Adiponectin concentration and insulin indicators following overfeeding in identical twins
- First Online:
- Cite this article as:
- Ukkola, O., Terán-García, M., Tremblay, A. et al. J Endocrinol Invest (2008) 31: 132. doi:10.1007/BF03345579
- 32 Downloads
Low adiponectin levels have been associated with high body mass index, low insulin sensitivity, and diabetes. Objective: To assess the relationships between changes in serum adiponectin concentration and adiposity, glucose, and insulin in response to long-term overfeeding in identical twins and to calculate the twin resemblance in serum adiponectin concentrations. Subjects and design: Twenty-four sedentary young men [mean (±SD) age, 21±2 yr] who constituted 12 pairs of healthy identical twins were studied for metabolic and adiponectin changes in response to overfeeding. Intervention: Subjects were overfed by 84,000 kcal over a 100-day period. Outcome measures: The overfeeding study provides an opportunity to examine the relationships between adiponectin and changes in body weight, adiposity, plasma glucose and insulin. Results: Serum adiponectin concentration correlated positively with body weight (r= 0.41, p=0.05) at baseline but not with indicators of adiposity or with visceral fat. No relationship existed between baseline adiponectin concentration and body weight or adiposity gains with overfeeding. However, serum adiponectin decreased significantly by −2.35±0.48 μg/ml (p=0.001) in response to overfeeding. Baseline adiponectin levels correlated negatively with changes in plasma fasting glucose levels (r=−0.53, p=0.01) and homeostasis model assessment index (r=−0.41, p=0.05), independently of fat mass. The intrapair coefficient for twin resemblance (r=0.75, p=0.001) strongly suggests that baseline serum adiponectin concentration is a familial trait. Conclusions: These data provide evidence that adiponectin concentration is a familial trait in normal-weight individuals, that it decreases when challenged by positive energy balance, and that its overfeeding-induced variations are correlated with glucose and insulin levels.