Analysis of the Association Between MicroRNA Biogenesis Gene Polymorphisms and Venous Thromboembolism in Koreans

Venous thromboembolism (VTE) involves the formation of a blood clot, typically in the deep veins of the leg or arm (deep vein thrombosis), which then travels via the circulatory system and ultimately lodges in the lungs, resulting in pulmonary embolism. A number of microRNAs (miRNAs) are well-known regulators of thrombosis and thrombolysis, and mutations in miRNA biogenesis genes, such as DICER1, DROSHA have been implicated in miRNA synthesis and function. We investigated the genetic association between polymorphisms in four miRNA biogenesis genes, DICER1 rs3742330A > G, DROSHA rs10719T > C, RAN rs14035C > T and XPO5 rs11077A > C, and VTE in 503 Koreans: 300 controls and 203 patients. Genotyping was assessed with polymerase chain reaction-restriction fragment length polymorphism assays. We detected associations between polymorphisms in RAN and XPO5 and VTE prevalence (RAN rs14035CC + CT versus TT: p = 0.018; XPO5 rs11077AA + AC versus CC: p < 0.001). Analysis of allele combinations of all four polymorphisms (DICER1, DROSHA, RAN, XPO5) revealed that A-T-T-A was associated with decreased VTE prevalence (p = 0.0002), and A-T-C-C was associated with increased VTE prevalence (p = 0.027). Moreover, in subjects with provoked VTE, the DROSHA rs10719T > C, polymorphism was associated with increased disease prevalence (TT versus TC + CC: p < 0.039). Our study demonstrates that RAN and XPO5 polymorphisms are associated with risk for VTE in Korean subjects.


Introduction
Venous thromboembolism (VTE), which involves deep vein thrombosis (DVT) followed by pulmonary embolism (PE), is the third most common cardiovascular disease following myocardial infarction and stroke [1]. DVT is a disease that occurs when blood clots form in the deep veins, usually in the legs. The thrombus can then become dislodged and migrate to form a blockage in the pulmonary veins, with 45.4% of DVT patients ultimately developing PE [2,3]. The incidence of VTE is higher in western populations than in Asians, although recently, annual increases in the incidence of VTE have been noted in Asian individuals, reaching levels as high as that of western people under certain circumstances [4,5]. VTE can be classified into two groups, known as provoked and unprovoked [6]. The provoked group includes patients with a known cause of a hypercoagulable state, including female hormones (estrogens or oral contraception), surgery (post-operative period, particularly for orthopedic susceptibility, and the RAN rs14035 polymorphic locus was significantly linked to post-stroke mortality [34]. Based on these findings, in this study, we selected four genes (DICER1, DROSHA, RAN, XPO5) among several miRNA biogenesis genes. The DROSHA gene is located on 15p13. 3 and DROSHA is a nuclear ribonulcease III (RNase III) enzyme, which cleaves a stem-loop pri-miRNAs to hairpin-shaped pre-miRNAs. DICER1 gene is located on 14q32.13 and DICER1 is another RNase III, which cleaves the pre-miRNA to intermediate miRNA duplex [35]. RAN gene is located on 12q24. 33 and RAN is a GTPase that makes a pre-miRNA/XPO5/Ran-GTP ternary complex and transports pre-miRNA to cytoplasm. XPO5 gene is located on 6p21.1 and XPO5 is a member of the karyopherin family and exports pre-miRNAs from the nucleus to cytoplasm [36]. Moreover, we investigated 3 -UTR polymorphisms of this four miRNA biogenesis gene, including DICER1 rs3742330A > G, DROSHA rs10719T > C, RAN rs14035C > T, and XPO5 rs11077A > C, which are independently or complementarily associated with disease prevalence and clinical outcomes in VTE patients.

Characteristics of the Study Population.
The demographic characteristics for VTE patients and controls are listed in Table 1. There were no significant differences in age, hypertension, hyperlipidemia, or smoking status between the two groups. In contrast, VTE was more prevalent among men, and the patient group included significantly more subjects with diabetes.

Genotype Frequencies of miRNA Biogenesis Genes
We investigated the prevalence of the DICER1 rs3742330A > G, DROSHA rs10719T > C, RAN rs14035C > T, and XPO5 rs11077A > C gene polymorphisms in VTE patients and control subjects. Table 2 shows the genotype distributions in each of these groups. We found that the XPO5 rs11077A > C polymorphism was associated with an increased risk for VTE (AA versus AC: adjusted odds ratio (AOR) = 2.522, 95% confidence interval (CI) = 1.564-4.067, p < 0.001; AA versus AC + CC: AOR = 2.493, 95% CI = 1.552-4.003, p < 0.001). In contrast, the RAN CT genotype was more frequent in control group than VTE patients, and the RAN rs14035C > T polymorphism was associated with reduced VTE risk (CC versus CT: AOR = 0.630, 95% CI = 0.425-0.935, p = 0.022; CC versus CT + TT: AOR = 0.627, 95% CI = 0.427-0.922, p = 0.018). The XPO5 rs11077A > C polymorphism showed the same increased AOR pattern in subjects with both provoked and unprovoked VTE, whereas the RAN rs14035C > T polymorphism also demonstrated a reduced AOR in subjects with unprovoked VTE. The DICER1 rs3742330A > G and DROSHA rs10719T > C polymorphisms were not significantly correlated with either total or unprovoked VTE occurrence. However, for provoked VTE, the DROSHA rs10719T > C polymorphism was associated with an increased disease risk (TT versus TC + CC: AOR = 2.460, 95% CI = 1.048-5.774, p = 0.039).

Genotype Combinations of miRNA Biogenesis Gene Polymorphisms
We then analyzed allele combinations of the four miRNA biogenesis genes polymorphisms in VTE patients and control subjects (Table 4).
When total VTE patients were compared with controls, the C-C (

Discussion
Hemostasis has the dual role of retaining blood flow and preventing inappropriate coagulation [11]. Previous studies have demonstrated that miRNAs and miRNA biogenesis genes (e.g., DICER1 and DROSHA) play important roles in vascular biology and hemostasis. For example, the platelet miRNA profiles are associated with platelet functions, miRNA biogenesis gene may alter platelet miRNA expression. When DICER1 gene, one of the miRNA biogenesis genes, was deleted, platelet miRNA profile was changed, and platelet reactivity was enhanced [37,38]. In particular, miRNA-145 is a significant factor in the prevention of thrombosis production and is present at decreased levels in VTE patients [14]. Despite the fact that the miRNA biogenesis gene haves been studied in other diseases [39][40][41]; however, the miRNA modulation mechanisms in VTE patients remain unclear. These miRNA biogenesis genes, including DICER1, DROSHA, RAN, and XPO5, play critical roles in miRNA production. As such, numerous other studies have found that miRNA biogenesis gene knockdown leads to a global reduction in mature miRNA levels. Those miRNAs for which decreased levels have been demonstrated are shown in Table S1 [23][24][25][26][27]. In this study, we analyzed the association between polymorphisms in each of the four miRNA biogenesis genes (DICER1, DROSHA, RAN, XPO5) and the risk of VTE in Korean subjects. From this analysis, we found that the RAN rs14035C > T and XPO5 rs11077A > C polymorphisms are associated with total VTE and unprovoked VTE susceptibility. Additionally, in our Genotype combination and allele combination data, we found that RAN rs14035 T allele was associated with decreased risk of VTE and XPO5 rs11077 C allele was associated with increased risk of VTE [36].
RAN is small Ras-related GTP-binding protein that plays an important role in the cell cycle. Specifically, RAN functions in the transport of molecules from the nucleus to the cytoplasm or from the cytoplasm to the nucleus through the nuclear pore complex, in a GTP-dependent manner [42]. In miRNA biology, RAN associates with a protein known as XPO5 to transport pre-miRNA precursors from the nucleus to the cytoplasm. Exportin (XPO) proteins are members of the karyopherin β family of transport factors, and XPO5 is one of the nucleo-cytoplasmic exportins. RAN, XPO5, and the pre-miRNA molecule form a complex, which functions to transport the pre-miRNA through the nuclear pore complex and into the cytoplasm. Additionally, there is a correlation between the expression of RAN and XPO5. In breast cancer, when RAN GTPase was overexpressed, XPO5 level was also significantly increased.
The XPO5 gene has been found to play a role in carcinogenesis, as it was reported that certain cancers (e.g., non-small-cell lung cancer, esophageal squamous cell carcinoma, gastric cancer, hepatocellular carcinoma, and thyroid cancer) express levels of XPO5 that are distinct from those detected in normal cells [43][44][45][46][47][48]. In particular, a number of studies have reported that XPO5 polymorphisms at the rs11077 locus affect disease development and patient survival in various cancers. For example, the rs11077AA genotype displayed a trend for high expression in ESCC tissues, and these high XPO5 expression levels were also associated with high survival rates among ESCC patients [45]. In addition, the XPO5 rs11077 polymorphism is associated with XPO5 stability and miRNA expression levels. One study reported that when a polymorphism rs11077 is present in XPO5, precursor miRNAs are trapped in the nucleus, and processing efficiency decreases. This results in reduced miRNA levels, leading to diminished miRNA target gene inhibition [27]. In another case, using a Renilla luciferase 3 -UTR reporter assay, the rs11077 CC genotype was shown to promote reduced Renilla expression system, and knockdown of XPO5 expression lead to reduced miRNA levels [43]. In addition, XPO5 functions to regulate expression of DICER1, another miRNA biogenesis factor that cleaves the pre-miRNA with a hairpin structure to miRNA; miRNA duplex form in the cytoplasm. Notably, it was shown that decreased levels of XPO5, lead to decreased DICER1 expression [49]. Overall, these data suggest that SNPs in miRNA biogenesis genes affect the expression of mature miRNAs and consequently, may also influence miRNA-mediated regulation.
Our data further indicate that the DROSHA rs10719T > C polymorphism is associated with the incidence of provoked VTE. DROSHA has previously been shown to play a role in vascular smooth muscle cell survival and angiogenesis [50], and provoked VTE results from risk factors associated with pregnancy, surgery, and hormones. Therefore, these findings suggest that in provoked VTE, DROSHA is significantly associated with vascular inflammation and recovery.
There are a number of limitations to this case-control study. For example, both the VTE patient and control sample sizes were small and the study population was comprised only of Korean individuals. We will therefore need to validate these data in other ethnic groups. In addition, although we found a significant association between the RAN rs14035 and XPO5 rs11077 polymorphisms and VTE, a proposed mechanism for the role of these polymorphisms in VTE prevalence is lacking. Thus, additional studies, potentially analyzing the effects of other miRNA biogenesis genes, will be needed to clarify the nature of the association between these polymorphisms and VTE.

Ethics Statement
All study protocols for this genetic analysis were reviewed and approved by The Institutional Review Board of CHA Bundang Medical Center in 10 January 2012 (IRB-number: 2005-002) and followed the recommendations of the Declaration of Helsinki (Fifth revision. 7, October, 2000). Written informed consent was obtained from all study participants prior to enrollment.

Study Population
A set of 203 patients with a recent (<6 months) objective diagnosis of DVT or PE, who visited the CHA Bundang Medical Center (Seongnam, Korea) between March 2006 and January 2011, were enrolled in the study. This VTE group included only patients with symptomatic VTE and excluded individuals with asymptomatic VTE. VTE was classified as provoked if the patient presented with at least one of the following risk factors: recent surgery (<3 months), recent trauma/fracture (<3 months), immobilization (>7 days), malignancy, stroke, severe medical disease, autoimmune disease, pregnancy, use of oral contraceptives, or known hypercoagulable disease. VTE was classified as unprovoked if these risk factors were absent. The 300 control group was chosen among patients visiting the CHA Bundang Health Promotion Center for periodic health examinations, who had no medical history of VTE. Vascular risk factors were assessed using medical records and laboratory data at the patient's first hospital appointment. The data included in this study consisted of age, sex, hypertension, diabetes, lipidemia. Hypertension was diagnosed when a subject had a high baseline blood pressure (≥systolic blood pressure (SBP) 140 mmHg or diastolic blood pressure (DBP) 90 mmHg) or patient had been taken anti-hypertensive medication. Diabetes mellitus was diagnosed when subject's high fasting plasma glucose level was higher than 126 mg/dL or had been taken anti-diabetic drugs. Smoking indicates current smoker at the time of the examination. Hyperlipidemia was defined as fasting serum total cholesterol of 240 mg/dL or a history of anti-hyperlipidemic agent treatment [51].

Genotyping
Peripheral blood samples were collected in blood collection tube and were ethylenediaminetetraacetic acid (EDTA) treated. Blood samples were centrifuged at 3000 rpm for 15 min, the buffy coat layer was collected, and leukocytes were separated. DNA was extracted from subject leukocytes using the G-DEX II Genomic DNA Extraction Kit (Intron Biotechnology, Seongnam, Korea), according to manufacturer instructions. We then assayed the four best-studied single nucleotide polymorphisms (SNPs) in the miRNA biogenesis genes, as determined by a literature search, which included the following 3 -UTR SNPs: DICER1 rs3742330A > G, DROSHA rs10719T > C, RAN rs14035C > T, and XPO5 rs11077A > C. These miRNA biogenesis gene polymorphisms were analyzed by the polymerase chain reaction-restriction fragment length polymorphism (PCR-RFLP) method, and the PCR conditions for these analyses are presented in Table S2.

Statistical Analysis
Clinical characteristics of the study subjects were compared using the independent sample t-test. Associations among VTE prevalence and the four miRNA biogenesis genotypes and allele combination frequencies were estimated by analyzing the odds ratios (ORs) and 95% confidence intervals (CIs) with multivariate logistic regression and the Fisher's exact test, respectively. Adjusted ORs (AORs) for the polymorphisms were determined using multiple logistic regression analysis based on gender, age, diabetes mellitus, hypertension, hyperlipidemia, and smoking status. The genotype distribution for each polymorphism was assessed for Hardy-Weinberg equilibrium deviations, and the genotype and allele frequency differences between groups were assessed using χ2 tests. p-values < 0.05 were considered to be statistically significant. HAPSTAT software, version 3.0 (University of North Carolina, Chapel Hill, NC, USA) (www.bios.unc.edu/Elin/hapstat/) was used to estimate haplotype frequencies for polymorphisms that were determined by multifactor dimensionality reduction analyses to have strong synergistic effects. Statistical analyses to measure the association between DICER1, DROSHA, RAN, and XPO5 polymorphisms and VTE prevalence were performed using MedCalc, version 18.9 (MedCalc Software, Mariakerke, Belgium) and GraphPad Prism, version 4.0 (GraphPad, San Diego, CA, USA) software. The false-positive discovery rate (FDR) correction was used to adjust for multiple comparison tests, and associations with FDR-adjusted p-values < 0.05 were considered to be significant.

Conclusions
In this study, we have identified an association between susceptibility to VTE and polymorphisms in the miRNA biogenesis genes, RAN rs14035C > T and XPO5 rs11077A > C. These findings may provide the basis for continued research efforts focusing on the roles of RAN and XPO5 in hemostasis and thrombus development. We propose that the RAN rs14035 and XPO5 rs11077 polymorphisms influence miRNA biosynthesis, and therefore, affect miRNA post-transcriptional regulation during hemostasis (i.e., platelet biogenesis, coagulation, and anticoagulation) and thrombus formation. However, the mechanisms underlying these relationships remain to be elucidated in future research.