ng of SER cRNA were brought to equilibrium with 125 I-EGF concentrations ranging from 5 ϫ 10 Ϫ10 to 1 ϫ 10 Ϫ8 M for 4 h at 0°C. Radioactivity bound to oocytes was determined, and Scatchard analysis was performed using EnzFitter software (Biosoft) and provided a K d value of 2.9 ϫ 10 Ϫ9 M (inset) for apparent affinity of SER for human EGF. Each point represents 10 oocytes, and receptor affinity was estimated by linear regression.

FIG. 5. EGF activation of SER in schistosome membranes.
A, Western blot analysis of schistosome membranes incubated without (Ϫ) or with (ϩ) 1 M human EGF and subjected to kinase assays conditions. A major tyrosine-phosphorylated protein of 190 kDa was revealed by PY20 antibodies in EGFtreated membranes. B, extracts from EGF-treated membranes in A were immunoprecipitated by either anti-SER or normal rabbit (NR) antibodies. Western blot analysis of immune complexes with PY20 antibodies identified the 190-kDa phosphorylated protein as SER. brate models, Drosophila (7,8,10) and C. elegans (6,9). The identification of an EGF-like receptor molecule in S. mansoni (SER) (11,13) led us to investigate its role in mitogenic and differentiation processes in the parasite and its possible involvement in host-parasite signaling. In this work, we analyzed the functional activity and ability of this receptor to bind host EGF and addressed the question of EGF signaling in S. mansoni development as well as in host-parasite interactions.
The SER protein shares sequence and structural homologies with other members of the EGF-R family. The tyrosine kinase domain of SER is well conserved and has previously been shown to be catalytically active (13). The recombinant protein, expressed in insect cells, was capable of tyrosine autophosphorylation; however, no information was given regarding its potential to bind and be activated by EGF ligands (13). For this reason, we have undertaken functional characterization of the SER receptor by analyzing its EGF-dependent activity and determining its affinity for human EGF. Because no technique is currently available to transfect schistosomes with efficacy, we have chosen to express the SER protein in two vertebrate cellular systems, MDCK epithelial cells and the oocytes of X. laevis. Transactivation in MDCK epithelial cells is a sensitive assay to analyze the functional activity of receptor tyrosine kinase by measuring their ability to induce a transcriptional response through specific Ras-dependent promoter elements (17). It was recently used to study signaling and biological activities of the MET receptor (27). This transactivation system requires a limited number of surface receptors, and the cotransfection with a luciferase vector containing ETS/AP1 binding sites allows a very sensitive detection of the activation of the Ras/MAPK pathways in the cell. The second system we have used offers other advantages, which are the high efficacy of protein expression and the facility to observe MAPK activation optically by the induction of GVBD in oocytes. Opresko and Wiley (19) have shown that the human receptor HER entirely retained its specific biological activities in the context of oocytes and was able to induce oocyte maturation. The fact that no endogeneous EGF-R was present in Xenopus oocytes was an additional reason to use this expression system for studying SER activity.
Results obtained in MDCK cells indicated that luciferase activity increased in SER-transfected cells as compared with control cells when stimulated by human EGF. The signal intensity was ligand dose-dependent, and a plateau was reached for nanomolar concentrations of EGF in both control and SERexpressing cells. This sensitive system allows the determination of the level of receptor activation and might be appropriate for further analyses of EGF ligand or specific inhibitor effects on SER activation. Such data, demonstrating that SER was able to trigger a Ras/MAPK pathway dependent on EGF, were then confirmed in the second expression system. We showed that EGF specifically stimulated maturation of oocytes expressing SER as a membrane protein, as demonstrated by analysis of labeled oocyte membrane extracts and immunopre- cipitation. In this model, the use of a dead kinase receptor allowed us to demonstrate the importance of SER kinase activity in the observed process. Without affecting the level of receptor expression in membranes, a single mutation in the Mg 2ϩ -binding site (DFG motif) of the SER thymidine kinase domain completely abolished the appearance of GVBD, confirming that receptor kinase activity was essential for oocyte maturation. Taken together, these results indicated that the recombinant parasite receptor tyrosine kinase was functionally active in the context of both epithelial cells and oocytes and strongly suggested that it could interact properly with the protein partners of the vertebrate Ras/ERK pathway. Western blot analysis of oocyte lysates clearly demonstrated that SER activation induces its phosphorylation and effectively triggers a Ras pathway and ERK protein phosphorylation, a process dependent on the existence of a functional kinase domain in the SER protein. At this stage, further analysis of the C-terminal region of SER is necessary to better characterize the activated signaling cascade. However, the absence of significant homology of the SER intracellular region to EGF-R homologues, with the exception of the kinase domain (11), makes such an analysis difficult. Indeed, although SER contains a large number of tyrosine residues in its C-terminal portion that represent potential phosphorylation sites, a precise sequence analysis of this region (residues 1333-1717) confirmed the absence of strictly conserved binding sites for the well-known EGF-R cascade partners (28). Further work is now concerned with tyrosine scanning and site-directed mutagenesis to identify functional sites for SER signaling and the characterization of potential partners able to activate ERK cascade in schistosomes.
Results of SER activation by human EGF already suggested a relatively high affinity of the parasite receptor for the vertebrate hormone. Indeed, human EGF concentrations sufficient to obtain optimal activation of the Ras/ERK pathway in MDCK cells or oocytes expressing SER were very similar to those required for optimal activation of the endogeneous canine receptor in epithelial cells or the human receptor expressed in oocytes. Binding experiments in oocytes confirmed that the calculated equilibrium affinity of SER expressed in oocytes was of the same order of magnitude (10 Ϫ9 M) as that of the human fibroblast receptor (26). To our knowledge, these results constitute the first direct demonstration that a schistosome growth factor receptor is able to bind the corresponding vertebrate ligand. The demonstration that the SER protein immunoprecipitated from EGF-treated membrane extracts was effectively phosphorylated on tyrosine residues also confirmed the ability of the native schistosome protein to bind and be activated by EGF. Previous studies have already shown the ability of SmRK1, a S. mansoni divergent member of the transforming growth factor ␤ receptor family, to activate a conserved pathway in response to human transforming growth factor ␤. However, in these transfection studies, no evidence was given for a direct interaction of transforming growth factor ␤ with SmRK1; a functional recruitment of the type I SmRK1 receptor by the mammalian type II receptor is more likely (29,30), with the mammalian receptor binding the ligand. Together, these data confirm the conservation of signaling pathways in helminths and support the hypothesis that host factors could regulate schistosome development by interaction with specific parasite receptors.
Further experiments were performed to analyze in vitro the effect of human EGF on S. mansoni parasites, considering in priority its influence on the metabolism of sporocysts. In vitrotransformed from miracidia, sporocysts are devoid of contamination by host components and constitute the most suitable stage for analysis of metabolic activities, such as protein and DNA synthesis. The procedures described have been already used with success to show that mollusc host molecules could stimulate proliferative activity in these infective schistosome larvae (31). Results presented in this work demonstrate that human EGF could increase protein and DNA synthesis as well as protein phosphorylation in primary sporocysts. A maximal effect is observed after incubation with 10 Ϫ8 M EGF; higher EGF doses are less efficient, probably because of toxicity. Similar assays performed on adult worms, which are more complex organisms, also indicated that human EGF could increase in vitro methionine incorporation 2-fold. Therefore, it seems that in both larval and adult stages, binding sites for exogenous EGF are accessible on the parasite. The observation that SER was predominantly present in schistosome muscles (13) would not be a priori in agreement with this hypothesis. However, considering the high affinity of SER for the human hormone, we could not exclude the possibility that SER molecules, present in a rather limited number on the surface and therefore not previously detected by immunofluorescence (13), could be sufficiently active to induce host EGF signaling in the context of host-parasite relationships. Our preliminary results showing that EGF was able to stimulate protein synthesis might be in agreement with a positive effect of host EGF on schistosome biology. Also, our results showing the stimulating effect of EGF on sporocysts indicated the possibility that the SER molecule was accessible to host components at this parasite stage and therefore suggested its potential recognition by mollusc EGF ligands and its importance in sporocyst differentiation.
In conclusion, our data demonstrate the conservation of EGF signaling in S. mansoni, i.e. the ability of the schistosome receptor to be activated by vertebrate EGF ligands and to activate a vertebrate classical MAPK pathway downstream, raising the question of the role of co-evolution in host-parasite adaptation processes.
As in the case of a variety of organisms, SER is supposed to play an important role in development and could thus be considered an interesting target for anti-schistosome chemotherapy. Three-dimensional structural analysis of the catalytic domain of SER and comparison with other EGF-R family members would be helpful for the discovery of SER-specific inhibitors that would probably represent new and efficient drugs against the parasite.