High-resolution localization of hyaluronic acid in the golden hamster oocyte-cumulus complex by use of a hyaluronidase-gold complex.код для вставкиСкачать
THE ANATOMICAL RECORD 228:370-382 (1990) High-Resolution Localization of Hyaluronic Acid in the Golden Hamster Oocyte-Cumulus Complex by Use of a Hyaluronidase-GoldComplex FREDERICK W.K. KAN Department of Anatomy, Faculty of Medicine, University of Montreal, Montreal, Quebec, Canada H3C 357 ABSTRACT The distribution of hyaluronic acid in the oocyte-cumulus complexes collected from the oviduct ampulla of superovulated hamsters was revealed by use of hyaluronidase coupled to colloidal gold. On thin sections of Lowicrylembedded oocyte-cumulus complexes, gold particles were associated specifically with interconnecting fibrillar materials that make up the cumulus matrix. Inside the cumulus cells, gold particles were found over the cisternal membrane of the rough endoplasmic reticulum, in the contents of lysosomes and multivesicular bodies, and over Golgi vesicles of some cumulus cells. A high concentration of gold labeling was observed over the peripheral condensed chromatin and perinucleolar components in the nucleus. The cell surface of the cumulus cells also appeared to be labeled. Gold particles, however, were absent over the mitochondria and lipid vacuoles. In the oocytes, labeling was found t o be associated mainly with rough endoplasmic reticulum and arrays of lamellar structures; cortical granules, mitochondria, and coated vesicles were essentially devoid of gold particles. Gold particles were also seen along the plasma membrane of the oocytes and within the perivitelline space. The zona pellucida was not labeled by hyaluronidase-gold. Different control experiments confirmed the specificity of the labeling. Digestion of thin sections with hyaluronidase prior to incubation with hyaluronidase-gold abolished the initial reaction, whereas treatment of thin sections with chondroitinase did not prevent labeling of oocyte-cumulus complexes by hyaluronidase-gold. Although the function of hyaluronic acid in the oocyte-cumulus complex a t the time of ovulation and fertilization is not known, the high concentration of this particular compound in the cumulus matrix and the cumulus cells and its specific locations in the perivitelline space and in the superovulated oocytes implicate the significance of its presence and warrant future investigations. In order to fertilize the egg, mammalian spermatozoa have to penetrate through two extracellular egg investments, the cumulus matrix and the zona pellucida. Although numerous studies have been done on the morphological, biochemical, and immunochemical properties of the zona pellucida (see reviews by Yanagimachi, 1981; Dunbar, 1983a,b; Dunbar and Wolgemuth, 1984; Wassarman et al., 19861, those of the cumulus matrix have been less characterized. Moreover, the distribution of hyaluronic acid within the egg investments and the precise intracellular location of this macromolecular component in the oocyte-cumulus complex are not known. Biochemically, hyaluronic acid has been shown to be the major component of the bovine cumulus matrix (Ball et al., 1982). In cattle and mice, the synthesis and deposition of hyaluronic acid in the oocyte-cumulus complex appear to be stimulated by follicle-stimulating hormone (Eppig, 1979; Ball et al., 1982), the action of which is believed to be mediated by cyclic adenosine monophosphate (AMP) (Eppig, 1979; Ball et al., 1982). Ultrastructurally, a recent report on the golden hamster cumulus matrix shows that this 0 1990 WILEY-LISS, INC egg investment consists of a network of fibrillar units with a considerable amount of interconnection (Yudin et al., 1988). Cytochemically, however, the distribution of hyaluronic acid in the oocyte-cumulus complex has not been thoroughly described, due mainly to the unavailability of a suitable marker for its detection a t the electron microscopic level. Previous studies on the localization of hyaluronic acid have relied on cationic dyes, such as Alcian blue, in conjunction with enzymatic digestion of tissue sections (Fisher and Solursh, 1977; Derby, 1978; Derby and Pintar, 1978; Markwald et al., 1978; Pintar, 1978; Singley and Solursh, 1980; Delgado and Zoller, 1987). However, these histochemical staining techniques are used mainly at the light microscopic level. Although ruthenium red has been utilized for the ultrastructural detection of hyaluronic Received July 11, 1989; accepted March 30, 1990. Address reprint requests to Dr. F.W.K. Kan, Department of Anatomy, University of Montreal, C.P. 6128, Succ. A, Montreal, Quebec, Canada H3C 357. 371 LOCALIZATION OF HYALURONIC ACID acid in the hamster oocyte-cumulus complex (Talbot, 1984), results reported in this study are scanty. Attempts have been made to employ the hyaluronic acid binding region and monoclonal antibodies against this component in cartilage proteoglycans as probes for the ultrastructural localization of hyaluronic acid in brain tissue sections (Ripellino et al., 1985, 1988, 1989). These attempts have yielded results of specificity. Radioautography has also been used to examine the preovulatory synthesis of proteoglycans by human oocytes and cumulus cells and their secretion into the extracellular matrices of the oocyte-cumulus complex (Tesarik and Kopecny, 1986). This approach, too, suffers the limitation of resolution and cannot specifically identify the labeled macromolecular component to be hyaluronic acid. Recently, hyaluronidase (HUD) adsorbed to colloidal gold particles has been used with success to detect hexuronic acid-containing macromolecules on thin sections of a variety of rat tissues (Londono and Bendayan, 1988). This HUD-gold method, with minor modification, was used in the present study for the high-resolution localization of hyaluronic acid in the superovulated hamster oocyte-cumulus complex. I report here the specific association of hyaluronic acid with interconnecting fibrillar materials of the cumulus matrix and the subcellular distribution of these macromolecular component in the cumulus cells and oocytes. In addition, the failure of labeling of the zona pellucida by HUD-gold confirms previous histochemical (Delgado and Zoller, 1987) and biochemical (Newport and Carrol, 1976; Dunbar et al., 1980) findings of absence of hyaluronic acid in this particular egg investment. MATERIALS AND METHODS Collection of Oocytes Sexually mature female golden hamsters (Mesocricetus auratus, 8 weeks old) were obtained from Charles Rivers Laboratories (St-Constant, Quebec, Canada). They were stimulated to superovulate by a n intraperitoneal (i.p.) injection of 25 IU pregnant mare’s serum gonadotropin (PMSG, Equinex; Ayerst, Montreal, Quebec, Canada) followed, 48 hours later by a n i.p. injection of 25 IU human chorionic gonadotropin (hCG, APL; Ayerst). The animals were killed by cervical dislocation 17 hours after injection with hCG. Their ventral abdominal wall was immediately cut open, and the oviducts were removed and placed in the well of a porcelain dish containing Dulbecco’s phosphate-buffered saline (PBS-D, pH 7.4, Gibco Inc., Burlington, Ontario, Canada). To collect oocyte-cumulus complexes, the ampullary portion of the oviduct was identified and torn open with fine steel tweezers under a dissecting microscope. The oocyte-cumulus complexes collected were fixed for 2 hours at room temperature by immersion in 2.5% glutaraldehyde in 0.1 M cacodylate buffer (pH 7.4). At the end of the fixation, the specimens were washed three times in 0.1 M cacodylate buffer and kept overnight at 4°C in the same buffer. For preparation of thin sections, the oocyte-cumulus complexes were dehydrated in a series of graded methanols, infiltrated, and embedded in Lowicryl K4M according to routine procedure. Superovulated oocytes were first located in 1 pm-thick sections of Lowicrylembedded specimens by light microscopy. Thin sections (pale gold interference color) were then cut with glass knives on a LKB ultramicrotome and mounted on 200 mesh nickel grids having a carbon-coated parlodion film and processed for HUD-gold labeling. Preparation of Colloidal Gold and Hyaluronidase VI-Gold Complex Colloidal gold particles of 15 nm diameter were prepared by the sodium citrate method as described by Frens (1973). Preparation of the HUD-gold complex was performed essentially as recently reported (Londono and Bendayan, 1988). The major difference was that HUD type VI (from bovine testes, EC 188.8.131.52, purchased from Sigma Chemical Co., St. Louis, MO) was used in the present study instead of HUD type IV-S as previously used by the above authors. Preliminary experiments indicated that a higher labeling intensity was obtained when HUD VI was used in place of type IV-S in preparation of the gold probe. In brief, for preparation of HUD VI-gold, 50 pg of HUD VI were dissolved in 0.1 ml of twice-distilled water, to which 10 ml of colloidal gold sol previously adjusted to pH 7.5 was added, and the two solutions were gently mixed. Then 0.1 ml of 1% polyethylene glycol (MW 20,000) was added and mixed well with the HUD VI-gold complex. The complex was centrifuged for 30 minutes at 26,500 rpm. The resulting red sediment was resuspended in 1ml of 0.05 M acetate buffer, pH 5, containing 0.02% polyethylene glycol. The HUD VI-gold complex was stored at 4°C in a capped, round-bottom polypropylene tube until use. Freshly prepared HUD VI-gold complex was normally used for incubation of sections within 3 days after preparation. Cytochemical Labeling Thin sections of Lowicryl-embedded oocyte-cumulus complexes were incubated for 5 minutes on a drop of 0.05 M acetate buffer, pH 5, followed by 30 minutes incubation at room temperature on a drop of the HUD VI-gold complex diluted 1:lO with the same buffer. After labeling with the HUD VI-gold complex, the sections were washed with acetate buffer, rinsed with twice-distilled water, and dried on a filter paper. The sections were stained with uranyl acetate and lead citrate for examination in a Philips 300 electron microscope operated at 80 KV. Cytochernical Controls The HUD VI-colloidal gold complex is a one-step post-embedding enzyme-gold labeling technique. Although the exact mechanism of the action of the enzyme-gold technique remains to be elucidated, it appears that upon incubation of the tissue section with a specific enzyme-gold complex, the enzyme molecules at the surface of the gold particle interact with their specific substrate molecules (in this case the hyaluronic acid) exposed at the surface of the section and that the biological activity of many enzyme-gold complexes previously studied was found to be retained (for reviews see Bendayan, 1984,1989). As for any other cytochemical technique, the specificity of the HUD VI-gold labeling was assessed through several control experiments listed below: 1. Lowicryl-embedded thin sections of oocytecumulus complexes were digested with the free enzyme 372 F.W.K. KAN (0.5 mg/ml of HUD VI) in the presence of protease inhibitors (Talbot, 1984) for 6 hours a t 37°C prior to labeling with the HUD VI-gold complex. In this control experiment, the nonlabeled enzyme (HUD VI) should extract the substrate molecules (hyaluronic acid) so that the subsequent enzyme-gold complex should yield little or no specific labeling. Protease inhibitors were added to prevent any activity that might occur due to possible protease contamination of the HUD VI. 2. Thin sections of oocyte-cumulus complexes were incubated with the HUD VI-gold complex, to which the specific substrate (1 mg/ml of hyaluronic acid, from bovine vitreous humor, Sigma Chemical Co.) was added. The free substrate should compete with the hyaluronic acid exposed on the surface of the section for binding t o the HUD VI-gold complex. 3. Thin sections were incubated with HUD VI-gold in presence of excess amount of the free enzyme (1 mg/ml of HUD VI). In this case, the free enzyme should compete with the HUD VI-gold complex for binding to the hyaluronic acid that is exposed on the tissue section. 4. Because testicular hyaluronidase degrades not only hyaluronic acid but also chondroitin and chondroitin sulfates as well (Leppi and Stoward, 1965; Zergibe, 1962), in order to ensure that HUD VI-gold complex detects mainly hyaluronic acid and not other glycosaminoglycans in the oocyte-cumulus complex, some thin sections were digested with chondroitinase ABC (1 mg/ml or 0.55 unit/ml, Sigma Chemical Co.) for 5 hours a t 37°C prior to labeling with HUD VI-gold complex as described above. with the cumulus matrix and was unlabeled by HUD VI-gold, indicating the absence of hyaluronic acid in this particular egg investment (Fig. 3). The junction between the inner limit of the cumulus matrix and the outer surface of the zona pellucida is illustrated a t high magnification in Figure 3, in which the heavily labeled cumulus matrix is shown in contrast to the zona pellucida, which was almost completely devoid of labeling by gold particles. The inner border of the unlabeled zona pellucida is sharply interrupted by the resurgence of gold labeling in the perivitelline space, which is a discrete extracellular compartment located between the inner surface of the zona pellucida and the plasma membrane of the oocyte (Fig. 3, lower inset). Gold labeling in the perivitelline space appears to be associated with cloudy materials that possess a higher staining density than that of the zona pellucida. Distribution of Hyaluronic Acid in the Cumulus Cells Cumulus cells in the cumulus matrix were found in clusters but more often in solitude. Occasionally, erythrocytes were found in apposition to the cumulus cells. HUD VI-gold labeling of thin sections of Lowicryl-embedded oocyte-cumulus complex revealed intense labeling of the cell surface of the cumulus cells (Figs. 4). Inside the cumulus cells, particles associated with ribosome-bound membrane of the rough endoplasmic reticulum (rER) were found (Fig. 4).Gold particles were predominant over the cisternal membrane of the rER, with occasional particles inside the lumen of the cisternae. Colloidal gold labeling was also observed over lysosome-like structures, multivesicular bodies, and RESULTS Golgi apparatus (Fig. 5). Mitochondria and lipid vacuLocalization of Hyaluronic Acid in the Cumulus Matrix oles in the cumulus cells were not labeled by HUD Lowicryl-embedded thin sections of oocyte-cumulus VI-gold (Figs. 1 , 2 , 4 , 5 ) .An unusually high concentracomplex, after labeling with HUD VI-gold, revealed tion of gold labeling was found in nuclear compartintense labeling of the cumulus matrix by gold parti- ments, in particular over the condensed peripheral cles (Fig. 1). The opacity of the gold particles facilitates chromatin (heterochromatin) and condensed perinucledelineation of the inner (Fig. 1) and outer (Fig. 2) olar chromatin (Figs. 1,2,4, 5). The diffused chromatin boundaries of this extracellular matrix. At low magni- (euchromatin) and the fibrillar region of the nucleolus fication, its structural organization resembles that of did not appear to be labeled. an “expanded spider-web’’made up of a network of interconnections of fibrillar materials woven into a Distribution of Hyaluronic Acid in the Superovulated Oocyte Thin sections of Lowicryl-embedded oocyte-cumulus sphere in which cumulus cells are suspended (not shown). The intense labeling of the fibrillar materials complex showed that the oocyte is surrounded by an by gold particles reflects the presence of a high concen- uniform thickness of zona pellucida, unlabeled by HUD tration of hyaluronic acid in the cumulus matrix. Indi- VI-gold, which in turn is enclosed by the heavily lavidual or clusters of cumulus cells are frequently seen beled cumulus matrix (Fig. 1). Externally, in addition a t the outer boundary of the cumulus matrix, which is to the labeling of the perivitelline space by HUD VIwell delineated by gold particles. Occasionally a cumu- gold as described earlier, gold particles were seen aslus cell is seen pressing against the matrix, creating an sociated with microvilli and along the plasma memoutward protrusion of the fibrillar network (Fig. 2). In brane (Fig. 3). Inside the oocyte, many cortical this part of the cumulus matrix, the stretched outer granules were found immediately below the plasma limit outlined by gold particles remained continuous membrane. These granules were not labeled (Fig. 3). and intact, showing no sign of disruption despite the Mitochondria, frequently appearing in clusters, were force exerted by the cumulus cell. This stretchable fea- also devoid of HUD VI-gold labeling (Fig. 3). The most ture of the cumulus matrix indicates the fibroelastic numerous and evident organelle found in the oocyte property of the interwoven strands, which appear to be proper were parallel arrays or swirls of cytoplasmic capable of sustaining weights and pressure exerted by lamellae (Figs. 1,6).Transverse section of these lamellar structures showed their concentric orientation the cells within the cumulus matrix. Labeling of the cumulus matrix by HUD VI-gold around clusters of mitochondria (Figs. 1 , 3 , 6 ) ,whereas stopped abruptly a t the inner boundary of the cumulus tangential section showed their appearance in crystalmatrix, which bordered along the outer surface of the line pattern (Fig. 6b). These cytoplasmic lamellae were zona pellucida (Figs. 1,3). The matrix of the zona pel- strongly labeled by HUD VI-gold (Fig. 6). Gold partilucida exhibited a more compact texture as compared cles, however, appeared not to be directly over these LOCALIZATION OF HYALURONIC ACID Fig. 1. Electron photomicrograph of an ultra-thin Lowicryl section of superovulated hamster oocyte stained for hyaluronic acid by use of the hyaluronidase VI-gold complex. The cumulus matrix (CM) delineated by arrowheads a t the left of CM is intensely labeled with gold particles. The zona pellucida (ZP) does not appear to be labeled. Immediately outside the oocyte, labeling is found over the microvilli 373 (arrows) and along the plasma membrane. Labeling is also localized to the perivitelline space (PV). Inside the oocyte, gold particles are associated mainly with the lamellar structures (smaller arrows), whereas cortical granules (c) and mitochondria (m) are unlabeled. CC, cumulus cell nucleus; Li, lipid droplets. X 8,000. 374 F.W.K. KAN Fig. 2.A cumulus cell (CC) is seen protruding outward from the cumulus network creating a bulge in the cumulus matrix (CM). X 9,500. particular structures; rather, they labeled fluffy materials t h a t are associated with the borders of the cytoplasmic lamellae (Fig. 6). This is evident in tangential section of lamellae in oocytes in which gold particles were not found over the crystalline lattice but were localized to the borders of the lamellae (Fig. 6b). Isolated cisternae of rER in the oocyte were also labeled (results not shown). Controls The specificity of the postembedding labeling over the cumulus matrix, in cumulus cells and oocytes, were confirmed by several controls. Addition of a n excess of the free enzyme (HUD VI, 1 mg/ml) to the HUD VIgold complex or addition of the specific substrate (hyaluronic acid, 1 mg/ml) prevented labeling of the oocyte-cumulus complex by gold particles (Figs. 7a,b). Predigestion of the Lowicryl sections of oocyte-cumulus complexes with free HUD VI (1mg/ml) in the presence of protease inhibitors (6 hours at 37°C) greatly reduced the labeling previously seen over the cumulus matrix (Fig. 7c). Treatment of thin sections of oocyte-cumulus complexes with chondroitinase ABC did not prevent label- 375 LOCALIZATION OF HYALURONIC ACID Fig. 3.High magnification of the oocyte-cumulus complex showing the transition from the cumulus matrix to the oocyte proper. The cumulus matrix (CM) is heavily labeled with gold particles. The labeling stops abruptly a t the outer surface of the zona pellucida (ZP), which, characterized by its compact feature, is not labeled by gold particles. The labeling reappears in the perivitelline space (PV) and is seen over the microvilli (arrowheads) and on the plasma membrane of the oocyte. Cortical granules (arrows) and mitochondria (m) are not labeled. However, intense labeling by gold particles is associated with the cytoplasmic lamellae. x 13,200. Upper inset: The intensely labeled cumulus matrix (CM) is shown here in contrast to the unlabeled zona pellucida (ZP). x 54,000. Lower inset: High magnification of a superovulated hamster oocyte showing a region of the perivitelline space (PV) labeled strongly by HUD VI-gold complex. The zona pellucida (ZP), cortical granules (arrowheads), mitochondria (m) and the contents of two coated vesicles (arrows) are devoid of labeling. x 21,300. ing of the cumulus matrix, cumulus cells, or oocytes by HUD VI-gold (Fig. 8). However, inside certain cumulus cells, there appeared to be a lessened accumulation of gold particles over the organelles. These results indicated that the reaction was due mainly to hyaluronic acid, and not to chondroitin or chondroitin sulfates, although a relatively small amount of the latter cornponents may also be present. DISCUSSION In this study, HUD VI coupled to gold particles was used for high-resolution mapping of hyaluronic acid in the golden hamster oocyte-cumulus complex. Four major findings resulted from this study: The first is that a high concentration of hyaluronic acid was detected in the cumulus matrix by use of the HUD VI-gold com- 376 F.W.K. KAN Fig. 4. Electron micrograph of a cumulus cell. Gold particles are localized over the plasma membrane (arrows). Inside the cell, labeling is seen along the cisternal membrane of the rough endoplasmic reticulum or in association with ribosomes (arrowheads). In the nucleus (N), labeling is preferentially associated with the peripheral chromatin and patches of condensed chromatin throughout the nucleoplasm. The perinucleolar condensed chromatin is also strongly labeled. Nu, nucleolus; CM, cumulus matrix. x 19,200. plex. Colloidal gold labeling reveals the structural cumulus matrix. Recent experiments performed in our organization of the cumulus matrix in the form of a laboratory with Limus flavus agglutinin (specific for network of interconnecting fibrillar structures, as de- sialic acid), Ricinus communis I (specific for D-galacscribed recently (Yudin et al., 1988). Similar to what tose), and wheat-germ agglutinin (specific for sialic has been described in the cumulus matrix of the mouse acid/N-acetyl-D-glucosamine) also showed that these (Eppig, 1979) and in cattle (Ball, 1982), hyaluronic acid lectins failed to label the hamster cumulus matrix is probably also the major component of glycosamino- (Roux and Kan, 1990). Collectively, the present findglycans in the golden hamster cumulus matrix. This is ings as well as those from our previous studies indicate evidenced by the absence or great reduction of gold that the hamster cumulus matrix is poor in glycoprolabeling in the cumulus matrix in all control experi- teins in general but contains copious hyaluronic acid. ments and by the fact that pretreatment of oocyte-cuAlthough biochemical studies have shown that mammulus complexes with chondroitinase ABC did not pre- malian oocyte-cumulus complexes are capable of synvent their labeling by HUD VI-gold. Previous studies thesizing hyaluronic acid in vitro (Eppig, 1979; Ball, in our laboratory using a monoclonal antibody against 1982) and a recent radioautographic study has sugan oviductal glycoprotein of high molecular weight gested that the cumulus cells and the oocyte in human (Kan et al., 1988,1989a) and Helix pomatia lectin (spe- are the cellular sources that supply glycosaminoglycific for N-acetyl-D-galactosamine residues) (Kan et cans to the cumulus matrix (Tesarik and Kopecny, al., 1989b) showed, respectively, the absence of the gly- 1986), the exact pathway of synthesis and secretion of coprotein and the corresponding sugar residues in the hyaluronic acid and other glycosaminoglycans is not LOCALIZATION O F HYALURONIC ACID 377 Fig. 5. Portion of a cumulus cell showing the vicinity of a Golgi complex (Gol). Multivesicular body (mvb) and lysosome-like structures (arrowheads) are strongly reactive to hyaluronidase VI-gold labeling. The Golgi apparatus (Gol) and its peripheral vesicles also show strong reactions. Gold particles in the nucleus (N) are localized over peripheral heterochromatin and condensed chromatin. m, mitochondria. X 29,000. Fig. 6. Transverse (a) and tangential (b) sections of lamellae in cytoplasm of a superovulated hamster oocyte. a: Shows the concentric orientation of the lamellae labeled by gold particles on their sides; b The peripheral labeling pattern is evident in tangentially cut section, which shows the interior crystalline lattice (asterisk) unlabeled by hyaluronidase VI-gold. Gold particles are localized to the peripheries of the lamellae (arrowheads). x 19,000. clear. Recently, a study with the HUD-gold technique (London0 and Bendayan, 1988) has localized the labeling to the rER membrane, plasma membrane, nuclei, and basement membrane components in the pancreas, duodenum, and kidney. The Golgi apparatus and its associated vesicles of certain cell types were also reported to be labeled. Similarly, in the present study, hyaluronic acid was also localized to the aforementioned structures and compartments. Presumably, the synthesis of hyaluronic acid in the cumulus cells takes place in the rER membrane with completion of its elongation in the Golgi apparatus. There is no firm evidence to date that hyaluronic acid is covalently linked to protein (Chakrabarti and Park, 1980; Roden, 1980); therefore, in the present study, it could not be deter- mined whether the labeling detected on the cell surface represents membrane proteins that carry hyaluronic acid or if it is due merely to the general electrostatic binding nature of glycosaminoglycans (Lindahl and Hook, 1978). In this study, hyaluronic acid was detected in lysosome-like structures and multivesiclular bodies, with more intense labeling over the latter. Internalization of cell-associated proteoglycans by receptor-mediated endocytosis has been reported (Fransson, 1987). Internalized proteoglycans are believed to appear first in endosomes (Hassel et al., 1986), which then divert the products to lysosomes, the Golgi apparatus, or the nucleus (Farquhar, 1985; Ishihara et al., 1987). The products may be recycled back to the cell surface via the Golgi complex (Fransson, 1987). It ap- 378 F.W.K. KAN Fig. 7. Control Lowicryl sections. a: Incubation of thin section with hyaluronidase VI-gold in the presence of excess enzyme (hyaluronidase, 1 mgiml). The labelings previously seen over the cumulus matrix (CM), cytoplasmic and nuclear components of the cumulus cell (CC), as well as the oocyte are abolished. P V perivitelline space. x 9,500. b: The lamellae (arrows), the perivitelline space (PV),.and the microvilli (arrowheads) are completely devoid of gold particles when the specific substrate (hyaluronic acid, 1 mgiml) was added to the incubation medium. x 17,500. c: Digestion of the thin section of oocyte-cumulus complex with free hyaluronidase VI (0.5 mgiml, 6 hours, 37°C) in the presence of protease inhibitors prior to labeling with hyaluronidase VI-gold also substantially reduces the labeling previously seen over the cumulus matrix (CM). ZP, zona pellucida. X 17,500. pears that hyaluronic acid associated with the plasma membrane of the cumulus cells could also be eventually internalized and degraded through the lysosomal system in a manner similar to that described above. In this case, free hyaluronic acid may be shed into the extracellular space and incorporated into the cumulus matrix or taken up by receptor-mediated endocytosis (for a review, see Fransson, 1987). The receptor for hy- LOCALIZATION OF HYALURONIC ACID 379 Fig. 8. Prior treatment of the oocyte-cumulus complexes with chondroitinase ABC does not affect labeling of the cumulus matrix (CM) by hyaluronidase VI-gold. a: Cytoplasmic lamellae (arrows) are intensely labeled; b: Condensed chromatin in the nucleus of a cumulus cell also are intensely labeled. ZP, zona pellucida. (a) x 12,000, (b) x 9,500. aluronate has been recently identified in adult Syrian hamsters (Tarone et al., 1984; Underhill et al., 1987) and localized to the basolateral surfaces of bronchial and bronchiolar epithelium as well as to the surfaces of pulmonary macrophages (Green et al., 1988). Alternatively, the high concentration of gold particles detected in the multivesicular bodies of the cumulus cells could be the result of post-translation regulation of produc- 380 F.W.K. KAN tion and secretion of hyaluronic acid by intracellular degradative mechanisms as suggested by several authors (Smith, 1979; Bienkowski, 1983; Nanci et al., 1987). Another interesting finding was the detection of hyaluronic acid in the nucleus, an observation corroborated with results obtained in other tissues using a similar HUD-gold complex (London0 and Bendayan, 1988). Several biochemical studies have indicated the involvement of glycosaminoglycans in nuclear functions (Stein et al., 1975; Bhavanandan and Davidson, 1975; for a review, see Lindahl and Hook, 1987). In particular, heparin has been shown to induce dispersion of chromatin structure (Arnold et al., 1972; Cook and Aikawa, 1973; Smith and Cook, 1977) and to stimulate the DNA template characteristic of chromatin and nuclei during DNA synthesis (Kraemer and Coffey, 1970). Although hyaluronic acid has been identified as the major component of nuclear glycosaminoglycans in isolated rat liver nuclei (Furukawa and Terayama, 1977), to date its function in the nucleus is still unclear. Most recently, immunocytochemical studies using antibodies to protein epitopes also demonstrated the presence of hyaluronic acid in the cytoplasm and nuclei of certain brain cells during brain development (Ripellino et al., 1989). The localization of hyaluronic acid to the peripheral condensed chromatin and to the condensed perinucleolar chromatin in the cumulus cells suggests that this macromolecular component could be involved in transcriptional activities. The exact role(s) of hyaluronic acid in the cumulus cells remains to be elucidated. The third major finding was the localization of hyaluronic acid associated with the superovulated oocyte. As mentioned earlier, the zona pellucida was not labeled by HUD VI-gold, whereas labeling was detected in the perivitelline space. A previous report on hamster oocyte-cumulus complex using ruthenium red for staining hyaluronic acid claimed to have detected reaction products in the upper third of the zona pellucida (Talbot, 1984). However, this staining property is likely due to extensions of the cumulus matrix in the upper porous region of the zona pellucida rather than to staining of the zona pellucida itself. Such occurrence is also found in the present study (shown in Fig. l),in which labeled filamentous extensions of the cumulus matrix are seen infiltrating the upper region of the zona pellucida. Morphologically,the perivitelline space of the superovulated oocytes is occupied by filamentous materials, which are highly compact and appear to be different from those of the zona pellucida. This marked difference is best illustrated in Figure 7a,b. Previous studies in our laboratory using a monoclonal antibody (Kan et al., 1988,1989a) and lectins (Kan et al., 1989b; Roux and Kan, 1990) demonstrated that the zona pellucida of the superovulated hamster oocytes is rich in glycoproteins, which, however, are absent in the perivitelline space. In the present study, the detection of hyaluronic acid in the latter suggests that in the superovulated oocyte there exists different structural components that are unique to the zona pellucida and the perivitelline space. Glycosaminoglycans, such as hyaluronic acid, heparin, and chondroitin sulfate, are known to stimulate acrosome reaction in the hamster (Meizel and Turner, 1985) and rabbit (Lenz et al., 1983). It has been suggested that one or more glycosaminoglycans that are present in the perivitelline space may be involved in acting synergistically with the zona to induce the acrosome reaction (Gwatkin, 1989).The cytochemical detection of hyaluronic acid in the perivitelline space in the present study seems to fit well into this scheme. Another noteworthy finding is the labeling of cytoplasmic lamellae by HUD VI-gold complex inside the oocyte. These lamellar structures were first described by Weakley (1966) in the hamster. Study of the fixation properties of these lamellar structures (Weakley, 1967) and freeze-fracture studies carried out on hamster oocytes (Suzuki and Yanagimachi, 1983; Koehler et al., 1985)indicated that the lamellae are largely protein in nature. Although the cytoplasmic lamellae are believed to be nutritive to embryo development (Nilsson, 1980), the exact chemical composition and functional significance of these structures remains obscure. In the present study, labeling by HUD VI-gold was observed a t the peripheries of the lamellae. This is evidenced on tangential section (Fig. 6b), which shows the absence of gold labeling over the crystalline lattice of the tangentially cut lamellae. Due to the polyanionic nature of glycosaminoglycans (Chakrabarti and Park, 1980), the binding of hyaluronic acid to the lamellae is presumably a result of electrostatic interaction. Future ultrastructural studies aided by cytochemical probes will likely shed light on the chemical composition and function of the cytoplasmic lamellae in relation to their association with hyaluronic acid. Mammalian sperm cells have to traverse the cumulus matrix and the zona pellucida before fertilization can occur. The presence of copious hyaluronic acid in the cumulus matrix, its absence in the zona pellucida, and its reappearance in the perivitelline space collectively point to the unique nature of the existence of this macromolecular component in the extracellular compartments within the oocyte-cumulus matrix. The localization of hyaluronic acid in different intracellular compartments of the cumulus cells reveals the possible intracellular traffic of this macromolecular component. The detection of hyaluronic acid along the peripheries of cytoplasmic lamellae has added new information to these prominent but less well-characterized structures. Based on these studies, I suggest that the hyaluronic acid associated with the cumulus matrix originates from the cumulus cells and that the hyaluronic acid found in the perivitelline space comes from the oocyte. Although the significance of this distribution and the exact function of hyaluronic acid remain to be unraveled, it is hoped that the findings of this study will pave the way for future investigations into the role of hyaluronic acid in sperm-cumulus-oocyte interaction, which is so critical to successful fertilization. ACKNOWLEDGMENTS The author thanks Dr. Gilles Bleau for comments on the manuscript, Ms. Christiane Rondeau, and Cecile Venne for their expert technical assistance and Mr. Jean Leveille for photographic work. Appreciation is also extended to Johanne Chainey for clerical assistance. This investigation was supported by the Fonds de la Recherche en Sante du Quebec (87 118) and LOCALIZATION OF HYALURONIC ACID CAFIR of Universite de Montreal. The author is a scholar from the Medical Research Council of Canada. 381 gates of oviductal origin to the zona pellucida of oocytes after ovulation. Anat. Rec., 226t37-47. Kan, F.W.K., S. St.-Jacques, and G. Bleau 1988 Immunoelectron microscopic localization of an oviductal antigen in hamster zona LITERATURE CITED pellucida by use of a monoclonal antibody. J. Histochem. Cytochern., 36t1441-1447. Arnold, E.A., D.H. Yawn, D.G. Brown, R.C. Wyllie, and D.S. Coffey Kan, F.W.K., S. St.-Jacques, and G. Bleau 1989a Immunocytochemi1972 Structural alteration in isolated rat liver nuclei after recal evidence for the-transfer of a n oviductal antigen to the zona moval of template restriction by polyanions. J. Cell Biol., 53: pellucida of hamster ova after ovulation. Biol. Reprod., 40:585737-757. 598. Ball, G.D., M.E. Bellin, R.L. Ax, and N.L. First 1982 GlycosaminoKoehler, J.K., J.M. Clark, and D. Smith 1985 Freeze-fracture obserglycans in bovine cumulus-oocyte complexes: Morphology and vations on mammalian oocytes. Am. J . Anat., 174t317-329. chemistry. Mol. Cell. Endocrinol., 28~113-122. Bendayan, M. 1984 Enzyme-gold electron microscopic cytochemistry: Kramer, R.J., and D.S. Coffey 1978 The interaction of natural and synthetic polyanions with mammalian nuclei. Biochim. Biophys. A new affinity approach for the ultrastructural localization of Acta, 224.553-567, macromolecules. J. Electron Microsc. Tech., lt349-372. Lenz, R.W., M.E. Bellin, and R.L. Az 1983 Rabbit spermatozoa unBendayan, M. 1989 The enzyme-gold cytochemical approach: A redergo an acrosome reaction in the presence of glycosaminoglyview. In: Colloidal Gold: Principles, Methods, and Applications. cans. Gamete Res., 8:11-19. Vol. 2. M.A. Hayat, ed. Academic Press, San Diego, pp. 118-145. Leppi, T.J., and P.J. Stoward 1965 On the use of testicular hyaluroniBhavanandan, V.P., and E.A. Davidson 1975 Mucopolysaccharides dase for identifying acid mucins in tissue sections. J. Histochem. associated with nuclei of cultured mammalian cells. Proc. Natl. Cytochem., 13t406-407. Acad. Sci. U.S.A., 72r2032-2036. Bienkowski, R.S. 1983 Intracellular degradation of newly synthesized Lindahl, U., and H. Hook 1978 Glycosaminoglycansand their binding secretory proteins. Biochem. J., 214rl-10. to biological macromolecules. Annu. Rev. Biochem., 47t385-417. Chakrabarti, B., and J.W. Park 1980 Glycosaminoglycans: Structure Londono, I., and M. Bendayan 1988 High resolution cytochemistry of and interaction. CRC Crit. Rev. Biochem., 8:225-313. neuraminic and hexuronic acid-containing macromolecules apCook, R.T., and M. Aikawa 1973 The effects of heparin on endogenous plying the enzyme-gold approach. J. Histochem. Cytochem., 36: DNA polymerase activity of rat liver nuclei and chromatin frac1005-1013. tions. Exp. Cell Res., 78t257-270. Markwald, R.R., T.P. Fitzharris, H. Bank, and D.H. Bernanke 1978 Delgado, M.V., and L.C. Zoller 1987 A quantitative and qualitative Structural analyses on the matrical organization of glycosaminocytochemical analysis of glycosaminoglycan content in the zona glycans in developing endocardia1 cushions. Dev. Biol., 62t292pellucida of hamster ovarian follicles. Histochemistry, 87r279316. 287. Meizel, S., and K.O. Turner 1985 Glycosaminoglycans stimulate the Derby, M.A. 1978 Analysis of glycosaminoglycans within the extraacrosome reaction of previously capacitated hamster sperm. J. cellular environments encountered by migrating neural crest Cell Biol., 99t261a. cells. Dev. Biol., 66t321-336. Nanci, A,, H.C. Slavkin, and C.E. Smith 1987 Immunocytochemical Derby, M.A., and J.E. Pintar 1978 The histochemical specificity of and radioautographic evidence for secretion and intracellular Stretomyces hyaluronidase and chondroitinase ABC. Histochem. degradation of enamel proteins by ameloblasts during the matuJ., 10t529-547. ration stage of amelogenesis in rat incisors. Anat. Rec., 217:107Dunbar, B.S. 1983a Morphological, biochemical and immunochemical 123. characterization of the mammalian zonae pellucidae. In: MechaNewport, A., and J. Carroll 1976 Structure and composition of the nism and Control of Animal Fertilization. J. Hartmann, ed. Aczona pellucida of the mouse oocyte. Biochem. SOC. Trans., 45396ademic Press, New York, pp. 139-175. 897. Dunbar, B.S. 1983b Characterization of antibodies to zonae pellucidae Nilsson, B.O. 1980 Comparative ultrastructure of the yolk material in antigens and their role in fertility. In: International Congress on preimplantation stages of the hamster, mouse, and rat embryos. Reproduction Immunology. T.G. Wegmann and T.G. Gill 111, ed. Gamete Res., 3t369-377. Oxford University Press, London, pp. 505-534. Pintar, J.E. 1978 Distribution and synthesis of glycosaminoglycans Dunbar, B.S., N.J. Wardrip, and J.L. Hedrick 1980 Isolation, physiduring quail neural crest morphogenesis. Dev. Biol., 67r444-464. cochemical properties, and macromolecular composition of zona Ripellino, J.A., M. Bailo, R.U. Margolis, and R.K. Margolis 1988 pellucida from porcine oocytes. Biochemistry, 19r356-365. Light and electron microscopic studies on the localization of hyDunbar, B.S., and D.J. Wolgemuth 1984 Structure and function of the aluronic acid in developing rat cerebellum. J. Cell Biol., 106: mammalian zona pellucida, a unique extracellular matrix. In: 845-855. Modern Biology, Vol. 3. B. Satir, ed. Alan R. Liss, New York, pp. Ripellino, J.A., M.M. Klinger, R.U. Margolis, and R.K. Margolis 1985 77-1 11. The hyaluronic acid binding region as a specific probe for the Epigg, J.J. 1979 FSH stimulates hyaluronic acid synthesis by oocytelocalization of hyaluronic acid in tissue sections. J. Histochem. cumulus cell complexes from mouse preovulatory follicles. NaCytochem., 33t1060-1066. ture. 281t483-484. Farquhar~M.G.~1985Process in unraveling pathways of Golgi traffic. Ripellino, J.A., R.U. Margolis, and R.K. Margolis 1989 Immunoelectron microscopic localization of hyaluronic acid-binding region Annu. Rev. Cell Biol., 1.447-488. and link protein epitopes in brain. J . Cell Biol., 108r1899-1907. Fisher, M., and M. Solursh 1977 Glgcosaminoglycan localization and role in the maintenance of tissue spaces i n the early chick em- Roden, L. 1980 Structure and metabolism of connective tissue. In: The Biochemistry of Glycoproteins and Proteoglycans. W. Lennarz, bryo. J . Embryol. Exp. Morphol., 42t195-207. ed. Plenum Press, New York, pp. 267-371. Fransson, L.-A. 1987 Structure and function of cell-associated proteoRoux, E., and F.W.K. Kan 1990 Changes of glycoconjugate contents of glycans. Trends Biochem. Sci., 12t406-411. the zona pellucida during oviductal transit of oocytes in the Frens, G. 1973 Controlled nucleation for the regulation of particle size golden hamster: a quantitative cytochemical study. J . Histochem. in monodisperse gold solution. Nature, 241t20-22. Cytochem. (submitted) Furukawa, K., and H. Terayama 1977 Isolation and identification of Singley, C.T., and M. Solursh 1980 The use of tannic acid for the glycosaminoglycans associated with purified nuclei from rat ultrastructural visualization of hyaluronic acid. Histochemistry, liver. Biochim. Biophys. Acta, 499t278-289. 65r93-102. Green, S.J., G. Tarone, and C.B. Underhill 1988 Distribution of hySmith, C. 1979 Ameloblasts: secretory and resorptive functions. J. aluronate receptors in the adult lung. J . Cell Sci., 89t145-156. Dent. Res., 58t695-706. Gwatkin, R.B.L. 1989 Zona binding sites of the spermatozoa. In: The Smith, M.R., and R.T. Cook 1977 Urea reduces the thermal stability Mammalian Egg Coat. J. Dietl, ed. Springer-Verlag Press, New of polyanion-treated chromatin. Biochem. Biophys. Res. ComYork, pp. 61-74. mun., 74r1475-1482. Hassel, J.R., J.H. Kimura, and V.C. Hascall 1986 Proteoglycan core protein families. Annu. Rev. Biochem., 55t539-567. Stein, G.S., R.M. Roberts, J.L. Davis, W.J. Head, J.L. Stein, C.L. Ishihara, M., N.S. Fedarko, and H.E. Conrad 1987 Involvement of Thrall, J. Van Veen, and D.W. Welch 1975 Are glycoproteins and phosphatidylinositol and insulin in the coordinate regulation of glycosaminoglycans components of the eukaryotic genome? Naproteoheparin sulfate metabolism and hepatocyte growth. J. Biol. ture, 258r639-641. Chem., 262t4708-4716. Suzuki. F.. and R. Yanaeimachi 1983 Freeze-fracture observations of ovulated hamster Gcytes with their cumulus cells. Cell Tissue Kan, F.W.K., E. Roux, S.St.-Jacques, and G. Bleau 1989b DemonRes., 231.265-274. stration by lectin-gold cytochemistry of transfer of glycoconju- 382 F.W.K. KAN Talbot, P. 1984 Hyaluronidase dissolves a component in the hamster zona pellucida. J . Exp. Zool., 229t309-316. Tarone, G., R. Ferracini, G. Galetto, and P. Comoglio 1984 A cell surface integral membrane glycoprotein of 85,000 mol wt (gp85) associated with Triton X-100 insoluble cell skeleton. J. Cell Biol., 99:512-519. Tesarik, J., and V. Kopecny 1986 Late preovulatory synthesis of proteoglycans by the human oocyte and cumulus cells and their secretion into the oocyte-cumulus complex extracellular matrices. Histochemistry, 85r523-528. Underhill, C.B., S.J. Green, P.M. Cologlio, and G. Tarone 1987 The hyaluronate receptor is identical to a glycoprotein of 85,000 Mr (gp85) as shown by a monoclonal antibody that interferes with binding activity. J. Biol. Chem., 262r13142-13146. Wassarman, P.M., J.D. Bleil, H.M. Florman, J.M. Greve, R.J. Roller, and G.S. Salzmann 1986 The mouse egg’sextracellular coat Synthesis, structure and function. In: Gametogenesis and the Early Embryo. J.G. Gall, ed. Alan R. Liss, New York, pp. 371-388. Weakley, B.S. 1966 Electron microscopy of the oocyte and granulosa cells in the developing ovarian follicles of the golden hamster (Mesocricetus auratus). J. Anat., 100.503-534. Weakley, B.S. 1967 Investigations into the structure and fixation properties of cytoplasmic lamellae in the hamster oocyte. Zeits. Zellforsch., 81t91-99. Yanagimachi, R. 1981 Mechanism of fertilization in mammals. In: Fertilization and Embryonic Development in Vitro. L. Mastrianni Jr. and J.D. Biggers, eds. Plenum, New York, pp. 81-182. Yudin, A.I., G.N. Cherr, and D.F. Katz 1988 Structure of the cumulus matrix and zona pellucida in the golden hamster: a new view of sperm interaction with oocyte-associated extracellular matrices. Cell Tissue Res., 251 555-564. Zergibe, F.T. 1962 The demonstration of the individual acid mucopolysaccharides in human aortas, coronary arteries and cerebral arteries. I. The methods. J. Histochem. Cytochem., IOt441-447.