Moderate Exercise Suppresses NF-κB Signaling and Activates the SIRT1-AMPK-PGC1α Axis to Attenuate Muscle Loss in Diabetic db/db Mice

Abstract

The clear mechanism of moderate exercise training (Ex) in attenuating muscle loss remains elusive in diabetes. We investigated the effects of moderate exercise training on diabetes-induced nuclear factor-κB (NF-κB) activation and mitochondrial dysfunction. Skeletal muscle size and atrophy signaling pathways were examined in type 2 diabetic db/db mice with or without moderate exercise training (5.2 m/min, 1 h/day, and 5 days/week for a total of 8 weeks). Exercise training decreased serum leptin, MCP-1, and resistin levels in db/db+Ex mice, but it did not reduce symptoms of insulin resistance including hyperglycemia, hyperinsulinemia, and impaired glucose tolerance. Moderate exercise training prevented the loss of muscle mass of tibialis anterior and gastrocnemius muscles in db/db+Ex mice. The average cross-sectional area of tibialis anterior muscle was increased significantly in db/db+Ex mice compared with untrained mice (830.6 vs. 676.5 μm²). Inhibition of MuRF-1 and K48-linked polyubiquitination was observed in db/db+Ex mice. Exercise training reduced activation of IκBα/NF-κB pathway and lowered IL-6, TNFα, F4/80 (macrophage marker) at mRNA level in db/db+Ex mice compared with untrained mice. Exercise training did not influence FoxO3a phosphorylation and its upstream regulator Akt. Exercise training increased SIRT1 and PGC1α expression and AMPKα and mitochondrial complex IV activities and upregulated genes involved in mitochondrial biogenesis/function including Nrf1, Tfam, and mitochondrial complexes I-V. In conclusion, moderate exercise training inhibits NFκB signaling and activates SIRT1-AMPKα-PGC1α axis, thereby attenuating type 2 diabetes-related muscle atrophy.

Keywords: MuRF-1; NF-κB signaling; diabetic db/db mice; mitochondrial function/biogenesis; moderate exercise.

FIGURE 1

Effect of moderate exercise training on body weight body, blood glucose, insulin, and glucose tolerance. Body weight (A, n = 11–12/group), fasting glucose (B, n = 6/group), serum insulin (C, n = 11–12/group), IPGTT (D, n = 6/group), and area under the curve (AUC) calculations (E, n = 6/group) in m/m, db/db, and db/db+Ex groups. Values presented are mean ± SEM. Significance (P < 0.05) among groups is denoted by different letters.

FIGURE 2

Effect of moderate exercise training on muscle size, muscle weight, and protein expression. Representative hematoxylin-eosin stained of tibialis anterior muscle (Aa–c) in m/m, db/db, and db/db+Ex groups (200×). Mean cross-sectional area of tibialis anterior muscle (B, n = 4/group). Tibialis anterior (C) and gastrocnemius (D) muscle weight (n = 9–12/group). Representative blots of MuRF-1 and K48-linkage polyubiquitin (E,F, n = 5–8). Values presented are mean ± SEM. Significance (P < 0.05) among groups is denoted by different letters.

FIGURE 3

Effect of moderate exercise training on protein expression, mRNA expression, and proinflammatory cytokines. Representative blots of NF-κB, phospho-NF-κB (Ser563), IκBα, and phospho-IκBα (Ser32), are shown (A,B). The protein levels in gastrocnemius muscles are presented mean ± SEM (n = 5/group). IL-6, TNFα, and F4/80 mRNA expression levels in gastrocnemius muscles are expressed as mean ratio to control after normalization with 18S mRNA levels. Fold differences were calculated using the ΔΔCt method. Values presented are mean ± SEM (C, n = 6–8/group, each in triplicate). Serum levels of IL-6 and TNFα (D, n = 11–12/each group). Representative blots of Akt, phospho-Akt (Thr308), FoxO3a, and phospho-FoxO3a (Ser253) are shown (E,F). The protein levels in gastrocnemius muscles are presented mean ± SEM (n = 5–7/group). Significance (P < 0.05) among groups is denoted by different letters. ^∗P < 0.05 vs. db/db.

FIGURE 4

Effect of moderate exercise training on protein expression, mRNA expression, and complex IV activity. Representative blots of SIRT1, PGC1α, AMPKα, and phospho-AMPKα (Thr172) are shown (A,B). The protein levels in gastrocnemius muscles are presented mean ± SEM (n = 5–7/group). Nrf1, Tfam, and mitochondrial complexes I–V mRNA expression levels in gastrocnemius muscles are expressed as mean ratio to control after normalization with 18S mRNA levels. Fold differences were calculated using the ΔΔCt method. Values presented are mean ± SEM (C,D, n = 6–8/group, each in triplicate). Cytochrome oxidase (Complex IV) activity in gastrocnemius muscle (E, n = 6/group). Significance (P < 0.05) among groups is denoted by different letters. ^∗P < 0.05 vs. db/db.