Intracellular free Ca²⁺ changes during physiological polyspermy in amphibian eggs

Cell activation or reinitiation of the cell cycle in response to various external signals (hormones, growth factors, neurotransmitters, mitogens, sperm, for instance) is in most cases marked by a transient increase in the activity of intracellular free calcium ions ((Ca²⁺]_i) (recently reviewed by Berridge and Irvine, 1989; Meyer, 1991). That calcium response is thought to represent a ubiquitous second messenger. Fertilization certainly represents the best system for studying these intracellular Ca²⁺ transients. Indeed, during fertilization, cell activation is initiated at a single particular point, that of the fusion between the fertilizing sperm and the egg. A transient increase in [Ca²⁺]_i accompanying egg fertilization (or artificial activation) has been recorded in various species (see references in the Discussion section). In fact, it appears that all deuterostome eggs display a transient Ca²⁺ wave at fertilization (reviewed by Jaffe, 1983; 1985). In most cases, the egg Ca²⁺ waves have been analyzed during monospermic fertilization, because it is the normal way of reproduction in all species studied so far. Groups that exhibit physiological polyspermy have polyploidy-preventing mechanisms operating inside the egg, rather than blocks to polyspermy acting at the egg’s outside as in physiologically monospermic species (reviewed by Jaffe and Gould, 1985; Jaffe and Cross, 1986). Amphibians represent a very interesting system since they have both styles of fertilization: anurans (frogs and toads) have monospermic eggs, while urodeles (salamanders and newts) have polyspermic eggs (reviewed by Elin-son, 1986). So far, [Ca²⁺]₁ variations associated with amphibian egg fertilization have been measured in some anurans but not in urodeles, with the exception, however, of a report in an abstract form by Kubota et al. (1989) who, using aequorin, observed a calcium release in the egg of Cynops pyrrhogaster.

Both the eggs of the anurans Xenopus laevis (Busa and Nuccitelli, 1985) and Discoglossus pictus (Nuccitelli et al., 1988) displayed a transient 3-fold increase in [Ca²⁺]_i at the time of fertilization. In anuran eggs, the period of increased [Ca²⁺]_i matches the period of cortical granule exocytosis quite well (Busa and Nucci-telli, 1985; Nuccitelli et al., 1988). In addition, the Cl^- conductance increase underlying the depolarizing phase of the fertilization potential in anuran eggs requires an increase in [Ca²⁺]_i (see references in Nuccitelli et al., 1988). It should be noted that both the fertilization potential (Cross and Elinson, 1980) and cortical granule exocytosis (Wyrick et al., 1974) serve as blocks to polyspermy in anuran eggs. One might postulate, on this basis, that urodele eggs do not need to develop a fertilization-associated Ca²⁺ transient, since they lack both types of block to polyspermy (Wartenberg and Schmidt, 1961; Charbonneau et al., 1983). However, urodele egg activation appears to depend on intracellular Ca²⁺, since it can be triggered by the calcium ionophore A23187 (Charbonneau et al., 1983), a universal egg activator (Steinhardt et al., 1974). In the present work, we report that polyspermic fertilization of Pleurodeles eggs is, in a substantial number of cases (28 out of 45), accompanied by a slow increase in [Ca²⁺]₁, around 0.15 μM from a resting level of 0.41 μM, taking place around 15 minutes after fertilization. Unlike that in anuran eggs, the increase in [Ca²⁺]_i associated with urodele egg fertilization was permanent or, in some cases, only slowly reversible. In addition, its amplitude was independent of the number of sperm interacting with the egg. These new data are discussed with respect to the corresponding changes in physiologically monospermic eggs and their possible function in regulating cell activation-associated events.

Mature individuals of Pleurodeles waltlii, purchased from the Centre d’Elevage des Amphibiens (CNRS, Université de Toulouse, France), were reared in our laboratory and fed daily on Trouw pellets, a complete food for fish (Trouvit, France). Females were induced to ovulate following injection of 600-800 international units of human chorionic gonadotropin (Organon, France). The first eggs were laid around 24 hours after injection, at 25-26°C, while the last eggs that could be used for successful fertilization were laid around 36 hours later. Eggs were either externally expressed by gentle squeezing of the female’s flanks or carefully picked up in the water tank immediately after laying. Jellied eggs were allowed to adhere onto the bottom of 4 ml tissue culture plastic dishes (60 × 15 mm) with a center well (Falcon, USA), serving as an experimental chamber for ion activity measurements (see below), for about 5 minutes, and immersed in Fl solution. The physiological solution used throughout the study was Fl solution, composed as follows (mM): 31.2 NaCl, 1.8 KC1, 1.0 CaCl₂, 0.1 MgCl₂, 2.0 NaHCO₃, 1.9 NaOH, buffered at pH 7.4-7.5 with 10.0 mM Hepes. Male spermiducts were surgically removed and stored in 4-fold concentrated Fl solution, at 4°C. Pure sperm fluid was obtained by squeezing, with forceps, a small portion of a spermiduct and was used immediately, either undiluted or diluted in Fl solution.

Unactivated jellied eggs were each impaled with a Ca²⁺-selective microelectrode and a membrane potential microelectrode. Ca²⁺-microelectrodes were made and calibrated as previously described (Grandin and Charbonneau, 1991a). The resin, contained in the microelectrode tip, used to detect intracellular Ca²⁺ activities, was a ready-to-use Ca²⁺ sensor (calcium ionophore I, Cocktail A; Fluka Chemical Corp., Switzerland), it is important to note that in order to measure [Ca²⁺]₁ with selective microelectrodes, it is necessary to impale each of the eggs with two microelectrodes: a potential microlectrode measuring only the membrane potential (Em) and a Ca²⁺-microelectrode measuring the intracellular activity of Ca²⁺ (pCa, the negative logarithm of intracellular free Ca²⁺ activity) plus the membrane potential. Membrane potential was subtracted from the total signal recorded by the Ca²⁺-microelectrode at the pen recorder input. Membrane potentials and ion-specific signals were recorded using high input amplifiers (Burr Brown, OPA 104) and connected to the ground via an F1 agarose bridge. Additional details concerning microelectrode impalement and the electrophysiological set-up can be found in Grandin and Charbonneau (1991b). After stabilization of both the electrical (Em traces, in all figures) and ionic parameters (pCa traces, in all figures), jellied eggs were inseminated as described below.

Sperm (pure or diluted in Fl solution) was added around the impaled eggs, in the recording chamber, using a P20 Pipetman. Depending on the concentration of living and motile sperm (assessed with a light microscope prior to insemination) for a given sperm sample, eggs were inseminated with various volumes of sperm suspension. Generally, addition of 20-30 μl of a 5- to 10-fold dilution of the spermiduct-expressed sperm suspension on top of the jelly layers of 5 or 6 eggs was sufficient to produce the appearance of about 10 sperm entry sites on each of these eggs. Addition of undiluted sperm under the same conditions could increase that number of sperm entry sites up to 20-30, or even more when all sperm was added on top of a single egg. Quality of the sperm, its rate of dilution in Fl solution, as well as the distance between the site of sperm addition and the egg, represented the three parameters that were found to influence polyspermy. The degree of polyspermy was evaluated by counting, under a stereomicroscope, the number of sperm entry sites at the egg surface, after mechanically dejellying with forceps. Successful egg activation was assessed by visualizing, in vivo under a stereomicroscope, the disappearance of the maturation spot (concomitant with blurring out of the extremity of the metaphase 2 spindle, a black dot in the center of the maturation spot) and a change of pigmentation from a rough, irregular appearance to a smooth appearance (see references in Elinson, 1986; see also Charbonneau et al., 1983), both events occurring 45-60 minutes after fertilization or artificial activation, at 21-23°C. Successful fertilization was assessed by visualizing, under a stereomicroscope, the sperm entry points appearing at the surface of the eggs. These could be detected around 5 minutes after insemination as large craters of about 15 μm in diameter, progressively expanding their size up to 45 μm and remaining visible for at least 3 hours (Picheral, 1977; Charbonneau et al., 1983). When further embryonic development proceeded normally, first cell division took place around 6 hours after insemination, at 21-23°C, followed by more rapid cell divisions every 90-100 minutes. Subsequent embryo development was assessed by comparing with previously described events (Gallien and Durocher, 1957).

In all polyspermic urodele species so far studied, 3 to 15 sperm have been found to penetrate each egg following natural insemination, that number sometimes appearing to be somewhat characteristic of each species (see references in Elinson, 1986). In Pleurodeles waltlii, 1 to 12 sperm have been seen to reach the egg surface following natural insemination (Picheral, 1977). It is also possible to inseminate eggs artificially under conditions resulting in a normal embryonic development. Amphibian eggs can still be successfully inseminated until around 1 hour following immersion in Fl solution (Hollinger and Corton, 1980), a period of time sufficient for stabilization of electrical and ionic parameters. Under the conditions used in this study, the first sperm entry points (see Materials and methods) were visualized 5-15 minutes after insemination, a 5-10 minute delay corresponding to a situation in which the first sperm reach the egg 2-3 minutes after insemination (see Picheral, 1977; Charbonneau et al., 1983). Various degrees of polyspermy (up to around 60 sperm per egg) could be obtained experimentally (see Materials and methods). In the present study, it was repeatedly observed, in all egg batches, that embryos displaying more than around 30 sperm entry sites failed to cleave subsequently. On the other hand, production of 1 to 30 sperm entry sites per egg resulted in normal embryonic development.

The [Ca²⁺]₁ level in unactivated jellied eggs of Pleurodeles impaled with two microelectrodes was found to be 0.41±0.13 μM (Standard Deviation, n=89). Two types of intracellular Ca²⁺ response to fertilization were observed in polyspermic Pleurodeles eggs. In 28 out of 45 cases in which fertilization took place (appearance of sperm entry sites at the egg surface and/or occurrence of cell division) [Ca²⁺]_i began to slowly increase 14.0±8.2 minutes (s.d., n=28) after insemination, reaching a plateau level 23.3±9.3 minutes (s.d., n=28) after insemination (Fig. 1). [Ca²⁺]_i did not promptly return back to the original level (Fig. 1). The amplitude of the fertilization-associated increase in [Ca²⁺]_i was 0.15±0.06 μM (s.d., n=28). In the 17 other cases, however, there was no detectable changes in [Ca²⁺]_i during at least 1.5 hours following insemination (Fig. 2). In control experiments in which eggs were inseminated with dead sperm or did not fertilize following insemination, there was no change in [Ca²⁺]_i (data not shown). As previously reported (Charbonneau et al., 1983; Iwao, 1985), there was no major change in the egg membrane potential during Pleurodeles fertilization (Figs 1, 2), which remained close to the level in the unactivated egg (–12.6±1.6 mV, s.d., n=89).

Both the eggs that displayed an increase in [Ca²⁺]₁ and the eggs that did not display such an increase, went on to divide and develop normally with similar efficiency. Thus, among the 17 eggs that did not show changes in [Ca²⁺]i at fertilization, all 13 in which embryonic development was followed developed normally (8 were followed during the first two or three cell divisions, 5 until the blastula stage) and displayed less than 30 sperm entry sites (6 eggs with 1-7 sperm entry sites, 2 with 14 and 30, and 9 eggs in which sperm entry sites, detected during recording, were clearly less than 30, but were not precisely counted afterwards). Embryonic development was followed in 16 of the 28 eggs of the other class, those showing an increase in [Ca²⁺]_i at fertilization. Eight of these 16 eggs developed normally until the blastula stage (all with less than 30 sperm entry sites), while the remaining 8 eggs failed to reach the first cell division (4 had less than 30 sperm entry sites, 4 other 30-60 sperm entry sites). Tlie only noticeable difference between the two classes of eggs was that all the 17 fertilized eggs lacking an increase in [Ca²⁺]_i had fewer than 30 sperm entry sites, whereas all 4 eggs with more than 30 sperm entry sites (and which did not develop as a consequence) displayed an increase in [Ca²⁺]_i at fertilization. Therefore, these highly polyspermic eggs represented the only subclass that could be found in the class of eggs showing an increase in [Ca²⁺]₁, but not in the class of eggs showing no change in [Ca²⁺]_i.

To detect any possible relationship between the number of sperm entry sites and the amplitude of the increase in [Ca²⁺]_i, data were arbitrarily classified according to the number of sperm entry sites counted on each egg (only those numbers that were precisely determined were taken into account). Comparison of these various classes clearly shows an absence of correlation between the degree of polyspermy and the amplitude of the fertilization-associated increase in [Ca²⁺]_i (Table 1).

The possible existence of a Ca²⁺ wave in Pleurodeles eggs was investigated by impaling single eggs with two Ca²⁺-microelectrodes (and two potential microelectrodes) and comparing the fertilization-associated [Ca²⁺]_i changes recorded by each of the two Ca²⁺-microelectrodes. In one such experiment, there was no increase in [Ca²⁺]₁ at either microelectrode. In three other experiments, [Ca²⁺]i increased (by 0.07, 0.09 and 0.15 μM) at only one of the two microelectrodes (Fig. 3A). In the remaining three experiments, [Ca²⁺]_i increased at both microelectrodes, but, more importantly, each time with a delay between the times of detection at each of the two microelectrodes. In one of these cases, the fertilization-associated increase in [Ca²⁺]_i had the same amplitude at both microelectrodes (0.06 μM), but began 7.8 minutes after insemination at one microelectrode and 19.2 minutes after insemination at the other microelectrode (Fig. 3B). In two other experiments, a 0.18 μM increase was detected at one microelectrode 6.0 and 9.0 minutes after insemination, respectively, and 0.25 μM and 0.13 μM increases were detected at the second microelectrode 13.2 and 22.2 minutes after insemination, respectively. However, due to the existence of numerous sperm entry sites, it was not possible to determine temporally and spatially the initiation of the [Ca²⁺]_i changes, with the exception of one experiment in which the first sperm entry sites, detected during recording, appeared very close to the Ca²⁺-microelectrode that first recorded the increase in [Ca²⁺]₁ (Fig. 3B).

From the experiments described above, it could not be determined whether the increase in [Ca²⁺]_i measured in the second place represented the same increase as that measured in the first place, spreading as a wave, or rather was triggered by different sperm with a spatially distinct location. To distinguish between the two possibilities, eggs were locally inseminated by positioning a micropipette, containing sperm and mounted on a micropositioner, very near one of the two Ca²⁺-microelectrodes. In 8 out of 14 such experiments, an increase in [Ca²⁺]_i (0.10±0.06 μM, s.d., n=8, from a basal level of 0.37±0.09 μM, s.d., n=8) was detected, 14.3 minutes (±5.6, s.d., n=8) after insemination, only in the region that was locally inseminated but not on the opposite side of the egg (Fig. 4), suggesting that the [Ca²⁺]_i changes cannot propagate, at least far from the site of initiation. In 2 of the 14 experiments, however, an increase in [Ca²⁺]_i first detected in the locally inseminated region was followed a few minutes later by a much smaller increase at the second microelectrode (Fig. 5). In these 2 experiments, this second increase in [Ca²⁺]_i, recorded 19.2 and 18.0 minutes after insemination, was in fact very small, 0.02 and 0.03 μM, respectively, and followed by 5.1 and 7.8 minutes, respectively, the increase detected in the first place (respectively 0.06 and 0.12 μM). It would be extremely hazardous to speculate that these [Ca²⁺]₁ increases detected in the second place reflect the existence of a propagating Ca²⁺ wave. Indeed, 0.02 μM changes in the [Ca²⁺]_i level would be barely detectable in normal conditions (see Fig. 5). The fact that these small increases occur at the right time following insemination may open the question concerning their real origin. However, there is nothing more that can be done, using the present technology, that might close the debate. Also, due to technical difficulties in realizing such experiments, one cannot totally rule out the possibility that some sperm do not reach very precisely the region of the egg they were assigned to by the experimentator. It should be noted, too, that in the experiment shown in Fig. 5 the distance between the two Ca²⁺-microelec-trodes was shorter than in the experiment shown in Fig. 4. Finally, in 4 out of 14 locally inseminated eggs which either displayed sperm entry sites or subsequently cleaved, there was no change at all in [Ca²⁺]₁ at either microelectrode. Data from 6 other eggs, other than those discarded for technical reasons, could not be taken into account, since these eggs did not show any fertilization-associated [Ca²⁺]_i changes or sperm entry sites, and did not cleave. Among the 14 eggs considered, 4 went at least to the early bud stages (Gallien and Durocher, 1957; they continued to develop but were not individually followed afterwards), 4 arrested at the gastrula or early neurula stages, and 6 did not cleave but displayed a typical [Ca²⁺]_i increase and sperm entry sites. Injuries around the sites of microelectrode impalement, that healed more or less efficiently according to egg batches, were particularly accentuated in this series in which each egg had 4 microelectrodes in it, and were probably responsible for the failure of these 6 eggs to cleave.

In conclusion, we believe that there is no Ca²⁺ wave in Pleurodeles eggs according to the parameters that have been defined by previous authors in other systems (see Discussion). However, our experiments of local insemination of eggs impaled with two Ca²⁺-microelec-trodes do not totally eliminate the possibility of the existence of [Ca²⁺]_i changes that rapidly vanish in amplitude as they propagate over short distances from around the sites of sperm entry.

The present study is the first to report on changes in intracellular free calcium ([Ca²⁺]_i) in eggs during physiological polyspermic fertilization. The relatively late, slow, small and slowly reversible increase in [Ca²⁺]_i that we have found associated with polyspermic fertilization of the egg of the urodele amphibian Pleurodeles waltlii strongly contrasts with the earlier, relatively rapid, larger and transient increase in [Ca²⁺]_i measured during monospermic fertilization of the egg of the anuran amphibian Xenopus laevis (Busa and Nuccitelli, 1985). Relative to the duration of the first mitotic cycle, 1.5 hours in Xenopus and 6 hours in Pleurodeles, the respective times of occurrence of the increase in [Ca²⁺]_i, around 2 minutes and 15 minutes after fertilization, are comparable although not exactly similar. If, in Xenopus eggs, it is clear that the Ca²⁺ wave starts from around the site of entry of the fertilizing sperm (Busa and Nuccitelli, 1985), this is less evident in Pleurodeles eggs in which several sperm can enter at any place of the egg surface.

A transient increase in [Ca²⁺]_i accompanying egg fertilization (or artificial activation) has been recorded in various echinoderms (Steinhardt et al., 1977; Zucker et al., 1978; Eisen and Reynolds, 1984; Eisen et al., 1984; Poenie et al., 1985; Yoshimoto et al., 1985, 1986; Swann and Whitaker, 1986; Tsien and Poenie, 1986; Hafner et al., 1988; Chiba et al., 1990; Whitaker and Patel, 1990), ascidians (Speksnijder et al., 1989, 1990a,b), the medaka fish (Ridgway et al., 1977; Gilkey et al., 1978; Yoshimoto et al., 1985, 1986), anuran amphibians (Busa and Nuccitelli, 1985; Kubota et al., 1987; Nuccitelli et al., 1988) and mammals (Cuthbertson et al., 1981; Cuthbertson and Cobbold, 1985; Igusa and Miyazaki, 1986; Miyazaki et al., 1986). All the species studied so far have naturally monospermic eggs, with the exception of the anuran amphibian Discoglos-sus which has recently been reported to present a partial physiological polyspermy (Talevi, 1989). By its characteristics, the Discoglossus [Ca²⁺]_i increase is clearly of the Xenopus type, although its duration is longer than that in Xenopus (Nuccitelli et al., 1988).

Two observations support the view that Pleurodeles egg activation depends on some intracellular release of Ca²⁺ ions. The first one is that Pleurodeles eggs can be artificially activated following application of the calcium ionophore A23187 (Charbonneau et al., 1983). The second is the present observation that an increase in [Ca²⁺]_i associated with egg fertilization can be recorded in a majority of cases. There are two explanations for the observed absence of fertilization-associated [Ca²⁺]_i changes in some experiments. One is that there are indeed cases in which [Ca²⁺]_i does not change. The second is that some [Ca²⁺]; variations cannot be detected because they are too small. In three experiments, an increase in [Ca²⁺]₁ was recorded only in a given region of the egg but not in the neighbouring region (Fig. 3A). These experiments suggest that the increase in [Ca²⁺]_i either does not have the same amplitude all over the egg (in which case the increase might be undetectable in some regions because too small) or does not propagate all over the egg (in which case there would be some regions unattained by the [Ca²⁺]_i increase). Experiments in which local inseminations were performed support the second conclusion. Several characteristics argue against the classification of the Pleurodeles [Ca²⁺]₁ increase as a Ca²⁺ wave: it is too small (0.15 μM versus at least 1 μM in most systems); it is permanent or slowly reversible rather than transient; its rate of propagation, if it did propagate, would be extremely low (Fig. 5). In fact, both the absence of a measurable increase in [Ca²⁺]_i in some fertilized eggs and the observation of localized changes in [Ca²⁺]_i (Figs 3A, 4) strongly support the existence of several local, non-propagating increases in [Ca²⁺]_i, each possibly triggered by one of the multiple sperm interacting with the egg. Such a possibility has already been evoked, and sometimes demonstrated, in a few other species. Thus, in golden hamster eggs, polyspermy, induced upon reinsemination, was found to be associated with the occurrence of additional Ca²⁺ waves generated at the site of additional sperm entry (Miyazaki et al., 1986). Immature starfish oocytes, lacking polyspermy blocks, displayed a Ca²⁺ transient in which rapid successive peaks might have corresponded to multiple sperm entries (Chiba et al., 1990).

What can be the characteristics of the Pleurodeles egg that make it incapable of propagating an intracellular Ca²⁺ signal? In fact, the Pleurodeles egg appears to behave, as far as Ca²⁺ propagation is concerned, as an immature Xenopus oocyte. Indeed, in Xenopus, the cortex of the unactivated egg possesses an extensive network of cortical endoplasmic reticulum (Campanella and Andreuccetti, 1977; Grey et al., 1974; Charbonneau et al., 1986), absent from the immature oocyte and supposed to be determinant in the propagation of the Ca²⁺ wave (Campanella et al., 1984; Charbonneau and Grey, 1984). In Pleurodeles, the cortex of the unactivated egg contains numerous cortical endoplasmic reticulum vesicles, but clearly these vesicles are simple vesicles that lack any connection between them (Charbonneau and Picheral, 1985), unlike the cortical endoplasmic reticulum in Xenopus eggs which forms long cistemae interconnecting the cortical granules between them and also makes junctions with the plasma membrane and the subcortical endoplasmic reticulum (Charbonneau et al., 1986). This might account for the absence of increase in [Ca²⁺]_i in some regions of the Pleurodeles egg, for instance when most of the spermegg interactions take place far from the region that is being measured.

In anuran eggs, both blocks to polyspermy are highly dependent on an intracellular release of Ca²⁺ (reviewed by Charbonneau and Grandin, 1989). In fact, due to the absence of cortical granule exocytosis and of a fertilization potential, two very rapid Ca²⁺-dependent reactions, the urodele egg might not need to develop as rapid and massive an increase in [Ca²⁺]_i as the one present in anuran eggs. There are probably many other Ca²⁺-dependent reactions associated with egg activation. In particular, the inactivation of MPF, a universal M-phase Promoting Factor (reviewed by Nurse, 1990), which takes place a few minutes after egg activation in Xenopus (Gerhart et al., 1984), can be triggered in vitro upon simple addition of Ca²⁺ (Meyerhof and Masui, 1977; Masui, 1982). This suggests that the Ca²⁺ transient in Xenopus eggs might serve to trigger MPF inactivation. One of the main roles of the permanent increase in [Ca²⁺]_i in Pleurodeles eggs might also be to trigger MPF inactivation, a reaction that starts around 45 minutes after egg activation (Grandin et al., 1991). If this was the case, Pleurodeles eggs might permit a detailed study of MPF inactivation since there is a long delay, around 30 minutes, between the increase in [Ca²⁺]_i and MPF inactivation.

We thank Dr Ray Kado (Gif-sur-Yvette, France) for the design of the electronic recording system, and Mr Jean-Paul Rolland for building it. This work was supported by grants from the Ligue Nationale contre le Cancer (Comité Départemental d’Ille-et-Vilaine), the Fondation pour la Recherche Médicale, the Fondation Langlois and the Association pour la Recherche sur le Cancer.