Family-Based Analysis of -675 4G/5G Polymorphism in the PAI-1 Gene in Chinese Polycystic Ovary Syndrome


 Background: Insulin resistance (IR) is one of the main pathophysiology of PCOS. Previous studies have shown that 4G/5G polymorphism in promoter region of plasminogen activator inhibitor-1 gene(PAI-1) can affect insulin sensitivity by elevating the level and activity of plasma PAI-1, involved in the formation of IR. In order to elucidate the relationship between 4G/5G polymorphism of PAI-1 gene and PCOS, we used transmission disequilibrium test (TDT) to study the nuclear family of PCOS. Purpose: This study used family-based analysis of TDT to determine whether an association exists between 4G/5G polymorphism of PAI-1 gene and PCOS and in Han Chinese in order to identify PAI-1 gene as a genetic susceptibility factor for PCOS. Methods: Eight hundred and fifty-five participants consisting of 285 trios (mother, father and offspring with PCOS) were recruited from the Center for Reproductive Medicine, Shandong University, from July 2007 to August 2014. 4G/5G polymorphism of PAI-1 gene was genotyped using direct sequencing protocol and TDT was used to analyze the association between PAI-1 gene and PCOS. Results: Though the 5G allele in PAI-1 gene was overtransmitted in families, no statistical significance existed and there was no association between PAI-1 gene and PCOS. The levels of FINS, HOMA-IR and TG were significantly different between obese and lean PCOS and the parents.Conclusion: No significant evidence of association or linkage was found between 4G/5G polymorphism of PAI-1 gene and PCOS, indicating that PAI-1 gene was unlikely to play a major role in the etiology of PCOS in our sample.


Introduction
Polycystic ovary syndrome (PCOS) is a complex disease involving multiple organs, reproductive-endocrine and metabolic abnormalities, which affect up to 4%-18% of women of reproductive age (1,2). The widely accepted pathophysiologic mechanisms were insulin resistance and hyperandrogenism, which could lead to a series of metabolic disorders and long-term complications such as type 2 diabetes mellitus, dyslipidemia, metabolic syndrome, premature arteriosclerosis and endometrial cancer, and so on (3)(4)(5)(6)(7).
Epidemiology studies suggested that the complex interactions of genetic factors and environmental resulted in the arises of PCOS (8-10). A large number studies have shown that 4G/5G polymorphism in promoter region of Plasminogen activator inhibitor-1 gene(PAI-1) can affect insulin sensitivity by elevating the level and activity of plasma PAI-1, participating in the formation of insulin resistance(IR). Current studies make no consensus on the relationship between PAI-1 gene 4G/5G polymorphism and PCOS. In order to make further con rmation of relationship between PAI-1 gene 4G/5G polymorphism and PCOS, we use transmission disequilibrium test(TDT)to study the 285 nuclear family (mother, father and offspring with PCOS, 855 participants in total) of PCOS. TDT can guarantee the coherency of the all objects in same genetic background, eliminates the differences of traditional case-control study on genetic and environmental factors, and can eliminate the false positive results.

Materials And Methods
Subjects A total of 855 participants, consisting of 285 trios (mother, father and offspring with PCOS) were recruited from the center for Reproductive Medicine, Shandong Provincial Hospital, A liated to Shandong University, between July 2007 to August 2014. They were all Han Chinese in Shandong province and signed Informed consent for molecular studies. The study was approved by the Ethics Committee of Shandong University. PCOS was diagnosed according to the 2003 Rotterdam criteria, at least two of the following three features: oligomenorrhea or amenorrhea, clinical or biochemical hyperandrogenism, and po1ycystic ovarian morphology on ultrasound. Other related disease, such as Cushing' syndrome, hypothyroidism or signi cant elevations of serum prolactin(PRL), non-classic 21-hydroxylase de ciency, adrenal congenital hyperplasia and estrogen-secreting tumors were excluded (11,12). Based on body mass index(BMI), PCOS patients were divided into two groups: group A (BMI ≥ 25kg/m 2 ) and group B (< 25kg/m 2 ). According to corresponding categories of PCOS patients,their parents were divided into group C versus D (fathers) and group E versus F(mothers) .

Methods
Anthropometrical variables, such as waist circumference(WC), hip circumference(HC), body height and weight were collected in all patients during the rst visit to the outpatient department. Blood samples of patients with PCOS were obtained on days 3-5 of their menstrual cycle. All probands underwent a 75g oral glucose tolerance test. Fasting blood glucose (FBG) and fasting insulin (FINS) and 2-h insulin levels were detected using chemiluminescence immunoassay(endocrine hormone determination kit; Haier Import and Export, Ningbo, China) and 2-h glucose level were detected with the use of the oxidase method (AU640 automatic biochemistry analyzer; Olympus company, Hamburg, Germany), and the serum levels of folliclestimulating hormone(FSH), luteinizing hormone(LH), total testosterone(TT) and estrogen were measured enzymatically on an automated biochemistry analyzer (Olympus 600, Clinical Chemistry Analyzer; Olympus Diagnostica GmbH, Co.Clare, Ireland). The homeostasis-model assessment for insulin resistance (HOMA-IR) was calculated according to the following equation: fasting blood glucose (FBG, mmol/L) ×fasting insulin (FINS, mU/L)/22.5. Serum total cholesterol (TC), triglycerides (TG), low-density lipoprotein (LDL) and high-density lipoprotein (HDL) cholesterol concentrations were measured by the precipitation and enzymatic method (Ft-7060; Beckman Coulter, Galway, Ireland).
Reactions were carried out under the following conditions: initial denaturation at 95℃ for 5 min followed by 35 cycles of denaturation at 95℃ for 30s, annealing for 30s at 60℃, extension at 72℃ for 45s, and nally 72℃ for 5 min. The polymerase chain reaction products were analyzed by melting curve rst.
Outliers lacking high peaks during melting curve were alternately analyzed by 1% agarose gel electrophoresis (AGE) and then sequenced on an automated sequencer (ABI PRISM 310; Applied Biosystems, Foster City, CA). All samples were double genotyped with 100% concordance.

Statistical analyses
Descriptive statistics for − 675 4G/5G polymorphisms of the PAI-1gene, including minor allele frequency (MAF) and Hardy-Weinberg equilibrium, were obtained by Haploview 4.2 (12). Then the association between the − 675 4G/5G polymorphisms in the PAI-1 gene and PCOS was tested by the TDT analysis, which was carried out by Haploview 4.2 too. Statistical signi cance was considered at the two-tailed level of P = 0.05. TDT is a valid chi-squared test for the linkage hypothesis regardless of population history. In the TDT test, by collecting related PCOS family trios, we analyzed the difference between the probability of parents-to-offspring transmission and the hypothesis of no association (probability of transmission 50%).
If the signi cant discrepancy existed, indicated an association between PAI-1 gene polymorphisms and PCOS.
The data of clinical and biochemical variables among different subgroups was analyzed by independentsamples t-test using the Statistical Package for Social Sciences (SPSS) version 22.0. Categorical data were expressed as frequencies and percentages. Descriptive characteristics were reported as mean ± SD. Genotypic and allelic distributions were compared using the Pearson chi-squared test and one-way analysis of variance in the PCOS group. P < 0.05 were considered statistically signi cant.

Clinical and metabolic characteristics
As shown in Table 1 The clinical and metabolic characteristics of PCOS's parents were shown in Table 3. WC, HC, SBP, DBP, FINS, LDL, TG, HOMA were signi cant different Both in group C and D, E and F (P < 0.05).   Clinical and metabolic characteristics of the patients with PCOS according to the genotype of -675 4G/5G polymorphisms(rs1799889) of the PAI-1 gene The clinical and metabolic characteristics of the patients with PCOS according to the genotype of PAI-1 (rs1799889) were shown in Table 4, no signi cant differences were observed between the different genotypes. There was no signi cance difference in genotypes of PAI-1 between the two groups (chi-squared = 2.7002, P = 0.2592), and no signi cance difference in allele frequencies existed too (chi-squared = 1.8023, P = 0.018).

Hardy-Weinberg equilibrium Test
In our study, the gene transmission of PAI-1(4G/5G, rs1799889) coincided with Hardy-Weinberg equilibrium, as shown in Table 6 .

Discussion
As a complex and polygene related neuroendocrine disease, polycystic ovary syndrome (PCOS) is in uenced by environmental and genetic factors, and show the general metabolic abnormalities and the abnormal ovarian local regulation. PCOS has short-term and long-term complications such as obesity, diabetes, dyslipidemia, infertility, and the risk of cardiovascular disease. Besides these, elevated activation of the blood coagulation system of PCOS has been reported (14).
PAI-1, was an important regulator of the endogenous brinolytic system. Recent studies reported that the 4G/5G polymorphism in the promoter of PAI-1 gene might relate to the increased risk of PCOS, while the conclusions were not consistent. Sales reported that PAI-1 4G/5G polymorphism might increase the susceptibility of PCOS in Brazil, Asians, Europe and Turkey, but not in Caucasian population (9,(15)(16).
Wang L reported that PAI-1 4G/5G polymorphism may be a major key in the pathogenesis of PCOS among chinses women (17). While TT Zhang had a contrary conclusion, their meta-analysis included a total of 11 case-control studies and found there were no association between PAI-1 4G/5G and PCOS risk under all models (18).
The discrepancy of the previous studies might ascribe to the population strati cation and environmental factors. To avoid these possible impact factors, we conducted the 285 family trios and used the transmission disequilibrium test (TDT) to verify the linkage between PAI-1 4G/5G polymorphism and PCOS. Our data showed that the HWEp was 0.4742 and the MAFs of the PAI-1 4G/5G was 0.325 by the TDT with no statistical signi cance (χ2 = 0.388, P = 0.5336). We failed to nd the association between PAI-1 4G/5G and PCOS. The conclusion was consistent with the results of TT Zhang (18). While our previous case-control studies showed that PAI-1 4G/5G polymorphism was associated with PCOS women in Han Chinese. The difference reminded us that we needed joint multicenter and collected more samples for further validation.
Michalska M reported that adipose tissue especially visceral adipose tissue could generate PAI-1 and led to an elevated blood level of PAI-1, speculating the association between obesity and PAI-1 (19). Ma found that PAI-1 and obesity rats had a direct causal relationship (20). One study found that obese PCOS patients had higher blood PAI-1 levels and IR was more common than non-obese PCOS patients (9). Although the relation between obesity and incidence of PCOS has not been well elucidated, the close relationship between infertility, hyperandrogenism, glucolipid metabolic abnormalities and obesity are validated. Our study also concentrated on the relationship between PAI-1 and obesity, we divided the subjects into obesity and non-obesity, we found that WC, HC, SBP, DBP, FINS, LDL, TG, HOMA in obesity patients and their parents were higher than non-obesity, and FINS, 2-h insulin, HOMA signi cantly increased especially (P < 0.001). while FBG had no signi cation between the two groups. The results veri ed that IR widely existed in both obese PCOS patients and their parents, and the relative insu ciency of insulin led to the compensatory increased insulin secretion and hyperinsulinemia. Additionally, we also speculated that obese of PCOS had family genetic predisposition.

Con icts of interest/Competing interests
There is no con ict of interest.
Availability of data and material

Data transparency
Code availability

Not applicable
Authors' contributions Xiaocui Song was responsible for the design of total study and results analysis. Li Ge mainly concentrated on the collection of samples and the manuscript writing. Dongsha Wang, Li Li and Dongmei Ma were responsible for the data input and relevant information collection of PCOS trios.

Ethics approval
The study was approved by the Ethics Committee of Shandong University.

Consent to participate
All participants signed informed consent.

Consent for publication
All authors approved the publication of the manuscript.