Adiponectin is expressed in and secreted from visceral fat, and its plasma level has been reported to correlate with insulin resistance and triglyceride metabolism in nondiabetic subjects (1,2). However, these relationships have not been evaluated in Japanese metabolically obese normal-weight (BMI <25 kg/m2 and visceral fat areas [evaluated by abdominal CT scanning] ≥100 cm2) men with normal glucose tolerance (NGT) (3–5).
The present study comprised 16 metabolically obese normal-weight men (aged 35.6 ± 1.8 [mean ± SE] years, BMI 23.8 ± 0.3 kg/m2, visceral fat areas 130.8 ± 5.2 cm2) and 15 age-matched normal men (BMI <25 and visceral fat areas <100 cm2) (aged 33.6 ± 1.8 years, BMI 20.9 ± 0.3 kg/m2, visceral fat areas 56.5 ± 5.1 cm2) with NGT.
The plasma levels of adiponectin were measured using a radioimmunoassay kit (Linco Research, St. Charles, MO).
Comparisons between metabolically obese normal-weight and normal subjects were done using the Mann-Whitney U test, and correlations were evaluated by Spearman’s rank correlation.
There were no significant differences in plasma levels of adiponectin between metabolically obese normal-weight (10.2 ± 1.3 ng/ml) and normal subjects (12.0 ± 0.8 ng/ml). The BMI (P < 0.01) and serum levels of triglyceride (1.67 ± 0.14 vs. 0.92 ± 0.09 mmol/l, P < 0.01) were significantly increased in metabolically obese normal-weight subjects compared with normal subjects. The glucose infusion rate (index of insulin resistance during the euglycemic-hyperinsulinemic clamp study) in metabolically obese normal-weight subjects (53.9 ± 3.4 μmol · kg−1 · min−1; P < 0.01) were significantly decreased compared with normal subjects (65.8 ± 2.7 μmol · kg−1 · min−1) (4,6).
The plasma levels of adiponectin were significantly correlated with glucose infusion rate (r = 0.509, P < 0.05), serum levels of triglyceride (r = −0.730, P < 0.01), and the visceral fat areas (r = −0.597, P < 0.05) in metabolically obese normal-weight subjects.
There were not significant correlations between plasma levels of adiponectin and glucose infusion rate (r = 0.146, P = 0.584), serum levels of triglyceride (r = −0.446, P = 0.095), or visceral fat areas (r = −0.214, P = 0.423) in normal subjects.
Visceral fat is an important determinant factor of the plasma level of adiponectin, which is known to exert an insulin-sensitizing effect (2,7). Unexpectedly, similar plasma levels of adiponectin and different glucose infusion rates were observed in metabolically obese normal-weight and normal subjects. The small number of patients may be the explanation for this unexpected result. Further study should be carried out in a larger population of Japanese metabolically obese normal-weight subjects.
Significant correlation between plasma levels of adiponectin and glucose infusion rate was observed in metabolically obese normal-weight subjects. Plasma adiponectin levels may play an important role in the development of insulin resistance in Japanese metabolically obese normal-weight subjects.
The plasma levels of adiponectin were significantly correlated with the serum levels of triglyceride in metabolically obese normal-weight subjects. Cnop et al. (2) demonstrated that association of adiponectin with increased visceral fat may shift the fate of apolipoprotein B away from degradation toward secretion from the liver, resulting in elevated triglyceride concentrations. This phenomenon might have occurred in our Japanese metabolically obese normal-weight subjects with NGT.
