Association of the Pro12Ala variant in the peroxisome proliferator-activated receptor-gamma2 gene with obesity in two Caucasian populations.

Abstract

The nuclear receptor, peroxisome proliferatora c t ivated receptor( P PA R ), is an important regulator of adipocyte differentiation and a modulator of intracellular insulin-signaling events (1). PPA R mRNA expression in both skeletal muscle and adipose tissue in vitro is induced by insulin (2,3). In skeletal muscle of obese subjects, PPA R mRNA is elevated in direct relation to BMI and fasting insulinemia (2); reports are mixed as to whether expression is increased in adipose tissue of obese subjects (3–5). In one report, activators of PPA R were shown to increase adiposity in a rodent model (6), while in another (7) they were found to increase the number of adipocytes, but not the mass of adipose tissue. Clinically, most studies have not found that PPA R –activating thiazolidinediones cause weight gain when administered to humans for treatment of diabetes (1). Alternate use of promoters and differential splicing of the human PPARgene results in two isoforms: PPAR1 and PPAR2. PPAR2 contains 28 additional amino acids at its NH2 terminus (1,8). PPAR1 and PPAR2 both are expressed in adipose tissue (1,3–5), and there appears to be no difference in the abilities of the two isoforms to participate in ligand-induced initiation of transcription of target genes or in ligand-induced adipocyte differentiation (9). Interestingly, however, it was demonstrated recently that P PA R could activate transcription in a ligand-independent fashion and that insulin potentiated this activity (9). F u r t hermore, the 2 isoform is much more potent at doing so than is the 1 isoform (9), suggesting a possible distinct role for PPA R 2 in obesity, insulin resistance, and diabetes. R e c e n t l y, we reported a naturally occurring variant in the human PPA R 2 gene that predicts substitution at amino acid 12 of alanine for the proline that is present in both normal human and mouse PPA R 2 sequences (10). This amino acid position is within the domain of PPA R 2 that enhances ligand-independent activation, as described by Werman et al. (9). Because the substitution of alanine for proline is nonconservative and could cause a significant change in protein structure, we hypothesize that it may alter the function of P PA R 2 such that individuals with this variant may be at an increased genetic risk for obesity and/or insulin resistance. To test this hypothesis, we genotyped and performed association studies of the Pro12Ala PPA R 2 variant in two independently recruited cohorts of unrelated, nondiabetic, adult Caucasian subjects from the Baltimore metropolitan area. One cohort of 517 subjects, with a distribution of being lean-to-moderately obese (mean BMI 26.5 kg/m, range 18.6–43.2 kg/m), is from the Baltimore Longitudinal Study on Aging (BLSA) and has been recruited continuously since 1958. The other cohort is of 169 very obese subjects (mean BMI 36.5 kg/m, range 24.2–76.8 kg/m), recruited prospectively from August 1994 through June 1996 from the Johns Hopkins University We i g h t Management Center (JHU-WMC). Subjects with diabetes by history or with fasting glucose ≥7.1 mmol/l (126 mg/dl) were excluded to avoid the well-known confounding effects of diabetes and its treatment on the obesity-related traits that were studied. All protocols were approved by the Institutional Review Board of the Johns Hopkins University, and informed written consent was obtained from all subjects. Genomic DNA was obtained from peripheral blood using standard methods, and the Pro12Ala PPA R 2 variant was detected by polymerase chain reaction-restriction fragment length polymorphism (PCR-RFLP) analysis. This PCR-RFLP analysis, previously described (10), uses a mutagenic PCR primer to introduce a B s t UI site only when a C → G substitution at nucleotide 34 of the PPA R 2 gene is present. Genotyping was repeated for all Ala 2 homozygotes, several Pro12Ala heterozygotes chosen randomly, and several Pro 2