Impact of Common Type 2 Diabetes Risk Polymorphisms in the DESIR Prospective Study Martine Vaxillaire, 1 Jacques Veslot, 1 Christian Dina, 1 Christine Proenc ¸a, 1 Ste ´ phane Cauchi, 1 Guillaume Charpentier, 2 Jean Tichet, 3 Fre ´de ´ ric Fumeron, 4,5 Michel Marre, 4,5,6 David Meyre, 1 Beverley Balkau, 7,8 and Philippe Froguel, 1,9 for the DESIR Study Group OBJECTIVE—The emerging picture of type 2 diabetes genetics involves differently assembled gene variants, each modestly increasing risk with environmental exposure. However, the rel- evance of these genes for disease prediction has not been extensively tested. RESEARCH DESIGN AND METHODS—We analyzed 19 com- mon polymorphisms of 14 known candidate genes for their contribution to prevalence and incidence of glucose intolerance in the DESIR (Data from an Epidemiological Study on the Insulin Resistance syndrome) prospective study of middle-aged Cauca- sian subjects, including 3,877 participants (16.8% with hypergly- cemia and 7.9% with diabetes after the 9-year study). RESULTS—The GCK (Glucokinase) -30A allele was associated with increased type 2 diabetes risk at the end of the follow-up study (adjusted OR 1.34 [95% CI 1.07–1.69]) under an additive model, as supported in independent French diabetic case sub- jects (OR 1.22, P = 0.007), with increased fasting glycemia (0.85% per A allele, P = 6  10 -5 ) and decreased homeostasis model assessment of -cell function (4%, P = 0.0009). IL6 (Interleukin- 6) -174 G/C interacts with age in disease risk and modulates fasting glycemia according to age (1.36% decrease over 56 years, P = 5  10 -5 ). These polymorphisms together with KCNJ11 (Kir6.2)-E23K and TCF7L2-rs7903146 may predict diabetes inci- dence in the DESIR cohort. Each additional risk allele at GCK, TCF7L2, and IL6 increased risk by 1.34 (P = 2  10 -6 ), with an OR of 2.48 (95% CI 1.59 –3.86), in carriers of at least four at-risk alleles compared with those with none or one risk allele. CONCLUSIONS—Our data confirm several at-risk polymor- phisms for type 2 diabetes in a general population and demonstrate that prospective studies are valuable designs to complement clas- sical genetic approaches. Diabetes 57:244–254, 2008 T ype 2 diabetes is a complex trait where common genetic variants having modest individual effects act together and interact with environmental factors to modulate the risk of the disease (1,2). Initial efforts to identify type 2 diabetes susceptibility genes favored the genome-wide linkage approach and candidate gene association studies (3). A few common polymorphisms have been widely replicated in popula- tions of different ethnic descents (rev. in 4), including: PPARG-P12A (5), the E23K single nucleotide polymor- phism (SNP) in KCNJ11 coding for Kir6.2 (6,7), two intronic polymorphisms in the protease calpain-10 (CAPN10, SNP-43 and -44) (8), and more recently TCF7L2 variants, which have the most important risk effect (9,10). An association of the G-30A polymorphism in the -cell– specific promoter of glucokinase (GCK) was reported with elevated fasting glycemia and birth weight (11), gesta- tional diabetes (12), impaired glucose regulation (13), and a higher prevalence of type 2 diabetes in patients with coronary artery disease (14). Common variation in adi- ponectin/ADIPOQ is associated with insulin sensitivity, type 2 diabetes, and vascular complications of obesity (15). Other variants in candidate genes for type 2 diabetes that showed association in at least two independent studies include the G482S coding SNP of PPARGC1A (encoding PGC-1) (16,17), SNPs at the distal P2 promoter of HNF4A (18 –20), UCP2 (uncoupling protein-2) G-866A promoter SNP (21,22), and variants in the transcription factor SREBF1, which associate with obesity and type 2 diabetes in French and Austrian populations (23,24), and in ADIPOR2 (encoding adiponectin receptor 2) (25,26). Inflammatory processes also play a pivotal role in meta- bolic diseases, and high plasma interleukin-6 levels were associated with increased type 2 diabetes risk (27,28). A recent well-powered joint analysis of the IL6 (G-174C) promoter variant, showing a weaker effect on IL6 tran- scription, reported a decreased risk for type 2 diabetes (29). Association with type 2 diabetes was also found for IL6R-D358A in Danish whites (30) and TNF G-308A pro- moter SNP in the Finnish Diabetes Prevention Study (28). The advent of genome-wide association studies, survey- ing 75% of common variation across the human genome, promises to greatly speed up the identification of novel and replicated susceptibility genes (31–35). For future application it will be important to quantify the contribu- tion of these risk alleles to overall diabetes risk. Case- control studies investigate one primary outcome, the disease by which cases are defined, whereas complex diseases require examining genetic and lifestyle risk fac- tors and intermediary phenotypes related to the disease From 1 UMR8090 and Institute of Biology, Lille 2 University, CNRS and Pasteur Institute, Lille, France; the 2 Department of Endocrinology-Diabetology, Centre Hospitalier Sud-Francilien, Corbeil-Essonnes, France; the 3 Regional Institute for Health, Tours, France; 4 INSERM U695, Paris, France; 5 Universite ´ Paris Diderot, Paris, France; the 6 Department of Endocrinology-Diabetology and Nutrition, Bichat Claude Bernard Hospital, Paris, France; 7 INSERM U780- IFR69, Villejuif, France; the 8 University of Paris-Sud, Villejuif, France; and the 9 Department of Genomic Medicine, Hammersmith Hospital, Imperial College London, London, U.K. Address correspondence and reprint requests to Dr. Martine Vaxillaire, CNRS UMR8090, Institut Pasteur de Lille, 1 rue du Professeur Calmette, BP 245, 59019 Lille, France. E-mail: martine.vaxillaire@good.ibl.fr. Received for publication 28 June 2007 and accepted in revised form 15 October 2007. Published ahead of print at http://diabetes.diabetesjournals.org on 31 Octo- ber 2007. DOI: 10.2337/db07-0615. AUC, area under the curve; DESIR, Data from an Epidemiological Study on the Insulin Resistance syndrome; FPG, fasting plasma glucose; HOMA, ho- meostasis model assessment; HOMA-B, HOMA of -cell function; HOMA-IR, HOMA of insulin resistance; IFG, impaired fasting glycemia; IL, interleukin; MAF, minor allele frequency; ROC, receiver operating characteristic; SNP, single nucleotide polymorphism Additional information for this article can be found in an online appendix at http://dx.doi.org/10.2337/db07-0615. © 2008 by the American Diabetes Association. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. ORIGINAL ARTICLE 244 DIABETES, VOL. 57, JANUARY 2008