- Email: [email protected]
Association study of AGER gene polymorphism and hypertension in Han Chinese population
Gene 498 (2012) 311–316
Contents lists available at SciVerse ScienceDirect
Gene journal homepage: www.elsevier.com/locate/gene
Short Communication
Association study of AGER gene polymorphism and hypertension in Han Chinese population Song Yang a, 1, Hairu Wang b, 1, Yichun Yang a, Wen Wang a, Jiandong Jiang a, Xianghai Zhao a, Qinglian Du a, Xuecai Wang c, Yingshui Yao d, Hongbing Shen b, Chong Shen b, e,⁎, Yanping Zhao f,⁎⁎ a
Department of Cardiology, Affiliated Yixing People's Hospital of Jiangsu University, People's Hospital of Yixing City, Yixing 214200, People's Republic of China Department of Epidemiology and Biostatistics, School of Public Health, Nanjing Medical University, Nanjing 210029, People's Republic of China Department of Clinical Laboratory, Affiliated Yixing People's Hospital of Jiangsu University, People's Hospital of Yixing City, Yixing 214200, People's Republic of China d Department of Preventive Medicine, Wannan Medical College, Wuhu 241001, People's Republic of China e Key Laboratory of Human Functional Genomics of Jiangsu Province, Nanjing Medical University, Nanjing 210029, People's Republic of China f Department of Neurology, Affiliated Yixing People's Hospital of Jiangsu University, People's Hospital of Yixing City, Yixing 214200, People's Republic of China b c
a r t i c l e
i n f o
Article history: Accepted 29 January 2012 Available online 13 February 2012 Keywords: Advanced glycation end products Receptor of advanced glycation end products Hypertension Association study
a b s t r a c t Background: Advanced glycation end products (AGEs) are produced by non-enzymatic glycation or glycoxidation of proteins, lipids and nucleic acids. The bond of AGEs and the receptor of AGE (AGER) in a pro-oxidant environment could induce immune and inflammation reaction involved in progress of microvascular disease. Accumulated evidence warrant further study on AGE–AGER pathway and genetic susceptibility to hypertension (HT). Methods: We designed a two-stage association study to evaluate the association of AGER polymorphism and HT. In stage 1, seven tagSNPs were tested in 524 cases and 531 controls and the significant SNPs (P b 0.05) would enter into stage 2 including 807 cases and 869 controls. Furthermore, joint analysis was performed for all 2731 subjects including 1331 cases and 1400 controls, and meta-analysis was applied to evaluate combined estimations from the subgroups of stage 1 and stage 2. Results: In stage 1, rs204994 had significant association with HT (P b 0.05) and enter stage 2. Neither joint analysis nor meta-analysis found statistical association of rs204994 with HT after adjusted for the covariates in the whole population. However, further stratification analysis found that rs204994 was significantly associated with HT in b 50 years and ≥50 years groups, ORs (95%CI) of dominant model were 1.623 (1.054–2.500) and 0.721 (0.546–0.952) respectively. No significant correlation was found between blood pressure and the polymorphisms of rs204994. Conclusions: Our data suggests that age might modulate the genetic effects of variation of rs204994 in AGER on HT and further replications in other populations and functional studies should be warranted. Crown Copyright © 2012 Published by Elsevier B.V. All rights reserved.
Abbreviations: AGEs, advanced glycation end products; AGER, receptor for AGE; Ang, angiotensin; BMI, body mass index; χ2, Chi-square; CHB, Chinese population in Beijing; CAD, coronary artery disease; DBP, diastolic blood pressure; esRAGE, endogenous secretory RAGE; EDTA, ethylenediamine tetraacetic acid; GLM, general linear model; GLU, glucose; HWE, Hardy–Weinberg equilibrium; HDL-C, high density lipoprotein cholesterol; HMGB1, high-mobility group box-1; HT, hypertension; LD, linkage disequilibrium; MAF, minor allele frequency; NF-κB, nuclear factor kappa-B; OR, odds ratio; CI, 95% confidence; PBX2, interval pre-B-cell leukemia homeobox 2; PCR-RFLP, polymerase chain reaction and restriction fragment length polymorphism; P, probability; RAS, renin–angiotensin system; ROS, reactive oxygen species; SNPs, single nucleotide polymorphisms; SHRSP, spontaneously hypertensive rats; SPSS, Statistical Product and Service Solutions; SBP, systolic blood pressure; tagSNPs, tagger SNPs; TLR, toll-like receptor; TFBS, transcript factor banding site; TGR, transgenic rat; TG, triglycerides; TC, total cholesterol. ⁎ Correspondence to: Chong Shen, Department of Epidemiology and Biostatistics, School of Public Health, Nanjing Medical University, No. 140 Hanzhong Road, Nanjing 210029 People's Republic of China. Tel.: + 86 25 86862815; fax: + 86 25 86527613. ⁎⁎ Correspondence to: Yanping Zhao, Department of Neurology, Affiliated Yixing People's Hospital, Jiangsu University, Yixing 214200, People's Republic of China. Tel.: + 86 510 87921000; fax: + 86 510 7902967. E-mail addresses: [email protected] (C. Shen), [email protected] (Y. Zhao). 1 Song Yang and Hairu Wang equally contributed to this work.
1. Introduction Advanced glycation end products (AGEs) are new compounds or modified structures produced among non-enzymatic glycation and glycoxidation of proteins, lipids and nucleic acids in a prooxidant environment (Schmidt et al., 1994; Bierhaus et al., 1998). Series of reports indicate that AGEs induced glycoxidative reaction plays an important role in the progress of aging disorders (McEniery et al., 2007; Farmer and Kennedy, 2009; Hallam et al., 2010; Sawabe, 2010). Both oxidative and carbonyl stress can cause severe damage to biological structures of proteins, lipids and nucleic acids and further induce inflammatory response (Schmidt et al., 1994). Accumulation of glycoxidation products would upregulate transcription and expression of cytokines, growth factors and adhesive molecules via AGE– AGER (receptor for AGE) interaction and subsequently increase classical acute phase reactants (e.g. C-reactive protein or orosomucoid) (Stitt et al., 2000; Kalousova et al., 2005).
0378-1119/$ – see front matter. Crown Copyright © 2012 Published by Elsevier B.V. All rights reserved. doi:10.1016/j.gene.2012.01.080
312
S. Yang et al. / Gene 498 (2012) 311–316
Recent evidence proved that the interaction of AGE-ligation and AGER could induce immune and inflammation through increasing production of cellular reactive oxygen species (ROS) and persistent activation of the proinflammatory transcription factor, nuclear factor kappa-B (NF-κB) (Schmidt et al., 1994; Schmidt et al., 2001; Bierhaus et al., 2005; Yan et al., 2009; Yao and Brownlee, 2010). Also, study on transgenic rat (TGR) model with angiotensin (Ang) II-dependent hypertension (HT) and renal damage has proved the interrelation between renin–angiotensin system (RAS) and the AGE/ AGER axis in promoting cardiovascular end-organ damage (Bohlender et al., 2005; Koyama et al., 2007b; Yamagishi et al., 2008; Ramasamy et al., 2009). On the other hand, high-mobility group box-1 (HMGB1) is a newly recognized potent innate “danger signal” to the proinflammatory response via AGER and members of the toll-like receptor (TLR) family (van Zoelen et al., 2008). The interaction of HMGB1 and AGER plays a critical role in initiating and sustaining the inflammatory response in inflammatory cardiomyopathy eventually leading to heart failure and coronary artery disease (CAD) (Gao et al., 2010; Volz et al., 2010). Additionally, studies observed that blood pressure and oxidative damage were significantly reduced in stroke-prone spontaneously hypertensive rats (SHRSP) after long-term administration of AGEs inhibitor OPB-9195, and suggested that AGE–AGER pathway might play an important role in the progression or maintenance of genetic hypertension and related diseases (Mizutani et al., 2002; Steckelings et al., 2009). Evidences from population also showed that serum carboxymethyllysine was associated with increased aortic pulse wave velocity and arterial stiffness in adults (Mizutani et al., 2002; Semba et al., 2009). Furthermore, hypertensive patients with lower tissue AGE accumulation presented a larger improvement in diastolic function in response to anti-hypertensive eprosartan. Other drugs such as ARBs could modulate AGER isoform expressions by correcting endothelial dysfunction and were already used for hypertension or diabetes treatment (Koyama et al., 2007b; Grossin et al., 2010; Hartog et al., 2010; Maeda et al., 2011). Those findings would be helpful to illustrate microvascular, macrovascular disease and diabetic complications. Meaningfully, well understanding AGER mediated vascular pathogenesis would lead new insight to prognosis evaluation and efficient therapeutic strategies for cardiovascular disease. As a complex polygenic disease, human essential hypertension is affected by varying combinations of genetic and environmental factors (Staessen et al., 2003). Those comprehensive evidences suggested that AGER mediated macrovascular impairment might significantly affect the pathogenesis of hypertension, and so, whether the AGER gene polymorphisms contribute genetic susceptibility to HT would be a concern. In the present study, we designed a two-stage association study, which could provide near-optimal power to detect the true marker conferring disease risk while substantially reducing the total number of marker evaluations (Satagopan and Elston, 2003; Satagopan et al., 2004) to evaluate the association of AGER gene with HT in Han Chinese population. 2. Methods 2.1. Subjects In total, 2731 adult subjects aged 18–62 years were recruited from a rural population of 14,469 subjects in Jiangsu province by an epidemiological stratification sampling approach. In the present study, 1331 hypertensive cases with systolic blood pressure (SBP) ≥140 mm Hg, and/or diastolic blood pressure (DBP) ≥90 mm Hg, or currently administering anti-hypertensive medication were included and the patients with a clinical history of secondary hypertension,
coronary heart disease, kidney disease and diabetes were excluded. 1400 age- and sex-matched normotensives with SBP b140 mm Hg and DBP b90 mm Hg were selected as controls from the same target study population. We designed a two-stage association study and randomly selected 524 cases and 531 controls as stage 1 subsample to screen positive single nucleotide polymorphisms (SNPs). In stage 2, the remaining sample of 1676 subjects included 807 cases and 869 controls. We take initial associations found in stage 1 as hypotheses and further test in stage 2. The ethics committee of Nanjing Medical University has approved the research protocol and all subjects accepted written informed consent during epidemiological interviews. Trained research staff administered a standard questionnaire to obtain information on demographic characteristics including age, gender, nation, education, occupation and status of smoking and drinking. Three BP measurements were obtained from each participant by certified observers according to a standard protocol recommended by the American Heart Association (Perloff et al., 1993). Characteristics of study population of stage 1, stage 2 and joint main sample with stage 1 plus stage 2 were listed in Table 1. 2.2. SNP selection AGER gene (GenBank ID: 177), also known as RAGE, is mapped on chromosome 6p21.3. AGER spans 3.278 kbps (32,148.711– 32,151.988 kbps) and consists of 11 exons. We searched SNPs covered AGER and upstream to test and found pre-B-cell leukemia homeobox 2 (PBX2) gene (GeneID: 5089) at upstream 486 bps of AGER. Thus, we selected candidate SNPs covered AGER gene and upstream 6 kbps including PBX2 that spans 5.454 kbps (32,152.482–32,157.935 kbps) and consists of 9 exons. We selected tagger SNPs (tagSNPs) from the data of Han Chinese population in Beijing, China (CHB) in HapMap (HapMap Data Rel 24/phase II Nov08, on NCBI B36 assembly, dbSNP b126). All tagSNPs were selected with minor allele frequency (MAF) ≥0.05 and r 2 ≥ 0.8 according to the linkage disequilibrium (LD) values. We applied a functional candidate strategy to search functional SNPs with the role of transcription, regulating or splicing, etc. on the website of selection tool for single nucleotide polymorphisms (FASTSNP, http://fastsnp.ibms. sinica.edu.tw) (Yuan et al., 2006). SNPs with predictive biological effects would be enforced to rerun as tagSNPs by the software of Haploview. Thus, eight tagSNPs that were located in putative functional regions of the gene, e.g., exons, promoter/regulatory region, transcript factor banding site (TFBS) or splicing regulation were listed in Table 2. All selected SNPs were validated with 96 DNA samples in genotype test and rs2269422 with lower MAF (0.010) in controls was excluded. Finally, seven tagSNPs were included in the genotyping experiments. 2.3. Blood sampling and SNP genotyping All the subjects were informed to keep an empty stomach over 12 h prior to E blood sampling in the morning. Blood for genotyping was taken into ethylenediamine tetraacetic acid (EDTA)-containing receptacles. SNP genotyping used an improved polymerase chain reaction and restriction fragment length polymorphism (PCR-RFLP) method with pink dye as indicator and high throughout electrophoresis method with 96-sample agarose gel block. Blood pressure and history of hypertension or medication status were mixed blindly for composition of random samples. Ninety-six randomly selected samples were genotyped twice for duplication accuracy and then sequencing analyses for 16 randomly selected samples were done respectively using a Big-dye sequence kit (Applied Biosystem, ABI model 377 genetic analyzer, Foster City, USA).
S. Yang et al. / Gene 498 (2012) 311–316
313
Table 1 Comparison of clinical characteristics between cases and controls in stage 1, stage 2 and joint analysis. Characteristics
Male (%) Age (years) SBP (mm Hg) DBP (mm Hg) BMI (kg/m2) TC (mmol/L) HDL-C (mmol/L) TG (mmol/L) GLU (mmol/L) Smokers (%) Drinkers (%)
Stage 1
Stage 2
Combined sample of stage 1 plus stage 2
Cases (n = 524)
Controls (n = 531)
Cases (n = 807)
Controls (n = 869)
Cases (n = 1331)
Controls (n = 1400)
34.0 50.78 ± 6.05 146.70 ± 18.52⁎,⁎⁎ 90.45 ± 10.54⁎,⁎⁎ 26.45 ± 3.58⁎ 4.37 ± 0.99⁎ 1.49 ± 0.35⁎ 1.87 ± 1.22⁎ 4.72 ± 1.36⁎
34.8 50.53 ± 6.12 110.16 ± 9.7⁎⁎ 67.50 ± 7.32⁎⁎ 21.18 ± 2.04 3.95 ± 0.70 1.75 ± 0.38 0.89 ± 0.32 4.17 ± 0.53 20.0 25.6
33.6 51.49 ± 7.08 151.17 ± 17.56⁎ 92.28 ± 10.79⁎ 26.38 ± 3.78⁎ 4.27 ± 0.94⁎ 1.50 ± 0.38⁎ 1.82 ± 1.29⁎ 4.73 ± 1.40⁎
35.2 50.87 ± 6.94 111.54 ± 10.17 68.53 ± 7.46 21.44 ± 2.03 3.9 ± 0.69 1.72 ± 0.34 0.89 ± 0.45 4.25 ± 0.56 18.9 27.3
33.7 51.22 ± 6.55 149.41 ± 18.7⁎ 91.56 ± 10.73⁎ 26.41 ± 3.70⁎ 4.31 ± 0.96⁎ 1.49 ± 0.37⁎ 1.85 ± 1.27⁎ 4.73 ± 1.38⁎
35.1 50.89 ± 6.64 111.02 ± 10.01 68.15 ± 7.42 21.34 ± 2.04 3.92 ± 0.70 1.73 ± 0.34 0.89 ± 0.41 4.22 ± 0.55 19.1 26.6
17.9 28.2
15.4 26.3
16.4 27.0
⁎ P b 0.0001 for comparison with corresponding controls. ⁎⁎ P b 0.05 for comparison with corresponding variables in stage 2.
2.4. Statistical analysis Unpaired Student's t-test was used to test the differences of all measured variables that were presented as the means ± SD between cases and controls. The Chi-square (χ 2) test was used to compare qualitative variables and the allele and genotype frequency distributions between cases and controls and a P value of 0.05 was defined to be statistically significant. Multiple unconditional logistic regression was applied to adjust for covariates including age, gender, body mass index (BMI), glucose (GLU), triglycerides (TG), total cholesterol (TC), high density lipoprotein cholesterol (HDL-C), smoking and drinking status. The criterion for SNPs entering into stage 2 study was that the SNPs had a two-tailed probability (P) value of b0.05 for the comparisons of allele or genotype frequency distributions between cases and controls in stage 1. Hardy–Weinberg equilibrium (HWE) was assessed by Fisher's exact χ 2 test using the program HWE in the control groups. In stage 2, joint analysis and meta-analysis were performed in the main study population (Cantor et al., 2010). Meta-analysis takes advantage of the high accuracy of effect size and increase statistical power. Random effect model, which may well be conservative compared with a fixed effects model when the primary goal is hypothesis testing, was selected and inverse-variance method was employed to estimate combined adjusted odds ratio (OR) and its 95% confidence interval (CI) (de Bakker et al., 2008). The two-sided Z-test with the corresponding P-value was summed across two studies weighing them by the per study sample size to test the significance of single SNP in joint study. We assessed heterogeneity among study populations with the Cochran's Q-statistic and I 2 statistic that measures the proportion of total variance in estimated ORs due to heterogeneity. A value of the I 2 statistic greater than 50% indicates the high degree of heterogeneity in estimated ORs across study populations.
Because of AGEs induced glycoxidative reaction has significant age-specified aging effects, and age often serves as a surrogate for a variety of interacting covariates and which may not only enhance gene discovery (Shi et al., 2009), in the present study, but also age stratification analysis was further performed. We divided age into b50 years and ≥ 50 years groups for SBP and DBP of human over 50 years often increase exponentially with age (Safar et al., 2004). Additionally, general linear model (GLM) was applied to compare blood pressure levels between genotypes. Statistical analyses as above were performed with Statistical Product and Service Solutions 13.0 (SPSS; SPSS Inc, Chicago, USA) and the statistics of meta-analysis were calculated in the STATA software package (College Station, Texas, USA). 3. Results 3.1. Clinical characteristics The characteristics of subjects included in the stage 1, stage 2 and joint analysis were listed in Table 1. Age and the proportions of men, smoking and drinking status were not significantly different between cases and controls in stage 1, stage 2 and main sample. As expected, the cases generally had higher SBP and DBP, BMI, TC, TG, and GLU levels and lower HDL-C than that in controls. Both SBP and DBP in stage 1 were a little lower than that of stage 2 (P b 0.05). 3.2. Association analyses In stage 1, all SNPs in controls were in HWE except rs2022059 (P = 0.006). Out of seven candidate SNPs, SNP rs204994 presented statistical correlations with HT and P values of recessive model and
Table 2 Biological information analysis for candidate SNPs in AGER gene. TagSNPs
Region
Allelesa
TFBS/AA code
Biological effect
MAFb,c
rs1800625 rs1800624 rs2022059 (PBX2) rs2070600 rs204994 (PBX2)d rs204993 (PBX2)d rs184003 rs2269422
5upstream 5upstream Intron Coding 5′ upstream 5′ upstream Intron Intron
A/G A/T C/G C/T G/A C/T C/T C/T
E2F – – v-Myb AML-1a, SREBP-1 MZF1 GATA-1, GATA-1 CRE-BP, deltaE, C/EBPa
Promoter/regulatory region Upstream Intronic enhancer Missense (non-conservative) splicing regulation Promoter/regulatory region Promoter/regulatory region Intronic enhancer Intronic enhancer
0.132/0.122 0.156/0.163 0.195/0.250 0.225/0.289 0.254/0.200 0.456/0.478 0.173/0.122 0.010/0.057
a b c d
Major/minor allele. MAF in the control. MAF in CHB. Variations of rs204994 and 204993 predicted as intronic enhancer for PBX2 gene.
314
S. Yang et al. / Gene 498 (2012) 311–316
Table 3 Association analyses of rs204994 and HT. Stage
Stage 1 Stage 2 Joint analysis
Genotype and OR (95%CI)a
Group
Case Control Case Control Case Control
WT
Ht + MT
GG
GA + AA
321 305 504 552 825 857
178 + 25 183 + 42 275 + 28 288 + 29 453 + 53 471 + 71
Meta-analysis
Allele Dominant
Major/minor
OR (95%CI)
G/A 0.857 (0.670–1.096) P = 0.220 1.047 (0.858–1.277) P = 0.651 0.935 (0.743–1.176) P = 0.564 0.915 (0.726–1.154) P = 0.455
0.782/0.218 0.795/0.295 0.790/0.210
0.818 (0.668–1.000) P = 0.049 1.038 (0.877–1.229) P = 0.665 0.947 (0.833–1.078) P = 0.413 0.899 (0.784–1.031) P = 0.129
WT wide type, Ht heterozygote, MT mutant type. a ORs were adjusted for age, sex, BMI, TC, TG, HDL-C, GLU, smoking and drinking.
allele were 0.020 and 0.049 respectively (Table 3). Thus, SNP rs204994 entered into stage 2 and further genotyped. The genotype and allele distributions of the seven SNPs between cases and controls were seen in Table S1 in supporting information. In stage 2, the association of rs204994 with HT wasn't replicated even after adjusted for the covariates, including age, sex, BMI, GLU, TC, TG, HDL-C, smoking and drinking status by logistic regression analysis. Meta-analysis with genetic variance of random effect model in the two stage samples presented well homogeneity (I 2 = 0 for additive, dominant and recessive models and I 2 = 30.4% for allele), and the results indicated that there was no statistical association of genotype or allele with HT (P > 0.05). 3.3. Stratification analysis Further age stratification analysis was conducted for the association of rs204994 and HT. In b50 years age group, A allele variation presented significant risk effect for HT and OR (95%CI) of dominant model of rs204994 and was 1.623 (1.054–2.500) and the association still presented statistical significance after adjusted for sex, TC, TG, HDL-C, GLU, BMI smoking and drinking. On the contrary, in ≥50 years age group, the genetic variation showed significant protective effects for HT and ORs (95%CI) of additive and dominant models and were 0.747 (0.592–0.943) and 0.721 (0.546–0.952) respectively, and also the association still presented statistical significance after adjusted for covariates. Table 4 listed the adjusted ORs and P values. However, there was no statistical correlation found in stratification analyses by sex, smoking and drinking even after adjusted for covariates (data were not listed here). 3.4. Analysis of quantitative trait for blood pressure Blood pressure levels were compared between GG, GA and AA genotypes of rs204994 and the results showed that no significant differences of SBP and DBP were found between the different genotypes either in subjects with hypertension history (n = 845) or in subjects without hypertension history (n = 1886) (Table 5). Additionally, no
significant interaction of age and gene was found for SBP or DBP (P > 0.1). 4. Discussion Many studies have reported that AGEs play important roles in aging disorders such as diabetes and diabetic cardiovascular complication via interaction of AGEs and cellular AGER (Bierhaus et al., 1998; Stitt et al., 2000; Koyama et al., 2007a; Yamagishi et al., 2008; Nakamura et al., 2009; Steckelings et al., 2009; Hallam et al., 2010). AGER is also profoundly associated with macrovascular complications through regulation of atherogenesis, angiogenic response, vascular injury, and inflammatory response (Koyama et al., 2007b) and nifedipine might provide an anti-inflammatory effect against AGEs in tubular cells by suppressing RAGE expression via PPARgamma activation (Matsui et al., 2010). Likewise, plasma endogenous secretory RAGE (esRAGE) may act as a protective factor against and a novel biomarker for the occurrence of metabolic syndrome and cardiovascular diseases (Koyama et al., 2007a; Nakamura et al., 2009). Therefore, the contribution of AGER to the vascular pathogenesis in human hypertension would deserve further investigation and that would advance our knowledge of the determinants of atherogenesis, angiogenic response, vascular injury and inflammatory response and disease. Recently, a few studies reported that AGER genetic polymorphism had significant association with diabetic CAD, CAD and stroke (dos Santos et al., 2005; Falcone et al., 2008; Gao et al., 2010), but less study investigates the genetic effect of AGER on HT except a relatively small sample study without correlative findings. In the present study, we conducted a two-stage case–control study and selected seven functional tagSNPs to assess the association of AGER gene with HT in Han Chinese population. The power was calculated with the software of Power and Sample Size Calculation (Dupont WD, Plummer WD: http://biostat.mc.vanderbilt.edu/twiki/ bin/ view/Main/PowerSampleSize). At the 5% significance level, we had 84.1% power to detect a MAF of 0.05 that confers OR of 1.4. Actually, rs1800625 has lower MAF 0.132 in controls and the power for OR equal to 1.3 is 92.8%. Furthermore, we use a meta-analysis, which due to increased power and validation may provide more
Table 4 Stratification analysis of rs204994 with hypertension by age. Stratification factor
Age b50 y ≥ 50 y a
Group
Case Control Case Control
Genotype and OR (95%CI)a WT
Ht + MT
GG
GA + AA
288 359 537 498
173 + 24 171 + 24 280 + 29 298 + 49
Adjusted for sex, BMI, TC, TG, HDL-C, GLU, smoking and drinking.
Additive
Dominant
Recessive
1.430 (0.993–2.059) P = 0.055 0.769 (0.611–0.968) P = 0.025
1.654 (1.071–2.554) P = 0.023 0.751 (0.571–0.989) P = 0.042
0.990 (0.349–2.806) P = 0.984 0.560 (0.295–1.065) P = 0.077
S. Yang et al. / Gene 498 (2012) 311–316 Table 5 Comparisons of quantitative traits of blood pressure between genotypes of rs204994 by treatment stratification. Group
Treatment (n) SBP (mm Hg) DBP (mm Hg) Non-treatment (n) SBP (mm Hg) DBP (mm Hg)
Genotype GG
GA
AA
525 146.97 ± 19.12 89.96 ± 10.97 1157 122.44 ± 22.29 75.17 ± 14.36
289 144.71 ± 19.38 89.09 ± 10.98 633 122.72 ± 22.88 74.76 ± 13.98
31 138.85 ± 18.35 89.24 ± 11.95 95 121.76 ± 21.08 75.37 ± 12.93
P valuea
0.118 0.712 0.677 0.867
P > 0.05 for all test of homogeneity of variances. a Adjusted for sex, BMI, TC, TG, HDL-C, GLU, smoking and drinking.
robust risk estimates and its 95%CI from direct joint analysis (de Bakker et al., 2008; Cantor et al., 2010), to assess heterogeneity among study populations and results indicated that genetic variance of random effect model in the two stage samples presented well homogeneity as described above. On the other hand, the ORs were adjusted for potential confounders such as age, gender, BMI, GLU, TC, TG, HDL-C, drinking and smoking status and that would effectively reduce the false positive findings in the present study. Some limitations might cause false positive association in the present study. Firstly, all the subjects came from a cross sectional population but not a follow-up cohort population. Secondly, false positive association would present in stage 1 if potential differentiation for covariates or population structure exists in the two stage samples. Thirdly, genetic heterozygosity that arose from the inner research population might devote different susceptibilities to HT in subgroups, and phenocopy for other disease often falsified results in association study. Finally, potential bias including information bias, selective bias and confounding bias often distort results for epidemiological association studies. Regardless of the limitation described as above, the present study identified that rs204994 of AGER had a positive association with HT in subgroup with b50 years for the first time. As a complex biological construction, age may serve as a surrogate for a variety of interacting covariates and render findings from across studies more comparable to each other (McEniery et al., 2007; Shi et al., 2009). Recently, several genes were reported to be involved in blood pressure regulation and the susceptibility to hypertension modulated by age (Dao et al., 2005; Lunetta et al., 2007; Shen et al., 2009; Shi et al., 2009). Furthermore, the statistical analysis strategy of age-genetic effects can prevent potential confounding bias modified by age-varying characteristics (Lunetta et al., 2007; Lasky-Su et al., 2008; Percy et al., 2009). According to the findings that SBP and DBP increase exponentially with age over 50 years (Safar et al., 2004), we divided age into b50 and ≥50 years groups to distinguish whether age might modulate the genetic effect of AGER on HT by stratification analysis (Dao et al., 2005; Lasky-Su et al., 2008). In the present study, the results indicated that rs204994 had positive association with HT in b50 years group but negative association in ≥50 years group and after adjusted for covariates respectively whereas no significant correlation was found between blood pressure and genetic polymorphism. Our results support that age might play an important modulating role in AGER genetic effect on the susceptibility to HT. Meanwhile, age's modification would blanket over real association for HT if just meta-analysis proposed for combined samples as above. Remarkably, rs204994 (G/A) is located at 5′ upstream promoter/ regulatory region of AGER and in the overlapping intron 6 of PBX2. Variation of G to A may predict buildups of TFBS of AML-1a and SREBP-1. Therefore, the positive association modulated by age found in the present study would be helpful to further research potential gene function to pathogenesis of vascular lesions. Meanwhile, it is desirable to verify the SNP's biological function contributing to AGER or PBX2 which was seldom reported to associate with vascular
315
disorders. Hopefully, our findings might facilitate further research for the benefits of using surrogate markers for pharmacogenetics of antihypertensive and prognosis evaluation. In conclusion, our finding suggests that age might modulate the genetic effects of AGER gene on the susceptibility to HT but not blood pressure trait. Further replication for association study or biological functional study would be warranted in the future. Supplementary materials related to this article can be found online at doi:10.1016/j.gene.2012.01.080. Acknowledgment This work was supported by the National Natural Science Foundation of China (Grant No. 30800947 and Grant No. 81072367), Health Research Program of Jiangsu Province (Grant No. H200839) and Natural Science Foundation of Jiangsu Province (Grant No. BK2011776). References Bierhaus, A., Hofmann, M.A., Ziegler, R., Nawroth, P.P., 1998. AGEs and their interaction with AGE-receptors in vascular disease and diabetes mellitus. I. The AGE concept. Cardiovasc. Res. 37, 586–600. Bierhaus, A., Humpert, P.M., Morcos, M., Wendt, T., Chavakis, T., Arnold, B., Stern, D.M., Nawroth, P.P., 2005. Understanding RAGE, the receptor for advanced glycation end products. J. Mol. Med. 83, 876–886. Bohlender, J., Franke, S., Sommer, M., Stein, G., 2005. Advanced glycation end products: a possible link to angiotensin in an animal model. Ann. N. Y. Acad. Sci. 1043, 681–684. Cantor, R.M., Lange, K., Sinsheimer, J.S., 2010. Prioritizing GWAS results: a review of statistical methods and recommendations for their application. Am. J. Hum. Genet. 86, 6–22. Dao, H.H., Essalihi, R., Bouvet, C., Moreau, P., 2005. Evolution and modulation of agerelated medial elastocalcinosis: impact on large artery stiffness and isolated systolic hypertension. Cardiovasc. Res. 66, 307–317. de Bakker, P.I., Ferreira, M.A., Jia, X., Neale, B.M., Raychaudhuri, S., Voight, B.F., 2008. Practical aspects of imputation-driven meta-analysis of genome-wide association studies. Hum. Mol. Genet. 17, R122–R128. dos Santos, K.G., Canani, L.H., Gross, J.L., Tschiedel, B., Pires Souto, K.E., Roisenberg, I., 2005. The –374A allele of the receptor for advanced glycation end products gene is associated with a decreased risk of ischemic heart disease in African-Brazilians with type 2 diabetes. Mol. Genet. Metab. 85, 149–156. Falcone, C., Geroldi, D., Buzzi, M.P., Emanuele, E., Yilmaz, Y., Fontana, J.M., Vignali, L., Boiocchi, C., Sbarsi, I., Cuccia, M., 2008. The –374T/A RAGE polymorphism protects against future cardiac events in nondiabetic patients with coronary artery disease. Arch. Med. Res. 39, 320–325. Farmer, D.G., Kennedy, S., 2009. RAGE, vascular tone and vascular disease. Pharmacol. Ther. 124, 185–194. Gao, J., Shao, Y., Lai, W., Ren, H., Xu, D., 2010. Association of polymorphisms in the RAGE gene with serum CRP levels and coronary artery disease in the Chinese Han population. J. Hum. Genet. 55, 668–675. Grossin, N., Boulanger, E., Wautier, M.P., Wautier, J.L., 2010. The different isoforms of the receptor for advanced glycation end products are modulated by pharmacological agents. Clin. Hemorheol. Microcirc. 45, 143–153. Hallam, K.M., Li, Q., Ananthakrishnan, R., Kalea, A., Zou, Y.S., Vedantham, S., Schmidt, A.M., Yan, S.F., Ramasamy, R., 2010. Aldose reductase and AGE–RAGE pathways: central roles in the pathogenesis of vascular dysfunction in aging rats. Aging Cell 9, 776–684. Hartog, J.W., van de Wal, R.M., Schalkwijk, C.G., Miyata, T., Jaarsma, W., Plokker, H.W., van Wijk, L.M., Smit, A.J., van Veldhuisen, D.J., Voors, A.A., 2010. Advanced glycation end-products, anti-hypertensive treatment and diastolic function in patients with hypertension and diastolic dysfunction. Eur. J. Heart Fail. 12, 397–403. Kalousova, M., Zima, T., Tesar, V., Dusilova-Sulkova, S., Skrha, J., 2005. Advanced glycoxidation end products in chronic diseases—clinical chemistry and genetic background. Mutat. Res. 579, 37–46. Koyama, H., Yamamoto, H., Nishizawa, Y., 2007a. Endogenous secretory RAGE as a novel biomarker for metabolic syndrome and cardiovascular diseases. Biomark. Insight 2, 331–339. Koyama, H., Yamamoto, H., Nishizawa, Y., 2007b. RAGE and soluble RAGE: potential therapeutic targets for cardiovascular diseases. Mol. Med. 13, 625–635. Lasky-Su, J., Lyon, H.N., Emilsson, V., Heid, I.M., Molony, C., Raby, B.A., Lazarus, R., Klanderman, B., Soto-Quiros, M.E., Avila, L., Silverman, E.K., Thorleifsson, G., Thorsteinsdottir, U., Kronenberg, F., Vollmert, C., Illig, T., Fox, C.S., Levy, D., Laird, N., Ding, X., McQueen, M.B., Butler, J., Ardlie, K., Papoutsakis, C., Dedoussis, G., O'Donnell, C.J., Wichmann, H.E., Celedon, J.C., Schadt, E., Hirschhorn, J., Weiss, S.T., Stefansson, K., Lange, C., 2008. On the replication of genetic associations: timing can be everything! Am. J. Hum. Genet. 82, 849–858. Lunetta, K.L., D'Agostino Sr., R.B., Karasik, D., Benjamin, E.J., Guo, C.Y., Govindaraju, R., Kiel, D.P., Kelly-Hayes, M., Massaro, J.M., Pencina, M.J., Seshadri, S., Murabito, J.M., 2007. Genetic correlates of longevity and selected age-related phenotypes: a genome-wide association study in the Framingham Study. BMC Med. Genet. 8 (Suppl 1), S13.
316
S. Yang et al. / Gene 498 (2012) 311–316
Maeda, S., Matsui, T., Takeuchi, M., Yamagishi, S., 2011. Co-treatment with azelinidipine and olmesartan inhibits advanced glycation end products (AGEs) elicited down-regulation of adiponectin mRNA levels in cultured adipocytes partly via its anti-oxidative property. Int. J. Cardiol. 146, 264–266. Matsui, T., Yamagishi, S., Takeuchi, M., Ueda, S., Fukami, K., Okuda, S., 2010. Nifedipine inhibits advanced glycation end products (AGEs) and their receptor (RAGE) interaction-mediated proximal tubular cell injury via peroxisome proliferatoractivated receptor-gamma activation. Biochem. Biophys. Res. Commun. 398, 326–330. McEniery, C.M., Wilkinson, I.B., Avolio, A.P., 2007. Age, hypertension and arterial function. Clin. Exp. Pharmacol. Physiol. 34, 665–671. Mizutani, K., Ikeda, K., Tsuda, K., Yamori, Y., 2002. Inhibitor for advanced glycation end products formation attenuates hypertension and oxidative damage in genetic hypertensive rats. J. Hypertens. 20, 1607–1614. Nakamura, K., Adachi, H., Matsui, T., Kurita, Y., Takeuchi, M., Yamagishi, S., 2009. Independent determinants of soluble form of receptor for advanced glycation end products in elderly hypertensive patients. Metabolism 58, 421–425. Percy, C.J., Brown, L., Power, D.A., Johnson, D.W., Gobe, G.C., 2009. Obesity and hypertension have differing oxidant handling molecular pathways in age-related chronic kidney disease. Mech. Ageing Dev. 130, 129–138. Perloff, D., Grim, C., Flack, J., Frohlich, E.D., Hill, M., McDonald, M., Morgenstern, B.Z., 1993. Human blood pressure determination by sphygmomanometry. Circulation 88, 2460–2470. Ramasamy, R., Yan, S.F., Schmidt, A.M., 2009. RAGE: therapeutic target and biomarker of the inflammatory response—the evidence mounts. J. Leukoc. Biol. 86, 505–512. Safar, M.E., Lajemi, M., Rudnichi, A., Asmar, R., Benetos, A., 2004. Angiotensinconverting enzyme D/I gene polymorphism and age-related changes in pulse pressure in subjects with hypertension. Arterioscler. Thromb. Vasc. Biol. 24, 782–786. Satagopan, J.M., Elston, R.C., 2003. Optimal two-stage genotyping in population-based association studies. Genet. Epidemiol. 25, 149–157. Satagopan, J.M., Venkatraman, E.S., Begg, C.B., 2004. Two-stage designs for genedisease association studies with sample size constraints. Biometrics 60, 589–597. Sawabe, M., 2010. Vascular aging: from molecular mechanism to clinical significance. Geriatr. Gerontol. Int. 10 (Suppl 1), S213–S220. Schmidt, A.M., Hori, O., Brett, J., Yan, S.D., Wautier, J.L., Stern, D., 1994. Cellular receptors for advanced glycation end products. Implications for induction of oxidant stress and cellular dysfunction in the pathogenesis of vascular lesions. Arterioscler. Thromb. 14, 1521–1528.
Schmidt, A.M., Yan, S.D., Yan, S.F., Stern, D.M., 2001. The multiligand receptor RAGE as a progression factor amplifying immune and inflammatory responses. J. Clin. Invest. 108, 949–955. Semba, R.D., Najjar, S.S., Sun, K., Lakatta, E.G., Ferrucci, L., 2009. Serum carboxymethyllysine, an advanced glycation end product, is associated with increased aortic pulse wave velocity in adults. Am. J. Hypertens. 22, 74–79. Shen, C., Lu, X., Li, Y., Zhao, Q., Liu, X., Hou, L., Wang, L., Chen, S., Huang, J., Gu, D., 2009. Emilin1 gene and essential hypertension: a two-stage association study in northern Han Chinese population. BMC Med. Genet. 10, 118. Shi, G., Gu, C.C., Kraja, A.T., Arnett, D.K., Myers, R.H., Pankow, J.S., Hunt, S.C., Rao, D.C., 2009. Genetic effect on blood pressure is modulated by age: the Hypertension Genetic Epidemiology Network Study. Hypertension 53, 35–41. Staessen, J.A., Wang, J., Bianchi, G., Birkenhager, W.H., 2003. Essential hypertension. Lancet 361, 1629–1641. Steckelings, U.M., Rompe, F., Kaschina, E., Unger, T., 2009. The evolving story of the RAAS in hypertension, diabetes and CV disease: moving from macrovascular to microvascular targets. Fundam. Clin. Pharmacol. 23, 693–703. Stitt, A.W., Burke, G.A., Chen, F., McMullen, C.B., Vlassara, H., 2000. Advanced glycation end-product receptor interactions on microvascular cells occur within caveolinrich membrane domains. FASEB J. 14, 2390–2392. van Zoelen, M.A., Yang, H., Florquin, S., Meijers, J.C., Akira, S., Arnold, B., Nawroth, P.P., Bierhaus, A., Tracey, K.J., van der Poll, T., 2008. Role of Toll-like receptors 2 and 4, and the receptor for advanced glycation end products (RAGE) in Hmgb1 induced inflammation in vivo. Shock 31, 280–284. Volz, H.C., Kaya, Z., Katus, H.A., Andrassy, M., 2010. The role of HMGB1/RAGE in inflammatory cardiomyopathy. Semin. Thromb. Hemost. 36, 185–194. Yamagishi, S., Nakamura, K., Matsui, T., Ueda, S., Fukami, K., Okuda, S., 2008. Agents that block advanced glycation end product (AGE)–RAGE (receptor for AGEs)-oxidative stress system: a novel therapeutic strategy for diabetic vascular complications. Expert Opin. Investig. Drugs 17, 983–996. Yan, S.F., Ramasamy, R., Schmidt, A.M., 2009. Receptor for AGE (RAGE) and its ligands— cast into leading roles in diabetes and the inflammatory response. J. Mol. Med. 87, 235–247. Yao, D., Brownlee, M., 2010. Hyperglycemia-induced reactive oxygen species increase expression of the receptor for advanced glycation end products (RAGE) and RAGE ligands. Diabetes 59, 249–255. Yuan, H.Y., Chiou, J.J., Tseng, W.H., Liu, C.H., Liu, C.K., Lin, Y.J., Wang, H.H., Yao, A., Chen, Y.T., Hsu, C.N., 2006. FASTSNP: an always up-to-date and extendable service for SNP function analysis and prioritization. Nucleic Acids Res. 34, W635–W641.
Contents lists available at SciVerse ScienceDirect
Gene journal homepage: www.elsevier.com/locate/gene
Short Communication
Association study of AGER gene polymorphism and hypertension in Han Chinese population Song Yang a, 1, Hairu Wang b, 1, Yichun Yang a, Wen Wang a, Jiandong Jiang a, Xianghai Zhao a, Qinglian Du a, Xuecai Wang c, Yingshui Yao d, Hongbing Shen b, Chong Shen b, e,⁎, Yanping Zhao f,⁎⁎ a
Department of Cardiology, Affiliated Yixing People's Hospital of Jiangsu University, People's Hospital of Yixing City, Yixing 214200, People's Republic of China Department of Epidemiology and Biostatistics, School of Public Health, Nanjing Medical University, Nanjing 210029, People's Republic of China Department of Clinical Laboratory, Affiliated Yixing People's Hospital of Jiangsu University, People's Hospital of Yixing City, Yixing 214200, People's Republic of China d Department of Preventive Medicine, Wannan Medical College, Wuhu 241001, People's Republic of China e Key Laboratory of Human Functional Genomics of Jiangsu Province, Nanjing Medical University, Nanjing 210029, People's Republic of China f Department of Neurology, Affiliated Yixing People's Hospital of Jiangsu University, People's Hospital of Yixing City, Yixing 214200, People's Republic of China b c
a r t i c l e
i n f o
Article history: Accepted 29 January 2012 Available online 13 February 2012 Keywords: Advanced glycation end products Receptor of advanced glycation end products Hypertension Association study
a b s t r a c t Background: Advanced glycation end products (AGEs) are produced by non-enzymatic glycation or glycoxidation of proteins, lipids and nucleic acids. The bond of AGEs and the receptor of AGE (AGER) in a pro-oxidant environment could induce immune and inflammation reaction involved in progress of microvascular disease. Accumulated evidence warrant further study on AGE–AGER pathway and genetic susceptibility to hypertension (HT). Methods: We designed a two-stage association study to evaluate the association of AGER polymorphism and HT. In stage 1, seven tagSNPs were tested in 524 cases and 531 controls and the significant SNPs (P b 0.05) would enter into stage 2 including 807 cases and 869 controls. Furthermore, joint analysis was performed for all 2731 subjects including 1331 cases and 1400 controls, and meta-analysis was applied to evaluate combined estimations from the subgroups of stage 1 and stage 2. Results: In stage 1, rs204994 had significant association with HT (P b 0.05) and enter stage 2. Neither joint analysis nor meta-analysis found statistical association of rs204994 with HT after adjusted for the covariates in the whole population. However, further stratification analysis found that rs204994 was significantly associated with HT in b 50 years and ≥50 years groups, ORs (95%CI) of dominant model were 1.623 (1.054–2.500) and 0.721 (0.546–0.952) respectively. No significant correlation was found between blood pressure and the polymorphisms of rs204994. Conclusions: Our data suggests that age might modulate the genetic effects of variation of rs204994 in AGER on HT and further replications in other populations and functional studies should be warranted. Crown Copyright © 2012 Published by Elsevier B.V. All rights reserved.
Abbreviations: AGEs, advanced glycation end products; AGER, receptor for AGE; Ang, angiotensin; BMI, body mass index; χ2, Chi-square; CHB, Chinese population in Beijing; CAD, coronary artery disease; DBP, diastolic blood pressure; esRAGE, endogenous secretory RAGE; EDTA, ethylenediamine tetraacetic acid; GLM, general linear model; GLU, glucose; HWE, Hardy–Weinberg equilibrium; HDL-C, high density lipoprotein cholesterol; HMGB1, high-mobility group box-1; HT, hypertension; LD, linkage disequilibrium; MAF, minor allele frequency; NF-κB, nuclear factor kappa-B; OR, odds ratio; CI, 95% confidence; PBX2, interval pre-B-cell leukemia homeobox 2; PCR-RFLP, polymerase chain reaction and restriction fragment length polymorphism; P, probability; RAS, renin–angiotensin system; ROS, reactive oxygen species; SNPs, single nucleotide polymorphisms; SHRSP, spontaneously hypertensive rats; SPSS, Statistical Product and Service Solutions; SBP, systolic blood pressure; tagSNPs, tagger SNPs; TLR, toll-like receptor; TFBS, transcript factor banding site; TGR, transgenic rat; TG, triglycerides; TC, total cholesterol. ⁎ Correspondence to: Chong Shen, Department of Epidemiology and Biostatistics, School of Public Health, Nanjing Medical University, No. 140 Hanzhong Road, Nanjing 210029 People's Republic of China. Tel.: + 86 25 86862815; fax: + 86 25 86527613. ⁎⁎ Correspondence to: Yanping Zhao, Department of Neurology, Affiliated Yixing People's Hospital, Jiangsu University, Yixing 214200, People's Republic of China. Tel.: + 86 510 87921000; fax: + 86 510 7902967. E-mail addresses: [email protected] (C. Shen), [email protected] (Y. Zhao). 1 Song Yang and Hairu Wang equally contributed to this work.
1. Introduction Advanced glycation end products (AGEs) are new compounds or modified structures produced among non-enzymatic glycation and glycoxidation of proteins, lipids and nucleic acids in a prooxidant environment (Schmidt et al., 1994; Bierhaus et al., 1998). Series of reports indicate that AGEs induced glycoxidative reaction plays an important role in the progress of aging disorders (McEniery et al., 2007; Farmer and Kennedy, 2009; Hallam et al., 2010; Sawabe, 2010). Both oxidative and carbonyl stress can cause severe damage to biological structures of proteins, lipids and nucleic acids and further induce inflammatory response (Schmidt et al., 1994). Accumulation of glycoxidation products would upregulate transcription and expression of cytokines, growth factors and adhesive molecules via AGE– AGER (receptor for AGE) interaction and subsequently increase classical acute phase reactants (e.g. C-reactive protein or orosomucoid) (Stitt et al., 2000; Kalousova et al., 2005).
0378-1119/$ – see front matter. Crown Copyright © 2012 Published by Elsevier B.V. All rights reserved. doi:10.1016/j.gene.2012.01.080
312
S. Yang et al. / Gene 498 (2012) 311–316
Recent evidence proved that the interaction of AGE-ligation and AGER could induce immune and inflammation through increasing production of cellular reactive oxygen species (ROS) and persistent activation of the proinflammatory transcription factor, nuclear factor kappa-B (NF-κB) (Schmidt et al., 1994; Schmidt et al., 2001; Bierhaus et al., 2005; Yan et al., 2009; Yao and Brownlee, 2010). Also, study on transgenic rat (TGR) model with angiotensin (Ang) II-dependent hypertension (HT) and renal damage has proved the interrelation between renin–angiotensin system (RAS) and the AGE/ AGER axis in promoting cardiovascular end-organ damage (Bohlender et al., 2005; Koyama et al., 2007b; Yamagishi et al., 2008; Ramasamy et al., 2009). On the other hand, high-mobility group box-1 (HMGB1) is a newly recognized potent innate “danger signal” to the proinflammatory response via AGER and members of the toll-like receptor (TLR) family (van Zoelen et al., 2008). The interaction of HMGB1 and AGER plays a critical role in initiating and sustaining the inflammatory response in inflammatory cardiomyopathy eventually leading to heart failure and coronary artery disease (CAD) (Gao et al., 2010; Volz et al., 2010). Additionally, studies observed that blood pressure and oxidative damage were significantly reduced in stroke-prone spontaneously hypertensive rats (SHRSP) after long-term administration of AGEs inhibitor OPB-9195, and suggested that AGE–AGER pathway might play an important role in the progression or maintenance of genetic hypertension and related diseases (Mizutani et al., 2002; Steckelings et al., 2009). Evidences from population also showed that serum carboxymethyllysine was associated with increased aortic pulse wave velocity and arterial stiffness in adults (Mizutani et al., 2002; Semba et al., 2009). Furthermore, hypertensive patients with lower tissue AGE accumulation presented a larger improvement in diastolic function in response to anti-hypertensive eprosartan. Other drugs such as ARBs could modulate AGER isoform expressions by correcting endothelial dysfunction and were already used for hypertension or diabetes treatment (Koyama et al., 2007b; Grossin et al., 2010; Hartog et al., 2010; Maeda et al., 2011). Those findings would be helpful to illustrate microvascular, macrovascular disease and diabetic complications. Meaningfully, well understanding AGER mediated vascular pathogenesis would lead new insight to prognosis evaluation and efficient therapeutic strategies for cardiovascular disease. As a complex polygenic disease, human essential hypertension is affected by varying combinations of genetic and environmental factors (Staessen et al., 2003). Those comprehensive evidences suggested that AGER mediated macrovascular impairment might significantly affect the pathogenesis of hypertension, and so, whether the AGER gene polymorphisms contribute genetic susceptibility to HT would be a concern. In the present study, we designed a two-stage association study, which could provide near-optimal power to detect the true marker conferring disease risk while substantially reducing the total number of marker evaluations (Satagopan and Elston, 2003; Satagopan et al., 2004) to evaluate the association of AGER gene with HT in Han Chinese population. 2. Methods 2.1. Subjects In total, 2731 adult subjects aged 18–62 years were recruited from a rural population of 14,469 subjects in Jiangsu province by an epidemiological stratification sampling approach. In the present study, 1331 hypertensive cases with systolic blood pressure (SBP) ≥140 mm Hg, and/or diastolic blood pressure (DBP) ≥90 mm Hg, or currently administering anti-hypertensive medication were included and the patients with a clinical history of secondary hypertension,
coronary heart disease, kidney disease and diabetes were excluded. 1400 age- and sex-matched normotensives with SBP b140 mm Hg and DBP b90 mm Hg were selected as controls from the same target study population. We designed a two-stage association study and randomly selected 524 cases and 531 controls as stage 1 subsample to screen positive single nucleotide polymorphisms (SNPs). In stage 2, the remaining sample of 1676 subjects included 807 cases and 869 controls. We take initial associations found in stage 1 as hypotheses and further test in stage 2. The ethics committee of Nanjing Medical University has approved the research protocol and all subjects accepted written informed consent during epidemiological interviews. Trained research staff administered a standard questionnaire to obtain information on demographic characteristics including age, gender, nation, education, occupation and status of smoking and drinking. Three BP measurements were obtained from each participant by certified observers according to a standard protocol recommended by the American Heart Association (Perloff et al., 1993). Characteristics of study population of stage 1, stage 2 and joint main sample with stage 1 plus stage 2 were listed in Table 1. 2.2. SNP selection AGER gene (GenBank ID: 177), also known as RAGE, is mapped on chromosome 6p21.3. AGER spans 3.278 kbps (32,148.711– 32,151.988 kbps) and consists of 11 exons. We searched SNPs covered AGER and upstream to test and found pre-B-cell leukemia homeobox 2 (PBX2) gene (GeneID: 5089) at upstream 486 bps of AGER. Thus, we selected candidate SNPs covered AGER gene and upstream 6 kbps including PBX2 that spans 5.454 kbps (32,152.482–32,157.935 kbps) and consists of 9 exons. We selected tagger SNPs (tagSNPs) from the data of Han Chinese population in Beijing, China (CHB) in HapMap (HapMap Data Rel 24/phase II Nov08, on NCBI B36 assembly, dbSNP b126). All tagSNPs were selected with minor allele frequency (MAF) ≥0.05 and r 2 ≥ 0.8 according to the linkage disequilibrium (LD) values. We applied a functional candidate strategy to search functional SNPs with the role of transcription, regulating or splicing, etc. on the website of selection tool for single nucleotide polymorphisms (FASTSNP, http://fastsnp.ibms. sinica.edu.tw) (Yuan et al., 2006). SNPs with predictive biological effects would be enforced to rerun as tagSNPs by the software of Haploview. Thus, eight tagSNPs that were located in putative functional regions of the gene, e.g., exons, promoter/regulatory region, transcript factor banding site (TFBS) or splicing regulation were listed in Table 2. All selected SNPs were validated with 96 DNA samples in genotype test and rs2269422 with lower MAF (0.010) in controls was excluded. Finally, seven tagSNPs were included in the genotyping experiments. 2.3. Blood sampling and SNP genotyping All the subjects were informed to keep an empty stomach over 12 h prior to E blood sampling in the morning. Blood for genotyping was taken into ethylenediamine tetraacetic acid (EDTA)-containing receptacles. SNP genotyping used an improved polymerase chain reaction and restriction fragment length polymorphism (PCR-RFLP) method with pink dye as indicator and high throughout electrophoresis method with 96-sample agarose gel block. Blood pressure and history of hypertension or medication status were mixed blindly for composition of random samples. Ninety-six randomly selected samples were genotyped twice for duplication accuracy and then sequencing analyses for 16 randomly selected samples were done respectively using a Big-dye sequence kit (Applied Biosystem, ABI model 377 genetic analyzer, Foster City, USA).
S. Yang et al. / Gene 498 (2012) 311–316
313
Table 1 Comparison of clinical characteristics between cases and controls in stage 1, stage 2 and joint analysis. Characteristics
Male (%) Age (years) SBP (mm Hg) DBP (mm Hg) BMI (kg/m2) TC (mmol/L) HDL-C (mmol/L) TG (mmol/L) GLU (mmol/L) Smokers (%) Drinkers (%)
Stage 1
Stage 2
Combined sample of stage 1 plus stage 2
Cases (n = 524)
Controls (n = 531)
Cases (n = 807)
Controls (n = 869)
Cases (n = 1331)
Controls (n = 1400)
34.0 50.78 ± 6.05 146.70 ± 18.52⁎,⁎⁎ 90.45 ± 10.54⁎,⁎⁎ 26.45 ± 3.58⁎ 4.37 ± 0.99⁎ 1.49 ± 0.35⁎ 1.87 ± 1.22⁎ 4.72 ± 1.36⁎
34.8 50.53 ± 6.12 110.16 ± 9.7⁎⁎ 67.50 ± 7.32⁎⁎ 21.18 ± 2.04 3.95 ± 0.70 1.75 ± 0.38 0.89 ± 0.32 4.17 ± 0.53 20.0 25.6
33.6 51.49 ± 7.08 151.17 ± 17.56⁎ 92.28 ± 10.79⁎ 26.38 ± 3.78⁎ 4.27 ± 0.94⁎ 1.50 ± 0.38⁎ 1.82 ± 1.29⁎ 4.73 ± 1.40⁎
35.2 50.87 ± 6.94 111.54 ± 10.17 68.53 ± 7.46 21.44 ± 2.03 3.9 ± 0.69 1.72 ± 0.34 0.89 ± 0.45 4.25 ± 0.56 18.9 27.3
33.7 51.22 ± 6.55 149.41 ± 18.7⁎ 91.56 ± 10.73⁎ 26.41 ± 3.70⁎ 4.31 ± 0.96⁎ 1.49 ± 0.37⁎ 1.85 ± 1.27⁎ 4.73 ± 1.38⁎
35.1 50.89 ± 6.64 111.02 ± 10.01 68.15 ± 7.42 21.34 ± 2.04 3.92 ± 0.70 1.73 ± 0.34 0.89 ± 0.41 4.22 ± 0.55 19.1 26.6
17.9 28.2
15.4 26.3
16.4 27.0
⁎ P b 0.0001 for comparison with corresponding controls. ⁎⁎ P b 0.05 for comparison with corresponding variables in stage 2.
2.4. Statistical analysis Unpaired Student's t-test was used to test the differences of all measured variables that were presented as the means ± SD between cases and controls. The Chi-square (χ 2) test was used to compare qualitative variables and the allele and genotype frequency distributions between cases and controls and a P value of 0.05 was defined to be statistically significant. Multiple unconditional logistic regression was applied to adjust for covariates including age, gender, body mass index (BMI), glucose (GLU), triglycerides (TG), total cholesterol (TC), high density lipoprotein cholesterol (HDL-C), smoking and drinking status. The criterion for SNPs entering into stage 2 study was that the SNPs had a two-tailed probability (P) value of b0.05 for the comparisons of allele or genotype frequency distributions between cases and controls in stage 1. Hardy–Weinberg equilibrium (HWE) was assessed by Fisher's exact χ 2 test using the program HWE in the control groups. In stage 2, joint analysis and meta-analysis were performed in the main study population (Cantor et al., 2010). Meta-analysis takes advantage of the high accuracy of effect size and increase statistical power. Random effect model, which may well be conservative compared with a fixed effects model when the primary goal is hypothesis testing, was selected and inverse-variance method was employed to estimate combined adjusted odds ratio (OR) and its 95% confidence interval (CI) (de Bakker et al., 2008). The two-sided Z-test with the corresponding P-value was summed across two studies weighing them by the per study sample size to test the significance of single SNP in joint study. We assessed heterogeneity among study populations with the Cochran's Q-statistic and I 2 statistic that measures the proportion of total variance in estimated ORs due to heterogeneity. A value of the I 2 statistic greater than 50% indicates the high degree of heterogeneity in estimated ORs across study populations.
Because of AGEs induced glycoxidative reaction has significant age-specified aging effects, and age often serves as a surrogate for a variety of interacting covariates and which may not only enhance gene discovery (Shi et al., 2009), in the present study, but also age stratification analysis was further performed. We divided age into b50 years and ≥ 50 years groups for SBP and DBP of human over 50 years often increase exponentially with age (Safar et al., 2004). Additionally, general linear model (GLM) was applied to compare blood pressure levels between genotypes. Statistical analyses as above were performed with Statistical Product and Service Solutions 13.0 (SPSS; SPSS Inc, Chicago, USA) and the statistics of meta-analysis were calculated in the STATA software package (College Station, Texas, USA). 3. Results 3.1. Clinical characteristics The characteristics of subjects included in the stage 1, stage 2 and joint analysis were listed in Table 1. Age and the proportions of men, smoking and drinking status were not significantly different between cases and controls in stage 1, stage 2 and main sample. As expected, the cases generally had higher SBP and DBP, BMI, TC, TG, and GLU levels and lower HDL-C than that in controls. Both SBP and DBP in stage 1 were a little lower than that of stage 2 (P b 0.05). 3.2. Association analyses In stage 1, all SNPs in controls were in HWE except rs2022059 (P = 0.006). Out of seven candidate SNPs, SNP rs204994 presented statistical correlations with HT and P values of recessive model and
Table 2 Biological information analysis for candidate SNPs in AGER gene. TagSNPs
Region
Allelesa
TFBS/AA code
Biological effect
MAFb,c
rs1800625 rs1800624 rs2022059 (PBX2) rs2070600 rs204994 (PBX2)d rs204993 (PBX2)d rs184003 rs2269422
5upstream 5upstream Intron Coding 5′ upstream 5′ upstream Intron Intron
A/G A/T C/G C/T G/A C/T C/T C/T
E2F – – v-Myb AML-1a, SREBP-1 MZF1 GATA-1, GATA-1 CRE-BP, deltaE, C/EBPa
Promoter/regulatory region Upstream Intronic enhancer Missense (non-conservative) splicing regulation Promoter/regulatory region Promoter/regulatory region Intronic enhancer Intronic enhancer
0.132/0.122 0.156/0.163 0.195/0.250 0.225/0.289 0.254/0.200 0.456/0.478 0.173/0.122 0.010/0.057
a b c d
Major/minor allele. MAF in the control. MAF in CHB. Variations of rs204994 and 204993 predicted as intronic enhancer for PBX2 gene.
314
S. Yang et al. / Gene 498 (2012) 311–316
Table 3 Association analyses of rs204994 and HT. Stage
Stage 1 Stage 2 Joint analysis
Genotype and OR (95%CI)a
Group
Case Control Case Control Case Control
WT
Ht + MT
GG
GA + AA
321 305 504 552 825 857
178 + 25 183 + 42 275 + 28 288 + 29 453 + 53 471 + 71
Meta-analysis
Allele Dominant
Major/minor
OR (95%CI)
G/A 0.857 (0.670–1.096) P = 0.220 1.047 (0.858–1.277) P = 0.651 0.935 (0.743–1.176) P = 0.564 0.915 (0.726–1.154) P = 0.455
0.782/0.218 0.795/0.295 0.790/0.210
0.818 (0.668–1.000) P = 0.049 1.038 (0.877–1.229) P = 0.665 0.947 (0.833–1.078) P = 0.413 0.899 (0.784–1.031) P = 0.129
WT wide type, Ht heterozygote, MT mutant type. a ORs were adjusted for age, sex, BMI, TC, TG, HDL-C, GLU, smoking and drinking.
allele were 0.020 and 0.049 respectively (Table 3). Thus, SNP rs204994 entered into stage 2 and further genotyped. The genotype and allele distributions of the seven SNPs between cases and controls were seen in Table S1 in supporting information. In stage 2, the association of rs204994 with HT wasn't replicated even after adjusted for the covariates, including age, sex, BMI, GLU, TC, TG, HDL-C, smoking and drinking status by logistic regression analysis. Meta-analysis with genetic variance of random effect model in the two stage samples presented well homogeneity (I 2 = 0 for additive, dominant and recessive models and I 2 = 30.4% for allele), and the results indicated that there was no statistical association of genotype or allele with HT (P > 0.05). 3.3. Stratification analysis Further age stratification analysis was conducted for the association of rs204994 and HT. In b50 years age group, A allele variation presented significant risk effect for HT and OR (95%CI) of dominant model of rs204994 and was 1.623 (1.054–2.500) and the association still presented statistical significance after adjusted for sex, TC, TG, HDL-C, GLU, BMI smoking and drinking. On the contrary, in ≥50 years age group, the genetic variation showed significant protective effects for HT and ORs (95%CI) of additive and dominant models and were 0.747 (0.592–0.943) and 0.721 (0.546–0.952) respectively, and also the association still presented statistical significance after adjusted for covariates. Table 4 listed the adjusted ORs and P values. However, there was no statistical correlation found in stratification analyses by sex, smoking and drinking even after adjusted for covariates (data were not listed here). 3.4. Analysis of quantitative trait for blood pressure Blood pressure levels were compared between GG, GA and AA genotypes of rs204994 and the results showed that no significant differences of SBP and DBP were found between the different genotypes either in subjects with hypertension history (n = 845) or in subjects without hypertension history (n = 1886) (Table 5). Additionally, no
significant interaction of age and gene was found for SBP or DBP (P > 0.1). 4. Discussion Many studies have reported that AGEs play important roles in aging disorders such as diabetes and diabetic cardiovascular complication via interaction of AGEs and cellular AGER (Bierhaus et al., 1998; Stitt et al., 2000; Koyama et al., 2007a; Yamagishi et al., 2008; Nakamura et al., 2009; Steckelings et al., 2009; Hallam et al., 2010). AGER is also profoundly associated with macrovascular complications through regulation of atherogenesis, angiogenic response, vascular injury, and inflammatory response (Koyama et al., 2007b) and nifedipine might provide an anti-inflammatory effect against AGEs in tubular cells by suppressing RAGE expression via PPARgamma activation (Matsui et al., 2010). Likewise, plasma endogenous secretory RAGE (esRAGE) may act as a protective factor against and a novel biomarker for the occurrence of metabolic syndrome and cardiovascular diseases (Koyama et al., 2007a; Nakamura et al., 2009). Therefore, the contribution of AGER to the vascular pathogenesis in human hypertension would deserve further investigation and that would advance our knowledge of the determinants of atherogenesis, angiogenic response, vascular injury and inflammatory response and disease. Recently, a few studies reported that AGER genetic polymorphism had significant association with diabetic CAD, CAD and stroke (dos Santos et al., 2005; Falcone et al., 2008; Gao et al., 2010), but less study investigates the genetic effect of AGER on HT except a relatively small sample study without correlative findings. In the present study, we conducted a two-stage case–control study and selected seven functional tagSNPs to assess the association of AGER gene with HT in Han Chinese population. The power was calculated with the software of Power and Sample Size Calculation (Dupont WD, Plummer WD: http://biostat.mc.vanderbilt.edu/twiki/ bin/ view/Main/PowerSampleSize). At the 5% significance level, we had 84.1% power to detect a MAF of 0.05 that confers OR of 1.4. Actually, rs1800625 has lower MAF 0.132 in controls and the power for OR equal to 1.3 is 92.8%. Furthermore, we use a meta-analysis, which due to increased power and validation may provide more
Table 4 Stratification analysis of rs204994 with hypertension by age. Stratification factor
Age b50 y ≥ 50 y a
Group
Case Control Case Control
Genotype and OR (95%CI)a WT
Ht + MT
GG
GA + AA
288 359 537 498
173 + 24 171 + 24 280 + 29 298 + 49
Adjusted for sex, BMI, TC, TG, HDL-C, GLU, smoking and drinking.
Additive
Dominant
Recessive
1.430 (0.993–2.059) P = 0.055 0.769 (0.611–0.968) P = 0.025
1.654 (1.071–2.554) P = 0.023 0.751 (0.571–0.989) P = 0.042
0.990 (0.349–2.806) P = 0.984 0.560 (0.295–1.065) P = 0.077
S. Yang et al. / Gene 498 (2012) 311–316 Table 5 Comparisons of quantitative traits of blood pressure between genotypes of rs204994 by treatment stratification. Group
Treatment (n) SBP (mm Hg) DBP (mm Hg) Non-treatment (n) SBP (mm Hg) DBP (mm Hg)
Genotype GG
GA
AA
525 146.97 ± 19.12 89.96 ± 10.97 1157 122.44 ± 22.29 75.17 ± 14.36
289 144.71 ± 19.38 89.09 ± 10.98 633 122.72 ± 22.88 74.76 ± 13.98
31 138.85 ± 18.35 89.24 ± 11.95 95 121.76 ± 21.08 75.37 ± 12.93
P valuea
0.118 0.712 0.677 0.867
P > 0.05 for all test of homogeneity of variances. a Adjusted for sex, BMI, TC, TG, HDL-C, GLU, smoking and drinking.
robust risk estimates and its 95%CI from direct joint analysis (de Bakker et al., 2008; Cantor et al., 2010), to assess heterogeneity among study populations and results indicated that genetic variance of random effect model in the two stage samples presented well homogeneity as described above. On the other hand, the ORs were adjusted for potential confounders such as age, gender, BMI, GLU, TC, TG, HDL-C, drinking and smoking status and that would effectively reduce the false positive findings in the present study. Some limitations might cause false positive association in the present study. Firstly, all the subjects came from a cross sectional population but not a follow-up cohort population. Secondly, false positive association would present in stage 1 if potential differentiation for covariates or population structure exists in the two stage samples. Thirdly, genetic heterozygosity that arose from the inner research population might devote different susceptibilities to HT in subgroups, and phenocopy for other disease often falsified results in association study. Finally, potential bias including information bias, selective bias and confounding bias often distort results for epidemiological association studies. Regardless of the limitation described as above, the present study identified that rs204994 of AGER had a positive association with HT in subgroup with b50 years for the first time. As a complex biological construction, age may serve as a surrogate for a variety of interacting covariates and render findings from across studies more comparable to each other (McEniery et al., 2007; Shi et al., 2009). Recently, several genes were reported to be involved in blood pressure regulation and the susceptibility to hypertension modulated by age (Dao et al., 2005; Lunetta et al., 2007; Shen et al., 2009; Shi et al., 2009). Furthermore, the statistical analysis strategy of age-genetic effects can prevent potential confounding bias modified by age-varying characteristics (Lunetta et al., 2007; Lasky-Su et al., 2008; Percy et al., 2009). According to the findings that SBP and DBP increase exponentially with age over 50 years (Safar et al., 2004), we divided age into b50 and ≥50 years groups to distinguish whether age might modulate the genetic effect of AGER on HT by stratification analysis (Dao et al., 2005; Lasky-Su et al., 2008). In the present study, the results indicated that rs204994 had positive association with HT in b50 years group but negative association in ≥50 years group and after adjusted for covariates respectively whereas no significant correlation was found between blood pressure and genetic polymorphism. Our results support that age might play an important modulating role in AGER genetic effect on the susceptibility to HT. Meanwhile, age's modification would blanket over real association for HT if just meta-analysis proposed for combined samples as above. Remarkably, rs204994 (G/A) is located at 5′ upstream promoter/ regulatory region of AGER and in the overlapping intron 6 of PBX2. Variation of G to A may predict buildups of TFBS of AML-1a and SREBP-1. Therefore, the positive association modulated by age found in the present study would be helpful to further research potential gene function to pathogenesis of vascular lesions. Meanwhile, it is desirable to verify the SNP's biological function contributing to AGER or PBX2 which was seldom reported to associate with vascular
315
disorders. Hopefully, our findings might facilitate further research for the benefits of using surrogate markers for pharmacogenetics of antihypertensive and prognosis evaluation. In conclusion, our finding suggests that age might modulate the genetic effects of AGER gene on the susceptibility to HT but not blood pressure trait. Further replication for association study or biological functional study would be warranted in the future. Supplementary materials related to this article can be found online at doi:10.1016/j.gene.2012.01.080. Acknowledgment This work was supported by the National Natural Science Foundation of China (Grant No. 30800947 and Grant No. 81072367), Health Research Program of Jiangsu Province (Grant No. H200839) and Natural Science Foundation of Jiangsu Province (Grant No. BK2011776). References Bierhaus, A., Hofmann, M.A., Ziegler, R., Nawroth, P.P., 1998. AGEs and their interaction with AGE-receptors in vascular disease and diabetes mellitus. I. The AGE concept. Cardiovasc. Res. 37, 586–600. Bierhaus, A., Humpert, P.M., Morcos, M., Wendt, T., Chavakis, T., Arnold, B., Stern, D.M., Nawroth, P.P., 2005. Understanding RAGE, the receptor for advanced glycation end products. J. Mol. Med. 83, 876–886. Bohlender, J., Franke, S., Sommer, M., Stein, G., 2005. Advanced glycation end products: a possible link to angiotensin in an animal model. Ann. N. Y. Acad. Sci. 1043, 681–684. Cantor, R.M., Lange, K., Sinsheimer, J.S., 2010. Prioritizing GWAS results: a review of statistical methods and recommendations for their application. Am. J. Hum. Genet. 86, 6–22. Dao, H.H., Essalihi, R., Bouvet, C., Moreau, P., 2005. Evolution and modulation of agerelated medial elastocalcinosis: impact on large artery stiffness and isolated systolic hypertension. Cardiovasc. Res. 66, 307–317. de Bakker, P.I., Ferreira, M.A., Jia, X., Neale, B.M., Raychaudhuri, S., Voight, B.F., 2008. Practical aspects of imputation-driven meta-analysis of genome-wide association studies. Hum. Mol. Genet. 17, R122–R128. dos Santos, K.G., Canani, L.H., Gross, J.L., Tschiedel, B., Pires Souto, K.E., Roisenberg, I., 2005. The –374A allele of the receptor for advanced glycation end products gene is associated with a decreased risk of ischemic heart disease in African-Brazilians with type 2 diabetes. Mol. Genet. Metab. 85, 149–156. Falcone, C., Geroldi, D., Buzzi, M.P., Emanuele, E., Yilmaz, Y., Fontana, J.M., Vignali, L., Boiocchi, C., Sbarsi, I., Cuccia, M., 2008. The –374T/A RAGE polymorphism protects against future cardiac events in nondiabetic patients with coronary artery disease. Arch. Med. Res. 39, 320–325. Farmer, D.G., Kennedy, S., 2009. RAGE, vascular tone and vascular disease. Pharmacol. Ther. 124, 185–194. Gao, J., Shao, Y., Lai, W., Ren, H., Xu, D., 2010. Association of polymorphisms in the RAGE gene with serum CRP levels and coronary artery disease in the Chinese Han population. J. Hum. Genet. 55, 668–675. Grossin, N., Boulanger, E., Wautier, M.P., Wautier, J.L., 2010. The different isoforms of the receptor for advanced glycation end products are modulated by pharmacological agents. Clin. Hemorheol. Microcirc. 45, 143–153. Hallam, K.M., Li, Q., Ananthakrishnan, R., Kalea, A., Zou, Y.S., Vedantham, S., Schmidt, A.M., Yan, S.F., Ramasamy, R., 2010. Aldose reductase and AGE–RAGE pathways: central roles in the pathogenesis of vascular dysfunction in aging rats. Aging Cell 9, 776–684. Hartog, J.W., van de Wal, R.M., Schalkwijk, C.G., Miyata, T., Jaarsma, W., Plokker, H.W., van Wijk, L.M., Smit, A.J., van Veldhuisen, D.J., Voors, A.A., 2010. Advanced glycation end-products, anti-hypertensive treatment and diastolic function in patients with hypertension and diastolic dysfunction. Eur. J. Heart Fail. 12, 397–403. Kalousova, M., Zima, T., Tesar, V., Dusilova-Sulkova, S., Skrha, J., 2005. Advanced glycoxidation end products in chronic diseases—clinical chemistry and genetic background. Mutat. Res. 579, 37–46. Koyama, H., Yamamoto, H., Nishizawa, Y., 2007a. Endogenous secretory RAGE as a novel biomarker for metabolic syndrome and cardiovascular diseases. Biomark. Insight 2, 331–339. Koyama, H., Yamamoto, H., Nishizawa, Y., 2007b. RAGE and soluble RAGE: potential therapeutic targets for cardiovascular diseases. Mol. Med. 13, 625–635. Lasky-Su, J., Lyon, H.N., Emilsson, V., Heid, I.M., Molony, C., Raby, B.A., Lazarus, R., Klanderman, B., Soto-Quiros, M.E., Avila, L., Silverman, E.K., Thorleifsson, G., Thorsteinsdottir, U., Kronenberg, F., Vollmert, C., Illig, T., Fox, C.S., Levy, D., Laird, N., Ding, X., McQueen, M.B., Butler, J., Ardlie, K., Papoutsakis, C., Dedoussis, G., O'Donnell, C.J., Wichmann, H.E., Celedon, J.C., Schadt, E., Hirschhorn, J., Weiss, S.T., Stefansson, K., Lange, C., 2008. On the replication of genetic associations: timing can be everything! Am. J. Hum. Genet. 82, 849–858. Lunetta, K.L., D'Agostino Sr., R.B., Karasik, D., Benjamin, E.J., Guo, C.Y., Govindaraju, R., Kiel, D.P., Kelly-Hayes, M., Massaro, J.M., Pencina, M.J., Seshadri, S., Murabito, J.M., 2007. Genetic correlates of longevity and selected age-related phenotypes: a genome-wide association study in the Framingham Study. BMC Med. Genet. 8 (Suppl 1), S13.
316
S. Yang et al. / Gene 498 (2012) 311–316
Maeda, S., Matsui, T., Takeuchi, M., Yamagishi, S., 2011. Co-treatment with azelinidipine and olmesartan inhibits advanced glycation end products (AGEs) elicited down-regulation of adiponectin mRNA levels in cultured adipocytes partly via its anti-oxidative property. Int. J. Cardiol. 146, 264–266. Matsui, T., Yamagishi, S., Takeuchi, M., Ueda, S., Fukami, K., Okuda, S., 2010. Nifedipine inhibits advanced glycation end products (AGEs) and their receptor (RAGE) interaction-mediated proximal tubular cell injury via peroxisome proliferatoractivated receptor-gamma activation. Biochem. Biophys. Res. Commun. 398, 326–330. McEniery, C.M., Wilkinson, I.B., Avolio, A.P., 2007. Age, hypertension and arterial function. Clin. Exp. Pharmacol. Physiol. 34, 665–671. Mizutani, K., Ikeda, K., Tsuda, K., Yamori, Y., 2002. Inhibitor for advanced glycation end products formation attenuates hypertension and oxidative damage in genetic hypertensive rats. J. Hypertens. 20, 1607–1614. Nakamura, K., Adachi, H., Matsui, T., Kurita, Y., Takeuchi, M., Yamagishi, S., 2009. Independent determinants of soluble form of receptor for advanced glycation end products in elderly hypertensive patients. Metabolism 58, 421–425. Percy, C.J., Brown, L., Power, D.A., Johnson, D.W., Gobe, G.C., 2009. Obesity and hypertension have differing oxidant handling molecular pathways in age-related chronic kidney disease. Mech. Ageing Dev. 130, 129–138. Perloff, D., Grim, C., Flack, J., Frohlich, E.D., Hill, M., McDonald, M., Morgenstern, B.Z., 1993. Human blood pressure determination by sphygmomanometry. Circulation 88, 2460–2470. Ramasamy, R., Yan, S.F., Schmidt, A.M., 2009. RAGE: therapeutic target and biomarker of the inflammatory response—the evidence mounts. J. Leukoc. Biol. 86, 505–512. Safar, M.E., Lajemi, M., Rudnichi, A., Asmar, R., Benetos, A., 2004. Angiotensinconverting enzyme D/I gene polymorphism and age-related changes in pulse pressure in subjects with hypertension. Arterioscler. Thromb. Vasc. Biol. 24, 782–786. Satagopan, J.M., Elston, R.C., 2003. Optimal two-stage genotyping in population-based association studies. Genet. Epidemiol. 25, 149–157. Satagopan, J.M., Venkatraman, E.S., Begg, C.B., 2004. Two-stage designs for genedisease association studies with sample size constraints. Biometrics 60, 589–597. Sawabe, M., 2010. Vascular aging: from molecular mechanism to clinical significance. Geriatr. Gerontol. Int. 10 (Suppl 1), S213–S220. Schmidt, A.M., Hori, O., Brett, J., Yan, S.D., Wautier, J.L., Stern, D., 1994. Cellular receptors for advanced glycation end products. Implications for induction of oxidant stress and cellular dysfunction in the pathogenesis of vascular lesions. Arterioscler. Thromb. 14, 1521–1528.
Schmidt, A.M., Yan, S.D., Yan, S.F., Stern, D.M., 2001. The multiligand receptor RAGE as a progression factor amplifying immune and inflammatory responses. J. Clin. Invest. 108, 949–955. Semba, R.D., Najjar, S.S., Sun, K., Lakatta, E.G., Ferrucci, L., 2009. Serum carboxymethyllysine, an advanced glycation end product, is associated with increased aortic pulse wave velocity in adults. Am. J. Hypertens. 22, 74–79. Shen, C., Lu, X., Li, Y., Zhao, Q., Liu, X., Hou, L., Wang, L., Chen, S., Huang, J., Gu, D., 2009. Emilin1 gene and essential hypertension: a two-stage association study in northern Han Chinese population. BMC Med. Genet. 10, 118. Shi, G., Gu, C.C., Kraja, A.T., Arnett, D.K., Myers, R.H., Pankow, J.S., Hunt, S.C., Rao, D.C., 2009. Genetic effect on blood pressure is modulated by age: the Hypertension Genetic Epidemiology Network Study. Hypertension 53, 35–41. Staessen, J.A., Wang, J., Bianchi, G., Birkenhager, W.H., 2003. Essential hypertension. Lancet 361, 1629–1641. Steckelings, U.M., Rompe, F., Kaschina, E., Unger, T., 2009. The evolving story of the RAAS in hypertension, diabetes and CV disease: moving from macrovascular to microvascular targets. Fundam. Clin. Pharmacol. 23, 693–703. Stitt, A.W., Burke, G.A., Chen, F., McMullen, C.B., Vlassara, H., 2000. Advanced glycation end-product receptor interactions on microvascular cells occur within caveolinrich membrane domains. FASEB J. 14, 2390–2392. van Zoelen, M.A., Yang, H., Florquin, S., Meijers, J.C., Akira, S., Arnold, B., Nawroth, P.P., Bierhaus, A., Tracey, K.J., van der Poll, T., 2008. Role of Toll-like receptors 2 and 4, and the receptor for advanced glycation end products (RAGE) in Hmgb1 induced inflammation in vivo. Shock 31, 280–284. Volz, H.C., Kaya, Z., Katus, H.A., Andrassy, M., 2010. The role of HMGB1/RAGE in inflammatory cardiomyopathy. Semin. Thromb. Hemost. 36, 185–194. Yamagishi, S., Nakamura, K., Matsui, T., Ueda, S., Fukami, K., Okuda, S., 2008. Agents that block advanced glycation end product (AGE)–RAGE (receptor for AGEs)-oxidative stress system: a novel therapeutic strategy for diabetic vascular complications. Expert Opin. Investig. Drugs 17, 983–996. Yan, S.F., Ramasamy, R., Schmidt, A.M., 2009. Receptor for AGE (RAGE) and its ligands— cast into leading roles in diabetes and the inflammatory response. J. Mol. Med. 87, 235–247. Yao, D., Brownlee, M., 2010. Hyperglycemia-induced reactive oxygen species increase expression of the receptor for advanced glycation end products (RAGE) and RAGE ligands. Diabetes 59, 249–255. Yuan, H.Y., Chiou, J.J., Tseng, W.H., Liu, C.H., Liu, C.K., Lin, Y.J., Wang, H.H., Yao, A., Chen, Y.T., Hsu, C.N., 2006. FASTSNP: an always up-to-date and extendable service for SNP function analysis and prioritization. Nucleic Acids Res. 34, W635–W641.