Elevated circulating levels of irisin and the effect of metformin treatment in women with polycystic ovary syndrome.

ORIGINAL

ARTICLE

Elevated Circulating Levels of Irisin and the Effect of Metformin Treatment in Women With Polycystic Ovary Syndrome Minyan Li, Mengliu Yang, Xiaoxin Zhou, Xia Fang, Wenjing Hu, Wei Zhu, Cong Wang, Dongfang Liu, Shengbing Li, Hua Liu, Gangyi Yang, and Ling Li Key Laboratory of Diagnostic Medicine (Ministry of Education) and Department of Clinical Biochemistry (M.L., L.L.), College of Laboratory Medicine, Chongqing Medical University, 400010 Chongqing, China; Department of Endocrinology (M.Y., X.Z., X.F., W.H., W.Z., C.W., D.L., S.L., G.Y.), The Second Affiliated Hospital, Chongqing Medical University, 400010 Chongqing, China; and Department of Pediatrics (H.L.), University of Mississippi Medical Center, 2500 North State Street, Jackson, Mississippi 39216-4505

Context: Polycystic ovary syndrome (PCOS) is an insulin resistance (IR) state, like obesity and type 2 diabetes mellitus (T2DM). Although previous studies have suggested a correlation between irisin and the metabolic parameters associated with obesity and T2DM, the results have been inconsistent. Objective: Our objective was to (1) determine circulating irisin levels in women with PCOS and control subjects, (2) examine the relationship of irisin and conventional markers of insulin resistance, and (3) examine irisin changes with interventions modulating IR in PCOS women. Patients and Design: This study was comprised of a series of cross-sectional and interventional studies of 178 PCOS and 123 healthy women from the general population and outpatients of the Internal Medicine Department at the Second Affiliated Hospital, Chongqing Medical University, China. Forty seven women with PCOS were randomly assigned to 6 months of oral metformin (850 mg bid). The oral glucose tolerance test (OGTT) and the euglycemic-hyperinsulinemic clamp (EHC) were performed to assess glucose tolerance and insulin sensitivity. Outcome measures were IR (AUCInsulin and M values) on an OGTT and EHC, irisin levels, and metabolic markers. Results: Circulating irisin was significantly higher in both overweight/obese (body mass index [BMI] ⱖ 25 kg/m2) and PCOS women (P ⬍ .01). Circulating irisin levels correlated with BMI, WHR, FAT%, total cholesterol (TC), triglyceride (TG), low-density lipoprotein cholesterol (LDL-C), AUCInsulin, homeostasis model assessment of insulin resistance (HOMA2-IR), M values, and free androgen index (FAI). During EHC, short-term hyperinsulinemia exhibited an inhibitory effect on irisin levels. After 6 months of metformin treatment, there was a significant decrease in circulating irisin in PCOS women following improved IR. Conclusions: These data suggest that irisin may be a useful marker of IR in PCOS women. (J Clin Endocrinol Metab 100: 1485–1493, 2015)

P

olycystic ovary syndrome (PCOS), a common endocrinopathy affecting 5–10% of women of reproductive age, is characterized by menstrual dysfunction and hyperandrogenism and is associated with insulin resis-

tance and pancreatic ␤-cell dysfunction, impaired glucose tolerance (IGT), type 2 diabetes mellitus (T2DM), dyslipidemia, and visceral obesity (1, 2). The consequent hyperinsulinemia is more prevalent in lean and obese women

ISSN Print 0021-972X ISSN Online 1945-7197 Printed in U.S.A. Copyright © 2015 by the Endocrine Society Received June 4, 2014. Accepted February 6, 2015. First Published Online February 12, 2015

Abbreviations: AUC Insulin, the area under the curve for insulin; BMI, body mass index; EHC, euglycemic-hyperinsulinemic clamp; FAI, free androgen index; FBG, fasting blood glucose; HOMA2-IR, homeostasis model assessment of insulin resistance; LDL-C, low-density lipoprotein cholesterol; OGTT, oral glucose tolerance test; PCOS, polycystic ovary syndrome; T2DM, type 2 diabetes mellitus; TC, total cholesterol; TG, triglyceride; WHR, waist-to-hip ratio.

doi: 10.1210/jc.2014-2544

J Clin Endocrinol Metab, April 2015, 100(4):1485–1493

jcem.endojournals.org

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 11 December 2016. at 03:45 For personal use only. No other uses without permission. . All rights reserved.

1485

1486

Li et al

Irisin and the Effect of Metformin in PCOS

with PCOS when compared with age-matched control women (3). Insulin resistance is found in most obese women with PCOS and in a significant proportion (30%) of lean women with the syndrome (4). Although the etiology of the syndrome remains enigmatic, the potential influence of environmental factors, insulin resistance and cytokine on PCOS development has recently been explored (5– 8). Recently, a novel myokine, named irisin, was identified in mice and human (9). Irisin is expressed in muscle as the type I membrane precursor protein FNDC5, which is proteolytically cleaved and secreted into the circulatory system. Exogenous administration of irisin using adenoviral delivery in mice triggered the development of a brownfat-like substance in specific deposits of white adipose tissue (WAT), and resulted in increased energy expenditure, improved glucose tolerance, and a modest but significant weight loss (9). More recently, some studies have reported that irisin levels are lower in the overweight, obese, and T2DM patients (10, 11). However, one study reported that irisin levels were significantly higher in gestational diabetes mellitus (GDM) patients as compared to controls (12). Irisin levels were also reported to positively correlate with body mass index (BMI) in some studies (11, 13, 14), although other studies reported a null (15) or even a negative correlation (10). Moreover, in nondiabetic individuals, circulating irisin correlated positively with age, BMI, total cholesterol (TC), triglycerides (TG), fasting blood glucose (FBG), and diastolic blood pressure (BP) (DBP) (16). Although previous human studies have suggested a correlation between irisin and metabolic parameters associated with obesity and T2DM, the results are inconsistent. Therefore, the relationship of circulating levels of irisin and insulin resistance in humans needs more indepth exploration. Of particular importance, the regulation of irisin in humans has yet to be demonstrated. In fact, to the best of our knowledge, no information about the relationship of circulating irisin with PCOS, the dynamic changes in irisin levels after therapeutic intervention, or acute hyperinsulinaemia in humans is available. Because PCOS is an insulin-resistance state without age-related impact factors, the aim of the current study is to investigate whether circulating irisin levels correlate with insulin resistance and whether a therapeutic intervention that improves insulin sensitivity is associated with an alteration in circulating irisin levels in tandem with associated changes to clinical, hormonal, and metabolic parameters in women with PCOS. Additionally, we also examined whether circulating irisin was affected by hyperinsulinemia during a euglycemic-hyperinsulinemic clamp (EHC, a hyperinsulinemic-euglycemic state). Fi-


nally, we examined the effect of metformin treatment on circulating irisin levels in women with PCOS.

Research Design and Methods Subjects One hundred seventy eight patients with PCOS were recruited for this study from outpatients attending the Internal Medicine Department at the Second Affiliated Hospital, Chongqing Medical University, Chongqing, China. All PCOS patients met all three criteria of the revised 2003 Rotterdam European Society of Human Reproduction and Embryology (ESHRE)/American Society of Reproductive Medicine (ASRM) PCOS Consensus Workshop Group diagnostic criteria (17). The three criteria are (1) oligo- and/or anovulation, (2) clinical and/or biochemical signs of hyperandrogenism, and (3) ultrasound appearance of polycystic ovaries with the exclusion of other known causes of hyperandrogenemia and ovulatory dysfunction including 21-hydroxylase deficiency, congenital adrenal hyperplasia, Cushing’s syndrome, and rogensecreting tumors, thyroid disease, and hyperprolactinemia. One hundred twenty three healthy women with regular periods and no hyperandrogenemia, hirsutism, or acne were recruited as the control group from the community or schools through advertisement, or routine medical check-up. Exclusion criteria for both groups included age ⬎35 years, BMI ⬎ 35 kg/m2, known cardiovascular disease, thyroid disease, neoplasms, smoking, prediabetes (including impaired glucose tolerance and impaired fasting glucose), diabetes, hypertension, and renal impairment (serum creatinine 120 ␮mol/L). None of PCOS and healthy women were on any hormone medications and medicines that affect insulin sensitivity within the past 6 months. All subjects gave their written informed consent before entering the study, which was conducted in accordance with the Declaration of Helsinki and approved by the national ethical committee.

Interventional studies A subset of 54 patients from the PCOS group received metformin treatment. Patients were excluded from participation if they were pregnant or planning to become pregnant. All participants gave their informed written consent about the side effects of metformin at the onset of interventional studies. Therapy was initiated after basal assessment and the dose of metformin was increased to a maintenance dose of 850 mg twice daily for 24 weeks. Only 47 women were investigated after 6 months of metformin treatment. The reasons for 7 subjects not completing study were gastrointestinal (GI) side effects, pregnancy, incompliance, and loss of contact. Women included were closely followed up for the period of the study. These women were also followed up during pregnancy for monitoring teratogenicity. Although no specific diet or exercise regimen was advised for this study, in line with our clinical practice, all women were informed about the relationship between PCOS, body weight, and insulin sensitivity and given standard advice concerning the beneficial effects of lifestyle modifications. All patients underwent anthropometric measurements before and after metformin treatment. Fasting blood samples for irisin measurements were obtained at 0800 hours on day 0 pretreatment, on week 12 and on day 2 after the last admission (24 weeks).


doi: 10.1210/jc.2014-2544


1487

At 0800 hours on the study days, after an 8 –10 hour overnight fast, an OGTT was performed on healthy and PCOS women. These subjects ingested 75 g glucose, and venous blood was drawn at 0, 30, 60, and 120 minutes for measurement of glucose and insulin.

No. EK-067-29) were evaluated using the parts of same sample. The values measured in the same subjects using the Phoenix Pharmaceuticals kit were slightly lower than those using the CUSABIO kit, but no statistical significance. Therefore, we have concluded that both Phoenix Pharmaceuticals and CUSABIO kits were appropriate for the analysis of clinical samples.

Euglycemic-hyperinsulinemic clamp

Anthropometric measurements

The EHC was performed in 178 women with PCOS and 123 healthy women as previously described (18, 19). Briefly, after an overnight fast, an IV catheter was placed in the antecubital vein to infuse insulin and glucose. Another catheter was placed retrograde in the dorsal vein of the contralateral hand for blood withdrawal. Regular human insulin (1 mU/kg minute) was infused for 2 h, and a variable infusion of 20% glucose was administered to maintain plasma glucose at the fasting level. During the procedure, plasma glucose levels were measured every 10 minutes to guide the glucose infusion. The rate of glucose disposal was defined as the glucose infusion rate (GIR) during the stable period of the clamp and was related to body weight (M value). Blood samples for irisin measurements were obtained at 0, 80, 100, 110, and 120 minutes. The samples were immediately cooled, and serum/plasma was prepared within 1 h and stored at ⫺80°C until assayed.

BMI was calculated as weight divided by height squared. The percentage of body fat (FAT %) was measured by bioelectrical impedance (BIA-101; RJL Systems). The homeostasis model assessment of insulin resistance (HOMA2-IR) was calculated using computer software (HOMA Calculator v2.2.2) (20). The area under the curve (AUC) for insulin (AUCInsulin) during the OGTT was calculated geometrically using the trapezoidal rule.

Oral glucose tolerance test (OGTT)

Biochemical and hormonal analysis Plasma glucose and HbA1c were measured by the glucose oxidase method and anion exchange high-performance liquid chromatography (HPLC), respectively. Insulin was measured by radioimmunoassay (RIA) using human insulin as a standard (Institute of Atomic Energy, China). Free fatty acid (FFA) was measured with a commercial kit (Randox Laboratories). TC, high-density lipoprotein cholesterol (HDL-C), low-density lipoprotein cholesterol (LDL-C), and triglyceride (TG) were analyzed enzymatically using an autoanalyzer (Hitachi). Serum hormonal concentrations including the luteinizing hormone (LH), follicle-stimulating hormone (FSH), testosterone and progesterone (Prog) were measured with well-established electrochemiluminescence immunoassay using COBAS E immunoassay analyzers (Roche Diagnostics GmbH). Total testosterone levels were measured with the coated tube RIA (DiaSorin, S. p. A, Salluggia, Italy and Diagnostic Products Corporation, respectively). Dehydroepiandrostenedione sulfate (DHEA-S) and the sex hormone binding globulin (SHBG) were performed using an automated analyzer (Abbott Architect; Abbott Laboratories). The free androgen index (FAI) was calculated as FAI ⫽ (testosterone/SHBG) ⫻ 100.

Measurements of plasma irisin Plasma irisin concentrations were measured in duplicate at the same time with assay kits (CUSABIO Life Science, Inc.). This test is an enzyme immunometric assay based on standard 96-well micrometer plates, and the sensitivity of the assay was 0.78 ng/ mL. In accordance with the manufacturer’s instructions, the ranges of the intra- and interassay coefficients of variation were 5.5–7.8 and 4.3– 8.2%, respectively. This kit has a measurement range of 3.12–200 ng/mL irisin. The assay has high sensitivity and excellent specificity for detection of human irisin with no significant cross-reactivity or interference. In addition, irisin ELISA kits from the CUSABIO (Catalog No. CSB- EQ027943HU) and Phoenix Pharmaceuticals (Catalog

Statistical analysis All analyses were performed with Statistical Package for Social Sciences version 15.0 (SPSS). Data were expressed as mean ⫾ SD. Comparisons between groups were performed by ANOVA, unpaired t test, or paired t test as appropriate. The association of irisin with PCOS was examined by multivariate logistic regression analysis. The distribution of irisin in the pooled data was further divided into tertiles, and the significant trends across increasing tertiles were estimated by row mean scores and the Cochran-Armitage trend test. In all statistical tests, P values ⬍.05 were considered as significant.

Results Clinical, hormonal, anthropometric, and metabolic parameters in normal and PCOS women Baseline clinical, anthropometric, and endocrine characteristics were listed in Table 1. The control and PCOS groups were similar in age, diastolic BP (DBP), FFA, DHEA-S, FSH, and Prog. BMI, waist-to-hip ratio (WHR), FAT%, BP, TG, TC, HDL-C, LDL-C, FBG, 2 h postglucose load blood glucose (2hPBG), HbA1c, LH, testosterone, and FAI were higher, whereas SHBG was lower in the PCOS women than in the controls (Table 1). In addition, four markers of insulin resistance, HOMA2-IR, FIns, 2-h plasma insulin after glucose overload (2hIns) and AUCinsulin were also higher in the PCOS women (Table 1). Circulating irisin level and its association with anthropometric and biochemical parameters in PCOS and normal subjects As shown in Figure 1A, overweight/obese women (BMI ⱖ 25 kg/ m2) had significantly higher circulating irisin levels than lean individuals in both normal and PCOS women (BMI ⬍ 25 kg/m2, both P ⬍ .01). Importantly, fasting circulating levels of irisin were significantly higher in PCOS women than in normal women (194.7 ⫾ 90.0 vs 168.6 ⫾ 70.4 ␮g/L; P ⬍ .01, Figure 1B). We next investigated the relationship between circulating irisin and


1488

Li et al

Table 1.



Clinical, Hormonal, and Metabolic Features of Women with PCOS and Controls

Variable

Controls (n ⴝ 123)

PCOS (n ⴝ 178)

Age (yr) BMI (kg/m2) WHR FAT (%) SBP (mmHg) DBP (mmHg) TG (mmol/liter) TC (mmol/liter ) HDL-C (mmol/liter) LDL-C (mmol/liter) FFA (umol/liter) FBG (mmol/liter) 2hPBG (mmol/liter) FIns (mU/liter) 2hIns (mU/liter) HBA1C AUCinsulin (mU*h/liter) HOMA2-IR DHEA-S (␮g/dl) LH (IU/liter) FSH (IU/liter) Prog (nmol/liter) SHBG (nmol/liter) TEST (nmol/liter ) FAI

25.7 ⫾ 2.3 20.5 ⫾ 2.6 0.80 ⫾ 0.06 26.8 ⫾ 5.4 108.6 ⫾ 9.0 73.1 ⫾ 8.7 0.78 (0.58 –1.20) 3.78 ⫾ 1.00 1.19 ⫾ 0.32 2.15 ⫾ 0.85 0.54 ⫾ 0.26 4.48 ⫾ 0.48 5.36 ⫾ 1.04 7.03 (6.10 – 8.71) 37.73 (23.06 –59.09) 5.15 ⫾ 0.25 92.85 (53.83–136.88) 1.00 (0.86 –1.27) 201.0 ⫾ 86.1 4.85 ⫾ 2.83 8.03 ⫾ 1.90 2.6 ⫾ 1.2 63.24 ⫾ 25.59 1.84 ⫾ 0.73 2.53 (1.83– 4.47)

26.1 ⫾ 4.5 24.8 ⫾ 4.4b 0.86 ⫾ 0.07b 35.6 ⫾ 8.9b 115.8 ⫾ 10.8b 76.8 ⫾ 7.0a 1.31 (0.92–1.92)b 4.49 ⫾ 1.02b 1.32 ⫾ 0.59a 2.62 ⫾ 0.79b 0.59 ⫾ 0.33 4.95 ⫾ 0.60b 7.30 ⫾ 1.98b 16.03 (9.11–21.42)b 115.65 (67.76 –173.90)b 5.43 ⫾ 0.33b 215.85 (152.42–315.15)b 2.29 (1.32–3.16)b 218.9 ⫾ 97.7 9.80 ⫾ 6.18b 7.85 ⫾ 3.08 2.9 ⫾ 1.6 44.82 ⫾ 32.30b 2.71 ⫾ 1.14b 7.51 (4.05–10.80)b

PCOS, polycystic ovary syndrome; BMI, body mass index; WHR, waist-to-hip ratio; FBG, fasting blood glucose; 2hPBG, 2 h post-glucose load blood glucose; SBP, systolic blood pressure; DBP, diastolic blood pressure; FIns, fasting plasma insulin; 2-hIns, 2-h plasma insulin after glucose overload; FAT%, visceral fat percentage; FFA, free fatty acid; TG, triglyeride; TC, total cholesterol; HDL-C, high-density lipoprotein cholesterol; LDL-C, lowdensity lipoprotein cholesterol; HOMA2-IR, HOMA-insulin resistance index; M, whole body glucose uptake rate; AUCinsulin, the area under the curve for insulin; Free androgen index (FAI) ⫽ T (nmol/liter)/SHBG (nmol/liter)⫻100. LH, luteinizing hormone; FSH, follicular stimulating hormone; DHEA-S, dehydroepiandrosterone-sulfate; SHBG, sex hormone-binding globulin; TEST, testosterone; Prog, Progesterone. Values were given as means⫾SD or median (interquartile Range). a, P ⬍ 0.05; b, P ⬍ 0.01 compared with controls.

various other parameters. Circulating irisin correlated positively with BMI, WHR, FAT%, TG, TC, LDL-C, AUCInsulin, HOMA2-IR, and FAI. All these correlations remained statistically significant after adjustment for age. We also performed multiple stepwise regressions to determine variables that had independent associations with circulating irisin. The results showed that only FAT%, TC, and FAI were independently related factors to circulating irisin (Table 2). The multiple regression equation was YIrisin ⫽ 67.371 ⫹ 1.753XFAT ⫹ 10.627XTC ⫹ 26.277XFAI. In addition, increasing irisin levels showed a significant linear trend and were independently associated with PCOS, especially when concentrations were analyzed by row mean scores difference and the Cochran-Armitage trend test (Supplemental Table 1). The relative risks for PCOS were significantly elevated along with increasing irisin quartiles (Figure 1C). Effects of EHC on circulating irisin in healthy or PCOS women We next asked whether circulating irisin levels are affected by euglycemic hyperinsulinemia. EHCs were performed in 123 normal women and 178 women with

PCOS. During the EHC, plasma insulin levels were increased from 7.9 ⫾ 3.3 to 103.1 ⫾ 20.2 mU/L in normal women, and from 18.6 ⫾ 5.4 to 86.2 ⫾ 22.5 mU/L in PCOS women. Blood glucose was clamped at euglycemic levels (4.5–5.5 mmol/L) by a variable rate infusion of 25% glucose without clinically significant hypoglycaemic events founded in these subjects. During the EHC, M values were markedly lower in PCOS women than in normal women (6.08 ⫾ 2.68 vs 10.10 ⫾ 2.58 mg/kg per min; P ⬍ .01), indicating insulin resistance in the PCOS women. In response to hyperinsulinemia, circulating irisin in normal women significantly and rapidly dropped from 178.2 ⫾ 77.4 to 136.6 ⫾ 55.3 ␮g/L at 80 minutes, then to 137.4 ⫾ 53.4 ␮g/L at 100 minutes, and to 141.6 ⫾ 55.2 ␮g/L ␮g/L at 110 minutes (P ⬍ .01 vs 0 minutes), and finally to 141.2 ⫾ 54.2 ␮g/L at 120 minutes (P ⬍ .01 vs 0 minutes). In PCOS women, circulating irisin also rapidly dropped from 217.2 ⫾ 65.7 to 150.2 ⫾ 65.2 ␮g/L at 80 minutes, then to 156.5 ⫾ 59.53 ␮g/L at 100 minutes and 156.5 ⫾ 44.4 ␮g/L at 110 minutes, and finally to 157.6 ⫾ 61.4 ␮g/L at 120 minutes (all P ⬍ .01 vs 0 minutes, Figure 1D).


doi: 10.1210/jc.2014-2544


1489

there was a concomitant improvement in insulin sensitivity as shown by the significant decrease in HOMA2-IR and the significant increase in M values (from 5.09 ⫾ 2.27 at pretreatment to 6.02 ⫾ 2.68 at post-treatment 3 months, and finally to 6.70 ⫾ 2.53 mg/kg per min at post-treatment 6 months, both P ⬍ .01 vs pretreatment; Figure 2B). Importantly, after 6 months of metformin treatment, circulating irisin in the PCOS women was significantly decreased following increasing insulin sensitivity (from 213.7 ⫾ 64.7 at pretreatment to 185.3 ⫾ 66.2 at post-treatment 3 months, and finally to 170.2 ⫾ 70.3 ␮g/L at posttreatment 6 months, P ⬍ .01 vs pretreatment; Figure 2C). However, Figure 1. Circulating irisin levels in study population. (A) Circulating irisin levels according to after both 3 and 6 months of metBMI (normal-weight: BMI ⬍ 25 kg/m2 and overweight/obese: BMI ⱖ 25 kg/m2 vs normal weight, formin treatment, the changes of iri*P ⬍ .01). (B) Circulating irisin levels in normal and PCOS women (vs controls *P ⬍ .01). (C) Prevalence of elevated PCOS in different quartiles of irisin: quartile 1, ⬍ 129.4 ␮g/L; quartile 2, sin concentration patterns during 129.4 –169.2 ␮g/L; quartile 3, 169.2–231.9 ␮g/L; quartile 4, ⬎ 231.9 ␮g/L (*P ⬍ .01 vs quartile EHC were similar with pretreatment ƒ 1). (D) Circulating irisin levels in both PCOS and healthy women during EHC ( P ⬍ .01 compared (Figure 2D). Furthermore, we anawith controls; *P ⬍ .01 vs compared with 0 minutes). lyzed the correlation between the change in circulating irisin (⌬ irisin) Importantly, in a combined population of PCOS and nor- before and after metformin therapy (6 months) and the mal women, there was a significant negative correlation changes in other covariates. Interestingly, ⌬ irisin was sigbetween irisin and M values (r ⫽ ⫺0.194, P ⬍ .01; Figure nificantly positively associated with ⌬AUCinsulin (r ⫽ 0.24, 2B). P ⬍ .05). Effects of metformin treatment on circulating irisin levels in PCOS women Metformin treatment was started in 54 women with PCOS. Seven subjects withdrew before the study could be completed because of GI adverse effects, pregnancy, incompliance, and loss of contact. The anthropometric, biochemical, and hormonal data pre- and post-treatment with metformin were shown in Supplemental Table 2. After 3 and 6 months of metformin treatment, BMI, FAT %, FIns, AUCInsulin, HbA1c, and HOMA2-IR in PCOS women declined significantly, whereas SHBG increased (P ⬍ .01 vs pretreatment). After 6 months of metformin treatment, BMI and FAT % further declined (P ⬍ .01, vs pretreatment), whereas SHBG further increased (P ⬍ .01 vs pretreatment). In addition, TG, TC, FBG, FIns, 2h-Ins, AUCInsulin, HbA1c, HOMA2-IR, TEST, LH, and FAI also declined significantly in PCOS women after 6 months of metformin treatment (P ⬍ .05 and P ⬍ .01; vs pretreatment; Supplemental Table 2). During an OGTT, insulin levels at 0, 30, 60, and 120 minutes were lower than before metformin treatment (P ⬍ .05 or P ⬍ .01, Figure 2A). Also,

Discussion Recent comprehensive studies in animals and human provide convincing evidence of a link between insulin resistance, T2DM, and irisin (12, 21, 22). However, human data have been inconsistent. PCOS is a known insulinresistant state, and thus far, no report has demonstrated the relationship of circulating irisin and insulin resistance in PCOS women. Therefore, this is the first report, to our knowledge, describing the plasma level of irisin in patients with PCOS. In this study, we have demonstrated that circulating irisin was significantly higher in untreated PCOS women compared with healthy women. Importantly, overweight/obese subjects also had significantly higher circulating irisin than lean subjects regardless of PCOS status. These findings are in contrast to two recent studies which indicated that circulating irisin is significantly lower in patients with T2DM (10, 11, 16). However, these populations mostly had abnormal glucose metabolism


1490

Li et al

Table 2. Studied



Linear and Multiple Regression Analysis of Variables Associated with Circulating Irisin Levels in Subjects Simple

Multiple

Variable

R

P

b

P

Age (yr) BMI (kg/m2) WHR FAT (%) SBP (mmHg) DBP (mmHg) TG (mmol/liter)a TC (mmol/liter) HDL-C (mmol/liter) LDL-C (mmol/liter) FFA (umol/liter) HBA1C AUCInsulin (mU⫻h/liter)a HOMA2-IRa DHEA-S (␮g/dl) LH (IU/liter) FSH (IU/liter) Prog (nmol/liter) FAIa

0.001 0.250 0.163 0.275 0.096 0.042 0.157 0.180 0.102 0.131 ⫺0.004 0.084 0.180 0.188 0.057 0.037 -0.033 0.004 0.212

.993 .000 .004 .000 .095 .472 .007 .002 .081 .025 .946 .145 .002 .001 .342 .535 .576 .951 .000

— — — 1.753 — — — 10.627 — — — — — — — — — — 26.277

— — — .003 — — — .019 — — — — — — — — — — .038

a Log transformed before analysis. In multiple linear stepwise regression analysis, values included for analysis were age, BMI, WHR, SBP, DBP, AUCInsulin, HbA1c, HOMA2-IR, FFA, TG, TC, HDL-C, LDL-C, FAT%, DHEA-S, LH, FSH, Prog and FAI.

rather than isolated insulin resistance and were predominately male. Thus, the isolated contribution of insulin resistance cannot be determined. The advantage of our study is the lack of potentially confounding pharmacotherapy and abnormal glucose metabolism that may affect insulin resistance and circulating irisin. In addition, our results are in agreement with studies of Hee Park et al. indicating a positive correlation between irisin and anthropometric variables including BMI, FAT%, and WHR (23). Additionally, we also found that circulating irisin correlated significantly with dyslipidemia (increased TG, TC, and LDL-C), and insulin resistance (increased AUCInsulin, HOMA2-IR, and FAI), suggesting that this hormone could play an important role in the delicate balance of energy metabolism and insulin resistance. However, reported results on the association of irisin with obesity and metabolic phenotypes are quite controversial. Briefly, while some authors support our findings with reports of a positive correlation between circulating irisin levels and BMI (13, 14, 16), others reported a negative correlation (10, 11). Similarly, in contrast to our findings, some authors demonstrated a negative correlation between irisin levels and insulin resistance (11), whereas others demonstrated no correlation (10, 16). It is possible that these conflicting data are due to the frequent (poly) pharmacotherapy and other confounding variables such as age, sex, race or the level of physical activity of the subjects, contributing considerably to the heterogeneity of the study populations. These discrepancies might also be related to

differences in assays used by different studies, which may provide different read outs for irisin levels. Although the nature of the present study does not permit us to determine the cause of increased circulating irisin in PCOS women, these findings suggest that increased circulating irisin is an adaptive response to compensate for the decreasing insulin sensitivity and disturbances in metabolism associated with PCOS as proposed previously (13). In addition, a positive association between irisin and BMI in PCOS women may also reflect a compensatory increase of irisin levels to counterbalance the increasing needs for irisin (irisin “tolerance” or “resistance”) and prevent the development of insulin resistance (similar to increasing insulin levels in obesity), whereas in states of fully developed insulin resistance and metabolic syndrome (such as T2DM), irisin levels may be lower reflecting a failure of higher irisin to compensate (similar to insulin levels in advanced T2DM). This hypothesis needs to be studied in the future. Previous research has not determined whether hyperinsulinemia might be responsible for increased irisin levels. To address this question, we performed an EHC study to control glucose levels and investigated the effect of hyperinsulinemia on circulating irisin in normal and PCOS women. Our results showed that short-term hyperinsulinemia in both healthy and PCOS women led to a rapid decrease of circulating irisin at the start of EHC. After that, irisin levels were constant during the steady-state of EHC. This result suggests an inhibitory effect of short-term hyperinsulinemia on irisin release and leads us to speculate


doi: 10.1210/jc.2014-2544


1491

Figure 2. Effects of metformin treatment on circulating irisin levels and insulin sensitivity in PCOS women. (A) Plasma insulin levels in PCOS subjects during an OGTT (vs 0 minutes, *P ⬍ .01; vs pretreatment ƒP ⬍ .05, P ⬍ .01). (B) M values in PCOS subjects during EHC (vs pretreatment **P ⬍ .01). (C) Circulating irisin levels in PCOS subjects after both 3 and 6 months of metformin treatment (vs pretreatment **P ⬍ .01). (D) Circulating irisin levels during EHC before and after metformin treatment (vs pretreatment P ⬍ .01).

that increased circulating irisin levels observed in PCOS women are likely due to the effects of metabolic stress, such as long term hyperinsulinemia, dyslipidosis, and adipocytokines in a state of insulin resistance. More importantly, in an interventional study we report for the first time that metformin therapy given to PCOS women for 6 months resulted in a significant decrease in

circulating irisin and glucose levels with a concomitant improvement in insulin sensitivity as shown by increasing M values. Although metformin has been widely reported to treat patients with PCOS (24 –26), our result suggests yet another beneficial effect of metformin on circulating cytokine levels. Admitting that, this result does not permit us to conclude that it was a direct effect of met-


1492

Li et al


formin on circulating irisin, or an indirect effect of the decrease in insulin levels, or improved insulin sensitivity, our study highlights metformin therapy as a confounding factor in T2DM and obesity-related metabolic disease with regards to the regulation of circulating irisin levels. This should alert investigators who are studying irisin biology to consider this in their analyses. In addition, this revelation may also apply to other forms of antidiabetes therapy. The main strengths of this study are (1) its prospective design with the inclusion of newly diagnosed PCOS women prevents pharmacotherapy and other confounding variables, such as age and sex; (2) the association between irisin and insulin is investigated by EHC, a gold standard for evaluating insulin sensitivity; (3) most importantly, the effect of metformin therapy on circulating irisin levels is evaluated. The present study has some limitations. First, our study population was limited to Chinese; therefore, our findings may not be directly applicable to all populations. Second, sample size was relatively small so that our studies were not powerful enough to account for potentially confounding factors in our analysis. In addition, our results could be improperly influenced by some outliers due to the sample size. Finally, the cross-sectional design and a lack of specific information on irisin protein levels in muscle tissue limit our ability to infer a causal relationship between increased circulating irisin level and PCOS. Nonetheless, this study is sufficient to demonstrate novel associations of circulating irisin with hormonal, metabolic parameters, and insulin resistance. Also, the cross-sectional design of the study cannot prove causality, but can certainly raise credible hypotheses to be confirmed and extended by future prospective cohort as well as mechanistic studies. In conclusion, we report novel findings of a significant increase of circulating irisin, a novel myokine, in women with PCOS as well as the effect of short-term hyperinsulinemia on circulating irisin during EHC. More importantly, we present novel data that metformin treatment possibly via its insulin sensitivity increasing effect, significantly decreases circulating irisin levels in women with PCOS. The physiologic and pathologic significance of our findings remain to be further elucidated.

Acknowledgments Address all correspondence and requests for reprints to: Ling Li, Key Laboratory of Diagnostic Medicine (Ministry of Education) and Department of Clinical Biochemistry, College of Laboratory Medicine, Chongqing Medical University, 400 010 Chongqing, China. E-mail: [email protected].


This work was supported by research grants from the National Natural Science Foundation of China (81270913, 81 070 640, 81 100 567,81 300 702,and81 300 670);DoctoralFundofMinistry of Education of China (20105503110002, 20125503110003); Natural Science Foundation Key Project of CQ cstc (cstc2012 jjB10022). All the authors contributed to the conception and design of the study and critical revision of the manuscript, and all authors gave final approval of the version to be published. M.L., M.Y., X.Z., and X.F. contributed equally to this project. M.L., M.Y., X.Z., X.F., W.H, W.Z., C.W., D.L., and S.L. had full access to all of the data and took responsibility for the integrity of the data and the accuracy of the data analysis. L.L. and G.Y. drafted the manuscript and acted as study supervisors. H.L. revised the manuscript. Registered No. of Clinical Trial: ChiCTR-OCS-13003185. Disclosure Summary: The authors have nothing to disclose.

References 1. Dunaif A. Insulin resistance and the polycystic ovary syndrome: mechanism and implications for pathogenesis. Endocr Rev. 1997; 18:774 – 800. 2. Wild RA, Painter PC, Coulson PB, Carruth KB, Ranney GB. Lipoprotein lipid concentrations and cardiovascular risk in women with polycystic ovary syndrome. J Clin Endocrinol Metab. 1985;61:946 – 951. 3. Diamanti-Kandarakis E. Insulin resistance in PCOS. Endocrine. 2006;30:13–17. 4. Kandaraki E, Chatzigeorgiou A, Livadas S, et al. Endocrine disruptors and polycystic ovary syndrome (PCOS): elevated serum levels of bisphenol A in women with PCOS. J Clin Endocrinol Metab. 2011; 96:E480 –E484. 5. Diamanti-Kandarakis E, Katsikis I, Piperi C, et al. Increased serum advanced glycation end-products is a distinct finding in lean women with polycystic ovary syndrome (PCOS). Clin Endocrinol (Oxf). 2008;69:634 – 641. 6. Diamanti-Kandarakis E, Piperi C, Korkolopoulou P, et al. Accumulation of dietary glycotoxins in the reproductive system of normal female rats. J Mol Med (Berl). 2007;85:1413–1420. 7. Hutchison SK, Harrison C, Stepto N, Meyer C, Teede HJ. Retinolbinding protein 4 and insulin resistance in polycystic ovary syndrome. Diabetes Care. 2008;31:1427–1432. 8. Tan BK, Adya R, Farhatullah S, Chen J, Lehnert H, Randeva HS. Metformin treatment may increase omentin-1 levels in women with polycystic ovary syndrome. Diabetes. 2010;59:3023–3031. 9. Boström P, Wu J, Jedrychowski MP, et al. A PGC1-␣-dependent myokine that drives brown-fat-like development of white fat and thermogenesis. Nature. 2012;481:463– 468. 10. Choi YK, Kim MK, Bae KH, et al. Serum irisin levels in new-onset type 2 diabetes. Diabetes Res Clin Pract. 2013;100:96 –101. 11. Moreno-Navarrete JM, Ortega F, Serrano M, et al. Irisin is expressed and produced by human muscle and adipose tissue in association with obesity and insulin resistance. J Clin Endocrinol Metab. 2013;98:E769 –E778. 12. Ebert T, Stepan H, Schrey S, et al. Serum levels of irisin in gestational diabetes mellitus during pregnancy and after delivery. Cytokine. 2014;65:153–158. 13. Huh JY, Panagiotou G, Mougios V, et al. FNDC5 and irisin in humans: I. Predictors of circulating concentrations in serum and plasma and II. mRNA expression and circulating concentrations in response to weight loss and exercise. Metabolism. 2012;61:1725– 1738. 14. Stengel A, Hofmann T, Goebel-Stengel M, Elbelt U, Kobelt P, Klapp


doi: 10.1210/jc.2014-2544

15. 16.

17.

18.

19.

20.

BF. Circulating levels of irisin in patients with anorexia nervosa and different stages of obesity– correlation with body mass index. Peptides. 2013;39:125–130. Timmons JA, Baar K, Davidsen PK, Atherton PJ. Is irisin a human exercise gene? Nature. 2012;488: E9 –E10; discussion E10 –E11. Liu JJ, Wong MD, Toy WC, et al. Lower circulating irisin is associated with type 2 diabetes mellitus. J Diabetes Complications. 2013;27:365–369. Rotterdam ESHRE/ASRM-Sponsored PCOS consensus workshop group. Revised 2003 consensus on diagnostic criteria and long-term health risks related to polycystic ovary syndrome (PCOS). Hum Reprod. 2004;19:41– 47. Jia Y, Yuan L, Hu W, et al. Zinc-finger BED domain-containing 3 (Zbed3) is a novel secreted protein associated with insulin resistance in humans. J Intern Med. 2014;275:522–533. Yang M, Liu R, Li S, et al. Zinc-␣2-glycoprotein is associated with insulin resistance in humans and is regulated by hyperglycemia, hyperinsulinemia, or liraglutide administration: cross-sectional and interventional studies in normal subjects, insulin-resistant subjects, and subjects with newly diagnosed diabetes. Diabetes Care. 2013; 36:1074 –1082. Levy JC, Matthews DR, Hermans MP. Correct homeostasis


21.

22.

23.

24.

25.

26.

1493

model assessment (HOMA) evaluation uses the computer program. Diabetes Care. 1998;21:2191–2192. Kurdiova T, Balaz M, Vician M, et al. Effects of obesity, diabetes and exercise on Fndc5 gene expression and irisin release in human skeletal muscle and adipose tissue: in vivo and in vitro studies. J Physiol. 2014;592:1091–1107. Spiegelman BM. Banting Lecture 2012: Regulation of adipogenesis: toward new therapeutics for metabolic disease. Diabetes. 2013;62: 1774 –1782. Park KH, Zaichenko L, Peter P, et al. Diet quality is associated with circulating C-reactive protein but not irisin levels in humans. Metabolism. 2014;63:233–241. Gilbert C, Valois M, Koren G. Pregnancy outcome after first-trimester exposure to metformin: a meta-analysis. Fertil Steril. 2006; 86:658 – 663. Cassina M, Donà M, Di Gianantonio E, Litta P, Clementi M. First-trimester exposure to metformin and risk of birth defects: a systematic review and meta-analysis. Hum Reprod Update. 2014; 20:656 – 669. Zheng J, Shan PF, Gu W. The efficacy of metformin in pregnant women with polycystic ovary syndrome: a meta-analysis of clinical trials. J Endocrinol Invest. 2013;36:797– 802.


Effects of Metformin Treatment on Homocysteine Levels and Metabolic Parameters of Women With Polycystic Ovary Syndrome.

Metformin use in women with polycystic ovary syndrome.

Circulating irisin and glucose-dependent insulinotropic peptide are associated with the development of polycystic ovary syndrome.

Elevated circulating levels of betatrophin are associated with polycystic ovary syndrome.

Assessment of ovarian stromal blood flow after metformin treatment in women with polycystic ovary syndrome.

Young women with polycystic ovary syndrome have raised levels of circulating annexin V-positive platelet microparticles.

Polycystic ovary syndrome (PCOS): metformin.

Homocysteine levels in Indian women with Polycystic Ovary Syndrome.

Effect of metformin on serum interleukin-6 levels in polycystic ovary syndrome: a systematic review.

Impact of metformin on C-reactive protein levels in women with polycystic ovary syndrome: a meta-analysis.

Dyslipidemia in women with polycystic ovary syndrome.

CTRP3 in women with polycystic ovary syndrome.

Metformin improved health-related quality of life in ethnic Chinese women with polycystic ovary syndrome.

Effects of metformin on pregnancy outcomes in women with polycystic ovary syndrome: A meta-analysis.

Impact of Treatment with Metformin on Adipocytokines in Patients with Polycystic Ovary Syndrome: A Meta-Analysis.

Prediction of metabolic syndrome in women with polycystic ovary syndrome.

Metformin increases pressure pain threshold in lean women with polycystic ovary syndrome.

Metformin and other insulin sensitizers in polycystic ovary syndrome.

The effect of metformin treatment in vivo on acute and long-term energy metabolism and progesterone production in vitro by granulosa cells from women with polycystic ovary syndrome.

Circulating microRNAs in patients with polycystic ovary syndrome.

Effects of metformin treatment on serum levels of C-reactive protein and interleukin-6 in women with polycystic ovary syndrome: a meta-analysis: A PRISMA-compliant article.

Self-Confidence in Women with and without Polycystic Ovary Syndrome.

Therapeutic effects of metformin and clomiphene in combination with lifestyle intervention on infertility in women with obese polycystic ovary syndrome.

High-sensitivity C-reactive protein (hs-CRP) levels and its relationship with components of polycystic ovary syndrome in Indian adolescent women with polycystic ovary syndrome (PCOS).