- Email: [email protected]
Identification and characterisation of three antigenic proteins from Cryptosporidium parvum sporozoites using a DNA library expressing poly-histidine tagged peptides1
International Journal for Parasitology 29 (1999) 1925±1933
Identi®cation and characterisation of three antigenic proteins from Cryptosporidium parvum sporozoites using a p DNA library expressing poly-histidine tagged peptides Fabio Tosini *, Simone CaccioÁ, Alessandra Tamburrini, Giuseppe La Rosa, Edoardo Pozio Laboratory of Parasitology, Istituto Superiore di SanitaÁ, viale Regina Elena 299, 00161 Rome, Italy Received 16 July 1999; received in revised form 6 September 1999; accepted 7 September 1999
Abstract To identify antigenic peptides of the parasite Cryptosporidium parvum, an expression library that allows for the production of chimeric proteins fused with a 6-histidine tag was made. The library was screened with C. parvum sporozoite rabbit anti-serum, and three positive clones (sa20, sa35, and sa40) were identi®ed. The corresponding recombinant proteins (SA20, SA35, and SA40) were expressed in Escherichia coli and puri®ed by metal-anity chromatography. The sequence of sa20 and sa35 clones did not show any homology with known genes or proteins, whereas the 5 0 end of the sa40 clone showed homology with two previously identi®ed C. parvum sequences. Hybridisations to intact chromosomes fractionated by pulsed-®eld gel electrophoresis revealed that the sa35 and sa40 sequences are localised on chromosome VII, whereas the sa20 sequence is localised on chromosome VI. Reverse transcriptase±PCR experiments showed the presence of mRNAs for sa35 and sa40 in the oocyst, whereas the sa20 mRNA was undetectable in this stage. The serological response to the three proteins was assayed in C. parvum-immunised rabbits and in immunocompetent individuals with cryptosporidiosis. The Western blot results indicated that rabbits, challenged with a sporozoite crude antigen or with an oocyst crude antigen, were highly responsive to these three antigens. Human serum samples showed a response to the three proteins, although the response to SA20 appeared to be unrelated to a recent C. parvum infection. These results suggest that the SA35 and the SA40 proteins may be useful in detecting C. parvum infections. # 1999 Australian Society for Parasitology Inc. Published by Elsevier Science Ltd. All rights reserved. Keywords: Antigens; Cryptosporidium parvum; Genomic library; Immuno-detection; Poly-histidine tagged peptides; RT±PCR
1. Introduction Note: Nucleotide sequences reported in this paper are available in the GenBank2, EMBL and DDBJ databases under the accession numbers AJ006592 (sa20), AJ006593 (sa35), and AJ132769 (sa40). * Corresponding author. Present address: Laboratory of Cellular Biology, Istituto Superiore di SanitaÁ, viale Regina Elena 299, 00161 Rome, Italy. Tel.: +39-06-4990-2308; fax: +39-06-4938-7065 E-mail address: [email protected] (F. Tosini). p
Cryptosporidium parvum is a protozoan parasite that causes enteritis in humans and in other mammalian species. Infection prevalently occurs in the gut epithelial mucosa, provoking severe diarrhoea, which, in an immunocompetent host, can resolve in about 10 days. Although infected humans can develop a protective immunore-
0020-7519/99/$20.00 # 1999 Australian Society for Parasitology Inc. Published by Elsevier Science Ltd. All rights reserved. PII: S 0 0 2 0 - 7 5 1 9 ( 9 9 ) 0 0 1 5 8 - 7
1926
F. Tosini et al. / International Journal for Parasitology 29 (1999) 1925±1933
sponse to C. parvum, they are susceptible to reinfection [1]. In immunocompromised hosts, such as persons with AIDS, cryptosporidiosis is a common cause of chronic diarrhoea, which may lead to death. Given the lack of an eective chemotherapy for C. parvum infection, several attempts have been made to develop passive immunisation. Positive results were obtained with the oral administration of immune colostrum in neonatal mice [2] and in humans [3]. The use of mAbs was also found to be eective in controlling C. parvum infection in mice [4]. The molecular basis of the pathogenesis and the molecules involved in the invasion mechanism of the host cell are poorly understood [5]. Moreover, few genes encoding for immunogenic proteins involved in the host±parasite interaction have been characterised. Therefore, the identi®cation of novel molecules recognised by the host immune system may contribute to understanding the mechanism of the infection. Identi®cation of some C. parvum antigens was successfully achieved with the immunological screening of expression libraries which express fusion proteins [6, 7]. The objective of our study was to identify and characterise antigenic molecules speci®c to the sporozoite stage. Sporozoites are responsible for the host cell invasion, and are free in the intestinal lumen before penetrating into the enterocyte [8]; thus, they can be blocked by the host immunoresponse [9].
2. Methods and materials 2.1. General microbiological and DNA recombinant techniques Microbiological techniques and media, preparation of plasmid DNA and isolation of restricted fragments all followed standard procedures [10]. DNA sequencing was performed using Sanger's procedure [11].
2.2. DNA extraction and construction of a genomic C. parvum expression library Puri®ed oocysts of C. parvum were obtained from infected calves as described previously [12]. The isolate (code ISSC6) originated from a naturally infected calf from Denmark. Total genomic DNA was prepared using previously described methods [13]. After partial digestion with BstyI endonuclease, DNA fragments were ligated in the BamHI site of the cloning vectors pQE30, pQE31, and pQE32 (Qiagen). The DNA fragments were inserted in three dierent vectors, which correspond to the three dierent open reading frames (ORF) for each cloning site, in order to increase the probability of obtaining the desired juxtapositions of the inserts with the 6His tag. The ligation reaction was used to transform Escherichia coli M15 cells, and approximately 3105 cells were plated on Luria Bertani agar plates with 50 mg ml ÿ 1 of ampicillin plus 25 mg ml ÿ 1 kanamycin to amplify single clones. Bacterial colonies were resuspended in LB medium using a rotating mixer; 50% glycerol was added and aliquots were stored at ÿ808C. The ®nal titer of the library corresponded to 21011 cfu ml ÿ 1 and recombinant colonies were estimated to represent 85% of the total. 2.3. Sporozoite puri®cation Puri®ed oocysts were suspended at 1107 ml ÿ 1 in ABS buer (0.1 M sodium acetate, pH 5.5; 0.145 M NaCl), washed, resuspended in ABS buer and placed on ice for 20 min. Oocysts were washed twice with PBS buer, resuspended in excystation buer (MEM medium plus 0.75% w/ v sodium taurocholate) and incubated at 378C for 30±60 min (excystation was followed by microscopy). The excystation mixture was then loaded onto a 50 ml syringe connected to a 25 mm micro®ltration device with a 3 mm ®lter (Costar Nuclepore) and the elution of puri®ed sporozoites was obtained by gravity ¯ow. Sporozoites were collected by centrifugation, resuspended in MEM medium and stored at 48C for 14±18 h prior to immunisation.
F. Tosini et al. / International Journal for Parasitology 29 (1999) 1925±1933
2.4. Rabbit immunisation Two New Zealand germ-free rabbits were inoculated i.v. without adjuvant with approximately 1108 sporozoites per inoculum. The challenge was repeated three times at intervals of 20 days, and the blood serum was collected 2 weeks after the last immunisation. Two other New Zealand germ-free rabbits were inoculated with an oocyst-soluble antigen as previously described [14]. Serum aliquots were stored at ÿ808C. 2.5. Library immunoscreening The expression library was screened using standard procedures [10]. Filters were washed three times with washing buer (TNT buer plus 0.1% BSA and 0.1% nonidet NP40), incubated with HRP-GAR antibody (BioRad) diluted 1:3000 for 1 h, and washed as above. Positive signals were visualised with 3 0 ,3 0 -diaminobenzidine tetrahydrochloride staining (Sigma). 2.6. Chromosomal mapping Preparation of agarose blocks, digestion of intact chromosomes with restriction enzymes, pulsed-®eld gel electrophoresis, and Southern hybridisations were performed according to previously described methods [13]. DNA probes for the sa20, sa35, and sa40 sequences were obtained by PCR ampli®cation with speci®c primers (see below). 2.7. Reverse transcriptase and PCR ampli®cation Total RNA was extracted from 2109 oocysts using the RNeasy mini puri®cation kit (Qiagen), following the manufacturer's protocol for yeast. The preparation was digested with DNAase RQ1 (Promega) for 1 h at 378C, extracted with phenol±chloroform, and precipitated with ethanol. Reverse transcriptase reaction was performed on 3 mg of total RNA with random hexamers as pri-
1927
mers, using the First-Strand cDNA Synthesis Kit (Pharmacia). The PCR ampli®cations consisted of 30 cycles (30 s at 948C; 60 s at 608C; 90 s at 728C) followed by one extension cycle (10 min at 728C) on a GeneAmp PCR System 2400 (PerkinElmer). Primers for the sa20 sequence were 5 0 AAAAGAAACCACATTGAGACTGAGC-3 0 and 5 0 -AATAGGGATTTTGGCGGAGG-3 0 . Primers for the sa35 sequence were 5 0 GAGAAGAAAACTAAAGCCAGCAG-3 0 and 5 0 -GAATGCAACAGTACCTGAAAGTAAG3 0. Primers for the sa40 sequence were 5 0 GACTGGAGAGCAAATCAAGGG-3 0 and 5 0 TTCCAGAATGATGAATCACGC-3 0 .
2.8. Protein puri®cation and immunoblotting analysis Puri®ed 6(His)-tagged peptides were obtained by anity chromatography on Ni-NTA agarose resin (Qiagen) and dialysed, according to methods described elsewhere [15]. Proteins were electrophoresed on 12% SDS±PAGE and transferred onto nitrocellulose. Blots were incubated for 1 h in a blocking buer (TNT buer plus 20% FCS) at room temperature, and then incubated for 1 h with dierent sera or mAb diluted in blocking buer. Seven serum samples were collected from immunocompetent persons during an outbreak of cryptosporidiosis that occurred in Northern Italy [16]; control serum samples were collected from 11 blood donors with no history of cryptosporidiosis. Rabbit and human sera were diluted 1:50, whereas the mAb for the 6His tag (Qiagen) was diluted 1:1000. Membranes were washed three times in washing buer. Speci®c HRP-conjugate antibodies (goat anti-rabbit, goat anti-human, and goat antimouse from BioRad) were diluted 1:3000 in blocking buer and incubated for 1 h with blots. Blots were washed three times and developed with 3 0 ,3 0 -diaminobenzidine tetrahydrochloride.
1928
F. Tosini et al. / International Journal for Parasitology 29 (1999) 1925±1933
3. Results 3.1. Isolation and puri®cation of three 6(His)tagged proteins from a genomic library of C. parvum The screening of the expression library with a rabbit serum speci®c for sporozoites revealed three positive clones. The expressed chimeric peptides were puri®ed by anity chromatography on Ni-NTA resin (Fig. 1). The approximate Mr values of these peptides were 20, 35, and 40 kDa; they were thus named `SA20', `SA35', and `SA40', respectively, whereas the corresponding clones were named `sa20', `sa35', and `sa40'. The SA35 preparation did not appear to be homogeneous, showing dierent bands (Fig. 1A, lane 4). However, the immunodetection with the speci®c antibody for the vector encoded epitope [MRGS 6(His) tag] on the whole lysate and on the puri®ed product (Fig. 1B, lanes 1, 2) demonstrated that the extra bands derived from the SA35 protein during the puri®cation process. 3.2. Characterisation of genomic clones encoding for SA20, SA35, and SA40 peptides Sequencing of the clones revealed the presence of three ORFs juxtaposed with the 6(His) encoding sequence. The sa20 clone contained an insert of 351 bp, which consisted of an uninterrupted ORF. The sa35 inserted sequence showed a continuous ORF of 906 bp. A search of the GenBank database did not reveal any sequences having homology with the sa20 and the sa35 sequences. The sa40 clone contained a large insert of approximately 6 kb, and 2506 bp at its 5 0 terminus were sequenced. The ®rst 750 bp constituted an ORF juxtaposed with the histidine tag. This coding sequence showed homology to two previously submitted C. parvum sequences (accession numbers: Y09042 and AF068065). Fig. 2 shows the aa sequences, including the vector encoded MRGS 6(His) tag, deduced from the corresponding inserted DNAs. To map the sequences on the C. parvum karyotype, which is composed of eight chromosomes distinguishable either by their size or by their
Fig. 1. Puri®cation of SA20, SA35, and SA40 peptides. (A) Twelve per cent SDS±PAGE. Lanes: M, molecular mass standards; 1, lysate from induced sa20 clone; 2, puri®ed SA20; 3, lysate from induced sa35 clone; 4, puri®ed SA35; 5, lysate from induced sa40 clone; 6, puri®ed SA40. (B) Immunoblot detected with antibody for MRGS 6(His) epitope showing the multiple bands of SA35. Lanes: 1, lysate from induced sa35 clone; 2, puri®ed SA35.
restriction fragments [13], the cloned sequences were hybridised to intact or to NotI- and S®Idigested chromosomes. The sa20 probe hybridised to chromosome VI of 1.36 Mbp and was associated with the 0.62 Mbp fragment, which originated by S®I digestion. The sa35 and sa40 probes hybridised to chromosome VII, and both were present on the 1.12 Mbp S®I fragment (Fig. 3). 3.3. Expression of mRNAs in the C. parvum oocysts The presence of mRNAs in the oocyst stage was assessed by RT±PCR assays using speci®c primer pairs for each cloned sequence. The expected amplicons, obtained using genomic DNA as template, are shown in Fig. 4. The mRNAs for sa35 and sa40 were present in the oocyst stage, whereas the mRNA for sa20 was not detected in this stage. To exclude genomic DNA carry over in the reverse transcribed template, the sample was tested in a PCR experiment with the primers described by Laxer et al. [17], which amplify a non-coding DNA sequence (Fig. 4, lanes marked `rep'). The positive ampli®cation with the RT template was detected using
F. Tosini et al. / International Journal for Parasitology 29 (1999) 1925±1933
1929
Fig. 2. Amino acid sequences of SA20, SA35, and SA40 as deduced from the DNA sequences of the sa20, sa35, and sa40 clones, respectively. The dierences observed at the C-terminus between the GP900 [20] and the SA40 proteins are shown. The arrow indicates an additional glycine residue found in the putative transmembrane motif (dotted line) of the SA40 protein. Underlined residues correspond to the totally dierent C-terminus of the SA40 protein caused by a single nucleotide insertion/deletion in the DNA encoding sequence.
primers speci®c for an ATPase gene of C. parvum (Fig. 4, lanes `ATPase'), which is expressed in the oocyst stage [18]. 3.4. Serologic responses to the antigenic peptides The Western blot results with rabbit serum samples were similar, regardless of the type of C. parvum antigen used. Thus, animals immunised with puri®ed sporozoites (Fig. 5, lanes 2 and 3) and those immunised with an oocyst-soluble antigen (Fig. 5, lane 4) both showed an immune response to the three puri®ed peptides. The human IgG response against the SA20 peptide appeared to exist independently of a C. parvum infection. The SA20 peptide was detected by 17 serum samples from persons with and without a history of cryptosporidiosis (data not
shown), and only one sample showed a very low reaction (Fig. 5, lane 5). In contrast, only serum samples from persons who had recently been infected showed a positive reaction to the SA35 and SA40 peptides (Fig. 5, lanes 7±13). All serum samples from infected persons recognised these two antigenic peptides, and smaller peptides originating from the processing of the SA35 major peptide were also detected (Fig. 5, lane 14).
4. Discussion The fact that the expression of recombinant proteins in our library was highly controlled facilitated the identi®cation of clones that expressed antigenic peptides. Furthermore, the
1930
F. Tosini et al. / International Journal for Parasitology 29 (1999) 1925±1933
Fig. 3. Chromosome mapping of sa20 and sa35 sequences. (A) Schematic reconstruction of the Cryptosporidium parvum karyotype [13]. (B) Southern blot hybridisation with the sa20 probe. (C) Southern blot hybridisation with the sa35 probe. In all panels, the left lane represents the intact chromosomes; the central lane, the NotI restriction pattern; and the right lane, the S®I restriction pattern. The sizes (in Mbp) of several Saccharomyces cerevisiae chromosomes are given on the left.
ectopic expression of C. parvum peptides fused with a 6-histidine tag allowed for the rapid isolation of the molecules of interest, avoiding the need for the subcloning steps. The sa20 and sa35 clones represent new C. parvum sequences encoding for antigenic proteins that do not show any similarity with other known proteins from other organisms. Our results suggest that the mRNAs for SA35 and SA40 are stored in the quiescent oocyst stage, and that these peptides could be synthesised during the excystation process. By contrast, since the mRNA encoding for SA20 is undetectable in the oocyst stage, the sa20 gene could be expressed before oocyst formation or later during sporulation. Since the mRNAs encoding for the
SA35 and the SA40 are present at the oocyst stage, these proteins may be involved in the invasion mechanism. However, the functions of the SA20 and the SA35 proteins are still unknown, and further studies are needed to identify their complete encoding sequences and to establish the cellular localisation of the proteins. Two recent studies have independently identi®ed the sequence encoding for SA40. In the ®rst report [19], this sequence was found to be a speci®c C. parvum cDNA clone, Cp3.4 (accession number: Y09042), from an oocyst±sporozoite expression library, and it was identi®ed as an oocyst protein. The 5 0 end of the sa40 clone was found to overlap with the 3 0 end of Cp3.4, thus including the SA40 antigen encoding sequence, and a perfect homology was observed at the aa level between these two sequences. The second study [20] demonstrated that Cp3.4 and, consequently, the sequence encoding for the sa40 clone, were strongly homologous to a genomic locus of 7.6 kb (accession number: AF068065). This locus corresponds to the single-copy gene encoding for a large glycoprotein (GP900) involved in the invasion of the host cell. This protein is localised in the micronemes and on the surface of the invasive forms of the parasite [20]. Although at the nucleotide level the sequences of sa40 and GP900 are almost identical, some mutations result in important dierences at the aa level (see Fig. 2). In fact, an additional codon results in the presence of an additional glycine residue in the putative transmembrane domain of SA40. Furthermore, a single nucleotide insertion/ deletion yields a frameshift of the encoding sequence, which results in a completely dierent C-terminus of the protein. These three sequences (GP900, Cp3.4, and sa40) were cloned from dierent isolates of animal origin, hence the dierences outlined above could re¯ect dierences among the dierent C. parvum isolates. It should be pointed out that GP900 is a highly immunogenic protein [21], and this could in part be due to the IgG response observed for the SA40 peptide, which is a fragment of GP900. Two (SA35 and SA40) of the three peptides identi®ed and characterised here are recognised by circulating antibodies as a consequence of a
F. Tosini et al. / International Journal for Parasitology 29 (1999) 1925±1933
1931
Fig. 4. Polymerase chain reaction ampli®cation of DNA (lanes D) and reverse-transcribed RNA (lanes R) from Cryptosporidium parvum oocysts. Lane M, molecular mass standards. All expected amplicons were obtained using genomic DNA as template, whereas only the sa35 and sa40 were ampli®ed from reverse-transcribed RNA. Lanes marked `rep' represent the negative control of the RT±PCR experiments, in that this is a non-coding repeated sequence [17]. Lanes marked `ATPase' represent the positive control, since the corresponding mRNA is known to be present in the oocyst stage [18].
recent C. parvum infection, whereas the serologic response to SA20 appears to be widely spread among humans irrespective of the presence of a recent infection. The immunoreactivity to some Cryptosporidium antigens in a large number of unaected individuals has recently been
reported [22, 23]. This could suggest a speci®c response to C. parvum antigens, since in immunocompetent persons the infection can be weak or completely asymptomatic [24]. Alternatively, C. parvum antigens may contain cross-reactive epitopes. The picture of seroprevalence among
Fig. 5. Western blots with the puri®ed SA20, SA35, and SA40 peptides. Antigens were probed with rabbit and human serum samples and with monoclonal antibody anti-MRGS 6(His). Upper panel, blots with SA20; central panel, blots with SA35; lower panel, blots with SA40. In all panels: lane 1, germ-free naive rabbit serum; lanes 2 and 3, serum from sporozoite-immunised rabbits; lane 4, serum from a rabbit immunised with an oocyst-soluble antigen; lanes 5 and 6, serum samples from healthy blood donors; lanes 7±13, serum samples from persons with cryptosporidiosis; lane 14, mAb for MRGS 6(His).
1932
F. Tosini et al. / International Journal for Parasitology 29 (1999) 1925±1933
infected and healthy humans has been a topic of debate, since data on serological response are strongly in¯uenced by the speci®c assay used for the evaluation of speci®c antibodies (i.e. ELISA or Western blot) and by the class of antibody detected (i.e. IgM, IgG, or IgA) [1, 25].
[10]
[11]
[12]
Acknowledgements [13]
This work was supported by the National AIDS research program, Ministero della SanitaÁ, Istituto Superiore di SanitaÁ, Rome, Italy (Grant No. 50B/F)
[14]
References
[15]
[1] Okhuysen PC, Chappell CL, Sterling CR, Jakubowski W, DuPont HL. Susceptibility and serologic response of healthy adults to reinfection with Cryptosporidium parvum. Infect Immun 1998;66:441±3. [2] Arrowood MJ, Mead JR, Mahrt JL, Sterling CR. Eects of immune colostrum and orally administered antisporozoite monoclonal antibodies on the outcome of Cryptosporidium parvum infections in neonatal mice. Infect Immun 1989;57:2283±8. [3] Okhuysen PC, Chappell CL, Crabb J, Valdez LM, Douglass ET, DuPont HL. Prophylactic eect of bovine anti-Cryptosporidium hyperimmune colostrum immunoglobulin in healthy volunteers challenged with Cryptosporidium parvum. Clin Infect Dis 1998;26:1324±9. [4] Enriquez FJ, Riggs MW. Role of immunoglobulin A monoclonal antibodies against P23 in controlling murine Cryptosporidium parvum infection. Infect Immun 1998;66:4469±73. [5] Steiner TS, Guerrant RL. The pathogenesis and host response to Cryptosporidium parvum. Curr Opin Infect Dis 1996;9:156±60. [6] Lally NC, Baird GD, McQuay SJ, Wright F, Oliver JJ. A 2359-base pair DNA fragment from Cryptosporidium parvum encoding a repetitive oocyst protein. Mol Biochem Parasitol 1992;56:69±78. [7] Petersen C, Gut J, Leech JH, Nelson RG. Identi®cation and initial characterization of ®ve Cryptosporidium parvum sporozoite antigen genes. Infect Immun 1992;60:2343±8. [8] Fayer R, Speer CA, Dubey JP. General biology of Cryptosporidium. In: Fayer R, editor. Cryptosporidium and cryptosporidiosis. Boca Raton: CRC Press, 1997;1± 42. [9] Perryman LE, Jasmer DP, Riggs MW, Bohnet SG, McGuire TC, Arrowood MJ. A cloned gene of
[16]
[17]
[18]
[19]
[20]
[21]
[22]
Cryptosporidium parvum encodes neutralization-sensitive epitopes. Mol Biochem Parasitol 1996;80:137±47. Sambrook J, Fritsch EF, Maniatis T. Molecular cloningÐa laboratory manual, 2nd ed. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press, 1989. Sanger F, Nicklen S, Coulson AR. DNA sequencing with chain-terminating inhibitors. Proc Natl Acad Sci USA 1977;74:5463±7. Arrowood M, Sterling CR. Isolation of Cryptosporidium parvum oocysts and sporozoites using discontinuous sucrose and isopycnic Percoll gradients. J Parasitol 1987;73:314±9. CaccioÁ S, Camilli R, La Rosa G, Pozio E. Establishing the Cryptosporidium parvum karyotype by NotI and S®I restriction analysis and Southern hybridization. Gene 1998;219:73±9. Gomez Morales MA, Pozio E, Croppo GP. Serodiagnosis of cryptosporidiosis in Italian HIV-positive patients by means of an oocyst soluble antigen in an ELISA. J Infect 1992;25:229±36. Tosini F, Venanzi S, Boschi A, Battaglia PA. The uvp1 gene on the R46 plasmid encodes a resolvase that catalyzes site-speci®c resolution involving the 5 0 conserved segment of the adjacent integron In1. Mol Gen Genet 1998;258:404±11. Pozio E, Rezza G, Boschini A et al. Clinical cryptosporidiosis and human immunode®ciency virus (HIV)-induced immunosuppression: ®ndings from a longitudinal study of HIV-positive and HIV-negative former injection drug users. J Infect Dis 1997;176:969±75. Laxer MA, Timblin BK, Patel RJ. DNA sequences for the speci®c detection of Cryptosporidium parvum by the polymerase chain reaction. Am J Trop Med Hyg 1991;45:688±94. Zhu G, Keithly JS. Molecular analysis of a P-type ATPase from Cryptosporidium parvum. Mol Biochem Parasitol 1997;90:307±16. Petry F, Shirley MW, Miles MA, McDonald V. Characterisation of a Cryptosporidium parvum-speci®c cDNA clone and detection of parasite DNA in mucosal scrapings of infected mice. Mol Biochem Parasitol 1998;95:21±31. Barnes DA, Bonnin A, Huang JX et al. A novel multidomain mucin-like glycoprotein of Cryptosporidium parvum mediates invasion. Mol Biochem Parasitol 1998;96:93±110. Petersen C, Gut J, Doyle PS, Crabb JH, Nelson RG, Leech JH. Characterization of a >900 000-Mr Cryptosporidium parvum sporozoite glycoprotein recognized by protective hyperimmune bovine colostral immunoglobulin. Infect Immun 1992;60:5132±8. Chappell CL, Okhuysen PC, Sterling CR, Wang C, Jakubowski W, Dupont HL. Infectivity of Cryptosporidium parvum in healthy adults with pre-existing anti-C. parvum serum immunoglobulin G. Am J Trop Med Hyg 1999;60:157±64.
F. Tosini et al. / International Journal for Parasitology 29 (1999) 1925±1933 [23] Moss DM, Chappell CL, Okhuysen PC et al. The antibody response to 27-, 17-, and 15-kDa Cryptosporidium antigens following experimental infection in humans. J Infect Dis 1998;178:827±33. [24] Frost FJ, Calderon RL, Muller TB et al. A two-year follow-up survey of antibody to Cryptosporidium in Jackson
1933
County, Oregon following an outbreak of waterborne disease. Epidemiol Infect 1998;121:213±7. [25] Frost FJ, de la Cruz AA, Moss DM, Curry M, Calderon RL. Comparisons of ELISA and Western blot assays for detection of Cryptosporidium antibody. Epidemiol Infect 1998;121:205±11.
Identi®cation and characterisation of three antigenic proteins from Cryptosporidium parvum sporozoites using a p DNA library expressing poly-histidine tagged peptides Fabio Tosini *, Simone CaccioÁ, Alessandra Tamburrini, Giuseppe La Rosa, Edoardo Pozio Laboratory of Parasitology, Istituto Superiore di SanitaÁ, viale Regina Elena 299, 00161 Rome, Italy Received 16 July 1999; received in revised form 6 September 1999; accepted 7 September 1999
Abstract To identify antigenic peptides of the parasite Cryptosporidium parvum, an expression library that allows for the production of chimeric proteins fused with a 6-histidine tag was made. The library was screened with C. parvum sporozoite rabbit anti-serum, and three positive clones (sa20, sa35, and sa40) were identi®ed. The corresponding recombinant proteins (SA20, SA35, and SA40) were expressed in Escherichia coli and puri®ed by metal-anity chromatography. The sequence of sa20 and sa35 clones did not show any homology with known genes or proteins, whereas the 5 0 end of the sa40 clone showed homology with two previously identi®ed C. parvum sequences. Hybridisations to intact chromosomes fractionated by pulsed-®eld gel electrophoresis revealed that the sa35 and sa40 sequences are localised on chromosome VII, whereas the sa20 sequence is localised on chromosome VI. Reverse transcriptase±PCR experiments showed the presence of mRNAs for sa35 and sa40 in the oocyst, whereas the sa20 mRNA was undetectable in this stage. The serological response to the three proteins was assayed in C. parvum-immunised rabbits and in immunocompetent individuals with cryptosporidiosis. The Western blot results indicated that rabbits, challenged with a sporozoite crude antigen or with an oocyst crude antigen, were highly responsive to these three antigens. Human serum samples showed a response to the three proteins, although the response to SA20 appeared to be unrelated to a recent C. parvum infection. These results suggest that the SA35 and the SA40 proteins may be useful in detecting C. parvum infections. # 1999 Australian Society for Parasitology Inc. Published by Elsevier Science Ltd. All rights reserved. Keywords: Antigens; Cryptosporidium parvum; Genomic library; Immuno-detection; Poly-histidine tagged peptides; RT±PCR
1. Introduction Note: Nucleotide sequences reported in this paper are available in the GenBank2, EMBL and DDBJ databases under the accession numbers AJ006592 (sa20), AJ006593 (sa35), and AJ132769 (sa40). * Corresponding author. Present address: Laboratory of Cellular Biology, Istituto Superiore di SanitaÁ, viale Regina Elena 299, 00161 Rome, Italy. Tel.: +39-06-4990-2308; fax: +39-06-4938-7065 E-mail address: [email protected] (F. Tosini). p
Cryptosporidium parvum is a protozoan parasite that causes enteritis in humans and in other mammalian species. Infection prevalently occurs in the gut epithelial mucosa, provoking severe diarrhoea, which, in an immunocompetent host, can resolve in about 10 days. Although infected humans can develop a protective immunore-
0020-7519/99/$20.00 # 1999 Australian Society for Parasitology Inc. Published by Elsevier Science Ltd. All rights reserved. PII: S 0 0 2 0 - 7 5 1 9 ( 9 9 ) 0 0 1 5 8 - 7
1926
F. Tosini et al. / International Journal for Parasitology 29 (1999) 1925±1933
sponse to C. parvum, they are susceptible to reinfection [1]. In immunocompromised hosts, such as persons with AIDS, cryptosporidiosis is a common cause of chronic diarrhoea, which may lead to death. Given the lack of an eective chemotherapy for C. parvum infection, several attempts have been made to develop passive immunisation. Positive results were obtained with the oral administration of immune colostrum in neonatal mice [2] and in humans [3]. The use of mAbs was also found to be eective in controlling C. parvum infection in mice [4]. The molecular basis of the pathogenesis and the molecules involved in the invasion mechanism of the host cell are poorly understood [5]. Moreover, few genes encoding for immunogenic proteins involved in the host±parasite interaction have been characterised. Therefore, the identi®cation of novel molecules recognised by the host immune system may contribute to understanding the mechanism of the infection. Identi®cation of some C. parvum antigens was successfully achieved with the immunological screening of expression libraries which express fusion proteins [6, 7]. The objective of our study was to identify and characterise antigenic molecules speci®c to the sporozoite stage. Sporozoites are responsible for the host cell invasion, and are free in the intestinal lumen before penetrating into the enterocyte [8]; thus, they can be blocked by the host immunoresponse [9].
2. Methods and materials 2.1. General microbiological and DNA recombinant techniques Microbiological techniques and media, preparation of plasmid DNA and isolation of restricted fragments all followed standard procedures [10]. DNA sequencing was performed using Sanger's procedure [11].
2.2. DNA extraction and construction of a genomic C. parvum expression library Puri®ed oocysts of C. parvum were obtained from infected calves as described previously [12]. The isolate (code ISSC6) originated from a naturally infected calf from Denmark. Total genomic DNA was prepared using previously described methods [13]. After partial digestion with BstyI endonuclease, DNA fragments were ligated in the BamHI site of the cloning vectors pQE30, pQE31, and pQE32 (Qiagen). The DNA fragments were inserted in three dierent vectors, which correspond to the three dierent open reading frames (ORF) for each cloning site, in order to increase the probability of obtaining the desired juxtapositions of the inserts with the 6His tag. The ligation reaction was used to transform Escherichia coli M15 cells, and approximately 3105 cells were plated on Luria Bertani agar plates with 50 mg ml ÿ 1 of ampicillin plus 25 mg ml ÿ 1 kanamycin to amplify single clones. Bacterial colonies were resuspended in LB medium using a rotating mixer; 50% glycerol was added and aliquots were stored at ÿ808C. The ®nal titer of the library corresponded to 21011 cfu ml ÿ 1 and recombinant colonies were estimated to represent 85% of the total. 2.3. Sporozoite puri®cation Puri®ed oocysts were suspended at 1107 ml ÿ 1 in ABS buer (0.1 M sodium acetate, pH 5.5; 0.145 M NaCl), washed, resuspended in ABS buer and placed on ice for 20 min. Oocysts were washed twice with PBS buer, resuspended in excystation buer (MEM medium plus 0.75% w/ v sodium taurocholate) and incubated at 378C for 30±60 min (excystation was followed by microscopy). The excystation mixture was then loaded onto a 50 ml syringe connected to a 25 mm micro®ltration device with a 3 mm ®lter (Costar Nuclepore) and the elution of puri®ed sporozoites was obtained by gravity ¯ow. Sporozoites were collected by centrifugation, resuspended in MEM medium and stored at 48C for 14±18 h prior to immunisation.
F. Tosini et al. / International Journal for Parasitology 29 (1999) 1925±1933
2.4. Rabbit immunisation Two New Zealand germ-free rabbits were inoculated i.v. without adjuvant with approximately 1108 sporozoites per inoculum. The challenge was repeated three times at intervals of 20 days, and the blood serum was collected 2 weeks after the last immunisation. Two other New Zealand germ-free rabbits were inoculated with an oocyst-soluble antigen as previously described [14]. Serum aliquots were stored at ÿ808C. 2.5. Library immunoscreening The expression library was screened using standard procedures [10]. Filters were washed three times with washing buer (TNT buer plus 0.1% BSA and 0.1% nonidet NP40), incubated with HRP-GAR antibody (BioRad) diluted 1:3000 for 1 h, and washed as above. Positive signals were visualised with 3 0 ,3 0 -diaminobenzidine tetrahydrochloride staining (Sigma). 2.6. Chromosomal mapping Preparation of agarose blocks, digestion of intact chromosomes with restriction enzymes, pulsed-®eld gel electrophoresis, and Southern hybridisations were performed according to previously described methods [13]. DNA probes for the sa20, sa35, and sa40 sequences were obtained by PCR ampli®cation with speci®c primers (see below). 2.7. Reverse transcriptase and PCR ampli®cation Total RNA was extracted from 2109 oocysts using the RNeasy mini puri®cation kit (Qiagen), following the manufacturer's protocol for yeast. The preparation was digested with DNAase RQ1 (Promega) for 1 h at 378C, extracted with phenol±chloroform, and precipitated with ethanol. Reverse transcriptase reaction was performed on 3 mg of total RNA with random hexamers as pri-
1927
mers, using the First-Strand cDNA Synthesis Kit (Pharmacia). The PCR ampli®cations consisted of 30 cycles (30 s at 948C; 60 s at 608C; 90 s at 728C) followed by one extension cycle (10 min at 728C) on a GeneAmp PCR System 2400 (PerkinElmer). Primers for the sa20 sequence were 5 0 AAAAGAAACCACATTGAGACTGAGC-3 0 and 5 0 -AATAGGGATTTTGGCGGAGG-3 0 . Primers for the sa35 sequence were 5 0 GAGAAGAAAACTAAAGCCAGCAG-3 0 and 5 0 -GAATGCAACAGTACCTGAAAGTAAG3 0. Primers for the sa40 sequence were 5 0 GACTGGAGAGCAAATCAAGGG-3 0 and 5 0 TTCCAGAATGATGAATCACGC-3 0 .
2.8. Protein puri®cation and immunoblotting analysis Puri®ed 6(His)-tagged peptides were obtained by anity chromatography on Ni-NTA agarose resin (Qiagen) and dialysed, according to methods described elsewhere [15]. Proteins were electrophoresed on 12% SDS±PAGE and transferred onto nitrocellulose. Blots were incubated for 1 h in a blocking buer (TNT buer plus 20% FCS) at room temperature, and then incubated for 1 h with dierent sera or mAb diluted in blocking buer. Seven serum samples were collected from immunocompetent persons during an outbreak of cryptosporidiosis that occurred in Northern Italy [16]; control serum samples were collected from 11 blood donors with no history of cryptosporidiosis. Rabbit and human sera were diluted 1:50, whereas the mAb for the 6His tag (Qiagen) was diluted 1:1000. Membranes were washed three times in washing buer. Speci®c HRP-conjugate antibodies (goat anti-rabbit, goat anti-human, and goat antimouse from BioRad) were diluted 1:3000 in blocking buer and incubated for 1 h with blots. Blots were washed three times and developed with 3 0 ,3 0 -diaminobenzidine tetrahydrochloride.
1928
F. Tosini et al. / International Journal for Parasitology 29 (1999) 1925±1933
3. Results 3.1. Isolation and puri®cation of three 6(His)tagged proteins from a genomic library of C. parvum The screening of the expression library with a rabbit serum speci®c for sporozoites revealed three positive clones. The expressed chimeric peptides were puri®ed by anity chromatography on Ni-NTA resin (Fig. 1). The approximate Mr values of these peptides were 20, 35, and 40 kDa; they were thus named `SA20', `SA35', and `SA40', respectively, whereas the corresponding clones were named `sa20', `sa35', and `sa40'. The SA35 preparation did not appear to be homogeneous, showing dierent bands (Fig. 1A, lane 4). However, the immunodetection with the speci®c antibody for the vector encoded epitope [MRGS 6(His) tag] on the whole lysate and on the puri®ed product (Fig. 1B, lanes 1, 2) demonstrated that the extra bands derived from the SA35 protein during the puri®cation process. 3.2. Characterisation of genomic clones encoding for SA20, SA35, and SA40 peptides Sequencing of the clones revealed the presence of three ORFs juxtaposed with the 6(His) encoding sequence. The sa20 clone contained an insert of 351 bp, which consisted of an uninterrupted ORF. The sa35 inserted sequence showed a continuous ORF of 906 bp. A search of the GenBank database did not reveal any sequences having homology with the sa20 and the sa35 sequences. The sa40 clone contained a large insert of approximately 6 kb, and 2506 bp at its 5 0 terminus were sequenced. The ®rst 750 bp constituted an ORF juxtaposed with the histidine tag. This coding sequence showed homology to two previously submitted C. parvum sequences (accession numbers: Y09042 and AF068065). Fig. 2 shows the aa sequences, including the vector encoded MRGS 6(His) tag, deduced from the corresponding inserted DNAs. To map the sequences on the C. parvum karyotype, which is composed of eight chromosomes distinguishable either by their size or by their
Fig. 1. Puri®cation of SA20, SA35, and SA40 peptides. (A) Twelve per cent SDS±PAGE. Lanes: M, molecular mass standards; 1, lysate from induced sa20 clone; 2, puri®ed SA20; 3, lysate from induced sa35 clone; 4, puri®ed SA35; 5, lysate from induced sa40 clone; 6, puri®ed SA40. (B) Immunoblot detected with antibody for MRGS 6(His) epitope showing the multiple bands of SA35. Lanes: 1, lysate from induced sa35 clone; 2, puri®ed SA35.
restriction fragments [13], the cloned sequences were hybridised to intact or to NotI- and S®Idigested chromosomes. The sa20 probe hybridised to chromosome VI of 1.36 Mbp and was associated with the 0.62 Mbp fragment, which originated by S®I digestion. The sa35 and sa40 probes hybridised to chromosome VII, and both were present on the 1.12 Mbp S®I fragment (Fig. 3). 3.3. Expression of mRNAs in the C. parvum oocysts The presence of mRNAs in the oocyst stage was assessed by RT±PCR assays using speci®c primer pairs for each cloned sequence. The expected amplicons, obtained using genomic DNA as template, are shown in Fig. 4. The mRNAs for sa35 and sa40 were present in the oocyst stage, whereas the mRNA for sa20 was not detected in this stage. To exclude genomic DNA carry over in the reverse transcribed template, the sample was tested in a PCR experiment with the primers described by Laxer et al. [17], which amplify a non-coding DNA sequence (Fig. 4, lanes marked `rep'). The positive ampli®cation with the RT template was detected using
F. Tosini et al. / International Journal for Parasitology 29 (1999) 1925±1933
1929
Fig. 2. Amino acid sequences of SA20, SA35, and SA40 as deduced from the DNA sequences of the sa20, sa35, and sa40 clones, respectively. The dierences observed at the C-terminus between the GP900 [20] and the SA40 proteins are shown. The arrow indicates an additional glycine residue found in the putative transmembrane motif (dotted line) of the SA40 protein. Underlined residues correspond to the totally dierent C-terminus of the SA40 protein caused by a single nucleotide insertion/deletion in the DNA encoding sequence.
primers speci®c for an ATPase gene of C. parvum (Fig. 4, lanes `ATPase'), which is expressed in the oocyst stage [18]. 3.4. Serologic responses to the antigenic peptides The Western blot results with rabbit serum samples were similar, regardless of the type of C. parvum antigen used. Thus, animals immunised with puri®ed sporozoites (Fig. 5, lanes 2 and 3) and those immunised with an oocyst-soluble antigen (Fig. 5, lane 4) both showed an immune response to the three puri®ed peptides. The human IgG response against the SA20 peptide appeared to exist independently of a C. parvum infection. The SA20 peptide was detected by 17 serum samples from persons with and without a history of cryptosporidiosis (data not
shown), and only one sample showed a very low reaction (Fig. 5, lane 5). In contrast, only serum samples from persons who had recently been infected showed a positive reaction to the SA35 and SA40 peptides (Fig. 5, lanes 7±13). All serum samples from infected persons recognised these two antigenic peptides, and smaller peptides originating from the processing of the SA35 major peptide were also detected (Fig. 5, lane 14).
4. Discussion The fact that the expression of recombinant proteins in our library was highly controlled facilitated the identi®cation of clones that expressed antigenic peptides. Furthermore, the
1930
F. Tosini et al. / International Journal for Parasitology 29 (1999) 1925±1933
Fig. 3. Chromosome mapping of sa20 and sa35 sequences. (A) Schematic reconstruction of the Cryptosporidium parvum karyotype [13]. (B) Southern blot hybridisation with the sa20 probe. (C) Southern blot hybridisation with the sa35 probe. In all panels, the left lane represents the intact chromosomes; the central lane, the NotI restriction pattern; and the right lane, the S®I restriction pattern. The sizes (in Mbp) of several Saccharomyces cerevisiae chromosomes are given on the left.
ectopic expression of C. parvum peptides fused with a 6-histidine tag allowed for the rapid isolation of the molecules of interest, avoiding the need for the subcloning steps. The sa20 and sa35 clones represent new C. parvum sequences encoding for antigenic proteins that do not show any similarity with other known proteins from other organisms. Our results suggest that the mRNAs for SA35 and SA40 are stored in the quiescent oocyst stage, and that these peptides could be synthesised during the excystation process. By contrast, since the mRNA encoding for SA20 is undetectable in the oocyst stage, the sa20 gene could be expressed before oocyst formation or later during sporulation. Since the mRNAs encoding for the
SA35 and the SA40 are present at the oocyst stage, these proteins may be involved in the invasion mechanism. However, the functions of the SA20 and the SA35 proteins are still unknown, and further studies are needed to identify their complete encoding sequences and to establish the cellular localisation of the proteins. Two recent studies have independently identi®ed the sequence encoding for SA40. In the ®rst report [19], this sequence was found to be a speci®c C. parvum cDNA clone, Cp3.4 (accession number: Y09042), from an oocyst±sporozoite expression library, and it was identi®ed as an oocyst protein. The 5 0 end of the sa40 clone was found to overlap with the 3 0 end of Cp3.4, thus including the SA40 antigen encoding sequence, and a perfect homology was observed at the aa level between these two sequences. The second study [20] demonstrated that Cp3.4 and, consequently, the sequence encoding for the sa40 clone, were strongly homologous to a genomic locus of 7.6 kb (accession number: AF068065). This locus corresponds to the single-copy gene encoding for a large glycoprotein (GP900) involved in the invasion of the host cell. This protein is localised in the micronemes and on the surface of the invasive forms of the parasite [20]. Although at the nucleotide level the sequences of sa40 and GP900 are almost identical, some mutations result in important dierences at the aa level (see Fig. 2). In fact, an additional codon results in the presence of an additional glycine residue in the putative transmembrane domain of SA40. Furthermore, a single nucleotide insertion/ deletion yields a frameshift of the encoding sequence, which results in a completely dierent C-terminus of the protein. These three sequences (GP900, Cp3.4, and sa40) were cloned from dierent isolates of animal origin, hence the dierences outlined above could re¯ect dierences among the dierent C. parvum isolates. It should be pointed out that GP900 is a highly immunogenic protein [21], and this could in part be due to the IgG response observed for the SA40 peptide, which is a fragment of GP900. Two (SA35 and SA40) of the three peptides identi®ed and characterised here are recognised by circulating antibodies as a consequence of a
F. Tosini et al. / International Journal for Parasitology 29 (1999) 1925±1933
1931
Fig. 4. Polymerase chain reaction ampli®cation of DNA (lanes D) and reverse-transcribed RNA (lanes R) from Cryptosporidium parvum oocysts. Lane M, molecular mass standards. All expected amplicons were obtained using genomic DNA as template, whereas only the sa35 and sa40 were ampli®ed from reverse-transcribed RNA. Lanes marked `rep' represent the negative control of the RT±PCR experiments, in that this is a non-coding repeated sequence [17]. Lanes marked `ATPase' represent the positive control, since the corresponding mRNA is known to be present in the oocyst stage [18].
recent C. parvum infection, whereas the serologic response to SA20 appears to be widely spread among humans irrespective of the presence of a recent infection. The immunoreactivity to some Cryptosporidium antigens in a large number of unaected individuals has recently been
reported [22, 23]. This could suggest a speci®c response to C. parvum antigens, since in immunocompetent persons the infection can be weak or completely asymptomatic [24]. Alternatively, C. parvum antigens may contain cross-reactive epitopes. The picture of seroprevalence among
Fig. 5. Western blots with the puri®ed SA20, SA35, and SA40 peptides. Antigens were probed with rabbit and human serum samples and with monoclonal antibody anti-MRGS 6(His). Upper panel, blots with SA20; central panel, blots with SA35; lower panel, blots with SA40. In all panels: lane 1, germ-free naive rabbit serum; lanes 2 and 3, serum from sporozoite-immunised rabbits; lane 4, serum from a rabbit immunised with an oocyst-soluble antigen; lanes 5 and 6, serum samples from healthy blood donors; lanes 7±13, serum samples from persons with cryptosporidiosis; lane 14, mAb for MRGS 6(His).
1932
F. Tosini et al. / International Journal for Parasitology 29 (1999) 1925±1933
infected and healthy humans has been a topic of debate, since data on serological response are strongly in¯uenced by the speci®c assay used for the evaluation of speci®c antibodies (i.e. ELISA or Western blot) and by the class of antibody detected (i.e. IgM, IgG, or IgA) [1, 25].
[10]
[11]
[12]
Acknowledgements [13]
This work was supported by the National AIDS research program, Ministero della SanitaÁ, Istituto Superiore di SanitaÁ, Rome, Italy (Grant No. 50B/F)
[14]
References
[15]
[1] Okhuysen PC, Chappell CL, Sterling CR, Jakubowski W, DuPont HL. Susceptibility and serologic response of healthy adults to reinfection with Cryptosporidium parvum. Infect Immun 1998;66:441±3. [2] Arrowood MJ, Mead JR, Mahrt JL, Sterling CR. Eects of immune colostrum and orally administered antisporozoite monoclonal antibodies on the outcome of Cryptosporidium parvum infections in neonatal mice. Infect Immun 1989;57:2283±8. [3] Okhuysen PC, Chappell CL, Crabb J, Valdez LM, Douglass ET, DuPont HL. Prophylactic eect of bovine anti-Cryptosporidium hyperimmune colostrum immunoglobulin in healthy volunteers challenged with Cryptosporidium parvum. Clin Infect Dis 1998;26:1324±9. [4] Enriquez FJ, Riggs MW. Role of immunoglobulin A monoclonal antibodies against P23 in controlling murine Cryptosporidium parvum infection. Infect Immun 1998;66:4469±73. [5] Steiner TS, Guerrant RL. The pathogenesis and host response to Cryptosporidium parvum. Curr Opin Infect Dis 1996;9:156±60. [6] Lally NC, Baird GD, McQuay SJ, Wright F, Oliver JJ. A 2359-base pair DNA fragment from Cryptosporidium parvum encoding a repetitive oocyst protein. Mol Biochem Parasitol 1992;56:69±78. [7] Petersen C, Gut J, Leech JH, Nelson RG. Identi®cation and initial characterization of ®ve Cryptosporidium parvum sporozoite antigen genes. Infect Immun 1992;60:2343±8. [8] Fayer R, Speer CA, Dubey JP. General biology of Cryptosporidium. In: Fayer R, editor. Cryptosporidium and cryptosporidiosis. Boca Raton: CRC Press, 1997;1± 42. [9] Perryman LE, Jasmer DP, Riggs MW, Bohnet SG, McGuire TC, Arrowood MJ. A cloned gene of
[16]
[17]
[18]
[19]
[20]
[21]
[22]
Cryptosporidium parvum encodes neutralization-sensitive epitopes. Mol Biochem Parasitol 1996;80:137±47. Sambrook J, Fritsch EF, Maniatis T. Molecular cloningÐa laboratory manual, 2nd ed. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press, 1989. Sanger F, Nicklen S, Coulson AR. DNA sequencing with chain-terminating inhibitors. Proc Natl Acad Sci USA 1977;74:5463±7. Arrowood M, Sterling CR. Isolation of Cryptosporidium parvum oocysts and sporozoites using discontinuous sucrose and isopycnic Percoll gradients. J Parasitol 1987;73:314±9. CaccioÁ S, Camilli R, La Rosa G, Pozio E. Establishing the Cryptosporidium parvum karyotype by NotI and S®I restriction analysis and Southern hybridization. Gene 1998;219:73±9. Gomez Morales MA, Pozio E, Croppo GP. Serodiagnosis of cryptosporidiosis in Italian HIV-positive patients by means of an oocyst soluble antigen in an ELISA. J Infect 1992;25:229±36. Tosini F, Venanzi S, Boschi A, Battaglia PA. The uvp1 gene on the R46 plasmid encodes a resolvase that catalyzes site-speci®c resolution involving the 5 0 conserved segment of the adjacent integron In1. Mol Gen Genet 1998;258:404±11. Pozio E, Rezza G, Boschini A et al. Clinical cryptosporidiosis and human immunode®ciency virus (HIV)-induced immunosuppression: ®ndings from a longitudinal study of HIV-positive and HIV-negative former injection drug users. J Infect Dis 1997;176:969±75. Laxer MA, Timblin BK, Patel RJ. DNA sequences for the speci®c detection of Cryptosporidium parvum by the polymerase chain reaction. Am J Trop Med Hyg 1991;45:688±94. Zhu G, Keithly JS. Molecular analysis of a P-type ATPase from Cryptosporidium parvum. Mol Biochem Parasitol 1997;90:307±16. Petry F, Shirley MW, Miles MA, McDonald V. Characterisation of a Cryptosporidium parvum-speci®c cDNA clone and detection of parasite DNA in mucosal scrapings of infected mice. Mol Biochem Parasitol 1998;95:21±31. Barnes DA, Bonnin A, Huang JX et al. A novel multidomain mucin-like glycoprotein of Cryptosporidium parvum mediates invasion. Mol Biochem Parasitol 1998;96:93±110. Petersen C, Gut J, Doyle PS, Crabb JH, Nelson RG, Leech JH. Characterization of a >900 000-Mr Cryptosporidium parvum sporozoite glycoprotein recognized by protective hyperimmune bovine colostral immunoglobulin. Infect Immun 1992;60:5132±8. Chappell CL, Okhuysen PC, Sterling CR, Wang C, Jakubowski W, Dupont HL. Infectivity of Cryptosporidium parvum in healthy adults with pre-existing anti-C. parvum serum immunoglobulin G. Am J Trop Med Hyg 1999;60:157±64.
F. Tosini et al. / International Journal for Parasitology 29 (1999) 1925±1933 [23] Moss DM, Chappell CL, Okhuysen PC et al. The antibody response to 27-, 17-, and 15-kDa Cryptosporidium antigens following experimental infection in humans. J Infect Dis 1998;178:827±33. [24] Frost FJ, Calderon RL, Muller TB et al. A two-year follow-up survey of antibody to Cryptosporidium in Jackson
1933
County, Oregon following an outbreak of waterborne disease. Epidemiol Infect 1998;121:213±7. [25] Frost FJ, de la Cruz AA, Moss DM, Curry M, Calderon RL. Comparisons of ELISA and Western blot assays for detection of Cryptosporidium antibody. Epidemiol Infect 1998;121:205±11.