IGF-1 prevents ANG II-induced skeletal muscle atrophy via Akt- and Foxo-dependent inhibition of the ubiquitin ligase atrogin-1 expression

Abstract

Congestive heart failure is associated with activation of the renin-angiotensin system and skeletal muscle wasting. Angiotensin II (ANG II) has been shown to increase muscle proteolysis and decrease circulating and skeletal muscle IGF-1. We have shown previously that skeletal muscle-specific overexpression of IGF-1 prevents proteolysis and apoptosis induced by ANG II. These findings indicated that downregulation of IGF-1 signaling in skeletal muscle played an important role in the wasting effect of ANG II. However, the signaling pathways and mechanisms whereby IGF-1 prevents ANG II-induced skeletal muscle atrophy are unknown. Here we show ANG II-induced transcriptional regulation of two ubiquitin ligases atrogin-1 and muscle ring finger-1 (MuRF-1) that precedes the reduction of skeletal muscle IGF-1 expression, suggesting that activation of atrogin-1 and MuRF-1 is an initial mechanism leading to skeletal muscle atrophy in response to ANG II. IGF-1 overexpression in skeletal muscle prevented ANG II-induced skeletal muscle wasting and the expression of atrogin-1, but not MuRF-1. Dominant-negative Akt and constitutively active Foxo-1 blocked the ability of IGF-1 to prevent ANG II-mediated upregulation of atrogin-1 and skeletal muscle wasting. Our findings demonstrate that the ability of IGF-1 to prevent ANG II-induced skeletal muscle wasting is mediated via an Akt- and Foxo-1-dependent signaling pathway that results in inhibition of atrogin-1 but not MuRF-1 expression. These data suggest strongly that atrogin-1 plays a critical role in mechanisms of ANG II-induced wasting in vivo.

Fig. 1.

Regulation of gene expression in skeletal muscle of ANG II-infused wild-type and myosin light chain (MLC)/mIgf-1 mice. ANG II or saline minipumps were implanted to FVB or MLC/mIgf-1 mice, and gastrocnemius muscles were collected 1, 4, and 7 days after implantation. Gastrocnemius muscle weight of each animal was measured 7 days after minipump implantation (E), and atrogin-1 (A), muscle ring finger-1 (MuRF-1; B), IGF-1 (C), and IGF-1 receptor (IGF-1R; D) expression was quantified by quantitative RT-PCR on days 1, 4, and 7 of infusion. Gastrocnemius muscles of FVB and MLC/mIgf-1 mice were collected 1 day after ANG II or saline infusion, and atrogin-1 (F) and MuRF-1 (G) expression was measured by quantitative RT-PCR. Means are ± SE; n = 5 to 6. *P < 0.05; **P < 0.01.

Fig. 2.

Efficiency and deleterious effect of plasmid electroporation into mouse gastrocnemius muscle. A: enhanced green fluorescent protein (EGFP) fluorescence after electroporation. EGFP encoding plasmid was electroporated to gastrocnemius muscle, and fluorescence was observed 3, 5, 7, and 14 days after electroporation. Whole gastrocnemius muscle is shown in the picture. Contralateral control gastrocnemius muscle was electroporated with empty vector. B: EGFP expression in gastrocnemius muscle after EGFP encoding plasmid electroporation was assessed by immunoblotting. C: IGF-1 mRNA expression was analyzed by Northern blotting 5 days after electroporation. Endogenous and plasmid-derived IGF-1 transcripts are indicated by white and black arrow, respectively. I, IGF-1 plasmid; C, control plasmid. D: plasma human IGF-1 (hIGF-1) levels were quantified before and after electroporation (EP) of IGF-1 encoding plasmid. Plasma samples collected from mice infused (inf) with 1.5 mg·kg⁻¹·day⁻¹ human IGF-1 were used as control. E and F: atrogin-1 and MuRF-1 expression was quantified 1 day after electroporation by quantitative RT-PCR. Gastrocnemius muscles were collected from empty plasmid vector electroporated (inj + EP), plasmid vector injected without electric pulse (inj), or electric pulse applied without plasmid injection (EP) skeletal muscles. Neither plasmid injection nor electric pulse administration was performed in control muscles. G and H: atrogin-1 and MuRF-1 expression were quantified 1, 3, 7, and 14 days after electroporation. Empty plasmid vector was injected in both of the muscles, and electric pulse was applied only to the EP gastrocnemius muscles. Means are ± SE; n = 5. *P < 0.05; **P < 0.01.

Fig. 3.

Electroporation of IGF-1 to skeletal muscle inhibits ANG II-induced skeletal muscle wasting and atrogin-1 but not MuRF-1 expression. C57BL/6 mice were electroporated with plasmid encoding IGF-1 or control empty vector into each gastrocnemius muscle, followed by 2-wk recovery period (day −14 to day 0). ANG II or saline infusion was started on day 0, and body weight (A) and food intake (B) were measured daily. Saline infused mice were pair fed. Gastrocnemius muscle weight of plasmid electroporated and ANG II or sham infused mice were measured on day 7 of infusion (C). Atrogin-1 (D) and MuRF-1 (E) expression in plasmid electroporated gastrocnemius muscle was quantified by quantitative RT-PCR after 1 day of ANG II infusion. F: activity of 5 and 1 kbp upstream promoter regions of atrogin-1 and MuRF-1 were measured by dual luciferase reporter gene assay. Firefly luciferase activity of gastrocnemius muscle collected 1 day after ANG II minipump implantation was determined, and Renilla luciferase under the control of thymidine kinase was used as an internal electroporation control. Luciferase activity was calculated as the ratio of firefly and Renilla luciferase bioluminescence. Means are ± SE; n = 5. *P < 0.05; **P < 0.01. AU, arbitrary units.

Fig. 4.

IGF-1 inhibition of ANG II muscle wasting and atrogin-1 expression is blocked by dominant-negative form of Akt (dnAkt). A: plasmids encoding wild-type Akt (wtAkt) or dnAkt were electroporated to gastrocnemius muscle, and phosphorylated and total Akt levels were analyzed by immunoblotting 7 days after electroporation. B–D: plasmid encoding IGF-1, together with wtAkt or dnAkt encoding vectors, was electroporated to gastrocnemius muscle, and ANG II or saline minipumps were implanted 2 wk after electroporation. Gastrocnemius muscles were collected 1 or 7 days after infusion. The weight of each plasmid electroporated gastrocnemius muscle (B) was measured 1 wk after minipump implantation. Atrogin-1 (C) and MuRF-1 (D) expression in gastrocnemius muscle was quantified by quantitative RT-PCR 1 day after minipump implantation. Means are ± SE; n = 5. *P < 0.05; **P < 0.01.

Fig. 5.

IGF-1 inhibition of ANG II muscle wasting and atrogin-1 expression is blocked by constitutively active form of Foxo-1 (caFoxo-1). A: plasmids encoding wild-type (wt)Foxo-1 or caFoxo-1 were electroporated to gastrocnemius muscle, and phosphorylated and total Foxo-1 levels were analyzed by immunoblotting 7 days after electroporation. B–D: plasmid encoding IGF-1, together with wtFoxo-1 or caFoxo-1 encoding vectors, was electroporated to gastrocnemius muscle, and ANG II or saline minipumps were implanted 2 wk after electroporation. Gastrocnemius muscles were collected 1 or 7 days after infusion. Weight of each plasmid electroporated gastrocnemius muscle (B) was measured 1 wk after minipump implantation. Atrogin-1 (C) and MuRF-1 (D) expression in gastrocnemius muscle was quantified by quantitative RT-PCR 1 day after minipump implantation. Means are ± SE; n = 5. *P < 0.05; **P < 0.01.

Fig. 6.

Signaling pathways involved in ANG II-induced skeletal muscle atrophy and its prevention by IGF-1. ANG II has been shown to cause dephosphorylation of Akt. Activation of Foxo results in transcriptional upregulation of atrogin-1. IGF-1 activates Akt resulting in phosphorylation and inactivation of Foxo and inhibition of atrogin-1 transcription. Although MuRF-1 is also activated by ANG II in skeletal muscle, its transcription is not affected by IGF-1 or Akt-Foxo signaling.