We have previously demonstrated that ovarian function and fertility can be preserved in sheep after castration by autotransplantation of cryopreserved strips of ovarian cortex. In the current experiments we have investigated the long term survival of such grafts by detailed measurements of ovarian function for a period of nearly 2 yr after autotransplantation. After ovariectomy and transplantation of frozen/thawed grafts, the concentrations of FSH and LH rose to castrate levels for about 14 weeks before falling gradually to reach near-normal levels at about 60 weeks. In the breeding season from October 1994 to March 1995, all ewes had 5-10 estrous cycles that were similar in length to those in the 4 control ewes. Luteal function as indicated by the progesterone concentration was identical before and 11 months after transplantation. In contrast, the basal concentrations of FSH and LH were persistently raised throughout the luteal phase, but showed a normal decline during the follicular phase. The concentration of inhibin A in ovarian venous plasma measured at the end of the experiment 22 months after transplantation was significantly lower than that in control ewes (mean +/- SE, 409 +/- 118 vs. 1914 +/- 555 pg/ml; P < 0.004). Transplantation of frozen/thawed ovarian tissue to SCID mice demonstrated that about 28% of primordial follicles survived the procedure. All of the ovaries transplanted into sheep contained large antral follicles and/or cysts, but very few primordial oocytes when recovered at autopsy after 22 months. These results demonstrate that despite a drastic reduction in the total number of primordial follicles, cyclical ovarian function is preserved in sheep after autotransplantation of frozen/thawed ovarian tissue and provide experimental confirmation that such a technique could provide a means of preserving fertility in women undergoing chemo- or radiotherapy for malignant disease.

The paper presents an update of our 1993 model of ovarian follicular development in ruminants, based on knowledge gained from the past 15 years of research. The model addresses the sequence of events from follicular formation in fetal life, through the successive waves of follicular growth and atresia, culminating with the emergence of ovulatory follicles during reproductive cycles. The original concept of five developmental classes of follicles, defined primarily by their responses to gonadotrophins, is retained: primordial, committed, gonadotrophin-responsive, gonadotrophin-dependent and ovulatory follicles. The updated model has more extensive integration of the morphological, molecular and cellular events during folliculogenesis with systemic events in the whole animal. It also incorporates knowledge on factors that influence oocyte quality and the critical roles of the oocyte in regulating follicular development and ovulation rate. The original hypothetical mechanisms determining ovulation rate are retained but with some refinements; the enhanced viability of gonadotrophin-dependent follicles and increases in the number of gonadotrophin-responsive follicles by increases in the throughput of follicles to this stage of growth. Finally, we reexamine how these two mechanisms, which are thought not to be mutually exclusive, appear to account for most of the known genetic and environmental effects on ovulation rate.

A model for folliculogenesis is proposed that is based as far as possible on a knowledge of physiological, rather than anatomical, changes taking place during follicle development. The model is therefore functional, rather than descriptive, and consists of five classes of follicles that have been defined by their dependency and sensitivity to gonadotrophins. These classes are: primordial, committed, gonadotrophin-responsive, gonadotrophin-dependent and ovulatory. The model is an attempt to encourage discussion and to promote the integration of morphological models of folliculogenesis with recent advances in the molecular endocrinology of the ovarian follicle. Two hypotheses for the mechanisms that determine ovulation rate are developed in light of the model. In the first, multiple ovulation results when the viability of gonadotropin-dependent follicles is enhanced. In the second, multiple ovulation is caused by increasing the number of gonadotrophin-responsive follicles available for further development; this results from the increasing rate of folliculogenesis and the throughput of follicles. The final section of this paper examines how these two hypothetical mechanisms, which are not mutually exclusive, appear to account for most of the known genetical and environmental effects on ovulation rate of sheep. In particular, the effects of nutrition, genotype, exogenous gonadotrophins, immunity to both oestrogens and androgens, and immunity to inhibin are discussed.