Tohoku J. Exp. Med., 2011, 224, 119-125Lymphotoxin-Alpha Gene SNPs and CAD in Chinese 119 Haplotype-Based Association of Four Lymphotoxin-Alpha Gene Polymorphisms with the Risk of Coronary Artery Disease in Han Chinese Yan Liu,1 Haihui Sheng,2 Lin Lu,1 Zhijun Wu,1 Qiujin Chen,1 Huasheng Xiao2 and Wei Jin1 1Department of Cardiology, Ruijin Hospital, Shanghai Jiaotong University School of Medicine, Shanghai, P.R. China 2National Engineering Center for Biochip at Shanghai, Shanghai, P.R. China Lymphotoxin-alpha (LTA), a pro-inflammatory cytokine, has been implicated in the pathogenesis of coronary atherosclerosis. Meanwhile, association of some single nucleotide polymorphisms (SNPs) of LTA gene with coronary artery disease (CAD) has been evaluated; however, the results are irreproducible. We therefore investigated the relationship between four SNPs of LTA gene and CAD in Han Chinese: G+10A (rs1800683, 5′-untranslated region), A+80C (rs2239704, 5′-untranslated region), T+496C (Cys13Arg, rs2229094, exon 2), and C+804A (Thr26Asn, rs1041981, exon 3). Genotyping was performed in 438 CAD patients and 330 healthy controls. Single-locus analysis showed that the genotype and allele frequencies of G+10A polymorphism exhibited marginal differences between CAD patients and controls, although no statistical significance was observed after the Bonferroni correction. Logistic regression analysis revealed that GG genotype of G+10A polymorphism was significantly associated with the risk of CAD under the dominant mode, whereas no significant association was detected between A+80C polymorphism and CAD. In contrast, individuals carrying TT or TC genotype of T+496C polymorphism showed a decreased CAD risk relative to those with CC genotype under the recessive mode. Likewise, CC genotype of C+804A polymorphism was associated with a protective effect on CAD under the dominant mode. Further, in haplotype analysis, the haplotype G-C-T-C (in order of rs1800683, rs2239704, rs2229094 and rs1041981) was significantly associated with a decreased risk of CAD after assigning the most common haplotype A-C-T-A as a reference. In conclusion, we show a protective effect of the haplotype G-C-T-C on the occurrence of CAD, suggesting the involvement of LTA in CAD pathogenesis. Keywords: coronary artery disease; gene; haplotype; lymphotoxin-alpha; polymorphism Tohoku J. Exp. Med., 2011, 224 (2), 119-125. © 2011 Tohoku University Medical Press Coronary artery disease (CAD) is a complex multifac- the risk of CAD (Jang et al. 2007; Koch et al. 2007; Palikhe torial disorder to which genetic and environmental factors et al. 2007). Meanwhile, family-based linkage analyses contribute interactively; it has become one of the leading (PROCARDIS Consortium 2004) and genome-wide associ- causes of death in China. Ample evidence exists proving ation studies (Iida et al. 2003; Koch et al. 2005; Ozaki and the involvement of inflammatory processes in all phases of Tanaka 2005; Tanaka and Ozaki 2006) had implicated some atherosclerotic lesion progression linked to various cyto- single nucleotide polymorphisms (SNPs) of LTA gene, such kines (Berliner et al. 1995; Ross 1999). Lymphotoxin- as A allele of C+804A polymorphism and C allele of alpha (LTA), a pro-inflammatory cytokine, plays an impor- T+496C polymorphism, were in predisposition to CAD or tant part in activating inflammatory and immunomodulatory MI, but the results were not often reproducible (Clarke et processes, suggesting its implication in the pathogenesis of al. 2006; Asselbergs et al. 2007; Park et al. 2007; Sedlacek atherosclerosis and CAD (Schreyer et al. 2002; Ozaki et al. et al. 2007; Palikhe et al. 2008; Ryan et al. 2008; Gao et al. 2004). 2010; Li et al. 2010). The LTA gene, firstly identified in 2002, was found to Suna et al. (2008) reported that LTA stimulation might be independently associated with myocardial infarction up-regulate various cytokines’ expression associated with (MI) in Japanese (Ozaki et al. 2002). Subsequently, several signal transduction, cell adhesion and chemoattraction, such studies confirmed the relationship between LTA gene and as vascular cell adhesion molecule 1 (VCAM-1), E-selectin Received December 27, 2010; revision accepted for publication May 11, 2011. doi: 10.1620/tjem.224.119 Correspondence: Wei Jin, M.D., PH.D., Department of Cardiology, Ruijin Hospital, Shanghai Jiaotong University School of Medicine, 197 Ruijin Er Road, Shanghai 200025, P.R. China. e-mail: [email protected] 119 120 Y. Liu et al. and monocyte chemotactic protein 1 (MCP 1) in human Genotyping analysis endothelial cells via the activation of the transcription Genomic DNA was extracted from peripheral blood leukocytes nuclear factor of kappa light polypeptide gene enhancer in by standard phenol-chloroform extraction. The genotypes of four B-cells (NFkB) as in tumor necrosis factor (TNF) receptor evaluated SNPs were determined by the MassARRAY technology signaling by TNF-α (a family member of LTA). Studies on platform (Sequenom Inc., San Diego, California, USA). The test was the role of LTA in the procedure of monocyte adhesion to based on the principle of allele-specific primer extension reaction and the read-out of the result was realized on the matrix-assisted laser endothelial cells, which was an initial and crucial step of desorption/ionization time-of-flight mass spectrometry (MALDI-TOF atherosclerosis, also raised the contribution of LTA genetic MS) analytic system. All the polymerase chain reaction (PCR) prim- polymorphisms to the regulation of those cytokines (Suna ers and extension primers were designed by MassARRAY Assay et al. 2008, 2009). Asselbergs et al. (2007) revealed that the Design 2.0 software according to reference sequence (NT_007592). A allele of C+804A polymorphism was associated with ele- To provide a significant improvement in overall performance, 10-mer vated plasma level of VCAM-1 in CAD cases. It is thus of tag (5′-ACGTTGGATG-3′) was added to the 5′ ends of each PCR interest to confirm which or how many SNPs of LTA gene primer (Table 2). might have functional potentials to affect the final bioavail- The amplification mixture of total volume 5 μL contained: 1 μL ability of LTA, and further the development of CAD. of genomic DNA (10 ng/μL), 0.325 μL MgCl2 (25 mmol/L), 0.25 μL It has been postulated that some SNPs in the putative dNTP (10 mmol/L), 0.625 μL 10 × buffer, 0.125 μL of each primer (10 promoter, transcriptional-regular regions or coding regions μmol/L) and 0.03 μL Hotstar® Taq DNA polymerase (all from Qiagen could affect the efficacy of transcription or the function of Inc., Valencia, California, USA). The amplification conditions were: protein, while haplotype analysis, which focuses on SNPs 95°C for 15 min, followed by 45 cycles at 95°C for 20 s, 56°C for in their combination simultaneously, has a higher complex- 30 s, 72°C for 1 min, then a final extension at 72°C for 3 min. ity level than single-locus analysis and more power to Preparation for extend reaction included incubation of 5 μL PCR explore the association between candidate genes and com- products with 0.3 μL shrimp alkaline phosphatase (SAP, Sequenom Inc., San Diego, California, USA) and 0.17 μL hME buffer at 37°C plex disease (Rioux et al. 2001; Tabor et al. 2002; Buckland for 20 min, followed by heat inactivation at 85°C for 5 min. et al. 2004). In the present study, to explore the association The extension mixture contained, in a final volume of 9 μL: of SNPs, both individually and as haplotypes, with the risk 7 μL of purified PCR products, 0.2 μL homogeneous MassEXTEND of CAD in Han Chinese, we focused on four SNPs: G+10A Mix (Sequenom Inc., San Diego, California, USA), 9 μM of each (rs1800683) and A+80C (rs2239704) in the 5′-untranslated extend primer and 0.018 μL MassEXTEND enzyme (Sequenom Inc., region encoded by exon 1, T+496C in exon 2 (rs2229094, a San Diego, California, USA). The extension conditions were: 94°C non-synonymous coding SNP that substitutes arginine for for 2 min, followed by 55 cycles at 94°C for 5 min, 52°C for 5 min, cysteine at codon 13, Cys13Arg), and C+804A in exon 3 72°C for 5 min, then a final extension at 72°C for 3 min. PCR (rs1041981, a non-synonymous coding SNP resulting in a products were purified with 3 mg of clean Resin (Sequenom Inc., San threonine to asparagine amino acid substitution at codon Diego, California, USA) and then sequenced using MassARRAY 26, Thr26Asn). Analyzer, version 3.0.1 (Sequenom Inc., San Diego, California, USA) on 384-well SpectroCHIP (Sequenom Inc., San Diego, California, Materials and Methods USA). Study population Our study population comprised 768 unrelated Han Chinese Statistical methods who were admitted to Ruijin Hospital, Shanghai Jiaotong University Continuous variables were expressed as mean ± standard School of Medicine when they were experiencing various symptoms deviation (S.D.) and compared by the unpaired Student's t-test. or for a medical checkup from June 2006 to June 2008. Participants Genotype/allele frequencies were calculated by the counting method were divided into 2 groups. The CAD group contained 438 patients and the differences in genotype/allele frequencies between CAD 2 aged 61.11 ± 9.66 years and the diagnosis of CAD was established group and healthy controls were analyzed by χ test. Each genotype angiographically in the presence of more than 50% stenosis in at least was assessed by logistic regression analysis assuming additive (major one of the three major coronary arteries or major branches. Patients homozygotes vs. heterozygotes vs. minor homozygotes), dominant with simple spasm of coronary arteries, myocardial bridge or other (major homozygotes vs. heterozygotes plus minor homozygotes) and non-coronary atherosclerotic lesions were excluded. The healthy recessive (major homozygotes plus heterozygotes vs.