Abstract
Polycystic ovary syndrome (PCOS) and type-2 diabetes mellitus (T2DM) share common genetic features. Transcription factor 7-like-2 (TCF7L2) is consistently studied T2DM susceptibility locus. However, limited studies on TCF7L2 have failed to demonstrate any link with the PCOS risk. Therefore, we investigated the association of TCF7L2 polymorphic variant (rs12255372) with the PCOS risk. We recruited 120 PCOS cases, diagnosed as per Rotterdam 2003 criteria, and an equal number of age-matched controls. Besides a detailed clinical assessment, subjects underwent biochemical and hormonal profiling. Genotyping for rs12255372 was done by PCR–RFLP. Conditional logistic regression was used to calculate odds ratios (ORs) and 95% confidence intervals (95%CIs) of genotype–phenotype correlations. The PCOS cases reported fewer menstrual cycles per year and exhibited signs of hyperandrogenism. The heterozygous genotype of rs12255372 was strongly associated with the PCOS risk (OR = 2.00; 95%CI: 1.07–3.76). Unlike controls, only 3 cases harbored TT genotype, and the PCOS risk persisted in the dominant model (GT + TT) as well. Moreover, we found a synergistic effect modification by the variant genotype in the subjects who had family histories of T2DM, hirsutism, or menstrual irregularities. We report a significant association of the TCF7L2 polymorphic variant rs12255372 with the PCOS risk.
Keywords: Diabetes, Polymorphism, Kashmir, India, Polycystic ovary syndrome, TCF7L2, Variant
Introduction
Polycystic ovary syndrome (PCOS), a highly prevalent disorder, represents the most common endocrine–metabolic disorder in reproductive-aged women. It is characterized by chronic anovulation, hyperandrogenism, and polycystic ovarian morphology [1]. In women of reproductive age, the global age-standardized incidence rate of PCOS has increased upto ~ 83/105 population in 2017 [2]. The prevalence of PCOS in India is reported between 3.7 to 22.5%, however its prevalence in Kashmir, India is as high as ~ 28.9% by NIH criteria and 34.3% by AE-PCOS criteria [3]. The significant disparity in the reported prevalence from different ethnicities has been attributed to the varied disease phenotype and non-uniform diagnostic approaches used for PCOS [4]. The deleterious impact of this disorder is not limited to female reproductive function, but their metabolic functions as well. Women with PCOS are at an increased risk for glucose intolerance, insulin resistance (IR) and type 2 diabetes mellitus (T2DM), metabolic syndrome, infertility, and cardiovascular events [1, 5]. Owing to its high prevalence and complexity in nature, currently, like T2DM, PCOS is a significant clinical and public health challenge throughout the world [6]. India is being considered as the diabetic capital of the world and young Indian women with PCOS presented impaired glucose tolerance [7]. Regardless of their body mass index (BMI), women with PCOS have a four-fold increased risk of T2DM with an early disease onset [8].
PCOS, a complex, polygenic, and multifactorial disorder is proposed to be an orchestration of gene–gene and gene-environment interactions. The syndrome is highly inherited and the risk increases up to 40% in families with a history of PCOS [9]. Given the strong etio-pathophysiological overlap between T2DM and PCOS, known T2DM predisposing genes emerge as strong drivers in the development of PCOS. The T2DM associated genetic loci including TCF7L2 (Transcription Factor- 7-Like 2) have been reported to contribute to PCOS pathogenesis [10]. TCF7L2 gene encodes a high-mobility group box-containing transcription factor that binds to DNA directly. The most potent locus for T2DM risk, TCF7L2 was identified by linkage and genome-wide association studies(GWAS) and then replicated in multiple ethnic populations [11]. The TCF7L2 forms a fundamental part of the Wnt β-Cat signaling pathway, a complex network of interacting partners producing diverse effects in the cell. Studies evaluating the physiological roles of genetic variations have found that TCF7L2 variants (rs290487) can contribute to insulin resistance (IR) [12], linking Wnt-β-Cat signaling with IR—a hallmark of PCOS.
While the genetic variations in the TCFL2 have been persistently associated with T2DM, the results on their association with PCOS are mixed and inconclusive. Owing to such an ambiguity in the results, the association of TCF7L2 variations with the development of PCOS remains unclear. Thus, we aimed to investigate the association of rs12255372 polymorphic variant of TCF7L2 with the PCOS risk among women with PCOS from Kashmir, India.
Methods
Recruitment of Subjects
The study was carried out in the department of endocrinology, in a tertiary care hospital of North India. Women presenting with features suggestive of PCOS including oligomenorrhea, hirsutism, acne, obesity as main complaints were evaluated. Out of 252 women informed about the study, 150 women qualified as PCOS cases, and 102 women were ineligible for the study. Among 150 qualified PCOS cases, 30 were excluded because they consumed some medications (n = 12) or did not agree to participate (n = 18). The diagnosis of PCOS was done according to Rotterdam 2003 criteria [13]. The women aged between 18 and 40 years were enrolled from January 2017 to December 2020. As reported previously [14, 15], recruited women underwent a detailed clinical examination followed by anthropometrical measurement and subsequent biochemical and hormonal tests during their respective follicular phases. The hirsutism assessment was done by Ferriman–Gallwey (FG score) [16].
The control group consisted of apparently healthy women (N = 120), age-matched (± 2 years) for respective cases, recruited from the community and various educational institutes as a part of screening-cum-community outreach programs periodically conducted by the Departments of Clinical Research, and Endocrinology. The admissibility measures for the control group included healthy volunteers with regular menstrual cycles, no apparent signs of clinical or biochemical hyperandrogenism. None of the controls was on hormonal therapy (including oral contraceptives) for the previous three months or longer or on medication known to affect carbohydrate metabolism or endocrine parameters for at least three months before enrolment for the study. Subjects who had a history of endocrine or autoimmune disorders were excluded from the control group in the study. All the recruited controls underwent similar clinical and laboratory evaluation to that of cases. The study was reviewed and approved by the Institute Ethical Committee (No. IEC19/16). Informed written consent was taken from each participant before recruitment.
Biochemical and Hormonal Assessment
The blood samples were collected post an overnight fast on the 2nd–3rd day of their menstrual cycle. The biochemical parameters including oral glucose tolerance test, lipid profile, and kidney function test, were estimated on a fully automated biochemistry analyzer (DiaSys Respons®910, Germany) using standard kits following the manufacturer's protocol.
The serum concentration of luteinizing hormone (LH), follicle-stimulating hormone (FSH), total testosterone, thyroid-stimulating hormone (TSH), Prolactin was measured using COBAS e-411(Roche Diagnostics, USA). Fasting serum insulin levels were measured by using ELISA (Elecsys Insulin, Roche Diagnostics GmBH. Cobas-411 analyzer) following manufacturer’s instructions, and IR was assessed using homeostatic model assessment insulin resistance index (HOMA-IR), fasting glucose to insulin ratio (FGIR), and quantitative insulin sensitivity check index (QUICKI). The HOMA index was calculated as [fasting insulin (mIU/mL) × fasting glucose (mg/dL)]/405. The QUICKI was calculated as 1/[log fasting insulin (mIU/mL) + log fasting glucose (mg/dL)]. The FGIR was calculated by using the formula fasting glucose × fasting insulin.
Genotyping
Genomic DNA was isolated using the standard phenol–chloroform method, followed by its purity and integrity assessment. The rs12255372 polymorphism of theTCF7L2 was genotyped by Polymerase Chain Reaction- Restriction Fragment Length Polymorphism using the already published primers:
Forward 5′-CTGGAAACTAAGGCGTGAGG-3′, Reverse5′ GGGTCGATGTTGTTGAGCTT-3 (Sigma Aldrich) [17]. A final reaction volume of 25μL for the polymerase chain reaction was constituted, which contained ~ 100–160 ng of genomic DNA, 0.25 μM of each primer, 0.5 mM of each deoxynucleotide triphosphate, 0.25 U of Taq DNA polymerase (Fermantas, MBI, Vilnius, Lithuania). The PCR was carried out on a Thermal Cycler (Agilent Technologies, USA) under the following conditions: 95 °C for 5 min, followed by 35 cycles of 95 °C for 30 s, 54 °C for30 seconds, 72 °C for 30 s, and a final extension of 72 °C for 5 min. The amplicons (346 bp) were then digested with the restriction enzyme Tsp509I at 65 °C for 3 h. The digested products were separated by electrophoresis on a 3% agarose gel containing ethidium bromide and visualized using a Gel doc system (Alpha Innotech, USA). The wildtype PCR fragment had two inherent sites for Tsp509I, and rs12255372(G > T) would create a 3rd site in the fragment. Therefore, the wild-type (GG) genotype was characterized by the presence of three bands at 143 bp, 99 bp, and 104 bp. The mutant (TT) genotype was identified by the presence of 126 bp, 99 bp, 104 bp, and 17 bp. The heterozygous genotype was characterized by the presence of 5 bands at 143 bp, 126 bp, 104 bp, 99 bp, and 17 bp. Pertinent to mention that in all the conditions 99 bp and 104 did not separate and showed a single band at ~ 100 bp and 17 bp was not visible on the gel.
Statistical Analysis
All continuous variables were presented as mean ± standard deviation and categorical variables as numbers and percentages. Clinical, anthropometric, hormonal, and metabolic variables were compared between PCOS cases and controls by unpaired student t-test and categorical variables were compared by chi-square test. Conditional logistic regression was used to calculate unadjusted and adjusted odds ratios (ORs) and corresponding 95% confidence intervals (CIs) for the association of TCF7L2 polymorphism with PCOS risk. Besides effect modification by the studied genetic variation (gene–clinical phenotype interaction) was also assessed. To avoid possible confounding, results were adjusted for age, levels of testosterone, LH, FSH, family history of T2DM and hirsutism, FG score, age of menarche, and BMI. All statistical analysis was done using Stata software, version 16 (STATA Corp., College Station, TX, USA). Two-sided p values < 0.05 were considered statistically significant.
Results
A total of 120 cases and age-matched controls for each case were recruited in the current study. The data on clinical history and anthropometry of PCOS women and controls is given in Table 1. The mean age (± SD) of cases and controls was 24.2(± 4.09) and 23.37 (± 3.03) respectively. The women with PCOS reported less number of menstrual cycles per year when compared to controls (7.7 ± 3.23) vs. 11.96 ± 0.20); p < 0.001). The BMI, mean FG score, levels of serum total testosterone, LH, and indices of lipid profile were significantly higher in cases than controls (p < 0.001). The PCOS cases had higher serum fasting insulin levels and presented IR as depicted by their HOMA(p < 0.001; (Table 1). Unlike the control group, more cases reported a family history of T2DM, (59.32% vs. 25.21%), hirsutism (21.67% vs. 5.83%), or maternal history of irregular cyclicity (32.5% vs.15.0%) (Table 2).
Table 1.
Anthropometric, hormonal, metabolic, and biochemical parameters in PCOS cases and controls
| Parameters | PCOS (N = 120) Mean () |
Control (N = 120) Mean () |
p value |
|---|---|---|---|
| Mean age (years) | 24.2 (± 4.09) | 23.37 (± 3.03) | 0.077 |
| Age at menarche (years) | 12.97 (± 1.31) | 13.49 (± 1.23) | 0.002 |
| Menstrual cycles/year | 7.7(± 3.23) | 11.96(± 0.20) | 0.000 |
| Mean FG score | 11.32 ( ± 4.40) | 5.97 (± 0.99) | 0.000 |
| Height (cm) | 157.14 (± 5.13) | 156.46 (± 5.36) | 0.318 |
| Weight (Kg) | 63.04 (± 10.44) | 53.82 (± 9.49) | 0.001 |
| BMI (Kg/m2) | 25.58 (± 4.33) | 21.99 (± 3.72) | 0.000 |
| Waist circumference (cm) | 89.68 (± 10.50) | 82.33 (± 8.49) | 0.000 |
| Systolic blood pressure (mmHg) | 116.02 (± 15.33) | 113.51 (± 10.75) | 0.147 |
| Diastolic blood pressure (mmHg) | 78.14 (± 7.52) | 77.95 (± 8.47 ) | 0.855 |
| Serum fasting blood glucose (mg/dl) | 88.18 (± 10.05) | 83.67 (± 9.23) | 0.000 |
| Serum AST/OT (IU/L) | 24.00 (± 8.34) | 24.79 (± 7.08) | 0.448 |
| Serum ALT/PT(IU/L) | 26.28 (± 14.42) | 22.43 (± 10.51) | 0.021 |
| Serum ALP (IU/L) | 93.44 (± 27.13) | 88.71 (± 35.46) | 0.270 |
| Serum albumin (g/dL) | 4.54 (± 0.43) | 4.71 (± 0.54) | 0.008 |
| Serum urea (mg/dL) | 22.78 (± 7.23) | 21.73 (± 4.92) | 0.225 |
| Serum creatinine (mg/dL) | 0.84 (± 0.17) | 0.76 (± 0.18) | 0.005 |
| Serum uric acid (mg/dL) | 5.26 (± 1.31) | 4.39 (± 0.96) | 0.000 |
| Serum total cholesterol (mg/dL) | 171.55 (± 32.20) | 156.89 (± 28.59) | 0.000 |
| Serum triglycerides (mg/dL) | 126.97 (± 60.55) | 118.67 (± 46.26) | 0.242 |
| Serum HDL (mg/dL) | 49.36 (± 16.78) | 44.99 (± 12.15) | 0.026 |
| Serum LDL (mg/dL) | 95.11 (± 20.65) | 82.91 (± 20.53) | 0.000 |
| Serum total T4 (ug/dL) | 8.59 (± 2.01) | 8.13 (± 1.70) | 0.094 |
| Serum TSH (ug/dL) | 3.51 (± 1.84) | 2.79 (± 1.24) | 0.000 |
| Serum LH (mIU/ml) | 8.49 (± 4.83 ) | 6.09 (± 3.24) | 0.000 |
| Serum FSH (mIU/ml) | 6.49 (± 1.79) | 6.67 (± 2.50) | 0.553 |
| LH: FSH ratio | 1.37(± 0.80) | 1.10 (± 0.873) | 0.024 |
| Serum total testosterone (ng/mL) | 63.53 (± 25.62) | 24.79 (± 9.37) | 0.000 |
| Serum prolactin (ng/dL) | 16.73 (± 8.00) | 12.55 (± 6.74) | 0.000 |
| Serum fasting insulin (mIU/ml) | 19.16 (± 13.82) | 6.26 (± 3.25) | 0.000 |
| HOMA IR | 4.03(± 2.92) | 1.31 (± 0.71) | 0.000 |
| QUICKI | 0.32 (± 0.04) | 0.38 (± 0.03) | 0.000 |
| FGIR | 6.59 (± 5.18) | 17.13 (± 8.93) | 0.000 |
Values are given as mean ± SD
PCOS Polycystic ovary syndrome, FG Ferrimen-Gallway Score, BMI Body Mass Index, TSH thyroid stimulating hormone, LH Luteinising Hormone, FSH Follicle Stimulating Hormone, AST/OT Aspartate aminotransferase, ALT/PT Alanine aminotransferase, ALP alkaline phosphatase, LDL Low Density Lipoprotein, HDL High-Density Lipoprotein, IU International Units, HOMA IR homeostasis model assessment estimated insulin resistance, QUICKI qualitative insulin sensitivity check index, FGIR fasting glucose insulin ratio
The numbers may not add up to the total due missing information in some variables. A p-value of < 0.05 was considered statistically significant (in bold numbers)
Table 2.
Stratification of subjects based on PCOS phenotypes, their heritability and allele frequency
| Parameters | PCOS (N = 120) | Control (N = 120) | p value |
|---|---|---|---|
| Family history of T2DM | |||
| Absent | 48 (40.68%) | 89 (74.79%) | |
| Present | 70 (59.32%) | 30(25.21%) | 0.000 |
| Family history of hirsutism | |||
| Absent | 94 (78.33%) | 113(94.17%) | |
| Present | 26(21.67%) | 7 (5.83%) | 0.001 |
| Family history of menstrual irregularities | |||
| Absent | 81 (67.50%) | 102(85.00%) | |
| Present | 39 (32.50%) | 18 (15.00%) | 0.002 |
| Acne | |||
| Absent | 50(42.02%) | 69 (57.5%) | |
| Present | 69 (57.98%) | 50(41.67%) | 0.022 |
| Alopecia | |||
| Absent | 68 (56.67%) | 102 (85%) | 0.000 |
| Present | 52 (43.33) | 18 (15%) | |
| Acanthosis nigricans | |||
| Absent | 102(87.18%) | 111 (94.07%) | |
| Present | 15(12.82%) | 7 (5.93%) | 0.082 |
| Allele frequency (rs12255372) | |||
| G | 161 (67.1%) | 185(77.1%) | |
| T | 79 (32.9%) | 55 (22.9%) | 0.019 |
T2DM;: type 2 diabetes mellitus., Chi square test was used to calcite p values. A p-value of < 0.05 was considered statistically significant (in bold numbers)
The minor allele (T) frequency for rs12255372 reprted in gnomAD, 1000G and IndiGenome data bases is 0.27, 0.21 and 0.219 respectively [18]. In the current study, more cases harbored the mutant allele (T) than the respective controls (32.9% vs. 22.9%; p = 0.019) (Table 2). Among the PCOS cases, 60.83% were heterozygous (GT) compared to controls (45.83%). We found a two fold risk of PCOS in the subjects who harbored GT genotype (OR = 2.00; 95% CI:1.07–3.76). Unlike controls, the homozygous mutant genotype (TT) was present in 3(2.5%) cases and the risk associated with the genotype persisted in the dominant model (GT + TT) as well (OR = 2.01; 95% CI: 1.08–3.78). We observed a synergistic effect on the PCOS risk in the subjects, who had a family history of T2DM (OR = 11.35; 95% CI: 3.34–38.55) or maternal history menstrual irregularity (OR = 6.65; 95% CI: 2.18–20.33), and harbored variant genotype (GT + TT) of TCF7L2 (rs12255372); albeit with wider CI’s due to low numbers in the model. However, the risk associated with the variant genotype did not change much in the subjects who had acne or Acanthosis nigricans (Table 3). On stratifying the data based on the genotypes, the earlier age of menarche in the PCOS cases was independent of the genotype harbored.
Table 3.
Association of polymorphic variant (rs12255372) of TCF7L2 gene with PCOS and its effect modification on disease phenotypes
| Genotype/ and phenotype | Cases (N = 120) |
Controls (N = 120) |
Unadjusted OR (95% CI) | Adjusted OR (95% CI)a | p value |
|---|---|---|---|---|---|
| rs12255372 | |||||
| GG (Wildtype) | 44 (36.7) | 65 (54.17) | Referent | Referent | |
| GT (Heterozygous) | 73(60.83) | 55 (45.83) | 2.00 (1.16–3.42) | 2.00(1.07–3.76) | 0.030 |
| TT (Mutant) | 3 (2.50) | 0 (0.00) | 1.68 (–) | – | 0.039 |
| GT + TT (Variant) | 76 (63.3) | 55(45.83) | 2.05 (1.20–3.50) | 2.01(1.08–3.78) | 0.029 |
| Family history of diabetes mellitus | |||||
| F/H DM−− + Wildtype | 19 (16.10) | 43 (36.13) | Referent | Referent | |
| F/H DM−− + Variant | 29 (24.58) | 46 (38.66) | 1.38 (0.64–2.96) | 1.15 (0.47–2.80) | 0.756 |
| F/H DM++ + Wild type | 25 (21.19) | 21 (17.65) | 2.61 (1.09 –6.26) | 2.39 (0.85 –6.75) | 0.098 |
| F/H DM++ + Variant | 45 (38.14) | 9 (7.56) | 11.87 (4.10–34.42) | 11.35 (3.34–38.55) | 0.000 |
| Family history of hirsutism | |||||
| F/H Hirsutism−− + Wildtype | 35 (29.17) | 59 (49.17) | Referent | Referent | |
| F/H Hirsutism−− + Variant | 59 (49.17) | 54 (45.00) | 1.69 (0.97–2.95) | 1.65 (0.92–2.93) | 0.088 |
| F/H Hirsutism ++ + Wild type | 9 (7.50) | 6 (5.00) | 2.64 (0.77–9.07) | 2.61 (0.75–9.05) | 0.130 |
| F/H Hirsutism ++ + Variant | 17 (14.17) | 1 (0.83) | 22.97 (2.96–178.14) | 19.23 (2.46 –150.25) | 0.005 |
| History of menstrual irregularity | |||||
| Wild type + > 9 cycles/year | 18 (15.0) | 65 (54.2) | Referent | Referent | |
| Variant + > 9 cycles/year | 26 (21.7) | 0 (0.0) | – | – | < 0.0001 |
| Wil type + ≤ 9 cycles/year | 26 (21.7) | 55 (45.8) | 1.71 (0.67–4.35) | 1.67 (0.59–3.98) | 0.257 |
| Wild type + ≤ 9 cycles/year | 50 (41.6) | 0 (0.0) | – | – | < 0.0001 |
| Family history of menstrual irregularity (F/H MI) | |||||
| F/H MI−− + Wild type | 27 (22.50) | 52 (43.33) | Referent | Referent | |
| F/H MI−− + Variant | 54 (45.00) | 50 (41.67) | 2.08 (1.08–3.99) | 2.05 (1.05–4.00) | 0.034 |
| F/H MI++ + Wild type | 17 (14.17) | 13 (10.83) | 3.30 (1.27–8.55) | 3.05 (1.17–7.94) | 0.022 |
| F/H MI++ + Variant | 22 (18.33) | 5 (4.17) | 8.05 (2.67–24.31) | 6.65 (2.18–20.33) | 0.001 |
| Acne vulgaris | |||||
| Acne−− + Wild type | 22 (18.49) | 37 (30.83) | Referent | Referent | |
| Acne−− + Variant | 28 (25.53) | 32 (26.67) | 1.29 (0.61–2.75) | 1.23 (0.57–2.68) | 0.591 |
| Acne++ + Wild type | 22 (18.49) | 28 (23.33) | 1.14 (0.51–2.54) | 1.07 (0.47–2.42) | 0.873 |
| Acne++ + Variant | 47(39.50) | 23 (19.17) | 3.05 (1.47–6.35) | 2.87 (1.35–6.11) | 0.006 |
| Acanthosis nigricans | |||||
| Acanthosis−− + Wild type | 38 (32.48) | 61 (51.69) | Referent | Referent | |
| Acanthosis−− + Variant | 64 (54.70) | 50 (42.37) | 1.89 (1.09–3.27) | 1.70 (0.86–3.37) | 0.126 |
| Acanthosis++ + Wild type | 6 (5.13) | 4 (3.39) | 2.72 (0.60–12.26) | 1.21 (0.18–8.10) | 0.841 |
| Acanthosis++ + Variant | 9 (7.69) | 3 (2.54) | 4.23 (1.09–16.44) | 2.22 (0.51–9.72) | 0.289 |
++Present −−Absent N: number of individuals. The Chi-square test (× 2) was used to calculate p values for categorical variables. F/H Family history, T2DM: type 2 diabetes mellitus. Variant: GT + TT
aThe results were adjusted for age, levels of testosterone, LH, FSH, family history of T2DM and hirsutism, FG score, age of menarche, and BMI. A p-value of < 0.05 was considered statistically significant (in bold numbers)
Discussion
Our results indicate that the polymorphic variant rs12255372 of the TCF7L2 gene is associated with increased susceptibility to PCOS. Moreover, we found a synergistic effect modification by the variant genotype in the subjects who had family histories of either T2DM, hirsutism, or menstrual irregularities. To our knowledge, this is the first report of an association between PCOS and TCF7L2 variant in a high-risk Kashmiri population, where the prevelence of PCOS as high as 35.3% has been reported [3].
The TCF7L2 gene has been linked with several fundamental cellular processes from β-cell dysfunction and adipogenesis to different cancers. Most of the TCF7L2 variants studied are distributed in the noncoding regions, indicating that they modulate the expression of TCF7L2 [19]. Although the exact mechanism of TCF7L2 mediated pathophysiology of PCOS is poorly understood, the mutations in the TCF7L2 gene might upregulate the expression of TCF7L2 in pancreatic cells, which may enhance the proinsulin/insulin ratios and pancreatic β-cell apoptosis, and decrease insulin secretion, thereby orchestrating the PCOS risk [20]. Moreover, genetic variations in TCF7L2 have been found to contribute to IR -a hallmark of PCOS [12].
Given the significant overlap between genetic and disease phenotype of T2DM and PCOS [10], proposed roles of TCF7L2 in the glucose homeostasis, and pancreatic β-cell function, can be a plausible explanation for the enhanced PCOS risk in the women that harbored the variant genotype in our cohort. However, further mechanistic studies are warranted to elucidate TCF7L2 (rs12255372) mediated PCOS pathophysiology. TCF7L2 is one of the few GWAS-identified susceptibility loci consistently associated with T2DM in diverse ethnicities. To date, many molecular studies carried out so far in diverse populations that evaluated the association between TCF7L2 polymorphisms and PCOS risk have reported inconsistent results. While an earlier metanalysis found a significant association of TCF7L2 variants (rs7903146) with PCOS risk [21], recent metanalysis [22] did not find any association of rs12255372 with the PCOS risk. The variation in the results is attributed to the varied disease phenotypes, non-uniform diagnostic criteria, study design, and ethnic differences.
Similar to the earlier reports elsewhere [23] and from our group [24], we also found evidence of IR in our patients in the current study. While obesity and PCOS have an independent and a synergistic relationship with IR, studies have shown that the majority of women with PCOS, irrespective of their BMI, have a form of IR, intrinsic to the syndrome. The pathogenesis of IR in PCOS is complex and is poorly understood, though it is propounded that there is a post-binding defect in insulin receptor signalling [25].
Maternal family history of PCOS is a risk factor for PCOS in daughters. Based on familial case clustering, PCOS is deemed to be a heritable disorder [26]. The high prevalence of PCOS or its clinical phenotypes including hyperandrogenemia, hirsutism, reproductive dysfunction, polycystic ovaries, among first-degree relatives is indicative of familial and genetic influences on the disorder [27]. We found an enhanced risk in the subjects who had a family history of T2DM, hirsutism, or maternal history of menstrual irregularities and harbored the variant genotype of rs12255372 suggesting heritability associated with the later in the PCOS women. However, paternal genotyping for rs12255372 of the subjects was not done in our study to rule out the presence of de novo mutations.
Unlike controls, young women with PCOS experience menarche at an earlier age (albeit with a wider age range), and there is sparse literature available studying the underlying predictors for age at menarche in PCOS [28]. Although, we observed a significant difference in the age of menarche between PCOS cases and controls (p = 0.002), the earlier age of menarche in the PCOS cases was independent of the genotype harbored in our cohort.
As expected, we found that PCOS cases presented elevated serum total testosterone levels compared to the respective controls [2, 26]. Hyperandrogenemia may result from elevated testosterone production from polycystic ovaries or elevated LH levels [29]. The increased testosterone also may be stimulated by elevated insulin levels [30]. An earlier study reported that the genetic variation in TCF7L2 can contribute to IR [12]. Therefore, it is tempting to speculate that variant genotype of rs12255372 might contribute to IR, which in turn can elevate the testosterone levels in women and modulate the risk of PCOS. However, more replicative, and mechanistic studies are needed to validate and unveil the causal role, (if any).
In addition to the study being carried out in a population harboring a preserved genetic pool with a high prevalence of T2DM and PCOS, following a uniform PCOS diagnostic criteria by the limited staff, and adjustments of the results for multiple potential confounders are among the major strengths of the study. The study limitations include its modest sample size, and case–control study associated recall-bias may also be a concern in this study, though the latter is unlikely to affect the outcome.
Conclusion
We report a significant association of TCF7L2 polymorphic variant rs12255372 with the PCOS risk. The association of TCF7L2 polymorphism with risk PCOS is poorly understood and needs to be replicated in other studies with a larger sample size.
Acknowledgements
The authors thank all the participants for volunteering in the study. The authors also thank the Multi-disciplinary Research Unit, SKIMS, Srinagar funded by the Department of Health Research, Govt of India, for providing necessary research facilities for carrying out this study.
Author contribution
MA Ganie conceived the study; MA Ganie, BA Ganai and M Godha designed the study. R Rashid and A Rashid collected the data, R Rashid and M J Makhdoomi performed experiments. IA Shah, R Rashid and MA Ganie analyzed and interpreted the data. R Rashid, IA Shah, and MA Ganie wrote the first draft of the manuscript. All the authors reviewed and approved the final draft of the manuscript.
Declarations
Conflict of interest
All the authors declare that they do not have any competing financial and/or non-financial or any conflicting interests about the work described.
Footnotes
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Azziz R. Polycystic ovary syndrome. Obstet Gynecol. 2018;132(2):321–336. doi: 10.1097/AOG.0000000000002698. [DOI] [PubMed] [Google Scholar]
- 2.Sanchez-Garrido MA, Tena-Sempere M. Metabolic dysfunction in polycystic ovary syndrome: pathogenic role of androgen excess and potential therapeutic strategies. Mol Metab. 2020;35:100937. doi: 10.1016/j.molmet.2020.01.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Ganie MA, Rashid A, Sahu D, Nisar S, Wani IA, Khan J. Prevalence of polycystic ovary syndrome (PCOS) among reproductive age women from Kashmir valley: a cross-sectional study. Int J Gynecol Obstet. 2020;149(2):231–236. doi: 10.1002/ijgo.13125. [DOI] [PubMed] [Google Scholar]
- 4.Witchel SF, Oberfield SE, Peña AS. Polycystic ovary syndrome: pathophysiology, presentation, and treatment with emphasis on adolescent girls. J Endocr Soc. 2019;3(8):1545–1573. doi: 10.1210/js.2019-00078. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Jeanes YM, Reeves S. Metabolic consequences of obesity and insulin resistance in polycystic ovary syndrome: diagnostic and methodological challenges. Nutr Res Rev. 2017;30(1):97–105. doi: 10.1017/S0954422416000287. [DOI] [PubMed] [Google Scholar]
- 6.Barthelmess EK, Naz RK. Polycystic ovary syndrome: current status and future perspective. Front Biosci (Elite Ed) 2014;6:104. doi: 10.2741/e695. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Ganie MA, Dhingra A, Nisar S, Sreenivas V, Shah ZA, Rashid A, et al. Oral glucose tolerance test significantly impacts the prevalence of abnormal glucose tolerance among Indian women with polycystic ovary syndrome: lessons from a large database of two tertiary care centers on the Indian subcontinent. Fertil Steril. 2016;105(1):194–201.e3. doi: 10.1016/j.fertnstert.2015.09.005. [DOI] [PubMed] [Google Scholar]
- 8.Rodgers R, Avery J, Moore V, Davies M, Azziz R, Stener-Victorin E, et al. Complex diseases and co-morbidities: polycystic ovary syndrome and type 2 diabetes mellitus. Endocrine Connections. 2019;8(3):R71–R75. doi: 10.1530/EC-18-0502. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Fitzgerald S, DiVasta A, Gooding H. An update on PCOS in adolescents. Curr Opin Pediatr. 2018;30(4):459–465. doi: 10.1097/MOP.0000000000000636. [DOI] [PubMed] [Google Scholar]
- 10.Bogari NM. Genetic construction between polycystic ovarian syndrome and type 2 diabetes. Saudi J Biol Sci. 2020;27(10):2539–2543. doi: 10.1016/j.sjbs.2020.05.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.del Bosque-Plata L, Martínez-Martínez E, Espinoza-Camacho MÁ, Gragnoli C. The role of TCF7L2 in type 2 diabetes. Diabetes. 2021;70:1220–1228. doi: 10.2337/db20-0573. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Liu P-H, Chang Y-C, Jiang Y-D, Chen WJ, Chang T-J, Kuo S-S, et al. Genetic variants of TCF7L2 are associated with insulin resistance and related metabolic phenotypes in Taiwanese adolescents and Caucasian young adults. Oxford: Oxford University Press; 2009. [DOI] [PubMed] [Google Scholar]
- 13.ESHRE TR, Group A-SPCW. Revised 2003 consensus on diagnostic criteria and long-term health risks related to polycystic ovary syndrome. Fertility and sterility. 2004;81(1):19–25. [DOI] [PubMed]
- 14.Rashid R, Shah IA, Asrar MM, Godha M, Ganai BA, Ganie MA. Family history of menstrual irregularity or diabetes mellitus enhances the susceptibility to polycystic ovary syndrome among subjects harboring rs7903146 genetic variant of TCF7L2. J Diabetes Metab Disord. 2022;21(1):769–776. doi: 10.1007/s40200-022-01050-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Makhdoomi MJ, Shah IA, Rashid R, Rashid A, singh S, Shah ZA, et al. Effect modification of LHCGR gene variant (rs2293275) on clinico-biochemical profile, and levels of luteinizing hormone in polycystic ovary syndrome patients. Biochem Genet. 2023;Ahead of print. [DOI] [PubMed]
- 16.Yildiz BO, Bolour S, Woods K, Moore A, Azziz R. Visually scoring hirsutism. Hum Reprod Update. 2010;16(1):51–64. doi: 10.1093/humupd/dmp024. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Nanfa D, Sobngwi E, Atogho-Tiedeu B, Noubiap JJN, Donfack OS, Mofo EPM, et al. Association between the TCF7L2 rs12255372 (G/T) gene polymorphism and type 2 diabetes mellitus in a Cameroonian population: a pilot study. Clin Transl Med. 2015;4(1):1–5. doi: 10.1186/s40169-015-0058-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Jain A, Bhoyar RC, Pandhare K, Mishra A, Sharma D, Imran M, et al. IndiGenomes: a comprehensive resource of genetic variants from over 1000 Indian genomes. Nucleic Acids Res. 2021;49(D1):D1225–D1232. doi: 10.1093/nar/gkaa923. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.del Bosque‐Plata L, Hernández‐Cortés EP, Gragnoli C. The broad pathogenetic role of TCF7L2 in human diseases beyond type 2 diabetes. J Cell Physiol. 2021;237(1):301–312. doi: 10.1002/jcp.30581. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Biyasheva A, Legro RS, Dunaif A, Urbanek M. Evidence for association between polycystic ovary syndrome (PCOS) and TCF7L2 and glucose intolerance in women with PCOS and TCF7L2. J Clin Endocrinol Metab. 2009;94(7):2617–2625. doi: 10.1210/jc.2008-1664. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Shen W-J, Li T-R, Hu Y-J, Liu H-B, Song M. Relationships between TCF7L2 genetic polymorphisms and polycystic ovary syndrome risk: a meta-analysis. Metab Syndr Relat Disord. 2014;12(4):210–219. doi: 10.1089/met.2014.0004. [DOI] [PubMed] [Google Scholar]
- 22.Wang X, Wang K, Yan J, Wu M. A meta-analysis on associations of FTO, MTHFR and TCF7L2 polymorphisms with polycystic ovary syndrome. Genomics. 2020;112(2):1516–1521. doi: 10.1016/j.ygeno.2019.08.023. [DOI] [PubMed] [Google Scholar]
- 23.Rasool SUA, Ashraf S, Nabi M, Rashid F, Masoodi SR, Fazili KM, et al. Insulin gene VNTR class III allele is a risk factor for insulin resistance in Kashmiri women with polycystic ovary syndrome. Meta Gene. 2019;21:100597. doi: 10.1016/j.mgene.2019.100597. [DOI] [Google Scholar]
- 24.Ganie MA, Hassan S, Nisar S, Shamas N, Rashid A, Ahmed I, et al. 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) Gynecol Endocrinol. 2014;30(11):781–784. doi: 10.3109/09513590.2014.924099. [DOI] [PubMed] [Google Scholar]
- 25.Diamanti-Kandarakis E, Dunaif A. Insulin resistance and the polycystic ovary syndrome revisited: an update on mechanisms and implications. Endocr Rev. 2012;33(6):981–1030. doi: 10.1210/er.2011-1034. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Rosenfield RL, Ehrmann DA. The pathogenesis of polycystic ovary syndrome (PCOS): the hypothesis of PCOS as functional ovarian hyperandrogenism revisited. Endocr Rev. 2016;37(5):467–520. doi: 10.1210/er.2015-1104. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Bruni V, Capozzi A, Lello S. The role of genetics, epigenetics and lifestyle in polycystic ovary syndrome development: the state of the art. Reprod Sci. 2021;1–12.29(3):668–679. doi: 10.1007/s43032-021-00515-4. [DOI] [PubMed] [Google Scholar]
- 28.Carroll J, Saxena R, Welt CK. Environmental and genetic factors influence age at menarche in women with polycystic ovary syndrome. J Pediatr Endocrinol Metab. 2012;25(5–6):459–466. doi: 10.1515/jpem-2012-0047. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Ashraf S, Nabi M, Rashid F, Amin S. Hyperandrogenism in polycystic ovarian syndrome and role of CYP gene variants: a review. Egypt J Med Hum Genet. 2019;20(1):1–10. doi: 10.1186/s43042-019-0031-4. [DOI] [Google Scholar]
- 30.Nestler JE, Jakubowicz DJ, Falcon de Vargas A, Brik C, Quintero N, Medina F. Insulin stimulates testosterone biosynthesis by human thecal cells from women with polycystic ovary syndrome by activating its own receptor and using inositolglycan mediators as the signal transduction system. J Clin Endocrinol Metab. 1998;83(6):2001–2005. doi: 10.1210/jcem.83.6.4886. [DOI] [PubMed] [Google Scholar]