Osteoporosis International

, Volume 18, Issue 6, pp 733–741

Genetic markers for ancestry are correlated with body composition traits in older African Americans

  • J. R. Shaffer
  • C. M. Kammerer
  • D. Reich
  • G. McDonald
  • N. Patterson
  • B. Goodpaster
  • D. C. Bauer
  • J. Li
  • A. B. Newman
  • J. A. Cauley
  • T. B. Harris
  • F. Tylavsky
  • R. E Ferrell
  • J. M. Zmuda
  • for the Health ABC study
Original Article

DOI: 10.1007/s00198-006-0316-6

Cite this article as:
Shaffer, J.R., Kammerer, C.M., Reich, D. et al. Osteoporos Int (2007) 18: 733. doi:10.1007/s00198-006-0316-6

Abstract

Summary

Individual-specific percent European ancestry was assessed in 1,277 African Americans. We found significant correlations between proportion of European ancestry and several musculoskeletal traits, indicating that admixture mapping may be a useful strategy for locating genes affecting these traits.

Introduction

Genotype data for admixed populations can be used to detect chromosomal regions influencing disease risk if allele frequencies at disease-related loci differ between parental populations. We assessed evidence for differentially distributed alleles affecting bone and body composition traits in African Americans.

Methods

Bone mineral density (BMD) and body composition data were collected for 1,277 African and 1,790 European Americans (aged 70–79). Maximum likelihood methods were used to estimate individual-specific percent European ancestry for African Americans genotyped at 37 ancestry-informative genetic markers. Partial correlations between body composition traits and percent European ancestry were calculated while simultaneously adjusting for the effects of covariates.

Results

Percent European ancestry (median = 18.7%) in African Americans was correlated with femoral neck BMD in women (r = −0.18, p < 10−5) and trabecular spine BMD in both sexes (r = −0.18, p < 10−5) independently of body size, fat, lean mass, and other covariates. Significant associations of European ancestry with appendicular lean mass (r = −0.19, p < 10−10), total lean mass (r = −0.12, p < 10−4), and total body fat (r = 0.09, p < 0.002) were also observed for both sexes.

Conclusions

These results indicate that some population differences in body composition may be due to population-specific allele frequencies, suggesting the utility of admixture mapping for identifying susceptibility genes for osteoporosis, sarcopenia, and obesity.

Keywords

Admixture mapping Body composition Genetic ancestry Linkage analysis Osteoporosis Single nucleotide polymorphisms (SNPs) 

Supplementary material

Copyright information

© International Osteoporosis Foundation and National Osteoporosis Foundation 2007

Authors and Affiliations

  • J. R. Shaffer
    • 1
  • C. M. Kammerer
    • 1
  • D. Reich
    • 2
    • 3
  • G. McDonald
    • 2
    • 3
  • N. Patterson
    • 3
  • B. Goodpaster
    • 4
  • D. C. Bauer
    • 5
  • J. Li
    • 5
  • A. B. Newman
    • 1
  • J. A. Cauley
    • 1
  • T. B. Harris
    • 6
  • F. Tylavsky
    • 7
  • R. E Ferrell
    • 1
  • J. M. Zmuda
    • 1
  • for the Health ABC study
  1. 1.Graduate School of Public HealthUniversity of PittsburghPittsburghUSA
  2. 2.Department of GeneticsHarvard Medical SchoolBostonUSA
  3. 3.Broad Institute of Harvard and MITCambridgeUSA
  4. 4.Department of EndocrinologyUniversity of PittsburghPittsburghUSA
  5. 5.University of California San FranciscoSan FranciscoUSA
  6. 6.National Institute of AgingBethesdaUSA
  7. 7.University of Tennessee Health Science CenterMemphisUSA