Animal Models and the Developmental Origins of Polycystic Ovary Syndrome: Increasing Evidence for the Role of Androgens in Programming Reproductive and Metabolic Dysfunction

Polycystic ovary syndrome (PCOS) has been recognized as an important endocrine disorder for more than 70 yr, and its prevalence in the modern population is estimated to be between 5 and 10% of women of reproductive age (1). It is by far the most common cause of oligo- or anovulation and accounts for more than 75% of cases of anovulatory infertility (2, 3). It is also the leading cause of hirsutism, androgenetic alopecia, and persistent acne, all of which are distressing and debilitating disorders in women. In the last 30 yr, increasing emphasis has been placed on metabolic dysfunction in women with PCOS. Insulin resistance and compensatory hyperinsulinemia are central to the metabolic abnormalities of PCOS (3, 4), and the syndrome is estimated to carry a 2- to 4-fold increase in the risk of developing type 2 diabetes mellitus (5). The increasing pandemic of obesity threatens to make that figure an underestimate; obesity has a preferentially adverse effect on metabolic function in women with PCOS compared with similarly obese women without PCOS. Cardiovascular risk factors are also significantly increased in PCOS, although it remains unclear whether the incidence of cardiovascular events is similarly increased (5). Despite the plethora of clinical and experimental data regarding this very common endocrine disorder, our knowledge of the etiology of the syndrome is, at best, incomplete. There is strong evidence for a genetic cause (6–9), but environmental factors, particularly the effect of diet, clearly play an important part in the presentation and severity of PCOS if not in the origin of the disorder itself (10). One important strategy for better understanding the etiology of PCOS has been the use of animal models. In general, the main principle underlying studies in animals has been that many of the characteristic features of PCOS arise as a result of exposure to excess androgen, the biochemical hallmark of PCOS. The observation that androgen treatment of immature rats results in abnormal antral follicle development and disrupted ovulation was made in the mid-1970s (11). Since then, there have been many studies of the effects of excess androgen on ovarian function in rodents (12–15). The relevance of rodent models in understanding the ovarian disorders in PCOS is limited, however, by the fact that ovarian development (and the nature of ovulatory cycles) in rodents differs markedly from that in women. Rats and mice are poly-ovulators and women mono-ovulators. Another important aspect of rodent models is the effect of timing of androgen exposure on reproductive phenotype. Tyndall and colleagues (16) have recently shown that whereas testosterone treatment has minimal effect on adult ovarian function if given in late postnatal life, treatment in both fetal and postnatal life causes streak ovaries in rats. Over the last 15 yr, studies of larger species as models for PCOS have been explored and have given significant insight into the genesis of PCOS, particularly the notion that the syndrome has its origins during intrauterine development (17). Here the studies of Abbott and colleagues (17–19) in the rhesus monkey and of Padmanabhan and co-workers (20–24) in the sheep have proved particularly informative with regard to understanding the role of androgens in development of both reproductive and metabolic disturbances in women. Animals exposed to high levels of androgen in utero (by giving very large doses of androgen to the mothers) develop, as adults, disrupted ovarian cycles and abnormalities of early follicle development (22, 25) that mimic those observed in women with PCOS (26, 27). Importantly, these animals also develop metabolic abnormalities, including insulin resistance and impaired glucose tolerance, again reminiscent of those seen in women with PCOS. In a recent variation of this protocol, Duncan and colleagues (28) have shown that direct fetal exposure to exogenous androgen also results in metabolic abnormalities, including hepatic dysfunction. Thus, the hypothesis of androgen programming, in utero or during postnatal development, being an important aspect of the origin of PCOS has gained currency. It has also become clear that in vivo processing of testosterone to either dihydrotestosterone or estradiol may produce differential effects on reproductive and metabolic pathways (23). A recent study in the rhesus monkey suggests that, as might be deduced, developmental programming by androgens involves epigenetic regulation (29), an area which would benefit from further research. Of course, two of the many questions that are yet to be answered regarding the etiology of PCOS, are where, in the human female, does the excess androgen come from and when are its effects exerted? It is very unlikely that maternal androgen excess in human pregnancy affects the fetus because of the twin barriers of high levels of SHBG and placental aromatase activity. One hypothesis is that the ovary is genetically predisposed to hypersecrete androgens, a function that may be manifest during fetal life (30) but may also be triggered during temporary activation of LH secretion during infancy or, more probably, during reawakening of the hypothalamic-pituitary-gonadal axis during puberty (17) (fig 1). Proposed role of testosterone in the developmental origin of PCOS in women. On the basis of clinical studies and animal models, it is suggested that exposure to excess androgens at or before puberty leads to downstream abnormalities of reproductive and metabolic function. In the case of women with PCOS, the increased androgen exposure may result from a genetic predisposition of the ovary to hypersecrete androgens. Additional development of the features of PCOS is determined by environmental factors (particularly dietary) and other genetic influences on adipose biology and glucose/insulin homeostasis. [Adapted from D. H. Abbott et al.: Developmental origin of polycystic ovary syndrome: a hypothesis. J Endocrinol 174:1, 2002 (17), with permission. © Society for Endocrinology and European Society of Endocrinology; and S. Franks: Polycystic ovary syndrome in adolescents. Int J Obes (Lond) 32:1035 (2008), with permission. © Nature Publishing Group.] But to return to the utility of rodent models in understanding the etiology of PCOS, in this issue of Endocrinology, there is an important paper from van Houten et al., (31) that illustrates why such models continue to be useful and relevant, particularly with regard to unraveling the complex effects of excess androgen on metabolic function. This paper is the latest of a number of recent publications [see review by Walters and colleagues (15)] that have looked in detail at both reproductive and metabolic consequences of androgen excess. In animals treated prepubertally with dihydrotestosterone (a nonaromatizable androgen), van Houten and colleagues have made a careful analysis of the ovarian follicle population and found abnormalities of antral follicles (as might be expected), but in contrast to findings in prenatally androgenized sheep and in human PCOS, there were no differences between experimental and control animals in the proportions of healthy or atretic preantral follicles. Their focus in the metabolic studies is on glucose/insulin homeostasis (as in previously published studies) and especially on fat cell morphology and function. Intriguingly, the findings of enlarged adipocytes and decreased serum levels of adiponectin are similar to those in women with PCOS (32, 33). In conclusion, although rodent models of PCOS may not be ideal with respect to understanding the ovarian abnormalities of PCOS (and are less informative than large-animal models), they may still provide important insight into the reproductive dysfunction of women with PCOS. And particularly in terms of examining the relationship between excess androgen and metabolic dysfunction, rodent models would appear to have an increasingly important part to play. This was supported by Medical Research Council (UK) Program Grant G0802782. Disclosure Summary: The author has nothing to disclose. Polycystic ovary syndrome.