Calcium oscillations and mammalian egg activation

Authors

  • Christopher Malcuit,

    1. Department of Veterinary and Animal Sciences, University of Massachusetts, Amherst, Massachusetts
    Current affiliation:
    1. Cyagra, Inc., 200 Westboro Rd., N. Grafton, MA 01536.
    Search for more papers by this author
  • Manabu Kurokawa,

    1. Department of Veterinary and Animal Sciences, University of Massachusetts, Amherst, Massachusetts
    Current affiliation:
    1. Department of Pharmacology and Cancer Biology, Duke University Medical Center, Durham, NC 27710.
    Search for more papers by this author
  • Rafael A. Fissore

    Corresponding author
    1. Department of Veterinary and Animal Sciences, University of Massachusetts, Amherst, Massachusetts
    • Department of Veterinary and Animal Sciences, Paige Laboratory, University of Massachusetts, Amherst, MA 01003.
    Search for more papers by this author

Abstract

Fertilization in all species studied to date induces an increase in the intracellular concentration of free calcium ions ([Ca2+]i) within the egg. In mammals, this [Ca2+]i signal is delivered in the form of long-lasting [Ca2+]i oscillations that begin shortly after fusion of the gametes and persist beyond the time of completion of meiosis. While not fully elucidated, recent evidence supports the notion that the sperm delivers into the ooplasm a trigger of oscillations, the so-called sperm factor (SF). The recent discovery that mammalian sperm harbor a specific phospholipase C (PLC), PLCζ has consolidated this view. The fertilizing sperm, and presumably PLCζ promote Ca2+ release in eggs via the production of inositol 1,4,5-trisphosphate (IP3), which binds and gates its receptor, the type-1 IP3 receptor, located on the endoplasmic reticulum, the Ca2+ store of the cell. Repetitive Ca2+ release in this manner results in a positive cumulative effect on downstream signaling molecules that are responsible for the completion of all the events comprising egg activation. This review will discuss recent advances in our understanding of how [Ca2+]i oscillations are initiated and regulated in mammals, highlight areas of discrepancies, and emphasize the need to better characterize the downstream molecular cascades that are dependent on [Ca2+]i oscillations and that may impact embryo development. © 2005 Wiley-Liss, Inc.

In mammals, ovulation occurs once fully grown oocytes have progressed to an arrested state in the metaphase stage of the second meiotic division (MII). The process leading up to this arrest, termed oocyte maturation, renders the oocyte fully competent to respond to the forthcoming signaling events that take place during fertilization. Part of this preparation period entails the recruitment and translation of stored maternal mRNAs that encode for effectors of the pathways involved in cell cycle progression and calcium (Ca2+) release, and a structural reorganization of key calcium-sensitive elements and organelles (Fujiwara et al., 1993; Mehlmann et al., 1995, 1996; Shirashi et al., 1995; Machaca, 2004). The culmination of these events results in a perfectly harmonized system that is poised to respond in the most sensitive and efficient manner to the incoming sperm.

Exit from the MII arrest and meiotic resumption, referred to as “egg activation,” is made possible by the fertilizing sperm, which evokes in the egg an increase in the concentration of intracellular free calcium ions ([Ca2+]i). This rise in [Ca2+]i is necessary and sufficient for the completion of all the events of egg activation (Schultz and Kopf, 1995), including exocytosis of the cortical granule material to block polyspermy (Cran et al., 1988), resumption of the meiotic cell cycle through Ca2+-dependent destruction of cyclin B (Hyslop et al., 2004), pronuclear formation, recruitment of maternal mRNAs, and initiation of mitotic divisions that unveil the complete developmental program (Schultz and Kopf, 1995; Ducibella et al., 2002).

Although the precise mechanism by which the sperm initiates the Ca2+ release that is responsible for triggering embryonic development has not yet been made fully clear, it has been established in all species studied to date that it involves the activation of the phosphoinositide (PI) pathway (Turner et al., 1984; Stith et al., 1994). Activation of the PI pathway in eggs (Fig. 1) results in the production of inositol 1,4,5-trisphosphate (IP3) and 1,2-diacylglycerol (DAG) via the hydrolysis of phosphatidyl 4,5-bisphosphate (PIP2) by a phosphoinositide-specific phospholipase C (PLC) isoform (Parrington et al., 1998; Rongish et al., 1999; Rice et al., 2000). Both products of the PI pathway are involved in shaping [Ca2+]i responses. Increase in the intracellular concentrations of IP3 is responsible for mediating Ca2+ release by binding and gating its receptor, the type I IP3 receptor (IP3R-1, Miyazaki et al., 1993), a tetrameric ligand-gated Ca2+ channel located on the endoplasmic reticulum (ER) membrane, the main Ca2+ store of the cell (Koch, 1990; Berridge, 2002). Production of DAG, either directly (Bazzi and Nelsestuen, 1989), or indirectly via activation of protein kinase C (PKC), may be involved in the regulation of Ca2+ influx (Halet et al., 2004). Notably, in spite of the universal requirement for a [Ca2+]i increase in egg activation, the Ca2+ release that accompanies fertilization varies widely among species. For instance, in some lower vertebrates and marine animals this Ca2+ release takes the form of a single, rather long (10 min) transient, whereas mammalian eggs show persistent and repetitive changes in [Ca2+]i (reviewed in Stricker, 1999) beginning shortly after fertilization (Lawrence et al., 1997) and lasting for several hours. Interestingly, in mouse eggs, sperm-initiated [Ca2+]i responses cease in concurrence with pronuclear formation (Jones et al., 1995; Marangos et al., 2003; Jellerette et al., 2004), while in other mammalian species oscillations persist throughout the first cell cycle (Fissore et al., 1992; Nakada et al., 1995). Therefore, the differences in [Ca2+]i responses among phyla and even within classes suggest, at the very least, evolutionary divergence of the mechanism(s) leading to activation of the PI pathway.

Figure 1.

Proposed mechanisms of [Ca2+]i oscillations during mammalian fertilization. Upon gamete fusion, the fertilizing sperm delivers into the ooplasm a soluble substance, presently thought to be PLCζ capable of hydrolyzing PIP2 into the two signaling molecules, IP3 and DAG. IP3 is involved in intracellular Ca2+ release by binding and gating its receptor, the IP3R-1, located in the ER, the Ca2+ store of the cell, while DAG may play an integral role in signaling events at the plasma membrane. DAG together with the elevations in [Ca2+]i may further target and activate PKC to the plasma membrane, where PKC may regulate Ca2+ influx to refill the intracellular stores making possible the persistence of oscillations. The channels and mechanisms responsible for the influx of Ca2+ are unknown, although the transient receptor potential (TRP) channels and the store-operated Ca2+ entry mechanism have been both demonstrated in mammalian eggs.

Consistent with the fundamental role of Ca2+ in egg activation and the protracted nature of this response in mammals, the egg's Ca2+ releasing machinery undergoes significant reorganization and modification(s) prior to fertilization. For instance, during oocyte maturation, the spatial distribution of the ER and of the IP3R-1 changes from the perinuclear region of the germinal vesicle (GV) to distinct focal clusters at the cortex of the MII egg (Mehlmann et al., 1996; FitzHarris et al., 2003). Additionally, at this time, the overall mass of IP3R-1 is increased due to the recruitment and translation of IP3R-1 mRNA (Miyazaki et al., 1993; Shiraishi et al., 1995; Mehlmann et al., 1996; Xu et al., 2003). Lastly, while not well understood, direct modification of IP3R-1 or of other proteins involved in Ca2+ homeostasis such as Ca2+ pumps and/or plasma membrane Ca2+ channels by the prevailing kinase environment in the maturing oocyte may also take place during this prelude to fertilization. Collectively, these adaptations are thought to lead to a maximal (Fujiwara et al., 1993; Mehlmann and Kline, 1994) and spatially organized series of oscillations (Kline et al., 1999; Deguchi et al., 2000), the occurrence of which is pivotal for the initiation of normal development (Ozil, 1990; Vitullo and Ozil, 1992; Miyazaki et al., 1993; Ozil and Huneau, 2001; Ducibella et al., 2002; Ozil et al., 2005). This review will discuss recent advances in understanding how the fertilizing sperm initiates [Ca2+]i oscillations in mammalian eggs, the cellular and molecular regulation of Ca2+ release in eggs, and the requirement for the Ca2+ regime in egg activation.

THE NATURE OF SPERM-INDUCED Ca2+ RELEASE IN EGGS

There has been much debate and speculation as to the mechanism(s) that triggers activation of the PI pathway and [Ca2+]i oscillations during mammalian fertilization, and this has led to the proposal of several theories to explain this phenomenon. One early hypothesis, referred to as the “conduit hypothesis,” proposed that sperm fusion allows Ca2+ to passively enter the egg. However, the findings that the initial fertilization Ca2+ responses proceed unaltered in the absence of extracellular Ca2+ ([Ca2+]e) (Stricker, 1996), clearly demonstrated that [Ca2+]e is not necessary for the initiation of [Ca2+]i oscillations. The “receptor hypothesis” suggests that upon sperm-egg membrane contact, receptor-ligand interactions on the surface of the gametes relay intracellular signaling events that initiate Ca2+ release in the egg. One of the signaling cascades thought to be engaged by the interaction of gametes is that mediated by protein tyrosine kinases (PTKs). Specifically, the Src-family of PTKs (SFKs) may activate PLCγs (Carroll et al., 1997; Shearer et al., 1999; Giusti et al., 1999a,b, 2000; Sato et al., 2000, 2003; Tokmakov et al., 2002), thereby triggering Ca2+ release through the production of IP3. In accordance with this notion, PTK and PLCγ activity are upregulated shortly following fertilization in echinoderm and fish eggs (Ciapa and Epel, 1991; Rongish et al., 1999; O'Neill et al., 2004), and fertilization-triggered Ca2+ signaling in egg extracts was reconstituted by adding “activated” membrane raft fractions (Sato et al., 2003), which suggests the presence of a receptor-mediated SFK/PLCγ activation model in Xenopus fertilization. Furthermore, inhibition of PLCγ activation by a dominant negative approach using over expression of PLCγ SH2 domains prevented the sperm-induced [Ca2+]i rise in sea urchin (Carroll et al., 1997) and starfish eggs (Runft et al., 2004). Nonetheless, extensive pharmacological studies (Kurokawa et al., 2004) along with dominant negative approaches (Mehlmann et al., 1998; Mehlmann and Jaffe, 2005) and injection of recombinant PLCγ (Mehlmann et al., 2001) failed to show any involvement of this pathway in evoking [Ca2+]i responses in mammalian eggs. This is in spite the fact that PLCγ and SFK isoforms are expressed in mouse eggs (Dupont et al., 1996; Mehlmann and Jaffe, 2005), and that stimulation of its activity in these cells by exogenous expression of tr-kit, a sperm tyrosine kinase which activates Fyn, a SFK present in mouse and rat eggs (Talmor et al., 1998), was able to induce exit from meiosis and pronuclear formation (Sette et al., 2002). Therefore, while available evidence argues against a role for SFKs in the initiation of [Ca2+]i oscillations in mammalian fertilization, possible downstream effects of this signaling cascade on other events of eggs activation need further investigation (Talmor-Cohen et al., 2004). Lastly, extensive studies, which initially relied on the injection of activators and inhibitors of α subunits of Gq proteins, implicated the activation of PLCβ isoforms in the initiation of Ca2+ release in mammalian fertilization (Miyazaki, 1988; Fissore and Robl, 1994; Moore et al., 1994). However, the subsequent findings demonstrating that inhibition of Gαq subunits by injection of a function-blocking antibody was without effect on fertilization-induced [Ca2+]i oscillations (Williams et al., 1998), in conjunction with the apparent normal fertility in most strain of mice lacking one of the PLCβ isoforms (Choi et al., 2001), suggest that the contribution of this pathway to the initiation of oscillations in mammalian fertilization is negligible (Figs. 1 and 2).

Figure 2.

The effect of the Ca2+ regime on egg activation. A: Depiction of typical Ca2+ traces from fertilized mouse and cow eggs. Although the interval between [Ca2+]i oscillations differ significantly between these two mammalian species (15–20 min in the mouse vs. 30–40 min in the cow), the Ca2+ stimulus is transduced through a conserved mechanism to achieve egg activation and embryonic development. B: Within minutes of sperm-egg fusion, the egg displays a series of long-lasting oscillations in [Ca2+]i. In the mouse, [Ca2+]i oscillations are entrained with the cell cycle, and cease shortly before the formation of the pronuclei. A subsequent rise is observed at the time of PNBD and, on occasions, oscillations are observed during the first embryonic mitosis (mitosis I). Egg activation events exhibit different [Ca2+]i requirements, with the first several spikes being responsible for the exocytosis of cortical granules and block to polyspermy (I), whereas additional [Ca2+]i transients are needed to induce the resumption of meiosis including second meiotic polar body extrusion and to initiate the recruitment of maternal mRNAs (II). Further stimulation appears necessary to promote pronuclear formation and the initiation of embryonic mitosis (III). The relative Ca2+ requirement for the completion of each egg activation event is represented by horizontal bars (dark areas within the bar denotes ∼ the number of [Ca2+]i rises needed to initiate each event; shaded areas denote ∼ the number of [Ca2+]i rises required to complete each event; modified from Ducibella et al., 2002). Egg activation events in which a direct link to the Ca2+ activation stimulus needs further demonstration are marked with a question mark (?) inside the corresponding bar. Each [Ca2+]i rise (black trace) induces a concomitant increase in the activity of CaMKII (blue trace), which is thought to be one of the molecules that translate the Ca2+ stimulus into events of egg activation. The activity of CaMKII outside the first hour post-fertilization is currently unknown (denoted by a question mark).

Given the inability of the aforementioned experimental paradigms to recapitulate the initiation of [Ca2+]i oscillations in mammals, consensus began to coalesce on the need for a novel mechanism to explain the initiation of oscillations in these species. The hypothesis that emerged, “the fusion hypothesis,” proposed that upon gamete fusion the sperm delivers a factor, commonly referred to as the sperm factor (SF), into the ooplasm capable of activating the PI pathway and oscillations (Parrington et al., 1996). The initial, and sole, experimental support for this hypothesis was the demonstration that injection of sperm extracts into mammalian eggs was able to replicate the pattern of oscillations initiated by the sperm (Stice and Robl, 1990; Swann, 1990; Stricker, 1997; Wu et al., 1997). Curiously, the first demonstration of this mechanism was obtained in sea urchin eggs (Dale et al., 1985), a species in which, paradoxically, the hypothesis under discussion may not account for the mechanism of fertilization (see above). The advent of intracytoplasmic sperm injection (ICSI, Palermo et al., 1992), a technique by which an intact sperm delivered into the ooplasm is capable of initiating fertilization-like [Ca2+]i responses (Tesarik and Testart, 1994) and embryo development to term, consolidated the concept that a sperm product was responsible for initiating oscillations in mammalian eggs. While the success of ICSI unquestionably implicates the sperm as the harbor of the factor that triggers oscillations, close examination of the ICSI-induced [Ca2+]i responses underscores the notion that events that take place prior to or during interaction of the gamete membranes are pivotal for normal fertilization. For instance, in mouse eggs, ICSI-initiated oscillations occur less frequently after the first hour and show premature termination (Kurokawa and Fissore, 2003). More revealing still are the findings in large domestic species where fertilization by ICSI fails altogether to initiate [Ca2+]i oscillations (Bedford et al., 2004; and unpublished data [Malcuit, Kurokawa, and Fissore]).

WHAT IS THE SPERM FACTOR?

Despite the fact that the identity of the molecule(s) responsible for the SF activity has yet to be fully elucidated, glimpses of the properties of the putative molecule(s) have emerged from fertilization and biochemical studies (Wu et al., 1997, 2001; Kimura et al., 1998; Jones et al., 2000; Perry et al., 2000; Rice et al., 2000). First, while early studies assumed that all SF activity was rapidly released into the ooplasm (Stice and Robl, 1990; Swann, 1990), subsequent studies revealed that the Ca2+-inducing activity was also present in detergent-resistant sperm domains, most likely the sperm perinuclear theca (Kimura et al., 1998; Perry et al., 1999, 2000; Knott et al., 2003; Sutovsky et al., 2003). Consistent with the concept of SF distribution to several sperm compartments was the demonstration that complete release of SF activity into the ooplasm required ∼2 h (Knott et al., 2003). Second, consistent with its perinuclear localization in the sperm, in vitro fertilization and ICSI studies showed that sperm's Ca2+-releasing activity could be recovered after fertilization, as it associated with the pronuclei of the developing zygotes (Kono et al., 1995, 1996; Knott et al., 2003). Third, in vitro PLC assays using sperm extracts showed that these extracts possessed high PLC activity, nearly twice as high as the activity present in other tissues known to express several PLCs isoforms (Jones et al., 2000; Rice et al., 2000). Importantly, the PLC activity of sperm extracts is prominent even in the presence of basal [Ca2+]e, which is very relevant given that this molecule is expected to initiate oscillations in mammalian eggs, which at the time of fertilization show [Ca2+]i basal levels of ∼0.1 µM. Therefore, since several PLC isoforms are expressed in mammalian sperm (Choi et al., 2001; Fukami et al., 2001; Parrington et al., 2002), these enzymes surfaced as logical candidates to be the SF. Importantly, injection of recombinant proteins representing most of the isoforms expressed in sperm failed to initiate oscillations, or it did so at non-physiological concentrations (Mehlmann et al., 2001). Furthermore, chromatographic fractionation of sperm extracts revealed that none of the known PLCs were present in the fractions with [Ca2+]i oscillation-inducing activity (Wu et al., 2001; Parrington et al., 2002). Hence, if a sperm PLC were to be the SF, it had to be a novel PLC. Towards this end, a novel sperm-specific PLC, PLCζ (Saunders et al., 2002), was screened out of mouse expressed sequence tags. Initial studies revealed that PLCζ exhibits [Ca2+]i oscillation-inducing activity ascribed, thus far, only to the sperm or SF. Hence, PLCζ may be the long sought-after SF.

PLCζ: THE TRIGGER OF [Ca2+]i OSCILLATIONS DURING MAMMALIAN FERTILIZATION

Since its discovery from mouse testis (Saunders et al., 2002), PLCζ has also been identified and cloned in humans and primates (Cox et al., 2002), and highly homologous sequences have been reported for several other species including the bovine and porcine (Genbank accession numbers GI:55669158 and GI:32400660, respectively). PLCζ is to date, the most elementary of PLC isoforms identified. In concurrence with the modular organization of other PLCs (Rhee, 2001), PLCζ consists of four Ca2+-binding EF hands, X and Y catalytic domains, and the Ca2+-dependent phospholipid-binding C2 domain (Essen et al., 1997). Notably, PLCζ lacks the typical pleckstrin homology (PH) domain, which has been found in all previously identified PLC isoforms (Williams and Katan, 1996). In support of its purported role as the SF, injection of recombinant PLCζ (Fujimoto et al., 2004; Kouchi et al., 2004) or PLCζ cRNA has been shown to evoke sperm-like- oscillations in mouse (Cox et al., 2002; Saunders et al., 2002; Fujimoto et al., 2004; Kouchi et al., 2004; Larman et al., 2004), human (Rogers et al., 2004), and bovine eggs (Malcuit et al., 2005). In addition, in vitro PLC assays using recombinant PLCζ revealed high enzymatic activity at basal [Ca2+]e concentrations (Kouchi et al., 2004). Moreover, PLCζ cRNA-induced oscillations that ceased at ∼ the time of pronuclear formation, which is comparable to what is observed after natural fertilization (Larman et al., 2004; Yoda et al., 2004) and, like the sperm, zygotes activated by injection of PLCζ cRNA showed high rates of in vitro development to the blastocyst stage (Cox et al., 2002; Saunders et al., 2002). Lastly, a recent report in mouse sperm has localized PLCζ to the post-acrosomal region of mouse sperm (Fujimoto et al., 2004), which is the first area thought to come in contact with the ooplasm after the fusion of gametes (Sutovsky et al., 2003). Collectively, the evidence in support of PLCζ as the mammalian SF is compelling. Nevertheless, important questions remain to be addressed regarding the expression, localization, and storage of PLCζ in sperm, its mechanism of release into the ooplasm, and mechanism(s) of activation once in the egg. Additionally, it needs to be demonstrated whether or not PLCζ represents the sole inducer of [Ca2+]i oscillations during fertilization in all mammals. In regards to the latter, two recent reports are worth discussing. First, using transgenic-mediated RNA interference, male mice were created with decreased levels of PLCζ protein. In vitro fertilization (IVF) using sperm from these males resulted in attenuated [Ca2+]i responses, and one of the males produced greatly reduced litter sizes despite inducing normal rates of in vitro embryonic development (Knott et al., 2005). Second, a recent report showed that following fractionation of sperm extracts, the presence of immunoreactive 72-kDa PLCζ correlated with the ability of these fractions to induce egg activation (Fujimoto et al., 2004). Importantly, several of the active fractions were devoid of 72-kDa PLCζ, and others had greatly reduced amounts of immunoreactive? PLCζ. Nevertheless, no mechanism(s) was proposed to account for the egg activation capacity in the fractions devoid of 72-kDa PLCζ. Perhaps the answers to these questions will only be fully realized once a PLCζ knockout is created.

REGULATION OF SPERM/PLCζ-INDUCED [Ca2+]i OSCILLATIONS

Having been initiated several minutes after sperm-egg fusion (Lawrence et al., 1997), oscillations in mammalian eggs persist for several hours at regular intervals of ∼15–30 min (Fig. 2A); in the mouse, oscillations cease at the time of pronuclear formation, revealing an association of Ca2+ release with the cell cycle (Jones et al., 1995; Nixon et al., 2000; Marangos et al., 2003). The mechanism(s) that underlies this correspondence have not yet been fully elucidated, although eggs seem to use several means to synchronize the cell cycle and [Ca2+]i oscillations. It is well established that the sperm's [Ca2+]i oscillation-inducing factor is targeted to the forming pronucleus of the developing zygote, and that, consistent with this, pronuclei of fertilized zygotes, but not of parthenogenetic zygotes, possess the ability to initiate [Ca2+]i oscillations when fused with MII eggs (Kono et al., 1995). Moreover, a [Ca2+]i transient ensues when the pronucleus of fertilized zygotes breaks down (PNBD) as the embryo enters the first mitosis (mitosis I) (Kono et al., 1996; Day et al., 2000). Therefore, the recent finding that PLCζ is targeted to the pronucleus by a stretch of basic amino acids in the linker domain of the protein implicated pronuclear sequestration of PLCζ as one of the mechanism entraining the cell cycle and [Ca2+]i oscillations in mouse zygotes (Larman et al., 2004). Demonstration of PLCζ localization was performed by fluorescent antibody labeling of a Myc-tagged version of PLCζ and this approach was also used to disclose the information that, upon PNBD, [Ca2+]i responses and Myc-PLCζ reappeared in the cytoplasm. In this study, mutation of a single amino acid of the nuclear localization signal abrogated pronuclear localization and extended oscillations. Interestingly, real time evaluation of PLCζ distribution using a yellow fluorescent protein “Venus”-fused PLCζ reported that [Ca2+]i oscillations ceased prior to the accumulation of PLCζ in the pronucleus (Yoda et al., 2004). Moreover, the findings in the bovine demonstrating that [Ca2+]i oscillations persist through interphase (Fissore et al., 1992; Nakada et al., 1995), despite the recently reported conservation of key basic amino acids in the linker region of bovine PLCζ is not consistent with the notion that PLCζ sequestration represents the sole mechanism coordinating the cell cycle with oscillations.

Besides the discrepancy of PLCζ distribution and cessation of oscillations reported by Yoda et al. (2004), other findings raised the possibility that other cellular mechanisms may assist in the cell cycle coordination of oscillations in the mouse. For instance, close examination of the pattern of sperm-initiated oscillations shows a gradual, but significant, increase in the intervals between rises well before formation of the pronuclei (Kurokawa and Fissore, 2003). What is more, interphase stage injections of either IP3 (Jones et al., 1995; FitzHarris et al., 2003) or adenophostin A, a non-hydrolyzable IP3 analogue, failed to initiate oscillations at this stage (Jellerette et al., 2004). Together, we interpreted the results to mean that the Ca2+ release machinery, and most likely the IP3R-1 is less responsive to IP3 and this is responsible, at least in part, for the termination of oscillations as zygotes transition into interphase (Jellerette et al., 2004).

Several mechanisms could account for decreased responsiveness of IP3R-1 in zygotes. Following fertilization, the concentration of IP3R-1 in zygotes steadily declines to stabilize at ∼50% of the initial IP3R-1 mass (Parrington et al., 1998; Brind et al., 2000; Jellerette et al., 2000). This IP3R-1 downregulation, which is driven by binding of IP3 to its receptor (Zhu and Wojcikiewicz, 2000; Khan and Joseph, 2003), might severely alter the ability to sustain oscillations as fertilization progresses. Consistent with this notion, it was recently demonstrated that elimination of ∼50% of IP3R-1 mass resulted in premature termination of oscillations in mouse eggs (Xu et al., 2003), although the impact that the gradual decline of receptor mass has on fertilization associated oscillations has not been thoroughly tested. The responsiveness of the IP3R-1 in zygotes may be also compromised by the declining activity of M-phase kinases (FitzHarris et al., 2003; Jellerette et al., 2004). For instance, several M-phase phosphorylation consensus sites have been identified in IP3R-1 and seem to undergo active modifications during the cell cycle in somatic cells (Malathi et al., 2003) and in mouse eggs/zygotes (Jellerette et al., 2004). M-phase kinases, and most specifically cdk1, may also affect IP3R-1 conductivity and presence of oscillations by regulating the organization of the ER. The presence of ER and IP3R-1 clusters, which is associated with active cdk1, correlates with a greater Ca2+-releasing ability of IP3R-1 and hence, a propensity for oscillations in a variety of eggs (Stricker et al., 1998; FitzHarris et al., 2003; Stricker and Smythe, 2003; Machaca, 2004). Nonetheless, whether this reflects a specific effect of receptor clustering or is due, at least in part, to other cell cycle-associated cellular changes remains to be determined. Lastly, the effect of M-phases kinases on Ca2+ influx pathways, as reportedly takes place in Xenopus oocytes (Machaca and Haun, 2002), cannot be discounted as playing a role in the presence and persistence of oscillations. Therefore, the cell-cycle association of [Ca2+]i oscillations in mammalian eggs may represent the coordinated modification of several cellular mechanisms; elucidating the precise contribution of each then, should be a task for the near future.

[Ca2+]i OSCILLATIONS AND ACTIVATION OF DEVELOPMENT

In spite of the realization that the activation of embryonic development in mammals relies on the initiation of [Ca2+]i oscillations, there is remarkably little known as to what are the “molecular effectors” in the egg that translate the transient [Ca2+]i elevations into cellular events. It is well established that unlike echinoderm and amphibian fertilization, initiation of multiple rises is required to promote complete exit from meiosis in mammals (for review see Stricker, 1999). Subsequent studies, first in Xenopus eggs, have lent support to the notion that [Ca2+]i elevations mediate meiotic resumption by activating the type II Ca2+/calmodulin-dependent protein kinase (CaMKII), which by an as yet unknown mechanism, leads to activation of the anaphase promoting complex (APC) (Yamamoto et al., 2005) presumably through destruction of a member of the early mitotic inhibitor (Emi) proteins (Reimann et al., 2001; Reimann and Jackson, 2002; Tung et al., 2005). Interestingly, in mammals, each [Ca2+]i rise results in a concomitant activation of CaMKII (Markoulaki et al., 2003, 2004), and this is envisioned to be responsible for the steady increase in the rate of cyclin degradation through maintenance of APC activity (Dupont, 1998; Marangos and Carroll, 2004b). The APC, an E3 ubiquitin ligase, targets cell cycle proteins such as cyclin B for proteasomal degradation following their poly-ubiquitination (King et al., 1995; Yu et al., 1996; Nixon et al., 2002; Marangos and Carroll, 2004a,b). The loss of cylin B, the regulatory subunit of MPF (Gautier et al., 1990) and limiting factor for the maintenance of the metaphase state, results in the exit from M-phase and thus the resumption of meiosis. Interestingly, this Ca2+-mediated cell-cycle transition is thought to be unique to eggs (Marangos and Carroll, 2004b), and requires stable levels of cytostatic factor (CSF) (Hyslop et al., 2004]), which is itself, unique to eggs (Masui, 1974).

If the only function of Ca2+ oscillations were to promote exit of meiosis, then elevations of [Ca2+]i according to disparate, but not excessive protocols, should lead to equal rates of activation and development. Nonetheless, elegant studies by Ozil and co-workers have demonstrated this not to be the case. In their studies, activation protocols that adhered more closely to the sperm-like patterns of oscillations resulted in higher rates of implantation and development (Ozil, 1990; Ozil and Huneau, 2001; Ozil et al., 2005). In these studies the amplitude, frequency, and number of [Ca2+]i rises were modulated by electroporation of [Ca2+]e into the ooplasm of rabbit and mouse eggs. Each parameter tested had a marked effect on the rates of egg activation as well as post-implantation development. In line with the developmental benefits of Ca2+, more recent studies revealed that different events of egg activation have dissimilar requirements regarding their initiation and completion, with the completion of each event needing higher number of rises (Ducibella et al., 2002; Ozil et al., 2005). Especially relevant to embryo development, is the possible impact of the [Ca2+]i activating stimulus on the recruitment of specific maternal RNAs (Ducibella et al., 2002; Ozil et al., 2005), the recruitment and translations of which may be critical for the activation of the zygotic genome (Aoki et al., 2003). For instance, it was shown that “recruitment of maternal RNAs,” as evidenced by the detection of new protein synthesis, was initiated by the administration of eight [Ca2+]e pulses but only became fully apparent, or “fertilization-like,” in zygotes receiving a total of 24 [Ca2+]e pulses (Fig. 2B and Ducibella et al., 2002). While these results imply a correspondence between [Ca2+]i oscillations and protein synthesis, it is important to discriminate whether the observed changes in protein profiles were not simply due to progression into zygotic interphase, which seemed to exhibit similar requirements for [Ca2+]e pulses (Ducibella et al., 2002). Nonetheless, the approach used in this study, followed by identification of the proteins regulated by Ca2+ pulses, could prove invaluable in elucidating yet uncharacterized signaling pathways during egg activation.

How then does the egg transduce these compounding [Ca2+]i rises into an effector? It has recently been shown that just as concentrations of [Ca2+]i oscillate in response to IP3 production, so too does the activity of CaMKII (Tatone et al., 2002; Markoulaki et al., 2003, 2004). For instance, the activity of CaMKII not only increased in response to augmenting levels of [Ca2+]i, but oscillated in close synchrony with each [Ca2+]i transient after fertilization, as determined in single eggs by simultaneous monitoring of [Ca2+]i levels and kinase activity (Markoulaki et al., 2003, 2004). Hence, given that each [Ca2+]i transient during fertilization accounts for a single event during which CaMKII acts upon its targets, it can be speculated that the compounding effect of each [Ca2+]i rise is required for brief periods of CaMKII activity (De Koninck and Schulman, 1998), the culmination of which results in successful egg activation. Importantly, a recent report in hippocampal dendrites has demonstrated that CaMKII directly phosphorylates the cytoplasmic polyadenylation element binding protein (CPEB) that is directly responsible for the polyadenylation, recruitment, and translation of mRNAs (Atkins et al., 2004). A similar mechanism, therefore, could potentially function in eggs. This would provide a direct link between the results mentioned above in which repetitive Ca2+ pulses modulated protein expression patterns from stored maternal mRNAs (Ducibella et al., 2002; Ozil et al., 2005). It is worth noting that other kinases such as PKC (Halet et al., 2004), or proteins such as actin (McAvey et al., 2002) and calreticulin (Tutuncu et al., 2004), may also serve as effectors of [Ca2+]i rises during egg activation, although additional research is required to realize the extent of their participation.

[Ca2+]i OSCILLATIONS AND NUCLEAR REPROGRAMMING

Besides the possible aforementioned impact that the Ca2+ activating stimulus may have on zygotic gene expression via recruitment and translation of specific maternal mRNAs, an equally intriguing possibility is that the fertilization-initiated [Ca2+]i responses are also involved in the nuclear remodeling of the parental genomes, which commences soon after fertilization (Dean et al., 2003). After fertilization, the gametes undergo extensive reconfiguration, which is required for activation of the zygotic genome (for review see Latham, 1999). Of the two gametes, the paternal genome undergoes the most rapid and dramatic change, as is subjected to the exchange of protamines for maternal histones and this occurs in concert with sperm nuclear decondensation (Adenot et al., 1997; Santos et al., 2002). Additionally, the paternal genome undergoes near complete DNA demethylation of CpG dinucleotides before reaching the pronuclear stage (Barton et al., 2001; Santos et al., 2002). While the precise nature of the molecular mechanisms that underlie the transformation of the sperm nucleus into a pronucleus is not fully elucidated, given that this reconfiguration coincides with presence of [Ca2+]i oscillations, a role for Ca2+-dependent processes in such events cannot be discounted.

Nuclear remodeling and epigenetic modification of the zygotic genome also entails covalent modifications of the nucleosomes' core histones (Stein et al., 1997; Strahl and Allis, 2000). These changes involve phosphorylation, acetylation, and methylation of histones as well as the removal by opposing enzymes of these modifications from the target amino acids in these proteins, which are the basic amino acids lysine and arginine (Fischle et al., 2003; Sarmento et al., 2004). While histone methylases have been identified in somatic cells and eggs, the enzymes that remove these modifications are not presently well known. Importantly, mouse oocytes and eggs reportedly abundantly express a peptidylarginine demethyliminase-like molecule, ePAD, which functionally demethylates histone by the conversion of methylarginine to citrulline, and the subsequent release of methylimine (Wright et al., 2003; Sarmento et al., 2004; Wang et al., 2004). Relevant to the possible role of Ca2+ on nuclear reprogramming, is the finding that PADs are Ca2+ sensitive enzymes (Vossenaar et al., 2004; Wang et al., 2004). Therefore, it will be important to address the issue of the Ca2+-dependence of ePAD for its functional role on nuclear remodeling and reprogramming of the fertilized zygote, and to determine if maximal ePAD activity requires [Ca2+]i oscillations similar to those seen during fertilization.

CONCLUSIONS AND FUTURE DIRECTIONS

The mechanisms of fertilization-induced [Ca2+]i oscillations and egg activation are finally becoming unraveled. We are just now learning how [Ca2+]i oscillations are induced by the sperm and how these [Ca2+]i rises are transduced into egg activation events. We must continue to reveal the downstream pathways that translate [Ca2+]i rises into cellular and molecular events, and determine if a link exist between [Ca2+]i oscillations and nuclear reprogramming. Elucidation of these pathways and mechanisms will make it possible to determine whether altered [Ca2+]i oscillations, such as those that may occur in zygotes generated by assisted reproductive techniques, may affect the health of the offspring later in life. This is of the utmost importance in light of the popularity of ICSI as a means for assisted reproduction in fertility clinics around the globe. Not only will the study of these events help elucidate the precise mechanisms by which the sperm acts during fertilization, but will also facilitate the development of more efficient exogenous activation methods that may more closely unleash the molecular events induced by fertilization, such that we may ensure that all offspring produced by such means begin their development in the most physiological way possible.

Ancillary