Chapter 23 / Animal Models, Fetal Programming of PCOS 259
259
From: Contemporary Endocrinology: Androgen Excess Disorders in Women:
Polycystic Ovary Syndrome and Other Disorders, Second Edition Edited by: R. Azziz et al. © Humana Press Inc., Totowa, NJ
23
Animal Models and Fetal Programming of the Polycystic Ovary Syndrome
David H. Abbott, Daniel A. Dumesic, Jon E. Levine, Andrea Dunaif, and Vasantha Padmanabhan
SUMMARY
At least 28 animal models provide insight into the etiological and pathophysiological basis of polycystic ovary syndrome (PCOS). About 50% of them, however, either do not show sufficient traits meriting designation of a PCOS phenotype or exhibit alternate features mimicking other disor- ders, such as hyperprolactinemia. In contrast, animal models of fetal programming through androgen excess show remarkable resilience and reliability in replicating PCOS, including metabolic defects in males, and therefore strongly implicate a fetal etiology in the developmental origins of PCOS. This chapter reviews the relevance of animal models for PCOS and their potential value for providing insight into the etiology and pathophysiology of this disorder.
Key Words: Polycystic ovary syndrome; rhesus monkeys; androgens; prenatal; fetal programming; animal model; anovulation; hyperandrogenism.
1. INTRODUCTION
The etiology and pathophysiology of polycystic ovary syndrome (PCOS) in women are poorly understood. PCOS is multifaceted and includes reproductive, metabolic, and general health disorders (Table 1). The syndrome is strongly familial in origin, with 67–93% of daughters born to women with PCOS developing the PCOS syndrome as adults (1–3). Clinical or biochemical manifestation of androgen excess is the most reliably transmitted PCOS trait (4). PCOS is the most common endocrin- opathy of women in their reproductive years (5,6), with a prevalence of 6–7% (7), and one of its most troubling general health disorders is early-onset type 2 diabetes (8,9).
Nevertheless, PCOS has a heterogeneous and unpredictable clinical presentation (5,6), and most
putative gene candidates studied to date have been unable to adequately explain its phenotype (10),
suggesting that PCOS has multiple (albeit undiscovered) genetic origins modified by environmental
factors and perhaps fetal programming (11). Therefore, the development and application of animal
models for PCOS provide timely insight into the origins and pathophysiological mechanisms that are
difficult to resolve from human studies. This chapter reviews the many animal models proposed as
relevant to PCOS, with particular emphasis on models that reliably and reproducibly emulate the
PCOS syndrome through androgen excess fetal programming.
260 Abbott et al.
Table 1
Common Signs and Symptoms of PCOS A. Consensus diagnostic criteriaa
Two out of three of the following:
1. Clinical or biochemical hyperandrogenism, as determined by elevated circulating levels of total or unbound testosterone or hirsutism
2. Intermittent or absent menstrual cycles 3. Polycystic ovaries (as visualized by ultrasound)
The following conditions must also be excluded: classical and nonclassical congenital adrenal hyperplasia, Cushing’s syndrome, thyroid dysfunciton, hyperprolactinemia, androgen-secreting tumors, and drug-induced androgen excess.
B. PCOS signs and symptoms outside those required for diagnosis (some, all, or none of these may be present in an individual)
Reproductive and endocrine
Luteinizing hormone (LH) hypersecretion Reduced steroid-negative feedback on LH release Increased recruitment and persistence of ovarian follicles
Ovarian hyperresponsiveness to gonadotropic therapy for in vitro fertiliztion (IVF) High rates of miscarriage
Endometrial hyperplasia and cancer Adrenal hyperandrogenism Gestational diabetes Metabolic
Insulin resistance and compensatory hyperinsulinemia Imparied glucose tolerance
Type 2 diabetes
Obesity (including abdominal adiposity)
Pancreatic impairments in insulin responses to glucose Hyperlipidemia
General health disorders Cardiovascular disease Sleep apnea
Acne
Chronic inflammation
Intra-uterine growth retardation
aThese criteria (13) are extented from the previous 1990 NIH Consensus diagnosis (12) which specified the first two criteria, alone, as a basis for PCOS diagnosis, following exclusion of conditions that mimic PCOS (listed above). Revi- sions of these criteria by Azziz (14) suggest requiring criterion 1 together with either criteria 2 or 3, whereas revisions by Chang (15) relegate criterion 3 to nondiagnostic status. PCOS, polycystic ovary syndrome. (Modified from ref. 11.)
2. BACKGROUND 2.1. Overview of PCOS
Because PCOS is a diagnosis of exclusion, animal models need to replicate traits consistent with
PCOS without having features that might mimic other clinical diseases. Such a goal is difficult and
compounded by recent changes in the manner in which PCOS is clinically diagnosed. For example,
the 1990 National Institutes of Health (NIH) consensus diagnosis for PCOS specifies
hyperandrogenism accompanied by oligo- or amenorrhea, excluding conditions that mimic PCOS,
such as classic and nonclassical congenital adrenal hyperplasia, thyroid dysfunction,
hyperprolactinemia, androgen-producing tumors, and drug-induced androgen excess (12).
Chapter 23 / Animal Models, Fetal Programming of PCOS 261
The Revised 2003 Rotterdam consensus diagnosis adds polycystic ovaries, as diagnosed by tranvaginal ultrasound (TVUS), to the above 1990 NIH consensus diagnosis for PCOS, specifying that two of the three criteria are required to diagnose PCOS (13). A recent reappraisal of the 1990 NIH and Revised 2003 Rotterdam consensus diagnoses (14) combines criteria from both consenses and specifies PCOS as androgen excess accompanied by oligo- or amenorrhea or by TVUS-con- firmed polycystic ovaries while still excluding clinical conditions mimicking PCOS. An additional suggestion (15) is to continue using the 1990 NIH consensus for the diagnosis of PCOS, while rel- egating sonographic imaging of polycystic ovaries to confirmatory, rather than diagnostic, status.
Equally complex, individual women with PCOS can exhibit different combinations of diagnostic criteria with varying degrees of severity, can show abnormalities other that those used for diagnosing PCOS (5,6), and can experience onset of symptoms at puberty (16,17), which can resolve during middle age (18).
Thus, for an animal model to truly approximate the complexities of the PCOS phenotype, it must not only exhibit ovarian hyperandrogenism, oligo- or amenorrhea, and/or an increased number of medium-sized ovarian follicles in the absence of features mimicking other clinical diseases, but also show an extraordinary array of relevant traits, permitting a heterogeneous phenotype.
2.2. Animal Models for PCOS
A variety of mammalian species have been employed as animal models of PCOS, ranging from rodents to nonhuman primates. Each species has differences in reproductive function compared to humans, and such reproductive differences need to be considered when translating experimental find- ings into clinical applications (19). For example, rats and mice undergo spontaneous ovulation approximately every 4–5 days (not every 26–34 days), and they complete follicle luteinization and form corpora lutea only if mating occurs, with the luteotropic support of prolactin (not luteinizing hormone [LH]).
2.2.1. Animal Models for PCOS That Produce Large Ovarian Follicular Cysts
Much has been made of the experimental induction of ovarian cysts in animal models for PCOS.
Unlike PCOS ovarian morphology, follicular cysts in animal models are frequently large (preovula- tory sized or larger) and are thus not a diagnostic trait for the PCOS syndrome. These animal models, while providing insight into the development of cystic ovarian morphology, may not be relevant in resolving the etiology and pathophysiology of PCOS. Models inducing large ovarian follicular cysts are found in a variety of experimental treatment paradigms (summarized in Table 1), with and with- out additional induction of PCOS and non-PCOS diagnostic traits. In this context it is important to note that rodents are generally multiovular and thus normally exhibit multifollicular ovarian mor- phology.
2.2.2. Animal Models for PCOS That Fail to Demonstrate PCOS Diagnostic Traits
When animal models lack elevated androgen levels or exhibit regular ovulatory cycles, they fail to demonstrate the basic tenets of PCOS. Those that fall into this category (Table 1) include treatment of adult females with constant exposure to light (probably the oldest animal model proposed for PCOS), acute estrogen, and valproic acid, as well as treatment of newborn females with testosterone. Estradiol valerate treatment of female rats, while inhibiting ovulatory cycles, also induces traits that are unchar- acteristic of PCOS: growth hormone excess and hypothalamic degeneration. Such animal models hold little promise for understanding etiological and pathophysiological mechanisms of PCOS.
2.2.3. Animal Models for PCOS That Exhibit Diagnostic Traits But Show Other Traits Excluding a PCOS Diagnosis
Four animal models, while exhibiting PCOS diagnostic traits, show other traits that are inconsis-
tent with PCOS in women,for example, thyroid dysfunction in hypothyroid rats treated with human
chorionic gonadotropin (hCG) (Table 2). The remaining three models, transgenic overexpression of
262 Abbott et al.
Table 2 Characteristics of Animal Models of PCOS Relevant to PCOS Diagnostic Criteria and Criteria for Diagnostic Exclusion from PCOS Criteria for PCOS diagnosis Criteria that exclude a PCOS diagnosis Multiple, Intermittent ormedium-sizedAdrenal 17Androgenabsent ovulatory ovarian antralOHPThyroidModel characteristics ModelSpeciesexcesscyclesfolliclesexcessdysfunctionHyperprolactinemianot found in PCOSRef. Models with insufficient traits for a PCOS diagnosis and no known exclusion traits Constant lightRat–+–N/A?+Many large, cystic ovarian35 follicles, loss of endogenous endocrine rhythms Estradiol valerateRat?+–N/A?+Elevated growth hormone,68 large ovarian follicular cysts, hypothalamic degeneration Chronic estrogenRat?++N/A??22 Acute estrogenGuinea pig–––???Large ovarian follicular cysts69 Valproic acid treatmentRat––+N/A??70
262
(anticonvulsant and fatty acid analog) Valproic acid treatmentRhesus monkey––––??71 PMSG treatment (pregnantRat?+–N/A??Large ovarian follicular cysts72 mare serum gonadotropin) Neonatal testosteroneNewborn rat–+–N/A??Large ovarian follicular cysts,73 treatmentsmall ovaries Models with PCOS diagnostic traits, but also diagnostic traits that exclude a PCOS diagnosis Hypothyroid treatmentRat+?+N/A++Hypothyroidism67 with hCG Transgenic overexpressionMouse++–N/A?+?Ovarian cancer, renal pathology,20 of LHlarge, blood-filled ovarian follicular cysts bDHEA treatmentImmature rat++–N/A?+Large ovarian follicular cysts21,22 bAdult rat++–N/A?+Large ovarian follicular cysts RU486 treatment (antipro-Rat++–N/A?+Large ovarian follicular cysts23 gestagenic and anti-with precocious luteinization glucocorticoid)Chapter 23 / Animal Models, Fetal Programming of PCOS 263
Models with PCOS diagnostic traits and no known exclusion traits NymphomaniaCow++–???Excessive sexual behavior22 Immunization againstRat+a+–N/A?–Large ovarian follicular cysts24 testosterone Letrozole treatmentRat+++N/A??30 (aromatase inhibitor) Aromatase knockout (ArKO)Mouse++–N/A??Ovaries and uterus immature31 Dexamethasone treatmentPig+++???Glucocorticoid excess29 (synthetic glucocorticoid) Chronic testosteroneImmature rat+b++N/A??Large ovarian follicular cysts25 treatment Chronic testosteroneRhesus monkey+b+–???33 treatment Chronic androstenedioneRhesus monkey+b(+)–???34 treatmentSeasonal anovulation only Chronic estrone treatmentRhesus monkey++–???35 Chronic hCG treatmentRat++?N/A??Large ovarian follicular cysts26 Chronic insulin and hCGRat++–N/A??Large ovarian follicular cysts27 treatment Chronic IGF-1 and hCGRat++–N/A??Large ovarian follicular cysts28 treatment Fetal testosterone treatmentMouse++????Virilized genitalia48 Fetal testosterone treatmentRat++????Virilized genitalia49 Fetal testosterone treatmentSheep+c++???Virilized genitalia only in females42 exposed during early gestation Fetal testosterone treatmentRhesus monkey+++–––Virilized genitalia only in females11,40 exposed during early gestation aLow serum levels of free unbound testosterone. bExogenous androgen excess. cFunctional ovarian hyperandrogenism (V. Padmanabhan, unpublished results). N/A, not applicable—rodent adrenal glands do not synthesize corticosterone from 17OHP. PCOS, polycystic ovary syndrome; hCG, human chorionic gonadotropin; LH, luteinizing hormone; DHEA, dehydroepiandrosterone; IGF, insulin-like growth factor.
263
264 Abbott et al.
LH (20), and treatment with either dehydroepiandrosterone (21,22) or RU486 (an antiprogestagen and antiglucocorticoid) (23), all induce hyperprolactinemia, a well-known mimic of PCOS in women (12,13). All but the hypothyroid rats also exhibit large ovarian follicular cysts (Table 2). These ani- mal models might thus provide useful information in determining mechanisms underlying the patho- physiology of conditions that resemble PCOS, but are not PCOS itself.
2.2.4. Animal Models for PCOS That Exhibit Diagnostic Traits and Lack Traits Excluding a PCOS Diagnosis
The remaining 16 animal models all demonstrate traits consistent with those for PCOS diagnosis, and none exhibit additional traits for exclusion (Table 2). The nymphomaniac cow is the only natu- rally occurring animal model for PCOS (22), but its unpredictable occurrence and unknown mecha- nism have yet to make it useful. Although manipulation of adult female rats by immunization against testosterone (24), chronic treatment with testosterone (25), hCG (26), insulin and hCG (27), and insulin-like growth factor-1 and hCG (28) all induce hyperandrogenic females that have intermittent or absent ovulatory cycles, the treatments induce large ovarian follicular cysts, unlike the smaller- sized cysts found in women with PCOS. Treatment of adult female pigs with dexamethasone, a syn- thetic glucocorticoid, induces PCOS diagnostic criteria provided that adrenergic innervation to the ovaries remains intact (29), yet glucocorticoid excess is not a common symptom accompanying PCOS. Use of an aromatase inhibitor, letrozole, on adult female rats produces a phenotype remark- ably similar to that of PCOS, including LH hypersecretion (30). Not surprisingly, however, serum levels of estradiol are greatly diminished, a steroidogenic abnormality not found in PCOS. Although complete aromatase knockout female mice are hyperandrogenic and anovulatory, they do not repre- sent a PCOS phenotype because their ovaries and uteri fail to mature (31). In this context it is interesting to note that fetal female monkeys exposed to a highly specific aromatase inhibitor have greatly di- minished ovarian follicular development that is prevented by simultaneous treatment with estradiol (32).
Chronic exposure of adult female monkeys to testosterone (33), androstenedione (34), or estrone (35) induce females with hyperandrogenism and intermittent or absent menstrual cycles (Table 1).
None of these adult models, however, exhibit traits commonly associated with PCOS beyond the diagnosis, such as PCOS-like ovarian morphology, LH hypersecretion, or metabolic dysfunction (36).
Such chronic manipulation of the adult steroid hormone environment, including that induced by aromatase inhibition, has generated four animal models that more closely approximate the symptomology of PCOS than the models previously discussed. Neither these nor the previous mod- els, however, match the replication of PCOS phenotype generated by animal models of fetal andro- gen excess that reprogram differentiation and development of multiple organ systems. Acute exposure of normal adult female monkeys to testosterone, accelerates the early stages of ovarian follicular development (37) and may mimic accelerated the early follicular development found in women with PCOS associated with diminished intraovarian expression of anti-Müllerian hormone (38).
2.3. Animal Models of Androgen Excess Fetal Programming of PCOS
Compared with other animal models for PCOS, models of fetal programming have one clear advan- tage: simultaneous exposure of multiple organ systems to a specific developmental insult during differ- entiation and maturation (Fig. 1). Altered structure and function is commonly permanent, such as fetal androgen excess virilization of the female urogenital tract, resulting in expression of fetal pro- gramming in adulthood (11). The best known example of fetal programming is described by Barker and colleagues (39), in which human fetal undernutrition and low birthweight are associated with adult cardiovascular disease, hypertension, insulin resistance, and type 2 diabetes, some key hall- marks of PCOS outside of those required for its diagnosis.
In the previous edition of this book, Abbott and colleagues (40) provided the first description of a
fetal programming model for PCOS: the prenatally androgenized female rhesus monkey. As illus-
trated in Fig. 1 and in Tables 2 and 3, early gestation exposure of the female monkey fetus to fetal
Chapter 23 / Animal Models, Fetal Programming of PCOS 265
male levels of testosterone results in adult females with PCOS traits. Early gestation in the rhesus monkey provides a stage in development when multiple organ systems regulating reproductive and metabolic function are undergoing differentiation. A single insult during such a sensitive stage of development can permanently alter disparate organ systems, producing a phenotypic mimic of PCOS.
2.3.1. Fetal Androgen Programming of Reproductive Defects
Prenatally androgenized female monkeys exhibit heterogeneity in their presentation of PCOS traits, an inherent complexity in the human syndrome. Approximately 70% of prenatally androgenized female monkeys exposed to androgen excess during early gestation have serum testosterone levels in excess of the mean value in normal adult female monkeys of the same age, weight, and body mass index; approximately 40% are anovulatory (~10 times the normal rate, while the remainder have mostly intermittent menstrual cycles); and approximately 40% have increased numbers of medium- sized ovarian antral follicles (approximately twice the normal incidence) (41). Differing degrees of virilization of both internal and external genitalia, however, are found in all female monkeys exposed to androgen excess during early gestation. Because virilized genitalia is not a feature of PCOS in women, exposure of female monkeys to androgen excess during late gestation, when the urogenital tract is no longer responsive to androgen reprogramming, produces a closer PCOS phenotype that retains heterogeneity of trait expression, but without genital virilization (Fig. 1) (11,41). Such results from early and late gestationally exposed prenatally androgenized female monkeys, confirmed in prenatally androgenized ewes (42), suggest that exposure of human female fetuses to androgen excess during the latter part of gestation (second to third trimesters) from genetic and/or environmental factors might induce organ system reprogramming, causing PCOS. Thus, the key element in the etiology and initial pathophysiology of PCOS may be appropriately timed fetal hyperandrogenism derived from hyperandrogenic fetal ovaries (43), fetal adrenal cortex (44), or from hyperandrogenemia of PCOS mothers (45) reflected in the fetal circulation (46).
Fig. 1. Gestational progression of aspects of differentiation and maturation of hypothalamic–pituitary–
ovarian function and pancreas and E-cell function in rhesus monkeys. The timing of exposure of females to androgen excess (early or late in gestation) is indicated in relation to fetal developmental progress. GnRH, gonadotropin-releasing hormone; LH, luteinizing hormone; FSH, follicle-stimulating hormone. (Modified from ref.11.)
266 Abbott et al.
Recent findings from prenatally androgenized female sheep (47), mice (48), and rats (49) confirm and extend those obtained from prenatally androgenized female monkeys (Tables 2 and 3). All express PCOS diagnostic traits, as well as traits commonly associated with the syndrome (e.g., LH hypersecretion). Anovulation in prenatally androgenized female mice (48) and rats (49) is induced by fetal exposure to the nonaromatizable androgen dihydrotestosterone (DHT), suggesting an androgen receptor-mediated neuroendocrine defect. In prenatally androgenized monkeys, although prenatal DHT exposure can induce similar behavioral outcomes to those achieved by testosterone (11), prena- tal DHT-exposed animals are not available for PCOS studies. Certainly in prenatally androgenized female mice, treatment of such adults with the antiandrogen flutamide restores ovulatory cycles and implicates adult excess androgen, acting via the androgen receptor, in the neuroendocrine mecha- nism of anovulation (48).
Table 3
Common Signs and Symptoms Associated With PCOS and Shown by Animal Models of Androgen Excess Fetal Programming of PCOS
Monkey Mousea Ratb Sheepc (early/late gestation)d Reproductive and endocrine
Ovarian hyperandrogenism + + + +
Intermittent or absent ovulatory cycles + + + +
Multiple medium-sized ovarian follicles ??? ??? + +
LH hypersecretion + + + +
Reduced steroid negative feedback on LH ??? ??? + +
Ovarian endocrine hyper-responsiveness to ??? ??? ??? (???) Poor embryo
gonadotropic hyperstimulation for IVF development
High rates of miscarriage
Endometrial hyperplasia and cancer ??? ??? ??? (+) Hyperplasia
Adrenal hyperandrogenism N/A N/A N/A +
Gestational diabetes ??? ??? ??? ???
Metabolic
Insulin resistance and compensatory hyperinsulinemia ??? + + +
Impaired glucose tolerance ??? + ??? +
Type 2 diabetes ??? ??? ??? +
Obesity (including abdominal adiposity) ??? ??? ??? +
Pancreatic impairments in insulin responses to glucose ??? + ??? +
Hyperlipidemia ??? ??? ??? +
General health disorders
Cardiovascular disease ??? ??? + ???
Sleep apnea ??? ??? ??? ???
Chronic inflammation ??? ??? ??? ???
Low birthweight ??? + + –
Heterogeneity of PCOS trait expression ??? ??? + +
N/A, not applicable because adrenal glands from nonprimate species do not normally synthesize androgens.
aFrom ref. 48.
bFrom ref. 49 and J. E. Levine, unpublished results.
cFrom refs. 51,56, and 62, and V. Padmanabhan, unpublished results.
dFrom refs. 11,41,50, and 66 and D. H. Abbott, unpublished results.
LH, luteinizing hormone; IVF, in vitro fertilization.
Chapter 23 / Animal Models, Fetal Programming of PCOS 267
2.3.2. Fetal Androgen Programming of Oocyte Quality
Beyond ovulatory dysfunction, diminished oocyte quality in both prenatally androgenized female monkeys and women with PCOS provides an additional barrier to fertility (50). Following controlled ovarian hyperstimulation for in vitro fertilization (IVF), retrieved oocytes from either early or late gestation exposed, prenatally androgenized female monkeys exhibit reduced competence as defined by the ability of the resulting diploid zygotes to reach the blastocyst stage. In women with PCOS, diminished quality of retrieved oocytes contributes to implantation failure and pregnancy loss. Both prenatally androgenized female monkeys and PCOS women exhibit abnormal intrafollicular ste- roidogenic responses to controlled ovarian hyperstimulation, which in the former is associated with an inability to normally suppress circulating insulin levels between the first day of the recombinant human follicle-stimulating hormone (rhFSH) treatment and the day of oocyte retrieval (50). Women with PCOS, however, are hyperresponsive to rhFSH and exhibit intrafollicular hyperandrogenism at oocyte retrieval following recombinant human chorionic gonadotropin (rhCG) injection. Prenatally androgenized female monkeys, however, are hyporesponsive to rhFSH relative to normal females, and early gestation-exposed female monkeys exhibit diminished intrafollicular androgen and estro- gen levels along with an exaggerated shift in intrafollicular steroidogenesis from androgen and estro- gen to progesterone at oocyte retrieval following rhCG injection. The early gestation-exposed female monkey response to controlled ovarian hyperstimulation, therefore, is reminiscent of that shown by normoandrogenic ovulatory women with reduced ovarian responsiveness to rhFSH (50). Collectively, in addition to ovulatory defects, the timing of fetal androgen excess exposure impairs oocyte quality through ovarian and/or metabolic dysfunction, with such oocyte defects possibly having transgenerational consequences for female offspring of prenatally androgenized monkeys and for daughters of women with PCOS.
2.3.3. Fetal Androgen Programming of Metabolic Defects
Insulin resistance and diminished pancreatic insulin response to glucose are integral defects in the development of type 2 diabetes in women with PCOS and are exacerbated by obesity (5,6). Prena- tally androgenized female monkeys (11), sheep (51), and rats (J. E. Levine, unpublished results) all exhibit such insulin dysfunction, while early gestation-exposed androgenized female monkeys also exhibit abdominal obesity, hyperlipidemia, and an increased incidence of type 2 diabetes (Table 3).
Insulin sensitivity in early, but not late, gestationally exposed, prenatally androgenized female mon- keys is reduced to that found in normal male monkeys and in normal females during the luteal phase of the menstrual cycle, and a similar degree of insulin resistance is found in prenatally androgenized sheep (51). Such parallels in metabolic dysfunction between androgen excess fetal programming models of PCOS and women with PCOS provide strong evidence for a fetal origin of metabolic defects in both cases, possibly through fetal programming of preferential accumulation of abdominal fat (Fig. 2).
2.3.4. Fetal Androgen Programming of General Health Disorders
Poor intrauterine growth and low birthweight are associated with the development of precocious
puberty and PCOS in northern Spanish women (52) and with PCOS pregnancies in Chilean women
(53), but not in larger studies of Finnish (54) and Dutch (55) women. Prenatally androgenized female
sheep (56) and rats (57) exhibit clear evidence of intrauterine growth restriction and low birthweight,
whereas prenatally androgenized female monkeys do not (41). Prenatally androgenized sheep and
rats may thus provide more suitable animal models for women with PCOS who have placental insuf-
ficiency. Perhaps not surprisingly, in this context, prenatally androgenized sheep have enlarged left
ventricles of the heart, kidneys, and adrenal glands suggestive of developing cardiovascular disease
(42), and prenatally androgenized female rats have increased mortality (58).
268 Abbott et al.
2.3.5. Comparison of Androgen Excess Fetal Programming Models for PCOS
Table 3 illustrates the comparability of androgen excess fetal programming models for PCOS.
Although most PCOS traits to date have been identified in prenatally androgenized female monkeys (11), increasing numbers of reports are describing PCOS traits in prenatally androgenized females in nonprimate species (42,49,48). The only clear difference in symptomology between primate and nonprimate models involves low birthweight in the latter and not the former, indicating an otherwise remarkable degree of concurrence among models. Because low birthweight is found in some (52,53), but not all (54,55) populations of women with PCOS, this heterogeneity in animal models may prove extremely useful in identifying mechanisms underlying the different fetal phenotypes found in PCOS.
Rodent models, nevertheless, may not be ideal for ovarian phenotype determination as they normally develop multiple large follicles prior to ovulation, unlike sheep or monkeys. On the other hand,
Fig. 2. Diagrammatic representation of our hypothesis for early gestation, fetal androgen excess program- ming of adult polycystic ovary syndrome traits. Genetic or environmental mechanisms induce fetal hyperandrogenism (see text) that result in permanent changes in both reproductive and metabolic function.Reproductive consequences include (1) altered hypothalamic–pituitary function leading to luteinizing hormone (LH) hypersecretion, (2) ovarian hyperandrogenism that may or may not be the result of LH hypersecretion, (3) reduced steroid hormone negative-feedback regulation of LH, which may be a component of the initial perma- nent alteration in hypothalamic–pituitary function, and (4) increased anovulation. Metabolic consequences in- clude (1) increased abdominal adiposity, which may be responsible for increased circulating total free fatty acid levels, (2) impaired pancreatic insulin secretory response to glucose, (3) impaired insulin action and compensa- tory hyperinsulinemia, (4) hyperglycemia, and (5) increased incidence of type 2 diabetes. Insulin resistance and compensatory hyperinsulinemia may be functionally implicated in the anovulatory mechanism. (Modified from ref.41.)
Chapter 23 / Animal Models, Fetal Programming of PCOS 269
transgenic rodent models hold much promise for determining fetal programming consequences of specific gene changes, such as those involving neuroendocrine regulation of ovarian function (48).
Sheep models are particularly useful for understanding abnormal ovarian follicular recruitment and persistence through sequential ultrasonography, repeated blood sampling, and ovarian manipulation.
Nonhuman primate models, with more than 90% genetic similarity to humans and the closest symptomology to PCOS, provide the most straightforward translation of experimental findings into improved clinical application.
The only major difference in trait defects between prenatally androgenized female rhesus mon- keys and women with PCOS involves their responses to ovarian hyperstimulation for IVF. Women with PCOS are hyperandrogenic and hyperresponsive to rhFSH, whereas prenatally androgenized female monkeys are not (Table 3). It is tempting to speculate that this differential response to gona- dotropin stimulation may reflect different regulatory mechanisms governing hyperandrogenism in ovarian theca cells of women with PCOS vs prenatally androgenized female monkeys, perhaps re- flecting an intrinsic, genetically determined hyperandrogenism in the former (59), but not the latter.
2.4. Animal Models for a Male Phenotype of PCOS
Close male relatives of women with PCOS present with metabolic dysfunction similar to that of their female kin (60–62). Consistent with a fetal origins hypothesis for PCOS, male monkeys ex- posed to androgen excess during gestation develop insulin resistance and diminished insulin response to glucose as adults (63). Because fetal monkey androgen treatments induce circulating testosterone levels in male and female fetuses only within the normal range for fetal males (64), fetal program- ming of male (and female?) metabolic dysfunction may be induced by mechanisms beyond andro- gen-mediated action in the fetus, perhaps involving an estrogenic metabolite of testosterone, or the action of androgen on the placenta or the mother. In support of potential estrogenic involvement in PCOS fetal programming, exposure of fetal ewes to bisphenol-A, a plasticizer and estrogen mimic, leads to intrauterine growth restriction and LH surge defects similar to those produced by fetal expo- sure to testosterone (V. Padmanabhan et al., unpublished).
3. CONCLUSIONS
A variety of animal models have been proposed for PCOS, but only models of androgen excess fetal programming have reliably produced the spectrum and heterogeneity of traits that closely re- flect PCOS in women (Tables 2 and 3). Animal models of fetal programming may thus provide the elusive etiology and initial pathophysiology for PCOS in women and for metabolic dysfunction in their close male kin. Such fetal models suggest that permanent alteration of gene expression in mul- tiple organ systems may provide a closer approximation to PCOS phenotypes than differing geno- types alone.
4. FUTURE AVENUES OF INVESTIGATION
Until the etiology of PCOS is determined and the cellular and molecular mechanisms of its patho-
physiology are understood, animal models for the syndrome will be essential. Animal models will
continue to lead the way in identifying a probable fetal origin for PCOS until sampling from human
fetuses becomes a low-risk routine procedure and pregnancies at risk for PCOS are identified pro-
spectively. Such PCOS animal models will also continue to be essential in ascertaining the otherwise
ethically unattainable goal of normalized oocyte and embryo quality in women with PCOS. Fetal
programming models thus hold promise for determining the gestational timing of crucial changes in
specific organ and system functions that ultimately result in specific PCOS defects and for determin-
ing transcriptional, translational, and posttranscriptional levels of dysfunction that can be targeted for
therapeutic amelioration during pre- and postconception.
270 Abbott et al.
KEY POINTS
• Animal models provide insight into the etiology and pathophysiology of PCOS unattainable in human studies.
• Many animal models either fail to exhibit PCOS diagnostic traits or exhibit traits that resemble other clinical disorders, including thyroid dysfunction and hyperprolactinemia (Table 2).
• Androgen excess fetal programming of PCOS in female mice, rats, sheep, and monkeys (Table 3) pro- vides compelling evidence for fetal origins of the syndrome in humans.
• Androgen excess fetal programming of a male PCOS metabolic phenotype suggests fetal origins for male PCOS beyond the direct effects of androgens on the fetus.
• The intrauterine environment conducive to PCOS may be induced by a variety of genetic or environmental factors, or a combination of both, resulting in fetal androgen excess.
ACKNOWLEDGMENTS
We thank the many staff members of our respective laboratories and institutions for their multiple contributions to the work reported here and A.D.M. Abbott for his contribution towards compilation of
Table 2. This work was supported by NIH grants P50 HD044405, U01 HD044650, R01 RR013635,R21 RR014093, T32 AG000268, P51 RR000167, R01 HD041098, and P01 HD044232 and was partly conducted at a facility constructed with support from Research Facilities Improvement Program grant numbers RR15459 and RR020141.
REFERENCES
1. Hague WM, Adams J, Reeders ST, Peto TEA, Jacobs HS. Familial polycystic ovaries: a genetic disease? Clin Endocrinol 1988;29:593–605.
2. Govind A, Obhrai MS, Clayton RN. Polycystic ovaries are inherited as an autosomal dominant trait: analysis of 29 polycystic ovary syndrome and 10 control families. J Clin Endocrinol Metab 1999;84:38–43.
3. Legro RS. The genetics of polycystic ovary syndrome. Am J Med 1995;16;98(1A):9S–16S.
4. Legro RS, Spielman R, Urbanek M, Driscoll D, Strauss JF 3rd, Dunaif A. Phenotype and genotype in polycystic ovary syndrome. Recent Prog Horm Res 1998;53:217–256.
5. Ehrmann DA, Barnes RB, Rosenfield RL. Polycystic ovary syndrome as a form of functional ovarian hyperandrogenism due to dysregulaton of androgen secretion. Endocr Rev 1995;16:322–353.
6. Dunaif A. Insulin resistance and the polycystic ovary syndrome: mechanism and implications for pathogenesis. Endocr Rev 1997;18:774–800.
7. Azziz R, Woods KS, Reyna R, Key TJ, Knochenhauer ES, Yildiz BO. The prevalence and features of the polycystic ovary syndrome in an unselected population. J Clin Endocrinol Metab 2004;89:2745–2749.
8. Arslanian SA, Lewy VD, Danadian K. Glucose intolerance in obese adolescents with polycystic ovary syndrome: roles of insulin resistance and beta-cell dysfunction and risk of cardiovascular disease. J Clin Endocrinol Metab 2001;86:66–71.
9. Peppard HR, Marfori J, Iuorno MJ, Nestler JE. Prevalence of polycystic ovary syndrome among premenopausal women with type 2 diabetes. Diabetes Care 2001;24:1050–1052.
10. Legro RS, Strauss JF. Molecular progress in infertility: polycystic ovary syndrome. Fertil Steril 2003;78:569–576.
11. Abbott DH, Barnett DK, Bruns CM, Dumesic DA. Androgen excess fetal programming of female reproduction: a developmental aetiology for polycystic ovary syndrome? Hum Reprod Update 2005;11:357–374.
12. Zawadzki JA, Dunaif A. Diagnostic criteria for polycystic ovary syndrome: towards a rational approach. In: Dunaif A, Givens JR, Haseltine FP, Merriam GR, eds. Polycystic Ovary Syndrome. Boston: Blackwell Scientific, 1992:377–384.
13. The Rotterdam ESHRE/ASRM-Sponsored PCOS consensus workshop group. Revised 2003 consensus on diagnostic criteria and long-term health risks related to polycystic ovary syndrome (PCOS). Hum Reprod 2004;19:41–47.
14. Azziz R. Diagnostic criteria for polycystic ovary syndrome: a reappraisal. Fertil Steril 2005;83:1343–1346.
15. Chang RJ. A practical approach to the diagnosis of polycystic ovary syndrome. Am J Obstet Gynecol 2004;191:713–717.
16. Arslanian SA, Witchell S. Premature pubarche, insulin resistance, and adolescent polycystic ovary syndrome. In: Chang RJ, Heindel JJ, Dunaif, A, eds. Polycystic Ovary Syndrome. New York: Marcel Dekker, Inc., 2002:37–53.
17. Rosenfield RL, Ghai K, Ehrmann DA, Barnes RB. Diagnosis of the polycystic ovary syndrome in adolescence: com- parison of adolescent and adult hyperandrogenism. J Pediatr Endocrinol Metab 2000;13(Suppl 5):1285–1289.
18. Taylor AE, McCourt B, Martin KA, et al. Determinants of abnormal gonadotropin secretion in clinically defined women with polycystic ovary syndrome. J Clin Endocrinol Metab 1997;82:2248–2256.
19. Abbott DH, Foong SC, Barnett DK, Dumesic DA. Nonhuman primates contribute unique understanding to anovulatory infertility in women. ILAR J 2004;45:116–131.
Chapter 23 / Animal Models, Fetal Programming of PCOS 271
20. Risma KA, Hirshfield AN, Nilson JH. Elevated luteinizing hormone in prepubertal transgenic mice causes hyperandrogenemia, precocious puberty, and substantial ovarian pathology. Endocrinology 1997;138:3540–3547.
21. Ward RC, Costoff A, Mahesh VB. The induction of polycystic ovaries in mature cycling rats by the administration of dehydroepiandrosterone (DHA). Biol Reprod 1978;18:614–623.
22. Mahesh VB, Mills TM, Bagnell CA, Conway BA. Animal models for study of polycystic ovaries and ovarian atresia.
In: Mahesh VB, Dhindsa DS, Anderson E, Katra SP, eds. Regulation of Ovarian and Testicular Function. New York:
Plenum Press, 198, pp. 237–257.
23. Ruiz A, Aguilar R, Tébar M, Gaytán F, Sánchez-Criado E. RU486-treated rats show endocrine and morphological responses to therapies analogous to responses of women with polycystic ovary syndrome treated with similar therapies.
Biol Reprod 1996;55:1284–1291.
24. Hillier SG, Groom GV, Boyns AR, Cameron EH. Development of polycystic ovaries in rats actively immunised against T-3-BSA. Nature 1974;250:433–434.
25. Beloosesky R, Gold R, Almog B, et al. Induction of polycystic ovary by testosterone in immature female rats: modula- tion of apoptosis and attenuation of glucose/insulin ratio. Int J Mol Med 2004;14:207–215.
26. Bogovich K. Induction of ovarian follicular cysts in the pregnant rat by human chorionic gonadotropin. Biol Reprod 1991;45:34–42.
27. Poretsky L, Clemons J, Bogovich K. Hyperinsulinemia and human chorionic gonadotropin synergistically promote the growth of ovarian follicular cysts in rats. Metabolism 1992;41:903–910.
28. Damario MA, Bogovich K, Liu HC, Rosenwaks Z, Poretsky L. Synergistic effects of insulin-like growth factor-I and human chorionic gonadotropin in the rat ovary. Metabolism 2000;49:314–320.
29. Dzienis A, Majewski M, Wojtkiewicz J, Piskula M, Jana B. Adrenergic innervation and steroidogenic activity of cystic porcine ovaries. Rocz Akad Med Bialymst 2004;49(Suppl 1):114–116.
30. Kafali H, Iriadam M, Ozardah I, Demir N. Letrozole-induced polycystic ovaries in the rat: a new model for cystic ovarian disease. Arch Med Res 2004;35:103–108.
31. Fisher CR, Graves KH, Parlow AF, Simpson ER. Characterization of mice deficient in aromatase (ArKO) because of targeted disruption of the cyp19 gene. Proc Natl Acad Sci USA 1998;95:6965–6970.
32. Zachos NC, Billiar RB, Albrecht ED, Pepe GJ. Developmental regulation of baboon fetal ovarian maturation by estro- gen. Biol Reprod 2002;67:1148–1156.
33. Faiman C, Reyes FI, Dent DW, Fuller GB, Hobson WC, Thliveris JA. Effects of long-term testosterone exposure on ovarian function and morphology in the rhesus monkey. Anat Rec 1988;222:245–251.
34. Billiar RB, Richardson D, Anderson E, Mahajan D, Little B. The effect of chronic and acyclic elevation of circulating androstenedione or estrone concentrations on ovarian function in the rhesus monkey. Endocrinology 1985;16:2209–2220.
35. Mahajan DK. Polycystic ovarian disease: animal models. Endocrinol Metab Clin North Am 1988;17:705–732.
36. Billiar RB, Richardson D, Schwartz R, Posner B, Little B. Effect of chronically elevated androgen or estrogen on the glucose tolerance test and insulin response in female rhesus monkeys. Am J Obstet Gynecol 1987;157:1297–1302.
37. Vendola K, Zhou J, Wang J, Famuyiwa OA, Bievre M, Bondy CA. Androgens promote oocyte insulin-like growth factor I expression and initiation of follicle development in the primate ovary. Biol Reprod 1999;61:353–357.
38. Stubbs SA, Hardy K, Da Silva-Buttkus P, Stark J, Webber LJ, Flanagan AM, Themmen AP, Visser JA, Groome NP, Franks S. Anti-Mullerian hormone (AMH) protein expression is reduced during the initial stages of follicle develop- ment in human polycystic ovaries. J Clin Endocrinol Metab 2005;90:5536–5543.
39. Barker DJP. Mothers, Babies and Health in Later Life. Edinburgh: Churchill Livingstone, 1994.
40. Abbott DH, Dumesic DA, Eisner, Kemnitz JW, Goy RW. The prenatally androgenized female rhesus monkey as a model for polycystic ovarian syndrome. In: Azziz R, Nestler JE, Dewailly D, eds. Androgen Excess Disorders in Women. Philadelphia: Lippincott-Raven Press, 1997:369–382.
41. Abbott DH, Bruns CM, Barnett DK, Dumesic DA. Fetal programming of polycystic ovary syndrome. In: Kovacs WG, Norman RL, eds. Polycystic Ovary Syndrome, 2nd ed. Cambridge: Cambridge University Press, 2006.
42. Padmanabhan V, Manikkam M, Recabarren S, Foster D. Prenatal testosterone programs reproductive and metabolic dysfunction in the female. Mol Cell Endocrinol 2005;246:165–174.
43. Barbieri RL, Saltzman DH, Torday JS, Randall RW, Frigoletto FD, Ryan KJ. Elevated concentrations of the beta- subunit of human chorionic gonadotropin and testosterone in the amniotic fluid of gestations of diabetic mothers. Am J Obstet Gynecol 1986;154:1039–1043.
44. Barnes RB, Rosenfield RL, Ehrmann DA, et al. Ovarian hyperandrogynism as a result of congenital adrenal virilizing disorders: evidence for perinatal masculinization of neuroendocrine function in women. J Clin Endocrinol Metab 1994;79:1328–1333.
45. Sir-Petermann T, Maliqueo M, Angel B, Lara HE, Perez-Bravo F, Recabarren SE. Maternal serum androgens in preg- nant women with polycystic ovarian syndrome: possible implications in prenatal androgenization. Hum Reprod 2002;17:2573–2579.
46. Gitau R, Adams D, Fish NM, Glover V. Fetal plasma testosterone correlates positively with cortisol. Arch Dis Child Fetal Neonatal Ed 2005;90:F166–F169.
272 Abbott et al.
47. West C, Foster DL, Evans NP, Robinson J, Padmanabhan V. Intrafollicular activin availability is altered in prenatal- androgenized lambs. Mol Cell Endocrinol 2001;185:51–59.
48. Sullivan SD, Moenter SM. Prenatal androgens alter GABAergic drive to gonadotropin-releasing hormone neurons:
implications for a common fertility disorder. Proc Natl Acad Sci USA 2004;101:7129–7134.
49. Foecking EM, Szabo M, Schwartz NB, Levine JF. Neuroendocrine consequences of prenatal androgen exposure in the female rat: absence of luteinizing hormone surges, suppression of progesterone receptor gene expression, and accelera- tion of the gonadotropin-releasing hormone pulse generator. Biol Reprod 2005;72:1475–1483.
50. Dumesic DA, Schramm RD, Abbott DH. Early origins of polycystic ovary syndrome (PCOS). Reprod Fertil Dev 2005;17:349–360.
51. Recabarren SE, Padmanabhan V, Codner E, et al. Postnatal developmental consequences of altered insulin sensitivity in female sheep treated prenatally with testosterone. Am J Physiol Endocrinol Metab 2005;E807–E806.
52. Ibanez L, Potau N, Zampolli M, et al. Hyperinsulinemia in postpubertal girls with a history of premature pubarche and functional ovarian hyperandrogenism. J Clin Endocrinol Metab 1996:81:1237–1243.
53. Sir-Petermann T, Hitchsfeld C, Maliqueo M, et al. Birth weight in offspring of mothers with polycystic ovarian syn- drome. Hum Reprod 2005;20:2122–2126.
54. Laitinen J, Taponen S, Martikainen H, et al. Body size from birth to adulthood as a predictor of self-reported polycystic ovary syndrome symptoms. Int J Obes Relat Metab Disord 2003;27:710–715.
55. Sadrzadeh S, Klip WA, Broekmans FJ, et al.; OMEGA Project Group. Birth weight and age at menarche in patients with polycystic ovary syndrome or diminished ovarian reserve, in a retrospective cohort. Hum Reprod 2003:18:2225–2230.
56. Manikkam M, Crespi EJ, Doop DD, et al. Fetal programming: prenatal testosterone excess leads to fetal growth retar- dation and postnatal catch-up growth in sheep. Endocrinology 2004;145:790–798.
57. Slob AK, den Hamer R, Woutersen PJ, van der Werff ten Bosch JJ. Prenatal testosterone propionate and postnatal ovarian activity in the rat. Acta Endocrinol (Copenh) 1983;103:420–427.
58. Wolf CJ, LeBlanc GA, Gray LE Jr. Interactive effects of vinclozolin and testosterone propionate on pregnancy and sexual differentiation of the male and female SD rat. Toxicol Sci 2004;78:135–143.
59. Wood JR, Ho CK, Nelson-Degrave VL, McAllister JM, Strauss JF III. The molecular signature of polycystic ovary syndrome (PCOS) theca cells defined by gene expression profiling. J Reprod Immunol 2004;63:51–60.
60. Yildiz BO, Yarali H, Oguz H, Bayraktar M. Glucose intolerance, insulin resistance, and hyperandrogenemia in first degree relatives of women with polycystic ovary syndrome. J Clin Endocrinol Metab 2003;88:2031–2036.
61. Fox R. Prevalence of a positive family history of type 2 diabetes in women with polycystic ovarian disease. Gynecol Endocrinol 1999;13:390–393.
62. Sir-Petermann T, Angel B, Maliqueo M, Carvajal F, Santos JL, Perez-Bravo F. Prevalence of Type II diabetes mellitus and insulin resistance in parents of women with polycystic ovary syndrome. Diabetologia 2002;45:959–964.
63. Bruns CM, Baum ST, Colman RJ, et al. Insulin resistance and impaired insulin secretion in prenatally androgenized male rhesus monkeys. J Clin Endocrinol Metab 2004;89:6218–6223.
64. Resko JA, Buhl AE, Phoenix CH. Treatment of pregnant rhesus macaques with testosterone propionate: observations on its fate in the fetus. Biol Reprod 1987;37:1185–1191.
65. Sarma HN, Manikkam M, Herkimer C, Dell’Orco J, Foster DL, Padmanabhan V. Fetal programming: excess prenatal testosterone reduces postnatal LH, but not FSH responsiveness to estradiol negative feedback in the female. Endocri- nology 2005;146:4281–4291.
66. Dumesic DA, Schramm RD, Peterson E, Paprocki AM, Zhou R, Abbott DH. Impaired developmental competence of oocytes in adult prenatally androgenized female rhesus monkeys undergoing gonadotropin stimulation for in vitro fertilization. J Clin Endocrinol Metab 2002;87:1111–1119.
67. Bagavandoss P, England B, Asirvatham A, Bruot BC. Transient induction of polycystic ovary-like syndrome in imma- ture hypothyroid rats. Proc Soc Exp Biol Med 1998;219:77–84.
68. Brawer JR, Munoz M, Farookhi R. Development of the polycystic ovarian condition (PCO) in the estradiol-valerate treated rat. Biol Reprod 1986;35:647–655.
69. Quant LM, Hutz RJ. Induction by estradiol 17E of polycystic ovaries in the guinea pig. Biol Reprod 1993;48:1088–1094.
70. Lagace DC, Nachtigal MW. Valproic acid fails to induce polycystic ovary syndrome in female rats. Prog Neuropsychopharmacol Biol Psychiatry 2003;27:587–594.
71. Ferin M, Morrell M, Xiao E, et al. Endocrine and metabolic responses to long-term monotherapy with antiepileptic drug valproate in the normally cycling rhesus monkey. J Clin Endocrinol Metab 2003;88:2908–2915.
72. Furudate S, Nakano T. PMSG-induced persistent estrus in rats as a model for polycystic ovary disease: characteristics and restoration to the normal cycle. Exp Anim 1989;38:121–126.
73. Jones HM, Vernon MW, Rush ME. Systematic studies invalidate the neonatally androgenized rat as a model for poly- cystic ovary disease. Biol Reprod 1987;36:1253–1265.
