Winckler Wendy, Graham Robert R, de Bakker Paul I W, Sun Maria, Almgren Peter, Tuomi Tiinamaija, Gaudet Daniel, Hudson Thomas J, Ardlie Kristin G, Daly Mark J, Hirschhorn Joel N, Groop Leif, Altshuler David
Department of Molecular Biology, Massachusetts General Hospital, Boston, MA 02114, USA.
Diabetes. 2005 Mar;54(3):886-92. doi: 10.2337/diabetes.54.3.886.
Two recent publications reported association of common polymorphisms in the P2 promoter of hepatocyte nuclear factor 4alpha (HNF4alpha) (the MODY1 gene) with risk for type 2 diabetes. We attempted to reproduce this putative association by genotyping 11 single nucleotide polymorphism (SNPs) spanning the HNF4alpha coding region and the P2 promoter in >3,400 patients and control subjects from Sweden, Finland, and Canada. One SNP that was consistently associated in the two previous reports (rs1884613, in the P2 promoter region) also trended in the same direction in our sample, albeit with a lower estimated odds ratio (OR) of 1.11 (P = 0.05, one-tailed). We genotyped this SNP (rs1884613) in an additional 4,400 subjects from North America and Poland. In this sample, the association was not confirmed and trended in the opposite direction (OR 0.88). Meta-analysis of our combined sample of 7,883 people (three times larger than the two initial reports combined) yielded an OR of 0.97 (P = 0.27). Finally, we provide an updated analysis of haplotype structure in the region to guide any further investigation of common variation in HNF4alpha. Although our combined results fail to replicate the previously reported association of common variants in HNF4alpha with risk for type 2 diabetes, we cannot exclude an effect smaller than that originally proposed, heterogeneity among samples, variation in as-yet-unmeasured genotypic or environmental modifiers, or true association secondary to linkage disequilibrium (LD) with as-yet-undiscovered variant(s) in the region.
最近的两篇出版物报道了肝细胞细胞核因子4α（HNF4α）（即MODY1基因）P2启动子中的常见多态性与2型糖尿病风险之间的关联。我们试图通过对来自瑞典、芬兰和加拿大的3400多名患者及对照者进行基因分型，以再现这种假定的关联，这些患者及对照者的样本涵盖了HNF4α编码区和P2启动子的11个单核苷酸多态性（SNP）。在前两篇报告中始终相关的一个SNP（rs1884613，位于P2启动子区域）在我们的样本中也呈相同趋势，尽管估计的优势比（OR）较低，为1.11（P = 0.05，单尾）。我们对另外来自北美和波兰的4400名受试者进行了该SNP（rs1884613）的基因分型。在这个样本中，该关联未得到证实，且呈相反趋势（OR 0.88）。对我们7883人的合并样本（比最初两篇报告的样本总和大三倍）进行荟萃分析，得出OR为0.97（P = 0.27）。最后，我们提供了该区域单倍型结构的最新分析，以指导对HNF4α常见变异的任何进一步研究。尽管我们的综合结果未能重现先前报道的HNF4α常见变异与2型糖尿病风险之间的关联，但我们不能排除比最初提出的效应更小的影响、样本间的异质性、尚未测量的基因型或环境修饰因子的变异，或与该区域尚未发现的变异存在连锁不平衡（LD）导致的真正关联。
