Novel approaches to the development and assessment of an ovine model of polycystic ovary syndrome

Abstract

Polycystic ovary syndrome (PCOS) is a common reproductive, endocrine and metabolic disorder present in women of reproductive age. Despite the widespread prevalence and heritability of PCOS, the heterogeneous and polygenic traits have made the successful identification of candidate genes difficult. Animal models have been developed on the premise that early exposure to sex steroids can programme epigenetic changes that predispose the fetus to the adult features of PCOS. Past research has modelled ovarian dysfunction, endocrine abnormalities and metabolic perturbances in rodent, non-human primate and sheep PCOS models, through the enhanced neonatal or prenatal exposure to the male sex hormone, testosterone. The modelling of PCOS in a large domestic species such as the sheep is advantageous due to similar biological reproductive function as the human. In this regard the sheep has been extensively used to model PCOS by the treatment of pregnant ewes from early to midgestation with androgens such as testosterone propionate (TP). These experiments have demonstrated the fetal programming effects of androgens on offspring that go on to develop PCOS-like characteristics in adulthood. One of the caveats of assessing steroid effects in this way is the effect of the placenta in mediating the transfer of these hormones. TP is an aromatisable androgen and thus some of its effects in the fetus may be attributable to placental by-products such as estrogens. This thesis describes the development and assessment of a novel model of prenatal androgenisation. Two models were compared: the indirect maternal exposure to TP (the current model) and the direct fetal injection of TP. In directly treating the fetus this allowed control over the dose of TP administered and avoidance of secondary effects that androgens may exert in the mother that could be transferred to the fetus. For the maternal model, pregnant Scottish Greyface ewes were administered TP twice weekly from day (d)62-102 of a 147 day gestation. For the fetal model, fetuses were injected twice while the ewe was anaesthetised with graded doses of TP during the same period of treatment as the maternal model. The effects of prenatal androgenisation were assessed in the female fetus shortly after treatment and also in young adult sheep. Fetal ovarian and adrenal steroidogenic gene expression was monitored and found to be altered in response to elevated levels of sex steroids. At d90 the morphology of the developing ovary was not changed by prenatal androgens. In the adult a detailed ovarian and endocrine assessment was undertaken, by examination of ovarian morphology, hormone levels, ovulatory cycles, hypothalamic pituitary ovarian function and follicle steroidogenesis, during the first breeding season. In addition, the metabolic effects of prenatal androgens were monitored by measuring body fat, insulin and glucose homeostasis and liver function. Neither maternal nor fetal prenatal androgenisation during mid-gestation resulted in a perturbed hormonal milieu or polycystic ovaries in young adults. These treatments did however programme a clear ovarian phenotype demonstrated by the increased capacity of follicles to secrete androgens, independently of an abnormal endocrine environment and disordered folliculogenesis. Furthermore, animals that were exposed maternally to TP developed fatty liver and had increased insulin secretion in response to glucose load. A major outcome of this study was the finding that the fetally injected control animals were phenotypically different than the maternal control animals. In fact, some of the reproductive and metabolic features of maternal TP exposure were found in the fetal control group. This unexpected finding has raised the possibility that it is the fetal exposure to stress, that is secondary to elevated maternal androgens, rather than androgens per se that is responsible for at least some of the multitude of anomalies encountered in PCOS.