BackgroundSchistosomiasis is a major parasitic disease affecting >200 million people in the developing world, and 400 million people are at risk for infection. This study aimed to identify and compare proteins recognized by serum samples from schistosome-exposed individuals before and after curative praziquantel treatment
MethodsProteins recognized by pooled serum samples from Schistosoma haematobium–exposed Zimbabweans were determined by 2-dimensional Western blotting and identified by mass spectrometry
ResultsSerum samples recognized 71 spots, which resolved to 26 different characterized proteins. Eleven of these proteins have not previously been shown to be immunogenic in natural human infection or in experimental models of schistosomiasis, making them novel antigens in the parasite. Pretreatment serum samples recognized 59 spots, which resolved to 21 different identified proteins. Posttreatment serum samples recognized an additional 12 spots, which resolved to 8 different identified proteins. Of these 8 proteins, 3 had putative isoforms recognized before treatment, and 5 (calreticulin, tropomyosin 1, tropomyosin 2, paramyosin, and triose phosphate isomerase) did not
ConclusionsThis study is the most comprehensive characterization of S. haematobium antigens to date and describes novel antigens in all schistosome species. Posttreatment results are consistent with praziquantel treatment inducing quantitative and qualitative changes in schistosome-specific antibody responses
Schistosomiasis is second to malaria in public health importance [1] in tropical and subtropical countries in Africa, the Middle East, and South America. Schistosoma haematobium the causative agent of urinary schistosomiasis, is primarily an African parasite and is found in 53 countries in the Middle East and Africa, including the islands of Madagascar and Mauritius. A recent survey of sub-Saharan Africa indicated that, of 682 million individuals, 70 million had hematuria and 32 million had dysuria associated with S. haematobium infection [2]. Furthermore, it was estimated that 18 million individuals had pathological changes in the bladder wall, and 10 million individuals had hydronephrosis
Schistosomes induce variable levels of resistance to reinfection in humans and other animals [3 –9]. The development of naturally acquired immunity against schistosomes is slow. This has been attributed partly to the need for the immune system to be exposed to sufficient parasite antigens and partly to effective immune avoidance mechanisms by the parasites [10]. It is therefore important to characterize and study the parasite proteins that interact with the host’s immune system and the outcome of that interaction. The primary strategy for control of schistosomiasis is treatment of infected individuals with antihelminth drugs. Praziquantel is widely used and is effective against the 3 primary schistosome species affecting humans (i.e., S. mansoni, S. japonicum and S. haematobium), whereas oxamniquine is effective against S. mansoni only. Although these drugs are effective, there is a continuing search—driven partly by concern over the development of drug resistance and partly by the desire for a preventative rather than a curative intervention [11]—for alternative or complementary methods of control, ranging from molluscides (to kill the intermediate snail host) to vaccine development
Several studies have identified schistosome immunogenic proteins by screening expression libraries with serum samples from infected or vaccinated animals [12 –16]. However, there are limitations associated with this approach: for example, it cannot detect immunogenic epitopes arising from posttranslation modifications. The proteomic approach uses native antigens and can readily incorporate serological reactivity through Western blot techniques [17]. The recent publication of S. mansoni and S. japonicum expressed sequence tag (EST) data [18, 19] has allowed proteomic technology to be used for the identification of schistosome proteins [20] and to be systematically applied for the first time in the identification of schistosome antigens. We report here the application of proteomic procedures to the characterization of immunogenic proteins in adult male and female S. haematobium worms. Of the 3 primary schistosomes that infect humans, this species is the least studied from an immunological perspective. For example, the large field study of 10 vaccine candidate proteins recently conducted by the World Health Organization (WHO) focused solely on S. mansoni [21]. At present, there is only 1 candidate vaccine antigen for S. haematobium (28-kDa glutathione-S-transferase [GST]) [22]. Because immune responses to S. haematobium differ from those to S. mansoni [23] (and may differ from those to S. japonicum) and because phylogenetic analyses show that S. haematobium is more closely related to the animal schistosomes S. mattheei and S. bovis than to the other human schistosomes [24], it is imperative to study immunogenic proteins and acquired immunity against this important species
The aim of this study was to identify and characterize major immunogenic proteins for S. haematobium. Serum samples from individuals exposed to schistosomes were used to screen soluble extracts from adult parasites, and responses before and after treatment with praziquantel were compared. The rationale for this comparative analysis was that treatment with praziquantel has been shown to alter schistosome-specific immune responses, and this alteration results in qualitative and quantitative changes associated with resistance to infection [25 –28]. We can therefore test the hypothesis that changes in antibody responses after treatment are partly due to changes in the antigen profile recognized by the immune system
Parasite material Freeze-dried adult S. haematobium soluble worm antigen preparation (SWAP) was obtained from the Theodor Bilharz Institute (Giza, Egypt). The parasite strain was used in previous immunoepidemiological studies [29], and the soluble fraction was used in immunological assays. To prepare this fraction, worms were perfused in saline buffer, washed in PBS (pH 7.4), homogenized, centrifuged to obtain the soluble fraction, and freeze-dried in aliquots (∼5 mg/mL) that were reconstituted with distilled water as required
Study subjects Serum samples were obtained from villagers in the Mashonaland East province of Zimbabwe, where S. haematobium is endemic. Only permanent inhabitants of the study area who had never been treated for any helminth infection were eligible for inclusion in the study. Permission to conduct the study was obtained from the provincial medical director. After an explanation of the study aims and procedures was given to the community, an initial parasitological (using stool and urine samples) and serological (using blood samples) survey of all compliant participants was conducted. Stool samples were processed in accordance with the Kato Katz procedure [30] to detect S. mansoni eggs and other intestinal helminths, whereas the urine filtration method [31] was used to detect S. haematobium eggs in urine samples. After collection of the samples, all participants were offered treatment with the recommended dose of praziquantel (40 mg/kg of body weight). Participants who would not accept treatment on religious grounds or were absent on treatment days but wished to remain part of the study cohort were classified as untreated control subjects. Parasitological and serological samples were collected in the same manner 12 weeks after treatment. To be included in the study cohort, participants had to meet all of the following criteria: (1) provide at least 2 urine and 2 stool samples on consecutive days at both time points; (2) be negative for intestinal helminths, including S. mansoni at both time points; (3) be confirmed to be negative for S. haematobium eggs at the second time point if they had been treated; and (4) provide a blood sample at both time points. A total of 174 individuals (5–42 years old) met these criteria; 112 individuals (5–42 years old) formed the treated cohort, and 62 individuals (5–39 years old) formed the untreated cohort. Pretreatment infection levels were similar in the 2 cohorts (60% prevalence; mean infection intensity, 32 eggs/10 mL of urine)
Gel electrophoresis Two different 2-dimensional gel separations were performed in parallel: the first contained 100 μg of SWAP (to be used for Western blotting) and the second contained 200 μg of SWAP (to be used for protein identification). Isoelectric focusing instrumentation, immobilized Ph gradient (IPG) buffers, and related reagents were purchased from Amersham, unless otherwise indicated. In the first-dimension electrophoresis, the antigen was mixed with rehydration solution (7 mol/L urea, 2 mmol/L thiourea, 4% CHAPS, 65 mmol/L dithiothreitol [DTT], and trace bromophenol blue) and IPG buffer (pH 3–10) to give a total sample volume of 250 μL, and then the sample was loaded into a 13-cm gel holder with a 13-cm gel strip (linear pH 3–10). The gel strip was rehydrated, and the proteins were focused on an IPGPhor machine by use of the following protocol: 12–14 h of rehydration at 20 V and a 5-h voltage-focusing procedure (1 h at 500 V, 1 h at 1000 V, and 3 h at 8000 V). The strips were then incubated in 5 mL of equilibration buffer (50 mmol/L Tris, 6 mol/L urea, 2% SDS, and 30% glycerol [pH 8.8]) containing 30 mmol/L DTT for 15 min and in equilibration buffer containing 135 mmol/L iodoacetamide for another 15 min. Second-dimension electrophoresis was performed on a 12% polyacrylamide 13-cm gel in a Hoefer SE600 system using SDS buffer. The proteins on the gel used for protein identification were stained with Coomassie blue to visualize them, whereas proteins on the gel used for Western blotting were transferred onto a nitrocellulose membrane, as described below
Immunoblotting Proteins were transferred from the gel onto a nitrocellulose membrane using a semidry system (Hoefer) in transfer buffer (Invitrogen) containing 10% methanol at 30 V for 1 h. The membrane was stained with Ponceau S solution (Sigma) to check transfer efficiency and then was blocked at room temperature for 1 h in Tris-buffered saline (TBS) blocking buffer (Pierce) and 0.05% Tween 20. After blocking, the membrane was subjected to 2 separate 10-min washes with TBS, 0.05% Tween 20, and 0.5% Triton-X 100 (TBS/TT). A pool of pretreatment serum samples (diluted 1:100 in TBS blocking buffer and 0.02% Tween 20) was added to the membrane, and the membrane was incubated overnight at 4°C and then was washed 3 times for 10 min each time in TBS/TT. Horseradish peroxidase–conjugated rabbit anti–human IgG (Dako) was diluted 1:4000 in TBS blocking buffer, and 0.05% Tween 20 was added. The membrane was incubated at room temperature for 1 h and then was washed 4 times for 10 min each time in TBS/TT and 1 time for 10 min in TBS alone. The proteins were visualized using the chemiluminescence product ECL Plus (Amersham), in accordance with the manufacturer’s instructions. Films were exposed to the blots for 5 s and then were developed, and spots were matched to those on the Coomassie blue–stained gel. After visualization, the membrane was stripped of the ECL Plus reagent, secondary antibody, and serum samples, in accordance with the protocol provided by the manufacturer. The same membrane was then probed using posttreatment serum samples. A previous assay showed that the stripping procedure removed all proteins not directly bound to the nitrocellulose membrane, as indicated by the lack of ECL reactivity with a stripped membrane. This procedure did not remove any of the parasite proteins, as evidenced by probing the same membrane with 3 serum samples successively (i.e., a pretreatment serum sample, then a negative control serum sample, and then the same pretreatment serum sample). The gel electrophoresis and Western blotting were repeated for all samples, to confirm the patterns that were obtained
Image analysis Images from the Western blots were electronically scanned with Image Master 2-dimensional gel image analysis software (version 3; Amersham) and used for matching. Predicted matches were also visually verified. Spots on the Coomassie blue–stained gel that matched those on the Western blots were excised and then were analyzed by mass spectrometry (MS)
Mass spectrometry Plugs of 1.4 mm were excised from the gels and were subjected to in-gel trypsin digestion in an Ettan Spot Handing Workstation (GE Healthcare), in accordance with standard protocols (Amersham). The resulting tryptic peptides were solubilized in 0.5% formic acid and were fractionated by nanoflow high-performance liquid chromatography on a C18 reverse phase column (GE Healthcare), and elution was performed with a continuous linear gradient of 40% acetonitrile for 20 min. The eluates were analyzed by online electrospray tandem MS (MS/MS) by use of a Qstar Pulsar mass spectrometer (Applied Biosystems). A 3-s survey scan preceded each MS/MS data-collection cycle of 4 product ion scans of 3 s each, and this gave a duty cycle of 15 s. Data were submitted for an MS/MS ion search via the Mascot search engine (Matrix Science), and both locally established databases for S. mansoni EST sequences and the present nonredundant National Center for Biotechnology Information (NCBI) database were searched
Two-dimensional gel electrophoresis analysis Two-dimensional gel electrophoresis resulted in separation of S. haematobium SWAP into ∼150 discrete spots that were visible after standard staining with Coomassie blue (figure 1). Additional spots could be detected by silver staining, but the quality of the mass spectra obtained for identification of the proteins was higher for the Coomassie blue–stained spots, and this technique resulted in superior data. Moreover, no spot that was subsequently shown to be reactive by Western blotting failed to be stained by Coomassie blue
Coomassie blue–stained 2-dimensional gel showing spots matched to the Western blots. Spots on the gel were excised and identified. Molecular weight markers (in kilodaltons) are given on the right
Western blot analysis To determine which proteins were recognized by the serum samples collected before and after praziquantel treatment, a Western blot assay was optimized on the basis of the results of the 2-dimensional gel electrophoresis. Initial assays using anti–human IgA, IgG, and IgM reagents showed that IgG detected the maximum number of spots and that IgA and IgM did not identify any spots that were not detected by IgG. Therefore, for the full Western blot analysis, anti–human IgG was used. This analysis showed that a total of 71 spots visualized on the Coomassie blue–stained gel reacted with human serum samples from S. haematobium–exposed individuals
MS/MS analysis The 71 spots identified as serologically reactive by Western blotting were excised from the Coomassie blue–stained gel and were subjected to in-gel trypsin digestion. Subsequently, the tryptic peptides were analyzed by MS/MS, and the peptide data obtained were used to search EST databases. Although there are relatively few S. haematobium peptide sequences available, most of the spots were successfully matched to S. mansoni or S. japonicum proteins whose peptide sequences are available in public databases. The identifications made for these 71 spots are shown in table 1. The identity given for each spot corresponds to the top hit score (the Mascot output statistic) that had a MOWSE score >30 (MOWSE scores are logarithmic, so that a hit score with a MOWSE score of 30 represents P=.05, a hit score with a MOWSE score of 40 represents P=.005, etc.). If the MOWSE score was <30, then the identification was rejected, and the spot was designated as being a nonsignificant hit. Predicted molecular weights (MWs) and isoelectric points (pIs) of each identified protein (not the spot) as well as the species they come from are also given in table 1. The majority of the spots corresponded to S. mansoni proteins
The MS/MS analysis revealed cases in which different spots were derived from the same protein: for example, spots 63, 64, and 65 are all heat-shock protein 70 (HSP70), as are spots resolving to the same protein but with different accession numbers (e.g., spots 31–34, which are all actin). Some of the recognized proteins occur as multiple isoforms differing by pIs, MW, or both. For example, there are at least 3 GST isoforms differing by pIs, and there are several enolase isoforms differing by both MW and pIs. In this analysis, it is not possible to define the precise nature of these differences, because sequence data are not yet available from the S. haematobium orthologues
Identity of proteins recognized by serum samples The proteins recognized included abundantly expressed proteins (as indicated by the size/intensity of the spot in figure 1), such as glyceraldehyde-3- phosphate dehydrogenase (GAPDH), and in most, but not all, cases, the size of the spot on the Western blot image was related to the size of the spot on the Coomassie blue–stained gel. For example, spot 21 (GAPDH) in figure 1 is also a very large spot in figures 2 and 3
Western blot analyses of serological reactivity of serum samples from the treated cohort, comparing pre- (A) and posttreatment (B) responses. A Spots reacting with serum samples collected at baseline (before treatment). Boxes represent areas where additional spots in panel B are absent. B Spots reacting with serum samples collected 12 weeks after treatment. Boxes highlight the additional spots recognized after treatment. Molecular weight markers (in kilodaltons) are given on the right
Western blot analyses of serological reactivity of serum samples from the untreated cohort. Serum samples were collected at the same time points as those used for the treated cohort. A Spots reacting with serum samples collected at baseline (before treatment in the treated cohort). Boxes represent areas where additional spots in figure 2B are absent. B Spots reacting with serum samples collected 12 weeks after treatment in the treated cohort. Boxes represent areas where additional spots in figure 2B are absent
Of the 71 spots recognized by the serum samples, all but 13 gave rise to protein identifications. Of the 58 identified spots, 2 were found to be ESTs whose proteins have not yet been characterized, whereas the remaining 56 spots resolved to 26 different proteins. The 26 proteins have been grouped by molecular function in table 2. They include structural/muscle proteins (which are most numerous), enzymes (mostly components of the glycolytic pathway), chaperone proteins, and binding proteins. Only GST has been studied in S. haematobium whereas, to our knowledge, the remaining 25 proteins are identified in S. haematobium here for the first time. Moreover, table 2 shows 14 proteins that have not been previously shown to be immunogenic in natural human infection with any schistosome species, and 11 of these have not previously been shown to be immunogenic in experimental models
Summary of proteins, classified by molecular function and published immunological status, recognized by serum samples
Enhanced reactivity after praziquantel treatment A comparative study between pretreatment and posttreatment serum samples was conducted to determine if treatment altered responses to the proteins. The pretreatment and posttreatment Western blot assays were conducted on the same membrane, to exclude any variation that might arise from the use of different antigen preparations. These assays showed that protein recognition patterns of serum samples from the 2 time points differed, as is shown in figure 2. Treatment enhanced the recognition of specific proteins by serum samples. Of the 71 spots, pretreatment serum samples recognized 59 spots representing 21 different identified proteins, as is shown in table 1. Serum samples collected 12 weeks after treatment recognized an additional 12 spots representing 8 identified proteins. Of the 12 additional spots, 3 had similar identities to spots of different MWs or pIs that had been recognized by pretreatment serum samples—for example, actin, actin-binding/filaminlike protein, and GST are likely to be different isoforms. Five proteins (calreticulin, tropomyosin 1, tropomyosin 2, paramyosin, and triose phosphate isomerase) were recognized only by posttreatment serum samples and did not have isoforms already recognized by pretreatment serum samples
In addition to these qualitative changes, there were also quantitative changes in protein recognition, as was indicated by increases in the intensity of recognition for some spots after treatment. This was most apparent in spots 1 (fatty acid–binding protein), 8 (GST), 31–34 (actin), 41–43 (enolase), and 63–65 (HSP70)
Serum samples from untreated participants showed no changes in protein identification patterns at the 2 time points (baseline and 12 weeks later), as is shown in figure 3B . In addition, at both time points, reactivity of serum samples from untreated individuals was similar to that of serum samples from treated individuals at the start of the study, except that they reacted with spots 67 and 68 (actin-binding/filaminlike protein), as is shown in figures 2A and 3A
Despite being the most prevalent and widespread schistosome species affecting humans in Africa [1], S. haematobium is the least studied with respect to parasite-specific immune responses and antigen characterization. In particular, few specific antigens have been identified or used for immunoepidemiological research. The present study gives the most comprehensive analysis to date of adult worm antigens in this species and for schistosomes in general. The analysis focused on the adult stage, which is the most long-lived developmental stage and a target for immune elimination in S. haematobium and S. bovis [57]. Of the 150 spots visualized on the Coomassie blue–stained gel, 71 were detected by their reactivity with total IgG antibodies in pooled serum samples. This number does not include proteins recognized by a minority of serum samples and for which reactivity could not be detected after dilution. The 13 spots that did not have significant hits to known proteins or ESTs (despite having been processed twice) may not be similar to presently known proteins or may have given mass spectra that were too unclear for identification
Studies are now under way to identify the spots that were not serologically recognized, particularly those that are abundant in the proteome and might play an important role in host immune evasion/modulation [58]. The recognition patterns of the individual IgG subclasses will be investigated, because they, together with the other isotypes, will help to characterize immune responses to the antigens we have defined here
Several proteins recognized were homologues of, or were similar to, presently known vaccine candidates characterized in S. mansoni and/or S. japonicum. For example, the serum samples reacted with homologues of 9 of 10 World Health Organization WHO S. mansoni vaccine candidate antigens [21] in the S. haematobium proteome. The sequence for the remaining WHO vaccine candidate antigen (PN18-cyclophilin) has not yet been published in the literature [21], but cyclophilins are members of the immunophilin family and are related to the immunophilin p50 recognized by the serum samples used in the present study
The S. haematobium proteins that reacted with the serum samples included those whose homologues are abundant in EST databases of S. mansoni [18] and S. japonicum [19] as well as those abundant in the soluble fraction of the adult S. mansoni proteome [20]. Several of these proteins are conserved among invertebrates, and some are vaccine candidates for other helminth species. For example, paramyosin, which was first shown to be protective against schistosomiasis [59], is a long-standing vaccine candidate for filariasis [60], cysticercosis [61, 62], and S. mansoni and S. japonicum infection [15, 32, 34]. Most of the recognized proteins have been localized to the parasite tegument in S. mansoni and S. japonicum and are therefore accessible to the immune system. Only 1 integral tegumental protein, a homologue of S. mansoni fatty acid–binding protein Sm14, was identified by the serum samples. This is not surprising, because tegumental proteins are poorly soluble and were underrepresented in the aqueous fraction used in the present study. For some proteins with several isoforms that reacted with the serum samples (e.g., GST for spots 7 and 8), the reactivity differed between isoforms, and this suggests that the processes generating these isoforms may alter the immunogenicity of the proteins
Treatment with praziquantel enhanced the reactivity of serum samples by increasing the number of proteins recognized and the intensity of the recognition of proteins before treatment. Most of the additional proteins recognized after treatment are associated with the parasite musculature or glycolytic metabolism, and this indicates that treatment made these proteins preferentially available. This finding is consistent with the hypothesis that treatment renders different parasite proteins accessible to the host immune system. In the worm, praziquantel induces paralysis followed by destruction of the tegument, and death is believed to result from synergistic action with the host immune system [63 –65]. Experimental work in mice has shown that praziquantel treatment exposes tegumental antigens such as actin [66, 67]. Therefore, the results of the present study are consistent with the hypothesis that there is a change in the antigen profile presented to the host immune system after treatment. Previous studies have shown changes in antibody responses to crude antigens after treatment but did not differentiate between changes arising from different amounts of antigen and those arising from different types of antigen [25, 28, 68, 69]. The present study clearly shows that both quantitative and qualitative changes in antigen recognition do occur after treatment
In conclusion, the present study has identified 27 S. haematobium proteins that react with serum samples from a population of Zimbabweans exposed to the parasite. Several of these antigens are novel in all schistosome species and require further immunological investigation and characterization at the antibody and T cell response levels. The study has also shown that treatment with praziquantel alters responses to individual proteins both qualitatively (new proteins/isoforms being recognized) and quantitatively (increases in reactivity with individual proteins)
We are grateful for the cooperation of the Ministry of Health and Child Welfare in Zimbabwe, the provincial medical director of Mashonaland East, the environmental health workers, and the residents, teachers, and schoolchildren in Mutoko and Rusike. We are also grateful for the technical assistance from staff at the Blair Research Institute and the technical advice from Rachel Curwen (University of York, United Kingdom)
↵Potential conflicts of interest: none reported
Financial support: Medical Research Council, United Kingdom (grant G81/538); Carnegie Trust for the Universities of Scotland; Wellcome Trust
↵Present affiliation: Centre de Recherche en Infectiologie, Centre Hospitalier Universitaire de Quebec, Pavillion CHUL, Sainte Foy, Quebec, Canada
IDSA Members: For your free access to this journal, log in via the IDSA members area.
Open access options for authors visit Oxford Open
This journal enables compliance with the NIH Public Access Policy