Do Animal Models of Polycystic Ovary Syndrome Help to Understand Its Pathogenesis and Management? Yes, but Their Limitations Should be Recognized

Polycystic ovary syndrome (PCOS) is the commonest endocrinopathy in women and is a major cause of physical and emotional morbidity (1, 2). Not only does it have profound implications for fertility, but it also causes hirsutism, acne, and alopecia and, very importantly, is now recognized to have a significant metabolic phenotype that is associated with increased risk of type 2 diabetes and cardiovascular disease (3, 4). PCOS is, by nature, a heterogeneous disorder, and the underlying cause (or causes) remains unclear. There is firm evidence for a strong genetic component to the syndrome (5–7), but environmental factors, particularly dietary, are also important. Obese women with PCOS are more symptomatic and have a more abnormal endocrine and metabolic profile than lean women with the syndrome (8), features that are closely related to the exaggerated insulin resistance and hyperinsulinemia (compared with women with simple obesity) in these subjects (3). In this issue of Endocrinology, Shi and colleagues (9) have described a rodent model that may represent a paradigm for study of cardiovascular risk factors in association with PCOS. The JCR:LA-cp rat has a defect in the leptin receptor that renders it hyperphagic with a predisposition to obesity. Adult cp/cp animals are significantly fatter than wild-type littermates and have indices of insulin resistance, dyslipidemia, and elevated serum concentrations of testosterone with associated ovarian dysfunction (irregular estrus cycles). They have ovarian follicular abnormalities that, in some respects, resemble polycystic ovaries in humans; i.e. they have multiple cystic follicles. In essence, this is a genetically obese animal in which the associated metabolic abnormalities appear to lead to ovarian dysfunction. This raises the question of whether, in women, simple obesity and the resultant insulin resistance/hyperinsulinemia can be a primary cause of PCOS. This seems unlikely in the majority of women with PCOS. Certainly rare genetic disorders such as insulin receptor mutations can result in ovarian hyperandrogenism (10), but it is more debatable whether simple obesity can cause PCOS in the absence of an underlying ovarian abnormality. Adolescent girls who are obese have higher than normal androgen levels (11) and disturbed gonadotropin secretion (12), but it is not clear whether these girls have an underlying predisposition to ovarian hyperandrogenism. It is more likely that obesity and hyperinsulinemia act to amplify preexisting ovarian dysfunction. This is most clearly evident in women, known to have polycystic ovaries but who may have mild symptoms and moderate hyperandrogenemia, whose symptoms and biochemical abnormalities worsen after weight gain (2, 4). Thus the heterogeneity of the syndrome is further complicated by the ability of diet to change the phenotype, even within individuals. It is hardly surprising, therefore, that there remains controversy about the diagnostic criteria for PCOS (13, 14). Nevertheless, even in women with the classic form of PCOS (i.e. those with both oligo-/anovulation and androgen excess, who are most likely to have insulin resistance), there are considerable variations in metabolic features. Although most obese women in this category are insulin resistant, lean women with anovulation and androgen excess may have normal insulin sensitivity (15). Thus, overall, it appears that in human PCOS, obesity and insulin resistance act as important amplifying factors in PCOS rather than as the primary cause. A number of animal models for PCOS have emerged over several decades, including rodents, sheep, and primates. The advantages of using the rat or mouse include the ability to provide a constant genetic background and controlled environment as well as capitalizing on the short life cycle to explore both reproductive and long-term metabolic effects. However, interpretation of any ovarian abnormalities has to be made in the light of the differences between rodents and primates in ovarian follicular formation and recruitment and in the ovulation cycle. Multiple cystic follicles in anovulatory rodents do not necessarily equate, in morphology or function, to those in women with polycystic ovaries. Nevertheless, the rodent may still be a useful model to study, for example, the specific effects of androgens on early follicle development, on control of gonadotropin secretion (16), and, in particular, to examine the metabolic effects of exposure to excess androgen (17, 18). As in the Rhesus monkey and sheep (see below), the administration of large doses of androgen either to pregnant females (prenatal androgenization) or in postnatal life may result in the very metabolic abnormalities that are characteristic of many women with PCOS. The prenatally androgenized (PA) sheep and Rhesus monkey have provided considerable insight into the role of androgen in genesis of both reproductive and metabolic abnormalities (19). These findings support the hypothesis that exposure to excess androgen during key windows of development (including in utero) lead to many of the features of the syndrome in adulthood. In common with humans (and in contrast to rodents), both are monoovulatory species that have completed ovarian follicle formation during fetal life. The PA Rhesus monkey during adulthood displays abnormal ovarian morphology, ovarian hyperandrogenism (and adrenal androgen excess), hypersecretion of LH, insulin resistance, and anovulation in relation to increased body weight (20–22). Studies to date in the PA sheep have yielded similar findings with respect to reproductive and metabolic sequelae (23). Indeed, PA models, whether using rodents, sheep, or primates, probably best relate to endocrine and metabolic sequelae of PCOS but are less informative in terms of the primary defect. We are still left with the question: where does excess androgen come from and when? There is evidence for a primary ovarian defect in human PCOS, but there is no very firm evidence that the typical ovarian abnormalities of PCOS (24), aberrant preantral as well as antral follicle development or stromal hyperplasia, can accurately be reproduced in any animal model. Nevertheless, early preantral follicle development has been shown to be abnormal in the PA sheep (25, 26), suggesting that androgens may have a role in the etiology of this salient feature of the polycystic ovary. Interestingly, abnormalities of ovarian cyclicity in the ewe can be exaggerated by excessive weight gain (27), mimicking the observations in women with PCOS. Excess androgen may be both the consequence and the cause of the ovarian abnormality. We have proposed that PCOS has its origins in early life and that the polycystic ovary is genetically predisposed to hypersecrete androgens, certainly at puberty but also in the early postnatal period or even during fetal development (20, 28). The dogma has been that the fetal ovary is not steroidogenically active, but recent data indicate that theca cells are present in the human fetal ovary and clearly express 17-hydroxylase, the key enzyme in androgen biosynthesis (29). Excess androgen production by the ovary in human fetal or adult life may, in turn, contribute to ovarian follicular abnormalities in adulthood. In summary, animal models are very useful in understanding the pathogenesis of PCOS (and particularly its metabolic consequences), but we cannot expect any one model to provide all the answers; each has its advantages and disadvantages. To date, the most informative models seem to be those that involve prenatal exposure to excess androgen, suggesting that androgen programs the development of PCOS in adult life. Key questions remain: where (and when) does the disorder start in women with PCOS, what predisposes to metabolic disturbance, and what interventions can be offered (and when) to reduce the risk of diabetes and cardiovascular disease in later life? Disclosure Summary: The author has no conflict of interest to declare. Prenatally androgenized; polycystic ovary syndrome.