An increase in intracellular Ca2+ concentration ([Ca2+]i) has been shown to drive sea-urchin embryos and some fibroblasts through nuclear-envelope breakdown (NEBD) and the metaphase-to-anaphase transition. Mitotic Ca2+ transients can be pan-cellular global events or localized to the perinuclear region. It is not known whether Ca2+ is a universal regulator of mitosis or whether its role is confined to specific cell types. To test the hypothesis that Ca2+ is a universal regulator of mitosis, we have investigated the role of Ca2+ in mitosis in one-cell mouse embryos. Fertilized embryos generate Ca2+ transients during the first mitotic division. Imposing a Ca2+ transient by photorelease of inositol (1,4,5)-trisphosphate [Ins(1,4,5)P3] resulted in acceleration of mitosis entry, suggesting that a [Ca2+]i increase is capable of triggering mitosis. Mitotic Ca2+ transients were inhibited using three independent approaches: injection of intracellular Ca2+ buffers; downregulation of Ins(1,4,5)P3 receptors; and removal of extracellular Ca2+. None of the interventions had any effects on the timing of NEBD or cytokinesis. The possibility that NEBD is driven by localized perinuclear Ca2+ transients was examined using two-photon microscopy but no Ca2+-dependent increases in fluorescence were found to precede NEBD. Finally, the second mitotic division took place in the absence of any detectable [Ca2+]i increase. Thus, although an induced [Ca2+]i increase can accelerate mitosis entry, neither cytosolic nor perinuclear [Ca2+] increases appear to be necessary for progression through mitosis in mouse embryos.

An increase in the intracellular concentration of Ca2+ ([Ca2+]i) has long been known to be the stimulus for cell-cycle resumption at fertilization (Fulton and Whittingham, 1978; Whitaker and Swann, 1993; Stricker, 1999). This unambiguous role for Ca2+ in cell-cycle progression at fertilization led to the discovery that Ca2+ plays a role in cell-cycle progression in several different cell types. Evidence supporting a role for Ca2+ in mitosis includes the fact that Ca2+ transients have been recorded at key mitotic transitions including, nuclear-envelope breakdown (NEBD) (Poenie et al., 1985; Kao et al., 1990; Tombes et al., 1990), the metaphase-anaphase transition (Poenie et al., 1986; Ratan et al., 1988; Tombes and Borisy, 1989; Groigno and Whitaker, 1998) and cytokinesis (Fluck et al., 1991; Snow and Nuccitelli, 1993; Chang and Meng, 1995; Muto et al., 1996). These cell-cycle Ca2+ transients are stimulated by cell-cycle-associated changes in the level of the Ca2+-mobilizing second messenger inositol-(1,4,5)-trisphosphate [Ins(1,4,5)P3] (Ciapa et al., 1994).

The case for a role for Ca2+ in mitosis is strongest in early sea-urchin embryos. Ca2+ chelators such as EGTA and BAPTA prevent Ca2+ release and block NEBD (Steinhardt and Alderton, 1988; Twigg et al., 1988). Conversely, NEBD can be induced by treatments that increase [Ca2+]i (Wilding et al., 1996; Twigg et al., 1988; Steinhardt and Alderton, 1988). Similar findings have been made at the metaphase-anaphase transition, which can be reversibly inhibited by Ca2+ buffers or by inhibiting Ins(1,4,5)P3 receptors (InsP3Rs) (Groigno and Whitaker, 1998). Inhibition of the Ca2+ signalling pathway by interfering with the activity of Ca2+/calmodulin-dependent protein kinase II results in similar phenotypes to Ca2+ buffers, implicating this kinase in transducing many of the effects of Ca2+ in mitosis (Baitinger et al., 1990; Torok et al., 1998). Thus, at least in sea-urchin embryos, there is a compelling body of evidence suggesting that Ca2+ is both necessary and sufficient for NEBD and anaphase onset (Whitaker and Larman, 2001).

In other cell types, some observations suggest that Ca2+ might not be a universal signal for driving mitosis. Most notably, mitosis occurs in the apparent absence of Ca2+ transients in some cell lines (Tombes and Borisy, 1989; Whitaker and Patel, 1990; Kao et al., 1990; Whitaker and Larman, 2001). Also, in cells that do generate mitotic Ca2+ oscillations, inhibition of the Ca2+ increases by intracellular Ca2+ buffers, removal of extracellular Ca2+ or removal of serum does not hinder cell division (Tombes and Borisy, 1989; Tombes et al., 1990; Kao et al., 1990).

One hypothesis which has been proposed to explain these apparent inconsistencies is that, in the absence of global Ca2+ transients, mitotic events are driven by spatially restricted Ca2+ microdomains that are inaccessible to conventional epifluorescence microscopy (see Hepler, 1994; Kao et al., 1990; Whitaker and Patel, 1990; Kono et al., 1996). This idea was supported by early reports that [Ca2+] is elevated at spindle poles before anaphase in PTK and endosperm cells (Keith et al., 1985; Ratan et al., 1986). More recently, using confocal imaging and ratiometric Ca2+ indicators, localized Ca2+ increases were detected in the perinuclear region just before NEBD in sea-urchin embryos (Wilding et al., 1996). Consistent with the ability to generate local Ca2+ signals is the observation that, shortly before NEBD, the endoplasmic reticulum aggregates around the nucleus (Terasaki, 2000). The presence of a large source and sink of Ca2+ around the nucleus might provide a basis for the generation of local Ca2+ gradients. These studies have therefore raised the possibility that Ca2+ is a universal regulator of mitosis but that it can act in a highly spatially restricted manner.

One-cell mouse embryos provide a useful model with which to investigate the role and mechanism of generation of mitotic Ca2+ transients. Several studies have reported Ca2+ oscillations during mitosis (Tombes et al., 1992; Kono et al., 1996; Day et al., 2000; Tang et al., 2000; Gordo et al., 2002). In some cases, these oscillations have been reported to continue through mitosis, albeit at a low and variable frequency (Tombes et al., 1992; Kono et al., 1996; Day et al., 2000; Marangos et al., 2004). Consistent with the results from sea-urchin embryos, AM loading with the Ca2+ buffer BAPTA inhibits NEBD (Tombes et al., 1990; Kono et al., 1996). However, recent studies have shown that the NEBD precedes the first mitotic Ca2+ transient and that parthenogenetic embryos produced by activating oocytes using strontium proceed through mitosis in the absence of any detectable Ca2+ transients (Kono et al., 1996; Marangos et al., 2003). By analogy with the sea-urchin embryo, one possible explanation for this discrepancy is that progression through mitosis in parthenogenetic embryos is driven by localized Ca2+ elevations.

In light of these recent observations, we have readdressed the role of Ca2+ in mitosis in mouse embryos. We used multiple strategies to disrupt or impose Ca2+ transients in order to test the role of Ca2+ in progression through mitosis in mouse embryos. In addition, we have used two-photon microscopy in an attempt to reveal evidence for localized Ca2+ release in parthenogenetic embryos. Our data suggest that neither global nor local Ca2+ release is necessary for mitosis in early mouse embryos. However, Ca2+ is sufficient to accelerate mitosis, a function that appears to be conserved from sea-urchin to mammalian embryos.

Collection and handling of embryos

One-cell embryos were recovered from 21-24-day-old female MF1 mice (Harlan, Bicester, UK) to which had been administered, by intraperitoneal injection, 7.5 international units (IU) pregnant mare's serum gonadotrophin and 5 IU human chorionic gonadotropin (hCG; Intervet) at a 48 hour interval. Females were mated with F1 males of proven fertility at the time of hCG administration. One-cell embryos were collected 24-28 hours after hCG administration. Embryos were released into HEPES-buffered KSOM (H-KSOM) (Lawitts and Biggers, 1993), which contains 95 mM NaCl, 2.5 mM KCl, 0.35 mM KH2PO4, 0.2 mM MgSO4, 10 mM sodium lactate, 0.2 mM glucose, 0.2 mM sodium pyruvate, 4 mM NaHCO3, 21 mM mM HEPES, 1.7 mM CaCl2, 1 mM glutamine, 0.01 mM tetrasodium EDTA, 0.03 mM streptomycin sulphate, 0.16 mM penicillin G and 1 mg ml-1 bovine serum albumin (BSA). Embryos were washed three times and transferred to a drop of H-KSOM under oil. All embryo manipulations were performed at 37°C. Two-cell-stage embryos were recovered 48 hours after hCG administration and mating by the same method. Experiments investigating the effects of Ca2+ buffers were performed using one-cell embryos that were obtained by in vitro fertilization as previously described (FitzHarris et al., 2003). To generate parthenogenetic embryos, metaphase-II-arrested oocytes were recovered from females that had not been mated and transferred to medium containing 7% ethanol at 25°C for 7 minutes, 18 hours after hCG administration. Oocytes were subsequently washed three times in fresh ethanol-free medium. Ca2+-free medium was used in some experiments and contained 1 mM EGTA in place of Ca2+ (Tombes et al., 1992).

Microinjection

Embryos were pressure injected using a micropipette and Narishige manipulators mounted on a Leica DM IRB inverted microscope (Leica, Wetzlar, Germany). Embryos were placed in a drop of H-KSOM covered with mineral oil to prevent evaporation. Cells were immobilized using a holding pipette while the injection pipette was pushed through the zona pellucida until contact was made with the oocyte plasma membrane. A brief overcompensation of negative capacitance caused the pipette tip to penetrate the cell. Microinjection was performed using a fixed-pressure pulse through a pico-pump (WPI, Sarasota, FL, USA). Injection volumes were estimated at 5% of total cell volume by cytoplasmic displacement.

Whole-cell fluorescence microscopy and photolysis of `caged' Ins(1,4,5)P3

[Ca2+]i was monitored using Fura-2/dextran or Fura-red (Molecular Probes, Eugene, OR, USA). Fura-2/dextran (10 kDa) was microinjected to an estimated final concentration of 2-4 μM. Indicator-loaded embryos were placed in a drop of H-KSOM under oil in a chamber and placed on an Axiovert microscope fitted with a 20× air objective lens (Zeiss, Welwyn Garden City, UK). Illumination was performed using a monochromator (TILL Photonics, Gräfelfing, Germany) to select appropriate wavelengths. Fura-2 was illuminated at 340 nm and 380 nm, and emitted light was collected using a 510 nm dichroic mirror and a 520 nm band-pass filter. Fluorescence was detected using a cooled CCD camera (MicroMax, Princeton Instruments). The monochromator, filter wheel and camera were all controlled using Metafluor software (Universal Imaging, Downington, PA, USA), and background subtraction was performed online. Ca2+ measurements are presented as the ratio of emission in response to illumination at 340 nm and 380 nm. To monitor the presence of nuclei during the second mitotic division, blastomeres were coinjected with Fura-2/dextran and a fluorescein-isothiocyanate-labelled BSA tagged with a nuclear targeting signal (FITC-NLS-BSA; kindly provided by M. Jackman, Gurdon Institute, University of Cambridge, Cambridge, UK). Experiments examining the second mitotic division were performed using an excitation filter wheel; Fura-2 was excited at 340 nm and 380 nm, and fluorescein at 490 nm. As before, emitted light was collected with a 520 nm band-pass filter. Fura-red was used to monitor [Ca2+]i in experiments involving flash photolysis (see below). Embryos were loaded with 4 μM Fura-red-AM for 10 minutes. Furared was illuminated at 427 nm and 490 nm, and emitted light collected with a 600 nm long-pass filter.

'Caged' Ins(1,4,5)P3 [cIns(1,4,5)P3; Molecular Probes] was used to trigger Ca2+ release in one-cell embryos. cIns(1,4,5)P3 was microinjected to an estimated final concentration of 50 μM. Photorelease was performed 30-60 minutes after microinjection by a brief, timed illumination at 360 nm. The duration of ultraviolet (UV) exposure was controlled using the Metafluor software. We and others have shown previously that using this protocol, repeated UV exposures trigger Ca2+ transients of similar amplitude (Jones and Nixon, 2000; FitzHarris et al., 2003). In experiments in which cIns(1,4,5)P3 was used to compare the sensitivity of Ca2+ release between populations of embryos, both experimental groups were placed on the stage together and simultaneously exposed to UV light. Thus, comparisons were made between groups that were injected with the same pipette of cIns(1,4,5)P3, loaded with Fura-red and illuminated at the same time and under identical conditions.

Two-photon imaging

In experiments involving two-photon imaging, embryos were microinjected with Ca2+-Green/dextran (10 kDa), rhodamine-dextran (10 kDa) or a mixture of Ca2+-Green/dextran (10 kDa) and a larger rhodamine-dextran (70 kDa), each to an estimated final concentration of 50-100 μM. Injected embryos were placed in a chamber shortly before the expected time of NEBD and maintained at 37°C. The zona pellucida was removed by brief exposure to acidified Tyrode's solution (Sigma) to permit adherence to the coverslip. The chamber was mounted on the stage of a Nikon E600 FN upright microscope equipped with a 60× water immersion objective (Nikon CFI Fluor 60×W) and the one-cell parthenogenetic embryos were imaged with a BioRad two-photon laser-scanning microscope. This consisted of an MRC 1024 scan head mounted on the upright microscope and a Spectra-Physics Tsunami titanium-sapphire mode-locked laser pumped by a Millennia V green laser. Fluorescence was excited with a wavelength of 775 nm. The exciting beam was focussed via the 60× objective, which also collected the emitted fluorescence. The intensity of excitation was adjusted by means of neutral-density filters and was generally less than 7 mW at the specimen. Emitted fluorescence was detected by a photomultiplier mounted as a non-scanning external detector. For experiments that used two dyes, the emitted light was split into its red and green components by a dichroic mirror before being collected by separate photomultipliers. As for single-dye experiments, these photomultipliers were mounted as non-scanning external detectors. In some experiments, bright-field images were also collected via a photodiode mounted beneath the microscope stage. For each experiment, a sequence of x-y axis images in a single plane were collected at 5-10 second intervals. Images were analysed after acquisition with Metamorph (Universal Imaging). Fluorescence intensity was measured from specific regions of interest (ROIs) as indicated, with background subtraction as necessary. In the two-dye experiments, there was no significant contamination of the green signal from the rhodamine but there was a significant bleed through of the green signal into the red channel.

Fig. 1.

Repetitive global mitotic Ca2+ oscillations are sperm specific. [Ca2+]i was monitored during the first mitotic division using Fura-2/dextran and epifluorescence microscopy. (A) Ca2+ oscillations were detected at NEBD (open arrow) in all fertilized embryos and further mitotic transients were detected in eight of nine embryos (mean number of transients per embryo was 3.9±0.54). No transients were detected following cytokinesis (closed arrow). (B) A typical mitotic Ca2+ transient is presented as a series of pseudocoloured images, warmer colours indicating an increase in [Ca2+]i. Note that the increase in Ca2+ occurs throughout the embryo. (C) No Ca2+ transients are detected at NEBD or during mitosis in parthenogenetically activated embryos (n=6). A large increase in Ca2+ was subsequently detected when mitotic parthenotes were challenged with ionomycin.

Fig. 1.

Repetitive global mitotic Ca2+ oscillations are sperm specific. [Ca2+]i was monitored during the first mitotic division using Fura-2/dextran and epifluorescence microscopy. (A) Ca2+ oscillations were detected at NEBD (open arrow) in all fertilized embryos and further mitotic transients were detected in eight of nine embryos (mean number of transients per embryo was 3.9±0.54). No transients were detected following cytokinesis (closed arrow). (B) A typical mitotic Ca2+ transient is presented as a series of pseudocoloured images, warmer colours indicating an increase in [Ca2+]i. Note that the increase in Ca2+ occurs throughout the embryo. (C) No Ca2+ transients are detected at NEBD or during mitosis in parthenogenetically activated embryos (n=6). A large increase in Ca2+ was subsequently detected when mitotic parthenotes were challenged with ionomycin.

Monitoring mitotic Ca2+ transients using Fura-2/dextran

In the first series of experiments, the characteristics of mitotic Ca2+ release in mouse one-cell embryos was examined using Fura-2/dextran and epifluorescence microscopy. Embryos were periodically examined using bright-field optics to determine the timing of NEBD and cleavage to the two-cell stage. In all fertilized embryos examined, mitosis entry was accompanied by a global increase in [Ca2+] that has been shown previously to be similar in magnitude to Ca2+ transients at fertilization (Marangos et al., 2003) (Fig. 1A,B). Additional mitotic Ca2+ transients were observed in eight of nine embryos (3.9±0.5 Ca2+ transients). No transients were detected in interphase, either before NEBD or after nucleus formation in the two-cell embryo. By contrast, Ca2+ transients were not detected during mitosis in ethanol-activated parthenogenetic embryos (Fig. 1C; n=6), similar to following activation by Sr2+ (Kono et al., 1996). Therefore, consistent with previous reports, mouse embryos exhibit repetitive mitotic Ca2+ transients that are confined to M phase and are specific to fertilized embryos (Kono et al., 1996; Marangos et al., 2003).

Precocious mitosis entry can be initiated by photorelease of Ins(1,4,5)P3

Mitosis entry can be triggered in sea-urchin embryos and fibroblasts by treatments that stimulate an increase in [Ca2+]i (Steinhart and Alderton, 1988; Twigg et al., 1988; Kao et al., 1990). However, microinjection of Ca2+-EGTA solutions is apparently not sufficient to stimulate NEBD in the mouse zygote (Tombes et al., 1992). Here, we addressed this question using photolysis of cIns(1,4,5)P3, which has the advantage of allowing Ins(1,4,5)P3 to be increased in a population of embryos simultaneously and at a time divorced from the microinjection procedure. Photorelease of cIns(1,4,5)P3 28-29 hours after hCG, when less than 10% of embryos had undergone NEBD, had no effect upon the timing of mitosis entry in fertilized embryos (not shown). Previous studies on sea-urchin embryos demonstrated that Ins(1,4,5)P3 was effective only during a brief window before NEBD (Twigg et al., 1988). In mouse embryos, there is considerable asynchrony in the population, resulting in a window of several hours during which the population undergoes mitosis. Therefore, to increase the proportion of embryos in which NEBD might have acquired a sensitivity to Ca2+, we released Ins(1,4,5)P3 when half of the embryos had already undergone NEBD (32-34 hours after hCG). First, we verified that addition of cIns(1,4,5)P3 resulted in an increase in [Ca2+]i and that [Ca2+]i was not affected by the UV flash (Fig. 2A). Monitoring the timing of NEBD revealed that photorelease of Ins(1,4,5)P3 induced an increase in the rate of NEBD in this population of embryos (Fig. 2B). Thus, an increase in [Ca2+]i close to the time of NEBD is sufficient to accelerate entry into mitosis in one-cell mouse embryos.

Relationship between Ca2+ and mitosis in mouse embryos

Effect of intracellular Ca2+ buffers on progression through mitosis

Two studies have previously reported that entry into the first mitotic division in mouse can be inhibited by the membrane-permeable Ca2+ chelator BAPTA-AM (Tombes et al., 1992; Kono et al., 1996). To investigate further the Ca2+ dependency of mitosis, we microinjected buffers with different Ca2+ affinities into one-cell embryos to a concentration of approximately 10 mM. In initial experiments, fertilized and parthenogenetic embryos were microinjected with BAPTA (Kd 160 nM), Br2BAPTA (Kd 1.6 μM) or EGTA (Kd 80 nM). Surprisingly, none of these buffers had any significant effect on the ability of embryos to undergo NEBD (Fig. 3A). However, consistent with previous studies, NEBD was efficiently inhibited by loading embryos with BAPTA-AM (10 μM, 30 minutes).

To verify that Ca2+ transients were inhibited in buffer-injected embryos, we examined Ca2+ release in fertilized embryos after injection of cIns(1,4,5)P3 (Fig. 3B). In BAPTA-injected embryos, there was no perturbation of baseline [Ca2+] at 10 millisecond, 100 millisecond and 1000 millisecond exposures to UV, and only a minor perturbation (approximately 10% of control) was seen at the highest level of Ins(1,4,5)P3 released. Similar results were obtained using parthenogenetic embryos (data not shown). Finally, to determine whether BAPTA prevented endogenous mitotic oscillations, [Ca2+]i was monitored in BAPTA-injected and control (Fura-2/dextran only) embryos. Control embryos exhibited between 2 and 12 Ca2+ transients during mitosis (6.3±1.1 transients per embryo, n=19), whereas no Ca2+ transients were detected in BAPTA-injected embryos (Fig. 3C). Close examination of traces from BAPTA-injected embryos (n=14) revealed small fluctuations in the Fura-2 ratio that occurred close to the time of NEBD (Fig. 3Cii, inset). These small fluctuations do not appear to be an artefact because no detectable changes were seen at NEBD in parthenogenetic embryos. Therefore, although capable of preventing repetitive large Ca2+ oscillations, 10 mM BAPTA was apparently insufficient to prevent all detectable Ca2+ release at NEBD.

Fig. 2.

Photorelease of cIns(1,4,5)P3 triggers precocious mitosis entry. Fertilized one-cell embryos were microinjected with cIns(1,4,5)P3 or water (vehicle) 32-34 hours after hCG, at which time approximately half of the embryos had undergone NEBD. Remaining interphase embryos were placed side-by-side on the microscope stage (cohorts of five to seven of each group) and exposed to UV light for 1 second. (A) An example of one such experiment in which six cIns(1,4,5)P3-injected and six control (vehicle) embryos were simultaneously exposed to UV light. Notice the occurrence of Ca2+ transients in all cIns(1,4,5)P3-injected embryos but not in controls. (B) Embryos were subsequently examined at 15 minute intervals to determine the timing of NEBD. The rate of mitosis entry was accelerated in cIns(1,4,5)P3-injected embryos compared with controls (n=24 for both groups). Inset shows the 15 minute and 30 minute time points as bar charts. C, control; I, cIns(1,4,5)P3. Data shown are from one of two similar replicates.

Fig. 2.

Photorelease of cIns(1,4,5)P3 triggers precocious mitosis entry. Fertilized one-cell embryos were microinjected with cIns(1,4,5)P3 or water (vehicle) 32-34 hours after hCG, at which time approximately half of the embryos had undergone NEBD. Remaining interphase embryos were placed side-by-side on the microscope stage (cohorts of five to seven of each group) and exposed to UV light for 1 second. (A) An example of one such experiment in which six cIns(1,4,5)P3-injected and six control (vehicle) embryos were simultaneously exposed to UV light. Notice the occurrence of Ca2+ transients in all cIns(1,4,5)P3-injected embryos but not in controls. (B) Embryos were subsequently examined at 15 minute intervals to determine the timing of NEBD. The rate of mitosis entry was accelerated in cIns(1,4,5)P3-injected embryos compared with controls (n=24 for both groups). Inset shows the 15 minute and 30 minute time points as bar charts. C, control; I, cIns(1,4,5)P3. Data shown are from one of two similar replicates.

Fig. 3.

Increased cytoplasmic Ca2+-buffering does not prevent the first mitotic division. (A) Fertilized and parthenogenetic embryos were microinjected with BAPTA, Br2BAPTA, EGTA (final concentrations 10 mM) or injection buffer, or were loaded with BAPTA-AM (10 μM) 1-2 hours before the predicted time of NEBD. Data are expressed as the percentage of embryos that had undergone NEBD within 20 hours of fertilization or parthenogenetic activation. Only BAPTA-AM treatment caused a significant inhibition of NEBD compared with controls (P<0.01, χ2 test). A minimum of two replicates were performed for each treatment. (B) One-cell embryos were microinjected with cIns(1,4,5)P3 and BAPTA (final concentration 10 mM; n=18) or cIns(1,4,5)P3 only (n=13). Injected embryos were loaded with Fura-red and [Ca2+]i was monitored during 10 millisecond, 100 millisecond, 1000 millisecond and 3000 millisecond exposures of UV light at 2 minute intervals. The peak change in Fura-red emission ratio was significantly reduced by BAPTA in response to 100 millisecond (#, P<0.05), 1000 millisecond and 3000 millisecond exposures (*, P<0.01). (C) BAPTA-injected embryos fail to exhibit repetitive Ca2+ transients in mitosis (n=19). Notice, however, that small fluctuations in Fura-2 baseline were detected at NEBD (ii, inset) (open arrow). Control embryos imaged at the same time exhibited characteristic Ca2+ oscillations (6.3±1.1 transients per embryo, n=14). No transients were detected following cytokinesis (closed arrow).

Fig. 3.

Increased cytoplasmic Ca2+-buffering does not prevent the first mitotic division. (A) Fertilized and parthenogenetic embryos were microinjected with BAPTA, Br2BAPTA, EGTA (final concentrations 10 mM) or injection buffer, or were loaded with BAPTA-AM (10 μM) 1-2 hours before the predicted time of NEBD. Data are expressed as the percentage of embryos that had undergone NEBD within 20 hours of fertilization or parthenogenetic activation. Only BAPTA-AM treatment caused a significant inhibition of NEBD compared with controls (P<0.01, χ2 test). A minimum of two replicates were performed for each treatment. (B) One-cell embryos were microinjected with cIns(1,4,5)P3 and BAPTA (final concentration 10 mM; n=18) or cIns(1,4,5)P3 only (n=13). Injected embryos were loaded with Fura-red and [Ca2+]i was monitored during 10 millisecond, 100 millisecond, 1000 millisecond and 3000 millisecond exposures of UV light at 2 minute intervals. The peak change in Fura-red emission ratio was significantly reduced by BAPTA in response to 100 millisecond (#, P<0.05), 1000 millisecond and 3000 millisecond exposures (*, P<0.01). (C) BAPTA-injected embryos fail to exhibit repetitive Ca2+ transients in mitosis (n=19). Notice, however, that small fluctuations in Fura-2 baseline were detected at NEBD (ii, inset) (open arrow). Control embryos imaged at the same time exhibited characteristic Ca2+ oscillations (6.3±1.1 transients per embryo, n=14). No transients were detected following cytokinesis (closed arrow).

Ins(1,4,5)P3R downregulation prevents mitotic Ca2+ release

A second strategy of preventing mitotic Ca2+ transients involved depletion of Ins(1,4,5)P3Rs. Adenophostin A (Ad-A) is a potent Ins(1,4,5)P3R agonist that has previously been found to trigger extensive Ins(1,4,5)P3R downregulation within 4 hours of microinjection (Takahashi et al., 1993; He et al., 1999; Brind et al., 2000). In previous experiments, we have shown that injection of Ad-A during oocyte maturation prevents Ca2+ release at fertilization. Here, we used Ins(1,4,5)P3R-depleted embryos to investigate the role of Ins(1,4,5)P3Rs in mitosis.

First, we used cIns(1,4,5)P3 to confirm that Ad-A treatment inhibits Ca2+ release in one-cell embryos. As expected, Ins(1,4,5)P3-induced Ca2+ release was dramatically inhibited in Ad-A-treated embryos compared with controls (Fig. 4A). Furthermore, Ad-A-injected embryos (n=9) showed no evidence of mitotic Ca2+ transients during mitosis (Fig. 4B). As before, global Ca2+ transients were recorded in all controls (n=12; 2.3±0.5 transients per embryo). Despite abolishing all measurable Ca2+ release during mitosis, Ad-A-injected embryos proceeded through mitosis with kinetics indistinguishable from controls (Fig. 4Ci). To confirm that anaphase had proceeded, the resultant two-cell embryos were loaded with Hoechst 33342. Each blastomere in all of the embryos examined was found to contain chromatin, indicating that chromosome disjunction at anaphase had taken place (Fig. 4Cii; control, n=19; Ad-A, n=12). These experiments show that mitotic Ca2+ transients are dependent on the Ins(1,4,5)P3R, and that, in fertilized embryos, progression through mitosis can take place in the absence of measurable Ca2+ transients.

Fig. 4.

Adenophostin-A treatment prevents mitotic Ca2+ oscillations in fertilized embryos. (A) cIns(1,4,5)P3 was used to test the responsiveness of Ca2+ release in one-cell embryos 4 hours after Ad-A injection (2.5 μM). Peak Fura-red ratio change was significantly reduced in Ad-A injected embryos (n=16) compared with controls (uninjected, n=15; vehicle, n=6) in response to 300 millisecond, 1200 millisecond and 3000 millisecond exposures of UV light (*, P<0.01), indicating that Ad-A treatment inhibits Ins(1,4,5)P3-mediated Ca2+ release. (B) [Ca2+]i was monitored in Ad-A-treated embryos during mitosis using Fura-2/dextran. Mitotic Ca2+ oscillations were detected in all controls (Fura-2/dextran only; n=9) but not following Ad-A treatment (n=12). Open arrow, NEBD; closed arrow, cytokinesis. (C) Ad-A treatment had no effect upon the timing of NEBD or cytokinesis as determined using bright-field optics (data from two similar replicates; buffer, n=30; Ad-A, n=34). (D) Subsequent Hoechst labelling revealed chromatin within both blastomeres, indicating that chromosome disjunction at anaphase was uninhibited (two replicates, n=12 for Ad-A; n=19 for control).

Fig. 4.

Adenophostin-A treatment prevents mitotic Ca2+ oscillations in fertilized embryos. (A) cIns(1,4,5)P3 was used to test the responsiveness of Ca2+ release in one-cell embryos 4 hours after Ad-A injection (2.5 μM). Peak Fura-red ratio change was significantly reduced in Ad-A injected embryos (n=16) compared with controls (uninjected, n=15; vehicle, n=6) in response to 300 millisecond, 1200 millisecond and 3000 millisecond exposures of UV light (*, P<0.01), indicating that Ad-A treatment inhibits Ins(1,4,5)P3-mediated Ca2+ release. (B) [Ca2+]i was monitored in Ad-A-treated embryos during mitosis using Fura-2/dextran. Mitotic Ca2+ oscillations were detected in all controls (Fura-2/dextran only; n=9) but not following Ad-A treatment (n=12). Open arrow, NEBD; closed arrow, cytokinesis. (C) Ad-A treatment had no effect upon the timing of NEBD or cytokinesis as determined using bright-field optics (data from two similar replicates; buffer, n=30; Ad-A, n=34). (D) Subsequent Hoechst labelling revealed chromatin within both blastomeres, indicating that chromosome disjunction at anaphase was uninhibited (two replicates, n=12 for Ad-A; n=19 for control).

Removal of extracellular Ca2+ prevents mitotic Ca2+ release

A third way to manipulate Ca2+ release during mitosis is to remove extracellular Ca2+. It has been shown previously that mouse embryos are not dependent on extracellular Ca2+ for progression through the first mitotic division (Tombes et al., 1992). However, it was not known whether removal of extracellular Ca2+ inhibits mitotic Ca2+ transients. We therefore monitored [Ca2+]i during mitosis in Ca2+-free medium and found no Ca2+ transients in any of the embryos examined (Fig. 5A; n=14). In addition, we found the timing of mitosis was not affected by Ca2+-free medium and that normal chromatid separation had taken place despite the absence of any Ca2+ transients (Fig. 5B; n=13).

Monitoring [Ca2+]i during NEBD in fertilized and parthenogenetic embryos using two-photon microscopy

The experiments described above demonstrate that global mitotic Ca2+ transients are not necessary for progression through mitosis in fertilized embryos. However, these epifluorescence studies cannot exclude the possibility that mitosis entry is triggered by localized [Ca2+] elevations, such as those detected before NEBD in sea-urchin embryos (Wilding et al., 1996). Such localized Ca2+ changes might also drive mitosis entry in parthenogenetic mouse embryos, which do not generate global Ca2+ transients at NEBD. We therefore examined parthenogenetic embryos for evidence of localized Ca2+ release around the time of mitosis entry.

Mammalian embryos are susceptible to photodamage by visible light (Daniel, 1964; Hegele-Hartung et al., 1991) and we have previously found that time-lapse confocal microscopy prevents mitosis entry in zygotes in almost all cases (G.F. and J.C., unpublished). Two-photon microscopy uses two-photon excitation of the fluorophore, allowing the use of longer-wavelength light and restricting excitation to the focal point. This technique has been demonstrated to be less damaging than confocal microscopy when used with two-cell hamster embryos (Squirrell et al., 1999). We have established conditions in which mitosis proceeds during imaging with a two-photon microscope (sampling frequency, 5-10 seconds) (Fig. 6A). In this example the fertilized embryo undergoes three Ca2+ transients, each transient resulting in a threefold increase in fluorescence of Ca2+-Green/dextran. We have used this set up to investigate whether parthenogenetic mitotic embryos generate local perinuclear Ca2+ transients during NEBD.

In the first series of experiments, Ca2+-Green/dextran (10 kDa) was used to monitor [Ca2+]i during mitosis entry in parthenogenetic embryos. Bright-field images were collected to determine the timing of NEBD. Strikingly, two-photon imaging revealed a small, transient increase in fluorescence within the nucleus close to the time of NEBD (Fig. 6B; n=16). To test whether the observed increase in Ca2+-Green/dextran fluorescence was dependent on Ca2+ rather than a change in the environment in the nucleus, a similar series of experiments was performed with a Ca2+-insensitive fluorescent probe, rhodamine-dextran (10 kDa). In embryos injected with rhodamine-dextran, a similar small increase in fluorescence was observed in the nucleus at NEBD (Fig. 6C; n=5). The Ca2+ ionophore ionomycin caused large global increases in fluorescence in embryos injected with Ca2+-Green but not in embryos injected with rhodamine, confirming that rhodamine could be used to reliably differentiate between Ca2+-dependent and Ca2+-independent events. As such, it is unlikely that the fluorescence change detected using Ca2+-Green/dextran can be attributed to Ca2+.

Fig. 5.

Mitotic transients are dependent on extracellular Ca2+. Pronucleate embryos were transferred to Ca2+-free medium containing 1 mM EGTA 28-29 hours after hCG administration, at which time ∼10% had undergone NEBD. (A) No change in the Fura ratio was detected at NEBD or during mitosis in Ca2+-free medium in any case (n=14). (B) Transfer to Ca2+-free medium had no effect on the timing of NEBD and cytokinesis (data from two similar replicates; n=46 for Ca2+-containing medium, n=51 for Ca2+-free medium). (C) Hoechst labelling revealed chromatin within the nucleus of each resulting blastomere, indicating that chromosome disjunction is not dependent on extracellular Ca2+ (Ca2+-free medium, n=13; control, n=6). Open arrow, NEBD; closed arrow, cytokinesis.

Fig. 5.

Mitotic transients are dependent on extracellular Ca2+. Pronucleate embryos were transferred to Ca2+-free medium containing 1 mM EGTA 28-29 hours after hCG administration, at which time ∼10% had undergone NEBD. (A) No change in the Fura ratio was detected at NEBD or during mitosis in Ca2+-free medium in any case (n=14). (B) Transfer to Ca2+-free medium had no effect on the timing of NEBD and cytokinesis (data from two similar replicates; n=46 for Ca2+-containing medium, n=51 for Ca2+-free medium). (C) Hoechst labelling revealed chromatin within the nucleus of each resulting blastomere, indicating that chromosome disjunction is not dependent on extracellular Ca2+ (Ca2+-free medium, n=13; control, n=6). Open arrow, NEBD; closed arrow, cytokinesis.

To determine accurately the timing of these localized fluorescence increases with respect to NEBD, we monitored mitosis entry in parthenogenetic one-cell embryos coinjected with Ca2+-Green/dextran (10 kDa) and a 70 kDa rhodamine-dextran. The 70 kDa rhodamine was excluded from the nucleus during interphase and therefore provided a precise indication of the timing of nuclear permeabilization (Fig. 7A). Analysis of relative fluorescence intensities revealed that the localized increase in Ca2+-Green fluorescence occurs at precisely the same time that the rhodamine-dextran entered the nucleoplasm (Fig. 7B; n=5). Therefore, the local increase in Ca2+-Green fluorescence that we detected at NEBD is Ca2+ independent and occurs concomitantly with, rather than before, permeabilization of the nuclear membrane.

Second cell division occurs in the absence of Ca2+ transients

Finally, it has not previously been established whether Ca2+ transients accompany the second (or subsequent) mitotic divisions in mouse. To address this question, two-cell-stage embryos were co-injected with Fura-2/dextran, and with 1-3 μM FITC-BSA-NLS (Jackman et al., 2002). FITC-BSA-NLS localizes to nuclei and therefore permitted visualization of NEBD and nuclear-envelope reformation (NER). Embryos were placed on the microscope stage shortly before the predicted time of mitosis entry (approximately 55 hours after hCG administration) and monitored throughout mitosis (Fig. 8). No changes in the Fura-2 ratio were detected in blastomeres of mitotic two-cell embryos, indicating that Ca2+ transients are restricted to the first mitotic division of fertilized embryos and providing further evidence that they are not a necessary trigger for mitosis.

Fig. 6.

Two-photon microscopy detects mitotic Ca2+ transients in fertilized embryos and reveals Ca2+-independent increases in indicator fluorescence within the nucleus at mitosis entry in parthenogenetic embryos. Ca2+ transients are readily detectable in a fertilized mitotic embryo using two-photon microscopy (A). (top) A series of images taken during the first transient plotted underneath. The mitotic Ca2+ transients cause a threefold increase in fluorescence. Ca2+-Green/dextran (B) and rhodamine-dextran (C) were monitored during NEBD in parthenotes. Bright-field optics were used to determine the timing of NEBD (Bi, top). In addition, the exclusion of the indicators from the nucleolus allowed mitosis entry to be established from the fluorescence images as the time at which the nucleolus disappeared (Bi, arrowheads indicate the position of the nucleolus). Notice the striking increase in fluorescence that occurs within the nucleus at the time of NEBD in both Ca2+-Green/dextran- and rhodamine/dextran-injected embryos (Bi,Ci). (Bii,Cii) The same images as in Bi and Ci but illustrating the regions of interest used for data analysis (grey, peripheral cytoplasm; black, nucleus). Analysis of fluorescence intensities using these regions of interest confirms that the fluorescence increase is restricted to the nucleus in both cases (Biii,Ciii). Lower-case letters indicate which image corresponds to given points on the graph. Notice also that ionomycin addition resulted in a dramatic fluorescence increase in Ca2+-Green/dextran-injected, but not rhodamine/dextran-injected, embryos.

Fig. 6.

Two-photon microscopy detects mitotic Ca2+ transients in fertilized embryos and reveals Ca2+-independent increases in indicator fluorescence within the nucleus at mitosis entry in parthenogenetic embryos. Ca2+ transients are readily detectable in a fertilized mitotic embryo using two-photon microscopy (A). (top) A series of images taken during the first transient plotted underneath. The mitotic Ca2+ transients cause a threefold increase in fluorescence. Ca2+-Green/dextran (B) and rhodamine-dextran (C) were monitored during NEBD in parthenotes. Bright-field optics were used to determine the timing of NEBD (Bi, top). In addition, the exclusion of the indicators from the nucleolus allowed mitosis entry to be established from the fluorescence images as the time at which the nucleolus disappeared (Bi, arrowheads indicate the position of the nucleolus). Notice the striking increase in fluorescence that occurs within the nucleus at the time of NEBD in both Ca2+-Green/dextran- and rhodamine/dextran-injected embryos (Bi,Ci). (Bii,Cii) The same images as in Bi and Ci but illustrating the regions of interest used for data analysis (grey, peripheral cytoplasm; black, nucleus). Analysis of fluorescence intensities using these regions of interest confirms that the fluorescence increase is restricted to the nucleus in both cases (Biii,Ciii). Lower-case letters indicate which image corresponds to given points on the graph. Notice also that ionomycin addition resulted in a dramatic fluorescence increase in Ca2+-Green/dextran-injected, but not rhodamine/dextran-injected, embryos.

Ca2+ transients are dispensable for mitosis in mouse embryos

This study was designed to investigate the role of Ca2+ during mitosis in early mouse embryos. Our data provide three major lines of evidence that indicate that Ca2+ is not a necessary trigger for entry into or progression through mitosis in mouse embryos. First, we inhibited the endogenous mitotic transients in fertilized one-cell embryos using three independent approaches. Each of these failed to prevent NEBD, anaphase or cytokinesis. Second, Ca2+ transients were not detected throughout the second mitotic division or during the first mitosis in parthenogenetic embryos. Third, we used two-photon microscopy to investigate the hypothesis that, in the absence of a global Ca2+ transient, mitosis is driven by local perinuclear Ca2+ transients. These experiments provided no evidence for localized Ca2+ release before mitosis entry. These studies suggest that, at least in early mouse embryos, an increase in Ca2+ is not necessary for entry into and progression through mitosis.

Fig. 7.

The nuclear increase in indicator fluorescence occurs concomitantly with nuclear-membrane permeabilization. Parthenogenetic embryos were co-injected with Ca2+-Green/dextran and a 70 kDa rhodamine-dextran. The 70 kDa rhodamine-dextran was excluded from the nucleus, allowing NEBD to be determined accurately as the time at which the rhodamine signal entered the nucleoplasm (Aa′-d′). The changes in fluorescence of the two indicators were examined using regions of interest placed in the cytoplasm and nucleus (Bi). Lower-case letters indicate which image corresponds to given points on the graph. Notice that the nuclear increase in Ca2+-Green fluorescence occurs at precisely the same time that the 70 kDa rhodamine-dextran enters the nucleoplasm (Bii).

Fig. 7.

The nuclear increase in indicator fluorescence occurs concomitantly with nuclear-membrane permeabilization. Parthenogenetic embryos were co-injected with Ca2+-Green/dextran and a 70 kDa rhodamine-dextran. The 70 kDa rhodamine-dextran was excluded from the nucleus, allowing NEBD to be determined accurately as the time at which the rhodamine signal entered the nucleoplasm (Aa′-d′). The changes in fluorescence of the two indicators were examined using regions of interest placed in the cytoplasm and nucleus (Bi). Lower-case letters indicate which image corresponds to given points on the graph. Notice that the nuclear increase in Ca2+-Green fluorescence occurs at precisely the same time that the 70 kDa rhodamine-dextran enters the nucleoplasm (Bii).

Fig. 8.

Mitotic Ca2+ transients cannot be detected during the second embryonic division. Two-cell embryos were co-injected with Fura-2/dextran and FITC-NLS-BSA to monitor [Ca2+]i and the presence of nuclei, respectively, and transferred to the microscope stage shortly before the second mitotic division. Notice the disappearance of the nucleus at mitosis entry and the formation of two new nuclei following cytokinesis. No Ca2+ transients were seen during the second mitotic division (n=9). A significant increase in Fura-2 ratio occurred when blastomeres were subsequently challenged with ionomycin (5 μM). Images were acquired at 10-second intervals.

Fig. 8.

Mitotic Ca2+ transients cannot be detected during the second embryonic division. Two-cell embryos were co-injected with Fura-2/dextran and FITC-NLS-BSA to monitor [Ca2+]i and the presence of nuclei, respectively, and transferred to the microscope stage shortly before the second mitotic division. Notice the disappearance of the nucleus at mitosis entry and the formation of two new nuclei following cytokinesis. No Ca2+ transients were seen during the second mitotic division (n=9). A significant increase in Fura-2 ratio occurred when blastomeres were subsequently challenged with ionomycin (5 μM). Images were acquired at 10-second intervals.

Given previous studies in mouse and sea-urchin embryos, this result is surprising. We and others have previously found that BAPTA-AM effectively inhibited NEBD in mouse zygotes. This was the case in fertilized embryos that generate Ca2+ transients and in parthenogenetic embryos that do not (Tombes et al., 1992; Kono et al., 1996) (present study). Similarly, in sea-urchin embryos and fibroblasts, BAPTA inhibits mitosis despite the absence of detectable Ca2+ transients in many cases (Kao et al., 1990; Wilding et al., 1996). These inconsistencies could be reconciled after it was found that local perinuclear [Ca2+] increases, which were not detectable using epifluorescence microscopy, were detectable using ratiometric confocal microscopy in sea-urchin embryos (Wilding et al., 1996). This result therefore raised the possibility that an increase in [Ca2+] is indeed a universal trigger for progression through mitosis.

In the present study, we have tested this possibility in parthenogenetic mouse embryos, in which we would predict that localized increases in [Ca2+] provide the trigger for mitosis. Despite initial enthusiasm when an increase in fluorescence was seen in the nucleus, we subsequently found that this increase was also detectable with Ca2+-insensitive rhodamine-dextran. Furthermore, the increase did not precede NEBD but was coincident with it. Thus, we could not find any evidence for local Ca2+ transients driving mitosis in the mouse zygote. We cannot discount the possibility that Ca2+ changes are present below the level of detection of our system, but the dynamic range of Ca2+-Green/dextran and the ability to detect small Ca2+-independent changes in fluorescence suggests that any such changes would need to be more localized or more transient than previously described. These data suggest that, although an increase in [Ca2+]i is both necessary and sufficient for triggering mitosis in sea-urchin embryos, it is apparently dispensable for mitotic progression in mouse embryos.

Our data showing that microinjection of BAPTA, Br2BAPTA or EGTA has no effect on mitosis suggest that BAPTA-AM loading has effects independent of cytosolic Ca2+ buffering that are able to inhibit mitosis. One possibility is that compartmentalization of BAPTA-AM into organelles such as the endoplasmic reticulum might disrupt other cellular functions such as protein synthesis (Brostrom and Brostrom, 2003). Indeed, BAPTA-AM has been shown to inhibit protein synthesis, probably as a result of depleting intracellular Ca2+ stores (Preston and Berlin, 1992; Lawrence et al., 1998). Because mitosis entry in mouse zygotes is critically dependent upon the manufacture of new proteins (Howlett, 1986), BAPTA-AM might prevent NEBD by disturbing lumenal Ca2+ homeostasis, rather than by preventing cytosolic Ca2+ changes.

The search for local nuclear Ca2+ transients has not been performed in many cell types but several recent studies indirectly support the conclusion that Ca2+ transients are dispensable for mitosis. In fertilized embryos, we have shown that the detectable Ca2+ transients start minutes after NEBD (Marangos et al., 2003; Larman et al., 2004). In addition, the metaphase-to-anaphase transition during meiosis I takes place in the absence of any measurable increase in cytosolic [Ca2+] (Hyslop et al., 2004; Marangos and Carroll, 2004). Finally, in some (but not all) fibroblast cell lines, mitosis proceeds normally in the absence of Ca2+ transients. Questions have been raised about potential problems with compartmentalization of indicators, which can confound interpretation of Ca2+ records (al Mohanna et al., 1994; Carroll et al., 1994). However, a more recent study using a cytosolic chameleon, which avoids such pitfalls, also failed to report an increase in [Ca2+] during mitosis (Whitaker and Larman, 2001). Therefore, though there is strong evidence for a necessary role for Ca2+ in mitosis in sea-urchin embryos, it appears that, in mammalian embryos and several cell lines, Ca2+ does not play an obligatory role in mitosis. Ca2+ transients might act as a pacing mechanism for the rapid cell cycles of echinoderms, and this mechanism might have been supplanted in the longer, more complex cell cycles of mammalian cells.

Ca2+ might not be necessary, but is sufficient, to accelerate mitosis entry

A more fundamental role for Ca2+ in embryonic cell cycles is indicated by our observation that the ability of a Ca2+ transient to trigger mitosis entry is conserved between sea-urchin and mouse embryos. This sensitivity to Ca2+ was revealed in mouse embryos by an acceleration of NEBD in response to Ins(1,4,5)P3-induced Ca2+ release and could only be uncovered in a population of embryos that remained in interphase once 50% had entered mitosis. A narrow window of sensitivity is also seen in sea-urchin embryos and suggests that a Ca2+-sensitive pathway becomes available only once NEBD is imminent (Twigg et al., 1988). A good candidate for the Ca2+-sensitive switch is activation of Cdc25. Cdc25 is necessary for the activation of Cdk1/cyclin-B at mitosis entry and has been shown to be sensitive to Ca2+ (Patel et al., 1999). In sea urchins, the Ca2+-sensitive step acts as a cell-cycle checkpoint, arresting progression into mitosis until certain conditions are established (Whitaker and Larman, 2001). It therefore remains an intriguing possibility that conserved Ca2+-sensitive events drive NEBD in mammalian embryos when the normal cell cycle is perturbed.

Fluorescence changes do not always mean a Ca2+ transient

These studies reveal some of the potential caveats of using fluorescent probes and high-resolution imaging in dynamic systems. Most remarkably, a compelling increase in fluorescence in the nucleus seen with a Ca2+-sensitive indicator was mimicked by an indicator that does not respond to [Ca2+]. The explanation for the small increase in fluorescence at NEBD remains obscure but it is well known that the fluorescence properties of fluorophores are sensitive to environmental factors such as viscosity and polarity (Tsien, 1989; Poenie, 1990; Roe et al., 1990; Busa, 1992). In addition, some cytoplasmic proteins might modify fluorescence by direct binding to fluorophores (Highsmith et al., 1986; Konishi et al., 1988). Because NEBD is accompanied by sweeping changes in cellular organization such as nuclear entry and exit of proteins, and a dramatic reorganization of organelle structure (Terasaki, 2000; FitzHarris et al., 2003), the fluorescence changes we observe probably reflect changes in the constitution of the nucleoplasm as it mixes with the cytosol. Convincing spatially restricted perinuclear Ca2+ transients have previously been reported in sea-urchin embryos (Wilding et al., 1996) and in HeLa cells (Lipp et al., 1997), the spatial organization of which are clearly distinct from the fluorescence changes we have reported here.

Maternal and paternal contributions to mitotic Ca2+ signalling in mouse embryos

The mechanisms by which mitotic Ca2+ transients are generated in mouse embryos are gradually becoming apparent. Our Ad-A experiments indicate that the Ins(1,4,5)P3 receptor is essential for mitotic Ca2+ release. We have previously shown that there is an increase in the sensitivity of Ins(1,4,5)P3R-mediated Ca2+ release at mitosis entry in both fertilized and parthenogenetic embryos (FitzHarris et al., 2003), suggesting a maternal cell-cycle-related modification of the Ca2+-releasing machinery. Recent studies suggest the cell-cycle-dependent control might be mediated by Ins(1,4,5)P3R phosphorylation (Jellerette al., 2004).

However, maternal factors are clearly not sufficient for the generation of mitotic transients because oscillations are specific to fertilized embryos (Kono et al., 1996) (present study). Indeed, in several studies, even the mitotic Ca2+ transients are not always able to be detected (Tombes et al., 1992; Day et al., 2000; Tang et al., 2000; Gordo et al., 2002). Recently, it has emerged that phospholipase Cζ (PLCζ), the sperm-borne factor responsible for initiating Ca2+ oscillations and egg activation (Saunders et al., 2002; Cox et al., 2002; Rogers et al., 2004), becomes compartmentalized within the developing pronuclei following fertilization (Larman et al., 2004; Yoda et al., 2004). Given the strict relationship between pronucleus formation and cessation of the oscillations, it has been proposed that sequestration of PLCζ terminates the oscillations, perhaps by isolating the enzyme from its substrate, and that mitotic Ca2+ release reflects the subsequent liberation of PLCζ back into the cytoplasm (Marangos et al., 2003; Larman et al., 2004). This model is consistent with the findings that the first mitotic transient occurs several minutes after the first signs that NEBD is imminent (Kono et al., 1996; Marangos et al., 2003; Larman et al., 2004) and that pronuclei from fertilized, but not from parthenogenetic, embryos possess a Ca2+-releasing activity capable of activating unfertilized eggs (Kono et al., 1995; Zernicka-Goetz et al., 1995). Thus, it appears that mitotic Ca2+ transients are generated by the release of PLCζ into a highly responsive M-phase cytoplasm. Our data showing that the second mitosis lacks global Ca2+ transients suggest that PLCζ might be inactivated by the time of second mitosis or that the cytoplasm is no longer capable of supporting Ca2+ release. PLCζ activity appears to be the limiting factor, because nuclei transferred from pronucleate, but not late two-cell- or four-cell-stage, embryos can activate recipient eggs (Kono et al., 1995; Zernicka-Goetz et al., 1995). Thus, the Ca2+ transients during the first mitotic division are the final act in a Ca2+ signalling pathway initiated some 16-20 hours earlier, at the time of sperm-egg fusion.

Role of mitotic Ca2+ release in mouse embryos

Although dispensable for mitosis, there is some evidence that mitotic Ca2+ transients serve a further developmental role. Sr2+ exposure during the first mitosis induces oscillations in parthenogenotes and resulting blastocysts have more cells in the inner cell mass (Bos-Mikich et al., 1995). Improved development has also been noted following exposure of one- or two-cell (but not four-cell) embryos to low concentrations of ethanol (Leach et al., 1993), a treatment that might increase [Ca2+]i. A similar phenomenon has been described following fertilization, when Ca2+ oscillation regimen can influence long-term developmental potential (Ozil and Huneau, 2001; Ducibella et al., 2002). Because these experiments imply a role for Ca2+ as a determinant of developmental potential several cell divisions later, mitotic Ca2+ transients might modify gene expression via Ca2+-dependent transcription factors (Mellstrom and Naranjo, 2001). However, it should be realized that blastocyst formation can be achieved by ethanol-activated parthenogenotes (Kaufman and Sachs, 1978), which we have now shown do not generate mitotic Ca2+ transients, and that the developmental outcome of inhibiting endogenous transients has not yet been established.

This work was supported by an MRC Career Establishment Grant to JC. The two-photon imaging was supported by a Bio-imaging Initiative Grant from the BBSRC to C.R. G.F. was supported by a Reproductive Medicine Studentship from the Department of Obstetrics and Gynaecology, and The Assisted Conception Unit at University College London (London, UK). We thank K. Swann, S. Patel, S. Bolsover, J. Baltz and M. Whitaker for discussion and comments on the manuscript.

al Mohanna, F. A., Caddy, K. W. and Bolsover, S. R. (
1994
). The nucleus is insulated from large cytosolic calcium ion changes.
Nature
367
,
745
-750.
Baitinger, C., Alderton, J., Poenie, M., Schulman, H. and Steinhardt, R. A. (
1990
). Multifunctional Ca2+/calmodulin-dependent protein kinase is necessary for nuclear envelope breakdown.
J. Cell Biol.
111
,
1763
-1773.
Bos-Mikich, A., Whittingham, D. G. and Jones, K. T. (
1997
). Meiotic and mitotic Ca2+ oscillations affect cell composition in resulting blastocysts.
Dev. Biol.
182
,
172
-179.
Brind, S., Swann, K. and Carroll, J. (
2000
). Inositol 1,4,5-trisphosphate receptors are downregulated in mouse oocytes in response to sperm or adenophostin A but not to increases in intracellular Ca2+ or egg activation.
Dev. Biol.
223
,
251
-265.
Brostrom, M. A. and Brostrom, C. O. (
2003
). Calcium dynamics and endoplasmic reticular function in the regulation of protein synthesis: implications for cell growth and adaptability.
Cell Calcium
34
,
345
-363.
Busa, W. B. (
1992
). Spectral characterization of the effect of viscosity on Fura-2 fluorescence: excitation wavelength optimization abolishes the viscosity artifact.
Cell Calcium
13
,
313
-319.
Carroll, J., Swann, K., Whittingham, D. and Whitaker, M. (
1994
). Spatiotemporal dynamics of intracellular [Ca2+]i oscillations during the growth and meiotic maturation of mouse oocytes.
Development
120
,
3507
-3517.
Chang, D. C. and Meng, C. (
1995
). A localized elevation of cytosolic free calcium is associated with cytokinesis in the zebrafish embryo.
J. Cell Biol.
131
,
1539
-1545.
Ciapa, B., Pesando, D., Wilding, M. and Whitaker, M. (
1994
). Cell-cycle calcium transients driven by cyclic changes in inositol trisphosphate levels.
Nature
368
,
875
-878.
Cox, L. J., Larman, M. G., Saunders, C. M., Hashimoto, K., Swann, K. and Lai, F. A. (
2002
). Sperm phospholipase Czeta from humans and cynomolgus monkeys triggers Ca2+ oscillations, activation and development of mouse oocytes.
Reproduction
124
,
611
-623.
Daniel, J. C., Jr (
1964
). Cleavage of mammalian ova inhibited by visible light.
Nature
201
,
316
-317.
Day, M. L., McGuinness, O. M., Berridge, M. J. and Johnson, M. H. (
2000
). Regulation of fertilization-induced Ca2+ spiking in the mouse zygote.
Cell Calcium
28
,
47
-54.
Ducibella, T., Huneau, D., Angelichio, E., Xu, Z., Schultz, R. M., Kopf, G. S., Fissore, R., Madoux, S. and Ozil, J. P. (
2002
). Egg-to-embryo transition is driven by differential responses to Ca2+ oscillation number.
Dev. Biol.
250
,
280
-291.
FitzHarris, G., Marangos, P. and Carroll, J. (
2003
). Cell cycle-dependent regulation of structure of endoplasmic reticulum and inositol 1, 4, 5-trisphosphate-induced Ca2+ release in mouse oocytes and embryos.
Mol. Biol. Cell
14
,
288
-301.
Fluck, R. A., Miller, A. L. and Jaffe, L. F. (
1991
). Slow calcium waves accompany cytokinesis in medaka fish eggs.
J. Cell Biol.
115
,
1259
-1265.
Fulton, B. P. and Whittingham, D. G. (
1978
). Activation of mammalian oocytes by intracellular injection of calcium.
Nature
273
,
149
-151.
Gordo, A. C., Kurokawa, M., Wu, H. and Fissore, R. A. (
2002
). Modifications of the Ca2+ release mechanisms of mouse oocytes by fertilization and by sperm factor.
Mol. Hum. Reprod.
8
,
619
-629.
Groigno, L. and Whitaker, M. (
1998
). An anaphase calcium signal controls chromosome disjunction in early sea urchin embryos.
Cell
92
,
193
-204.
He, C. L., Damiani, P., Ducibella, T., Takahashi, M., Tanzawa, K., Parys, J. B. and Fissore, R. A. (
1999
). Isoforms of the inositol 1, 4, 5-trisphosphate receptor are expressed in bovine oocytes and ovaries: the type-1 isoform is down-regulated by fertilization and by injection of adenophostin A.
Biol. Reprod.
61
,
935
-943.
Hegele-Hartung, C., Schumacher, A. and Fischer, B. (
1991
). Effects of visible light and room temperature on the ultrastructure of preimplantation rabbit embryos: a time course study.
Anat. Embryol.
183
,
559
-571.
Hepler, P. K. (
1994
). The role of calcium in cell division.
Cell Calcium
16
,
322
-330.
Highsmith, S., Bloebaum, P. and Snowdowne, K. W. (
1986
). Sarcoplasmic reticulum interacts with the Ca2+ indicator precursor Fura-2-AM.
Biochem. Biophys. Res. Commun.
138
,
1153
-1162.
Howlett, S. K. (
1986
). A set of proteins showing cell cycle dependent modification in the early mouse embryo.
Cell
45
,
387
-396.
Hyslop, L. A., Nixon, V. L., Levasseur, M., Chapman, F., Chiba, K., McDougall, A., Venables, J. P., Elliott, D. J. and Jones, K. T. (
2004
). Ca2+-promoted cyclin B1 degradation in mouse oocytes requires the establishment of a metaphase arrest.
Dev. Biol.
269
,
206
-219.
Jackman, M., Kubota, Y., den Elzen, N., Hagting, A. and Pines, J. (
2002
). Cyclin A- and cyclin E-CDK complexes shuttle between the nucleus and the cytoplasm.
Mol. Biol. Cell
13
,
1030
-1045.
Jellerette, T., Kurokawa, M., Lee, B., Malcuit, C., Yoon, S. Y., Smyth, J., Vermassen, E., DeSmedt, H., Parys, J. B. and Fissore, R. A. (
2004
). Cell cycle-coupled [Ca2+]i oscillations in mouse zygotes and function of the inositol 1,4,5-trisphosphate receptor-1.
Dev. Biol.
274
,
94
-109.
Jones, K. T. and Nixon, V. L. (
2000
). Sperm-induced Ca2+ oscillations in mouse oocytes and eggs can be mimicked by photolysis of caged inositol 1,4,5-trisphosphate: evidence to support a continuous low level production of inositol 1,4,5-trisphosphate during mammalian fertilization.
Dev. Biol.
225
,
1
-12.
Kao, J. P., Alderton, J. M., Tsien, R. Y. and Steinhardt, R. A. (
1990
). Active involvement of Ca2+ in mitotic progression of Swiss 3T3 fibroblasts.
J. Cell Biol.
111
,
183
-196.
Kaufman, M. H. and Sachs, L. (
1978
). Complete preimplantation development in culture of parthenogenetic mouse embryos.
J. Embryol. Exp. Morphol.
35
,
179
-190.
Keith, C. H., Ratan, R., Maxfield, F. R., Bajer, A. and Shelanski, M. L. (
1985
). Local cytoplasmic calcium gradients in living mitotic cells.
Nature
316
,
848
-850.
Konishi, M., Olson, A., Hollingworth, S. and Baylor, S. M. (
1988
). Myoplasmic binding of Fura-2 investigated by steady-state fluorescence and absorbance measurements.
Biophys. J.
54
,
1089
-1104.
Kono, T., Carroll, J., Swann, K. and Whittingham, D. G. (
1995
). Nuclei from fertilized mouse embryos have calcium-releasing activity.
Development
121
,
1123
-1128.
Kono, T., Jones, K. T., Bos-Mikich, A., Whittingham, D. G. and Carroll, J. (
1996
). A cell cycle-associated change in Ca2+ releasing activity leads to the generation of Ca2+ transients in mouse embryos during the first mitotic division.
J. Cell Biol.
132
,
915
-923.
Larman, M. G., Saunders, C. M., Carroll, J., Lai, F. A. and Swann, K. (
2004
). Cell cycle-dependent Ca2+ oscillations in mouse embryos are regulated by nuclear targeting of PLCzeta.
J. Cell Sci.
117
,
2513
-2521.
Lawitts, J. A. and Biggers, J. D. (
1993
). Culture of preimplantation embryos.
Methods Enzymol.
225
,
153
-164.
Lawrence, Y., Ozil, J. P. and Swann, K. (
1998
). The effect of a Ca2+ chelator and heavy-metal-ion chelators upon Ca2+ oscillations and activation at fertilisation in mouse eggs suggests a role for repetitive Ca2+ increases.
Biochem. J.
335
,
335
-342.
Leach, R. E., Stachecki, J. J. and Armant, D. R. (
1993
). Development of in vitro fertilized mouse embryos exposed to ethanol during the preimplantation period: accelerated embryogenesis at subtoxic levels.
Teratology
47
,
57
-64.
Lipp, P., Thomas, D., Berridge, M. J. and Bootman, M. D. (
1997
). Nuclear calcium signalling by individual cytoplasmic calcium puffs.
EMBO J.
16
,
7166
-7173.
Marangos, P. and Carroll, J. (
2004
). Fertilization and InsP3-induced Ca2+ release stimulate a persistent increase in the rate of degradation of cyclin B1 specifically in mature mouse oocytes.
Dev. Biol.
272
,
26
-38.
Marangos, P., FitzHarris, G. and Carroll, J. (
2003
). Ca2+ oscillations at fertilization in mammals are regulated by the formation of pronuclei.
Development
130
,
1461
-1472.
Mellstrom, B. and Naranjo, J. R. (
2001
). Mechanisms of Ca2+-dependent transcription.
Curr. Opin. Neurobiol.
11
,
312
-319.
Muto, A., Kume, S., Inoue, T., Okano, H. and Mikoshiba, K. (
1996
). Calcium waves along the cleavage furrows in cleavage-stage Xenopus embryos and its inhibition by heparin.
J. Cell Biol.
135
,
181
-190.
Ozil, J. P. and Huneau, D. (
2001
). Activation of rabbit oocytes: the impact of the Ca2+ signal regime on development.
Development
128
,
917
-928.
Patel, R., Holt, M., Philipova, R., Moss, S., Schulman, H., Hidaka, H. and Whitaker, M. (
1999
). Calcium/calmodulin-dependent phosphorylation and activation of human Cdc25-C at the G2/M phase transition in HeLa cells.
J. Biol. Chem.
274
,
7958
-7968.
Poenie, M. (
1990
). Alteration of intracellular Fura-2 fluorescence by viscosity: a simple correction.
Cell Calcium
11
,
85
-91.
Poenie, M., Alderton, J., Tsien, R. Y. and Steinhardt, R. A. (
1985
). Changes of free calcium levels with stages of the cell division cycle.
Nature
315
,
147
-149.
Poenie, M., Alderton, J., Steinhardt, R. and Tsien, R. (
1986
). Calcium rises abruptly and briefly throughout the cell at the onset of anaphase.
Science
233
,
886
-889.
Preston, S. F. and Berlin, R. D. (
1992
). An intracellular calcium store regulates protein synthesis in HeLa cells, but it is not the hormone-sensitive store.
Cell Calcium
13
,
303
-312.
Ratan, R. R., Shelanski, M. L. and Maxfield, F. R. (
1986
). Transition from metaphase to anaphase is accompanied by local changes in cytoplasmic free calcium in Pt K2 kidney epithelial cells.
Proc. Natl. Acad. Sci. USA
83
,
5136
-5140.
Ratan, R. R., Maxfield, F. R. and Shelanski, M. L. (
1988
). Long-lasting and rapid calcium changes during mitosis.
J. Cell Biol.
107
,
993
-999.
Roe, M. W., Lemasters, J. J. and Herman. B. (
1990
). Assessment of Fura-2 for measurements of cytosolic free calcium.
Cell Calcium
11
,
63
-73.
Rogers, N. T., Hobson, E., Pickering, S., Lai, F. A., Braude, P. and Swann, K. (
2004
). Phospholipase Czeta causes Ca2+ oscillations and parthenogenetic activation of human oocytes.
Reproduction
128
,
697
-702.
Saunders, C. M., Larman, M. G., Parrington, J., Cox, L. J., Royse, J., Blayney, L. M., Swann, K. and Lai, F. A. (
2002
). PLCζ: a sperm-specific trigger of Ca2+ oscillations in eggs and embryo development.
Development
129
,
3533
-3544.
Snow, P. and Nuccitelli, R. (
1993
). Calcium buffer injections delay cleavage in Xenopus laevis blastomeres.
J. Cell Biol.
122
,
387
-394.
Squirrell, J. M., Wokosin, D. L., White, J. G. and Bavister, B. D. (
1999
). Long-term two-photon fluorescence imaging of mammalian embryos without compromising viability.
Nat. Biotechnol.
17
,
763
-767.
Steinhardt, R. A. and Alderton, J. (
1988
). Intracellular free calcium rise triggers nuclear envelope breakdown in the sea urchin embryo.
Nature
332
,
364
-366.
Stricker, S. A. (
1999
). Comparative biology of calcium signaling during fertilization and egg activation in animals.
Dev. Biol.
211
,
157
-176.
Takahashi, M., Kagasaki, T., Hosoya, T. and Takahashi, S. (
1993
). Adenophostins A and B: potent agonists of inositol-1, 4, 5-trisphosphate receptor produced by Penicillium brevicompactum. Taxonomy, fermentation, isolation, physico-chemical and biological properties.
J. Antibiot.
46
,
1643
-1647.
Tang, T. S., Dong, J. B., Huang, X. Y. and Sun, F. Z. (
2000
). Ca2+ oscillations induced by a cytosolic sperm protein factor are mediated by a maternal machinery that functions only once in mammalian eggs.
Development
127
,
1141
-1150.
Terasaki, M. (
2000
). Dynamics of the endoplasmic reticulum and Golgi apparatus during early sea urchin development.
Mol. Biol. Cell
11
,
897
-914.
Tombes, R. M. and Borisy, G. G. (
1989
). Intracellular free calcium and mitosis in mammalian cells: anaphase onset is calcium modulated, but is not triggered by a brief transient.
J. Cell Biol.
109
,
627
-636.
Tombes, R. M., Simerly, C., Borisy, G. G. and Schatten, G. (
1992
). Meiosis, egg activation, and nuclear envelope breakdown are differentially reliant on Ca2+, whereas germinal vesicle breakdown is Ca2+ independent in the mouse oocyte.
J. Cell Biol.
117
,
799
-811.
Torok, K., Wilding, M., Groigno, L., Patel, R. and Whitaker, M. (
1998
). Imaging the spatial dynamics of calmodulin activation during mitosis.
Curr. Biol.
8
,
692
-699.
Tsien, R. Y. (
1989
). Fluorescent probes of cell signalling.
Annu. Rev. Neurosci.
12
,
227
-253.
Twigg, J., Patel, R. and Whitaker, M. (
1988
). Translational control of InsP3-induced chromatin condensation during the early cell cycles of sea urchin embryos.
Nature
332
,
366
-369.
Whitaker, M. and Patel, R. (
1990
). Calcium and cell cycle control.
Development
108
,
525
-542.
Whitaker, M. and Swann, K. (
1993
). Lighting the fuse at fertilization.
Development
117
,
1
-12.
Whitaker, M. and Larman, M. G. (
2001
). Calcium and mitosis.
Semin. Cell Dev. Biol.
12
,
53
-58.
Wilding, M., Wright, E. M., Patel, R., Ellis-Davies, G. and Whitaker, M. (
1996
). Local perinuclear calcium signals associated with mitosis-entry in early sea urchin embryos.
J. Cell Biol.
135
,
191
-199.
Yoda, A., Oda, S., Shikano, T., Kouchi, Z., Awaji, T., Shirakawa, H., Kinoshita, K. and Miyazaki, S. (
2004
). Ca2+ oscillation-inducing phospholipase C zeta expressed in mouse eggs is accumulated to the pronucleus during egg activation.
Dev. Biol.
268
,
245
-257.
Zernicka-Goetz, M., Ciemerych, M. A., Kubiak, J. Z., Tarkowski, A. K. and Maro, B. (
1995
). Cytostatic factor inactivation is induced by a calcium-dependent mechanism present until the second cell cycle in fertilized but not in parthenogenetically activated mouse eggs.
J. Cell Sci.
108
,
469
-474.