c.*84G>A Mutation in CETP Is Associated with Coronary Artery Disease in South Indians

Background Coronary artery disease (CAD) is one of the leading causes of mortality worldwide. It is a multi-factorial disease and several studies have demonstrated that the genetic factors play a major role in CAD. Although variations in cholesteryl ester transfer protein (CETP) gene are reported to be associated with CAD, this gene has not been studied in South Indian populations. Hence we evaluated the CETP gene variations in CAD patients of South Indian origin. Methods We sequenced all the exons, exon-intron boundaries and UTRs of CETP in 323 CAD patients along with 300 ethnically and age matched controls. Variations observed in CETP were subjected to various statistical analyses. Results and Discussion Our analysis revealed a total of 13 variations. Of these, one3’UTRvariant rs1801706 (c.*84G>A) was significantly associated with CAD (genotype association test: OR = 2.16, 95% CI: 1.50–3.10, p = 1.88x10-5 and allelic association test: OR = 1.92, 95% CI: 1.40–2.63, p = 2.57x10-5). Mutant allele “A” was observed to influence the higher concentration of mRNA (p = 7.09×10−3, R2 = 0.029 and β = 0.2163). Since expression of CETP has been shown to be positively correlated with the risk of CAD, higher frequency of “A” allele (patients: 22.69% vs.controls: 13%) reveals that c.*84G>A is a risk factor for CAD in South Indians. Conclusions This is the first report of the CETP gene among South Indians CAD patients. Our results suggest that rs1801706 (c.*84G>A) is a risk factor for CAD in South Indian population.


Introduction
Coronary artery disease (CAD) is the leading cause of mortality worldwide [1]. CAD and its clinical manifestations are etiologically complex, with approximately equal contributions from genetic and environmental factors [2,3]. Genetic risk scores derived from several functionally relevant single nucleotide polymorphisms (SNPs) or haplotypes in several genes may help in predicting CAD [4]. Associations of only a few common SNPs with CAD have been consistently replicated in several studies [5]. Both genome-wide association studies and candidate-gene approaches have identified a number of novel chromosome loci or genes that are associated with CAD [6,7]. However, they account for relatively small portion of the overall CAD risk; therefore, there is a need for identification of novel loci or genes for CAD. Cholesteryl ester transfer protein (CETP) gene is localized on chromosome 16, which is of 21995 base pairs (bp) in size and consists of 16 exons (Gene ID 1071).CETP is a hydrophobic glycoprotein which plays a major role in RCT (Reverse Cholesterol Transport) from tissues to the liver. By enabling transfer of cholesteryl esters from high-density lipoproteins (HDL) to low density lipoproteins (LDL) and very low density lipoproteins (VLDL) lipoproteins CETP enables remodeling of plasma lipoproteins [8]. The elevated level of HDL cholesterol (HDL-C) is shown to be a protective factor for coronary artery disease from many epidemiological studies [9].Our earlier studies have suggest that Indian populations are unique in their origin and practicing endogamy for the past thousands of years, hence expected to have unique set of mutation which led to several disease, cardiac disease in particular [10][11][12].
Among the fastest growing non-communicable diseases, cardiovascular diseases (CVDs) are expected to cause largest number of mortality and morbidity within India [13]. Indians possess a unique lipid profile characterized by high triglycerides, low high-density lipoprotein (HDL), and increased lipoprotein (a) levels [14]. CETP plays a major role in HDL metabolism and this gene possesses several SNPs that have been reported to be associated with plasma HDL concentrations. However, this gene has not been analysed on Indian population, hence this study was aimed to investigate whether CETP gene variations influence CAD in South Indian population.

Sample details
The study subjects composed of 323 CAD patients with coronary atherosclerosis and 300 age and ethnically matched healthy controls from South India. Blood samples were collected from Yashoda Hospital and Nizam Institute of Medical Sciences (NIMS), Hyderabad, India. Clinical manifestation of coronary atherosclerosis was evaluated by precutaneous coronary angiography, by a panel of experienced cardiologists. All ethnically matched control individuals were free from CAD, as determined by medical history, clinical examinations, or electrocardiography. Among CAD patients, 70% were males and 30% were females, while among the healthy control group 76% were male and 24% were female. The mean age of healthy controls was 65.26±10.30 years while that of patients was 56.21±10.45 years. In CAD patients; 18.57% were tobacco chewers, while it was 10% among the healthy controls. The demographic details of the individuals included in the study are given in Table 1.
We also utilized the genotype data of rs1801706/c. Ã 84G>A of 1000 genome project's samples from Ensembl genome browser (version 84) (asia.ensembl.org).

Sample collection and DNA isolation
Prior to collection of blood samples, CAD patients were subjected to physical/clinical examinations such as 12 lead ECG and lipid profile, etc. Blood samples of healthy controls from the same ethnic background, without hypertension or CAD based on electrocardiograph were collected. A total of 10 ml intravenous blood samples of both cases and controls were collected in EDTA vaccutainer, after obtaining informed written consent. Genomic DNA was isolated from all the samples using standard protocol [15]. This study followed the principles outlined in the Declaration of Helsinki (WMA World Medical Association Declaration of Helsinki), and was approved by the Institutional Ethics Committee of Yashoda Hospital, Hyderabad; Nizam's Institute of Medical Sciences (NIMS), Hyderabad; and CSIR-Centre for Cellular and Molecular Biology (CCMB), Hyderabad, India.

Genotyping
The reference genomic sequence of CETP (ENSG00000087237) was obtained from the Ensemble database (asia.ensembl.org). Primers to amplify all the exons, and exon-intron boundaries of CETP were designed using Primer3 web version 4.0 ( Table 2). PCRs (polymerase chain reactions) were performed using GeneAmp 9700 (Applied Biosystems, Foster City, USA) using Emerald Amp GT PCR master mix (TaKaRa) according to the manufacturer's protocol. After PCR, Amplicons were size fractionated using2% agarose gel, stained with ethidium bromide and observed under UV transilluminator. Subsequently, amplicons were treated with Exo-SAP (USB Corp., USA), and sequenced using a BigDye Terminator (v3.1) cycle sequencing kit (Applied Biosystems, Foster City, USA) on an ABI 3730XL DNA analyzer. Sequences obtained were assembled with the reference sequences using AutoAssembler software (Applied Biosystems, Foster City, USA). Variations observed were noted for further analysis.

Statistical analysis and functional validation
Allele and genotype frequencies were calculated by the allele counting method. Statistical comparisons were carried out by Plink software [16].The P values less than 0.05 were considered for statistical significance. To explore the Hardy-Weinberg equilibrium (HWE), we consider Further, to explore the functional significant of mutant allele "A" of rs1801706 (c. Ã 84G>A), we utilized the genome expression dataset GSE6536 of HapMap population from GEO (gene expression omnibus) database [17] and genotype data of CETP with ±10 kb flanking region from ftp://ftp.ncbi.nlm.nih.gov/hapmap/genotypes/2009-01_phaseIII/plink_format/. Further, we extracted the population-wise normalized expression value of CETP specific probe GI_4557442 from above downloaded GSE6536 dataset and performed quantitative trait association analysis using Plink software [16]. To explore the group-wise differences of mRNA level, we performed t-test using R. Moreover, to conclude the relation between higher mRNA concentrations with CAD patients, we utilized the relationship discussed in previous reports. On the basis of Barkowski, RS et. al. and Tan, MH [18,19]; Hence, the genetic risk of CAD will be directly proportional to the mRNA concentration of CETP;

Results
We have investigated the exons, exon-intron boundaries and UTR of CETP in 323 individuals with CAD and 300 ethnically matched controls. In total, we found thirteen variants (SNPs), of which one was in splice regions  Table 3 and Fig 1).  Among these 13 variants, 6 were in HWE equilibrium (p>0.05) ( Table 3). Of which, rs1801706 (c. Ã 84G>A) was significantly associated with patients group. The 3' UTR variant c. Ã 84G>A (G/A) showed a genotype distribution (%) of 60.37 (GG), 34.67 (GA) and 4.95 (AA) among individuals with CAD; whereas in controls the frequencies were 76.66 (GG), 20.66 (GA) and 2.66 (AA). Association analysis with genotype showed significant association with CAD (OR = 2.16, 95% CI: 1.50-3.10, p = 1.88x10 -5 ). The allelic distribution showed 77.70% of 'G' and 22.29% of ' A' in cases and among the controls the G showed 87% and ' A' showed 13% with significant association of the ' A' allele with OR = 1.92, 95% CI; 1.40-2.63, p-value 2.57x10 -5 . Both genotype and allele distribution of CETP SNPs among cases and controls are given in Table 3.Since, we did not observed LD differences between the cases and controls; we did not proceed for haplotype analysis.

Functional validation of rs1801706/c.*84G>A
To functionally validate the mutant allele "A" (c. Ã 84G>A), we have utilized the whole genome gene-expression data of 210 HapMap samples [17,20] and performed QTL analysis with genotype information of same individuals from HapMap project, with additive model ( Table 4) and observed variant rs1801706 in association with CETP mRNA level with p-value 7.09×10 −3 (R 2 = 0.029 and β = 0.2163) ( Table 4). The genotype and normalized mRNA intensity/expression value for c. Ã 84G>A is given in S1 Table. Interestingly, both heterozygous (GA) and mutant homozygous (AA) genotype were found to influence the higher level of mRNA (Fig 2).We also explored pair-wise comparison of mRNA expression between genotypes and observed that mRNA expression level of genotype AA vs. AG (p-value = 0.3452) and genotype AA vs. GG (p-value = 0.2632) were not significant, while genotype AG vs. GG was significantly different (p-value = 0.0018).

Prevalence of rs1801706/c.*84G>A
Further, we explored the frequency spectrum of rs1801706 (c. Ã 84G>A) in other world populations, including Indians. We utilized the genotype dataset from 1000 genome project. We  Table 5).

Discussion
CAD is caused by multiple genetic and environmental factors [2].CETP plays a central role in human lipoprotein metabolism, as it facilitates the removal of excess cholesterol from the body via LDL receptor-mediated uptake in the liver and excretion into the bile [21]. Our earlier study on the -629 promoter of CETP gene had shown a significant association between CAD patients and controls [22]. Considering the crucial role of CETP in lipid metabolism, we investigated the association of genetic variants of the CETP with risk of coronary artery disease in patients from South India.
In the present study, we observed a 3' UTR variant, rs1801706 (c. Ã 84G>A) is associated with the CAD in South Indians, however, we did not find association of a few previously reported SNPs. A genome-wide linkage analysis conducted on healthy American woman cohort had analyzed over 350,000 SNPs and found only SNPs flanking or in the CETP gene were associated with both HDL-C and risk of incident CAD [23].Papp et.al. observed that rs5883T/rs9930761C was associated with increased HDL-C levels in males [24]. Although our study group had 76% males, we did not find any significant association with rs5883. Studies on C>T/In9 (rs289714) was earlier shown to be associated with undesirable changes in adiposity and HDL-C levels when exposed to excessive calorie consumption [25,26], however, we did not find significant association between cases and the controls. Studies on Caucasians and African Americans [27] showed association of CETP variations with myocardial infarction (MI).
The rs1800777 (Arg 451Gln) is located within the lipid-binding region of CETP protein and possibly may result in the loss of positive charge, varying the binding efficiency of CETP to cholesteryl esters. Lu et al. reported that rs1800777was associated with lower plasma HDL cholesterol levels [28], whereas Moleres et al. reported that this SNP is strongly associated with adiposity indexes [29]. However, in the present study both the genotype and allele frequency of this variant did not show any association with CAD. Interestingly, we found rs1801706 (c. Ã 84G>A) was significantly associated with CAD, which is in agreement with Whitehall II et al.
Since, associated variant rs1801706 (c. Ã 84G>A) was observed in UTR region, we predicted that mutant allele "A" might be affecting mRNA expression of CETP. Here, we were not aware that mutant allele "A" increasing or decreasing the expression. But, using Eq 2, it can be further predicted that mutant allele "A" should increase the expression because risk of CAD increases c.*84G>A Mutation in CETP Is Associated with Coronary Artery Disease in South Indians Table 4. SNPs present within ±10kb of CETP and their respective association p-value with normalized mRNA expression level. Only rs1801706 (c. *84G>A) was significantly associated and highlighted in bold. with expression level and patients were having higher frequency of "A". Interestingly, in QTL analysis, we observed that our prediction is true. It is well known that inhibition of CETP decrease LDL level while increase the HDL, viceversa might be true. We can predict that high expression of CETP, due to mutant allele "A", is responsible for high LDL level in CAD patients. This might be the reason of plague formation in blood vessels and further causing coronary artery disease.

Chr
Further, data obtained for rs1801706 (c. Ã 84G>A) from the 1000 genome project revealed that only 3 populations were having frequency of<0.1; (1) 0.086 in CDX (Chinese Dai), (2) 0.078 in MXL (Mexicans) and (3) 0.029 in PEL (Peruvians) ( Table 5). Populations with Indian ancestry (GIH, STU, ITU and BEB) have higher frequency of this allele compared to the controls. This might be due to the admixture of migrant Indians with local populations, who might have higher frequency of rs1801706-A allele (Europeans population in 1000 genome project).

Conclusion
In conclusion, our study revealed rs1801706 (c. Ã 84G>A), a functionally relevant variant in 3' UTR of CETP, is strongly associated with CAD in South Indian. Interestingly, mutant allele "A" was found to be associated with higher concentration of CETP mRNA. Since, CETP involve in the conversion of HDL to LDL/VLDL, we are tempted to conclude that rs1801706-A increases the risk of CAD by increasing rate of conversion of HDL to LDL/VLDL, through changing the half life of CETP mRNA.