Association of Polymorphisms in Plasminogen Activator Inhibitor-1 (PAI-1), Tissue Plasminogen Activator (tPA), and Renin (REN) with Recurrent Pregnancy Loss in Korean Women

Recurrent pregnancy loss (RPL) is defined as two or more consecutive pregnancy losses prior to 20 weeks of gestational age. Various factors, including immune dysfunction, endocrine disorders, coagulation abnormality, and genetic disorders influence RPL. In particular, plasminogen activator inhibitor-1 (PAI-1), tissue plasminogen activator (tPA), and renin (REN) have important roles in the thrombotic and thrombolytic systems, and abnormal expression of these genes have a reported negative correlation with pregnancy maintenance. Moreover, some polymorphisms of the three genes are related to expression levels and thrombotic disorder. Therefore, we investigated whether polymorphisms of PAI-1, tPA, and REN are linked to RPL. Genotyping of the six polymorphisms (PAI-1 rs11178, rs1050955, tPA rs4646972, rs2020918, REN rs1464816, and rs5707) was performed using polymerase chain reaction (PCR)-restriction fragment length polymorphism and associations of the polymorphisms with RPL were evaluated by statistical analysis. The polymorphism PAI-1 rs1050955 GA+AA was associated with decreased RPL risk (AOR, 0.528; 95% CI 0.356–0.781; p = 0.001) as was the REN 10795 rs5707 GG genotype (AOR, 0.487; 95% CI 0.301–0.787; p = 0.003). In contrast, the tPA rs4646972 II genotype correlated with increased RPL risk (AOR, 1.606; 95% CI, 1.047–2.463; p = 0.030). This study provides evidence that tPA Alu rs4646972 may contribute to the risk of idiopathic RPL, but PAI-1 12068 rs1050955 and REN 10795 rs5707 are associated with a decreased risk of RPL. Therefore, these alleles may be useful as biomarkers to evaluate the risk of RPL.


Introduction
About 10-12% of pregnant women experience early miscarriages, and between 1% and 4% of all pregnant women experience recurrent pregnancy loss (RPL) [1]. While the etiologies of RPL are unclear in a majority of RPL patients, many are suspected to have thrombophilic problems [2]. Hypercoagulability, resulting from an increase in coagulation factors, a decrease in blood flow, and a decrease in the activity levels of natural anticoagulants, leads to an increased susceptibility to venous thromboembolism in pregnant women. Hypofibrinolysis and thrombophilia can be risk factors in RPL and infertility [3]. An imbalance in the dynamic equilibrium between fibrin formation and fibrinolysis will result in a prothrombotic state [4] and thrombophilia, with the formation of a microthrombosis at the embryonic implantation site, resulting in implantation failure and RPL [5].
Systemic or local inflammation play an important role for normal embryo implantation and normal pregnancy development [6]. In particular, the homeostatic balance between pro-inflammatory Th1 cytokines including interferon-γ (IFN-γ) and tumor necrosis factorα (TNF-α) and anti-inflammatory Th2 such as interleukin-10 (IL-10) is very important for successful pregnancy [7]. TNF-α is encoded on chromosome 6 and produced by NK cells, CD4+, CD8+, macrophages, monocytes, mast cells, neutrophils, and endothelial cells. Nuclear transcription factor-κB (NF-κB) regulates gene expressions for cytokines including IL-1, IL-6, IL-8, and TNF-α and cell adhesion molecules such as ICAM-1, VCAM-1, therefore promoting vascular inflammation and contributing the activation of the hemostatic pathway. Dysregulation of TNF-α leads to vascular dysfunction and pro-coagulant statuses that are related to RPL by regulating IL-6, PAI-1 while decreasing the expression of thrombomodulin [8].
PAI-1, also known as the SERPINE1 gene, contains nine exons and encodes the primary inhibitor of t-PA. The PAI-1 rs1050955 polymorphism, located at an miRNA binding site of the PAI-1 gene, may change the expression level of the gene [9]. The gene for plasminogen activator inhibitor-1 (PAI-1), encoding a principal inhibitor of tissue plasminogen activator (tPA) and urokinase (uPA), has been mapped to chromosome 7 (7q21. [3][4][5][6][7][8][9][10][11][12][13][14][15][16][17][18][19][20][21][22]. Increased concentrations of PAI-1, which plays a role in the conversion of plasminogen to the proteolytic enzyme plasmin [10], affect extracellular proteolysis during ovulation and embryo implantation [11]. The balance between PAI-1 and tPA determines the extent of plasminogen activation and fibrin degradation. The PAI-1 polymorphism -675 ID, 4G/5G is correlated with an increased risk of RPL [5]. Moreover, the PAI-1 -675 4G/5G and tPA Alu polymorphisms have been associated with thrombotic disorders such as RPL [12], strokes [13], and myocardial infarction [14]. The fibrinolytic system is mainly regulated by tPA and its inhibitor PAI-1, but also by the renin-angiotensin system (RAS), which is a major regulator of blood pressure, sodium balance, and urinary excretion of potassium. The renin gene (REN), located on chromosome 1 (1q 32-1q 42), includes a sequence for steroid receptors in its promoter region [15]. Polymorphism in REN are associated with high blood pressure and heart failure [16] and in pregnancy complications such as preterm delivery, premature rupture of the membrane, and preeclampsia [15][16][17][18]. RAS control the fibrinolytic system through regulation of plasma PAI-1 and tPA, and this controlling may be affected by genetic variations in RAS and fibrinolytic systems [19,20]. Since interactions between the fibrinolytic system and the RAS could affect the risk of RPL, we hypothesized that polymorphisms in REN, PAI-1, and tPA would be associated with the risk of RPL.

Results
Analysis of the demographic characteristics and laboratory test values for the RPL and control groups revealed significant differences in the gestation age at the time of the termination of pregnancy and the PT between the two groups ( Table 1).

Discussion
Successful embryo implantation depends on the regulation of coagulation and fibrinolysis to avoid excess fibrin accumulation in placental vessels and intervillous spaces [21,22]. Overexpression of PAI-1 is associated with thrombus formation in various types of blood vessels [23] and adverse pregnancy complications including miscarriage, stillbirth, fetal growth restriction, and placental abruption [24]. We found that in Korean women with a history of RPL, PAI-1 polymorphisms were associated with idiopathic RPL and with increased plasma concentrations of vascular risk factors such as homocysteine, folate, and urate [25]. Also, chronically elevated PAI-1 levels due to the PAI-1 -675 4G allele may affect folliculogenesis and induce ischemic damage to the ovary, leading to primary ovarian insufficiency [26]. High levels of PAI-1 and PLT as well as shorter PT and aPTT are associated with hypofibrinolytic status [27,28]. We demonstrated a negative relationship between RPL and PAI-1 12068 rs1050955 GA+AA or REN 10795 rs5707 GG, which may be hemophilic or hyperfibrinolytic genotypes that prolong the PT and aPTT.
The miRNA, by base-pairing with target mRNAs at the 3 -untranslated region (UTR), leads to translational repression of target genes. The GA+AA genotype in PAI-1 rs1050955 was associated with a decreased risk of RPL compared with the GG genotype. PAI-1 rs2227631 and rs1050955 likely modulate different transcriptional levels of PAI-1, but further studies will be needed to better understand the mechanisms that control PAI-1. Since the PAI-1 10692 rs11178 polymorphism did not show an association with susceptibility to RPL, this polymorphism may regulate PAI-1 differently than the rs1050955 polymorphism, even though both polymorphisms are located in the 3 -UTR of PAI-1.
The tPA Alu (I/D) polymorphism, which is characterized by the presence or absence of a 311-bp Alu retrotransposon in the 8th intron, was more prevalent among RPL patients compared to the control group. Homozygous carriers of the tPA Alu polymorphism show an increase in the release of tPA from vascular endothelial cells [29]. The amount of tPA may be an important factor in the early stages of placental development and placental separation from maternal tissue at delivery [30]. An impaired fibrinolytic system could cause recurrent pregnancy loss through a defect in trophoblast development or excess fibrin deposition in early placentation [31]. This impairment of fibrinolysis could result from significantly higher levels of tPA in patients with tPA Alu rs4646972 II compared to Alu rs4646972 DD or DI [29]. The free tPA that is released into blood from endothelial cells immediately combines with circulating PAI-1 forming a PAI-1/tPA complex, whose concentrations in plasma correlate strongly with levels of tPA antigen and PAI-1 activity [32]. The importance of the fibrinolytic system in RPL is supported by the work of Magdoud et al. [33] who found that the −675 4G/5G polymorphism in the PAI-1 promoter (a) induced high PAI-1 expression levels, (b) suppressed fibrinolysis and promoted thrombosis through inhibition of plasminogen activator, and (c) was associated with an increased risk of RPL. The increased RPL risk that is associated with the tPA Alu polymorphism might be explained by enhanced local recombination, which is characteristic of all Alu elements, if the tPA Alu polymorphism is in linkage disequilibrium with an unidentified polymorphism in a coding area of the tPA gene that directly affects the protein [34]. However, we did not identify an association of any clinical parameters with tPA Alu rs4646972 polymorphism.
The GG genotype of REN 10975 rs5707 polymorphism was associated with a decreased risk of RPL in women with either 2 or ≥3 pregnancy losses. Since the REN gene is an attractive genetic marker candidate for hypertension associated disease, the REN rs5707 polymorphism has been studied in the development of coronary artery disease and preeclampsia, but those studies did not reveal a significant relationship [16,35]. Therefore, the association of REN rs5707 polymorphism with RPL is more valuable, and further studies to validate the function of the REN rs5707 in RPL are needed.
Several gene polymorphisms affect the risk of RPL, such as matrix metalloproteinase genes [36], the methylenetetrahydrofolate reductase (MTHFR) gene [37], complement factors D and H [38], and genome wide methylation analysis [39], reflecting the complicated mechanisms that underlie the development of RPL. Previous studies have investigated the association of high levels of homocysteine and low levels of folate with the risk of RPL [40,41]. Increased levels of homocysteine and/or decreased levels of folate in the blood have been related to reduced expression of the enzyme MTHFR that catalyzes the process of converting 5,10-methylenetetrahydrofolate to 5-methyltetrahydrofolate and is a critical regulatory enzyme involved in folate metabolism. Kim et al. investigated the association of four 3 -UTR of the MTHFR gene with the risk of RPL [37]. They reported that the 2572C>A, 4869C>G, 5488C>T, and 6685T>C MTHFR polymorphisms were significantly associated with RPL susceptibility. In particular, MTHFR 4869C>G and 5488C>T genotypes were related to elevated homocysteine levels and reduced folate levels in plasma. We have identified an association between the fibrinolytic system and susceptibility to RPL; however, this study has some limitations. First, this study did not reveal a relationship between RPL and clinical factors such as homocysteine, folate, and PLT. Second, the mechanisms by which the PAI-1, tPA, and REN polymorphisms affect the development of RPL are not clear. Third, the study population included only Korean women, so we do not know whether these same gene polymorphism results apply to other populations. Fourth, we cannot rule out the cases where RPL is caused by other causes, such as chromosomal abnormalities, infection, etc., that have not been evaluated despite the tests for other causes of RPL. Future studies with a larger and more heterogeneous population are needed to confirm our understanding of the influence of the fibrinolytic system on the development of RPL.
In conclusion, to the best of our knowledge, this is the first study to evaluate the association of PAI-1, tPA, and REN gene polymorphisms with the susceptibility of Korean women to RPL. Our findings suggest that polymorphisms in PAI-1, tPA, and REN may contribute to RPL and can be developed as biomarkers to evaluate a woman's susceptibility to RPL.

Subjects
This study included 206 women in the control group and 334 women with RPL at the Infertility Medical Center of CHA Bundang Medical Center. RPL patients had experienced two or more consecutive spontaneous abortions confirmed by human chorionic gonadotropin levels, ultrasonography, and physical examinations. Patients who used alcohol or smoked were excluded from this study. The women in the control group had a history of at least one successful, naturally-conceived pregnancy and no history of pregnancy loss. Hormonal tests were performed to determine hormonal causes of miscarriage such as thyroid disease, hyperprolactinemia, or luteal insufficiency. Anatomical abnormalities associated with RPL were determined by hysteroscopy, hysterosalpingography, computed tomography (CT), or magnetic resonance imaging (MRI). Possible autoimmune diseases as the cause of miscarriage, such as antiphospholipid syndrome or lupus, were evaluated using lupus anticoagulant and anticardiolipin antibodies. The institutional review board in CHA Bundang Medical Center approved this study and written informed consent was obtained from all participants (IRB number: 2010-01-123).

Genotyping
Genomic DNA (gDNA) was extracted from anticoagulated peripheral blood using a G-DEX II Genomic DNA Extraction Kit (Intron Biotechnology, Seongnam, Korea) according to the manufacturer's instructions. The concentration and purity of gDNA, which were dissolved in 100 µL of DNA rehydration buffer (Intron Biotechnology, Seongnam, Korea), were measured using nanodrop 1000 (Thermo Fisher Scientific, Waltham, MA, USA). The concentration of all gDNA were above 100 ng/µL and the purity of the gDNA ranged from 1.8 to 2.1 at the 260/280 nm ratio and from 1.9 to 2.2 at the 260/230 nm ratio. gDNA samples were diluted in DNA rehydration buffer to a concentration of 100 ng/µL. The genotypes of the six polymorphisms were determined by polymerase chain reaction (PCR)-restriction fragment length polymorphism (RFLP) analysis with 1 µL of gDNA sample. The six primer sets were designed for easy genotype discrimination using RFLP (Table S4). The PCR was performed with 8 negative controls using distilled water instead of DNA samples per 96 tubes. The PAI-1 10692 rs11178 polymorphism was detected using the forward primer (5 -AGA TCT GTC TCC AAG ACC TTG-3 )  . The annealing temperature for the two polymorphisms in tPA was 58 • C and for the other polymorphisms was 55 • C. For RFLP analysis, the PCR products of PAI-1 10692, PAI-1 12068, tPA -7351, REN 6567, and REN 10795 were digested with BtsCI, BstZ17I, BanII, MluCI, and AvaI, respectively, for 16 h at the recommended temperature. The enzyme was inactivated, and the DNA bands were visualized by electrophoresis in a 3% agarose gel containing DNA staining solution. The RFLP patterns of each polymorphism are presented in Table S4. For each polymorphism, 30% of the PCR analyses were randomly selected and repeated for validation of the PCR-RFLP assay.

Assessment of Clinical Characteristics and Blood Coagulation Status from Blood Samples
Plasma homocysteine, folate, platelet (PLT), activated partial thromboplastin (aPTT), natural killer (NK) cell, prothrombin time (PT), uric acid, total cholesterol, PAI-1, and vascular endothelial growth factor (VEGF) levels were determined in RPL patients after 12 h of fasting. Homocysteine was determined by a fluorescence polarization immunoassay with the Abbott IMx analyzer (Abbott Laboratories, Abbott Park, IL, USA). Folate levels were determined by a competitive immunoassay with ACS:180 (Bayer Diagnostics, Tarrytown, NY, USA). The uric acid and total cholesterol levels were measured with commercially available enzymatic colorimetric tests using the Roche/Hitachi Modular Pre-analytics Plus system (Roche Diagnostics, Mannheim, Germany). PLT, aPTT, PT, and PAI-1 were determined as described previously [36].

Measurement of Peripheral NK Cell Proportions
Estimations of NK cells were determined by flow cytometry with CellQuest software (BD FACSCalibur; BD Biosciences, Seoul, Korea). Fluorescently labeled (fluorescein isothiocyanate, phycoerythrin [PE], peridinin chlorophyll protein, and allophycocyanin) monoclonal antibodies specific for CD3, CD16, and CD56 were purchased from BD Biosciences. Anti-NKG2A-PE antibodies were obtained from Immunotech (Beckman Coulter, Fullerton, CA, USA). All antibodies were used in 1:1000 dilution. Peripheral blood mononuclear cells (2.5 × 10 5 ) were stained for cell-surface antigen expression at 4 • C in the dark for 30 min, washed twice in 2 mL of phosphate-buffered saline containing 1% bovine serum albumin and 0.01% sodium azide (FACS wash buffer) and subsequently fixed in 200 µL of a 1% formaldehyde solution (Sigma-Aldrich, St. Louis, MO, USA) before sorting.

Data Analysis
The two-sample t-test and the Mann-Whitney U test were used for the clinical profiles. The associations between PAI-1, t-PA, and REN gene polymorphisms and RPL susceptibility were determined by age-adjusted odds ratios (AOR) and 95% confidence intervals (CIs). Allele frequencies were determined to confirm deviations from the Hardy-Weinberg equilibrium. The allele combinations for the polymorphisms were estimated by the chi-square test and were adjusted using the false discovery rate correction. The association between each PAI-1, t-PA, and REN gene polymorphism and each clinical value (homocysteine, folate, PLT, aPTT, NK cell, PT, uric acid, total cholesterol, PAI-1, and VEGF) for the RPL patients were assessed by ANOVA and Kruskal-Wallis tests. We considered the p < 0.05 of all statistic tests as statistically significant. Analyses were performed using GraphPad Prism 4.0 (GraphPad Software Inc., San Diego, CA, USA) and Medcalc version 12.7.1.0 (Medcalc Software, Mariakerke, Belgium).

Supplementary Materials:
The following are available online at https://www.mdpi.com/article/10 .3390/jpm11121378/s1, Table S1: Genotype frequencies of PAI-1, t-PA and REN gene polymorphisms according to the number of RPL; Table S2: Genotype combination analysis of PAI-1, t-PA and REN gene polymorphisms between RPL patients and controls; Table S3: Allele combination analysis of PAI-1, tPA and REN gene polymorphisms in RPL and controls; Table S4: Information for PCR and RFLP analysis of each variants.