MOLECULAR REPRODUCTION A N D DEVELOPMENT 31:135-143 (1992)
Perivitelline Space of Mammalian Oocytes: Extracellular Matrix of Unfertilized Oocytes and Formation of a Cortical Granule Envelope Following Fertilization P. DANDEKAR' AND P. TALBOT2 'Department of Obstetrics, Gynecology, and Reproductive Sciences, University of California School of Medicine, San Francisco, California; 'Department of Biology, University of California, Riverside, California
ABSTRACT Extracellular matrices (ECM) present around unfertilized and fertilized mammalian oocytes were studied ultrastructurally in samples prepared in the presence of ruthenium red to facilitate stabilization of extracellular materials. Unfertilized mouse, hamster,and human oocytes have an ECM comprising granules and filaments in their perivitelline spaces (PVS).This matrix is more abundant in the human than in hamsters and mice. The granule/filament matrix appears identical to the matrix seen between cumulus and corona radiata cells following ruthenium red processing and previously shown to comprise protein and hyaluronic acid. By including ruthenium red during fixation, it is possible to demonstrate the existence of cortical granule exudate in the PVS of fertilized oocytes from hamsters, mice, and humans. Much of the cortical granule exudate is trapped in the PVS and forms a new coat around the fertilized oocyte. This material is particulate when stained with ruthenium red and appears to be uniformly dispersed around the entire oocyte surface. We refer to this new coat as the cortical granule envelope. This envelope is observed in the PVS of all developmental stages up to and including blastocysts in all three species. Following hatching of mouse and hamster blastocysts, the cortical granule envelope is no longer present. Possible functions of this envelope are discussed. Key Words: Cortical reaction, ECM, PVS, Polyspermy
INTRODUCTION The perivitelline space (PVS) of unfertilized opossum and pig oocytes contains a matrix comprising granules and filaments (Talbot and DiCarlantonio, 1984a; Kopecny et al., 1984).This matrix can be demonstrated by fixing oocytes in the presence of ruthenium red. In the opossum, the granules and filaments can be removed by trypsin and Streptomyces hyaluronidase treatment, respectively, indicating that the granules comprise protein and the hyaluronic acid filaments (Talbot and DiCarlantonio, 1984a). Following fertilization, cortical granules release their contents exocytotically into the PVS (Szollozi, 1967) in a n event that has been termed the cortical reaction (Braden et al., 1954). The cortical granule exu-
0 1992 WILEY-LISS, INC.
date is thought to block polyspermy at the level of the oolemma andlor by altering the properties of the zona pellucida (reviewed by Cherr and Ducibella, 1990). The contents of the cortical granules are periodic acid-Schiff (PAS) positive (Yanagimachi and Chang, 1961; Guraya, 1969) and bind several lectins (Cherr et al., 1988; Ducibella et al., 1988; Lee et al., 1988). After fertilization, hamster oocytes release heparin binding placental protein (HBPP), which may come from the cortical granules (Sinosich et al., 1990). In addition, mammalian cortical granules appear to contain an ovoperoxidase that has been implicated in zona hardening (Gulyas and Schmell, 1980; Schmell and Gulyas, 1980) and several proteinases. Experimental evidence suggests that a trypsin-like proteinase is released from the cortical granules and modifies the zona pellucida to block polyspermy (Barros and Yanagimachi, 1971; Gwatkin e t al., 1973; Wolf and Hamada, 1977). This may be plasminogen activator, which is secreted from fertilized mouse oocytes (Huarte et al., 1985). More recently, Moller and Wassarman (1989) have isolated a proteinase from mouse oocytes that is released a t fertilization and modifies the zona protein, ZP2, to produce ZP2f, a form that is not capable of binding sperm. The authors implicate this proteinase in blocking polyspermy by inhibiting sperm binding to ZP2 and by bringing about zona hardening. The proteinase described by Moller and Wassarman (1989) is not blocked by trypsin inhibitors and appears to be distinct from that described in earlier studies. Although the proteinases are probably of cortical granule origin, this has not yet been shown cytochemically. The only cortical granule constituent that has been isolated and characterized from mammalian oocytes is a 75 kd protein found in mouse cortical granules (Pierce et al., 1990).This protein (75p) was shown cytochemically to be localized in the cortical granules and released into the PVS and zona pellucida after fertilization. Its function is a s yet unknown.
Received August 15, 1991; accepted September 23, 1991. Address reprint requests to P. Talbot, Department of Biology, University of California, Riverside, CA 92521.
136
P. DANDEKAR AND P. TALBOT
In spite of the probable importance of the PVS matrices in fertilization, we know relatively little about them. The granuleifilament matrix of unfertilized oocytes has not been identified in any species other than pig and opossum, and the fate of the cortical granule exduate after its release into the PVS is not fully understood. Some cortical granule materials, such as the ovoperoxidase and proteinases, presumably diffuse into the zona pellucida to bring about zona hardening and to block polyspermy. However, in several preliminary studies, cortical granule exudate was observed in the PVS after the cortical reaction (Szollozi, 1967; Gordon et al., 1975; Baranska et al., 1975; Talbot and DiCarlantonio, 198413).The purposes of the present study were to determine if a granuleifilament matrix is present in the PVS of unfertilized oocytes in other mammalian species and to examine the fate of the cortical granule exudate through early developmental stages up to and including blastocyst hatching. Ruthenium red was used to visualize extracellular matrices in the PVS, and the study was conducted using oocytes from mice, hamsters, and humans.
MATERIALS AND METHODS Hamsters Mature female golden hamsters were obtained, maintained, and fed as previously described (Talbot and DiCarlantoni, 1 9 8 4 ~ )Unfertilized . oocyte-cumulus complexes (OCC)were collected in Earle’s balanced salt solution (EBSS) from the oviducts of unmated females. In vivo fertilization (IVF)was set up as described previously (Talbot and DiCarlantonio, 198413). At various times after mating, reproductive tracts were removed from females and flushed through the infundibulum with EBSS to recover fertilized oocytes and developmental stages up to and including hatched blastocysts. Mice Four- to six-week-old, random, bred CFW (Swiss Webster) mice (Charles River, Portage, MI) were superovulated and mated; unfertilized oocytes and one-cell embryos were collected. Some one-cell embryos were cultured to hatched blastocysts, as described previously (Dandekar and Glass, 1987). Briefly, unfertilized oocytes and one-cell embryos in cumulus masses were released from swollen ampullae. To free oocytes and embryos, cumulus masses were treated with 0.1% hyaberonidase for 2-5 min and washed four to six times with T, medium. To verify fertilization, embryos were checked for the presence of two pronuclei and polar bodies. Unfertilized oocytes, one-cell embryos, and blastocysts were processed with ruthenium red, as described previously (Talbot and DiCarlantonio, 1984b). Humans Human oocytes that failed to fertilize were obtained from a n in vitro fertilization program. Details of the procedures used in this program have been described (Dandekar et al., 1990). Oocytes that failed to fertilize,
triploid embryos, and abnormally developing embryos also were processed with ruthenium red (Talbot and DiCarlantonio, 198413).
Processing for Electron Microscopy The developmental stage of each sample was determined using a dissecting microscope. Each sample was then processed for transmission electron microscopy (TEM)using previously described methods that include ruthenium red in the fixatives and postglutaraldehyde washes (Talbot, 1984; Talbot and DiCarlantonio, 1 9 8 4 ~ )Blocks . were thin sectioned using glass or diamond knives on a n MT-2B ultramicrotome. Sections were picked up on copper grids and examined either without staining or following heavy metal staining in a Hitachi H-500 or a Phillips 300 TEM. RESULTS The ultrastructure of the ECM in the PVS surrounding unfertilized oocytes and a t early developmental stages, up to and including blastocysts, was examined in the mouse (Figs. 1, 5, 6, 13, 14), hamster (Figs. 2, 9, 12,15),and human (Figs. 3,4,7,8,10,11). All material was fixed in the presence of ruthenium red to help stabilize extracellular materials. Unfertilized Oocytes In all three species, the PVS surrounding unfertilized oocytes is relatively small and contains a n ECM comprising granules and filaments (Figs. 1-4). This matrix is identical to the granuleifilament matrix previously described in the extracellular spaces of the cumulus layer, corona radiata, and outer pores of the zona pellucida (Talbot and DiCarlantonio, 1984b; Dandekar et al., 1987). The granules and filaments interconnect with each other, and some appear to bind to the 00lemma (Fig. 2). The granuleifilament matrix is most abundant in the PVS surrounding human oocytes; it is relatively sparse and difficult to demonstrate in the PVS of hamsters and mice. Fertilized Oocytes In vivo-fertilized oocytes (monospermic) were obtained from mice and hamsters after natural matings. In the case of the human, only in vitro-fertilized oocytes
Fig. 1. Overview of the cortex (C), perivitelline space (PVSJ, and zona pellucida (ZPJ of a n unfertilized mouse oocyte. The PVS is small and contains a sparse ECM comprising granules (arrowheads) and filaments (not evident a t this low magnification). A similar ECM is evident on the outer surface of the zona pellucida, where it is much more abundant. x 17,300. Fig. 2. Higher magnification view of the PVS surrounding an unfertilized hamster oocyte. The granules (arrowheads) and filaments (FJ of the ECM are shown. The granules are smaller than usual hecause this oocyte was lightly trypsinized before fixation. ZP, zona pellucida; C, cortex of oocyte. ~ 5 5 , 4 0 0 . Figs. 3.4. Unfertilized human oocyte showing the cortex (CJ, PVS, and part of the zona pellucida (ZPJ. The PVS contains an ECM comprising granules (GI and filaments (F).The filaments are best seen at higher magnification (Fig. 4). Figure 3, x 17,600. Figure 4, x59,lOO.
Figs. 1-4.
138
P. DANDEKAR AND P. TALBOT
Figs. 5-9.
PERIVITELLINE SPACE MATRIX OF MAMMALIAN OOCYTES that exhibited polyspermy or abnormal cleavage were processed for electron microscopy. Following fertilization, the PVS of all species increases significantly in size (Figs. 5, 7,9). This increase in size is accompanied by loss of most cortical granules from the periphery of the oocytes. Coincident with the cortical reaction, numerous small, dense particles (cortical granule particles; CGP) appear in the PVS (Figs. 5-9). Cortical granules caught in the process of exocytosis (Figs. 6, 7) are releasing numerous small dense particles identical to those that are already dispersed (Figs. 4 , 9 ) in the PVS of fertilized oocytes. In all species, the CGP disperse apart from each other after their release and could be found evenly distributed in the PVS (Figs. 5, 9, 10, 12, 13). Following their dispersion, the CGP formed a new vestment, the cortical granule envelope (CGE), in the PVS. One abnormally developing human embryo reached the blastocyst stage; some CGP remained together in a coherent spherical mass in the PVS around this blastocyst (Fig. 11). While the majority of CGP are free in the PVS, some appear to be bound to the oolemma (Figs. 5 , 6 , 9 ) . In the mouse, some particles diffuse into the adjacent pores of the zona pellucida (Fig. 5); however, the number of particles in the zona pellucida is always small compared with that observed in the PVS. In the human and hamster, the zona pellucida facing the PVS lacks the large pores characteristic of the mouse zona, and CGP did not appear to penetrate into the zonae of these two species.
Later Stages of Development Cleavage stages, morulae, and blastocysts were collected from mice and hamsters following IVF. IVF hu-
Fig. 5. Overview of the cortex (C),PVS, and zona pellucida (ZP)of a fertilized mouse oocyte. The PVS is larger than in unfertilized oocytes and contains numerous small electron-dense particles (arrowheads) released by the cortical granules. Some of these particles have diffused into the pores of the zona pellucida. No cortical granules remain in the cortex of the oocyte in this view. x61,SOO. Fig. 6. Surface of a fertilized mouse oocyte showing a n invagination, which probably represents a cortical granule undergoing exocytosis. The electron-dense particles (arrowhead 1)in this fused cortical granule had access to ruthenium red and appear identical to the particles (arrowhead 2) of the CGE. x61,lOO. Fig. 7. Fertilized human oocyte showing exocytosis of a cortical granule (CG). Its contents appear identical to CGP (not shown) dispersed in the PVS. In the human, the CGP often appeared cohesive after their release (arrowhead). Remnants of the granulelfilament matrix seen around unfertilized oocytes were still visible around this oocyte but are not shown in this area. X65,900. Fig. 8. Fertilized human oocyte showing CGP (arrowheads) after their release into the PVS. The particles are cohesive and often remain clustered together in spheres about the size of intact cortical granules. In this oocyte, the cortical granule exudate has not yet dispersed, and the PVS has not yet increased in size. ~ 6 1 , 6 5 4 . Fig. 9. Fertilized hamster oocyte that has reached the two-cell stage. The PVS is large and contains numerous dense particles of cortical granule origin (arrowheads). Some of these particles are bound to the oolemma (0). ZP, zona pellucida. x29,OOO.
139
man oocytes that showed abnormal cleavage or fragmentation were allowed to develop in vitro, and various stages, including a blastocyst, were fixed for electron microscopy. In all stages up to and including unhatched blastocysts, CGP could be found in the PVS of humans (Figs. 10-111, hamsters (Fig. 121, and mice (Figs. 13, 14). In blastocysts which have a normal zona pellucida and do not appear close to hatching, the PVS is large and the particles are dispersed (Figs. 10, 12, 13).However, in blastocysts surrounded by a thin zona pellucida and close to hatching, the PVS is reduced in size, and the CGP often are aggregated (Fig. 14). Particles are not present around hatched mouse (not shown) or hamster (Fig. 15)blastocysts.
DISCUSSION Two main points have emerged from this study. First, the PVS of unfertilized oocytes contains a granule/ filament matrix that is indistinguishable morphologically from the ECM described previously between corona radiata cells and cumulus cells of the hamster, mouse, and human OCC (Talbot and DiCarlantonio, 198413;Dandekar et al., 1987,1988)and between cumulus cells of three hydromyine rodents (McGregor et al., 1989). Second, the cortical reaction results in the elaboration of a new coat around the fertilized mammalian oocyte. This coat is confined to the PVS and is not readily apparent in electron micrographs unless special methods, such a s ruthenium red processing, are used. This coat comprises numerous electron-dense particles that are larger than those observed in the PVS around unfertilized oocytes. This coat remains around ail early developmental stages up to and including blastocysts. However, it is not present following blastocyst hatching. We propose that this coat be referred to as the cortical granule envelope (CGE). This envelope appears to be analogous to the hyaline layer, which originates from the cortical granules of fertilized sea urchin oocytes (Hylander and Summers, 1982). The granulelfilament matrix we observed in the PVS of unfertilized oocytes is structurally similar to that surrounding opossum (Talbot and DiCarlantonio, 1984a) and pig (Kopecny et al., 1984) oocytes. This ECM may be secreted by the corona radiata cells and released at the tips of their processes, which Szollozi (1967) has shown extend through the zona pellucida. Alternatively, i t may be synthesized and secreted by the oocyte, a s Kopecny e t al. (1984) have shown for the pig. The granules of this matrix can be digested by trypsin and hence are proteinacous, while the filaments can be removed by Streptomyces hyaluronidase, which is specific for hyaluronic acid (HA) (Talbot and DiCarlantonio, 1984a; Dandekar et al., 1988). HA in the PVS might present a significant blockade to the fertilizing sperm, since HA in other systems has been shown to inhibit membrane fusion (Vollet and Roth, 1974; Kujawa and Tepperman, 1983; Orkin et al., 1985). Sperm hyaluronidase, which is released during the acrosome reaction at the zona pellucida surface, could diffuse through the zona and remove HA from the
140
P. DANDEKAR AND P. TALBOT
Figs. 10-15.
PERIVITELLINE SPACE MATRIX OF MAMMALIAN OOCYTES oolemma. Alternatively, some hyaluronidae may be retained by acrosome-reacted sperm (Brown, 1975; Morton, 1976; Harrison, 1988) and carried into the PVS to perform this function. This ECM is much more abundant and is easily demonstrated in the human than in the two rodents. Following formation of the CGE, it was not possible to observe the granule/filament matrix characteristic of unfertilized oocytes. This could be because i t was masked by the CGE or because i t was degraded during fertilization. The new extracellular coat that is present in the PVS after fertilization appears to originate from the cortical granules. This is supported by the fact that this coat is not present prior to fertilization, when cortical granules are still within the oocyte, but appears coincident with the cortical reaction. Some cortical granules caught in the process of exocytosis contained ruthenium red-positive granules identical to those dispersed in the PVS. Moreover, the ruthenium red-reactive granules of the PVS were initially released in spherical clusters the size of cortical granules. Others who have labeled mammalian oocytes with ruthenium red have drawn similar conclusions; i.e., the ruthenium red-positive material observed in the PVS after fertilization comes from the cortical granules (Szollozi, 1967; Gordon et al., 1975). Previous studies (see Introduction) have shown that the cortical granules are rich in carbohydrates. Demonstration of the CGE by ruthenium red staining would be consistent with carbohydrate groups being present in the CGE. Ruthenium red is a polycation and would be expected to interact with negatively charged sugar residues (Luft, 1971). The cortical granule components of mammalian oocytes that have been studied previously are presumed to diffuse into the zona pellucida, where their substrates would be located (e.g., ovoperoxidase, proteinase). The results of our study do not preclude this possibility. However, we have shown that a significant amount of cortical granule material, perhaps most of it, remains in the PVS following fertilization. In the
Figs. 10, 11. PVS surrounding human blastocysts. The CGP (arrowheads) are still present a t this stage ofdevelopment. In Figure 11, some of these particles are aggregated in a sphere. ZP, zona pellucida; T, trophoblast. Figure 10, ~51,500.Figure 11, ~52,500. Fig. 12. PVS surrounding a hamster blastocyst, which is surrounded by a normal appearing zona pellucida (ZP).The PVS is still large and contains numerous dispersed CGP (arrowheads). T, trophoblast. ~32,000. Fig. 13. Mouse blastocyst surrounded by a normal-appearing zona pellucida (ZP).About 20% of zona thickness is shown. The PVS contains numerous dispersed CGP (arrowheads).T, trophoblast. x 28,600. Fig. 14. Mouse blastocyst fixed shortly before the presumed time of hatching. The zona pellucida (ZP)is very thin (about 95% of its thickness is shown), and the PVS has been reduced in size. CGP (arrowheads) are still present in the PVS and the ZP but are highly aggregated. T, trophoblast. ~43,800. Fig. 15. Hamster blastocyst fixed shortly after hatching. The CGP are no longer present around the blastocyst or on the surface membrane of the trophoblast (T). x56,822.
141
mouse, some of this material appears to enter the relatively large pores of the zona that are closest to the oocyte; however, even in the mouse, most of the ruthenium red-stainable material is confined to the PVS. There are several possible explanations for the retention of the CGP in the PVS. Electrostatic forces may prevent it from diffusing into the zona, which would also be expected to have a large number of negatively charged groups. The molecules of the CGE may also form a cross-linked network or gel that is not apparent in our electron micrographs. This could prevent the components of the CGE from migrating into the zona pellucida. Some inferences may be drawn regarding the relationship between the ruthenium red-stainable material and cortical granule material labeled by other methods. LCA-gold and concanavalin (Con A) (both specific for a-D-mannose) label microvilli of activated hamster (Cherr e t al., 1988) and rabbit (Gordon et al., 1975) oocytes in a fairly uniform pattern. In addition, LPA (specific for D-mannose) and UEA I (specific for fucose) are thought to label cortical granule exudate of mouse oocytes (Lee et al., 1988). However, in all these studies, labeling was observed even when oocytes were zona free, and thus most of the ruthenium red stainable material would presumably not have been present. Since some ruthenium red-positive granules seem to be attached to the oolemma in a uniform pattern, it is possible that these granules do correspond to the lectin-positive material detected by others. The 75 kd protein that Pierce et al. (1990) have isolated and characterized from mouse cortical granules may be part of the CGE. This idea is supported by the observations that their antibody to 75p binds to material in the PVS of fertilized oocytes and that the antibody was generated by immunizing rabbits with whole mouse blastocysts. However, antibodies to 75p also show significant labeling of the zona pellucida. Several functions may be postulated for the CGE. The negative charges of the CGE could draw water into the PVS and explain the increase in size observed in the PVS following fertilization. The cortical granule exudate appears fairly cohesive immediately after its release; “spheres” of cortical granule material are frequently observed following fertilization. Dispersal of the exudate may represent a gradual hydration of the cortical granule material and subsequently swelling of the PVS. In support of this idea, the cortical granule exudate of some human oocytes had not dispersed at the time of fixation (see, e.g., Fig. 81, and the PVS of such oocytes remained small. The CGE may also facilitate cleavage of blastomeres and may be necessary for early development. In sea urchins, the hyaline layer, which appears to be analogous to the CGE of mammals, is necessary for proper development (Citkowitz, 1971). Moreover, isolated mouse and pig blastocysts have been shown to develop better in vitro when cultured in the presence of ECM molecules such as fibronectin or laminin (Wilton and Trounson, 1989; Saito and Nieman, 1991). Although the reasons for enhanced growth and development of
142
P. DANDEKAR AND P. TALBOT
mammalian blastomeres in the presence of these ECM components are not yet clear, these results suggest that the CGE, which is an ECM, could affect early development. It is also possible that the CGE is directly involved in blocking polyspermy a t the level of the PVS or the oolemma. Mouse and hamster oocytes have a slow (nonelectrical) block to polyspermy at the level of the oolemma (Wolf, 1978; Sato, 1979; Barros and Yanagimachi, 1972; Stewart-Savage and Bavister, 19881, which could be due to the cortical granule material observed a t this site. It is also possible that the CGE affects sperm interaction with the oocyte. It could impede sperm-oocyte fusion by sterically hindering gamete contact or by enzymatically modifying a n incoming sperm. It is worth noting that the human embryos used in this study were all polyspermic (three pronuclei), and the cortical granule spheres appeared to disperse slowly around these embryos. This could indicate that the CGE did not form quickly enough to set up a block to polyspermy. Cran and Cheng (1986) have also noted increased polyspermy in pig oocytes, in which cortical granule material disperses slowly in the PVS. Moreover, the slow dispersion in pigs appears to be related to suboptimal calcium concentrations in the culture medium. Our results with the human may indicate that the composition of media used in human IVF laboratories does not always favor rapid formation of the CGE, and this can lead to polyspermy.
ACKNOWLEDGMENTS This study was supported in part by a grant from the Academic Senate (UC Riverside). We are grateful to David Demers, Vanessa Hsieh, and Jocelyn Wu for their help in preparing the plates. We also thank Drs. Mary Martin and Robert Glass for providing resources for this project through the In Vitro Fertilization Program at the University of California, San Francisco, and for their helpful comments on the manuscript.
REFERENCES Baranska W, Konwinski M, Kujawa M (1975): Fine structure of the zona pellucida of unfertilized egg cells and embryos. J Exp Zoo1 192:193-202. Barros C, Yanagimachi R (1971):Induction ofzona reaction in golden hamster eggs by cortical granule material. Nature 233:268-269. Barros C, Yanagimachi R (1972):Polyspermy-preventing mechanisms in the golden hamster egg. J Exp Zoo1 180:251-266. Braden AWH, Austin CR, David HA (1954):The reaction of the zona pellucida to sperm penetration. Aust J Biol Sci 7:391-409. Brown CR (1975):Distribution of hyaluronidase in the ram spermatozoon. J Reprod Fertil45:537-539. Cherr GN, Drobnis EZ, Katz DF (1988):Localization of cortical granule constituents before and after exocytosis in the hamster egg. J Exp Zoo1 246:81-93. Cherr GN, Ducibella T (1990): Activation of the mammalian egg: Cortical granule distribution, exocytosis, and the block to polyspermy. In BD Bavister, J Cummins, ERS Roldan (eds): “Fertilization in Mammals.” Norwell, MA: Serono Symposia, pp 309-334. Citkowitz E (1971):The hyaline layer: Its isolation and role in echinoderm development. Dev Biol24:348-362. Cran DG, Cheng WTK (1986): The cortical reaction in pig oocytes during in vivo and in vitro fertilization. Gamete Res 13541-251,
Dandekar P, Aggeler J, Talbot P (1987): Distribution of hyaluronic acid in the human oocyte cumulus complex. J.Cell. Biol. 105:295A. Dandekar P, Aggelar J , Talbot P (1988):Sensitivity of extracellular matrix in the human oocyte cumulus complex to enzymes. Abstract 4th Int Cong Cell Biol Montreal, p 247. Dandekar PV, Glass RH (1987): Development of mouse embryos is affected by strain and culture medium. Gam. Res. 17:279-285. Dandekar PV, Martin MC, Glass RH (1990): Polypronuclear embryos alter in vitro fertilization. Fertil Steril 53:510-514. Ducibella T, Anderson E, Albertini DF, Aalberg J, Rangarajan S (1988): Quantitative studies of changes in cortical granule number and distribution in the mouse oocyte during meiotic maturation. Dev Biol 130:184-197. Gordon M, Fraser LR, Dandeker PV (1975): The effect of ruthenium red and concanavalin A on the vitelline surface of fertilized and unfertilized rabbit ova. Anat Rec 181:95-112. Gulyas BJ, Schmell ED (1980): Ovoperoxidase activity in ionophore treated mouse eggs. I. Electron microscopic localization. Gamete Res 3267-278. Guraya SS (1969): Histochemical nature of cortical granules in the human egg. 2 Zellforsch 94:32-35. Gwatkin RBL, Williams DT, Hartmann JF, Kniazuk M (1973):The zona reaction of hamster and mouse eggs: Production in vitro by a trypsin-like protease from cortical granules. J Reprod Fertil32:259265. Harrison RAP (1988):Hyalurondiase in ram semen: Quantitative determination and isolation of multiple forms. Biochem J 252865874. Huarte J , Belin D, Vassalli J D (1985): Plasminogen activator in mouse and rat oocytes: Induction during meiotic maturation. Cell 43551-558, Hylander BL, Summers RG (1982): An ultrastructural immunocytochemical localization of hyalin in the sea urchin egg. Dev Biol 93:368-380. Kopecny V, Flechon J E , Motlik J , Piuko J (1984):Glycosaminoglycan synthesis by pig oocytes during meiotic maturation: Fine structure autoradiography and cytochemistry. Abstract 8th Eur Cong Electron Microsc Budapest. Kujawa MJ, Tepperman K (1983): Culturing chick muscle cells on glycosaminoglycansubstrates: Attachment and differentiation. Dev Biol 99:277-286. Lee SH, Ahuja KK, Gilburt DJ, Whittingham DG (1988):The appearance of glycoconjugates associated with cortical granule release during mouse fertilization. Development 102:595-604. Luft JH (1971):Ruthenium red and violet. I. Chemistry, purification, methods of use in microscopy and mechanisms of action. Anat Rec 171:347-368. McGregor L, Flaherty SP, Breed WG (1989): Structure of the zona pellucida and cumulus oophorus in three species of native Australian rodents. Gamete Res 23:279-287. Moller CC, Wassarman PM (1989): Characterization of a proteinase that cleaves zona pellucida glycoprotein ZP2 following activation of mouse eggs. Dev Biol 132:103-112. Morton DB (1976)Lysosomal enzymes in mammalian spermatozoa. In J D Dingle, RT Dean (eds): “Lysosomes in Biology and Pathology, Vol. 5” Amsterdam: ElsevieriNorth-Holland, pp 203-255. Orkin RW, Knudson W, Toole BP (1985): Loss of hyaluronate-dependent coat during myoblast fusion. Dev Biol 107527-530. Pierce KE, Siebert MC, Kopf GS, Schultz RM, Calarco PG (1990): Characterization and localization of a mouse egg cortical granule antigen prior to and following fertilization or egg activation. Dev Biol 141:381-392. Saito S, Niemann H (1991): Effects of extracellular matrices and growth factors in the development of isolated porcine blastomeres. Biol Reprod 44:927-936. Sat0 K (1979): Polyspermy preventing mechanisms in mouse eggs fertilized in vitro. J Exp Zoo1 210:353-359. Schmell ED, Gulyas BJ (1980): Ovoperoxidase activity in ionophore treated mouse eggs. 11. Evidence for the enzyme’s role in hardening of the zona pellucida. Gamete Res 3:279-290. Sinosich MJ, Lanzendorf SE, Hodgen GD (1990): Immunocytochemical studies on hamster oocyte activator. Mol Reprod Dev 26:331336.
PERIVITELLINE SPACE MATRIX OF MAMMALIAN OOCYTES Steward-Savage J, Bavister BD (1988): A cell surface block to polyspermy occurs in golden hamster eggs. Dev Biol 128:150-157. Szollosi D (1967): Development of cortical granules and the cortical reaction in rat and hamster eggs. Anat Rec 159:431-446. Talbot P (1984):Hyaluronidase dissolves a component in the hamster zona pellucida. J Exp Zoo1 229:309-316. Talbot P, DiCarlantonio G (1984a): Ultrastructure of opposum oocyte investing coats and their sensitivity to trypsin and hyaluronidase. Dev Biol 103:159-167. Talbot P, DiCarlantonio G (1984b): The oocyte-cumulus complex: Ultrastructure of the extracellular components in hamsters and mice. Gamete Res 10:127-142. Talbot P, DiCarlantonio G (1984~):Architecture of the hamster oocyte-cumulus complex. Gamete Res 9:261-272. Turner TT, Reich GW (1987): Influence of proteins in rat cauda epididymidal lumen fluid on cauda sperm motility. Gamete Res 18:267-278.
143
Vollett JJ, Roth LE (1974):Cell fusion by nascent-membrane induction and divalent cation treatment. Cytobiologie 9:249-262. Wilton U,Trounson A 0 (1989): Biopsy of preimplantation mouse embryos: development of micromanipulated embryos and proliferation of single blastomeres in vitro. Biol Reprod 40:145-152. Wolf DP (1978): The block to sperm penetration in zona-free mouse eggs. Dev Biol 64:l-10. Wolf DP, Hamada M (1977):Induction ofzonal and oolemmal blocks to sperm penetration in mouse eggs with cortical granule exudate. Biol Reprod 17:350-354. Yanagimachi R, Chang MC (1961):Fertilizable life of golden hamster ova and their morphological changes a t the time of losing fertilizability. J Exp Zoo1 148:185-204.
Perivitelline Space of Mammalian Oocytes: Extracellular Matrix of Unfertilized Oocytes and Formation of a Cortical Granule Envelope Following Fertilization P. DANDEKAR' AND P. TALBOT2 'Department of Obstetrics, Gynecology, and Reproductive Sciences, University of California School of Medicine, San Francisco, California; 'Department of Biology, University of California, Riverside, California
ABSTRACT Extracellular matrices (ECM) present around unfertilized and fertilized mammalian oocytes were studied ultrastructurally in samples prepared in the presence of ruthenium red to facilitate stabilization of extracellular materials. Unfertilized mouse, hamster,and human oocytes have an ECM comprising granules and filaments in their perivitelline spaces (PVS).This matrix is more abundant in the human than in hamsters and mice. The granule/filament matrix appears identical to the matrix seen between cumulus and corona radiata cells following ruthenium red processing and previously shown to comprise protein and hyaluronic acid. By including ruthenium red during fixation, it is possible to demonstrate the existence of cortical granule exudate in the PVS of fertilized oocytes from hamsters, mice, and humans. Much of the cortical granule exudate is trapped in the PVS and forms a new coat around the fertilized oocyte. This material is particulate when stained with ruthenium red and appears to be uniformly dispersed around the entire oocyte surface. We refer to this new coat as the cortical granule envelope. This envelope is observed in the PVS of all developmental stages up to and including blastocysts in all three species. Following hatching of mouse and hamster blastocysts, the cortical granule envelope is no longer present. Possible functions of this envelope are discussed. Key Words: Cortical reaction, ECM, PVS, Polyspermy
INTRODUCTION The perivitelline space (PVS) of unfertilized opossum and pig oocytes contains a matrix comprising granules and filaments (Talbot and DiCarlantonio, 1984a; Kopecny et al., 1984).This matrix can be demonstrated by fixing oocytes in the presence of ruthenium red. In the opossum, the granules and filaments can be removed by trypsin and Streptomyces hyaluronidase treatment, respectively, indicating that the granules comprise protein and the hyaluronic acid filaments (Talbot and DiCarlantonio, 1984a). Following fertilization, cortical granules release their contents exocytotically into the PVS (Szollozi, 1967) in a n event that has been termed the cortical reaction (Braden et al., 1954). The cortical granule exu-
0 1992 WILEY-LISS, INC.
date is thought to block polyspermy at the level of the oolemma andlor by altering the properties of the zona pellucida (reviewed by Cherr and Ducibella, 1990). The contents of the cortical granules are periodic acid-Schiff (PAS) positive (Yanagimachi and Chang, 1961; Guraya, 1969) and bind several lectins (Cherr et al., 1988; Ducibella et al., 1988; Lee et al., 1988). After fertilization, hamster oocytes release heparin binding placental protein (HBPP), which may come from the cortical granules (Sinosich et al., 1990). In addition, mammalian cortical granules appear to contain an ovoperoxidase that has been implicated in zona hardening (Gulyas and Schmell, 1980; Schmell and Gulyas, 1980) and several proteinases. Experimental evidence suggests that a trypsin-like proteinase is released from the cortical granules and modifies the zona pellucida to block polyspermy (Barros and Yanagimachi, 1971; Gwatkin e t al., 1973; Wolf and Hamada, 1977). This may be plasminogen activator, which is secreted from fertilized mouse oocytes (Huarte et al., 1985). More recently, Moller and Wassarman (1989) have isolated a proteinase from mouse oocytes that is released a t fertilization and modifies the zona protein, ZP2, to produce ZP2f, a form that is not capable of binding sperm. The authors implicate this proteinase in blocking polyspermy by inhibiting sperm binding to ZP2 and by bringing about zona hardening. The proteinase described by Moller and Wassarman (1989) is not blocked by trypsin inhibitors and appears to be distinct from that described in earlier studies. Although the proteinases are probably of cortical granule origin, this has not yet been shown cytochemically. The only cortical granule constituent that has been isolated and characterized from mammalian oocytes is a 75 kd protein found in mouse cortical granules (Pierce et al., 1990).This protein (75p) was shown cytochemically to be localized in the cortical granules and released into the PVS and zona pellucida after fertilization. Its function is a s yet unknown.
Received August 15, 1991; accepted September 23, 1991. Address reprint requests to P. Talbot, Department of Biology, University of California, Riverside, CA 92521.
136
P. DANDEKAR AND P. TALBOT
In spite of the probable importance of the PVS matrices in fertilization, we know relatively little about them. The granuleifilament matrix of unfertilized oocytes has not been identified in any species other than pig and opossum, and the fate of the cortical granule exduate after its release into the PVS is not fully understood. Some cortical granule materials, such as the ovoperoxidase and proteinases, presumably diffuse into the zona pellucida to bring about zona hardening and to block polyspermy. However, in several preliminary studies, cortical granule exudate was observed in the PVS after the cortical reaction (Szollozi, 1967; Gordon et al., 1975; Baranska et al., 1975; Talbot and DiCarlantonio, 198413).The purposes of the present study were to determine if a granuleifilament matrix is present in the PVS of unfertilized oocytes in other mammalian species and to examine the fate of the cortical granule exudate through early developmental stages up to and including blastocyst hatching. Ruthenium red was used to visualize extracellular matrices in the PVS, and the study was conducted using oocytes from mice, hamsters, and humans.
MATERIALS AND METHODS Hamsters Mature female golden hamsters were obtained, maintained, and fed as previously described (Talbot and DiCarlantoni, 1 9 8 4 ~ )Unfertilized . oocyte-cumulus complexes (OCC)were collected in Earle’s balanced salt solution (EBSS) from the oviducts of unmated females. In vivo fertilization (IVF)was set up as described previously (Talbot and DiCarlantonio, 198413). At various times after mating, reproductive tracts were removed from females and flushed through the infundibulum with EBSS to recover fertilized oocytes and developmental stages up to and including hatched blastocysts. Mice Four- to six-week-old, random, bred CFW (Swiss Webster) mice (Charles River, Portage, MI) were superovulated and mated; unfertilized oocytes and one-cell embryos were collected. Some one-cell embryos were cultured to hatched blastocysts, as described previously (Dandekar and Glass, 1987). Briefly, unfertilized oocytes and one-cell embryos in cumulus masses were released from swollen ampullae. To free oocytes and embryos, cumulus masses were treated with 0.1% hyaberonidase for 2-5 min and washed four to six times with T, medium. To verify fertilization, embryos were checked for the presence of two pronuclei and polar bodies. Unfertilized oocytes, one-cell embryos, and blastocysts were processed with ruthenium red, as described previously (Talbot and DiCarlantonio, 1984b). Humans Human oocytes that failed to fertilize were obtained from a n in vitro fertilization program. Details of the procedures used in this program have been described (Dandekar et al., 1990). Oocytes that failed to fertilize,
triploid embryos, and abnormally developing embryos also were processed with ruthenium red (Talbot and DiCarlantonio, 198413).
Processing for Electron Microscopy The developmental stage of each sample was determined using a dissecting microscope. Each sample was then processed for transmission electron microscopy (TEM)using previously described methods that include ruthenium red in the fixatives and postglutaraldehyde washes (Talbot, 1984; Talbot and DiCarlantonio, 1 9 8 4 ~ )Blocks . were thin sectioned using glass or diamond knives on a n MT-2B ultramicrotome. Sections were picked up on copper grids and examined either without staining or following heavy metal staining in a Hitachi H-500 or a Phillips 300 TEM. RESULTS The ultrastructure of the ECM in the PVS surrounding unfertilized oocytes and a t early developmental stages, up to and including blastocysts, was examined in the mouse (Figs. 1, 5, 6, 13, 14), hamster (Figs. 2, 9, 12,15),and human (Figs. 3,4,7,8,10,11). All material was fixed in the presence of ruthenium red to help stabilize extracellular materials. Unfertilized Oocytes In all three species, the PVS surrounding unfertilized oocytes is relatively small and contains a n ECM comprising granules and filaments (Figs. 1-4). This matrix is identical to the granuleifilament matrix previously described in the extracellular spaces of the cumulus layer, corona radiata, and outer pores of the zona pellucida (Talbot and DiCarlantonio, 1984b; Dandekar et al., 1987). The granules and filaments interconnect with each other, and some appear to bind to the 00lemma (Fig. 2). The granuleifilament matrix is most abundant in the PVS surrounding human oocytes; it is relatively sparse and difficult to demonstrate in the PVS of hamsters and mice. Fertilized Oocytes In vivo-fertilized oocytes (monospermic) were obtained from mice and hamsters after natural matings. In the case of the human, only in vitro-fertilized oocytes
Fig. 1. Overview of the cortex (C), perivitelline space (PVSJ, and zona pellucida (ZPJ of a n unfertilized mouse oocyte. The PVS is small and contains a sparse ECM comprising granules (arrowheads) and filaments (not evident a t this low magnification). A similar ECM is evident on the outer surface of the zona pellucida, where it is much more abundant. x 17,300. Fig. 2. Higher magnification view of the PVS surrounding an unfertilized hamster oocyte. The granules (arrowheads) and filaments (FJ of the ECM are shown. The granules are smaller than usual hecause this oocyte was lightly trypsinized before fixation. ZP, zona pellucida; C, cortex of oocyte. ~ 5 5 , 4 0 0 . Figs. 3.4. Unfertilized human oocyte showing the cortex (CJ, PVS, and part of the zona pellucida (ZPJ. The PVS contains an ECM comprising granules (GI and filaments (F).The filaments are best seen at higher magnification (Fig. 4). Figure 3, x 17,600. Figure 4, x59,lOO.
Figs. 1-4.
138
P. DANDEKAR AND P. TALBOT
Figs. 5-9.
PERIVITELLINE SPACE MATRIX OF MAMMALIAN OOCYTES that exhibited polyspermy or abnormal cleavage were processed for electron microscopy. Following fertilization, the PVS of all species increases significantly in size (Figs. 5, 7,9). This increase in size is accompanied by loss of most cortical granules from the periphery of the oocytes. Coincident with the cortical reaction, numerous small, dense particles (cortical granule particles; CGP) appear in the PVS (Figs. 5-9). Cortical granules caught in the process of exocytosis (Figs. 6, 7) are releasing numerous small dense particles identical to those that are already dispersed (Figs. 4 , 9 ) in the PVS of fertilized oocytes. In all species, the CGP disperse apart from each other after their release and could be found evenly distributed in the PVS (Figs. 5, 9, 10, 12, 13). Following their dispersion, the CGP formed a new vestment, the cortical granule envelope (CGE), in the PVS. One abnormally developing human embryo reached the blastocyst stage; some CGP remained together in a coherent spherical mass in the PVS around this blastocyst (Fig. 11). While the majority of CGP are free in the PVS, some appear to be bound to the oolemma (Figs. 5 , 6 , 9 ) . In the mouse, some particles diffuse into the adjacent pores of the zona pellucida (Fig. 5); however, the number of particles in the zona pellucida is always small compared with that observed in the PVS. In the human and hamster, the zona pellucida facing the PVS lacks the large pores characteristic of the mouse zona, and CGP did not appear to penetrate into the zonae of these two species.
Later Stages of Development Cleavage stages, morulae, and blastocysts were collected from mice and hamsters following IVF. IVF hu-
Fig. 5. Overview of the cortex (C),PVS, and zona pellucida (ZP)of a fertilized mouse oocyte. The PVS is larger than in unfertilized oocytes and contains numerous small electron-dense particles (arrowheads) released by the cortical granules. Some of these particles have diffused into the pores of the zona pellucida. No cortical granules remain in the cortex of the oocyte in this view. x61,SOO. Fig. 6. Surface of a fertilized mouse oocyte showing a n invagination, which probably represents a cortical granule undergoing exocytosis. The electron-dense particles (arrowhead 1)in this fused cortical granule had access to ruthenium red and appear identical to the particles (arrowhead 2) of the CGE. x61,lOO. Fig. 7. Fertilized human oocyte showing exocytosis of a cortical granule (CG). Its contents appear identical to CGP (not shown) dispersed in the PVS. In the human, the CGP often appeared cohesive after their release (arrowhead). Remnants of the granulelfilament matrix seen around unfertilized oocytes were still visible around this oocyte but are not shown in this area. X65,900. Fig. 8. Fertilized human oocyte showing CGP (arrowheads) after their release into the PVS. The particles are cohesive and often remain clustered together in spheres about the size of intact cortical granules. In this oocyte, the cortical granule exudate has not yet dispersed, and the PVS has not yet increased in size. ~ 6 1 , 6 5 4 . Fig. 9. Fertilized hamster oocyte that has reached the two-cell stage. The PVS is large and contains numerous dense particles of cortical granule origin (arrowheads). Some of these particles are bound to the oolemma (0). ZP, zona pellucida. x29,OOO.
139
man oocytes that showed abnormal cleavage or fragmentation were allowed to develop in vitro, and various stages, including a blastocyst, were fixed for electron microscopy. In all stages up to and including unhatched blastocysts, CGP could be found in the PVS of humans (Figs. 10-111, hamsters (Fig. 121, and mice (Figs. 13, 14). In blastocysts which have a normal zona pellucida and do not appear close to hatching, the PVS is large and the particles are dispersed (Figs. 10, 12, 13).However, in blastocysts surrounded by a thin zona pellucida and close to hatching, the PVS is reduced in size, and the CGP often are aggregated (Fig. 14). Particles are not present around hatched mouse (not shown) or hamster (Fig. 15)blastocysts.
DISCUSSION Two main points have emerged from this study. First, the PVS of unfertilized oocytes contains a granule/ filament matrix that is indistinguishable morphologically from the ECM described previously between corona radiata cells and cumulus cells of the hamster, mouse, and human OCC (Talbot and DiCarlantonio, 198413;Dandekar et al., 1987,1988)and between cumulus cells of three hydromyine rodents (McGregor et al., 1989). Second, the cortical reaction results in the elaboration of a new coat around the fertilized mammalian oocyte. This coat is confined to the PVS and is not readily apparent in electron micrographs unless special methods, such a s ruthenium red processing, are used. This coat comprises numerous electron-dense particles that are larger than those observed in the PVS around unfertilized oocytes. This coat remains around ail early developmental stages up to and including blastocysts. However, it is not present following blastocyst hatching. We propose that this coat be referred to as the cortical granule envelope (CGE). This envelope appears to be analogous to the hyaline layer, which originates from the cortical granules of fertilized sea urchin oocytes (Hylander and Summers, 1982). The granulelfilament matrix we observed in the PVS of unfertilized oocytes is structurally similar to that surrounding opossum (Talbot and DiCarlantonio, 1984a) and pig (Kopecny et al., 1984) oocytes. This ECM may be secreted by the corona radiata cells and released at the tips of their processes, which Szollozi (1967) has shown extend through the zona pellucida. Alternatively, i t may be synthesized and secreted by the oocyte, a s Kopecny e t al. (1984) have shown for the pig. The granules of this matrix can be digested by trypsin and hence are proteinacous, while the filaments can be removed by Streptomyces hyaluronidase, which is specific for hyaluronic acid (HA) (Talbot and DiCarlantonio, 1984a; Dandekar et al., 1988). HA in the PVS might present a significant blockade to the fertilizing sperm, since HA in other systems has been shown to inhibit membrane fusion (Vollet and Roth, 1974; Kujawa and Tepperman, 1983; Orkin et al., 1985). Sperm hyaluronidase, which is released during the acrosome reaction at the zona pellucida surface, could diffuse through the zona and remove HA from the
140
P. DANDEKAR AND P. TALBOT
Figs. 10-15.
PERIVITELLINE SPACE MATRIX OF MAMMALIAN OOCYTES oolemma. Alternatively, some hyaluronidae may be retained by acrosome-reacted sperm (Brown, 1975; Morton, 1976; Harrison, 1988) and carried into the PVS to perform this function. This ECM is much more abundant and is easily demonstrated in the human than in the two rodents. Following formation of the CGE, it was not possible to observe the granule/filament matrix characteristic of unfertilized oocytes. This could be because i t was masked by the CGE or because i t was degraded during fertilization. The new extracellular coat that is present in the PVS after fertilization appears to originate from the cortical granules. This is supported by the fact that this coat is not present prior to fertilization, when cortical granules are still within the oocyte, but appears coincident with the cortical reaction. Some cortical granules caught in the process of exocytosis contained ruthenium red-positive granules identical to those dispersed in the PVS. Moreover, the ruthenium red-reactive granules of the PVS were initially released in spherical clusters the size of cortical granules. Others who have labeled mammalian oocytes with ruthenium red have drawn similar conclusions; i.e., the ruthenium red-positive material observed in the PVS after fertilization comes from the cortical granules (Szollozi, 1967; Gordon et al., 1975). Previous studies (see Introduction) have shown that the cortical granules are rich in carbohydrates. Demonstration of the CGE by ruthenium red staining would be consistent with carbohydrate groups being present in the CGE. Ruthenium red is a polycation and would be expected to interact with negatively charged sugar residues (Luft, 1971). The cortical granule components of mammalian oocytes that have been studied previously are presumed to diffuse into the zona pellucida, where their substrates would be located (e.g., ovoperoxidase, proteinase). The results of our study do not preclude this possibility. However, we have shown that a significant amount of cortical granule material, perhaps most of it, remains in the PVS following fertilization. In the
Figs. 10, 11. PVS surrounding human blastocysts. The CGP (arrowheads) are still present a t this stage ofdevelopment. In Figure 11, some of these particles are aggregated in a sphere. ZP, zona pellucida; T, trophoblast. Figure 10, ~51,500.Figure 11, ~52,500. Fig. 12. PVS surrounding a hamster blastocyst, which is surrounded by a normal appearing zona pellucida (ZP).The PVS is still large and contains numerous dispersed CGP (arrowheads). T, trophoblast. ~32,000. Fig. 13. Mouse blastocyst surrounded by a normal-appearing zona pellucida (ZP).About 20% of zona thickness is shown. The PVS contains numerous dispersed CGP (arrowheads).T, trophoblast. x 28,600. Fig. 14. Mouse blastocyst fixed shortly before the presumed time of hatching. The zona pellucida (ZP)is very thin (about 95% of its thickness is shown), and the PVS has been reduced in size. CGP (arrowheads) are still present in the PVS and the ZP but are highly aggregated. T, trophoblast. ~43,800. Fig. 15. Hamster blastocyst fixed shortly after hatching. The CGP are no longer present around the blastocyst or on the surface membrane of the trophoblast (T). x56,822.
141
mouse, some of this material appears to enter the relatively large pores of the zona that are closest to the oocyte; however, even in the mouse, most of the ruthenium red-stainable material is confined to the PVS. There are several possible explanations for the retention of the CGP in the PVS. Electrostatic forces may prevent it from diffusing into the zona, which would also be expected to have a large number of negatively charged groups. The molecules of the CGE may also form a cross-linked network or gel that is not apparent in our electron micrographs. This could prevent the components of the CGE from migrating into the zona pellucida. Some inferences may be drawn regarding the relationship between the ruthenium red-stainable material and cortical granule material labeled by other methods. LCA-gold and concanavalin (Con A) (both specific for a-D-mannose) label microvilli of activated hamster (Cherr e t al., 1988) and rabbit (Gordon et al., 1975) oocytes in a fairly uniform pattern. In addition, LPA (specific for D-mannose) and UEA I (specific for fucose) are thought to label cortical granule exudate of mouse oocytes (Lee et al., 1988). However, in all these studies, labeling was observed even when oocytes were zona free, and thus most of the ruthenium red stainable material would presumably not have been present. Since some ruthenium red-positive granules seem to be attached to the oolemma in a uniform pattern, it is possible that these granules do correspond to the lectin-positive material detected by others. The 75 kd protein that Pierce et al. (1990) have isolated and characterized from mouse cortical granules may be part of the CGE. This idea is supported by the observations that their antibody to 75p binds to material in the PVS of fertilized oocytes and that the antibody was generated by immunizing rabbits with whole mouse blastocysts. However, antibodies to 75p also show significant labeling of the zona pellucida. Several functions may be postulated for the CGE. The negative charges of the CGE could draw water into the PVS and explain the increase in size observed in the PVS following fertilization. The cortical granule exudate appears fairly cohesive immediately after its release; “spheres” of cortical granule material are frequently observed following fertilization. Dispersal of the exudate may represent a gradual hydration of the cortical granule material and subsequently swelling of the PVS. In support of this idea, the cortical granule exudate of some human oocytes had not dispersed at the time of fixation (see, e.g., Fig. 81, and the PVS of such oocytes remained small. The CGE may also facilitate cleavage of blastomeres and may be necessary for early development. In sea urchins, the hyaline layer, which appears to be analogous to the CGE of mammals, is necessary for proper development (Citkowitz, 1971). Moreover, isolated mouse and pig blastocysts have been shown to develop better in vitro when cultured in the presence of ECM molecules such as fibronectin or laminin (Wilton and Trounson, 1989; Saito and Nieman, 1991). Although the reasons for enhanced growth and development of
142
P. DANDEKAR AND P. TALBOT
mammalian blastomeres in the presence of these ECM components are not yet clear, these results suggest that the CGE, which is an ECM, could affect early development. It is also possible that the CGE is directly involved in blocking polyspermy a t the level of the PVS or the oolemma. Mouse and hamster oocytes have a slow (nonelectrical) block to polyspermy at the level of the oolemma (Wolf, 1978; Sato, 1979; Barros and Yanagimachi, 1972; Stewart-Savage and Bavister, 19881, which could be due to the cortical granule material observed a t this site. It is also possible that the CGE affects sperm interaction with the oocyte. It could impede sperm-oocyte fusion by sterically hindering gamete contact or by enzymatically modifying a n incoming sperm. It is worth noting that the human embryos used in this study were all polyspermic (three pronuclei), and the cortical granule spheres appeared to disperse slowly around these embryos. This could indicate that the CGE did not form quickly enough to set up a block to polyspermy. Cran and Cheng (1986) have also noted increased polyspermy in pig oocytes, in which cortical granule material disperses slowly in the PVS. Moreover, the slow dispersion in pigs appears to be related to suboptimal calcium concentrations in the culture medium. Our results with the human may indicate that the composition of media used in human IVF laboratories does not always favor rapid formation of the CGE, and this can lead to polyspermy.
ACKNOWLEDGMENTS This study was supported in part by a grant from the Academic Senate (UC Riverside). We are grateful to David Demers, Vanessa Hsieh, and Jocelyn Wu for their help in preparing the plates. We also thank Drs. Mary Martin and Robert Glass for providing resources for this project through the In Vitro Fertilization Program at the University of California, San Francisco, and for their helpful comments on the manuscript.
REFERENCES Baranska W, Konwinski M, Kujawa M (1975): Fine structure of the zona pellucida of unfertilized egg cells and embryos. J Exp Zoo1 192:193-202. Barros C, Yanagimachi R (1971):Induction ofzona reaction in golden hamster eggs by cortical granule material. Nature 233:268-269. Barros C, Yanagimachi R (1972):Polyspermy-preventing mechanisms in the golden hamster egg. J Exp Zoo1 180:251-266. Braden AWH, Austin CR, David HA (1954):The reaction of the zona pellucida to sperm penetration. Aust J Biol Sci 7:391-409. Brown CR (1975):Distribution of hyaluronidase in the ram spermatozoon. J Reprod Fertil45:537-539. Cherr GN, Drobnis EZ, Katz DF (1988):Localization of cortical granule constituents before and after exocytosis in the hamster egg. J Exp Zoo1 246:81-93. Cherr GN, Ducibella T (1990): Activation of the mammalian egg: Cortical granule distribution, exocytosis, and the block to polyspermy. In BD Bavister, J Cummins, ERS Roldan (eds): “Fertilization in Mammals.” Norwell, MA: Serono Symposia, pp 309-334. Citkowitz E (1971):The hyaline layer: Its isolation and role in echinoderm development. Dev Biol24:348-362. Cran DG, Cheng WTK (1986): The cortical reaction in pig oocytes during in vivo and in vitro fertilization. Gamete Res 13541-251,
Dandekar P, Aggeler J, Talbot P (1987): Distribution of hyaluronic acid in the human oocyte cumulus complex. J.Cell. Biol. 105:295A. Dandekar P, Aggelar J , Talbot P (1988):Sensitivity of extracellular matrix in the human oocyte cumulus complex to enzymes. Abstract 4th Int Cong Cell Biol Montreal, p 247. Dandekar PV, Glass RH (1987): Development of mouse embryos is affected by strain and culture medium. Gam. Res. 17:279-285. Dandekar PV, Martin MC, Glass RH (1990): Polypronuclear embryos alter in vitro fertilization. Fertil Steril 53:510-514. Ducibella T, Anderson E, Albertini DF, Aalberg J, Rangarajan S (1988): Quantitative studies of changes in cortical granule number and distribution in the mouse oocyte during meiotic maturation. Dev Biol 130:184-197. Gordon M, Fraser LR, Dandeker PV (1975): The effect of ruthenium red and concanavalin A on the vitelline surface of fertilized and unfertilized rabbit ova. Anat Rec 181:95-112. Gulyas BJ, Schmell ED (1980): Ovoperoxidase activity in ionophore treated mouse eggs. I. Electron microscopic localization. Gamete Res 3267-278. Guraya SS (1969): Histochemical nature of cortical granules in the human egg. 2 Zellforsch 94:32-35. Gwatkin RBL, Williams DT, Hartmann JF, Kniazuk M (1973):The zona reaction of hamster and mouse eggs: Production in vitro by a trypsin-like protease from cortical granules. J Reprod Fertil32:259265. Harrison RAP (1988):Hyalurondiase in ram semen: Quantitative determination and isolation of multiple forms. Biochem J 252865874. Huarte J , Belin D, Vassalli J D (1985): Plasminogen activator in mouse and rat oocytes: Induction during meiotic maturation. Cell 43551-558, Hylander BL, Summers RG (1982): An ultrastructural immunocytochemical localization of hyalin in the sea urchin egg. Dev Biol 93:368-380. Kopecny V, Flechon J E , Motlik J , Piuko J (1984):Glycosaminoglycan synthesis by pig oocytes during meiotic maturation: Fine structure autoradiography and cytochemistry. Abstract 8th Eur Cong Electron Microsc Budapest. Kujawa MJ, Tepperman K (1983): Culturing chick muscle cells on glycosaminoglycansubstrates: Attachment and differentiation. Dev Biol 99:277-286. Lee SH, Ahuja KK, Gilburt DJ, Whittingham DG (1988):The appearance of glycoconjugates associated with cortical granule release during mouse fertilization. Development 102:595-604. Luft JH (1971):Ruthenium red and violet. I. Chemistry, purification, methods of use in microscopy and mechanisms of action. Anat Rec 171:347-368. McGregor L, Flaherty SP, Breed WG (1989): Structure of the zona pellucida and cumulus oophorus in three species of native Australian rodents. Gamete Res 23:279-287. Moller CC, Wassarman PM (1989): Characterization of a proteinase that cleaves zona pellucida glycoprotein ZP2 following activation of mouse eggs. Dev Biol 132:103-112. Morton DB (1976)Lysosomal enzymes in mammalian spermatozoa. In J D Dingle, RT Dean (eds): “Lysosomes in Biology and Pathology, Vol. 5” Amsterdam: ElsevieriNorth-Holland, pp 203-255. Orkin RW, Knudson W, Toole BP (1985): Loss of hyaluronate-dependent coat during myoblast fusion. Dev Biol 107527-530. Pierce KE, Siebert MC, Kopf GS, Schultz RM, Calarco PG (1990): Characterization and localization of a mouse egg cortical granule antigen prior to and following fertilization or egg activation. Dev Biol 141:381-392. Saito S, Niemann H (1991): Effects of extracellular matrices and growth factors in the development of isolated porcine blastomeres. Biol Reprod 44:927-936. Sat0 K (1979): Polyspermy preventing mechanisms in mouse eggs fertilized in vitro. J Exp Zoo1 210:353-359. Schmell ED, Gulyas BJ (1980): Ovoperoxidase activity in ionophore treated mouse eggs. 11. Evidence for the enzyme’s role in hardening of the zona pellucida. Gamete Res 3:279-290. Sinosich MJ, Lanzendorf SE, Hodgen GD (1990): Immunocytochemical studies on hamster oocyte activator. Mol Reprod Dev 26:331336.
PERIVITELLINE SPACE MATRIX OF MAMMALIAN OOCYTES Steward-Savage J, Bavister BD (1988): A cell surface block to polyspermy occurs in golden hamster eggs. Dev Biol 128:150-157. Szollosi D (1967): Development of cortical granules and the cortical reaction in rat and hamster eggs. Anat Rec 159:431-446. Talbot P (1984):Hyaluronidase dissolves a component in the hamster zona pellucida. J Exp Zoo1 229:309-316. Talbot P, DiCarlantonio G (1984a): Ultrastructure of opposum oocyte investing coats and their sensitivity to trypsin and hyaluronidase. Dev Biol 103:159-167. Talbot P, DiCarlantonio G (1984b): The oocyte-cumulus complex: Ultrastructure of the extracellular components in hamsters and mice. Gamete Res 10:127-142. Talbot P, DiCarlantonio G (1984~):Architecture of the hamster oocyte-cumulus complex. Gamete Res 9:261-272. Turner TT, Reich GW (1987): Influence of proteins in rat cauda epididymidal lumen fluid on cauda sperm motility. Gamete Res 18:267-278.
143
Vollett JJ, Roth LE (1974):Cell fusion by nascent-membrane induction and divalent cation treatment. Cytobiologie 9:249-262. Wilton U,Trounson A 0 (1989): Biopsy of preimplantation mouse embryos: development of micromanipulated embryos and proliferation of single blastomeres in vitro. Biol Reprod 40:145-152. Wolf DP (1978): The block to sperm penetration in zona-free mouse eggs. Dev Biol 64:l-10. Wolf DP, Hamada M (1977):Induction ofzonal and oolemmal blocks to sperm penetration in mouse eggs with cortical granule exudate. Biol Reprod 17:350-354. Yanagimachi R, Chang MC (1961):Fertilizable life of golden hamster ova and their morphological changes a t the time of losing fertilizability. J Exp Zoo1 148:185-204.
