Regulation of skeletal muscle oxidative capacity and muscle mass by SIRT3

Abstract

We have previously reported that the expression of mitochondrial deacetylase SIRT3 is high in the slow oxidative muscle and that the expression of muscle SIRT3 level is increased by dietary restriction or exercise training. To explore the function of SIRT3 in skeletal muscle, we report here the establishment of a transgenic mouse model with muscle-specific expression of the murine SIRT3 short isoform (SIRT3M3). Calorimetry study revealed that the transgenic mice had increased energy expenditure and lower respiratory exchange rate (RER), indicating a shift towards lipid oxidation for fuel usage, compared to control mice. The transgenic mice exhibited better exercise performance on treadmills, running 45% further than control animals. Moreover, the transgenic mice displayed higher proportion of slow oxidative muscle fibers, with increased muscle AMPK activation and PPARδ expression, both of which are known regulators promoting type I muscle fiber specification. Surprisingly, transgenic expression of SIRT3M3 reduced muscle mass up to 30%, likely through an up-regulation of FOXO1 transcription factor and its downstream atrophy gene MuRF-1. In summary, these results suggest that SIRT3 regulates the formation of oxidative muscle fiber, improves muscle metabolic function, and reduces muscle mass, changes that mimic the effects of caloric restriction.

Figure 1. Creation of muscle-specific SIRT3 transgenic mice.

(A): Diagram of the transgene construct. The SIRT3-M3-FLAG transgene was under the control of the 6.5kb muscle creatine kinase (MCK) promoter/enhancer with the human growth hormone polyadenylation site at the 3′ end. (B): The mRNA expression of the SIRT3-M3-FLAG transgene in heart, quadriceps muscle and white adipose tissue, was measured using real-time RT-PCR. The results were normalized with cyclophilin expression and presented as relative to WT controls. n = 5. (C): The SIRT3M3-FLAG transgene product was detected in the quadriceps muscle lysates by Western blot analysis using anti-SIRT3 or anti-FLAG antibodies. (D): Body weight of male WT and MCK-SIRT3M3 mice. n = 6–9. (E): Body weight of female WT and MCK-SIRT3M3 mice. n = 5–9. **P<0.01 between WT and MCK-SIRT3M3 mice.

Figure 2. Metabolic characterization of MCK-SIRT3M3 transgenic mice.

(A): Daily food intake of 6-month old WT and MCK-SIRT3M3 mice. (B): Total locomotor activity at daytime and nighttime of 6-month old male WT and MCK-SIRT3M3 mice. (C): Total locomotor activity at daytime and nighttime of 6-month old female WT and MCK-SIRT3M3 mice. (D): Correlation of energy expenditure and lean body mass, for female WT and MCK-SIRT3M3 mice. (E and F): Respiratory exchange rate (RER) of WT and MCK-SIRT3M3 mice. n = 6. *P<0.05, ***P<0.001 between WT and MCK-SIRT3M3 mice.

Figure 3. Oxidative capacity and muscle strength of MCK-SIRT3M3 transgenic mice.

(A): Running distance of WT and MCK-SIRT3M3 mice on treadmill. (B): Work performed of WT and MCK-SIRT3M3 mice on treadmill. (C, D and E): Oxygen consumption, heat production, and respiratory exchange rate (RER) of male WT and MCK-SIRT3M3 mice during treadmill. n = 5–7. (F): Holding time of WT and MCK-SIRT3M3 mice on inverted grid mesh test. (G): Transition time of WT and MCK-SIRT3M3 mice climbing on string test. n = 6–10. *P<0.05, **P<0.01 between WT and MCK-SIRT3M3 mice.

Figure 4. Fiber type characterization of muscle from MCK-SIRT3M3 transgenic mice.

(A): Protein levels of three different myosin heavy chains (MHC) in quadriceps muscle, evaluating by Western blotting. The relative levels of MHC proteins were normalized by tubulin levels. (B): ATPase staining of quadriceps muscle.

Figure 5. The analysis of proteins responsible for muscle fiber determination in MCK-SIRT3M3 transgenic mice.

(A): Phosphorylation of AMPK and ACC in quadriceps muscle of WT and MCK-SIRT3M3 mice. (B): Protein levels of PPARγ and PPARδ in quadriceps muscle of WT and MCK-SIRT3M3 mice.

Figure 6. Mitochondrial respiration rates and citrate synthase activity in muscles of MCK-SIRT3M3 transgenic mice.

(A): Oxygen consumption rates in isolated mitochondria from muscles of WT and MCK-SIRT3M3 mice at 5 months of age (n = 6). Respiration parameters were recorded using an Oroboros O2k oxygraph. Resting respiration (state 4) and maximal ADP-stimulated respiration (state 3) were presented. (B): Respiratory control ratio (RCR) was calculated as the ratio of oxygen consumption at state 3 over oxygen consumption at state 4. (C): Citrate synthase (CS) activity in gastrocnemius muscle extracts from WT and MCK-SIRT3 transgenic mice at 5 months of age (n = 6). CS activity was measured according to Srere . **P<0.01 between WT and MCK-SIRT3M3 mice.

Figure 7. Transgenic expression of SIRT3M3 decreased skeletal muscle mass.

(A): Tibia length of WT and MCK-SIRT3M3 mice. (B): Comparison of representative samples of dissected skeletal muscle (Quad, quadriceps; EDL, extensor digitorum longus; TA, tibialis anterior; Gastroc, gastrocnemius) between MCK-SIRT3M3 mice and litter-mate control mice. (C and D): Muscle weights from 6–8 m old WT and MCK-SIRT3M3 mice, for male and female. n = 6–8. **P<0.01, ***P<0.001 between WT and MCK-SIRT3M3 mice.

Figure 8. Fiber size of muscle from MCK-SIRT3M3 transgenic mice.

(A): H&E staining of quadriceps and gastrocnemius muscle from 3–4 month-old WT and MCK-SIRT3M3 mice. (B): Fiber cross-section area of quadriceps and gastrocnemius muscle from 4.5 m old male WT and MCK-SIRT3M3 mice. n = 3–5. *P<0.05, **P<0.01 between WT and MCK-SIRT3M3 mice.

Figure 9. MCK-SIRT3M3 mice had increased muscle FOXO1 expression.

(A): Total and phosphorylated FOXO1 protein level in quadriceps muscle from WT and MCK-SIRT3M3 mice. (B): Total FOXO1 protein level in nuclear and cytosol fraction of quadriceps muscle from WT and MCK-SIRT3M3 mice. (C): q-PCR analyses of MuRF-1 and atrogen-1 in quadriceps muscle from WT and MCK-SIRT3M3 mice. n = 6. *P<0.05 between WT and MCK-SIRT3M3 mice.