Angiopoietin-like protein 4 (ANGPTL4) represents an adipokine with metabolic effects within adipose tissue, such as inhibition of lipoprotein lipase activity and stimulation of lipolysis. These effects were convincingly demonstrated in mice. Therefore, we asked whether genetic variation within the ANGPTL4 gene contributes to prediabetic phenotypes, such as dyslipidemia, insulin resistance, or beta-cell dysfunction, in white subjects at an increased risk for type 2 diabetes mellitus. We genotyped 629 subjects with and without a family history of diabetes for the 4 single nucleotide polymorphisms (SNPs) rs4076317, rs2278236, rs1044250, and rs11672433 and performed correlational analyses with metabolic traits. For metabolic characterization, all subjects underwent an oral glucose tolerance test; a subset was additionally characterized by hyperinsulinemic-euglycemic clamp. The 4 SNPs rs4076317, rs2278236, rs1044250, and rs11672433 cover 100% of common genetic variation (minor allele frequency>or=0.05) within the ANGPTL4 gene (r2>or=0.8). None of these SNPs revealed significant correlation with anthropometric data (sex, age, body mass index, body fat, and waist-hip ratio) or with family history of diabetes. Furthermore, no reliable correlations were found with fasting triglycerides, fasting nonesterified fatty acids, and area under the curve of nonesterified fatty acids during oral glucose tolerance test or with parameters of insulin sensitivity and insulin secretion. Finally, haplotype analysis revealed the existence of 8 common diplotypes. None of these, however, was significantly correlated with insulin sensitivity, insulin secretion, or plasma lipid measures. We conclude that common genetic variation within the ANGPTL4 gene may not play a major role in the development of prediabetic phenotypes in our white population.

Visceral fat mass is a strong and independent predictor of obesity-related disorders. To date, little is known about the genetic determinants of regional body fat distribution in humans. As candidates of regional fat distribution, we investigated the fat mass- and obesity-associated gene, the peroxisome proliferator-activated receptor-delta gene, and the retinoic acid receptor responder 2 (RARRES2) gene. We studied whether genetic variation within these genes contributes to the development of disproportionate visceral obesity and obesity-related traits, such as insulin resistance and beta-cell dysfunction. We genotyped 337 subjects with an increased risk for type 2 diabetes mellitus for tagging single nucleotide polymorphisms (SNPs) in the 3 genes and performed association analyses with anthropometric data and parameters of insulin sensitivity and beta-cell function. All subjects underwent an oral glucose tolerance test; a subset was additionally characterized by a hyperinsulinemic-euglycemic clamp. Body fat distribution was assessed by nuclear magnetic resonance imaging. The fat mass- and obesity-associated gene SNP rs8050136 was nominally associated with body mass index (P = .0130), but not with body fat distribution, after appropriate adjustment. Magnetic resonance imaging-quantified visceral fat mass was significantly associated with RARRES2 SNP rs17173608 and nominally associated with RARRES2 SNP rs10278590 in nonobese subjects (P = .0002 and P = .0423, respectively), with carriers of the minor alleles displaying lower visceral adipose tissue mass. Besides, the minor allele of SNP rs17173608 was nominally associated with a lower waist-to-hip ratio (P = .0295). In obese subjects, these associations were not detected. No associations were found between the peroxisome proliferator-activated receptor-delta gene and measures of whole-body adiposity and of body fat distribution. All SNPs were associated neither with insulin sensitivity nor with insulin secretion. Common genetic variation within RARRES2 is associated with increased visceral fat mass in nonobese subjects. In generalized obesity, this genetic effect may be masked by the close association between whole-body obesity and visceral fat mass.