The transcriptional coactivator PGC-1alpha mediates exercise-induced angiogenesis in skeletal muscle

Abstract

Peripheral arterial disease (PAD) affects 5 million people in the US and is the primary cause of limb amputations. Exercise remains the single best intervention for PAD, in part thought to be mediated by increases in capillary density. How exercise triggers angiogenesis is not known. PPARgamma coactivator (PGC)-1alpha is a potent transcriptional co-activator that regulates oxidative metabolism in a variety of tissues. We show here that PGC-1alpha mediates exercise-induced angiogenesis. Voluntary exercise induced robust angiogenesis in mouse skeletal muscle. Mice lacking PGC-1alpha in skeletal muscle failed to increase capillary density in response to exercise. Exercise strongly induced expression of PGC-1alpha from an alternate promoter. The induction of PGC-1alpha depended on beta-adrenergic signaling. beta-adrenergic stimulation also induced a broad program of angiogenic factors, including vascular endothelial growth factor (VEGF). This induction required PGC-1alpha. The orphan nuclear receptor ERRalpha mediated the induction of VEGF by PGC-1alpha, and mice lacking ERRalpha also failed to increase vascular density after exercise. These data demonstrate that beta-adrenergic stimulation of a PGC-1alpha/ERRalpha/VEGF axis mediates exercise-induced angiogenesis in skeletal muscle.

Fig. 1.

Exercise-induced angiogenesis requires PGC-1α. (A) PGC-1α −/− (total-body deletion) mice fail to run on in-cage running wheels. A sample tracing of wheel activity, in revolutions per minute, is shown for both WT and PGC-1α −/− mice. (B) Mice lacking PGC-1α specifically in skeletal muscle (MKO mice) do run on in-cage running wheels. Right, sample tracings of wheel activity. Left, average distance run. n = 5 per group. (C) Capillary density from wild-type and PGC-1α MKO mice, either after 14 days of voluntary running, or sedentary controls. Left, representative immunostains for CD31 (endothelial-specific PECAM) in green. Insets show immunostains for laminin of same section, highlighting muscle fiber outlines. Right, quantification of microvascular density. n = 5 per group. Data are presented as mean ± SEM. *, P < 0.05 vs. control. †, P < 0.05 vs. WT exercised.

Fig. 2.

β-adrenergic signaling and exercise induce expression of PGC-1α from an alternative promoter. (A) Schema of PGC-1α alternative and proximal promoters. See text for details. Homology between mammalian species (rat, human, dog, horse, monkey, chicken) is indicated. (B) Tissue distribution of total PGC-1α mRNA (Top) and mRNA initiated at the alternative promoter (Bottom), as determined by qPCR. (C) Relative expression in quadriceps of total PGC-1α mRNA (Left), PGC-1α mRNA originating at the proximal promoter (Middle), or the alternative promoter (Right), after running on voluntary wheels for the indicated time. (D) Relative expression of total PGC-1α mRNA (Left), PGC-1α mRNA originating at the proximal promoter (Middle), or the alternative promoter (Right), 6 h after i.p. injection of clenbuterol (200 μg/kg). (E) Anti-PGC-1α Western blot analysis of quadriceps extracts from mice after 16 h of voluntary running (+) vs. control (−). NS, nonspecific band. (F) Anti-PGC-1α Western blot analysis of quadriceps extracts from WT and MKO mice 6 h after PBS (−) or clenbuterol (−) injection. NS, nonspecific band. (G) Expression of alternative PGC-1α in quadriceps, at the indicated time after clenbuterol injection. (H) PGC-1α expression 6 h after clenbuterol or saline injection in wild-type or PGC-1α MKO mice. Exons 3–5 are deleted in the MKO mice, while both the 3′ UTR and the alternative promoter remain intact. n = 3 per group for A–H. Data are presented as mean ± SEM. *, P < 0.05 vs. control. †, P < 0.05 vs. WT clenbuterol treated.

Fig. 3.

β-adrenergic signaling activates the alternative PGC-1α promoter, via activation of a conserved cAMP responsive element. (A) Luciferase activity in muscles electroporated with alternative (Left) or proximal (Right) PGC-1α promoter-luciferase constructs, 6 h after i.p. injection with clenbuterol (1 mg/kg) or saline control. n = 6 muscles per group. (B) Luciferase activity in muscles electroporated with the indicated deletion of the alternative PGC-1α promoter-luciferase construct, 6 h after injection with clenbuterol or control. n = 6 muscles per group. Homology among mammals, and CRE sequences (in red), are shown at Bottom. Data are presented as mean ± SEM. *, P < 0.05 vs. control. †, P < 0.05 vs. intact promoter clenbuterol treated.

Fig. 4.

β-adrenergic stimulation induces an angiogenic program in skeletal muscle via PGC-1α. (A) mRNA expression of the indicated angiogenic factors in quadriceps of wild-type mice 6 h after injection of clenbuterol (1 mg/kg). (B) mRNA expression of the same factors as in A in quadriceps of wild-type (WT) and MCK-PGC-1α (TG) mice. (C) Total PGC-1α (Left) and VEGF (Right) mRNA expression in quadriceps of mice after 16 h of voluntary running, injected with either propranolol or PBS (as indicated) 30 min before initiating the running. (D) Relative mRNA expression of VEGF (Left) and PDK4 (Right), in quadriceps of wild-type (WT) and PGC-1α −/− mice, 6 h after PBS (gray) or clenbuterol (green) injection. (E) As in D, with PGC-1α MKO mice. (F) Affymetrix microarray analysis of 2 WT and 2 −/− animals in D. Left, all genes with present calls and induced >2-fold by clenbuterol injection are shown. Red and blue indicate elevated and reduced expression, respectively. Right, representative genes. n = 3 per group for all except E. Data are presented as mean ± SEM. *, P < 0.05 vs. control. †, P < 0.05 vs. WT clenbuterol treated. ††, P < 0.05 vs. 16 h run.

Fig. 5.

Exercise-induced angiogenesis requires ERRα. (A) Relative mRNA expression of VEGF (Top) and PDGF-B (Bottom), in differentiated primary skeletal myocytes isolated from wild-type (ERR +/+) or ERR −/− mice, 48 h after infection with adenovirus expressing PGC-1α or GFP control. n = 3 per group. (B) Relative mRNA expression of VEGF (and alternative PGC-1α: Inset) in quadriceps of wild-type (WT) and ERRα −/− mice, 6 h after PBS (gray) or clenbuterol (red) injection. (C) Capillary density, as determined in Fig. 1, from wild-type and ERRα −/− mice, either after 14 days of voluntary running, or sedentary controls. Top, representative immunostains. Bottom, quantification of capillaries/HPF (Left) and capillaries/fiber (Right). n = 5 per group. (D) Proposed model for part of the mechanism underpinning exercise-induced angiogenesis. See text for details. All data are presented as mean ± SEM. *, P < 0.05 vs. control. †, P < 0.05 vs. WT cells + PGC-1α. ††, P < 0.05 vs. WT exercised mice.