Association of Insulin Receptor and Peroxisome Proliferator-Activated Receptor γ Gene Polymorphisms with Phenotypic Features and Insulin Resistance in Women with Polycystic Ovary Syndrome

Abstract

Introduction: Polycystic Ovary Syndrome (PCOS), the most prevalent endocrine disorder among reproductive-aged women, is characterized by three diagnostic phenotypic features: anovulation/ oligomenorrhea (AO/OM), polycystic ovarian morphology (PCOM), and hyperandrogenism (HA). Various ethnic groups differ in the prevalence and metabolic characteristics of PCOS. Insulin resistance (IR) is an important covariate in PCOS; its prevalence is reported to vary from 44 to 70% and Asian PCOS women are known to be more prone to IR. Both PCOS and IR are under significant genetic control with substantial heterogeneity among various racial groups. Several genes of the insulin signalling pathway are targeted as candidate genes linked with PCOS, among those INSR and PPARG genes are among the major ones. His1085C/T polymorphism of the INSR gene and His447C/T polymorphism in the PPARγ gene have been found to be significantly associated with PCOS in various populations including Indian women. In facility-based studies on the Bangalee population (the 3rd largest ethnic group in the world after the Han Chinese and Arabs) the prevalence of PCOS has been found to vary from 6.11%-92.16%. In reality, the prevalence as well as clinical, metabolic and endocrine aspects of PCOS have not yet been well studied in this population. The prevalence of IR has varied widely from 16% to 77%, but standardized methodology was not followed in most studies. To our knowledge, no genetic analysis has also been reported to explain the various phenotypic features of PCOS in this population. Given the critical role played by insulin resistance in the pathophysiology as well as in the management and prevention of PCOS His1085C/T polymorphism of the INSR gene and His447C/T polymorphism in the PPARγ gene were chosen to be explored in this study for their possible association with phenotypic features and IR among Bangalee women with PCOS. The specific objectives were: a. to explore the proportion of IR and its covariates in PCOS; b. to find out the independent association of individual diagnostic phenotypic features of PCOS with IR; c. to compare the frequency of the two different single nucleotide polymorphisms (SNPs) between PCOS and healthy women; and d. to explore the independent association of the investigated SNPs with phenotypic features and IR on adjustment of the effects of major confounding variables. Methods: Following a case-control design an observational analytic study was conducted on age-matched 111 non-PCOS and 141 PCOS subjects [age, years, median (range): 24 (19-34) in non-PCOS vs 23 (15-34) in PCOS groups], both recruited from community and hospital-based settings through purposive sampling. PCOS was diagnosed as per modified Rotterdam criteria which require the presence of at least 2 of the 3 phenotypic features: AM/PM, HA and PCOM. Subjects with pregnancy, lactation, smoking habits, and other potential modifiers of endocrine/ ovarian function (such as metabolic or other disorders and use of oral contraceptives and relevant drugs) were excluded. Hirsutism was detected and graded using the Ferriman-Gallwey score (FG-score) with a cut-off value of ≥ 8 as a marker of clinical HA. All patients undertook a standard 75-g oral glucose tolerance test (OGTT). Serum levels of hormones [c-peptide (as surrogate for insulin), total testosterone and sex hormone binding globulin (SHBG)] were assayed by chemiluminescent immunoassay (CLIA) technique; other serum parameters [Glucose (Gl) and Lipid Profile] were estimated by automation-based standard techniques. HA was defined in terms of biochemical HA, as women having free androgen index (FAI) >7.1, computed by the formula TT (nmol/L)/SHBG (nmol/l) ×100. IR was calculated by using the C-peptide-modified formula (HOMA1-IR = 1.5 + FPG × FCP / 2800) with a value > 2.4 considered to represent IR. A lower abdominal ultrasound with folliculometry was done using a Voluson E6 machine and a standard counting procedure was followed to detect PCOM. Genomic DNA was extracted from peripheral blood using a standard DNA isolation &and purification kit and PCR amplification was done with the following regions of the targeted genes:- INSR - Cytogenic location 19p13.3-p13.2, primary target rs1799817; and b. PPARγ - Cytogenic location 3q25, primary target rs3856806. The resulting amplified products (800 bp segments up- and down-streams with primary targets in the middle) were sequenced by Sanger sequencing using an automated Analyzer and the data were analyzed using bioinformatics tools. Statistical analyses were done by relevant bi- and multi-variate statistics using SPSS for Windows (v 26) software. The study was approved by the Ethics Committees of the Department of Biochemistry of the University of Dhaka and BSMMU. Results: PCOS subjects had significantly higher (p=0.037) median BMI (23.98 vs 23.40) and higher (p=0.019) Waist-Hip Ratio (WHR) (0.94 vs 0.92) as compared to non-PCOS subjects. PCOS subjects also had significantly higher median 2hBG (p=0.004)], TChol (p=0.045), and TG (p=0.002) as compared to Non-PCOS subjects. PCOS subjects had significantly higher median total Testosterone [U/l: 1.4 (0.47-5.20) vs 1.1 (0.45-2.91), p=0.001] and median HOMA-IR [%:2.5 (1.7-29.1) vs 2.2 (1.7-4.6), p<0.001] as compared to their Non-PCOS counterparts. The proportion of subjects with IR was 52% and 28% in the PCOS and Non-PCOS groups, respectively. HA was present in much higher proportion among insulin-resistant PCOS subjects (83%) as compared to the non-insulin-resistant ones (46%), with a highly significant difference (x2=17.11; p <0.001). IR was found to be significantly correlated with BMI (p<0.001), TG (p=0.007), FAI (p<0.001) and 2hBG (p=0.002) in the PCOS group. On binary logistic regression analysis, BMI (p<0.001) and FAI (p=0.001) showed significant association with IR when the effects of age, WHR, TC, and TG were adjusted. Regarding His1085C/T alleles in the INSR Gene, the proportions of C and T alleles were 56% and 42% among Non-PCOS and 69% and 41% among PCOS subjects (p=ns). No significant difference in allele distribution was found for any phenotypic feature of PCOS or for IR, either in the Non-PCOS or PCOS group. No significant association of His1085C/T polymorphism of the INSR gene with individual phenotypic features of PCOS or with IR was found on binary logistic regression analysis. Regarding His 447C/T Polymorphism of the PPARγ Gene, the proportions of C and T alleles were 83% and 17% among Non-PCOS and 82% and 18% among PCOS subjects (p=ns). Except for Hirsutism, no significant difference in allele distribution was found for any phenotypic feature of PCOS or for IR, either in the Non-PCOS or PCOS group; only in the PCOS group, Hirsutism showed a significant positive association (p=0.05) with the polymorphism. No significant association of His 447C/T Polymorphism of PPARγ Gene with individual phenotypic features of PCOS or with IR was found on binary logistic regression analysis. On multiple linear regression analysis, a significant (p=0.02) association of IR with His447C/T polymorphism in the PPARγ Gene was found on adjusting the effects of age, BMI. WHR, TC, TG, and total testosterone. Conclusions: The present data lead to the following conclusions: i. Overweight and obesity (both central and peripheral) are notable features of PCOS among young to early middle-aged Bangalee women; ii. Substantial proportions of non-PCOS young women are already overweight/ obese as well as hyperandrogenic signifying their potential risk of being PCOS in future; iii. Insulin resistance seems to be present among more than fifty percent of young PCOS and more than a quarter of young non-PCOS subjects; iv. Strong independent association of insulin resistance with hyperandrogenism is present in PCOS irrespective of age and lipidemic status; v. His1085C/T polymorphism of the INSR gene does not seem to have a significant association with any of the characteristic phenotypic features of PCOS or with insulin resistance among Bangalee women; vi. Among Bangalee women, both clinical hyperandrogenism (an important phenotypic feature in diagnosing PCOS) and insulin resistance are associated with His447C/T polymorphism in the PPARγ gene; and vii. There is no independent association of biochemical hyperandrogenism with PPARG gene polymorphism among Bangalee women; however, the polymorphism contributes to clinical hyperandrogenism through the mediation of insulin resistance in these women.