Aims/hypothesis: Obesity and insulin resistance may be associated with altered expression and secretion of adipokines. Physical activity can markedly improve insulin sensitivity, but the association with adipokines remains largely unknown. In this study, we examined the effects of physical activity on the subcutaneous white adipose tissue (scWAT) secretome and its relationship to insulin sensitivity.
Methods: As reported previously, we enrolled 26 sedentary, middle-aged men (13 dysglycaemic and overweight; 13 normoglycaemic and of healthy weight) into a 12 week, supervised, intensive physical exercise intervention that included two endurance and two resistance sessions each week. Insulin sensitivity was measured as the glucose infusion rate from a euglycaemic-hyperinsulinaemic clamp. In our previous study, we measured maximum oxygen uptake, upper- and lower-body strength and a range of circulating biomarkers, and quantified adipose tissue depots using MRI and magnetic resonance spectroscopy. We have now performed global mRNA sequencing, microarrays and RT-PCR of scWAT and skeletal muscle biopsies, and quantified selected plasma adipokines by ELISA.
Results: Insulin sensitivity increased similarly in both dysglycaemic (45%) and normoglycaemic (38%) men after 12 weeks of exercise, as reported previously. mRNA sequencing of scWAT revealed 90 transcripts that responded to exercise in dysglycaemic men, whereas only marginal changes were observed in normoglycaemic men. These results were validated using microarrays and RT-PCR. A total of 62 out of 90 transcripts encoded secreted proteins. Overall, 17 transcripts were upregulated and 73 transcripts were downregulated. Downregulated transcripts included several macrophage markers, and were associated with inflammatory and immune-related pathways. Levels of these immune-related transcripts were enhanced in dysglycaemic men vs normoglycaemic men at baseline, but were normalised after the exercise intervention. Principal component and correlation analyses revealed inverse correlations between levels of these immune-related transcripts and insulin sensitivity at baseline, after the intervention, and for the change between baseline and after the intervention. In addition, levels of these transcripts at baseline could predict exercise-induced improvements in insulin sensitivity. Adipokine levels in scWAT (but not in skeletal muscle) were significantly correlated with corresponding plasma adipokine concentrations, as exemplified by leptin, high-molecular-weight adiponectin and secreted frizzled-related protein 4 (SFRP4). SFRP4 mRNA was the most exercise-responsive transcript in scWAT from dysglycaemic men, and plasma SFRP4 concentrations were reduced in dysglycaemic men, but not in normoglycaemic men, after 12 weeks of exercise.
Conclusions/interpretation: This study indicates that scWAT may be an important mediator of exercise-induced improvements in insulin sensitivity, especially in overweight dysglycaemic individuals at increased risk of developing type 2 diabetes.
Keywords: Adipokines; Humans; Insulin sensitivity; Long-term exercise; Prediabetes; Subcutaneous adipose tissue.