Role of Phospholipase C- (cid:1) Domains in Ca 2 (cid:2) -dependent Phosphatidylinositol 4,5-Bisphosphate Hydrolysis and Cytoplasmic Ca 2 (cid:2) Oscillations*

The sperm-specific phospholipase C- (cid:1) (PLC (cid:1) ) elicits fer-tilization-like Ca 2 (cid:2) oscillations and activation of embryo development when microinjected into mammalian eggs (Saunders, C. M., Larman, M. G., Parrington, J., Cox, L. J., Royse, J., Blayney, L. M., Swann, K., and Lai, F. A. (2002) Development ( Camb. ) 129, 3533–3544; Cox, L. J., Larman, M. G., Saunders, C. M., Hashimoto, K., Swann, K., and Lai, F. A. (2002) Reproduction 124, 611–623). PLC (cid:1) may represent the physiological stimulus for egg activation and development at mammalian fertilization. PLC (cid:1) is the small-est known mammalian PLC isozyme, comprising two EF hand domains, a C2 domain, and the catalytic X and Y

A series of pre-programmed biochemical events are triggered during fertilization. The earliest signaling event in the activation of an egg by a sperm is a large, transient increase in intracellular free calcium ion concentration ([Ca 2ϩ ] i ) (1,2). In many nonmammalian species, such as sea urchin, the observed Ca 2ϩ increase in the egg comprises a single transient, but in mammals and some marine invertebrates there is a series of repetitive Ca 2ϩ oscillations (3,4). The frequency and duration of these Ca 2ϩ oscillations vary between species (1). In response to this Ca 2ϩ signal, the fertilized egg completes meiosis and initiates the process of embryonic development (5).
Several lines of evidence implicate the inositol 1,4,5trisphosphate (IP 3 ) 1 signaling pathway (6) as the origin of the Ca 2ϩ signals in mammalian eggs. IP 3 is produced by hydrolysis of phosphatidylinositol 4,5-bisphosphate in a reaction that is catalyzed by phosphoinositide-specific phospholipase (PI-PLC) (6). Liberated IP 3 then causes Ca 2ϩ release by binding to IP 3 receptors located on the endoplasmic reticulum of eggs and oocytes (7,8). The essential role of IP 3 and the IP 3 receptor in fertilization is illustrated by studies in mouse and hamster eggs, where Ca 2ϩ oscillations at fertilization can be inhibited by microinjection of antibodies that inhibit the IP 3 receptor (9) or by down-regulation of IP 3 receptors (10,11). In addition, it has been shown that sustained injection of IP 3 , the repeated photorelease of caged IP 3 , or the microinjection of the IP 3 analogue adenophostin can all lead to a series of Ca 2ϩ oscillations in eggs (8,12,13). Hence, in mammalian eggs, IP 3 is both necessary and potentially sufficient to explain the Ca 2ϩ oscillations observed at fertilization. However, the precise mechanism employed by a sperm to generate an IP 3 increase in the egg is not established.
In mammalian eggs there is evidence for the existence of different types of PI-PLCs. The stimulation of egg-derived PLC isoforms of the ␤ or ␥ class by receptor tyrosine kinases (14) or by guanine nucleotide-binding proteins (G-proteins) (15) can lead to Ca 2ϩ release in different species of eggs. However, there is evidence that neither the tyrosine kinase nor the G-protein pathways are necessary for Ca 2ϩ release at fertilization in mouse eggs (2). In contrast to classical transmembrane receptor signaling, there is evidence that mammalian eggs can be activated by a sperm protein component that is introduced into the egg after gamete membrane fusion (1,4). Egg microinjection of sperm extracts, or whole sperm, triggers Ca 2ϩ oscillations similar to fertilization in a range of different species (17)(18)(19)(20)(21)(22)(23). The injection of such a sperm factor also activates the development of eggs and hence could represent the physiological agent that triggers egg activation and embryo development at fertilization (24).
We demonstrated that the ability of soluble mammalian sperm extracts to cause Ca 2ϩ oscillations can be explained by the presence of a sperm-specific PLC activity (25,26). This activity was shown to be due to a novel mammalian PI-PLC, phospholipase C (PLC), that we isolated from a spermatid cDNA library (27). Microinjection of cRNA encoding the mouse (27), human, and cynomolgus monkey PLC (28) into mouse eggs triggered Ca 2ϩ oscillations similar to those observed at fertilization. Furthermore, Ca 2ϩ oscillations are abolished when PLC is immunodepleted from native sperm extracts (27). A recent study showed that microinjection of recombinant PLC, synthesized using a baculovirus expression system, could also trigger Ca 2ϩ oscillations in mouse eggs (29). One unusual feature of PLC is that it is effective at causing Ca 2ϩ oscillations in eggs at very low concentrations (e.g. 10 fg/egg) (24,27,29). In contrast, other studies have shown that PLC isoforms of the ␤, ␥, or ␦ class are either ineffective (25) or at least much less effective than PLC at causing Ca 2ϩ release when microinjected into eggs (2,29,30). The specific reasons why sperm PLC is much more effective than other PLC isoforms at causing Ca 2ϩ oscillations in eggs are currently unknown.
The unique functional features of PLC may be attributable to its distinct domain structure. There are five subfamilies of PI-PLCs (␤, ␥, ␦, ⑀, and ) classified on the basis of their sequence homology (27,31). The regions of greatest sequence identity are the X and Y domains, which form the active site for enzymatic cleavage of PIP 2 in all PI-PLCs. In addition, all isoforms have in common a C2 domain, which binds to lipids such as phosphatidylserine. In studies with PI-PLC␦1, the C2 domain has been suggested to orientate the enzyme to the membrane containing PIP 2 , and this association is Ca 2ϩ -dependent (31). PI-PLCs also contain EF hands, but the role of these Ca 2ϩ binding domains in PLC isozymes is not clear. These domains usually bind one Ca 2ϩ ion each, although variants that do not bind Ca 2ϩ ions have been identified (31,32). PLC is most closely related to PLC␦ but is distinct in that it lacks a PH domain (Fig. 1A), a common feature present in most PI-PLCs (27). The PH domain characteristically targets the PLC to phosphoinositides or other regulatory proteins such as G-protein ␤␥ subunits in the plasma membrane (31)(32)(33).
The aim of this study is to examine the importance of the EF hand and C2 domains with respect to PLC enzymatic activity in vitro and their involvement in initiating Ca 2ϩ oscillations in eggs. A series of recombinant PLC domain-deletion constructs were prepared, together with the well characterized PLC␦1 (34,35) for comparison, and were tested by using an in vitro PLC activity assay (hydrolysis of PIP 2 ) and also an in situ assay by microinjection into eggs of the corresponding cRNA constructs (Ca 2ϩ oscillations). Our studies show that although the XY catalytic domain is minimally sufficient for in vitro enzymatic hydrolysis of PIP 2 , all of the disparate domains, i.e. EF hands and C2 domains, are required for PLC to initiate the trademark Ca 2ϩ oscillations observed in mammalian eggs upon fertilization by sperm.
The rat PLC␦1 clone (GenBank TM accession number M20637) was kindly provided by M. Katan (Cancer Research UK, Centre for Cell and Molecular Biology, London, UK). Appropriate primers were designed to incorporate a 5Ј-SalI site and a 3Ј-NotI site at the ends of PLC␦1. PLC␦1 was amplified by PCR and cloned into pGEX-5X2 by using these restriction sites. The following primers were used for PLC␦1: 5Ј-CT-TCGTCGACCATGGACTCGGGTAGGGAC-3Ј (forward) and 5Ј-CACC-GCGGCCGCTTAGTCCTGGATGGAGATCTTC-3Ј (reverse); for ⌬PH␦1 the forward primer was 5Ј-TTCGTCGACTGGGCTCCATGGACCAGC-GGCAGAAGC-3Ј(forward) and reverse primer was the same as for full-length PLC␦1. For ⌬PH-PLC␦1 the primers incorporated a 5Ј-EcoRI site, TTCAGAATTCCCATGGACTCGGGTAGGGACTT-3Ј (forward), and a 3Ј-SalI site, 5Ј-TGTCGACCAGCCTTTCGCAAGCAGGAGTGA-AT-3Ј (reverse). Each of the above expression vector constructs was confirmed by dideoxynucleotide sequencing (Prism Big Dye kit; ABI Prism® 3100 Genetic Analyzer, Applied Biosystems, Warrington, UK).
For quantitation of expression of microinjected cRNAs in mouse oocytes, PLC␦1, PLC, and various deletion constructs were tagged via their C termini with firefly luciferase. PCR amplification of the PLCs was performed with the original templates, described above, using Pfu polymerase. PLC constructs were amplified using a 5Ј primer incorporating an EcoRI site and a 3Ј primer that ablated the stop codon and created a NotI site. PLC␦1 was amplified using a 5Ј primer containing a SalI site and a 3Ј primer that removed the stop codon and incorporated a NotI site. PCR products were cloned into pCR3 (Invitrogen) via the appropriate restriction sites. The luciferase open reading frame was amplified from pGL2 (Promega) with primers incorporating NotI sites, and the product was cloned into the NotI site of the pCR3-PLC vector. Each construct was confirmed by dideoxynucleotide sequencing, as described above. cRNA was generated as described below, using the mMessage mMachine T7 kit (Ambion) and polyadenylated using the poly(A) tailing kit (Ambion), as per the manufacturer's instructions.
Protein Expression and Purification-For bacterial expression of GST fusion proteins, Escherichia coli (BL21 or Rosetta cells, Novagen) was transformed with the appropriate pGEX-5X-2 construct and cultured at 30°C until the A 600 reached 0.5, and then protein expression was induced for 4 h at 25°C with 0.5 mM isopropyl ␤-D-thiogalactopyranoside (Promega). Cells were harvested by centrifugation at 6000 ϫ g for 10 min, resuspended in phosphate-buffered saline (PBS: 137 mM NaCl, 2.7 mM KCl, 4.3 mM Na 2 HPO 4 ⅐7H 2 O, 1.4 mM KH 2 PO 4 , pH 7.4) containing 20% glycerol, 2 mM dithiothreitol, and protease inhibitor mixture (Roche Applied Science), and then sonicated three times for 20 s on ice. After 15 min of centrifugation at 15,000 ϫ g, 4°C, soluble GST fusion proteins were purified by affinity chromatography using glutathione-Sepharose TM 4B (GS4B) beads following standard procedures (Amersham Biosciences). Eluted proteins were dialyzed overnight (Pierce; SnakeSkin 10,000 molecular weight cut-off) at 4°C against 4 liters of PBS, then concentrated with centrifugal concentrators, and stored at Ϫ80°C in buffer containing 40% glycerol, 2 mM dithiothreitol, and EDTA-free protease inhibitors (Roche Applied Science).
Assay of PLC Activity-PIP 2 hydrolytic activity of recombinant PLC␦1, PLC, and their deletion constructs was assayed as described previously (36), with some modifications. The final volume of the assay mixture was 50 l containing 100 mM NaCl, 0.4% sodium cholate (w/v), 2 mM CaCl 2 , 4 mM EGTA, 20 g of bovine serum albumin, 5 mM 2-mercaptoethanol, and 20 mM Tris-HCl buffer, pH 6.8. The final concentration of PIP 2 in the reaction mixture was 220 M, containing 0.05 Ci of [ 3 H]PIP 2 . The assay conditions were optimized for linearity, requiring a 10-min incubation of 20 pmol of PLC protein sample at 25°C. Reactions were stopped by addition of 0.25 ml of chloroform/ methanol/concentrated HCl (100:100:0.6 v/v) followed by 0.075 ml of concentrated HCl. The mixture was vortexed and centrifuged at 2000 ϫ g for 2 min, and then 0.2 ml of the upper aqueous phase was removed and added to 10 ml of Optiphase "Hisafe 3" scintillation mixture (Wallac), and the radioactivity was determined by liquid scintillation spectrofluorimetry (Packard Tri-Carb 2100TR). In assays to determine dependence on PIP 2 concentration, 0.05 Ci of [ 3 H]PIP 2 was mixed with cold PIP 2 to give the appropriate final concentration. In assays examining the Ca 2ϩ sensitivity, Ca 2ϩ buffers were prepared by EGTA/CaCl 2 admixture, as described previously (37). In the assays to monitor pH dependence of the PLCs, Tris-HCl buffers were prepared over the range of pH 5.2-8. 6.
cRNA Synthesis and in Vitro Translation-Full-length mouse PLC and the domain-deletion constructs used for RNA synthesis were am-plified by PCR from the original clone cDNA using appropriate primers and Pfu polymerase, as described above. PCR products were cloned into pCR-Blunt-TOPO (Invitrogen), sequence-verified, and then subcloned into the pTarget vector (Promega). cRNAs encoding the wild type PLC and deletion constructs were produced from the linearized pTarget constructs using the Ribomax RNA synthesis kit (Promega) in the presence of 3 mM m 7 G(5Јppp(5Ј)G (37°C, 2 h). Synthesized cRNA products were analyzed by agarose gel electrophoresis, isopropyl alcoholprecipitated, and resuspended in diethyl pyrocarbonate-treated water containing 4 units of RNasin l Ϫ1 (Promega). Integrity of complementary RNAs were assayed by in vitro expression of proteins (Reticulocyte Lysate System, Promega) in the presence of [ 35 S]methionine (Amersham Biosciences) and analyzed by 8% SDS-PAGE and autoradiography, as described previously (27).
Preparation and Handling of Gametes-Experiments were carried out with mouse eggs in Hepes-buffered saline (H-KSOM) as described previously (38). All compounds were from Sigma unless stated otherwise. Female mice were superovulated by injection of human chorionic gonadotrophin (Intervet). Eggs were collected 13.5-14.5 h later (5) and maintained in 100-l droplets of H-KSOM under mineral oil at 37°C. Microinjection of the eggs was carried out 14.5-15.5 h after human chorionic gonadotrophin injection (27).
Microinjection and Measurement of Intracellular Ca 2ϩ and Luciferase Expression-Mouse eggs were washed in H-KSOM and microinjected as described previously (27) with cRNA diluted in injection buffer (120 mM KCl, 20 mM Hepes, pH 7.4). The volume injected was estimated from the diameter of cytoplasmic displacement caused by the bolus injection. All injections were 3-5% of the egg volume. In experiments with untagged PLC, Ca 2ϩ changes were monitored with a CCD-based imaging system using a Zeiss Axiovert 100 microscope, with illumination from a monochromator (Photonics) controlled by MetaFluor version 4.0 (Universal Imaging Corp.). Eggs were loaded for 10 min with 4 M Fura Red-AM (Molecular Probes) and dissolved in Me 2 SO ϩ 5% (w/v) pluronic acid, and the loading medium was supplemented with sulfinpyrazone, which helps prevent compartmentalization and extrusion of the dye (5).
For experiments with luciferase-tagged PLC, eggs were microinjected with the appropriate cRNA mixed with an equal volume of 1 mM Oregon Green BAPTA dextran (Molecular Probes) in KCl Hepes buffer. Eggs were then maintained in H-KSOM with 100 M luciferin and imaged on a Nikon TE2000 or Zeiss Axiovert 100 microscope equipped with a cooled intensified CCD camera (Photek Ltd., UK). Ca 2ϩ was monitored in these eggs for 4 h after injection by measuring the Oregon Green BAPTA dextran fluorescence with low level excitation light from a halogen lamp. At the end of Ca 2ϩ measurements, the same set of eggs was then monitored for luminescence by integrating light emission (in the absence of fluorescence excitation) for 20 min using the same intensified CCD camera. The fluorescence signals were typically 10 -100 times greater than the luminescence signals. Ca 2ϩ measurements for an egg were considered valid only if the same egg was also luminescent. Groups of eggs verified as being luminous were then collected and placed in a test tube containing PBS with 1 mM MgATP ϩ 100 M luciferin, which was held in a custom-made luminometer equipped with a cooled S20 photomultiplier tube (Electron Tubes Ltd., UK). The eggs were then lysed with 0.5% Triton X-100, and the steady state light was compared with that emitted from calibrated amounts of recombinant firefly luciferase (Sigma). The amount of luciferase activity measured for each group of eggs was then divided by the number of luminous eggs to obtain the mean value for protein expression of each type of luc-PLC.
SDS-PAGE and Western Blotting-Recombinant proteins were separated by SDS-PAGE as described previously (27). Separated proteins were transferred onto polyvinylidene difluoride membrane, incubated overnight at 4°C in Tris-buffered saline, 0.1% Tween 20 (TBS-T) containing 5% nonfat milk powder, and probed with anti-GST polyclonal antibody (1:5000 dilution). Detection of horseradish peroxidase-coupled secondary antibody was achieved using Super Signal West Dura (Pierce) and a Bio-Rad ChemiDoc gel documentation system for image capture.

Expression and Enzymatic Characterization of Recombinant
PLC-PLC␦1 and PLC were expressed in E. coli BL21 cells and purified by GST affinity chromatography, as described under "Experimental Procedures." Empirically determined, optimal protein production required maintaining bacterial cultures at 37°C until A 600 of 0.5-0.6, followed by induction of protein expression upon addition of 0.5 mM isopropyl 1-thio-␤-D-galactopyranoside and 4 h of vigorous shaking at 25°C. Fig.  1A schematically depicts the distinct domains of PLC␦1 and PLC and illustrates their close similarity because of common structural features, specifically the two EF hands, catalytic X and Y domains, and C2 domain, with the notable exception of an N-terminal PH domain absent from PLC (27). Fig. 1B shows the GST affinity-purified recombinant proteins analyzed by SDS-PAGE followed by Coomassie Brilliant Blue staining and immunoblot detection using an anti-GST antibody. The predicted molecular mass, including the GST tag (26 kDa), for GST-PLC␦1 and GST-PLC was 111 and 100 kDa, respectively. The corresponding proteins with appropriate molecular mass were observed as the top band in both Coomassie Brilliant Blue staining and on the immunoblot (Fig. 1B). The lowest band of 26 kDa is consistent with cleaved GST, which along with the much fainter intermediate molecular mass bands detected by the anti-GST antibody are probably the result of protein degradation occurring through the bacterial expression and purification processes. Hydrolysis of [ 3 H]PIP 2 to [ 3 H]IP 3 to monitor the enzymatic activities of recombinant PLC␦1 and PLC (36) was optimized for PLC activity by varying a series of parameters, including incubation time, reaction temperature, and protein concentrations. Linearity of [ 3 H]PIP 2 cleavage was obtained with 20 pmol of recombinant protein incubated with 220 M [ 3 H]PIP 2 for 10 min at 25°C (data not shown). Fig. 2A shows the effect of total PIP 2 concentration on PLC␦1 and PLC enzyme activity. For both PLC␦1 and PLC, the maximum hydrolytic enzyme activity was obtained at 660 M PIP 2 , with specific activity values of 1884 and 770 nmol/min/mg measured, respectively. The Michaelis-Menten constant, K m , calculated by a Lineweaver-Burk reciprocal plot for both recombinant proteins was very similar, with PLC having a K m value of 87 M in comparison to 75 M for PLC␦1 (Fig. 2B). Although PLC␦1 and PLC had common enzymatic properties with regard to PIP 2 , the Ca 2ϩ dependence of their activities was markedly different (Fig. 3A). PLC was activated between 0.01 and 0.1 M Ca 2ϩ , whereas the threshold for PLC␦1 was 0.1 M, with maximum activity at about 100 M. The EC 50 was 82 nM (Hill constant, 4.3) for PLC and 6 M (Hill constant, 1.5) for PLC␦1 (calculated from Fig. 3A). PLC showed maximal activity over a broad pH range, varying between 5.2 and 6.0, in contrast with PLC␦1, which displayed an optimum pH at 6.0 (Fig. 3B).
Enzymatic Analysis of Domain-truncated PLC in Vitro-To examine the role of the distinct structural domains on enzymatic activity, four domain-deletion constructs of PLC were expressed and purified as GST fusion proteins. Fig. 4A schematically illustrates the full-length PLC and the various domain-truncated PLC versions that have one or both EF hands removed (⌬EF1 and ⌬EF1,2, respectively), the C2 domain deleted (⌬C2) or all the above domains absent (XY), and their specific amino acid sequence coordinates. Following expression in E. coli and purification by affinity chromatography, samples of each protein were analyzed by SDS-PAGE followed by Coomassie Brilliant Blue staining and immunoblotting using anti-GST polyclonal antibody. Fig. 4B shows that the major protein band, with mobility corresponding to the predicted molecular mass for each construct, was present for all four of the domaintruncated proteins (left panel), and these major bands were also recognized by the corresponding anti-GST immunoblot (right panel), confirming the appropriate expression of all domain- truncated PLC proteins. Enzyme activity assays performed for each of the recombinant proteins, determined by using the standard [ 3 H]PIP 2 hydrolysis assay, showed that every domain-deletion construct retained some of the enzymatic activity present with the full-length PLC. The histogram of Fig. 4C plots the enzyme specific activity values obtained for each protein and reveals that the PLC proteins lacking either one or both EF hand domains or the C2 domain retained about 70% of the activity of the full-length PLC protein. Even the XY catalytic domain alone exhibited well over half of the activity of the full-length PLC. These data suggest that the PLC catalytic site alone, comprising the X and Y domains, is capable of binding and hydrolyzing PIP 2 and that the C2 and EF hand domains are not essential for enzymatic activity in vitro.
Activity Analysis of PLC Domain Deletions Expressed in Mouse Eggs-Because the recombinant deletion constructs of PLC displayed robust enzymatic activity in vitro, further experiments were conducted to determine whether this is matched by their ability to trigger Ca 2ϩ oscillations when the corresponding cRNA is injected into mouse eggs (Fig. 5A). Injection of 0.2 mg/ml cRNA encoding the full-length PLC (WT in Fig. 5) was used as the positive control, and this caused a robust series of Ca 2ϩ oscillations in all eggs, as reported previously (27,28). The Ca 2ϩ oscillations started about 20 min after injection, and spiking behavior persisted at an interval of about 3 min for over 2 h. In contrast, when the cRNAs for ⌬EF1-PLC, ⌬EF1,2-PLC, or ⌬C2-PLC were microinjected into eggs under the same conditions, there were no Ca 2ϩ oscillations observed in any of the injected eggs (Fig. 5B). All the cRNAs used in these experiments were synthesized in parallel with batches of control (wild type PLC) cRNA, and as demonstrated previously (27,28), the cRNA used for these microinjection experiments was able to generate proteins of the correct size when expressed in an in vitro reticulocyte lysate expression system (data not shown). Consequently, these data suggest that the absence of either of the EF hand domains or the C2 domain of PLC prevents it from being able to trigger Ca 2ϩ oscillations in intact eggs.
Quantitation of PLC Expression in Mouse Eggs-To verify that the domain-deleted versions of PLC were faithfully expressed as proteins in cRNA-microinjected eggs, we generated luciferase-tagged versions of these constructs to enable quantitation of relative protein expression. Fig. 6A shows that eggs injected with cRNA encoding the wild type PLC-luciferase (PLC-luc) fusion construct caused a series of Ca 2ϩ oscillations in eggs. This indicates that a fusion tag at the C terminus of PLC can also retain the ability to generate Ca 2ϩ oscillations, as has been previously shown for N-terminal tags (27). At the end of 4 h of monitoring the changes in Ca 2ϩ , we measured the light emitted from the same set of eggs (in the absence of fluorescence excitation), and we found that they were luminescent (Fig. 6B). Every mouse egg injected with PLC-luc cRNA that showed clear expression of luciferase activity after 4 h (n ϭ 19) had also exhibited prior robust Ca 2ϩ oscillations. When eggs were injected with ⌬EF1-PLC-luc cRNA, however, none of the eggs showed any Ca 2ϩ increase (n ϭ 26), although they were visibly expressing luciferase activity, as confirmed by the intense luminescence detected at the end of the experiment (Fig. 6, C and D). Similarly, upon injection of ⌬C2-PLC-luc cRNA, all 25 eggs failed to show any Ca 2ϩ oscillations, although they exhibited strong luciferase luminescence (Fig. 6, E  and F). The exact level of luminescence in cells can depend upon the amount of luciferase protein, the intracellular pH, and the ATP concentration (39). Consequently, we quantified the relative expression of the luciferase fusion protein by lysing the mouse eggs in a buffer with a fixed concentration of ATP. Upon lysing the eggs in a luminometer, we determined that at the end of the experiment (4 h), a mean value of 0.19 pg of wild type PLC-luc protein was expressed per egg (n ϭ 19). With ⌬EF1-PLC-luc, a mean of 0.98 pg of protein was expressed per egg (n ϭ 26), and with ⌬C2-PLC-luc a mean of 2.7 pg of protein was expressed per egg (n ϭ 25). These data show that the two PLC-luc domain-deletion constructs did not cause any Ca 2ϩ oscillations but were demonstrably being expressed in the eggs that were injected with cRNA. The ⌬EF1-PLC-luc and ⌬C2-PLC-luc were expressed at levels that were 5-and 14-fold that of PLC-luc, respectively. This high level of expression of domain-deletion constructs more than compensates for the reduced specific activity of these proteins (ϳ70%) relative to full-length PLC (Fig. 4). Because the threshold for PLC to cause a Ca 2ϩ oscillation is around 50 fg (27), the two domaindeletion constructs were expressed at levels that are 20 -50 times the amount required to cause Ca 2ϩ oscillations with PLC.
Effect of [Ca 2ϩ ] on the Activity of PLC Domain Deletions-The activity assays described above suggest that the X and Y catalytic domains alone are sufficient for in vitro enzymatic activity (Fig.  4), although all domains appear to be essential for Ca 2ϩ oscillations activity in intact eggs (Figs. 5 and 6). To examine the role of selected domains on the Ca 2ϩ sensitivity of PLC activity, we tested the ability of the domain-deletion constructs of PLC to hydrolyze [ 3 H]PIP 2 at different Ca 2ϩ concentrations ranging from 0.1 mM to 0.1 nM (Fig. 7). Fig. 7A illustrates the Ca 2ϩ dependence of specific PIP 2 hydrolytic activity for the full-length PLC and each of the truncated proteins, and these are shown normalized to the maximum specific activity in Fig. 7B. Table I summarizes the EC 50 and Hill coefficients of PLC␦1, PLC, and domain-deletion constructs. Deletion of EF1 increased the EC 50 by 10-fold and reduced the Hill coefficient from 4.135 to 2.204. Deletion of both EF hands led to a dramatic increase of the EC 50 of PLC from 82 nM to 30 M and a decrease of the Hill coefficient from 4.315 to 0.642. Deletion of the C2 domain did not change the EC 50 of PLC but reduced the Hill coefficient from 4.315 to 1.139. Finally, deletion of both EF hands and C2 domain (PLC-XY) drastically changed the EC 50 and Hill coefficient (62 and 0.327 M, respectively). These results suggest the EF hands may play a direct modulatory role in the Ca 2ϩ regulation of PLC activity. DISCUSSION PLC was identified as a sperm-specific PLC that is highly effective in causing Ca 2ϩ oscillations and activation in mouse eggs (27). It appears to be responsible for the previously described PLC activity and Ca 2ϩ -releasing activity present in sperm extracts (27). PLC has also been identified as the protein factor that is responsible for causing Ca 2ϩ oscillations after injection of whole sperm into mouse eggs (24). This "sperm factor" is proposed to enter the egg upon fusion with the sperm at fertilization and to be responsible for activating embryo development in mammals (27,28). Despite its significance in cell signaling at fertilization, the regulation of PLC activity and the mechanism by which PLC locates and targets the substrate in eggs are currently unknown (40). The aims of this study were to compare the enzymatic properties of the bacterially expressed, full-length PLC and PLC␦1 in vitro. We then determined which structural domains within the PLC sequence are essential for its enzymatic function and whether this correlates with its unique ability to cause sustained Ca 2ϩ oscillations in mouse eggs.
The dependence of PLC and PLC␦1 enzyme activity on PIP 2 concentration indicated that the K m values for these closely related isoforms were very similar (Fig. 2) and in reasonable agreement with the K m value obtained for recombinant PLC␦3 (142 M) in another study (41). This suggests that the enzymes have similar affinity for their substrate PIP 2 , which is consistent with the presence of highly conserved X and Y active site domains found throughout the PI-PLC family.
To determine the role of different parts of PLC in PIP 2 hydrolyzing activity, we generated a series of truncated proteins with deletions of selected structural domains (Fig. 4). Our data showed that the recombinant X ϩ Y catalytic domains alone are able to retain considerable enzyme activity (ϳ70%) compared with that of the full-length PLC. Furthermore, the proteins with deletions of either the EF hands or the C2 domains also showed a similar PLC activity in vitro compared with that of full-length PLC. Thus, the absence of both EF  hands and the C2 domain is unable to ablate PLC enzyme activity when the X ϩ Y catalytic domains are presented with PIP 2 in micellar form. These data are consistent with experiments on other PLC isoforms (34). In order to assess the relevance of PLC structural domains in triggering Ca 2ϩ oscillations in intact cells, various truncated forms of PLC were prepared, and individual mouse eggs were microinjected with cRNA corresponding to each domain-deletion construct (Fig. 5). Microinjection of PLC cRNA into eggs is used in preference to recombinant PLC protein because it avoids the problems of maintaining protein activity and solubility and to negate potential cross-contamination from bacterial proteins during the purification process. In previous work with recombinant PLCs, we have found that "purified" preparations can be also contaminated with IP 3 , and this can give the misleading impression that microinjection of a purified PLC directly triggers Ca 2ϩ oscillations in intact eggs (25). In contrast, the use of microinjected cRNA ensures the sample delivered into the egg is free from such contamination; moreover, it is a robust and reliable method because we have demonstrated previously that the mouse egg is capable of efficiently translating the cRNA encoding the active sperm PLC protein (27,28,42,43). By using cRNA injection, we found that expression of the PLC domain-deletion constructs in mouse eggs did not lead to the generation of any Ca 2ϩ oscillations (Fig. 5). Because the in vitro assays demonstrated that these shortened versions possess PLC enzyme activity (Fig. 4), we can infer that the absence of either of the EF hands or the C2 domain of PLC causes it to be unable to hydrolyze PIP 2 under the conditions found in intact cells.
One issue that remains to be addressed with the use of microinjected cRNA samples into eggs is the level of expression achieved with each of the various PLC truncations. For example, if the full-length PLC is expressed with much greater efficiency than the PLC domain deletions, then microinjection of equivalent amounts of cRNA would not be directly comparable. Thus, to examine whether expression of specific domaindeletion constructs was selectively impaired relative to fulllength PLC, a series of luciferase-tagged fusion constructs for the domain deletions was microinjected into eggs. Although at levels of luciferase expression for the two domain deletions that were significantly higher than that of the full-length PLC-luc (Fig. 6), no Ca 2ϩ oscillations were produced by any of the domain-deleted PLC-luc constructs. It was notable that the Ca 2ϩ oscillations observed upon injection of the full-length PLC-luc cRNA occurred about 1 h after injection, and at the end of the experiment (4 h) the level of expressed PLC-luc protein was determined to be 190 fg/egg. If we assume a linear increase in PLC-luc protein during the 4 h of recording, then we can estimate that the amount of PLC-luc required to initiate Ca 2ϩ oscillations is around 50 fg. This value is similar to our previous estimate that 20 -50 fg of PLC is required to initiate Ca 2ϩ oscillations in eggs (27), as well as the estimate that 10 -40 fg of venusGFP-PLC is required to initiate Ca 2ϩ release (44). These data therefore suggest that the PLC-luc has a similar efficiency is generating IP 3 in eggs to other N-terminally tagged PLC fusion proteins.
One potential reason for the striking difference between the activity observed in vitro and in intact eggs could be the way in which Ca 2ϩ regulates PLCs. Most mammalian PLCs show some stimulation of activity with increasing Ca 2ϩ concentration, but PLCs are often only stimulated by Ca 2ϩ concentrations that are much higher than those found in resting cells. This is even true for PLC␦1, which has been reported previously to be one of the most Ca 2ϩ -sensitive of the PLC isozymes, with marked stimulation by Ca 2ϩ concentrations in the micro-molar range (45). Our assays of PLC␦1 are consistent with this earlier work, because we found the EC 50 value for Ca 2ϩ stimulation of PLC␦1 to be about 6 M (Fig. 3). In contrast to PLC␦1, PLC is about 100 times more sensitive to Ca 2ϩ with an EC 50 of 50 -80 nM, which is well within the range of reported resting Ca 2ϩ concentration in eggs (3,4). The difference between these isoforms means that PLC is not only likely to show significant activity at resting Ca 2ϩ levels, but it will be maximally active at 1 M Ca 2ϩ , although PLC␦1 will not be fully activated until Ca 2ϩ reaches 30 M. As well as the high Ca 2ϩ sensitivity, we found that the dependence of PLC activity on Ca 2ϩ had a Hill coefficient of 4.3, suggesting the binding of 4 Ca 2ϩ molecules/ protein. This is greater than the previous value of 0.9 observed by Kouchi et al. (29) in their study of PLC expressed in insect cells. The reason for the difference is unclear, but it is not likely to be due to systematic differences in the assay, because the calculated EC 50 values and the Hill coefficients for the Ca 2ϩ dependence of PLC␦1 were very similar between the present data and those of Kouchi et al. (1.5 and 1.7, respectively). The difference in Ca 2ϩ sensitivity and Hill coefficient between PLC␦1 and PLC is of interest because it could be one source of explanation for why PLC is so effective at causing Ca 2ϩ oscillations in mouse eggs, and yet PLC␦1 is reported to be either ineffective (25) or at least much less effective (29). This idea is supported by our current observations of the behavior of the PLC domain-deletion constructs in eggs.
It is clear from our data that PLC is not effective in stimulating Ca 2ϩ oscillations when it lacks one or both of its EF hand domains (Figs. 5 and 6). This result is consistent with previous observations with a short form of PLC that lacks the first 110 amino acids at the N terminus (29), which corresponds to removal of the first EF hand and part of the second (Fig. 4A). The basic ability to hydrolyze PIP 2 in vitro would be preserved for these deletion constructs, so the lack of Ca 2ϩ oscillationinducing activity in eggs injected with EF hand domain deletions of PLC may be explained by their differential response to Ca 2ϩ regulation. EF hands appear to play a vital role in the Ca 2ϩ sensitivity of PLC activity, because deletion of both EF hands dramatically changed the EC 50 of PLC from 82 nM to 30 M. The N-terminal truncation of both EF hands would ablate the ability of this domain deletion to generate IP 3 in an intact cell with a Ca 2ϩ level of around 100 nM. Even deletion of the first EF hand domain raised the EC 50 for Ca 2ϩ to Ͼ700 nM, which is well above the resting Ca 2ϩ level in an egg. The Hill coefficients were also decreased by deletion of one or both of the EF hand domains. The Hill coefficient describes the minimum number of interacting active sites required for enzyme function, suggesting that upon removal of the EF hands the minimum number of sites is reduced from ϳ4 to ϳ1. This in turn suggests that Ca 2ϩ binding to the EF hands is important for the interaction of the X-Y domain with PIP 2 substrate and thus for PLC enzyme activity.
It is of interest that the loss of the C2 domain from PLC also leads to the inability to cause Ca 2ϩ oscillations in intact eggs, although the EC 50 value for Ca 2ϩ stimulation was unchanged. This result suggests two potential explanations. One possibility is linked to the significant change in the Hill coefficient, as the C2 domain removal caused a marked reduction in the Hill coefficient for Ca 2ϩ stimulation from ϳ4 to 1. This loss of cooperativity in Ca 2ϩ stimulation could be important for generating Ca 2ϩ oscillations (see below). The other possibility is that the C2 domain plays an important role in targeting PLC to the correct subcellular source of PIP 2 in eggs. C2 domains display functional diversity and can be involved in binding to lipids, or proteins, in a way that can be Ca 2ϩ -dependent (46). The precise physiological mechanism by which IP 3 produc-tion leads to the exquisite pattern of Ca 2ϩ oscillations at fertilization in mammalian eggs is unknown. By analogy with mechanisms proposed in somatic cells, the positive and delayed negative feedback effect of Ca 2ϩ on the IP 3 receptor may be an important part of the mechanism for generating regular Ca 2ϩ oscillations (6). The observation that Ca 2ϩ oscillations similar, but not identical, to those occurring at fertilization can be stimulated by the sustained introduction of IP 3 alone into eggs supports this idea (12,13). However, because of the high Ca 2ϩ dependence of PLC activity, it is plausible that some Ca 2ϩinduced IP 3 formation occurs in eggs after the introduction of PLC. Such a positive feedback loop of Ca 2ϩ release and IP 3 production has been suggested to play an important role in generating Ca 2ϩ oscillations (47,48). This feedback phenomenon could provide an explanation for the enhanced sensitivity of the egg to Ca 2ϩ -induced Ca 2ϩ release, which occurs after fertilization, or sperm factor injection (12,16). The nonlinear dependence of PLC-induced IP 3 production on Ca 2ϩ levels could also help generate the rapid rising phase of Ca 2ϩ release that is a specific feature of fertilized eggs (3,45). Consequently, it is possible that a combination of Ca 2ϩ -stimulated PLC activity and the properties of the IP 3 receptor combine to generate the Ca 2ϩ oscillations at fertilization in mammals.