Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
E European Society of Human Reproduction and Embryology Human Reproduction Update 1999, Vol. 5, No.4 pp. 302–313 The glycocalyx of the sperm surface Sabine Schröter, Caroline Osterhoff, Wendy McArdle and Richard Ivell1 IHF Institute for Hormone and Fertility Research, University of Hamburg, Grandweg 64, 22529 Hamburg, Germany The surface of mammalian spermatozoa is covered by a dense coating of carbohydrate-rich molecules forming a 20–60 nm thick glycocalyx. The majority of sugar residues are attached to proteins which are either integrated within the sperm membrane, or are more or less loosely associated with it. It is estimated that there may be several hundred different glycoproteins comprising the glycocalyx, some of which are synthesized within the testis. Others, however, are produced by the epithelia of the efferent ducts, epididymis and possibly other accessory glands, and become associated with the spermatozoa post-testicularly during transit through, and storage in, the male tract. The acquisition of the mature glycocalyx is associated with the attainment of full sperm fertilizing ability. Until its complete molecular structure is elucidated, the complex function of the glycocalyx remains obscure, though it may be related to membrane maturation and immunoprotection in the female tract, as well as to sperm–zona binding and fertilization. Key words: cloned human antigens/epididymis/glycoconjugates/sperm maturation/sperm surface TABLE OF CONTENTS Introduction Structure and maturation of the sperm glycocalyx Origin and regulation of the glycoconjugates on the sperm surface Analysis of specific proteins forming part of the human sperm glycocalyx Glycosyl-modifying enzymes The function of the sperm glycocalyx Conclusion Acknowledgements References 302 303 304 305 308 310 311 311 311 Introduction The mammalian spermatozoon is a highly specialized epithelial cell whose main function exists free of its generative epithelial layer. It undergoes meiotic and mitotic differentiation within the testis, thereby acquiring a large number of proteins uniquely expressed in male germ cells. Before leaving the testis, ~90–95% of the cytoplasm is jettisoned in the form of the cytoplasmic droplet. The resulting gamete has been subjected to very rigorous natural selection to design a vehicle capable of transporting the haploid male nucleus to the oviduct of a recipient female coincident with the presence there of a mature oocyte, and to penetrate and fertilize that oocyte. The male gamete must therefore be specialized to be stored within the male tract for periods of time varying from several hours to several weeks; it must be able to penetrate the mucus secretions 1To whom correspondence should be addressed of the female reproductive tract and retain and metabolize sufficient energy to reach the oocyte in the oviduct; and finally it must be equipped with selective mechanisms to bind and penetrate the zona pellucida. It then has to fuse with and traverse the oocyte plasmalemma and undergo nuclear decondensation and zygote formation. Additionally, it has to be resilient to the immune defence mechanisms of the female tract, yet offer selection possibilities to sort good from poor quality spermatozoa, and be furnished with mechanisms which would prohibit interspecies fertilization. A number of these functions are also shared with certain small invading parasites, such as trypanosomes. It may therefore be helpful also to look at these organisms to identify which sperm components and mechanisms may be involved in some of these very special functions (McConville and Ferguson, 1993). It has been estimated that the outer membrane of ejaculated human spermatozoa includes at least 300 different proteins, as detectable by vectorial labelling of intact spermatozoa with 125I or biotin (Naaby Hansen et al., 1997). Some of these are glycosylated and go to form what is generally referred to as the sperm glycocalyx. Additionally, the glycocalyx comprises carbohydrate residues linked to lipids and other molecular structures. Not all such molecules are tightly bound to the sperm surface membrane. Whereas some component moieties are intercalated or anchored within the lipid bilayer [e.g. transmembrane- or glycosylphosphatidylinositol (GPI)-anchored proteins], others are only superficially associated with the membrane via polar groups (e.g. protein–protein or lectin–sugar interactions) or by hydrophobic interaction. The majority (92%) of extracellular membrane proteins of all cells are glycosylated, The sperm glycocalyx as a search in 1996 of the SWISSPROT protein database was able to show (Gahmberg and Tolvanen, 1996). From experiments using a variety of different methodologies (e.g. N-glycosidase F- and neuraminidase digestion, lectin-binding or periodate oxidation; reviewed in Benoff, 1997) it would appear that, also for spermatozoa, carbohydrates dominate the extracellular surface. Taking these observations together, it seems that almost all interactions between the spermatozoon and its environment must initially involve an interaction with the sperm glycocalyx. The aim of this review is to focus on certain aspects of the structure and function of the sperm glycocalyx with particular relevance to the human. Structure and maturation of the sperm glycocalyx The sperm glycocalyx represents the interface between the male gamete and the extracellular milieu. The female gamete is also encompassed by a heavily glycosylated matrix, the zona pellucida. However, unlike the oocyte zona pellucida, the sperm glycocalyx appears to be quite different both in dimensions and complexity. Whereas the thickness of the zona pellucida varies from 1–2 µm in marsupials to 16 µm in the pig (Yanagimachi, 1994), the glycocalyx of the spermatozoon appears to be considerably thinner. In the guinea pig, electron microscopic analysis of sectioned sperm heads shows a heteromorph layer varying in thickness from 20 to 60 nm apposed to the plasma membrane (Bearer and Friend, 1990). For comparison, the typical glycocalyx of a human erythrocyte, as an example of a typical somatic cell, extends only about 10 nm from the plasma membrane, itself being about 7 nm in thickness (Rademacher et al., 1988). Also, the sperm glycocalyx differs in complexity from the glycosylated investment of the oocyte. Only three glycoproteins (ZP1, ZP2 and ZP3; Yanagimachi, 1994) make up the main part of the zona pellucida, whereas it seems probable that the sperm glycocalyx comprises somewhere between 50 and 150 different glycoconjugates. Additionally, unlike the zona pellucida, numerous studies have shown that the sperm surface is not homogeneous but is composed of different functional domains, reflecting a distinctive and regulable spatial distribution of the glycocalyx components. It is suggested that this microgeographic heterogeneity and/or compartmentalization is part of the key to understanding the complex biological functions of the spermatozoon during its post-testicular existence (Bearer and Friend, 1990; Yanagimachi, 1994; Moore, 1996). In contrast to this apparent complexity of the sperm surface, labelling experiments on intact rat spermatozoa with galactose oxidase/NaB3H4 show a very high selectivity: in caudal spermatozoa, only one molecule of 32 kDa seems to be accessible to the action of the galactose oxidase, which could not be detected on testicular (Brown et al., 1983) or caput spermatozoa (Zeheb and Orr, 1984; Jones, 1989). This 303 molecule alone is thus supposed to form the major component of the outer part of the sperm glycocalyx. This 32 kDa molecule carries a GPI-anchor (Moore et al., 1989) and is highly glycosylated. Homologous gene products showing highly similar expression patterns have been found in all mammals investigated so far, including mouse (Kubota et al., 1990; Kirchhoff, 1994) and rat (Eccleston et al., 1994; Kirchhoff, 1994, 1996). Since this molecule appears on the sperm surface when spermatozoa acquire their fertilizing capacity, it was called the ‘major maturation antigen’ (Zeheb and Orr, 1984; Moore et al., 1989) or sperm membrane glycoprotein (SMemG; Eccleston et al., 1994). Spermatozoa are terminally differentiated cells and appear unable to synthesize proteins de novo. Kirchhoff and Hale (1996) have therefore postulated a cell-to-cell transfer of GPI-anchored proteins from the epithelium of the epididymal duct onto the sperm surface. During the epididymal passage, the acquisition of this highly glycosylated antigen seems to be one of the major events which form the sperm glycocalyx. The composition of the sperm glycocalyx has been investigated by a number of authors (Koehler, 1981; Lee and Wong, 1986; De Cerezo et al., 1996). The majority have made use of specific lectins, which bind to different types of sugar side-chain (Cummings, 1994), and thus together provide a picture of the sugar moieties forming the sperm surface. Whereas the majority of authors have looked at lectin-binding to ejaculated spermatozoa, some have used this technique to assess the changes in the glycocalyx during post-testicular sperm maturation, particularly during transit through the epididymis (Eddy et al., 1985; Margagee et al., 1988; Kumar et al., 1990; Bains et al., 1993; Arenas et al., 1996). Much confusion in the literature has arisen through the use of differing sperm preparation protocols prior to lectin application. A particular fixation protocol might expose epitopes not detectable on native spermatozoa, for example, on the inner acrosomal membrane, and disrupt the outer structures of the glycocalyx, thereby removing some sugar epitopes and exposing others (Haas et al., 1988). Additionally, lectin specificity can vary with lectin concentration, temperature and various other technical conditions, and needs to be checked by the appropriate use of competing sugars (Zeng and Gabius, 1992; Gabriel et al., 1994). A further factor of relevance is the secondary or tertiary structure of a particular glycan, which may strongly enhance the interaction with the corresponding lectin. This can lead to the very high affinities reported for some lectins (Koehler, 1981). In addition to the possible variability in lectin detection due to the use of differing protocols, there may also be considerable species differences in the lectin patterns reported, and it has been suggested that this could represent part of the cross-species fertilization barrier. Additionally, lectin binding may even vary within a species, in, for example, seasonal breeders (Calvo et al., 1995). Nevertheless, carbohydrate–lectin interactions of this type are considered to 304 S.Schröter et al. be one of the most important components of cell–cell recognition (Gahmberg and Tolvanen, 1996). Because of this inherent variability, attention has often focused on features which appear to be shared in common across a wide range of species. Runnebaum et al. (1995) have described ConA-binding (Concanavalin A; Canavalia ensiformis lectin) molecules on the acrosomal region of the sperm plasma membrane which are recognized by a specific monoclonal antibody across several species. This would imply that this epitope on the sperm surface is conserved and thus of physiological importance. Similar ConA-binding has even been demonstrated on the surface of Drosophila spermatozoa (Perotti and Pasini, 1995). A further feature associated with sperm maturation also shared in common by many species is the general increase in negative surface charge during maturation (Yanagimachi et al., 1972; Holt, 1980; Koehler, 1981; Eddy et al., 1985). This charge increase is probably due to the acquisition of sialic acid residues or acidic sperm antigens, and can be demonstrated by the increase in electrophoretic mobility of spermatozoa following epididymal transit (Bedford, 1963). It also correlates with a rise in WGA-binding (wheat germ agglutinin, Triticus vulgaris lectin; specific for sialic acid and N-acetylglucosamine, GlcNac) on the sperm surface (Arenas et al., 1996). Indeed, Fierro et al. (1996) showed a strong WGA-binding over the whole surface of ejaculated human spermatozoa, with presumably the thick layer of negatively charged sugars reported by Tsuji and colleagues (Tsuji et al., 1988; Tsuji, 1995) being responsible for this binding. Furthermore, Lassalle and Testart (1994) found a dose-dependent decrease in WGA-binding after digestion of motile human ejaculated spermatozoa with a neuraminidase (from Arthrobacter urefaciens) which preferentially cleaves terminal 2,6-linked sialic acid residues. In the rat, the major epididymally derived sperm surface proteins D and E are sialoproteins expressed in the proximal epididymis (Brooks, 1983). Kumar et al. (1990) showed that binding of WGA over the acrosomal region of the living sperm head increased during epididymal transit in five different species (rat, mouse, rabbit, hamster, goat) with a maximum in the corpus region. In the rhesus monkey, Navaneetham et al. (1996) also showed that in unfixed epididymal spermatozoa there was an increase in WGA-binding in the corpus epididymides. The increase in WGA-binding is probably caused by an increase in GlcNac and/or sialic acid residues; however, this could be due either to a local glycosylation reaction on pre-existing substrates on the sperm membrane, or to the transfer of new sialylated glycoconjugates to the spermatozoa during transit. Focusing on detection of lectin within the human epididymis, which appears to show less pronounced regionalization compared to other mammals, Arenas et al. (1996, 1998), using lectin-histochemistry, indicated a considerable increase in lectin-binding molecules along the epididymis. In the caudal lumen, epididymal fluid was positive for all lectins studied and therefore should contain β-GlcNac/sialic acid (WGA; wheat germ agglutinin, Triticus vulgaris lectin), α-mannose (ConA; Canavalia ensiformis lectin), α-fucose (UEA-1; Ulex europaeus lectin), α-GalNAc (DBA; Dolichos biflorus lectin), β-GalNAc (SBA; soybean agglutinin; Glycine max lectin) and β-Gal (PNA; peanut agglutinin, Arachis hypogaea lectin). In contrast, in the efferent ducts, the lumen was only positive for WGA-reacting components. Thus the epididymis is a site of high extracellular glycosyl metabolism, presumably involving the secretory activity of the principal cells lining the epididymis. However, whereas some of this activity may involve the spermatozoa in transit directly, a large part of this may have less to do with the spermatozoa than with the composition and properties of the epididymal luminal fluid, where the spermatozoa are retained prior to ejaculation. Origin and regulation of the glycoconjugates on the sperm surface Very little is known about non-protein glycoconjugates on the sperm surface, for example, glycolipid components of the glycocalyx. Most attention has been given to the glycoprotein components of the sperm membrane. In general, there appear to be at least three different types of glycoprotein interaction with the sperm membrane. Some glycoproteins are made within the testis and have defined transmembrane domains, with extracellular regions glycosylated on the outer cell surface. These would include typical hormone-receptor proteins and similar integral membrane proteins. Whereas the majority of such proteins will be of spermatozoal origin, it cannot be excluded that even this class of protein might be transferred post-testicularly from cells of the male tract through the agency of prostasome or seminosome apocrine vesicles (Aumuller et al., 1997). A second class of glycoproteins may be produced after completion of spermiogenesis either within the testis (for example, from Sertoli cells or from epithelial cells of the rete testis) or post-testicularly in the epididymis or subsequent parts of the male tract. Such glycoproteins could undergo polar interactions with integral polar moieties on the sperm membrane. These might be in the form of protein–protein interactions or lectin–sugar interactions. Some of these interactions might be specific and of high affinity, others may be of relatively low affinity, the glycoproteins being later released from the sperm surface, for example, during capacitation (e.g. HE4, see below; or the so-called ‘acrosome reaction inhibiting glycoprotein’, described by Drisdel et al., 1995). The spermadhesins also fall into this category (Töpfer-Petersen, 1999). One of the first indications for this sort of sperm surface interaction came from the work of Tezon et al. (1985). These authors showed that washing ejaculated human spermatozoa with high concentrations of salt removed a group of proteins which, when used as immunogen to raise The sperm glycocalyx 305 the outer membrane lipid layer (Kirchhoff, 1996; Kirchhoff and Hale, 1996). Analysis of specific proteins forming part of the human sperm glycocalyx Figure 1. Denaturing polyacrylamide gel electrophoresis of proteins eluted in 0.6 M NaCl from fresh human semen (lane 1) or from semen stored at 4°C overnight in the presence of proteinase inhibitors (lane 2). Lanes 3–12: lectin blot of the same proteins using different lectins as indicated, following incubation or not with N-glycosidase F. (GNA, Galanthus nivalis lectin; SNA, Sambucus nigra lectin; DSA, Datura stramonium lectin; MAA, Maackia amurensis lectin; PNA, Arachis hypogaea lectin, peanut agglutinin.) antibodies, indicated their source to be within the epididymis rather than within the testis. We have repeated these experiments and showed that treatment of isotonic saline-washed, ejaculated human spermatozoa with 0.6 M NaCl removed a group of ~20 proteins with molecular weights ranging from 10 from 100 kDa (Figure 1, left panel). Subsequent incubation of these proteins with lectins showed that the majority were indeed glycosylated, and that this glycosylation was largely eliminated by pretreatment with N-glycosidase F (Figure 1, right panel) to remove the N-glycosidic side chain from the primary amino acid substrate. Use of these proteins to raise monoclonal antibodies confirmed the observations of Tezon et al. (1985) for at least two of the proteins, i.e. that these originated within the epididymis and not the testis (McArdle, 1991). A third group of proteins, however, appears to undergo a secondary integration into the sperm plasma membrane. The best described of these are those furnished with a glycosylphosphatidylinositol (GPI) anchor structure, such as CD52 and CD59 (see below). In some way, which is still unclear, but may involve direct membrane–membrane or membrane–seminosome interactions, these molecules are transferred from the site of original synthesis, for example, in the principal cells of the epididymis, to the sperm plasma membrane, where the lipophilic GPI-anchor intercalates into As mentioned above, some proteins that could be removed by washing ejaculated human spermatozoa with moderately high salt concentrations were shown to have their origins in the epididymis. In an alternative approach, proteins secreted from epididymal epithelial cell cultures, and which were metabolically labelled with tritiated amino acids or [35S]methionine, were shown to relocate to co-cultured spermatozoa (Tezon et al., 1985; Ross et al., 1990; reviewed for other species also in Cooper, 1986; Moore, 1990). Precise information on the nature and synthesis of some of the human glycoproteins associated with the sperm surface and post-testicular sperm maturation have only emerged following the systematic molecular cloning of testicularly and epididymally expressed genes (e.g. PH-20: Lin et al., 1993; fertilin (PH-30): Gupta et al., 1996; Jury et al., 1997; ARP: Hayashi et al., 1996; HE2: Osterhoff et al., 1994; HE5: Kirchhoff, 1996). Although sperm surface glycoproteins have now been characterized from a number of species, there still appears to be a high degree of species diversity in expression. One example is the selenium-dependent glutathione peroxidase (GPX5) which was originally cloned from the mouse (Ghyselinck et al., 1991). This gene product is very variable between species and does not appear to occur in the human (Beiglböck et al., 1998; Hall et al., 1998). Here, we concentrate on those expressed in human tissues, and discuss other species only for comparative purposes. The most important of these glycoproteins are listed in Table I in order of their first appearance in the male tract. PH-20 PH-20 is a sperm glycoprotein of ~64 kDa in the human (Sabeur et al., 1997) and appears to be highly conserved across species. It was originally discovered in the guinea pig (Primakoff et al., 1988), as an integral sperm protein made in the testis, though it is anchored within the sperm plasmalemma and in the intra-acrosomal membrane (IAM) via a GPI-anchor. PH-20 has been shown to relocate during epididymal transit (Phelps and Myles, 1987). In the testis it is localized over the entire sperm surface, whereas it moves during passage through the epididymis to be localized in the posterior head region (hence PH). There is a further redistribution during the acrosome reaction when the PH-20 epitopes on the IAM also become exposed (Myles and Primakoff, 1984; Phelps et al., 1990). There appears to be some species specificity in regard to the precise domain localization on the sperm surface; in the mouse, monkey and human, PH-20 appears initially to be restricted to the whole head region. 306 S.Schröter et al. The human and cynomolgus monkey PH-20 cDNAs have been cloned, and share 90% identity with each other and 59% with the guinea pig PH-20 (Lin et al., 1993). In particular, all 12 cysteine residues are conserved. This general conservation suggests that in all species this protein has a similar important function. The N-terminal domain of the molecule contains a hyaluronidase activity in the species examined, and it is assumed that this enables the spermatozoa to penetrate the cumulus cell layer surrounding the oocyte. Additionally, proteolytic processing of the PH-20 molecule has been reported following the acrosome reaction and is clearly subsequent to its insertion into the sperm plasma membrane. The resulting fragments, however, remain associated via disulphide bridges. It has been speculated that this processing may be part of the secondary binding to the zona pellucida, though the mechanisms are still poorly understood (Myles and Primakoff, 1997). Fertilin (PH-30) Fertilin is an α–β heterodimer, which like PH-20, was also originally described from the guinea pig. Both subunits represent closely related transmembrane proteins, and appear to be involved in gamete fusion (Blobel and White, 1992; reviewed in Myles and Primakoff, 1997). Also like PH-20, fertilin appears to exhibit a species-specific domain localization on the sperm surface. Both primate and human fertilin cDNAs have been cloned from the testis. However, at least the α-gene appears to be non-functional in the human (Jury et al., 1997). Fertilin α and β were the first identified members of the ADAM (a disintegrin and metalloprotease) family. It is therefore possible that other members of this family may also be involved in gamete fusion, for example in the mouse (Cho et al., 1996). Another member of this family has been cloned from the efferent ducts and caput epididymides of the monkey, the so-called epididymal apical protein 1 (EAP-1; Perry et al., 1992). However, this has only been located on the apical membrane of the epididymal principal cells and not as yet on the sperm surface. A growing number of ADAMs have been found in the testis (Wolfsberg et al., 1995), suggesting a more broad-based function of these heterodimeric transmembrane glycoproteins. CD59 The 19 kDa CD59 glycoprotein appears to be present on the whole sperm surface of ejaculated, capacitated and also acrosome-reacted spermatozoa, as well as in tissue sections of the testis, as determined using immunohistochemistry (Fenichel et al., 1994). This GPI-anchored molecule is widely distributed on cell surfaces throughout the body, and was originally found on the membranes of human blood cells (erythrocytes and leukocytes), and also on endothelial cells and ductular epithelia (Meri et al., 1991). The CD59 cDNA was cloned from human lymphocytes by Sawada et al. (1990). CD59, also known as ‘protectin’, effectively inhibits lysis by the membrane attack complex of the homologous complement system (Lachmann, 1991; Meri et al., 1991; Walsh et al., 1991). Thus, the sperm CD59 glycocalyx antigen is also suggested to participate in the protection of male gametes against complement-mediated damage as they travel through the female genital tract. Furthermore, sperm CD59 is speculated to be involved in the gamete adhesion process during fertilization (Fenichel et al., 1994). Table I. Cloned human glycoproteins which bind sequentially to the plasma membrane (PM) of spermatozoa during their transit through the male reproductive tract Name of cloned glycoprotein Origin (site of first synthesis) Size in amino acids (aa) (+ signal peptide, SP) Mr of glycoprotein Type of glycosylation (consensus sites for N- and O-glycosylation Sperm domain localization Proposed functions/comments Reference PH-20 testis 455 aa (+35 SP + 19 GPI-anchor SP) 64 kDa 7 N-glycosylation sites >20 O-glycosyl. sites ejaculated and capacitated spermatozoa: entire head acrosome reacted: IAM→ 2 functions: hyaluronidase activity→cumulus penetration IAM-anchored PH-20→secondary binding to zona pellucida Lin et al. (1993) CD59 testis 103 aa (+25 SP + 22 GPI-anchor SP) 19 kDa 1 N-glycosylation site (verified) 9 O-glycosylation sites whole sperm surface protection from complement? Sawada et al. (1990) SOB1 testis 853 aa 100 kDa 5 N-glycosylation sites 4 O-glycosylation sites ejaculated spermatozoa: subequatorial involved in sperm–oocyte binding Lefèvre et al. (1998) ARP efferent ducts 230 aa (+19 SP) 30 kDa 1 N-glycosylation site verified by N-glycanase F (carbohydrate ≈4 kDa) postacrosomal on sperm head involved in fusion of gametes? functional counterpart of rodent AEG Hayashi et al. (1996) HE2 caput epididymidis 81 aa (+22 SP) 1 N-glycosylation site 6 O-glycosylation sites ejaculated spermatozoa: acrosomal & equatorial capacitated spermatozoa: mainly equatorial LRE-motif: cell adhesion? Osterhoff et al. (1994) HE5 corpus epididymidis 12 aa (+24 SP + 25 GPI-anchor SP) 1 N-glycosylation site (verified) 20–25 kDa whole sperm surface lesser over chromosome GPI-anchor: increase in PM stability? carbohydrates: immunosuppression? Kirchhoff et al. (1993) HE4 cauda epididymidis 95 aa (+30 SP) 1 N-glycosylation site 8 O-glycosylation sites epididymis: associated to sperm surface lost during capacitation protease inhibitor? Kirchhoff et al. (1991) IAM = intra-acrosomal membrane; AEG = acidic epididymal glycoprotein; LRE = Leu-Arg-Glu; GPI = glycosylphosphatidylinositol. The sperm glycocalyx SOB1 SOB1 is a glycoprotein of testicular origin (Boué et al., 1992; Lefèvre et al., 1998) that is localized on the subequatorial region of epididymal and ejaculated spermatozoa using the monoclonal antibody CA6 as probe (Boué et al., 1992, 1995). Western blotting of extracts of human, macaque and rodent spermatozoa revealed a size for the SOB1 protein of 100 kDa, agreeing well with the size predicted from the cloned cDNA (94.8 kDa). The small difference in size compared to Western blotting is likely to be due to glycosylation of the sperm antigen (Lefèvre et al., 1998). Northern blot analysis shows that the testis-specific expression of SOB1 is conserved among mouse, rat, rabbit, dog and cat. SOB1 reveals 32% homology with the mouse sperm fibrous sheath component (Fulcher et al., 1995). The SOB1-recognizing Fab fragments of CA6 were also shown to inhibit specifically human sperm binding to hamster and human oocytes in a dose-dependent manner, suggesting that the SOB1 glycoprotein is involved in oocyte binding (Boué et al., 1995). ARP To date, no functional counterpart to the guinea pig fertilin (PH-30) has been described for the human. However, a potential candidate could be the human glycoprotein ARP [acidic epididymal glycoprotein (AEG)-related protein], which has also been suggested to be involved in the sperm–egg interaction (Hayashi et al., 1996). ARP is localized on the posterior region of the sperm head, similar to the localization of the guinea pig PH-proteins. However, ARP is synthesized post-testicularly by the epithelium of the efferent ducts, and expression persists in the caput and corpus regions of the epididymis (Hayashi et al., 1996). Western blot analysis showed a protein of ~30 kDa (with a weak secondary band at 26 kDa) to be present throughout the epididymal lumen, in seminal plasma and on the surface of washed ejaculated spermatozoa. N-Glycosidase F treatment reduced the size of the protein by ~4 kDa and implied only a single sugar side-chain attached to Asn211 close to the C-terminus of the polypeptide. ARP is a novel member of the CRISP (cysteine-rich secretory protein) family (Haendler et al., 1993), which includes a number of gene products specific to the male genital tract (Kratzschmar et al., 1996). ARP is also related to another well-characterized secretory protein, the acidic epididymal glycoprotein (AEG) or protein DE (reviewed in Brooks, 1987) described in rodents. Although ARP is not the true homologue of the rodent AEG (Hayashi et al., 1996), it resembles this protein in as much as it also binds to the post-acrosomal region of the sperm head. However, since a true AEG counterpart has not yet been identified in the human, ARP may prove to be the functional homologue of the rodent molecule. In rats, AEG is considered to be involved in sperm–egg fusion: specific antibodies block fertilization in vivo and in vitro (Cuasnicu et al., 1984; Perez Martinez et al., 1995). Purified AEG binds to the egg surface and inhibits sperm–egg fusion in a dose-dependent manner 307 (Rochwerger et al., 1992). After in-vitro and in-vivo capacitation, AEG migrates from the post-acrosomal to the equatorial region of the sperm head, the region which first fuses with the oolemma (Rochwerger and Cuasnicu, 1992). HE2 HE2 is an abundant gene product from the proximal human epididymis, comprising 0.2% of the total epididymal mRNA pool, and was identified by tissue-specific differential screening of a human epididymal cDNA library (Kirchhoff et al., 1990; Osterhoff et al., 1994). Both the mRNA and protein can be detected in the principal cells of the proximal epididymal epithelium, and immunoreactive epitopes are also detected on fixed epididymal and ejaculated spermatozoa over the acrosome and equatorial segment. Following in-vitro capacitation, the immunoreactivity is restricted to the equatorial segment (C.Osterhoff, unpublished data). The HE2 mRNA encodes a small polypeptide which after removal of a signal peptide has 81 amino acids (size ~9 kDa), including a single N-glycosylation site and six possible sites for O-glycosylation. Preliminary Western blot analysis of sperm protein extracts, however, using antibodies raised against different epitopes, suggests a much larger size of ~37–38 kDa (C.Osterhoff, unpublished data). Whether this is due to a considerable glycosylation or to possible complexation with other molecules, or to dimerization, is still an open question. An interesting feature of the HE2 molecule is that it appears to be retained with relatively high affinity by ejaculated spermatozoa, and yet there are no obvious structural features which could account for this. There is, however, an LRE-motif (-Leu-Arg-Glu-) within the primary sequence that has been implicated in cell adhesion (Hunter et al., 1989; Kawaoka et al., 1990). This motif is located within a region of the HE2 molecule which shows a weak but significant homology to the influenza virus haemagglutinin sequence responsible for the attachment of the virus to cell receptors. It thus seems possible that HE2 may also be involved in sperm–egg interaction like fertilin, which has a related TDE motif and a similar localization on the sperm head. To date, HE2 has only been identified in the human and recently also in the chimpanzee (Pan troglodytes) epididymis (Young and Fröhlich, 1998). In the chimpanzee epididymis, the homologous HE2 mRNA expression seems to be androgendependent. Homologous molecules have not yet been found in other mammalian species, thus prohibiting detailed experimental studies on its possible function. CD52/HE5 When the HE5 gene transcript was first cloned from the human epididymis (Kirchhoff et al., 1993), it was shown to be identical in its peptide backbone to the lymphocyte antigen CD52 (Xia et al., 1991). This peptide backbone is unusual in being remarkably short (only 12 amino acids) and results from cleavage 308 S.Schröter et al. of a secretory signal peptide from the N-terminus and a GPI-anchor signal sequence from the C-terminus of the primary translation product. The peptide is additionally furnished with a substantial N-glycosylation attached to an asparagine residue. Preliminary studies show that the epididymal CD52 has a different glycosylation from that on lymphocytes (S.Schröter, unpublished data). Antibody studies showed this molecule to represent quantitatively the most important glycoprotein on the surface of ejaculated human spermatozoa. Although initial studies could find no rodent homologue for this molecule, studies at the genomic level showed that HE5 is indeed the genomic homologue of the main maturation-associated, sperm surface antigen in rats, SMemG (Moore et al., 1989; Eccleston et al., 1994), and in mice (Kubota et al., 1990; Kirchhoff, 1994). Non-coding regions and exon–intron structures are highly homologous at the genomic level. Nevertheless, comparisons of CD52 peptide sequences revealed that the post-translational peptide backbone is quite different between species, though all homologues possess a conserved putative N-glycosylation site, conserved displaced GPI-anchor signal peptides, and conserved displaced signal peptides. Because the CD52 antigen is attached to the sperm surface by intercalation of the GPI-anchor into the outer leaflet of the sperm plasma membrane, it can potentially cover the whole of the sperm surface. HE5/CD52 is expressed relatively late during epididymal transit, with maximal expression in the distal corpus and cauda regions, coinciding with the major increase in negative sperm surface charge alluded to above. Antibody studies confirm that it indeed covers the whole of the sperm surface, and that this does not appear to alter following in-vitro capacitation (Osterhoff et al., 1995; Yeung et al., 1997; S.Schröter, unpublished data). Molecular modelling of the lymphocyte CD52 molecule (Figure 2) shows that the glycosylation probably makes up ~60% of the molecule (Xia et al., 1991; Treumann et al., 1995). The sperm CD52, like the lymphocyte molecule, is also highly glycosylated (Schröter et al., 1997) and should similarly form a highly flexible structure on the sperm surface (Figure 2). It is easy to imagine how a molecule like this, being expressed late during sperm maturation and in large amounts, could effectively coat the sperm surface with a highly fluid glycosyl matrix, thus masking other aspects of the sperm surface which may have been exposed and/or modified earlier during post-testicular transit. A comparison of all the known CD52 molecules from five mammalian species (Figure 3) shows that, although the dimensions of the sperm glycocalyx may be altered, the principal features of the CD52 molecule, with a putative N-terminal N-glycosylation site upon a flexible GPI-anchored peptide backbone remain unchanged. The sequence comparisons of the homologous cDNAs have revealed a conserved consensus site for N-glycosylation in all examined species. This strongly suggests that the in-vivo product is in fact N-glycosylated in all mammals. In the case of the human sperm CD52 counterpart, the existence of an extensive N-glycosylation site has been convincingly demonstrated (Schröter et al., 1997). Also, Zeheb and Orr (1984) have found that the rat CD52 molecule binds to a ConA–agarose column, specific for N-glycosidic mannose-rich chains. Whereas the human sperm CD52 appears to possess no sites for O-glycosylation (Xia et al., 1993), the rat CD52 is probably highly O-glycosylated (Eccleston et al., 1994). For the CD52 homologues from other species, there is still no direct experimental information on the degree of O- or N-glycosylation. HE4 HE4 is an example of a protein with a relatively loose association with the sperm glycocalyx. The cloned cDNA sequence of this human epididymal protein predicts a small (95 amino acid) polypeptide with two whey-acidic-protein (WAP) domains represented by a duplicated pattern of four cysteine residues (Seemuller et al., 1986; Kirchhoff et al., 1991). It also includes a single N-glycosylation site. Antibodies raised against the biotechnologically produced protein confirm the in-situ hybridization analysis of the mRNA, i.e. that this molecule is produced by the epithelial cells in low to moderate amounts throughout the epididymis, though with highest levels in the distal corpus and cauda. The antibodies also show that HE4 is associated with freshly ejaculated spermatozoa, but during incubation in appropriate capacitation medium, the HE4 protein diffuses away from the gametes, indicating only a low affinity binding. This is particularly noticeable during in-vitro capacitation, when the HE4 immunoreactivity is rapidly lost from the sperm surface (Osterhoff et al., 1995). These properties are very similar to those expected of a decapacitation factor. However, the WAP-domain structure of HE4 also links it to a group of secretory protease inhibitors found in the female and the male reproductive tract in some species (Seemuller et al., 1986; Coronel and Lardy, 1992). While this is not evidence for protease inhibitor interactions, it does imply that HE4 may be involved in other specific protein–protein interactions at the sperm surface. HE4 appears to have homologues in the epididymides of most mammalian species, at either the immunohistochemical or RNA hybridization levels; cross-reactive products have been identified in the human, dog, pig, horse and bull (Kirchhoff et al., 1990, 1991; Uhlenbruck et al., 1993; Ellerbrock et al., 1994). Glycosyl-modifying enzymes In this review we do not focus on the enzymatic action of glycosyl-modifying enzymes, as there are excellent recent reviews elsewhere covering this topic in more detail (Benoff, 1997; Tulsiani et al., 1997). Tulsiani et al. (1998) summarize their results from in-vitro studies in the rat model on the role of epididymal luminal fluid glycosyltransferases (as synthetic enzymes, mainly sialyl- and fucosyltransferase) and glycosidases (as hydrolytic enzymes), together comprising >80% of total enzyme activities in the epididymal luminal fluid. They showed that after incubation of live caput The sperm glycocalyx Figure 2. Dynamic molecular modelling of one glycoform of the lymp examined on the basis of a Metropolis Monte Carlo simulation (court tion of the peptide backbone (green) with its N-glycosylation (light glycosylphosphatidylinositol (GPI)-anchor has been omitted in the c sible positions of the sialic acids which terminate the arms of the tetra are highly flexible and can only be interpreted as dynamic structures spermatozoa with [14C]fucose, the labelled fucose was incorporated into an endogenous 85 kDa sperm glycoprotein, thus providing an example for the action of a synthetic enzyme. In-vitro activity of a hydrolytic enzyme was shown for a purified β-D-galactosidase from epididymal luminal fluid acting on a 135–150 kDa antigen from live caput but not caudal spermatozoa as substrate. This antigen was suggested to be degalactosylated during the incubation with the pure enzyme, indirectly shown by the decrease in PNA binding. This lectin has a high affinity for the terminal galactosyl residues of Galβ(1–3)GalNAc. These in-vitro results imply a role for glycoprotein-modifying enzymes in the modification of rat sperm plasma membrane glycoproteins during epididymal maturation, though the in-vivo function of glycosyl-modifying enzymes remains to be demonstrated. A soluble galactosyltransferase has also been described from human seminal plasma (Ross et al., 1993). This highly active 45 kDa component was detected in the caput epididymal 309 Figure 3. Schematic representation of the glycosylphosphatidylinositol (GPI)-anchored HE5/CD52 antigen in comparison with its corresponding species homologues. The GPI-anchored molecules are inserted into the outer leaflet of the sperm plasma membrane bilayer on the sperm surface. The proportions of the various components have been drawn to scale, following the calculations of Rademacher et al. (1988) and Jentoft (1990). The symbols for N- and O-glycosylation sites are filled in if their existence has been experimentally proven (see text). The schematic drawing of the glycosylation is based on a typical complex type N-glycan. intraluminal fluid, and is therefore another potential candidate involved in the modification of the human sperm glycocalyx during epididymal maturation. Soluble glycosyl-modifying enzymes of the luminal fluid, acting within the male reproductive tract, may not be the only enzymes responsible for the changes of the sperm surface glyco-architecture. A number of intrinsic sperm surface components also appear to show enzymatic activity (reviewed by Tulsiani et al., 1997), or at least bind to glycosidic structures on the zona pellucida. These have been described for several mammalian species, including the mouse and human, and include an α-D-mannosidase, mannose-binding protein, and a selectin-like molecule. Since carbohydrate-binding molecules are among those proposed as receptors for the zona pellucida 310 S.Schröter et al. (Tulsiani et al., 1997), these glycosyl-modifying enzymes, whether soluble or sperm-bound, could encourage sperm maturation and enable sperm glycoconjugates to interact with their complementary molecules on the zona pellucida during fertilization. The function of the sperm glycocalyx N-Glycans form a prominent part of the sperm glycocalyx, and probably contribute greatly to its physicochemical properties. One of the most striking features of N-glycans is their extraordinary flexibility, with a very high degree of rotation and movement possible about each glycosidic bond. Additionally, there is usually considerable microheterogeneity; on the same backbone molecule N-glycans may vary markedly in chain length, branching or monosaccharide composition. Analysis of such a glycan structure in solution can be modelled by the application of a MMC (Metropolis Monte Carlo) simulation (Figure 2; courtesy of Kay Boettcher) to just one of the diverse carbohydrate structures of the lymphocyte CD52: a tetra-antennary polylactosamine-containing N-glycan with terminal sialic acids, as proposed by Treumann et al. (1995). The coloured dots show all possible positions for the terminal sialic acids, and clearly demonstrate the mobility and flexibility of this structure. Statistically, such a mobility would also enhance the probability of entering a conformation which could be recognized by lectin-like structures which, themselves being proteins, are much less flexible. Thus it is important to note that glycosylation is four-dimensional, rather than three-dimensional, with the added dimension (time) enhancing recognition by specific, receptor-type molecules. It has been suggested that specific oligosaccharide moieties of certain glycoproteins may be immunosuppressive (El Ouagari et al., 1995; Clark et al., 1996a,b), though mechanisms as to how this may work are far from clear. However, carbohydrates can also definitely be immunogenic, as for example, in the AB(H)O-system of blood factors, and certainly spermatozoa are able to raise a relatively high antibody titre, (as seen in vasectomized men). Important in this context is the fact that a T-cell-dependent immune response against conjugated carbohydrates is only possible against molecules that still possess a peptide moiety, since only this part of the molecule can be processed by antigen-presenting cells (Austyn, 1989). Equally, however, monoclonal antibodies raised against sperm surface membranes tend to be directed preferentially against carbohydrate moieties (Isojima et al., 1990; Kameda et al., 1991, 1992; Kurpisz and Alexander, 1995; Tsuji, 1995). Thus it is this part that appears to be ’seen’ by the antibodies. The regionalization of sperm maturation provides another possible clue to the role of the glycocalyx. It is evident that all such membrane modifications do not occur within the testis, nor simultaneously in any one region of the male tract. It is particularly noticeable that there is a geographic hierarchy in the modification of the sperm membrane, with most notably the major sperm surface antigen, CD52, being added last, following other epididymal interactions. Given the high degree of carbohydrate mobility discussed above, it seems logical to consider that the addition of this antigen is likely to occlude later interactions between luminal molecules and other proteins on the sperm surface. Such interactions are probably easier prior to the advent of CD52. This molecule and similar antigens may thus also serve to protect the sperm surface from non-specific pre-emptive interactions in the female tract, prior to specific surface changes occurring at fertilization. Another, less obvious, role could be in influencing the biophysical properties of the sperm membrane. Possibly more than most other cell types, the properties of the sperm outer membrane are very important with regard to all the functions and interactions of the male gamete, e.g. capacitation, sperm motility, sperm survival in the male and female tracts, the acrosome reaction, zona binding and penetration, and gamete fusion. This has been underscored by the numerous studies which involve the change in membrane fluidity and membrane cholesterol content during sperm maturation (Jones, 1989; Bearer and Friend, 1990), and most recently by the dramatic effects of free radicals and the oxidative status of the sperm membrane (reviewed in Aitken, 1995). In general, lipid oxidation and factors causing a reduction of membrane fluidity lead to a loss of sperm function. It therefore seems likely that molecules intercalating into the sperm membrane, such as glycolipids or GPI-anchored glycoproteins, could substantially influence local membrane properties, including ion flux across the membrane, and possibly also motility parameters. A further biophysical property of a dense hydrophilic covering of the sperm cells could be to prevent the spermatozoa from non-specific interactions with proteins—acting as a natural repellent. Similar non-adhesive glycocalyx-like surfaces were artificially constructed by Holland et al. (1998), who coated graphite with oligosaccharide surfactant polymers possessing dextran and alkanoyl side-chains. These polymers were shown to assemble on the graphite surface via their hydrophobic domains, while the dextran side-chains protruded into the aqueous phase and created a glycocalyx-like surface. This surface was effective in suppressing protein adsorption (from human plasma protein solution). Finally, there is a striking superficial resemblance between the glycocalyx of the sperm surface and that characterized on the surface of some successful protozoan parasites. Trypanosoma brucei also uses highly glycosylated GPI-anchored molecules (VSG, variant surface glycoprotein) to protect itself within the host bloodstream (Ferguson et al., 1994). This GPI-anchored glycocalyx allows a high packing density of glycosylated molecules, which by being variant offer protection against the host immune system and protect underlying membrane molecules which might offer a target for host resistance. Other GPI-anchored proteins, like the 35/60 kDa antigen from Trypanosoma cruzi, are involved in specific host–parasite interactions (and finally appear to be involved in infectivity, i.e. The sperm glycocalyx specific membrane–membrane interaction (Ferguson et al., 1994). A major difference, however, is the induced variability of trypanosome glycoproteins, which are able to switch structure once an immune reaction has occurred, thus making the parasites effectively invisible to the host immune system. Nevertheless, the resemblance of the sperm surface to that of the protozoan parasite in terms of structure (GPI-anchored molecules, dense sugar surface) and function (protection, binding specificity, hierarchical interactions with host cells) could be very informative about the mechanisms involved in glycocalyx function. Conclusion The spermatozoon has been likened to a spaceship carrying a very important genetic load across a vast distance of hostile space, to deliver it in perfect condition to its target planet. Like the outer hull of that spaceship, the sperm glycocalyx has to be protective against a variety of potentially inimical influences, and yet contain complex functions involved in informational exchange, motility control, docking and delivery of its contents. In the past few years we have begun to understand the shape and form of some of the components of the sperm glycocalyx, including for some molecules full primary structure. Only when the complete structures of most of the sperm surface components are known, however, will we begin to understand the mechanisms underlying these complex functions and their pathologies manifest as male infertility. Acknowledgements We should particularly like to thank Professor Freimut Leidenberger for continuous support and encouragement of our research into post-testicular sperm maturation. Very special thanks are due also to Dr Christiane Kirchhoff for very helpful discussion and advice, and to Kay Boettcher for the Metropolis Monte Carlo simulation. Much of our research reported here was supported by generous grants from the Deutsche Forschungsgemeinschaft (Iv7/4–6 and Ki317/5) and from the German Federal Ministry for Education, Research and Technology as part of a concerted project ’Fertilitätsstörungen’ (01Ky9103). References Aitken, R.J. (1995) Free radicals, lipid peroxidation and sperm function. Reprod. Fertil. Dev., 7, 659–668. Arenas, M.I., de Miguel, M.P., Bethencourt, F.R. et al. (1996) Lectin histochemistry in the human epididymis. J. Reprod. Fertil., 106, 313–320. Arenas, M.I., Madrid, J.F., Bethencourt, F.R. et al. (1998) Identification of N- and O-linked oligosaccharides in the human epididymis. J. Histochem. Cytochem., 46, 1185–1188. Aumuller, G., Renneberg, H., Schiemann, P.J. et al. (1997) The role of apocrine released proteins in the post-testicular regulation of human sperm function. Adv. Exp. Med. Biol., 424, 193–219. Austyn, J.M. (1989) Antigen processing and immunostimulation. In Male, D. (ed.), Antigen-Presenting Cells. IRL Press at Oxford University Press, Oxford, pp. 46–58. 311 Bains, H.K., Pabst, M.A. and Bawa, S.R. (1993) Changes in the lectin binding sites on the testicular, epididymal, vas, and ejaculated spermatozoon surface of dog. Andrologia, 25, 19–24. Bearer, E.L. and Friend, D.S. (1990) Morphology of mammalian sperm membranes during differentiation, maturation, and capacitation. J. Electron Microsc. Tech., 16, 281–297. Bedford, J.M. (1963) Changes in the electrophoretic properties of rabbit spermatozoa during passage through the epididymis. Nature, 200, 1178–1180. Beiglböck, A., Pera, I., Ellerbrock, K. and Kirchhoff, C. (1998) Dog epididymis-specific mRNA encoding secretory glutathione peroxidase-like protein. J. Reprod. Fertil., 112, 357–367. Benoff, S. (1997) Carbohydrates and fertilization: an overview. Mol. Hum. Reprod., 3, 599–637. Blobel, C.P. and White, J.M. (1992) Structure, function and evolutionary relationship of proteins containing a disintegrin domain. Curr. Opin. Cell Biol., 5, 760–765. Boué, F., Lassalle, B., Duquenne, C. et al. (1992) Human sperm proteins from testicular and epididymal origin that participate in fertilization: modulation of sperm binding to zona-free hamster oocytes, using monoclonal antibodies. Mol. Reprod. Dev., 33, 470–480. Boué, F., Duquenne, C., Lassalle, B. et al. (1995) FLB1, a human protein of epididymal origin that is involved in the sperm–oocyte recognition process. Biol. Reprod., 52, 267–278. Brooks, D.E. (1983) Influence of incubation conditions, tunicamycin and castration on incorporation of [3H]mannose and [3H]fucose into rat epididymal glycoproteins in vitro. J. Reprod. Fertil., 67, 97–105. Brooks, D.E. (1987) Androgen-regulated epididymal secretory proteins associated with post-testicular sperm development. Ann. N.Y. Acad. Sci., 513, 179–194. Brown, C.R., von Glos, K.I. and Jones, R. (1983) Changes in plasma membrane glycoproteins of rat spermatozoa during maturation in the epididymis. J. Cell Biol., 96, 256–264. Calvo, A., Pastor, L.M., Horn, R. and Pallares, J. (1995) Histochemical study of glycoconjugates in the epididymis of the hamster (Mesocricetus auratus). Histochem. J., 27, 670–680. Cho, C., Primakoff, P., White, J.M. and Myles, D.G. (1996) Chromosomal assignment of four testis-expressed mouse genes from a new family of transmembrane proteins (ADAMs) involved in cell–cell adhesion and fusion. Genomics, 34, 413–417. Clark, G.F., Oehninger, S., Patankar, M.S. et al. (1996a) A role for glycoconjugates in human development: the human feto-embryonic defence hypothesis. Hum. Reprod., 11, 467–473. Clark, G.F., Oehninger, S. and Seppälä, M. (1996b) Role for glycoconjugates in cellular communication in the human reproductive system. Mol. Hum. Reprod., 2, 513–517. Cooper, T.G. (1986) The Epididymis, Sperm Maturation and Fertilisation. Springer Verlag, Berlin. Coronel, C.E. and Lardy, H.A. (1992) Functional properties of caltrin proteins from seminal vesicle of the guinea pig. Mol. Reprod. Dev., 33, 74–80. Cuasnicu, P.S., Gonzalez Echeverria, F., Piazza, A.D. et al. (1984) Antibodies against epididymal glycoproteins block fertilizing ability in rat. J. Reprod. Fertil., 72, 467–471. Cummings, R.D. (1994) Use of lectins in analysis of glycoconjugates. Methods Enzymol., 230, 66–86. de Cerezo, J.M., Marquinez, A.C., Sarchi, M.I. and Cerezo, A.S. (1996) Fucosylated glycoconjugates of the human spermatozoon. Comparison of the domains of these glycoconjugates with the alpha-fucosyl binding sites, and with lactosaminic glycoconjugates and beta-D-galactosyl binding site domains. Biocell, 20, 11–20. Drisdel, R.C., Mack, S.R., Anderson, R.A. and Zaneveld, L.J. (1995) Purification and partial characterization of acrosome reaction inhibiting glycoprotein from human seminal plasma. Biol. Reprod., 53, 201–208. Eccleston, E.D., White, T.W., Howard, J.B., and Hamilton, D.W. (1994) Characterization of a cell surface glycoprotein associated with maturation of rat spermatozoa. Mol. Reprod. Dev., 37, 110–119. Eddy, E.M., Vernon, R.B. and Muller, C.H. (1985) Immunodissection of sperm surface modifications during epididymal maturation. Am. J. Anat., 174, 225–237. 312 S.Schröter et al. Ellerbrock, K., Pera, I., Hartung, S. and Ivell, R. (1994) Gene expression in the dog epididymis: a model for human epididymal function. Int. J. Androl., 17, 314–323. El Ouagari, K., Teissiers, J. and Benoist, H. (1995) Glycophorin A protects K562 cells from natural killer cell attack. J. Biol. Chem., 270, 26970–26975. Fenichel, P., Cervoni, F., Hofmann, P. et al. (1994) Expression of the complement regulatory protein CD59 on human spermatozoa: characterization and role in gametic interaction. Mol. Reprod. Dev., 38, 338–346. Ferguson, M.A., Brimacombe, J.S., Cottaz, S. et al. (1994) Glycosyl-phosphatidylinositol molecules of the parasite and the host. Parasitology, 108, S45–S54. Fierro, R., Foliguet, B., Grignon, G. et al. (1996) Lectin-binding sites on human sperm during acrosome reaction: modifications judged by electron microscopy/flow cytometry. Arch. Androl., 36, 187–196. Fulcher, K.D., Mori, C., Welch, J.E. et al. (1995) Characterization of Fsc1 cDNA for a mouse sperm fibrous sheath component. Biol. Reprod., 52, 41–49. Gabriel, L.K., Franken, D.R., van der Horst, G. and Kruger, T.F. (1994) Localization of wheat germ agglutinin lectin receptors on human sperm by fluorescence microscopy: utilization of different fixatives. Arch. Androl., 33, 77–85. Gahmberg, C.G. and Tolvanen, M. (1996) Why mammalian cell surface proteins are glycoproteins. Trends Biochem. Sci., 21, 308–311. Ghyselinck, N.B., Jimenez, C. and Dufaure, J.P. (1991) Sequence homology of androgen-regulated epididymal proteins with glutathione peroxidase in mice. J. Reprod. Fertil., 93, 461–466. Gupta, S.K., Alves, K., Palladino, L.O. et al. (1996) Molecular cloning of the human fertilin beta subunit. Biochem. Biophys. Res. Commun., 224, 318–326. Haas, G.G. Jr., DeBault, L.E., D’Cruz, O. and Shuey, R. (1988) The effect of fixatives and/or air-drying on the plasma and acrosomal membranes of human sperm. Fertil. Steril., 50, 487–492. Haendler, B., Kratzschmar, J., Theuring, F. and Schleuning, W.D. (1993) Transcripts for cysteine-rich secretory protein-1 (CRISP-1; DE/AEG) and the novel related CRISP-3 are expressed under androgen control in the mouse salivary gland. Endocrinology, 133, 192–198. Hall, L., Williams, K., Perry, A.C.F. et al. (1998) The majority of human glutathione peroxidase type 5 (GPX5) transcripts are incorrectly spliced: implications for the role of GPX5 in the male reproductive tract. Biochem. J., 333, 5–9. Hayashi, M., Fujimoto, S. and Takano, H. (1996) Characterization of a human glycoprotein with a potential role in sperm-egg fusion: cDNA cloning, immunohistochemical localization, and chromosomal assignment of the gene (AEGL1). Genomics, 32, 367–374. Holland, N.B., Yongxing, Q., Ruegsegger, M. and Marchant, R.E. (1998) Biomimetic engineering of non-adhesive glycocalyx-like surfaces using oligosaccharide surfactant polymers. Nature, 392, 799–801. Holt, W.V. (1980) Surface-bound sialic acid on ram and bull spermatozoa: deposition during epididymal transit and stability during washing. Biol. Reprod., 23, 847–857. Hunter, D.D., Porter, B.E., Bulock, J.W. et al. (1989) Primary sequence of a motor neuron-selective adhesive site in the synaptic basal lamina protein S-laminin. Cell, 59, 905–913. Isojima, S., Tsuji, Y. and Kameda, K. (1990) Coating antigens of human seminal plasma involved in fertilization. In Alexander, N., Griffin, D., Spieler, J.M. and Waites, G.M.H. (eds), Gamete Interaction: Prospects for Immunocontraception. Wiley-Liss, New York, pp. 155–174. Jentoft, N. (1990) Why are proteins O-glycosylated ? Trends Biochem. Sci., 15, 291–294. Jones, R.C. (1989) Membrane remodelling during sperm maturation in the epididymis. Oxford Rev. Reprod. Biol., 11, 285–337. Jury, J.A., Frayne, J. and Hall, L. (1997) The human fertilin alpha gene is non-functional: implications for its proposed role in fertilization. Biochem. J., 321, 577–581. Kameda, K., Takada, Y., Hasegawa, A. et al. (1991) Sperm immobilizing and fertilization-blocking monoclonal antibody 2C6 to human seminal plasma antigen and characterization of the antigen epitope corresponding to the monoclonal antibody. J. Reprod. Immunol., 20, 27–41. Kameda, K., Tsuji, Y., Koyama, K. and Isojima, S. (1992) Comparative studies of the antigens recognized by sperm-immobilizing monoclonal antibodies. Biol. Reprod., 46, 349–357. Kawaoka, Y., Yamnikova, S., Chambers, T.M. et al. (1990) Molecular characterization of a new hemagglutinin, subtype H14, of influenza A virus. Virology, 179, 759–767. Kirchhoff, C. (1994) A major messenger ribonucleic acid of the rodent epididymis encodes a small glycosylphosphatidylinositol-anchored lymphocyte surface antigen. Biol. Reprod., 50, 896–902. Kirchhoff, C. (1996) CD52 is the ‘major maturation-associated’ sperm membrane antigen. Mol. Hum. Reprod., 2, 9–17. Kirchhoff, C. and Hale, G. (1996) Cell-to-cell transfer of glycosylphosphatidylinositol-anchored membrane proteins during sperm maturation. Mol. Hum. Reprod., 2, 177–184. Kirchhoff, C., Osterhoff, C., Habben, I. and Ivell, R. (1990) Cloning and analysis of mRNAs expressed specifically in the human epididymis. Int. J. Androl., 13, 155–167. Kirchhoff, C., Habben, I., Ivell, R. and Krull, N. (1991) A major human epididymis-specific cDNA encodes a protein with sequence homology to extracellular proteinase inhibitors. Biol. Reprod., 45, 350–357. Kirchhoff, C., Krull, N., Pera, I. and Ivell, R. (1993) A major mRNA of the human epididymal principal cells, HE5, encodes the leucocyte differentiation CDw52 antigen peptide backbone. Mol. Reprod. Dev., 34, 8–15. Koehler, J.K. (1981) Lectins as probes of the spermatozoon surface. Arch. Androl., 6, 197–217. Kratzschmar, J., Haendler, B., Eberspaecher, U. et al. (1996) The human cysteine-rich secretory protein (CRISP) family. Primary structure and tissue distribution of CRISP-1, CRISP-2 and CRISP-3. Eur. J. Biochem., 236, 827–836. Kubota, H., Okazaki, H., Onuma, M. et al. (1990) Identification and gene cloning of a new phosphatidyl-linked antigen expressed on mature lymphocytes. J. Immunol., 145, 3924–3931. Kumar, G.P., Laloraya, M. and Agrawal, P. (1990) The involvement of surface sugars of mammalian spermatozoa in epididymal maturation and in-vitro sperm-zona recognition. Andrologia, 2, 184–194. Kurpisz, M. and Alexander, N.J. (1995) Carbohydrate moieties on sperm surface: physiological relevance. Fertil. Steril., 63, 158–165. Lachmann, P.J. (1991) The control of homologous lysis. Immunol. Today, 12, 312–315. Lassalle, B. and Testart, J. (1994) Human zona pellucida recognition associated with removal of sialic acid from human sperm surface. J. Reprod. Fertil., 101, 703–711. Lee, C.Y. and Wong, E. (1986) Developmental studies of sperm surface antigens using sperm-specific monoclonal antibodies. J. Reprod. Immunol., 9, 275–287. Lefèvre, A., Duquenne, C. and Finaz, C. (1998) SOB1, a human gene coding for a sperm protein with potential oocyte binding activity. 10th European Testis Workshop, Capri, 1998, Abstr. F1. Lin, Y., Kimmel, L.H., Myles, D.G. and Primakoff, P. (1993) Molecular cloning of the human and monkey sperm surface protein PH-20. Proc. Natl. Acad. Sci. USA, 90, 10071–10075. Margagee, S.F., Kunze, E. and Hammerstedt, R.H. (1988) Changes in lectin-binding features of ram sperm surfaces associated with epididymal maturation and ejaculation. Biol. Reprod., 38, 667–685. McArdle, W.L. (1991) Regulation of Epididymal Cell Function In Vitro and Characterisation of In Vivo Sperm Binding Proteins of Likely Epididymal Origin. PhD thesis, Hatfield University, UK. McConville, M.J. and Ferguson, M.A.J. (1993) The structure, biosynthesis and function of glycosylated phosphatidylinositols in the parasitic protozoa and higher eukaryotes. Biochem. J., 294, 305–324. Meri, S., Waldmann, H. and Lachmann, P.J. (1991) Distribution of protectin (CD59), a complement membrane attack inhibitor, in normal human tissues. Lab. Invest., 65, 532–537. Moore, D., White, T.W. and Ensrud, K.M. (1989) The major maturation glycoprotein on rat cauda epididymal sperm surface is linked to the The sperm glycocalyx membrane via phosphatidylinositol. Biochem. Biophys. Res. Commun., 160, 460–468. Moore, H.D.M. (1990) Development of sperm–egg recognition processes in mammals. J. Reprod. Fertil., Suppl. 42, 71–78. Moore, H.D.M. (1996) The influence of the epididymis on human and animal sperm maturation and storage. Hum. Reprod., 11, 103–110. Myles, D.G. and Primakoff, P. (1984) Localized surface antigens of guinea pig sperm migrate to new regions prior to fertilization. J. Cell Biol., 99, 1634–1641. Myles, D.G. and Primakoff, P. (1997) Why did the sperm cross the cumulus? To get to the oocyte. Functions of the sperm surface proteins PH-20 and fertilin in arriving at, and fusing with, the egg. Biol. Reprod., 56, 320–327. Naaby-Hansen, S., Flickinger, C.J. and Herr, J.C. (1997) Two-dimensional gel electrophoretic analysis of vectorially labeled surface proteins of human spermatozoa. Biol. Reprod., 56, 771–787. Navaneetham, D., Sivashanmugam, P. and Rajalakshmi, M. (1996) Changes in binding of lectins to epididymal, ejaculated, and capacitated spermatozoa of the rhesus monkey. Anat. Rec., 245, 500–508. Osterhoff, C., Kirchhoff, C., Krull, N. and Ivell, R. (1994) Molecular cloning and characterization of a novel human sperm antigen (HE2) specifically expressed in the proximal epididymis. Biol. Reprod., 50, 516–525. Osterhoff, C., Kirchhoff, C. and Ivell, R. (1995) Fate of two novel human sperm antigens HE4 and HE5 during in vitro capacitation. Colloque INSERM, 236, 393–394. Perez Martinez, S., Conesa, D. and Cuasnicu, P.S. (1995) Potential contraceptive use of epididymal proteins: evidence for the participation of specific antibodies against rat epididymal protein DE in male and female fertility inhibition. J. Reprod. Immunol., 29, 31–45. Perotti, M.E. and Pasini, M.E. (1995) Glycoconjugates of the surface of the spermatozoa of Drosophila melanogaster: a qualitative and quantitative study. J. Exp. Zool., 272, 311–318. Perry, A.C., Jones, R., Barker, P.J. and Hall, L. (1992) A mammalian epididymal protein with remarkable sequence similarity to snake venom haemorrhagic peptides. Biochem. J., 286, 671–675. Phelps, B.M. and Myles, D.G. (1987) The guinea pig sperm plasma membrane protein, PH-20, reaches the surface via two transport pathways and becomes localized to a domain after an initial uniform distribution. Dev. Biol., 123, 63–72. Phelps, B.M., Koppel, D.E., Primakoff, P. and Myles, D.G. (1990) Evidence that proteolysis of the surface is an initial step in the mechanism of formation of sperm cell surface domains. J. Cell Biol., 111, 1839–1847. Primakoff, P., Lathrop, W., Woolman, L. et al. (1988) Fully effective contraception in male and female guinea pigs immunized with the sperm protein PH-20. Nature, 335, 543–546. Rademacher, T.W., Parekh, R.B. and Dwek, R.A. (1988) Glycobiology. Annu. Rev. Biochem., 57, 785–838. Rochwerger, L. and Cuasnicu, P.S. (1992) Redistribution of a rat sperm epididymal glycoprotein after in vitro and in vivo capacitation. Mol. Reprod. Dev., 31, 34–41. Rochwerger, L., Cohen, D.J. and Cuasnicu, P.S. (1992) Mammalian sperm–egg fusion: the rat egg has complementary sites for a sperm protein that mediates gamete fusion. Dev. Biol., 153, 83–90. Ross, P., Kan, F.W., Antaki, P. et al. (1990) Protein synthesis and secretion in the human epididymis and immunoreactivity with sperm antibodies. Mol. Reprod. Dev., 26, 12–23. Ross, P., Vigneault, N., Provencher, S. et al. (1993) Partial characterization of galactosyltransferase in human seminal plasma and its distribution in the human epididymis. J. Reprod. Fertil., 98, 129–137. Runnebaum, I.B., Schill, W.B. and Töpfer Petersen, E. (1995) ConA-binding proteins of the sperm surface are conserved through evolution and in sperm maturation. Andrologia, 27, 81–90. Sabeur, K., Cherr, G.N., Yudin, A.I. et al. (1997) The PH-20 protein in human spermatozoa. J. Androl., 18, 151–158. Sawada, R., Ohashi, K., Anaguchi, H. et al. (1990) Isolation and expression of the full-length cDNA encoding CD59 antigen of human lymphocytes. DNA Cell Biol., 9, 213–220. 313 Schröter, S., Kirchhoff, C., Yeung, C.H. et al. (1997) Purification and structural analysis of sperm CD52, a GPI-anchored membrane protein. Adv. Exp. Med. Biol., 424, 233–234. Seemuller, U., Arnhold, M., Fritz, H. et al. (1986) The acid-stable proteinase inhibitor of human mucous secretions (HUSI-I, antileukoprotease). Complete amino acid sequence as revealed by protein and cDNA sequencing and structural homology to whey proteins and Red Sea turtle proteinase inhibitor. FEBS Lett., 199, 43–48. Tezon, J.G., Ramella, E., Cameo, M.S. et al. (1985) Immunochemical localization of secretory antigens in the human epididymis and their association with spermatozoa. Biol. Reprod., 32, 591–597. Töpfer-Petersen, E. (1999) Carbohydrate-based interactions on the route of a spermatozoon to fertilization. Hum. Reprod. Update, 5, 314–329. Treumann, A., Lifely, M.R., Schneider, P. and Ferguson, M.A. (1995) Primary structure of CD52. J. Biol. Chem., 270, 6088–6099. Tsuji, Y. (1995) Carbohydrate antigens recognized by anti-sperm antibodies. In Kurpisz, M. and Fernandez, F. (eds), Immunology of Human Reproduction, M. Bios Scientific Publishers, Oxford, UK, pp. 23–32. Tsuji, Y., Clausen, H., Nudelman, E. et al. (1988) Human sperm carbohydrate antigens defined by an antisperm monoclonal antibody derived from an infertile woman bearing antisperm antibodies in her serum. J. Exp. Med., 168, 343–356. Tulsiani, D.R.P., Orgebin-Christ, M.C. and Skudlarek, M.D. (1998) Role of luminal fluid glycosyltransferases and glycosidases in the modification of rat sperm plasma membrane glycoproteins during epididymal maturation. J. Reprod. Fertil., Suppl. 53, 85–97. Tulsiani, D.R.P., Yoshida-Komiya, H. and Araki, Y. (1997) Mammalian fertilization: a carbohydrate-mediated event. Biol. Reprod., 57, 487–494. Uhlenbruck, F., Sinowatz, F., Amselgruber, W. et al. (1993) Tissue-specific gene expression as an indicator of epididymis-specific functional status in the boar, bull and stallion. Int. J. Androl., 16, 53–61. Walsh, L.A., Tone, M. and Waldmann, H. (1991) Transfection of human CD59 complementary DNA into rat cells confers resistance to human complement. Eur. J. Immunol., 21, 847–850. Wolfsberg, T.G., Straight, P.D., Gerena, R.L. et al. (1995) ADAM, a widely distributed and developmentally regulated gene family encoding membrane proteins with a disintegrin and metalloprotease domain. Dev. Biol., 169, 378–383. Xia, M.Q., Tone, M., Packman, L. et al. (1991) Characterization of the CAMPATH-1 (CDw52) antigen: biochemical analysis and cDNA cloning reveal an unusually small peptide backbone. Eur. J. Immunol., 21, 1677–1684. Xia, M.Q., Hale, G., Lifely, M.R. et al. (1993) Structure of the CAMPATH-1 antigen, a glycosylphosphatidylinositol-anchored glycoprotein which is an exceptionally good target for complement lysis. Biochem. J., 293, 633–640. Yanagimachi, R. (1994) Mammalian Fertilization. In Knobil, E. and Neill, J.D. (eds), The Physiology of Reproduction. Raven Press, New York, pp. 189–317. Yanagimachi, R., Noda, Y.D., Fujimoto, M. and Nicolson, G.L. (1972) The distribution of negative surface charges on mammalian spermatozoa. Am. J. Anat., 135, 497–519. Yeung, C.H., Schröter, S., Wagenfeld, A. et al. (1997) Interaction of the human epididymal protein CD52 (HE5) with epididymal spermatozoa from men and cynomolgus monkeys. Mol. Reprod. Dev., 48, 267–275. Young, L.G. and Fröhlich, O. (1998) HE2/EP2, an androgen-dependent protein of the chimpanzee (Pan troglodytes) epididymis. J. Reprod. Fertil., Suppl. 53, 215–220. Zeheb, R. and Orr, G.A. (1984) Characterization of maturation-associated glycoprotein on the plasma membrane of rat caudal epididymal sperm. J. Biol. Chem., 259, 839–848. Zeng, F.Y. and Gabius, H.J. (1992) Sialic acid-binding proteins: characterization, biological function and application. Z. Naturforsch. Ser. C, 47, 641–53. Received on July 20, 1998; accepted on March 23, 1999