Skip to main content
NCBI home page
Proceedings of the National Academy of Sciences of the United States of America logoLink to Proceedings of the National Academy of Sciences of the United States of America
. 2002 Mar 5;99(5):2854–2859. doi: 10.1073/pnas.052707599

Binding of rainbow trout sperm to egg is mediated by strong carbohydrate-to-carbohydrate interaction between (KDN)GM3 (deaminated neuraminyl ganglioside) and Gg3-like epitope

Song Yu *,†,‡, Naoya Kojima *,†,§, Sen-itiroh Hakomori *,†,, Shigeharu Kudo ‖, Sadako Inoue **, Yasuo Inoue **,
PMCID: PMC122437  PMID: 11880637

Abstract

KDNα2→3Galβ4Glcβ1Cer [(KDN)GM3] is a major (≈90%) component of total gangliosides found in sperm of rainbow trout (Oncorhynchus mykiss) and was shown to be present prominently at the sperm head by immunochemical staining with its specific mAb kdn3G. Liposomes containing (KDN)GM3 adhere specifically to GalNAcβ4Galβ4Glcβ1Cer (Gg3Cer)-coated plastic plates. Interaction between (KDN)GM3 and Gg3Cer was much stronger than that previously observed between Neu5Acα2→3Galβ4Glcβ1Cer and Gg3Cer. (KDN)GM3–Gg3Cer interaction did not require the presence of Ca2+ and Mg2+, but was enhanced in the presence of Mn2+. Fresh trout sperm adhered specifically to Gg3Cer-coated plates under physiological conditions, and the binding was inhibited by pretreatment of sperm with mAb kdn3G. The presence of Gg3 or Gg3-related epitope structure in the specific area surrounding the micropyle, through which sperm enter the egg, was confirmed by immunogold labeling under electron microscopy. These findings suggest that initial sperm-egg adhesion during the process of fertilization occurs when sperm adhere to the area surrounding the micropyle through specific interaction between (KDN)GM3 on the sperm head and Gg3 epitope (GalNAcβ4Galβ1→) expressed at a defined region of the egg surface membrane.

Glycoconjugates are highly expressed at the cell surface, displaying a morphologically distinctive form of plasma membrane termed glycocalyx (1). Some glycans with extensive structural variation, such as polysialic acid of neural cell adhesion molecule, may be required for “coarse and fine tuning” of adhesive behavior of cells (2). Numerous studies have suggested that the glycocalyx may be involved in cell adhesion, although it was only recently that endogenous lectins, particularly selectins (3–5) and siglecs (6), were assigned as binding sites of defined glycoconjugates to mediate cell-to-cell adhesion among blood cells, endothelial cells, and immunocytes. Adhesion of cells in tissues may be mediated by a group of endogenous lectins termed galectins (7), although unambiguous evidence for this concept has not been presented.

Prominent changes in glycosylation and glycosylation-dependent cell adhesion are typically observed during ontogenic and oncogenic processes (8). Compaction of morula, the first cell adhesion event during embryogenesis, and its model process, displayed as autoaggregation of teratocarcinoma, are mediated by carbohydrate (chydr)-to-chydr interaction (9, 10) (see Discussion). Adhesion of mouse melanoma B16 to mouse endothelial cells is mediated by GM3-to-Galβ4Glcβ1Cer (LacCer) interaction, and B16 metastasis to lung may be mediated by GM3-to-Gg3 interaction (11–13).

A series of studies indicate that sponge (Microciona prolifera) cell aggregation factor is a macromolecular proteoglycan having repetitive glycan. Antibodies to the glycan inhibited the aggregation, suggesting that it takes place between glycans (14, 15), although this may not completely exclude the possibility of glycan–protein interaction. The fact that autoaggregation of sponge cells is mediated by chydr-to-chydr interaction was verified by determination of specific structures involved, i.e., pyruvated Galβ4GlcNAcβ3Fuc (16), and recently by self-recognition of 3-sulfated GlcNacβ3-l-Fucα1→R (17). Specific interaction of various other chydr with chydr was verified by quantitative determination using surface plasmon resonance spectroscopy (18)‡‡ or molecular force microscopy (20) (see Discussion).

We (S.I. and Y.I.) discovered in 1986 a novel deaminated analogue of sialic acid, 2-keto-3-deoxy-d-glycero-d-galacto-nononic acid (KDN) (21), and subsequently GM3 ganglioside containing KDN, (KDN)GM3 (KDNα2→3Galβ4Glcβ1Cer), as a major component of rainbow trout sperm (22). A mAb (kdn3G) was established against this KDN-ganglioside (23). In this article, we demonstrate: (i) specific staining of the sperm head by mAb kdn3G; (ii) strong interaction of (KDN)GM3 with Gg3 epitope, and (iii) presence of Gg3 or Gg3-related epitope structures in the region of the trout egg surface surrounding the micropyle, a tiny opening in the vitelline envelope (VE) through which sperm enter (24). Based on these findings, we propose that the basic biological process of fertilization in trout takes place through a specific type of chydr-to-chydr interaction.

Materials and Methods

Materials.

(KDN)GM3 was obtained from rainbow trout (Oncorhynchus mykiss) sperm as reported (22). Gg3Cer was prepared as described (25). IgG anti-(KDN)GM3 mAb (kdn3G) (23) and mouse anti-Gg3 IgM mAb 2D4 (26) were prepared as described. Fresh rainbow trout sperm were obtained from the Okutama Fish Farm, Tokyo Metropolitan Government, kept on ice, and used within 12 h after collection.

Indirect Immunofluorescence of Rainbow Trout Sperm with mAb kdn3G.

Mature spermatozoa of rainbow trout were fixed with 2% paraformaldehyde/0.1 M sodium cacodylate (pH 7.2) for 1.5 h on ice and washed three times with a solution of 5% sucrose, 0.1% NaN3, and 0.1 M sodium cacodylate buffer (pH 7.2). The fixed sperm were incubated with mAb kdn3G (7.5 μg/ml of IgG) for 1 h at room temperature. A (KDN)GM3-negative hybridoma supernatant was used as a control. Immunoreacted sperm were washed three times with 2% BSA in PBS, treated in the dark with FITC-labeled goat anti-(mouse IgG) IgG (affinity-purified; Cappel), and diluted 1:10 (100 μg/ml) with 2% BSA at room temperature. After 1 h, the sperm were washed three times with PBS, and fluorescence was observed with an Olympus BHT-PFK epifluorescence microscope.

Preparation of Liposomes.

Preparation of ganglioside-containing liposomes was based on the single compartment method (9). In brief, cholesterol (Sigma), dimyristoyl-l-α-phosphatidylcholine (C14:0), (Sigma), and purified glycosphingolipids (GSLs) were mixed in chloroform/methanol (2:1, vol/vol) solution in a ratio of 4:7:3 (wt/wt) and evaporated to dryness under a nitrogen stream. For preparation of radiolabeled liposomes, 0.8 nmol or 1.9 × 105 cpm of labeled [1,2,6,7-3H(N)]-cholesterol (Du Pont-New England Nuclear) (55,000 cpm/μg lipids) was added to the above mixture. These lipid mixtures, dried completely in test tubes, were added with 2 ml PBS with or without bivalent cation (see below), mixed vigorously with a Vortex mixer for 1 min, sonicated for 30 min, and left at room temperature for 30 min. The sonication/incubation procedure was repeated three times, and colloidal suspensions containing GSL-liposomes with concentration 35 μg/ml in PBS were obtained. Liposomes without GSLs were prepared by the same procedure. We used PBS with final concentrations of 137 mM NaCl, 2.7 mM KCl, 1.5 mM KH2PO4, and 8.1 mM Na2HPO4. To this was added 0.9 mM CaCl2, 0.5 mM MgCl2, or 0.1 mM MnCl2. To eliminate the effect of bivalent cations on chydr-dependent binding, 10 or 50 mM EDTA was used as described (9).

Labeling of Fresh Rainbow Trout Sperm with Fluorogenic Substrate Bis(carboxymethyl)carboxyfluorescein Acetoxymethyl Ester (BCECF-AM).

The fluorogenic substrate BCECF-AM can be incorporated into intact cells, resulting in fluorescence of cells (27, 28). Mature trout sperm were centrifuged (2,000 rpm, 5 min) to remove seminal plasma. After washing with salmon Ringer's solution (SRS) (29) (134.5 mM NaCl, 3.8 mM KCl, 3.2 mM CaCl2, adjusted to pH 7.0 with 0.1 M NaHCO3), the sperm were resuspended in SRS (1 × 107/ml). To this sperm suspension was added 5 μl of BCECF-AM per ml. Stock solution of BCECF-AM/DMSO was 1 mg/3 ml (Funakoshi, Tokyo). The suspension was incubated for 30 min at 37°C with gentle shaking (125 rpm/min) and centrifuged to remove supernatant. The sperm were washed twice with SRS and finally suspended in SRS containing 0.5 mM MgCl2 and 0.1 mM MnCl2, pH 7.0.

Treatment of Rainbow Trout Sperm with mAb.

For pretreatment of trout sperm with mAb kdn3G, fresh sperm (1 × 108/ml) were incubated separately in 5 ml SRS for 20, 40, or 60 min on ice with mAb concentration 0.1, 1, 5 or 20 μg/ml. Goat IgG antibody vs. mouse Ig(G+M) (Tago) was used as a control. The sperm were washed twice with SRS and resuspended in SRS containing 0.5 mM MgCl2 and 0.1 mM MnCl2.

Coating of GSLs on Polystyrene Plates.

GSLs were dissolved in a small amount of chloroform/methanol (2:1, vol/vol), and ethanol was added to make a final GSL concentration of 50 μg/ml. After vortex mixing and brief sonication, 50 μl containing various amounts of GSLs was placed in polystyrene wells (96-well flat-bottom plastic plate, 0.3 ml/well volume; no. 2581, Costar) and incubated to complete dryness at 37°C (about 2 h). GSL-coated solid phase was incubated with 1% BSA/PBS at room temperature for 2 h to block nonspecific binding and washed twice with PBS before use.

Upper and lower phases of extracts of VE and membranous fraction (MF) from trout eggs, prepared as below, were placed in each well at concentration 0–10 μg neutral sugar per well, and plates were dried and blocked as described above.

Assay for Liposome Binding to GSL-Coated Solid Surface.

Each GSL-coated or noncoated well of 96-well plates was added with 100 μl of 3H-labeled freshly prepared liposomes, incubated at room temperature for 16 h, then washed repeatedly with PBS. At each washing, the plate was inverted and shaken gently to remove PBS solution. The last washing was tested to confirm absence of contaminating radioactivity. Radioactivity of liposomes remaining on solid phase was measured by scintillation counter (Beckman LS 3801) using ACS-II (Amersham Pharmacia) as a scintillant.

Assay for Binding of Sperm to GSL-Coated Surface.

To each GSL-coated or noncoated well was added 100 μl (1 × 108/ml) of BCECF-labeled sperm, unlabeled sperm, or sperm treated with mAb kdn3G or anti-mouse Ig(G+M) (as control), and incubated at 4°C. Binding of sperm to Gg3Cer-coated plates was examined at 1, 3, 6, 9, 12, and 16 h. Wells were then washed three or four times with SRS. For BCECF-labeled sperm, fluorescence remaining on the solid phase was directly measured at 530 nm for emission at excitation wavelength 490 nm on a microplate reader (Corona Electric MTP-32). Alternatively, intact sperm were incubated with GSL-coated wells. After washing the wells, bound sperm on the solid phase were fixed with 20 μl 30% ethanol for 5 min at room temperature, and 10 μl 4′,6-diamidino-2-phenylindole (10 μg/ml) was added to each well. The wells were incubated for 10 min and measured at 450 nm for emission at excitation wavelength 365 nm. Sperm treated with mAbs were also tested by this method.

Preparation of VE and MF from Unfertilized Eggs.

Fresh unfertilized trout eggs (440 g) were crushed with a pestle and squeezed through Tetoron gauze. VE on gauze was washed until free from the bulk of yolk and cytoplasm by suspending in 0.8% NaCl (1.0 liter), and filtered through gauze. The filtrate combined with washing solution was centrifuged at 10,000 g for 50 min at 4°C, and the supernatant was further centrifuged at 100,000 g for 1 h at 2°C. The precipitate was used as the MF.

Aqueous and Organic Phases from VE and MF Extract and Their Binding Ability.

VE and MF prepared as above were twice extracted with isopropanol/hexane/water (55:25:20) (30), extract was evaporated to dryness, and residue was extracted in chloroform/methanol (2:1) and Folch-partitioned three times (31) to separate upper (aqueous) from lower (organic) phase. These phases were evaporated to dryness, and residue was dissolved in ≈100 μl chloroform/methanol (2:1) and diluted with ≈5 ml ethanol with sonication. Neutral sugar content of an aliquot (5–10 μl) of this ethanol solution was determined by orcinol-sulfuric acid spotted on TLC. Various quantities of aliquots containing 0–10 μl neutral sugar were added to each well and air-dried. To each well was added (KDN)GM3-liposome or sperm suspension, and adhesion was determined.

Immunoelectron Microscopy.

Trout eggs were cut into approximately equal halves with a razor blade, and VE was purified by repeated washing with 5 mM Tris⋅HCl buffer (pH 7.0) containing 5 mM EDTA (Kanto Chemical, Tokyo) and 1 mM phenylmethanesulfonyl fluoride (Sigma). The purified VE was incubated with 1% BSA (Paesel-Lorei, Hanau, Germany), 0.1% gelatin (Sigma), and 0.1% skim milk (Difco) in PBS (pH 7.5) for 2 h at room temperature, then incubated with anti-Gg3 mAb 2D4 for 3 h at 28–30°C. The 2D4 was used at a dilution of 1:5 or 1:10 with PBS containing 1% BSA, 0.1% gelatin, and 0.1% skim milk. The VE incubated with mAb was rinsed several times in PBS, and antibody-binding sites were detected by 3-h incubation in goat anti-mouse IgM F(ab′)2 conjugated to colloidal gold (10 nm) (BioCell Laboratories). After washing with PBS, osmication, and dehydration, the VE was embedded in Quetol 812. The control consisted of incubation of ultrathin sections with goat anti-mouse IgM F(ab′)2-gold conjugate without previous incubation with primary mAb. Ultrathin sections were stained with uranyl acetate and viewed by electron microscopy.

TLC and Western Blot Immunostaining of Trout Egg VE and MF Extract.

Material present in the aqueous and the organic phases prepared as above was analyzed by (i) TLC with various solvents and visualized by orcinol-sulfuric acid reaction and immunostaining (32, 33); and (ii) extraction of VE and MF with 50 mM Hepes (pH 7.5), containing 2 mM EDTA, 50 mM NaCl, 2% Triton X-100, 1 mM DTT, and 1 mM phenylmethanesulfonyl fluoride, and subjected to SDS/PAGE and Western blotting by the method of Towbin et al. (34). Immunostaining was performed by using mAb 2D4 as described (33, 34).

Results

Specific Adhesion of (KDN)GM3-Liposome to Gg3Cer-Coated Plastic Surface.

(KDN)GM3-liposome (35 μg/ml) was incubated in PBS containing 0.9 mM CaCl2 and 0.5 mM MgCl2 with polystyrene surfaces coated with Gg3Cer, LacCer, Neu5Acα2→ 3Galβ4Glcβ1Cer [(Neu5Ac)GM3], or (KDN)GM3. (KDN)GM3-liposome bound strongly to Gg3Cer-coated plastic surface (Fig. 1), and this binding was shown to be dose-dependent. No such strong binding of (KDN)GM3 was observed to LacCer-, (Neu5Ac)GM3-, or (KDN)GM3-coated surfaces. Control liposomes made from only phosphatidylcholine and cholesterol did not adhere to Gg3Cer-coated surface. These results indicate that (KDN)GM3–Gg3Cer interaction is highly specific.

Figure 1.

Figure 1

Open in a new tab

Interaction of radiolabeled (KDN)GM3-liposome with GSL-coated polystyrene surfaces. (KDN)GM3-liposome (35 μg/ml) in PBS was incubated in 96-well plate coated with various GSLs in various amounts (0–2.5 μg/well). Radioactivity of (KDN)GM3-liposome bound to GSL-coated plate was determined by scintillation counter (Beckman LS 3801). (KDN)GM3-liposome was incubated with: Gg3Cer (●), LacCer (○), (Neu5Ac)GM3 (■), and (KDN)GM3 (□).

Interaction between (KDN)GM3 and Gg3Cer was much stronger than that observed previously between (Neu5Ac)GM3 and Gg3Cer (35), as indicated by several findings: (i) in the absence of MnCl2 (Neu5Ac)GM3-liposome at 35 μg/ml did not bind to Gg3Cer-coated plate, whereas distinct binding was seen for (KDN)GM3-liposome at this concentration; (ii) in the presence of 0.1 mM MnCl2 (Neu5Ac)GM3-liposome at 35 μg/ml adhered to Gg3Cer-coated plate less strongly (about 25%) than (KDN)GM3-liposome; and (iii) at higher liposome density (1.0 mg/ml), interaction between (KDN)GM3-liposome and Gg3Cer-coated plate was about 4-fold stronger than that between (Neu5Ac)GM3 and Gg3Cer.

Effect of Bivalent Cations on (KDN)GM3–Gg3Cer Interaction.

Binding of (KDN)GM3-liposome to Gg3Cer-coated plastic surface was doubled by addition of Mn2+ to PBS (not shown). The effect of addition of Ca2+ or Mg2+ was smaller than that of Mn2+. Presence of EDTA (50 mM) had no significant effect on the interaction (not shown).

Specific Adhesion of Trout Sperm to Gg3Cer-Coated Plastic Surface and Localization of (KDN)GM3 at Trout Sperm Head.

Fresh sperm were incubated in SRS supplemented with 0.5 mM MgCl2 and 0.1 mM MnCl2 with GSL-coated plastic plates at 4°C for 1, 3, 6, 9, 12, or 16 h, and bound sperm were quantitated as described in Materials and Methods. Incubation at 4°C for 6–9 h was optimal for observation and quantitation of binding. Sperm adhered to Gg3Cer-coated plate (Fig. 2A), but not to LacCer-, (Neu5Ac)GM3-, or (KDN)GM3-coated plates (Fig. 2B). The binding of sperm to Gg3Cer-coated plates is based on specific localization of (KDN)GM3 at trout sperm head, as indicated by specific fluorescence immunostaining of sperm head with mAb kdn3G (Fig. 2C). Light microscopic imaging of the same sperm is shown in Fig. 2D. This finding confirms our previous observation of such specific localization (23).

Figure 2.

Figure 2

Open in a new tab

Adhesion of fresh sperm to GSL-coated plastic plates, and localization of (KDN)GM3 at the sperm head. (A) Trout sperm (107) prelabeled with BCECF-AM were added to each well of a 96-well plate coated with various amounts (0–50 μg/well) of Gg3Cer, LacCer, (Neu5Ac)GM3, or (KDN)GM3, and incubated at 4°C for 9 h. Fluorescence remaining on the solid phase was directly measured at 530 nm for emission upon excitation at 490 nm. Relative fluorescence intensity represents sperm numbers bound onto plates coated with Gg3Cer (●), LacCer (○), (Neu5Ac)GM3 (■), and (KDN)GM3 (□). (B) Fresh sperm (107) were incubated with Gg3Cer-, LacCer-, (Neu5Ac)GM3-, and (KDN)GM3-coated wells (0–3.0 μg/well). Adherent sperm were stained with 4′,6-diamidino-2-phenylindole and measured at 450 nm for emission upon excitation at 365 nm. (C) Specific fluorescence at sperm heads stained by mAb kdn3G/FITC-labeled secondary goat anti-mouse IgG antibody (see Materials and Methods). (D) The same sperm examined in C were observed under a light microscope. (Magnifications: ×3,600.)

Binding of sperm to Gg3Cer-coated plate (Fig. 3A), as observed under light microscopy, was much higher than that to LacCer-coated plate (Fig. 3B), and there was essentially no binding to (Neu5Ac)GM3- (Fig. 3C) or (KDN)GM3-coated plates (Fig. 3D).

Figure 3.

Figure 3

Open in a new tab

Light microscopic observation of trout sperm adhered onto solid surfaces coated with various GSLs. Sperm were incubated on GSL-coated solid surfaces (3.0 μg/well) at 4°C for 9 h and observed under a light microscope (Olympus, IMT-2) after washing. Plates were coated with (A) Gg3Cer, (B) LacCer, (C) (Neu5Ac)GM3, or (D) (KDN)GM3. (Scale bar = 16 μm.)

Inhibition of Sperm Binding to Gg3Cer-Coated Plastic Surface by mAb kdn3G.

Trout sperm were pretreated with mAb kdn3G and incubated with Gg3Cer-coated plastic plates. Binding of sperm to the plates was inhibited by kdn3G (Fig. 4B). The mAb had no effect on adhesion of anti-mouse (IgG + IgM) IgG-treated sperm to the same surface (Fig. 4A). These results indicate that (KDN)GM3 on the sperm surface is involved in the adhesive interaction between sperm and Gg3Cer-coated solid surface.

Figure 4.

Figure 4

Open in a new tab

Microscopic observation of the inhibitory effect of mAb kdn3G on specific binding of trout sperm to Gg3Cer-coated plastic plates. Fresh sperm were pretreated on ice for 1 h with 5 μg/ml of (A) goat IgG antibody directed to mouse IgG+IgM or (B) mAb kdn3G, and incubated with Gg3Cer-coated plastic plate at 4°C for 6 h. (Scale bar = 16 μm.)

Adhesion of (KDN)GM3-Liposome or Trout Sperm to the Extracts from Trout Egg VE and MF.

We tested the ability of (KDN)GM3-liposome and fresh sperm to adhere to plastic surface coated with aqueous and organic phase fractions prepared from VE and MF. Only aqueous phase material extracted from both VE and MF showed significant binding to (KDN)GM3-liposome, whereas organic phase material showed much lower or no significant binding (Fig. 5A). Similarly, only aqueous phase material coated on plastic surface bound to sperm, whereas organic phase material did not (Fig. 5B). These results indicate possible occurrence of ligand(s) or receptor for sperm surface (KDN)GM3 on VE and in MF.

Figure 5.

Figure 5

Open in a new tab

Adhesion of (A) (KDN)GM3-liposome and (B) trout sperm to plastic surfaces coated with the aqueous and organic phases of Folch extract (see Materials and Methods) from VE and MF. The surface of plastic wells was coated with 0–10 μg (as neutral sugar) materials of aqueous (○) and organic (■) phases obtained from VE, and aqueous (●) and organic (□) phases obtained from MF. (KDN)GM3-liposome and fresh sperm were incubated at room temperature for 16 h and at 4°C for 6 h, respectively. Amounts of liposomes and numbers of sperm remaining on the plastic surfaces after washing were determined as described in Fig. 1.

Preliminary Demonstration of Gg3 Epitope in Extracts of VE and MF.

TLC immunostaining with anti-Gg3 mAb 2D4, applied to aqueous and organic phases of VE and MF, indicated the presence of an immobile or extremely slow-migrating band in the aqueous phase (data not shown). No band corresponding to Gg3Cer was detectable. Western blot analysis showed several 2D4-positive bands, including high molecular weight components, in extracts of both VE and MF (data not shown). These results suggest the presence of polyglycosylceramide or hydrophobic glycoproteins having GalNAcβ4Galβ1→ epitope in VE and MF of unfertilized eggs.

Location of Gg3-Related Epitope at Trout Egg Membrane (VE), as Revealed by Immunoelectron Microscopy.

The location and distribution pattern of Gg3-related epitopes were studied based on binding of gold sol coated with anti-Gg3 mAb 2D4 and control gold sol without primary antibody, as described in Materials and Methods. Of three layers of VE, the surface of the second layer is exposed at the area surrounding the micropyle vestibule, where there is no first layer. This surface of exposed second layer expresses a high level of Gg3-related structure bound to the gold sol coated with 2D4 (Fig. 6A and Inset 1). In contrast, the vestibular area of micropyle having no first or second layer, but only third layer, did not express this Gg3-like epitope (Fig. 6A). No staining of gold sol without primary 2D4 was observed at any of the locations at second layer in the surrounding area, nor at vestibule of micropyle (Fig. 6B and Inset 2). This pattern suggests that sperm may bind to the exposed second layer at the surrounding area through their heads, and one sperm reaches the micropyle (see schematic diagram in Fig. 6C). The third layer, particularly its inner surface, expresses Gg3-like epitope, except in the micropyle area (data not shown).

Figure 6.

Figure 6

Open in a new tab

(A) Binding of gold sol particles coated with anti-Gg3 mAb 2D4, to the exposed second layer (ex) of VE in the area surrounding the micropyle, but not to the third layer of the micropyle vestibule (vm) (magnification: ×57,000). Detailed structure of ex in A is shown at higher magnification (×570,000) in Inset 1. (B) Control of same area as in A, treated with gold sol particles without coating of primary antibody (×43,700). Detailed structure of ex in B is shown at higher magnification (×437,000) in Inset 2. Note that only exposed second layer (ex) showed binding to anti-Gg3. (C) Schematic diagram of VE structure around the micropyle (m). Three layers (1, 2, 3) cover the entire egg, forming VE. Layers 1 and 2 are discontinuous in the area surrounding the micropyle and its vestibule (vm). Layer 2 contains Gg3 epitope and is exposed only in the area indicated by red; therefore, sperm heads are able to bind only in this area, presumably through Gg3 epitope/(KDN)GM3 interaction. Because many sperm thereby congregate in this area, there is a good probability for one of them to enter the micropyle, as shown, leading to fertilization.

Discussion

Interaction between chydr to chydr has been unequivocally demonstrated based on physical changes (optical rotation, 13C-NMR, viscosity, etc.) of agarose upon its gelification (36), or agarose-to-β1→4 glycan interaction (37). Such obvious physical interaction observed in vitro has not been considered for biological interactions in vivo.

Because of well-documented observations that cell-to-substratum interaction is based primarily on cell adhesion receptors, e.g., integrin family (38), and that cell-to-cell interaction occurs through homotypic binding between Ig-like receptors (39) or between cadherins (40, 41), development of a concept of chydr-dependent adhesion was much delayed, despite recent evidence for cell-cell adhesion mediated by endogenous lectin (selectin, siglec, galectin) (3–7). Much less attention has been paid to the idea of cell adhesion based on chydr-to-chydr interaction in vivo.

Starting from different historical backgrounds, a few lines of study on cell adhesion have incidentally arrived at similar conclusions, i.e., a specific chydr can bind to a complementary chydr, in some cases based on homotypic autorecognition, in other cases based on heterotypic interaction. Some examples with different historical backgrounds are as follows.

A line of studies along embryogenesis.

Autoaggregation of mouse or human teratocarcinoma provides a model of the “compaction” process in preimplantation embryo. A high expression of Galβ4(Fucα3)GlcNAcβ1 (Lex) (SSEA-1) in mouse morula (42–44), and inhibitability of compaction by multivalent Lex (45) were observed. A consequent search for endogenous lectins recognizing Lex (46) led to the concept of autorecognition of Lex by Lex (9). Lex is minimally expressed in primate embryos; instead, globo-series GalNAcβ3Galα4Galβ4Glcβ1Cer (Gb4) and Galβ3GalNAcβ3Galα4Galβ4Glcβ1Cer and other extended globo-epitopes (SSEA3, SSEA4) are highly expressed in primate embryos and human teratocarcinoma cell lines 2102 and Tera-2 (47). Autoaggregation of 2102 cells is clearly mediated by interaction of Gb4 with Galβ3GalNAcβ3Galα4Galβ4Glcβ1Cer (Gb5), and Gb4 with Galβ4GlcNAcβ3Galβ4Glcβ1Cer (nLc4) (10).

A line of studies along species-specific aggregation and sorting of sponge cells.

This well-known marine developmental biology model is based on species-specific aggregation factors (48). Biochemical analysis of the factors identified it as proteoglycan-type macromolecules, in which repeated glycan moiety of proteoglycan may cause cell aggregation, suggesting glycan–glycan interaction (14, 15, 49). In this series of studies, the possibility of protein involvement in cell aggregation was difficult to completely rule out, until the clear identification of specific oligosaccharides that shows autorecognition (17) (see Introduction).

A line of studies along melanoma cell adhesion and invasiveness.

Mouse melanoma B16 cells express unusually high level of GM3 and show adhesion and motility enhancement on plates coated with Gg3Cer, LacCer, or (weakly) Gb4, based on interaction of these GSLs with GM3 (GM3-dependent adhesion). B16 cells adhere to mouse endothelial cells that express LacCer and Gb4. Gg3-like component was detectable by surface labeling of mouse lung microvasculature. Therefore, lung metastasis of B16 cells is based on GM3-dependent adhesion (11, 12), and lung metastasis is inhibited by GM3-liposome or Gg3-liposome (13).

A line of studies along GSL interaction in myelin membrane.

Two major GSLs present in myelin membrane are GalCer and sulfatide, which show strong interaction between their head groups in the presence of Ca2+ (50, 51). This interaction is strongly enhanced when α-hydroxy group is present at fatty acid. Biological significance of this interaction is not clear, but it probably helps to stabilize the myelin membrane.

We now add an important line of studies supporting specific chydr–chydr interaction in sperm-egg binding as a basis of fertilization. Although the data are so far limited to trout, they open up an exciting possibility that such basic biological processes as fertilization are based on chydr–chydr interaction. We showed previously that (KDN)GM3 is specifically localized at the sperm head (23). We now present strong evidence that (KDN)GM3 at the sperm head is involved in binding to Gg3 epitope expressed at the egg cell surface, i.e.: (i) specific strong interaction of (KDN)GM3 with Gg3 epitope; (ii) sperm head interaction with Gg3Cer-coated plate, and its inhibition by anti-Gg3 mAb 2D4; (iii) binding of sperm or (KDN)GM3-liposome to aqueous phase material from lipid extract of VE; and (iv) specific localization of Gg3 or Gg3-like epitope in the area surrounding the micropyle at the egg surface. This specific localization leads to clustering of sperm in this region, thereby facilitating entry of one particular sperm into the narrow opening, as illustrated in Fig. 6C. Detailed structure of fish egg envelope and micropyle have been described (24).

In all of these lines of study, chydr–chydr interaction was clearly observed when GSL at the cell surface was used as target, and when GSL was used as probe to demonstrate GSL-to-GSL interaction. In the present case, however, (KDN)GM3 is GSL, but Gg3 epitope may be carried in polyvalent fashion by polyglycosylceramide or glycoprotein in VE, which allows strong binding to (KDN)GM3 present at sperm head. Microdomains bearing GSLs or mucin-type glycoproteins having O-linked glycans are organized with signal transducer molecules such as cSrc, Src family kinases, and small G proteins. Thus, chydr-dependent adhesion activates signal transducers to modify cellular phenotype (52–54). Therefore, binding of (KDN)GM3 on trout sperm head to Gg3 epitope carried by glycoprotein at the egg surface may induce various signaling processes.

In the process of sea urchin fertilization, a macromolecular sperm-binding glycoprotein having large O-linked glycans is expressed at the egg surface. The O-glycan mediates genus-nonspecific sperm-egg binding, whereas a specific peptide domain within the glycoprotein mediates genus-specific binding (19). It is possible that the fertilization process in trout is based on additional mechanisms cooperating with (KDN)GM3–Gg3 epitope interaction as described in this article.

Chydr–chydr interaction has been determined on a quantitative basis, using various biophysical methods. Interaction of GM3 monolayer (Langmuir layer) with various multivalent oligosaccharides bound to polystyrene was measured by surface plasmon resonance spectroscopy. Ka value for GM3 binding to Gg3-polystyrene was 2.5 × 106 M−1. The value for binding to Lac-polystyrene was 7.7 × 104 M−1. Ka value for binding of (KDN)GM3 monolayer to Gg3 was much lower than the value for GM3: 2.5 × 105 M−1 (18). This finding is in conflict with our liposome binding data showing that binding of (KDN)GM3 liposome to Gg3 epitope was much stronger than that of GM3. The discrepancy may be caused by the fact that our determination is based on binding of liposome containing phosphatidylcholine and cholesterol, which is similar to the natural state.

Recently, accurate molecular force of Lex-to-Lex interaction was determined by molecular force microscopy. Molecular force of a single Lex-to-Lex interaction was determined as 20 ± 4 pN (20). Because ≈400 pN may be necessary for cell–cell adhesion, 20 Lex-Lex molecular pairs may be sufficient to cause Lex-dependent cell adhesion. In this case, molecular force of Lex-to-Lex binding is not affected by presence or absence of Ca2+ (20), even though Ca2+ is necessary for observing a clear Lex-dependent cell or liposome adhesion. Ca2+ is assumed to provide a cell surface environment, rather than affect molecular force for interaction. Interestingly, the presence or absence of Ca2+ has no effect on (KDN)GM3-to-Gg3 interaction or GalGb4–Gb4 interaction (10).

Acknowledgments

S.Y. and Y.I. are grateful to Dr. Ken Kitajima for his unfailing help at the initial stage of this work at the University of Tokyo. The major part of this study was originally carried out at the Department of Biophysics and Biochemistry, University of Tokyo, and The Biomembrane Institute, Seattle, WA. We thank Dr. Stephen Anderson for scientific editing and figure preparation. This work was supported by the Mizutani Foundation for Glycoscience (Japan) and National Science Council and National Health Research Institutes (Taiwan) (Grant NSC 90-2311-B-001-096 and NHRI-GT-EX89B805P to Y.I.) and by the National Institutes of Health/National Cancer Institute (Grants OIG CA42505, CA80054, and CA82167 to S.H.).

Abbreviations

BCECF-AM

bis(carboxymethyl)carboxyfluorescein acetoxymethyl ester

chydr

carbohydrate

Gb4

GalNAcβ3Galα4Galβ4Glcβ1Cer

Gg3Cer

GalNAcβ4Galβ4Glcβ1Cer

GSL

glycosphingolipid

KDN

2-keto-3-deoxy-d-glycero-d-galacto-nononic acid

(KDN)GM3

KDNα2→3Galβ4Glcβ1Cer

LacCer

Galβ4Glcβ1Cer

Lex

Galβ4(Fucα3)GlcNAcβ1

MF

membranous fraction

(Neu5Ac)GM3

Neu5Acα2→3Galβ4Glcβ1Cer

SRS

salmon Ringer's solution

VE

vitelline envelope

Footnotes

‡‡

Hernaiz, M. J., Gallardo, J. A., de la Fuente, J. M., Barrientos, A. G. & Penades, S. (2001) Glycoconj. J.18, 31 (abstr.).

References

  • 1.Varki A, Cummings R, Esko J, Freeze H, Hart G, Marth J. Essentials of Glycobiology. Plainview, NY: Cold Spring Harbor Lab. Press; 1999. [PubMed] [Google Scholar]
  • 2.Inoue S, Inoue Y. J Biol Chem. 2001;276:31863–31870. doi: 10.1074/jbc.M103336200. [DOI] [PubMed] [Google Scholar]
  • 3.Varki A. Proc Natl Acad Sci USA. 1994;91:7390–7397. doi: 10.1073/pnas.91.16.7390. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Lowe J B. Kidney Int. 1997;51:1418–1426. doi: 10.1038/ki.1997.194. [DOI] [PubMed] [Google Scholar]
  • 5.Rosen S D. Am J Pathol. 1999;155:1013–1020. doi: 10.1016/S0002-9440(10)65201-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Crocker P R, Varki A. Immunology. 2001;103:137–145. doi: 10.1046/j.0019-2805.2001.01241.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Cooper D N, Barondes S H. Glycobiology. 1999;9:979–984. doi: 10.1093/glycob/9.10.979. [DOI] [PubMed] [Google Scholar]
  • 8.Hakomori S, Kannagi R. J Natl Cancer Inst. 1983;71:231–251. [PubMed] [Google Scholar]
  • 9.Eggens I, Fenderson B A, Toyokuni T, Dean B, Stroud M R, Hakomori S. J Biol Chem. 1989;264:9476–9484. [PubMed] [Google Scholar]
  • 10.Yu S, Withers D A, Hakomori S. J Biol Chem. 1998;273:2517–2525. doi: 10.1074/jbc.273.5.2517. [DOI] [PubMed] [Google Scholar]
  • 11.Kojima N, Hakomori S. J Biol Chem. 1991;266:17552–17558. [PubMed] [Google Scholar]
  • 12.Kojima N, Shiota M, Sadahira Y, Handa K, Hakomori S. J Biol Chem. 1992;267:17264–17270. [PubMed] [Google Scholar]
  • 13.Otsuji E, Park Y S, Tashiro K, Kojima N, Toyokuni T, Hakomori S. Int J Oncol. 1995;6:319–327. doi: 10.3892/ijo.6.2.319. [DOI] [PubMed] [Google Scholar]
  • 14.Misevic G N, Finne J, Burger M M. J Biol Chem. 1987;262:5870–5877. [PubMed] [Google Scholar]
  • 15.Misevic G N, Burger M M. J Biol Chem. 1990;265:20577–20584. [PubMed] [Google Scholar]
  • 16.Spillmann D, Hard K, Thomas-Oates J, Vliegenthart J F G, Misevic G, Burger M M, Finne J. J Biol Chem. 1993;268:13378–13387. [PubMed] [Google Scholar]
  • 17.Haseley S R, Vermeer H J, Kamerling J P, Vliegenthart J F G. Proc Natl Acad Sci USA. 2001;98:9419–9424. doi: 10.1073/pnas.151111298. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Matsuura K, Kitakouji H, Sawada N, Ishida H, Kiso M, Kitajima K, Kobayashi K. J Am Chem Soc. 2000;122:7406–7407. .. [Google Scholar]
  • 19.Lennarz W J. In: Sialobiology and Other Novel Forms of Glycosylation. Inoue Y, Lee Y C, Troy F A, editors. Osaka: Gakushin; 1999. pp. 1–5. [Google Scholar]
  • 20.Tromas C, Rojo J, de la Fuente J M, Barrientos A G, Garcia R, Penades S. Angew Chem Int Ed. 2001;40:3052–3055. doi: 10.1002/1521-3773(20010817)40:16<3052::AID-ANIE3052>3.0.CO;2-Q. [DOI] [PubMed] [Google Scholar]
  • 21.Nadano D, Iwasaki M, Endo S, Kitajima K, Inoue S, Inoue Y. J Biol Chem. 1986;261:11550–11557. [PubMed] [Google Scholar]
  • 22.Yu S, Kitajima K, Inoue S, Inoue Y. J Biol Chem. 1991;266:21929–21935. [PubMed] [Google Scholar]
  • 23.Yu S, Kitajima K, Inoue Y. Glycobiology. 1993;3:31–36. doi: 10.1093/glycob/3.1.31. [DOI] [PubMed] [Google Scholar]
  • 24.Kudo S. Annot Zool Jpn. 1982;55:224–235. [Google Scholar]
  • 25.Seyama Y, Yamakawa T. J Biochem (Tokyo) 1974;75:837–842. doi: 10.1093/oxfordjournals.jbchem.a130455. [DOI] [PubMed] [Google Scholar]
  • 26.Young W W J, MacDonald E M S, Nowinski R C, Hakomori S. J Exp Med. 1979;150:1008–1019. doi: 10.1084/jem.150.4.1008. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Rotman B, Papermaster B W. Proc Natl Acad Sci USA. 1966;55:134–141. doi: 10.1073/pnas.55.1.134. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Rink T J, Tsien R Y, Pozzan T. J Cell Biol. 1982;95:189–196. doi: 10.1083/jcb.95.1.189. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Janna H, Yamamoto T S. J Exp Zool. 1984;230:449–463. [Google Scholar]
  • 30.Kannagi R, Nudelman E D, Levery S B, Hakomori S. J Biol Chem. 1982;257:14865–14874. [PubMed] [Google Scholar]
  • 31.Folch J, Lees M B, Stanley G H S. J Biol Chem. 1957;226:497–509. [PubMed] [Google Scholar]
  • 32.Ye J, Kitajima K, Inoue Y, Inoue S, Troy F A. Methods Enzymol. 1994;230:460–484. doi: 10.1016/0076-6879(94)30029-1. [DOI] [PubMed] [Google Scholar]
  • 33.Yu S, Kitajima K, Inoue S, Muto Y, Kasama T, Handa S, Inoue Y. Biochemistry. 1993;32:9221–9229. doi: 10.1021/bi00086a030. [DOI] [PubMed] [Google Scholar]
  • 34.Towbin H, Staehelin T, Gordon J. Proc Natl Acad Sci USA. 1979;76:4350–4354. doi: 10.1073/pnas.76.9.4350. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Kojima N, Hakomori S. J Biol Chem. 1989;264:20159–20162. [PubMed] [Google Scholar]
  • 36.Arnott S, Fulmer A, Scott W E, Dea I C M, Moorhouse R, Rees D A. J Mol Biol. 1974;90:269–284. doi: 10.1016/0022-2836(74)90372-6. [DOI] [PubMed] [Google Scholar]
  • 37.Dea I C M, Rees D A. Carbohydr Polymers. 1987;7:183–224. [Google Scholar]
  • 38.Hynes R O. Cell. 1992;69:11–25. doi: 10.1016/0092-8674(92)90115-s. [DOI] [PubMed] [Google Scholar]
  • 39.Edelman G M, Crossin K L. Annu Rev Biochem. 1991;60:155–190. doi: 10.1146/annurev.bi.60.070191.001103. [DOI] [PubMed] [Google Scholar]
  • 40.Takeichi M. Curr Opin Cell Biol. 1993;5:806–811. doi: 10.1016/0955-0674(93)90029-p. [DOI] [PubMed] [Google Scholar]
  • 41.Takeichi M, Nakagawa S, Aono S, Usui T, Uemura T. Philos Trans R Soc London B. 2000;355:885–890. doi: 10.1098/rstb.2000.0624. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Solter D, Knowles B B. Proc Natl Acad Sci USA. 1978;75:5565–5569. doi: 10.1073/pnas.75.11.5565. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Hakomori S, Nudelman E D, Levery S B, Solter D, Knowles B B. Biochem Biophys Res Commun. 1981;100:1578–1586. doi: 10.1016/0006-291x(81)90699-9. [DOI] [PubMed] [Google Scholar]
  • 44.Gooi H C, Feizi T, Kapadia A, Knowles B B, Solter D, Evans M J. Nature (London) 1981;292:156–158. doi: 10.1038/292156a0. [DOI] [PubMed] [Google Scholar]
  • 45.Fenderson B A, Zehavi U, Hakomori S. J Exp Med. 1984;160:1591–1596. doi: 10.1084/jem.160.5.1591. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46.Rosenman S J, Fenderson B A, Hakomori S. In: Glycoconjugates in Medicine. Ohyama M, Muramatsu T, editors. Tokyo: Professional Postgraduate Services; 1988. pp. 43–50. [Google Scholar]
  • 47.Kannagi R, Cochran N A, Ishigami F, Hakomori S, Andrews P W, Knowles B B, Solter D. EMBO J. 1983;2:2355–2361. doi: 10.1002/j.1460-2075.1983.tb01746.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48.Humphreys T. Dev Biol. 1963;8:27–47. doi: 10.1016/0012-1606(63)90024-1. [DOI] [PubMed] [Google Scholar]
  • 49.Varner J A, Burger M M, Kaufman J F. J Biol Chem. 1988;263:8498–8508. [PubMed] [Google Scholar]
  • 50.Stewart R J, Boggs J M. Biochemistry. 1993;32:10666–10674. doi: 10.1021/bi00091a017. [DOI] [PubMed] [Google Scholar]
  • 51.Koshy K M, Boggs J M. J Biol Chem. 1996;271:3496–3499. doi: 10.1074/jbc.271.7.3496. [DOI] [PubMed] [Google Scholar]
  • 52.Iwabuchi K, Yamamura S, Prinetti A, Handa K, Hakomori S. J Biol Chem. 1998;273:9130–9138. doi: 10.1074/jbc.273.15.9130. [DOI] [PubMed] [Google Scholar]
  • 53.Iwabuchi K, Handa K, Hakomori S. J Biol Chem. 1998;273:33766–33773. doi: 10.1074/jbc.273.50.33766. [DOI] [PubMed] [Google Scholar]
  • 54.Handa K, Jacobs F, Longenecker B M, Hakomori S. Biochem Biophys Res Commun. 2001;285:788–794. doi: 10.1006/bbrc.2001.5225. [DOI] [PubMed] [Google Scholar]

Articles from Proceedings of the National Academy of Sciences of the United States of America are provided here courtesy of National Academy of Sciences

RESOURCES