Effects of Aerobic and Resistance Training on Circulating Micro-RNA Expression Profile in Subjects With Type 2 Diabetes

Context: Structured exercise programs are of great benefit for the treatment of type 2 diabetes (T2DM). However, whether aerobic (AER) or resistance (RES) exercise training exerts specific epigenetic changes through the expression profile of circulating miRNAs (c-miRNAs) is still largely unknown. Objective: To assess whether the c-miRNAs profile changes after either AER or RES training in subjects with T2DM. Design: Twenty-four patients with T2DM randomized to AER or RES training protocols were randomly selected from the Resistance vs. Aerobic Exercise in Type 2 Diabetes (RAED2) Trial (N-AER = 12; N-RES = 12). The baseline and post-training levels of 179 c-miRNAs were initially measured by RT-PCR in 6 individuals (N-AER = 3; N-RES = 3). C-miRNAs exhibiting >= 40% fold change variation and/or nominal significance from baseline were measured in the whole group. Results: Nineteen c-miRNAs were eventually assessed in the whole group. Compared with baseline, the post-training levels of miR-423-3p, miR-451a, and miR-766-3p were significantly up-regulated, irrespective of exercise type (P < 0.0026; 0.05/19), and targeted upstream pathways relevant to fatty acids biosynthesis and metabolic regulation. MiR-451a and miR-423-3p were significantly correlated with fat loss (rho = 0.45 and 0.43, respectively) and resulted, alone or in combination, in being predictors of fat loss in generalized linear regression models including exercise type as covariate. Only the association with miR-451a eventually retained significance after further correction for age, sex, body mass index, and HbA1c. Conclusions: Exercise training in T2DM is associated with substantial c-miRNAs profile changes, irrespective of exercise type and other relevant metabolic covariates. The mechanistic significance of the observed relationship between fat loss and the epigenetic modifications induced by exercise warrants further investigation in larger datasets.