BACKGROUND:

In the last several decades obesity has emerged as a major public health threat. While prevention through lifestyle change is the only long-term solution, better understanding of the physiologic mechanisms would greatly assist development of drugs and targeted prevention. Obesity is a highly heritable condition and while genes must account for a substantial proportion of individual susceptibility they have eluded detection. Powerful new genetic and genomic tools now permit comprehensive evaluation of candidate genes, including all genes under linkage peaks. These tools include new genomic resources (the human genome sequence, databases of common SNPs, and the haplotype map), rapid and inexpensive discovery and genotyping and new analytic methods (haplotype-based association and admixture mapping).

DESIGN NARRATIVE:

In a large African American family set Dr. Zhu and colleagues have obtained strong linkage evidence for obesity on chromosome 3q (combined LOD score = 3.7). A prime candidate (adiponectin) lies near this peak. They propose to follow up that finding by combining the epidemiologic data with high-throughput genotyping and move from linkage to association analysis. The results for this QTL will be evaluated within the available environmental factors to assess potential gene-environment and gene-gene interactions. The available phenotypes include body composition, resting metabolic rate, physical activity, plasma insulin, glucose, and leptin. In a family-based design they will examine the linkage peak centered on position 188 cM on chromosome 3q (20 Cm 1-LOD support interval), with the following step-wise strategy: (a) Genotype 200 single nucleotide polymorphisms (SNPs) in this region on 300 families (1,000 individuals); (b) Conduct linkage, linkage disequilibrium and admixture mapping to potentially further narrow the region; and (c) Conduct resequencing and haplotype-based association studies for all candidate genes under the peak. Statistical analysis incorporating intermediate phenotypes and environmental covariates will be used to characterize potential gene x gene or gene x environment interactions. Replication will be tested in additional populations of African and European origin.