Recent genome-wide association (GWA) studies have identified a number of novel genetic determinants of blood lipid concentrations in Europeans. However, it is still unclear whether these loci identified in the Caucasian GWA studies also exert the same effect on lipid concentrations in the Chinese population.
Methods and Results
We conducted a replication study assessing associations between SNPs at 15 loci and blood lipid and lipoprotein concentrations in two Chinese cohorts, comprising 2533 and 2105 individuals respectively. SNPs in APO(A1/C3/A4/A5), TIMD4-HAVCR1, DOCK7, TRIB1, ABCA1, and TOMM40-APOE showed strong associations with at least one lipids trait, and rs174546 in FADS1/2/3 showed modest association with triglyceride in the Chinese population.
Citation: Zhang Z, Tao L, Chen Z, Zhou D, Kan M, Zhang D, et al. (2011) Association of Genetic Loci with Blood Lipids in the Chinese Population. PLoS ONE 6(11): e27305. https://doi.org/10.1371/journal.pone.0027305
Editor: Nikolas Nikolaidis, California State University Fullerton, United States of America
Received: April 7, 2011; Accepted: October 13, 2011; Published: November 3, 2011
Copyright: © 2011 Zhang et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This work was supported by the 973 Program (2010CB529600, 2007CB947300, 2006CB910601), the 863 Program (2006AA02A407, 2009AA022701), the Shanghai Municipal Commission of Science and Technology Program (09DJ1400601), the National Key Project for Investigational New Drug 2008ZX09312-003, and the Shanghai Leading Academic Discipline Project (B205), and the National Natural Science Foundation of China (No. 909190492, 30972529). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Plasma lipids and lipoproteins concentrations are important risk factors for atherosclerosis and related vascular diseases . Twin and family studies suggest that about 50% of the variation in plasma lipid and lipoprotein levels is genetically determined .
A search for the genetic contributions to variation in plasma lipid and lipoprotein levels has been ongoing for several decades . Since 2007, genome-wide association (GWA) studies have obtained great success and have implicated common variants in numerous loci and genes as being the genetic influences underlying lipid and lipoprotein levels , , , , , , , . Kathiresan's group  recently performed a comprehensive meta-analysis of the GWA studies and identified 95 loci significantly associated with blood lipids. These associations were primarily found in European ancestry. Following studies successfully replicated most of these loci in East Asian population , , , , , . However, some of the loci showed no significant association with lipid and lipoprotein levels in Asian population, including the MAFB, NCAN/CLIP2/PBX4 and MVK/MMAB loci , , . It is therefore important to confirm whether known loci have consistent effects across ethnic groups.
The purpose of this study was to replicate the previously reported genetic loci in the Chinese population. We firstly evaluated association between lipid levels and 15 loci selected from the three recent GWAS reports , ,  in Cohort1, comprising 2, 533 Chinese individuals, and selected 10 out of the 15 loci on the basis of the strength of statistical evidence. We then tested association of the 10 loci with lipid traits in Cohort2, comprising 2,105 individuals, to confirm the findings in Cohort1. Finally, we combined the two cohorts results together, since both the two cohort were from Shanghai, with similar genetic background.
Clinically, the most important plasma lipids and lipoproteins are triglycerides (TG), total cholesterol (TC), high density lipoprotein (HDL) cholesterol and low density lipoprotein (LDL) cholesterol. Several studies have suggested that the lipid ratio (TC/HDL-C) has greater independent predictive value for coronary heart disease (CHD) and cardiovascular events than either total cholesterol or LDL cholesterol levels , . We therefore focused on five lipid traits: TG, TC, HDL cholesterol, LDL cholesterol and the TC/HDL ratio in this study.
The ethics committee of the Shanghai Institute for Biological Sciences approved this study. Written consents were given by the patients.
Participants in the present study comprised two groups, Cohort1 and Cohort2. Cohort1 was primarily designed for a case-control study of type 2 diabetes (including 1,360 non-type 2 diabetes controls and 1,173 type 2 diabetes patients). Cohort2 was a community-based prospective epidemiologic cohort of 2,105 subjects. Individuals known to be on lipid-lowering therapy were excluded. Both cohorts were recruited from Shanghai, China. The characteristics of participants are summarized in Table 1.
For all individuals, height, weight, hip and waist circumference and blood pressure were measured by trained medical professionals using a standardized protocol. Body mass index (BMI) was calculated as weight (kg)/[height (m)]2. Blood samples were collected after an overnight fast. Total cholesterol, HDL cholesterol, LDL cholesterol, triglycerides, and fasting plasma glucose (FPG) were measured enzymatically according to standard methods on the Roche modular P800 autoanalyzer (Roche, Mannheim, Germany) with the appropriate reagents (Roche Diagnostics CmbH, Mannheim, Germany).
Selection of candidate variants
We selected 15 out of 55 single nucleotide polymorphisms (SNPs) that achieved genome-wide statistical significance in three recently published GWA studies , , . Three criteria were adopted to choose SNPs:
- Only one lead SNP was selected for each locus. For instance, three GWAS reported seven SNPs in the FADS1/2/3 cluster, including rs174570,  rs174537, rs102275, rs174556, rs1535, rs174546,  and rs174547 . Considering all were in a region with extremely high degree of linkage disequilibrium (D' = 1.0, r2 = 1.0, The International HapMap Project), we selected one variant, rs174546, to represent this region.
- We selected only those SNPs with a minor allele frequency higher than 5% in Chinese population to ensure that this study had enough statistical power.
- Loci had been studied in Chinese population  were excluded.
High-molecular-weight genomic DNA was prepared from venous blood using the QuickGene 610 L Automatic DNA/RNA Extraction System (Fijifilm, Tokyo, Japan). All genotyping experiments were done using TaqMan technology on an ABI7900 system (Applied Biosystems, Foster City, California). The standard 5 µl polymerase chain reaction (PCR) reactions were carried out using TaqMan Universal PCR Master Mix reagent kits under the guidelines provided. Genotype data were obtained in 97.5% of the DNA samples. Replicate quality control samples (5% samples) were included and genotyped with 100% concordance.
SHEsis  was used to perform the Hardy-Weinberg Equilibrium (HWE) test. We assumed an additive model of inheritance, and conducted multiple linear regressions to assess the effect of the number of the specified allele of each SNP on five traits—concentrations of TC, TG, LDL cholesterol, HDL cholesterol and the TC/HDL ratio. Age, gender, body mass index (BMI), and type 2 diabetes status were included in the multiple linear regression models as covariant. Plasma TG and TC/HDL-C were logarithmically transformed before linear regression due to skewed distributions. We did not exclude the type 2 diabetes subjects, in line with previous genome-wide studies , . Considering the potential correlation between diabetes and lipid levels, we included the diabetes status in the association model as a covariant. Multiple testing corrections were performed in stage 2 and combined analysis, P<0.001 was considered as significant, given there were 10 SNPs and 5 traits analyzed.
Stage 1 replication
15 SNPs in 15 loci were genotyped in Cohort 1 comprising 2533 individuals at stage 1. Genotype distribution of each SNP did not deviate from Hardy-Weinberg equilibrium at the 5% level. Results of the multiple linear regression analysis adjusted for age, gender, BMI, and type 2 diabetes status are shown in Table 2.
Of the 15 SNPs tested, 8 SNPs including rs10889353 in DOCK7, rs1501908 in TIMD4-HAVCR1, rs2954029 in TRIB1, rs1883025 in ABCA1, rs964184 in APO(A1/C3/A4/A5), rs2338104 in MMAB-MVK, rs2650000 in HNF1A, and rs157580 in TOMM40-APOE, showed significant association (P<0.05) with at least one lipid level trait.
However, 7 other loci including rs10903129 located in TMEM57, rs12670798 in DNAH11, rs4936883 in LIPG, rs2304130 in NCAN, rs7120118 in NR1H3, rs6102059 in MAFB, and rs174546 in FADS1/2, which had previously been reported to be associated with plasma lipid levels in European ancestry, did not show evidence for association with any of the five lipid traits in the our Chinese sample. Given that the FADS1/2/3 cluster was reported to be associated with lipid concentrations in all of the three GWAS papers, we included this locus in the stage 2 replication. Rs6102059 in MAFB showed a relatively low P value with TC (P<0.09), therefore was also included.
Stage 2 replication
The 10 selected SNPs were subsequently genotyped in Cohort2, and 4 of them showed significant association with lipid traits after multiple testing correction (P<0.001), including variants in TIMD4-HAVCR1, TRIB1, ABCA1 and APO(A1/C3/A4/A5) (Table 3).
Considering the two cohorts were from the same place, we combined the two cohorts together to enhance the statistic power (Table 4). We found three variants, including rs10889353 in DOCK7 (Combined P∼6.5×10−4), rs2954029 in TRIB1 (Combined P∼5.8×10−6) and rs1883025 in ABCA1 (Combined P∼4.0×10−4), associated with total cholesterol concentration. Three variants, including rs10889353 in DOCK7 (Combined P∼5.9×10−4), rs2954029 in TRIB1 (Combined P∼2.3×10−7) and rs964184 in APO(A1/C3/A4/A5) (Combined P∼2.8×10−28), showed significant association with triglyceride concentrations. Two variants including rs1883025 in ABCA1 (Combined P∼2.0×10−5) and rs964184 in APO(A1/C3/A4/A5) (Combined P∼3.0×10−11), showed association with HDL cholesterol concentrations. Three variants including rs2954029 in TRIB1 (Combined P∼7.1×10−4), rs964184 in APO(A1/C3/A4/A5) (Combined P∼4.6×10−6) and rs157580 in TOMM40-APOE (Combined P∼2.0×10−8) showed significant association with LDL cholesterol concentrations. Three variants including rs1501908 in TIMD4-HAVCR1 (Combined P∼1.9×10−5), rs2954029 in TRIB1 (Combined P∼6.0×10−6) and rs964184 in APO(A1/C3/A4/A5) (Combined P∼2.5×10−9) showed association with the ratio of total cholesterol to HDL cholesterol.
We found marginal association between rs174546 in the FADS1/2/3 and triglycerides (Combined P<0.01) (Table 4). We found no significant association between rs2338104 in MMAN-MVK, rs2650000 in HNF1A or rs6102059 in MAFB and plasma lipid levels either in Cohort2 or in the combined Cohorts.
In this study, we investigated whether the results of three independent genome-wide European association studies on plasma lipid and lipoprotein levels were replicatable in the Chinese population. Of the 15 loci selected from the European GWAS reports, 7 loci were successfully replicated.
The most significant association was found between rs964184 in the APO(A1/C3/A4/A5) cluster and triglycerides (Combined P∼2.8×10−28). This variant was also found to be associated with HDL cholesterol, LDL cholesterol, and TC/HDL. Differences in the G allele frequency (0.22 vs. 0.14) and effects on TG (0.14 vs. 0.30)  were found between Chinese and Europeans, suggesting a higher risk allele frequency and weaker effect in Chinese population. The APO(A1/C3/A4/A5) cluster encodes important regulators of fasting lipids, and there is considerable evidence suggesting that variants in this region are associated with altered lipid metabolism , . Fine mapping in this region may help us to find the functional variant.
Another similar case is rs157580 in TOMM4-APOE, which encodes Apolipoprotein E, a main apoprotein of the chylomicron, essential for the normal catabolism of triglyceride-rich lipoprotein constituents. We found the A allele of rs157580 was associated with decreased LDL, TC and increased TG in Chinese, which was different from European population (A allele was associated with increased LDL, TC and TG) (Table 4 and Table S1) . Two other SNPs rs4420638 and rs439401 in this region were reported to be associated with blood lipid profile in both Europeans and Chinese , , , and rs439401 (not included in this study) also showed different effect direction on LDL and TC between the two ethnic groups. Given the different allele frequencies (A allele of rs157580, 0.44 vs. 0.67) and different linkage equilibrium patterns of this region (Figure 1) between Chinese and Europeans, these discordant results across ethnic groups could be explained by different linkage patterns between the causal variants and the tag SNPs that were studied. Nevertheless, these results confirmed the involvement of variants of this gene cluster in the lipid metabolism.
Linkage disequilibrium (LD) structure of SNPs near the TOMM40-APOE locus in HapMap CHB and HapMap CEU. The scheme of the representation of LD is according to the default setting of Haploview Software.
We found variant in ABCA1 associated with TC, TG and HDL. The association with TG was newly identified and retained significant after adjusting by other lipid traits (data not shown), suggesting variant in ABCA1 gene played a wider effect on blood lipid profile in the Chinese population. Recently, Acuna-Alonzo et al  reported a functional ABCA1 gene variant exclusive to Native American and descent populations is associated with low HDL cholesterol levels and shows evidence of positive selection. These findings suggest the importance of ABCA1 genetic variants in lipid metabolism.
We also found variant in DOCK7 associated with TC and TG, variant in TIMD4-HAVCR1 associated with TC, TG, LDL and TC/HDL, variant in TRIB1 associated with TC, TG, LDL, and TC/HDL, with similar effects and same effect directions on blood lipid traits as previous studies in Europeans , , , . Our study confirmed that these loci are implicated in lipid metabolism in the Chinese as well as the European populations.
The FADS1/2/3 cluster locates on 11q12 encoding fatty acid desaturases, which convert polyunsaturated fatty acids into cell signaling metabolites and are functionally involved in lipid metabolism. Previous European studies found variants in the FADS1/2/3 cluster to be associated with plasma concentrations of TG, TC, HDL cholesterol and LDL cholesterol , , , . We found rs174546 in this locus marginally associated with TG (P∼0.01) in Chinese population, with same effect direction to that in Europeans. This result is consistent with another East Asian study, which reported a SNP in FADS1/2 is associated with TG in Japanese and associated with LDL in Mongolian. The linkage disequilibrium patterns of this region in the Chinese population somewhat differ with Europeans. For instance, the LD value between rs174546 and rs174570 is much higher in Chinese (D' = 1.0, r2 = 1.0, HapMap, CHB) than in Europeans (D' = 1.0, r2 = 0.32, HapMap, CEU). The different linkage disequilibrium pattern may therefore explain different association profile across the two ethnic groups.
Rs2338104 in MMAB-MVK was reported to be associated with HDL cholesterol in studies by Willer  and Kathiresan , but in our study this variant showed no association with HDL cholesterol, which is consistent with a large scale Japanese study . Similarly, variants in NR1H3, LIPG, DNAH11, HNF1A and MAFB didn't show significant association with blood lipid traits in our study. It is not necessarily the case that these loci do not influence lipid phenotypes in Chinese. One possible reason is that because of the modest effect sizes of the individual genetic variants on lipid traits our sample size is not enough to detect the association. Another possible reason is the different linkage disequilibrium pattern in Europeans and East Asian population. It is possible that these genes may influence lipid levels through other polymorphisms in East Asian populations. Fine mapping these regions by deep sequencing or additional screening of dense arrays would be needed to reveal association between these genes and lipid levels in the Chinese population.
In conclusion, we successfully replicated association between 7 loci and plasma lipid concentrations in the Chinese population. Our study confirmed the implication of APO(A1/C3/A4/A5), TOMM40-APOE, ABCA1, DOCK7, TIMD4-HAVCR1, TRIB1 and FADS1/2 in plasma lipid and lipoprotein concentrations in Chinese population.
We warmly thank all the participants for their cooperation in this study and gratefully acknowledge all the physicians and nurses who assisted in studying the patients.
Conceived and designed the experiments: LH YL. Performed the experiments: ZZ LT ZC. Analyzed the data: ZZ LT. Contributed reagents/materials/analysis tools: ZZ LT ZC. Wrote the paper: ZZ LT. Revised the manuscript: D. Zhou MK D. Zhang CL.
- 1. Lewington S, Whitlock G, Clarke R, Sherliker P, Emberson J, et al. (2007) Blood cholesterol and vascular mortality by age, sex, and blood pressure: a meta-analysis of individual data from 61 prospective studies with 55,000 vascular deaths. Lancet 370: 1829–1839.
- 2. Hunt SC, Hasstedt SJ, Kuida H, Stults BM, Hopkins PN, et al. (1989) Genetic heritability and common environmental components of resting and stressed blood pressures, lipids, and body mass index in Utah pedigrees and twins. Am J Epidemiol 129: 625–638.
- 3. Hegele RA (2009) Plasma lipoproteins: genetic influences and clinical implications. Nat Rev Genet 10: 109–121.
- 4. Aulchenko YS, Ripatti S, Lindqvist I, Boomsma D, Heid IM, et al. (2009) Loci influencing lipid levels and coronary heart disease risk in 16 European population cohorts. Nat Genet 41: 47–55.
- 5. Kathiresan S, Melander O, Anevski D, Guiducci C, Burtt NP, et al. (2008) Polymorphisms associated with cholesterol and risk of cardiovascular events. N Engl J Med 358: 1240–1249.
- 6. Kathiresan S, Melander O, Guiducci C, Surti A, Burtt NP, et al. (2008) Six new loci associated with blood low-density lipoprotein cholesterol, high-density lipoprotein cholesterol or triglycerides in humans. Nat Genet 40: 189–197.
- 7. Kathiresan S, Willer CJ, Peloso GM, Demissie S, Musunuru K, et al. (2009) Common variants at 30 loci contribute to polygenic dyslipidemia. Nat Genet 41: 56–65.
- 8. Kooner JS, Chambers JC, Aguilar-Salinas CA, Hinds DA, Hyde CL, et al. (2008) Genome-wide scan identifies variation in MLXIPL associated with plasma triglycerides. Nat Genet 40: 149–151.
- 9. Sabatti C, Service SK, Hartikainen AL, Pouta A, Ripatti S, et al. (2009) Genome-wide association analysis of metabolic traits in a birth cohort from a founder population. Nat Genet 41: 35–46.
- 10. Saxena R, Voight BF, Lyssenko V, Burtt NP, de Bakker PI, et al. (2007) Genome-wide association analysis identifies loci for type 2 diabetes and triglyceride levels. Science 316: 1331–1336.
- 11. Willer CJ, Sanna S, Jackson AU, Scuteri A, Bonnycastle LL, et al. (2008) Newly identified loci that influence lipid concentrations and risk of coronary artery disease. Nat Genet 40: 161–169.
- 12. Teslovich TM, Musunuru K, Smith AV, Edmondson AC, Stylianou IM, et al. (2010) Biological, clinical and population relevance of 95 loci for blood lipids. Nature 466: 707–713.
- 13. Liu Y, Zhou D, Zhang Z, Song Y, Zhang D, et al. (2011) Effects of genetic variants on lipid parameters and dyslipidemia in a Chinese population. J Lipid Res 52: 354–360.
- 14. Nakayama K, Bayasgalan T, Yamanaka K, Kumada M, Gotoh T, et al. (2009) Large scale replication analysis of loci associated with lipid concentrations in a Japanese population. J Med Genet 46: 370–374.
- 15. Pan LA, Chen YC, Huang H, Zhang L, Liu R, et al. (2009) G771C Polymorphism in the MLXIPL Gene Is Associated with a Risk of Coronary Artery Disease in the Chinese: A Case-Control Study. Cardiology 114: 174–178.
- 16. Nakayama K, Bayasgalan T, Tazoe F, Yanagisawa Y, Gotoh T, et al. (2010) A single nucleotide polymorphism in the FADS1/FADS2 gene is associated with plasma lipid profiles in two genetically similar Asian ethnic groups with distinctive differences in lifestyle. Hum Genet 127: 685–690.
- 17. Park MH, Kim N, Lee JY, Park HY (2011) Genetic loci associated with lipid concentrations and cardiovascular risk factors in the Korean population. J Med Genet 48: 10–15.
- 18. Ingelsson E, Schaefer EJ, Contois JH, McNamara JR, Sullivan L, et al. (2007) Clinical utility of different lipid measures for prediction of coronary heart disease in men and women. JAMA 298: 776–785.
- 19. Shi YY, He L (2005) SHEsis, a powerful software platform for analyses of linkage disequilibrium, haplotype construction, and genetic association at polymorphism loci. Cell Res 15: 97–98.
- 20. Delgado-Lista J, Perez-Jimenez F, Ruano J, Perez-Martinez P, Fuentes F, et al. (2010) Effects of variations in the APOA1/C3/A4/A5 gene cluster on different parameters of postprandial lipid metabolism in healthy young men. J Lipid Res 51: 63–73.
- 21. Gomez P, Perez-Martinez P, Marin C, Camargo A, Yubero-Serrano EM, et al. (2010) APOA1 and APOA4 gene polymorphisms influence the effects of dietary fat on LDL particle size and oxidation in healthy young adults. J Nutr 140: 773–778.
- 22. Acuna-Alonzo V, Flores-Dorantes T, Kruit JK, Villarreal-Molina T, Arellano-Campos O, et al. (2010) A functional ABCA1 gene variant is associated with low HDL-cholesterol levels and shows evidence of positive selection in Native Americans. Hum Mol Genet 19: 2877–2885.