Oolemma Receptors and Oocyte Activation

Abstract

At fertilization the sperm triggers a series of intracellular calcium oscillations that are pivotal to oocyte activation and development. Although the biological significance of the characteristic intracellular calcium (Ca²⁺_i) oscillations is not fully understood, calcium ions are known to be involved in cortical granule release and in controlling cell cycle progression. Two different hypotheses attempt to explain how sperm initiate (Ca²⁺_i) oscillations in mammalian oocytes. One hypothesis is that spermatozoa interact with a receptor located in the plasma membrane of the oocyte, which results in induction of pathways leading to activation. This receptor is coupled to a GTP-binding protein or to have tyrosine kinase activity and have the ability to induce activation of phospholipase C (PLC). In turn, PLC stimulates the hydrolysis of phosphatidyl inositol (4,5)–bisphosphate (PIP2) to produce diacylglycerol (DAG) and 1,4,5 inositol trisphosphate (IP3), a common Ca²⁺ releasing compound. Most studies used to develop the mammalian model of oocyte activation have been performed in the mouse. There is a paucity of information from other mammalian models. The predominant mouse model of oocyte activation is that there is a soluble factor (PLC-zeta) delivered to the cytosol after fertilization that induces oocyte activation. However, as data in other mammals is collected, substantial evidence is beginning to support the existence of other more complex oocyte activation pathways in both murine and non-murine systems. Indeed, activation may involve redundant processes, each of which acting alone may be able to induce aspects of oocyte activation. Recent findings demonstrate the involvement of receptors that are known to associate in large, multimeric complexes. This fact leads one to speculate that the process of oocyte activation by the sperm cell is a highly complex and elaborate process that likely involves many more players than perhaps was initially expected.

KEYWORDS:

• activation
• bovine
• oocyte
• receptor
• sperm factor

Abbreviations
PLC:	=	phospholipase C
PIP2:	=	phosphatidyl inositol (4,5)-bisphosphate
DAG:	=	diacylglycerol
IP3:	=	1,4,5 inositol trisphosphate
Ca2+:	=	intracellular calcium
G-protein:	=	GTP-binding protein
SF:	=	sperm factor
PLCζ:	=	phospholipase C zeta
ICSI:	=	intracytoplasmic sperm injections
PAWP:	=	postacrosomal sheath WW domain binding protein
ECM:	=	extracellular matrix
RGD:	=	Arg-Gly-Asp
6-DMAP:	=	6-dimethylaminopurine
ADAM1:	=	sperm fertilin alpha (also known as A disintegrin and A metalloprotease 1)
ADAM2:	=	fertilin beta
ADAM3:	=	cyritestin
BOMP:	=	bacterial outer membrane
CRISP1:	=	cystein-rich secretory protein 1
GPI:	=	glycosyl phosphatidylinositol
CTK:	=	cytoplasmic tyrosine kinase
FAK:	=	focal adhesion kinase

INTRODUCTION

After the sperm cell penetrates the zona pellucida it moves through the perivitelline space and comes into contact with the vitelline membrane. By this time the sperm has undergone the acrosome reaction and thus ligands present on the inner acrosomal membrane are exposed and come into contact with receptors present on the vitelline membrane. These receptors bind the gametes together and likely transduce a signal across the vitelline membrane leading to activational events within the oocyte. Oocyte activation is defined by a series of metabolic changes that occur within the oocyte and include transient release of intracellular calcium [Ca²⁺_i], cortical granule release, resumption of meiosis, and extrusion of the second polar body. There are also receptor interactions that lead to fusion of the sperm and oocyte membranes.

At fertilization, the sperm induces repetitive intracellular calcium transients that are critical to oocyte activation and development in every species that has been studied [Berridge and Galione 1988; Kline and Kline 1992]. Calcium is a divalent cation that commonly functions as a second messenger, relaying signals downstream so that a cell can respond to various stimuli. The cell has established critical mechanisms, including ATP driven ion channels and ion exchange channels, which precisely maintain intracellular calcium at very low levels. In this way, calcium is either sequestered within the endoplasmic reticulum or in the extracellular space. This low intracellular and high extracellular calcium environment establishes a strong gradient, such that, when an appropriate signal is received, a very rapid influx of calcium occurs. Several enzymes within the cell, including calmodulin and calpain, are designed to respond to this concomitant increase in calcium. These enzymes can, in turn, activate other enzymes thus propagating the signal cascade. The biological significance of the changes in Ca²⁺_i concentration as it relates to oocyte activation is not fully understood, however, calcium ions are known to be involved in cortical granule release which leads to a block to polyspermy and in the control of cell cycle progression [Kline and Kline 1992].

Two different hypotheses attempt to explain how sperm initiate Ca²⁺_i oscillations in mammalian oocytes. One hypothesis is that spermatozoa interact with a receptor located in the plasma membrane of the oocyte. This receptor is postulated to be coupled to either a GTP-binding protein (G-protein) or to have tyrosine kinase activity and the ability to activate phospholipase C (PLC) which, in turn, stimulates the production of diacylglycerol (DAG) and 1,4,5 inositol trisphosphate (IP3), a common Ca²⁺ releasing compound, from phosphatidyl inositol (4,5)-bisphosphate (PIP2).

The second hypothesis proposes that a factor from the sperm (SF) is released into the oocyte and, by interacting with unknown cytosolic targets, generates Ca²⁺_i oscillations. Several lines of evidence support this theory. First, injection of a sperm preparation into hamster and mouse oocytes triggered Ca²⁺_i oscillations similar to those observed during fertilization [Swann 1994]. Secondly, in most species, sperm-oocyte fusion appears to precede the initiation of oscillations [Jones et al. 1998; Lawrence et al. 1997]. Finally, injection of sperm directly, incon junction with intracytoplasmic sperm injection (ICSI), into the cytoplasm of human oocytes results in initiation of Ca²⁺_i oscillations and subsequent embryonic development [Tesarik et al. 1994]. Phospholipase C zeta (PLCζ) has been proposed to be the SF involved in activation of murine oocytes and has been shown to initiate Ca²⁺_i oscillations and parthenogenetic development to the blastocyst stage [Saunders et al. 2002]. These results suggest that oocyte activation, in mice and humans, may be mediated by a soluble factor independent of a receptor on the plasma membrane of the oocyte.

The most compelling results to date in bovine supporting an SF hypothesis are recent reports by Ross et al. [2008, 2009] indicating that the injection of cRNA coding for bovine PLCζ transcript induces activation of bovine oocytes and somatic cell nuclear transfer embryos. The activation response in these two reports is defined as highly efficient development (comparable to IVF controls) to the blastocyst stage and calcium transients that closely resemble those observed at fertilization. However, the main obstacle to establishing this hypothesis as a native mechanism of bovine activation is the inability of any SF isolates to induce complete activation of bovine oocytes [Malcuit et al. 2006]. Putative bovine SFs from sperm lysates have been shown to induce activation responses in murine oocytes, but fail to induce similar responses when used in bovine oocytes [Tang et al. 2000]. The deficiency in the SF hypothesis is further illustrated by the fact that, contrary to the results observed in human and mouse oocytes following intracytoplasmic sperm injections (ICSI), ICSI-treated bovine oocytes fail to develop [Chung et al. 2000; Suttner et al. 2000; Oikawa et al. 2005; Abdalla et al. 2009; Devito et al. 2009]. It is well established that PLC is involved in the activation process of several species including the bovine [Runft et al. 2002; Carroll et al. 1999, 1997; Tokmakov et al. 2002]. Therefore, it is not surprising that injection of PLC cRNA, induces efficient activation of the bovine oocyte. The experiment that must be done to confirm that it is only PLCζ that induces activation is to provide other isotypes of PLC to access their potential for activation or reduce egg PLC isoforms to evaluate the resulting impact on activation. If other cRNA isotypes of PLC induce a response similar to that observed following PLCζ injection, this would argue that PLCζ-triggered activation may not be a physiologic response, or at a minimum, not the only event in activation.

In the mouse, reductions in PLCβ1, an egg PLC isoform, results in a concomitant reduction in calcium transient amplitude, but had no impact on duration or frequency following fertilization [Igarashi et al. 2007]. In these same experiments, over-expression of PLCβ1 resulted in an altered initial transient following fertilization and subsequent changes in duration and frequency of subsequent transients. Together, these data indicate the potential of a more active role of egg-PLC isoforms in the activation mechanism than previously thought.

A recent report identified a novel alkaline extractable protein named PAWP (postacrosomal sheath WW domain binding protein) that is located within the post acrosomal sheath of the bovine sperm head. Microinjection of PAWP induces pronuclear formation in bovine, porcine, and rhesus monkey oocytes [Wu et al. 2007]. This data indicates that protein factors in addition to PLC are able to induce activation events, either directly as a SF or through a membrane associated event. A need for more than one factor could explain the failure of bovine SF isolates and ICSI protocols to effectively induce a complete activation response and development. Possibly, PLC activity must be preceded by a membrane-associated event.

The debate between these two hypotheses has certainly not ended. There is data to indicate that a SF can induce activation and limited development in some species. However, activation in the bovine appears to be refractory to initiation by putative SF. All reports to date indicate SF fails to induce par thenogenetic development in bovine oocytes and injection of bovine sperm into bovine oocytes fail to induce subsequent development [Malcuit et al. 2006]. It is understood that the sperm cell, when it undergoes the acrosome reaction (a fusion of cell membranes that release factors allowing the sperm to penetrate the zona pellucida, and exposes ligands for binding with the oocyte vitelline membrane) experiences many events which are similar to oocyte activation, including transient release of Ca²⁺_i [Kirkman-Brown et al. 2004, 2002, 2000; Marquez and Suarez 2004]. Additionally, there is significant data in the bovine as well as other species, indicating that the sperm induces activation of the oocyte through a receptor-mediated interaction. Current models for oocyte activation in both echinoderm and Xenopus oocytes do not appear to include PLCζ, but seem to include Src family kinases and PLC isoforms that are endogenous to the oocyte [Stith et al. 2005; Townley et al. 2005]. The following is a review of possible mediators of oocyte activation particularly pertaining to the bovine model.

OOCYTE RECEPTORS

Literature indicates that there are two primary putative receptor families, integrins and tetraspannins, which are thought to have some interaction with sperm ligands.

Integrins

Integrin molecules are ubiquitously synthesized cell surface adhesion receptors that form a family of transmembrane glycoproteins with heterodimeric structure [Hynes 1992]. Mammalian systems include genes for 18 different alpha and 8 different beta subunits that combine in various permutations to form 24 known integrin receptors. The specific subunit pair that composes the integrin receptor is a major determinant of ligand specificity. Integrin receptors function in cellular processes such as attachment, migration, and cell differentiation. Integrins are known to be involved in the onset and continuation of several disease states, including tumor metastasis, viral infections, and osteoporosis. Because of their role in prominent diseases such as cancer and osteoporosis they are a well-understood family of cell adhesion receptors. Integrins facilitate attachment of the cell to the extracellular matrix (ECM) impacting much of the cellular architecture and function, and act as two-way signaling molecules [Sjaastad and Nelson 1997]. Integrins have a short cytoplasmic domain that links to cytoskeletal elements, and when adhesion occurs will induce ‘outside in’ signaling mechanisms, which can feedback ‘inside out’ signaling to regulate integrin function, cytoskeletal assembly, cell behavior, and protein synthesis [Hynes 1992].

In addition to integrin subunits conferring specificity, the presence of specific integrin recognition sequences in the ligand determines which integrins will bind the ligand. Fibronectin, collagen, vitronectin, as well as pathogens like viruses and certain snake venom proteins influence cells by binding and signaling through integrins. Although there is diversity in the type of ligand that binds cells through integrins, there is a common element to the ligand motif. Most integrins bind to an element that contains an aspartic acid residue (RGD, ECD, LDV, KGD, RTD, and KQAGD) and they recognize the RGD sequence, which appears in ECM proteins and cell surface molecules [Ruoslahti and Pierschbacher 1987], and has been implicated in fertilization [Campbell et al. 2000; Sessions et al. 2006].

An important feature of integrin receptor binding is the requirement of divalent cations. Media commonly used for in vitro fertilization protocols require Ca²⁺ and Mg²⁺ for proper fertilization to occur. Campbell et al. [2000] also demonstrated that the synthetic RGD-containing sequence was unable to generate release of intracellular calcium (a hallmark of fertilization) when used in the absence of divalent cations.

Integrins have been shown to be involved in the process of fertilization [Almeida et al. 1995; Bronson and Fusi 1990a, 1990b]. In 1990, Bronson and Fusi showed that addition of RGD-containing peptides in a heterologous system (human sperm and zona-free hamster oocytes) or a homologous system (hamster sperm and zona-free hamster oocytes) resulted in the complete inhibition of fertilization [Bronson and Fusi 1990a, 1990b]. Integrins were then found on the plasma membrane of unfertilized murine oocytes and, using a combination of antibody inhibition, peptide inhibition, and somatic cell transfection experiments, the integrin α6β1 complex was shown to serve as a sperm receptor [Almeida et al. 1995]. Pig oocytes also consistently express αVβ1 integrin on the vitelline membrane, but not other integrin molecules [Linfor and Berger 2000]. These results provide circumstantial evidence that a αVβ1 pig oocyte integrin might interact with a ligand on the sperm plasma membrane during fertilization. Bovine oocytes have also been shown to express integrins on their vitelline membrane. The αV, α6, α4, α2, β1, and β3 integrin subunits were all found on the bovine oocyte vitelline membrane [Pate et al. 2007]. Prior results demonstrated that a short synthetic RGD-containing peptide could induce activation and parthenogenetic development in bovine oocytes [Campbell et al. 2000]. Because RGD is a known integrin binding ligand, this provides clear evidence in support of integrin-mediated bovine oocyte activation.

In contrast to the experiments mentioned above, a number of publications have indicated that there is no role for integrins, especially α6β1 in fertilization [Miller et al. 2000; Barraud-Lange et al. 2007]. Knockout mouse models of α6 and β1, as well as antibody inhibition experiments in which αV and β3 subunits were blocked demonstrated no inhibition of binding or fusion between gametes. It was thus concluded that no integrin subunits known to be on the mouse oocyte or known to bind A Disintegrin and A Mettalloprotease domain (ADAMs) is involved in fertilization [He et al. 2003].

Tetraspannins

Integrins bind their ligands relatively loosely compared to other receptors, but are present in much higher concentrations on the surface of cells. In order to compensate for the relatively loose binding, these integrin receptors will cluster together at the site of attachment to increase the strength of binding to the ligands. The complexes of integrin and associated molecules that form these clusters are thought to be organized by members of the tetraspannin family, called a tetraspannin web [Ellerman et al. 2003]. Tetraspannins are large proteins with 4 hydrophobic regions and therefore span the plasma membrane four times, with 2 extracellular and 3 cytoplasmic domains. The family contains about 30 members in mammals, several of which have been identified in the mouse oocyte (CD9, CD81, CD82, CD151, SAS, Tspan-3, and Tspan-5). CD9, a tetraspannin signaling molecule known to associate with β1 integrins [Chen et al. 1999], has been shown to be involved in the process of sperm-oocyte fusion. Oocytes from female CD9–/–mice were unable to fuse with sperm, and hence were infertile [Miller et al. 2000; Miyado et al. 2000]. In a subsequent study, sperm fusion to CD9-/-mouse oocytes was restored following expression of human CD9 and mouse CD81 in these animals [Kaji et al. 2002]. CD9 has been identified and localized in pig oocytes and it has been determined that both binding and fusion of sperm occur in anti-CD9 antibody treated oocytes [Li et al. 2004]. It has also been demonstrated that avidity binding between fertilin beta (ADAM 2) and α6β1 requires cooperation between α6β1 and CD9 [Chen et al. 1999]. Though tetraspannins were not originally thought to have a role as receptors, it has recently been demonstrated that a member of the immunoglobulin (IgSF)/CEA subfamily that binds to CD9 called PSG17, is involved in the process of sperm-oocyte fusion [Ellerman et al. 2003]. The oocyte receptor complex appears to be far more complex than originally thought and is yet to be completely understood.

Sperm Ligands

There are several families of proteins that are involved in sperm-oocyte interactions leading to oocyte activation. Both fertilin α and β, and snake venom disintegrins, are members of a growing family of proteins known as ADAMs. To date there are 15 ADAM family members described and sequenced at the cDNA level in the guinea pig, monkey, mouse, rabbit, rat, and human [Wolfsberg et al. 1995a, 1995b]. Fertilin α and β, formerly known as PH-30α and PH-30β, are now referred to as ADAMs 1 and 2 [Huang 1998; Wolfsberg et al. 1995a, 1995b]. Every member of this family has a proteolytic, an adhesion, a fusion, an EGF-like, and signaling domains [Wolfsberg et al. 1995a, 1995b]. The specific identity of a disintegrin molecule, or other RGD containing protein on the bovine sperm inner acrosomal membrane is still to be determined. Although RGD-containing bovine inner acrosomal membrane proteins that interact with oocyte membrane receptors have been identified, their specific biological significance has not yet been elucidated [Pate et al. 2008]. Identification of sperm ligands and intracellular signaling molecules would potentially increase the efficiency of in vitro embryo production by nuclear transfer and could be used as a target for inhibiting fertilization in mammalian populations where this is desirable.

Disintegrins

Disintegrins are peptides of about 70 amino acids that contain many cysteine residues that are involved in intramolecular disulphide bonding. They also contain an Arg-Gly-Asp (RGD) tripeptide sequence that acts as the recognition site of adhesion proteins like integrins [Huang 1998]. Disintegrins were first discovered as a component of some snake venoms that operate by inhibiting platelet aggregation by binding to integrin receptors on platelets and thereby blocking binding to fibrinogen. The discription of the effect of snake venom disintegrins was a significant development in the study of integrin-ligand interactions [Huang 1998; 38]. Furthermore, these antagonists of ligand-integrin binding are approximately 500–1,000 fold more potent inhibitors of cell adhesion than synthetic linear RGD peptides derived from fibronectin and fibrinogen (known integrin ligands that contain the RGD sequence [Hynes 1992]). This immense increase in binding affinity is indicative of a functionally optimal 3-dimensional conformation of the RGD-containing protein [Huang 1998], and is important in the function of disintegrins in bovine sperm-oocyte interactions.

Although the RGD domain plays a pivotal role in determining the functional characteristics of the protein, the amino acid composition in and around the RGD tripeptide appears to have an impact on the biological activity and functional properties of RGD-containing peptides. Naturally occurring snake venom disintegrins (RGD-containing) appear to be able to impact fertilization. RGD-containing snake venom disintegrins were used in conjunction with an in vitro fertilization protocol and were shown to competitively inhibit fertilization, while control venoms without the RGD tripeptide have no effect [White et al. 2006]. Parthenogenetic development results point to several interesting possibilities about the action of RGD-containing peptides and the mechanism and/or factors involved in oocyte activation [Campbell et al. 2000]. Despite the primary role of the RGD domain in determining functional characteristics of the ligand protein, it is becoming clear that secondary contact sites, separate and distinct from the RGD tripeptide, contribute to the binding and functional properties of RGD-containing peptides [Bronson and Fusi 1990a; Iwao and Fujimura 1996; Lu et al. 1994]. Even though the linear RGD-containing peptide used in this work was able to block fertilization and induce intracellular Ca²⁺ transients, a cyclic RGD peptide or modifying the amino acid environment proximal to the tripeptide RGD may enhance the biological activity of this peptide and may negate the observed requirement for 6-DMAP in the parthenogenetic activation protocol [Campbell et al. 2000].

The ability of the RGD-containing peptide to initiate intracellular Ca²⁺ transients but not full activation as evidenced by parthenogenetic development may suggest the need for two factors to produce “complete” oocyte activation. Based on the totality of the results in the bovine, a reasonable conclusion is that intracellular calcium transients induced by the ligand-receptor pair works in conjunction with a soluble factor produced by and released from the sperm into the oocyte cytoplasm upon fusion with the oocyte, or working with existing PLC isoforms found within the oocyte. This soluble factor might mimic the effects of 6-DMAP, which inhibits protein phosphorylation, and suggest that two signaling pathways are required for “complete” activation [Moses et al. 1995; Moses and Masui 1995]. It seems highly probable that multiple sites or receptors may be required for the “complete” activation of the bovine oocyte. It is also possible that this requirement for 6-DMAP could be negated by using a more appropriate 3-D structure than the short synthetic linear peptide used by Campbell et al. [2000].

There is a strong correlation between biological activity and the surrounding 3–D structure of the disintegrin [White et al. 2006]. Bovine oocytes were responsive to naturally occurring RGD-containing snake venom (kistrin, echistatin, and elegantin) as evidenced by their ability to induce parthenogenetic development to the blastocyst stage regardless of the presence or absence of 6-dimethylaminopurine (6-DMAP). In contrast, the non-RGD-containing venom (erabutoxin b) failed to induce parthenogenetic development at all concentrations evaluated.

Substitutions within the RGD tripeptide sequence, as well as in regions flanking the RGD sequence, affect the ability of RGD to parthenogenetically activate bovine oocytes [Sessions et al. 2006]. It is apparent that the amino acids in and around the RGD motif have an effect on the functional properties of the sequence as it relates to fertilization. The peptide resulting in the strongest inhibition of fertilization, PRGDMPPDD, contained two consecutive negatively charged side chains near its negatively charged c-terminus of the peptide, indicating the negatively charged carboxy terminus might allow for a more stable conformation resulting in a better association with the integrin receptor [Sessions et al. 2006]. Also, the overall acidic nature of the peptide may have resulted in increased biological activity and better conformation that enabled more efficient binding to the integrin receptor. Sessions et al. [2006] also indicated that synthetic peptides with weakly hydrophobic residues in close proximity to each other and dispersed throughout the peptide resulted in better inhibition of sperm binding. Though evidence exists to demonstrate the importance of surrounding regions in integrin interactions, to date no 3-D disintegrin structure has been shown to be clearly optimal for binding and activation of bovine oocytes.

We recently reported [Pate et al. 2008] a list of sperm proteins that interact with receptors on the oocyte plasma membrane. Among those of greatest interest are a bacterial outer membrane (BOMP)-like protein and RIKEN cDNA hypothetical protein. BOMP proteins such as activin mediate the invasion of non-phagocytic eukaryotic cells through beta 1 integrins and utilize focal adhesion kinase and Src-family kinase pathways [Alrutz and Isberg 1998]. The RIKEN protein is postulated to have an RGD sequence. Both of these, could mediate integrin binding and signaling events during bovine oocyte activation. Work to determine the specific biological activity of these and other candidate proteins is ongoing.

Fertilin

One of the mammalian sperm proteins thought to be involved in adhesion and fusion of gametes is fertilin. Fertilin is a heterodimeric membrane protein composed of an α and β subunit [Blobel et al. 1992]. The fertilin ligand has been linked to sperm-oocyte binding and fusion. Sperm from mice lacking fertilin β are deficient in their ability to adhere to and fuse with oocytes [Frayne and Hall 1999; Nishimura et al. 2001]. Fertilin β on murine sperm is also known to bind the α6β1 integrin, and requires the CD9 as a co-receptor [Chen et al. 1999].

The involvement of sperm fertilin alpha (also known as A Disintegrin and A Metalloprotease 1, ADAM1), fertilin beta (ADAM2), cyritestin (ADAM3), and CRISP 1 (Cystein-Rich Secretory Protein 1) in sperm-oocyte adhesion are demonstrated by studies in which monoclonal antibodies, short peptides, and native or recombinant proteins were used [Frayne and Hall 1999; Nishimura et al. 2001]. It was demonstrated that fertilin beta-/- and cyritestin-/- mice had a marked inability to adhere to the vitelline membrane. In a few cases where sperm were able to adhere to the vitelline membrane, there was no noticeable reduction in fusion between the gametes [Nishimura et al. 2001]. This implicates both fertilin beta and cyritestin as mediators of gamete binding, but not fusion. Because of the disintegrin domain contained within the sequence of these sperm proteins, it has been hypothesized that integrin receptors on the oocyte vitelline membrane are likely to be involved in the process of sperm-oocyte binding, though it is not known whether there is any role in fusion.

Izumo

Recently, a novel protein from the immunoglobulin superfamily was discovered to have a role in fusion of mouse sperm and oocytes. This protein, named Izumo, is a sperm-specific, 56.4–kDa antigen that can be blocked with antibodies to prevent fusion. It was also demonstrated, using Izumo-/- mice, that fusion of gametes did not occur [Inoue et al. 2005; Sleight et al. 2005]. Although little is presently known about this specific antigen, it has been found in human sperm as well, and may prove to be a critical player in gamete interactions as more information is gathered.

Cyritestin

Several functional studies have provided evidence that fertilin α, fertilin β, and cyritestin participate in the sperm-oocyte interaction [Yuan et al. 1997]. Antibodies to these proteins bind to sperm and inhibit fertilization in IVF assays. Recombinant forms of fertilin α, fertilin β, and cyritestin bind to the mouse oocytes plasma membrane, and inhibit sperm-oocyte binding resulting in a reduction of fertilization. Cyritestin (ADAM3) is an ADAM that was first identified in the mouse and monkey [Wolfsberg et al. 1995a, 1995b]. Studies indicate that the disintegrin domain of fertilin functions in sperm-egg adhesion leading to fusion. These data indicate that a second ADAM family member, cyritestin, may function with fertilin in sperm-egg plasma membrane adhesion and fusion. [Yuan et al. 1997].

GPI-Anchored Proteins

A glycosyl phosphatidylinositol (GPI)-anchored protein may be required for fertilization. Studies have shown that removal of GPI-anchored proteins with phosphatidylinositol-specific phospholipase C (PI–PLC) results in the extensive inhibition of sperm-oocyte adhesion and fusion [Coonrod et al. 1999a, 1999b]. The GPI-anchored proteins appear to have some interaction with the bovine oocyte plasma membrane during fertilization [Pate et al. 2008]. Angiotensin-converting enzyme has been demonstrated to have GPI GPI-anchored protein releasing activity and is necessary for proper sperm penetration of the zona pellucida in mice and is localized to the acrosomal region of mouse sperm [Saha et al. 2000]. A role for this protein in bovine fertilization is being determined.

The specific identity of a disintegrin, ADAM, or other RGD containing protein on the sperm inner acrosomal membrane is still to be determined. The identification of sperm membrane proteins that bind to the vitelline membrane protein should lead to an understanding of intracellular signals, processes, and pathways that are involved in bovine oocyte activation. Identification of sperm ligands and intracellular signaling molecules will potentially increase the efficiency of in vitro embryo production by nuclear transfer, increase efficiency of ICSI, or help in the reduction of species in which overpopulation is a concern. Intracellular signaling pathways will help to clarify and confirm previously published data [Campbell et al. 2000; Sessions et al. 2006] that has demonstrated receptor-mediated oocyte activation in the bovine model.

Intracellular Signaling Pathways

When integrins bind to form cell-matrix or cell-cell interactions they cluster together. As the integrins cluster on the plasma membrane to initiate a signal, a direct consequence of this process is the recruitment of a cytoplasmic tyrosine kinase (CTK) referred to as focal adhesion kinase (FAK). The association of integrins to various different intracellular elements results in the recruitment and clustering of FAK enzymes following binding to their ligands; making FAK an important component of the integrin-regulated signaling mechanism. These two proteins may have a role in communication across cell-matrix and cell-cell junctions. FAK is known to be involved in the regulation of N-cadherin-based cell-cell adhesion [Schaller 2004; Yano et al. 2004]. FAK molecules cross-phosphorylate each other on tyrosine residues which act as attachment sites for various CTKs from the Src family; the Src family kinases are responsible for phosphorylating other tyrosines on FAK as well as other proteins that have been recruited to the focal adhesion, thereby activating them. FAK is considered to be a regulator of focal adhesions and once activated, initiate many intracellular signaling pathways [Parsons et al. 2000].

It is hypothesized that integrins mediate sperm-oocyte interactions and thus a variety of CTKs, including FAK and the Src family, are implicated for a possible role in oocyte activation. Tyrosine kinase involvement in oocyte activation pathways has also been detected in mouse oocytes [Mori et al. 1992], pig oocytes [Kim et al. 1999], and Xenopus eggs [Abassi et al. 2000; Abassi and Foltz 1994; Moore and Kinsey 1995]. Although the majority of results to date indicate that one or more tyrosine kinases are involved, it is not yet unclear which specific kinase is involved, and their complete role in mammalian fertilization remains to be elucidated.

The largest family of cell-surface receptors in eukaryotes is G-protein-linked receptors. When extracellular signaling molecules bind receptors, the receptors undergo a conformational change that activates G-proteins. G-proteins are trimers composed of α, β, and γ subunits. There are several known isoforms of alpha subunits, which are used to classify the various G-protein signaling trimers. Activation of a G-protein follows induction of the α subunit to exchange a bound GDP molecule for a GTP molecule resulting in a production of an α subunit and a βγ subunit. Each type of α subunit and each βγ subunit can act as a signaling molecule, targeting specific enzymes. G_s α can activate Ca²⁺ channels, while G_o βγ can inactivate Ca²⁺ channels. Several subunits can also activate phospholipase isoforms. In 1994 it was reported that injection of guanosine 5′–0–(2-thiodiphosphate) (GDPβ[S]), a G-protein antagonist, into fertilized rabbit oocytes resulted in inhibition of intracellular Ca²⁺ oscillations [Fissore and Robl 1994]. GDPβ[S] is a non-hydrolizable GDP analog that competitively inhibits G-protein activation by GTP. It was hypothesized that G-proteins were possibly involved in the production of IP3 [Fissore and Robl 1994]. Acetylcho-line, known to interact with plasma membrane-coupled G-protein receptors [Williams et al. 1992], and injection of GTPγ[S], an activator of G-proteins [Swann and Whitaker 1990], elicits Ca²⁺_i oscillations. An exogenously added rat M1 muscarinic receptor mediated porcine oocyte activation by a G protein coupled signal transduction pathway and led to oocyte activation [Kim et al. 1998]. The Gβγ subunit is responsible for the modulation of IP3 binding to IP3 receptors (IP3R) and stabilizes IP3Rs in a channel conformation [Zeng et al. 2003]. Zeng et al. [2003] suggested Gβγ as an alternative to IP3 in activating IP3R. Our preliminary data also indicate that G-proteins are functional in bovine oocyte development, as oocytes microinjected with the GDPβ[S] inhibitor do not cleave as often as control groups. More specific inhibitors of G-protein sub-units can be used to determine which subunits are specifically involved in fertilization pathways.

We recently published data that implicated tyrosine kinases, particularly Src-family kinases and focal adhesion kinase (FAK) in bovine oocyte activation. Of particular note is the involvement of FAK because of its relationship to integrin signaling pathways. FAK levels were reported to be relatively high in mature bovine oocytes compared to 2, 4, or 8-cell embryos. FAK levels then rise sharply during morula and blastocyst stages as would be expected as cell-cell interactions become necessary for differentiation of blastomeres. The fact that FAK levels are high in the mature oocyte could be an indication that they are necessary for fertilization. The requirement for FAK in bovine oocyte activation was demonstrated by using a dominant competitive inhibitor (FRNK) of FAK to block release of intracellular calcium (Table 1; [Pate et al. 2009]). Pretreatment of bovine oocytes with FRNK at a final concentration of 12 μM (estimated IC₅₀ = 120 pM) completely blocked calcium transients following fertilization by sperm. Further evidence of the involvement of FAK in the activation mechanism was observed following pretreatment of bovine oocytes with specific antibody directed against FAK which also inhibited calcium transients in fertilized oocytes to unfertilized control levels [Pate et al. 2009]. The apparent involvement of FAK, as well as other tyrosine kinases, specifically from the Src family, is consistent with the integrin-mediated oocyte activation hypothesis.

Table 1.  Involvement of FAK: Focal Adhesion Kinase; PLC: phospholipase C; and SRC: proto-oncogenic tyrosine kinase in Induction of Activation-Associated Calcium Transients After Fertilization by Sperm. FRNK: Focal Adhesion-Related Non-Kinase; PP1: 1 –(1,1 Dimethylethyl)–1–(4–methylphenyl)-1 H-pyrazolo[3, 4–d]-pyrimidin-4-amine.

Inhibiting Agent Microinjected into Oocytes	Concentration	Percentage of Oocytes Failing to Elicit Calcium Response
FRNK (Dominant-negative inhibitor of FAK)	1.2 pM	28%
	120 pM	60%
	12 μM	96%
U73122 (Phospholipase C inhibitor)	40 nM	47%
	10 μM	72%
	40 μM	100%
Monoclonal Anti- FAK Antibody	1:10,000 dilution	36%
	1:100 dilution	88%
PP1 (SRC family kinase inhibitor)	180 nM	26%
	360 nM	50%
	540 nM	74%

CONCLUSIONS

A significant amount of data has been generated to elucidate the specific molecules that are involved in interactions between gametes of many species. Most of this data was collected using mouse, Xenopus, and ascidian models. Many of the findings have been contradictory, and there is certainly no clear-cut demonstration that specific molecules are or are not involved in sperm-oocyte binding, fusion, activation, or development. Recent findings demonstrate the involvement of receptors that are known to associate in large, multimeric complexes. This fact partially explains why it has been so difficult to specifically identify molecules and their particular role. It also leads one to speculate that the process of oocyte activation by the sperm cell is a highly complex and elaborate process that involves many more players than perhaps was initially expected. Investigation of the bovine sperm proteome should reveal putative binding molecules and the biological activity of these ligands could be identified, shedding light on this critical mechanism.

Declaration of Interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.