| pubmed-article:12879219 | pubmed:abstractText | We examined the association of candidate gene polymorphisms with bone mineral density (BMD) and bone mineral content (BMC) in a cohort of 428 healthy non-Hispanic white children participating in the Iowa Bone Development Study, a longitudinal study of determinants of bone accrual in childhood. BMD and BMC measurements of the hip, spine and whole body were made using a Hologic 2000 Plus densitometer in 228 girls and 200 boys ages 4.5-6.5 years. Genotypes at 14 loci representing eight candidate genes [type I collagen genes (COL1A1 and COL1A2), osteocalcin, osteonectin, osteopontin, vitamin D receptor (VDR), estrogen receptor (ER), androgen receptor (AR)] were determined. Gender-specific and gender-combined prediction models for bone measures that included age, weight, height (and gender) were developed using multiple linear regression analysis. COL1A2 and osteocalcin genotypes were identified as having the strongest and most consistent association with BMD/BMC measures. Osteonectin, osteopontin and VDR translation initiation site polymorphisms were associated with some individual bone measures, but none of the associations was as consistent as those identified for the COL1A2 and osteocalcin genes. No association was identified with COL1A1 (RsaI and Sp1), VDR (BsmI) and ER polymorphisms (PvuII, XbaI, TA) and BMD/BMC. However, we identified significant gene-by-gene interaction effects involving the ER and both VDR and osteocalcin, which were associated with BMD/BMC. Our data suggest that genetic variation at multiple genetic loci is important in bone accrual in children. Moreover, the combination of genotypes as several loci may be as important as a single genotype for determining BMD and BMC. | lld:pubmed |
