Polycystic ovarian disease: heritability and heterogeneity

Human Reproduction Update, Vol.7, No.1 pp. 3±7, 2001 Polycystic ovarian disease: heritability and heterogeneity P.G.Crosignani 1 and A.E.Nicolosi Department of Obstetrics and Gynecology, University of Milano, Milano, Italy 1 To whom correspondence should be addressed. E-mail: piergiorgio.crosignani@unimi.it The polycystic appearance of the ovary is the distinguishing characteristic of polycystic ovary syndrome (PCOS) but there is a wide range of other clinical and biochemical features, e.g. elevated serum concentrations of androgens, insulin, LH and decreased insulin sensitivity. The high prevalence of affected individuals and the wide range of related phenotypes can be explained by the interaction of a small number of key genes with environmental factors. Heritability of PCOS has been inferred from studies of the syndrome in various populations (ethnic groups, twins and PCOS families). The data suggest that the condition is passed down through either sex, according to an autosomal dominant model of genetic transmission. To date, speci c gene mutations affecting androgen synthesis, insulin secretion and insulin activity explain most of the endocrine and metabolic symptoms, while environmental risk factors (either during prenatal or post-natal life), seem to convert an occult PCOS into a clinically manifest syndrome. Key words: anovulation/hirsutism/insulin resistance/obesity/polycystic ovarian disease TABLE OF CONTENTS Polycystic ovarian disease PCOS phenotypes in various populations Mechanism of heritability PCOS: environmental risk factors References Polycystic ovarian disease Polycystic ovaries are often seen in healthy women but are more frequent in women with abnormal cycles and hyperandrogenism (Figure 1). The polycystic appearance of the ovary is the distinguishing sign of polycystic ovary syndrome (PCOS) but there is a wide range of clinical and biochemical features, e.g. elevated serum concentrations of androgens, insulin, LH and decreased insulin sensitivity. These conditions are frequently associated with obesity. Since insulin resistance in PCOS patients is predominantly extra-splanchnic (Dunaif et al., 1992), the fasting blood sugar is normal. According to another group (Franks et al., 1997), ovarian morphology is the essential marker of the syndrome and the wide range of related phenotypes can be explained by the interaction of a small number of key genes with environmental factors. Since these symptoms are found in up to 10% of young women, PCOS is certainly the most frequent endocrine disorder diagnosed in these subjects. Despite the high prevalence of isolated polycystic ovarian morphology (22%), the syndrome may be accompanied by minimal clinical manifestations and, in particular, no uniformly deleterious effect on fertility has been reported (Clayton et al., 1992). A controlled comparative study of patients undergoing an IVF programme found no signi cant difference in pregnancy and live birth rates between women with and without polycystic ovaries (MacDougall et al., 1993). Nevertheless, in a large group of PCOS patients, a high prevalence of primary (46%) and secondary (26%) infertility was found (Balen et al., 1995), while another group (Regan et al., 1990) found an elevated rate of miscarriages in patients with raised LH concentrations in which PCOS was inferred. PCOS phenotypes in various populations In view of the high prevalence of affected individuals, a genetic cause of the syndrome was suggested 30 years ago (Cooper et al., 1968). This has been investigated in several studies on PCOS phenotypes in different populations. Ethnic groups PCOS women (n = 75) from three different ethnic groups were studied (Carmina et al., 1992) and it was concluded that although obesity and hirsutism varied with genetic and environmental factors, the prevalence of adrenal androgen excess and insulin resistance appeared fairly uniform. More recent studies, however, found that ethnicity and PCOS were associated with independent and additive defects of insulin action in Caribbean±Hispanic women (Dunaif et al., 1992). There were also differences in Ó European Society of Human Reproduction and Embryology 3

P.G.Crosignani and A.E.Nicolosi insulin and glucose responses to glucose provocation tests between white and Indian women with the syndrome (Norman et al., 1995). Other authors (Legro et al., 1999) suggest that PCOS is a more important risk factor than ethnicity or race for glucose intolerance in young women. Twins PCOS has been occasionally reported in identical twins (McDonough et al., 1972; Hutton and Clark, 1984). An Australian study found a 50% incidence of PCOS in 34 female twin pairs studied. The high degree of discordance in sonographic ovarian imaging between twins suggests a complex inheritance pathway and/or an important role of environmental factors in the genetic transmission mechanism (Jahnfar et al., 1995). These authors suggest that the high prevalence of PCOS among twins may be explained by factors acting in prenatal life. Families Ovarian morphology, menstrual irregularities and signs of hyperandrogenism were the main symptoms investigated by studies on the familial clustering of PCOS cases, and premature balding was the male phenotype frequently found in the male relatives. All the studies showed a substantially Figure 1. Prevalence of the polycystic appearance of the ovary in different groups of women. Data taken from (1) Adams et al. (1986); (2) Clayton et al. (1992); and (3) O'Driscoll et al. (1994). increased risk of PCOS in rst-degree female relatives of PCOS patients (Table I). Two additional studies on PCOS pedigrees (Franks et al., 1997; Govind et al., 1999), suggest the condition is passed down through either sex according to an autosomal dominant model of genetic transmission (Tables II and III). Mechanism of heritability Ovarian steroidogenesis Although the secretion of androgens by the adrenal glands may be increased, the main source of androgen excess in PCOS is the ovary (Franks et al., 1989). Oestrogen production may be linked either to abnormal stimulation of the ovary or to an intrinsic defect of ovarian steroidogenesis or increased extraovarian conversion of androgen in fat tissue. Abnormal ovarian stimulation Sensitive immunoassay systems have revealed the existence of microheterogeneity of human LH in a large population (Pettersson and Soderholm, 1991; Pettersson et al., 1992). Analysis of the structure of the LHb gene in women with immunologically anomalous LH revealed two nucleotide substitutions in codons 8. This common genetic variant induces higher LH bioactivity than the wild form (Haavisto et al., 1995). While abnormal LH secretion may cause anovulation and luteal insuf ciency, leading to PCOS, the frequency of LH mutations in women with PCOS is not different from that in healthy women, so the presence of the variant does not explain the abnormal steroidogenesis in polycystic ovaries (Elter et al., 1999). Follistatin binds to activin and affects its stimulatory activity on FSH secretion. A follistatin gene mutation in PCOS patients may play a role in the functional impairment of the FSH±granulosa cell axis. While evidence for such a link between the follistatin gene and PCOS has been found in a large study (Urbanek et al., 1999), the association of PCOS with a polymorphism of the gene encoding follistatin was not con rmed by another recent study (Liao et al., 2000) Intrinsic ovarian defects It is well known that there is a primary abnormality in the theca cells of PCOS patients leading to excessive production of progesterone and androgen (Gilling-Smith et al., 1994, 1997). Therefore the abnormal steroidogenesis observed in Table I. Familial clustering of polycystic ovarian syndrome (PCOS) Publication Population studied (n) Inherited symptoms Ferriman and Purdie (1979) Hirsute women (284) Compared with controls (179): higher prevalence of hirsutism Oligomenorrhoeic with enlarged ovary (45) oligomenorrhoea and infertility in rst-degree female relatives Lunde et al. (1989) Women with polycystic ovaries Polycystic ovary phenotype in 31.4% of rst-degree relatives after wedge resection (132) (3.2% in the controls) Hague et al. (1988) Women with PCOS and their relatives (50) 51% risk of PCOS in rst-degree female relatives. Association with congenital adrenal hyperplasia. Carey et al. (1993) PCOS: family pedigrees (10) 67% of mothers and 87% of sisters affected. 4

PCOS: heritability and heterogeneity PCOS is related to an intrinsic abnormality of the theca cells rather than to abnormal gonadotrophin stimulation (IbanÄez et al., 1996). This nding prompted a study of the cholesterol side chain cleavage gene (CYP11a) as a possible cause of the deranged steroidogenesis. The segregation of CYP11a in 20 PCOS families was studied (Gharani et al., 1997). The most common polymorphism of the gene (indicated as 216±) was signi cantly associated with PCOS families. A non-parametric linkage (NPL) analysis using polymorphic markers in that region similarly suggested that the steroid synthesis gene CYP11a is a very important locus for the genetic susceptibility of PCOS hyperandrogenism (NPL score 3.03, P = 0.003) (Gharani et al., 1997). Adrenal and ovarian hyperandrogenism The increased ovarian and adrenal steroidogenic activity in PCOS can also be caused by enhanced lyase activity, exclusively by the cytochrome P450 C17a. Serine phosphorylation of this enzyme system selectively increases its enzymatic activity, leading to hypersecretion of ovarian and adrenal androgen, with no rise in adrenocorticotrophic hormone (ACTH) or other steroidogenic activity (Zhang et al., 1995). Insulin resistance Insulin resistance is another common feature in women with PCOS. The cause is still unknown. Interestingly only women with an endocrine syndrome of hyperandrogenism and chronic anovulation appear to be insulin resistant and at high risk of glucose intolerance (Dunaif et al., 1987; Robinson et al., 1993). Table II. Clinical ndings in sisters with and without polycystic ovaries (PCO) from proband families (Govind et al. 1999) PCO no PCO (n = 35) (n = 18) Menstrual cycle Regular 16 (46%) 15 (84%)* Oligomenorrhoea 19 (54%) 3 (16%) Ovarian volume (mm 3 ) Right 6.7 (3.2±24) 5.2 (2±4±7.7)** Left 6.9 (2.9±23) 4.5 (1.2±7.1)** *PCO versus no PCO (P < 0.02). **PCO versus no PCO (P < 0.001). There appears to be a genetic target cell defect as a cause of the metabolic condition (Holte, 1996). The same hyperphosphorylation process described for cytochrome P450 C17 lyase activity, leading to adrenal and ovarian hyperandrogenism, has been implicated as the cause of a speci c post-receptor defect of transduction of the insulin signal in broblasts (Dunaif et al., 1995). In these patients, autophosphorylation of the serine (rather than tyrosine) residue impairs insulin signal transduction and contributes to the 50% insulin resistance observed. Thus a single molecular defect leading to the activation of a serine kinase might explain the two main biochemical disturbances in these patients: hirsutism and insulin resistance. Abnormal insulin secretion Hyperinsulinaemia has been reported in patients with PCOS and the syndrome is one of the major risk factors for non-insulindependent diabetes mellitus (NIDDM) ((Holte et al., 1995). The b cell dysfunction is not obesity-dependent and in the majority of PCOS women is not associated with glucose intolerance (Dunaif and Finegood, 1996). The direct role of the insulin gene in the aetiology of hyperinsulinaemia was investigated in three groups of PCOS patients (one of which included 17 families with several affected individuals). All three populations showed an association between class III alleles at the variable number of tandem repeats (VNTR) 5 to the insulin gene and PCOS (Bennett et al., 1995). The association was stronger in anovulatory patients, who more frequently have hyperinsulinaemia. A non-parametric linkage analysis in the PCOS families showed excess allele sharing at the same locus (NPL score 3.250, P = 0.002) (Waterworth et al., 1997). The authors concluded that the VNTR 5 region to the insulin gene is a major locus for PCOS-associated hyperinsulinaemia. Table IV summarizes the pathophysiology for inherited susceptibility. Table III. Premature male-pattern baldness in polycystic ovarian syndrome (PCOS) and control families (Govind et al., 1999) Families PCOS Control Evaluable men 50 20 Men with premature male-pattern baldness 10 (22%) 1 (5%)* *P = 0.006. Table IV. Polycystic ovarian syndrome (PCOS): pathophysiology for the inheritable susceptibilities Gene Molecular lesion Target Phenotype CYP 11a locus 216 genotype Ovarian theca cells Hyperandrogenism (gene coding cholesterol side chain cleavage) Autosomal dominant gene Point mutation encoding Adrenal and ovarian P450 C17aLYASE Hyperandrogenism serine hyperphosphorylation Activated insulin receptor Insulin resistance 11 p 15.5 locus (insulin gene) Class III alleles at VNTR Pancreatic b cells Hyperinsulinaemia VNTR = variable number of tandem repeats. 5

P.G.Crosignani and A.E.Nicolosi PCOS: environmental risk factors Prenatal life Following a retrospective study on PCOS patients (Cresswell et al., 1997), the existence of speci c prenatal risk factors for the post-pubertal expression of the PCOS phenotype was suggested. These authors found two distinct groups of patients with polycystic ovaries: (i) those who had above-average birthweight and (ii) those born to overweight mothers. The second group comprised women of normal weight who had high plasma LH, but normal testosterone, concentrations. These women were born after term (40 weeks gestation). On the basis of these ndings, the authors suggest that the two forms of PCOS have different origins in intrauterine life. Obese, hirsute women with polycystic ovaries have higher than normal ovarian secretion of androgens, associated with high birthweight and maternal obesity. Thin women with polycystic ovaries have altered hypothalamic control of LH release resulting from prolonged gestation. Post-natal risk factors Chronic anovulation The role of chronic anovulation as an environmental risk factor for PCOS is suggested by several pathophysiological mechanisms where androgen, LH and sex hormone binding globulin (SHBG) play key roles. A subgroup of patients with PCO and hypogonadotrophic anovulation has also been described (Shoham et al., 1992). Obesity Obesity is an independent risk factor for chronic anovulation (Grodstein et al., 1994) and body fat distribution (waist to hip ratio) seems more important than weight itself (Zaadstra et al., 1993). In obese women, the two main mechanisms leading to anovulation are similar to those in patients with PCOS: (i) excess of LH and androgen secretion; and (ii) hyperinsulinaemia and insulin resistance. In fact, short-term fasting reduces LH secretion in normal weight women (Olson et al., 1995). In overweight women, caloric restriction lowers insulin concentrations and raises SHBG concentrations (Kiddy et al., 1992), while in severely obese patients post-gastroplasty recovery of ideal weight restores normal glucose and insulin metabolism (Letiexhe et al., 1995). In contrast, in women with PCOS, obesity worsens the syndrome; in these patients insulin resistance appears to be directly related to the body mass index (BMI) (Pasquali et al., 1986), while weight reduction in obese women with PCOS lowers LH hypersecretion and reverses insulin insensitivity (Kiddy et al., 1990; BuÈtzow et al., 1996). Experimental data suggest there may be some genetic control of appetite, body weight and reproductive function (Friedman, 1997). Conclusions PCOS can be considered a complex, heterogeneous metabolic syndrome triggered or maintained by the combined effect of inheritable genetic susceptibilities and environmental risk factors. The complicated panel of genetic and post-conceptional mechanisms leading to PCOS is still poorly understood. Acknowledgements I am grateful to Howard Jacobs for the constructive and critical revision of the manuscript and Simonetta Vassallo for secretarial assistance. References Adams, J., Polson, D.W. and Franks, S. (1986) Prevalence of polycystic ovaries in women with anovulation and idiopathic hirsutism. Br. Med. J. (Clin. Res. Ed.), 293, 355±359. Balen, A.H., Conway, G.S., Kaltsas, G. et al. (1995) Polycystic ovary syndrome: the spectrum of the disorder in 1741 patients. Hum. Reprod., 10, 2107±2111. Bennett, S.T., Lucassen, A.M., Gough, S.C.L. et al. (1995) Susceptibility to human type 1 diabetes at IDDM2 is determined by tandem repeat variation at the insulin gene minisatellite locus. Nature Genet., 9, 284±292. BuÈtzow, T.L., Lehtovirta, M.T., VaÈinaÈmoÈ, U. et al. (1996) The effect of weight reduction on gonadotrophin, insulin and androgen metabolism in hyperandrogenic overweight infertile women. [Abstr. No. 100] Hum. Reprod., 11 (Abstr. Book), 47±48. Carey, A.H., Chan, K.L., Short, F. et al. (1993) Evidence for a single gene effect causing polycystic ovaries and male pattern baldness. Clin. Endocrinol., 38, 653±658. Carmina, E., Koyama, T., Chang, L. et al. (1992) Does ethnicity in uence the prevalence of adrenal hyperandrogenism and insulin resistance in polycystic ovary syndrome? Am. J. Obstet. Gynecol., 167, 1807±1812. Clayton, R.N., Ogden, V., Hodgkinson, J. et al. (1992) How common are polycystic ovaries in normal women and what is their signi cance for the fertility in the general population? Clin. Endocrinol., 37, 127±134. Cooper, H., Spellacy, W., Prem, K. and Cohen, W. (1968) Hereditary factors in the Stein±Leventhal syndrome. Am. J. Obstet. Gynecol., 100, 371±387. Cresswell, J.L., Barker, D.J.P., Osmond, C. et al. (1997) Fetal growth, length of gestation, and polycystic ovaries in adult life. Lancet, 350, 1131±1135. Dunaif, A., Graf, M., Mandeli, J. et al. (1987) Characterization of groups of hyperandrogenic women with acanthosis nigricans, impaired glucose tolerance, and/or hyperinsulinemia. J. Clin. Endocrinol. Metab., 65, 499±507. Dunaif, A., Segal, K.R., Shelley, D.R. et al. (1992) Evidence for distinctive and intrinsic defects in insulin action in polycystic ovary syndrome. Diabetes, 41, 1257±1266. Dunaif, A., Xia, J., Book, C.B. et al. (1995) Excessive insulin receptor serine phosphorylation in cultured broblasts and in skeletal muscle. A potential mechanism for insulin resistance in the polycystic ovary syndrome. J. Clin. Invest., 96, 801±810. Dunaif, A. and Finegood, D.T. (1996) b cell dysfunction independent of obesity in the polycystic ovary syndrome. J. Clin. Endocrinol. Metab., 81, 942±947. Elter, K., Erel, C.T., Cine, N. et al. (1999) Role of the mutations Trp8 Arg and Ile15 Thr of the human luteinizing hormone beta-subunit in women with polycystic ovary syndrome. Fertil. Steril., 71, 425±430. Ferriman, D. and Purdie, A.W. (1979) The inheritance of polycystic ovarian disease and a possible relationship to premature balding. Clin. Endocrinol., 11, 291±300. Franks, S. (1989) Polycystic ovary syndrome: a changing perspective. Clin. Endocrinol., 31, 87±120. Franks, S., Gharani, N., Waterworth, D. et al. (1997) The genetic basis of polycystic ovary syndrome. Hum. Reprod., 12, 2641±2648. Friedman, J.M. (1997) The alphabet of weight control. Nature, 385, 119±20. Gharani, N., Waterworth, D.M., Batty, S. et al. (1997) Association of the steroid synthesis gene CYP11a with polycystic ovary syndrome and hyperandrogenism. Hum. Mol. Genet., 6, 397±402. Gilling-Smith, C., Willis, D.S., Beard, R.W. and Franks, S. (1994) Hypersecretion of androstenedione by isolated theca cells from polycystic ovaries. J. Clin. Endocrinol. Metab., 79, 1158±1165. Gilling-Smith, C., Story, E.H., Rogers, V. and Franks, S. (1997) Evidence for a primary abnormality of thecal cell steroidogenesis in the polycystic ovary syndrome. Clin. Endocrinol., 47, 93±99. Govind, A., Obhrai, M.S. and Clayton, R.N. (1999) Polycystic ovaries are inherited as an autosomal dominant trait: analysis of 29 polycystic ovary syndrome and 10 control families. J. Clin. Endocrinol. Metab., 84, 38±43. Grodstein, F., Goldman, M.B. and Cramer, D.W. (1994) Body mass index and ovulatory infertility. Epidemiology, 5, 247±250. Haavisto, A.M., Pettersson, K., Bergendahl, M. et al. (1995) Occurrence and 6

PCOS: heritability and heterogeneity biological properties of a common genetic variant of luteinizing hormone. J. Clin. Endocrinol. Metab., 80, 1257±1263. Hague, W.M., Adams, J., Reeders, S.T. et al. (1988) Familial polycystic ovaries: a genetic disease? Clin. Endocrinol. (Oxf.), 29, 593±605. Holte, J. (1996) Disturbances in insulin secretion and sensitivity in women with the polycystic ovary syndrome. J. Clin. Endocrinol. Metab., 10, 221±247. Holte, J., Bergh, T., Berne, C. et al. (1995) Restored sensitivity but persistently increased early insulin secretion after weight loss in obese women with polycystic ovary syndrome. J. Clin. Endocrinol. Metab., 80, 2586±2593. Hutton, C. and Clark, F. (1984) Polycystic ovarian syndrome in identical twins. Postgrad. Med. J., 60, 64±65. IbanÄez, L., Hall, J.E., Potau, N. et al. (1996) Ovarian 17-hydroxyprogesterone hyperresponsiveness to gonadotropin-releasing hormone (GnRH) agonist challenge in women with polycystic ovary syndrome is not mediated by luteinizing hormone hypersecretion: evidence from GnRH agonist and human chorionic gonadotropin stimulation testing. J. Clin. Endocrinol. Metab., 81, 4103±4107. Jahnfar, S., Eden, J.A., Warren, P. et al. (1995) A twin study of polycystic ovary syndrome. Fertil. Steril., 63, 478±86. Kiddy, D.S., Sharp, P.S., White, D.M. et al. (1990) 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.), 32, 213±20. Kiddy, D.S., Hamilton-Fairley, D., Bush, A. et al. (1992) Improvement in endocrine and ovarian function during dietary treatment of obese women with polycystic ovary syndrome. Clin. Endocrinol. (Oxf.), 36, 105±111. Legro, R.S., Kunselman, A.R., Dodson, W.X. and Dunaif, A. (1999) Prevalence and predictors of risk for type 2 diabetes mellitus and impaired glucose tolerance in polycystic ovary syndrome: a prospective, controlled study in 254 affected women. J. Clin. Endocrinol. Metab., 84, 165±168. Letiexhe, M.R., Scheen, A.J., Gerard, P.L. et al. (1995) Postgastroplasty recovery of ideal body weight normalizes glucose and insulin metabolism in obese women. J. Clin. Endocrinol. Metab., 80, 364±369. Liao, W.X., Roy, A.C. and Ng, S.C. (2000) Preliminary investigation of follistatin gene mutations in women with polycystic ovary syndrome. Mol. Hum. Reprod., 6, 587±590. Lunde, O., Magnus, P., Sandvik, L. and Hoglo, S. (1989) Familial clustering in the polycystic ovarian syndrome. Gynecol. Obstet. Invest., 28, 23±30. MacDougail, M.J., Tan, S.-L., Balen, A. and Jacobs, H.S. (1993) A controlled study comparing patients with and without polycystic ovaries undergoing in-vitro fertilization. Hum. Reprod., 8, 233±237. McDonough, P.G., Mahesh, V.B. and Ellegood, J.O. (1972) Steroid, follicle stimulating hormone, and luteinizing hormone pro les in identical twins with polycystic ovaries. Am. J. Obstet. Gynecol., 113, 1072±1078. Norman, R.J., Mahabeer, S. and Masters, S. (1995) Ethnic differences in insulin and glucose response to glucose between white and Indian women with polycystic ovary syndrome. Fertil. Steril., 63, 58±62. O'Driscoll, J.B., Mamtora, H., Higginson, J. et al. (1994) A prospective study of the prevalence of clear-cut endocrine disorders and polycystic ovaries in 350 patients presenting with hirsutism or androgenic alopecia. Clin. Endocrinol., 41, 231±236. Olson, B.R., Cartledge, T., Sebring, N. et al. (1995) Short-term fasting affects luteinizing hormone secretory dynamics but not reproductive function in normal-weight sedentary women. J. Clin. Endocrinol. Metab., 80, 1187±1193. Pasquali, R., Fabbri, R., Venturoli, S. et al. (1986) Effect of weight loss and antiandrogenic therapy on sex hormone blood levels and insulin resistance in obese patients with polycystic ovaries. Am. J. Obstet. Gynecol., 154, 139±144. Pettersson, K. and Soderholm, J.R. (1991) Individual differences in lutropin immunoreactivity revealed by monoclonal antibodies. Clin. Chem., 37, 333±340. Pettersson, K., Ding, Y.-Q. and Huhtaniemi, I. (1992) An immunologically anomalous luteinizing hormone variant in a healthy woman. J. Clin. Endocrinol. Metab., 74, 164±171. Regan, L., Owen, E.J. and Jacobs, H.S. (1990) Hypersecretion of luteinising hormone, infertility, and miscarriage. Lancet, 336, 141±144. Robinson, S., Kiddy, D., Gelding, S.V. et al. (1993) The relationship of insulin insensitivity to menstrual pattern in women with hyperandrogenism and polycystic ovaries. Clin. Endocrinol. (Oxf.), 39, 351±355. Shoham, Z., Conway, G.S., Patel, A. and Jacobs, H.S. (1992) Polycystic ovaries in patients with hypogonadotropic hypogonadism: similarity of ovarian response to gonadotropin stimulation in patients with polycystic ovarian syndrome. Fertil. Steril., 58, 37±45. Urbanek, M., Legro, R.S., Driscoll, D.A. et al. (1999) Thirty-seven candidate genes for polycystic ovary syndrome: Strongest evidence for linkage is with follistatin. Proc. Natl Acad. Sci. USA, 15, 8573±8578. Waterworth, D.M., Bennett, S.T., Gharani, N. et al. (1997) Linkage and association of insulin gene VNTR regulatory polymorphism with polycystic ovary syndrome. Lancet, 349, 986±90. Zaadstra, B.M., Seidell, J.C., Van Noord, P.A. et al. (1993) Fat and female fecundity: prospective study of effect of body fat distribution on conception rates. Br. Med. J., 306, 484±487. Zhang, L.H., Rodriguez, H., Ohno, S. and Miller, W.L. (1995) Serine phosphorylation of human P450c17 increases, 17,20-lyase activity: implications for adrenarche and the polycystic ovary syndrome. Proc. Natl Acad. Sci. USA, 92, 10619±10623. Received on May 18, 2000; accepted on September 29, 2000 7