Novel interactions of adiponectin with the endocrine system and inflammatory parameters.

Abstract

Several markers of chronic immune activation have been found in association with obesity and insulin resistance. We aimed to study the interaction of adiponectin with chronic inflammation and known components of the insulin resistance syndrome. Insulin sensitivity (minimal model analysis) and plasma soluble fractions of TNF-alpha receptor 1 (sTNFR1) and 2 (sTNFR2), adrenal and thyroid function, and adiponectin were evaluated in 68 apparently healthy subjects. An additional group of type 2 diabetic patients (n = 19) similarly studied, except for insulin sensitivity, were also included in the analysis. As reported by others, serum adiponectin concentrations were higher in women than in men (13.55 +/- 9.79 vs. 8.64 +/- 7.83 mg/liter; P = 0.018). They were also higher in healthy subjects compared with diabetic patients (10.35 +/- 8.48 vs. 7.41 +/- 8.31 mg/liter; P = 0.021). As expected also, circulating adiponectin was significantly associated with waist to hip ratio (r = -0.28; P = 0.013), diastolic blood pressure (r = -0.25; P = 0.027), fasting plasma high-density lipoprotein cholesterol (r = 0.35; P = 0.001), triglycerides (r = -0.37; P = 0.001), and insulin sensitivity (r = 0.30; P = 0.011). Additionally, subjects in the higher quartile of circulating adiponectin had lower sTNFR2 concentrations (3.05 vs. 4.37 microg/liter; P = 0.012), a trend to lower sTNFR1 concentrations (1.76 vs. 2.20 microg/liter; P = 0.055), higher concentration of serum morning cortisol (16.86 vs. 13.52 microg/dl; P = 0.027), and higher serum free T(4) levels (1.31 vs. 1.20 ng/dl; P = 0.038). Multiple regression analysis models were constructed to predict adiponectin concentrations. Predictive variables in these models included insulin sensitivity, waist to hip ratio and free T(4), contributing to 17%, 10%, and 8% of adiponectin variance, respectively, These findings suggest that circulating adiponectin differentially modulates insulin action and that thyroid-axis, inflammatory cytokines, and the adrenal cortex might intervene in this modulation.