DESCRIPTION (provided by applicant): The long-term objective of this project is to better define risk factors for age-related bone loss and fractures, For the past 15 years, this research program has focused on three specific themes; 1) fracture epidemiology by establishing the Framingham Fracture Registry, 2) the genetics of bone traits, 3) the influence of dietary factors on the skeleton, including the interaction between genes, diet and other environmental factors on bone density. The next five years will expand on these three themes using both previously collected data as well as new data to be generated. The first aim is to examine the association between lower extremity lean mass and total lean mass, previously measured by dual x-ray absorptiometry (DXA) and the risk for hip and non-vertebral fracture using the Framingham Fracture Registry of the Original and Offspring Cohorts. The second aim is to measure lumbar spine volumetric trabecular bone mineral density (vTBMD) and vertebral cross-sectional area (CSA) using QCT scans of the spine acquired in 3,529 members of the Offspring and 3rd generation Cohorts. These new measurements will be used in aim 3 to perform a genome-wide association study (GWAS) using 3529 subjects from the Offspring and Gen3 Cohorts who will have genotyping data from the "Framingham Heart Study SNP Health Association Resource" (FHS SHARe Project) that will provide 550K dense single nucleotide polymorphisms (SNPs) genome wide. We will also separately perform a second GWAS on 3507 Original and Offspring participants with DXA measures of areal BMD (aBMD) using the same FHS SHARe Project genotyping data. Following the GWAS analyses using vTBMD and CSA, a staged replication will be performed in two separate cohorts: the Family Heart Study - SCAN (FHS-SCAN) and the AGES Reykjavik Study (AGES). These two cohorts have identical QCT phenotypes and will undertake new genotyping to replicate findings from our GWAS. This staged replication will lead to the identification of the 5 most significant non-overlapping (r2<0.8) SNPs/candidate regions across all three cohorts that will then be further explored with additional genotyping in the Framingham Study. We will also replicate our GWAS results using DXA-derived aBMD phenotypes in silico with results from the identical phenotypes in the Rotterdam Study GWAS. The final aim will be to investigate the interaction of SNPs in four genes related to age-associated accumulation of marrow fat and osteoblast dysfunction (PPAR3, Runx2, HIVEP3, and WWP1) with dietary fat intake on vTBMD, and the interaction of four genes related to methylation pathways (MTHFR, MTR, MTRR, CBS) with plasma folate, vitamin B12 and B6 levels on vTBMD. The findings from this study will provide important new data on the influence of leg muscle mass on fracture risk. This will also be the first GWAS in the U.S. to use dense SNP genotyping data along with state-of-the-art skeletal imaging phenotypes to find genes related to bone density. Finally, using dietary assessments and B-vitamin levels, this study will characterize important potential interactions between genes and nutrients as they affect bone density. PUBLIC HEALTH RELEVANCE This research is relevant to public health in several ways. First it will determine how muscle mass in the legs contributes to fracture risk. Second it will identify genes as well as nutritional factors such as the B-vitamins and dietary fat that work together to increase the risk for osteoporosis.