The Dynamics of PKC-Induced Phosphorylation Triggered by Ca2+ Oscillations in Mouse Eggs

Fertilization of mammalian eggs is characterized by a series of Ca2+ oscillations triggered by a phospholipase C activity. These Ca2+ increases and the parallel generation of diacylglycerol (DAG) stimulate protein kinase C (PKC). However, the dynamics of PKC activity have not been directly measured in living eggs. Here, we have monitored the dynamics of PKC-induced phosphorylation in mouse eggs, alongside Ca2+ oscillations, using fluorescent C-kinase activity reporter (CKAR) probes. Ca2+ oscillations triggered either by sperm, phospholipase C zeta (PLCζ) or Sr2+ all caused repetitive increases in PKC-induced phosphorylation, as detected by CKAR in the cytoplasm or plasma membrane. The CKAR responses lasted for several minutes in both the cytoplasm and plasma membrane then returned to baseline values before subsequent Ca2+ transients. High frequency oscillations caused by PLCζ led to an integration of PKC-induced phosphorylation. The conventional PKC inhibitor, Gö6976, could inhibit CKAR increases in response to thapsigargin or ionomycin, but not the repetitive responses seen at fertilization. Repetitive increases in PKCδ activity were also detected during Ca2+ oscillations using an isoform-specific δCKAR. However, PKCδ may already be mostly active in unfertilized eggs, since phorbol esters were effective at stimulating δCKAR only after fertilization, and the PKCδ-specific inhibitor, rottlerin, decreased the CKAR signals in unfertilized eggs. These data show that PKC-induced phosphorylation outlasts each Ca2+ increase in mouse eggs but that signal integration only occurs at a non-physiological, high Ca2+ oscillation frequency. The results also suggest that Ca2+-induced DAG formation on intracellular membranes may stimulate PKC activity oscillations at fertilization. J. Cell. Physiol. 228: 110–119, 2013. © 2012 Wiley Periodicals, Inc.

Intracellular Ca 2þ oscillations driven by hydrolysis of phosphatidylinositol 4,5-bisphosphate (PIP 2 ) to generate inositol 1,4,5-trisphosphate (InsP 3 ) and diacylglycerol (DAG) are one of the most common trans-membrane signal transduction systems used by cells (Berridge, 1993). At fertilization in mammals, the sperm causes a prolonged series of low frequency Ca 2þ oscillations that are driven by InsP 3 production in the unfertilized egg (Miyazaki et al., 1993;Kurokawa et al., 2004;Swann and Yu, 2008). Substantial evidence suggests that the sperm causes these Ca 2þ oscillations by introducing a novel, sperm-specific phospholipase C enzyme isoform, phospholipase C zeta (PLCz), into the egg after gamete membrane fusion Swann and Yu, 2008;Nomikos et al., 2012). The exclusive introduction of PLCz into eggs by microinjecting its cognate cRNA has been shown to precisely mimic the characteristic series of sperminitiated Ca 2þ oscillations observed at fertilization Saunders et al., 2002). These distinctive oscillations in cytosolic free Ca 2þ appear to involve a positive feedback loop consisting of InsP 3 -induced Ca 2þ release and Ca 2þ -dependent production of InsP 3 by PLCz .
The phenomenon of Ca 2þ oscillations initiated at fertilization in mouse eggs have been shown to be the specific trigger for egg activation events, including granule exocytosis, exit from metaphase II arrest, and entry into first mitotic division (Kline and Kline, 1992). A major issue that remains unresolved is how the intrinsically repetitive nature of the sperm-activated Ca 2þ signals is specifically transduced into downstream egg activation events. It has been suggested that the fertilized egg is able to either, integrate the total Ca 2þ flux, or count the number of Ca 2þ spikes, or else read the frequency of Ca 2þ oscillations (Meyer and Stryer, 1991;Ducibella et al., 2002;Ducibella and Fissore, 2007). So far, recruitment of maternal mRNA and embryo development to term have been found to be affected by the number of Ca 2þ transients recorded in mouse eggs (Ozil and Swann, 1995;Ducibella et al., 2002;Ozil et al., 2006). The integral of Ca 2þ increases in the egg has also been correlated with activation rate in the mouse (Ozil et al., 2005). The main essential target for Ca 2þ oscillations in mouse fertilization is calmodulin-dependent protein kinase II (CaMKII; Ducibella and Fissore, 2007) and assays of CaMKII at fertilization suggest that its kinase activity oscillates in near synchrony with Ca 2þ oscillations (Markoulaki et al., 2004). However, it is not known whether protein phosphorylation driven by CaMKII responds in a manner that is able to either count or integrate Ca 2þ oscillations.
Another protein kinase that has been shown to increase in activity at fertilization is protein kinase C (PKC; Gallicano et al., 1997;Tatone et al., 2003;Kalive et al., 2010). PKC stimulation alone is not sufficient for egg activation, but it could play a significant role since addition of the PKC activator, PMA (phorbol myristate acetate), to mouse eggs can cause activation, and the presence of pseudo-substrate inhibitors have been reported to interfere with activation at fertilization (Gallicano et al., 1993(Gallicano et al., , 1997Moses and Kline, 1995). PKC could also play an important role in causing Ca 2þ influx at fertilization, which is important for maintaining Ca 2þ oscillations . There are 10 mammalian PKC isoforms, classified into three major subfamilies (Mellor and Parker, 1998;Newton, 2003): the conventional PKCs (cPKC) a, bI, bII, and g are stimulated by both Ca 2þ and DAG; in contrast, novel PKCs (nPKC) d, e, h, and u are regulated by DAG but are Ca 2þindependent. Atypical PKCs (aPKC) z and i/l are neither regulated by Ca 2þ nor by DAG. Isoforms from all three subfamilies have been found to be expressed in mammalian eggs (Jones, 1998;Luria et al., 2000;Pauken and Capco, 2000;Halet, 2004;Baluch and Capco, 2008). A specific role for PKC may have a particular relevance for eggs because PKC can act as a decoder of Ca 2þ oscillations (Oancea and Meyer, 1998;Cullen, 2003;Violin et al., 2003). This decoding phenomenon can involve the sequential binding of Ca 2þ and DAG to the C2 and C1 domains of cPKCs, respectively, turning the kinase into its activated state with translocation to the plasma membrane (Oancea and Meyer, 1998;Violin et al., 2003). The cPKCs, PKCa, and PKCbI translocate to the plasma membrane during fertilization in mouse eggs (Luria et al., 2000). Significantly, GFPtagged versions of PKCa or g were found to translocate in response to individual Ca 2þ transients, and following decline of Ca 2þ to basal levels, the GFP-PKCs return to the cytosol . Hence, PKC activation/translocation does not appear to outlast the Ca 2þ transients, although phosphorylation events specifically induced by the activated PKC might last for longer than the Ca 2þ transients. However, in vitro PKC kinase assays performed on egg lysates are not able to accurately monitor phosphorylation occurring in a single egg with sufficient time resolution (Gallicano et al., 1997;Tatone et al., 2003). Consequently, it remains unknown whether each cycle of PKC activity-induced phosphorylation is able to significantly outlast the duration of each Ca 2þ transient at fertilization.
In addition to Ca 2þ -dependent cPKC, unconventional PKCs also contribute to PKC activity at fertilization. In particular, PKCd is implicated as being the isoform responsible for a significant proportion of the biochemically measurable PKC increase occurring at fertilization (Tatone et al., 2003). PKCd is known to be phosphorylated during oocyte maturation and then becomes dephosphorylated during the early stages of egg activation (Viveiros et al., 2001(Viveiros et al., , 2003. The phosphorylation event is essential for PKCd activation and, since PKCd is required for oocyte maturation, it was suggested that the PKCd phosphorylation reflects its activation state. However, up to now there have been no studies that have measured PKCd-specific activity in eggs in real time. PKC-induced phosphorylation has been monitored dynamically in cells using a CKAR, a probe that undergoes changes in fluorescence resonance energy transfer (FRET) in response to phosphorylation. CKAR consists of a pseudosubstrate that is specific to PKC fused to a FHA2 domain that binds phosphothreonine. This fusion protein is in turn flanked by a cyan fluorescent protein (CFP) and yellow fluorescent protein (YFP) at either end. A change in FRET between the CFP and YFP is caused by changes in CKAR conformation when the PKC-specific substrate is phosphorylated and bound by the FHA2-binding domain (Violin et al., 2003). CKAR has been shown to be subject to phosphorylation and dephosphorylation in cells (Violin et al., 2003;Gallegos et al., 2006). Myristoylated CKAR, which is targeted specifically to the plasma membrane, has been shown to undergo oscillations in FRET signal in response to Ca 2þ transients in cell lines. The FRET response in the plasma membrane of cells was delayed with respect to Ca 2þ transients by 10-15 sec. In contrast, the cytoplasmic CKAR did not show any oscillations in FRET signal during Ca 2þ oscillations (Violin et al., 2003). In mouse eggs, it is unknown whether PKC activity might show a similar Ca 2þ response pattern to that exhibited in somatic cells. Hence, in the present study, we have monitored the dynamics of PKC-induced phosphorylation during Ca 2þ oscillations in mouse eggs using both the cytoplasmically located CKAR, and its membrane-targeted form, MyrPalm-CKAR. In addition, we monitored phosphorylation of dCKAR, which specifically responds to PKCd activation (Kajimoto et al., 2010). Our data show that there are distinct oscillation patterns in PKC activity within the cytoplasm and the plasma membrane that occur in response to physiological Ca 2þ transients in mouse eggs. The stimulation of PKC activity outlasts each Ca 2þ transient by several minutes and appears to involve both cPKCs and PKCd. Our data suggest that in mouse fertilization, the Ca 2þ signal-induced DAG formation may play a precise role in generating oscillations in PKC activation-mediated phosphorylation.
Gamete collection and handling MF1 female mice were super-ovulated by intraperitoneal injection of 7.5 i.u. of PMSG (pregnant mare's serum gonadotrophin; Folligon) followed 48 h later by 10 i.u. of hCG (human chorionic gonadotropin; Folligon; Saunders et al., 2002). Eggs (13-16 h post-hCG) were released from the oviduct into warmed M2 medium (Sigma, Dorset, UK). Oocytes were held in drops of M2 medium under paraffin oil in Falcon tissue culture dishes. Cumulus cells were removed by a brief exposure to hyaluronidase and the zona pellucida removed by exposure to acid Tyrode's solution (Sigma). For all fluorescence recordings, the eggs were placed in drops of HEPES-buffered KSOM (HKSOM) media . For media with Sr 2þ , HKSOM media was used where the CaCl 2 was omitted and replaced with 10 mM SrCl 2 . Spermatozoa were expelled from the cauda epididymis of male CBA/C57 mice into 1 ml of T6 medium containing 16 mg/ml BSA, and incubated under oil for 2-3 h at 378C and 5% CO 2 to capacitate. For in vitro fertilization (IVF) experiments, approximately 10 ml of sperm suspension was added to the dish containing the eggs.

cRNA synthesis and microinjection
Complementary RNA (1 mg/ml) encoding CKAR, MyrPalm-CKAR (Violin et al., 2003), dCKAR (Kajimoto et al., 2010), and mouse PLCz  were synthesized and polyadenylated using mScript TM mRNA Production System (Epicentre, Calbiochem, Nottingham, UK) following the manufacturer's instructions. Microinjection of cRNA into mature mouse eggs was performed as previously described , followed by a 3 h incubation at 378C to allow the cRNA to be transcribed at detectable levels of expression.

Measurement of CKAR and intracellular Ca 2R
Zona-free MII eggs were kept in HKSOM under mineral oil at 378C on the heated stage chamber of an inverted microscope (Nikon UK, Kingston upon Thames, UK). For Ca 2þ measurements, Rhoddextran was co-injected with CKAR or PLCz cRNAs. One of the issues when measuring FRET together with Ca 2þ concentration changes is that the fluorescence spectra from YFP and CFP overlap with some fluorescent Ca 2þ indicators (e.g., Fura2, Fluo3, and FuraRed). This potential for fluorescence signal ''spill-over'' can distort FRET ratios. In contrast, Rhod-dextran is a longwavelength Ca 2þ indicator with a fluorescence excitation and emission maxima of 530 and 576 nm, respectively, and it is the dextran-coupled version of Rhod2 that is retained in the cytoplasm. Fluorescence was captured using a 20 Â 0.75 NA objective at 10sec intervals with a cooled CCD camera (Coolsnap HQ 2 , Photometrics, Tucson, AZ) and In Vivo software. The excitation light source was a white LED lamp (OptoLEDLite, Cairn Research Ltd., Faversham, UK). Filters (10 nm bandwidth) were controlled using filter wheels (Lambda 10-3; Sutter Instruments, Novato, CA). FRET signals were measured by taking the ratio of emission at 470 and 535 nm with excitation at 430 nm. Rhod-dextran was excited with 550 nm light and emission collected at 600 nm. A multi-band filter (XF2054, from GlenSpectra, Stanmore, UK, part of HORIBA Scientific) was placed in the dichroic filter block housing to allow excitation and emission at the selected wavelengths. Image J and SigmaPlot software (Systat Software, Inc., Hounslow, UK) were used for data analysis. Data are plotted as the ratio of cyan fluorescence to yellow emission and values normalized to the percentage change from the start of the experiment. Specific PKC inhibitors Gö 6976 and rottlerin, were used to act upon conventional PKCa and PKCb, and the novel PKCd isoforms, respectively.

Confocal imaging
CKAR and MyrPalm-CKAR cRNA were expressed in eggs as described above, and dCKAR was co-injected with PLCz and incubated for 2-8 h. Single snapshots were taken of the subcellular distribution corresponding to CKAR/MyrPalm-CKAR/dCKAR FRET probes using a confocal microscope (TCS SP5; Leica, Milton Keynes, UK), under a 20Â (0.75 NA) lens, and an argon laser. FRET signal was determined using the same ratiometric settings as described above and data were analyzed using Image J.

Statistical analysis
The % CKAR FRET changes of individual signal increases were calculated based on the mean of three random spikes taken from each oscillating egg trace and divided by the total number of eggs. The ''n'' refers to the total number of eggs examined for each experiment type.

Monitoring PKC-induced phosphorylation in eggs
We found that CKAR was effectively expressed in mature mouse eggs, following microinjection of its cRNA, as indicated by fluorescence detected in the CFP and YFP channels. PKC activity, as reflected by the phosphorylation of CKAR, was monitored by measuring the ratio of the CFP to YFP signal intensity and plotted as the percentage change over the starting ratio versus time. These data are presented as the inverse of FRET efficiency, since there is an increase in this ratio with increased phosphorylation (Violin et al., 2003). Confocal images of CKAR show that it is distributed widely throughout the cytosol, with the possible exception of some exclusion by organelles (Fig. 1a). In contrast, MyrPalm-CKAR was detected specifically in the plasma membrane (Fig. 1a). This distinct localization is consistent with the fact that MyrPalm-CKAR contains seven residues of the Lyn kinase fused to the N-terminus that targets the probe to the membrane via myristoylation and palmitoylation post-translational modifications (Zacharias et al., 2002). Figure 1b shows that the addition of the potent PKC activator, PMA (200 nM), caused a CFP/YFP signal increase which reached saturation after addition of the phosphatase inhibitor, calyculin A (16.9 AE 0.25%, n ¼ 4). Similar results were seen with MyrPalm-CKAR (data not shown). These data suggest that CKARs can be successfully expressed in mouse eggs and respond to stimuli that specifically activate PKC, and that endogenous phosphatases are continuously active in reducing the level of PKC-induced phosphorylation (Violin et al., 2003). It also reveals that the full dynamic range of the CKAR expressed in mouse eggs involves changes of <10% of the resting signal.
We tested the Ca 2þ -dependence of the CKAR response by the addition of the Ca 2þ pump inhibitor, thapsigargin, and the Ca 2þ ionophore, ionomycin, both of which cause monotonic Ca 2þ increases in mouse eggs. Figure 1c shows that addition of thapsigargin triggered a CKAR signal increase of only 3.30 AE 0.84% (n ¼ 17), whereas the subsequent addition of ionomycin effected a 7.78 AE 1.96% (n ¼ 17) FRET increase. The greater FRET increase produced by ionomycin was correlated with a larger amplitude Ca 2þ transient. The Ca 2þ -induced CKAR signal increase was diminished, although not abolished, in the presence of cPKC inhibitor, Gö 6976 (10 mM), with either thapsigargin (1.92 AE 0.79%, n ¼ 8) or ionomycin (4.56 AE 1.20%, n ¼ 8; Fig. 1d). These Gö 6976-mediated inhibition data suggest that there is a partial contribution of cPKC-mediated phosphorylation to the CKAR response. It should be noted that Gö 6976 is the only commonly used PKC inhibitor that is non-fluorescent and does not interfere with the FRET signal. As reported previously, other broad-spectrum PKC inhibitors (e.g., Gö 6983 or BIM) are fluorescent and cannot be readily used to inhibit CKAR responses without interfering with the CFP or YFP fluorescence signals required for FRET analysis (Gallegos et al., 2006).

PKC-induced phosphorylation at fertilization
PKC-induced phosphorylation was monitored during IVF, using both CKAR and MyrPalm-CKAR, and their FRET signal change measured every 10 sec, alongside the occurrence of cytosolic Ca 2þ oscillations. Figure 2a shows that Ca 2þ oscillations following IVF of mouse eggs occurred in near synchrony with oscillatory increases in the cytoplasmic CKAR signal. The plasma membrane-localized MyrPalm-CKAR also showed comparable patterns of oscillatory FRET signal changes, similar in form to the cytoplasmic CKAR probe (Fig. 2b). The oscillatory increases in CKAR signal were small and typically displayed less than a 5% ratio change. This change was entirely due to CKAR since control IVF experiments conducted in the absence of Rhod-dextran still showed oscillatory CKAR increases, and measuring Ca 2þ oscillations in the absence of CKAR showed no discernable oscillations in the CFP/YFP channel ( Supplementary Fig. S1). The overall duration of Ca 2þ oscillations in fertilizing eggs was not different between eggs with or without CKAR or MyrPalm-CKAR (Table 1). Hence all of the eggs studied, stopped their Ca 2þ oscillations on schedule. Since the cessation of the Ca 2þ signal is due to the formation of pronuclei, this suggests that the timing of egg activation events was unaffected by the presence of CKAR or MyrPalm-CKAR. In these IVF experiments, the CKAR response was not blocked by the presence of Gö6976 in fertilized eggs (Fig. 2c). Moreover, no inhibitory effect was seen following Gö 6976 addition upon the CKAR oscillations induced by either Sr 2þ or PLCz (data not shown). Thus, it is unclear whether this oscillatory CKAR phosphorylation signal change occurring upon mouse fertilization involves the direct activation of cPKCs by each Ca 2þ transient.
A consistent feature of the CKAR-mediated oscillations at fertilization was that each of the FRET transients showed a different time-course relative to the Ca 2þ transients. Figure 3 shows a series of three Ca 2þ transients during IVF at a higher time resolution. The amplitude for the cytosolic CKAR (4.36 AE 0.53%, n ¼ 30) was relatively larger than that for MyrPalm-CKAR (3.42 AE 0.24%, n ¼ 31). However, for both CKAR and MyrPalm-CKAR, the peak of the FRET transient occurred 10-30 sec after the peak of the Ca 2þ signal. In addition, the CKAR response displayed a slower decline, and did not return to baseline until $5 min after the Ca 2þ transient had finished. However, it was notable that each CKAR signal increase had returned to near baseline value prior to initiation of the next Ca 2þ spike, and hence there was no sign of progressive accumulation of the CKAR signal in Figure 2 or 3.
To determine if any long-term integration of response could occur, we tested the effects of higher frequency oscillations. Injecting high concentrations of PLCz cRNA has been shown to cause high-frequency Ca 2þ oscillations in mouse eggs . In Figure 4, eggs were microinjected with a calibrated amount of PLCz (0.1 mg/ml pipette concentration) that was specifically chosen to generate high-frequency Ca 2þ oscillations. With either CKAR or MyrPalm-CKAR there was an increase in FRET signal that did not fully return to baseline between Ca 2þ oscillations, hence the CKAR response appeared to integrate with time. However, even in these cases the FRET signal could decline considerably as the frequency of oscillations decreased (as in Fig. 4b). These data suggest that only high-frequency Ca 2þ oscillations are able to produce a significant integration of PKC-induced phosphorylation in eggs.

PKCd activity in eggs
Previous studies have implicated a role for PKCd in egg activation at fertilization, so we conducted similar experiments to those described above in mouse eggs, using a newly developed PKCd isoform-specific probe, dCKAR (Kajimoto et al., 2010). The dCKAR was expressed in mouse eggs throughout the cytoplasm, and persisted for at least 8 h after PLCz cRNA injection, with some dCKAR signal being present in the pronucleus (Fig. 5a). There was only a very small increase in the dCKAR signal upon PMA addition ( Fig. 5b; 2.04 AE 0.29%, n ¼ 6), compared to eggs injected with conventional CKAR (11.22 AE 0.83%, n ¼ 4). Only when we added the phosphatase inhibitor calyculin A, did the dCKAR signal show a significant response (5.13 AE 0.53%) similar to that of CKAR (5.68 AE 0.25%), although the time course was slow. These results suggest that PKCd cannot be readily activated by PMA in mouse eggs. This could be explained if this isoform of PKC already has some activity in unfertilized mature MII mouse eggs (Viveiros et al., 2003). To test this hypothesis, rottlerin, a known PKCd-specific inhibitor, was added to mature unfertilized mouse eggs. Figure 5c shows that this inhibitor caused a significant decrease in the dCKAR signal Responses are shown from typical eggs in response to thapsigargin (20 mM) and then ionomycin (5 mM). In (d) the conditions are the same as in (c)but the egg was incubated in the presence of conventional PKCs inhibitor Gö 6976 (10 mM) before the addition of thapsigargin and ionomycin. The ''n'' numbers refer to the total number of eggs examined for each experiment type. [Color figure can be seen in the online version of this article, available at http://wileyonlinelibrary.com/journal/jcp] (12.74 AE 2.15%). Rottlerin also caused a drop in the CKAR signal in unfertilized eggs, indicating that the majority of PKC activity in mouse eggs may derive from that of PKCd ( Supplementary Fig. S2a). It was also noted that rottlerin addition caused a small Ca 2þ increase in mouse eggs (Fig. 5c) and an immediate decrease in the oscillating CKAR signal in fertilized eggs, which was followed by a gradual increase in the cytosolic free Ca 2þ levels and eventually, the cessation of the sperm-induced Ca 2þ oscillation ( Supplementary Fig. S2b). These changes might be due to an effect upon Ca 2þ influx (Xu, 2007), but is unlikely to account for the decrease in dCKAR signal, since this was delayed in comparison to the rottlerininduced FRET change. These data support the idea that PKCd is already active to some extent in an unfertilized mouse egg. However, in contrast to the unfertilized egg, PMA caused a dCKAR signal increase in fertilizing eggs when it was added during the course of Ca 2þ oscillations (5.93 AE 1.66%; Fig. 5d) at about 2 h after sperm addition. This PMA-induced signal increase suggests that whilst PKCd is mostly active in unfertilized mouse eggs, it may decline in activity during the early stages of egg activation. Interestingly, there were small oscillations in the dCKAR signal at fertilization before we added PMA (Fig. 5d), which suggests that some PKCd activity can still be further stimulated in the fertilizing mouse egg.
Since PKCd can only be stimulated by DAG and not by Ca 2þ directly, we examined dCKAR signals during Ca 2þ oscillations caused by two different stimuli, Sr 2þ or PLCz. Figure 6 shows that both Sr 2þ -and PLCz-induced Ca 2þ oscillations lead to dCKAR signal increases similar in form to those seen with IVF in Figure 3. However, the dCKAR phosphorylation signal oscillations (Fig. 6b) generated by Sr 2þ were even more delayed compared with those observed for conventional CKAR (Fig. 6a). In addition, the dCKAR signal peak occurred after the Ca 2þ levels in the egg had declined, leading to dCKAR oscillations that were out of phase with Ca 2þ oscillations. There is also a slight reduction in the Sr 2þ -induced signal changes with dCKAR (2.32 AE 0.22%, n ¼ 15) compared to conventional CKAR (3.63 AE 0.59%, n ¼ 23). In contrast, Figure 6c,d shows that PLCz caused Ca 2þ oscillations and similar FRET oscillatory signal increases using either cytosolic CKAR (4.97 AE 0.33%) or dCKAR (4.47 AE 0.46%). These signal changes are comparable in amplitude and pattern to those seen at fertilization. These data suggest that PKCd responds differently to Sr 2þ compared with fertilization by sperm or with PLCz.

Discussion
Mammalian fertilization is characterized by a sperm-induced series of Ca 2þ oscillations in the egg that are critical for the physiological activation of embryo development (Kline and Kline, 1992;Ducibella et al., 2002). Previous studies have shown that PKC activity is increased at fertilization in mouse eggs (Gallicano et al., 1997;Tatone et al., 2003). There is also evidence that PKC plays a role in normal meiotic resumption after fertilization in the mouse (Gallicano et al., 1993(Gallicano et al., , 1997Moses and Kline, 1995). In this study, we have specifically set   out to study the precise relationship between Ca 2þ oscillations and PKC activity, since this topic has not been previously addressed. This is due to the absence of precise time resolution when using cell extract-based biochemical assays of PKC activity. This study shows for the first time the dynamic changes in PKC-induced phosphorylation events during fertilization in a living mammalian egg. We have achieved this by using FRETbased probes for PKC-induced phosphorylation (CKARs) to measure these dynamic changes alongside Ca 2þ oscillations. CKAR and its subcellular-targeted derivatives have shown to be specific for monitoring PKC-induced phosphorylation and are subject to dephosphorylation by cellular phosphatases (Violin et al., 2003;Gallegos et al., 2006). The expression of CKAR or MyrPalm-CKAR did not appear to have any inhibitory effect upon egg activation since Ca 2þ oscillations terminated similar to controls and the cessation of oscillations in mouse eggs is due to pronuclear formation (Marangos et al., 2003). Our data show that PKC-induced phosphorylation events can outlast the duration of individual Ca 2þ spikes by several minutes. Significantly, this prolongation of phosphorylation relative to the Ca 2þ signal can occur in both the cytoplasmic compartment and the plasma membrane. Furthermore, experiments using the PKCd isoform-specific probe, dCKAR (Kajimoto et al., 2010), suggest that Ca 2þ can stimulate PKCs both through the generation of DAG, as well as via direct Ca 2þ -dependent binding and activation.

Integration of Ca 2R oscillations by PKC-induced phosphorylation
One of the outstanding Ca 2þ signaling issues in eggs is how oscillatory Ca 2þ changes are translated, and possibly integrated, into changes in the activity of relevant target enzymes (Ducibella and Fissore, 2007). Previous studies in somatic cell lines have found that cPKC, such as PKCg, can act as an integration module for decoding Ca 2þ oscillations that are associated with phosphoinositide turnover, by virtue of its ability to become activated and translocate to the plasma membrane (Oancea and Meyer, 1998). This integration of Ca 2þ oscillations relies upon DAG production within the plasma membrane causing prolonged membrane residence of cPKC. Fertilization also stimulates translocation of PKCa, b, and g to the plasma membrane (Raz et al., 1998;Luria et al., 2000;Baluch et al., 2004). At fertilization, however, whilst each Ca 2þ transient leads to plasma membrane translocation of cPKC-GFP in mouse eggs, the cPKC-GFPs return to the cytoplasm within $10 sec of the cytosolic-free Ca 2þ returning to resting levels . This implies only a very limited integration of PKC activity, which could be due to very limited accumulation of DAG in the plasma membrane during fertilization Yu et al., 2008). Nevertheless, a PKC signal in the cell may persist for a longer period because phosphorylated substrates may outlast the PKC translocation process. The dynamics of phosphorylation events induced by PKCs has been monitored in somatic cell lines using CKAR, and derivatives of CKAR targeted to sub-cellular compartments (Cullen, 2003;Violin et al., 2003). It is not known whether the phosphorylation of CKAR precisely reflects the phosphorylation and dephosphorylation of endogenous PKC substrates. However, CKAR has been shown to be a specific substrate for PKC and subject to dephosphorylation by the same type of phosphatases as endogenous PKC substrates. It is important to note that the increase in CKAR signal we see at fertilization is relatively small ($5%), but this reflects the intrinsic limitation of the probe, rather than the response of the cell, which is likely to involve a much larger change in phosphorylation. In fact FRET probes of the same class as CKAR all show small changes in signal. In a previous study in eggs, for example, a CFP/YFP-based probe for InsP 3 showed <5% increases in eggs after PLCz injection, despite the fact that InsP 3 probably increases by several fold (Shirakawa et al., 2006).
In the study in somatic cells by Violin et al. (2003), it was found that phosphorylation of plasma membrane CKAR outlasted the cytoplasmic Ca 2þ spikes by 10-15 sec. Whilst significant in cell lines, this degree of integration would not be sufficient in mammalian fertilized eggs, where each Ca 2þ transient lasts for approximately 1 min and are typically spaced 10 min apart. We found that the phosphorylation of CKAR in both the cytoplasm and plasma membrane is maintained for about 5 min after each Ca 2þ transient during repetitive oscillations in mouse eggs, which is significantly longer than the 10-15 sec observed in previous somatic cell studies using CKAR. However, this extended phosphorylation time-course still results in the PKC signal in eggs returning to near-resting levels within the 10 min before the next Ca 2þ transient begins. This response profile might be sufficient for the PKCstimulated Ca 2þ influx that occurs after each Ca 2þ transient in mouse eggs (McGuinness et al., 1996), but it does not provide the basis for explaining longer-term effects. In contrast, we were able to see a clear accumulation of the CKAR response when we injected high concentrations of PLCz to deliberately cause high-frequency Ca 2þ oscillations. This result suggests that the degree of CKAR phosphorylation can be varied in response to the frequency of Ca 2þ oscillations. However, this cumulative effect is only observed with a Ca 2þ oscillation frequency well above that observed physiologically at fertilization. Hence, it appears unlikely that the primary phosphorylation events induced by PKC activation are able to integrate the lower frequency Ca 2þ oscillations occurring during normal fertilization.
Ca 2R -induced DAG formation as the stimulus for PKC In our experiments, there is a distinct increase in the CKAR signal observed in response to each Ca 2þ transient. The elevations in free Ca 2þ concentration could stimulate this PKC activity increase by two potential mechanisms; by direct binding of Ca 2þ to the C2 domain or by stimulating PLC-mediated DAG production, which then binds to the PKC C1 domain. Ca 2þstimulated DAG production is likely to occur in fertilizing mammalian eggs because it has been shown that sperm PLCz activity is very sensitive to increases in cytosolic Ca 2þ levels (Nomikos et al., 2005). In addition, Ca 2þ -dependent InsP 3 production has been shown to be part of the mechanism of Ca 2þ oscillations and this implies that oscillatory increases in both InsP 3 and DAG occur during each Ca 2þ transient . Our data suggest that PKCs may be stimulated directly by Ca 2þ , but that Ca 2þ -induced DAG formation may also form a significant component of the PKC response. All of the stimuli that cause an elevation of Ca 2þ in eggs lead to an increase in the CKAR signal. The CKAR response was not effectively blocked by the cPKC inhibitor, Gö 6976, with the exception of thapsigargin, which only causes a small increase in Ca 2þ . Surprisingly, we found no effect of Gö 6976 on fertilization-induced CKAR increases. Either Gö 6976 may not be fully effective at inhibiting PKC in mouse eggs, or it could also suggest that the conventional isoforms of PKC are partially involved in stimulating some of the CKAR in response to Ca 2þ elevation. This second idea is further supported by the finding, that despite its pre-existing basal activity in unfertilized eggs, dCKAR can be further stimulated by the Ca 2þ transients induced by fertilization, PLCz and Sr 2þ -containing media. The presumed mechanism for Ca 2þ to stimulate dCKAR is via DAG production. Sr 2þ media is of particular interest because it is thought to act via stimulating InsP 3 receptors to release Ca 2þ (Marshall and Taylor, 1994;Zhang et al., 2005). Unlike the sperm and PLCz, Sr 2þ medium does not lead to any detectable down-regulation of InsP 3 receptors and so is not expected to cause significant PIP 2 hydrolysis (Brind et al., 2000;Jellerette et al., 2000). Our data show that Sr 2þ -induced Ca 2þ oscillations are accompanied by some dCKAR signal, implying that these Ca 2þ increases alone can cause some DAG production. This Sr 2þ -mediated mechanism could involve Ca 2þ stimulation of other egg-derived PLCs such as PLCb1, which appears to be stimulated to some extent at fertilization in mouse eggs (Igarashi et al., 2007). It was, however, noted that the amplitude and time course of dCKAR stimulation was different between Sr 2þ and PLCz. The Sr 2þ response was smaller and more delayed with respect to the Ca 2þ transient than that with PLCz, which, in turn, could be due to a delay in DAG production. Previous studies have found that Ca 2þ ionophores induced DAG accumulation in the plasma membrane with a delay of a few minutes in unfertilized eggs Yu et al., 2008). This implies that the Ca 2þ -induced stimulation of PLCz generates DAG much more rapidly than that provided by Ca 2þ stimulation of other egg-derived PLCs.

Basal PKC activity in eggs
Previous studies have suggested that there might be a basal level of PKC activity present in mouse eggs or muscle cells (Nicolas et al., 1998;Akabane et al., 2007). PKCd has been shown to be phosphorylated at an activating residue in mature mouse eggs, and hence PKCd may already be active at the MII stage (Viveiros et al., 2003). Our data are consistent with this idea, since the PKCd-specific inhibitor, rottlerin, caused a clear decrease in the dCKAR signal in an unfertilized egg. Addition of PMA only caused a minimal increase, although there was a large increase in the dCKAR signal when added over an hour into the activation process. This result is consistent with previous reports showing that PKCd dephosphorylation occurs after egg activation (Viveiros et al., 2003). Nevertheless, there were still small increases in the dCKAR signal associated with Ca 2þ transients at fertilization, suggesting that PKCd substrates are not completely phosphorylated in an unfertilized egg.

The nature of cytoplasmic PKC activity oscillations
One of the most remarkable results of the current study was that a PKC-induced response is detected with both the plasma membrane-targeted and cytoplasmic CKAR. Previous studies of PKC in live somatic cells have shown that agonists can lead to DAG production, although PKC oscillations only occur in the plasma membrane (Oancea and Meyer, 1998;Violin et al., 2003). To date, the evidence for a PKC-induced phosphorylation response that outlasts oscillating Ca 2þ transients (by $15 sec) is within the plasma membrane (Violin et al., 2003). The previous dynamic PKC imaging in mouse eggs has also entirely concerned short-term translocation to the plasma membrane Yu et al., 2008). Our new data show that longer-lasting phosphorylation increases occur in fertilizing mouse eggs, both in the cytoplasm and the plasma membrane. In fact, the CKAR signal is notably stronger in the cytoplasm than at the plasma membrane. This suggests that the majority of DAG formation and subsequent PKC stimulation occurs at sites within the egg cytoplasm in response to Ca 2þ transients. In somatic cells, agonist stimulation can lead to DAG generation in the Golgi membranes as well as the plasma membrane (Gallegos et al., 2006). Internal membrane organelles in mouse eggs could therefore also be a potential source of DAG at fertilization. In accord with this possibility, we have recently found that mouse eggs contain a significant amount of PIP 2 specifically located in internal vesicles (Yu et al., 2012). Moreover, these discrete intracellular vesicles appear to be the precise target of PLCz-induced PIP 2 hydrolysis. Therefore, it is distinctly possible that the sperm-delivered PLCz enables Ca 2þ -dependent DAG formation on intracellular PIP 2 -containing vesicles, facilitating repetitive PKC stimulation throughout the egg cytoplasm. Further experiments could address this possibility by the use of DAG-specific probes targeted to intracellular vesicles.