Genome-wide association scans identified CTNNBL1 as a novel gene for obesity

Abstract

Obesity is a major public health problem with strong genetic determination; however, the genetic factors underlying obesity are largely unknown. In this study, we performed a genome-wide association scan for obesity by examining approximately 500 000 single-nucleotide polymorphisms (SNPs) in a sample of 1000 unrelated US Caucasians. We identified a novel gene, CTNNBL1, which has multiple SNPs associated with body mass index (BMI) and fat mass. The most significant SNP, rs6013029, achieved experiment-wise P-values of 2.69 x 10(-7) for BMI and of 4.99 x 10(-8) for fat mass, respectively. The SNP rs6013029 minor allele T confers an average increase in BMI and fat mass of 2.67 kg/m(2) and 5.96 kg, respectively, compared with the alternative allele G. We further genotyped the five most significant CTNNBL1 SNPs in a French case-control sample comprising 896 class III obese adults (BMI > or = 40 kg/m(2)) and 2916 lean adults (BMI < 25 kg/m(2)). All five SNPs showed consistent associations with obesity (8.83 x 10(-3) < P < 6.96 x 10(-4)). Those subjects who were homozygous for the rs6013029 T allele had 1.42-fold increased odds of obesity compared with those without the T allele. The protein structure of CTNNBL1 is homologous to beta-catenin, a family of proteins containing armadillo repeats, suggesting similar biological functions. beta-Catenin is involved in the Wnt/beta-catenin-signaling pathway which appears to contribute to maintaining the undifferentiated state of pre-adipocytes by inhibiting adipogenic gene expression. Our study hence suggests a novel mechanism for the development of obesity, where CTNNBL1 may play an important role. Our study also provided supportive evidence for previously identified associations between obesity and INSIG2 and PFKP, but not FTO.

Figure 1.

Association signals for all genotyped SNPs in the CTNNBL1 gene in US Caucasians. The empty diamonds represent association signals for BMI and the filled diamonds represent association signals for fat mass. The Y axis is −log₁₀(P) values. The haplotype block maps were constructed using our own genotype data showing pairwise LD (D′).

Figure 2.

BMI (kg/m²) and fat mass (kg) distribution in different genotype groups of rs6013029 in US Caucasians.

Figure 3.

The results of association tests for BMI and fat mass before and after adjusting for genotype of rs6013029. (A) Results of association analyses for BMI. (B) Results of association analyses for fat mass. Filled squares represent association signals in original association analyses. Empty diamonds represent association signals after adjusting genotype at the rs6013029 as a covariate.

Figure 4.

Q–Q plots for BMI (A) and fat mass (B) association −log₁₀(P) values for the US Caucasian sample and P-values expected under the null distribution for the GWA SNPs.