Serum and adipocyte resistin in polycystic ovary syndrome with insulin resistance

Kok-Min Seow1,2,6, Chi-Chang Juan2,3,4, Liang-Yi Wu4, Yung-Pei Hsu4, Wei-Ming Yang3, Yieh-Loong Tsai1, Jiann-Loung Hwang1 and Low-Tone Ho3,4,5,7

1 Department of Obstetrics and Gynecology, Shin Kong Wu Ho-Su Memorial Hospital, 2 Institute of Clinical Medicine and 3 Institute of Physiology, National Yang-Ming University, 4 Department of Medical Research & Education and 5 Department of Internal Medicine, Taipei Veterans General Hospital, Taipei, Taiwan and 6 School of Medicine, Fu-Jen Catholic University, Hsinchuang, Taipei Hsien, Taiwan

7 To whom correspondence should be addressed at: Department of Medical Research and Education, Taipei Veteran General Hospital, No. 201, Shih-Pai Road, Taipei, Taiwan. e-mail: ltho{at}vghtpe.gov.tw


    Abstract
 Top
 Abstract
 Introduction
 Materials and methods
 Reagents
 Results
 Discussion
 References
 
BACKGROUND: The aim of this study was to investigate the relationship between resistin and insulin resistance in patients with polycystic ovary syndrome (PCOS). METHODS: We compared serum resistin levels in 17 PCOS women and 10 lean, healthy, age-matched non-PCOS women and also compared levels of insulin receptor (IR), phosphatidylinositol-3 kinase (PI3-kinase), glucose transporter 4 (GLUT4) protein and resistin mRNA in adipocytes isolated from the omental adipose tissue of five of the PCOS patients and five age- and weight-matched, non-PCOS controls, to look for local defects in insulin action in PCOS. RESULTS: The PCOS group was hyperinsulinaemic and displayed an impaired insulin response in a 75 g oral glucose tolerance test and an abnormal homeostasis model insulin resistance index. Serum resistin levels were similar in PCOS patients and controls; however, resistin mRNA levels were 2-fold higher in adipocytes from PCOS patients. No correlation was found between serum resistin levels and either the BMI or testosterone levels. Western blot analysis showed that adipocyte levels of insulin receptor, PI3-kinase, and GLUT4 were respectively decreased by 56, 39.4 and 54% in PCOS patients compared with controls. CONCLUSIONS: These results suggest that overexpression of the resistin gene in adipocytes may be a local determinant factor in the pathogenesis of PCOS.

Key words: adipose tissue/LH/PCOS/resistin/signal transduction


    Introduction
 Top
 Abstract
 Introduction
 Materials and methods
 Reagents
 Results
 Discussion
 References
 
Polycystic ovary syndrome (PCOS) is a heterogeneous syndrome characterized by oligomenorrhoea or amenorrhoea, hyperandrogenism, and multiple small subcapsular cystic follicles in the ovary on ultrasonography (Franks, 1995Go). It affects ~5–10% of women of reproductive age (Dunaif, 1997Go), and ~16–80% of the affected women are obese (Dunaif, 1992Go). PCOS is frequently associated with insulin resistance accompanied by compensatory hyperinsulinaemia and obesity (Burghen et al., 1980Go). Insulin resistance is thought to play an important role in the aetiology of PCOS (Chang et al., 1983Go; Shoupe et al., 1983Go). In vitro and in vivo studies have shown that, in women with PCOS, the sensitivity of insulin to glucose metabolism is subnormal and that modest hyperinsulinaemia prevails (Dunaif et al., 1991Go). The administration of insulin sensitizing agents, such as metformin, may increase insulin sensitivity and thus induce ovulation (Nestler et al., 1998Go).

Several studies have investigated the possible cellular mechanism underlying insulin resistance in PCOS employing the major insulin target tissue, the adipocyte. Marsden et al. (1994Go) reported decreased numbers of insulin receptors and severe impairment of insulin action in adipocytes from amenorrhoeic women with PCOS. Ciaraldi (1997Go) showed that adipocytes from PCOS patients have low levels of adenosine. Rosenbaum et al. (1993Go) found that the decrease in the maximal rate of adipocyte glucose uptake is secondary to a significant decrease in GLUT4 glucose transporter protein. These results suggest that insulin resistance in PCOS represents post-binding defects in signal transduction and may involve defects in more than one step in the signalling pathway.

The hormone resistin, a novel 12.5 kDa cysteine-rich protein, also known as resistin-like molecule (RELM) or ‘found in inflammatory zone’ (FIZZ), is secreted by adipocytes (Steppan et al., 2001Go). Serum resistin levels are significantly increased in insulin-resistant mice and genetic or diet-induced obese mice (Steppan et al., 2001Go). In addition, neutralization of endogenous resistin with antibodies significantly suppresses hyperglycaemia in diet-induced obese mice by improving insulin sensitivity (Steppan et al., 2001Go). Furthermore, resistin gene expression in obese mice is down-regulated by thiazolidinedione (TZD) compounds, a new class of antidiabetic drugs with an insulin-sensitizing action. These results suggest that resistin may be an important link between increased fat mass and insulin resistance.

In an animal study, resistin mRNA levels were found to be decreased in insulin-resistant rats (Juan et al., 2001Go), while human studies have shown resistin mRNA levels to be very low in freshly isolated adipocytes and undetectable in pre-adipocytes, endothelial cells and vascular smooth muscle cells (Nagaev and Smith, 2001Go; Savage et al., 2001Go; Sentinelli et al., 2002Go). McTernan (2002Go) also reported that resistin mRNA was detectable in human adipocytes and that its levels were 4-fold higher in adipocytes from abdominal tissue depots than in those from the thigh. Thus, resistin may serve as a hormone link between central obesity and insulin resistance in humans and may regulate insulin sensitivity and glucose tolerance.

This study was performed to compare serum levels of resistin in women with PCOS and normal controls, and to detect its expression in human adipocytes using the semi-quantitative RT–PCR. We also examined whether the decreased adipocyte insulin response seen in PCOS patients might be attributable to local defects of insulin action in the signal transduction pathway, including defects in the insulin receptor, phosphatidylinositol 3-kinase (PI3-kinase), and the GLUT4 glucose transporter.


    Materials and methods
 Top
 Abstract
 Introduction
 Materials and methods
 Reagents
 Results
 Discussion
 References
 
Subjects
Seventeen women (10 obese and seven lean: obesity defined as BMI ≥ 27 kg/m2, lean as BMI < 27 kg/m2) who fulfilled the inclusion criteria for PCOS given below were enrolled in this study. All were in good health and had not taken oral contraceptives within the last 3 months. The protocol was reviewed and approved by the Institutional Review Board of both the Shin Kong Wu Ho-Su Memorial Hospital and Taipei Veteran General Hospital. All patients entered this study only after informed written consent was obtained.

PCOS was defined by clinical, laboratory and ultrasound criteria. The clinical criteria included oligomenorrhoea (menstrual interval >6 weeks) or amenorrhoea (no menstrual loss for ≥ 6 months) dating from menarche. All subjects were hirsute with a score of >10 on the Ferriman–Gallwey score (Ferriman and Gallwey, 1961Go). None of the subjects had acanthosis nigricans. The biochemical criteria were an increased LH concentration (> 6 mIU/ml, normal follicular range 1–6 mIU/ml), a normal FSH concentration, and elevated total serum testosterone levels (>0.8 ng/ml, normal range 0.06–0.86 ng/ml). The ultrasound criteria were enlarged ovaries with an increased stroma and >10 subcapsular follicles of 3–8 mm diameter, arranged peripherally around a dense core of stroma, as described by Adams et al. (1986Go). Serum prolactin and thyroid hormone levels were checked in all patients and were within the normal limits. Cushing’s syndrome and congenital adrenal hyperplasia were excluded by an overnight dexamethasone test (1 mg) and a morning serum 17-hydroxyprogesterone level of >1500 ng/dl.

Ten healthy, lean, age-matched women served as controls. None was hirsute, and all had a normal regular cycling menstrual period. None was taking oral contraceptives. All had a normal appearance of the ovaries on ultrasound and normal LH and FSH levels, and none had elevated androgen levels.


    Reagents
 Top
 Abstract
 Introduction
 Materials and methods
 Reagents
 Results
 Discussion
 References
 
Human monoclonal antibodies against human insulin receptor {beta}-subunit or PI3-kinase were purchased from Upstate Biotechnology Inc. Polyclonal rabbit anti-GLUT4 antibody was obtained from Biogenesis Inc. Human anti-{beta}-actin antibody was from Chemicon Inc. The resistin enzyme immunoassay (EIA) kit was purchased from Phoenix Pharmaceuticals Inc. (USA).

Oral glucose tolerance test and homeostasis model insulin resistance index
A 2 h oral glucose tolerance test (OGTT) with 75 g of glucose load was performed after an overnight fast during the early follicular phase (days 3–7) on all women. Four blood samples were collected at 0, 30, 60 and 120 min and the plasma stored at –20°C until assayed for glucose and insulin. The homeostasis model (HOMA) insulin resistance index was calculated using the formula: fasting glucose (mmol/l)xfasting insulin (µIU/ml)/22.5 (Matthews et al., 1985Go).

Hormonal profile
The hormonal profile (FSH, LH, E2 and testosterone) was measured in a blood sample collected on the same day as the OGTT. For the patients with amenorrhoea, 75 mg progesterone was given i.m. to induce withdrawal bleeding and their blood samples were also collected in the follicular phase.

Blood collection for resistin
Blood samples (7 ml) were collected in Lavender Vacutainer tubes containing EDTA. The blood was then transferred to a centrifuge tube containing aprotinin (0.6 TIU/ml of blood) and rocked several times to inhibit proteinease activity, then the sample was centrifuged at 1600 g for 15 min at 4°C and the plasma stored at –70°C until assayed for resistin. The plasma sample was diluted 2-fold with EIA buffer to perform the EIA.

Resistin assay
Resistin was measured in a fasting blood sample, collected on the same day as the OGTT, using an EIA kit with a lower limit of sensitivity of 1.21 ng/ml (range: 0–500 ng/ml) and intra- and inter-assay coefficients of variation (CV) of <5 and <14% respectively.

Adipocyte cell isolation
Adipose tissue (5–6 g) was obtained by laparoscopic surgery from the omental fat tissue of five controls undergoing gynaecological surgery and five PCOS patients undergoing ovarian drilling therapy. The adipose tissue was transported to the laboratory in Krebs–Ringer bicarbonate (KRB) buffer (in mmol/l: NaCl 118, MgSO4 1.2, CaCl2 1.3, NaHCO3 2.5, pH adjusted to 7.4 at 37°C) within 30 min and adipocytes isolated following the procedure described by Rodbell (1964Go) with minor modifications. The adipocyte tissue was finely minced, then disintegrated into cells by gentle shaking for 60 min at 37°C in KRB buffer containing 1 mmol/l pyruvate, 1% bovine serum albumin (BSA), and 0.1% collagenase (10 ml per g of tissue).

The cell suspension was then filtered through a nylon mesh (400 µmol/l) and centrifuged at 100 g for 1 min at room temperature. The supernatant was harvested and washed three times with 50 ml of KRB buffer containing 1 mmol/l pyruvate and 1% BSA. To count the cells and measure their size, an aliquot of diluted cell suspension was fixed overnight in collidine buffer containing 2% osmium tetroxide prior to microscopic photography. The mean cell volume and surface area were calculated from the known lipocrit of the cell suspension. Half of the adipocytes were used immediately for western blots and the rest stored at –70°C for mRNA extraction.

RNA extraction
Total RNA was extracted from the adipocytes isolated from each patient using a Tri Reagent kit (Molecular Research Center, Inc., USA). The integrity of the total RNA was examined by 1% agarose gel electrophoresis, and the RNA concentration was determined by UV absorbance at 260 nm (Genequant RNA/DNA calculator; Pharmacia, LKB Biochrom, UK).

Quantification of resistin mRNA
Levels of resistin and {beta}-actin mRNA in adipocytes were assessed by RT–PCR based on the known cDNA sequences for resistin and {beta}-actin; we designed the two pairs of primer listed in Table I. For RT, 5 µg of RNA template was incubated for 30 min at 37°C with 1 µl of RNase-free deoxyribonuclease (DNase) and 7 µl of H2O, followed by incubation for 10 min at 70°C with 1 µl of 25 mmol/l EDTA to inactivate the DNase, then 2 µl of 100 ng/µl of oligo(dT) primer (Promega, USA) and 7.1 µl of H2O were added and the sample incubated for 10 min at 70°C. RT was then carried out for 50 min at 42°C in a total volume of 50 µl containing 1xRT buffer, 10 mmo/l dithiothreitol, 0.5 mmol/l dNTP (HT Biotechnology Ltd), 16 IU of human placental ribonuclease inhibitor (HT Biotechnology Ltd), and 1 IU of Super RT reverse transcriptase (HT Biotechnology Ltd, UK). The RT mixtures were then heated for 3 min at 95°C to inactivate the reverse transcriptase. The resistin cDNA was amplified by touch-down PCR using a DNA Thermal Cycler (PTC-100; MJ Research, USA). The total volume of 25 µl contained 2 µl of the cDNA template, 500 nmol/l each of the sense and antisense primers, 0.625 IU of FailSafe Enzyme, and a final concentration of 1xFailSafe buffer D (both from Epicentre Technologies, USA). In a preliminary run, we found that a minimum of 30 PCR cycles was required to produce an optimal amount of nucleic acids for measuring on an agarose gel.


View this table:
[in this window]
[in a new window]
 
Table I. List of primers used for semiquantitative RT–PCR
 
As a reference for the quantitative gene expression, the housekeeping gene {beta}-actin was simultaneously amplified under conditions identical to those for resistin PCR. For example, to determine the relative quantities of resistin mRNA, 5 µl each of resistin and {beta}-actin PCR products amplified from the same RT template solution were combined and electrophoresed on a 2% agarose gel in 0.04 mol/l Tris–acetate/0.001 mol/l EDTA buffer for 40 min. After staining the gel with ethidium bromide (0.5 µg/ml) for 15 min, the resistin and {beta}-actin cDNA bands were measured by densitometry, using Image Master VDS and Image Quant Analysis Software (Amersham Pharmacia Biotech, Hong Kong).

Western blots
Homogenates of isolated adipocytes were centrifuged at 6000 g for 10 min at –4°C, then the proteins (50 µg) in the supernatant were separated by 7.5 or 10% sodium dodecylsulphate polyacrylamide gel electrophoresis (SDS–PAGE), and transferred to a polyvinylidene fluoride membrane. The membrane was blocked with 5 % skimmed milk in phosphate-buffered saline, for 1 h at room temperature, then reacted with antibodies against the insulin receptor {beta}-subunit, PI3-kinase, or GLUT4, followed by secondary antibody (Jackson Immunoresearch), and enhanced chemiluminescence (ECL) reagent. Chemiluminescence were quantified using a Personal Densitometer (Molecular Dynamics, USA) with a range of 0.01–4.0 OD units.

Statistics
All results are expressed as the mean ± SD. Differences between the PCOS and control groups were compared by Student’s t-test for independent samples. Correlations between variables were evaluated by Pearson’s correlation coefficient and regression analysis. P < 0.05 was considered significant.


    Results
 Top
 Abstract
 Introduction
 Materials and methods
 Reagents
 Results
 Discussion
 References
 
As shown in Table II, the ages of the patients with PCOS ranged from 22 to 37 years (28.9 ± 5.10, mean ± SD) and those of the control group from 19 to 32 years (mean 25.4 ± 4.33; no significant difference). The body mass index values for the PCOS and control groups were 26.7 ± 4.6 and 19.5 ± 1.27 kg/m2 respectively (P < 0.0001).


View this table:
[in this window]
[in a new window]
 
Table II. Clinical and endocrine characteristics of women with polycystic ovary syndrome (PCOS) and controls
 
Insulin resistance in PCOS
Table II shows that, although fasting insulin levels were higher in our PCOS patients than in the controls, the difference was not statistically significant. In addition, there was no difference in fasting glucose levels. Nevertheless, after a 75 g oral glucose challenge, the 2 h insulin levels increased from 6.1 ± 3.94 to 24.3 ± 12.15 mIU/ml in the PCOS patients, a highly significant difference compared to the normal controls. Forty-seven per cent of the PCOS patients fulfilled World Health Organization (1985Go) criteria for impaired glucose tolerance on the basis of serum glucose levels at 2 h after a 75 g glucose load. In addition, 52.9% of the PCOS women showed insulin resistance (HOMA >2.7), which was confirmed by western blot analysis showing that adipocyte levels of insulin receptor, PI3-kinase and GLUT4 were decreased by 56, 39.4 and 55.4% respectively in the PCOS patients compared with the controls (Figure 1).



View larger version (79K):
[in this window]
[in a new window]
 
Figure 1. Western blot analysis of adipocytes from women with polycystic ovary syndrome and controls. Blots were probed with antibody against human insulin receptor {beta}-subunit (a), phosphatidylinositol-3 kinase (b), or glucose transporter 4 (c). Left: western blot; right quantitative data. The experiment was performed twice with similar results.

 
Gonadotrophins and sex steroids in PCOS
The basal serum hormone levels in women with PCOS and the control group are shown in Table II. Mean plasma LH levels, testosterone levels, and estradiol levels were all significantly higher in the PCOS group.

Correlations between variables
Both fasting insulin and glucose levels were significant negatively correlated with the FSH or LH levels. Correlations were found between estradiol levels and fasting insulin levels, 2 h glucose levels, or 2 h insulin levels (P < 0.05). No correlations were found between testosterone levels and fasting glucose levels, fasting insulin levels, 2 h glucose levels, or 2 h insulin levels. No correlation was found between serum resistin levels and BMI (R = 0.184, P = 0.479), testosterone levels (R = 0.022, P = 0.934), fasting insulin levels (R = 0.153, P = 0.557), or 2 h insulin levels (R = 0.324, P = 0.204).

Serum resistin levels and resistin gene expression in adipocyte
No difference was found between serum resistin levels in the two groups (P = 0.601). However, resistin mRNA levels were 2-fold higher in adipocytes from PCOS patients than in those from normal controls (Resistin/{beta}-actin: control 0.3; PCOS: 0.6; Figure 2).



View larger version (35K):
[in this window]
[in a new window]
 
Figure 2. (a) RNA extracted from the omental adipose tissue of women with polycystic ovary syndrome and normal controls; 28S and 18S RNA are shown. (b) Adipocyte resistin gene expression (412 bp) in the two groups. (c) Human {beta}-actin (528 bp) gene expression in the two groups.

 

    Discussion
 Top
 Abstract
 Introduction
 Materials and methods
 Reagents
 Results
 Discussion
 References
 
In the present study, we found that adipocytes from women with PCOS had a lower content of insulin receptor, PI3-kinase, and GLUT4 glucose transporter. This suggests that insulin resistance in PCOS represents post-binding defects in signal transduction and that there are multiple defects in insulin action in PCOS that affect metabolism (Stuart et al., 1986Go; Freidenberg et al., 1988Go; Rosenbaum et al., 1993Go; Marsden et al., 1994Go; Ciaraldi et al., 1997Go). These results differ from those of Book (1999Go) who reported a defect in metabolic, but not mitogenic, signalling pathways in PCOS fibroblasts, with no difference in insulin receptor or PI3K activity between fibroblasts from PCOS patients and controls. However, the number of PCOS patients in which the cellular aspects were examined in the present study was small and a further study is required.

The combined results of the clinical diagnosis methods (OGTT and HOMA index) and the western blotting experiments showed that the PCOS patients were insulin-resistant. Although fasting insulin levels were higher in the PCOS group, the difference was not statistically significant. After challenge with 75 g of oral glucose, both the 2 h insulin and 2 h glucose levels were significantly higher in the PCOS group than in the controls, showing that the PCOS women were hyperinsulinaemic with glucose intolerance. Taking into account the additional evidence of the HOMA insulin resistance index and the signal transduction pathway defects in adipocytes from PCOS patients, we conclude that our patients were insulin-resistant. In fact, western blotting analysis may be the most accurate test for identifying insulin-resistant PCOS patients for targeted treatment with insulin-sensitizing agents.

Resistin has been described as a potential link between obesity and insulin resistance (Steppan et al., 2001Go) and resistin mRNA levels have been shown to be increased in 3T3-L1 cells and to be down-regulated by antidiabetic thiazolidinediones (TZD) drugs, which bind to peroxisome proliferator activated receptor-{gamma} (PPAR-{gamma}) in fat cells (Steppan et al., 2001Go). Resistin might therefore be expected to play a role in PCOS, and serum levels in these patients might be expected to be elevated. However, we found that serum resistin levels in PCOS did not differ from those in normal cycling women of a similar age and with a similar BMI. It should be noted that none of the 17 PCOS women studied displayed hyper-resistinaemia. Furthermore, serum resistin levels did not correlate with either fasting insulin or 2 h insulin levels in the PCOS group. These results may indicate that serum resistin levels did not affect insulin sensitivity in PCOS. However, a difference was seen in resistin mRNA levels in adipocytes, these being twice as high in PCOS patients as in controls, suggesting that resistin may have a local paracrine action in the regulation of insulin resistance in obesity and PCOS.

The poor correlation found between serum resistin levels and BMI in this study conflicts with the report by Steppan (2001Go) of increased levels of serum resistin in ob/ob and db/db mice. Steppan’s observation is inconsistent with the aforementioned expectation that insulin resistance should correlate positively with increased serum resistin levels. Furthermore, adipocyte resistin mRNA levels were increased in our women with PCOS, who had high testosterone levels. This agrees with the results of Ling et al. (2001Go) who showed that resistin mRNA levels are increased 2.6-fold in control male mice with elevated testosterone levels. Testosterone administration to obese males can improve insulin sensitivity (Marin et al., 1992Go), but may produce hyperinsulinaemia in women (Nestler et al., 1988Go; Mortola and Yen, 1990Go).

Women with PCOS had significantly higher estradiol levels than normal controls. This result fits with the hypothesis proposed by Fox et al. (1991Go) that estrogenization is a key feature of PCOS. Furthermore, there was a strong association between estradiol levels and fasting insulin levels or 2 h insulin levels. This association is of special interest because the estradiol levels could be a surrogate marker for hyperinsulinaemic insulin resistance in PCOS. The increased estradiol levels are thought to be due to the combined secretion of estradiol and estrone by numerous follicles, resulting in substantial release (Wajchenberg et al., 1988Go) and the peripheral aromatization of circulating androgens in adipose tissue (Marsden et al., 1999Go).

The modest hyperandrogenism characteristic of PCOS has been proposed to be associated with insulin resistance (Dunaif et al., 1991Go; Rittmaster et al., 1993Go). However, androgen administration to healthy females does not produce insulin resistance of the same magnitude as that seen in PCOS (Polderman et al., 1994Go). In our study group, we did not find any correlation between testosterone levels and fasting or 2 h insulin levels. These results demonstrate that testosterone levels are of limited use in predicting insulin resistance in PCOS.

In conclusion, we have shown that insulin resistance in PCOS women involves both receptor and post-receptor defects, including defects in PI3-kinase and the GLUT4 glucose transporter. Serum resistin levels in PCOS women were similar to those in the control group. Nevertheless, the increased levels of resistin mRNA in PCOS adipocytes suggest that resistin may be a determinant factor in the local pathogenesis of PCOS. Estradiol, serum levels of which were significantly elevated in the PCOS group, could be a surrogate marker for hyperinsulinaemic insulin resistance in PCOS.


    Acknowledgements
 
This study was supported by grants from the Shin Kong Wu Ho-Su Memorial Hospital (No. 8302-92-2477) and the Taipei Veteran General Hospital (No. VGH 92-C329).


    References
 Top
 Abstract
 Introduction
 Materials and methods
 Reagents
 Results
 Discussion
 References
 
Adams J, Polson DW and Franks S (1986) Prevalence of polycystic ovaries in women with anovulation and idiopathic hirsutism. Br Med J 293,355–359.[ISI][Medline]

Burghen GA, Givens, JR and Kitabchi, AE (1980) Correlation of hyperandrogenism with hyperinsulinism in polycystic ovarian disease. J Clin Endocrinol Metab 50,113–116.[Abstract]

Book CB and Dunaif A (1999) Selective insulin resistance in the polycystic ovary syndrome. J Clin Endocrinol Metab 84,3110–3116.[Abstract/Free Full Text]

Chang RJ, Nakamura RM, Judd HL and Kaplan SA (1983) Insulin resistance in nonobese patients with polycystic ovarian disease. J Clin Endocrinol Metab 57,356–359.[Abstract]

Ciaraldi TP, Morales AJ, Hickman MG, Odom-Ford R, Olefsky JM and Yen SS (1997) Cellular insulin resistance in adipocytes from obese polycystic ovary syndrome subjects involves adenosine modulation of insulin sensitivity. J Clin Endocrinol Metab 82,1421–1425.[Abstract/Free Full Text]

Collison M, Campbell IW, Salt IP, Dominiczak AF, Connell JM, Lyall H and Gould GW (2000) Sex hormones induce insulin resistance in 3T3-L1 adipocytes by reducing cellular content of IRS proteins. Diabetologia 43,1374–1380.[CrossRef][ISI][Medline]

Dunaif A (1992) Polycystic ovary syndrome and obesity. In Bjorntorp P, Brodoff BN (eds) Obesity. JB Lippincott Co, Philadelphia, 594 pp.

Dunaif A (1997) Insulin resistance and the polycystic ovary syndrome: mechanism and implications for pathogenesis. Endocr Rev 18,774–800.[Abstract/Free Full Text]

Dunaif A, Green G, Phelps RG, Lebwohl M, Futterweit W and Lewy L (1991) Acanthosis nigricans, insulin action and hyperandrogenism: clinical, histological and biochemical findings. J Clin Endocrinol Metab 73,590–595.[Abstract]

Ferriman D and Gallwey J (1961) Clinical assessment of body hair growth in women. J Clin Endocrinol Metab 21,1440–1445.[ISI]

Fox R, Corrigan E, Thomas PG and Hull MG (1991) Oestrogen and androgen states in oligo-amenorrhoeic women with polycystic ovaries. Br J Obstet Gynecol 98,294–299.[ISI][Medline]

Franks S (1995) Polycystic ovary syndrome. N Engl J Med 333,853–861.[Free Full Text]

Freidenberg GR, Reichart D, Olefsky JM and Henry RR (1988) Reversibility of defective adipocyte insulin receptor kinase activity in non-insulin-dependent diabetes mellitus. Effect of weight loss. J Clin Invest 82,1398–1406.[ISI][Medline]

Juan CC, Au LC, Fang VS, Kang SF, Ko YH, Kuo SF, Hsu YP, Kwok CF and Ho LT (2001) Suppressed gene expression of adipocyte resistin in an insulin-resistant rat model probably by elevated free fatty acids. Biochem Biophys Res Commun 289,1328–1333.[CrossRef][ISI][Medline]

Ling C, Kindblom J, Wennbo H and Billig H (2001) Increased resistin expression in the adipose tissue of male prolactin transgenic mice and in male mice with elevated androgen levels. FEBS Lett 507,147–150.[CrossRef][ISI][Medline]

Marin P, Krotkiewski M and Bjorntorp P (1992) Androgen treatment of middle-aged, obese men: effects on metabolism, muscle and adipose tissues. Eur J Med 6,329–336.

Marsden PJ, Murdoch AP and Taylor R (1994) Severe impairment of insulin action in adipocytes from amenorrheic subjects with polycystic ovary syndrome. Metabolism 43,1536–1542.[ISI][Medline]

Marsden PJ, Murdoch AP and Taylor R (1999) Adipocyte insulin action following ovulation in polycystic ovarian syndrome. Hum Reprod 14,2216–2222.[Abstract/Free Full Text]

Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF and Turner RC (1985) Homeostasis model assessment: insulin resistance and [beta]-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia 28,412–419.[ISI][Medline]

McTernan CL, McTernan PG, Harte AL, Levick PL, Barnett AH and Kumar S (2002) Resistin, central obesity and type 2 diabetes. Lancet 359,46–47.[CrossRef][ISI][Medline]

Mortola JF and Yen SS (1990) The effects of oral dehydroepiandrosterone on endocrine-metabolic parameters in postmenopausal women. J Clin Endocrinol Metab 71,696–704.[Abstract]

Nagaev I and Smith U (2001) Insulin resistance and type 2 diabetes are not related to resistin expression in human fat cells or skeletal muscle. Biochem Biophys Res Commun 285,561–564.[CrossRef][ISI][Medline]

Nestler JE, Barlascini CO, Clore JN and Blackard WG (1988) Dehydroepiandrosterone reduces serum low density lipoprotein levels and body fat but does not alter insulin sensitivity in normal men. J Clin Endocrinol Metab 66,57–61.[Abstract]

Nestler JE, Jakubowicz DJ, Evans WS and Pasquali R (1998) Effects of metformin on spontaneous and clomiphene-induced ovulation in the polycystic ovary syndrome. N Engl J Med, 338,1876–1880.[Abstract/Free Full Text]

Polderman KH, Gooren JG, Asscherman H, Bakker A and Heine RJ (1994) Induction of insulin resistance by androgens and estrogens. J Clin Endocrinol Metab 79,265–271.[Abstract]

Rittmaster RS, Deshwal N and Lehman L (1993) The role of adrenal hyperandrogenism, insulin resistance and obesity in the pathogenesis of polycystic ovarian syndrome. J Clin Endocrinol Metab 76,1295–1300.[Abstract]

Rodbell M (1964) Metabolism of isolated fat cell. I. Effects of hormone on glucose metabolism and lipolysis. J Biol Chem 239,375–380.[Free Full Text]

Rosenbaum D, Haber RS and Dunaif A (1993) Insulin resistance in polycystic ovary syndrome: decreased expression of GLUT-4 glucose transporters in adipocytes. Am J Physiol 264,E197–E202.[ISI][Medline]

Savage DB, Sewter CP, Klenk ES, Segal DG, Vidal-Puig A, Considine RV and O’Rahilly S (2001) Resistin/FIZZ3 expression in relation to obesity and peroxisome proliferator-activated receptor-[gamma] action in humans. Diabetes 50,2199–2202.[Abstract/Free Full Text]

Sentinelli F, Romeo S, Arca M, Filippi E, Leonetti F, Banchieri M, Di Mario U and Baroni MG (2002) Human resistin gene, obesity and type 2 diabetes: mutation analysis and population study. Diabetes 51,860–862.[Abstract/Free Full Text]

Shoupe D, Kumar DD and Lobo RA (1983) Insulin resistance in polycystic ovary syndrome. Am J Obstet Gynecol 147,588–592.[ISI][Medline]

Steppan CM, Bailey ST, Bhat S, Brown EJ, Banerjee RR, Wirght CM, Patel HR, Ahima RS and Lazar MA (2001) The hormone resistin links obesity to diabetes. Nature 409,307–312.[CrossRef][ISI][Medline]

Stuart CA, Peters EJ, Prince MJ, Richards G, Cavallo A Meyer WJ 3rd (1986) Insulin resistance with Acanthosis nigricans: the role of obesity and androgen excess. Metabolism 35,197–205.[ISI][Medline]

Wajchenberg BL, Achando SS, Mathor MM, Czeresnia CE, Giannella Neto D and Kirschner MA (1988) The source(s) of estrogen production in hirsute women with polycystic ovarian disease as determined by simultaneous adrenal and ovarian venous catheterization. Fertil Steril 49,56–61.[ISI][Medline]

World Health Organization (1985) Diabetes Mellitus: Report of a WHO Study Group. Technical Report Series, no. 727. World Health Organization, Geneva.

Submitted on June 2, 2003; resubmitted on August 12, 2003; accepted on September 19, 2003.