Androgen receptor gene CAG repeat polymorphism in women with polycystic ovary syndrome

Transcription

1 Androgen receptor gene CAG repeat polymorphism in women with polycystic ovary syndrome Jin Ju Kim, M.D., a Seon Ha Choung, B.S., a Young Min Choi, M.D., a,b Sang Ho Yoon, M.D., c Seok Hyun Kim, M.D., a,b and Shin Yong Moon, M.D. a,b a Department of Obstetrics and Gynecology, Seoul National University College of Medicine, Seoul; b Institute of Reproductive Medicine and Population, Medical Research Center, Seoul National University College of Medicine, Seoul; and c Department of Obstetrics and Gynecology, Dongguk University International Hospital, Goyang, South Korea Objective: To evaluate the role of the androgen receptor (AR) gene CAG repeat in women with polycystic ovary syndrome (PCOS). Design: Case control study. Setting: University department of obstetrics and gynecology. Patient(s): Women with (n ¼ 114) or without (n ¼ 205) PCOS. Intervention(s): Peripheral blood sampling was done for DNA analysis and serum hormone measurements. Main Outcome Measure(s): CAG repeat length and serum androgen levels. Result(s): No statistically significant CAG repeat length differences were found between patients and controls. We conducted a detailed analysis after dividing PCOS patients according to their free testosterone levels. The high free testosterone group had a statistically significantly longer mean biallelic average ( vs ), short ( vs ), and long ( vs ) allelic lengths than the normal free testosterone group. In PCOS patients, a statistically significant correlation was found between biallelic average length and free testosterone concentration, either unadjusted or after adjustment. Conclusion(s): The AR gene CAG repeat polymorphism may contribute to the serum concentration of free testosterone in PCOS patients. A subset of PCOS patients with relatively longer CAG repeats (less AR activity) tended to show a higher serum androgen concentration. (Fertil Steril Ò 2008;90: Ó2008 by American Society for Reproductive Medicine.) Key Words: Androgen receptor, CAG repeat, polycystic ovary syndrome Polycystic ovary syndrome (PCOS) is a common endocrine disorder in women of reproductive age, with a prevalence that varies from 4% to 7% (1, 2), and hyperandrogenism is a central feature. Androgens act through the androgen receptor (AR), which belongs to the nuclear receptor superfamily of transcription factors. Like other steroid receptors, the AR protein has three major domains: an N-terminal transactivation domain, a DNA-binding domain, and an androgenbinding domain. The transactivation activity of the receptor resides in its N-terminal domain. The AR gene, located on Xq , is composed of eight exons and is known to show more frequent sequence variation than other steroid receptor genes (3). In particular, its N- terminal transactivation domain contains a polyglutamine stretch that is coded by a CAG triplet of polymorphic length, beginning at codon 58 in exon 1. The inverse correlation between the number of CAG repeats and AR function has been described in many in vivo and in vitro studies (4 6). Received May 9, 2007; revised and accepted October 25, Supported by a grant from Seoul National University Hospital Research Fund (# ), Seoul, South Korea. Reprint requests: Young Min Choi, M.D., Department of Obstetrics & Gynecology, The Institute of Reproductive Medicine and Population, Medical Research Center, Seoul National University College of Medicine, 28 Yungun-dong, Chongno-ku, Seoul , South Korea (FAX: ; ymchoi@snu.ac.kr). The CAG repeat length polymorphism has been related to several diseases associated with low or high androgenic activity in both men and women. Shorter CAG tracts, a more transcriptionally active feature, have been associated with increased prostate cancer risk (7 9), higher cancer grade at diagnosis (10), and earlier onset in Caucasians (11). On the other hand, longer CAG repeat lengths appear to result in reduced AR activity. Abnormal elongation of CAG repeat number (>40 CAG repeats) has been linked to spinal and bulbar muscular atrophy (Kennedy disease), which is also associated with androgen insensitivity, decreased virilization, testicular atrophy, reduced sperm production, and infertility (12). Although within the normal polymorphic range, some studies have shown that relatively long tracts (R28 CAG repeats) are associated with an increased risk of male infertility due to impaired sperm production (5, 13). In men, it also has been reported that serum androgen levels are modulated by AR activity based upon the length of its polyglutamine tract: free testosterone increased with an increase in CAG repeat length (14). Data on the functional importance of the CAG repeat in women is sparse, but conditions including hirsutism (15), androgenic alopecia (16), and breast cancer (17) have been studied. Recent studies have shown an association between CAG repeat length and the subset of anovulatory patients with low serum androgen levels (18). Though the etiology of PCOS has not been elucidated, a number of studies have suggested that genetic factors 2318 Fertility and Sterility â Vol. 90, No. 6, December /08/$34.00 Copyright ª2008 American Society for Reproductive Medicine, Published by Elsevier Inc. doi: /j.fertnstert

2 play important roles in its etiology and pathogenesis (19, 20). Moreover, hyperandrogenism is a core feature of PCOS, and candidate genes that predispose an individual to the development of PCOS include the AR gene. The present study aimed to evaluate whether there is a difference in the AR gene CAG repeat length between PCOS patients and controls. We also assessed whether serum androgen levels are affected by AR activity in women with PCOS. MATERIALS AND METHODS Subjects We recruited 114 PCOS patients at the Department of Obstetrics and Gynecology at Seoul National University Hospital from 2004 to A diagnosis of PCOS was based on the 2003 American Society for Reproductive Medicine/European Society for Human Reproduction and Embryology (ASRM/ESHRE) consensus meeting guideline (21). Hirsutism was defined as a score of 8 or more on the Ferriman Gallwey scale (22), and oligoovulation was defined as a menstrual cycle of more than 35 days. None of the patients had features of Cushing disease or drug-induced hirsutism. The controls were 205 healthy premenopausal women. Among them, 106 women were patients being treated for uterine, tubal, or male factor infertility. The remaining 99 control women visited our outpatient clinic for an annual Papanicolaou smear test as a part of a cancer screening. The 2003 Rotterdam criteria were used to exclude PCOS in the controls before they entered the study. All control women had regular menstrual cycles (21 to 35 days) and Ferriman Gallwey scores below 8. They all had a transvaginal pelvic ultrasound examination performed to evaluate ovarian morphology and were excluded if they had polycystic ovary morphologic features. The patients and controls had not taken hormone medications, including oral contraceptives, for at least 6 months, and all the participants were screened to exclude hyperprolactinemia and thyroid dysfunction. The review board for human research of Seoul National University Hospital approved this project, and written informed consent was obtained from each woman. Hormonal Assays Biochemical data were available only for PCOS patients. Serum luteinizing hormone (LH), follicle-stimulating hormone (FSH), and estradiol (E 2 ) were measured on cycle day 3 of the follicular phase or during the period of amenorrhea in amenorrheic PCOS patients. Serum total testosterone (reference range: ng/ml), 17-hydroxyprogesterone (17-OHP) (reference range: ng/ml), dehydroepiandrosterone sulfate (DHEAS) (reference range: ng/ml), and sex hormone binding globulin (SHBG) (reference range: nmol/l) were measured using commercial kits (Diagnostic Products Corporation, Los Angeles, CA). Free testosterone was also directly measured by the radioimmunoassay (RIA) method using the same commercial kits (Diagnostic Products Corporation). For free testosterone, intra-assay and interassay coefficients of variation were less than 17% (intra-assay) and 8.0% to 18.3% (interassay), according to the manufacturer. All PCOS patients were assessed for glucose tolerance and insulin resistance with measurement of fasting and 2-hour glucose and insulin levels after a 75-g-glucose load. Plasma insulin (reference range: 2 25 mu/ml) was measured using commercial kits (BioSource Europe S.A., Nivelles, Belgium) with an intra-assay variation of 1.6% to 2.2% and inter-assay variation of 6.1% to 6.5%. In addition, the homeostatic model assessment (HOMA) was calculated as follows: HOMA ¼ (mg/dl glucose) (mu/ml insulin)/405. Polymerase Chain Reaction Genomic DNA was isolated and extracted from peripheral blood leukocytes using Wizard DNA extraction kits (Promega, Madison, WI). The total volume of the polymerase chain reaction (PCR) reaction mixture was 25 ml and contained 0.05 mg of genomic DNA, 10 mm Tris-HCl, ph 8.3, 50 mm KCl, 1.5 mm MgCl 2, 200 mm dntps (deoxynucleotide triphosphates), IU of Taq polymerase (Takara, Shiga, Japan), and 0.08 mm of each upstream and downstream primer. The oligonucleotide primers used for PCR were as follows: upstream primer 6FAM-5 0 TGC GCG AAG TGA TCC AGA AC3 0, and downstream primer 5 0 CTT GGG GAG AAC CAT CCT CA3 0. Following an initial denaturation step (5 minutes at 94 C), samples were subjected to 35 cycles of PCR at 95 C for 1 minute, 60 C for 1 minute, and 72 C for 1 minute with a final extension of 10 minutes at 72 C. The PCR product sizes were determined using GeneScan software (Perkin Elmer Applied Biosystems, Foster City, CA). The number of CAG repeats was calculated in relation to a series of standards obtained by automatic sequencing in three homozygous subjects. Statistical Analysis The AR gene CAG repeat length distributions were compared in PCOS patients and controls. Because two X-linked AR alleles are present, we adopted three modes of allele representation: using the mean value of the two alleles (biallelic averages) in each person, and using the short and long alleles of each patient separately. Because a subset (30% to 50%) of PCOS patients have normal serum androgen concentrations (23) and it has been shown that these patients may have an AR gene with a low number of CAG repeats (more active AR) (18), the patients were categorized into high free testosterone and normal free testosterone groups using a free testosterone cut-off value of 2.0 pg/ml (24). The CAG repeat lengths of the AR gene and other continuous variables were compared using the Student s t-test or Mann-Whitney U test. Correlation analysis was performed using Pearson s correlation coefficient. To explore the effect of the AR gene CAG repeat length on free testosterone concentration, a multiple linear regression model was Fertility and Sterility â 2319

3 constructed with free testosterone as a dependent variable and age, body mass index, HOMA, LH/FSH ratio, and biallelic average repeat length as independent variables. All data analyses were done using SPSS software (version 12.0; SPSS Inc., Chicago, IL). In all tests, P<.05 was considered statistically significant. Continuous data are reported as mean standard deviation (SD). RESULTS The AR Gene CAG Repeat Length Polymorphism in PCOS Patients and Controls The mean age of PCOS patients was statistically significantly lower ( vs years, P<.001) and mean body mass index was statistically significantly higher ( vs , P¼.017) than controls. The CAG repeats were successfully determined in all subjects. Allele sizes ranged from 14 to 31 among the PCOS patients, and from 12 to 32 among controls, which reflects the normal polymorphic range (11 38) reported for Chinese women (18). No differences were found in the mean numbers of CAG repeats between PCOS patients and controls, whether analyzed using biallelic averages ( in PCOS vs in controls) or separately for short ( in PCOS vs in controls) and long alleles ( in PCOS vs in controls). The AR Gene CAG Repeat Length Polymorphism in Normal and High Free Testosterone PCOS Groups The clinical and endocrine characteristics of the PCOS patients are depicted in Table 1. They were categorized into a high free testosterone group and a normal free testosterone group using a free testosterone cut-off value of 2.0 pg/ml (24). Examination of allele distributions revealed statistically significant differences in the AR gene CAG repeat length between these two groups (Table 2). The high free testosterone women had statistically significantly greater mean lengths than patients with normal free testosterone, as determined using biallelic averages ( vs , P¼.006) and short ( vs , P¼.046) and long alleles ( vs , P¼.013). In addition, alleles shorter than 16 repeats were recorded only in the normal free testosterone group, and the longest alleles (more than 30 repeats) were found only in the high free testosterone group. No differences were observed between the two groups for body mass index, waist-to-hip ratio, LH, FSH, SHBG, HOMA, postprandial 2-hour glucose, and insulin circulating levels, suggesting the differences observed above were not likely to be due to chance. Pearson s correlation analysis revealed a statistically significant correlation between free testosterone and biallelic average CAG repeat length (r ¼ 0.309, P¼.002). To analyze the impact of the AR gene CAG repeat length on free testosterone, a multiple linear regression model was constructed with free testosterone as a dependent variable (Table 3). A positive correlation was observed between TABLE 1 Clinical and biochemical features of the PCOS patients (n [ 114). Characteristic Value Age (years) Body mass index (kg/m 2 ) Waist-to-hip ratio Hirsutism score Total testosterone (ng/ml) Free testosterone (pg/ml) OHP (ng/ml) SHBG (nmol/l) DHEAS (ng/ml) LH (miu/ml) FSH (miu/ml) LH/FSH ratio E 2 (pg/ml) FBS (mg/dl) Fasting insulin (mu/ml) HOMA Postprandial 2-hour glucose (mg/dl) Postprandial 2-hour insulin (mu/ml) Note: All values are mean SD. 17-OHP, 17-hydroxyprogesterone; DHEAS, dehydroepiandrosterone sulfate; E 2, estradiol; FBS, fasting plasma glucose; FSH, follicle-stimulating hormone; HOMA, homeostatic model assessment; LH, luteinizing hormone; SHBG, sex hormone binding globulin. Kim. AR gene CAG repeats in PCOS. Fertil Steril CAG repeat length and free testosterone concentration, either unadjusted or after adjustment for age, body mass index, LH/FSH ratio, and HOMA (using centered data to avoid high multicollinearity). DISCUSSION Our study determined whether the pattern of (CAG) n polymorphism differed between normal women and women with PCOS and evaluated the relationship between CAG repeat and serum androgen levels in PCOS patients. According to our data, the AR gene exon 1 (CAG) n repeat length distribution in PCOS patients was not different from that of controls, indicating that the AR gene is not a major determinant of PCOS. However, more detailed analysis, performed after dividing PCOS patients into two groups according to free testosterone concentration, revealed statistically significant differences. The high free testosterone group had a statistically significantly longer mean biallelic average ( vs , P¼.006) and short ( vs., , P¼.046) and long ( vs , P¼.013) allelic lengths than the normal free testosterone group. However, the 2320 Kim et al. AR gene CAG repeats in PCOS Vol. 90, No. 6, December 2008

4 TABLE 2 Clinical features, biochemical variables and the AR exon 1 (CAG) n repeat lengths in polycystic ovary syndrome patients with below or above normal free testosterone levels. Normal free T (n [ 64) High free T (n [ 38) P value a Age (years) NS Body mass NS index (kg/m 2 ) Waist-to-hip ratio NS Hirsutism score Total testosterone (ng/ml) <.001 Free testosterone (pg/ml) < OHP (ng/ml) SHBG (nmol/l) NS DHEAS (ng/ml) <.001 LH (miu/ml) NS FSH (miu/ml) NS LH/FSH ratio NS E 2 (pg/ml) NS FBS (mg/dl) NS Fasting insulin (mu/ml) NS HOMA NS Postprandial 2-hour glucose NS (mg/dl) Postprandial 2-hour insulin (mu/ NS ml) CAG repeat CAG repeat (short allele) CAG repeat (long allele) Note: All values are mean SD. 17-OHP, 17-hydroxyprogesterone; DHEAS, dehydroepiandrosterone sulfate; E 2, estradiol; FBS, fasting plasma glucose; FSH, follicle-stimulating hormone; HOMA, homeostatic model assessment; LH, luteinizing hormone; NS, not statistically significant; SHBG, sex hormone binding globulin. a Student s t-test. Kim. AR gene CAG repeats in PCOS. Fertil Steril CAG length differences were only about one repeat for all alleles analyzed. Our results are similar to the findings of Mifsud et al. (18), who found no differences between PCOS patients and controls in terms of CAG repeat length; however, among PCOS patients with normal or elevated testosterone concentrations, they found the former to have a significantly smaller CAG repeat number only when the short allele in each patient was considered. In their results, CAG repeat length difference was also 1.6 units on average among patients with normal or elevated testosterone concentrations ( vs , respectively, P¼.004). It is unclear to what extent one CAG repeat (one glutamine unit) affects AR activity. Hsiao et al. (6) reported the identification of an AR coactivator, ARA 24, that can interact with the AR N-terminal polyglutamine region and enhance AR transactivation. They found that the interaction of the AR N-terminal domain with ARA 24 diminishes as polyglutamine length increases. The coactivator activity of ARA 24 to AR also diminished with polyglutamine expansion. They concluded that CAG repeat expansion may result in poorer interaction and weaker activation of the AR coactivator, and could contribute to weaker AR transactivation. Yeh et al. (25) reported that the AR coactivator had been shown to enhance AR-mediated transcriptional activity up to 10- fold, a level that androgen-ar alone cannot reach. Thus, we would infer that even a modest change in repeat number and a subsequent change in the interaction with coactivator could have a large or cumulative effect on AR transcriptional activity. According to our data, progressive expansion of biallelic average length was associated with a linear increase in serum free testosterone concentration. Because an inverse Fertility and Sterility â 2321

5 TABLE 3 Multiple linear regression models with free testosterone as the dependent variable. Variables Standardized b-coefficient P value Unadjusted model (r 2 ¼ 0.095) CAG repeat length Adjusted model I (r 2 ¼ 0.238) Age Body mass index <.001 CAG repeat length Adjusted model II (r 2 ¼ 0.270) Age Body mass index LH/FSH HOMA CAG repeat length Note: FSH, follicle-stimulating hormone; HOMA, homeostatic model assessment; LH, luteinizing hormone. Kim. AR gene CAG repeats in PCOS. Fertil Steril relationship had been shown between CAG repeat length and androgenicity (4, 5), we would infer that the positive correlation found between serum androgen and CAG repeat number is the result of a compensatory change or a feedback mechanism dependent on AR activity; that is, depressed AR activity triggers a mechanism that results in increased androgen secretion and vice versa. However, if serum levels of androgen are related to AR, it is unclear whether the influence of AR on serum androgen is mediated by the hypothalamus-pituitaryovary axis or by its direct effect on ovarian theca cells. In view of the modest correlation (r ¼ 0.309) between them, other factors are probably involved. The transcriptional activation of AR is affected not only by polymorphisms in the AR gene, but also by a number of other factors, including tissue levels of dihydrotestosterone (26), estradiol (27, 28), insulin-like growth factors (29, 30), and AR coactivators (6, 25). Thus, these too may also affect the results. We used free testosterone to determine the presence of biochemical hyperandrogenism. Because a laboratory-specific reference range was not available, we used 2.0 pg/ml as a cut-off value according to Speroff and Fritz (24). Byun et al. (31) investigated the prevalence of PCOS in college students from Korea, and, using the same commercial kit, they defined hyperandrogenism as a free testosterone level above the 95th percentile of that for normally cycling, nonhirsute women. In their results, the mean serum level of free testosterone was pg/ml in 133 normally cycling, nonhirsute women. Likewise, Lee et al. (32) found that the mean serum level of free testosterone was pg/ml in 84 normally cycling, nonhirsute Korean women. Therefore, we think that 2.0 pg/ml, according to Speroff and Fritz (24), is more appropriate as a cut-off reference in our population than the manufacturer s suggestion (2.57 pg/ml in 47 Canadian women, aged from 20 to 39 years). However, we conducted additional analyses using different cut-off values. When the cut-off was increased to 2.57 pg/ml, which was the manufacturer s suggestion, we still observed statistically significant differences. The high free testosterone patients (n ¼ 21) had statistically significantly greater mean lengths than the patients with normal free testosterone (n ¼ 81), as determined using biallelic averages ( vs , P¼.002) and short alleles ( vs , P¼.008) and long alleles ( vs , P¼.013). When the cut-off was decreased to 1.72 pg/ml, which was the mean concentration in PCOS patients in our data, statistically significant differences were still maintained (data not shown). These results may be due to a positive correlation between biallelic average length and free testosterone concentration. In every female cell, one X-chromosome becomes inactive. We did not investigate whether alleles with specific numbers are preferentially inactivated and nonrandom inactivation could have altered our results. But it is doubtful that an X-inactivation feature in peripheral blood leukocytes may reflect the inactivation present in ovarian tissue. In the random inactivation state, averages of both CAG repeat numbers could represent that of target tissue; in our data, the mean biallelic average length was most significantly different between the high and normal free testosterone PCOS patients. The mean age of PCOS patients was statistically significantly lower than that of controls ( vs years, P<.001). This discrepancy was due to the chief complaints of patients and controls. Namely, in women with PCOS, most patients visited the hospital because of irregular menstruation, and most were unmarried, young women. However, the controls largely visited the hospital for cancer screening or infertility and were mainly married women. Nevertheless, we would be surprised if this age discrepancy influenced our results. Our study has not answered the question of whether the AR gene CAG repeat length polymorphism is a direct or major determinant of PCOS or not, but the main findings of our study suggest that the AR gene (CAG) n repeat polymorphism may contribute to the serum concentration of free testosterone in PCOS patients and it may act as a modulator of serum androgen concentration. A subset of PCOS patients with relatively longer CAG repeats in the AR gene (functionally less active AR activity) tend to show higher serum androgen concentration, and they thus may develop symptoms despite decreased AR activity Kim et al. AR gene CAG repeats in PCOS Vol. 90, No. 6, December 2008

6 REFERENCES 1. Knochenhauer ES, Key TJ, Kahsar-Miller M, Waggoner W, Boots LR, Azziz R. Prevalence of the polycystic ovary syndrome in unselected black and white women of the southeastern United States: a prospective study. J Clin Endocrinol Metab 1998;83: Asuncion M, Calvo RM, San Millan JL, Sancho J, Avila S, Escobar- Morreale HF. A prospective study of the prevalence of the polycystic ovary syndrome in unselected Caucasian women from Spain. J Clin Endocrinol Metab 2000;85: Tilley WD, Marcelli M, McPhaul MJ. Recent studies on the androgen receptor: new insights into old questions. Mol Cell Endocrinol 1990;68:C Mahtre AN, Trifiro MA, Kaufman M, Kazemi-Esfarjani P, Figlewicz D, Rouleau G, et al. Reduced transcriptional regulatory competence of the androgen receptor in X-linked spinal and bulbar muscular atrophy. Nat Genet 1993;5: Tut TG, Ghadessy FJ, Trifiro MA, Pinsky L, Yong EL. Long polyglutamine tracts in the androgen receptor are associated with reduced trans-activation, impaired sperm production and male infertility. J Clin Endocrinol Metab 1997;82: Hsiao PW, Lin DL, Nakao R, Chang C. The linkage of Kennedy s neuron disease to ARA24, the first identified androgen receptor polyglutamine region-associated coactivator. J Biol Chem 1999;274: Giovannucci E, Stampfer MJ, Krithivas K, Brown M, Dahl D, Brufsky A, et al. The CAG repeat within the androgen receptor gene and its relationship to prostate cancer. Proc Natl Acad Sci USA 1997;94: Hakimi JM, Schoenberg MP, Rondinelli RH, Piantadosi S, Barrack ER. Androgen receptor variants with short glutamine or glycine repeats may identify unique subpopulations of men with prostate cancer. Clin Cancer Res 1997;3: Stanford JL, Just JJ, Gibbs M, Wicklund KG, Neal CL, Blumenstein BA, et al. Polymorphic repeats in the androgen receptor gene: molecular markers of prostate cancer risk. Cancer Res 1997;57: Sartor O, Zheng Q, Eastham JA. Androgen receptor gene CAG length varies in a race specific fashion in men without prostate cancer. Urology 1999;53: Hardy DO, Scher HI, Bogenreider T, Sabbatini P, Zhang ZF, Nanus DM, et al. Androgen receptor CAG repeat lengths in prostate cancer: correlation with age of onset. Clin Endocrinol Metab 1996;81: La Spada AR, Wilson EM, Lubahn DB, Harding AE, Fischbeck KH. Androgen receptor gene mutations in X-linked spinal and bulbar muscular atrophy. Nature 1991;352: Yong EL, Loy CJ, Sim KS. Androgen receptor gene and male infertility. Hum Reprod Update 2003;9: Mifsud A, Choon AT, Fang D, Yong EL. Prostate-specific antigen, testosterone, sex-hormone binding globulin and androgen receptor CAG repeat polymorphisms in subfertile and normal men. Mol Hum Reprod 2001;7: Legro RS, Shahbahrami B, Robo RA, Kovacs BW. Size polymorphisms of the androgen receptor among female Hispanics and correlation with androgenic characteristics. Obstet Gynecol 1994;83: Sawaya ME, Shalita AR. Androgen receptor polymorphisms (CAG repeat lengths) in androgenic alopecia, hirsutism and acne. J Cutan Med Surg 1998;3: Rebbeck TR, Kantoff PW, Krithivas K, Neuhausen S, Blacjwood MA, Godwin AK, et al. Modification of BRCA1-associated breast cancer risk by the polymorphic androgen receptor CAG repeat. Am J Hum Genet 1999;64: Mifsud A, Ramirez S, Yong EL. Androgen receptor gene CAG trinucleotide repeats in anovulatory infertility and polycystic ovaries. J Clin Endocrinol Metab 2000;85: Franks S, Gharani N, McCarthy M. Candidate genes in polycystic ovary syndrome. Human Reprod Update 2001;7: Legro RS, Strauss JF. Molecular progress in infertility: polycystic ovary syndrome. Fertil Steril 2002;87: Balen AH, Laven JS, Tan SL, Dewailly D. Ultrasound assessment of the polycystic ovary: international consensus definitions. Human Reprod Update 2003;9: Ferriman D, Gallwey JD. Clinical assessment of body hair growth in women. J Clin Endocrinol Metab 1961;21: Franks S. Polycystic ovary syndrome: a changing perspective. Clin Endocrinol (Oxf) 1989;31: Speroff L, Fritz MA. Clinical gynecologic endocrinology and infertility. 7th ed. Philadelphia: Lippincott Williams & Wilkins, 2005; Yeh S, Chang C. Cloning and characterization of a specific coactivator, ARA70, for the androgen receptor in human prostate cells. Proc Natl Acad Sci USA 1996;93: Cude KJ, Dixon SC, Guo Y, Lisella J, Figg WD. The androgen receptor: genetic considerations on the development and treatment of prostate cancer. J Mol Med 1999;77: Yeh S, Miyamoto H, Shima H, Chang C. From estrogen to androgen receptor: a new pathway for sex hormone in prostate. Proc Natl Acad Sci USA 1998;95: McAbee MD, Doncarlos LL. Estrogen, but not androgens, regulates androgen receptor messenger ribonucleic acid expression in the developing male rat forebrain. Endocrinology 1999;140: Culig Z, Hobisch A, Cronauer M, Hittmair A, Radmayr C, Bartsch G, et al. Activation of the androgen receptor by polypeptide growth factors and cellular regulators. World J Urol 1995;13: Gupta C. Modulation of androgen receptor (AR) mediated transcriptional activity by EGF in the developing mouse reproductive tract primary cells. Mol Cell Endocrinol 1999;152: Byun EK, Kim HJ, Oh JY, Hong YS, Sung YA. The prevalence of polycystic ovary syndrome in college students from Seoul. J Korean Soc Endocrinol 2005;20: Lee JY, Cha YJ, Hur SE, Kwon HS, Lee SJ, Sohn IS, et al. Association of the COMT gene polymorphism with the risk of PCOS in Korean women. Kor J Fertil Steril 2006;33: Fertility and Sterility â 2323