Sulfated polysaccharides from the egg jelly layer are species-specific inducers of acrosomal reaction in sperms of sea urchins.

We have characterized the fine structure of sulfated polysaccharides from the egg jelly layer of three species of sea urchins and tested the ability of these purified polysaccharides to induce the acrosome reaction in spermatozoa. The sea urchin Echinometra lucunter contains a homopolymer of 2-sulfated, 3-linked α-L-galactan. The species Arbacia lixula and Lytechinus variegatus contain linear sulfated α-L-fucans with regular tetrasaccharide repeating units. Each of these sulfated polysaccharides induces the acrosome reaction in conspecific but not in heterospecific spermatozoa. These results demonstrate that species specificity of fertilization in sea urchins depends in part on the fine structure of egg jelly sulfated polysaccharide.

Successful fertilization by free-spawning organisms such as sea urchins can occur only if a series of constraints are overcome before the sperm ever makes contact with the egg (1). First, males and females must synchronize the time of release of their gametes (2). Once spawned, the sperm must find and interact with an egg of the correct species. A further event necessary for successful fertilization is induction of the acrosome reaction in the sperm (3,4), which involves fusion of the acrosomal vesicle membrane with the plasma membrane. This results in exocytosis of the vesicle contents, which include proteases and bindin. Concomitantly, actin in the subacrosomal region of the sperm polymerizes and causes extension of the tip of the sperm. As a consequence of these two events, bindin is localized to the outside of the tip of the process where it can then interact with an egg protein (3,4).
The acrosome reaction is induced when the sperms contact the egg jelly layer. The sea urchin egg is surrounded by a transparent gelatinous layer composed mainly of sulfated fucan, sialoprotein, and other glycoproteins or peptides (5). Previous attempts to identify the acrosome reaction inducer in sea urchin egg jelly have suggested that all the activity resides in the sulfated fucan (6,7). In addition, these authors suggested that the jelly coat preparations of some species of sea urchins are totally species-specific and induce the acrosome reaction only in homologous sperm. On the basis of these observations, it was suggested that the specificity of induction of the acrosome reaction might reside in structural differences in the carbohydrate linkages and/or location and degree of sulfation of the polysaccharide (6,7). Recent studies suggest that a glycoprotein or peptide, tightly associated with the sulfated fucan, was also involved in acrosome reaction induction (8 -11).
In this study, we isolated, purified, and characterized the fine structure of the sulfated polysaccharides from the egg jelly coat of three species of sea urchins. These compounds have simple, well-defined repeating structures that from each species present a particular pattern of sulfate substitution. Purified polysaccharide from the egg jelly induces the acrosome reaction in sperm from the same species of sea urchin and not from different species. This result is of considerable significance for the study of fertilization processes since it is the first time that fully described carbohydrate structures have been shown to regulate part of the process at such a specific level.

Sulfated Polysaccharides from the Jelly Coat of Sea Urchin
Extraction-Mature species of sea urchins were collected in Guanabara Bay (Rio de Janeiro, Brazil) and gametes were isolated by intracelomic injection of 0.5 M KCl (ϳ5 ml/specimen). Eggs were collected in a solution containing 450 mM NaCl, 9 mM KCl, 48 mM MgSO 4 ⅐7H 2 O, 10 mM CaCl 2 , and 6 mM NaHCO 3 . The egg jelly was separated by pH shock, as described previously (6). The acidic polysaccharides were extracted from the jelly coat by papain digestion and partially purified by cetylpyridinium chloride and ethanol precipitation as described (12).
Purification-The crude polysaccharides (ϳ100 mg) from the jelly coat of the sea urchins were applied to a DEAE-cellulose column (15 ϫ 2 cm), equilibrated with 50 mM sodium acetate buffer (pH 5.0), and washed with 250 ml of the same buffer. The column was eluted in three different steps. Initially, the column was eluted by a linear gradient prepared by mixing 50 ml of 50 mM sodium acetate buffer (pH 5.0) with 50 ml of 1.0 M NaCl in the same buffer. Then the column was washed with 100 ml of the sodium acetate buffer containing 1.0 M NaCl. Finally, the column was eluted by a linear gradient prepared by mixing 100 ml of 1.0 M NaCl with 100 ml of 5.0 M NaCl, both in the same sodium acetate buffer. The flow rate of the column was 15 ml/h, and fractions of 3.5 ml were collected in the different elution steps. Fractions were checked for fucose (or galactose) and sialic acid by the Dubois et al. reaction (13) and by the Ehrlich assay (14), respectively, and by their metachromasia (15). The NaCl concentration was estimated by conductivity. Fractions containing the sulfated ␣-L-fucan or the sulfated ␣-Lgalactan were pooled, dialyzed against distilled water, and lyophilized.
The DEAE-cellulose-purified sulfated polysaccharide (ϳ40 mg) was applied to a Sephacryl S-400 column (90 ϫ 1.5 cm) and eluted with 50 mM sodium acetate buffer (pH 5.0) at a flow rate of 8 ml/h. Fractions of 1.5 ml were collected and assayed by the reaction of Dubois et al. (13) and by their metachromasia (15). The column was calibrated using blue dextran and cresol red as markers of V o and V t , respectively.
Chemical Analyses-Total galactose was measured by the method of * This work was supported by grants from Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq: FNDCT and PADCT), Financiadora de Estudos e Projetos (FINEP) and Mizutani Foundation for Glysoscience (to P. A. S. M.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Dubois et al. (13) and total fucose measured by the method of Dische and Shettles (16). After acid hydrolysis of the polysaccharides (5.0 M trifluoroacetic acid for 5 h at 100°C), sulfate was measured by the BaCl 2 /gelatin method (17). The percentages of hexoses and 6-deoxyhexoses in the acid hydrolysates were estimated by paper chromatography in 1-butanol:pyridine:water (3:2:1, v/v) for 48 h and by gas-liquid chromatogaphy of derived alditol acetates (18). Optical rotations were measured with a digital polarimeter (Perkin-Elmer model 243-B).
Oxidation with L-Fucose Dehydrogenase-Fucose obtained by acid hydrolysis of the sulfated fucans (5.0 M trifluoroacetic acid for 5 h at 100°C), and authentic samples of D-or L-fucose (20 g of each) were incubated with 0.2 units of porcine liver L-fucose dehydrogenase, as described (19).
Oxidation with D-Galactose Oxidase-Galactose obtained by acid hydrolysis of the sulfated galactan (see above) and authentic samples of Dor L-galactose (20 g of each) were incubated with 1 unit of Dactylium dendroides D-galactose oxidase, as described (20).
Desulfation and Methylation of the Polysaccharides-Desulfation of the sulfated polysaccharides was performed by solvolysis in dimethyl sulfoxide, as described previously for desulfation of other types of polysaccharides (12,20). The native and desulfated polysaccharides (ϳ5 mg) were subjected to three rounds of methylation as described (21), with the modifications suggested by Patankar et al. (22). The methylated polysaccharides were hydrolyzed with 6 M trifluoroacetic acid for 5 h at 100°C, reduced with borohydride, and the alditols were acetylated with acetic anhydride:pyridine (1:1, v/v) (18). The alditol acetates of the methylated sugars were dissolved in chloroform and analyzed in a gas chromatography/mass spectrometry unit.
NMR Spectroscopy-1 H spectra were recorded at 500 MHz and 13 C spectra at 125 MHz using a Varian Unity 500 spectrometer. The po-lysaccharide sample (ϳ15 mg) was converted to the sodium salt by passage through a column 10 ϫ 1 cm of DOWEX 50-X8 Na ϩ form, and all samples were dissolved in approximately 0.7 ml of 99.8% D 2 O. The spectra were recorded at 60°C with suppression of the HOD signal by presaturation. 13 C-spectra were recorded with full proton decoupling. Two dimensional double-quantum filtered COSY, TOCSY, 1 and NOESY experiments were performed using pulse sequences supplied by Varian. TOCSY spectra were run with a spin-lock field of about 10 kHz and a mixing time of 80 ms; NOESY spectra were run with a mixing time of 100 ms. All chemical shifts were relative to internal or external trimethylsilylpropionic acid.
Effects of the Sulfated Polysaccharides as Inducers of the Acrosome Reaction-The effects of the various sulfated polysaccharides as inducers of the acrosome reaction in conspecific and heterospecific spermatozoa were assayed essentially as described (6). Sperms of each species were prepared by intracelomic injection of 0.5 M KCl (12). The reaction mixtures contain ϳ10 7 sperms and ϳ100 g/ml (as galactose or fucose content) various sulfated polysaccharides in 200 l of filtered sea water. After incubation at 20°C for 3 min, an equal volume of cold 6% glutaraldehyde in sea water was added, and the acrosome reaction was monitored by direct counting, using transmission electron microscopy to identify the distinct morphological changes characteristic of the acrosome reaction, of at least 100 sperms for each point.

RESULTS AND DISCUSSION
Sulfated L-Fucans or Sulfated L-Galactans Are Found in the Egg Jelly Coat of Sea Urchins-Sulfated polysaccharides were purified from the jelly coat of three species of sea urchins. Purification was achieved by anion exchange chromatography on a DEAE-cellulose column (Fig. 1, A-C). A peak rich in sialic acid was completely eluted by ϳ1.0 M NaCl. A second peak, eluted at higher salt concentration, corresponds to the sulfated polysaccharide. The purity of these polysaccharides was con- firmed by gel filtration chromatography on Sephacryl S-400 (not shown).
We found that egg jelly coats of the species Lytechinus variegatus and Arbacia lixula contain sulfated L-fucans as has also been reported for other sea urchin species (6, 7), but surprisingly, the jelly coat of the sea urchin Echinometra lucunter contains a sulfated L-galactan (Table I). 2 These sulfated polysaccharides have a similar molecular mass, as indicated by their migration on polyacrylamide gel electrophoresis (not shown). The variation in electrophoretic mobilities observed among sulfated polysaccharides from different species of sea urchins (Fig. 1D) may be accounted for in part by slight differences in the sulfate:sugar molar ratio, but this variation may also reflect other important structural differences (24), as discussed below.
The Egg Jelly of the Sea Urchin E. lucunter Contains a Homopolymer of 2-Sulfated, 3-Linked ␣-L-Galactan-Methylation of the native sulfated L-galactan from E. lucunter yields 4,6-di-O-methylgalactose, whereas 2,4,6-tri-O-methylgalactose is the predominant methyl ether derivative from the desulfated L-galactan (Table II). This indicates a linear polysaccharide composed of 3-linked and 2-sulfated L-galactopyranoside residues, which structure was confirmed by the 1 H and 13 C NMR spectra. The 1 H spectrum (Fig. 2, A-C) was assigned by means of COSY and TOCSY spectra (not shown). On desulfation of the L-galactan, alterations in chemical shifts of proton signals are consistent with 2-sulfation: Ϫ0.57 ppm for H2, Ϫ0.09 for H3, and Ϫ0.03 for H4 (Fig. 2, A-C), confirming that C2 is the position of sulfation. The 13 C spectrum of the native polysaccharide (Fig. 2D) contains 6 resonances (indicating that the sample is a homopolymer): an anomeric signal at 97.16 ppm, unsubstituted C-6 at 63.8 ppm, glycosylated or sulfated carbons at 76.1 and 75.9 ppm, and two other ring carbons at 73.9 and 69.1 ppm. In the 13 C spectrum of the desulfated L-galactan (Fig. 2E), a single substituted carbon at 77.25 ppm is observed, as expected. Tentative assignments indicated in Fig. 2E are based on comparison with literature values (25).
The Egg Jelly of the Sea Urchins A. lixula and L. variegatus Contain Linear Sulfated ␣-L-Fucans with Regular Tetrasaccharide Repeating Units-The structure of the sulfated ␣-L-fucan from the egg jelly coat of L. variegatus was described in our previous study (24). This sulfated ␣-L-fucan is essentially a linear polymer, composed of a regular repeating sequence of 4 residues, as follows: The structure of the sulfated ␣-L-fucan from the sea urchin A. lixina has not previously been investigated. As for the polysaccharide from L. variegatus, we observed that the high-field 1 H NMR spectrum of the sulfated L-fucan from A. lixula contains four anomeric residues in equal proportions by integration (Fig.  3A). Double quantum filtered COSY and TOCSY spectra (not shown) confirm that the four anomeric residues correspond to four spin systems, each consistent with ␣-fucose. The spin systems can be traced, giving the values of Table III. Strong downfield shifts of H2 of residues A and B relative to H2 of C and D indicate that two of the residues are sulfated at C2. Thus, the sulfated ␣-L-fucan from A. lixina also has a tetrasaccharide repeat unit but consists of two residues sulfated at the O-2 position and two unsulfated residues.
The order of the four residues was determined by the NOESY spectrum (Fig. 4). As in the NOESY spectra of the other fucan from the sea urchin L. variegatus (24) or from the sea cucumber (24,26), cross-peaks can be seen from H1 of each residue to protons on one and only one of the other residues (besides, of course, nOes to other protons in the same residue). This suggests that the four residues make up a linear tetrasaccharide repeating unit, as in the case for the other echinoderm fucans we have studied (24,26). The pattern of the nOes is, however, different. In the fucans from the sea urchin L. variegatus (24) and from a sea cucumber (26) studied previously, the major inter-residue nOes were between H1 and H3 of the next residue. In the sulfated L-fucan from A. lixula, nOes can be seen from each H1 to one of the H6 signals, from H1 of residues A and C to the envelope containing signals from H4 of residues A and B (4.01 ppm), and from H1 of residues B and D to the overlapping H4 signals of residues C and D (3.92 ppm) (Fig. 4). This pattern of nOes is indicative of 134 linkages (27) rather than the 133 linkages previously seen in other similar compounds (24,26) and indicates the repeating 134-linked structure -B-D-C-A-in which two consecutive 2-O-sulfated residues are followed by two unsulfated residues to give the structure shown in Fig. 5A.
Methylation analysis (Table II) confirms the occurrence of 134 linkages in the sulfated L-fucan from A. lixula; 62% of 2,3-di-O-methylfucose and 38% of 3-methylfucose were formed from the native polysaccharide. Although the proportions of the methylated derivatives are not exactly as expected, they are consistent with a polysaccharide composed of 4-linked fucose residues, half of them sulfated at the O-2 position and half unsulfated units. 3 The 1 H spectrum of the polysaccharide resulting from successive desulfation processes (Fig. 3, B and C) shows a reduc- Fucose occurs entirely in the L-enantiomeric form since this sugar is totally oxidized by L-fucose dehydrogenase.
b The enantiomeric form of L-galactose was determined by the resistance of this sugar to oxidation by D-galactose oxidase. In addition, its specific rotation of ϳϪ81°is similar to that recorded for a mutarotated solution of authentic L-galactose (20). The methodology used to characterize L-galactose is described in Refs. 20 and 23. tion in the intensity of anomeric residues at 5.2-5.3 ppm and a corresponding increase in intensity at 5.04 ppm, indicating that the fully desulfated polysaccharide is a homopolymer of identical fucose residues, as expected from our proposed structure. The 13 C spectrum of the native ␣-L-fucan from A. lixula shows 4 signals of anomeric carbons (Fig. 3D). Most of the ring carbons resonate at about 69 -71 ppm and overlap heavily. The 13 C spectrum of the desulfated polysaccharide (Fig. 3E) contains one major anomeric signal at 103.1 ppm; again, most of the ring carbons overlap heavily. The 13 C spectra corroborate our proposition that the sulfated polysaccharide from A. lixula is composed of ␣-L-fucose units with a single type of linkage and a tetrasaccharide repeat unit defined by a specific pattern of sulfation.     Purified Sulfated Polysaccharides from the Jelly Coat Induce the Acrosome Reaction in Conspecific but Not in Heterospecific Spermatozoa-Once we had isolated, purified, and characterized the fine structure of the sulfated polysaccharides from the egg jelly layer of three species of sea urchins, we were in a position to test the ability of these polysaccharides to induce the acrosome reaction in conspecific and heterospecific spermatozoa. Transmission electron microscopy can be used to differentiate sperms that have undergone the acrosome reaction from those that have not, by clear and unambiguous morphological differences, so that the extent of the acrosome reaction among a sample of sperm may be monitored by direct counting. Effectively, we observed a species-specific induction of the acrosome reaction in the sperms of the three sea urchin species (Fig. 5B). Thus, the sulfated ␣-L-galactan from the egg jelly coat of E. lucunter induces the acrosome reaction exclusively in sperm of this species (No. 2 in Fig. 5B). Even the similar sulfated ␣-L-galactan from the ascidian Herdmania monus, which is also linear but composed of 4-linked and 3-sulfated ␣-L-galactose units (28), does not induce the acrosome reaction in this species of sea urchin.
The acrosome reaction in sperms of L. variegatus is induced exclusively by the sulfated ␣-L-fucan from the egg jelly coat of this species (No. 3 in Fig. 5B). Sulfated ␣-L-fucans from A. lixula or from the sea cucumber Ludwigothurea grisea, although composed of regular repeating tetrasaccharide units but with a different sulfation pattern (compare Nos. 1, 3, and 4 in Fig. 5A), do not induce acrosome reaction in sperms of L. variegatus (No. 3 in Fig. 3B). We also observed a conspecific but not heterospecific induction of the acrosome reaction by the sulfated ␣-L-fucan from A. lixula (No. 1 in Fig. 5B).
Further purification of the sulfated ␣-L-fucan from L. variegatus on a gel filtration column yields a single and narrow peak (Fig. 6), which induces the acrosome reaction in sperms of this species of sea urchin at the same extent reported in Fig. 5B. This experiment indicates that a sulfated polysaccharide itself, and not a contaminant present in the preparation, is in fact the inducer of acrosome reaction. CONCLUSION Specificity in fertilization is the result of a series of interactions between molecules located on the surfaces of the egg and of the sperm. This is especially relevant in free-spawning or-ganisms and constitutes a barrier to prevent interspecific crosses, and consequently the formation of hybrids. It may be that the induction of acrosome reaction by sulfated polysaccharides from the egg jelly coat is the first level of recognition during fertilization in sea urchin species while the more specific interaction of egg receptor with sperm serves as a second level of recognition (4).
Our results indicate that the acrosome reaction in sea urchin is mediated by sulfated polysaccharide and in fact is regulated by the structure of the saccharide chain and its sulfation pattern. This constitutes an unusually clear-cut example of a biological event regulated by sulfated polysaccharides. Variations in the structure of these polymers among the species of sea urchins may represent one of the barriers which prevent interspecific crosses.
Approximately 900 species of sea urchins have been described (29). To act as a recognition molecule for the acrosome reaction in such a large number of species, the sulfated polysaccharides from the egg jelly must have a wide capacity for structural variation. In fact, an array of sulfated esters in the eight possible sulfation positions of the tetrasaccharide repeat units of a linear sulfated L-fucan (as those reported in Fig. 5) allows 255 combinations. Different positions of the glycosidic linkages (2-, 3-, or 4-linked), as we reported for the sulfated L-fucans from the sea urchins L. variegatus and A. lixula, increase the possible combinations to 765. In addition, the presence of sulfated L-galactan in some other species, for example E. lucunter, expands the variation possibilities for sulfated polysaccharides from the egg jelly of sea urchins.
Recent studies have suggested that glycoproteins or peptides, tightly associated with the sulfated polysaccharides of the egg jelly coat, are in fact the inducers of the acrosome reaction in sea urchins (8 -11). It is unlikely that glycoproteins or peptides would resist the drastic protease digestion conditions we have used to extract sulfated polysaccharides from the egg jelly or would remain associated with the sulfated polysaccharides at the NaCl concentration we used to elute these polymers from the DEAE-cellulose column (Fig. 1A-C). But, if this is the case, and the sulfated polysaccharides are not the inducers of the acrosome reaction but the "carriers" of glycoproteins or peptides, which directly induce the process, our proposition is still valid. That is, there is a species-specific variation in the structure of the sulfated polysaccharide from the sea urchin egg jelly coat, and these polymers are either directly involved in the induction of the acrosome reaction or are specific "carriers" of acrosome reaction-inducing molecules. We believe the relative importance of these glycoproteins or peptides and of the sulfated polysaccharides as inducers of the acrosome reaction in sea urchins requires further investigation.