1Santiago Rodriguez, 1Tom R. Gaunt, 1Xiao-he Chen, 1Shuwen Huang, 2Julie A. Gilg, 2Derek G. Cook, 2Peter H. Whincup, 3George J. Miller, 4Holly E. Syddall, 4Elaine Dennison, 4David I.W. Phillips, 4Cyris Cooper, 1Ian N.M. Day
1Human Genetics Division, University of Southampton, School of Medicine, Duthie Buuilding (MP808), Southampton General Hospital, Tremona Road, Southampton SO16 6YD, UK, 2Department of Community Health Sciences, St George's Hospital Medical School, Cranmer Terrace, London SW17 0RE, UK, 3Medical Research Council Cardiovascular Research Group, Wolfson Institute of Preventive Medicine, Barts and the London Queen Mary's School of Medicine and Dentistry, Charterhouse Square, London EC1M 6BQ, UK, 4Fetal Origins of Adult Disease Division, MRC Environmental Epidemiology Unit, School of Medicine, Southampton University Hospital, Tremona Road, Southampton SO16 6YD, UK
Disentangling the role of genetic factors in complex traits is important to understanding pathogenesis and potentially to disease prediction of common human diseases and pleiotropy (i.e. multiple effects of a genetic factor on different phenotypes) is one key aspect to consider. In 2004, we identified a common tagging haplotype strongly associated (P=0.00001) with low weight in unrelated UK Caucasian men of the NPHSII cohort (Rodríguez et al., 2004, Hum Mol Genet 13: 715-725). This extended haplotype (designated IGF2-INS-TH*5), spans the IGF2-INS-TH cluster in human 11p15 (a strong candidate region influencing cardiovascular risk traits), and has been observed in more than 25% of individuals from two independent samples representative of the general UK population (NPHSII and Hertfordshire Cohort Study, HCS). In follow-up studies, we have replicated the weight-lowering effect of the *5 haplotype in HCS men, and we have found nominal associations between *5 and a wide range of phenotypes influencing metabolic syndrome risk and smoking (Rodríguez et al., 2006, Eur J Hum Genet 14: 109-16; Rodríguez et al., 2006, Pharmacogenetics and Genomics 16: 15-23). In summary, *5 associated significantly with lower weight (P=0.00001), lower BMI (P=0.00001), lower LBM (P=0.025) and lower plasma TG (P=0.021), and with borderline significance with younger age of first regular smoking (AFRS) (P=0.064) in NPHSII. In HCS, *5 associated with lower weight (P=0.015), lower BMI (P=0.005), lower waist circumference (P=0.001), lower waist:hip ratio (P=0.0001), lower proinsulin (P=0.015), lower baseline insulin (P=0.043), lower insulin 120 min after a glucose tolerance test (GTT) (P=0.004), lower insulin area under the curve after a GTT (P=0.015), lower glucose 30 min after a GTT (P=0.043), lower insulin resistance as measured by HOMA-%S (P=0.019), and younger AFRS in Hertfordshire (P=0.044). Multiple regression results suggest that some of the associations may be secondary to others (e.g. most, but not all, of the insulin, glucose and TG effects observed for *5 are substantially dependent on obesity indices). In contrast, other *5 effects (e.g. low weight and smoking) are independent. Taken together, our results suggest that the *5 haplotype may have a pleiotropic effect on different cardiovascular risk traits, by influencing both independent chain of events and interrelated intermediary phenotypes.
Other abstracts in same session