Pronounced effects of acute endurance exercise on gene expression in resting and exercising human skeletal muscle

Abstract

Regular physical activity positively influences whole body energy metabolism and substrate handling in exercising muscle. While it is recognized that the effects of exercise extend beyond exercising muscle, it is unclear to what extent exercise impacts non-exercising muscles. Here we investigated the effects of an acute endurance exercise bouts on gene expression in exercising and non-exercising human muscle. To that end, 12 male subjects aged 44-56 performed one hour of one-legged cycling at 50% W(max). Muscle biopsies were taken from the exercising and non-exercising leg before and immediately after exercise and analyzed by microarray. One-legged cycling raised plasma lactate, free fatty acids, cortisol, noradrenalin, and adrenalin levels. Surprisingly, acute endurance exercise not only caused pronounced gene expression changes in exercising muscle but also in non-exercising muscle. In the exercising leg the three most highly induced genes were all part of the NR4A family. Remarkably, many genes induced in non-exercising muscle were PPAR targets or related to PPAR signalling, including PDK4, ANGPTL4 and SLC22A5. Pathway analysis confirmed this finding. In conclusion, our data indicate that acute endurance exercise elicits pronounced changes in gene expression in non-exercising muscle, which are likely mediated by changes in circulating factors such as free fatty acids. The study points to a major influence of exercise beyond the contracting muscle.

Figure 1. Experimental design.

The timeline of the study (A) and set-up of the endurance exercise bout (B). After 2 familiarization trails and 2 exercise tests, subjects performed 1 hour of submaximal one-legged endurance exercise. Before and after the exercise bout muscle biopsies and venous blood samples were taken, and another blood sample was taken 2 hours after the end of exercise.

Figure 2. Exercise increases heart rate and plasma levels of FFA, insulin, cortisol and noradrenaline.

Heart rate reserve (%) was calculated based on the heart rate measured during the exercise (N = 12). Plasma glucose, triglyceride, free fatty acids, lactate, insulin, cortisol, adrenaline and noradrenaline were measured before and after exercise (T0 and T1; N = 12) and after 2 hours of recovery (T3; N = 12). a = p<0.05 compared to T0, b = p<0.5 compared to T3, c = p<0.1 compared to T0, p<0.1 compared to T3, repeated measures ANOVA. Depicted is mean ± SEM.

Figure 3. Exercise mainly causes upregulation of gene expression in both the exercising and non-exercising leg.

(A) Venn diagram of significantly regulated genes and their overlap. (B) Flowchart of microarray analysis. Heatmaps of all significant genes in the non-exercising (C) and exercising leg (D) N = 9, IQR  =  interquartile range.

Figure 4. Top 20 of most highly induced genes in exercising and non-exercising leg.

A) Left panel shows the top 20 of upregulated genes in the exercising leg (N = 9), right panel the corresponding genes in the non-exercising leg. B) Left panel shows the top 20 of upregulated genes in the non-exercising leg (N = 7), right panel the corresponding genes in the exercising leg. Green is a signal log ratio of −3, red a signal log ratio of 3. Values are displayed per subject to visualize inter-individual differences. FC = fold change, *  = p<0.05, ^#  = p<0.1 between exercising and non-exercising leg.

Figure 5. Induction of transcription factor pathways by exercise.

Transcription factor pathways related to growth, stress response, cAMP signalling and hypoxia were induced by exercise. Transcription factor pathways were identified for the exercising leg using IPA and are displayed in a bar diagram. Genes induced by exercise for the different transcription factors can be found in table S1. Transcription factors with a z-score above 1.5 (or under −1.5) are considered as biologically relevant.

Figure 6. ClueGO network analysis.

Analysis shows significant regulation of several GO categories involved in skeletal muscle development, angiogenesis, inflammation and MAPK cascade in the exercising leg (A; N = 9) and basal metabolism and signalling in the non-exercising leg (B; N = 7). The nodes represent significantly changed GO categories. Lines represent the overlap between different categories. All nodes with a large overlap have a similar colour.