Keywords

8.1 Introduction

Each human female is born with a population of oocytes estimated to range from several hundred thousand to several million (Wallace and Kelsey 2010). Although this may seem like a large number, it pales in comparison to the millions of sperm produced daily by a young adult male (Oatley and Brinster 2012). Moreover, although some controversy remains, the weight of available evidence indicates that, unlike males, female mammals do not possess a population of germ-line stem cells that give rise to new oocytes postnatally (Eggan et al. 2006; Lei and Spradling 2013; Zhang et al. 2014a; Zhang et al. 2015). Oocytes are thus relatively rare, and each presents a precious reproductive opportunity. It is not surprising that each has been allocated its own structural and developmental niche—a follicle. Each follicle contains a single oocyte together with somatic granulosa cells, which are specialized companion cells that surround the oocyte and provide support to it as well as producing steroid hormones, and an outer layer of thecal cells that serve crucial roles in steroid synthesis and in ovulation. The follicular compartment functions as a single physiological unit whose purpose is to produce a healthy egg that upon ovulation can be fertilized and will give rise to a new organism, thus enabling the female germ cell to fulfill its reproductive potential. Here, based mainly on findings using the murine model, we will discuss the cellular interactions that enable the follicle to achieve this goal, focusing on the communication between the germ and somatic cells,

8.2 Overview of Oogenesis and Folliculogenesis

During embryonic development in the female, oogonia proliferate to generate a population of germ cells. These synchronously enter meiosis and progress to diplotene of prophase I where they become arrested (Borum 1961). By the time of birth, somatic pre-granulosa cells within the ovary have surrounded oocytes to generate primordial follicles, each consisting of one oocyte (~15 μm diameter in the mouse) enclosed by a single layer of squamous granulosa cells (Findlay et al. 2015; Grive and Freiman 2015; Jorgensen 2013; Pepling 2012). However, for reasons not yet fully understood, a significant proportion of the oocyte population is lost by apoptosis during late embryogenesis (Ene et al. 2013; Hutt 2015; Malki et al. 2014), reducing the number of primordial follicles that form or remain at birth. Primordial follicles thus constitute the ovarian reserve pool, from which follicles are continually recruited during reproductive life into the growth phase to produce fertilizable eggs. Recruitment progressively depletes the ovarian reserve and when, as a consequence of aging, disease, or environmental toxins, the supply of primordial follicles drops below a functional threshold, the female is no longer fertile.

When a primordial follicle enters the growth phase, the enclosed oocyte begins to increase in size. The granulosa cells undergo a change in shape and proliferate mitotically so that they continue to fully enclose the surface of the expanding oocyte. Such follicles, consisting of a newly growing oocyte surrounded by a single layer of granulosa cells, are termed primary. During this stage, the oocyte begins to secrete a family of glycoproteins that assemble into an extracellular coat, termed the zona pellucida, which physically separates the oocyte from the bodies of the adjacent granulosa cells. As the granulosa cells continue to proliferate, they generate multiple layers around the growing oocyte. In addition, thecal cells become deposited around the outside of the basement membrane surrounding the follicle. These follicles, now termed secondary, also begin to acquire their own vasculature (Suzuki et al. 1998). Eventually, a fluid-filled cavity termed the antrum appears within the follicle, defining it as antral. Once the antrum expands, it separates the granulosa cells into two populations—the mural granulosa cells that line the inside of the follicular wall and the cumulus granulosa cells that surround the oocyte (Fig. 8.1). Near the time of antral formation, the oocyte, having increased >100-fold in volume and accumulated enormous reserves of transcripts, proteins, and organelles needed to sustain early embryo development, stops increasing in size. The oocyte nonetheless continues to undergo molecular differentiation and it is only at the late antral stage that it acquires the ability—known as developmental competence—to develop as an embryo following ovulation and fertilization.

Fig. 8.1
figure 1

Postnatal oocyte development. Upper: Oocyte and follicular growth. Oocytes in primordial follicles are surrounded by a small number of squamous granulosa cells. Incompletely understood signals cause primordial follicles to transition to the primary stage, characterized by a growing oocyte enclosed by cuboidal, mitotically active granulosa cells. When multiple granulosa cell layers have been generated and a basement membrane and thecal cell layer have been recruited, the follicles are termed secondary. The appearance of a small cavity, termed the antrum, within the granulosa cell layer defines antral follicles; oocytes at this stage are (nearly) fully grown. As the antrum enlarges, the mural and cumulus granulosa cell populations become clearly separated, and the fully grown Graafian follicle is ready to ovulate. Lower: Meiotic maturation. Upon the initiation of maturation, the nucleus (germinal vesicle) breaks down and the chromosomes condense and migrate to the periphery of the cell where they become assembled on the first meiotic spindle. Half of the products of meiosis I are discarded in the first polar body, and the oocyte chromosomes are assembled on the second meiotic spindle in preparation for meiosis II, which will occur after fertilization. The maturing oocyte also undergoes many cytoplasmic changes, not illustrated. During maturation, in a process termed cumulus expansion, the cumulus cells secrete proteins that generate an extracellular matrix that separates them from each other and from the oocyte and facilitates fertilization

Full size image

The final stage of oocyte development is meiotic maturation. Maturation is physiologically triggered by the preovulatory surge of luteinizing hormone (LH) that also triggers cumulus expansion and ovulation, thus coordinating the completion of oocyte development with its deposition into the reproductive tract to be fertilized (Eppig 1979; Espey 1980; Meinecke and Meinecke-Tillmann 1979). During maturation, the cell cycle, which has been arrested at late prophase I throughout growth, resumes and in most mammals, the oocyte completes meiosis I before becoming arrested at metaphase of meiosis II (Fig. 8.1). In addition to these nuclear events, many cytoplasmic changes also occur during maturation. These include changes in the ultrastructure of the endoplasmic reticulum (Mehlmann et al. 1995; Payne and Schatten 2003), migration of the cortical granules to just underneath the plasma membrane (Liu et al. 2003; Nicosia et al. 1977), and accumulation of the mitochondria around the meiotic spindle (Dalton and Carroll 2013; Van Blerkom et al. 2003). Many previously silent mRNAs become translationally activated, whereas others become silenced and in some cases degraded (Clarke 2012; Kang and Han 2011). During maturation, physical contact between the oocyte and granulosa cells, which has persisted since primordial follicle formation, is irreversibly lost. By the end of maturation, the mature germ cell, now termed an egg, is ovulated and can be fertilized.

8.3 Intra-follicular Cell Contact and Bidirectional Communication

The ovarian follicle is among the few physiological structures in mammals whose development can be sustained for a relatively long period of time in vitro and for which the quality of the cells that grow or differentiate in vitro can be rigorously tested—in this case, by the ability of the oocyte to give rise to an embryo. Thus, the ovarian follicle is an ideal model system to study mechanisms that regulate the formation and function of complex tissues. The relatively simple structure and cellular composition of the ovarian follicle, together with the development of culture conditions that enable oocyte and follicular growth to be reproduced in vitro, have provided researchers with an unparalleled opportunity to discover how the different cell types within the follicle cooperate to produce a healthy egg. We will focus on postnatal growth and development of the oocyte and follicle. We refer readers to recent reviews for studies of pre- and perinatal stages (Findlay et al. 2015; Grive and Freiman 2015; Jorgensen 2013; Pepling 2012).

Experimental studies have revealed a dynamic bidirectional communication network that links the germ-line and somatic cells of the follicular niche and is governed by both direct cell–cell contact and paracrine signaling. This communication is absolutely indispensable for the production of a healthy oocyte. When granulosa–oocyte complexes at an early stage of growth are removed from the ovarian follicle and placed in culture, the oocyte can complete normal development, as witnessed by its ability to give rise to offspring following fertilization (Eppig and Schroeder 1989; Hasegawa et al. 2006; Yamamoto et al. 1999). This remarkable result demonstrates that, with the possible exception of the earliest stages of growth, which have not been tested, the granulosa cells alone are sufficient to provide all of the external support needed by the oocyte to acquire developmental competence. If the granulosa cells are removed, however, the oocyte fails to grow, even when the two cell types are cocultured. Thus, physical contact with the granulosa is essential for oocyte growth.

Because the zona pellucida becomes assembled around the oocyte at an early stage of growth, thereby physically separating it from the bodies of the granulosa cells, an adaptation is required to allow the two cellular compartments to remain in physical contact. This is achieved by structures termed transzonal projections (TZPs). These narrow cytoplasmic fingers extend from the granulosa cells to the oocyte plasma membrane. TZPs link the surrounding follicle cells to the oocyte in most, and possibly all, mammalian species (Anderson and Albertini 1976; Hertig and Adams 1967), and similar structures have been observed in numerous nonmammalian species (Kessel et al. 1985; Schroeder 1981). Most TZPs contain an actin backbone, but a small fraction contains microtubules instead (Albertini and Rider 1994; De Smedt and Szollosi 1991). At the tips of the TZPs, where they contact the oocyte plasma membrane, are located gap junctions (which may also link TZPs to each other) (Anderson and Albertini 1976; Albertini and Rider 1994). These intercellular channels, composed of proteins known as connexins, permit the passage of molecules up to about 1 kDa in size (Kidder and Mhawi 2002). Connexin37 is the principal gap junction component in oocytes, and deletion of its encoding gene, Gja4, completely blocks gap junctional communication between the two cell types. As a result, oocytes of Gja4 −/− mice fail to complete growth and do not become competent to develop as embryos (Simon et al. 1997). Thus, oocyte development depends crucially on gap junctional communication with the surrounding follicular granulosa cells.

On the other hand, oocyte-derived paracrine factors (ODPFs) regulate a wide range of granulosa cell activities (Emori and Sugiura 2014). For instance, in the absence of the growth-differentiation factor 9 (GDF9), a member of the transforming growth factor β (TGF β) family provided by the oocyte, the granulosa cells exhibit numerous abnormalities including a failure to proliferate to generate multilayered secondary follicles (Carabatsos et al. 1998; Dong et al. 1996). They also are unable to support normal oocyte development. These observations collectively underscore the essential nature of the bidirectional interactions between the somatic and germ-line compartments during folliculogenesis. By coordinating oocyte and follicular development from the time of growth initiation until the final stages of development, this communication network ensures that ovulation releases a developmentally competent egg. In the following, we discuss how these bidirectional interactions govern the various stages of follicle development.

8.4 Initiation of Oocyte Growth

At the time of primordial follicle formation, the association of the small oocytes with the somatic granulosa cells protects them from undergoing cell death (Pepling and Spradling 2001; Pesce et al. 1997; Rajah et al. 1992). Thus, interactions between the germ-line and somatic compartments of the follicle are indispensable even at the earliest stages of folliculogenesis, and indeed presage an ongoing interaction that is essential to yield a fertilizable egg. The first indication that a primordial follicle has been activated to begin growth is a morphological transition of the granulosa cells from a squamous to cuboidal morphology (Braw-Tal 2002; Hirshfield 1991). The external signals (if they exist) that initiate this transition remain unknown; however, it is associated with the increase in mitotic activity of the cells that is essential to enable them to fully enclose the growing oocyte (Da Silva-Buttkus et al. 2008). This morphological transition is apparently not indispensable for the initiation of oocyte growth though, because the oocytes of mice lacking forkhead box protein L2 (Foxl2) begin to grow even though the granulosa cells remain squamous (Schmidt et al. 2004; Uda et al. 2004). These oocytes grow only modestly before becoming arrested, however, suggesting that maintaining oocyte growth depends on this transition and/or continued proliferation of the granulosa cells.

The early morphological change observed in the granulosa cells of newly growing follicles suggests that follicular activation may initiate in this cellular compartment, and recent work supports this model. Mechanistic target of rapamycin complex-1 (mTORC1) regulates growth of many cell types, typically by increasing the rate of protein synthesis (Brunn et al. 1997). When Raptor (Rptor), which encodes an essential component of mTORC1, was selectively deleted in granulosa cells, the primordial follicles did not initiate growth (Zhang et al. 2014b). This was not due to cell death, as the follicles remained intact within the ovary for several weeks, although they subsequently degenerated. Conversely, when tuberous sclerosis 1 (Tsc1), which encodes an inhibitor of mTORC1 signaling, was deleted, the granulosa cells of most primordial follicles became cuboidal and the enclosed oocytes grew robustly (Zhang et al. 2014b). These results suggest that an increase in protein synthesis within the granulosa cells could be the signal that initiates oocyte growth (Fig. 8.2).

Fig. 8.2
figure 2

Oocyte–granulosa communication and the initiation of growth. In a primordial follicle, TSC1/2 are active in granulosa cells, inhibiting mTORC1 activity and therefore limiting protein synthesis. In the oocyte, the transcription factors, FOXO3A and LHX8, prevent the oocytes from growing, possibly by inhibiting PI3-kinase activity. Signals that increase mTORC1 activity and therefore protein synthesis in granulosa cells are associated with a squamous-to-cuboidal morphological transition, mitotic proliferation, and increased production of KITL. This activates KIT on the oocyte plasma membrane, leading to increased protein synthesis. FOXO3A becomes excluded from the nucleus and activity of other transcription factors including LHX8 may be altered. Other growth factors from the granulosa cells may also be able to initiate oocyte growth, and triggering oocyte growth can also stimulate the morphological transition and mitotic proliferation of the granulosa cell (dotted lines)

Full size image

Several lines of evidence implicate KIT ligand (KITL) as a key growth-promoting signal for the oocyte. KITL is expressed by developing granulosa cells and binds to its receptor KIT on the oocyte, causing its phosphorylation (Manova et al. 1993) and activation of downstream signaling via the phosphatidylinositide (PI) 3-kinase pathway (Liu et al. 2006; Manova et al. 1993; Reddy et al. 2005, 2008). Early studies showed that adding KITL to cultures of ovarian fragments obtained from early postnatal animals induced follicle growth whereas an anti-KIT antibody known to block its function, or chemical inhibitors of KIT, reduced the number of growing follicles (Packer et al. 1994; Parrott and Skinner 1999; Yoshida et al. 1997). Similarly, fibroblasts expressing membrane-associated KITL induce growth of oocytes in the absence of granulosa cells (Thomas et al. 2008). Consistent with these results, deletion of Tsc1 in granulosa cells of primordial follicles (which activates oocyte growth—see above) is associated with a substantial increase in KITL expression. Conversely, deletion within the oocyte of genes encoding negative regulators of PI3-kinase signaling, including phosphatase and tensin homolog (Pten), Tsc1, and Tsc2, is sufficient to cause the oocytes of most primordial follicles to begin growth (Adhikari et al. 2009, 2010).

Despite this considerable evidence favoring a key role for KIT signaling in the initiation of oocyte growth, the picture is not entirely clear. Mice carrying an Y719F mutation in KIT are unable to signal through the PI3-kinase pathway (John et al. 2009). Although females are subfertile, a small number of oocytes grow and give rise to apparently normal offspring (John et al. 2008; Kissel et al. 2000; Zhang et al. 2014b). This result seems to clearly indicate that oocyte growth does not absolutely require KIT activity or, by implication, KITL produced by the granulosa cells. Numerous other growth factors have been implicated in the activation of primordial follicles, including leukemia inhibiting factor (LIF), nerve growth factor (NGF), basic fibroblast growth factor (bFGF), keratinocyte growth factor (KGF), platelet-derived growth factor (PDGF), bone morphogenetic proteins (BMP), and connective tissue growth factor (CTGF) (Dissen et al. 2001; Kezele et al. 2005; Lee et al. 2001; Nilsson and Skinner 2001; Nilsson et al. 2006; Schindler et al. 2010). One possibility is that activation of KIT signaling by KITL plays an important role under normal conditions but that, when this pathway is disabled, other signals initiate oocyte growth. It is also worth noting that the relative importance of different growth factors may vary among species (Gilchrist et al. 2008).

Although the evidence described above favors granulosa cell-derived signals as initiators of primordial follicle activation and oocyte growth, several intriguing observations suggest that this is not an exclusive mechanism. Within the oocyte, the transcriptional regulator, forkhead box O 3A (FOXO3A), plays a central role in regulating the initiation of growth. FOXO3A is predominantly nuclear in oocytes within primordial follicles, but becomes translocated to the cytoplasm in growing oocytes (Li et al. 2010). As FOXO3A is thought to function primarily as a transcriptional repressor, this result suggests that it represses genes whose products promote oocyte growth (Brunet et al. 1999). In females lacking Foxo3a, the oocytes within virtually all primordial follicles begin to grow shortly after birth (Castrillon et al. 2003; Hosaka et al. 2004). In these follicles, the granulosa cells surrounding these oocytes become cuboidal and proliferate. Similarly, when the suppressors of the PI3-kinase pathway are genetically deleted in the oocyte, as described above, not only do the oocytes begin to grow, but the granulosa cells again become cuboidal and proliferate (Castrillon et al. 2003; Hosaka et al. 2004; John et al. 2008). Thus, it seems that activation of primordial follicles can be triggered by signals originating either in the granulosa cells or in the oocyte and that each cell type is capable of responding appropriately to signals received from the other so that oocyte and follicular growth are initiated in a coordinated manner (Fig. 8.2).

Other transcriptional regulators also play key roles in regulating the initiation of oocyte growth. LIM homeobox gene 8 (Lhx8) encodes a transcription factor that is abundantly expressed in oocytes in primordial follicles, as well as at later stages. When Lhx8 was globally deleted, oocytes within primordial follicles failed to grow, suggesting an important role in oocyte growth (Choi et al. 2008a; Pangas et al. 2006). Recently, however, transgenic mice expressing Cre recombinase under the control of the promoter of Gdf9 were used to selectively inactivate Lhx8 in oocytes within primordial follicles (Ren et al. 2015). Many oocytes began to grow shortly after birth, accompanied by an increase in phosphorylated protein kinase B (AKT), indicating that PI3-kinase signaling was active and that FOXO3A translocated from the nucleus to the cytoplasm. Strikingly, however, the granulosa cells failed to undergo the squamous-cuboidal transition and the oocytes did not grow beyond ~30 μm in diameter (Ren et al. 2015). These results not only suggest that that LHX8 normally acts to maintain oocytes in a quiescent nongrowing condition, but may hint that signaling pathways that can relay the growth signal from the oocyte to the granulosa are dependent on LHX8. The spermatogenesis- and oogenesis-specific basic helix-loop-helix 1/2 (SOHLH1/2) proteins may play a similar role. These are abundant in oocytes of primordial follicles, but less detectable in growing oocytes. When either Sohlh1 or Sohlh2 is deleted, there is a rapid loss of oocytes and follicles. Some oocytes begin to grow, but the granulosa cells remain squamous (Choi et al. 2008b; Pangas et al. 2006), likely explaining why oocyte growth fails to progress. Taken together, the studies demonstrate that the entry of primordial follicles into the growth pool is an intricately coordinated process that can be regulated by multiple factors in both the oocyte and the granulosa cells.

8.5 Maintenance of Oocyte Growth and Follicle Development

8.5.1 Metabolic Support

The enormous increase in size of the growing oocyte makes it among the largest cells in the body. Growth is associated with an accumulation of a large stock of mRNAs and proteins, which direct late oocyte development and early embryonic development after fertilization, and of organelles such as mitochondria. This activity imposes a substantial metabolic burden on the oocyte (Collado-Fernandez et al. 2012). The oocyte does not shoulder this alone but outsources some essential tasks to the granulosa cells that surround it, allowing it to minimize its own energy expenditure (Fig. 8.3).

Fig. 8.3
figure 3

Oocyte–granulosa cell communication during growth. The oocyte is unable to efficiently produce cholesterol or pyruvate or to take up certain amino acids such as alanine. The granulosa cells produce these essential factors and transfer them to the oocyte, via gap junctions in the case of pyruvate and amino acids. Conversely, the oocyte secretes several growth factors that are required for granulosa cell proliferation and/or differentiation. GDF9 and BMP15 may be presented to the granulosa cells as homo- or heterodimers

Full size image

Follicular growth is associated with an increased consumption of glucose and oxygen, both in vivo (Harris et al. 2009) and in vitro (Boland et al. 1994; Harris et al. 2007). Many studies have shown that mammalian oocytes, including human (Tsutsumi et al. 1990) and mouse (Brinster 1971; Eppig 1976; Fagbohun and Downs 1992; Tsutsumi et al. 1992; Zuelke and Brackett 1992), cannot efficiently metabolize glucose to generate pyruvate. Indeed, oocytes are relatively deficient in expression of mRNAs encoding glycolytic enzymes including Idoa, Eno1, Ldha, Pfkp, Pkm2, and Tpi1, whereas these genes are highly expressed by surrounding granulosa cells (Sugiura et al. 2005; Sugiura et al. 2007). Granulosa cells metabolize glucose and then transfer pyruvate to the oocyte (Brinster 1971; Tsutsumi et al. 1990). Nevertheless, one study traced the uptake of glucose by granulosa cells via glucose transporters and its transfer into the oocyte through gap junctions (Wang et al. 2012). Glucose transferred to the oocyte may be metabolized by non-glycolytic pathways such as the pentose phosphate or the hexosamine biosynthesis pathways (Collado-Fernandez et al. 2012). Hence, the oocyte relies on the somatic compartment to obtain its source of ATP. As reduced ATP content in the oocyte has been linked to defects of the meiotic spindle (Zhang et al. 2006), as well as diminished developmental competence (Igarashi et al. 2005; Van Blerkom et al. 1995), this cooperation between the oocyte and granulosa cells is essential for normal embryo development and fertility.

The protein synthetic activity of the oocyte evidently depends on a large supply of amino acids. But the oocyte is only poorly able to take up certain amino acids including alanine, glycine, and proline in vitro. Uptake significantly improves in the presence of surrounding granulosa cells, which transfer the amino acids through gap junctions (Colonna and Mangia 1983; Haghighat and Van Winkle 1990). Consistent with these observations, certain transporters including solute carrier (SLC) 38A3, whose substrates include histidine and alanine (Eppig et al. 2005), and SLC7A6, a cationic amino acid transporter (Corbett et al. 2014), are present only in granulosa cells. In addition, although other amino acid transport systems are active in denuded oocytes, they are functionally improved by the presence of surrounding granulosa cells (Colonna and Mangia 1983; Haghighat and Van Winkle 1990; Pelland et al. 2009). By supplying the oocyte with the amino acids for which it lacks transporters, granulosa cells thus greatly increase the efficiency of oocyte protein synthesis.

The oocyte also depends on the granulosa cells to obtain cholesterol. Many enzymes that act in the cholesterol biosynthetic pathway are expressed at relatively low levels in oocytes. Consistently, acetate is inefficiently converted to cholesterol in granulosa-free oocytes (Su et al. 2008). Oocytes also lack the ability to import cholesterol as they do not express receptors for HDL cholesterol and LDL cholesterol (Sato et al. 2003; Trigatti et al. 1999). Oocytes are thus deficient in both cholesterol synthesis and uptake of extracellular cholesterol. Instead, the surrounding granulosa cells, which express the necessary enzymes, synthesize cholesterol and transfer it to the oocyte (Su et al. 2008). Accumulation of cholesterol in oocytes is essential for normal embryo development (Comiskey and Warner 2007), as early preimplantation embryos are also unable to synthesize cholesterol (Pratt 1982). Thus, the metabolic cooperation between the granulosa cells and the oocyte, which ensures the deposition of cholesterol into the oocyte, is critical for the oocyte to acquire embryonic developmental competence.

The molecular deficiencies of the oocyte thus force it to be strictly dependent in several metabolic respects on the granulosa cells. However, the oocyte does not merely receive these metabolites from the granulosa cells. Rather, it instructs the granulosa cells to synthesize and/or supply the nutrients it needs to meet its metabolic requirements (Fig. 8.3). A clue to this was the observation that expression of several enzymes of the glycolytic pathway (Emori et al. 2013; Sugiura et al. 2005; Sugiura et al. 2007) and that of several amino acid transporters (Emori et al. 2013; Eppig et al. 2005) are all enriched in the cumulus cells adjacent to the oocyte as compared to mural granulosa on the inner wall of the follicle. It was then discovered that their expression, as well as that of several enzymes in the cholesterol biosynthetic pathway, is enhanced by oocytes or ODPFs, including GDF9 and BMP15 (Emori et al. 2013; Eppig et al. 2005; Nakamura et al. 2015; Sugiura et al. 2005, 2007; Su et al. 2008). These observations highlight the dynamic and bidirectional nature of the signaling between the oocyte and granulosa cells that allow the growing oocyte to meet its high metabolic demands.

8.5.2 Estradiol Synthesis

Steroidogenesis occurs in the adrenal gland, ovary, testis, and placenta. In the ovary, this process has long been considered as an example of intercellular cooperation, as exemplified by the “two-cell, two-gonadotropin” model describing cooperation between the granulosa and theca cells (Fortune and Armstrong 1977). Steroidogenesis involves the uptake and metabolism of cholesterol by steroidogenic enzymes and is driven by follicle-stimulating hormone (FSH) and LH (Conley et al. 1994; Parker and Schimmer 1995; Richards 1994; Richards et al. 2002; Tian et al. 1995). Because theca cells do not express the FSH receptor (Camp et al. 1991; Zeleznik et al. 1974), the theca–granulosa cooperation is necessary for estrogen synthesis.

Steroidogenesis is initiated by the import of cholesterol into mitochondria by steroidogenic acute regulatory protein (StAR) (Arakane et al. 1996; Clark et al. 1994; Stocco 2001). Cholesterol is then cleaved by the mitochondrial enzyme, cytochrome P450 (CYP) side-chain cleavage (CYP11A) producing pregnenolone. CYP17, 3-β-hydroxysteroid dehydrogenase (HSD), and 17β-HSD metabolize pregnenolone to produce androgens. These are then converted by aromatase (CYP19) to estradiol (Miller 1988; Peter and Dubuis 2000; Simpson et al. 1994). LH, whose receptor is on theca as well as granulosa cells (Camp et al. 1991; Erickson et al. 1979; Piquette et al. 1991; Segaloff et al. 1990; Zeleznik et al. 1974), drives androgen synthesis (Fortune and Armstrong 1977; Richards et al. 1987; Smyth et al. 1994), while FSH stimulates aromatase activity in granulosa cells (Dorrington et al. 1975; Erickson et al. 1979; Hickey et al. 1988; Fitzpatrick and Richards 1991; Whitelaw et al. 1992), enabling the production of estrogen from the theca-derived androgens. In combination with other follicular factors such as insulin-like growth factor 1 (IGF-1), LH (Devoto et al. 1999; McGee et al. 1996; Schoppee et al. 2002; Sekar et al. 2000; Simone et al. 1993; Willis et al. 1996; Zhang et al. 2000) and FSH (Adashi 1994; Balasubramanian et al. 1997; Glister et al. 2001; Hsu and Hammond 1987; Khalid et al. 2000; Monniaux and Pisselet 1992; Pescador et al. 1997; Silva and Price 2002; Spicer et al. 2002; Yoshimura 1998) increases expression of these steroidogenic enzymes in the theca and the granulosa cells, thus promoting androgen and estrogen synthesis, respectively, both in vivo and in vitro.

Hence, the combined actions of theca cells and granulosa cells enable steroidogenesis in the developing ovarian follicle. Recent findings suggest, however, that ovarian steroidogenesis is regulated by three, rather than two, follicular cell types. The removal of an oocyte from a granulosa–oocyte complex in vitro by oocytectomy leads to reduction in estradiol levels, suggesting that the oocyte promotes its production (Vanderhyden et al. 1993). Oocytes have been shown to enhance estradiol production in nonmammalian species as well (Sretarugsa and Wallace 1997; Yoshimura et al. 1994). In contrast, other studies have shown that oocyte signaling decreases the expression of steroidogenic enzymes (Diaz et al. 2007; Pangas et al. 2006; Wigglesworth et al. 2015) and suppresses FSH-stimulated estradiol synthesis (Glister et al. 2001; Vitt et al. 2000). These seemingly contradictory results may indicate that oocyte signaling promotes estradiol synthesis while also inhibiting the premature luteinization of the granulosa cells (Coskun et al. 1995; el-Fouly et al. 1970; Nekola and Nalbandov 1971) through independent pathways (Vanderhyden et al. 1993; Vanderhyden and Tonary 1995). The mechanism of how the oocyte achieves such a balanced effect remains to be elucidated.

The follicle itself is a target of the estradiol, as well as of the androgens, that it produces (Berisha et al. 2002; LaVoie et al. 2002; Rosenfeld et al. 2001; Walters et al. 2012). Estradiol drives folliculogenesis (Couse et al. 2005; Emmen et al. 2005; Krege et al. 1998) and prevents follicular atresia (Britt et al. 2000; Dupont et al. 2000; Richards 1980; Robker and Richards 1998; Rosenfeld et al. 2001). It promotes the survival of granulosa cells as well as the oocyte by inhibiting the expression of pro-apoptotic genes (Billig et al. 1993; Lund et al. 1999; Quirk et al. 2006; Toda et al. 2001). It also increases expression of IGF-1, thus promoting expression of FSH and LH targets within the follicle (Hsu and Hammond 1987; Khalid et al. 2000). Moreover, estradiol has also been shown to enhance the expression of natriuretic peptide receptor B (NPR2; see below), thus enabling the maintenance of meiotic arrest (Zhang et al. 2011). Hence, the cooperation between the various follicular compartments, resulting in the production of estradiol, is crucial for survival and development of the follicle as a whole.

8.6 Terminal Stages of Oocyte Growth and Development

8.6.1 Nuclear Differentiation of the Oocyte

Growing oocytes are highly active transcriptionally (Bachvarova 1985), likely reflecting that they must manufacture not only the RNA needed to sustain growth itself but also the RNA needed during the transcriptionally inactive periods of meiotic maturation and early embryonic development. In the mouse, the oocyte RNA content increases 300-fold during growth (Schultz et al. 1979; Sternlicht and Schultz 1981). Upon reaching full size, however, the oocyte dramatically reduces its transcriptional activity to an undetectable level (Bouniol-Baly et al. 1999; Christians et al. 1999; Worrad et al. 1994). Coincident with transcriptional arrest, the chromatin, which was previously dispersed throughout the nucleus in a configuration termed non-surrounded nucleolus (NSN), becomes partially condensed and prominently arrayed around the nucleolus in a configuration termed surrounded nucleolus (SN) (Mattson and Albertini 1990; Wickramasinghe et al. 1991). This conformational change, termed the NSN-SN transition, occurs in oocytes of many mammalian species (Fuhrer et al. 1989; Hinrichs et al. 1993; Parfenov et al. 1989). Although the events of chromatin remodeling and transcriptional silencing are temporally correlated, they do not strictly depend on each other (Abe et al. 2010; Andreu-Vieyra et al. 2010; De La Fuente 2006). For example, the dissociation of RNA polymerase II from DNA, an indication of transcriptional silencing, is not dependent on the NSN-SN transition (Abe et al. 2010).

The NSN-SN is associated with a number of events of late oocyte development, including a change in the nuclear concentration of the transcription factors, specificity protein 1 (SP1) and TATA box-binding protein (TBP), and a rearrangement of cytoplasmic components including the appearance of microtubule-organizing centers (Lodde et al. 2008; Mattson and Albertini 1990; Wickramasinghe et al. 1991). Hence, it is considered to be a reliable and consistent marker of oocyte developmental progression and generally strongly correlates with the ability of the oocyte to develop as an embryo following fertilization (Wickramasinghe et al. 1991; Zuccotti et al. 2002). Intriguingly, the granulosa cells play an important but not fully defined role in mediating the NSN-SN transition (De La Fuente 2006; Luciano et al. 2011). When the FSH analogue, equine chorionic gonadotropin, was injected into females and the cumulus cell–oocyte complexes recovered 2 days later from antral follicles, an increased number had undergone the NSN-SN transition and had ceased transcriptional activity as compared to oocytes obtained from non-injected females. This effect was not observed in oocytes that were not closely associated with cumulus cells at the time of recovery. Similarly, incubation of bovine cumulus–oocyte complexes in the presence of FSH promoted the NSN-SN transition together with increased gap junctional communication. As the FSH receptors are located on the granulosa cells, these observations imply that the granulosa send signals to the oocyte that promote its nuclear differentiation.

8.6.2 Maintenance of Meiotic Arrest

Throughout its growth, the oocyte remains arrested at late prophase I of the meiotic cell cycle. Upon the initiation of meiotic maturation, it completes the first meiotic division and then becomes arrested at metaphase of meiosis II. Physiologically, maturation is triggered by LH (Dekel et al. 1979; Eppig 1980), thereby coordinating it with ovulation. However, experiments in which oocytes of antral follicles are isolated and placed in culture showed that they acquire the ability to mature to metaphase II about midway through growth, which is well before the follicle acquires the ability to ovulate in response to LH (Sorensen and Wassarman 1976). This result established the concept that the follicular environment actively prevents oocytes from undergoing maturation until this inhibition is relieved by LH. The granulosa cells play a central role in mediating this inhibition.

Maturation is triggered by a protein complex, known as metaphase-promoting factor (MPF), that consists of a catalytic subunit, cyclin-dependent kinase 1 (CDK1), and a regulatory subunit, cyclin B1 (Brunet and Maro 2005). The activity of this complex is regulated by phosphorylation of specific sites on CDK1. The WEE1/MYT1 kinases phosphorylate two sites, which inhibits CDK1 activity, whereas the CDC25 phosphatases dephosphorylate these sites, thereby activating CDK1 (Lew and Kornbluth 1996; Lincoln et al. 2002). The relative activities of WEE/MYT1 and CDC25 thus control CDK1 activity. These two regulators are themselves controlled by phosphorylation. Phosphorylation by cyclic (c) AMP-dependent kinase A activates WEE1 and inhibits CDC25 (Han et al. 2005; Oh et al. 2010; Pirino et al. 2009; Zhang et al. 2008). Hence, high protein kinase A activity maintains CDK1 in an inactive state.

As would be anticipated based on the foregoing, cAMP levels have long been known to be high in follicle-enclosed oocytes and to drop rapidly when oocytes are removed from the follicle and placed in culture (Schultz et al. 1983). This suggests that the follicular environment prevents oocyte maturation by maintaining a high level of cAMP within the oocyte (Cho et al. 1974; Dekel and Beers 1978). Because the growth of antral follicles is promoted by FSH, whose receptors are located on the granulosa cells, and gonadotropins can increase intracellular cAMP, an attractive hypothesis was that the granulosa cells transferred cAMP to the oocyte via the communicating gap junctions. Subsequent work showed, however, that the oocyte is able to synthesize cAMP (Olsiewski and Beers 1983; Urner et al. 1983) through the activity of a G protein related receptor, GPR3 (Mehlmann et al. 2002, 2004). Importantly, when Gpr3 was deleted from oocytes, they underwent maturation within the follicle in the absence of LH signaling (Mehlmann et al. 2004), providing convincing evidence that the oocyte-generated cAMP is needed to maintain high protein kinase A activity and inhibit maturation.

Although the granulosa cells may not be a primary source of cAMP for the oocyte, they nonetheless play a critical role in maintaining its meiotic arrest. cAMP in the oocyte is metabolized by phosphodiesterase (PDE)-3A (Bornslaeger et al. 1984; Dekel and Beers 1978, 1980; Jensen et al. 2002; Richard et al. 2001; Shitsukawa et al. 2001; Tsafriri et al. 1996; Vivarelli et al. 1983). Within the follicle, the granulosa cells transfer cyclic GMP, an inhibitor of PDE3A, to the oocyte, again via the gap junctions (Norris et al. 2009; Richard and Baltz 2014; Zhang et al. 2010). This helps to maintain a high steady-state level of cAMP within the oocyte. Intriguingly, the production of cGMP illustrates the cooperativity between the mural and cumulus granulosa cells. The mural cells secrete C-type natriuretic peptide C (CNP, encoded by Nppc), which binds to NPR2 located on the plasma membrane of the cumulus cells. NPR2 is associated with guanylate cyclase and thus upon its activation increases the production of cGMP (Fig. 8.4) (Franciosi et al. 2014; Kawamura et al. 2011; Zhang et al. 2010, 2011). The essential role of this cGMP is demonstrated by mice bearing mutations in either Npr2 or Nppc. Their oocytes are unable to maintain meiotic arrest and thus undergo maturation within the ovarian follicle independently of LH signaling. Thus, by supplying an inhibitor of PDE3A activity, the granulosa cells maintain a high level of cAMP in the oocyte, which activates protein kinase A and keeps CDK1 in an inactive state.

Fig. 8.4
figure 4

Oocyte–granulosa communication and the regulation of meiotic maturation. Prior to maturation, CNP secreted by the granulosa cells activates NPR2 receptors that are coupled to guanylate cyclase, thereby activating cGMP synthesis. cGMP is transferred to the oocyte via gap junctions, where it inhibits PDE3A. This allows the level of cAMP in the oocyte, generated by the activity of Gs-coupled GPR3, to remain high. Protein kinase A is therefore active and CDK1 is inactive. Upon binding of LH to its receptor, LHGCR, on the mural granulosa cells, EGFR ligands (amphiregulin, beta-cellulin, and epiregulin) are released by the mural granulosa cells and activate EGFR on both mural and cumulus granulosa cells. Signaling through LHGCR and EGFR causes cGMP levels in the granulosa to fall via phosphorylation and activation of PDE5A, dephosphorylation of NPR2, and reduced Nppc transcription and CNP production. Transcription of Areg, Btc, and Ereg are also increased. The specific contributions of LHGCR and EGFR signaling to each of these events remain to be fully defined. The net result is a loss of cGMP from the oocyte, thus relieving the inhibition of PDE3A. Shortly after the initiation of maturation, gap junctional communication between granulosa cells and oocytes is disrupted

Full size image

Once again, the oocyte plays a key role in facilitating this granulosa cell function. GDF9 and BMP15 promote the expression of Nppc and Npr2 (Lee et al. 2013; Wigglesworth et al. 2015; Zhang et al. 2010, 2011) and also enhance the expression of Impdh, an enzyme required for the production of cGMP (Wigglesworth et al. 2013). The oocyte thus plays a key role in maintaining its meiotic arrest. This cooperation between the two compartments, by preventing LH-independent meiotic maturation within the follicle, ensures that the production of a fertilizable egg is temporally coordinated with its ovulation.

8.7 Ovulation-Associated Events

8.7.1 Meiotic Maturation

Meiotic maturation, the final stage of oocyte development, is triggered by the LH surge (Lei et al. 2001; Neal and Baker 1975). The LH receptors are expressed only on the mural granulosa cells within the follicle (as well as the thecal cells on the outside of the follicle). Therefore, this signal must be relayed to the oocyte. The discovery of the central role of granulosa cell-derived cGMP in maintaining meiotic arrest enabled the mechanism by which the LH signal is relayed to be deciphered. Less than 5 min after adding LH to antral follicles in vitro, the level of cGMP in the mural granulosa cells falls (Shuhaibar et al. 2015). This is followed by a decrease in cGMP in the cumulus granulosa cells and in the oocyte. Importantly, the rapid decrease in cumulus cell and oocyte cGMP is prevented by chemical inhibitors of gap junction permeability. cGMP decreases later in the cumulus cells and oocytes even when gap junctions are not functional, suggesting that multiple mechanisms may contribute to the decrease (Shuhaibar et al. 2015).

Several processes that could contribute to the fall in cGMP have been identified. Upon LH stimulation, NPR2 becomes rapidly dephosphorylated and inactivated, thus inhibiting the synthesis of new cGMP (Egbert et al. 2014; Robinson et al. 2012). In addition, PDE5A, which degrades cGMP, becomes phosphorylated and activated (Egbert et al. 2014). This likely increases the rate of degradation of existing cGMP. Following these early events, the amount of Nppc mRNA falls, which could lead to a reduction in the amount of CNP (Kawamura et al. 2011; Lee et al. 2013; Liu et al. 2014; Robinson et al. 2012). Finally, gap junctional communication becomes impaired both between granulosa cells and between the granulosa cells and oocyte, thus preventing the transfer of cGMP into the germ cell. It has been proposed that LH triggers a direct loss of cGMP in the mural granulosa cells that is followed by a flow of cGMP from the cumulus cells and oocyte to the mural cells to equalize the concentration of cGMP through the follicle (Shuhaibar et al. 2015). Thus, gap junctional communication is required for the initial reduction in cGMP within the cumulus cells and oocyte. Later, however, the mechanisms described above may directly reduce cGMP in cumulus cells, leading to a further fall in the oocyte cGMP.

The mechanism by which LH activates the multiple pathways that reduce cGMP remains to be fully elucidated. Upon binding of LH to its receptor, the mural granulosa cells release three members of the epidermal growth factor (EGF) family—amphiregulin (AREG), beta-cellulin (BTC), and epiregulin (EREG) (Conti et al. 2012; Park et al. 2004). As both the mural and cumulus granulosa cells express EGF receptors, this suggests that LH could act by transactivating EGF receptor (EGFR) signaling (Fig. 8.4). This provides a pathway by which the LH signal received by the mural granulosa cells could be relayed to the cumulus granulosa and ultimately the oocyte, and numerous observations support a central role for EGFR signaling in maturation. First, EGF is able to induce maturation of follicle-enclosed oocytes and expansion of the cumulus layer in the absence of LH (Ashkenazi et al. 2005; Downs and Chen 2008; Park et al. 2004). Second, the EGF-like ligands released by the granulosa cells in response to LH, as well as EGF, trigger the decrease in follicular cGMP independently of LH (Hsieh et al. 2011; Norris et al. 2010). Third, in mutants carrying the hypomorphic Egfr wa2/wa2 allele or lacking Areg or Ereg, the oocytes fail to mature in response to LH (Hsieh et al. 2007, 2011).

Yet, other observations imply that LH may act independently of EGFR signaling. The extremely rapid drop in cGMP in the mural granulosa cells, which is detectable within a minute of adding LH in vitro (Shuhaibar et al. 2015), suggests that the gonadotropin acts directly on the granulosa cells. Consistent with this, LH induces a rapid drop in cGMP in the oocytes of Egfr wa2/wa2 mice, although some EGFR activity remains in this mutant (Hsieh et al. 2011). In addition, the decrease in Nppc mRNA during maturation does not appear to require EGFR activity (Liu et al. 2014). It may be that LH acts through EGFR-dependent and -independent pathways to achieve a robust and stable decrease in cGMP throughout the preovulatory follicle, thus enabling activation of PDE3A in the oocyte and the consequent resumption of meiosis.

In addition to their role in the initiation of maturation, the granulosa cells also influence the maturation process itself. It is well established that cumulus cell-free oocytes can complete nuclear maturation to metaphase II in vitro, but develop poorly as embryos as compared to oocytes that mature within cumulus–oocyte complexes (Chang et al. 2005; De La Fuente et al. 1999). This suggests that cytoplasmic events of maturation may depend on or occur more efficiently in the presence of the cumulus cells. As noted earlier, a subset of mRNAs that had previously been synthesized and stored become translationally activated during maturation (Chen et al. 2011; Paynton and Bachvarova 1994; Su et al. 2007). Among these is Tpx2, encoding a protein required for proper assembly and function of the meiotic spindle (Chen et al. 2013). Although TPX2 increases when cumulus-free oocytes are matured in vitro, this increase is greater in cumulus-enclosed oocytes than in cumulus-free oocytes and is further enhanced in the presence of AREG. Translational activation is associated with an AREG-dependent increase in the activity of mTORC (mechanistic target of rapamycin complex) in the oocyte. Further evidence of the importance of signals from the cumulus cells comes from oocytes of mice deficient in either Egfr or Areg (Chen et al. 2013). These show reduced translational activation during maturation and impaired progression to the two-cell stage following fertilization and give rise to relatively small litters. These results indicate that the EGFR-mediated increase in translational activity during meiotic maturation is developmentally important and are consistent with previous reports that oocytes matured in vitro in the presence of EGF (De La Fuente et al. 1999; Richani et al. 2014) manifested increased developmental competence. Because gap junctional communication between the oocyte and granulosa cells is lost relatively early during the maturation process, it will be important to identify when and by what pathway the granulosa cells exert their beneficial effects on maturing oocytes.

8.7.2 Cumulus Expansion

As a result of LH-driven cumulus expansion, the egg is ovulated embedded in a mucified mass of cumulus granulosa cells. Cumulus expansion facilitates transport of the cumulus–oocyte complex through the oviduct (Chen et al. 1993) and is also essential for fertilization, likely by permitting the sperm to pass through the cumulus layers to reach the oocyte (Eisenbach 1999; Salustri et al. 2004; Van Soom et al. 2002). Cumulus expansion is characterized by the physical dispersal of the cumulus granulosa cells as an extracellular matrix becomes deposited around the individual cells (Fig. 8.5). This matrix consists mostly of the glycosaminoglycan, hyaluronan (HA), as well as HA-binding proteins (Chen et al. 1996; Fulop et al. 1997a; Hess et al. 1998; Sato et al. 2001; Yoshioka et al. 2000) and proteoglycans (McArthur et al. 2000) that maintain the stability and structure of the matrix. The integrity of the mucified matrix also requires proteins such as pentraxin 3 (PTX3) and tumor necrosis factor-inducible gene 6 (TNFAIP6) (Baranova et al. 2014; Fulop et al. 2003; Ievoli et al. 2011; Salustri et al. 2004; Sanggaard et al. 2008; Scarchilli et al. 2007) which mediate the interactions between the components of the matrix. These factors, along with other proteins that are needed for cumulus expansion including Ptgs2 (prostaglandin-endoperoxide synthase 2) (Davis et al. 1999; Ochsner et al. 2003b; Sirois et al. 1992), hyaluronan synthase 2 (Has2) (Sugiura et al. 2009), and Tnfaip6 (Fulop et al. 2003; Ochsner et al. 2003a; Sugiura et al. 2009), are produced by the cumulus cells in response to LH surge (Fig. 8.5) (Eppig 1981; Fulop et al. 1997b; Salustri et al. 1999).

Fig. 8.5
figure 5

Oocyte–granulosa communication and cumulus expansion. The combined influences of EGFR activation by its ligands and SMAD signaling activated by oocyte-derived factors such as GDF9 and BMP15 trigger the increased transcription in the cumulus cells of genes whose products are secreted and assembled into a matrix. This process, termed cumulus expansion, is required for efficient fertilization

Full size image

Although cumulus expansion occurs in response to the LH stimulus, which is transduced at least in part through the EGFR pathway, it also requires a signaling input from the oocyte. If the oocyte is microsurgically removed from a cumulus–oocyte complex, the remaining oocytectomized cumulus cell shell cannot undergo expansion in response to EGFR ligands. It can, however, expand if it is cocultured with oocytes or with oocyte-conditioned medium (Buccione et al. 1990; Salustri et al. 1990; Vanderhyden et al. 1990). Many studies using the mouse have demonstrated that signaling by ODPFs is required for upregulation of expansion-related mRNAs and for cumulus expansion (Diaz et al. 2006; Dragovic et al. 2005, 2007; Joyce et al. 2001; Mazerbourg et al. 2004; Moore et al. 2003; Sasseville et al. 2010; Su et al. 2003, 2010; Vanderhyden et al. 2003). Other studies have suggested that the ODPFs are needed for the activation of mitogen-activated protein kinase (MAPK) (Su et al. 2003) and for maintaining the activity of EGFR (Su et al. 2010) in the cumulus cells. The activity of the ODPFs appears to be potentiated by heparan sulfate proteoglycans (HSPG) acting as co-receptors (Watson et al. 2012). The oocyte thus enables cumulus expansion both indirectly through promoting EGFR activity and directly by increasing the expression of expansion-related transcripts.

Although the ability of the oocyte to induce cumulus expansion in vitro is also a feature of other species such as pig and rat (Prochazka et al. 1998; Vanderhyden 1993), in the cow, sheep, and pig, oocyte signaling is not required to induce cumulus expansion (Nagyova et al. 1999; Prochazka et al. 1991; Ralph et al. 1995; Varnosfaderani et al. 2013). Interestingly, oocytes of these species are capable of inducing the expansion of mouse cumulus granulosa cells (Vanderhyden 1993). Whether signaling from the oocyte is required during cumulus expansion in humans remains to be determined. In summary, in the mouse at least, the somatic and germ-line compartments of the follicle cooperate to induce the cumulus cells to expand and mucify in response to the LH signal, thus fulfilling an essential prerequisite for fertilization.

8.8 Conclusion

As in many tissues and organs, growth and differentiation of the ovarian follicle requires a precisely regulated program of intercellular communication and cooperation. We have described several important events during ovarian follicle growth that exemplify this cooperation, specifically between the oocyte and the granulosa cells. Although the gap junction-mediated support from the granulosa to the oocyte is best understood, it is plausible that the granulosa also support the growing oocyte through gap junction-independent pathways. For example, recent work suggests that bovine granulosa cells can transfer mRNAs and long noncoding RNAs to the oocyte (Macaulay et al. 2014). Yet uncharacterized secreted or membrane-associated molecules of the granulosa may also be important. In addition to the events we have discussed, other aspects of folliculogenesis also depend on intercellular cooperation, including the transition from a primary to a secondary follicle (Braw-Tal 2002; Dong et al. 1996; Galloway et al. 2000; Parrott and Skinner 1999; Yoshida et al. 1997), formation of the TZPs (Carabatsos et al. 1998), thecal cell recruitment (Dong et al. 1996; Liu et al. 2015; Spicer et al. 2008), differentiation of the granulosa into mural and cumulus cell compartments, and granulosa cell luteinization (Eppig et al. 1997; Glister et al. 2003; Li et al. 2000; Nekola and Nalbandov 1971; Vanderhyden 1993). All of these processes serve, directly or indirectly, the growth and development of the oocyte and thereby enable the ovulation of a mature and developmentally competent egg ready to give rise to a new organism.