Irisin is more strongly predicted by muscle oxidative potential than adiposity in non-diabetic men.

Numerous controversies surround the peptide hormone irisin. Although implicated as a myokine promoting the browning of adipose tissue in rodents, its roles in humans remain unclear. Contradictory results have also been found with respect to the relationships between adiposity or metabolic health and plasma irisin levels in humans. We investigated the relationship between irisin levels and body composition (hydrostatic weighing), insulin sensitivity (hyperinsulinemic-euglycemic clamp), fitness level (ergocycle VO2max) and skeletal muscle metabolic profile in 53 men (aged 34-53 years) from four groups: sedentary non-obese controls (body mass index [BMI] <25 kg/m(2)), sedentary obese (BMI >30 kg/m(2)), sedentary obese glucose-intolerant, and non-obese highly trained endurance active. Baseline plasma irisin levels were significantly different between groups, being lowest in trained men (140.6 ± 38.2 ng/mL) and highest in metabolically deteriorated glucose-intolerant subjects (204.0 ± 50.5 ng/mL; ANOVA p = 0.01). Including all subjects, irisin levels were positively associated with adiposity (e.g. fat mass, r = 0.430, p < 0.01) and negatively associated with fitness (r = -0.369, p < 0.01), insulin sensitivity (M/I, r = -0.355, p < 0.01) and muscle citrate synthase (CS) activity (r = -0.482, p < 0.01). Most correlations lost statistical significance when excluding active individuals, except for insulin resistance (r = -0.413, p < 0.01) and CS (r = -0.462, p < 0.01). Multiple regression analyses reveal CS as the strongest independent predictor of irisin levels (r(2) range 0.214 to 0.237). We conclude that muscle oxidative potential is an important factor linked to circulating irisin levels.