The aims of this study were; 1) to assess clinical and subclinical factors associated with insulin-stimulated glucose uptake and 2) compare skeletal muscle lipid profile (untargeted lipidomics) in a sub-cohort of insulin sensitive vs. insulin resistant subjects. 66 obese adults (F=46 [BMI=34±3kg/m2], M=20 [BMI=33±2.9 kg/m2]) with homogeneous body composition completed a 2h hyperinsulinemic-euglycemic clamp, and stable isotope dilution methods to assess glucose rate of disappearance (Glucose Rd: index of insulin sensitivity), fatty acid rate of appearance (FA Ra), and hepatic glucose production (Glucose Ra) under basal and hyperinsulinemic conditions. We also performed an array of metabolic measurements (e.g., hepatic and visceral fat accumulation [via MRI], fat oxidation, lipid profile, circulating cytokines). Covariate analysis (LASSO) identified insulin-mediated suppression of FA Ra, basal glucose Ra, and fat oxidation as the cluster of measurements best predicting Glucose Rd (r=0.76, p<0.01). Insulin-mediated suppression of FA Ra was also the strongest independent predictor of Glucose Rd (r=0.51; p<0.01). As a result of greater adipose tissue anti-lipolytic response to insulin, lower FA delivery to skeletal muscle may improve insulin-mediated glucose uptake via lower muscle lipid accumulation. Importantly, we found the accumulation of several long-chain acylcarnitine species (measure of incomplete β-oxidation: C18:0,18:1,20:1,20:3) to be lower (p<0.05) in muscle samples from an insulin sensitive sub-cohort (Glucose Rd > 8.5 µmol/(min•FFM•[Insulin]); n=10) compared with an insulin resistant sub-cohort (< 7.2 µmol/(min•FFM•[Insulin]); n=8).

In conclusion, our findings indicate that greater adipose tissue anti-lipolytic response to insulin is an important predictor of insulin-mediated glucose uptake - and acylcarnitine accumulation in skeletal muscle may mediate this effect.


M. W. Schleh: None. B. J. Ryan: None. C. Ahn: None. A. Ludzki: None. P. Varshney: None. J. B. Gillen: None. J. F. Horowitz: None.


National Institutes of Health (R01DK077966, P30DK089503, T32DK007245, F32DK117522); Canadian Institutes of Health Research (DFS146190, 338735)

Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. More information is available at