Wallace Chris, Newhouse Stephen J, Braund Peter, Zhang Feng, Tobin Martin, Falchi Mario, Ahmadi Kourosh, Dobson Richard J, Marçano Ana Carolina B, Hajat Cother, Burton Paul, Deloukas Panagiotis, Brown Morris, Connell John M, Dominiczak Anna, Lathrop G Mark, Webster John, Farrall Martin, Spector Tim, Samani Nilesh J, Caulfield Mark J, Munroe Patricia B
Clinical Pharmacology and The Genome Centre, Barts and The London, Queen Mary's School of Medicine and Dentistry, London, EC1M 6BQ, UK.
Am J Hum Genet. 2008 Jan;82(1):139-49. doi: 10.1016/j.ajhg.2007.11.001.
Many common diseases are accompanied by disturbances in biochemical traits. Identifying the genetic determinants could provide novel insights into disease mechanisms and reveal avenues for developing new therapies. Here, we report a genome-wide association analysis for commonly measured serum and urine biochemical traits. As part of the WTCCC, 500,000 SNPs genome wide were genotyped in 1955 hypertensive individuals characterized for 25 serum and urine biochemical traits. For each trait, we assessed association with individual SNPs, adjusting for age, sex, and BMI. Lipid measurements were further examined in a meta-analysis of genome-wide data from a type 2 diabetes scan. The most promising associations were examined in two epidemiological cohorts. We discovered association between serum urate and SLC2A9, a glucose transporter (p = 2 x 10(-15)) and confirmed this in two independent cohorts, GRAPHIC study (p = 9 x 10(-15)) and TwinsUK (p = 8 x 10(-19)). The odds ratio for hyperuricaemia (defined as urate >0.4 mMol/l) is 1.89 (95% CI = 1.36-2.61) per copy of common allele. We also replicated many genes previously associated with serum lipids and found previously recognized association between LDL levels and SNPs close to genes encoding PSRC1 and CELSR2 (p = 1 x 10(-7)). The common allele was associated with a 6% increase in nonfasting serum LDL. This region showed increased association in the meta-analysis (p = 4 x 10(-14)). This finding provides a potential biological mechanism for the recent association of this same allele of the same SNP with increased risk of coronary disease.
许多常见疾病都伴有生化特征紊乱。确定其遗传决定因素可为疾病机制提供新见解，并揭示开发新疗法的途径。在此，我们报告了一项针对常见血清和尿液生化特征的全基因组关联分析。作为WTCCC的一部分，对1955名高血压患者的50万个全基因组单核苷酸多态性（SNP）进行了基因分型，这些患者具有25种血清和尿液生化特征。对于每种特征，我们评估了与单个SNP的关联，并对年龄、性别和体重指数进行了校正。在一项2型糖尿病扫描的全基因组数据荟萃分析中，对脂质测量结果进行了进一步研究。在两个流行病学队列中对最有前景的关联进行了检测。我们发现血清尿酸与葡萄糖转运蛋白SLC2A9之间存在关联（p = 2×10⁻¹⁵），并在两个独立队列GRAPHIC研究（p = 9×10⁻¹⁵）和TwinsUK（p = 8×10⁻¹⁹）中得到了证实。高尿酸血症（定义为尿酸>0.4 mmol/l）的优势比为每一个常见等位基因拷贝1.89（95%置信区间= 1.36 - 2.61）。我们还重复验证了许多先前与血清脂质相关的基因，并发现了先前公认的低密度脂蛋白（LDL）水平与编码PSRC1和CELSR2基因附近的SNP之间的关联（p = 1×10⁻⁷）。常见等位基因与非空腹血清LDL升高6%相关。该区域在荟萃分析中显示出更强的关联性（p = 4×10⁻¹⁴）。这一发现为同一SNP的相同等位基因近期与冠心病风险增加之间的关联提供了潜在的生物学机制。
