- Email: [email protected]
Molecular characterization of P-29, a metacestode-specific component of Echinococcus granulosus which is immunologically related to, but distinct from, antigen 5
Molecular and Biochemical Parasitology 105 (2000) 177 – 184 www.elsevier.com/locate/parasitology
Molecular characterization of P-29, a metacestode-specific component of Echinococcus granulosus which is immunologically related to, but distinct from, antigen 5 Gualberto Gonza´lez a,*, Pablo Spinelli a, Carmen Lorenzo a, Ulf Hellman b, Alberto Nieto a, Antony Willis c, Gustavo Salinas a a Ca´tedra de Inmunologı´a-Facultad de Quı´mica, Monte6ideo, Uruguay Ludwig Institute for Cancer Research, Uppsala Branch, Uppsala, Sweden c MRC Immunochemistry Unit, Department of Biochemistry, Uni6ersity of Oxford, Oxford, UK b
Received 9 June 1999; received in revised form 14 June 1999; accepted 20 August 1999
Abstract In this work the characterization of P-29, a novel 29 kDa antigen from Echinococcus granulosus is reported. E. granulosus was identified while looking for parasite antigens distinct from those present in hydatid cyst fluid. A monoclonal antibody (mAb 47H.PS) prepared against protoscolex components revealed that P-29 is localized to the tegument and rostellum of protoscoleces, and to the germinal layer of the cyst, but it is absent in hydatid cyst fluid or adult worm extracts. Several internal fragments of P-29 showed sequence identity to the amino acid sequence encoded by Eg6, a partial gene sequence reported to code for an epitope of antigen 5 (Ag5), one of the major diagnostic antigens of the parasite. We confirmed that Eg6 encodes a sub-fragment of P-29 by mapping the epitope of mAb 47H.PS, and isolating the full length P-29 cDNA. Since Eg6 had been postulated to encode a fragment of Ag5, we specifically studied the relationship of P-29 and Ag5 by: (i) examining the cross-reactivity displayed by different mAbs; (ii) comparison of their peptide finger prints; and (iii) a comparative study of their diagnostic value. Our results prove unequivocally that P-29 and Ag5 are immunologically related, but different proteins, raising several questions on the current knowledge of Ag5. © 2000 Elsevier Science B.V. All rights reserved. Keywords: E. granulosus; E. multilocularis; Hydatid disease; Diagnosis; Antigen 5
Abbre6iations: SPE, somatic protoscolex extract; Ag5, E. granulosus antigen 5; Eg6, partial cDNA sequence coding for a protein fragment immunologically related to Ag5; Em6, E. multilocularis partial cDNA sequence homologous to Eg6. Note: Nucleotide sequence data reported in this paper is available in the EMBL, GenBank™ and DDJB data bases under the accession number AF078931. * Corresponding author. Tel.: + 598-2-4801196; fax: + 598-2-4874320. E-mail address: [email protected] (G. Gonza´lez) 0166-6851/00/$ - see front matter © 2000 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 6 - 6 8 5 1 ( 9 9 ) 0 0 1 6 6 - 8
178
G. Gonza´lez et al. / Molecular and Biochemical Parasitology 105 (2000) 177–184
1. Introduction Hydatid cyst fluid has been the usual source of Echinococcus granulosus antigens for the immunodiagnosis of hydatid disease, while other parasite structures such as protoscoleces, cyst membranes or oncospheres have received little attention. The major antigens of hydatid cyst fluid are antigen B and antigen 5 (Ag5), and their diagnostic value has been thoroughly studied [1,2]. While there have been important advances in the understanding of the molecular organization of AgB [3–5], the characterization of Ag5 has been only partial. The antigen is identified as such by the formation of a characteristic arc of precipitation in immunoelectrophoresis with sera from infected patients. This arc, known as arc 5, was one of the first tools used in the immunodiagnosis of hydatid disease and gave its name to the antigen [6].Using antibodies obtained from this arc, Lightowler et al. [1] identified the components of Ag5 in SDS-gels. Under non-reducing conditions, Ag5 occurs as a band of about 67 kDa which dissociates, in the presence of reducing agents, into two bands of 38 and 24 kDa, respectively. Another component of about 57 kDa was identified by using monoclonal antibodies (mAbs) [7]. Apart from some N-terminal amino acid sequences published by Zhang and McManus [8], little is known about the primary structure of these components. Using disease-specific sera highly reactive to Ag5, Facon et al. [9] isolated a partial cDNA sequence termed Eg6. The recombinant protein fragment encoded by Eg6, was recognized by the mAb EG 02 154/12 specific for Ag5 [10]. In addition, antibodies eluted from this recombinant protein recognized the 38 kDa subunit of Ag5. Based on this cross-reactivity evidence, the authors suggested that Eg6 codes for an antigenic epitope of Ag5. Supporting their hypothesis, a 34 amino acid synthetic peptide derived from Eg6 was recognized by most of the sera reactive to Ag5 and inhibited the binding of mAb EG 02 154/12 to Ag5 [11]. Likewise, using antibodies eluted from a 38 kDa protein component of Echinococcus multilocularis fertile cysts, Siles et al. [12] recently isolated Em6, a partial cDNA sequence which displays 98% sequence identity with Eg6. On these basis, the authors suggested that Em6 is the
E. multilocularis homologous of Eg6, and concluded that it may represent a fragment of E. multilocularis Ag5. However, in either case, no direct evidence of the true relationship between Eg6 (or Em6) and Ag5 was presented. Here we report the characterization of a novel 29 kDa protein of E. granulosus, termed P-29, which was isolated as a result of our attempts to find alternative antigens distinct from those present in hydatid cyst fluid. The protein is restricted to the metacestode stage of the parasite and appears to be valuable in diagnosis. We also found evidence that Eg6 encodes a fragment of P-29, and therefore analyzed the relationship between P-29 and Ag5. We demonstrated that they are immunologically related, but different proteins. 2. Materials and methods
2.1. Parasite antigens Bovine hydatid cysts were collected from cattle killed at abattoirs in Uruguay. The hydatid fluid and protoscoleces were drained from the cyst, the protoscoleces were decanted and the supernatant was supplemented with 500 mg/l of NaN3, 1 mM EDTA and kept at 4°C until used. Hydatid cyst walls were removed from both fertile and non-fertile cysts. Prepatent adult worms (30 days post-infection) were obtained from infected dogs as described by [13]. Somatic extracts of protoscoleces, membrane layer or adult worms were prepared by sonication. Periodate treatment was performed according to Woodward et al. [14]. P-29 and Ag5 were purified by immunochromatography using the mAbs described below.
2.2. mAbs mAb EG 02 154/12 reactive with Ag5 [10] was a kind gift of Dr M. Chamekh. mAb 47H.PS was obtained as follows: a mouse was primed intraperitoneally with 20 mg of somatic protoscolex extract (SPE) in Freund’s complete adjuvant, and boosted after 3 and 6 weeks with the same antigen using Freund’s incomplete adjuvant. Three days after the last booster the mouse was killed and the splenocytes fused with SP2/0 cells, basically as de-
G. Gonza´lez et al. / Molecular and Biochemical Parasitology 105 (2000) 177–184
scribed in [15]. Cultures producing mAbs reactive with SPE were selected by ELISA.
2.3. Mapping of the mAb 47H.PS epitope This was accomplished by using the ‘PEPSCAN’ approach [16] (Zeneca, UK). Eight-mer-overlapping peptides derived from the sequence of Eg6 were assayed on the synthesis pins with mAb 47H.PS following the instructions of the supplier.
2.4. SDS-PAGE, immunoblot and immunolabelling Electrophoresis and immunoblot were performed essentially as described by Laemmli [17] and Towbin [18], respectively. Immunocytochemical labelling with mAb 47H.PS was performed as described by Jackson and Blythe [19] on 6 mm parasite sections, blocked with 1% normal goat serum and developed with diaminobenzidine-H/hydrogen peroxide.
2.5. Amino acid sequencing and fingerprinting of P-29 and Ag5 Fragmentation of immunopurified P-29 or Ag5 was performed in solution with trypsin as described in reference [20] or by in situ digestion in SDS-gels according to Hellman et al. [21]. Briefly, the band was excised from the SDS-gel, washed and dried under nitrogen. This material was then rehydrated, during which process Lys-C (Boeringer-Manheim, Germany) was added. After overnight incubation at 30°C, and several extraction steps with 0.1% trifluoracetic acid, 60% acetonitrile, the supernatant was run on a mRPC C2/C18 SC 2.1/10 column, operated in SMART System (Pharmacia, Sweden). Elution was performed by a linear gradient from 0 – 40% acetonitrile in 0.065% trifluoroacetic acid. Selected fractions were sequenced in an Applied Biosystems Protein Sequencer, Model 470a (USA).
2.6. Isolation of the full length cDNA encoding P-29 Eg6 sense (5’-GAGGTGCGAAAAGACGAAAGTGAC-3’) and antisense (5’-GCAAGTCTTCTCAAAGATAGTAAGAG-3’) oligonucle-
179
otides derived from the cDNA sequence of Eg6 [9] were used in combination with lambda gt11 reverse and forward, respectively, to amplify the 5’-and 3’ends of Eg6 by PCR [22] from an E. granulosus protoscolex cDNA library. The PCR products obtained were isolated (GeneClean, BIO101) and cloned into pGEMTM (Promega) according to standard techniques [23]. The 5’- and 3’-end cDNA clones were then sequenced using AmpliTaqTM FS according to standard protocols on an ABI Prism 377 DNA sequencer (Sequi-net, CO). In order to obtain the 5’-end of the cDNA, 5’ RACE (rapid amplification of cDNA ends) was carried out using a commercial kit (Gibco). Briefly, reverse transcription was performed from TRIzol (Gibco) prepared total protoscolex RNA using a P-29 antisense primer (GAGCAGTACCCAGTTTGTCGGTA), the cDNA tailed using the terminal deoxynucleotidyl transferase (TdT) and subsequently amplified by PCR with a second P-29 antisense primer (GGTCGATCTCATGGTAAAG) and the abridged anchor primer provided by the kit. The PCR product thus obtained was cleaned up (Promega) and directly sequenced as above.
2.7. Human serum specimens The following serum samples were used in this study: Thirty nine sera from surgically confirmed cystic hydatid patients, taken before surgery (the samples were not preselected on the basis of previous serologic information); 19 sera from healthy donors and 51 serum samples from patients with the following diseases: alveolar hydatid disease (17), Taenia solium cysticercosis (24), schistosomiasis (five), lymphatic filariasis (three), toxocariasis (two).
2.8. ELISA ELISA was performed as described by Barbieri et al. [2] in plates coated with 250 ng of antigen per well. Serum samples (diluted 1/500) were assayed by triplicates (total IgG), and were considered positives when their mean absorbance reading was higher than that of the cut-off of the assay, which was defined as the mean absorbance value of the 19 healthy donor sera assayed, plus three standard deviations.
180
G. Gonza´lez et al. / Molecular and Biochemical Parasitology 105 (2000) 177–184
3. Results and discussion
3.1. Identification of P-29 In order to obtain mAbs specific for protoscolex components, we immunized mice with SPE and screened for positive clones against this antigen. Only a minor subset of the positive clones reacted against periodate treated SPE, which is in agreement with the reported immunodominance of the protoscolex carbohydrates [24]. mAb 47H.PS was selected because it defines a proteinous epitope and because of its advantageous isotype (IgG). In immunoblot experiments, mAb 47H.PS recognized a single band of about 29 kDa in SPE (under both reducing and non-reducing conditions). This component, that we termed P-29, was isolated from SPE using mAb 47H-PS coupled to Sepharose (Fig. 1). P-29 was also present in cyst membrane extracts (fertile and non-fertile), but could not be detected in highly concentrated samples of hydatid cyst fluid, or adult worm extracts (Fig. 1). These results were further confirmed by the im-
munohistochemistry studies shown in Fig. 2. Immunoreactivity was observed in the germinal layer of the hydatid cyst membrane, and also in the tegumentary syncytium, rostellum and suckers of protoscoleces, while was absent in the adult worm sections. These results would indicate that P-29 is a developmentally regulated component of the metacestode of E. granulosus, which is not secreted in vivo. In addition, since P-29 is localized to those regions of the protoscolex that are crucial for the interaction with the host mucosal epithelium, it may constitute and interesting molecule to be assayed for protection.
3.2. Characterization of P-29 In order to find out about the relationship of P-29 with known proteins we prepared internal fragments by proteolysis of immunopurified P-29 with trypsin, or by in situ digestion in gels after SDS-PAGE with Lys-C. The fragments were separated by reverse phase chromatography and four main peaks were sequence (XXTATEXFVDINIAS, VGQESLTIFEK, LGTALEQVASQS and XFDGLSVQLLDLI). Comparison of these sequences with the protein databases showed that three of them were identical to regions of the deduced amino acid sequence of Eg6 (the cDNA clone postulated to represent an Ag5 epitope [9]). This strongly suggests that Eg6 codes for a subfragment of P-29. The remaining fragment (XFDGLSVQLLDLI) could latter be assigned to another region of P-29 (see below).
3.3. Isolation of the full length cDNA encoding P-29
Fig. 1. SDS-PAGE and immunoblot analysis of immunopurified P-29. Samples were analyzed on 12% gel under non-reducing conditions, and silver stained (panel a) or blotted into nitrocellulose sheets and probed with mAb 47H.PS (panel b). Lane 1a: molecular weight markers; lane 2a: immunopurified P29; lane 3a: SPE; lane 1b: SPE; lane 2b: hydatid cyst fluid; lane 3b: fertile cyst membrane extract; lane 4b: adult worm extract.
The molecular characterisation of P-29 was carried out by isolating the full length cDNA encoding this protein, using the strategy described in Section 2. The combined sequences of the 5’ end RACE and cDNAs amplified from the library was 852 nucleotide long, containing a 714 bp open reading frame from an initiator ATG codon at position 25. The deduced amino acid sequence of 238 residues has a predicted molecular mass of 27.2 kDa (Fig. 3). The fact that a single PCR product was obtained by 5%end RACE, the conser-
G. Gonza´lez et al. / Molecular and Biochemical Parasitology 105 (2000) 177–184
181
Fig. 2. Immunohistochemistry of E. granulosus sections. Staining with mAb 47H.PS (panel b) and negative controls (panel a). Protoscolex (1a and 1b), non-fertile cyst membrane (2a and 2b), and adult worm sections (3a and 3b). Capital letters are used to indicate some parasite structures: suckers, S; rostellum, R; tegument, T; germinal layer, G; laminar layer, L. Positive labelling is evident in the rostellum, suckers and tegument of protoscoleces, as well as the germinal layer of the cyst. No staining is observed in the adult worm sections.
vation of the motif for transcription initiation by eukaryotic ribosomes around the first ATG codon [25], and the calculated molecular mass from the deduced amino acid sequence compared to the electrophoretic mobility of the native protein, strongly indicates that the full length cDNA encoding P-29 had been isolated. Furthermore, although the N-terminal amino acid sequence of the native protein was blocked, it was possible to assign all the peptide fragments sequenced from the native P-29 to the deduced amino acid sequence of the cDNA, Fig. 3. No hydrophobic leader sequence was observed, suggesting that P29 is unlikely to be a secreted protein, in agreement with its absence in hydatid cyst fluid.
No significant homology was observed with nucleotide or amino acid sequences stored in the existing databases, except for its full identity with Eg6, and almost complete identity (98%) with Em6. At this point, it was clear that Eg6, the partial gene sequence described as an immunogenic epitope of Ag5 [9], was a subfragment of the gene coding for P-29, while Em6 (also a partial cDNA sequence) represented the homologous P29 gene for E. multilocularis.
3.4. Relationship between P-29 and Ag5 As we have seen, different groups using antibodies reactive with the 38 kDa subunit of Ag5
182
G. Gonza´lez et al. / Molecular and Biochemical Parasitology 105 (2000) 177–184
Fig. 3. Alignment of the deduced amino acid sequence of P-29 and Em6. The subfragment of P-29 corresponding to deduced amino acid sequence of Eg6 is italicized, and the proteolytic fragments submitted to amino acid sequencing are shown in bold. The epitope defined by mAb 47H.PS (DAFQKNKE) is underlined. The nucleotide sequence of P-29 cDNA has been deposited in GenBank™ (accession number AF078931).
isolated cDNA sequences enconding fragments of P-29. The question is whether P-29 is a fragment of the Ag5 38 kDa subunit, or a different protein? To assess this matter we first analyzed by immunoblot the cross-reactivity of the mAbs EG 02 154/12 and 47H.PS against SPE and Ag5, under non-reducing conditions (mAb EG02 154/12 epitope is lost by reduction) (Fig. 4). As expected, mAb EG 02 154/12, which has previously been shown to recognize Eg6 recombinant protein and Ag5 [9], reacted strongly against the 67 and 57 kDa subunits of Ag5, and also specifically recognized P-29 from the SPE components. On the other hand, mAb 47H.PS did not cross-react with Ag5, constituting the first evidence that P-29 and Ag5 are different proteins. In order to assess the significance of this finding, we characterized the epitope defined by 47H.PS by PEPSCAN, and found that it is delimited to the stretch DAFQKNKE (Fig. 3); therefore, this defined region of P-29 is absent in the structure of Ag5. More definitive evidence about the fact that P-29 and Ag5 are different proteins was obtained by comparing the peptide fingerprint of P-29 and that of the 38 kDa subunit of Ag5, (Fig. 5). We chose to work with the 38 kDa subunit of Ag5 because it has been shown [9] that human serum antibodies eluted from Eg6 recombinant protein only reacted with this subunit. The obvious differences between the elution profiles of their internal fragments clearly showed that they are different proteins. This was also supported by the sequencing of the peptide fragments indicated in Fig. 5. Peaks 1 and 2 corresponded to the sequences
LGTALEQVASQS and XXTATEXFVDINIAS of P-29, and peaks 3 and 4 (internal fragments of the Ag5 38 kDa subunit) corresponded to the sequences FFLGK and AANYNASEGLTRLK, respectively, which are not present in P-29. Asearch for homology of Ag5 fragments did not
Fig. 4. Immunoblot of SPE and immunopurified Ag5 probed with mAb EG 02 154/12 and mAb 47H.PS. SPE, lanes 1 and 3; Ag5, lanes 2 and 4; mAb EG 02 154/12 and mAb 47H.PS were used to stain lanes 1 and 2, and 3 and 4, respectively. The position of the molecular mass markers is shown on the left.
G. Gonza´lez et al. / Molecular and Biochemical Parasitology 105 (2000) 177–184
183
It also explains the failures of cloning Ag5 by means of immunoscreening, and illustrates the need of alternative cloning strategies, such as those based on the use of degenerate primers designed from internal fragments of Ag5.
3.5. Diagnostic 6alue of P-29
Fig. 5. Fingerprinting of P-29 and Ag5. Reverse phase elution profile of peptides obtained by in situ digestion in SDS-gels of P-29 (top) and Ag5 38 kDa subunit (bottom). The acetonitrile gradient profile is represented by straight lines. Peaks submitted to amino acid sequencing are indicated by numbers. Table 1 Diagnostic performances of Ag5, P-29 and peptide 89-122 for immunodiagnosis of hydatid disease by ELISA (total IgG, number of positive sera/number of sera assayed) Infection
Ag5
P-29
Peptide 89–122
E. granulosus E. multilocularis Taenia solium Schistosoma mansoni Filariasis Toxocariasis Normal controls
21/39 9/17 0/24 0/5 0/3 0/2 0/29
24/39 8/17 8/24 0/5 0/3 0/2 0/29
17/39 0/17 0/24 0/5 0/3 0/2 0/29
show similarities to amino acid sequences deposited in data banks. Given this new information, it was possible to deduce that the puzzle about the relationship of Eg6 and Em6 with Ag5 was brought about by the existence of cross-reacting epitopes between P-29 and the 38 kDa subunit of Ag5, especially, that defined by mAb EG 02 154/12. The fact that P-29 and Ag5 are different proteins raises several points regarding the current knowledge of Ag5, one of the major antigens for diagnosis of hydatid disease. First of all, Ag5 has not been molecularly characterized yet. Secondly, much of the information derived from studies carried out using antibodies to Ag5 could be equivocal as a result of the cross reactivity between both Ag5 and P-29; such as the data on the immunolocalization of Ag5 which has been the subject of many reports [26 – 28].
Finally, in view of the fact that Eg6 recombinant protein and its derived peptide 89–122 have been indicated as useful reagents for diagnosis of hydatid disease [2,9,11], we also established the diagnostic value of P-29, Ag5, and peptide 89–122 by measuring total IgG (Table 1). Though a slightly better sensitivity was obtained by using P-29 (62% versus 54% for Ag5), a higher degree of cross-reactivity was observed. Attempts to improve the diagnostic specificity of P-29 by destroying potential sugar epitopes with periodate [29] or measuring specific IgE did not work (not shown). Then we focused our attention in the region of P-29 that has been shown to be involved in the cross-reactivity with Ag5 [11]. This is represented by the peptide 89–122 (residues 144-177 of P-29), first described by Chamekh et al. [11], which they demonstrated mimicked the Ag5 epitope defined by mAb EG 02 154/12. Table 1 displays the reactivity of the sera against this peptide. Though the sensitivity decreased, the peptide performed with excellent specificity (100%). Therefore, the diagnostic value of P-29 seems to be limited to the region of the molecule involved in the cross-reactivity with Ag5. In conclusion, we have characterized at the molecular level a novel component of E. granulo-sus, which is highly abundant in, and restricted to, the metacestode stage of the parasite. A remarkable feature of this antigen is the strong degree of crossreactivity with Ag5, which has been misleading in previous studies on this antigen, particularly on topics such as its localization in parasite structures, cloning strategies and diagnostic value. We believe that the awareness about the fact that P-29 and Ag5 are different proteins solves the lasting confusion about P-29/Ag5 and will foster further studies on these antigens. Acknowledgements This work was supported by the Zaffaroni Foun-
184
G. Gonza´lez et al. / Molecular and Biochemical Parasitology 105 (2000) 177–184
dation, The Swedish Agency for Research Cooperation with Developing Countries (SAREC), the Consejo Nacional de Investigaciones Cientı´ficas y Te´cnicas (CONICYT), the Comisio´n Sectorial de Investigacio´n Cientı´fica (CSIC), Universidad de la Repu´blica, Uruguay, and the Wellcome Trust. References [1] Lightowlers MW, Liu DY, Haralambous A, Rickard MD. Subunit composition and specificiy of the major cyst fluid antigens of Echinococcus granulosus. Mol Biochem Parasitol 1989;37:171 – 82. [2] Barbieri M, Ferna´ndez V, Gonza´lez G, Martı´nez V, Nieto A. Diagnostic evaluation of a synthetic peptide derived from a novel antigen B subunit as related to other available peptides and native antigens used for serology of cystic hydatidosis. Parasite Immunol 1997;20:51–61. [3] Frosch P, Hartmann M, Mu¨hlschlegel F, Frosch M. Sequence heterogeneity of the echinococcal antigen B. Mol Biochem Parsitol 1994;64:171–5. [4] Ferna´ndez V, Ferreira HB, Ferna´ndez C, Zaha A, Nieto A. Molecular characterization of a novel 8 kDa subunit of Echinococcus granulosus antigen. Mol Biochem Parasitol 1995;77:247 – 50. [5] Gonza´lez G, Nieto A, Ferna´ndez C, O8 rn A, Wernstedt C, Hellman U. Two different 8 kDa monomers are involved in the oligomeric organization of the native Echinococcus granulosus antigen B. Parasite Immunol 1996;18:587–96. [6] Capron A, Vernes A, Biguet J. Le diagnostic immuno-e´lectrophore´tique de l’hydatidose. In: Le Kyste Hydatique du Foie. Paris: SIMEP, 1967:27–40. [7] Di Felice G, Pini E, Afferni C, Vicari G. Purification and partial characterization of the major antigen of Echinococcus granulosus (antigen 5) with monoclonal antibodies. Mol Biochem Parasitol 1986;20:133–42. [8] Zhang LH, McManus DP. Purification and N-terminal amino acid sequencing of Echinococcus granulosus antigen 5. Parasite Immunol 1996;12:597–606. [9] Facon B, Chamekh M, Dissous C, Capron A. Molecular cloning of an Echinococcus granulosus protein expressing an immunogenic epitope of antigen 5. Mol Biochem Parasitol 1991;45:233 – 40. [10] Chamekh M, Facon B, Dissous C, Haque A, Capron A. Use of a monoclonal antibody specific for a protein epitope of Echinococcus granulosus antigen 5 in a competitive antibody radioimmunoassay for diagnosis of hydatid disease. J Immunol Methods 1990;134:129–37. [11] Chamekh M, Gras-Masse H, Bossus M, Facon B, Dissous C, Tartar A, Capron A. Diagnostic value of a synthetic peptide derived from Echinococcus granulosus recombinant protein. J Clin Invest 1992;89:458–64. [12] Siles MM, Gottstein BG, Felleisen RS. Identification of a differentially expressed Echinococcus multilocularis protein Em6 potentially related to antigen 5 of Echinococcus granulosus. Parasite Immunol 1998;20:473–81. [13] Jenkins DJ, Rickard Md. Specific antibody responses in dogs experimentally infected with Echinococcus granulosus.
Am J Trop Med Hyg 1986;35:345 – 9. [14] Woodward MP, Young WW Jr, Boodgood RA. Detection of monoclonal antibodies specific for carbohydrate epitopes using periodate oxidation. J Immunol Methods 1985;78:143 – 5. [15] Gonza´lez G, O8 rn A, Cazzulo JJ, Gro¨nvik K. Production and characterization of monoclonal antibodies against the major cysteine proteinase of Trypanosoma cruzi. Scand J Immunol 1994;40:389 – 94. [16] Geysen HM, Rodda SJ, Mason TJ, Tribbick G, Schoofs PG. Strategies for epitope analysis using peptide synthesis. J Immunol Methods 1987;102:259 – 74. [17] Laemmli UK. Cleavage of the structural proteins during the assembly of the head of bacteriophage T4. Nature 1970;227:680 – 5. [18] Towbin H, Staehelin T, Gordon J. Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedures and some applications. Proc Natl Acad Sci 1979;76:4350 – 4. [19] Jackson P, Blythe D. Immunolabelling techniques for light microscopy. In: Beesley JE, editor. Immunocytochemistry. Oxford: University Press, 1993:15 – 41. [20] McIlhinney RA, McGlone K, Willis AC. Purification and partial sequencing of myristoyl-CoA:protein N-myristoyltransferase from bovine brain. Biochemistry 1993;290:405 – 10. [21] Hellman U, Wernstedt C, Gonez J, Heldin CH. Improvement of an ‘In-Gel’ digestion procedure for the micropreparation of internal protein fragments for amino acid sequencing. Anal Biochem 1995;224:451 – 5. [22] Saiki RK, Gelfand DH, Stoffel S, Scharf S, Higuchi R, Horn G, Mullis K, Erlich H. Primer-directed enzymatic amplification of DNA with a thermostable DNA polymerase. Science 1988;238:487 – 91. [23] Sambrook J, Fritsch E, Maniatis T. Plasmid vectors. In: Sambrook J, Fritsch E, Maniatis T, editors. Molecular Cloning: A Laboratory Manual, 2nd ed. Cold Spring Harbor, NY: Laboratory Press, 1989:1.1 – 1.101. [24] Migues M, Baz A, Nieto A. Carbohydrates on the surface of Echinococcus granulosus protoscoleces are immunodominant in mice. Parasite Immunol 1996;18:559 – 69. [25] Kozak M. The scanning model for translation: an update. J Cell Biol 1989;108:229 – 41. [26] Sa´nchez F, Garcı´a J, March F, Carden˜osa N, Coll M, Mun˜oz P, Auladell C, Prats G. Ultrastructural localization of major hydatid fluid antigens in brood capsules and protoscoleces of Echinococcus granulosus of human origin. Parasite Immunol 1993;15:441 – 7. [27] Davies C, Rickard MD, Bout DT, Smith JD. Ultrastructural immunocytochemical localization of two hydatid fluid antigens (antigen 5 and antigen B) in the brood capsules of Echinococcus granulosus and E. multilocularis. Parasitology 1978;77:143 – 52. [28] Jones MK, Zhang LH, Leggatt GR, Stenzel DJ, McManus DP. The ultrastructural localization of Echinococcus granulosus antigen 5. Parasitology 1996;113:213 – 22. [29] Sterla S, Ljungstro¨m I, Nieto A. Modified ELISA for hydatid serodiagnosis: the potential of periodate treatment and phosphorylcholine inhibition. Serodiagn Immunother Infect Dis 1997;8:145 – 8.
Molecular characterization of P-29, a metacestode-specific component of Echinococcus granulosus which is immunologically related to, but distinct from, antigen 5 Gualberto Gonza´lez a,*, Pablo Spinelli a, Carmen Lorenzo a, Ulf Hellman b, Alberto Nieto a, Antony Willis c, Gustavo Salinas a a Ca´tedra de Inmunologı´a-Facultad de Quı´mica, Monte6ideo, Uruguay Ludwig Institute for Cancer Research, Uppsala Branch, Uppsala, Sweden c MRC Immunochemistry Unit, Department of Biochemistry, Uni6ersity of Oxford, Oxford, UK b
Received 9 June 1999; received in revised form 14 June 1999; accepted 20 August 1999
Abstract In this work the characterization of P-29, a novel 29 kDa antigen from Echinococcus granulosus is reported. E. granulosus was identified while looking for parasite antigens distinct from those present in hydatid cyst fluid. A monoclonal antibody (mAb 47H.PS) prepared against protoscolex components revealed that P-29 is localized to the tegument and rostellum of protoscoleces, and to the germinal layer of the cyst, but it is absent in hydatid cyst fluid or adult worm extracts. Several internal fragments of P-29 showed sequence identity to the amino acid sequence encoded by Eg6, a partial gene sequence reported to code for an epitope of antigen 5 (Ag5), one of the major diagnostic antigens of the parasite. We confirmed that Eg6 encodes a sub-fragment of P-29 by mapping the epitope of mAb 47H.PS, and isolating the full length P-29 cDNA. Since Eg6 had been postulated to encode a fragment of Ag5, we specifically studied the relationship of P-29 and Ag5 by: (i) examining the cross-reactivity displayed by different mAbs; (ii) comparison of their peptide finger prints; and (iii) a comparative study of their diagnostic value. Our results prove unequivocally that P-29 and Ag5 are immunologically related, but different proteins, raising several questions on the current knowledge of Ag5. © 2000 Elsevier Science B.V. All rights reserved. Keywords: E. granulosus; E. multilocularis; Hydatid disease; Diagnosis; Antigen 5
Abbre6iations: SPE, somatic protoscolex extract; Ag5, E. granulosus antigen 5; Eg6, partial cDNA sequence coding for a protein fragment immunologically related to Ag5; Em6, E. multilocularis partial cDNA sequence homologous to Eg6. Note: Nucleotide sequence data reported in this paper is available in the EMBL, GenBank™ and DDJB data bases under the accession number AF078931. * Corresponding author. Tel.: + 598-2-4801196; fax: + 598-2-4874320. E-mail address: [email protected] (G. Gonza´lez) 0166-6851/00/$ - see front matter © 2000 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 6 - 6 8 5 1 ( 9 9 ) 0 0 1 6 6 - 8
178
G. Gonza´lez et al. / Molecular and Biochemical Parasitology 105 (2000) 177–184
1. Introduction Hydatid cyst fluid has been the usual source of Echinococcus granulosus antigens for the immunodiagnosis of hydatid disease, while other parasite structures such as protoscoleces, cyst membranes or oncospheres have received little attention. The major antigens of hydatid cyst fluid are antigen B and antigen 5 (Ag5), and their diagnostic value has been thoroughly studied [1,2]. While there have been important advances in the understanding of the molecular organization of AgB [3–5], the characterization of Ag5 has been only partial. The antigen is identified as such by the formation of a characteristic arc of precipitation in immunoelectrophoresis with sera from infected patients. This arc, known as arc 5, was one of the first tools used in the immunodiagnosis of hydatid disease and gave its name to the antigen [6].Using antibodies obtained from this arc, Lightowler et al. [1] identified the components of Ag5 in SDS-gels. Under non-reducing conditions, Ag5 occurs as a band of about 67 kDa which dissociates, in the presence of reducing agents, into two bands of 38 and 24 kDa, respectively. Another component of about 57 kDa was identified by using monoclonal antibodies (mAbs) [7]. Apart from some N-terminal amino acid sequences published by Zhang and McManus [8], little is known about the primary structure of these components. Using disease-specific sera highly reactive to Ag5, Facon et al. [9] isolated a partial cDNA sequence termed Eg6. The recombinant protein fragment encoded by Eg6, was recognized by the mAb EG 02 154/12 specific for Ag5 [10]. In addition, antibodies eluted from this recombinant protein recognized the 38 kDa subunit of Ag5. Based on this cross-reactivity evidence, the authors suggested that Eg6 codes for an antigenic epitope of Ag5. Supporting their hypothesis, a 34 amino acid synthetic peptide derived from Eg6 was recognized by most of the sera reactive to Ag5 and inhibited the binding of mAb EG 02 154/12 to Ag5 [11]. Likewise, using antibodies eluted from a 38 kDa protein component of Echinococcus multilocularis fertile cysts, Siles et al. [12] recently isolated Em6, a partial cDNA sequence which displays 98% sequence identity with Eg6. On these basis, the authors suggested that Em6 is the
E. multilocularis homologous of Eg6, and concluded that it may represent a fragment of E. multilocularis Ag5. However, in either case, no direct evidence of the true relationship between Eg6 (or Em6) and Ag5 was presented. Here we report the characterization of a novel 29 kDa protein of E. granulosus, termed P-29, which was isolated as a result of our attempts to find alternative antigens distinct from those present in hydatid cyst fluid. The protein is restricted to the metacestode stage of the parasite and appears to be valuable in diagnosis. We also found evidence that Eg6 encodes a fragment of P-29, and therefore analyzed the relationship between P-29 and Ag5. We demonstrated that they are immunologically related, but different proteins. 2. Materials and methods
2.1. Parasite antigens Bovine hydatid cysts were collected from cattle killed at abattoirs in Uruguay. The hydatid fluid and protoscoleces were drained from the cyst, the protoscoleces were decanted and the supernatant was supplemented with 500 mg/l of NaN3, 1 mM EDTA and kept at 4°C until used. Hydatid cyst walls were removed from both fertile and non-fertile cysts. Prepatent adult worms (30 days post-infection) were obtained from infected dogs as described by [13]. Somatic extracts of protoscoleces, membrane layer or adult worms were prepared by sonication. Periodate treatment was performed according to Woodward et al. [14]. P-29 and Ag5 were purified by immunochromatography using the mAbs described below.
2.2. mAbs mAb EG 02 154/12 reactive with Ag5 [10] was a kind gift of Dr M. Chamekh. mAb 47H.PS was obtained as follows: a mouse was primed intraperitoneally with 20 mg of somatic protoscolex extract (SPE) in Freund’s complete adjuvant, and boosted after 3 and 6 weeks with the same antigen using Freund’s incomplete adjuvant. Three days after the last booster the mouse was killed and the splenocytes fused with SP2/0 cells, basically as de-
G. Gonza´lez et al. / Molecular and Biochemical Parasitology 105 (2000) 177–184
scribed in [15]. Cultures producing mAbs reactive with SPE were selected by ELISA.
2.3. Mapping of the mAb 47H.PS epitope This was accomplished by using the ‘PEPSCAN’ approach [16] (Zeneca, UK). Eight-mer-overlapping peptides derived from the sequence of Eg6 were assayed on the synthesis pins with mAb 47H.PS following the instructions of the supplier.
2.4. SDS-PAGE, immunoblot and immunolabelling Electrophoresis and immunoblot were performed essentially as described by Laemmli [17] and Towbin [18], respectively. Immunocytochemical labelling with mAb 47H.PS was performed as described by Jackson and Blythe [19] on 6 mm parasite sections, blocked with 1% normal goat serum and developed with diaminobenzidine-H/hydrogen peroxide.
2.5. Amino acid sequencing and fingerprinting of P-29 and Ag5 Fragmentation of immunopurified P-29 or Ag5 was performed in solution with trypsin as described in reference [20] or by in situ digestion in SDS-gels according to Hellman et al. [21]. Briefly, the band was excised from the SDS-gel, washed and dried under nitrogen. This material was then rehydrated, during which process Lys-C (Boeringer-Manheim, Germany) was added. After overnight incubation at 30°C, and several extraction steps with 0.1% trifluoracetic acid, 60% acetonitrile, the supernatant was run on a mRPC C2/C18 SC 2.1/10 column, operated in SMART System (Pharmacia, Sweden). Elution was performed by a linear gradient from 0 – 40% acetonitrile in 0.065% trifluoroacetic acid. Selected fractions were sequenced in an Applied Biosystems Protein Sequencer, Model 470a (USA).
2.6. Isolation of the full length cDNA encoding P-29 Eg6 sense (5’-GAGGTGCGAAAAGACGAAAGTGAC-3’) and antisense (5’-GCAAGTCTTCTCAAAGATAGTAAGAG-3’) oligonucle-
179
otides derived from the cDNA sequence of Eg6 [9] were used in combination with lambda gt11 reverse and forward, respectively, to amplify the 5’-and 3’ends of Eg6 by PCR [22] from an E. granulosus protoscolex cDNA library. The PCR products obtained were isolated (GeneClean, BIO101) and cloned into pGEMTM (Promega) according to standard techniques [23]. The 5’- and 3’-end cDNA clones were then sequenced using AmpliTaqTM FS according to standard protocols on an ABI Prism 377 DNA sequencer (Sequi-net, CO). In order to obtain the 5’-end of the cDNA, 5’ RACE (rapid amplification of cDNA ends) was carried out using a commercial kit (Gibco). Briefly, reverse transcription was performed from TRIzol (Gibco) prepared total protoscolex RNA using a P-29 antisense primer (GAGCAGTACCCAGTTTGTCGGTA), the cDNA tailed using the terminal deoxynucleotidyl transferase (TdT) and subsequently amplified by PCR with a second P-29 antisense primer (GGTCGATCTCATGGTAAAG) and the abridged anchor primer provided by the kit. The PCR product thus obtained was cleaned up (Promega) and directly sequenced as above.
2.7. Human serum specimens The following serum samples were used in this study: Thirty nine sera from surgically confirmed cystic hydatid patients, taken before surgery (the samples were not preselected on the basis of previous serologic information); 19 sera from healthy donors and 51 serum samples from patients with the following diseases: alveolar hydatid disease (17), Taenia solium cysticercosis (24), schistosomiasis (five), lymphatic filariasis (three), toxocariasis (two).
2.8. ELISA ELISA was performed as described by Barbieri et al. [2] in plates coated with 250 ng of antigen per well. Serum samples (diluted 1/500) were assayed by triplicates (total IgG), and were considered positives when their mean absorbance reading was higher than that of the cut-off of the assay, which was defined as the mean absorbance value of the 19 healthy donor sera assayed, plus three standard deviations.
180
G. Gonza´lez et al. / Molecular and Biochemical Parasitology 105 (2000) 177–184
3. Results and discussion
3.1. Identification of P-29 In order to obtain mAbs specific for protoscolex components, we immunized mice with SPE and screened for positive clones against this antigen. Only a minor subset of the positive clones reacted against periodate treated SPE, which is in agreement with the reported immunodominance of the protoscolex carbohydrates [24]. mAb 47H.PS was selected because it defines a proteinous epitope and because of its advantageous isotype (IgG). In immunoblot experiments, mAb 47H.PS recognized a single band of about 29 kDa in SPE (under both reducing and non-reducing conditions). This component, that we termed P-29, was isolated from SPE using mAb 47H-PS coupled to Sepharose (Fig. 1). P-29 was also present in cyst membrane extracts (fertile and non-fertile), but could not be detected in highly concentrated samples of hydatid cyst fluid, or adult worm extracts (Fig. 1). These results were further confirmed by the im-
munohistochemistry studies shown in Fig. 2. Immunoreactivity was observed in the germinal layer of the hydatid cyst membrane, and also in the tegumentary syncytium, rostellum and suckers of protoscoleces, while was absent in the adult worm sections. These results would indicate that P-29 is a developmentally regulated component of the metacestode of E. granulosus, which is not secreted in vivo. In addition, since P-29 is localized to those regions of the protoscolex that are crucial for the interaction with the host mucosal epithelium, it may constitute and interesting molecule to be assayed for protection.
3.2. Characterization of P-29 In order to find out about the relationship of P-29 with known proteins we prepared internal fragments by proteolysis of immunopurified P-29 with trypsin, or by in situ digestion in gels after SDS-PAGE with Lys-C. The fragments were separated by reverse phase chromatography and four main peaks were sequence (XXTATEXFVDINIAS, VGQESLTIFEK, LGTALEQVASQS and XFDGLSVQLLDLI). Comparison of these sequences with the protein databases showed that three of them were identical to regions of the deduced amino acid sequence of Eg6 (the cDNA clone postulated to represent an Ag5 epitope [9]). This strongly suggests that Eg6 codes for a subfragment of P-29. The remaining fragment (XFDGLSVQLLDLI) could latter be assigned to another region of P-29 (see below).
3.3. Isolation of the full length cDNA encoding P-29
Fig. 1. SDS-PAGE and immunoblot analysis of immunopurified P-29. Samples were analyzed on 12% gel under non-reducing conditions, and silver stained (panel a) or blotted into nitrocellulose sheets and probed with mAb 47H.PS (panel b). Lane 1a: molecular weight markers; lane 2a: immunopurified P29; lane 3a: SPE; lane 1b: SPE; lane 2b: hydatid cyst fluid; lane 3b: fertile cyst membrane extract; lane 4b: adult worm extract.
The molecular characterisation of P-29 was carried out by isolating the full length cDNA encoding this protein, using the strategy described in Section 2. The combined sequences of the 5’ end RACE and cDNAs amplified from the library was 852 nucleotide long, containing a 714 bp open reading frame from an initiator ATG codon at position 25. The deduced amino acid sequence of 238 residues has a predicted molecular mass of 27.2 kDa (Fig. 3). The fact that a single PCR product was obtained by 5%end RACE, the conser-
G. Gonza´lez et al. / Molecular and Biochemical Parasitology 105 (2000) 177–184
181
Fig. 2. Immunohistochemistry of E. granulosus sections. Staining with mAb 47H.PS (panel b) and negative controls (panel a). Protoscolex (1a and 1b), non-fertile cyst membrane (2a and 2b), and adult worm sections (3a and 3b). Capital letters are used to indicate some parasite structures: suckers, S; rostellum, R; tegument, T; germinal layer, G; laminar layer, L. Positive labelling is evident in the rostellum, suckers and tegument of protoscoleces, as well as the germinal layer of the cyst. No staining is observed in the adult worm sections.
vation of the motif for transcription initiation by eukaryotic ribosomes around the first ATG codon [25], and the calculated molecular mass from the deduced amino acid sequence compared to the electrophoretic mobility of the native protein, strongly indicates that the full length cDNA encoding P-29 had been isolated. Furthermore, although the N-terminal amino acid sequence of the native protein was blocked, it was possible to assign all the peptide fragments sequenced from the native P-29 to the deduced amino acid sequence of the cDNA, Fig. 3. No hydrophobic leader sequence was observed, suggesting that P29 is unlikely to be a secreted protein, in agreement with its absence in hydatid cyst fluid.
No significant homology was observed with nucleotide or amino acid sequences stored in the existing databases, except for its full identity with Eg6, and almost complete identity (98%) with Em6. At this point, it was clear that Eg6, the partial gene sequence described as an immunogenic epitope of Ag5 [9], was a subfragment of the gene coding for P-29, while Em6 (also a partial cDNA sequence) represented the homologous P29 gene for E. multilocularis.
3.4. Relationship between P-29 and Ag5 As we have seen, different groups using antibodies reactive with the 38 kDa subunit of Ag5
182
G. Gonza´lez et al. / Molecular and Biochemical Parasitology 105 (2000) 177–184
Fig. 3. Alignment of the deduced amino acid sequence of P-29 and Em6. The subfragment of P-29 corresponding to deduced amino acid sequence of Eg6 is italicized, and the proteolytic fragments submitted to amino acid sequencing are shown in bold. The epitope defined by mAb 47H.PS (DAFQKNKE) is underlined. The nucleotide sequence of P-29 cDNA has been deposited in GenBank™ (accession number AF078931).
isolated cDNA sequences enconding fragments of P-29. The question is whether P-29 is a fragment of the Ag5 38 kDa subunit, or a different protein? To assess this matter we first analyzed by immunoblot the cross-reactivity of the mAbs EG 02 154/12 and 47H.PS against SPE and Ag5, under non-reducing conditions (mAb EG02 154/12 epitope is lost by reduction) (Fig. 4). As expected, mAb EG 02 154/12, which has previously been shown to recognize Eg6 recombinant protein and Ag5 [9], reacted strongly against the 67 and 57 kDa subunits of Ag5, and also specifically recognized P-29 from the SPE components. On the other hand, mAb 47H.PS did not cross-react with Ag5, constituting the first evidence that P-29 and Ag5 are different proteins. In order to assess the significance of this finding, we characterized the epitope defined by 47H.PS by PEPSCAN, and found that it is delimited to the stretch DAFQKNKE (Fig. 3); therefore, this defined region of P-29 is absent in the structure of Ag5. More definitive evidence about the fact that P-29 and Ag5 are different proteins was obtained by comparing the peptide fingerprint of P-29 and that of the 38 kDa subunit of Ag5, (Fig. 5). We chose to work with the 38 kDa subunit of Ag5 because it has been shown [9] that human serum antibodies eluted from Eg6 recombinant protein only reacted with this subunit. The obvious differences between the elution profiles of their internal fragments clearly showed that they are different proteins. This was also supported by the sequencing of the peptide fragments indicated in Fig. 5. Peaks 1 and 2 corresponded to the sequences
LGTALEQVASQS and XXTATEXFVDINIAS of P-29, and peaks 3 and 4 (internal fragments of the Ag5 38 kDa subunit) corresponded to the sequences FFLGK and AANYNASEGLTRLK, respectively, which are not present in P-29. Asearch for homology of Ag5 fragments did not
Fig. 4. Immunoblot of SPE and immunopurified Ag5 probed with mAb EG 02 154/12 and mAb 47H.PS. SPE, lanes 1 and 3; Ag5, lanes 2 and 4; mAb EG 02 154/12 and mAb 47H.PS were used to stain lanes 1 and 2, and 3 and 4, respectively. The position of the molecular mass markers is shown on the left.
G. Gonza´lez et al. / Molecular and Biochemical Parasitology 105 (2000) 177–184
183
It also explains the failures of cloning Ag5 by means of immunoscreening, and illustrates the need of alternative cloning strategies, such as those based on the use of degenerate primers designed from internal fragments of Ag5.
3.5. Diagnostic 6alue of P-29
Fig. 5. Fingerprinting of P-29 and Ag5. Reverse phase elution profile of peptides obtained by in situ digestion in SDS-gels of P-29 (top) and Ag5 38 kDa subunit (bottom). The acetonitrile gradient profile is represented by straight lines. Peaks submitted to amino acid sequencing are indicated by numbers. Table 1 Diagnostic performances of Ag5, P-29 and peptide 89-122 for immunodiagnosis of hydatid disease by ELISA (total IgG, number of positive sera/number of sera assayed) Infection
Ag5
P-29
Peptide 89–122
E. granulosus E. multilocularis Taenia solium Schistosoma mansoni Filariasis Toxocariasis Normal controls
21/39 9/17 0/24 0/5 0/3 0/2 0/29
24/39 8/17 8/24 0/5 0/3 0/2 0/29
17/39 0/17 0/24 0/5 0/3 0/2 0/29
show similarities to amino acid sequences deposited in data banks. Given this new information, it was possible to deduce that the puzzle about the relationship of Eg6 and Em6 with Ag5 was brought about by the existence of cross-reacting epitopes between P-29 and the 38 kDa subunit of Ag5, especially, that defined by mAb EG 02 154/12. The fact that P-29 and Ag5 are different proteins raises several points regarding the current knowledge of Ag5, one of the major antigens for diagnosis of hydatid disease. First of all, Ag5 has not been molecularly characterized yet. Secondly, much of the information derived from studies carried out using antibodies to Ag5 could be equivocal as a result of the cross reactivity between both Ag5 and P-29; such as the data on the immunolocalization of Ag5 which has been the subject of many reports [26 – 28].
Finally, in view of the fact that Eg6 recombinant protein and its derived peptide 89–122 have been indicated as useful reagents for diagnosis of hydatid disease [2,9,11], we also established the diagnostic value of P-29, Ag5, and peptide 89–122 by measuring total IgG (Table 1). Though a slightly better sensitivity was obtained by using P-29 (62% versus 54% for Ag5), a higher degree of cross-reactivity was observed. Attempts to improve the diagnostic specificity of P-29 by destroying potential sugar epitopes with periodate [29] or measuring specific IgE did not work (not shown). Then we focused our attention in the region of P-29 that has been shown to be involved in the cross-reactivity with Ag5 [11]. This is represented by the peptide 89–122 (residues 144-177 of P-29), first described by Chamekh et al. [11], which they demonstrated mimicked the Ag5 epitope defined by mAb EG 02 154/12. Table 1 displays the reactivity of the sera against this peptide. Though the sensitivity decreased, the peptide performed with excellent specificity (100%). Therefore, the diagnostic value of P-29 seems to be limited to the region of the molecule involved in the cross-reactivity with Ag5. In conclusion, we have characterized at the molecular level a novel component of E. granulo-sus, which is highly abundant in, and restricted to, the metacestode stage of the parasite. A remarkable feature of this antigen is the strong degree of crossreactivity with Ag5, which has been misleading in previous studies on this antigen, particularly on topics such as its localization in parasite structures, cloning strategies and diagnostic value. We believe that the awareness about the fact that P-29 and Ag5 are different proteins solves the lasting confusion about P-29/Ag5 and will foster further studies on these antigens. Acknowledgements This work was supported by the Zaffaroni Foun-
184
G. Gonza´lez et al. / Molecular and Biochemical Parasitology 105 (2000) 177–184
dation, The Swedish Agency for Research Cooperation with Developing Countries (SAREC), the Consejo Nacional de Investigaciones Cientı´ficas y Te´cnicas (CONICYT), the Comisio´n Sectorial de Investigacio´n Cientı´fica (CSIC), Universidad de la Repu´blica, Uruguay, and the Wellcome Trust. References [1] Lightowlers MW, Liu DY, Haralambous A, Rickard MD. Subunit composition and specificiy of the major cyst fluid antigens of Echinococcus granulosus. Mol Biochem Parasitol 1989;37:171 – 82. [2] Barbieri M, Ferna´ndez V, Gonza´lez G, Martı´nez V, Nieto A. Diagnostic evaluation of a synthetic peptide derived from a novel antigen B subunit as related to other available peptides and native antigens used for serology of cystic hydatidosis. Parasite Immunol 1997;20:51–61. [3] Frosch P, Hartmann M, Mu¨hlschlegel F, Frosch M. Sequence heterogeneity of the echinococcal antigen B. Mol Biochem Parsitol 1994;64:171–5. [4] Ferna´ndez V, Ferreira HB, Ferna´ndez C, Zaha A, Nieto A. Molecular characterization of a novel 8 kDa subunit of Echinococcus granulosus antigen. Mol Biochem Parasitol 1995;77:247 – 50. [5] Gonza´lez G, Nieto A, Ferna´ndez C, O8 rn A, Wernstedt C, Hellman U. Two different 8 kDa monomers are involved in the oligomeric organization of the native Echinococcus granulosus antigen B. Parasite Immunol 1996;18:587–96. [6] Capron A, Vernes A, Biguet J. Le diagnostic immuno-e´lectrophore´tique de l’hydatidose. In: Le Kyste Hydatique du Foie. Paris: SIMEP, 1967:27–40. [7] Di Felice G, Pini E, Afferni C, Vicari G. Purification and partial characterization of the major antigen of Echinococcus granulosus (antigen 5) with monoclonal antibodies. Mol Biochem Parasitol 1986;20:133–42. [8] Zhang LH, McManus DP. Purification and N-terminal amino acid sequencing of Echinococcus granulosus antigen 5. Parasite Immunol 1996;12:597–606. [9] Facon B, Chamekh M, Dissous C, Capron A. Molecular cloning of an Echinococcus granulosus protein expressing an immunogenic epitope of antigen 5. Mol Biochem Parasitol 1991;45:233 – 40. [10] Chamekh M, Facon B, Dissous C, Haque A, Capron A. Use of a monoclonal antibody specific for a protein epitope of Echinococcus granulosus antigen 5 in a competitive antibody radioimmunoassay for diagnosis of hydatid disease. J Immunol Methods 1990;134:129–37. [11] Chamekh M, Gras-Masse H, Bossus M, Facon B, Dissous C, Tartar A, Capron A. Diagnostic value of a synthetic peptide derived from Echinococcus granulosus recombinant protein. J Clin Invest 1992;89:458–64. [12] Siles MM, Gottstein BG, Felleisen RS. Identification of a differentially expressed Echinococcus multilocularis protein Em6 potentially related to antigen 5 of Echinococcus granulosus. Parasite Immunol 1998;20:473–81. [13] Jenkins DJ, Rickard Md. Specific antibody responses in dogs experimentally infected with Echinococcus granulosus.
Am J Trop Med Hyg 1986;35:345 – 9. [14] Woodward MP, Young WW Jr, Boodgood RA. Detection of monoclonal antibodies specific for carbohydrate epitopes using periodate oxidation. J Immunol Methods 1985;78:143 – 5. [15] Gonza´lez G, O8 rn A, Cazzulo JJ, Gro¨nvik K. Production and characterization of monoclonal antibodies against the major cysteine proteinase of Trypanosoma cruzi. Scand J Immunol 1994;40:389 – 94. [16] Geysen HM, Rodda SJ, Mason TJ, Tribbick G, Schoofs PG. Strategies for epitope analysis using peptide synthesis. J Immunol Methods 1987;102:259 – 74. [17] Laemmli UK. Cleavage of the structural proteins during the assembly of the head of bacteriophage T4. Nature 1970;227:680 – 5. [18] Towbin H, Staehelin T, Gordon J. Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedures and some applications. Proc Natl Acad Sci 1979;76:4350 – 4. [19] Jackson P, Blythe D. Immunolabelling techniques for light microscopy. In: Beesley JE, editor. Immunocytochemistry. Oxford: University Press, 1993:15 – 41. [20] McIlhinney RA, McGlone K, Willis AC. Purification and partial sequencing of myristoyl-CoA:protein N-myristoyltransferase from bovine brain. Biochemistry 1993;290:405 – 10. [21] Hellman U, Wernstedt C, Gonez J, Heldin CH. Improvement of an ‘In-Gel’ digestion procedure for the micropreparation of internal protein fragments for amino acid sequencing. Anal Biochem 1995;224:451 – 5. [22] Saiki RK, Gelfand DH, Stoffel S, Scharf S, Higuchi R, Horn G, Mullis K, Erlich H. Primer-directed enzymatic amplification of DNA with a thermostable DNA polymerase. Science 1988;238:487 – 91. [23] Sambrook J, Fritsch E, Maniatis T. Plasmid vectors. In: Sambrook J, Fritsch E, Maniatis T, editors. Molecular Cloning: A Laboratory Manual, 2nd ed. Cold Spring Harbor, NY: Laboratory Press, 1989:1.1 – 1.101. [24] Migues M, Baz A, Nieto A. Carbohydrates on the surface of Echinococcus granulosus protoscoleces are immunodominant in mice. Parasite Immunol 1996;18:559 – 69. [25] Kozak M. The scanning model for translation: an update. J Cell Biol 1989;108:229 – 41. [26] Sa´nchez F, Garcı´a J, March F, Carden˜osa N, Coll M, Mun˜oz P, Auladell C, Prats G. Ultrastructural localization of major hydatid fluid antigens in brood capsules and protoscoleces of Echinococcus granulosus of human origin. Parasite Immunol 1993;15:441 – 7. [27] Davies C, Rickard MD, Bout DT, Smith JD. Ultrastructural immunocytochemical localization of two hydatid fluid antigens (antigen 5 and antigen B) in the brood capsules of Echinococcus granulosus and E. multilocularis. Parasitology 1978;77:143 – 52. [28] Jones MK, Zhang LH, Leggatt GR, Stenzel DJ, McManus DP. The ultrastructural localization of Echinococcus granulosus antigen 5. Parasitology 1996;113:213 – 22. [29] Sterla S, Ljungstro¨m I, Nieto A. Modified ELISA for hydatid serodiagnosis: the potential of periodate treatment and phosphorylcholine inhibition. Serodiagn Immunother Infect Dis 1997;8:145 – 8.