AMPK Suppresses Vascular Inflammation In Vivo by Inhibiting Signal Transducer and Activator of Transcription-1

Abstract

Activation of AMPK suppresses inflammation, but the underlying mechanisms remain poorly understood. This study was designed to characterize the molecular mechanisms by which AMPK suppresses vascular inflammation. In cultured human aortic smooth muscle cells, pharmacologic or genetic activation of AMPK inhibited the signal transducer and activator of transcription-1 (STAT1), while inhibition of AMPK had opposite effects. Deletion of AMPKα1 or AMPKα2 resulted in activation of STAT1 and in increases in proinflammatory mediators, both of which were attenuated by administration of STAT1 small interfering RNA or fludarabine, a selective STAT1 inhibitor. Moreover, AMPK activation attenuated the proinflammatory actions induced by STAT1 activators such as interferon-γ and angiotensin II (AngII). Mechanistically, we found that AMPK activation increased, whereas AMPK inhibition decreased, the levels of mitogen-activated protein kinase phosphatase-1 (MKP-1), an inducible nuclear phosphatase, by regulating proteasome-dependent degradation of MKP-1. Gene silencing of MKP-1 increased STAT1 phosphorylation and prevented 5-aminoimidazole-4-carboxyamide ribonucleoside-reduced STAT1 phosphorylation. Finally, we found that infusion of AngII caused a more severe inflammatory response in AMPKα2 knockout mouse aortas, all of which were suppressed by chronic administration of fludarabine. We conclude that AMPK activation suppresses STAT1 signaling and inhibits vascular inflammation through the upregulation of MKP-1.

Figure 1

AMPK activation inhibits STAT1 signaling in HASMCs. A: Confluent HASMCs were treated with AICAR (2 mmol/L) at the indicated time points. Phosphorylated AMPK at Thr172 (P-AMPK) and phosphorylated STAT1 at Tyr701 (P-STAT1) in cell lysates were analyzed by Western blotting. B: HASMCs were treated with varying concentrations of AICAR for 16 h. Cell lysates were subjected to Western analysis of P-AMPK and P-STAT1. C: Phosphorylation of AMPK and STAT1 was detected in HASMCs treated with indicated concentrations of compound C for 16 h. D: STAT1 and AMPK phosphorylation in response to compound C (20 μmol/L) was measured by Western blotting (n = 4). *P < 0.05 vs. control (Con). E: Immunofluorescence staining of P-STAT1 in HASMCs treated with or without compound C. The photographs are representative of three independent experiments. F: HASMCs were treated with AICAR (2 mmol/L) or compound C (20 μmol/L) for 16 h, and DNA-binding activity of STAT1 was determined by EMSA. G and H: Densitometric analysis of STAT1 DNA-binding activity (n = 4). *P < 0.05 vs. control.

Figure 2

AICAR treatment attenuates IFN-γ–enhanced STAT1 activity and suppresses AngII-induced proinflammatory cytokine expression. A: HASMCs were pretreated with AICAR (2 mmol/L) for 16 h and then treated with IFN-γ (100 ng/mL) for 30 min. Phosphorylation of AMPK (P-AMPK) and STAT1 (P-STAT1) was examined by Western blotting (n = 3). *P < 0.05 vs. control; †P < 0.05 vs. IFN-γ. B: Subcellular distribution of P-STAT1 was determined by immunocytochemistry. The photographs are representative of three independent experiments. C: After being treated with AICAR for 16 h, HASMCs were treated with or without AngII (1 μmol/L) for 24 h. AMPK and STAT1 phosphorylation were measured by Western blotting. D: Subcellular distribution of P-STAT1 was determined by immunocytochemistry. The photograps are representative of three independent experiments. E and F: HASMCs were treated with or without AngII (1 μmol/L) for 24 h after being treated with AICAR for 16 h. IL-6 and MCP-1 in conditioned media were measured by ELISA (n = 4).*P < 0.05 vs. control (Con); †P < 0.05 vs. AngII.

Figure 3

MKP-1 mediates suppression of STAT1 by AMPK. A: Confluent HASMCs were treated with AICAR (2 mmol/L) for 4 h or compound C (Comp C) for 8 h. AMPK phosphorylation (P-AMPK) and MKP-1 protein levels were detected by Western blotting (n = 5). *P < 0.05 AICAR vs. control (Con). B: HASMCs were transfected with GFP, AMPK-CA, or AMPK-DN adenovirus for 48 h. STAT1 phosphorylation (P-STAT1) and MKP-1 protein levels were detected by Western blotting (n = 4).*P < 0.05 vs. GFP; †P < 0.05 vs. AMPK-CA. C: WT and AMPK-deficient MASMCs were transfected with adenovirus encoding GFP (Ad-GFP) or MKP-1 (Ad-MKP-1) for 48 h. MKP-1 protein levels and STAT1 phosphorylation were determined by Western blotting (n = 3).*P < 0.05 vs. WT/GFP; †P < 0.05 AMPKα1^−/− and AMPKα2^−/− plus Ad-GFP vs. AMPKα1^−/− and AMPKα2^−/− plus Ad-MKP-1. D: MKP-1 protein levels and STAT1 and P-STAT1 were analyzed by Western blotting in HASMCs transfected with control (C)-siRNA or MKP-1 siRNA (n = 3). *P < 0.05 vs. C-siRNA. E: HASMCs were transfected with C-siRNA or MKP-1 siRNA for 48 h and then treated with AICAR (2 mmol/L) for 4 h. Cell lysates were subjected to Western analysis of STAT1 phosphorylation (n = 3). *P < 0.05 vs. C-siRNA; †P < 0.05 vs. AICAR/C-siRNA. F: HASMCs were treated with compound C (20 μmol/L) for the indicated time points. MKP-1 protein levels in cell lysates and P-STAT1 in nuclear (n) fractions were analyzed by Western blotting. Histone H2AX was used as a loading control for nuclear fractions.

Figure 4

Deletion of AMPK reduces MKP-1 protein levels by increasing proteasome-mediated degradation of MKP-1. A: HASMCs were treated with AICAR (2 mmol/L) for 4 h. The activity of 26S proteasome in cell lysates was assayed as described in research design and methods. B: MKP-1 protein levels were detected by Western blotting (n = 5). *P < 0.05 vs. control (Con). C: MKP-1 mRNA was measured by quantitative real-time PCR (n = 5). D: WT and AMPK-deficient MASMCs were treated with or without MG-132 (0.5 μmol/L) for 4 h. The activity of 26S proteasome in cell lysates was detected (n = 4). *P < 0.05 vs. WT control; †P < 0.05 vs. AMPKα1^−/−; ‡P < 0.05 vs. AMPKα2^−/−. E: MKP-1 protein levels were determined by Western blotting (n = 4). *P < 0.05 vs. WT control; †P < 0.05 vs. AMPKα1^−/−; ‡P < 0.05 vs. AMPKα2^−/−.

Figure 5

Deletion of AMPK increases STAT1 phosphorylation, nuclear translocation, and DNA-binding activity in MASMCs. A: MASMCs were isolated from WT, AMPKα1^−/−, or AMPKα2^−/− mice, and cell lysates were subjected to Western blot to determine the phosphorylation of STAT1 (P-STAT1) (n = 4). *P < 0.05 vs. WT. B: Subcellular distribution of P-STAT1 was determined by immunocytochemistry. The photographs are representative of three independent experiments. C: DNA-binding activity of STAT1 was measured by EMSA. D: Densitometric analysis of STAT1 DNA-binding activity (n = 4). *P < 0.05 vs. WT. E: Anti-STAT1 antibody was added to the reaction mixtures to test the specificity of the interaction.

Figure 6

STAT1 inhibition attenuates expression of inflammatory mediators in AMPK-deficient MASMCs. A: MASMCs were transfected with control (C)-siRNA or STAT1 siRNA for 48 h. STAT1 phosphorylation (P-STAT1) was evaluated by Western blotting. B: Protein levels of iNOS and COX-2 were analyzed by Western blotting (n = 3). *P < 0.05 vs. WT; †P < 0.05 AMPKα1^−/− and AMPKα2^−/− vs. AMPKα1^−/− and AMPKα2^−/− plus STAT1 siRNA. C: MASMCs were treated with fludarabine (50 μmol/L) for 16 h, and P-STAT1 was determined by Western blotting. D: Western blot analysis was used to determine the expression of iNOS and COX-2 in cell lysates (n = 4). *P < 0.05 vs. WT; †P < 0.05 AMPKα1^−/− and AMPKα2^−/− vs. AMPKα1^−/− and AMPKα2^−/− plus fludarabine. IL-6 (E) and MCP-1 protein (F) levels in conditioned media from WT and AMPK-deficient MASMCs were determined by ELISA (n = 3). *P < 0.05 vs. WT; †P < 0.05 AMPKα1^−/− and AMPKα2^−/− vs. AMPKα1^−/− and AMPKα2^−/− plus fludarabine.

Figure 7

Deletion of AMPKα2 reduces MKP-1 protein levels and enhances STAT1 phosphorylation in mouse aortas. A: Protein levels of MKP-1 in mouse aortas were measured by Western blotting. B: Aorta sections were stained with MKP-1 antibody, and the stained area was quantified. The photomicrographs are representative of five independent mouse aortas. C: STAT1 phosphorylation (P-STAT1) in mouse aortas was measured by Western blotting. D: Aorta sections were stained with P-STAT1 antibody, and the stained areas were quantified and expressed as a percentage of total tissue area. ♣P < 0.05 vs. WT control; †P < 0.05 vs. WT/AngII; *P < 0.05 vs. AMPKα2^−/−/control; #P < 0.05 AMPKα2^−/−/AngII vs. WT/AngII; ‡P < 0.05 vs. AMPKα2^−/−/AngII. Flud, fludarabine.

Figure 8

Inhibition of STAT1 by fludarabine (Flud) attenuates AngII-enhanced inflammatory response in WT and AMPKα2^−/− mice. A: Aorta sections were stained with CD68 antibody and the stained areas were quantified and expressed as a percentage of total tissue area. Expression of VCAM-1 (B), MCP-1 (C), iNOS (D), TNF-α (E), and IFN-γ (F) was analyzed by using immunohistochemistry. The photomicrographs are representative of five independent mouse aortas. Total RNA were extracted from mouse aortas, and mRNA levels of VCAM-1 (B), MCP-1 (C), iNOS (D), TNF-α (E), and IFN-γ (F) were determined by quantitative real-time PCR. ♣P < 0.05 vs. WT control; †P < 0.05 vs. WT/AngII; *P < 0.05 vs. AMPKα2^−/−/control; #P < 0.05 AMPKα2^−/−/AngII vs. WT/AngII; ‡P < 0.05 vs. AMPKα2^−/−/AngII.