G-T haplotype established by rs3785889-rs16941382 in GOSR2 gene is associated with coronary artery disease in Chinese Han population

Objectives: The aim of the present study is to assess the association between the human GOSR2 gene and coronary artery disease using a haplotype-based casecontrol study in Chinese Han population. Methods: A total of 283 coronary artery disease patients and 280 controls were genotyped for the human GOSR2 gene (rs197932, rs3785889, rs197922, rs17608766, and rs16941382). Data were analyzed for three separate groups: the total subjects, men, and women. Results: For the total subjects, the frequency of the G-T haplotype established by rs3785889-rs16941382 was significantly higher in the coronary artery disease patients as compared to the control subjects (P=0.009). Multiple logistic regression analysis also confirmed that the subjects with G-T haplotype established by rs3785889-rs16941382 (homozygote) were found having significantly higher chance suffering from coronary artery disease than the ones without this haplotype (OR=1.887, P=0.007). Conclusions: The G-T haplotype established by rs3785889-rs16941382 may be a risk genetic marker for coronary artery disease patients in Chinese Han population.


INTRODUCTION
Coronary artery disease (CAD) is a widely spread disease worldwide, its pathogenesis includes both environmental and genetic factors [1], many hereditary features were reported to show significant connections with CAD in previous investigations [2][3].
Golgi Snap receptor complex member 2 (GOSR2) is one of the Golgi-associated soluble N-ethylmaleimidesensitive factor attachment receptor (SNARE) proteins, its main function included the transportation via Golgi complex in different cells [4]. The function also includes interaction with target-localized SNAREs (t-SNAREs), with which the insulin, leptin, angiotensinogen and other macromolecules would go through Golgi compartments [5][6][7].
One study based on American population reported that the A allele in rs197922 of GOSR2 gene was a risk factor for myocardial infarction (OR=1.17, P=0.032) [8]. Pan et al [9] did not find A allele was an significant risk or protective factor for MI, however, base on Japanese population, the T-G-G haplotype of rs197932-rs3785889-rs197922 in GOSR2 was found to show beneficial effects against the occurrences of myocardial infarction. However, in Chinese Han population, there is no such study to evaluate the relationship between GOSR2 gene and CAD. Therefore, the present paper is intended to discover the possible connection between the human GOSR2 gene and CAD in Chinese Han population. Table 1 presented the dimorphic and laboratory data in patients with or without CAD. For total subjects, fasting glucose and incidence of diabetes in CAD patients were higher than those in the control subjects significantly (P<0.05). For men, body mass index, high density lipoprotein (HDL) and fasting glucose in CAD patients were different with those in control subjects significantly (P<0.05). In women, body mass index, triglycerides, HDL, fasting glucose, creatinine and incidence of diabetes in CAD patients were different with those in control subjects significantly (P<0.05). Table 2 presents the genotyping and allele frequency as well as the frequencies in dominant and recessive models between the CAD patients and control subjects for each SNP. For each single nucleotide polymorphism (SNP), the frequency of genotype was consistent with Hardy-Weinberg equilibrium (data not shown). The genotype distributions showed no significant difference for the all 5 SNPs in GOSR2. For total, men and women subjects, the distribution frequency showed no significant difference in the allele distribution as well as the dominant and recessive model distribution.

RESULTS
The linkage disequilibrium data of the selected 4 SNPs are presented in Table 3. All the 5 SNPs were confirmed to be placed in the same haplotype block according to the | D'| values. As the frequency of C allele in SNP4 was too low for the haplotype construction. Meanwhile, since r 2 for SNP3-SNP5 was 0.672, which was above 0.5 in the participants, indicating that SNP3 and SNP5 may not be applied for constructing haplotypes at the same time. Considering that the genotype distribution difference in SNP5 was more significant than that in SNP3, we chose SNP1, SNP2, and SNP5 to construct the haplotype. Table 4 presented thehaplotype-based analysis via the establishment of haplotype using SNP1-SNP2-SNP5, SNP1-SNP2, SNP1-SNP5, SNP2-SNP5. In total subjects, the overall SNP2-SNP5 haplotype distribution differed significantly between the CAD patients and the controls (P=0.031). In total subjects, the G-T haplotype frequency constructed via SNP2-SNP5 in the CAD patients was higher than that in the controls (0.468 vs 0.389, P = 0.009). Meanwhile, in men, the G-T haplotype frequency constructed via SNP2-SNP5 in CAD patients was also higher than that in control subjects (0.459 vs 0.360, P = 0.021). As in women, the same distribution patterns were not noticed.
The software SNPAlyze version 8.2 was applied to construct the diplotypes for every single patient. The H13 (homozygote) was taken into the logistic regression analysis as one potential risk factor, other confounding such as diabetes mellitus, hypertension, low density

DISCUSSION
GOSR2 is a Golgi associated soluble factor attachment receptor (SNARE) protein, it built bridges to allow and promote membrane fusion via the endosomal and secretory pathways [10,11]. A twisted parallel bundle constructed by four amphipathic helices were contained in the SNARE complex [12,13]. Although the specific function of SNAREs, such as transport vesicles functions of docking and fusion, is still uncertain, some studies confirmed that SNARE complexes played the key part in intracellular membrane fusion. The trafficking function of cardiovascular related macromolecules may be in the central stage for the association with multiple cardiovascular diseases such as CAD and hypertension [9,14].
In one previous study based on the Japanese population, TT genotype is the only genotype in the rs17608766 of GOSR2 [9]. In this study based on Chinese Han population, we found 2 subjects had the CT genotype in the CAD group and 3 subjects had the CT genotype in the control group in Chinese Han population. However, we did not notice the significant difference on CT and TT genotype distribution between the CHD patients and control subjects. One study based on American population reported that the A allele in rs197922 of GOSR2 gene was a risk factor for myocardial infarction (MI) (OR=1.17, P=0.032) [8]. However, the study base on the Japanese population reported no significant difference for A allele in rs197922 between the control subjects and CAD patients. In this study based on Chinese Han population, the frequency of A allele in rs197922 of GOSR2 gene didn't show any significant difference between the control subjects and CAD patients (P=0.713). Our results were consistent with the findings in Japanese population and were not consistent with the findings in American population. The differences may mainly be due to the races, since the Japanese population was more likely to be consistent in the origin with Chinese populations.
When the linkage disequilibrium among the SNPs appeared to be weak, the haplotypes analysis could be superior than individual SNPs analysis since it is consist with the human hereditary features [15]. In the previously described study based on Japanese population, after constructing all the possible haplotypes, they discovered that the frequency distribution of T-G-G haplotype, which was constructed with rs197932-rs3785889-rs197922, was lower in the myocardial infarction men when compared with that in the control men (P=0.040). After adjusting the confounding factors, they found that the T-G-G haplotype including homozygous and heterozygous diplotypes may have protective effects on myocardial infarction in Japanese men (OR=0.455, P=0.041) [9]. In this present study, we did not use the rs197932-rs3785889-rs197922 to establish the halotype since the genotype distribution difference in rs16941382 was more significant than that in rs197922 in Chinese population. Therefore, we constructed the haplotype using rs197932-rs3785889-rs16941382 and found a risk haplotype G-T of SNP2-SNP5 in total and male population. With the logistic regression analysis, the haplotype (G-T) is targeted as an novel risk factor for CAD in Chinese Han population (OR=1.887, P=0.007).
The mechanism that GOSR2 is associated with CAD is still not clear. Based on limited data, we supposed that three possible mechanisms may be involved in the association between the GOSR2 and CAD. First, the genome-wide association study (GWAS) showed that rs17608766 in GOSR2 had age-dependent effects on BP [16]. The association with blood pressure may accelerate the arteriosclerosis which would consequently make the some haplotypes in GOSR2 gene the risk factor of CAD. Second, Ghanbari M et al [17] identified that GOSR2 was one of the miR-4513 target genes, which may show multiple effects on lipid and glucose metabolism, blood pressure regulation as well as atherosclerosis. Then the association between GOSR2 and coronary artery disease may be established via miRNA-4513 pathway. Third, GOSR2 has the function of coding for vesicular-SNARE (v-SNAREs), which was perceived as transporter of vesicles into the Golgi complex [18]. v-SNAREs would have influence on target-localized SNAREs (t-SNAREs), allowing the macromolecule substances to move directly across Golgi membrane [19,20]. Due to its trafficking function, the GOSR2 gene could be associated with the cardiovascular diseases which are highly associated with macromolecules such as insulin, leptin, and angiotensinogen.
Our study has several strengths. First, all case and control subjects underwent the coronary angiography, which is the golden standard for the diagnosis of CAD with the combination of clinical or electrocardiogram results. Second, the control subjects may have other SNP, single nucleotide polymorphism. | D'| values of >0.5 to assign SNP locations to one haplotype block. SNPs with an r 2 value > 0.5 were selected as tagged. | D'| above the diagonal and r 2 below the diagonal. The shadowed portion indicates | D' | > 0.5 and r 2 >0.5. cardiovascular or metabolic or endocrine diseases, for example, they may have hypertension, diabetes mellitus (DM), and lipids disorders, which meant that the subjects in control group may share the same risk factors as subjects in CAD group except for that the results of coronary angiography were normal. It could decrease the effects of environmental factors of CAD, increase the homogeneity of the two groups and focus mainly on the effects of genetic factors. There are some limitations in the present study. In Chinese women of this study, the G-T haplotype frequency constructed by SNP2-SNP5 was higher in CAD patients  when compared with controls, but the P value did not reach significance (P=0.181). We believed the enlargement in women samples may lead to more significant difference. At last, the association between the human GOSR2 gene and CAD has been addressed for the first time in the Chinese Han population. The data indicates that in Chinese Han population, the G-T haplotype established by rs3785889-rs16941382 of the human GOSR2 gene might be the risk marker for coronary artery disease.

Ethics statement
Written informed consent was obtained from all participants. All participants explicitly provided permission for DNA analyses as well as collection of relevant clinical data. This study was approved by the Ethics Committee of People's Hospital of Shaanxi Province. It was conducted according to the standards of the Declaration of Helsinki. Authors have access to information that could identify individual participants during or after data collection.

Subjects
The subjects were from the Han population who lived in Shaanxi. All patients and controls had a differential diagnosis for chest pain encountered in the Cardiac Catheterization Department of People's Hospital of Shaanxi Province from January in 2015 to December in 2016. The procedures of coronary angiography were undertaken by highly skilled physicians using the Judkins approach. The findings of coronary angiography were interpreted by at least two experienced imaging specialists and the final diagnosis of CAD was made according to the angiography report.
We randomly recruited 283 Han patients with CAD and 280 age matched controls. The CAD was defined as the presence of at least one significant coronary artery stenosis of more than 50% luminal diameter in coronary angiography. All control subjects also underwent the coronary angiography and had no coronary artery stenosis and did not show clinical or electrocardiogram evidence of myocardial infarction or CAD. Control subjects were not healthy individuals, some of them had hypertension, some of them had diabetes mellitus (DM), and some of them had lipids disorders, which meant that the subjects in control group were exposed to the same risk factors as subjects in CAD group except that the results of coronary angiography were normal.
The data of traditional coronary risk factors, including blood lipids, hypertension and DM were collected from all study participants. The diagnosis of hypertension was established if patients were on antihypertensive medication or if the mean of 3 measurements of systolic blood pressure (SBP) ≥140 mm Hg or diastolic blood pressure (DBP) ≥90 mm Hg, respectively. Diabetes mellitus was defined on the basis of the World Health Organization (WHO) criteria. In addition, individuals with fasting plasma glucose ≥7.0 mmol/L or with a history of diabetes or treatment with insulin were considered diabetic. All patients with impaired renal function, malignancy, connective tissue disease, valvular disease or chronic inflammatory disease were excluded.

SNP selection
GOSR2 gene in human has three transcript variants to encode three different isoforms (isoform A, isoform B, isoform C). The isoform A of GOSR2 gene has the longest length among the three isoforms, it consists of 212 amino acids and is located on chromosome 17q21. This gene is approximately 18.25 kilobase pairs (kbp) and contains six exons separated by five introns.
In this study, we screened the data on the International HapMap Project website (http://hapmap.ncbi. nlm.nih.gov/index.html.en) for the Tag SNPs of GOSR2 gene. SNPs with relatively high minor allele frequencies (MAF) have been shown to be useful as genetic markers in genetic association studies, so we selected three SNPs (rs3785889, rs197932, and rs16941382) which had a MAF of >0.1 as the tag SNPs in Chinese population. rs197932 is located 26 kbp upstream from the start codon in exon 1, while rs16941382 is located 25 kbp downstream from the stop codon in exon six. Meanwhile, we also include rs17608766 and rs197922 which may be associated with blood pressure and coronary heart disease respectively in GOSR2 gene [8,21].

Genotyping
Venous blood samples were collected from all participants in fasting state, and genomic DNA was extracted from the peripheral blood leukocytes using phenol and chloroform extraction method in Central Laboratory of People's Hospital of Shaanxi Province from January in 2015 to December in 2016 and stored in -80°C refrigerator for further genotyping [22,23].
The DNA samples were carried on dry ice to Laboratory Department of Shanghai GENESKY Biological Technology Co., Ltd. (http://www.geneskies. com/) via airplane for genotyping in January in 2017. Genotyping was performed using the TaqMan SNP Genotyping Assay (Applied Biosystems). The primers