SIRT1 is required for AMPK activation and the beneficial effects of resveratrol on mitochondrial function

Abstract

Resveratrol induces mitochondrial biogenesis and protects against metabolic decline, but whether SIRT1 mediates these benefits is the subject of debate. To circumvent the developmental defects of germline SIRT1 knockouts, we have developed an inducible system that permits whole-body deletion of SIRT1 in adult mice. Mice treated with a moderate dose of resveratrol showed increased mitochondrial biogenesis and function, AMPK activation, and increased NAD(+) levels in skeletal muscle, whereas SIRT1 knockouts displayed none of these benefits. A mouse overexpressing SIRT1 mimicked these effects. A high dose of resveratrol activated AMPK in a SIRT1-independent manner, demonstrating that resveratrol dosage is a critical factor. Importantly, at both doses of resveratrol no improvements in mitochondrial function were observed in animals lacking SIRT1. Together these data indicate that SIRT1 plays an essential role in the ability of moderate doses of resveratrol to stimulate AMPK and improve mitochondrial function both in vitro and in vivo.

Figure 1. Improved mitochondrial function and increased mitochondrial biogenesis in response to resveratrol treatment requires SIRT1

(A) Mitochondrial membrane potential and (B) ATP content in C2C12 cells treated with 25 μM resveratrol and 10 μM EX-527 for 24h. (C) Representative immunoblot for SIRT1 and tubulin in C2C12 cells infected with SIRT1 or non-targeting shRNA. (D) Mitochondrial membrane potential and (E) ATP content in C2C12 cells infected with SIRT1 or non-targeting shRNA and treated with 25 μM resveratrol for 24h. (F) Mitochondrial DNA content analyzed by means of quantitative PCR in C2C12 cells treated with 10 μM EX-527 or (G) infected with SIRT1 or non-targeting shRNA and treated with 25 μM resveratrol. Relative expression values were normalized to untreated cells. (H) C2C12 cells infected with SIRT1 or non-targeting shRNA, and expressing Flag-HA-PGC-1α were treated with resveratrol 25 μM for 24h and PGC-1α acetylation was tested in Flag immunoprecipitates. Total PGC-1α was evaluated on total extracts as input. (I) PGC-1α, NRF-1, NDUFS8, SDHb, Uqcrc1, COX5b, ATP5a1 mRNA analyzed by means of quantitative RT-PCR in C2C12 cells infected with SIRT1 or non-targeting shRNA after 24h treatment with 25 μM resveratrol. Relative expression values were normalized to untreated cells. Values are expressed as mean ± SEM. (*p < 0.05 versus DMSO).

Figure 2. Generation of adult inducible SIRT1 KO mice revealed that ability of resveratrol to improve mitochondrial function requires SIRT1 in vivo

(A) Schematic representation of induction of SIRT1 KO and treatment with the different diets. (B) Representative immunoblot for SIRT1 and tubulin in skeletal muscle and heart of WT and SIRT1 KO mice. (C) State 3 respiration of isolated mitochondria from skeletal muscle of WT and SIRT1 KO mice on experimental diets (n=8) (*p<0.05 versus WT SD, #p<0.05 versus WT HFD). (D) State 4 respiration of isolated mitochondria from skeletal muscle of WT and SIRT1 KO mice on experimental diets (n=8) (E) FCCP-induced respiration of isolated mitochondria from skeletal muscle of WT and SIRT1 KO mice on experimental diets (n=8) (*p<0.05 versus WT SD, #p<0.05 versus WT HFD). (F) Mitochondrial membrane potential of isolated mitochondria from skeletal muscle of WT and SIRT1 KO mice on experimental diets (n=8) (*p<0.05 versus WT SD, #p<0.05 versus WT HFD). (G) Cellular ATP content from gastrocnemius of WT and SIRT1 KO mice on experimental diets (n=8) (*p<0.05 versus WT SD, #p<0.05 versus WT HFD, +p<0.05 versus WT HFD).

Figure 3. Resveratrol induces a shift toward more oxidative fibers in a SIRT1 dependent manner

(A–B) MyHCI, MyHCIIa, MyHCIIb and MyHCIIx mRNA analyzed by quantitative RT-PCR in gastrocnemius (A) and soleus (B) of WT and SIRT1 KO mice on experimental diets. Relative expression values were normalized to WT SD mice (n=4) *p<0.05 versus WT HFD, #p<0.05 versus WT SD, +p<0.05 versus WT SD). (C) Representative immunoblot for MyHCIIa, MyHCIIb and tubulin in gastrocnemius of WT and SIRT1 KO mice on experimental diets. (D) Representative MyHCIIa immunostaining in gastrocnemius of WT and SIRT1 KO mice on experimental diets. Values are expressed as mean ± SEM.

Figure 4. Resveratrol improves mitochondrial biogenesis in skeletal muscle of WT but not SIRT1 KO mice

(A) Citrate synthase activity measured in gastrocnemius from WT and SIRT1 KO mice on experimental diets (n=8) (*p<0.05 versus WT SD, #p<0.05 versus SIRT1 KO SD, +p<0.05 versus WT HFD). (B) NDUFS8, SDHb, Uqcrc1, COX5b and ATP5a1 mRNA analyzed by quantitative RT-PCR in gastrocnemius of WT and SIRT1 KO mice on experimental diets. Relative expression values were normalized to WT SD mice. (n=4 experiments *p<0.05 versus WT HFD). (C) Mitochondrial DNA content analyzed by quantitative PCR in gastrocnemius of WT and SIRT1 KO mice on experimental diets. Relative expression values were normalized to WT SD mice. (n=8 experiments *p<0.05 versus WT SD, #p<0.05 versus WT HFD). (D) Electronic microscopy analysis of gastrocnemius from WT and SIRT1 KO mice on experimental diets and the respective mitochondrial area quantification (n=4) (*p<0.05 versus WT HFD). (E) PGC-1α, PGC-1β, NRF-1, NRF-2, TFAM, TFB1M and TFB2M mRNA analyzed by means of quantitative RT-PCR in gastrocnemius of WT and SIRT1 KO mice on experimental diets. Relative expression values were normalized to WT SD mice. (n=4) (*p<0.05 versus WT HFD, #p < 0.05 versus WT SD, +p<0.05 versus WT SD). Values are expressed as mean ± SEM.

Figure 5. Mice overexpressing SIRT1 mimic the effects of resveratrol on mitochondrial function and biogenesis

(A) Representative immunoblot for SIRT1 and tubulin in gastrocnemius of 6 months old WT and SIRT1 Tg mice. (B) State 3 respiration of isolated mitochondria from skeletal muscle of WT and SIRT1 Tg mice (n=6). (C) State 4 respiration of isolated mitochondria from skeletal muscle of WT and SIRT1 Tg mice (n=6) (D) FCCP-induced respiration of isolated mitochondria from skeletal muscle of WT and SIRT1 Tg mice (n=6). (E) Mitochondrial membrane potential of isolated mitochondria from skeletal muscle of WT and SIRT1 Tg mice (n=6). (F) Mitochondrial DNA content analyzed by means of quantitative PCR in skeletal muscle of WT and SIRT1 Tg mice. Relative expression values were normalized to WT mice. (n=6). (G) NDUFS8, SDHb, Uqcrc1, COX5b and ATP5a1 mRNA analyzed by quantitative RT-PCR in gastrocnemius of WT and SIRT1 Tg mice. Relative expression values were normalized to WT mice. (n=6). (H) PGC-1α, PGC-1β, NRF-1, NRF-2, TFAM, TFB1M and TFB2M mRNA analyzed by quantitative RT-PCR in gastrocnemius of WT and SIRT1 Tg mice. Relative expression values were normalized to WT mice. (n=6). Values are expressed as mean ± SEM (*p < 0.05 versus WT).

Figure 6. Resveratrol activates AMPK through SIRT1 dependent and independent mechanisms depending on dose

(A) Representative immunoblot for AMPKα and tubulin in C2C12 cells infected with AMPKα or non-targeting shRNA. (B) ATP content in C2C12 cells infected with AMPKα or non-targeting shRNA and treated with 25 μM resveratrol for 24h (n=5) (*p<0.05 versus DMSO). (C) Mitochondrial DNA content analyzed by quantitative PCR in C2C12 cells infected with AMPKα or non-targeting shRNA and treated with 25 μM resveratrol for 24h. Relative expression values were normalized to WT. (n=5) (*p<0.05 versus DMSO). (D) ATP content in C2C12 cells infected with AMPKα or non-targeting shRNA and treated with adenovirus overexpressing SIRT1 or empty vector (n=5) (*p<0.05 versus empty DMSO). (E) Mitochondrial DNA content analyzed by quantitative PCR in C2C12 cells infected with AMPKα or non-targeting shRNA and with adenovirus overexpressing SIRT1 or empty vector. Relative expression values were normalized to control. (n=5) experiments (*p<0.05 versus empty DMSO). (F) Representative immunoblot for p-AMPK (Thr172) and total AMPK in gastrocnemius of WT and SIRT1 KO mice on experimental diets. (G) Quantification of AMPK activity evaluated by the ratio of p-AMPK and AMPK in gastrocnemius of WT and SIRT1 KO mice on experimental diets (n=8). (H) NAD+ content in gastrocnemius of WT and SIRT1 KO mice on experimental diets (n=4) (*p<0.05 versus WT SD, #p<0.05 versus WT HF, +p<0.05 versus SIRT1 KO HF). (J) Representative immunoblot for p-AMPK (Thr172), and total AMPK in gastrocnemius of WT and SIRT1 Tg mice. (K) Quantification of AMPK activity evaluated by the ratio of quantification of p-AMPK and AMPK in gastrocnemius of WT and SIRT1 Tg mice (n=6) (*p<0.05 versus WT). Values are expressed as mean ± SEM.

Figure 7. Resveratrol activates AMPK in a SIRT1 dependent manner through deacetylation of LKB1

(A) Representative immunoblot for p-AMPK (Thr172) and total AMPK in C2C12 cells infected with SIRT1 or non-targeting shRNA and treated with 10, 25 or 50 μM resveratrol for 24h. (B) ATP content in C2C12 cells treated with 25 or 50 μM resveratrol for 24h (n=4) (*p<0.05 versus DMSO). (C) Mitochondrial membrane potential in C2C12 cells treated with 25 or 50 μM resveratrol for 24h (n=4) (*p<0.05 versus DMSO). (D) ATP content in C2C12 cells treated with 25 or 50 μM resveratrol for 1, 4, 6 and 12h (n=4) (*p<0.05 versus DMSO). (E) Representative immunoblot for for p-AMPK (Thr172), and total AMPK in C2C12 cells treated with 25 or 50 μM resveratrol for 1, 4, 6 and 12h. (F) NAD⁺ content in C2C12 cells treated with 25 or 50 μM resveratrol for 1, 4, 6 and 12h (n=4) (*p<0.05 versus 50 μM DMSO, #p<0.05 versus 25 μM DMSO). (G) Representative immunoblot for for p-AMPK (Thr172), and total AMPK in primary myoblasts isolated from wild type and SIRT1 knockout mice and treated with 500 μM AICAR for 24h. (H) Mitochondrial DNA content analyzed by quantitative PCR in primary myoblasts isolated from wild type and SIRT1 knockout mice and treated with 500 μM AICAR for 24h. Relative expression values were normalized to control. (n=3) experiments (*p<0.05 versus empty DMSO). (I) ATP content in primary myoblasts isolated from wild type and SIRT1 knockout mice and treated with 500 μM AICAR for 24h (n=3) (*p<0.05 versus DMSO). (J) C2C12 cells infected with SIRT1 or non-targeting shRNA, and expressing Flag-LKB1 were treated with resveratrol 25 μM for 24h and LKB1 acetylation was tested in Flag immunoprecipitates. Total LKB1 was evaluated in total extracts as input. (K) Moderate doses of resveratrol activate AMPK and stimulate mitochondrial biogenesis in a SIRT1-dependent manner that results in improvement of mitochondrial function. Values are expressed as mean ± SEM.