The ability of hamster oolemma to fuse with spermatozoa: Its acquisition during oogenesis and loss after fertilization

Fertilization begins with membrane fusion between the spermatozoon and oocyte and ends with the mingling of the male and female genome. Although the morphology of sperm-oocyte membrane fusion in mammals has been studied extensively using a variety of species, the mechanisms involved in this process are not well understood (for reviews, see Yanagimachi, 1988a,b). There has been no definition so far of the point in oogenesis at which the fusibility of the oocyte’s plasma membrane (oolemma) appears. If this can be identified, it may be possible then to begin to look for the key elements in the oolemma that are required for fusion. The questions that we ask in this study are (1) when during oogenesis does the oolemma gain the ability to fuse with spermatozoa and (2) when after fertilization does this ability disappear?

Syrian (golden) hamsters were raised and maintained in a temperature-controlled room with a 14 h light phase (05.00–19.00 h) and 10h dark phase (19 00–05.00h). The ages of females and males at the time of experiments were 1 week to 4 months and 4 to 6 months, respectively.

All reagents were purchased from Sigma Chemical Co. (St Louis, MO) unless otherwise stated.

The medium used for the induction of capacitation and the acrosome reaction of spermatozoa was a modified Tyrode’s solution, m-TALP-3, with the following composition: 101.02mM NaCl, 2 68mM KC1, 1.80mM CaCl₂, 0.49mM MgCl₂.6H₂O, 0.36HIM NaH₂PO4.H₂O, 35 70mM NaHCO₃, 4.50mM D-glucose, 1.0HIM sodium pyruvate, 9.0 mM sodium lactate, 0.5mM hypotaunne, 0.05HIM (–) epinephrine, and 15 mg ml⁻¹ bovine serum albumin (BSA) (Fr. V, Calbiochem, La Jolla, CA). This medium was supplemented with 50/igml^-1 gentamicin sulfate and 0.1 mM EDTA The pH of this medium was approximately 7.4 when equilibrated with 5 % CO₂ in air at 37°C. Tissue culture medium TC199 (with modified Earle’ base, Sigma Chem., M-2520) containing 26 mM NaHCO₃, 25 mM Hepes, 4 mg ml⁻¹ BSA and 50 μg ml⁻¹ gentamicin sulfate was used for isolation of ovarian oocytes. The osmolality of this medium was about 290 mOsmol. The pH of the medium was adjusted to 7.2 by adding a small quantity of 1 M NaOH This medium was also used for treating and inseminating oocytes, fertilized eggs and preimplantation embryos

Spermatozoa from the cauda epididymis were incubated in m-TALP-3 medium as described by Uto et al (1988). The concentration of spermatozoa in the incubation medium was approximately 4–5×10⁶ml⁻¹ After incubation for 4 to 5 h at 37 °C under 5 % CO₂ in air, 40 to 50 % of spermatozoa in the entire population were ‘spontaneoulsy’ acrosome-reacted and most of them were hyperactivated.

Oocytes were isolated from ovaries of immature (1- to 4-week-old) and mature (>2-month-old) females, using the procedures described by Roy and Greenwald (1985) with slight modifications. Ovaries were minced and incubated in 2 ml of albumin-free TC199 containing 2 5 mg ml⁻¹ collagenase (type II, 330 CD units mg⁻¹) and 400 units ml⁻¹ DNAase (type I; 2050 Kumtz units mg⁻¹). After 20 mm of incubation at 37°C under constant shaking, 1% BSA was added and the mixture was filtered through a 200 μm nylon mesh to remove tissue debris. The filtrate was centrifuged at, 190g for 2min, the supernatant discarded and the pellet suspended in 2 ml of TC199 (37°C) containing 2.4mM EGTA Three minutes later, the suspension was centrifuged (190g, 2min), and the pellet was resuspended in TC199. When vortexed for 2 min (using Maxi Mix, Thermolyn Corp., Dubuque, IA), naked (zona-pellucida-free) oocytes of various sizes were seen in the medium (collagenase has the ability to dissolve the zona). The naked oocytes were rinsed and kept in TC199 medium for 10-30 min before they were inseminated Ovaries of some immature females (1- to 3-week-old) were fixed and processed for electron microscopy (Okada et al 1986) to examine the status of the oocyte surface (e.g. the presence or absence of oolemma microvilli and zona pellucida).

Mature oocytes were collected from oviducts of superovulated females between 15 and 17 h after injection of human chorionic gonadotrophin (Fleming and Yanagimachi, 1980). Eggs at the pronuclear stage and embryos at 2-, 4- and 8-cell stages were collected by flushing the oviducts and/or uteri of naturally mated females without gonadotropin treatment (for the time relationship between mating and embryonic development, see Sato and Yanagimachi, 1972). Oviductal oocytes were first freed from the cumulus cells by treating them for 10 min with TC199 containing 0.1 % bovine testicular hyaluronidase (300 USP units mg⁻¹; ICN Biochemicals, Costa Mesa, CA) They were then treated with 2.5 mg ml⁻¹ collagenase (330 CD units mg⁻¹) for 6-9 min to remove the zona pellucida Zona-free oocytes thus obtained were thoroughly rinsed and kept in TC199 for 10-30 min before they were inseminated. Pronuclear eggs and preimplantation embryos were already free from cumulus cells. Therefore, hyaluronidase treatment was unnecessary prior to the collagenase treatment.

Zona-free oocytes/eggs/embryos were placed in 300μl of fresh TC 199 medium which had been previously placed under mineral oil (Squibb and Sons, Princeton, NJ) in a plastic Petn dish Insemination was carried out by adding a drop of sperm suspension containing actively motile, acrosome-reacted spermatozoa. The final concentration of spermatozoa was approximately S×10⁴ ml⁻¹. The medium, mineral oil and dishes were all kept at 37°C before, during and after insemination.

When an acrosome-reacted hamster spermatozoon attaches to the plasma membrane of the mature unfertilized oocyte, its vigorous tail movement stops abruptly in about 20 s. This is the earliest visible indication of sperm-oocyte fusion (Yanagimachi, 1977, 1988b). To determine the time between sperm attachment to the oocyte and cessation of sperm tail movement, several oocytes (10 to 80 μm m diameter) were individually placed in a drop (10 μl) of TC199 medium under mineral oil in a plastic dish. The medium, mineral oil and dishes were all kept at 37°C before and during insemination. Upon insemination of an oocyte, the dish was brought onto the warm (37 °C) stage of a phase-contrast microscope (× 100) to examine for the interaction of spermatozoa and the oocyte. The time interval between the contact of the first and/or second spermatozoon(-oa) to an oocyte and complete cessation of their tail movement was recorded. After incubation for about 1h at 37 °C, each oocyte was reexamined for the number of attached spermatozoa, and then pipetted five times to see if these spermatozoa could be removed from the oocyte surface. The inner diameter of the pipette was about twice as large as the diameter of the oocyte. Some oocytes were processed for transmission electron microscopy (Usui and Yanagimachi, 1976) for evidence of sperm incorporation into the ooplasm.

The dye transfer technique (Hinkley et al. 1986; StewartSavage and Bavister, 1988) was also used to detect spermoocyte fusion. Oocytes, pronuclear eggs or preimplantation embryos were incubated in TC199 medium containing 1 μgml⁻¹ Hoechst 33342. After incubation for 30 mm at 37°C, they were rinsed thoroughly in TC199 and inseminated with acrosome-reacted spermatozoa as described previously 5 min later, the oocytes (or pronuclear eggs or embryos) were fixed for 5 mm with 2.5 % glutaraldehyde in 0 1 M cacodylate buffer (pH 7.4). They were rinsed with the buffer, mounted between a slide and coverslip, and examined with a UV microscope for evidence of sperm-oocyte fusion. An oocyte, pronuclear egg or embryo was recorded as ‘fused’ when at least one brightly fluorescent sperm nucleus was seen in the cortex The validity of the dye transfer technique was tested by inseminating mature zona-free oocytes with uncapacitated, acrosome-intact spermatozoa, which are known to be incapable of fusing with the oocytes (Yanagimachi and Noda, 1970; Yanagimachi, 1981, 1988a, b).

Zona-free ovarian oocytes were grouped according to their size (diameter) and exposed to solubilized lectins to see if lectin-binding properties of the plasma membrane change as the oocyte grows. Lectins tested were: Concanavahn A (ConA), Pisum sativum agglutinin (PSA), wheat germ agglutinin (WGA) and Ricinus communis agglutinin (RCA 120) Since these lectins are known to bind to the oolemma of mature oviductal oocyte (Yanagimachi and Nicolson, 1976), agglutination of immature ovarian oocytes with these mature oocytes in the presence of lectins must indicate the presence of lectin-binding sites on the testing ovarian oocytes.

Agglutination was scored according to the method of Yanagimachi and Nicolson (1976). Briefly, one zona-free oviductal oocyte (80 μm in diameter) and five ovarian oocytes of any one specified size range (e.g. 20–30 μm) were placed in serially diluted solutions of lectins in Tris-buffered sahne (0 8% NaCl-10mM Tns-HCl, pH 7.4) containing 4 mg ml⁻¹ BSA. 5 min later, all the oocytes were aggregated by pushing them together with a needle. 10min later, the dish was agitated to disperse the oocytes. Agglutination was scored (++) when all of the smaller oocytes remained firmly attached to the large oocyte despite repeated agitation, (+) when there were more agglutinated than unagglutinated, (±) when there were more unagglutinated than agglutinated, and (−) when all the oocytes dispersed completely. Binding of FTTC-conjugated ConA to the oocyte surface was examined according to Yanagimachi and Nicolson (1976).

Oviductal oocytes were freed from zonae pellucidae with 2.5 mg ml⁻¹ collagenase. Some of these zona-free oocytes were treated for 30min at 25–30°C with 1 mg ml⁻¹ trypsin (bovine pancreatic, 2×cryst., 3000 NF units mg⁻¹; ICN Biochermcals) in TC199, and then exposed to FITC-ConA (0.2mg ml⁻¹) for 15 min to determine whether the trypsin-treated oolemma remains capable of bmding with ConA. Other oocytes were exposed first to FITC-ConA for 15 min then to trypsin for 30 min to see if trypsin could remove ConA that had been bound to the oolemma.

Oviductal oocytes were freed from zonae pellucidae by 2 min treatment with 0.1% trypsin. A group of zona-free oocytes was treated further for 30 min at 25–30 °C with either 1mg ml⁻¹ trypsin or 1mgml⁻¹ pronase (84 units mg⁻¹; Calbiochem., La Jolla, CA) in TC199 medium. Another group of zona-free oocytes was treated for 15 mm with 0 2 mg ml⁻¹ ConA The oocytes were then thoroughly rinsed and inseminated as described. Sperm-oocyte fusion was assessed 1 h later by determining the number of decondensed sperm nuclei m each oocyte.

The morphology of maturing ovarian oocytes is shown in Fig. 1. Ovaries of 1-week-old females contained many small primary oocytes (10–20μm in diameter) and relatively fewer larger oocytes (20–26 μm in diameter) (inset of Fig. 1). Only rarely, smaller oocytes were surrounded by follicle cells or had zona pellucida and microvilli (Fig. 1A). All of the oocytes larger than 20μm in diameter were surrounded either partially or completely by a single layer of loosely arranged follicle cells. Ovaries of 2-week-old females contained many oocytes in various stages of growth (insets of Fig. 1B-D). In addition to small oocytes without follicle cells, there were many larger oocytes surrounded by one to three layers of cuboidal follicle cells. Those surrounded by a single layer of the follicle cells measured 20–30 μm in diameter. Formation of the microvilli and zona pellucida was about to begin (Fig. 1B) or in progress (Fig. 1C). Oocytes with two to three layers of follicle cells, as large as 45 μm in diameter, had numerous microvilli and a distinct zona pellucida (Fig. 1D). Not until about three weeks after birth (weaning) did fully grown oocytes (80 μm in diameter) appear in the ovary. The ovaries of these weaning and sexually mature females contained oocytes of all sizes, ranging from about 15 μm to 80 μm in diameter.

When zona-free oviductal oocytes (80 μm in diameter) were inseminated, acrosome-reacted spermatozoa came into contact with oocyte surface one by one. The head of each spermatozoon attached firmly to the oocyte surface, and tail continued to beat vigorously. In about 20 s, the tail movement ceased rather abruptly. Table 1 summarizes the response of spermatozoa to zona-free ovarian oocytes of various sizes. When the smallest primary oocytes (10–15 μm in diameter) were inseminated, most of the spermatozoa that collided with the oocyte swam away. Some spermatozoa remained on the oocyte surface, beating their tails vigorously during 10 min of observation. When examined 1 h later, either none or only a few spermatozoa were seen on the oocyte surface. Those on the oocyte surface were removed readily by pipetting. Spermatozoa could attach firmly to larger oocytes. They stopped their tail movement soon after attachment. As the size of the oocyte increased, the number of attached spermatozoa increased and the time interval between the initial sperm-oocyte contact and the cessation of sperm tail movement decreased (Table 1).

Sperm-oocyte fusion was confirmed by either the dye transfer technique or electron microscopy. When oviductal zona-free oocytes were loaded with Hoechst 33342, inseminated with acrosome-reacted spermatozoa, fixed and examined with a UV microscope, many sperm nuclei fluoresced (Fig. 2) Acrosome-intact spermatozoa, which could not fuse with oocytes (Yanagimachi and Noda, 1970; Yanagimachi, 1981, 1988a,b), did not fluoresce under the same conditions (Fig. 3). None of the smallest oocytes (less than 15 μ in diameter) showed evidence of sperm-oocyte fusion, although spermatozoa occasionally attached to their surfaces. Fig. 4A-E show that all of the growing oocytes can fuse with spermatozoa. Fig. 5, one of 16–20μm oocytes inseminated in vitro, shows a sperm nucleus clearly within the ooplasm. The number of spermatozoa fusing with an oocyte was correlated with oocyte size, more spermatozoa fusing with larger oocyte. Even pronuclear eggs and blastomeres of 2- to 4-cell embryos could fuse with spermatozoa (Fig. 4F-H), although the number of spermatozoa fused with these cells was far less than that of spermatozoa fused with mature oocytes. Blastomeres of 8-cell embryos were unable to fuse with spermatozoa (Fig. 4I).

Table 2 summarizes the results of two experiments. In the first experiment, growing oocytes of various sizes (20–60 μm in diameter) were mixed with either fully grown ovarian oocytes (80 μm) or oviductal oocytes (80 μm) and inseminated in two different dishes, but using the same sperm suspension at the same sperm concentration. In the second experiment, oviductal oocytes, pronuclear eggs and 2-cell embryos were inseminated in the same dish. These two experiments were performed on different days using spermatozoa from two different males. This table shows that the fusibility of the oolemma increases as the oocyte grows and is reduced drastically after fertilization. The difference in the mean number of spermatozoa that fused with an oviductal oocyte (7.5 in Exp. I and 11.6 in Exp. II) was most likely due to the difference in the concentration of acrosome-reacted spermatozoa in insemination media.

Changes in the surface properties of growing oocytes were also detected by a Con A-mediated agglutination test. The ability of oocytes to agglutinate by ConA was increased as they grew (Table 3), indicating that Con A-binding sites on the oocyte plasma membrane increase with oocyte growth. Contrary to this, the density of PSA-, WGA- and RCA-binding sites tended to decrease with oocyte growth (data not shown). Since ConA detected changes in the surface properties of the growing oocyte most clearly, binding of FITC-conjugated ConA to growing oocytes was then studied. As shown in Fig. 6, Con A-bindmg to the oolemma increased in proportion to the size of the oocyte (Fig. 6A-D). The oolemma maintained the Con A-binding ability after fertilization (Fig. 6E-G), but lost it almost completely by the 8-cell stage (Fig. 6H).

Oviductal oocytes treated with ConA, trypsin or pronase were fully capable of fusing with spermatozoa (Table 4). ConA that had bound to the oolemma of oviductal oocytes was not removed by trypsin or pronase treatment.

In this study, we assessed sperm-oocyte fusion by light and electron microscopy. The presence or absence of a sperm nucleus(-ei) within the ooplasm, seen by electron microscopy, was the most convincing way to evaluate sperm fusion. The light (phase-contrast) microscopic approach using living oocytes was simpler and quicker, but it required certain precautions. Although the presence of decondensed sperm nucleus(-ei) in the oocyte can be considered unequivocal evidence of spermoocyte fusion, the absence of decondensed sperm nucleus(-ei) does not necessarily imply its failure. For example, the nuclei of spermatozoa that have fused with immature oocytes at the germinal vesicle stage remain condensed (Usui and Yanagimachi, 1976) Firm sperm attachment followed by ‘sudden’ cessation of sperm tail movement appears to be good, though indirect, evidence of fusion. In fact, only when spermatozoa stopped their tail movement within 1–3 min of contact with the oolemma, were sperm nuclei detected within the ooplasm by electron microscopy. In the present study, we routinely used the dye (Hoechst 33342) transfer technique which was originally developed by Hinkley et al. (1986) for sea urchin gametes. It proved to be a simple and reliable technique for detection of sperm-egg fusion here, provided that the dye-loaded oocytes/eggs/embryos were fixed within a short time (e.g. 5 min) after insemination. This short interval is of critical importance, at least for the hamster, because the dye in the living oocyte/pronuclear eggs/embryos diffuses slowly into the surrounding medium, staining the nuclei of unfused spermatozoa. Perhaps, the hamster oocytes lose the dye much faster than the oocytes of other species like the mouse (Conover and Gwatkin, 1988).

It was somewhat unexpected that hamster oocytes gained the ability to fuse with spermatozoa as soon as they started to grow. The smallest oocytes that were capable of fusing with spermatozoa were about 20 μm in diameter. These oocytes had just begun to form microvilli and to secrete zona pellucida material (see Fig. 1B). The number of spermatozoa fusing with each oocyte increased as the oocyte grew (Table 2). This is expected as the surface area of the oocyte increases with the growth of the oocyte. We think that the fusibility of the oolemma increases exponentially as the oocyte grows because the number of spermatozoa fusing with a relatively small oocyte (e.g. 25 μm in diameter) remained almost constant (1 to 2) regardless of sperm concentration in the insemination medium, whereas it increased dramatically in larger oocytes (e.g. >40μm in diameter) as sperm concentration was increased (unpublished preliminary data).

It is unclear how the oolemma gains the ability to fuse with spermatozoa. The appearance of microvilli coincides with the acquisition of fusiblity of the oolemma, but we do not think that they are directly related. The oolemma of >8-cell hamster preimplantation embryos has numerous microvilli (Grant et al. 1977), yet it is unable to fuse with spermatozoa (the present study). It is possible that some specific molecule (protein, glycoprotein or glycolipid) is synthesised by the oocyte or/and granulosa cells and inserted in the oolemma during the growth phase of the oocyte to make the oolemma competent to fuse with spermatozoa. As the oocyte secretes zona molecules to retain spermatozoa in the initial phase of fertilization (Grave et al. 1982; Wassarman, 1988,1990), it may insert a component into the oolemma to ‘hold’ the spermatozoa that come into contact with it after passing through the zona pellucida and perivitelhne space. It will be the subject of future investigations to determine (1) if such sperm-holding substances really exist on/in the oolemma, (2) if so, what is the chemical nature of such substances, (3) how they are inserted in the oolemma, and (4) when and how the genes governing the production of such molecules are turned on and off during oogenesis.

In the present study, a close correlation was evident between oolemma’s Con A-binding ability and fusibility, both being first detected in approximately 20 μm diameter oocyte and become undetectable by the 8-cell stage of embryonic development. Thus, the Con A-binding molecule and the fusion-mediating molecule might be the same. If this is the case, then masking of Con A-binding sites with ConA might prevent spermegg fusion. However, as we already know, masking Con A-binding sites with ConA does not prevent the oolemma from fusing with spermatozoa (Nicolson et al. 1975; Gordon and Dandekar, 1976; confirmed in this study). It is conceivable that as ConA binds to and masks the putative fusion-mediator, the lectin then acts as a substitutive fusion-mediating molecule. Since both hamster oocyte and sperm plasma membranes carry Con A-binding sites (Nicolson et al. 1975; Yanagimachi and Nicolson, 1976; Kinsey and Koehler, 1978), binding of the lectin to both the oocyte and sperm plasma membranes could facilitate fusion rather than preventing it. Perhaps, ConA is not specific enough to identify the putative fusion-mediating molecules on/in the oolemma. We will have to find a more specific probe for identification of the molecule in question. It is interesting that treatment of oocytes with proteinase (trypsin or pronase) affects neither Con A-binding ability nor fusibility of the oolemma (Table 4). Apparently, proteinases cannot remove ConA that has already bound to the oolemma. Therefore, it is conceivable that (a) the protein moiety of the putative fusion-mediating molecules is ‘cryptic’ (embedded in membrane lipid bilayers) and not easily accessible to proteinase digestion, or (b) the molecules in question are glycolipids rather than glycoproteins.

Soon after sperm fusion, cortical granules of the mature oocyte are released by exocytosis. The granules contain hydrolytic enzymes and saccharide components (Gulyas, 1980), and at least a part of its material covers the surface of the oolemma of fertilized eggs (Lee et al. 1988; Cherr et al. 1988). While this could be responsible, in part, for the reduced fusibility of the oolemma after fertilization, destruction (digestion) or modification of the putative fusion-mediating molecules by cortical granule enzymes and/or saccharides is another possibility.

In the rabbit, the oolemma becomes completely refractory to further spermatozoa after its fusion with the first (fertilizing) spermatozoon. In other words, the oolemma of the rabbit oocyte has a very strong and rapid polyspermy-blocking mechanism (Austin, 1961). The reduction or loss of the fusibility of oolemma after fertilization is also reported in many other species including the hamster (Cherr et al. 1988; StewartSavage and Bavister, 1988; this study). However, as shown by Usui and Yanagimachi (1976) and confirmed in the present study, the oolemmae of fertilized hamster eggs and even those of 2- to 4-cell embryos are still capable of fusing with spermatozoa when the eggs and embryos are challenged by many acrosome-reacted spermatozoa. Apparently, covering the oolemma with cortical granule material is not enough to render the oolemma completely refractory to excess spermatozoa, at least for the hamster. Only after the embryos had reached the 8-cell stage did the oolemma of hamster embryo becomes totally refractory to spermatozoa. It is likely that the putative fusion-mediating molecules are lost permanently from the oolemma of developing embryos because no embryonic cells from more advanced stages of development can undergo true membrane fusion with spermatozoa.

This work was supported by a grant from National Institute of Child Health and Human Development (HD-03402). We are indebted to Dr Gilbert S. Greenwald (University of Kansas), Dr J Michael Bedford (Cornell University) and Dr Thomas T. F. Huang (University of Hawaii) for their in valuable advice.