Skeletal muscle Rac1 mediates exercise training adaptations towards muscle glycogen resynthesis and protein synthesis

Abstract

Long-term exercise training elicits tremendous health benefits; however, the molecular understanding is incomplete and identifying therapeutic targets has been challenging. Rho GTPases are among the most regulated groups of proteins after exercise in human skeletal muscle, yet, unexplored candidates for mediating the effects of exercise training. We found that the Rho GTPase Rac1 was activated acutely after multiple exercise modalities in human skeletal muscle. Loss of Rac1 specifically in muscle attenuated contraction-induced muscle protein synthesis, diminished improvements in running capacity, and prevented muscle hypertrophy after exercise training in mice. Additionally, Ncf1∗ mice revealed that Rac1 regulated glycogen resynthesis via a NOX2-dependent mechanism. Molecularly, Rac1 was required for contraction-induced p38MAPK signaling towards HSP27, MNK1, and CREB phosphorylation. In vivo muscle-targeted overexpression of a hyperactive Rac1-mutant elevated reactive oxidant species production during exercise but did not affect muscle mass. Using mass spectrometry-based proteomics, we found that loss or gain of Rac1 muscle protein affected pathways related to cytoskeleton organization, muscle adaptation, and large ribosomal subunits. Thus, skeletal muscle Rac1 mediates both molecular and functional adaptation to exercise training.

Keywords: Contraction; Exercise training; Glycogen; Metabolism; Muscle hypertrophy; Protein synthesis; Rac1; Skeletal muscle.

Graphical abstract

Fig. 1

Post-contraction protein synthesis and glycogen resynthesis require Rac1, which regulates p38 MAPK singling. (A) Schematic illustration of Rac1-PAK1 signaling at rest and in situ contracted (CTX) mouse quadriceps muscle. (B) Experimental setup of the investigation of post-contraction (4 h, 4h-CTX) intramyocellular signaling in littermate control (Con) and inducible muscle-specific Rac1 knockout (Rac1 imKO) mice. (C) Relative increase in protein synthesis (puromycin incorporation measured by Western blot) and (D) glycogen synthesis 4 h after contraction in quadriceps muscle. (E) Signaling events leading to protein synthesis in muscle. The effect of in situ contraction on: (F) mTORC1 signaling, (G) ERK MAPK phosphorylation, (H) p38 MAPK phosphorylation, and (I) p38 MAPK signaling. Representative blots are shown in the bottom right corner. Acute contraction in mice (A): n = 8. Post-contraction in mice (B–G), Control mice: n = 12, Rac1 imKO mice: n = 18. Significant differences basal leg vs. contraction are indicated; ∗/∗∗/∗∗∗ = p < 0.05/p < 0.01/p < 0.001. Significant differences Rac1 imKO vs. control are indicated as; #/# #/# # # = p < 0.05/p < 0.01/p < 0.001. Data are presented as boxplots or bar plots, mean + SEM incl. individual values.

Fig. 2

Rac1 affects post-contraction protein synthesis independent of NOX2 activity. (A) Illustration of the activation of the NOX2 complex and subsequent O₂⁻-production with and without p47phox mutation (∗Ncf1). (B) Contraction protocol in ∗Ncf1mice. (C) Relative increase in protein synthesis (puromycin incorporation measured by Western blot) in quadriceps muscle from control mice and mice carrying a mutation in the ∗Ncf1 gene (p47^phox). (D) Muscle glycogen after contraction. (E) Phosphorylation of mTORC1 substrates, (F) p38 MAPK and ERK1/2 MAPK, and (G) p38 MAPK substrates. Representative blots are shown in the bottom right corner. Control: n = 8, ∗Ncf1: n = 7. Significant differences from basal vs. contraction are indicated; ∗/∗∗/∗∗∗ = p < 0.05/p < 0.01/p < 0.001. Data are presented as boxplots or mean + SEM incl. individual values.

Fig. 3

Rac1 signaling is increased by several exercise modalities in skeletal muscle of young healthy men. (A) Overview of the exercise interventions. (B) Rac1 signaling (pPAK1 T423) in vastus lateralis skeletal muscle of young healthy men before (Pre), immediately after (Post), and 3 h into the recovery (Reco) from endurance, sprint, or resistance exercise. (C) pHSP27 S82. (D) pMNK1 T197/202. (E) PAK1 and Rac1 protein content. n = 8. Representative blots are presented at the bottom of the figure. Significant differences between pre and post/reco are indicated as; ∗ = p < 0.05, ∗∗ = p < 0.01, ∗∗∗ = p < 0.001. Significant differences between modalities vs. endurance are indicated as; # # = p < 0.01, # # # = p < 0.001.

Fig. 4

Exercise training increases Rac1 protein content in human and mouse skeletal muscle. (A) Effect of single-leg exercise training (ET) on Rac1 protein content in human vastus lateralis skeletal muscle (n = 9). (B) Effect of voluntary wheel running (ET) on gastrocnemius Rac1 protein content in mice. Mouse study: n = 19–22, color of dots indicate two separate studies. Significant differences after training intervention are indicated; ∗ = p < 0.05. Data are presented as mean + SEM incl. individual values, connected when applicable.

Fig. 5

Rac1 mediates critical adaptations to exercise training. (A) Lack of skeletal muscle Rac1 during exercise training (ET) was investigated using inducible skeletal muscle-specific Rac1 knockout mice (Rac1 imKO, compared to control littermate mice (Con). (B) Running distance during the intervention. Effect of exercise training on (C) exercise capacity. (D) change in exercise capacity relative to sedentary (Sed) mice, (E) muscle glycogen, (F) body mass and composition. (G) muscle weight. (H) Effect of ET on protein content of proteins involved in muscle metabolism including representative blots. For control mice: n = 8–13. For Rac1 imKO mice: n = 5–13. Significant effects of training are indicated; ∗/∗∗/∗∗∗ = p < 0.05/p < 0.01/p < 0.001. Significant effects of Rac1 imKO compared to controls are indicated as; # = p < 0.05, # # = p < 0.01, # # # = p < 0.001. Data are presented as mean +SEM incl. individual values. (I) Schematic illustration of whole proteomic analyses of mouse muscle from control (Con) and skeletal muscle-specific Rac1 knockout mice (Rac1 imKO). (J) Volcano plot of the main effect of exercise training (ET) in mouse muscle. (K) Volcano plot and (L) ClueGo enrichment analysis of muscle from trained mice. For the ClueGo analyses: Right-sided hypergeometric test, Bonferroni FDR <0.05 for the proteins regulated >0.2 fold change, functionally related terms are grouped, and color coded based on overlapping proteins. The most significantly regulated terms in each group are labeled. Sedentary mice: n = 5, trained mice: n = 4–5.

Fig. 6

H-Rac1 increases exercise-induced ROS production and enriches the proteome in proteins related to the cytoskeleton organization, large ribosomal subunits, and mitochondria. (A) Constitutively active G12V Rac1 (H-Rac1) was overexpressed in muscle using recombinant adeno-associated virus. (B) Experimental design. (C) immunoblotting validation of H-Rac1 overexpression. (D) DCF-oxidation (oxidant production) at rest and during exercise in tibialis anterior muscle. (E) Effect of 8 weeks of H-Rac1 overexpression on muscle protein content measured by immunoblotting in gastrocnemius muscle. (F) Schematic presentation of proteomic analyses (n = 6). Volcano plot of the muscle proteome in muscle overexpressing H-Rac1. (G) ClueGo enrichment analysis of GO terms of the proteins elevated >0.2 Fold Change by H-Rac1. Right-sided hypergeometric test, Bonferroni FDR <0.05, functionally related terms are grouped, and color coded based on overlapping proteins. The most significantly regulated terms in each group are labeled. (H) Effect of H-Rac1 on muscle weight. For H-Rac1 studies: n = 4–10. Significant differences of acute exercise are indicated as; ∗ = p < 0.05. Significant differences of H-Rac1 overexpression are indicated as; # = p < 0.05, # # = p < 0.01, # # # = p < 0.001. Data are presented as mean +SEM incl. individual values, paired samples connected with lines when applicable. LC-MS/MS, liquid chromatography–tandem mass spectrometry.