| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Regular Article |
a Department of Molecular and Cellular Biology, Baylor College of Medicine, Houston, Texas 77030
ABSTRACT
A lactosaminoglycan-associated antigen is associated with a carbohydrate moiety of all three zona pellucida (ZP) glycoproteins of pig and rabbit but is absent in the mouse and rat. A monoclonal antibody (PS1) recognizing this determinant was obtained by immunizing mice with a porcine ZP glycoprotein isoform purified by two-dimensional polyacrylamide gel electrophoresis. Conditions known to remove O-linked or sialic acid carbohydrate moieties (alkaline reduction; O-glycanase or neuraminidase enzymatic cleavage) did not remove the carbohydrate epitope. However, treatment with endo-ß-glycosidase, endoglycosidase F, or combinations of neuraminidase plus ß-galactosidase, totally removed the determinant, indicating that it is associated with a poly-N-acetyllactosaminoglycan structure present on an N-linked oligosaccharide. Molecular morphology studies using immunofluorescence and confocal microscopy techniques demonstrate that the PS1 antigen is localized at the surface of the ZP. Confirmation of this localization was obtained through studies that show that this antibody will inhibit homologous sperm binding to the pig ZP. Additional analyses using modular contrast microscopy and immunocytochemistry demonstrate that this carbohydrate-associated antigen is localized in discrete layers throughout the ZP matrix. These studies are the first to demonstrate the presence of a lactosaminoglycan type carbohydrate moiety in all three ZP proteins using a monoclonal antibody that appears to be involved in sperm recognition and structural organization.
fertilization, follicle, follicular development, reproductive immunology
INTRODUCTION
The zona pellucida (ZP) is the unique extracellular glycoprotein matrix that is formed between the oocyte and differentiating granulosa cells during oogenesis. The complete secretion and assembly of the three glycoproteins that constitute the ZP matrix takes a few days in rodents but may take several weeks or months in other mammalian species [1–3]. The ZP plays an important role in the fertilization process in that the sperm must bind to and penetrate it while leaving the matrix intact to protect the cleaving blastocyst and developing embryo until implantation [4, 5].
The ZP matrix of most species is comprised of three major glycoprotein families, ZP1, ZP2, and ZP3 [6, 7]. It is well established that ZP glycoproteins exhibit extreme heterogeneity both in charge and molecular weight primarily due to extensive post-translational modifications involving both N-linked and O-linked glycosylation [8–11]. Thus, the comparative nomenclature of ZP proteins among mammalian species based on electrophoretic mobility patterns has made it difficult to compare directly the function of the different proteins. With the recent isolation of cDNAs for ZP proteins of different species and in conjunction with antibodies against specific ZP proteins, it has now been possible to evaluate the genetic relationship of these proteins (Table 1) [12, 13]. Although the molecular sizes of individual ZP proteins vary, within each family there is considerable conservation of the number and position of cysteine residues as well as potential N-linked glycosylation sites [12, 13]. However, there are differences in these proteins between the mouse and nonrodent mammalian species relative to their site of synthesis and functional properties [3, 13].
|
The extracellular assembly of the ZP matrix is developmentally regulated, because the ZP is synthesized and deposited between the oocyte and granulosa cells during specific stages of oogenesis [1–3, 14]. However, there is considerable controversy regarding the site of biosynthesis of ZP proteins. In mice, ZP proteins have been reported to be expressed coordinately only in the oocytes [15, 16]. In contrast our studies have demonstrated the expression of ZP proteins in the granulosa cells of growing follicles of mouse ovary ([13]; unpublished communication). Expression of ZP mRNA and/or protein in the granulosa cells has also been demonstrated in other species such as the rabbit, pig, cow, marmoset, rhesus monkey, and human [2, 3, 14]. These studies suggest that both oocyte and certain granulosa cells contribute to the formation of ZP in mammalian ovaries.
Studies to determine the function of ZP proteins have largely concentrated on their sperm receptor activity. It is becoming increasingly apparent that the molecular mechanism of sperm-ZP interaction varies among mammalian species. The initial stages of sperm-ZP interaction referred to as the attachment of sperm followed by binding of sperm to the egg are thought to occur through interactions with carbohydrate moieties while the subsequent tight binding of sperm may be the result of sperm interaction with the carbohydrate and protein of the ZP [17, 18]. A number of ZP ligands associated with the ZP3 protein have been implicated in the initial sperm-ZP interaction in the mouse. These include: a class of O-linked oligosaccharides with terminal 
-galactose covalently linked to ZP3 [19, 20]; N-acetylglucosaminyl residues present on O-linked oligosaccharides of mZP3 that bind to galactosyltransferase present on mouse spermatozoa [21], and substrates for fucosyltransferase [22]. However, Nagdas et al. [23] have demonstrated the presence of trisaccharide sugar chain with terminal N-acetylglucosaminyl residue on O-linked oligosaccharide of mouse ZP3 but not terminal -galactose residues while lectin-binding studies have demonstrated the presence of -galactose [24, 25]. In addition N-linked polylactosaminyl glycans have been demonstrated in mouse ZP3 [23, 26, 27] that could contain various terminal sugar residues that may be involved in sperm binding.
In the pig, N-linked carbohydrates and not the O-linked carbohydrates present on ZP are involved in sperm-ZP interaction [28, 29]. In addition, it has been shown that tri-or tetra-antennary, neutral, complex-type N-linked carbohydrate chain(s) localized in the N-terminal region (Asn220) of pig ZP1 are involved in sperm binding [28, 30, 31]. More recently, Yurewicz et al. [32] have demonstrated that heterocomplexes of ZP1 and ZP3 proteins are involved in sperm-ZP interaction in the pig. It therefore appears that carbohydrate moieties of the ZP glycoproteins may be involved in some aspects of species-specificity in sperm-ZP interaction, and that, multiple molecular domains (involving carbohydrates and/or protein backbone) on a single ZP protein and/or multiple ZP proteins may interact with sperm membrane proteins as has been shown in the pig and rabbit [32–34].
Because monoclonal antibodies recognize distinct antigenic determinants, they have been used as tools to provide specific details of the structures as well as to define the functional domain(s) of molecules. In line with this, studies to determine the sperm receptor function of ZP proteins have utilized monoclonal antibodies specific to ZP proteins [35–39]. We have, however, generated a unique monoclonal antibody [40] that is specific to a carbohydrate moiety which is associated with the acidic isoforms of all three ZP glycoproteins of some, but not all, mammalian species. In the present study we have further characterized this carbohydrate antigen and have used the monoclonal antibody to study the molecular morphology and function of the carbohydrate moieties of ZP glycoproteins.
MATERIALS AND METHODS
Isolation and Characterization of ZP Glycoproteins
Pig ovaries were obtained frozen from slaughterhouses, and rabbit ovaries were obtained from Pel Freez Biological (Rogers, AR). Large-scale isolation of intact ZP and solubilization of ZP proteins were carried out as previously described [41]. One-dimensional polyacrylamide gel electrophoresis (1D-PAGE) and high-resolution two-dimensional polyacrylamide gel electrophoresis (2D-PAGE) was carried out as previously described [6, 42]. Electrophoretic transfer of ZP glycoproteins to nitrocellulose membranes and immunoblot analysis was accomplished as described by Timmons et al. [40].
Antibody Preparation
Pig ZP was heat solubilized (HSPZ) as previously described [43]. Rabbits were immunized with HSPZ to prepare polyclonal antisera against the major ZP glycoproteins according to the method described by Wood et al. [41].
The monoclonal antibody (PS1) was made against a single isoform spot from a silver-stained 2D-PAGE gel as previously described [40, 44]. Briefly, heat-solubilized pig ZP glycoproteins were separated using 2D-PAGE, and proteins were stained with the color-based silver stain [42]. A protein spot corresponding to a single charge species of the major pig ZP glycoprotein (see Fig. 1B) was cut from the acrylamide gel and homogenized in phosphate-buffered saline before being emulsified in complete Freund adjuvant and used to immunize mice. More recent studies demonstrate that the isoform family associated with the major low molecular weight porcine ZP contains both ZP1 and ZP3 proteins based on the mouse ZP nomenclature (Table 1). Spleen cell fusions from immunized mice were carried out with P3U1 mouse myeloma cells [45]. Screening for positive hybridoma growth was performed using ELISA. Male Balb/c mice were injected intraperitoneally with 5 x 106 hybridoma cells, and ascites fluid was collected 10–14 days later. The IgG fraction of the ascites fluid was isolated using a DEAE-Sephacel column eluted with a NaCl gradient. Antibody purity was evaluated by immunoblot analysis on 1D- and 2D-PAGE of ZP proteins.
|
Chemical and Enzymatic Deglycosylation of ZP Glycoproteins
The description of specificities, the commercial sources for the chemicals and enzymes, and the specific experimental conditions used to study deglycosylated ZP glycoproteins are summarized in Table 2. The deglycosylated products were analyzed by 1D- and 2D-PAGE followed by silver staining and/or immunoblotting to identify specific ZP glycoproteins.
|
Sperm-ZP Interaction Studies
The interaction of ejaculated boar spermatozoa with intact pig ZP was observed using the methods previously described [46]. Briefly, boar sperm were washed twice with minimal essential medium (Gibco Labs, Grand Island, NY). A 50-µl suspension containing 1 x 105 sperm was added to each sample of intact pig ZP, either untreated or incubated overnight (4°C) with antisera, ascites fluid, or purified IgG, and washed with media. After a 2-h incubation in a 3% CO2 incubator, supernumerary spermatozoa were removed by micropipetting using an Eppendorf pipettor (200 µl) through three changes of media in 500-µl spot plates. Sperm adherence to the ZP was monitored microscopically and the number of sperm bound to ZP determined. The average number (±SD) was determined for each sample group (10–20 ZP) and % inhibition was determined relative to the number of sperm bound in the control samples (P3U1 IgG). Statistical analysis was performed between treatment groups and the control group using Student t-test to determine the statistical significance of sperm binding.
Immunohistochemical Localization of PS1 Monoclonal Antibody
Immunohistochemical analysis was performed as previously described [47]. Briefly, portions of fresh ovaries from sexually immature and mature rabbits were immersed in OCT embedding compound (Miles Scientific, Elkhart, IN) and frozen in a dry ice-ethanol bath. Four-micron-thick cryostat sections were cut at -20°C and stored on acid-alcohol-washed slides at -70°C until needed. The sections were fixed for 10 min in methanol and rehydrated in PBS for 10 min. Alternatively, sections were fixed with 2% formaldehyde in buffered PBS. Sections were imcubated with PS1 ascites fluid (1:100 dilution) followed by 1:50 dilution of peroxidase-conjugated rabbit anti-mouse antiserum (Miles, Kankakee, IL). The signal was developed using the chromophore 3-3'-diaminobenzidine tetrahydrochloride (DAB). The slides were then mounted in an aqueous mounting solution (Aqua-PolyMount; Polysciences, Warrington, PA).
Confocal Microscopy
Rabbit oocytes were isolated from large ovarian follicles and were incubated with mouse PS1 antibody (1:100 dilution) along with guinea pig anti-rabbit ZP1 antibody (1:200 dilution; formerly referred to as guinea pig anti-rabbit 55 kDa [48]). Following washing, eggs were incubated in Texas red-labeled (Zymed, San Francisco, CA) rabbit anti-mouse IgG (1:200 dilution) and fluorescein-labeled (Zymed) goat anti-guinea pig antibody (1:400 dilution). Oocytes were analyzed using laser scanning confocal microscopy (Molecular Dynamics) to evaluate the molecular morphology of the PS1 carbohydrate determinant as it is related to the localization of the ZP1 protein.
RESULTS
Characterization of Carbohydrate Determinant by 2D-PAGE Immunoblot Analysis
To determine which ZP glycoproteins contained the antigenic determinant recognized by the PS1 monoclonal antibody, pig ZP glycoproteins were separated by 2D-PAGE and electrophoretically transferred to nitrocellulose, probed with PS1 antibody, and visualized with 125I-protein A. Although all three major pig ZP glycoprotein families are recognized by the monoclonal antibody PS1, only the most acidic isoforms of each protein are recognized (Fig. 1A). To confirm that the more basic isoforms of the major glycoprotein (ZP1 and ZP3) were, in fact, transferred but not recognized by the antibody, the identical blot was reprobed with a polyclonal antibody, rabbit anti-heat-solubilized pig ZP (RHSPZ) that recognizes all isoforms of the three major pig ZP proteins (Fig. 1B). These acidic protein species appear to have undergone the most extensive post-translational modification, as indicated by their increase in acidic charge and increase in molecular weight heterogeneity as analyzed by 2D-PAGE.
Characterization of the Carbohydrate Determinant by Chemical Treatment of ZP Glycoproteins
Following chemical deglycosylation of pig ZP glycoproteins using trifluoromethane sulfonic acid (TFMS) (which removes both N- and O-linked oligosaccharides from glycoproteins, leaving the N-linked sugar N-acetylglucosamine attached to the asparagine residue in the protein chain), the determinant is no longer detected by the PS1 antibody by immunoblot analysis (Table 3). Alkaline reduction of these glycoproteins with NaOH (which removes O-linked oligosaccharides) left the PS1 determinant intact (Table 3), even though there was an apparent decrease in relative molecular weight of the treated glycoproteins, indicating the removal of some carbohydrate.
|
Characterization of the Carbohydrate Determinant by Enzymatic Deglycosylation
Endoglycosidase treatment
Treatment of ZP glycoproteins with endo-ß-galactosidase hydrolyzes (ß-1-4) galactosidic bonds and therefore hydrolyzes unbranched poly-N-acetyllactosaminoglycans, but leaves the high-mannose core of N-linked oligosaccharides and most O-linked oligosaccharides intact. However, the enzyme removes a sufficient number of charged residues to significantly reduce the charge and molecular weight heterogeneity of both pig and rabbit ZP glycoproteins, allowing better separation of individual proteins by 1D-PAGE (Fig. 2). Although the untreated intact rabbit and pig ZP glycoproteins react with the PS1 monoclonal antibody (Fig. 2A, lanes 1 and 4), the enzyme-treated samples were not recognized (lanes 2 and 3), suggesting that the antigenic determinant recognized by PS1 is not present in the core oligosaccharide or the protein-carbohydrate linkage. As a control, the immunoblot was reprobed with polyclonal antisera (RHSPZ) that recognized all three of the major glycoproteins of the pig and rabbit ZP (Fig. 2B). This demonstrates that the ZP proteins had been transferred to the nitrocellulose and proves that lack of immunoreactivity was not due to lack of protein transfer to the nitrocellulose. Digestion of HSRZ or HSPZ with the enzyme O-glycanase that hydrolyses Gal ß-1-3GalNAc linkage to O-linked oligosaccharides caused a reduction in apparent molecular weight on 1D-PAGE, but the antigenic determinant remained intact (Table 3).
|
To investigate further the nature of the carbohydrate antigen recognized by PS1 monoclonal antibody, pig ZP was treated with endoglycosidase F that acts on N-linked oligosaccharides and cleaves high mannose, hybrid, and biantennary complex structures. The products were separated by 1D- and 2D-PAGE, electrophoretically transferred to nitrocellulose, and the blot was probed with PS1. The results show that, endoglycosidase F also destroyed the PS1 determinant (Fig. 3, A and B). Again, the immunoblot was reprobed with the polyclonal antibody to demonstrate protein transfer (Fig. 3, C and D).
|
Exoglycosidase treatment
The effects of partial deglycosylation with three exoglycosidases on the PS1 determinant in pig ZP glycoproteins (HSPZ) were analyzed (Fig. 4A). Neuraminidase (removes sialic acid residues) treatment alone (lane 2) had no effect on antibody binding, while treatment with ß-galactosidase (lane 3) or ß-N-acetyl-glucosaminidase (lane 4) alone appeared to decrease the binding of PS1 to HSPZ. However, after incubation of the ZP glycoproteins with a combination of neuraminidase with ß-galactosidase (lane 5), or ß-N-acetylglucosaminidase with ß-galactosidase (lane 6) or neuraminidase with ß-galactosidase and ß-N-acetylglucosaminidase (lane 7), PS1 recognition of ZP glycoproteins was no longer detectable. Figure 4B illustrates the transfer of total ZP protein, as determined by detection with the polyclonal antibody RHSPZ. Table 3 summarizes the results from immunoblot analysis of deglycosylated ZP glycoproteins using the PS1 monoclonal antibody that reveals that the PS1 determinant is associated with poly-N-acetyllactosaminyl residues of N-linked oligosaccharides.
|
Inhibition of Sperm Binding to the ZP by PS1 Monoclonal Antibody
Table 4 illustrates the results of a series of assays measuring the ability of intact pig ZP to bind homologous spermatozoa. If incubations were carried out using control media, or P3U1 myeloma ascites fluid, approximately 50 sperm bound per oocyte/ZP. Preincubation of intact ZP with ascites fluid containing a monoclonal antibody R5 that recognizes a protein determinant (ZP3) in mammalian ZP [44] does not inhibit sperm binding. Preincubation of the ZP with rabbit anti-HSPZ IgG that recognizes all three major glycoproteins inhibits sperm binding, even at 0.25 mg/ml (100% inhibition). Preincubation of intact ZP with PS1 ascites fluid, or PS1 IgG at 1.0 mg/ml, also inhibits sperm binding (98% inhibition), and a concentration of 0.025 mg/ml is sufficient to significantly inhibit sperm binding (71%).
|
Structural Localization of PS1 Antigen
When viewed with a light microscope using modulation contrast optics, pig and rabbit ZP appear to be smooth, amorphous structures as illustrated in Figure 5A. After treatment with either P3U1 control ascites fluid or other monoclonal antibody (R5) that recognizes ZP3 protein determinant, the morphology is unchanged (data not shown). However, after treatment with PS1, the ZP matrix exhibits distinct concentric rings surrounding the oocyte (Fig. 5B). The appearance of these rings suggests that there is a specific structural organization of the PS1 determinant within the ZP matrix. The number of rings appears to vary among individual ZP and among species. These rings are also evident in the ZP of large follicles of rabbit ovaries that are frozen and sectioned using a cryostat (Fig. 5C). As controls, sections of mouse and rat ovaries have been used to evaluate the presence of this antigen in other species. Regardless of the fixative used, this antibody did not detect the antigen in the oocyte or in the ZP of either of these rodent ovaries (data not shown).
|
Confocal microscopy further demonstrates that the PS1 carbohydrate epitope is accessible to the antibody at the ZP surface as are the ZP1 peptide epitopes (Fig. 6; F') and is therefore likely to be involved in the initial sperm-ZP interaction. The observations that sperm are inhibited from binding to the ZP by this antibody provide additional evidence that this epitope is located at the surface of the ZP.
|
DISCUSSION
Identification of a Lactosaminoglycan Moiety in the ZP
We describe the first studies to localize a carbohydrate antigen determinant associated with poly-N-acetyllactosaminoglycans present in multiple glycoproteins of the ZP of the rabbit and pig. In order to characterize the epitope recognized by the PS1 monoclonal antibody used in these studies, initial characterization of the ZP carbohydrate moiety was accomplished with selective chemical and enzymatic digestion of the oligosaccharides. The data presented here indicate that the antigenic determinant for the PS1 monoclonal antibody resides on an N-linked, and not an O-linked, oligosaccharide. These conclusions are drawn from the observations that TFMS and endoglycosidase F will remove the determinant from ZP glycoproteins, while alkaline reduction and O-glycanase have no effect. Further evidence suggests that the PS1 determinant is contained in a poly-N-acetyllactosaminoglycan that is part of the carbohydrate antigen, as prolonged incubation of the ZP glycoproteins with endo-ß-galactosidase removes the determinant while decreasing the charge and molecular-weight heterogeneity of the glycoprotein. Similar results are obtained with various exoglycosidase treatment of ZP protein. The marked decrease in molecular weight of ZP proteins would suggest that multiple monosaccharide residues are removed from the ZP by enzyme treatments.
Data from 1D-PAGE analysis of ZP glycoprotein after neuraminidase treatment show that removal of NANA (sialic acid) residues susceptible to enzymatic cleavage does not affect PS1 binding (Fig. 4, lane 2). Analysis of the treated glycoproteins by 2D-PAGE indicates the loss of NANA by a decrease in the negative charge on the glycoproteins with no detectable decrease in their molecular weight (data not shown). Treatment with either ß-galactosidase or N-acetylglucosaminidase (also has some ß-N-acetylgalacosaminidase activity) alone (Fig. 4, lanes 3 and 4) can remove sufficient quantities of the terminal carbohydrate chain to alter the charge and molecular weights of the ZP glycoproteins, as seen in the 2D-PAGE gel patterns (data not shown) but does not remove the PS1 determinant. However, a combination of neuraminidase and ß-galactosidase is sufficient to remove the PS1 determinant (Fig. 4, lane 6), suggesting that it is most likely associated with a poly-N-acetyllactosaminoglycan branch of the N-linked oligosaccharide. The presence of N-linked polylactosaminoglycan residues has been shown in the pig [49, 50] and also in the mouse [23, 26, 27]. One possible explanation for the removal of PS1 epitope by a combination of neuraminidase and ß-galactosidase treatment and not by ß-galctosidase alone is that sialic acid residues may be masking the sites of ß-galactosidase are causing stearic hindrance, thus preventing the action of ß-galactosidase. With the treatment of pig ZP first with neuraminidase, sialic acid residues are removed upon which ß-galactosidase is able to act resulting in the loss of the epitope that is recognized by the monoclonal antibody.
Although the carbohydrate content of the rabbit ZP glycoproteins has not been reported, that of the total pig ZP and of isolated pig ZP glycoproteins has been determined [6, 50, 51]. These oligosaccharide chains contain nearly equimolar amounts of galactose and N-acetylglucosamine, and lesser amounts of mannose, fucose, sialic acid, and N-acetylgalactosamine. More recently, both O-linked and N-linked carbohydrate structures of the porcine ZP have been outlined in detail [28, 51–53].
Molecular Morphology of the ZP Using the PS1 Antibody
Using a monoclonal antibody specific to carbohydrate moiety, the present studies demonstrate defined molecular layers within the well-organized extracellular matrix of the ZP (Figs. 5 and 6). The results suggest that the carbohydrate moiety may be involved in the organization of this unique matrix. The time frame for in vivo synthesis of ZP proteins and complete assembly of the matrix during the development of the ovarian follicle is not known for most species. This process likely takes place over a period of weeks or months in most nonrodent species, including the human [54, 55]. It is estimated that this process takes days to weeks in the rabbit and involves both the oocyte and the ovarian follicular cells [2, 10, 14].
Of interest, however, is that this antibody does not recognize a similar epitope in the ZP of the mouse (data not shown). It is well known that rodent ZP have physicochemical properties distinct from other mammalian species and that dissolution of the ZP matrix is much more susceptible to variation in pH and proteolysis than are the ZP of other mammalian species including the rabbit and pig [8, 14].
The appearance of these layers or rings suggests that the PS1 moiety may be responsible for the specific structural organization of the ZP matrix. Additional preliminary immunohistochemical localization studies also demonstrate some unique features of the appearance and localization of this carbohydrate antigen. When the ovarian tissue is fixed with formaldehyde that cross-links proteins, the antigen is observed in oocytes at all stages of growth as well as in the mature ZP matrix. However, if the tissue is fixed with methanol, the antigen appears to be extracted from the oocyte cytoplasm, although it remains associated with the mature ZP matrix (data not shown). Because methanol fixation would likely remove lipids during fixation, these studies suggest that the antigen may be associated with an intracellular lipid. Because the antigen is not extractable following assembly of the mature extracellular matrix, it is apparent that it may be associated with the ZP glycoproteins in the matrix during follicular development. These observations are consistent with the idea that sulfated glycolipids are important in extracellular matrix formation and interactions [56, 57].
Inhibition of In Vitro Sperm Binding with PS1 Antibody
The present studies are also the first to identify a monoclonal antibody specific to a carbohydrate determinant that inhibits primary sperm-egg interaction. Monoclonal antibody directed against an oligosaccharide sequence (GalNAcß1 
4Galß1 4) that is present in the inner region of the mouse zona has been reported to inhibit secondary binding and subsequent penetration of the sperm but not the initial binding of the sperm to the ZP [58, 59]. Using this monoclonal antibody, the carbohydrate sequence has been shown to be present on mouse ZP2 and ZP3 [59]. In the present study, the presence of the PS1 carbohydrate determinant on all three ZP proteins of pig and rabbit but not in mouse ZP suggests species difference in glycosylation and also suggests that the PS1 determinant is involved in both sperm recognition and structural organization of the ZP. However, it has not been possible to evaluate the binding of acrosome-intact sperm with all three ZP proteins containing the PS1 epitope because such studies will have to be carry out using purified native glycosylated proteins. Because of the extensive molecular weight heterogeneity, it is not possible to purify native glycosylated ZP proteins from any species containing the PS1 epitope (pig, rabbit, or human) [40, 47], unless they are deglycosylated first [50]. Studies on sperm binding in porcine model have used deglycosylated ZP1 and ZP3 native proteins [32, 33, 60] or nonglycosylated recombinant ZP2 protein expressed in bacteria [61], both of which do not contain the PS1 epitope. In these studies, ZP1 and ZP2 have been demonstrated to be involved in the primary binding while recombinant ZP2 has been shown to bind to acrosome reacted sperm. Although these studies indicate that PS1 epitope may not be necessary for sperm binding, it is more likely that there are multiple levels of interaction between the sperm and egg (carbohydrates and/or proteins) and that this epitope may be involved in one such level of interaction.
There are several reports in the literature of monoclonal antibodies against ZP components that interfere with primary sperm binding [35–39, 62]. However, all these antibodies are directed against the protein component of the ZP. The PS1 monoclonal antibody used in the present study is specific to a carbohydrate determinant associated with N-linked oligosaccharides of all three ZP protein families, suggesting that N-linked oligosaccharides of pig ZP are involved in sperm-egg interaction. This is in agreement with the reported observations that it is the N-linked oligosaccharides and not O-linked oligosaccharides of pig ZP that are involved in sperm binding [28–31]. Recently, it has been demonstrated that N-linked oligosaccharides of Xenopus egg envelope ZPC (ZP3) are involved in sperm binding [63]. In the present study, the N-linked carbohydrate determinant recognized by the PS1 monoclonal antibody contains poly-N-acetyllactosaminoglycan, and it appears that these residues may be involved in the initial interaction with the ZP. This is in contrast to that reported by Sacco et al. [60] and Yonezawa et al. [29], that polylactosamine residues are not involved in sperm binding, although it appears there is a decrease in sperm binding after endo-ß glycosidase treatment of ZP3. In addition, Yonezawa et al. have suggested that removal of N-acetylpolylactosamine from O-linked carbohydrate chains may destroy the sperm receptor activity of O-glycans while the role of polylactosamine residues on N-linked neutral carbohydrate chains is not clear [25]. It is also possible that different terminal residues on N-linked polylactosaminoglycan may be involved in sperm-ZP interaction as has been suggested for mouse ZP3 [23]. Oligosaccharides containing polylactosaminoglycan residues with varied terminal residues have been demonstrated in several glycoproteins [64, 65].
While it is possible that the PS1 monoclonal antibody could inhibit sperm binding to the ZP by stearic hindrance, preincubation of the ZP with other monoclonal antibodies such as R5 [44] (which binds to the protein backbone of the ZP3 glycoprotein of the porcine and rabbit [40]), does not interfere with sperm binding (data not shown). Therefore, the mechanism for PS1 inhibition of sperm binding is likely due to interference with ZP carbohydrate-sperm interaction. Regardless of the role of this carbohydrate epitope in sperm-ZP interaction, these studies demonstrate that this carbohydrate determinant is located at the surface of the ZP as well as within discrete layers throughout the ZP.
In summary, these studies demonstrate that an N-linked oligosaccharide containing a lactosaminoglycan moiety that is recognized by a monoclonal antibody is associated with each of the three major pig ZP glycoproteins. In addition to the use of this antibody in the analysis of the structural architecture of the ZP, this antibody should be important in further studies to evaluate the role of carbohydrates in sperm-ZP interaction, as well as to follow the formation of this unique extracellular matrix (ZP) during ovarian follicular development.
ACKNOWLEDGMENTS
The authors thank Drs. Don Marcus, Daniel Carson, and Cindy Farach for their many helpful conversations and constructive comments during preparation of this manuscript. We thank Dr. Michael Mancini for generating computer images from confocal microscopy. The authors acknowledge the technical expertise of Ms. Hitomi Kimura and Ms. Claire Lo.
FOOTNOTES
First decision: 21 March 2001.
1 This work was supported by grants to B.S.D. from the National Institutes of Health (HD-12587) and the Mellon Foundation. 
2 Correspondence: Bonnie S. Dunbar, Department of Molecular and Cellular
Biology, Baylor College of Medicine, One Baylor Plaza, Houston,
TX 77030. FAX: 713 798 7341; bdunbar{at}bcm.tmc.edu
Accepted: May 1, 2001.
Received: February 27, 2001.
REFERENCES
glycoprotein mediates binding of the biotin-labeled Mr 55,000 family (ZP3) to boar sperm membrane vesicles. Mol Reprod Dev 1993; 36:382-389[CrossRef][Medline] 4Galß1 4 on the mouse zona pellucida. J Reprod Immunol 1992; 21:241-256[CrossRef][Medline]-glycoprotein component of the Mr = 55 000 family. Biol Reprod 1989; 41:523-532[Abstract]This article has been cited by other articles:
|
|
 |
|
  P. C.N. Chiu, B. S.T. Wong, C.L. Lee, R. T.K. Pang, K.-F. Lee, S.B. Sumitro, S.K. Gupta, and W. S.B. Yeung Native human zona pellucida glycoproteins: purification and binding properties Hum. Reprod., June 1, 2008; 23(6): 1385 - 1393. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Shen, T. Stalf, C. Mehnert, U. Eichenlaub-Ritter, and H.-R. Tinneberg High magnitude of light retardation by the zona pellucida is associated with conception cycles Hum. Reprod., June 1, 2005; 20(6): 1596 - 1606. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Bogner, K.-D. Hinsch, P. Nayudu, L. Konrad, C. Cassara, and E. Hinsch Localization and synthesis of zona pellucida proteins in the marmoset monkey (Callithrix jacchus) ovary Mol. Hum. Reprod., July 1, 2004; 10(7): 481 - 488. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
