Twin Research and Human Genetics

Articles

Sex-Limited Genome-Wide Linkage Scan for Body Mass Index in an Unselected Sample of 933 Australian Twin Families

Belinda K. Cornesa1, Sarah E. Medlanda2, Manuel A. R. Ferreiraa3, Katherine I. Morleya4, David L. Duffya5, Bastiaan T. Heijmansa6, Grant W. Montgomerya7 and Nicholas G. Martina8 c1

a1 Genetic Epidemiology, Queensland Institute of Medical Research, Australia; School of Medicine, Central Clinical Division, University of Queensland, Australia.

a2 Genetic Epidemiology, Queensland Institute of Medical Research, Australia; School of Psychology, University of Queensland, Australia.

a3 Genetic Epidemiology, Queensland Institute of Medical Research, Australia; School of Medicine, Central Clinical Division, University of Queensland, Australia.

a4 Genetic Epidemiology, Queensland Institute of Medical Research, Australia; Institute for Molecular Bioscience, University of Queensland, Australia.

a5 Genetic Epidemiology, Queensland Institute of Medical Research, Australia.

a6 Molecular Epidemiology, Leiden University Medical Centre, the Netherlands.

a7 Genetic Epidemiology, Queensland Institute of Medical Research, Australia.

a8 Genetic Epidemiology, Queensland Institute of Medical Research, Australia. Nick.Martin@qimr.edu.au

Abstract

Genes involved in pathways regulating body weight may operate differently in men and women. To determine whether sex-limited genes influence the obesity-related phenotype body mass index (BMI), we have conducted a general non- scalar sex-limited genome-wide linkage scan using variance components analysis in Mx (Neale, 2002). BMI measurements and genotypic data were available for 2053 Australian female and male adult twins and their siblings from 933 families. Clinical measures of BMI were available for 64.4% of these individuals, while only self-reported measures were available for the remaining participants. The mean age of participants was 39.0 years of age (SD 12.1 years). The use of a sex-limited linkage model identified areas on the genome where quantitative trait loci (QTL) effects differ between the sexes, particularly on chromosome 8 and 20, providing us with evidence that some of the genes responsible for BMI may have different effects in men and women. Our highest linkage peak was observed at 12q24 (–log10p = 3.02), which was near the recommended threshold for suggestive linkage (–log10p = 3.13). Previous studies have found evidence for a quantitative trait locus on 12q24 affecting BMI in a wide range of populations, and candidate genes for non- insulin-dependent diabetes mellitus, a consequence of obesity, have also been mapped to this region. We also identified many peaks near a –log10p of 2 (threshold for replicating an existing finding) in many areas across the genome that are within regions previously identified by other studies, as well as in locations that harbor genes known to influence weight regulation.

(Received September 16 2005)

(Accepted September 28 2005)

Correspondence:

c1 Address for correspondence: Professor Nicholas G. Martin, Genetic Epidemiology, Queensland Institute of Medical Research, Post Office Royal Brisbane Hospital, Brisbane, QLD 4029, Australia.

Metrics