rep 356 Smitz - Reproduction

57 downloads 75 Views 1MB Size Report
Applications of these in vitro techniques include fertility preservation for humans, conservation of rare animals and development of oocyte banks for research ...
Reproduction (2002) 123, 185–202

Review

The earliest stages of folliculogenesis in vitro Johan E. J. Smitz and Rita G. Cortvrindt Follicle Biology Laboratory, Vrije Universiteit Brussel, Laarbeeklaan 101, B-1090 Brussels, Belgium

In recent years several follicle culture systems have been pioneered in different mammalian species for studying ovarian folliculogenesis and culturing immature oocytes. Applications of these in vitro techniques include fertility preservation for humans, conservation of rare animals and development of oocyte banks for research purposes. Immature female gametes in the ovarian cortex can be cryopreserved for later use if culture techniques are available afterwards to promote growth and maturation. This review focuses on biochemical and biophysical factors related to oocyte culture in mice, the only animal in which live offspring have been produced after folliculogenesis in vitro. The advantage of using mice for these studies is that, in parallel to development of follicle culture systems, essential knowledge on folliculogenesis can be obtained from knockout mouse models. Recent experiments in mice stressed the principal role of the oocyte in follicle development and the strict timing of the biological processes underlying oogenesis in vitro. In large domestic animals and humans, study of oocyte culture is confounded by the constitutively prolonged nature of ovarian follicle development. In humans, only some aspects of follicle development have been studied because of the limited availability of suitable material for experimentation, technical difficulties related to manipulation of very small structures and lack of knowledge on physiological regulation of the early stages of follicle growth. Only a few reports describe ovarian follicular growth in vitro. In this review, relevant information on hormonal and growth factor regulation of the earliest stages of follicle growth in mammals is reviewed. Techniques are becoming available for the precise isolation of distinct classes of follicle and powerful molecular biology techniques can be used in studies of ovarian tissue culture.

The availability of oocytes is the limiting factor in the development of new reproductive techniques. The duration of the fertile period in a female’s life is determined by the size of the primordial follicle pool formed during fetal life and by the rate of depletion of the pool after birth. Mechanisms of regulation of follicle formation, oocyte attrition and follicle development and atresia are only partially understood and more information on these processes could be obtained by culturing follicles in vitro at the earliest stages of formation and growth. In small mammals, the ovarian physiology of mice and knockout mice models has been studied extensively and culture of well-defined stages of preantral and antral follicles has provided some insight into the complex interplay between somatic and germ cells during growth and maturation. It is possible to grow female gametes fully in vitro: in rodents, live offspring have been obtained after extended

Email: [email protected]

follicle culture periods of a few weeks. However, even in mice, much effort is still required to refine the culture systems. For domestic animals and humans, the technology is still far from successful; there are only a few reports of limited success using in vitro culture of large preantral follicles that have progressed further developmentally. These larger follicles contain oocytes that are almost fully grown; however, the meiotic competence of these follicles remains low, indicating that the culture methods require optimization. As the store of female gametes is maintained within the primordial follicles, the real challenge is to establish the cultures of primordial follicles. Unravelling the regulation of the transition from primordial to primary follicle is the key to accessing an almost unlimited source of oocytes for biomedical use. This review summarizes current knowledge on mammalian folliculogenesis in vivo and discusses achievements of in vitro systems for culture of preantral follicle stages, focusing on mice and humans.

© 2002 Society for Reproduction and Fertility 1470-1626/2002

186

J. E. J. Smitz and R. G. Cortvrindt

Formation of the primordial follicle Primordial germ cells arise at about 3 weeks after fertilization and migrate by amoeboid movements from the epithelium of the yolk sac via the connective tissue of the hindgut to the genital ridge. During their journey, the primordial germ cells multiply rapidly. In humans, the genital ridge is formed at about 3.5–4.5 weeks of gestation by a knot of mesenchyme overlaid by coelomic epithelium. At about week 7 of gestation the cells derived from the mesonephros and the coelomic epithelium migrate inward from the genital ridge, forming primitive medullary cords and sex cords, respectively. The cords, which are anatomically less well defined in females than in males, are compacted in the cortical region of the gonadal ‘anlage’. The primitive cords are colonized by the primordial germ cells. In the female fetus, the primordial germ cells are named oogonia upon arrival in the primitive gonad. The oogonia are still interconnected by cytoplasmic bridges. The primitive medullary cords degenerate to be replaced by highly vascularized ovarian stroma. The cord cells proliferate and mesenchymal cells condense around the oogonia to form the individual primordial follicles. Follicle formation in humans begins in the inner part of the ovary, near the rete ovarii, between week 16 and week 18 of fetal life. In the primordial follicles, the mesenchymal cells secrete an outer basement membrane and the same cells will give rise to granulosa cells in the growing follicle. Meanwhile, the mitotic activity of the oogonia ceases and the oogonia enter meiosis (Byskov, 1986). In mammals, this process of envelopment of oogonia occurs either before birth (humans, cows, sheep, cavia) or shortly thereafter (mice, rats, hamsters). A meiosisinitiating factor is derived from the cells of the ingrowing mesonephric tissue when the first meiotic division is initiated. Once the meiotic process is initiated, the oogonial germ cells are defined as primary oocytes. As a consequence of initiation of meiosis, multiplication is prohibited and the store of female gametes is set definitively at that stage of life. Once primordial follicles are formed the first meiotic division is arrested at the diplotene stage. The chromosomes decondense and are packed within a nucleus known as the germinal vesicle. In humans, the follicle may remain at this stage of development for 40–50 years up to the moment at which a signal initiates oocyte and follicle growth. All remaining oogonia that are not surrounded by somatic cells are expelled from the ovary. The maximum number of germ cells is present at about 4–5 months after conception, and decreases from 8 ⫻ 106 to 1–2 ⫻ 106 at birth by a process called ‘attrition’ (Baker, 1963). During prepubertal life continuous initiation of follicle growth leads to further depletion of the store of gametes available. There are few reports on the number of primordial follicles in fetuses. The data from Baker (1963), Block (1953) and Sforza et al. (1993) indicate that there is a trend towards increasing numbers of primordial follicles with gestational age. The limited data on early follicle growth stages in human fetal ovaries show that at birth in humans < 1% of follicles have developed further than the primary

stage (Kurilo, 1981). Antral follicle stages can be observed in human fetal ovaries (Fig. 1).

Cryostorage of ovarian cortical tissue After the pioneering work by Roger Gosden in the early 1990s (Carroll, 1990), researchers started to investigate storage of immature female gametes (Newton, 1998; Gook et al., 1999). At any stage of premenopausal life, there are only a few follicles at advanced stages of development that could be used for storage. In an ovary of a woman aged 20, there are only approximately 100 follicles in the growing pool and approximately 30 of these are at the antral stage (Gougeon, 1984). Therefore, it is more advantageous to cryopreserve the large number of gametes present at the earliest stages of follicle development: the resting primordial stages (Gougeon et al., 1994; Lass et al., 1997). By using slow freezing protocols it is possible to recover 70–80% of the initial number of primordial follicles that are frozen (Hovatta et al., 1996; Newton et al., 1996; Oktay et al., 1997; Gook et al., 1999). The cryopreservation technique requires sophisticated equipment such as a programmable freezing machine. Therefore, investigation of new cryopreservation media that permit rapid freezing by simply plunging the tissue into liquid nitrogen would be worthwhile (Zhang et al., 1995). This development would enable more clinics to offer fertility preservation to women undergoing treatments, such as chemotherapy, radiotherapy or ovariectomy, that have devastating effects on the follicle (Donnez and Bassil, 1998; Yeung et al., 1998; Smitz and Cortvrindt, 1999). After initial cryopreservation experiments with mice (Carroll et al., 1991; Cortvrindt et al., 1996a; Newton et al., 1996) and sheep (Gosden et al., 1994; Aubard et al., 1999) the survival and potential for further development of frozen ovarian primordial follicles was evaluated by transplanting ovarian tissue pieces in immunocompromised mice (Newton et al., 1996; Shaw et al., 1997; Oktay et al., 2000). Homologous transplantation was attempted in humans by Nugent et al. (1998) and Oktay et al. (2000). The transplantation experiments on human tissues indicate that cryopreservation does not compromise further development of frozen oocytes, but that there is a need for refinement of the grafting conditions before moving to clinical applications. Most primordial follicles die by ischaemia after clamping of ovarian arteries and by the slow revascularization. However, transplantation of ovarian tissue will not be possible for all malignant indications as a result of the risk of re-transferring the cancer cells. For these indications, in vitro systems for folliculogenesis should be developed (Shaw et al., 1996; Meirow et al., 1999).

Requirements for experimental material for follicle culture Primordial follicles that will survive the sampling and cryopreservation conditions are suitable for use in the ovarian tissue culture laboratory. Ideally, a sufficiently high

Folliculogenesis in vitro

(a)

(b)

(c)

(d)

(e)

(f)

Fig. 1. Overview of follicle formation and growth in the human fetal ovary. (a) Section through a fetal ovary at week 18 of gestation. Follicles are not yet organized. The large clear cells with the largest nuclei, observed mainly in the central portion of the section, are oogonia. (b) Part (a) at larger magnification. Oogonia are dispersed within mesenchymal cells. Cells from the coelomic epithelium form cords. At this stage oogonia are still interconnected by cytoplasmic bridges and appear as strings (arrow). The fetal ovary is delineated by coelomic epithelium. (c) Ovary from a fetus at week 24 of gestation. Oogonia can be observed which are entirely surrounded by pregranulosa cells and form the primordial follicle. The tissue already shows lytic changes. Follicles are embedded within mesenchymal tissue, which is highly vascularized. (d) Ovary from a fetus at week 25 of gestation. Follicles at the primordial stage with flattened pregranulosa cells are present. There is a follicle containing a growing oocyte (arrow) showing the cuboidal transformation of the pregranulosa cells. (e) Fetal ovary (week 30 of gestation) showing lytic changes. Primordial follicles are observed. In the right upper quadrant there is a follicle with three to four layers (arrow) surrounded by a theca interna containing a collapsed oocyte. (f) Fetal ovary (week 34 of gestation). The left half of the figure contains large (atretic) multilayered follicles, the right half contains very dense tissue populated with primordial follicles and oogonia. Scale bars represent (a) 20 µm, (b) 10 µm, (c,d) 25 µm, (e) 30 µm, (f) 150 µm.

187

J. E. J. Smitz and R. G. Cortvrindt

188

(a)

(b)

of the ovarian vascular tree and use of the tissue should be kept as short as possible. The tissue should be kept in a physiological solution with antibiotics on melting ice to reduce autolytic processes. It is impossible to estimate the number of follicles within the cortical tissue by observation under a dissection microscope, as there are a number of morphological structures that can be mistaken for follicle structures (curled vessels). Cortvrindt and Smitz (2001) reported a method permitting a rapid evaluation of a biopsy in terms of the number of follicles, ovarian tissue viability and even morphology of the follicles (Fig. 2).

(c)

Classification of small ovarian follicles and morphological aspects

Fig. 2. (a,b) A piece of human ovarian cortex stained with calcium AM and ethidium homodimer and scanned by confocal laser microscopy, which reveals living (green) and dead cells (red), respectively. Analysis by fluorescence microscopy 45 min after applying the fluorescent stain reveals the follicles as bright green dots. Bright red spots represent dead cells. (c) The ‘PicoGreen’ stain binds DNA of all cells within a 10 min incubation period, which permits recognition of follicle structures by scanning using a confocal laser scanning microscope. This rapid technique permits evaluation of oocyte size and stage of follicle development. A combination of both techniques illustrated permits a rapid assessment of follicle contents, density and stage of development. Scale bars represent 20 µm.

density of follicle units should be present to permit a fair comparative study of culture conditions. Ideal tissues for this purpose are fetal ovaries (from second and third trimester aborted fetuses), or ovaries from prepubertal girls as there is a large number of primordial follicles in this material (Fig. 1f). Experience with adult tissues shows that in donors over 30 years of age follicles are grouped in clusters and the low density of the primordial follicles makes the tissue less useful for determining optimum culture conditions (Lass et al., 1997; Campbell and Picton, 1999; Cortvrindt and Smitz, 2001). Several sources of ovarian tissue have been investigated in the last ten years including ovarian biopsies from infertile women at laparoscopy, strips of ovarian tissues from patients undergoing sterilization, biopsies during Caesarean section, ovariectomy pieces and wedge resection biopsies (Hovatta et al., 1997; Nugent et al., 1998; Cortvrindt and Smitz, 2001). Ovarian tissue can also be obtained from young patients requesting sex reversal (Van den Broecke et al., 2001). The optimum conditions for sampling and transportation of ovarian tissue before cryopreservation or culture in vitro have been studied (Shaw et al., 1997). The period between clamping

Several classifications for mammalian ovarian follicles (human, sheep, mouse) have been published (Pedersen and Peters, 1968; Lintern-Moore et al., 1974; Gougeon, 1996; Lundy et al., 1999). The classification proposed by Gougeon (1996) for human follicles is summarized (Table 1). The primordial follicle is defined as a primary oocyte surrounded by flattened granulosa cells. When a few of the flattened cells become cuboidal the follicle is classified as ‘transitionary’ or ‘intermediary’; these follicles are still considered to belong to the resting pool. A primary follicle is characterized by a full cuboidal granulosa cell layer surrounded by a basement membrane. Primary follicles are the first stage belonging the growing pool. Preantral follicles can be identified as a growing primary oocyte surrounded by several (up to six or seven) granulosa cell layers. Theca cells are recruited from the interstitial stromal cells and can be recognized as individual cells on the basement membrane in part of the primary follicles (Hirshfield, 1991; Lundy et al., 1999; Parrott and Skinner, 2000). As soon as the follicle reaches the secondary stage (two layers), a distinct theca cell layer is formed in all follicles (Gougeon, 1996; O’Shaugnessy, 1997; Lundy et al., 1999). Preantral follicle growth can be divided into two phases: a vascular and an avascular phase. After seven to eight doublings of the number of granulosa cells, mammalian follicles reach a diameter of 200 µm. Fluid-filled patchy spaces appear within the granulosa cells and the follicles are termed ‘early antral’. Follicles are termed ‘antral’ when the fluid-filled spaces have become coalescent into a large crescent cavity; granulosa cells then differentiate into mural and cumulus cells. Other morphological milestones can be used to describe the developmental stages. Formation of the zona pellucida during transformation of the primordial follicle into a primary follicle is another distinct reference point. The zona pellucida and theca interna layer are formed when the follicles are at the primary stage (Braw-Tal and Yossefi, 1997; O’Shaugnessy et al., 1997; Lundy et al., 1999; Yeh and Adashi, 1999). In a further stage of preantral follicle development, a theca externa is formed as a highly vascularized layer, providing the follicle with systemic endocrine factors that permit its exponential volumetric expansion (Fig. 3).

Folliculogenesis in vitro

Table 1. Morphometric characteristics of very small follicles in adult human ovary

Follicle Primordial Intermediary Small primary

Oocyte Follicle Oocyte nuclear diameter diameter diameter (µm) (µm) (µm) 35 38 46

32 32 33

16 16 17

189

(a)

Mean number of granulosa cells (range) 13 (7–23) 28 (9–50) 76 (23–223)

Adapted from Gougeon, 1996.

Do all oocytes from primordial follicles have the capacity to be ovulated? It is difficult to classify a follicle as being either at rest or at an early phase of growth using strictly morphological criteria. The transformation of flattened pregranulosa cells into cuboidal cells, initiating the growth of the oocytes, is an extremely slow process. Gougeon and Chainy (1987) referred to the transformation as the ‘maturation’ phase instead of the ‘growing’ phase. These authors consider that the follicular growth phase is initiated only when the oocyte of a primary follicle starts to grow rapidly. More recent data presented by Gougeon and Busso (2000) using proliferating cell nuclear antigen (PCNA) immunostaining or bromodeoxyuridine (BrdU) incorporation indicates that up to the primary stage most of these follicles are in the resting stage. It is still a matter of debate whether primordial follicles can be removed directly from the resting pool by atresia. Reynaud and Driancourt (2000) and Yuan and Giudice (1997) could not demonstrate apoptosis in resting follicles in sheep ovaries and in human ovarian specimens, respectively. Lee et al. (1999) studied expression of apoptosis-related genes in 35 human ovaries and described the sequence of expression of genes known to induce cell death by apoptosis for the earliest stages of follicular growth. Lee et al. (1999) could not demonstrate expression of Bcl-2 or Bax in primordial follicles. FAS and FAS ligand were present only in oocytes from primordial follicles, but not in their surrounding granulosa cells. In preantral follicles, neither Bcl-2 nor Bax could be detected in oocyte or granulosa cells, but both types of cell showed weak staining for FAS and FAS ligand (compared with intense FAS staining in primordial oocytes). Atretic follicles showed intense staining for Bax in granulosa and theca cells and moderate staining for FAS in oocyte, granulosa cells, theca cells and stroma. Weak staining for FAS ligand was observed in granulosa and theca cells, but not in stromal cells. There is a consensus that in early preantral follicles in vivo the process of atresia is initiated in the oocytes, whereas in the larger antral follicle population apoptosis emerges in the somatic compartment (Tilly et al., 1991; Yuan and Giudice, 1997). Recent observations from de Bruin et al. (2001) indicated that a necrotic process is the basis of atresia in resting follicles.

(b)

(d)

(c)

(e)

Fig. 3. The earliest follicle growth stages in the adult human ovary. (a) Initiation of follicle growth in the proximity of a capillary vessel at the corticomedullary border. The transition from the resting to the growing stage of the follicle is shown. In close proximity to the blood vessel there is a primordial follicle (left) and a large primary follicle (right). The upper left quadrant contains a follicle with granulosa cells transforming from flattened to cuboidal morphology. (b) Resting primordial follicle embedded within stromal cells. An oocyte at the germinal vesicle stage in surrounded by flattened pregranulosa cells encapsulated by a thin basal membrane. (c) Growing primordial follicle: the volume of the oocyte has increased as well as the surrounding granulosa cells, which have changed shape to cuboidal. Note the thickening of the basal membrane. (d) Primary follicle: growing oocyte surrounded by a dense layer of cuboidal granulosa cells encapsulated in a basal membrane. Note the proximity of a vessel (arrow). (e) Late preantral follicle with six to eight layers of granulosa cells and a well-organized theca cell layer. Scale bars represent (a) 30 µm, (b,c,d) 12 µm, (e) 20 µm.

190

J. E. J. Smitz and R. G. Cortvrindt

If all primordial follicles have an equal potential for development when exposed to optimum culture conditions, initiating an in vitro culture with primordial follicles could provide a very large source of developmentally competent oocytes. Cryopreservation of oocytes (primordial follicles) could circumvent the accelerated disappearance of the finite store of female gametes that occurs in vivo when females are exposed to environmental toxicants, tobacco, alcohol, disease, irradiation or adverse medication. The question remains whether the intrinsic quality of oocytes in resting follicles decreases with age. Although morphological changes in mitochondria, smooth endoplasmic reticulum and Golgi apparatus from oocytes in resting follicles are correlated with age, it is as yet uncertain whether this finding has any prognostic value for the future lower quality of oocytes (de Bruin et al., 2001).

Functional characteristics of preantral follicle growth The earliest stages of ovarian folliculogenesis are morphologically similar in different mammalian species. However, each species has its own specific timescale for development. In rodents (mice, rats) the time span between initiation of follicle growth and formation of the antral cavity is a few weeks; in large domestic animals it takes several months. During the preantral growth phase, the oocyte grows rapidly and reaches almost its maximum volume when the first accumulation of fluid is observed within the granulosa. In humans, Gougeon (1986) estimated that the maturation phase from primordial to primary follicle takes > 120 days. Once in the growing pool, the follicle requires 65 days to reach the early antral phase (follicle of 2–5 mm diameter), at which point it becomes dependent on gonadotrophins for further growth.

Regulation of early preantral follicle growth stages The fine regulation of the selection and initiation of growth of the primordial follicles out of the dormant pool remains enigmatic. Primordial follicles are located in a 2 mm thin, poorly vascularized layer under the tunica albuginea. The follicles showing initiation of growth are found in the corticomedullary border, which is highly vascularized (van Wezel and Rodgers, 1996) (Fig. 3). It is known that the earliest growth stage, including the transition from primordial to primary follicle, is gonadotropin-independent and is regulated mainly by intraovarian factors. The first trigger for initiation of growth has not been characterized and might arise in the oocyte or in the surrounding cells. These cells are of widespread origin: some are derived from haematopoietic stem cells, some belong to vascular structures and some to the autonomic nervous system. Oncogenes c-myc and erb-A localized in oocytes of the primordial follicle and pregranulosa cells might play a role in autonomous growth (Li et al., 1994; Sato et al., 1994). C-Kit, which is present in primordial oocytes, plays an important

role as inhibition of interaction of C-Kit with Kit ligand, originating in granulosa cells, blocks the initiation of growth (Yoshida et al., 1997). Studies in vitro in mice showed that Kit ligand induces growth in oocytes isolated from prepubertal animals (Packer et al., 1994). A recent review by Driancourt et al. (2000) emphasized the relevance of Kit–Kit ligand interaction as a key regulator of the initiation of growth from the primordial pool and transitions beyond the primary stage. There is evidence for mechanisms that hold the pool of primordial follicles under growth arrest. The tissues surrounding the primordial follicles undoubtedly play a crucial role in co-ordinated initiation of growth, as removal of the stromal compartment in vitro is accompanied by marked initiation of granulosa cell proliferation (Wandji et al., 1997; Fortune et al., 2000). Members of the transforming growth factor β (TGF-β) superfamily of secreted growth factors play important roles at several stages of ovarian folliculogenesis. Among factors of the TGF-β family, activin A is secreted by secondary follicles. In an in vitro model using rat follicles, Mizunuma et al. (1999) demonstrated that activin A can prevent the growth of primary follicles despite their exposure to extraovarian growth factors. Recent data from the knockout mouse model for anti-Müllerian hormone (AMH) and from follicle culture indicates that AMH inhibits recruitment of primordial follicles (Durlinger et al., 1999). In the human ovary, AMH protein was not detected in primordial follicles, but was found in the granulosa cells of primary follicles and was abundant in secondary and early antral follicles (De Vet et al., 2000). Some members of the TGF-β superfamily are expressed by the oocyte: growth differentiation factor 9 (GDF-9) and bone morphogenetic protein 15 (BMP-15), which are similar to GDF-9B and BMP-6 (for a review see Elvin et al., 2000). Although GDF-9 expression has been localized in other extraovarian tissues, BMP-15 mRNA has been found exclusively in the ovary. There is species-dependent variability in the expression of these factors. In mice, there is no GDF-9 expression in oocytes before the type 3a preantral follicle stage, but in pigs and cows GDF-9 has also been found in primordial follicles. The knockout mouse for GDF-9 showed arrest of granulosa proliferation and, despite exaggerated oocyte growth, absence of theca cell formation (Dong et al., 1996). The other TGF-β family members expressed by the oocyte were unable to substitute for GDF9. These oocytes cannot become cytoplasmically mature. Factors of vascular origin (pericytes), such as Thy-1 differentiation protein, could signal to the oocyte to remove a hypothetical arresting factor originating in the oocyte and acting upon the pregranulosa cells (Bukovsky et al., 1995). The hypothesis that oncogenes such as erbA (Maruo, 1995), myc (Li et al., 1994) and Wilms’ tumour gene (Hsu et al., 1995; Chun et al., 1999) may be involved in initiation of follicle growth is based on their localization in oocytes of primordial follicles and granulosa cells of the follicles at the earliest growth stages. Expression of Wilms’ tumour gene

Folliculogenesis in vitro

191

FSHr Oocyte diameter (µm)

LHr TC

Steroidogenic capacity

90

5000 GC

ERs ARs

70

Antral cavity

600 GC 50

Theca externa Vascularization

7–50 GC 30 Theca interna Zona pellucida 30

50

100 150 Follicle diameter (µm)

> 120 days Follicle maturation phase

25 days Follicle growth phase

200

60 days to ovulation

Fig. 4. Earliest stages of follicle growth in humans. The relationship between follicle diameter (x-axis) and oocyte diameter (y-axis) is shown and major milestones in follicle formation and differentiation are indicated. AR: androgen receptor; ER: oestrogen receptor; FSHr: FSH receptor; GC: granulosa cells; LHr: LH receptor; TC: theca cells.

(WT-1) decreases from the primordial pool of oocytes up to the antral stage follicles. Expression of WT-1 suppresses growth and differentiation genes (growth factors and receptors). There is increasing evidence that the following neurotrophins play a role in the initial stages of follicular development: nerve growth factor (NGF), brain-derived neurotrophic factor (BDNF), and neurotrophins 3 and 4. The mRNAs encoding these factors and their tyrosine kinase receptors (trk) are transiently present in the ovarian follicle during follicle formation and early follicular growth (Ojeda et al., 2000). A combined deficiency of neurotrophins 3 and 4 and BDNF resulted in a significant reduction in the number of follicles at all preantral stages (Ojeda et al., 1999). Among neurotrophins, vasoactive intestinal polypeptide (VIP) is also a potential growth initiator: VIP is expressed on nerves in the proximity of follicles at the earliest growth stages in cows (Hulshof et al., 1994) and monkeys (Schultea et al., 1992). In the bovine model, primordial oocytes express basic fibroblast growth factor (bFGF) (Van Wezel et al., 1995), which can interact with granulosa cells from early preantral follicles (Gospodarowicz and Bialecki, 1979). Furthermore, FGF-2 mRNA and FGF receptor have been localized in human fetal ovarian tissue (Yeh and Osathanondh, 1993). In humans, as soon as formation of a secondary follicle begins, granulosa cells develop FSH receptors (Oktay et al., 1997) and theca cells become differentiated around the basement membrane. As soon as theca cells form, LH receptor is

expressed (mice: Sokka et al., 1996; O’Shaughnessy et al., 1997). During formation of the secondary follicle there is a migration towards the medullary part of the ovary, in which early theca externa will be acquired, and vascularization of the follicle is initiated. Gonadotrophin receptors and steroid receptors also appear (Fig. 4). LH stimulates the theca cells to proliferate, differentiate and initiate androgen production. Although theca cells are an important target of LH action, the survival of theca cells is not dependent on LH, as demonstrated recently in LH receptor knockout mice (Zhang et al., 2001). It is mainly androstenedione, which is LH-dependent, which is derived from theca cells and acts in granulosa cells to cytodifferentiate the responses to FSH (Magoffin and Erickson, 1994). Androgen receptor immunoreactivity has been demonstrated in human and primate granulosa and theca cells (Hild Petito et al., 1991; Horie et al., 1992) and expression of the androgen receptor gene was most abundant in the growing preantral and antral follicle pool (primates: Weil et al., 1998). Treatment of primates with androgens (testosterone and dihydrotestosterone) in vivo promoted growth of small and medium-sized follicles (Vendola et al., 1998) and stimulated insulin-like growth factor I (IGF-I) expression in oocytes and expression of FSH receptor in granulosa cells (Vendola et al., 1999; Weil et al., 1999). From knockout mouse models and diseases in humans involving genes for gonadotrophin subunits and gonadotrophin receptors it was shown that the preantral follicular growth phase is largely gonadotrophin-independent

J. E. J. Smitz and R. G. Cortvrindt

192

Optimal ovulation

Early antral (180 µm)

Johnson et al. (1995) 4 days

Hartshorne et al. (1994)

Late preantral (150 µm) 6 days

Nayudu and Osborn (1992)

Early preantral (100—130 µm) 12 days

Cortvrindt et al. (1996, 1998a)

Early secondary (90—100 µm) 16 days (80—90 µm) 18 days

Unpublished observation from authors laboratory (2001)

(70—80 µm) 20 days Primordial (< 30 µm) Organ culture: 8 days

Eppig et al. (1995) 12 days

Fig. 5. Experimental evidence on follicle culture periods from published studies. Optimum maturation results are obtained only when the duration of culture is adapted to the stage of follicle development at isolation. Follicle development in vitro follows a strictly timed progression.

(Cattanach et al., 1977; Kendall et al., 1995; Dierich et al., 1998; Abel et al., 2000). In humans, follicles of 2–5 mm in diameter can be observed in the ovary in the absence of FSH (Kallman syndrome). However, the efficiency of growth during the preantral phase might be improved when baseline concentrations of gonadotrophins are made available.

The mouse model in follicle culture Rodent species (mice, rats, hamsters) have been used extensively for investigation of follicle growth in vitro. The advantage of using mice is that they have a short life cycle allowing experiments to be performed within a reasonable time frame. The production of live offspring after culture of early

stage follicles for extended periods has only been documented in mice (Eppig and Schroeder, 1989; Eppig and O’Brien, 1996; Cortvrindt et al., 1998b). It is well established that folliculogenesis and meiotic maturation are strictly timed processes (Fig. 5). Consequently, at isolation, each follicle unit is at a particular developmental stage. Thus follicle selection at isolation should ideally involve a very homogeneous population of follicles to obtain a large yield of competent oocytes at a particular time point after culture. Selection of follicles is possible under the microscope (⫻ 400) by measuring follicle diameter and classifying the number of granulosa cell layers surrounding the oocyte. For each initial follicle diameter at the start of culture there is an intrinsic optimum culture period in which to reach the optimum maturation stage. Initiating a

Folliculogenesis in vitro

193

Table 2. Characteristics of culture systems for mouse ovarian follicles Culture system (see Fig. 6)

Follicle components

System and follicle integrity

S1 S2 S3

Organ culture Oocyte/GC Oocyte/GC/TC

Groups Groups Individual

– Open (COC) Open

S4 S5

Oocyte/GC/TC Oocyte/GC/TC

Individual Individual

Open Closed

Sessile Sessile (1) Floating (2) Sessile Sessile Floating

Medium

Protein

Waymouth Waymouth α-MEM α-MEM α-MEM α-MEM

BSA BSA FCS FCS FCS Mouse serum

Hormones + growth factors

Ovulation stimulus

EGF – FSH/LH FSH/LH FSH/LH FSH

– FSH – hCG + EGF hCG + EGF –

BSA: bovine serum albumin; COC: cumulus–oocyte complexes; EGF: epidermal growth factor; FCS: fetal calf serum; GC: granulosa cells; hCG: human chorionic gonadotrophin; MEM: minimal essential medium; TC: theca cells.

Early preantral Follicle stage

Primordial

30

Primary

50

Late preantral

Antral

Secondary 100 130 150 180 200 µm

70

Follicle diameter at start

S1: Organ culture Eppig and O’Brien (1996)

S2: Granulosa oocyte culture Eppig and Schroeder (1989)

S3: Primary follicle culture Unpublished authors’ data (2001)

S4: Secondary follicle culture Cortvrindt et al. (1996)

S5: Late preantral follicle culture Nayudu and Osborn (1992) Fig. 6. Published culture systems for mouse follicles, designed in relation to the follicle diameter at the start of culture. The figure relates stage of follicle development to mean follicle diameter in mice.

culture with follicles that have a homogeneous diameter will greatly improve oocyte quality at the end of the culture. During the first wave of follicular growth, it is easy to obtain a large homogeneous population of follicles. As follicle formation occurs after birth in mice, the age of the juvenile mice can be selected in relation to the maximum developmental stage of the follicles that are required for each specific experiment. A group of 20–30 primary follicles per ovary can be isolated mechanically from 7-day-old mice. Approximately

40–50 early secondary follicles can be dissected with fine needles from ovaries of 14-day-old animals.

Culture systems for mouse ovarian follicles Different culture systems have been described for mouse ovarian follicles (Fig. 6) depending on the size classes of follicles at isolation. Details of a variety of culture systems have been published (Table 2) but in terms of culture morphology they

J. E. J. Smitz and R. G. Cortvrindt

194

(a)

(b)

(c)

(d)

(e)

(f)

(g)

(j)

(h)

(k)

(i)

(l)

Folliculogenesis in vitro

can be divided as follows: (i) cultures in which the aim is to keep the spherical structure intact; and (ii) cultures in which the follicles are allowed to remodel into sessile mound-like structures. The choice of culture system used should be dictated by the aims of the study. When the purpose is to use the somatic cell compartment as a ‘feeder layer’ for oocytes, a high yield can be obtained by allowing the follicles to remodel (Cortvrindt et al., 1996b) (Fig. 7). When the aim is to determine physiological intrafollicular relationships, it is necessary to maintain the cultured units as multilayered spheres (Nayudu and Osborn, 1992). In addition, when the culture is to be used to study ovulation, it is necessary to retain a normal follicular wall, that is, mural granulosa, basement membrane and theca cell layer (Rose et al., 1999). Choosing an artificial culture system will always imply a compromise. When the aim is to mimic normal physiology and to culture the follicles as multilayered spheres, the supply of essential nutritional and physical factors to the inner centre of the structure becomes critical (Johnson et al., 1995). Nutrients, growth factors and oxygen have to diffuse through a non-vascularized follicular wall. Experience with this culture model shows that decreased oocyte and follicle survival rates ensue, probably because it is impossible to compensate in vitro for the absence of the highly vascularized network that is present within a normal thecal wall. When the aim is to culture oocytes, the environment should be supportive of the growth of this large type of cell and should supply the essential nutrients. This support is feasible even with a non-spherical ‘open’ structure (Eppig and Schroeder, 1989). The accessibility of nutrients, hormones and gases is much better in an open structure and leads to an improved oocyte survival rate. Open culture systems can be refined by adapting the composition of the medium to preserve the normal differentiation status of the cells surrounding the oocyte. As emphasized by Eppig et al. (1997), the crosstalk between oocyte and cumulus cells is of major importance in determining oocyte developmental competence. Fine tuning of the culture components, including the types of cell (with or without thecal cells), gonadotrophin concentration and ratio, growth factors (insulin, IGF-I, epidermal growth factor (EGF)), protein

195

source (albumin, synthetic serum supplements, homologous serum, fetal bovine serum) and concentration, can lead to optimization of the culture system (Table 3). Maintaining the physical parameters such as temperature and pH changes under strict control is of significant importance to guarantee good culture outcome. When optimizing the culture conditions, it is important to recognize the balanced interrelationships between the biochemical and physical components of the system. It has been demonstrated that there is close interaction between the oxygen requirements of a particular system, the hormonal and growth factor composition and the protein source (Gosden and Byath-Smith, 1986; Eppig and Wigglesworth, 1995; Smitz et al., 1996). Follicle survival and oocyte competence are obviously interdependent. Oxygen supply to the deepest layers of the follicle is dependent on the requirements of the somatic cells, which are driven into proliferation by the relative amounts of insulin, IGFs and FSH. Follicle survival in vitro is FSH-dependent but the FSH concentration in the culture medium determines the metabolic rate at which the granulosa cells function. The metabolic requirements for oxygen will differ in relation to the basal FSH and insulin tonus. In the mouse follicle culture system described by Cortvrindt et al. (1996b), oocytes do not survive under reduced oxygen (5%) when the FSH concentration in the medium is 100 miu ml–1 (Smitz et al., 1996). Oocyte survival under low oxygen tension is possible only by reduction of the FSH concentration to 10 miu ml–1 (the minimum effective dose) (R Cortvrindt and J Smitz, unpublished). Oxygen provision also influences the generation of free radicals, which can be neutralized by serum components and scavengers. The study of Eppig and Wigglesworth (1995) illustrates the relationship between oxygen tension and the potential to generate normal blastocysts. It should be emphasized that, in this system, oocyte–granulosa cell complexes are cultured without serum and gonadotrophins. These conditions are compatible with a lower metabolic rate and hence lower oxygen requirements for membrane and steroid biosynthesis. Applying a supraphysiological oxygen tension to the culture system of Eppig and Wigglesworth (1995) might lead to production of free oxygen radicals, which interfere with key developmental processes and lead to arrest in development.

Fig. 7. In vitro growth of early preantral follicles, in vitro ovulation, preimplantation embryo culture and production of live offspring in mice. (a) A secondary follicle (diameter 115 µm) containing an oocyte at the germinal vesicle stage (diameter 55 µm) after isolation from a 14-day-old mouse. (b) At day 2 of culture the theca cells have attached the follicles to the bottom of the culture dish and somatic cells are proliferating. (c) At day 4 granulosa cells have pierced through the basal membrane and are overgrowing the theca cell monolayer. (d) At day 6 the original spherical structure is completely undetectable. The oocyte is maintained within a mound of granulosa cells surrounded by distally located theca cells. (e) At day 8 of culture the antral-like cavities are formed showing clear differentiation of granulosa cells in mural and cumulus cells. (f) On day 10 the antral-like cavity is recognized as a clear halo around the cumulus–oocyte complex. (g) On day 12 the antral-like cavity is expanded and an individual cumulus complex is visible. (h) On day 13 the expanded and mucified cumulus–oocyte complex has detached from the cultured follicle, which received 1.5 iu hCG l–1 on day 12. (i) Denudation of the somatic cells on day 13 reveals a first polar body extruded in the perivitelline space. Oocyte diameter is 72 µm. (j) An embryo at day 1 obtained from a fertilized oocyte grown in vitro. (k) Hatching blastocyst at day 5 after fertilization. (l) Normal live young obtained after culture of early preantral follicles.

196

J. E. J. Smitz and R. G. Cortvrindt

Table 3. Percentage success illustrated by stepwise implementation of components in basal culture medium for culture of mouse ovarian follicles Outcome parameters Meiotic competencea

Components of culture medium Rich basal medium Rich basal medium + 5% FCS Rich basal medium + 5% FCS + FSH Rich basal medium + 5% FCS + FSH + LH Rich basal medium + 5% FCS + FSH + LH + GF

Days 3 6 13 13 13

Follicle survivala GVBD 0 20 70 80 90

0 20 80 90 100

MII

Fertilizationb (two-cell)

Developmentc (blastocyst)

Live young

– 10 40 70 80

– – Variable 40 Stable 50 Stable 60

– – 40 60 70

– – Variable +/– Stable + ++

GF: growth factors, proteins, peptides, scavengers; GVBD: germinal vesicle breakdown; MII: metaphase II. aPercentage of plated follicles. bPercentage of surviving oocytes. cPercentage of fertilized oocytes.

Endpoints in follicle culture can vary widely: intact follicle survival, germinal vesicle breakdown, extrusion of the first polar body, fertilization, hatching blastocyst formation or live offspring. It is evident that for those performing oocyte culture the only valid endpoint on which to evaluate culture conditions is achieving a reasonable yield of healthy offspring. However live offspring is a distant endpoint. Hence there is a need to determine valid predictors of normal developmental competence at the oocyte. Further studies are required to evaluate the point at which developmental competence is determined by the source and the endocrine environment at harvesting. How important are factors such as the origin of the follicle (fetal, prepubertal, postpubertal, aged female), priming of the animal with gonadotrophins before follicle isolation, and nutritional and seasonal influences? Do all primordial follicles have the potential to generate healthy offspring or do some primordial follicles have certain inherent defects which are eliminated by the natural selection processes during further development? To what extent is the marked reduction in the number of immature female gametes determined by the hormonal environment? Is there an optimum environment for the oocyte in which it could progress from the resting stage to a fully grown oocyte that is developmentally competent? Can the natural selection processes be bypassed by artificially designed growth conditions in vitro? The use of ovaries from fetal or prepubertal animals for research purposes to answer these questions is very attractive since these ovaries contain a large population of primordial follicles. Suggestions of a low developmental competence of oocytes harvested from prepubertal animals originate from studies using antral follicles as the starting material, which have already progressed much further in development and possibly are already conditioned irreversibly towards atresia (Driancourt et al., 2001). The experiments of Eppig and O’Brien (1995) and Cortvrindt et al. (1998b) reported the production of live

offspring from follicles isolated from prepubertal animals (follicles were isolated from 12–14-day-old mice). As the mouse primordial follicle is formed between day 1 and day 4 after birth, these experiments prove that developmentally competent oocytes can be generated within 3–4 weeks after the formation of the follicle in which they are harboured. The first waves of growing follicles might not require the long transition phase (as seen in adult animals) for transformation of the resting primordial follicle into a growing primary follicle. For adult rats, this maturation period is estimated to be 30 days (Hirshfield, 1991; Hirshfield and DeSanti, 1995; for a review see McGee and Hsueh, 2000). The shorter period for maturation and growth of follicles from the first cohort could be related to the more advanced stage of the oocytes retained in the primordial follicles of the first wave. Data from Lundy et al. (1999) in sheep demonstrated the differences in the diameter of oocytes in primordial follicles. Alternatively, the differences observed in follicle growth characteristics could be explained by the difference in origins of the somatic cells (precursors of follicular cells: mesenchymal and mesangial cells of the mesonephros) during the process of oocyte envelopment ending in formation of the primordial follicle (McNatty et al., 2000). This observation provides the important message that within the same ovary follicle heterogenicity in somatic composition might promote growth at a faster rate. As suggested earlier by Eppig and O’Brien (1995), follicle culture could circumvent an ‘inadequate’ hormonal environment related to the age of the animal (prepubertal or perimenopausal). Experiments from our laboratory indicate that mouse follicles isolated at the early preantral stage from prepubertal animals perform equally well during culture in vitro (in terms of blastocyst formation) as do follicles of a similar developmental stage isolated from adult females (D Nogueira, R Cortvrindt and J Smitz, unpublished), whereas oocytes grown in vivo are less competent when they originate from prepubertal animals. Similar work using follicles from aged perimenopausal

Folliculogenesis in vitro

females could be conducted to test the hypothesis that oocyte quality is determined by the follicular environment rather than by age factors. In other mammals in which primordial follicles are formed during fetal life the question is even more important as the follicle reserve is largest at that stage of development. As a source for oocyte banking (for research purposes), a fetal ovary is most attractive as it contains the very large number of primordial follicles embedded in a very soft interstitial tissue that makes follicle harvest and extraction much easier. Although culture of ovarian follicles from mice has shown the potential to generate live offspring, refining the conditions to optimize yields remains a major challenge. The shortened life span of a mouse born after culture of a primordial follicle in vitro should prompt further studies on the life expectancy of such offspring (Eppig and O’Brien, 1998). The mouse model has great potential for use in the pursuit of the essential knowledge that will underpin more longlasting and extensive experiments in larger mammalian species.

Culture of preantral follicles from other mammalian species Several researchers have published results on follicle culture in other mammalian species such as rats, hamsters, pigs, cats, sheep and cows. When species that are larger than rodents are used, the first problem encountered is the isolation of small follicles, which are located within a dense interstitial tissue. For isolation, either an enzymatic (Greenwald and Moor, 1989; Roy and Treacy, 1993) or a mechanical approach is chosen, or enzymatic pretreatment can be followed by mechanical isolation using fine needles (Figueiredo et al., 1993; Abir et al., 1999). Although enzymatic digestion combined with other selection techniques (centrifugation or sieving) permits the harvest of large numbers of small follicles, survival rates of unilaminar follicles after in vitro culture are low (Figueiredo et al., 1993; Van den Hurk et al., 1998). The low survival rate may be due to the uncontrolled activity of the proteolytic enzymes, which can ultimately cause release of thecal cell layers, rupture of basement membrane and damage to or alteration of surface protein molecules. This damage leads to dispersion of granulosa cells, which will attach to the culture vessel and break their contact with the oocyte. The mechanical isolation of multilaminar preantral follicles using small needles is difficult and time consuming. In large mammals, there are a limited number of follicles at the large preantral stages within the ovary and these are embedded within a fibrous interstitial matrix, which precludes easy dissection. However, mechanical isolation has the advantage that it is possible to control the cell population included in the culture (Ralph et al., 1995; Abir

197

et al., 1997). As first reported by Wu et al. (1998), it is also possible to obtain small preantral follicles from the regular needle aspirates during oocyte retrievals for in vitro fertilization (IVF). Depending on the physical aspiration procedure and the type of ovary, between 1 and 40 (average 13) follicles with a diameter between 30 and 95 µm can be harvested (Mitchell et al., 1999). Wu et al. (1998) reported that such follicles could be cultured to antral stages and became meiotically competent, but such extensive development could not be repeated in another laboratory using similar or modified conditions (L. M. Mitchell and G. M. Hartshorne, personal communication). Attempts to culture primordial and primary follicles from cows (Van den Hurk et al., 1998) and humans (Abir et al., 1997; Hovatta et al., 1997, 1999) have resulted in massive follicle degeneration after a few days in culture. This led researchers to the idea of culturing the very small follicle units (primordial, transitionary, primary) in situ first. For this purpose ovarian cortex is sliced into pieces 100 µm in thickness and cultured just beneath the surface of the medium (to allow good oxygen and nutrient penetration and avoid necrosis) (Hovatta et al., 1997, 1999; Wright et al., 1999). This type of initial tissue culture could lead to initiation of growth and the subsequent culture steps could be performed using isolated follicle units. A review of published studies reveals that, at present, such systems do not lead to normal follicle development in vitro. The limited successes with culture of early preantral follicles from large mammals compared with the results achieved with follicles of equivalent stages from rodents may be explained by the fact that, initially, oocyte volume is proportionally much smaller in larger species than it is in rodents. This effect is revealed by comparison of the percentage of the maximum volume of the oocytes, for the same classes of follicle, for different species (Fig. 8). Use of several culture systems involving different culture media, growth factors and serum supplements provided the same observation: the resting follicles are activated, but degeneration occurs mainly within a few days (Braw-Tal and Yossefi, 1997; Hovatta et al., 1997; Wandji et al., 1997). Typical morphological changes are observed when culturing small ovarian tissue pieces in vitro (Fig. 9). In vivo, the transition and growth of primordial to primary follicles is characterized by slow transformation of the granulosa cells (a process of 120 days) from a flattened to a cuboidal appearance and initial growth of the nucleus; in vitro, these changes occur within a few days (Fortune et al., 2000). The regulatory mechanisms that co-ordinate growth of oocyte and cell division in the surrounding somatic cells are largely unknown. The culture environments and media compositions tested to date appear incapable of creating the right environment for inducing growth changes that are compatible with physiological growth patterns. Chaotic proliferation and altered differentiation of pregranulosa cells is observed in histological preparations from cultured tissue blocks. Oocyte growth is often insufficient or even undetectable and the development of a thecal cell layer at

J. E. J. Smitz and R. G. Cortvrindt

198

Small preantral follicles

Large preantral follicles (≥ 2 granulosa cell layers) 100 80

Rodent, pig

Rodent

60

Human

Human Pig Bovine

40

Bovine

20 0 Percentage of maximum oocyte diameter

Fig. 8. Percentage of oocyte diameter at start of culture, expressed as percentage of the maximum oocyte diameter at maturity, in small and large preantral follicles from different species.

the secondary follicle stage is rarely observed. In cultures of ovarian slices from baboon fetuses many follicles were observed in which the oocyte had increased in diameter but the surrounding cells had remained flattened (Fortune et al., 2000). The limited success to date testifies that signalling pathways are as yet inappropriately installed. Physical (temperature, pH, oxygen supply) or chemical (for example nutrients, growth factor balance) aspects of the culture systems are still inadequate and require optimization. In mice and rats, factors such as FSH and activin induce physiological responses that promoted follicle survival (Li et al., 1995; Cortvrindt et al., 1997; McGee et al., 1997; Yokota et al., 1997; Smitz and Cortvrindt, 1998). This effect was not observed unequivocally in larger mammals and humans. Although Fortune et al. (2000) found no evidence of growth-promoting effects of FSH or activin during the culture of bovine or baboon follicles from primary to secondary stages in vitro, Hovatta et al. (1997) and Wright et al. (1999) obtained improved follicle development by addition of FSH. Other factors added to culture media of human follicles that promote development in vitro include insulin (Wright et al., 1999) and GDF-9 (Hreinsson et al., 2001). Addition of Steel factor did not improve the growth of human follicles in vitro (Hovatta et al., 1997). The hypothesis was put forward that the media used for culture of cortical tissue pieces are too rich and lead to wholesale activation of primordial follicles. The normal environment of primordial follicles is poorly vascularized and oxygen tension is low in this compartment (Van Wezel and Rodgers, 1996). Fortune et al. (2000) recently developed a technique of grafting cortical tissue pieces between the chorioallantoic membrane and the yolk sac membrane of chicken eggs. This ‘in ovo’ grafting technique

prohibited the massive activation of the primordial follicles which is observed when follicles are cultured in vitro, providing evidence that the growth arrest of primordial follicles is not solely dependent on a medullary inhibitor, but is possibly linked to factors induced during the rapid neovascularization of the graft or to the environment created by bursal fluid within chicken embryonic membranes. This model could provide clues to the mechanisms regulating initiation of follicle growth.

Culture of preantral follicles from the growing pool in large mammals and humans Some studies have focused on the use of follicles isolated from ovarian tissue, which have already reached the early preantral growth stage. Ovaries from large mammals do not contain many follicles at this stage of growth. Hence the use of growing follicles as starting material will not be of major importance in oocyte storage programmes. As unilaminar follicles are present in large numbers in the human ovary, development of a culture system for these follicles would be more useful. Attempts to promote growth of secondary follicles in culture have been more successful: promising results have already been obtained in sheep. Newton et al. (1999) and Cecconi et al. (1999) reported growth of late preantral follicles in vitro and an increase in the volume of the oocytes, which were occasionally capable of resuming meiosis. Pig cumulus–oocytes complexes from preantral follicles have been grown for 2 weeks in collagen matrix and oocytes that reached 100 µm in diameter became meiotically competent in vitro (Hirao et al., 1994). However, the final yield from this culture system was very low: < 2% of oocytes completed the meiotic process. Attempts by Telfer et al. (2000) to design a culture system for pig follicles revealed that serum is an essential survival factor and that only 19% of follicles with a starting diameter of 210–260 µm could become meiotically competent. In cows, Gutierrez et al. (2000) maintained late preantral follicles in long-term culture for 1 month and reported an increase in follicle diameter but did not provide evidence of oocyte meiotic capacity. To date, few studies have focused on oocyte quality and there are insufficient data on oocyte meiotic competence after culture in vitro. In humans, although late preantral follicles can be kept viable in a culture dish over periods of a few weeks, there is as yet no consistent data documenting oocyte meiotic competence (Abir et al., 1997; Picton et al., 1999).

Conclusions and prospects Establishment of techniques for the culture of follicles in vitro will enable important progress in different fields of biology and medicine. Growth of primordial follicles in vitro followed by in vitro maturation of oocytes will provide mature oocytes for nuclear transfer programmes and cloning. Cryopreservation of female gonadal tissue is

Folliculogenesis in vitro

(a)

199

(b)

Fig. 9. In vitro culture of human ovarian tissue. (a) Freshly collected and fixed ovarian tissue from a transsexual patient aged 28. Primordial follicles are distributed in nests. (b) Frozen and thawed tissue from the same patient, cultured in vitro for 2 weeks. Two primordial follicles in which growth is arrested are present on the left-hand side. The follicle on the right contains a healthy oocyte which has increased slightly in volume and is surrounded by a single layer of cuboidal granulosa cells. No cell death is visible in the follicles, but pyknoses are present in the stromal compartment in the central and the bottom parts of the figure. Scale bars represent (a) 60 µm, (b) 30 µm.

already an established technique, which allows survival of > 50% of the stored gametes. Combination of cryostorage and follicle culture will offer new possibilities for preserving rare species. Strategies will be developed to protect the oocyte store from insults leading to sterility (environmental agents, chemotherapy, radiotherapy). The developments will bring new hope to young women undergoing cancer therapy who may be able to preserve their fertility. The ability to impact on the pace of spontaneous growth of follicles brings the potential to prolong the reproductive lifespan and prevent some of the secondary effects of the menopause and ageing. Today, the only available culture techniques are for animals such as mice, in which folliculogenesis is relatively short. Experiments to date with ovarian cortical pieces from larger animals, for example sheep, cows, pigs and humans, have provided disappointing results. Approaches using culture in vitro lead to the unco-ordinated growth of granulosa cells and oocytes and to the absence of normal differentiation of granulosa and theca cells. There is ample opportunity for improvement in this area: the technical requirements and the selection of essential biochemical determinants for the preantral stages of follicle development need to be characterized precisely. Mapping these stage-related biochemical determinants of follicle growth requires investigation of follicles at well-characterized growth stages using the new powerful molecular biology techniques such as gene chip technology. Interaction between those working in the fields of cell culture and molecular biology will lead to more prompt advances in this area. Unpublished work and studies published by the authors were supported in part by grants from the Fund for Research Flanders (Grant nos G.0166.98 and G.3C01.93N), from the IWT (grant no.

STWW 980343) and by Ares Serono International (Project no. GF 9405). The authors thank their clinical colleagues who provide the ovarian tissues for research: Paul Devroey, Jacques Donnez, Bart Fauser, Jan Gerris and Raphael Ron-El.

References Key references are identified by asterisks. Abel MH, Wootton AN, Wilkins V, Huhtaniemi I, Knight P and Charlton HM (2000) The effect of a null mutation in the FSH receptor gene on mouse reproduction Endocrinology 141 1795–1803 Abir R, Franks S, Mobberly MA, Moore PA, Margara RA and Winston RM (1997) Mechanical isolation and in vitro growth of preantral and small antral human follicles Fertility and Sterility 68 682–688 Abir R, Roizman P, Fisch B, Nitke S, Okon E, Orvieto R and Rafael ZB (1999) Pilot study of isolated early human follicles cultured in collagen gels for 24 hours Human Reproduction 14 1299–1301 Aubard Y, Piver P, Cogni Y, Fermeaux V, Poulin N and Driancourt MA (1999) Orthotopic and heterotopic autografts of frozen–thawed ovarian cortex in sheep Human Reproduction 14 2149–2154 Baker TG (1963) A quantitative and cytogenetical study of germ cells in human ovaries Proceedings Reproductive Society London (Biology) 158 417–433 Block E (1953) A quantitative morphological investigation of the follicular system in newborn female infants Acta Anatomica 17 201–206 Braw-Tal R and Yossefi S (1997) Studies in vivo and in vitro on the initiation of follicle growth in the bovine ovary Journal of Reproduction and Fertility 109 165–171 Bukovsky A, Caudle MR, Keenan JA, Wimalasena J, Foster JS and Van Meter SE (1995) Quantitative evaluation of the cell cycle-related retinoblastoma protein and localisation of Thy-1 differentiation protein and macrophages during follicular development and atresia, and in human corpora lutea Biology of Reproduction 53 1373–1384 Byskov AG (1986) Differentiation of mammalian embryonic gonad Physiological Reviews 66 71–117 Campbell BK and Picton HM (1999) Oocyte storage Current Obstetrics and Gynecology 9 203–209 Carroll J, Whittingham DG, Wood, MJ, Telfer E and Gosden RG (1990) Exraovarian production of mature viable mouse oocytes from frozen primary follicles Journal of Reproduction and Fertility 90 321–327 Carroll J, Whittingham DG and Wood MJ (1991) Effect of dibutyryl cyclic

200

J. E. J. Smitz and R. G. Cortvrindt

adenosine monophosphate on granulosa cell proliferation, oocyte growth and meiotic maturation in isolated mouse primary ovarian follicles cultured in collagen gels Journal of Reproduction and Fertility 92 197–207 Cattanach BM, Iddon CA, Charlton HM, Chiappa SA and Fink G (1977) Gonadotropin-releasing hormone deficiency in a mutant mouse with hypogonadism Nature 269 338–340 Cecconi S, Barboni B, Coccia M and Mattioli M (1999) In vitro development of sheep preantral follicles Biology of Reproduction 60 594–601 Chun SI, McGee EA, Hsu SY, Minami S, LaPolt PS, Yao HH, Gougeon A, Schomberg DW and Hsueh AJ (1999) Restricted expression of WT1 messenger ribonucleic acid in immature ovarian follicles: uniformity in mammalian and avian species and maintenance during reproductive senescence Biology of Reproduction 60 365–373 Cortvrindt R and Smitz J (2001) Fluorescent probes allow rapid and precise recording of follicle density and staging in human ovarian cortical biopsies Fertility and Sterility 75 588–593 Cortvrindt R, Smitz J and Van Steirteghem A (1996a) A morphological and functional study of the effect of slow freezing followed by complete in vitro maturation of primary mouse ovarian follicles Human Reproduction 11 2648–2655 Cortvrindt R, Smitz J and Van Steirteghem AC (1996b) In vitro maturation, fertilization and embryo development of immature oocytes from early preantral follicles from prepubertal mice in a simplified culture system Human Reproduction 11 2656–2666 Cortvrindt R, Smitz J and Van Steirteghem AC (1997) Assessment of the need for follicle stimulating hormone in early preantral mouse follicle in vitro culture Human Reproduction 12 759–768 Cortvrindt R, Hu Y, Liu J and Smitz J (1998a) A timed analysis of the nuclear maturation of oocytes in recombinant gonadotropin-supplemented early preantral mouse follicle culture Fertility and Sterility 70 1114–1125 Cortvrindt R, Liu J and Smitz J (1998b) Validation of a simplified culture system for primary mouse follicles by birth of live young. In Ares Serono Symposia, XI-th International Workshop on Development and Function of the Reproductive Organs. Artis Zoo, Amsterdam, The Netherlands de Bruin JP (2001) Development and Decline of the Human Ovarian Follicle Pool PhD Thesis, Utrecht University De Vet A, Laven JSE, Kramer P, Themmen APN and Fauser BCJM (2000) Stage-specific follicle expression of AMH in the human ovary Journal of Reproduction and Fertility Abstract Series 26 17 Dierich A, Sairam MR, Monaco L, Fimia GM, Gansmuller A, LeMeur M and Sassone-Corsi P (1998) Impairing follicle stimulating hormone (FSH) signaling in vivo: targeted disruption of the FSH receptor leads to aberrant gametogenesis and hormonal imbalance Proceedings National Academy of Sciences USA 95 13 612–13 617 Dong J, Albertini DF, Nishimori K, Kumar TR, Lu N and Matzuk MM (1996) Growth differentiation factor-9 is required during early ovarian folliculogenesis Nature 383 531–535 Donnez J and Bassil S (1998) Indications for cryopreservation of ovarian tissue Human Reproduction Update 4 248–259 Driancourt MA, Reynaud K, Cortvrindt R and Smitz J (2000) Roles of Kit and Kit ligand in ovarian function Reviews of Reproduction 5 143–152 Driancourt MA, Reynaud K and Smitz J (2001) Differences in follicular function of 3-month-old calves and mature cows Reproduction 121 463–474 Durlinger ALL, Kramer P, Karels B, de Jong FH, Uilenbroek JTJ, Grootegoed JA and Themmen APN (1999) Control of primordial follicle recruitment by anti-Müllerian hormone in the mouse ovary Endocrinology 140 5789–5796 Elvin JA, Yan C and Matzuk MM (2000) Oocyte-expressed TGF-β superfamily members in female fertility Molecular and Cellular Endocrinology 159 1–5 Eppig JJ and O’Brien M (1995) In vitro maturation and fertilization of oocytes isolated from aged mice: a strategy to rescue valuable genetic resources Journal of Assisted Reproduction and Genetics 12 269–273 *Eppig JJ and O’Brien MJ (1996) Development in vitro of mouse oocytes from primordial follicles Biology of Reproduction 54 197–207 Eppig JJ and O’Brien MJ (1998) Comparison of preimplantation developmental competence after mouse growth and development in vitro and in vivo. Theriogenology 49 415–422

Eppig JJ and Schroeder AC (1989) Capacity of mouse oocytes from preantral follicles to undergo embryogenesis and development to live young after growth, maturation, and fertilization in vitro. Biology of Reproduction 41 268–276 Eppig JJ and Wigglesworth K (1995) Factors affecting the developmental competence of mouse oocytes grown in vitro: oxygen concentration Molecular Reproduction and Development 42 447–456 Eppig JJ, Chesnel F, Hirao Y, O’Brien MJ, Pendola FL, Watanabe S and Wigglesworth K (1997) Oocyte control of granulosa cell development: how and why Human Reproduction 12 127–132 Figueiredo JR, Hulshof SCJ, Van den Hurk R, Ectors FJ, Fontes RS, Nusgens B, Bevers MM and Beckers JF (1993) Development of a new mechanical and enzymatical method for the isolation of intact preantral follicles from fetal, calf and adult bovine ovaries Theriogenology 40 789–799 Fortune JE, Cushman RA, Wahl CM and Kito S (2000) The primordial to primary follicle transition Molecular and Cellular Endocrinology 163 53–60 Gook D, Edgar DH and Stern C (1999) Effect of cooling rate and dehydration regimen on the histological appearance of human ovarian cortex following cryopreservation in 1,2-propanediol Human Reproduction 14 2061–2068 Gosden RG and Byath-Smith JG (1986) Oxygen concentration gradient across the ovarian follicular epithelium: model, predictions and implications Human Reproduction 2 65–68 Gosden RG, Baird DT, Wade JC and Webb R (1994) Restoration of fertility to oophorectomized sheep by ovarian autografts stored at –196⬚C Human Reproduction 9 597–603 Gospodarowicz D and Bialecki H (1979) Fibroblast and epidermal growth factors are mitogenic agents for cultured granulosa cells of rodent, porcine, and human origin Endocrinology 104 757–764 Gougeon A (1984) Caractères qualitatifs et quantitatifs de la population folliculaire dans l’ovaire humain adulte Contraception Fertilité Sexualité 12 527–535 Gougeon A (1986) Dynamics of follicular growth in the human: a model from preliminary results Human Reproduction 1 81–87 Gougeon A (1996) Regulation of ovarian follicular development in primates: facts and hypotheses Endocrine Reviews 17 121–155 Gougeon A and Busso D (2000) Morphologic and functional determinants of primordial and primary follicles in the monkey ovary Molecular and Cellular Endocrinology 163 33–41 Gougeon A and Chainy GBN (1987) Morphometric studies of small follicles in ovaries of women at different ages Journal of Reproduction and Fertility 81 433–442 *Gougeon A, Ecochard R and Thalabard JC (1994) Age-related changes of the population of human ovarian follicles: increase in the disappearance rate of non-growing and early-growing follicles in ageing women Biology of Reproduction 50 653–663 Greenwald GS and Moor RM (1989) Isolation and preliminary characterisation of pig primordial follicles Journal of Reproduction and Fertility 87 561–571 Gutierrez CG, Ralph JR, Telfer EE, Wilmut I and Webb R (2000) Growth and antrum formation of bovine preantral follicles in long-term culture in vitro. Biology of Reproduction 62 1322–1328 Hild-Petito S, West N, Brenner R and Stouffer R (1991) Localization of the androgen receptor in the follicle and corpus luteum of the primate ovary during the menstrual cycle Biology of Reproduction 44 561–568 Hirao Y, Nagai T, Kubo M, Miyano T, Miyake M and Kato S (1994) In vitro growth and maturation of pig oocytes Journal of Reproduction and Fertility 100 333–339 Hirshfield AN (1991) Theca cells may be present at the outset of follicular growth Biology of Reproduction 44 1157–1162 Hirshfield AN and DeSanti AM (1995) Patterns of ovarian cell proliferation in rats during the embryonic period and the first three weeks post partum Biology of Reproduction 53 1208–1221 Horie K, Takakura K, Fujiwawa H, Suginami H, Liao S and Mori T (1992) Immunohistochemical localization of the androgen receptor in the human ovary throughout the menstrual cycle in relation to oestrogen and progesterone receptor expression Human Reproduction 7 184–190 Hovatta O, Silye R, Krausz T, Abir R, Margara R, Trew G, Lass A and

Folliculogenesis in vitro Winston ML (1996) Cryopreservation of human ovarian tissue using dimethylsulphoxide and propanediol–sucrose as cryoprotectants Human Reproduction 11 1268–1272 Hovatta O, Silye R, Abir R, Krausz T and Winston RM (1997) Extracellular matrix improves the survival of human primordial and primary fresh and frozen–thawed ovarian follicles in long-term culture Human Reproduction 12 1032–1036 Hovatta O, Wright C, Krausz T, Hardy KK and Winston RML (1999) Human primordial, primary and secondary ovarian follicles in long-term culture: effect of partial isolation Human Reproduction 14 2519–2524 Hreinsson JG, Scott J, Rasmussen C, Swahn ML, Hsueh AJW and Hovatta O (2001) Growth differentiation factor-9 (GDF-9) promotes the growth, development and survival of human ovarian follicles in organ culture. In Final program and syllabus of the workshop Mammalian Oogenesis and Folliculogenesis: In Vivo and In Vitro Approaches ESHRE Campus 2001, Lisbon, Portugal, 6–7 April 2001 Hsu SY, Kubo M, Chun SY, Haluska FG, Housman DE ans Hsueh AJ (1995) Wilms’ tumor protein WT1 as an ovarian transcription factor: decreases in expression during follicular development and repression of inhibinalpha gene promotor Molecular Endocrinology 9 1356–1366 Hulshof SCJ, Dijkstra G, Van der Beek EM, Bevers MM, Figueiredo JR, Beckers JF and Van den Hurk R (1994) Immunocytochemical localisation of vasoactive intestinal peptide and neuropeptide Y in the bovine ovary Biology of Reproduction 50 553–560 *Johnson LD, Albertini DF, McGinnis LK and Biggers JD (1995) Chromatin organization, meiotic status and meiotic competence acquisition in mouse oocytes from cultured ovarian follicles Journal of Reproduction and Fertility 104 277–284 Kendall SK, Samuelson LC, Saunders TL, Wood RI and Camper SA (1995) Targeted disruption of the pituitary glycoprotein hormone alpha-subunit produces hypogonadal and hypothyroid mice Genes and Development 9 2007–2019 Kurilo LF (1981) Oogenesis in antenatal development in man Human Genetics 57 86–92 Lass A, Silye R, Abrams DC, Krausz T, Hovatta O, Margara R and Winston RML (1997) Follicular density in ovarian biopsy of infertile women: a novel method to assess ovarian reserve Human Reproduction 12 1028–1031 Lee BS, Yang WI, Park KH, Cho DJ, Song CH and Kim JW (1999) The expression of apoptosis related proteins, Bcl-2, Bax, Fas and Fas-ligand in the human ovarian follicles. In Frontiers in Endocrinology. Ovarian Function Research: Present and Future Vol. 21 pp 233–239 Eds S Fujimoto, EY Adashi, AJW Hsueh and JF Strauss. Serono Symposia Publications, Rome Li S, Maruo T, Ladines-Llave CA, Kondo H and Mochizuki M (1994) Stagelimited expression of myc oncoprotein in the human ovary during follicular growth, regression and atresia Endocrine Journals 41 83–92 Lintern-Moore S, Peters H, Moore GPM and Faber M (1974) Follicular development in the infant human ovary Journal of Reproduction and Fertility 39 53–64 Lundy T, Smith P, O’Connell A, Hudson NL and McNatty KP (1999) Populations of granulosa cells in small follicles of the sheep ovary Journal of Reproduction and Fertility 115 251–262 McGee EA and Hsueh AJW (2000) Initial and cyclic recruitment of ovarian follicles Endocrine Reviews 21 200–214 McGee E, Spears N, Minami S, Hsu SY, Chun SY, Billig H and Hsueh AJW (1997) Preantral ovarian follicles in serum-free culture: suppression of apoptosis after activation of the cyclic guanosine 3⬘,5⬘-monophosphate pathway and stimulation of growth and differentiation by folliclestimulating hormone Endocrinology 138 2417–2424 McNatty KP, Fidler AE, Juengel JL, Quirke LD, Smith PR, Heath DA, Lundy T, O’Connell A and Tisdall DJ (2000) Growth and paracrine factors regulating follicular formation and cellular function Molecular and Cellular Endocrinology 163 11–20 Magoffin DA and Erickson GF (1994) Control systems of theca-interstitial cells. In Molecular Biology of the Female Reproductive System pp 39–66 Ed. JK Findlay. Academic Press, San Diego, CA Maruo T (1995) Expression of oncogenes, growth factors and their receptors in follicular growth, regression and atresia: their roles in granulosa cell

201

proliferation and differentiation Nippon Sanka Fujinka Gakkai Zasshi 47 738–750 Meirow D, Fasouiliotis SJ, Nugent D, Schenker JG, Gosden RG and Rutherford AJ (1999) A laparoscopic technique for obtaining ovarian cortical biopsy specimens for fertility conservation in patients with cancer Fertility and Sterility 71 948–951 Mitchell LM, Kennedy CR and Hartshorne GM (1999) Collection and culture of human primordial and primary follicles arising from ovarian aspiration Molecular and Cellular Endocrinology 163 149 (Abstract) Mizunuma H, Liu X, Andoh K, Abe Y, Kobayashi J, Yamada K, Yokota H and Ibuki Y (1999) Activin from secondary follicles causes primary follicles to remain dormant at the preantral stage. In Frontiers in Endocrinology. Ovarian Function Research: Present and Future Vol. 21 pp 241–247 Eds S Fujimoto, EY Adashi, AJW Hsueh and JF Strauss. Serono Symposia Publications, Rome Nayudu PL and Osborn SM (1992) Factors influencing the rate of preantral and antral growth of mouse ovarian follicles in vitro. Journal of Reproduction and Fertility 95 349–362 Newton H (1998) The cryopreservation of ovarian tissue as a strategy for preserving the fertility of cancer patients Human Reproduction Update 4 237–247 *Newton H, Aubard Y, Rutherford A, Sharma V and Gosden R (1996) Low temperature storage and grafting of human ovarian tissue Human Reproduction 11 1487–1491 Newton H, Picton H and Gosden RG (1999) In vitro growth of oocyte– granulosa cell complexes isolated from cryopreserved ovine tissue Journal of Reproduction and Fertility 115 141–150 Nugent D, Newton H, Gosden RG and Rutherford AJ (1998) Investigation of follicle survival after human heterotopic autografting Human Reproduction 13 1998 (Abstract) Ojeda SR, Dissen GA and Romero C (1999) Role of neurotrophic factors in the control of ovarian development. In Frontiers in Endocrinology. Ovarian Function Research: Present and Future Vol. 21 pp 171–179 Eds S Fujimoto, EY Adashi, AJW Hsueh and JF Strauss. Serono Symposia Publications, Rome Ojeda SR, Romero C, Tapia V and Dissen GA (2000) Neurotrophic and cell–cell dependent control of early follicular development Molecular and Cellular Endocrinology 163 67–71 Oktay K, Nugent D, Newton H, Salha O, Chatterjee P and Gosden RG (1997) Isolation and characterization of primordial follicles from fresh and cryopreserved human ovarian tissue Fertility and Sterility 67 481–486 Oktay K, Newton H and Gosden RG (2000) Transplantation of cryopreserved human ovarian tissue results in follicle growth initiation in SCID mice Fertility and Sterility 73 599–603 *O’Shaughnessy PJ, McLelland D and McBride MW (1997) Regulation of luteinizing hormone-receptor and follicle-stimulating hormone-receptor messenger ribonucleic acid levels during development in the neonatal mouse ovary Biology of Reproduction 57 602–608 Packer AI, Hsu YC, Besmer P and Bacharova RF (1994) The ligand of the ckit receptor promotes oocyte growth Developmental Biology 161 194–205 Parrott JA and Skinner MK (2000) Kit ligand actions on ovarian stromal cells: effect on theca cell recruitment and steroid production Molecular Reproduction and Development 55 55–64 Pedersen T and Peters H (1968) Proposal for the classification of oocytes and follicles in the mouse ovary Journal of Reproduction and Fertility 17 555–557 Picton HM, Mkandla A, Salha O, Wynn P and Gosden RG (1999) Initiation of human primordial follicle growth in vitro in ultrathin slices of ovarian cortex Human Reproduction 14 Abstract Book 1 11 Ralph JH, Wilmut I and Telfer EE (1995) In vitro growth of bovine preantral follicles and the influence of FSH on follicle and oocyte diameter Journal of Reproduction and Fertility 15 6–7 Reynaud K and Driancourt MA (2000) Oocyte attrition Molecular and Cellular Endocrinology 163 101–108 Rose UM, Hanssen RG and Kloosterboer HJ (1999) Development and characterisation of an in vitro ovulation model using mouse ovarian follicles Biology of Reproduction 61 503–511

202

J. E. J. Smitz and R. G. Cortvrindt

Roy SK and Treacy BJ (1993) Isolation and long-term culture of human preantral follicles Fertility and Sterility 59 783–790 Sato A, Bo M, Maruo T, Yoshida S and Mochizuki M (1994) Stage-specific expression of c-myc messenger ribonucleic acid in porcine granulosa cells early in follicular growth European Journal of Endocrinology 131 319–322 Schultea TD, Dees WL and Ojeda SR (1992) Postnatal development of sympathetic and sensory innervation of the rhesus monkey ovary Biology of Reproduction 47 760–767 Sforza C, Ferrarion VF, dePol A, Marzona L, Forni M and Forabosco A (1993) Morphometric study of the human ovary during compartmentalization Anatomical Record 236 626–634 Shaw JM, Bowles J, Koopman P, Wood EC and Trounson AO (1996) Fresh and cryopreserved ovarian tissue samples from donors with lymphoma transmit the cancer to graft recipients Human Reproduction 11 1668–1673 Shaw JM, Dawson KJ and Trounson AO (1997) A critical evaluation of ovarian tissue cryopreservation and grafting as a strategy for preserving the human female germ line Reproduction of Medical Reviews 6 163–183 Smitz J and Cortvrindt R (1998) Effects of recombinant activin A on in vitro culture of mouse preantral follicles Molecular Reproduction and Development 50 294–304 Smitz J and Cortvrindt R (1999) Preservation of gonadal function after cancer therapy in women. In Frontiers in Endocrinology. Ovarian Function Research: Present and Future Vol. 21 pp 367–374 Eds S Fujimoto, EY Adashi, AJW Hsueh and JF Strauss. Serono Symposia Publications, Rome Smitz J, Cortvrindt R and Van Steirteghem A (1996) Normal oxygen atmosphere is essential for the solitary long-term culture of early preantral mouse follicles Molecular Reproduction and Development 45 466–475 Sokka TA, Hamalainen TM, Kaipia A, Warren DW and Huhtaniemi IT (1996) Development of luteinizing hormone action in the perinatal rat ovary Biology of Reproduction 55 663–670 Telfer EE, Binnie JP, McCaffery FH and Campbell BK (2000) In vitro development of oocytes from porcine and bovine primary follicles Molecular and Cellular Endocrinology 163 117–123 Tilly JL, Kowalsky KI, Johnson AL and Hsueh AJW (1991) Involvement of apoptosis in ovarian follicular atresia and postovulatory regression Endocrinology 129 2799–2801 Van den Broecke R, Van der Elst J, Liu J, Hovatta O and Dhont M (2001) The female-to-male transsexual patient: a source of human ovarian cortical tissue for experimental use Human Reproduction 16 145–147 Van den Hurk R, Spek ER, Hage WJ, Fair T, Ralph JH and Schotanus K (1998) Ultrastructure and viability of isolated bovine preantral follicles Human Reproduction Update 4 833–841 Van Wezel IL and Rodgers RJ (1996) Morphological characterisation of bovine primordial follicles and their environment in vivo. Biology of Reproduction 55 1003–1011 Van Wezel IL, Umapathysivain K, Tilley WD and Rodgers RJ (1995) Immunohistochemical localization of basic fibroblast growth factor in bovine ovarian follicles Molecular and Cellular Endocrinology 115 133–140

Vendola KA, Zhou J, Adesanya OO, Weil SJ and Bondy CA (1998) Androgens stimulate early stages of follicular growth in the primate ovary Journal of Clinical Investigation 1010 2622–2629 Vendola KA, Zhou J, Wang J, Famuyiwa OA, Bievre M and Bondy CA (1999) Androgens promote oocyte insulin-like growth factor I expression and initiation of follicle development in the primate ovary Biology of Reproduction 61 353–357 Wandji SA, Srsen V, Nathanielsz PW, Eppig JJ and Fortune JE (1997) Initiation of growth of baboon primordial follicles in vitro. Human Reproduction 12 1993–2001 Weil SJ, Vendola K, Zhou J, Adesanya OO, Wang J, Okafor J and Bondy CA (1998) Androgen receptor gene expression in the primate ovary: cellular localization, regulation, and functional correlations Journal of Clinical Endocrinology 83 2479–2485 Weil S, Vendola K, Zhou J and Bondy CA (1999) Androgen and folliclestimulating hormone interactions in primate ovarian follicle development Journal of Clinical and Endocrinology Metabolism 84 2951–2956 Wright CS, Hovatta O, Margara R, Trew G, Winston RML, Franks S and Hardy K (1999) Effects of follicle-stimulating hormone and serum substitution on the in vitro growth of human ovarian follicles Human Reproduction 14 1555–1562 Wu J, Zhang L and Liu P (1998) A new source of human oocytes: preliminary report on the identification and maturation of human preantral follicles from follicular aspirates Human Reproduction 13 2561–2563 Yeh J and Adashi EY (1999) The ovarian life cycle. In Reproductive Endocrinology: Physiology, Pathophysiology and Clinical Management 4th Edn pp 153–190 Eds SSC Yen, RB Jaffe and RL Barbieri. Sanders Company, Philadelphia Yeh J and Osathanondh R (1993) Expression of messenger ribonucleic acids encoding for basic fibroblast growth factor (bFGF) and alternatively spliced FGF receptor in human fetal ovary and uterus Journal of Clinical Endocrinology and Metabolism 77 1367–1371 Yeung SCJ, Chiu AC, Vassilopoulou-Sellin R and Gagel RF (1998) The endocrine effects of nonhormonal antineoplastic therapy Endocrine Reviews 19 144–172 Yokota H, Yamada K, Liu X, Kobayashi J, Abe Y, Mizunuma H and Ibuki Y (1997) Paradoxical action of activin A on folliculogenesis in immature and adult mice Endocrinology 138 4572–4576 Yoshida H, Takakura N, Kataoko H, Kunisada T, Okamura H and Nishikawa SI (1997) Stepwise requirement of ckit tyrosine kinase in mouse ovarian follicle development Developmental Biology 184 122–137 Yuan W and Giudice LC (1997) Programmed cell death in human ovary is a function of follicle and corpus luteum status Journal of Clinical Endocrinology and Metabolism 82 3148–3155 Zhang FP, Poutanen M, Wilbertz J and Huhtaniemi I (2001) Normal prenatal but arrested postnatal sexual development of luteinizing hormone receptor knockout (LuRKO) mice Molecular Endocrinology 15 172–183 Zhang J, Liu J, Xu KP, Liu B and Dimattina M (1995) Extracorporeal development and ultrarapid freezing of human fetal ova Journal of Assisted Reproduction and Genetics 12 361–368