SCHISTOSOMIASIS
Schistosomiasis is a chronic and potentially deadly parasitic disease caused by members of the helminth genus Schistosoma of which S. haematobium, S. mansoni and S. japonicum are the most widespread in humans. Approximately 207 million people in tropical and sub-tropical areas are infected with schistosomes and 779 million people are at risk of being infected (Gryseels et al. Reference Gryseels, Polman, Clerinx and Kestens2006; Steinmann et al. Reference Steinmann, Keiser, Bos, Tanner and Utzinger2006). Schistosomes have a complex life-cycle in which larval, adult worm and egg stages interact with the human host, each playing a role in immunology, immunopathology and maintenance of infection. Eggs deposited in the organs of the host cause a strong immune response leading to the formation of periovular granulomas which eventually may give rise to fibrosis and organ failure.
Two immunologically distinct phases develop after a schistosome infection, the acute and the chronic phase. Acute schistosomiasis occurs during a primary infection upon first contact with the parasite while chronic schistosomiasis develops when immune responses become modulated, as observed in individuals living in endemic areas with continuous exposure to the parasites. Development of pathology depends on individual immune responses and infection intensity, but may lead to life-threatening inflammatory and obstructive disease (Gryseels et al. Reference Gryseels, Polman, Clerinx and Kestens2006).
Despite the strong immune response mounted, schistosomes can survive for years in the human host (Caldas et al. Reference Caldas, Campi-Azevedo, Oliveira, Silveira, Oliveira and Gazzinelli2008). Effective treatment of schistosomiasis, at the individual level or in mass treatment programmes, is mainly by use of the chemical agent Praziquantel (PZQ) (Fenwick and Webster, Reference Fenwick and Webster2006; Gray et al. Reference Gray, Ross, Li and McManus2011), and alternative anti-schistosomal drugs are currently also being investigated (Utzinger et al. Reference Utzinger, N'goran, Caffrey and Keiser2010). However, drug treatment does not prevent reinfection and therefore a more definitive solution may be the development of a prophylactic vaccine inducing protection against schistosomiasis. Detailed epidemiological and immuno-epidemiological studies in endemic areas have shown that natural human immunity to Schistosoma does occur, although it takes many years of infection to develop (Wilkins et al. Reference Wilkins, Blumenthal, Hagan, Hayes and Tulloch1987; Butterworth et al. Reference Butterworth, Fulford, Dunne, Ouma and Sturrock1988b; Fulford et al. Reference Fulford, Butterworth, Sturrock and Ouma1992; Ouma et al. Reference Ouma, Fulford, Kariuki, Kimani, Sturrock, Muchemi, Butterworth and Dunne1998; Vereecken et al. Reference Vereecken, Naus, Polman, Scott, Diop, Gryseels and Kestens2007). Young children are far more susceptible to re-infection than older children and adults even in communities where adults are more exposed than children (Kabatereine et al. Reference Kabatereine, Vennervald, Ouma, Kemijumbi, Butterworth, Dunne and Fulford1999). Other studies have also shown a peak shift in prevalence and intensity of infection with age (Woolhouse et al. Reference Woolhouse, Taylor, Matanhire and Chandiwana1991; Fulford et al. Reference Fulford, Butterworth, Sturrock and Ouma1992; Muller-Graf et al. Reference Muller-Graf, Collins, Packer and Woolhouse1997; Mutapi et al. Reference Mutapi, Ndhlovu, Hagan and Woolhouse1997) which provides evidence for a gradually acquired protective immunity in Schistosoma infection (Woolhouse, Reference Woolhouse1998).
PROTECTIVE ROLE OF ANTIBODIES IN SCHISTOSOMA INFECTION
Several immunological parameters, including specific antibody and T cell responses, are predictive of the age-dependent immunity or susceptibility to re-infection after treatment (Butterworth et al. Reference Butterworth, Dunne, Fulford, Capron, Khalife, Capron, Koech, Ouma and Sturrock1988a; Khalife et al. Reference Khalife, Dunne, Richardson, Mazza, Thorne, Capron and Butterworth1989; Leenstra et al. Reference Leenstra, Acosta, Wu, Langdon, Solomon, Manalo, Su, Jiz, Jarilla, Pablo, McGarvey, Olveda, Friedman and Kurtis2006; Vereecken et al. Reference Vereecken, Naus, Polman, Scott, Diop, Gryseels and Kestens2007). In the mouse model, it has been shown that B cell-deficient mice singly vaccinated with radiation-attenuated S. mansoni cercariae are significantly less protected against challenge infection than vaccinated wildtype mice (Jankovic et al. Reference Jankovic, Wynn, Kullberg, Hieny, Caspar, James, Cheever and Sher1999), suggesting a role for antibodies in immunity. Furthermore, repeated vaccination increased protection in wildtype mice but not in B cell-deficient mice (Anderson et al. Reference Anderson, Coulson, Ljubojevic, Mountford and Wilson1999; Jankovic et al. Reference Jankovic, Wynn, Kullberg, Hieny, Caspar, James, Cheever and Sher1999). Serum transfer studies with rabbits, rats and mice showed that serum from vaccinated animals can protect against infection with S. mansoni when transferred to non-vaccinated animals (Ford et al. Reference Ford, Bickle, Taylor and Andrews1984; Bickle et al. Reference Bickle, Andrews, Doenhoff, Ford and Taylor1985; Mangold and Dean, Reference Mangold and Dean1986). Also in baboons and vervet monkeys vaccination with irradiated S. haematobium and S. mansoni cercariae, respectively, results in high antibody titers correlating with protection (Harrison et al. Reference Harrison, Bickle, Kiare, James, Andrews, Sturrock, Taylor and Webbe1990; Yole et al. Reference Yole, Reid and Wilson1996).
In natural human infection, protective IgE, IgG and IgA-mediated immune responses have been reported (Black et al. Reference Black, Muok, Mwinzi, Carter, Karanja, Secor and Colley2010). IgA and IgE are isotypes that can mediate antibody-dependent cellular cytotoxicity of schistosomula in vitro (Butterworth et al. Reference Butterworth, Sturrock, Houba, Mahmoud, Sher and Rees1975, Reference Butterworth, Remold, Houba, David, Franks, David and Sturrock1977; Capron et al. Reference Capron, Rousseaux, Mazingue, Bazin and Capron1978; Dunne et al. Reference Dunne, Richardson, Jones, Clark, Thorne and Butterworth1993) and high levels of IgE and IgA against adult worm antigens have been associated with increased resistance to re-infection after treatment (Dunne et al. Reference Dunne, Butterworth, Fulford, Kariuki, Langley, Ouma, Capron, Pierce and Sturrock1992; Hagan et al. Reference Hagan, Blumenthal, Dunn, Simpson and Wilkins1991; Naus et al. Reference Naus, van Dam, Kremsner, Krijger and Deelder1998; Walter et al. Reference Walter, Fulford, McBeath, Joseph, Jones, Kariuki, Mwatha, Kimani, Kabatereine, Vennervald, Ouma and Dunne2006; Vereecken et al. Reference Vereecken, Naus, Polman, Scott, Diop, Gryseels and Kestens2007; Jiz et al. Reference Jiz, Friedman, Leenstra, Jarilla, Pablo, Langdon, Pond-Tor, Wu, Manalo, Olveda, Acosta and Kurtis2009; Black et al. Reference Black, Muok, Mwinzi, Carter, Karanja, Secor and Colley2010). One of the most predominant antigens for protective IgE antibody responses is SmTAL1 (formerly known as Sm22·6) (Dunne et al. Reference Dunne, Butterworth, Fulford, Kariuki, Langley, Ouma, Capron, Pierce and Sturrock1992, Reference Dunne, Webster, Smith, Langley, Richardson, Fulford, Butterworth, Sturrock, Kariuki and Ouma1997; Fitzsimmons et al. Reference Fitzsimmons, McBeath, Joseph, Jones, Walter, Hoffmann, Kariuki, Mwatha, Kimani, Kabatereine, Vennervald, Ouma and Dunne2007). IgE response patterns against SmTAL1 resemble those against adult worm antigen (Walter et al. Reference Walter, Fulford, McBeath, Joseph, Jones, Kariuki, Mwatha, Kimani, Kabatereine, Vennervald, Ouma and Dunne2006) and are associated with reduced odds of re-infection (Dunne et al. Reference Dunne, Butterworth, Fulford, Kariuki, Langley, Ouma, Capron, Pierce and Sturrock1992, Reference Dunne, Webster, Smith, Langley, Richardson, Fulford, Butterworth, Sturrock, Kariuki and Ouma1997; Webster et al. Reference Webster, Fulford, Braun, Ouma, Kariuki, Havercroft, Gachuhi, Sturrock, Butterworth and Dunne1996; Pinot de Moira et al. Reference Pinot de Moira, Fulford, Kabatereine, Ouma, Booth and Dunne2010), indicating that anti-SmTAL1 IgE could be a factor in host defence or a marker for resistance against infection. Although mice also produce large amounts of IgE upon Schistosoma infection (De Oliveira Fraga et al. Reference De Oliveira Fraga, Lamb, Moreno, Chatterjee, Dvorak, Delcroix, Sajid, Caffrey and Davies2010) the role of this IgE is not clear. Some reports have shown a protective role in primary infection, while others show a detrimental or even the absence of a functional role for IgE (Amiri et al. Reference Amiri, Haak-Frendscho, Robbins, McKerrow, Stewart and Jardieu1994; King et al. Reference King, Xianli, Malhotra, Liu, Mahmoud and Oettgen1997; El et al. 1998), which may be partly explained by the lack of expression of the high affinity receptor FcεR1 on mouse eosinophils (De Andres et al. Reference De Andres, Rakasz, Hagen, McCormik, Mueller, Elliot, Metwali, Sandor, Britigan, Weinstock and Lynch1997).
With respect to IgG antibodies, immuno-epidemiological studies have suggested that IgG1 and IgG3 are correlated with protection to reinfection, while IgG2 has a dual function. In the presence of activated eosinophils IgG2 has an effector function, but in the presence of normal eosinophils it acts as a blocking antibody with detrimental consequences for the expression of protective immunity (Butterworth et al. Reference Butterworth, Dunne, Fulford, Capron, Khalife, Capron, Koech, Ouma and Sturrock1988a; Khalife et al. Reference Khalife, Dunne, Richardson, Mazza, Thorne, Capron and Butterworth1989). The latter observations have been further supported by in vitro studies of antibody-dependent cell-mediated cytotoxicity to schistosomula in which killing was mediated by IgG1 and IgG3, but not by IgG2 (Khalife et al. Reference Khalife, Dunne, Richardson, Mazza, Thorne, Capron and Butterworth1989). IgG4 can block protective IgE activity (Wachholz and Durham, Reference Wachholz and Durham2004), and a high level of IgG4 antibodies, or rather an increased IgG4/IgE ratio, is associated with susceptibility to re-infection (Hagan et al. Reference Hagan, Blumenthal, Dunn, Simpson and Wilkins1991; Dunne et al. Reference Dunne, Butterworth, Fulford, Kariuki, Langley, Ouma, Capron, Pierce and Sturrock1992; Viana et al. Reference Viana, Correa-Oliveira, Carvalho, Massara, Colosimo, Colley and Gazzinelli1995; Satti et al. Reference Satti, Lind, Vennervald, Sulaiman, Daffalla and Ghalib1996; Zhang et al. Reference Zhang, Wu, Chen, Hu, Xie, Qiu, Su, Cao, Wu, Zhang and Wu1997; Li et al. 2001; Jiz et al. Reference Jiz, Friedman, Leenstra, Jarilla, Pablo, Langdon, Pond-Tor, Wu, Manalo, Olveda, Acosta and Kurtis2009; Pinot de Moira et al. Reference Pinot de Moira, Fulford, Kabatereine, Ouma, Booth and Dunne2010). Elevated IL-10 levels are also considered a major risk factor for re-infection (Van den Biggelaar et al. Reference Van den Biggelaar, Borrmann, Kremsner and Yazdanbakhsh2002; Leenstra et al. Reference Leenstra, Acosta, Wu, Langdon, Solomon, Manalo, Su, Jiz, Jarilla, Pablo, McGarvey, Olveda, Friedman and Kurtis2006; Mutapi et al. Reference Mutapi, Winborn, Midzi, Taylor, Mduluza and Maizels2007; Caldas et al. Reference Caldas, Campi-Azevedo, Oliveira, Silveira, Oliveira and Gazzinelli2008), possibly by stimulating B cells to switch to IgG4 production (Wachholz and Durham, Reference Wachholz and Durham2004).
The individual target antigens of these antibodies, protective or not, are largely unknown. Relatively few protein antigens are exposed to the host in the schistosome tegument, but more antigens become exposed when schistosomes die and fall apart, possibly giving rise to more effective immune responses and the development of immunity against Schistosoma infection. Humans treated with PZQ, which disrupts the schistosome tegument thereby exposing underlying antigens to the host, develop a serological profile similar to that of resistant individuals (Mutapi et al. Reference Mutapi, Ndhlovu, Hagan, Spicer, Mduluza, Turner, Chandiwana and Woolhouse1998; Correa-Oliveira et al. Reference Correa-Oliveira, Caldas and Gazzinelli2000). IgE, IgG3 and IgA levels to adult worm antigens and SEA become higher after treatment with PZQ whereas IgG4 levels decrease (Webster et al. Reference Webster, Fallon, Fulford, Butterworth, Ouma, Kimani and Dunne1997a, Reference Webster, Fallon, Fulford, Butterworth, Ouma, Kimani and Dunneb; Mutapi et al. Reference Mutapi, Hagan, Woolhouse, Mduluza and Ndhlovu2003).
SCHISTOSOMA GLYCAN ANTIGENS
While major interest is focused on surface-exposed and/or secretory proteins of schistosome larval and adult worm stages to discover novel targets for immune intervention by vaccination, the antigenic carbohydrate chains (glycans) expressed on schistosome proteins have been receiving far less attention. Specific repertoires of glycans with numerous different structural characteristics are abundantly expressed on secreted and membrane-bound proteins of each schistosome life stage (Nyame et al. Reference Nyame, Kawar and Cummings2004; Hokke and Yazdanbakhsh, Reference Hokke and Yazdanbakhsh2005; Hokke et al. Reference Hokke, Deelder, Hoffmann and Wuhrer2007a) and these glycans are capable of, or involved in the activation and modulation of the host immune response (Velupillai et al. Reference Velupillai, dos Reis, dos Reis and Harn2000; Okano et al. Reference Okano, Satoskar, Nishizaki and Harn2001; Van der Kleij et al. Reference Van der Kleij, van Remoortere, Schuitemaker, Kapsenberg, Deelder, Tielens, Hokke and Yazdanbakhsh2002; Hokke and Yazdanbakhsh, Reference Hokke and Yazdanbakhsh2005; Van Die and Cummings, Reference Van Die and Cummings2010). Furthermore, high levels of antibodies against glycan antigens are generated during natural and experimental Schistosoma infection (Eberl et al. Reference Eberl, Langermans, Vervenne, Nyame, Cummings, Thomas, Coulson and Wilson2001; Hokke and Deelder, Reference Hokke and Deelder2001; Nyame et al. Reference Nyame, Lewis, Doughty, Correa-Oliveira and Cummings2003; Hokke et al. Reference Hokke, Deelder, Hoffmann and Wuhrer2007a, Reference Hokke, Fitzpatrick and Hoffmannb; Kariuki et al. Reference Kariuki, Farah, Wilson and Coulson2008), indicating that it would be worth examining the impact of these antibody responses on the development of immunity to schistosome infection.
Elaborate structural and biochemical studies on schistosome glycans expressed throughout the different life stages of the parasite have indicated that hundreds of different glycan structures are present within the N- and O-linked glycans and the glycolipids (Hokke et al. Reference Hokke, Deelder, Hoffmann and Wuhrer2007a). Most of these glycomics data have been generated by various mass spectrometric techniques from whole parasite extracts, although some studies have been performed on more restricted groups of proteins (e.g. secretions, or single purified glycoproteins). The occurrence of the different glycans and glycan motifs throughout the schistosome life stages within the mammalian host are summarized below.
Glycans of cercariae and schistosomula
A clear feature of both protein- and lipid-linked glycans of cercariae is the abundance the immunogenic glycan motif Galβ1-4(Fucα1-3)GlcNAc (Lewis X, LeX, see Table 1 for definition of the glycan elements). The majority of cercarial N-glycans are modified with a core β2-xylose as well as an α6-fucose, while no core α3-fucosylated glycans are detected in this life stage (Khoo et al. Reference Khoo, Huang and Lee2001; Hokke et al. Reference Hokke, Deelder, Hoffmann and Wuhrer2007a). Furthermore, LDN-based structures are present as part of the cercarial glycocalyx O-glycans, but in glycolipids and N-glycans they occur only in minor amounts (Wuhrer et al. Reference Wuhrer, Dennis, Doenhoff, Lochnit and Geyer2000; Huang et al. Reference Huang, Tsai and Khoo2001; Khoo et al. Reference Khoo, Huang and Lee2001; Hokke et al. Reference Hokke, Deelder, Hoffmann and Wuhrer2007a; Jang-Lee et al. Reference Jang-Lee, Curwen, Ashton, Tissot, Mathieson, Panico, Dell, Wilson and Haslam2007). Interestingly, the cercarial glycocalyx has been reported to carry complex O-glycans with repeating units of unique multi-fucosylated (Fucα1-2Fucα1-3, DF) LDN motifs (Khoo et al. Reference Khoo, Sarda, Xu, Caulfield, McNeil, Homans, Morris and Dell1995; Huang et al. Reference Huang, Tsai and Khoo2001). The multi-fucosylated LDN-motifs were found however in only low abundance on the cercarial excretory/secretory proteins (Jang-Lee et al. Reference Jang-Lee, Curwen, Ashton, Tissot, Mathieson, Panico, Dell, Wilson and Haslam2007). In addition to LeX, the cercarial glycolipids express the Fucα1-3Galβ1-4(Fucα1-3)GlcNAc (pseudo-LeY) motif (Wuhrer et al. Reference Wuhrer, Dennis, Doenhoff, Lochnit and Geyer2000), which to date has not been observed in other S. mansoni life stages.
Antibody responses to defined antigenic glycan elements in natural and experimental schistosome infections
blue square, N-acetylglucosamine; yellow square, N-acetylgalactosamine; green circle, mannose; yellow circle, galactose; red triangle, fuccose; open star, xylose; blue/white diamond, glucuronic acid.
a N-glycan core modifications which can be present separately or combined.
b Major expression of glycan elements on glycoproteins and/or glycolipids in cercariae (cer), schistosomula (som), adult worms (worms) and eggs. P, present as a protein conjugate; L, present as a lipid conjugate. Mainly based on Hokke et al. Reference Hokke, Deelder, Hoffmann and Wuhrer2007a, Reference Hokke, Fitzpatrick and Hoffmann2007b.
c Only antibody responses in infected mice (M), primates (P) and humans (H) are noted.
d Monoclonal antibodies against the glycan were obtained from S. mansoni infected mice. No elevated antibody levels were found in sera of infected humans.
The glycosylation of the schistosomula which develop after transformation of the penetrating cercariae is less thoroughly studied and limited data on glycan structures are available. While O-linked and lipid glycosylation have never been analysed, one mass spectrometric analysis of N-glycosylation of in vitro transformed 3-day old schistosomula exists (Hokke et al. Reference Hokke, Deelder, Hoffmann and Wuhrer2007a). In comparison to cercariae, the expression of LeX-containing glycans is reduced, and truncated N-glycans are more prevalent. Xylosylation of complex glycans is nearly absent, but a major fraction of truncated glycans still carries this motif. Monoclonal antibody (mAb) studies have indicated the presence of LeX, LDN and GalNAcβ1-4(Fucα1-3)GlcNAc (LDN-F) on the surface of schistosomulae (Koster and Strand, Reference Koster and Strand1994; Nyame et al. Reference Nyame, Lewis, Doughty, Correa-Oliveira and Cummings2003). As LDN and LDN-F motifs are not clearly detectable on N-glycans of schistosomula, these might be expressed by O-glycans and/or glycolipids.
Glycans of adult worms
Upon maturation of the larvae into adult worms, xylosylation and α3-core fucosylation of N-glycans decreases further, and also N-glycans with LeX motifs become less abundant. Instead, N-glycosylation of adult worms is mainly characterized by α6-core fucosylated, mono- and di-antennary glycans terminating with LDN (Wuhrer et al. Reference Wuhrer, Koeleman, Deelder and Hokke2006b; Hokke et al. Reference Hokke, Deelder, Hoffmann and Wuhrer2007a). Minor subsets include diantennary glycans with mixed LDN, N-acetyllactosamine (LN) and LeX termini as well as linear repeats of these structures (Wuhrer et al. Reference Wuhrer, Koeleman, Deelder and Hokke2006b). Although male and female glycans in general display a similar N-glycosylation profile, subtle differences in the minor glycan subsets are observed, with females expressing more LN/LeX-type glycans, whereas LDN/LDN-F-type glycans are more prevalent in males (Wuhrer et al. Reference Wuhrer, Koeleman, Fitzpatrick, Hoffmann, Deelder and Hokke2006c). Immunofluorescence studies using mAbs revealed that these gender-specific glycans were at least in part found on the tegument, which might have consequences for immune responses elicited by the two sexes. O-glycans could not be directly detected within the adult worm extract also used to characterize the N-glycans (Wuhrer et al. Reference Wuhrer, Koeleman, Deelder and Hokke2006b), but previously worms were shown to excrete the highly antigenic circulating cathodic antigen (CCA) and circulating anodic antigen (CAA) from the gut that carry long O-linked carbohydrate chains containing repeats of LeX units and a unique GlcA-substituted GalNAc polymer, respectively (Bergwerff et al. Reference Bergwerff, van Dam, Rotmans, Deelder, Kamerling and Vliegenthart1994; Van Dam et al. Reference Van Dam, Bergwerff, Thomas-Oates, Rotmans, Kamerling, Vliegenthart and Deelder1994). Worm glycolipids have to date been poorly defined in terms of glycosylation. However, using defined anti-glycan antibodies, TLC overlays of worm glycolipids indicated the presence of (multi-)fucosylated LDN structures including LDN-F and LDN-DF (Robijn et al. Reference Robijn, Wuhrer, Kornelis, Deelder, Geyer and Hokke2005), as well as the presence of LeX (Van Stijn et al. Reference Van Stijn, Meyer, van den Broek, Bruijns, van Kooyk, Geyer and van Die2010).
Glycans of eggs and miracidia
The glycan profile of eggs evidently differs from that of adult worms. Within the N-glycan pool, β2-core xylosylation as observed in the cercarial stage re-appears, and a set of α3-core fucosylated glycans can be detected (Khoo et al. Reference Khoo, Chatterjee, Caulfield, Morris and Dell1997; Hokke et al. Reference Hokke, Deelder, Hoffmann and Wuhrer2007a). As in most other life stages, antenna structures on N-, O- glycans for a large part consist of fucosylated LN and LDN motifs, including LeX and LDN-F structures (Khoo et al. Reference Khoo, Chatterjee, Caulfield, Morris and Dell1997; Wuhrer et al. Reference Wuhrer, Kantelhardt, Dennis, Doenhoff, Lochnit and Geyer2002). This is in line with the observations for major secretory egg-glycoproteins omega-1, IPSE/α1 and κ5 which carry these types of terminal motifs on a α3/α6-difucosylated N-glycan core, in the case of κ5 in a unique combination with core-linked xylose (Wuhrer et al. Reference Wuhrer, Balog, Catalina, Jones, Schramm, Haas, Doenhoff, Dunne, Deelder and Hokke2006a; Jang-Lee et al. Reference Jang-Lee, Curwen, Ashton, Tissot, Mathieson, Panico, Dell, Wilson and Haslam2007; Meevissen et al. Reference Meevissen, Wuhrer, Doenhoff, Schramm, Haas, Deelder and Hokke2010, Reference Meevissen, Balog, Koeleman, Doenhoff, Schramm, Haas, Deelder, Wuhrer and Hokke2011a). Another characteristic feature of egg glycans is the occurrence of multi-fucosylated antenna structures containing the Fucα1-2Fucα1-3 (DF) motif (Bergwerff et al. Reference Bergwerff, Thomas-Oates, Van Oostrum, Kamerling and Vliegenthart1992; Khoo et al. Reference Khoo, Chatterjee, Caulfield, Morris and Dell1997; Jang-Lee et al. Reference Jang-Lee, Curwen, Ashton, Tissot, Mathieson, Panico, Dell, Wilson and Haslam2007). On glycolipids, these motifs are expressed in the form of repeating -4(Fucα1-2Fucα1-3)GlcNAcβ1-units terminating with (Fucα1-2)0/1Fucα1-3GalNAcβ1- at the non-reducing end. Notably, egg glycolipids do not seem to express the LeX element (Robijn et al. Reference Robijn, Wuhrer, Kornelis, Deelder, Geyer and Hokke2005). The egg shell surface most likely contains both N- and O-glycans and the presence of fucosylated LDN and LeX was shown by monoclonal antibodies (Dewalick et al. Reference Dewalick, Bexkens, van Balkom, Wu, Smit, Hokke, de Groot, Heck, Tielens and van Hellemond2011), but the precise composition of these glycans remains unknown.
The N-glycans of miracidia, which constitute the major part of mature eggs are indeed very similar to the ones found in egg extracts (Hokke et al. Reference Hokke, Deelder, Hoffmann and Wuhrer2007a). However, eggs contain additional glycan structures such as those found on omega-1 and IPSE/α1, which are expressed in the sub-shell area within the egg (Hokke et al. Reference Hokke, Deelder, Hoffmann and Wuhrer2007a). The glycan structures of miracidial O-glycans and glycolipids have not been analysed yet, but, as observed for N-glycans, are expected to be largely similar to the respective egg glycans.
ANTIBODY RESPONSES AGAINST SCHISTOSOMA GLYCAN ELEMENTS
It has been shown that high levels of antibodies directed against glycan epitopes are present in sera from Schistosoma infected individuals, and an unidentified subset of these antibodies may be protective (Richter et al. Reference Richter, Incani and Harn1996; Nyame et al. Reference Nyame, Leppanen, Bogitsh and Cummings2000; Van Remoortere et al. Reference Van Remoortere, Hokke, van Dam, van Die, Deelder and van den Eijnden2000; Eberl et al. Reference Eberl, Langermans, Vervenne, Nyame, Cummings, Thomas, Coulson and Wilson2001; Naus et al. Reference Naus, van Remoortere, Ouma, Kimani, Dunne, Kamerling, Deelder and Hokke2003; Kariuki et al. Reference Kariuki, Farah, Wilson and Coulson2008). These observations led to the hypothesis that glycans may also form the basis for a vaccine that induces antibody-mediated protection against schistosomes. The molecular nature of the glycan epitopes recognized by antibodies in natural schistosomiasis infection serum is however still largely unknown. Below we summarize current knowledge of specific glycan targets of antibodies in schistosomiasis.
The Lewis X motif
Antibodies against the LeX element have been identified in sera from naturally infected humans (Nyame et al. Reference Nyame, Pilcher, Tsang and Cummings1996, Reference Nyame, Lewis, Doughty, Correa-Oliveira and Cummings2003; Van Remoortere et al. Reference Van Remoortere, van Dam, Hokke, van den Eijnden, van Die and Deelder2001), experimentally infected primates (Nyame et al. Reference Nyame, Pilcher, Tsang and Cummings1996; Eberl et al. Reference Eberl, Langermans, Vervenne, Nyame, Cummings, Thomas, Coulson and Wilson2001) and rodents (Richter et al. Reference Richter, Incani and Harn1996; Nyame et al. Reference Nyame, Pilcher, Tsang and Cummings1997; Van Remoortere et al. Reference Van Remoortere, Hokke, van Dam, van Die, Deelder and van den Eijnden2000; Van Roon et al. Reference Van Roon, van de Vijver, Jacobs, van Marck, van Dam, Hokke and Deelder2004). All species investigated generated mainly IgM antibodies against the LeX element although low levels of IgG and IgA were observed as well (Richter et al. Reference Richter, Incani and Harn1996; Nyame et al. Reference Nyame, Pilcher, Tsang and Cummings1997; Van Roon et al. Reference Van Roon, van de Vijver, Jacobs, van Marck, van Dam, Hokke and Deelder2004). LeX is expressed not only in monomeric but also in linear oligomeric and polymeric forms, the latter as the major immunogenic part of CCA (Van Dam et al. Reference Van Dam, Bergwerff, Thomas-Oates, Rotmans, Kamerling, Vliegenthart and Deelder1994). Interestingly, the response against monomeric LeX was much lower than the response against di- and trimeric LeX with a different temporal response pattern in mice. Both antibody responses peaked at 8 weeks after infection, but the anti-di- and tri-meric LeX response rapidly declined while the response against the monomeric LeX showed a gradual decline (Van Roon et al. Reference Van Roon, van de Vijver, Jacobs, van Marck, van Dam, Hokke and Deelder2004). Furthermore, the response pattern for the polymeric LeX containing CCA was different from the anti-mono-, di- and trimeric responses as it appeared later and is more prolonged (Van Roon et al. Reference Van Roon, van de Vijver, Jacobs, van Marck, van Dam, Hokke and Deelder2004). Together these observations indicate that specific and different antibody responses are generated against structurally related but non-identical presentations of the LeX antigen.
The LeX element is also present on a limited set of cell surface expressed and/or secretory glycolipids and glycoproteins of the human host, and it has been shown that anti-LeX antibodies from sera of infected humans are able to mediate complement-dependent cytolysis (Nyame et al. Reference Nyame, Pilcher, Tsang and Cummings1996, Reference Nyame, Pilcher, Tsang and Cummings1997). Whether this also leads to significant pathology in vivo is unknown but it is an important reason why LeX is not a straightforward potential glycan-vaccine candidate. Since schistosomes express different structural forms of LeX associated with different antibody responses, specific LeX conjugates or multimers may still be explored as vaccine candidates (Van Dam et al. Reference Van Dam, Bergwerff, Thomas-Oates, Rotmans, Kamerling, Vliegenthart and Deelder1994; Van Roon et al. Reference Van Roon, van de Vijver, Jacobs, van Marck, van Dam, Hokke and Deelder2004). It is worth noting that in particular the LeX motif, as part of synthetic or natural glycoconjugates, has also been shown to harbour immunomodulatory properties via interactions with antigen presenting cells such as DC and macrophages (Harn et al. Reference Harn, McDonald, Atochina and Da'dara2009). Several other schistosome glycans, including LDN and LDN-F, may also be involved in inducing innate and modulatory immune mechanisms in the host, in addition to being a direct target of the antibody response (Van Die and Cummings, Reference Van Die and Cummings2010; Meevissen et al. Reference Meevissen, Yazdanbakhsh and Hokke2011b).
LDN and its fucosylated derivatives
Antibody responses against LDN and fucosylated variants have been shown in humans (Van Remoortere et al. Reference Van Remoortere, van Dam, Hokke, van den Eijnden, van Die and Deelder2001; Naus et al. Reference Naus, van Remoortere, Ouma, Kimani, Dunne, Kamerling, Deelder and Hokke2003; Nyame, Reference Nyame, Lewis, Doughty, Correa-Oliveira and Cummings2003), primates and mice (Nyame et al. Reference Nyame, Leppanen, DeBose-Boyd and Cummings1999, Reference Nyame, Leppanen, Bogitsh and Cummings2000; Van Remoortere et al. Reference Van Remoortere, Hokke, van Dam, van Die, Deelder and van den Eijnden2000; Eberl et al. Reference Eberl, Langermans, Vervenne, Nyame, Cummings, Thomas, Coulson and Wilson2001). Schistosomes express the more exceptional glycan motifs F-LDN (Kantelhardt et al. Reference Kantelhardt, Wuhrer, Dennis, Doenhoff, Bickle and Geyer2002) and multifucosylated LDN (Khoo et al. Reference Khoo, Sarda, Xu, Caulfield, McNeil, Homans, Morris and Dell1995, Reference Khoo, Chatterjee, Caulfield, Morris and Dell1997; Wuhrer et al. Reference Wuhrer, Kantelhardt, Dennis, Doenhoff, Lochnit and Geyer2002) as well as the more widely expressed glycans LDN and LDN-F which are shared between schistosomes and mammalian hosts including humans (Hakomori et al. Reference Hakomori, Nudelman, Levery, Solter and Knowles1981; Fox et al. Reference Fox, Damjanov, Knowles and Solter1983; Spooncer et al. Reference Spooncer, Fukuda, Klock, Oates and Dell1984; Fukuda et al. Reference Fukuda, Dell, Oates, Wu, Klock and Fukuda1985; Van Kuik et al. Reference Van Kuik, de Waard, Vliegenthart, Klein, Carnoy, Lamblin and Roussel1991; Yan et al. Reference Yan, Chao and van Halbeek1993; Bergwerff et al. Reference Bergwerff, van Oostrum, Kamerling and Vliegenthart1995; Dell et al. Reference Dell, Morris, Easton, Panico, Patankar, Oehniger, Koistinen, Koistinen, Seppala and Clark1995; Khoo et al. Reference Khoo, Sarda, Xu, Caulfield, McNeil, Homans, Morris and Dell1995; Van den Eijnden et al. Reference Van den Eijnden, Neeleman, van der Knaap, Bakker, Agterberg and van Die1995). Antibody responses to these less specific LDN and LDN-F motifs are generally low and predominantly of the IgM type, while antibody responses against the more exceptional elements LDN-DF and F-LDN are more pronounced and are predominantly of the IgG isotype in infected humans and chimpanzees (Khoo et al. Reference Khoo, Chatterjee, Caulfield, Morris and Dell1997; Van Remoortere et al. Reference Van Remoortere, van Dam, Hokke, van den Eijnden, van Die and Deelder2001; Kantelhardt et al. Reference Kantelhardt, Wuhrer, Dennis, Doenhoff, Bickle and Geyer2002; Naus et al. Reference Naus, van Remoortere, Ouma, Kimani, Dunne, Kamerling, Deelder and Hokke2003). For LDN-DF it has been shown that isotypes can differ between age groups and groups infected with different Schistosoma species. Sera of S. japonicum individuals contained mainly IgG to LDN-DF, while S. mansoni-infected individuals show IgM antibody responses. Both isotypes were observed in high amounts in S. haematobium infected individuals (Van Remoortere et al. Reference Van Remoortere, van Dam, Hokke, van den Eijnden, van Die and Deelder2001). For all Schistosoma species, the antibody response against LDN-DF is higher in children than in adults, which may or may not be the result of the higher infection intensities generally observed in children. While S. mansoni infection is dominated by IgM in residents of an endemic area, an immigration study showed that young children upon their first year of exposure induce high IgG1 mediated responses against LDN-DF that decline within 2 years to levels comparable with children are exposed and infected more frequently (Naus et al. Reference Naus, van Remoortere, Ouma, Kimani, Dunne, Kamerling, Deelder and Hokke2003). This indicates that antibody levels as well as isotypes/subclasses to specific glycan antigens are correlated to specific infection characteristics.
Core-α3-Fuc and core-β2-Xyl
The core-α3-Fuc and core-β2-Xyl modifications occurring in schistosomes also occur in other invertebrates and in plants, but not in mammalian species. Sera from S. mansoni-infected mice contain antibodies that specifically reacted with complex type glycans containing these modifications since the antibodies bound to proteins of S. mansoni and Arabidopsis thaliana but not to proteins from mutant A. thaliana lacking the core antigens (Van Die et al. Reference Van Die, Gomord, Kooyman, van den Berg, Cummings and Vervelde1999). Interestingly, these antibodies were of the IgE isotype and not IgM or IgG which is mainly the case for all other glycan antigens described. The generation of IgE against core-α3-Fuc and/or core-β2-Xyl suggests that antibodies against these elements are generated later in infection during the development of a Th2 response (Faveeuw et al. Reference Faveeuw, Mallevaey, Paschinger, Wilson, Fontaine, Mollicone, Oriol, Altmann, Lerouge, Capron and Trottein2003). This is consistent with the presence of core-α3-Fuc in schistosomes on egg and miracidia glycans only (Khoo et al. Reference Khoo, Chatterjee, Caulfield, Morris and Dell1997; Hokke et al. Reference Hokke, Deelder, Hoffmann and Wuhrer2007a). In natural human infection, the major anti-glycan IgE response appears to be directed against the N-glycan core structure of the major egg glycoprotein κ5, which uniquely carries both β2-Xyl and α3/α6-core-difucosylation (Meevissen et al. Reference Meevissen, Balog, Koeleman, Doenhoff, Schramm, Haas, Deelder, Wuhrer and Hokke2011a). The core-β2-Xyl modification, often in combination with the commonly occurring α6-core Fuc is also present on cercariae and schistosomula glycans and is therefore exposed to the host immune system at an earlier stage after infection (Hokke et al. Reference Hokke, Deelder, Hoffmann and Wuhrer2007a, Reference Hokke, Fitzpatrick and Hoffmannb). It would be interesting to explore further IgE and other types of antibodies against core-α3-Fuc and core-β2-Xyl in infection in humans and determine if they play a role in immunity.
Antibody responses to Schistosoma glycans: protective or a smoke screen?
In immunization studies with radiation-attenuated cercariae which induce protection in animal models, strong antibody responses against glycans are observed (Richter et al. Reference Richter, Incani and Harn1996; Eberl et al. Reference Eberl, Langermans, Vervenne, Nyame, Cummings, Thomas, Coulson and Wilson2001; Kariuki et al. Reference Kariuki, Farah, Wilson and Coulson2008). It has been argued that in general these responses are merely a smoke screen rather than involved in the protective response, but it could also be hypothesised that responses to specific subset of glycan elements may be be linked to protective immunity. Cytolytic capacity and protective properties have been described for glycan-specific antibodies (Nyame et al. Reference Nyame, Pilcher, Tsang and Cummings1996, Reference Nyame, Pilcher, Tsang and Cummings1997, Reference Nyame, Lewis, Doughty, Correa-Oliveira and Cummings2003). IgE directed against glycolipids has been shown to be negatively correlated with egg output at 2 years post-treatment, indicating that it could play a role in resistance to reinfection (Van der Kleij et al. Reference Van der Kleij, Tielens and Yazdanbakhsh1999). However, IgM and IgG2 antibodies that reacted with carbohydrate epitopes expressed on the surface of schistosomula and eggs were negatively associated with resistance to reinfection which is probably due to blocking activity of the antibody isotypes (Butterworth et al. Reference Butterworth, Dunne, Fulford, Capron, Khalife, Capron, Koech, Ouma and Sturrock1988a). These observations for anti-glycan antibodies are in line with the protective IgE and blocking IgM/IgG2 responses described above. In vaccination/infection studies in chimpanzees (Eberl et al. Reference Eberl, Langermans, Vervenne, Nyame, Cummings, Thomas, Coulson and Wilson2001) and baboons (Kariuki et al. Reference Kariuki, Farah, Wilson and Coulson2008) it was noticed that anti-glycan antibodies were predominantly produced against cercarial and egg secretions of Schistosoma during the early phases of infection but that antibodies to peptide epitopes become more prominent during the chronic phase of infection when protective immune responses are generated (Eberl et al. Reference Eberl, Langermans, Vervenne, Nyame, Cummings, Thomas, Coulson and Wilson2001). These observations support the smoke screen theory which reasons that high antibody responses towards glycans are beneficial for the parasite rather than the host by subverting the immune system away from epitopes that could provoke protective immune responses. This hypothesis was further supported by the perception that immunization with eggs resulted in high antibody titers in mice, which were cross-reactive with cercarial and egg secretions, but did not result in increased protection (Kariuki et al. Reference Kariuki, Farah, Wilson and Coulson2008).
It will be necessary to study in more detail responses to individual glycan antigens rather than to crude glycoprotein mixtures, in particular to those that are expressed by larval and worm stages but not by eggs, before conclusions can be drawn about their detrimental or beneficial effects to host immunity.
SHOTGUN GLYCAN MICROARRAYS: EFFECTIVE TOOLS FOR STUDYING ANTIBODY-GLYCAN INTERACTIONS
The notion that high antibody titers against glycan elements do not correlate with protection per se makes the identification of potential vaccine candidates more complex and asks for more advanced strategies than simply identifying glycan elements towards which antibodies are generated. So far research has been focused on antibody responses against a limited set of synthetic glycoconjugates representing antigenic schistosome glycan elements such as listed in Table 1. It is very likely that in a Schistosoma infection antibodies are generated against a much wider range of glycan elements not tested so far, and that the larger underlying glycan structure as produced by the schistosome itself contributes to antibody specificity and affinity, and thereby also immunological functionality. Additionally, a protective immune response may be formed by the combined action of multiple antibodies against various glycans and glycoproteins and rather than against a single antigen. With the recent development of glycan microarrays it is possible to study antibody responses against multiple glycans simultaneously. Glycan microarrays contain small amounts of a large number of glycans presented on a surface to quantitatively measure their interaction with complementary molecules, analogous to arrays developed for gene transcription analysis or the study of protein-protein interactions (Bergwerff et al. Reference Bergwerff, van Dam, Rotmans, Deelder, Kamerling and Vliegenthart1994; Blixt et al. Reference Blixt, Head, Mondala, Scanlan, Huflejt, Alvarez, Bryan, Fazio, Calarese and Stevens2004; Bochner et al. Reference Bochner, Alvarez, Mehta, Bovin, Blixt, White and Schnaar2005; Gryseels et al. Reference Gryseels, Polman, Clerinx and Kestens2006; De Boer et al. Reference De Boer, Hokke, Deelder and Wuhrer2007; Lonardi et al. Reference Lonardi, Balog, Deelder and Wuhrer2010; Smith et al. Reference Smith, Song and Cummings2010). So far, glycan microarrays have been explored in particular for studying glycan ligands of mammalian lectins, for example in the framework of the Consortium of Functional Glycomics in the USA (www.functionalglycomics.org). In different versions, these glycan microarrays contain up to about 600 synthesized glycan structures and motifs mainly related to the mammalian glycome (Blixt et al. Reference Blixt, Head, Mondala, Scanlan, Huflejt, Alvarez, Bryan, Fazio, Calarese and Stevens2004; Bochner et al. Reference Bochner, Alvarez, Mehta, Bovin, Blixt, White and Schnaar2005; Smith et al. Reference Smith, Song and Cummings2010). While efficiently generating high quality data and new hypotheses, drawbacks of these arrays are the dependence on laborious and time-consuming synthesis of glycans to be printed on the arrays and the relatively limited glycan repertoire covered.
These limitations can be overcome with the development of shotgun glycan microarray approaches. Shotgun glycan microarrays have the potential to study protein-glycan interactions at the whole natural glycome level (De Boer et al. Reference De Boer, Hokke, Deelder and Wuhrer2007; Song et al. Reference Song, Lasanajak, Xia, Heimburg-Molinaro, Rhea, Ju, Zhao, Molinaro, Cummings and Smith2011). These arrays consist of glycans isolated directly from relevant cells or organisms thereby completely avoiding the need for synthetic glycan structures. Also a major advantage of the shotgun glycan microarray approach versus the conventional synthetic array approach is the inclusion of unique and unusual (e.g. pathogen specific) glycans that would not be available through chemical synthesis because these glycans have simply never been structurally identified. The natural glycans are obtained via routine analytical procedures and, after chromatographic separation, individual glycans from complex glycomes can be printed in an array format. When constructed of pathogen-derived glycans, such arrays can for instance be applied to determine specific anti-glycan antibodies by incubating the arrays with infection sera and subsequent detection with fluorescently labeled secondary antibodies Glycan microarrays have already been applied to examine anti-glycan antibodies in sera of healthy individuals (Oyelaran et al. Reference Oyelaran, McShane, Dodd and Gildersleeve2009), of S mansoni-infected subjects (De Boer et al. Reference De Boer, Hokke, Deelder and Wuhrer2008), and of patients with Lyme's disease (Song et al. Reference Song, Lasanajak, Xia, Heimburg-Molinaro, Rhea, Ju, Zhao, Molinaro, Cummings and Smith2011), indicating the potential of glycan arrays as a tool for anti-glycan antibody profiling. Comparing antibody response profiles in sera of infected, non-infected and resistant individuals and cohorts using these shotgun glycan microarrays could be a promising strategy to discover glycan antigens and vaccine candidates for various infectious diseases (Oyelaran and Gildersleeve, Reference Oyelaran and Gildersleeve2007).
SCHISTOSOMA GLYCANS AS VACCINE CANDIDATES
Despite the fact that antibody-mediated immune responses can confer resistance against re-infection with Schistosoma there is no effective vaccine available yet (Hotez et al. Reference Hotez, Bethony, Diemert, Pearson and Loukas2010; McManus and Loukas, Reference McManus and Loukas2008). Initial research with the radiation-attenuated schistosome vaccine showed high levels (up to 90%) of immunity in animal models against a challenge infection (Bickle, Reference Bickle2009) but with the current protein vaccine candidates such effective levels have not yet been reached (McManus and Loukas, Reference McManus and Loukas2008; Hotez et al. Reference Hotez, Bethony, Diemert, Pearson and Loukas2010). Although so far largely unexplored, glycans may also be considered as targets to develop vaccines to schistosome infections, and in fact other helminth infections in which antigenic glycans play equally prominent roles. Glycans are extensively surface exposed, they are present at a higher density than proteins, often in a multivalent pattern, and one single glycan can be expressed by multiple proteins thereby allowing the targeting of more than one protein by the same antibody.
Immunization with glycans generally induces poor antibody responses due to T-cell independent mechanisms that result in the production of IgM (Astronomo and Burton, Reference Astronomo and Burton2010). IgM is associated with blocking of immunity in Schistosoma infection (Butterworth et al. Reference Butterworth, Dunne, Fulford, Capron, Khalife, Capron, Koech, Ouma and Sturrock1988a) and not the isotype of choice in vaccine development. However, conjugation of glycans to a protein carrier results in T cell-dependent immune responses with subsequent production of IgG antibodies against the carbohydrate antigen. Such vaccines have been shown to confer protection against a variety of microbes including, Haemophilus influenza type b, Neisseria meningitidis, Salmonella typhi and Streptococcus pneumonia (reviewed in Astronomo and Burton, Reference Astronomo and Burton2010). By generating an anti-glycan antibody response of high class isotype that activates a protective mechanism via an appropriate specific glycan target a Schistosoma glycan-based vaccine may be feasible.
Although a significant amount of structural information on schistosome glycans is available, the identity of the proteins on which these glycans are expressed is known in only a few cases. Protein-specific information may in many cases easily be obtained by Western blot using monoclonal antibodies against glycan epitopes (Meevissen et al. Reference Meevissen, Wuhrer, Doenhoff, Schramm, Haas, Deelder and Hokke2010, Reference Meevissen, Balog, Koeleman, Doenhoff, Schramm, Haas, Deelder, Wuhrer and Hokke2011a). Vice versa, proteomic studies identifying promising vaccine targets often contain information on putative N- and O-glycosylation sites but lack identification of glycans that are expressed by the protein (Ribeiro de et al. 2000; Siddiqui et al. Reference Siddiqui, Phillips, Charest, Podesta, Quinlin, Pinkston, Lloyd, Pompa, Villalovos and Paz2003; Al-Sherbiny et al. Reference Al-Sherbiny, Osman, Barakat, El, Bergquist and Olds2003; Shalaby et al. Reference Shalaby, Yin, Thakur, Christen, Niles and LoVerde2003; Tran et al. Reference Tran, Pearson, Bethony, Smyth, Jones, Duke, Don, McManus, Correa-Oliveira and Loukas2006; Farias et al. Reference Farias, Cardoso, Miyasato, Montoya, Tararam, Roffato, Kawano, Gazzinelli, Correa-Oliveira, Coulson, Wilson, Oliveira and Leite2010). In particular, in studies on glycoprotein antigens it will be of importance to identify glycan structures on individual proteins. Antibodies to native glycoproteins may actually bind to the glycan structures rather than the protein. In addition, it is becoming clear that proper glycosylation of protein vaccines may be of utmost importance via its capacity to modulate and activate the anti-protein response induced by immunization (Okano et al. Reference Okano, Satoskar, Nishizaki and Harn2001; Singh et al. Reference Singh, Stephani, Schaefer, Kalay, Garcia-Vallejo, den Haan, Saeland, Sparwasser and van Kooyk2009, Reference Singh, Streng-Ouwehand, Litjens, Kalay, Burgdorf, Saeland, Kurts, Unger and van Kooyk2011; Van Montfort et al. Reference Van Montfort, Eggink, Boot, Tuen, Hioe, Berkhout and Sanders2011).
CONCLUDING REMARKS
In parallel with proteomics technologies, sensitive glycomic techniques are generating a wealth of structural information on glycans of schistosomes. Novel technologies such as glycan arrays now allow the screening of antibody responses to individual components of a pathogen glycome. Further research on schistosome glycan antigens will contribute to a better understanding of glycan-induced immune responses, and may contribute to the development of an effective vaccine against schistosomes and other infectious helminths.
