Modern trends Edward E. Wallach, M.D., Associate Editor Vol. 58, No.4, October 1992

FERTILITY AND STERILITY

Printed on acid-free paper in U.S.A.

Copyright" 1992 The American Fertility Society

Puberty and polycystic ovarian syndrome: the insulin/insulin-like growth factor I hypothesis

Frank Nobels, M.D.* Didier Dewailly, M.D.t Department of Endocrinology and Reproductive Function, Centre Hospitalier Regional de Lille, Lille, France

Objectives: To provide an up-to-date review of studies that have examined the physiological effects of insulin and insulin -like growth factor 1 (I G F -I) on ovarian growth, maturation, and steroid synthesis, their physiological role in puberty, and their pathophysiological role in polycystic ovarian syndrome (PCOS). To deduce from these data a hypothesis, explaining the pathogenetic connections between puberty and PCOS. Data Identification: The most relevant studies related to this topic have been identified through a computerized bibliographic search (MEDLINE) and through manual scanning of what has been published during recent years in the most important journals in the field of reproductive endocrinology. Results: Insulin and IGF -I stimulate ovarian growth and potentiate the actions of gonadotropins on ovarian steroid synthesis. Insulin also augments the bioactive concentrations of IGF-I and androgens through regulation of the synthesis of their respective binding proteins insulin -like growth factor-1 binding protein (IGFBP-1) and sex hormone-binding globulin (SHBG) in the liver. Insulin and IGF-I might also be able to increase the adrenal sensitivity to adrenocorticotropic hormone (ACTH). Insulin resistance with compensating hyperinsulinism is a common feature of PCOS. It is also a normal phenomenon during puberty. Polycystic ovarian syndrome often develops during puberty. Ultrasonographic investigations suggest that it is much more common during adolescence than generally assumed. Actually, there is a striking resemblance between the endocrine characteristics of puberty and some forms of PCOS. Both conditions are characterized by insulin resistance, hyperpulsatile gonadotropin secretion, hyperactive ovarian and adrenal androgen synthesis, and decreased levels of IGFBP-1 and SHBG. Conclusion: We propose the progressively increasing insulin levels and IGF-I activity during puberty as inducing factors in the development of PCOS {n susceptible subjects. Fertil Steril 1992;58:655-66 Key Words: Polycystic ovarian syndrome, puberty, insulin, insulin-like growth factor I, somatomedins, insulin-like growth factor binding proteins, sex hormone-binding globulin, growth, sex maturation, ovary

The association of menstrual disturbances, infertility, and hirsutism with enlarged cystic ovaries has been described originally by Stein and Leventhal (1). It defines the classical form of polycystic ovarian

Received May 5, 1992.

* Present address: Department of Endocrinology, Onze-LieveVrouwhospital, Aalst, Belgium. t Reprint requests: Didier Dewailly, M.D., Department of Endocrinology and Reproductive Function, Centre Hospitalier Regional de Lille, 6 Rue du Pro Laguesse, 59037 Lille Cedex, France. Vol. 58, No.4, October 1992

syndrome (PCOS). Typical endocrinological features are overproduction of ovarian androgens and excessive secretion of luteinizing hormone (LH), with an increased ratio of LH to follicle-stimulating hormone (FSH) (2). This full-blown syndrome is relatively rare. At present, however, polycystic ovaries (PCO) are often detected in patients with more subtle clinical and endocrinological expression (3) because of advances in ovarian imaging techniques (4). This results in a dramatic increase of the prevalence of the disease and a shift of the peak incidence Nobels and Dewailly

Puberty and pcas

655

to a younger age group (5, 6). These observations elicit substantial debate concerning the definition of the syndrome, the diagnostic value of endocrine investigations, and the pathogenetic mechanisms involved. Irregular menstrual cycles are common during adolescence. Evaluation of adolescents with menstrual disturbances reveals the frequent presence of hyperandrogenism and ovaries containing several cysts (5). The striking resemblance to PCOS suggests a pathogenetic link between the two conditions. Recent evidence reveals that insulin resistance, with compensating hyperinsulinemia, is a normal phenomenon during puberty (7-13). Furthermore, during the last years, several authors drew attention to the frequent coexistence of hyperinsulinism and PCO (14-23). A substantial body of clinical and experimental data, obtained in human and animal beings, shows that insulin and insulin-like growth factor I (IGF-I) are capable of stimulating ovarian growth and steroid synthesis (24, 25). Attempting to combine these observations, we provide a hypothesis that proposes insulin and/or IGF-I as the pathogenetic connections between puberty and PCOS, at least in some forms of this heterogenous disease. The first and second sections of this study will deal with the involvement of insulin and IGF -I in puberty and PCOS, respectively. Then, the similarities between the two conditions will be highlighted in the third section before we discuss the hypothesis in the fourth section. INSULIN AND IGF-I IN NORMAL PUBERTY Insulin Resistance During Puberty

In a cross-sectional study in 224 normal nonobese subjects, 1 to 20 years of age, Laron and co-workers (7) showed that fasting insulin and C peptide sharply increase at the onset of puberty, although glucose levels remain unchanged. Peak fasting insulin levels, frequently surpassing 75 pmoljL, are reached at midpuberty, regardless of chronological age. After adolescent maturation, the levels progressively decline to reach prepubertal values again in early adulthood. These data suggest that puberty is accompanied by a physiological insulin resistance. Several studies based on oral (8-10) or intravenous (11) glucose tolerance tests (OGTT or IGTT) or hyperglycemic clamps (12, 13) confirm these findings. Euglycemic clamps in combination with measurement of glucose kinetics localize this insulin resistance at the level of the peripheral tissues, 656

Nobels and Dewailly

Puberty and peas

without affecting the liver. The defect is restricted to glucose metabolism. Measurement of amino acid kinetics reveals no abnormalities (13). Several points of evidence suggest that the rise in plasma growth hormone (GH) concentrations during puberty is responsible for the defective insulin action. Administering GH infusions to adults imitates the pubertal insulin-resistant state, with a selective hampering of peripheral glucose uptake, without affecting the anabolic functions of insulin (26). During GH treatment in 16 short normal prepubertal children. Hindmarsch and Brook (27) observed a significant rise in fasting insulin concentrations. Because all subjects remained prepubertal, both· clinically and biochemically, an effect of sex steroids could be excluded. Amiel and associates (28) found a negative correlation between insulin sensitivity, determined by hyperinsulinemic clamps, and 24hour integrated GH concentrations in 14 nondiabetic and 9 diabetic prepubertal and pubertal children. These findings indicate GH as the major responsible factor for the insulin resistance of puberty. It remains unclear whether this is a direct effect of GH on peripheral tissues or whether it is mediated through local production of IGF-1. Role of Insulin in Somatic Growth

The opposing effects of two anabolic hormones during a period of active growth seem paradoxical. However, Amiel and associates (13) demonstrated that the insulin resistance is restricted to glucose metabolism, without affecting amino acid metabolism. This implies that the compensatory hyperinsulinemia actually amplifies protein anabolism, favoring growth. Moreover, a considerable amount of evidence has been accumulated throughout the last years, suggesting that insulin is an important modulator ofIGF-I action. Insulin-like growth factor I, a structural homologue of proinsulin, mediates the growth-stimulating effects of GH (29). Although the liver is the major production site, many tissues in the body express the peptide and its receptors (30). This suggests that, besides its classical endocrine role, it also acts as a part of an autocrine-paracrine system. Insulin-like growth factors are bound to several specific proteins in the circulation (31). At present, four different insulin-like growth factor binding proteins (IGFBP) have been identified and characterized (32). In particular, IGFBP-1 has generated much interest because it might be an important regulator of IGF-I function, acting as a competitive binding site and thus reducing the interaction ofIGF-I with its receptors (33-36). SevFertility and Sterility

eral studies provide evidence that the plasma levels ofIGFBP-1 are mainly controlled by insulin. Insulin inhibits the production of IGFBP-1 by a human hepatoma cell line, at concentrations corresponding to those found in portal blood (37, 38). In addition, a recent report indicates that insulin might also facilitate the transcapillary transport of IGFBP-1 (39). Insulin -like growth factor-1 binding protein concentrations are elevated in states of hypoinsulinemia, as in type 1 diabetes (40,41), fasting (42), and exercise (43), and suppressed in states of hyperinsulinemia, as in obesity (44), Cushing's syndrome (45), and in patients with insulinoma (40). Euglycemic hyperinsulinemic clamps cause a fast and significant decrease in IGFBP-11evels (40). The tight relationship between IGFBP-1 and insulin is further illustrated by the observation that the circadian variation of IGFBP-1 is inversely related to the secretory pattern of insulin (46). The fasting levels of IGFBP-1 progressively decline throughout childhood, attaining a nadir during puberty, and again showing a slight increase toward adulthood (47). This profile is opposite to the one found for total IGF-I, insulin, and C peptide. As such, the hyperinsulinemia of puberty may promote growth through direct anabolic effects and through suppression of IGFBP-1 concentrations. Moreover, animal data suggest that insulin is also necessary for IGF-I production. Insulin-deficient animals show reduced serum IGF-I levels that can be restored by insulin but not by GH treatment (48). At present, however, no direct effects of insulin on IGF-I production in vitro have been reported. Role of Insulin and IGF-I in Ovarian Growth and Maturation

A vast literature, mostly based on animal studies, supports the view that insulin and the related IGFI may play an important part in ovarian physiology. Two excellent reviews on this subject have been published by Poretsky and Kalin (24) and Adashi et al. (25). Active insulin and IGF-I receptors have been demonstrated in animal and human ovaries (24). Both hormones exert a mitogenic effect (4951) by stimulating the proliferation of bovine and porcine (but not murine) granulosa cells (GCs) in culture. In addition, they augment the ovarian steroidogenetic capacity by potentiating the effects of the gonadotropins. They stimulate FSH-induced estradiol (E 2 ) (52,53) and progesterone (P) (50-52) synthesis in GCs and LH-induced androstenedione (A) synthesis in theca and stroma cells (54-57). These actions are both dose- and time-dependent. Vol. 58, No.4, October 1992

Combining both hormones has no additive effects. The mechanisms of these replicative and cytodifferentiative functions of insulin and IGF-I remain a matter of study. Cyclic adenosine 3':5' monophosphate probably functions as a second messenger (57). Direct effects on the induction and activation of the cholesterol side chain cleavage (58) and aromatase enzymes (53, 59) have been reported. Furthermore, augmentation of the ability of FSH to generate LH receptors in GCs has been observed (25, 60). The above experiments have almost exclusively been conducted in prepubertal animals. Immature ovaries were preferred because they allow a relatively easy separation of the granulosa and theca-interstitial compartments for in vitro culture. Studies in human beings are relatively rare but seem to confirm the animal findings (61-63). In all these observations, IGF-I is active at physiological concentrations, whereas insulin has little or no effect at concentrations known to saturate its receptors. Only in theca cells a slight synergistic effect on LH-induced androgen synthesis has been shown at physiological insulin concentrations (57). Because the dose requirements for insulin generally exceed those achievable in vivo, the concept has emerged that only IGF-I is relevant to the regulation of ovarian function. The in vitro effects of insulin probably occur through interaction with IGF-I receptors. Insulin and IGF-I receptors share many structural and functional properties, allowing a certain degree of cross-reaction at supraphysiological concentrations (64, 65). Moreover, the presence of hybrid receptors, which contain a combination of a- and iJ-subunits of both receptors, has been described recently in other tissues than ovaries (66, 67). Because supraphysiological insulin concentrations are needed to interact with these receptors, one could argue that insulin cannot play an active role in ovarian physiology in vivo. Several artifacts, however, make it extremely difficult to interpret in vitro data. Hernandez and co-workers observed a substantial degree of insulin degradation and nonspecific adsorption to the substratum of the culture dishes (57). Moreover, the sustained exposure to high concentrations of insulin in the culture medium induces a significant down regulation of insulin receptors. Because these problems don't occur with IGF-I, a true comparison between the dose-effect relationships of both hormones is impossible under in vitro conditions. Insulin might also be able to accomplish an indirect effect in vivo by influencing the concentrations of IGFBP-1, potentiating IGF-I activity. ReNobels and Dewailly

Puberty and

peas

657

cently, IGFBP-1 production has been identified in animal and human GCs (68-70). So far, the possible role of insulin in the local regulation of this production has not been evaluated. No evidence that demonstrates IGF-I synthesis in human GCs exists as yet, although it has been well documented in rat and porcine GCs (71-73). This production is probably GH-dependent (30, 72). Studies in GC cultures reveal that GH is able to activate the steroidogenesis (74). It remains unknown whether it concerns a direct effect or one mediated through local IGF-I production. In short, a wealth of publications has elucidated a network of endocrine control mechanisms of ovarian function in which the major players are not only the gonadotropins but also metabolic hormones, such as GH, insulin, and IGF-I (Fig. 1). Moreover, the existence of an intraovarian autoregulation and paracrine regulation has been displayed recently in which diverse growth factors such as the IGFs and their binding proteins, and fibroblast (75), epidermal (EGF) (59, 76), and transforming growth factors (TGF) (77) playa part. The mitogenic effects of several of these polypeptides, including insulin and IGFI, may accomplish an important function in the growth and maturation of the prepubertal ovaries. Once puberty occurs, they potentiate the actions of the gonadotropins on the ovarian steroid synthesis. Insulin Augments Bioactive-Free Sex Steroid Levels

The role of insulin in sexual maturation seems even more complicated because several data indicate that it augments the bioactive-free sex steroid levels by decreasing the concentrations of the carrier protein sex hormone-binding globulin (SHBG). Sex hormone-binding globulin concentrations decline throughout puberty (78-82). Because androgens inhibit and estrogens (Es) stimulate hepatic SHBG production, alterations in the levels of these hormones have been held responsible for this phenomenon (83). However, several observations cast doubt on this theory. Sex hormone-binding globulin levels decrease in both sexes, despite higher E concentrations in girls (78-82). Cunningham and colleagues (84) observed a clear decline in SHBG levels in the second decade of life in four boys with untreated isolated gonadotropin deficiency and in two individuals with complete androgen insensitivity. Holly and co-workers (78) presented data suggesting that insulin is the major determinant ofthe pubertal SHBG drop. They obtained strongly negative correlations between falling levels of SHBG and rising levels of 658

N obels and Dewailly

Puberty and peas

/

ONSET OF PUBERTY ~

pulsatile GaRB secretion

.nriaa

pul_tile GH secretion

l~.-:·£ .L I"-~ -::"S~ ~ ~'BPI I

and differentiation~

~

~~+

+

SEXUAL MATURATION

SOMATIC GROWfH

Figure 1 Simplified scheme of the endocrine interactions that regulate the pubertal development. GnH = gonadotropins; GH = growth hormone; IGF-I = insulin-like growth factor I; SHBG = sex hormone-binding globulin; IGFBP-l = IGF-l binding protein.

insulin in both sexes in 69 adolescents. Moreover, the decline of the insulin-dependent IGFBP-1 was also strongly correlated with SHBG. Actually, there is a striking resemblance between the two binding proteins in essentially all situations tested. Both increase in states ofinsulinopenia (40-43, 85) and decrease in states of insulin excess (40,41,44,45,86, 87). Insulin is able to inhibit the production of both binding proteins by human hepatoma tissue (37,88). Despite all similarities, the temporal relation between SHBG and insulin is much slower than between IGFBP-1 and insulin. This is demonstrated by the absence of a circadian variation in SHBG and the observation that the rapid changes in insulin levels during an OGTT don't influence SHBG concentrations. These data suggest that insulin performs an additional regulatory function in sexual maturation and growth by influencing the two key binding proteins IGFBP-1 and SHBG. Thus, insulin and IGF -I fulfill a sophisticated regulatory function in synchronizing sexual maturation and growth. Because insulin exerts diverse metabolic effects, it is excellently suited to link both processes to the metabolic and nutritional well-being of the organism. This adaptation is extremely important in animals because, on one hand, it allows saving energy in times of shortage and, on the other hand, acceleration of maturation and therefore of the ability to procreate in times of food abundance. This theory can explain several clinical observations, for example, GH administration to GH-deficient children not only accelerates growth but also promotes genital development (89). Delayed puberty and growth spurt are typical features of malnutrition or badly controlled type I diabetes (90, 91). Hyperinsulinemia, associated with hyperphagia and obesity of hypothalamic origin, allows a normal or even Fertility and Sterility

excessive growth in patients with craniopharyngioma and documented GH deficiency (92). INSULIN AND IGF-I IN peos Insulin Resistance in Women With peos High circulating insulin concentrations have been reported in some forms of PCOS (24,93). Since the original report by Kahn and colleagues (94) in 1976, several disorders characterized by the association of hyperandrogenism, extreme insulin resistance, and acanthosis nigricans (HAIR-AN syndromes) have been identified. Regardless of the etiology of the insulin resistance, these patients invariably show signs of ovarian hyperstimulation, with hirsutism, clitoromegaly, elevated androgen levels, and enlarged cystic ovaries. These syndromes clearly represent only one extreme of the broad spectrum of PCOS. During the last years, however, several authors have described the frequent existence of insulin resistance in the more common forms of PCOS in obese as well as in lean subjects (14-23). The exact prevalence is still not precisely known because of differences in techniques and cutoff values used to demonstrate hyperinsulinism. Moreover, study populations are not comparable because of selection bias and the use of different definitions of PCOS (clinical, endocrinological, morphological). Conway and coworkers (21), using basal plasma insulin levels, found hyperinsulinism in 30% of their lean patients with PCOS. Falcone and associates (22), using IGTT with calculation of insulin resistance by minimal model analysis (95), reported a prevalence of 63%. In our experience with OGTT (23), 27% and 42% of our lean and obese patients, respectively, showed an increased area under the curve of plasma insulin levels (>2 SD above lean control subjects). The molecular abnormalities causing insulin resistance are well documented in several cases of HAIR-AN syndrome (94). Whereas they are still poorly understood in the more common forms of PCOS with hyperinsulinism. For example, it is still unknown whether the defects in insulin action are congenital or acquired. Role of Insulin Resistance in Ovarian Abnormalities of peos Recently, a remarkable clinical observation has delivered proof that hyperinsulinemia is the cause of the ovarian hyperstimulation in HAIR-AN syndrome (96). Polycystic ovarian syndrome with severe hyperandrogenism occurred in a young black female Vol. 58, No.4, October 1992

with type B insulin resistance because of anti-insulin receptor antibodies. After spontaneous resolution of her insulin resistance, the hyperandrogenism disappeared, and she delivered a normal female neonate 13 months later. Nestler and associates (97) showed that suppression of insulin production with diazoxide causes a significant decline of total testosterone (T) plasma levels in severely hyperinsulinemic women. Clinical studies in PCOS patients with less severe forms of insulin resistance are less convincing. Although significant correlations between plasma insulin and androgen levels were initially reported in small groups of selected patients (1420), recent data in larger patient populations with milder forms of the disease reveal a less tight relationship (21, 98). In vitro studies show that insulin and IGF-I stimulate the production of T and A by normal as well as PCO. Barbieri and co-workers (99) compared the effects of insulin in cultured fragments of stroma from three normal women and four patients with PCOS. They showed that insulin is able to accomplish direct stimulation in PCO, whereas it can only exert this effect through synergism with LH in normal ovaries. Moreover, more androgens were produced per unit of weight in tissue from polycystic as compared with normal ovaries. These data suggest that PCO are more sensitive to the stimulating effects of insulin. This needs to be confirmed in a larger series. Insulin Augments Bioactive-Free Steroid and IGFBP-l Levels in peos Several investigators report low levels of SHBG in lean and in obese women with PCOS, although the obese subjects tend to have the lowest concentrations (3, 100, 101). The levels of IGFBP-l follow the same pattern and are strongly correlated with those of SHBG (21, 102). Both binding proteins show negative correlations with insulin levels. Nestler and colleagues (97) proved that hyperandrogenism is not the cause of the low SHBG levels. They demonstrated in six hyperinsulinemic women with PCOS that reducing the androgen levels substantially by means of the long-acting gonadotropin-releasing hormone agonist (GnRH-a), leuprolide acetate, does not influence the SHBG concentrations. On the other hand, reduction of the insulin levels by means of diazoxide resulted in significant elevations of SHBG. In summary, insulin resistance is frequently present in PCOS. Direct and indirect effects of insulin on ovarian physiology are probably important in the Nobels and Dewailly

Puberty and peas

659

pathogenesis of some forms of the syndrome. Insulin plays also an active role in the processes of sexual maturation and growth during normal puberty. In fact, PCOS and puberty share many common features, suggesting pathogenetic links. PUBERTY AND PCOS: PATHOGENETIC LINKS Polycystic Ovarian Syndrome Originates in Puberty

Clinical observation teaches that PCOS often develops during adolescence (103). Excessive hair growth usually originates from before the onset of menstrual cycles. Menarche tends to be delayed and even primary amenorrhea may occur, although rare. Irregular cycles, although considered a normal phenomenon during the first gynecological years, frequently continue into adulthood. Interesting observations by Venturoli and coworkers (104) suggest that PCOS is much more common during adolescence than is generally assumed. They describe the frequent existence of PCO in adolescents with irregular menstrual cycles. These girls tend to have a lower frequency of ovulation, and higher LH, T, and A levels than adolescent girls with regular cycles. At ultrasonographic evaluation, they present enlarged PCO with stroma hyperplasia. These endocrine and morphological characteristics tend to improve with age (5). Long-term follow-up is needed to investigate which ofthem will ultimately develop PCOS. In a large longitudinal study of 200 schoolgirls, Apter and Vihko (105) revealed that the girls with the highest androgen concentrations during puberty showed the lowest fertility rates in the third decade of life. These studies suggest that anovulation, hyperandrogenism, and PCO may evolve from early puberty through adolescence into adulthood. Polycystic Ovarian Syndrome, a State of Hyperpuberty

Many of the endocrine changes that occur during puberty are also present in PCOS, often in an excessive way. First, morphological changes in ovaries of patients with PCOS may be considered as an exaggeration of those that occur during puberty. The presence of multicystic ovaries is considered a normal phenomenon during early puberty, and progressively disappears with the institution of ovulation. The ultrasonographic characteristics of PCO, which are enlarged ovaries containing numerous 660

Nobels and Dewailly

Puberty and peDS

small cysts and an increased amount of hyperdense stroma, must be differentiated from these multicystic ovaries (106). The latter are not enlarged and lack stroma hyperplasia. In several early studies this distinction was not made, which caused considerable confusion. However, several cases of true PCO have been described in girls during early puberty, even before menarche (107, 108). A few cases have been described in girls with precocious puberty (109). Probably multicystic and PCO are two extremes of the same spectrum of anatomical alterations. Second, the pattern of gonadotropin secretion in PCOS seems an excessive amplification of that of puberty. Elevated basal levels of LH, an increased ratio of LH:FSH, and a hyper-response of LH to exogenous GnRH are typical features of PCOS (2, 3, 103). Studies of pulsatile gonadotropin secretion reveal an increase in LH pulse amplitude. Several authors also report an elevated LH pulse frequency, but this is not uniformly accepted (3). Furthermore, an abnormal diurnal pattern of gonadotropin secretion, with high levels during the morning instead of the normal nocturnal rise, has been demonstrated in adolescent girls with PCOS (110). Third, adrenarche, another milestone of sexual maturation, is also excessively accentuated in PCOS. The activities of 17a-hydroxylase and 17-20-desmolase, which are actually two parts of a single enzyme cytochrome P450c17, increase during early puberty (111). This allows a higher production rate of C19-steroids by the adrenals, causing adrenarche. The same alterations occur in the ovaries and constitute the first step in the maturation toward adult ovarian steroid synthesis. Dehydroepiandrosterone sulfate, the most representative adrenal androgen, is frequently elevated in PCOS (112). It is not suppressed by inhibition of ovarian steroidogenesis with GnRH-a, proving its adrenal origin (113). Moreover, adrenocorticotropic hormone (ACTH) stimulation tests provoke excessive responses of adrenal C19steroid levels in patients with PCOS (114-118). This has been interpreted by most authors as heterozygous or late onset homozygous forms of congenital adrenal hyperplasia (116, 118-120). However, several reports refute this conclusion (115, 117, 120, 121). Because the responses of all steroids on the pathway beyond P tend to be exaggerated, these data can also be explained by increased adrenal P450c17a activity, as proposed by Rosenfield and coworkers (122). This is another example of the amplification of a normal pubertal maturation process. It is tempting to speculate that insulin and/or IGF-I might be able to stimulate both LH -dependent Fertility and Sterility

P450c17a activity in ovaries and ACTH-dependent activity in adrenals, as has been shown for IGF-I in bovine cells in culture (123). Finally, as discussed above, the decline of SHBG and IGFBP-1 plasma levels during puberty also occurs in an excessive manner in women with PCOS. This is clearly related to hyperinsulinism. In summary, clinical and epidemiologic data show that PCOS usually begins during early adolescence. The endocrine and morphological characteristics of PCOS often reflect in an exaggerated way what happens during puberty. Thus PCOS can be considered as a state of hyperpuberty. In particular, the presence of insulin resistance in a large subgroup of PCOS patients and its universal existence during puberty suggest a pathogenetic link. PUBERTY AND PCOS: THE INSULIN-IGF-I HYPOTHESIS

Based on the above observations, we postulate that excessive ovarian stimulation caused by the progressively rising insulin and IGF-I levels during puberty induces a PCOS in predisposed girls. Before we provide a detailed description of this hypothesis, several questions need to be answered. Since in vitro studies reveal that insulin and IGFI stimulate the synthesis not only of androgens by theca and interstitial cells (54-57) but also ofE (52, 53) and P (50-52) by GCs, one might wonder why high insulin and IGF-I levels are manifested clinically as hyperandrogenism. Poretsky (93) postulated that the increased ovarian concentrations of androgens induce follicular atresia, resulting in a gradual elimination of E- and P-secreting cells, which are progressively replaced by androgen-producing tissue. Another possible explanation is the presence of an intraovarian inhibitor of aromatase activity. The most likely candidates are EGF and the EGF-receptor agonist TGF-a, because they have potent inhibitory effects on rat and human steroidogenesis (59, 76, 77). Recent evidence suggests that TGF-a is synthesized in the ovary (77). Nevertheless, as yet no definite data exist proving its importance in PCOS nor its interaction with insulin and/or IGF-L The second point that needs to be discussed is the apparent paradox that insulin can remain active on ovarian steroidogenesis, whereas its effects are hampered in other tissues by insulin resistance. Several possible explanations can be proposed (93). Because insulin fulfills a vast array of functions, one could imagine selective defects. At the ovarian level, glucose transport might be hindered while effects on Vol. 58, No.4, October 1992

ovarian growth and steroid synthesis might remain undisturbed. In addition, organ-specific insulin resistance could occur. This is illustrated by low IGFBP-1 and SHBG levels in PCOS (21, 100-102) that are the result of insulin action on the liver. The suppression of IGFBP-1 concentrations allows insulin to exert indirect effects by potentiation of IGF1. Moreover, insulin is able to interact directly with IGF-I (64,65) or hybrid receptors (66, 67) that contain a combination of a- and {3-subunits of both receptors. Because their affinity for insulin is relatively weak, only supraphysiological concentrations can allow this interaction. Another subject of discussion is the position of LH hyperpulsatility in this hypothesis. Mechanick and Futterweit (124) proposed that excessive pubertal maturation of the hypothalamic structures involved in sex steroidal feedback produces a state of inappropriate gonadotropin secretion that is responsible for the ovarian hyperstimulation, leading to PCOS. However, in some patients with typical clinical and ultrasound features of PCOS, LH levels remain normal, even when a careful analysis ofpulsatile LH secretion is performed (3). In addition, Stanhope and colleagues (107) described the ultrasonographic detection of PCOS in a girl with delayed puberty because of gonadotropin deficiency, caused by a hypothalamic tumor. During treatment with pulsatile administration of a fixed low dose of GnRH, a pattern of high-amplitude LH pulses ensued, which was similar to that described in adolescent girls with PCOS (110). As such, although it cannot be excluded that in some cases the hypothalamus may be the primary generator of excessive GnRH, these findings favor an abnormal feedback signal from the ovary as the cause of the excessive gonadotropin secretion (125). Taking into account the above remarks, the following hypothesis can be postulated (Fig. 2). The HYPOTHALAMUS /PITUITARY ~-,+!:...-_-,

_,Z \ m~!t!.I.

byperlDsuhnemla

\

'

_je . .,ret;o. LH

lIbklacti •• IGF

1~"'"

,

~ ~ IGF-BPI

~ SHBG

OVARY

brOllinlar atresia

"~:~O:r~~~:s --+1testrone

+

HYPERANDROGENISM

Figure 2 puberty.

Hypothetical mechanism of onset of peDS during

Nobels and Dewailly

Puberty and peas

661

onset of pulsatile GH secretion during early puberty induces the release of IGF-I by the liver and most other tissues. This mediator of G H action is a potent stimulator of somatic and probably also gonadal growth. Furthermore, GH provokes insulin resistance, which selectively affects peripheral glucose but not amino acid metabolism. The resulting hyperinsulinemia favors protein anabolism, accomplishes direct mitogenic effects on the ovaries, and potentiates IGF-I action through modulation ofthe concentrations of IGFBP-l. Insulin inhibits the production of this binding protein in the liver and possibly also in other organs, e.g., the ovaries. Dnce the pulsatile secretion of gonadotropins sets off, both insulin and IGF-I synergistically augment their effects on ovarian steroid synthesis. This can be mediated through insulin, IGF-I, or hybrid receptors. The resulting ovarian hyperstimulation induces theca-cell hyperplasia, with excessive androgen production in predisposed girls. Furthermore, the bioavailability of the sex hormones is enhanced by insulin-dependent suppression of SHBG release by the liver. The increased ovarian androgen levels cause follicular atresia, impairing E2 production. Dn the other hand, circulating estrone levels are increased because of aromatization of A in adipose tissue (125). The altered endocrine milieu provokes increased pituitary LH secretion, which aggravates the theca-cell stimulation. The combination of theca-cell hyperplasia and the presence of follicles arrested in their maturation constitute the typical histologic features of PCDS. In the light of the above hypothesis, we propose that the transient hyperinsulinemia and the elevated IGF-I levels of puberty induce a PCDS-like state in a substantial number of adolescent girls. After puberty, the insulin and IGF-I levels progressively decline in most patients, resulting in normalization of the clinical and morphological picture. Dnly in a few cases PCDS persists either because hyperinsulinemia persists or because another pathogenic factor has taken over its role in the meantime. In the latter instance, hyperinsulinism only served as inducing event. This might explain why the role of insulin is difficult to demonstrate in several older women with PCDS. Because not all adolescents ultimately develop a PCDS, it is clear that genetic or environmental predisposition is necessary to express the disease. Several pedigrees of PCDS have been described (126, 127). Although the exact genetic basis remains undefined, several conclusions can be drawn. The syndrome may be transmitted as well through the paternal as through the maternal family. The type and 662

Nobels and Dewailly

Puberty and peas

degree of expression are variable within the same family, with some members presenting the fullblown syndrome, and others, e.g., only slight hirsutism. Diabetes mellitus, obesity, hypertension, hyperlipidemia, and ischemic heart disease are frequently associated within the same families. Insulin resistance is probably the factor that links all these entities, as has been proposed by Reaven (128) in his classical description of the syndrome X. As such, hyperinsulinemic subjects with PCDS should be considered at high risk for developing cardiovascular disease (129). Careful assessment of risk factors, with institution of preventive measures when indicated, is advisable. Nutritional intervention could be very important because patients with PCDS are frequently obese (103), and those of normal weight often present bulimia (130), characterized by alternate bouts of carbohydrate binge eating and fasting. These eating disorders can aggravate insulin resistance. CONCLUSION

We reviewed the data supporting the hypothesis that proposes the hyperinsulinism of puberty as the inciting event in the development of PCDS in susceptible subjects. We don't believe that this theory can explain all cases. However, it provides a pathogenetic concept for the important subgroup of hyperinsulinemic PCDS patients. The recognition of this entity is essential, for it fits in a broader syndrome, linking insulin resistance to increased atherogenesis and its associated cardiovascular pathology (128, 129). This has important therapeutic implications. Medication that worsens insulin resistance needs to be avoided. Treatment directed at correction of eating disorders should be encouraged, and preventive measures to decrease cardiovascular risks should be considered. This hypothesis can serve as a working model for future studies. It is obvious that many questions remain unanswered in this exciting new field of insulin research. Acknowledgment. Since the first submission of the manuscript, we have become aware of Dr. S. C. C. Yen's hypothesis, which is in full agreement with ours and which has been summarized elsewhere (see Yen SSC. Chronic anovulation caused by peripheral endocrine disorders. In: Yen SCC, Jaffe RB, editors. Reproductive endocrinology. Philadelphia: WB Saunders, 1991:576-630). Weare indebted to Ms. Els Borghys for her expert English linguistic revision. REFERENCES 1. Stein IF, Leventhal ML. Amenorrhea associated with bilateral polycystic ovaries. Am J Obstet Gynecol 1935;29: 181-91. 2. Rebar R, Judd HL, Yen SCC, Rakoff J, Vandenberg G, Naftolin F. Characterization of the inappropriate gonadotropin

Fertility and Sterility

secretion in polycystic ovary syndrome. J Clin Invest 1976;57:1320-9. 3. Franks S. Polycystic ovary syndrome: a changing perspective. Clin Endocrinol (OxO 1989;31:87-120. 4. Ardaens Y, Robert Y, Lemaitre L, Fossati P, Dewailly D. Polycystic ovarian disease: contribution of vaginal endosonography and reassessment of ultrasonic diagnosis. Fertil Steril 1991;55:1062-8. 5. Venturoli S, Porcu E, Fabbri R, Paradisi R, Ruggeri S, Bolelli G, et al. Menstrual irregularities in adolescents: hormonal pattern and ovarian morphology. Horm Res 1986;24:26979. 6. Brook CGD, Jacobs HS, Stanhope R. Polycystic ovaries in childhood. Br Med J 1988;296:878. 7. Laron Z, Aurbach-Klipper Y, Flasterstein B, Litwin A, Dickerman Z, Heding LG. Changes in endogenous insulin secretion during childhood as expressed by plasma and urinary C-peptide. Clin Endocrinol (Oxf) 1988;29:625-32. 8. Rosenbloom AL, Wheeler L, Bianchi R, Chin FT, Tiwary CM, Grgic A. Age-adjusted analysis of insulin responses during normal and abnormal glucose tolerance tests in children and adolescents. Diabetes 1975;24:820-8. 9. Bloch CA, Clemons P, Sperling MA. Puberty decreases insulin sensitivity. J Pediatr 1987;110:481-7. 10. Hindmarsh P, Di Silvio L, Pringle PJ, Kurtz AB, Brook CGD. Changes in serum insulin concentration during puberty and their relationship to growth hormone. Clin Endocrinol (Oxf) 1988;28:381-8. 11. Smith CP, Archibald HR, Thomas JM, Tarn AC, Williams AJK, Gale EAM, et al. Basal and stimulated insulin levels rise with advancing puberty. Clin Endocrinol (Oxf) 1988;28: 7-14. 12. Caprio S, Plewe G, Diamond MP, Simonson DC, Boulware SD, Sherwin RS, et al. Increased insulin secretion in puberty: a compensatory response to reductions in insulin sensitivity. J Pediatr 1989;114:963-7. 13. Amiel SA, Caprio S, Sherwin RS, Plewe G, Haymond MW, Tamborlane WV. Insulin resistance of puberty: a defect restricted to peripheral glucose metabolism. J Clin Endocrinol Metab 1991;72:277-82. 14. Burghen GA, Givens JR, Kitabchi AE. Correlation of hyperandrogenism with hyperinsulinism in polycystic ovarian disease. J Clin Endocrinol Metab 1980;50:113-6. 15. Chang RJ, Nakamura RM, Judd HL, Kaplan SA. Insulin resistance in non-obese patients with polycystic ovarian disease. J Clin Endocrinol Metab 1983;57:356-9. 16. Pasquali R, Casimirri F, Venturoli S, Paradisi R, Mattioli L, Capelli M, et al. Insulin resistance in patients with polycystic ovaries: its relationship to body weight and androgen levels. Acta Endocrinol (Copenh) 1983;104:110-6. 17. Shoupe D, Kumar DD, Lobo RA. Insulin resistance in polycystic ovary syndrome. Am J Obstet Gyneco11983;14 7:58892. 18. Stuart CA, Peters EJ, Prince MJ, Richards G, Cavallo A, Meyer WJ. Insulin resistance with acanthosis nigricans: the roles of obesity and androgen excess. Metabolism 1986;35: 197-205. 19. Dunaif A, Segal KR, Futterweit W, Dobrjansky A. Profound peripheral insulin resistance, independent of obesity, in polycystic ovary syndrome. Diabetes 1989;38:1165-74. 20. Mahabeer S, Jialal I, Norman RJ, Naiddo C, Reddi K, Joubert SM. Insulin and C-peptide secretion in non obese Vol. 58, No.4, October 1992

patients with polycystic ovarian disease. Horm Metab Res 1989;21:502-6. 21. Conway GS, Jacobs HS, Holly JMP, Wass JAH. Effects of luteinizing hormone, insulin, insulin-like growth factor-I, and insulin-like growth factor small binding protein 1 in the polycystic ovary syndrome. Clin Endocrinol (Oxf) 1990;33:593-603. 22. Falcone T, Finegood DT, Fantus G, Morris D. Androgen response to endogenous insulin secretion during the frequently sampled intravenous glucose tolerance test in normal and hyperandrogenic women. J Clin Endocrinol Metab 1990;71:1653-7. 23. Cortet-Rudelli C, Henric B, Racadot A, Fossati P, Dewailly D. Variabilite des relations entre insuline, LH, SBP et testosterone dans Ie syndrome des ovaires polymicrokystiques. Ann Endocrinol (Paris) 1991;52:191. 24. Poretsky L, Kalin MF. The gonadotropic function of insulin. Endocr Rev 1987;8:132-41. 25. Adashi EY, Resnick CE, D'Ercole AJ, Svoboda ME, Van Wyk JJ. Insulin-like growth factors as intraovarian regulators of granulosa cell growth and function. Endocr Rev 1985;6:400-20. 26. Bratush-Marrain PR, Smith D, DeFronzo RA. The effect of growth hormone on glucose metabolism and insulin secretion in man. J Clin Endocrinol Metab 1982;55:973-82. 27. Hindmarsh PC, Brook CGD. Effect of growth hormone on short normal children. Br Med J 1987;295:573-7. 28. Amiel SA, Sherwin RS, Simonson DC, Lauritano AA, Tamborlane WV. Impaired insulin action in puberty. A contributing factor to poor glycemic control in adolescents with diabetes. N Engl J Med 1986;315:215-9. 29. Rutanen E-M, Pekonen F. Insulin-like growth factors and their binding proteins. Acta Endocrinol (Copenh) 1990; 123: 7-13. 30. D'Ercole AJ, Stiles AD, Underwood LE. Tissue concentrations of somatomedin C: further evidence for multiple sites of synthesis and paracrine or autocrine mechanisms of action. Proc Natl Acad Sci USA 1984;81:935-9. 31. Baxter RC, Martin KL. Binding proteins for the insulinlike growth factors: structure, regulation and function. Prog Growth Factor Res 1989;1:49-68. 32. Cohen P, Fielder PJ, Hasegawa Y, Frisch H, Giudice LC, Rosenfeld RG. Clinical aspects of insulin-like growth factor binding proteins. Acta Endocrinol (Copenh) 1991;124:7485. 33. Clemmons DR, Elgin RG, Han VKM, Casella SJ, D'Ercole AJ, Van Wyk JJ. Cultured fibroblast monolayers secrete a protein that alters the cellular binding of somatomedin-C/ insulin-like growth factor I. J Clin Invest 1986;77:1548-56. 34. DeVroede MA, Tseng LY-H, Katsoyannis PG, Nissley SP, Rechler MM. Modulation of insulin-like growth factor I binding to human fibroblast monolayer cultures by insulinlike growth factor carrier proteins released into the incubation media. J Clin Invest 1986;77:602-13. 35. Ritvos 0, Ranta T, Jalkanen J, Suikkari A-M, Voutilainen R, Bohn H, et al. Insulin-like growth factor (lGF) binding protein from human decidua inhibits the binding and biological action of IGF-1 in cultured choriocarcinoma cells. Endocrinology 1988;122:2150-7. 36. Rutanen E-M, Pekonen F, Makinen T. Soluble 34 K binding protein inhibits the binding of insulin-like growth factor I to its cell receptors in human secretory phase endometrium:

Nobels and Dewailly

Puberty and

peas

663

37.

38.

39.

40.

41.

42.

43.

44.

45.

46.

47.

48.

49.

50.

664

evidence for autocrine/paracrine regulation of growth factor action. J Clin Endocrinol Metab 1988;66:173-80. Singh A, Hamilton-Fairley D, Koistinen R, Seppala M, James VH, Franks S, et al. Effects of insulin-like growth factor type I (IGF-l) and insulin on the secretion of sex hormone binding globulin and IGF-I binding protein (IBPI) by human hepatoma cells. J Endocrinol 1990;124:R1. Cotterill AM, Cowell CT, Silink M. Insulin and variation in glucose levels modify the secretion rates of the growth hormone-independent insulin-like growth factor binding protein-1 in the human hepatoblastoma cell line Hep G2. J Endocrinol 1989;123:R17. Bar RS, Boes M, Clemmons DR, Busby WH, Sandra A, Dake BL, et al. Insulin differentially alters transcapillary movement of intravascular IGFBP-1, IGFBP-2 and endothelial cell IGF-binding proteins in the rat heart. Endocrinology 1990;127:497-9. Suikkari AM, Koivisto VA, Rutanen EM, Yki-Jarvinen H, Karonen SL, Seppala M. Insulin regulates the serum levels of low molecular weight insulin-like growth factor-binding protein. J Clin Endocrinol Metab 1988;66:266-72. Brismar K, Gutniak M, Povoa G, Werner S, Hall K. Insulin regulates the 35 kDa IGF binding protein in patients with diabetes mellitus. J Endocrinol Invest 1988;11:599-602. Busby WH, Snyder DK, Clemmons DR. Radioimmunoassay of a 26,000-dalton plasma insulin-like growth factor binding protein: control by nutritional variables. J Clin Endocrinol Metab 1988;67:1225-30. Suikkari AM, Sane T, Seppala M, Jarvinen HY, Karonen SL, Koivist VA. Prolonged exercise increases serum insulinlike growth factor concentrations. J Clin Endocrinol Metab 1989;68:141-4. Weaver JU, Holly JMP, Kopelman PG, Noonan K, Giadom CG, White N, et al. Decreased sex hormone binding globulin (SHBG) and insulin-like growth factor binding protein (IGFBP-1) in extreme obesity. Clin Endocrinol (Oxf) 1990;33:415-22. Degerblad M, Povoa G, Thoren M, Wivall IL, Hall K. Lack of diurnal rhythm of low molecular weight insulin-like growth factor binding protein in patients with Cushing's disease. Acta Endocrinol (Copenh) 1989;120:195-200. Holly JMP, Biddlecombe RA, Dunger DB, Edge JA, Amiel SA, Howell R, et al. Circadian variation of GH -independent IGF -binding protein in diabetes mellitus and its relationship to insulin. A new role for insulin? Clin Endocrinol (Oxf) 1988;29:667-75. Holly JMP, Smith CP, Dunger DB, Edge JA, Biddlecombe RA, Williams AJK, et al. Levels of the small insulin-like growth factor-binding protein are strongly related to those of insulin in prepubertal and pubertal children but only weakly so after puberty. J EndocrinoI1989;121:383-7. Scheiwiller E, Guier HP, MerryweatherJ, Scandella C, Maerki W, Zapf J, et al. Growth restoration of insulin-deficient diabetic rats by recombinant human insulin-like growth factor 1. Nature 1986;323:169-71. Savion N, Liu GM, Laherty R, Gospodarowicz D. Factors controlling proliferation and progesterone production by bovine granulosa cells in serum-free medium. Endocrinology 1981;109:409-14. Baranao JLS, Hammond JM. Comparative effects of insulin and insulin-like growth factors on DNA synthesis and differentiation of porcine granulosa cell. Biochem Biophys Res Commun 1984;124:484-90. Nobels and Dewailly

Puberty and PCDS

51. Adashi EY, Resnick CE, Svoboda ME, Van Wyk JJ. Somatomedin-C synergizes with follicle-stimulating hormone in the acquisition of progestin biosynthetic capacity by cultured rat granulosa cells. Endocrinology 1985;116:2135-42. 52. Davoren JB, Hsueh AJW. Insulin enhances FSH-stimulated steroidogenesis by cultured rat granulosa cells. Mol Cell EndocrinoI1984;35:97-105. 53. Adashi EY, Resnick CE, Brodie AMH, Svoboda ME, Van Wyk JJ. Somatomedin-C-mediated potentiation of folliclestimulating hormone-induced aromatase activity of cultured rat granulosa cells. Endocrinology 1985;117:2313-20. 54. Erickson GF, Magoffin DA, Dyer CA, Hofeditz C. The ovarian androgen producing cells: a review of structure/function relationships. Endocr Rev 1985;6:371-99. 55. Cara JF, Rosenfield RL. Insulin-like growth factor I and insulin potentiate luteinizing hormone-induced androgen synthesis by rat ovarian thecal-interstitial cells. Endocrinology 1988;123:733-9. 56. Hernandez ER, Resnick CE, Svoboda ME, Van Wyk JJ, Payne DW, Adashi EY. Somatomedin-C/insulin-like growth factor I as an enhancer of androgen biosynthesis by cultured rat ovarian cells. Endocrinology 1988;122:1603-12. 57. Hernandez ER, Resnick CE, Holtzclaw WD, Payne DW, Adashi EY. Insulin as a regulator of androgen biosynthesis by cultured rat ovarian cells: cellular mechanism(s) underlying physiological and pharmacological hormonal actions. Endocrinology 1988;122:2034-43. 58. Veldhuis JD, Rodgers RJ, Dee A, Simpson ER. The insulinlike growth factor, somatomedin-C, induces the synthesis of cholesterol side-chain cleavage cytochrome P-450 and adrenodoxin in ovarian cells. J Bioi Chern 1986;261:2499502. 59. Steinkampf M, Mendelson C, Simpson E. Effects of epidermal growth factor and insulin-like growth factor I on the levels of mRNA encoding aromatase cytochrome p-450 of human ovarian granulosa cells. Mol Cell Endocrinol 1988;59:93-9. 60. Adashi EY, Resnick CE, Svoboda ME, Van Wyk JJ. Somatomedin-C enhances induction of luteinizing hormone receptors by follicle-stimulating hormone in cultured rat granulosa cells. Endocrinology 1985;116:2369-75. 61. Barbieri RL, Makris A, Ryan KJ. Insulin stimulates androgen accumulation in incubations of human ovarian stroma and theca. Obstet GynecoI1984;64(Suppl):73S-80S. 62. Garzo VG, Dorrington JH. Aromatase activity in human granulosa cells during follicular development and the modulation by follicle-stimulating hormone and insulin. Am J Obstet GynecoI1984;148:657-62. 63. Erickson GF, Garzo VG, Magoffin DA. Insulin-like growth factor- I regulates aromatase activity in human granulosa and granulosa-luteal cells. J Clin Endocrinol Metab 1989;69: 716-24. 64. Chernausek SD, Jacobs S, Van Wyk JJ. Structural similarities between human receptors for somatomedin C and insulin: analysis by affinity labeling. Biochemistry 1981;20: 7345-51. 65. Fradkin JE, Eastman RC, Lesniak MA, Roth J. Specificity spillover at the hormone receptor: exploring its role in human disease. N Engl J Med 1989;320:640-5. 66. Moxham CP, Duronio V, Jacobs S. Insulin-like growth factor I receptor beta-subunit heterogeneity. Evidence for hybrid tetramers composed of insulin-like growth factor I and in-

Fertility and Sterility

67.

68.

69.

70.

71.

72.

73.

74.

75.

76.

77.

78.

79.

80.

81.

sulin receptor heterodimers. J Bioi Chern 1989;264:1323844. Soos MA, Siddle K. Immunological relationships between receptors for insulin and insulin-like growth factor 1. Evidence for structural heterogeneity of insulin-like growth factor I receptor involving hybrids with insulin receptors. Biochem J 1989;263:553-63. Adashi EY, Resnick CE, Hernandez ER, Hurwitz A, Rosenfield RG. Follicle-stimulating hormone inhibits the constitutive release of insulin-like growth factor binding proteins by cultured rat ovarian granulosa cells. Endocrinology 1990;126:1305-7. Suikkari A-M, Jalkanen J, Koistinen R, Biitzow R, Ritvos 0, Ranta T, et al. Human granulosa cells synthesize low molecular weight insulin-like growth factor-binding protein. Endocrinology 1989;124:1088-90. Koistinen R, Suikkari A-M, Tiitinen A, Kontula K, Seppruii M. Human granulosa cells contain insulin-like growth factor-binding protein (IGFBP-1) mRNA. Clin Endocrinol (Oxf) 1990;32:635-40. Hammond JM, Baranao JLS, Skaleris D, Knight AB, Romanus JA, Rechler MM. Production of insulin-like growth factors by ovarian granulosa cells. Endocrinology 1985;117: 2553-5. Davoren JB, Hsueh AJW. Growth hormone increases ovarian levels of immunoreactive somatomedin-C/insulin-like growth factor-I in vivo. Endocrinology 1986;118:888-90. Hernandez ER, Roberts CT, LeRoith D, Adashi EY. Rat ovarian insulin-like growth factor-I (IGF-I) gene expression is granulosa cell-selective: 5' untranslated mRNA variant representation and hormonal regulation. Endocrinology 1989;125:572-4. Mason HD, Martikainen H, Beard RW, Anyaoku V, Franks S. Direct gonadotrophic effect of growth hormone on oestradiol production by human granulosa cells in vitro. J EndocrinoI1990;126:Rl-4. Hurwitz A, Hernandez ER, Resnick CE, Packman IN, Payne DW, Adashi EY. Basic fibroblast growth factor inhibits gonadotropin-supported ovarian androgen biosynthesis: mechanism(s) and site(s) of action. Endocrinology 1990;126: 3089-95. Hsueh AJW, Welsh TH, Jones PBC. Inhibition of ovarian progestin production by EGF in cultured rat granulosa cells. J Bioi Chern 1982;257:11268-73. Skinner MK, Coffey RJ. Regulation of ovarian cell growth through the local production of transforming growth factora by theca cells. Endocrinology 1988;123:2632-8. Holly JMP, Smith CP, Dunger DB, Howell RJS, Chard T, Perry LA, et al. Relationship between the pubertal fall in sex hormone binding globulin and insulin-like growth factor binding protein -1. A synchronized approach to pubertal development? Clin Endocrinol (Oxf) 1989;31:277-84. Apter D, Bolton NJ, Hammond GL, Vihko R. Serum sex hormone-binding globulin during puberty in girls and in different types of adolescent menstrual cycles. Acta Endocrinol (Copenh) 1984;107:413-9. Lee IR, Lawder LE, Townend DC, Wetherall JD, Hiihnel R. Plasma sex hormone binding globulin concentration and binding capacity in children before and during puberty. Acta Endocrinol (Copenh) 1985;109:276-80. Belgorosky A, Rivarola MA. Progressive increase in nonsex hormone-binding globulin bound-testerone from infancy to

Vol. 58, No.4, October 1992

82.

83. 84.

85.

86.

87.

88.

89.

90. 91.

92.

93.

94.

95.

96.

97.

98.

late prepuberty in boys. J Clin Endocrinol Metab 1987;64: 482-5. Belgorosky A, Rivarola MA. Progressive increase in nonsex hormone-binding globulin bound-testerone and estradiol from infancy to late prepuberty in girls.' J Clin Endocrinol Metab 1988;67:234-7. Anderson DC. Sex hormone-binding globulin. Clin Endocrinol (Oxf) 1974;3:69-96. Cunningham SK, Loughlin T, Culliton M, McKenna TJ. Plasma sex hormone-binding globulin levels decrease during the second decade of life irrespective of pubertal status. J Clin Endocrinol Metab 1984;58:915-8. Pugeat M, Garrel D, Estour B, Lejeune H, Kurzer MS, Tourniaire J, et al. Sex steroid-binding protein in nonendocrine diseases. Ann NY Acad Sci 1988;538:235-47. Haffner SM, Katz MS, Stern MP, Dunn JF. The relationship of sex hormones to hyperinsulinemia and hyperglycemia. Metabolism 1988;37:683-8. Peiris AN, Sothmann MS, Aiman EJ, Kissebah AH. The relationship of insulin to sex hormone-binding globulin: role of adiposity. Fertil Steril1989;52:69-72. Plymate SR, Matej LA, Jones RE, Friedl KE. Inhibition of sex hormone-binding globulin production in the human hepatoma (HepG2) cell line by insulin and prolactin. J Clin Endocrinol Metab 1988;67:460-4. Laron Z, Sarel R. Penis and testicular size in patients with growth hormone insufficiency. Acta Endocrinol (Copenh) 1970;63:625-33. Joslin EP, Root HF, White P. The growth, development and prognosis of diabetic children. JAMA 1925;85:420-8. Pugliese MT, Lifshitz F, Grad G. Fear of obesity. A cause of short stature and delayed puberty. N Engl J Med 1983;309: 513-8. Bucher H, Zapf J, Torresani T, Prader A, Froesch ER, Illig R. Insulin-like growth factors I and II, prolactin, and insulin in 19 growth hormone-deficient children with excessive, normal, or decreased longitudinal growth after operation for craniopharyngioma. N Engl J Med 1983;309:1142-6. Poretsky L. On the paradox of insulin-induced hyperandrogenism in insulin-resistant states. Endocr Rev 1991;12: 3-13. Kahn CR, Flier JS, Bar RS, Archer JA, Gorden P, Martin MM, et al. The syndromes of insulin resistance and acanthosis nigricans. Insulin-receptor disorders in man. N Engl J Med 1976;294:739-45. DeClue TJ, Shah SC, Marchese M, Malone J1. Insulin resistance and hyperinsulinemia induce hyperandrogenism in a young type B insulin-resistant female. J Clin Endocrinol Metab 1991;72:1308-11. Bergman RN. Lilly lecture 1989: toward physiological understanding of glucose tolerance: minimal-model approach. Diabetes 1989;38:1512-27. Nestler JE, Powers LP, Matt DW, Steingold KA, Plymate SR, Rittmaster RS, et al. A direct effect of hyperinsulinemia on serum sex hormone-binding globulin levels in obese women with the polycystic ovary syndrome. J Clin Endocrinol Metab 1991;72:83-9. Sharp PS, Kiddy DS, Reed MJ, Ahyaoku V, Johnston DG, Franks S. Correlation of plasma insulin and insulin-like growth factor- I with indices of androgen transport and metabolism in women with polycystic ovary syndrome. Clin Endocrinol (Oxf) 1991;35:253-7.

Nobels and Dewailly

Puberty and peas

665

99. Barbieri RL, Makris A, Randall RW, Daniels G, Kistner RW, Ryan KJ. Insulin stimulates androgen accumulation in incubations of ovarian stroma obtained from women with hyperandrogenism. J Clin Endocrinol Metab 1986;62:90410. 100. Plymate SR, Fariss BL, Bassett ML, Matej L. Obesity and its role in polycystic ovary syndrome. J Clin Endocrinol Metab 1981;52:1246-8. 101. Kiddy DS, Sharp PS, White DM, Scanlon MF, Mason HD, Bray CS, et al. Differences in clinical and endocrine features between obese and non -obese subjects with polycystic ovary syndrome: an analysis of 263 consecutive cases. Clin Endocrinol (Oxf) 1990;32:213-30. 102. Pekonen F, Laatikainen T, Buyalos R, Rutanen E. Decreased 34K insulin-like growth factor binding protein in polycystic ovarian disease. Fertil Steril 1989;51:972-5. 103. Yen SSC. The polycystic ovary syndrome. Clin Endocrinol (Oxf) 1980;12:177-208. 104. Venturoli S, Porcu E, Fabbri R, Magrini 0, Paradisi R, Pallotti G, et al. Postmenarchal evolution of endocrine pattern and ovarian aspects in adolescents with menstrual irregularities. Fertil Steril 1987;48:78-85. 105. Apter D, Vihko R. Endocrine determinants of fertility: serum androgen concentrations during follow-up of adolescents into the third decade of life. J Clin Endocrinol Metab 1990;71:970-4. 106. Orsini S, Salardi S, Pilu G, Bovicelli L, Cacciari E. Pelvic organs in premenarcheal girls: real-time ultrasonography. Radiology 1984;153:113-6. 107. Stanhope R, Adams J, Pringle JP, Jacobs HS, Brook CGD. The evolution of polycystic ovaries in a girl with hypogonadotrophic hypogonadism before puberty and during puberty induced with pulsatile gonadotrophin-releasing hormone. Fertil SteriI1987;47:872-5. 108. Rao JK, Chihal HJ, Johnson CM. Primary polycystic ovary syndrome in a premenarcheal girl: a case report. J Reprod Med 1985;30:361-5. 109. Root AW, Moshang T. Evolution of the hyperandrogenismpolycystic ovary syndrome from isosexual precocious puberty: report of two cases. Am J Obstet Gynecol 1984;149: 763-7. 110. ZumoffB, Freeman R, Coupey S, Saenger P, Markowitz M, Kream J. A chronobiologic abnormality in luteinizing hormone secretion in teenage girls with the polycystic ovary syndrome. N Engl J Med 1983;309:1206-9. 111. Dickerman Z, Grant DR, Faiman C, Winter JSD. Intraadrenal steroid concentrations in man: zonal differences and developmental changes. J Clin Endocrinol Metab 1984;59: 1031-6. 112. Hoffman DI, Klove K, Lobo RA. The prevalence and significance of elevated dehydroepiandrosterone sulfate levels in anovulatory women. Fertil Steril 1984;42:76-81. 113. Chang RJ, Laufer LR, Meldrum DR, Defazio J, Lu JKH, Vale WW, et al. Steroid secretion in polycystic ovarian disease after ovarian suppression by a long-acting gonadotropin-releasing hormone agonist. J Clin Endocrinol Metab 1983;56:897-903. 114. Lachelin GCL, Barnett M, Hopper BR, Brink G, Yen SCC. Adrenal function in normal women and women with the

666

Nobels and Dewailly Puberty and PCDS

polycystic ovary syndrome. J Clin Endocrinol Metab 1979;49:892-8. 115. Lucky AW, Rosenfield RL, Mc Guire J, Rudy S, Helke J. Adrenal androgen hyperresponsiveness to ACTH in women with acne and/or hirsutism: adrenal enzyme defects and exaggerated adrenarche. J Clin Endocrinol Metab 1986;62: 840-8. 116. Benjamin F, Deutsch S, Saperstein H, Seltzer V. Prevalence of and markers for the attenuated form of congenital adrenal hyperplasia and hyperprolactinemia masquerading as polycystic ovarian disease. Fertil Steril 1986;46:215-21. 117. Dewailly D, Vantyghem M-C, Lemaire C, Dufosse F, Racadot A, Fossati P. Screening heterozygotes for 21-hydroxylase deficiency among hirsute women: lack of utility of the adrenocorticotropin hormone test. Fertil Steril 1988;50:22832. 118. Siegel SF, Finegold DN, Lanes R, Lee PA. ACTH stimulation tests and plasma dehydroepiandrosterone sulfate levels in women with hirsutism. N Engl J Med 1990;323:84954. 119. Granoff AB, Chasalow FI, Blethen SL. 17-Hydroxyprogesterone responses to adrenocorticotropin in children with premature adrenarche. J Clin Endocrinol Metab 1985;60: 409-15. 120. Knorr D, Bidiingmaier F, Holler W, Kuhnle U, Meiler B, Nachmann A. Is heterozygosity for the steroid 21-hydroxylase deficiency responsible for hirsutism, premature pubarche, early puberty, and precocious puberty in children? Acta Endocrinol (Copenh) 1986;113(279 Suppl):284-9. 121. Dewailly D. ACTH stimulation tests in women with hirsutism. N Engl J Med 1991;324:564. 122. Rosenfield RL, Barnes RB, Cara JF, Lucky AW. Dysregulation of cytochrome P450c17a as the cause of polycystic ovarian syndrome. Fertil Steril 1990;53:785-91. 123. Penhoat A, Jaillard C, Saez JM. Synergistic effects of corticotropin and insulin -like growth factor I on corticotropin receptors and corticotropin responsiveness in cultured bovine adrenocortical cells. Biochem Biophys Res Commun 1989;165:355-9. 124. Mechanick JI, Futterweit W. The aberrant puberty hypothesis of polycystic ovarian disease: a review. Mt Sinai J Med 1986;53:310-4. 125. McKenna TJ. Pathogenesis and treatment of polycystic ovary syndrome. N Engl J Med 1988;318:558-62. 126. Givens JR. Familial polycystic ovarian disease. Endocrinol Metab Clin North Am 1988;17:1-17. 127. Hague WH, Adams J, Reeders ST, Peto TEA, Jacobs HS. Familial polycystic ovaries: a genetic disease? Clin Endocrinol (Oxf) 1988;29:593-605. 128. Reaven GM. Banting lecture 1988: role of insulin resistance in human disease. Diabetes 1988;37:1595-607. 129. De Fronzo RA, Ferrannini E. Insulin resistance. A multifaceted syndrome responsible for NIDDM, obesity, hypertension, dyslipidemia and atherosclerotic cardiovascular disease. Diabetes Care 1990;14:558-77. 130. McCluskey S, Evans C, Lacey JH, Pearce JM, Jacobs H. Polycystic ovary syndrome and bulimia. Fertil SteriI1991;55: 287-91.

Fertility and Sterility