The presence of intra-ovarian nerves was reported more than a century ago and successive investigations have demonstrated that the mammalian ovary is innervated by both sympathetic and sensory fibres distributed to the different compartments of the gland, such as blood vessels, ovarian stroma and follicle wall. Despite the extensive ovarian innervation and the experimental evidence indicating that neuromediators can influence sex steroid production, the role of the nervous system in the control of ovarian activity is still largely unknown.

Adducin is a cytoskeletal protein that can function in vitro to bundle F-actin and to control the assembly of the F-actin/spectrin cytoskeletal network. The Drosophila Adducin-like (Add) locus (also referred to as hu-li tai shao (hts)) encodes a family of proteins of which several are homologous to mammalian adducin (Ding et al., Proc. Natl. Acad. Sci. USA90, 2512–16, 1993; Yue & Spradling, Genes Dev.6, 2443–54, 1992). We report the identification of two novel adducin isoforms: a 95 × 103Mr form (ADD-95) and an 87 × 103Mr form (ADD-87). We present a detailed analysis of the distribution patterns of ADD-95 and ADD-87 during oogenesis and embryogenesis. The isoforms are co-expressed in several cell- and tissuetypes; however, only ADD-87 is present in mid- to late-stage oocytes. ADD-87 is present throughout the oocyte cortex at stages 9 and 10 of oogenesis but is detectable only at the anterior pole from stage 11 onward, correlated with localisation of Add-hts mRNA first to the cortex and then to the anterior pole of the oocyte. ADD-87 co-localises with F-actin and spectrin in the cortex of the oocyte through stage 10 of oogenesis, consistent with a possible role in cytoskeletal assembly or function predicted by mammalian studies.

In each oestrous cycle only a limited number of follicles are selected for ovulation whereas the remaining majority undergo atresia. The earliest and most prominent feature of atresia is the death of granulosa cells. Recent biochemical evidence has demonstrated that granulosa cell death during follicular atresia in swine (Tilly et al., 1992), bovine (Jolly et al., 1994) and rodent (Tilly et al., 1991) ovaries occurs by apoptosis, a process whereby cells die in a controlled manner. A biochemical event considered to be characteristic of apoptotic cell death is the intranucleosomal cleavage of genomic DNA into fragments 180–200 bp in size, which separate into a distinctive ladder-like pattern on agarose gel electrophoresis. Detection of this pattern of oligonucleosomes in DNA provides a marker of apoptotic cell death.

In mammals the ability of an oocyte to become fertilised is the result of a complex process occurring within the ovarian follicle which depends on the stagespecific expression of oocyte genes and the presence of granulosa cells (for a review see Buccione et al., 1990a). The coordinated development of germinal and somatic components of the follicle is regulated by two principal systems of interaction, based on the presence of gap junctions and on the production of paracrine factors. Gap junctions link granulosa cells to each other and to the oocyte (Anderson & Albertini, 1976), and represent a major route for the transfer of small molecules involved in oocyte metabolism (for a review see Mangia et al., 1992) and regulation of the arrest and resumption of meiosis (for a review see Eppig, 1993). The production of paracrine factors by granulosa cells has been suggested by the findings that these cells express the production of the Steel locus, the Steel factor (SLF) or kit ligand (KL; Motro et al., 1991; Manova et al., 1993), and that this factor promotes oocyte growth in vitro when used at high concentrations (Packer et al., 1994). Since KL is too large to be transmitted through gap junctions, it must necessarily be released in the extracellular environment before binding to the c-kit receptor present on oocyte membrane (Manova et al., 1990; Horie et al., 1991).

Nerve growth factor (NGF) belongs to a family of related target-derived proteins required for the survival and development of discrete neuronal populations in the central and peripheral nervous systems (Levi-Montalcini, 1987; Snider, 1994). Although initial observations led to the conclusion that the biological actions of neurotrophins are restricted to the nervous system (Thoenen, 1991; Raffioni et al. 1993), new evidence suggests that they, and in particular NGF, can also affect non-neuronal cells.

In a Graafian follicle, granulosa cells are classified into two principal cell subpopulations: cumulus cells, which are closely associated with the oocyte to form the cumulus cell-oocyte complex (COC), and mural granulosa cells, which are organised as a stratified epithelium at the periphery of the follicle. Following the preovulatory gonadotropin surge, cumulus cells lose contact with mural granulosa cells and start to synthesise and secrete a large amount of hyaluronan (HA), a glycosaminoglycan with high molecular weight and large hydrodynamic domains (Salustri et al., 1992). Proteins derived from serum (Chen et al., 1992, 1994) and synthesised by cumulus cells (Camaioni et al., 1993, 1996) organise the strands of HA into an intercellular elastic network that traps the cumulus cells and the oocyte in a unit which can not be mechanically dissociated – a process also referred to as cumulus expansion. At ovulation, the expanded COC is released through the ruptured follicle wall and transferred to the oviduct. The matrix in the expanded COC facilitates its extrusion from the follicle and its capture by oviductal fimbria, and provides, together with the cumulus cells, a suitable microenvironment for sperm penetration and fertilisation (for references see Salustri et al., 1993).

Investigations of strains of mice defective in germ cell development have revealed the importance of oocytes for the initial stages of folliculogenesis (Pellas et al., 1991; Huang et al., 1993). Various aspects of follicular development are dependent upon and/or influenced by the presence of oocytes, including granulosa cell proliferation (Vanderhyden et al., 1990, 1992) and cumulus expansion (Buccione et al., 1990; Salustri et al., 1990; Vanderhyden et al., 1990; Vanderhyden, 1993). We are investigating the possibility that oocytes influence one of the primary functions of granulosa cells: steroidogenesis. In many species, granulosa cells removed from preovulatory follicles luteinise in vitro (Channing et al., 1982), presumably due to loss of contact with follicular luteinisation inhibitory factor(s). Indeed, follicular fluid can prevent granulosa cell luteinisation in vitro (Ledwitz-Rigby et al., 1977). Follicular fluid, however, may simply be the medium for transport of factors secreted by oocytes to regulate granulosa cell activities.

Each oestrous cycle the ovary produces one or more oocytes, depending on the species considered, which are fully competent to sustain the development of a new individual and are defined as matured. Ovulation is the terminal phase of a lengthy and complex selection process which enables only a minute proportion of the oocytes present in the ovary to be ovulated and possibly fertilised. The use of exogenous gonadotrophins for the pharmacological stimulation of the ovaries has clearly indicated that oocytes naturally excluded by the selection process can also be matured and, if fertilised in vivo or in vitro, can generate normal offspring.

The ovulatory follicle has the dual role of releasing a viable oocyte capable of fertilisation and the production of key endocrine signals that result in mating behaviour, the induction of the preovulatory LH surge and ovulation. Further, following ovulation the ovulatory follicle must be capable of forming a viable corpus luteum if pregnancy is to be maintained. Follicle growth is therefore a developmental process during which the follicle progressively acquires a number of properties, each of which is an essential prerequisite for further development. Failure to acquire these properties at the correct time and in an exact sequence will lead to failure of the developmental process and to the deterioration of the follicle through atresia and degeneration of the oocyte.

As meiosis is initiated and the oogonium is transformed into a primary oocyte, the female germ cell becomes intimately invested by a single squamous layer of sex cord epithelium. As the follicle cell population expands during the initial stages of the ovarian cycle, oocyte and follicle cells become increasingly connected to one another by one of the most extensive populations of gap junctions documented in any epithelium (reviewed in Larsen & Wert, 1988).

Much effort has been focused on establishing optimal conditions for obtaining in vitro maturation of oocytes from different species with results comparable to those achieved after in vivo development (reviewed by Brackett, 1992). However, even though extraordinary progress has been made, the in vitro technology for oocyte maturation lags far behind that in vivo and improvements are needed to increase the quantity and quality of the embryos produced from these matured oocytes.

During embryogenesis germ cells originate from primordial germ cells (PGCs). The development of mammalian PGCs involves a number of complex events (formation and segregation of PGC precursors, PGC migration and proliferation) which lead to the differentiation of oocytes or prospermatogonia (for a review see De Felici et al., 1992). During recent years developments in methods for isolation, purification and culture of mouse PGCs have led to significant progress in the understanding of molecular mechanisms of migration, proliferation and differentiation of these cells (for reviews see De Felici, 1994; and De Felici & Pesce, 1994a). In this paper we describe the key role played by stem cell factor (SCF) in PGC development and early folliculogenesis.