INTRODUCTION

Top Abstract Introduction Materials and Methods Results and Discussion References

Schistosomes establish chronic infections in their hosts despite the presence of specific cellular and humoral immune responses. One of the most striking immunoevasive mechanisms displayed by these parasites is the acquisition of host products onto the tegument of the schistosome to mask its foreign status. Host molecules adsorbed onto the surface include immunoglobulins (20, 40, 44), major histocompatibility complex products (3, 36), ₂-microglobulin (₂m) (40), complement components (35, 38), ₂-macroglobulin (7, 19), C3 decay-accelerating factor (14), and glycolipids in the form of A, B, H, and Lewis blood group antigens (12).

More than 20 years ago, Schistosoma mansoni schistosomula were shown to selectively bind the Fc but not the Fab fragment of immunoglobulin G (IgG) (40). This was demonstrated by the adhesion of rosetted sheep erythrocytes to transformed schistosomula of S. mansoni in vitro and inhibition of adhesion by free Fc fragments. IgG from rat, mouse, and rabbit also inhibited rosette formation. Whereas most of the host molecules listed above are adsorbed by schistosomula, receptors for Fc and complement C3 have also been identified on the surface of adult S. mansoni (38).

Our understanding of the molecular interactions between schistosomes and their hosts has progressed significantly since these early studies, a situation facilitated by the availability of more than 15,000 schistosome expressed sequence tags deposited in the public databases (15). However, despite this information, schistosome surface receptors for host immune proteins have yet to identified. Here, we describe the surface biotinylation of live schistosomes and the subsequent purification of labeled Fc- and ₂m-binding proteins from the surface of S. mansoni. We identify paramyosin (Pmy) as the Fc-binding protein (Fc-Bp) and show that Pmy from the Asian schistosome, Schistosoma japonicum, also binds human Fc. Pmy is a promising vaccine candidate for schistosomiasis; while it has been shown to have an unusual immunomodulatory role by binding C1q of the complement cascade (25), this is the first report of schistosome Pmy binding to Fc, a finding with important implications for the use of this antigen as a vaccine target.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results and Discussion References

Parasites. Adult S. mansoni (Puerto Rican strain; both sexes) were perfused from BALB/c mice 7 to 10 weeks after infection with cercariae and washed thoroughly in phosphate-buffered saline (PBS). The viability of worms was examined microscopically; intact parasites were washed three times in RPMI 1640 supplemented with penicillin (100 U ml¹), streptomycin (100 µg ml¹), amphotericin  (Fungizone; 0.025 µg/ml), and 20 mM HEPES (RPMI-PSF) and then cultured overnight in the same medium at 37°C in an atmosphere of 5% CO₂-95% air. Medium was changed the following morning, and parasites were cultured for another 2 h, checked for viability and then surface biotinylated as outlined below. S. japonicum (Sorsogon, Philippines) adult parasites were perfused from mice and washed in PBS, and soluble adult worm protein was produced as previously described (43). S. mansoni schistosomula were obtained by mechanical transformation of cercariae according to published protocols (6) and cultured overnight at 37°C in RPMI-PSF in an atmosphere of 5% CO₂-95% air.

Surface biotinylation. Live adults or schistosomula of S. mansoni were surface biotinylated using sulfo-N-hydroxysuccinimide-LC-biotin (Pierce) as previously described (8). After removal of free biotin and washing, parasites were resuspended in Tris-buffered saline (TBS)-1% Triton X-100 and incubated on ice for 30 min with occasional agitation. The supernatants, hereafter referred to as bio-ad-TX (adult worms) or bio-som-TX (schistosomula), were then removed, centrifuged at 10,000 × g and dialyzed overnight against 500 volumes of TBS-10 mM CaCl₂. Protease inhibitors (all purchased from Sigma) were added to the extract at the final concentrations given in parentheses: E64 (10 µM), pepstatin A (5 µM), and PMSF (1 mM). The extract was then stored at 80°C until use. Two different surface extracts were prepared from adult worms, the first was from parasites that were perfused from mice and washed thoroughly in RPMI-PSF for 5 min before surface biotinylation (bio-ad-TX-0), and the second was from parasites that had been incubated in RPMI-PSF for 24 h as outlined above (bio-ad-TX-24). Efficacy of the biotinylation process was monitored by precipitating biotinylated proteins from extracts with streptavidin-agarose (8) and by probing bound and unbound proteins with streptavidin-peroxidase or various antisera as follows. Bound and unbound proteins were subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and transferred to nitrocellulose paper (NCP), and the membrane was blocked overnight in 5% skimmed milk in PBS at 4°C. Blots were probed with either horseradish peroxidase (HRP)-conjugated streptavidin (1 µg/ml; Sigma) or a rabbit antiserum (1:2,000) prepared against S. japonicum recombinant Pmy (rSj-Pmy) (18) for 1 h at room temperature (RT). Pmy has been localized to the musculature underlying the tegument of S. mansoni adult worms and has been shown by others not to incorporate biotin during surface labeling (8). Furthermore, S. mansoni and S. japonicum Pmy are 96% identitical at the amino acid level, and an antiserum raised to rSj-Pmy in our laboratory has been shown to cross-react with S. mansoni Pmy (42). Streptavidin blots were then washed with PBS-0.1% Tween 20 and developed using an enhanced chemiluminescence substrate (Amersham). Immunoblots were washed then probed with goat anti-rabbit IgG-HRP (1:2,000; Bio-Rad) for 1 h at RT, washed, and developed using chemiluminescence. Additional monitoring of the biotinylation process was performed using antibodies raised to recombinant tegumental antigen Sj22.6 (41) and recombinant aspartic protease (localized to the gut) (2) of S. japonicum.

Coupling of ligands to Sepharose. Five milligrams of human IgG Fc (Rockland Immunochemicals) was dialyzed overnight into coupling buffer (0.1 M NaHCO₃, 0.5M NaCl, pH 8.3) and bound to 2 ml (packed volume) of cyanogen bromide-activated Sepharose 4B (Pharmacia) according to the manufacturer's instructions. Human ₂m (500 µg; Sigma) was coupled to 500 µl of Sepharose as outlined above.

Purification of biotinylated Fc-Bp. Bio-ad-TX (100 µl) or bio-som-TX (100 µl) was added to Fc-Sepharose (50-µl packed volume) in 200 µl of TBS-0.1% Triton X-100 (TBS-TX) and rotated on an orbital shaker for 2 h at RT. The reactions were centrifuged, and the supernatant was retained as unbound protein. Sepharose was washed three times with TBS-TX (1 ml per wash), and bound proteins were eluted by resuspending the Sepharose in 100 µl of reducing SDS-PAGE sample buffer and boiling. Alternatively, after washing, the following buffers were tested for the ability to elute bound proteins from Fc-Sepharose under native conditions: NaCl (0.6, 1.0, and 2.0 M), 0.1 M glycine (pH 2.5), and TBS-10 mM EDTA. Supernatants and eluted proteins were electrophoresed on 10% polyacrylamide gels and transferred to NCP, nonreactive sites were blocked, and the membranes were probed with HRP-streptavidin as described earlier. We attempted to competitively inhibit binding of Pmy to Fc-Sepharose by pre-incubating biotinylated extracts with either rabbit antiserum to rSj-Pmy (2.5 µl) or soluble protein A (100 µg) for 1 h at RT before adding the extract to Fc-sepharose.

Purification of biotinylated ₂-m-binding proteins. Bio-som-TX (100 µl) was added to ₂m-Sepharose (50-µl settled bed volume) in 200 µl of TBS-TX, and bound proteins were identified as outlined above for Fc-Bp. Bound proteins were eluted by boiling in reducing SDS-PAGE loading dye.

Immunoprecipitations. To observe whether bio-ad-TX contained complexes of Pmy and host IgG, 50 µl (packed volume) of protein G-Sepharose (Sigma) was added to 100 µl of bio-ad-TX-0 and rotated at RT for 1 h. The sepharose was then centrifuged and washed three times with TBS-TX, and bound proteins were eluted by boiling in SDS-PAGE reducing sample buffer. Proteins were electrophoresed, transferred to NCP, and probed with anti-rSj-Pmy serum; this and all subsequent steps were performed as outlined above.

Preparation of Fab fragments. Rabbit immunoglobulins were purified from either 10 µl of anti-Pmy serum or 10 µl of normal rabbit serum by using protein G-Sepharose. Bound immunoglobulins were eluted in 0.1 M glycine (pH 2.8), and 0.1 M Tris was added until the solution reached pH 8.0. Immunoglobulins were buffer exchanged into 0.1 M sodium acetate (pH 5.5) in Amicon Centricon 10 concentrators (Millipore), and Fab was cleaved from Fc domains by using papain according to published protocols (13). Cleaved Fc regions and entire, uncleaved immunoglobulins were removed using protein G-Sepharose. The remaining, unbound Fab regions were tested for recognition of Pmy by dot blotting.

Microscopy. S. mansoni mechanically transformed schistosomula were fixed in 100% methanol and stored at 20°C. Worms were transferred to PBS at RT and processed for fluorescence microscopy as follows. All fluorescent media were processed in PBS. Schistosomula were incubated in fluorescein isothiocyanate (FITC)-conjugated Fc (17 µg/ml) for 24 h at 4°C and rinsed in multiple changes of PBS. Other schistosomula were incubated concurrently in the following media for 24 h: (i) unconjugated Fc fragment (17 µg/ml, Rockland), (ii) Cy3-conjugated rabbit anti-human Fc (17 µg/ml, Sigma), (iii) anti-Pmy serum diluted 1:200, (iv) normal rabbit serum diluted 1:200, (v) anti-Pmy Fab fragment, and (vi) normal rabbit Fab. After several rinses in PBS, all samples except sample (ii) were incubated in human Fc-FITC (17 µg/ml) for 24 h at 4°C. In parallel experiments, schistosomula were incubated in a 1:200 dilution of anti-Pmy serum for 24 h at 4°C, followed by incubation in Cy3-conjugated anti-rabbit IgG serum (Jackson Laboratories) diluted 1:300 in PBS. All samples for fluorescence were mounted in DABCO and examined using an Olympus BX60 fluorescence microscope.

	RESULTS AND DISCUSSION

Top Abstract Introduction Materials and Methods Results and Discussion References

Surface biotinylation of S. mansoni. Live S. mansoni adults, determined to be undamaged by microscopic analysis, were subjected to surface biotinylation as previously described (8). In their pioneering studies, Davies and Pearce (8) used Pmy as a marker for determining the efficacy of surface labeling. They probed biotinylated and nonbiotinylated proteins from detergent extracts of adult S. mansoni (separated by precipitating with streptavidin-agarose) with antiserum to Pmy and noted that the protein was found only in the nonbiotinylated extract. They concluded that their labeling was specific for exposed surface proteins only. In a similar experiment probing labeled and nonlabeled detergent extracts (Fig. 1A) with anti-rSj-Pmy serum, we found Pmy to be present in both surface and underlying tissue extracts of bio-ad-TX-24 (Fig. 1B). The exact localization of Pmy in adult schistosomes remains a matter of debate (11, 26, 28). When probing the nonbiotinylated proteins (i.e., those not bound by streptavidin-agarose) from bio-ad-TX-24, we obtained reactivity with proteins with molecular sizes of 97 kDa (Pmy) and approximately 200 kDa (Fig. 1B). The 200-kDa antigen is most likely to be myosin, a related muscle protein with sequence similarity to Pmy (4) but which has not been found on the surface of adult worms (45). We concur with this finding, as our Western blots of biotinylated and underlying, nonbiotinylated extracts using anti-rSj-pmy serum showed a band of about 200 kDa (presumed to be myosin) only in the unlabeled fraction of bio-ad-TX-24 (i.e., subtegumental tissues), while Pmy was present in both labeled and unlabeled detergent extracts. Pmy is found in the tegument of schistosomula, as well as in the musculature and postacetabular glands (11, 31). Antiserum to schistosome cathepsin D, an aspartic protease found in the gut of schistosomes (2), did not bind to surface-labeled proteins but did bind to underlying, nonlabeled proteins (not shown), further accentuating the specificity of the surface biotinylation procedure.

View larger version (61K): [in this window] [in a new window]   FIG. 1.   (A) Comparison of biotinylated and nonbiotinylated surface proteins of adult S. mansoni solubilized with Triton X-100 and separated using streptavidin (strept)-agarose. Proteins were then probed with peroxidase-labeled streptavidin (A). The left lane (unbound) contains proteins that were not precipitated with streptavidin-agarose, while the right lane (bound) contains biotinylated surface proteins that were precipitated. (B) Western blot of the extracts in panel A probed with antiserum raised to rSj-Pmy (-pmy), showing that Pmy is exposed to labeling procedures on the surface of live parasites. (C) Western blot of biotinylated and nonbiotinylated surface proteins of S. mansoni schistosomula precipitated with streptavidin-agarose and probed with -pmy serum. Here and in succeeding figures, positions of size markers are shown in kilodaltons at the left.

View larger version (34K): [in this window] [in a new window]   FIG. 2.   Biotinylated and nonbiotinylated surface proteins of adult S. mansoni were solubilized with Triton X-100 shortly after removal from the host (0 h) or after 24 h culture of in vitro after removal from the host (24 h). Surface-labeled proteins were precipitated using streptavidin-agarose and probed with antiserum to rSj-Pmy (A) or to murine IgG (B). The position of the light chain of IgG is shown at the right.

S. mansoni and S. japonicum Pmy binds to Fc. When bio-ad-TX-0, bio-ad-TX-24, and bio-som-TX were incubated with Fc-Sepharose, a single, biotinylated band remained bound after washing (Fig. 3A and B). The band corresponding to the Fc-Bp was less intense in bio-ad-TX-0 than in bio-ad-TX-24, indicating that more of the protein was present in the tegument after overnight incubation of parasites in vitro or that the majority of host IgG had been shed by 24 h, allowing a greater quantity of the Fc-Bp to bind to the resin. The Fc-Bp had the same molecular size as S. mansoni Pmy (97 kDa), and probing of the bound protein with anti-rSj-pmy serum revealed its identity as Pmy (Fig. 3C). In addition, when Pmy was removed from bio-ad-TX by immunoprecipitation with anti-rSj-Pmy serum, no biotinylated proteins bound to Fc-Sepharose (Fig. 3D). Pmy from keyhole limpet (Mollusca) also bound to human Fc (not shown), indicating that the binding process is not restricted to schistosome Pmy. Pmy from molluscs, cestodes, and trematodes has been shown to bind to collagen (22, 25, 34) and to the collagen-like region of the complement protein C1q (25). Pmy can be eluted from collagen using 0.6 M NaCl (22), and so we attempted to elute Pmy from Fc-Sepharose in a similar manner. Partial elution of bound Pmy was achieved with 2.0 M but not with 1.0 M NaCl (Fig. 3E). Recombinant S. mansoni Pmy expressed in bacteria (kindly provided by Chris King, Case Western Reserve University, Cleveland, Ohio) also bound to Fc-Sepharose (Fig. 3F), indicating that Pmy binds directly to Fc and not indirectly via a complex of host and parasite proteins.

View larger version (82K): [in this window] [in a new window]   FIG. 3.   (A) Detection of a 97-kDa biotinylated surface protein of adult S. mansoni that bound to human Fc-Sepharose, (seph). Extracts were obtained from parasites that were labeled after removal from the host (0 h) or after 24 h of culture in vitro (24 h). The blot was probed with peroxidase-labeled streptavidin. (B) Binding of a 97-kDa biotinylated protein from schistosomula to Fc. (C) Demonstration that the bound protein from panel A (24 h) is recognized by antiserum to rSj-Pmy (-pmy) but not by antiserum to another tegument protein, Sj22.6 (-Sj 22.6). Depletion of Pmy from biotinylated surface extracts by immunoprecipitation using anti-Pmy serum and protein G-Sepharose prior to incubation of the extract with Fc-Sepharose resulted in an absence of any Fc-BP detected using streptavidin-peroxidase (D). Pmy eluted from Fc-Sepharose in the presence of 2 M but not 1 M NaCl, as detected using streptavidin peroxidase (E). (F) Binding of recombinant Pmy from S. mansoni (kindly provided by C. King) but not of bovine serum albumin (BSA) control to Fc-Sepharose.

View larger version (70K): [in this window] [in a new window]   FIG. 4.   S. japonicum soluble extracts were incubated with Fc-Sepharose, and bound and unbound proteins were probed with antiserum to rSj-Pmy (A) or antiserum to soluble adult worm protein (B).

View larger version (58K): [in this window] [in a new window]   FIG. 5.   Adsorbed murine immunoglobulins and associated complexes were precipitated from surface extracts of adult S. mansoni by using protein G-Sepharose, and bound proteins were probed by Western blotting with antiserum to recombinant Pmy (-pmy) or recombinant Sj22.6 (-Sj22.6) as a control surface protein. Reactivity of the murine immunoglobulin heavy chain with the anti-rabbit IgG secondary antibody is highlighted.

View larger version (66K): [in this window] [in a new window]   FIG. 6.   Fluorescence micrograph showing binding of rabbit anti-Pmy serum to the surface of an S. mansoni schistosomulum (A). FITC-conjugated human Fc bound to the surface, oral cavity, and vestigial penetration glands of schistosomula (B), while preincubation of parasites with anti-Pmy Fab fragments ablated subsequent binding of Fc-FITC to the surface (C); normal rabbit Fab did not interfere with subsequent binding of Fc-FITC (D).

₂m does not bind to Pmy but does bind to other S. mansoni surface proteins. To control for nonspecific binding of Pmy to indiscriminate ligands or to Sepharose, we conjugated human ₂m to cyanogen bromide-activated Sepharose and repeated the binding assays described for Fc-Sepharose using bio-som-TX. Previously Torpier et al. (40) reported that ₂m binds to the surface of S. mansoni mechanically transformed schistosomula and inhibits Fc receptor-mediated rosette formation, prompting these authors to suggest that both Fc and ₂m bound to the same receptor on the surface of the parasite. Furthermore, ₂m associates with the intestinal Fc receptor in neonatal rats (37). We incubated bio-som-TX with ₂m-Sepharose and examined the biotinylated, bound proteins. Pmy did not bind to ₂m-Sepharose, but four bands with molecular sizes of approximately 38, 42, 110, and 200 kDa were eluted (Fig. 7). Bound proteins were probed with anti-rSj-pmy but were not recognized by the antibody (not shown), indicating that ₂m binds to a receptor other than Pmy on the parasite tegument and that this receptor appears to be a heteromultimer. The nature of this receptor awaits elucidation.

View larger version (68K): [in this window] [in a new window]   FIG. 7.   Biotinylated 2m-binding proteins from the surface of S. mansoni schistosomula. Bound proteins and unbound proteins were detected using HRP-streptavidin. Bound proteins are highlighted by arrows.

Concluding remarks. It has long been known that schistosomes adsorb host Fc onto their tegument, but the molecular mechanisms by which they achieve this process had not been elucidated. Here we have shown that Pmy, a vaccine candidate antigen that elicits protection against schistosomiasis in laboratory and domestic animals (27, 30, 33), is an Fc-Bp. Furthermore, we have shown that host ₂m does not bind to Pmy but does bind to other proteins on the surface of schistosomula. If schistosomes mask themselves from immune recognition by adsorbing host proteins such as immunoglobulins, antibodies directed at the surface receptors for these host proteins might interfere with adsorption and thus masking. Specific antiserum to Pmy interfered with binding of Fc to Pmy in vitro and to the tegument of live parasites, offering a plausible explanation for the mechanism of protection against S. mansoni infection induced by anti-Pmy immunization of mice (33) and against S. japonicum in natural reservoir hosts such as sheep (39) and buffalo (29). Protective immunization with Pmy resulting in decreased worm burdens in the S. mansoni murine model appears to be primarily cell mediated and is not transferred in serum (33). However, transfer of an IgE monoclonal antibody against Pmy conferred partial protection in mice challenged with S. japonicum (21), suggesting that specific antibodies might have blocked adsorption of nonspecific Fc domains and thus at least partially interfered with masking.