- Email: [email protected]
Frequency of the T228A polymorphism in the SORBS1 gene in children with premature pubarche and in adolescent girls with hyperandrogenism
FERTILITY AND STERILITY威 VOL. 80, NO. 1, JULY 2003 Copyright ©2003 American Society for Reproductive Medicine Published by Elsevier Inc. Printed on acid-free paper in U.S.A.
Frequency of the T228A polymorphism in the SORBS1 gene in children with premature pubarche and in adolescent girls with hyperandrogenism Selma F. Witchel, M.D.,a Ram N. Trivedi, Ph.D.,a and Candace Kammerer, Ph.D.b Children’s Hospital of Pittsburgh, University of Pittsburgh, Pittsburgh, Pennsylvania
Received July 31, 2002; revised and accepted November 25, 2002. Supported in part by grants from the National Institutes of Health R29HD34808 (SFW) and 5M01RR-00084 (GCRC), Pennsylvania Chapter of the American Heart Association (SFW), Genentech Foundation (SFW), and the American Heart Association (SFW). Presented in part at the Endocrine Society Annual Meeting, San Francisco, California, June 19 –22, 2002. Reprint requests: Selma F. Witchel, M.D., Division of Endocrinology, Department of Pediatrics, Children’s Hospital of Pittsburgh, 3705 Fifth Avenue, Pittsburgh, Pennsylvania 15213 (FAX: 412-692-5834; E-mail: selma.witchel@chp. edu). a Division of Endocrinology, Department of Pediatrics, Children’s Hospital of Pittsburgh, University of Pittsburgh. b Department of Human Genetics, University of Pittsburgh. 0015-0282/03/$30.00 doi:10.1016/S0015-0282(03) 00506-5
128
Objective: Because the metabolic actions of insulin are more impaired than the mitogenic pathways in polycystic ovary syndrome (PCOS), genes coding for proteins involved in insulin-mediated glucose transport can be considered as candidate genes. The sorbin and SH3-domain-containing-1 (SORBS1) gene codes for c-Cbl–associated protein (CAP) involved in insulin-mediated glucose uptake. An association study showed that a missense variant of the SORBS1 gene is protective against obesity and diabetes. We tested the hypothesis that the frequency of the protective allele would be decreased in children with premature pubarche and adolescent girls with hyperandrogenism. Design: Association study. Setting: Academic research environment. Patient(s): Children referred for the evaluation of premature pubarche (n ⫽ 79), adolescent girls with hyperandrogenism (n ⫽ 56), and healthy nondiabetic controls (n ⫽ 50). Intervention(s): None. Main Outcome Measure(s): Frequency of the T228A allele in our patients and the relationship of body mass index to presence or absence of the T228A variant in our patient population. Result(s): Using allele-specific restriction fragment length polymorphism, allele frequencies were found to be similar among the premature pubarche, hyperandrogenism, and control groups (6.0%, 4.6%, and 8.0%, respectively). No statistically significant relationships were found between the SORBS1 genotypes and body mass index or hormone status. Conclusion(s): This SORBS1 polymorphism does not play a major role in premature pubarche, hyperandrogenism, and/or polycystic ovary syndrome in our patient population. (Fertil Steril威 2003;80:128 –32. ©2003 by American Society for Reproductive Medicine.) Key Words: Premature pubarche, hyperandrogenism, polycystic ovary syndrome, gene, polymorphism
Women with polycystic ovary syndrome (PCOS) have an increased propensity to develop impaired glucose tolerance and diabetes mellitus (1–3). Adolescent girls with hyperinsulinemic ovarian hyperandrogenism and children with premature pubarche (PP), two syndromes that seem to precede the development of PCOS, are often obese and have hyperinsulinemia and/or insulin resistance. Preliminary data indicate that the prevalence of impaired glucose tolerance is also increased in adolescent girls with hyperandrogenism (4). Recently, lifestyle interventions have been demonstrated to modify the natural history of impaired glucose tolerance (5, 6). Identification
of children and adolescents who have an increased risk to develop PCOS, obesity, and/or diabetes mellitus as adults could enable earlier lifestyle interventions. Earlier interventions might alter the natural history of the condition for the child and, perhaps, have a secondary beneficial effect for the nuclear family (7). However, predictive risk factors need to be discerned to be able to accurately assess the efficacy of such lifestyle intervention studies. The familial clustering of PCOS, obesity, and diabetes implies that genes influence the development of these disorders (8, 9). Despite the fact that monogenic etiologies of obesity, diabetes, and insulin resistance have been rec-
ognized, these are uncommon causes (10). To date, the majority of the candidate genes investigated by mutation analysis, linkage, or case-control association studies have failed to be consistently associated with PCOS (11–17). Nevertheless, the frequency of heterozygosity for mutations in the 21-hydroxylase (CYP21) or type 2 3-hydroxysteroid dehydrogenase (HSD3B2) genes is increased in children with PP and adolescent girls with hyperandrogenism (18 – 21). However, heterozygosity for variants of steroidogenic enzyme genes alone is insufficient to elicit the phenotype of PCOS for two reasons. First, only 25% to 35% of affected individuals are heterozygous carriers of CYP21 or HSD3B2 variants (18 –21). Second, another group of heterozygotic carriers of CYP21 mutations, mothers of children with classic congenital adrenal hyperplasia, shows no increase in the frequency of clinical features of hyperandrogenism (22). Hence, we hypothesize that additional genetic and/or environmental factors are necessary for manifestation of the PP, hyperandrogenism, and PCOS phenotypes (18). Additional candidate genes include those coding for proteins involved in insulin-mediated glucose transport because in diabetes, obesity, and PCOS, the metabolic actions of insulin to mediate glucose transport are more impaired than the mitogenic actions of insulin. The binding of insulin to its cognate receptor initiates phosphorylation of the insulin receptor and downstream signaling proteins. One such protein is c-Cbl, which then interacts with the c-Cbl–associated protein (CAP) through one of three SH3 domains. CAP is a member of the sorbin family of proteins and is encoded by the sorbin and SH3-domain-containing-1 (SORBS1) gene. Following phosphorylation, the Cbl-CAP complex migrates from the vicinity of the insulin receptor to the lipid rafts, interacts with flotillin, and recruits CrkII where it complexes with the protein C3G, leading to activation of TC10 and stimulation of glucose uptake (23). Treatment with thiazolidinediones can increase SORBS1 expression in adipose tissue (24). This pathway is a second signal transduction pathway essential for insulin-mediated glucose uptake independent of phosphatidylinositol 3-kinase. The SORBS1 gene contains 34 exons and is located at chromosome 10q23.3-24.1 (25). Two single nucleotide variants predicted to alter amino acids have been identified; one variant, A3 G at nucleotide 682, is predicted to lead to a missense variant T228A. An association study among Chinese adults demonstrated that this polymorphism is protective against obesity and diabetes (26). To test our hypothesis that the frequency of the protective allele would be decreased in patients with PP and hyperandrogenism, we ascertained the frequency of the T228A variant in 79 children with PP, 56 girls with hyperandrogenism, and 50 healthy controls. FERTILITY & STERILITY威
MATERIALS AND METHODS Patients Seventy-nine children (73 girls and 6 boys) were referred to the Children’s Hospital of Pittsburgh for the evaluation of premature development of pubic hair (PP). By definition, pubic hair had developed prior to age 8 years in girls and before 9.5 years in boys. Among the girls, 6 were black, 2 were biracial (Asian-white), and 65 were white. Of the boys, 5 were white and one was black. Fifty-six girls were referred for the evaluation of oligo/amenorrhea and/or hirsutism; there were 53 white and 3 black girls. Height, weight, pubertal stage, and clinical manifestations of hyperandrogenism were recorded for all children and adolescents. Body mass index (BMI) was calculated as weight/height2 (kg/m2). Four pairs of sisters were included. For two pairs, one sister presented with adolescent hyperandrogenism and the younger sister had PP. Both sisters in the other two sibships presented with PP. For determination of allele frequencies, only the eldest sister was included. All four sister pairs are white. For all patients, congenital adrenal hyperplasia, Cushing’s syndrome, and sex-steroid secreting tumors were excluded by history, physical examination, and laboratory data. Congenital adrenal hyperplasia was excluded by ACTH stimulation tests and/or molecular genotype analyses of the 21hydroxylase (CYP21) and 3-hydroxysteroid dehydrogenase type 2 (HSD3B2) genes (18, 19, 21). Fifty healthy unrelated controls were included to determine the frequency of the T228A variant in our population. The control group included 10 children (8 girls and 2 boys) and 40 adults (26 women, and 13 men). Among the controls, 47 were white, 1 was black, 1 was Japanese, and 1 was East Indian. The control children were recruited to be normal controls for other ongoing studies. All had unremarkable medical histories and normal physical examinations. The adult controls were also recruited to be normal controls for other ongoing studies and had unremarkable past medical histories. All women had regular menses; none were infertile. This protocol is approved by the Human Rights Committee of the Children’s Hospital of Pittsburgh. Written informed consent was obtained from the parents of all child and adolescent patients. Assent was obtained from all children ⱖ7 years of age. Written informed consent was obtained all from healthy control subjects.
Hormone Determinations Blood samples were obtained from PP and hyperandrogenism patients for hormone determinations. Androstenedione concentrations were measured by radioimmunoassay as described elsewhere (27). Fasting insulin and glucose measurements were available for 39 patients. Homeostasis model assessment (HOMA) was calculated as fasting serum insulin (gU/mL) ⫻ fasting plasma glucose (mmol/L)/22.5 (28, 29). 129
Mutation Detection Studies Genomic DNA was extracted from peripheral blood leukocytes. Restriction fragment length polymorphism (RFLP) analysis was performed to characterize the T228A variant. Exon 7 was PCR amplified using primers, 7F (5⬘-TACCTCACTGCATGCCCACTCTC-3⬘) and 7R (5⬘-GACTGCTGGGAGGAGACATTCAGAA-3⬘) in a total volume of 12.5 L. Cycling parameters were denaturation at 96°C for 2 minutes, 35 cycles with 94°C for 30 seconds, 65°C for 30 seconds, 72°C for 30 seconds, and 72°C for 10 minutes. Following PCR amplification, the DNA was digested with KasI for 2 hours at 37°C. Digested DNA fragments were electrophoresed on a 1.5% agarose gel containing ethidium bromide and visualized on a ultraviolet transilluminator. The presence of the A3 G variant created a KasI restriction site. Hence, a single 518-bp (base pair) band indicated homozygosity for the T228 allele. The presence of two fragments, 360 and 158-bp, indicated homozygosity for the A228 variant allele. Three bands, 518, 360, and 158-bp, indicated heterozygosity (26). Genotype analyses for the G972R variant of the insulin receptor substrate-1 (IRS-1), W64R variant of the 3-adrenergic receptor (ADRB3), and the P12A variant of the peroxisome proliferator activator-␥2 (PPAR-␥2) genes were performed as previously described (27, 30, 31). Genotype analyses for CYP21 and HSD3B2 were performed as previously described (21, 31, 32, 33).
Statistical Analysis AbSTAT statistical software (Anderson-Bell, Boulder, CO) was used to perform descriptive analyses. Contingency chi-square analyses were used to compare allelic frequencies among the hyperandrogenism, PP, and normal groups. To assess for differences in mean BMI and hormone concentrations among individuals with specific genotypes, values were compared using independent t-tests and analysis of covariance.
RESULTS Patients The patients were divided into four groups based on chronologic age (Table 1) and sex. Group 1 consisted of 45 girls aged 3 to 8 years. All presented for evaluation of PP. Mean (⫾ SD) chronologic age was 6.7 ⫾ 0.9 years. Five were black, two biracial (Asian-white), and 38 white. Their BMI ranged from 14.12 to 25.20 kg/m2 with a mean BMI of 18.94 ⫾ 2.88 kg/m2. Twenty had BMI values less than the 85th percentile for age, 8 had BMI values greater than the 85th percentile and less than the 95th percentile, and 17 had BMI values greater than the 95th percentile for American girls based on the National Center for Health Statistics/ National Center for Chronic Disease Prevention and Health Promotion (2000) (http://www.cdc.gov/growthcharts) (34, 35). 130
Witchel et al.
SORBS1 gene in PP and HA
TABLE 1 Characteristics of children and adolescents for chronologic age (years), body mass index (kg/m2), and androstenedione concentrations (ng/dL). Trait mean Group 1
2 3
4
Genotype
n
Age
BMI
Androstenedione
All A/T T/T T/T All A/T T/T All A/T T/T
45 8 37 29 55 5 50 6 2 4
6.7 ⫾ 0.9 6.6 ⫾ 0.7 6.8 ⫾ 1.0 9.0 ⫾ 1.0 15.3 ⫾ 2.0 15.2 ⫾ 1.6 15.4 ⫾ 2.0 9.3 ⫾ 1.3 8.1, 8.3 9.8 ⫾ 1.2
18.9 ⫾ 2.8 20.7 ⫾ 1.7 18.6 ⫾ 3.0 23.04 ⫾ 6.3 31.5 ⫾ 9.1 34.8 ⫾ 15.4 31.2 ⫾ 8.4 22.7 ⫾ 6.4 16.0, 21.8 24.7 ⫾ 6.8
92 ⫾ 44 (44) 115 ⫾ 29 (8) 87 ⫾ 46 (36) 142 ⫾ 88 (29) 369 ⫾ 142 (53) 285 ⫾ 113 (5) 377 ⫾ 143 (48) 97.3 ⫾ 44 (6) 33, 78 (2) 118 ⫾ 35 (4)
Note: Results are listed by group and genotype. “All” refers to all patients regardless of genotype. “A/T” refers to heterozygotic carriers of the T228A variant. “T/T” refers to homozygotic carriers of the T228 allele. For androstenedione concentrations, the number of patients is provided in parentheses. Mean and standard deviation are listed. Individual values are provided if the number of patients in a group was less than 3. Witchel. SORBS1 gene in PP and HA. Fertil Steril 2003.
Group 2 consisted of 29 girls, aged 8.1 to 12.0 years. One girl presented with hyperandrogenism; the other 28 girls were referred for evaluation of PP. Mean chronologic age was 9.0 ⫾ 1.0 years. One was black and 28 were white. Their BMI ranged from 13.71 to 37.22 kg/m2, with a mean BMI of 23.04 ⫾ 6.29 kg/m2. Eleven had BMI values less than 85th percentile for age, 5 had BMI values greater than 85th percentile and less than the 95th percentile for age, and 13 had BMI values greater than 95th percentile for chronologic age. Group 3 consisted of 55 girls older than 12 years of age. All had experienced menarche and presented with complaints related to oligo/amenorrhea and/or hirsutism. Two girls reported onset of pubic hair before age 8 years and before the onset of breast development; the other girls could not recall age of pubertal onset or sequence of development. Mean chronologic age was 15.3 ⫾ 2.0 years. Three were black, and 52 white. Their BMI ranged from 18.80 to 55.08 kg/m2, with a mean BMI of 31.84 ⫾ 9.12 kg/m2. Fourteen had BMI values less than the 85th percentile, 14 had BMI values greater than the 85th percentile and less than the 95th percentile, and 27 had BMI values greater than the 95th percentile for chronologic age. Group 4 consisted of six boys, aged 8.1 to 12.0 years. Mean chronologic age was 9.3 ⫾ 1.3 years. One was black and five were white. Mean BMI was 22.73 kg/m2. Two boys had BMI values less than the 85th percentile and four boys had BMI values greater than the 95th percentile for chronologic age. Vol. 80, No. 1, July 2003
Genotype Results
Gene–Gene Interactions
Fifteen patients (11.4%) were heterozygous carriers of the T228A variant (see Table 1). None of the patients was homozygous for the variant allele. Of the four pairs of sisters, both sisters in one pair were heterozygous for the T228A variant; in the other three pairs, both sisters were homozygous for the T228 allele. Excluding the younger sister of each pair, frequency of the T228A allele was 14 out of 262 (5.3%). Including one pair of sisters, ethnic distribution of T228A variant carriers was 11 white, 3 black, and 1 biracial. There was no statistically significant difference in allele frequencies between whites and blacks.
Genotypes for three loci, the IRS-1, ADRB3, CYP21, and PPAR␥2 variants, were available for 43 girls in group 1, 27 girls in group 2, and 38 girls in group 3. Because these loci are involved in insulin resistance pathways, interactions between these loci and SORBS1 may be relevant to the development of PCOS. However, the numbers of double heterozygotes were insufficient for formal testing, and visual inspection did not reveal evidence for gene by gene interactions. For example, two patients were heterozygous for the G972R variant of the IRS-1 gene and the T228A variant. One patient was homozygous for the W64R variant of the ADRB3 gene and heterozygous for the T228A variant. Frequencies for the other variant alleles did not differ from published reports. For CYP21, 24 of 87 (27.6%) children were heterozygotic mutation carriers. Four patients were heterozygous for both CYP21 and SORBS1 variants. No gene– gene interactions were apparent.
Eight of 50 (16%) controls were heterozygous carriers of the T228A variant. Excluding the younger sisters, frequencies of the T228A allele were not significantly different among the PP, hyperandrogenism, and control groups, (6.0%, 4.6%, and 8.0%, respectively).
Phenotype-Genotype Correlation We also investigated whether there were any statistically significant relationships between SORBS1 genotypes and phenotypes within individuals with PP or hyperandrogenism. For group 1, the mean BMI was 18.56 ⫾ 3.00 kg/m2 for the 37 patients homozygous for the wild-type T228 allele. Mean BMI for the eight heterozygotic carriers was 20.68 ⫾ 1.68 kg/m2. Three had BMI values greater than the 85th percentile, but less than the 95th percentile. Five had BMI values greater than the 95th percentile for chronologic age. There were no statistically significant differences in BMI, chronologic age, breast development, pubic hair development, or androstenedione concentrations between the carriers and the noncarriers. All girls in group 2 were homozygous for the wild-type T228 allele. For group 3, mean BMI was 34.78 ⫾ 15.4 kg/m2 for the five T228A variant carriers. Two had BMI values less than the 85th percentile for chronologic age and three had BMI values greater than the 95th percentile. Mean BMI was 31.16 ⫾ 8.41 for the 50 patients homozygous for the T228 allele. There were no statistically significant differences in BMI, chronologic age, or androstenedione concentrations between carriers and noncarriers. There were no statistically significant relationships between the SORBS1 genotypes and BMI or hormone status within age groups. Using analysis of covariance in which chronological age was included as a covariate, we also tested whether BMI or hormone concentrations differed between carriers and noncarriers across age groups. As would be predicted from the mean of these phenotypes within the three age groups, there were no statistically significant mean difference across age groups. There were no statistically significant differences in the HOMA between T228A carriers (n ⫽ 4) and noncarriers (n ⫽ 45): 10.26 ⫾ 8.30 versus 8.72 ⫾ 8.30. FERTILITY & STERILITY威
DISCUSSION In a group of Chinese subjects, a single nucleotide polymorphism predicted to generate a missense mutation, T228A, in the human homologue of the CAP protein was found to be protective against diabetes and obesity (26). We hypothesized that, compared with a control population, the frequency of this variant would be lower among our patients with PP and hyperandrogenism because these disorders are associated with increased prevalence of obesity and diabetes. The frequency of the T228A allele was low within the PP, hyperandrogenism, and control groups and did not differ among them. In our control population, T228A allele frequency (8.0%) was comparable to that reported by Lin et al. (7.5%) (26). Importantly, no association was found between BMI and T228A carrier status for individuals with hyperandrogenism or PP. Indeed, in contrast to the results obtained in the Chinese study, most T228A carriers had higher BMI values than noncarriers (26). In addition, insulin resistance expressed as the HOMA was comparable between carriers and noncarriers. Therefore, in our population, the T228A variant does not appear to be protective against obesity. However, this conclusion is moderated because of our sample size and the low frequency of the T228A allele in the control group, our power to detect a decreased frequency among individuals with the disorder was low. Hence, we are unable to exclude the possibility that this variant or another variant in linkage disequilibrium exerts a minor effect. Nevertheless, our association study appears to exclude the T228A variant of the SORBS1 gene as having a major effect as a PCOS, obesity, and/or insulin resistance gene in children and adolescent girls living in western Pennsylvania. References 1. Legro RS, Kunselman AR, Dodson WC, Dunaif A. 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 1999;84:165–9.
131
2. Ehrmann DA, Barnes RB, Rosenfield RL, Cavaghan MK, Imperial J. Prevalence of impaired glucose tolerance and diabetes in women with polycystic ovary syndrome. Diabetes Care 1999;22:141–6. 3. Norman RJ, Masters L, Milner CR, Wang JX, Davies MJ. Relative risk of conversion from normoglycaemia to impaired glucose tolerance or non-insulin dependent diabetes mellitus in polycystic ovarian syndrome. Hum Reprod 2001;16:1995–8. 4. Palmert MR, Gordon CM, Kartashov AI, Legro RS, Emans SJ, Dunaif A. Screening for abnormal glucose tolerance in adolescents with polycystic ovary syndrome. J Clin Endocrinol Metab 2002;87:1017–23. 5. Knowler WC, Barrett-Connor E, Fowler SE, Hamman RF, Lachin JM, Walker EA, et al. Reduction in the incidence of type 2 diabetes with lifestyle intervention or metformin. N Engl J Med 2002;346:393–403. 6. Tuomilehto J, Lindstrom J, Eriksson JG, Valle TT, Hamalainen H, Ilanne-Parikka P, et al. Prevention of type 2 diabetes mellitus by changes in lifestyle among subjects with impaired glucose tolerance. N Engl J Med 2001;344:1343–50. 7. Sir-Petermann T, Angel B, Maliqueo M, Carvajal F, Santos JL, Pe´rezBravo F. Prevalence of type II diabetes mellitus and insulin resistance in parents of women with polycystic ovary syndrome. Diabetologia 2002;45:959 –64. 8. Kashar-Miller M, Azziz R. Heritability and the risk of developing androgen excess. J Steroid Biochem Mol Biol 1999;69:261–8. 9. Franks S, Gharani N, Waterworth D, Batty S, White D, Williamson R, et al. The genetic basis of polycystic ovary syndrome. Hum Reprod 1997;12:2641–8. 10. Arner P. Obesity—a genetic disease of adipose tissue. Br J Nutr 2000;83(Suppl 1):S9 –S16. 11. Gharani N, Waterworth DM, Batty S, White D, Gilling-Smith C, Conway GS, et al. Association of the steroid synthesis gene CYP11a with polycystic ovary syndrome and hyperandrogenism. Hum Mol Genet 1997;6:397–402. 12. Waterworth DM, Bennett ST, Gharani N, McCarthy MI, Hague S, Batty S, et al. Linkage and association of insulin gene VNTR regulatory polymorphism with polycystic ovary syndrome. Lancet 1997;349:986 – 90. 13. Urbanek M, Legro RS, Driscoll DA, Azziz R, Ehrmann DA, Norman RJ, et al. Thirty-seven candidate genes for polycystic ovary syndrome: strongest evidence for linkage is with follistatin. Proc Natl Acad Sci 1999;96:8573–8. 14. Calvo RM, Villuendas G, Sancho J, San Millan JL, Escobar-Morreale HF. Role of the follistatin gene in women with polycystic ovary syndrome. Fertil Steril 2000;75:1020 –3. 15. Talbot JA, Bicknell EJ, Rajkhowa M, Krook A, O’Rahilly S, Clayton RN. Molecular scanning of the insulin receptor gene in women with polycystic ovarian syndrome. J Clin Endocrinol Metab 1996;81:1979 – 83. 16. Cohen DP, Stein EM, Li Z, Matulis CK, Ehrmann DA, Layman LC. Molecular analysis of the gonadotropin-releasing hormone receptor in patients with polycystic ovary syndrome. Fertil Steril 1999;72:360 –3. 17. Gharani N, Waterworth DM, Williamson R, Franks S. 5⬘ polymorphism of the CYP17 gene is not associated with serum testosterone levels in women with polycystic ovaries [letter]. J Clin Endocrinol Metab 1996; 81:4174. 18. Witchel SF, Aston CE. The role of heterozygosity for CYP21 in the polycystic ovary syndrome. J Pediatr Endocrinol Metab 2000;13(Suppl 5):1315–7. 19. Escobar-Morreale HF, San Millan JL, Smith RR, Sancho J, Witchel SF. The presence of the 21-hydroxylase deficiency carrier status in hirsute
132
Witchel et al.
SORBS1 gene in PP and HA
20. 21. 22.
23. 24.
25.
26.
27.
28.
29.
30.
31. 32. 33. 34. 35.
women: phenotype-genotype correlations. Fertil Steril 1999;72:629 – 38. Blanche H, Vexiau P, Clauin S, Le Gall I, Fiet J, Mornet E, et al. Exhaustive screening of the 21-hydroxylase gene in a population of hyperandrogenic women. Hum Genet 1997;101:56 –60. Nayak S, Lee PA, Witchel SF. Variants of the type II 3-hydroxysteroid dehydrogenase gene in children with premature pubic hair and hyperandrogenic adolescents. Mol Genet Metab 1998;64:184 –92. Knochenhauer ES, Cordet-Rudelli C, Cunningham RD, Conway-Meyers BA, Dewailly D, Azziz R. Carriers of 21-hydroxylase deficiency are not at increased risk for hyperandrogenism. J Clin Endocrinol Metab 1997;82:479 –85. Baumann CA, Ribon V, Kanzaki M, Thurmond DC, Mora S, Shigematsu S, et al. CAP defines a second signalling pathway required for insulin-stimulated glucose transport. Nature 2000;407:202–7. Ribon V, Johnson JH, Camp HS, Saltiel AR. Thiazolidinediones and insulin resistance: peroxisome proliferator activated receptor ␥ activation stimulates expression of the CAP gene. Proc Natl Acad Sci 1998; 95:14751–6. Lin WH, Huang CJ, Liu MW, Chang HM, Chen YJ, Tai TY, Chuang LM. Cloning, mapping, and characterization of human sorbin and SH3 domain containing 1 (SORBS1) gene: a protein associated with c-Abl during insulin signaling in hepatoma cell line. Genomics 2001;74:12– 20. Lin W-H, Chiu KC, Chang H-M, Lee K-C, Tai T-Y, Chuang L-M. Molecular scanning of the human sorbin and SH-3-domain-containing-1 (SORBS1) gene: positive association of the T228A polymorphism with obesity and type 2 diabetes. Hum Mol Genet 2001;10:1753–60. Witchel SF, Fagerli J, Siegel J, Smith R, Mitwally MF, Lewy V, Arslanian S, Lee PA. No association between body mass index and 3-adrenergic receptor variant (W64R) in children with premature pubarche and adolescent girls with hyperandrogenism. Fertil Steril 2000;73:509 –15. Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DR, Turner RC. Homeostasis model assessment: insulin resistance and -cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia 1985;28:412–19. Bonora E, Targher G, Alberiche M, Bonadonna RC, Saggiani F, Zenere MB, et al. Homeostasis model assessment closely mirrors the glucose clamp technique in the assessment of insulin sensitivity. Diabetes Care 2000;23:57–63. Witchel SF, White C, Siegel ME, Aston CE. Inconsistent effects of the proline123alanine variant of the peroxisome proliferator-activated receptor-␥2 gene on body mass index in children and adolescent girls. Fertil Steril 2001;76:741–7. Witchel SF, Smith R, Tomboc M, Aston CE. Candidate gene analysis in premature pubarche and adolescent hyperandrogenism. Fertil Steril 2001;75:724 –30. Siegel SF, Hoffman EP, Trucco M. Molecular diagnosis of 21-hydroxylase deficiency: detection of four mutations on a single gel. Biochem Med Metab Biol 1994;51:66 –73. Siegel SF, Lee PA, Rudert WA, Swinyard M, Trucco M. Phenotype/ genotype correlations in 21-hydroxylase deficiency. Adolesc Pediatr Gynecol 1995;8:9 –16. Andres R, Elahi D, Tobin JD, Muller DC, Brant L. Impact of age on weight goals. Ann Intern Med 1985;103:1030 –3. Guo SS, Chumlea WC. Tracking of body mass index in children in relation to overweight in adulthood. Am J Clin Nutr 1999;70(Suppl): 145S–148S.
Vol. 80, No. 1, July 2003
Frequency of the T228A polymorphism in the SORBS1 gene in children with premature pubarche and in adolescent girls with hyperandrogenism Selma F. Witchel, M.D.,a Ram N. Trivedi, Ph.D.,a and Candace Kammerer, Ph.D.b Children’s Hospital of Pittsburgh, University of Pittsburgh, Pittsburgh, Pennsylvania
Received July 31, 2002; revised and accepted November 25, 2002. Supported in part by grants from the National Institutes of Health R29HD34808 (SFW) and 5M01RR-00084 (GCRC), Pennsylvania Chapter of the American Heart Association (SFW), Genentech Foundation (SFW), and the American Heart Association (SFW). Presented in part at the Endocrine Society Annual Meeting, San Francisco, California, June 19 –22, 2002. Reprint requests: Selma F. Witchel, M.D., Division of Endocrinology, Department of Pediatrics, Children’s Hospital of Pittsburgh, 3705 Fifth Avenue, Pittsburgh, Pennsylvania 15213 (FAX: 412-692-5834; E-mail: selma.witchel@chp. edu). a Division of Endocrinology, Department of Pediatrics, Children’s Hospital of Pittsburgh, University of Pittsburgh. b Department of Human Genetics, University of Pittsburgh. 0015-0282/03/$30.00 doi:10.1016/S0015-0282(03) 00506-5
128
Objective: Because the metabolic actions of insulin are more impaired than the mitogenic pathways in polycystic ovary syndrome (PCOS), genes coding for proteins involved in insulin-mediated glucose transport can be considered as candidate genes. The sorbin and SH3-domain-containing-1 (SORBS1) gene codes for c-Cbl–associated protein (CAP) involved in insulin-mediated glucose uptake. An association study showed that a missense variant of the SORBS1 gene is protective against obesity and diabetes. We tested the hypothesis that the frequency of the protective allele would be decreased in children with premature pubarche and adolescent girls with hyperandrogenism. Design: Association study. Setting: Academic research environment. Patient(s): Children referred for the evaluation of premature pubarche (n ⫽ 79), adolescent girls with hyperandrogenism (n ⫽ 56), and healthy nondiabetic controls (n ⫽ 50). Intervention(s): None. Main Outcome Measure(s): Frequency of the T228A allele in our patients and the relationship of body mass index to presence or absence of the T228A variant in our patient population. Result(s): Using allele-specific restriction fragment length polymorphism, allele frequencies were found to be similar among the premature pubarche, hyperandrogenism, and control groups (6.0%, 4.6%, and 8.0%, respectively). No statistically significant relationships were found between the SORBS1 genotypes and body mass index or hormone status. Conclusion(s): This SORBS1 polymorphism does not play a major role in premature pubarche, hyperandrogenism, and/or polycystic ovary syndrome in our patient population. (Fertil Steril威 2003;80:128 –32. ©2003 by American Society for Reproductive Medicine.) Key Words: Premature pubarche, hyperandrogenism, polycystic ovary syndrome, gene, polymorphism
Women with polycystic ovary syndrome (PCOS) have an increased propensity to develop impaired glucose tolerance and diabetes mellitus (1–3). Adolescent girls with hyperinsulinemic ovarian hyperandrogenism and children with premature pubarche (PP), two syndromes that seem to precede the development of PCOS, are often obese and have hyperinsulinemia and/or insulin resistance. Preliminary data indicate that the prevalence of impaired glucose tolerance is also increased in adolescent girls with hyperandrogenism (4). Recently, lifestyle interventions have been demonstrated to modify the natural history of impaired glucose tolerance (5, 6). Identification
of children and adolescents who have an increased risk to develop PCOS, obesity, and/or diabetes mellitus as adults could enable earlier lifestyle interventions. Earlier interventions might alter the natural history of the condition for the child and, perhaps, have a secondary beneficial effect for the nuclear family (7). However, predictive risk factors need to be discerned to be able to accurately assess the efficacy of such lifestyle intervention studies. The familial clustering of PCOS, obesity, and diabetes implies that genes influence the development of these disorders (8, 9). Despite the fact that monogenic etiologies of obesity, diabetes, and insulin resistance have been rec-
ognized, these are uncommon causes (10). To date, the majority of the candidate genes investigated by mutation analysis, linkage, or case-control association studies have failed to be consistently associated with PCOS (11–17). Nevertheless, the frequency of heterozygosity for mutations in the 21-hydroxylase (CYP21) or type 2 3-hydroxysteroid dehydrogenase (HSD3B2) genes is increased in children with PP and adolescent girls with hyperandrogenism (18 – 21). However, heterozygosity for variants of steroidogenic enzyme genes alone is insufficient to elicit the phenotype of PCOS for two reasons. First, only 25% to 35% of affected individuals are heterozygous carriers of CYP21 or HSD3B2 variants (18 –21). Second, another group of heterozygotic carriers of CYP21 mutations, mothers of children with classic congenital adrenal hyperplasia, shows no increase in the frequency of clinical features of hyperandrogenism (22). Hence, we hypothesize that additional genetic and/or environmental factors are necessary for manifestation of the PP, hyperandrogenism, and PCOS phenotypes (18). Additional candidate genes include those coding for proteins involved in insulin-mediated glucose transport because in diabetes, obesity, and PCOS, the metabolic actions of insulin to mediate glucose transport are more impaired than the mitogenic actions of insulin. The binding of insulin to its cognate receptor initiates phosphorylation of the insulin receptor and downstream signaling proteins. One such protein is c-Cbl, which then interacts with the c-Cbl–associated protein (CAP) through one of three SH3 domains. CAP is a member of the sorbin family of proteins and is encoded by the sorbin and SH3-domain-containing-1 (SORBS1) gene. Following phosphorylation, the Cbl-CAP complex migrates from the vicinity of the insulin receptor to the lipid rafts, interacts with flotillin, and recruits CrkII where it complexes with the protein C3G, leading to activation of TC10 and stimulation of glucose uptake (23). Treatment with thiazolidinediones can increase SORBS1 expression in adipose tissue (24). This pathway is a second signal transduction pathway essential for insulin-mediated glucose uptake independent of phosphatidylinositol 3-kinase. The SORBS1 gene contains 34 exons and is located at chromosome 10q23.3-24.1 (25). Two single nucleotide variants predicted to alter amino acids have been identified; one variant, A3 G at nucleotide 682, is predicted to lead to a missense variant T228A. An association study among Chinese adults demonstrated that this polymorphism is protective against obesity and diabetes (26). To test our hypothesis that the frequency of the protective allele would be decreased in patients with PP and hyperandrogenism, we ascertained the frequency of the T228A variant in 79 children with PP, 56 girls with hyperandrogenism, and 50 healthy controls. FERTILITY & STERILITY威
MATERIALS AND METHODS Patients Seventy-nine children (73 girls and 6 boys) were referred to the Children’s Hospital of Pittsburgh for the evaluation of premature development of pubic hair (PP). By definition, pubic hair had developed prior to age 8 years in girls and before 9.5 years in boys. Among the girls, 6 were black, 2 were biracial (Asian-white), and 65 were white. Of the boys, 5 were white and one was black. Fifty-six girls were referred for the evaluation of oligo/amenorrhea and/or hirsutism; there were 53 white and 3 black girls. Height, weight, pubertal stage, and clinical manifestations of hyperandrogenism were recorded for all children and adolescents. Body mass index (BMI) was calculated as weight/height2 (kg/m2). Four pairs of sisters were included. For two pairs, one sister presented with adolescent hyperandrogenism and the younger sister had PP. Both sisters in the other two sibships presented with PP. For determination of allele frequencies, only the eldest sister was included. All four sister pairs are white. For all patients, congenital adrenal hyperplasia, Cushing’s syndrome, and sex-steroid secreting tumors were excluded by history, physical examination, and laboratory data. Congenital adrenal hyperplasia was excluded by ACTH stimulation tests and/or molecular genotype analyses of the 21hydroxylase (CYP21) and 3-hydroxysteroid dehydrogenase type 2 (HSD3B2) genes (18, 19, 21). Fifty healthy unrelated controls were included to determine the frequency of the T228A variant in our population. The control group included 10 children (8 girls and 2 boys) and 40 adults (26 women, and 13 men). Among the controls, 47 were white, 1 was black, 1 was Japanese, and 1 was East Indian. The control children were recruited to be normal controls for other ongoing studies. All had unremarkable medical histories and normal physical examinations. The adult controls were also recruited to be normal controls for other ongoing studies and had unremarkable past medical histories. All women had regular menses; none were infertile. This protocol is approved by the Human Rights Committee of the Children’s Hospital of Pittsburgh. Written informed consent was obtained from the parents of all child and adolescent patients. Assent was obtained from all children ⱖ7 years of age. Written informed consent was obtained all from healthy control subjects.
Hormone Determinations Blood samples were obtained from PP and hyperandrogenism patients for hormone determinations. Androstenedione concentrations were measured by radioimmunoassay as described elsewhere (27). Fasting insulin and glucose measurements were available for 39 patients. Homeostasis model assessment (HOMA) was calculated as fasting serum insulin (gU/mL) ⫻ fasting plasma glucose (mmol/L)/22.5 (28, 29). 129
Mutation Detection Studies Genomic DNA was extracted from peripheral blood leukocytes. Restriction fragment length polymorphism (RFLP) analysis was performed to characterize the T228A variant. Exon 7 was PCR amplified using primers, 7F (5⬘-TACCTCACTGCATGCCCACTCTC-3⬘) and 7R (5⬘-GACTGCTGGGAGGAGACATTCAGAA-3⬘) in a total volume of 12.5 L. Cycling parameters were denaturation at 96°C for 2 minutes, 35 cycles with 94°C for 30 seconds, 65°C for 30 seconds, 72°C for 30 seconds, and 72°C for 10 minutes. Following PCR amplification, the DNA was digested with KasI for 2 hours at 37°C. Digested DNA fragments were electrophoresed on a 1.5% agarose gel containing ethidium bromide and visualized on a ultraviolet transilluminator. The presence of the A3 G variant created a KasI restriction site. Hence, a single 518-bp (base pair) band indicated homozygosity for the T228 allele. The presence of two fragments, 360 and 158-bp, indicated homozygosity for the A228 variant allele. Three bands, 518, 360, and 158-bp, indicated heterozygosity (26). Genotype analyses for the G972R variant of the insulin receptor substrate-1 (IRS-1), W64R variant of the 3-adrenergic receptor (ADRB3), and the P12A variant of the peroxisome proliferator activator-␥2 (PPAR-␥2) genes were performed as previously described (27, 30, 31). Genotype analyses for CYP21 and HSD3B2 were performed as previously described (21, 31, 32, 33).
Statistical Analysis AbSTAT statistical software (Anderson-Bell, Boulder, CO) was used to perform descriptive analyses. Contingency chi-square analyses were used to compare allelic frequencies among the hyperandrogenism, PP, and normal groups. To assess for differences in mean BMI and hormone concentrations among individuals with specific genotypes, values were compared using independent t-tests and analysis of covariance.
RESULTS Patients The patients were divided into four groups based on chronologic age (Table 1) and sex. Group 1 consisted of 45 girls aged 3 to 8 years. All presented for evaluation of PP. Mean (⫾ SD) chronologic age was 6.7 ⫾ 0.9 years. Five were black, two biracial (Asian-white), and 38 white. Their BMI ranged from 14.12 to 25.20 kg/m2 with a mean BMI of 18.94 ⫾ 2.88 kg/m2. Twenty had BMI values less than the 85th percentile for age, 8 had BMI values greater than the 85th percentile and less than the 95th percentile, and 17 had BMI values greater than the 95th percentile for American girls based on the National Center for Health Statistics/ National Center for Chronic Disease Prevention and Health Promotion (2000) (http://www.cdc.gov/growthcharts) (34, 35). 130
Witchel et al.
SORBS1 gene in PP and HA
TABLE 1 Characteristics of children and adolescents for chronologic age (years), body mass index (kg/m2), and androstenedione concentrations (ng/dL). Trait mean Group 1
2 3
4
Genotype
n
Age
BMI
Androstenedione
All A/T T/T T/T All A/T T/T All A/T T/T
45 8 37 29 55 5 50 6 2 4
6.7 ⫾ 0.9 6.6 ⫾ 0.7 6.8 ⫾ 1.0 9.0 ⫾ 1.0 15.3 ⫾ 2.0 15.2 ⫾ 1.6 15.4 ⫾ 2.0 9.3 ⫾ 1.3 8.1, 8.3 9.8 ⫾ 1.2
18.9 ⫾ 2.8 20.7 ⫾ 1.7 18.6 ⫾ 3.0 23.04 ⫾ 6.3 31.5 ⫾ 9.1 34.8 ⫾ 15.4 31.2 ⫾ 8.4 22.7 ⫾ 6.4 16.0, 21.8 24.7 ⫾ 6.8
92 ⫾ 44 (44) 115 ⫾ 29 (8) 87 ⫾ 46 (36) 142 ⫾ 88 (29) 369 ⫾ 142 (53) 285 ⫾ 113 (5) 377 ⫾ 143 (48) 97.3 ⫾ 44 (6) 33, 78 (2) 118 ⫾ 35 (4)
Note: Results are listed by group and genotype. “All” refers to all patients regardless of genotype. “A/T” refers to heterozygotic carriers of the T228A variant. “T/T” refers to homozygotic carriers of the T228 allele. For androstenedione concentrations, the number of patients is provided in parentheses. Mean and standard deviation are listed. Individual values are provided if the number of patients in a group was less than 3. Witchel. SORBS1 gene in PP and HA. Fertil Steril 2003.
Group 2 consisted of 29 girls, aged 8.1 to 12.0 years. One girl presented with hyperandrogenism; the other 28 girls were referred for evaluation of PP. Mean chronologic age was 9.0 ⫾ 1.0 years. One was black and 28 were white. Their BMI ranged from 13.71 to 37.22 kg/m2, with a mean BMI of 23.04 ⫾ 6.29 kg/m2. Eleven had BMI values less than 85th percentile for age, 5 had BMI values greater than 85th percentile and less than the 95th percentile for age, and 13 had BMI values greater than 95th percentile for chronologic age. Group 3 consisted of 55 girls older than 12 years of age. All had experienced menarche and presented with complaints related to oligo/amenorrhea and/or hirsutism. Two girls reported onset of pubic hair before age 8 years and before the onset of breast development; the other girls could not recall age of pubertal onset or sequence of development. Mean chronologic age was 15.3 ⫾ 2.0 years. Three were black, and 52 white. Their BMI ranged from 18.80 to 55.08 kg/m2, with a mean BMI of 31.84 ⫾ 9.12 kg/m2. Fourteen had BMI values less than the 85th percentile, 14 had BMI values greater than the 85th percentile and less than the 95th percentile, and 27 had BMI values greater than the 95th percentile for chronologic age. Group 4 consisted of six boys, aged 8.1 to 12.0 years. Mean chronologic age was 9.3 ⫾ 1.3 years. One was black and five were white. Mean BMI was 22.73 kg/m2. Two boys had BMI values less than the 85th percentile and four boys had BMI values greater than the 95th percentile for chronologic age. Vol. 80, No. 1, July 2003
Genotype Results
Gene–Gene Interactions
Fifteen patients (11.4%) were heterozygous carriers of the T228A variant (see Table 1). None of the patients was homozygous for the variant allele. Of the four pairs of sisters, both sisters in one pair were heterozygous for the T228A variant; in the other three pairs, both sisters were homozygous for the T228 allele. Excluding the younger sister of each pair, frequency of the T228A allele was 14 out of 262 (5.3%). Including one pair of sisters, ethnic distribution of T228A variant carriers was 11 white, 3 black, and 1 biracial. There was no statistically significant difference in allele frequencies between whites and blacks.
Genotypes for three loci, the IRS-1, ADRB3, CYP21, and PPAR␥2 variants, were available for 43 girls in group 1, 27 girls in group 2, and 38 girls in group 3. Because these loci are involved in insulin resistance pathways, interactions between these loci and SORBS1 may be relevant to the development of PCOS. However, the numbers of double heterozygotes were insufficient for formal testing, and visual inspection did not reveal evidence for gene by gene interactions. For example, two patients were heterozygous for the G972R variant of the IRS-1 gene and the T228A variant. One patient was homozygous for the W64R variant of the ADRB3 gene and heterozygous for the T228A variant. Frequencies for the other variant alleles did not differ from published reports. For CYP21, 24 of 87 (27.6%) children were heterozygotic mutation carriers. Four patients were heterozygous for both CYP21 and SORBS1 variants. No gene– gene interactions were apparent.
Eight of 50 (16%) controls were heterozygous carriers of the T228A variant. Excluding the younger sisters, frequencies of the T228A allele were not significantly different among the PP, hyperandrogenism, and control groups, (6.0%, 4.6%, and 8.0%, respectively).
Phenotype-Genotype Correlation We also investigated whether there were any statistically significant relationships between SORBS1 genotypes and phenotypes within individuals with PP or hyperandrogenism. For group 1, the mean BMI was 18.56 ⫾ 3.00 kg/m2 for the 37 patients homozygous for the wild-type T228 allele. Mean BMI for the eight heterozygotic carriers was 20.68 ⫾ 1.68 kg/m2. Three had BMI values greater than the 85th percentile, but less than the 95th percentile. Five had BMI values greater than the 95th percentile for chronologic age. There were no statistically significant differences in BMI, chronologic age, breast development, pubic hair development, or androstenedione concentrations between the carriers and the noncarriers. All girls in group 2 were homozygous for the wild-type T228 allele. For group 3, mean BMI was 34.78 ⫾ 15.4 kg/m2 for the five T228A variant carriers. Two had BMI values less than the 85th percentile for chronologic age and three had BMI values greater than the 95th percentile. Mean BMI was 31.16 ⫾ 8.41 for the 50 patients homozygous for the T228 allele. There were no statistically significant differences in BMI, chronologic age, or androstenedione concentrations between carriers and noncarriers. There were no statistically significant relationships between the SORBS1 genotypes and BMI or hormone status within age groups. Using analysis of covariance in which chronological age was included as a covariate, we also tested whether BMI or hormone concentrations differed between carriers and noncarriers across age groups. As would be predicted from the mean of these phenotypes within the three age groups, there were no statistically significant mean difference across age groups. There were no statistically significant differences in the HOMA between T228A carriers (n ⫽ 4) and noncarriers (n ⫽ 45): 10.26 ⫾ 8.30 versus 8.72 ⫾ 8.30. FERTILITY & STERILITY威
DISCUSSION In a group of Chinese subjects, a single nucleotide polymorphism predicted to generate a missense mutation, T228A, in the human homologue of the CAP protein was found to be protective against diabetes and obesity (26). We hypothesized that, compared with a control population, the frequency of this variant would be lower among our patients with PP and hyperandrogenism because these disorders are associated with increased prevalence of obesity and diabetes. The frequency of the T228A allele was low within the PP, hyperandrogenism, and control groups and did not differ among them. In our control population, T228A allele frequency (8.0%) was comparable to that reported by Lin et al. (7.5%) (26). Importantly, no association was found between BMI and T228A carrier status for individuals with hyperandrogenism or PP. Indeed, in contrast to the results obtained in the Chinese study, most T228A carriers had higher BMI values than noncarriers (26). In addition, insulin resistance expressed as the HOMA was comparable between carriers and noncarriers. Therefore, in our population, the T228A variant does not appear to be protective against obesity. However, this conclusion is moderated because of our sample size and the low frequency of the T228A allele in the control group, our power to detect a decreased frequency among individuals with the disorder was low. Hence, we are unable to exclude the possibility that this variant or another variant in linkage disequilibrium exerts a minor effect. Nevertheless, our association study appears to exclude the T228A variant of the SORBS1 gene as having a major effect as a PCOS, obesity, and/or insulin resistance gene in children and adolescent girls living in western Pennsylvania. References 1. Legro RS, Kunselman AR, Dodson WC, Dunaif A. 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 1999;84:165–9.
131
2. Ehrmann DA, Barnes RB, Rosenfield RL, Cavaghan MK, Imperial J. Prevalence of impaired glucose tolerance and diabetes in women with polycystic ovary syndrome. Diabetes Care 1999;22:141–6. 3. Norman RJ, Masters L, Milner CR, Wang JX, Davies MJ. Relative risk of conversion from normoglycaemia to impaired glucose tolerance or non-insulin dependent diabetes mellitus in polycystic ovarian syndrome. Hum Reprod 2001;16:1995–8. 4. Palmert MR, Gordon CM, Kartashov AI, Legro RS, Emans SJ, Dunaif A. Screening for abnormal glucose tolerance in adolescents with polycystic ovary syndrome. J Clin Endocrinol Metab 2002;87:1017–23. 5. Knowler WC, Barrett-Connor E, Fowler SE, Hamman RF, Lachin JM, Walker EA, et al. Reduction in the incidence of type 2 diabetes with lifestyle intervention or metformin. N Engl J Med 2002;346:393–403. 6. Tuomilehto J, Lindstrom J, Eriksson JG, Valle TT, Hamalainen H, Ilanne-Parikka P, et al. Prevention of type 2 diabetes mellitus by changes in lifestyle among subjects with impaired glucose tolerance. N Engl J Med 2001;344:1343–50. 7. Sir-Petermann T, Angel B, Maliqueo M, Carvajal F, Santos JL, Pe´rezBravo F. Prevalence of type II diabetes mellitus and insulin resistance in parents of women with polycystic ovary syndrome. Diabetologia 2002;45:959 –64. 8. Kashar-Miller M, Azziz R. Heritability and the risk of developing androgen excess. J Steroid Biochem Mol Biol 1999;69:261–8. 9. Franks S, Gharani N, Waterworth D, Batty S, White D, Williamson R, et al. The genetic basis of polycystic ovary syndrome. Hum Reprod 1997;12:2641–8. 10. Arner P. Obesity—a genetic disease of adipose tissue. Br J Nutr 2000;83(Suppl 1):S9 –S16. 11. Gharani N, Waterworth DM, Batty S, White D, Gilling-Smith C, Conway GS, et al. Association of the steroid synthesis gene CYP11a with polycystic ovary syndrome and hyperandrogenism. Hum Mol Genet 1997;6:397–402. 12. Waterworth DM, Bennett ST, Gharani N, McCarthy MI, Hague S, Batty S, et al. Linkage and association of insulin gene VNTR regulatory polymorphism with polycystic ovary syndrome. Lancet 1997;349:986 – 90. 13. Urbanek M, Legro RS, Driscoll DA, Azziz R, Ehrmann DA, Norman RJ, et al. Thirty-seven candidate genes for polycystic ovary syndrome: strongest evidence for linkage is with follistatin. Proc Natl Acad Sci 1999;96:8573–8. 14. Calvo RM, Villuendas G, Sancho J, San Millan JL, Escobar-Morreale HF. Role of the follistatin gene in women with polycystic ovary syndrome. Fertil Steril 2000;75:1020 –3. 15. Talbot JA, Bicknell EJ, Rajkhowa M, Krook A, O’Rahilly S, Clayton RN. Molecular scanning of the insulin receptor gene in women with polycystic ovarian syndrome. J Clin Endocrinol Metab 1996;81:1979 – 83. 16. Cohen DP, Stein EM, Li Z, Matulis CK, Ehrmann DA, Layman LC. Molecular analysis of the gonadotropin-releasing hormone receptor in patients with polycystic ovary syndrome. Fertil Steril 1999;72:360 –3. 17. Gharani N, Waterworth DM, Williamson R, Franks S. 5⬘ polymorphism of the CYP17 gene is not associated with serum testosterone levels in women with polycystic ovaries [letter]. J Clin Endocrinol Metab 1996; 81:4174. 18. Witchel SF, Aston CE. The role of heterozygosity for CYP21 in the polycystic ovary syndrome. J Pediatr Endocrinol Metab 2000;13(Suppl 5):1315–7. 19. Escobar-Morreale HF, San Millan JL, Smith RR, Sancho J, Witchel SF. The presence of the 21-hydroxylase deficiency carrier status in hirsute
132
Witchel et al.
SORBS1 gene in PP and HA
20. 21. 22.
23. 24.
25.
26.
27.
28.
29.
30.
31. 32. 33. 34. 35.
women: phenotype-genotype correlations. Fertil Steril 1999;72:629 – 38. Blanche H, Vexiau P, Clauin S, Le Gall I, Fiet J, Mornet E, et al. Exhaustive screening of the 21-hydroxylase gene in a population of hyperandrogenic women. Hum Genet 1997;101:56 –60. Nayak S, Lee PA, Witchel SF. Variants of the type II 3-hydroxysteroid dehydrogenase gene in children with premature pubic hair and hyperandrogenic adolescents. Mol Genet Metab 1998;64:184 –92. Knochenhauer ES, Cordet-Rudelli C, Cunningham RD, Conway-Meyers BA, Dewailly D, Azziz R. Carriers of 21-hydroxylase deficiency are not at increased risk for hyperandrogenism. J Clin Endocrinol Metab 1997;82:479 –85. Baumann CA, Ribon V, Kanzaki M, Thurmond DC, Mora S, Shigematsu S, et al. CAP defines a second signalling pathway required for insulin-stimulated glucose transport. Nature 2000;407:202–7. Ribon V, Johnson JH, Camp HS, Saltiel AR. Thiazolidinediones and insulin resistance: peroxisome proliferator activated receptor ␥ activation stimulates expression of the CAP gene. Proc Natl Acad Sci 1998; 95:14751–6. Lin WH, Huang CJ, Liu MW, Chang HM, Chen YJ, Tai TY, Chuang LM. Cloning, mapping, and characterization of human sorbin and SH3 domain containing 1 (SORBS1) gene: a protein associated with c-Abl during insulin signaling in hepatoma cell line. Genomics 2001;74:12– 20. Lin W-H, Chiu KC, Chang H-M, Lee K-C, Tai T-Y, Chuang L-M. Molecular scanning of the human sorbin and SH-3-domain-containing-1 (SORBS1) gene: positive association of the T228A polymorphism with obesity and type 2 diabetes. Hum Mol Genet 2001;10:1753–60. Witchel SF, Fagerli J, Siegel J, Smith R, Mitwally MF, Lewy V, Arslanian S, Lee PA. No association between body mass index and 3-adrenergic receptor variant (W64R) in children with premature pubarche and adolescent girls with hyperandrogenism. Fertil Steril 2000;73:509 –15. Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DR, Turner RC. Homeostasis model assessment: insulin resistance and -cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia 1985;28:412–19. Bonora E, Targher G, Alberiche M, Bonadonna RC, Saggiani F, Zenere MB, et al. Homeostasis model assessment closely mirrors the glucose clamp technique in the assessment of insulin sensitivity. Diabetes Care 2000;23:57–63. Witchel SF, White C, Siegel ME, Aston CE. Inconsistent effects of the proline123alanine variant of the peroxisome proliferator-activated receptor-␥2 gene on body mass index in children and adolescent girls. Fertil Steril 2001;76:741–7. Witchel SF, Smith R, Tomboc M, Aston CE. Candidate gene analysis in premature pubarche and adolescent hyperandrogenism. Fertil Steril 2001;75:724 –30. Siegel SF, Hoffman EP, Trucco M. Molecular diagnosis of 21-hydroxylase deficiency: detection of four mutations on a single gel. Biochem Med Metab Biol 1994;51:66 –73. Siegel SF, Lee PA, Rudert WA, Swinyard M, Trucco M. Phenotype/ genotype correlations in 21-hydroxylase deficiency. Adolesc Pediatr Gynecol 1995;8:9 –16. Andres R, Elahi D, Tobin JD, Muller DC, Brant L. Impact of age on weight goals. Ann Intern Med 1985;103:1030 –3. Guo SS, Chumlea WC. Tracking of body mass index in children in relation to overweight in adulthood. Am J Clin Nutr 1999;70(Suppl): 145S–148S.
Vol. 80, No. 1, July 2003