Association Study of Single-Nucleotide Polymorphisms on Chromosome 1p13, 1p32, 9p21 and 19p13 with Cardiovascular Diseases in Chinese Han Population: A Case-Control Study

Objectives: Previous research discovered single nucleotide polymorphism (rs2383206 and rs2383207) on chromosome 9p21 that is associated with coronary heart disease in a Chinese population. However, few data are available on the association of other single nucleotide polymorphism with cardiovascular disease in a Chinese population. This study aimed to determine whether the single nucleotide polymorphisms on chromosome 1p13, 1p32, 9p21 and 19p13 were associated with coronary artery disease in a Chinese population. Methods: We conducted a case-control study. Cases were coronary artery disease (n=670) between 2010 and 2015. Controls (n=1340) were randomly selected and frequency matched to cases on age and gender. All of the participants were selected to study 18 single nucleotides using allele-specific real-time polymerase chain reaction method. Results: Four single nucleotides in 9p21, two single nucleotides in 1p13 and one single nucleotide in 1p32 were associated with cardiovascular disease risk in Chinese population (Global P value for multiple logistic regression, <0.0001, respectively). rs10757274 showing the strongest association with cardiovascular disease. GG carriers of four SNPs (rs10757274, rs2383206, rs10757278 and rs1333049) in 9p21 had higher risk (Odds ratio=1.40, 95% Confidence interval: 1.10-1.79; Odds ratio=1.33, 95% Confidence interval: 1.04-1.69; Odds ratio=1.35, 95% Confidence interval: 1.07-1.72; Odds ratio=1.34, 95% Confidence interval: 1.06-1.71). Conclusion: rs10757274, rs2383206, rs10757278 in 9p21, rs562556 in 1p32, and rs646776 in 1p13 may serve as a novel genetic marker for the risk of significant cardiovascular disease in Chinese Han population.


Introduction
Cardiovascular disease (CVD) is the leading cause of death and disability-adjusted life-years worldwide, with increasing incidence and prevalence in low and middle income countries [1]. Non-modifiable risk factors include increasing age, male sex, and heredity. Modifiable risk factors include smoking, hypertension, dyslipidemia, obesity, physical inactivity, and diabetes [2][3][4].
In 2007, genome-wide association studies on CVD identified a series of associated single-nucleotide polymorphisms (SNPs) in an intergenic region of chromosome 9p21, near the CDKN2A and CDKN2B genes [5][6][7]. In Chinese Han population few studies have simultaneously explored the relationship of these and other genes on risk of coronary artery disease (CAD). Despite recent progress in identifying some novel genetic contributors to CVD, it is currently unknown how these recently discovered loci interact with the environment and what role such interactions play in the development of disease [8,9]. Investigation of gene-environment interactions are necessary to further our understanding of the underlying biology and pathophysiology of the disease, and could potentially be useful in improving cardiovascular risk stratification and thereby reducing clinical events [9][10][11]. Our study has been carried out in Beijing of China, showed that nine potentially modifiable risk factors account for most of the risk of CAD. We sought to investigate this further by first, exploring the risk genes in Chinese Han population and assessing whether the relationships between risk genes and CAD, and second, identifying some genetic and other risk factors interactions and comparing the variations in risk genes and CAD relationships between different stratification.
We examined the association of common genetic variation in 1p32 near PCSK9, 1p13 near CELSR2-PSRC1-SORT1, 9p21 near CDKN2A, CDKN2B, and 19p13 near LDLR with risk of incident acute myocardial infarction in a population-based case-control study.

Study populations
The setting for this study has been carried out in Beijing of China. The affiliated hospital of Capital Medical University coordinated the study. A case was defined as patients with angiographically confirmed narrowing of the coronary vessels by more than 50%, fatal or nonfatal myocardial infarction or unstable angina pectoris, as well as evidence of CAD by coronary angiography. Patients with congenital heart disease, cardiomyopathy, valvular heart disease, and renal or hepatic disease were excluded from the study. At least two age (±5 years) control (without a history of CVD) was recruited for each case. Every participant was assigned an index date. For cases, the index date was the admission date for CAD. For the controls, the index date was a random date within the year for which they were sampled. In addition, we excluded patients whose blood specimens did not yield genotype information. The study was approved by the ethics committee of Capital Medical University (No.2015SY52), and written informed consent was obtained from each subject before his/her inclusion in the study.

Epidemiological survey
Trained personnel administered the structured questionnaires and physical examinations in a standardized manner. We use of standardized questionnaires for all study participants to obtain the information, containing demographic factors, lifestyle, smoking history, alcohol use, family income and education, psychosocial factors, physical activities, personal and family history of CVD and risk factors. Blood pressure, height, weight, waist and hip circumferences were measured. Body mass index (BMI) was defined as weight (kg) divided by height squared (m 2 ). Questions were included about psychosocial conditions to identify psychological stress. General psychological stress was defined as experiencing stress at work or at home and was also assessed in the four categories.

Genetic variation
Single nucleotide polymorphisms (SNPs) and Haplotype were identified in two regions (1p13 near CELSR2-PSRC1-SORT1 and 9p21 near CDKN2A, CDKN2B). Among variants, SAS/genetics was selected maximally informative sets of SNPs to describe genetic variation in Chinese Han population using linkage disequilibrium and of 0.90.

Blood collection and genotyping
Non-fasting blood samples (20 ml) were drawn and centrifuged within 2 h of admission, frozen immediately after processing. Samples were shipped in nitrogen vapour tanks from every site to a blood storage site and stored at -70°C in freezers or -170°C in liquid nitrogen.
A total of 2010 participants (670 CAD cases and 1340 controls) were recruited for our study, and of these, all participants had DNA samples available. The SNPs was genotyped by allele-specific real-time polymerase chain reaction (PCR) using GeneAmp 5700 Sequence Detector (Applied Biosystem, Foster city, CA, USA). We obtained genotyping results on the study subjects.

Statistical analysis
Continuous variables differences between cases and controls were calculated by t test. Categorical variables differences were calculated by Pearson chi-squared analysis. Hardy-Weinberg equilibrium was also assessed by Pearson chi-squared analysis in the ALLELE procedure in SAS software package (Version 9.3; SAS Institute, Chicago, IL, USA)/ Genetics.
We assessed genotypes and allele frequencies difference between CAD and control group using Pearson chi-squared, Fisher's exact tests and Trend test in the CASECONTROL procedure of SAS/Genetics. Haplotype frequencies were inferred using the Expectation-Maximization algorithm available in the HAPLOTYPE procedure in SAS/Genetics. Haplotype figure was described by Haploview software.
Odds ratios estimates and 95% confidence intervals for each variant were calculated in logistic regression models. Adjusted odds ratios were performed by multiple logistic regressions, and it controls for gender, age, obesity, smoking, alcohol, type 2 diabetes, stroke history, stress, Apolipoprotein-A1 (ApoA1), and Apolipoprotein-b (ApoB). Data were analysed using LOGISTIC procedure in SAS/STAT. Wald test and likelihood-ratio test calculated the statistical significance. P<0.05 was reject level to indicate statistically significant differences.

Characteristics of participants
The general and biochemical characteristics between CAD cases and controls are detailed in (Table 1). The levels of body mass index (BMI), waist-to-hit ratio (WHR), ApoB, systolic blood pressure, and diastolic blood pressure and the percentages of general stress were higher in CAD group than control group (P<0.05-0.001). There were no significant differences the percentages of subjects who smoked cigarettes and consumed alcohol (P>0.05 for all).

SNPs genotype and their frequencies
We genotyped three informative SNPs in 1p13 near CELSR2-PSRC1-SORT1, eight SNPs in 9p21 near CDKN2A, CDKN2B and other seven SNPs. These SNPs were in Hardy-Weinberg equilibrium in CAD and controls except rs646776. In the Uni-variable Logistic regression, eight SNPs in 1p13 and 9p21 were associated with CAD. Next, we adjusted CAD risk factors BMI, smoking, drink, Hypertension, Diabetes, Stroke, CVD of family, stress. Upon adjustment, the effect of the rare alleles of the analysed SNPs remained highly significant and was similar to that seen in the unadjusted analysis. Five SNPs in 9p21 and two SNPs in 1p13 showed the association under the multiplicative model. rs10757274 showing the strongest association. For the G allele frequency at rs10757274 in CAD, was significantly higher than control group (50.6% vs. 46.1%, P=0.002, respectively). We found that each additional copy of the G allele at rs10757274 increased the risk of CAD

Stratification analysis
We conducted stratified analysis for the three SNPs (   Figure 1: Pair-wise LD among eight SNPs in 9p21 in the control group. The numbers inside the squares are ×100. Figure 1 shows the results of haplotype analysis for the SNPs examined. And then we calculated possible haplotype frequency of 8 loci in 9p21, using haploview and SAS software. Using the genotypes of 1340 controls, we defined the haploblock structure of SNPs within the region of 9p21 in the Chinese population. By defining a solid spine of LD as >0.90, we identified one haploblock in the 9p21 region.

Haplotype frequency
Next, we focused on the haplotypes within block with association. One haplotype, AAAC, was less frequent in CAD than control subjects (49.2% vs. 44.8%) and the other haplotype, GGGG, was more frequent (48.2% vs. 44.4%) ( Table 4). Our results indicated that haploblock on chromosome 9p21 have possible association with CAD in the Chinese population. rs646776 and rs599839 in 1p13 have been confirmed correlation in a large scale analysis [8,9]. We chose rs611917, rs646776 and rs602633 in our study, and found that rs602633 was higher associated with risk of CAD in China population. A could be risk allele, and heterozygosis appeared in lower risk of CAD. rs646776 and rs611917 had lower association with CAD, and rs602633 was in linkage disequilibrium with rs611917 (r 2 =0.9).
Overall, common variation of four SNPs (rs10757274, rs2383206, rs10757278 and rs1333049) in9p21 near CDKN2A, CDKN2B and two SNPs (rs602633 and rs646776) in 1p13 near CELSR2-PSRC1-SORT1 were associated with risk of CAD. We found no association of common variation of 7 SNPs in 1p32 near PCSK9 and 19p13 near LDLR with risk of CAD in Chinese population (
The region on chromosome 9p21 showed an association with the coronary heart disease (CHD) in previous studies. The associated SNPs were rs1333049 (18 researches, OR=1.23), rs10757274 (17researches, odds ratio (OR)=1.24), rs2383207 (6 researches, OR=1.28), rs2891168 (4 researches, OR=1.29), rs10757278 (2 researches, OR=1.27) [2]. In Asian race, rs1333049 (Japan, China), rs10757274 (China), rs10757278 (China) showed an association with the Myocardial infarction (MI) [2]. Our case-control study not only replicated the findings of the 3 SNPs on chromosome 9p21 that were associated with CAD, but also provided rs564398, rs2383206 that were possible associated with CAD in Chinese population. And SNP rs10757274 showed the strongest association. HapMap data suggests that rs10757274 and rs2383206 were located within 20 kb of each other on chromosome 9p21 and were in strong linkage disequilibrium (r2=0.89) in North American population, and the five SNPs (rs2383206, rs2383207, rs10757274, rs10757278, and rs1333049) are in 1 block, and the D' between each of the 5 SNPs was Chinese Han origin [5,16]. In our study, we identified one haploblock in the 9p21 region (rs10757274, rs2383206, rs10757278, and rs1333049) in Chinese population, and it's similar to previous North American population study. On the basis of the stringent analysis of these SNPs and replication in our research, we are confident that 9p21 region is a strong candidate locus for CAD susceptibility in the Chinese population.
CDKN2A and CDKN2B in the region of chromosome 9p21 are coding sequences for 2 cyclin-dependent kinase inhibitors. These genes involved in the regulation of the cell cycle and would be implicated, and transform growth factor (TGF)--induced growth inhibition, in the pathogenesis of atherosclerosis [16][17][18][19]. McPherson et al. re-sequenced the most proximal to the risk locus in CDKN2A and CDKN2B gene coding regions and found no association between the CHD risk and this locus [7]. However, some researchers recently found the relationship between the same region and increased susceptibility to type 2 diabetes (T2D) [20][21][22]. Many studies have identified of the underlying possible mechanism at this locus of CHD. These results imply that the region of 9p21 may be associated with many complex diseases. In our study, the interactions between smoking, alcohol, and BMI with rs0757274, rs2383206, and rs10757278 were not significant, so this interaction need to be replicated in other populations.
The strongest association with CHD after 9p21 is the CELSR2-PSRC1-SORT1 gene cluster. The risk allele at CELSR2-PSRC1-SORT1 has a frequency of 81% in European subjects and an additive odds ratio of 1.19 [23,24]. rs646776 and rs599839 in 1p13 have been confirmed correlation in a large scale analysis [12,13]. Few studies have examined the associations between SNPs on chromosome 1p13 and CVD risk in East Asian populations. We found that rs646776 and rs602633 were higher associated with risk of CAD in China population. However, SNP rs646776 genotype frequencies did not obey Hardy-Weinberg equilibrium, and we thought that there was error in our experimental test and sample survey. This is limitations of our study.
The relationship of rs646776 and other SNPs at 1p13.3 with low density lipoprotein-cholesterol (LDL-C) and ApoB in Genome-wide associations study may be associated with the reported roles of the SORT1 protein in vesicular transport [25]. A recent 176 whole blood samples study implied that there exist functional variants at this locus of SORT1 gene. And the risk allele decreased expression of SORT1 gene and increased plasma LDLC concentration and the risk of CHD [22]. The gene product of SORT1 is sortilin, a multiligand receptor that may aid in the clearance of low density lipoprotein (LDL) from plasma, mediated through direct interactions with the LDL receptor-related protein [26][27][28][29]. Another 400 liver samples study confirmed the relationship of the risk allele in genome-wide associations study with increased LDLC concentrations, and these concentrations decreased SORT1 and CELSR2 expression [30]. In our study, we only found that rs646776 was associated with risk of CAD. We should increase sample size and statistical power to validate the association in other population. Meanwhile we also exclude causal roles for CELSR2, PSRC1 genes, because these genes map neighbouring recombination hotspot from SORT1 closed to the strongest associations. 9p21 in CDKN2A, CDKN2B, and 1p13 near CELSR2-PSRC1-SORT1 also showed association with other disease. Ahluwalia et al. [31] tested the 11 biomarker-associated SNPs for association with T2D in Danish samples and found significant statistical associations for CELSR2 SNPs [OR=1.11 (1.05-1.18), P=1.2×10 -3 ], respectively. rs646776 near CELSR2 was associated with an increased fasting serum insulin like growth factor binding protein 1 (IGFBP1) level and a decreased risk of T2D [31]. Gaulton KJ et al. [32] observed four loci that each had two distinct association signals, CDKN2A-CDKN2B, diacylglycerol kinase beta (DGKB), melanocortin 4 receptor, (MC4R) and gastric inhibitory polypeptide receptor, (GIPR), with each locus represented by noncoding index variants. The index variants at the CDKN2A-CDKN2B locus represent the known T2D haplotype association signal mapping to a 12-kb intergenic recombination interval [32]. In our research, we adjusted influence of type 2 diabetes on this association using by multiple regression. We will focus on this research of association between these SNPs and other disease.
In our research, we address SNP-environment interactions which are thought to underlie many diseases and traits. Using logistic model didn't find the interaction of SNPs and life style. While most researchers are eager to look at the interaction of multiple genes and environment in causing disease, the statistical tools and algorithms that are needed are still in the early stages of development.

Conclusion
The SNP 10757274, rs2383206, rs10757278 in 9p21, rs562556 in 1p32, and rs646776 in 1p13 may serve as a novel genetic marker for the risk of significant cardiovascular disease in Chinese Han population.