Increased mitochondrial ROS generation mediates the loss of the anti-contractile effects of perivascular adipose tissue in high-fat diet obese mice

Abstract

Background and purpose: Obesity is associated with structural and functional changes in perivascular adipose tissue (PVAT), favouring release of reactive oxygen species (ROS), vasoconstrictor and proinflammatory factors. The cytokine TNF-α induces vascular dysfunction and is produced by PVAT. We tested the hypothesis that obesity-associated PVAT dysfunction was mediated by augmented mitochondrial ROS (mROS) generation due to increased TNF-α production in this tissue.

Experimental approach: C57Bl/6J and TNF-α receptor-deficient mice received control or high fat diet (HFD) for 18 weeks. We used pharmacological tools to determine the participation of mROS in PVAT dysfunction. Superoxide anion (O₂^.- ) and H₂ O₂ were assayed in PVAT and aortic rings were used to assess vascular function.

Key results: Aortae from HFD-fed obese mice displayed increased contractions to phenylephrine and loss of PVAT anti-contractile effect. Inactivation of O₂^.- , dismutation of mitochondria-derived H₂ O₂ , uncoupling of oxidative phosphorylation and Rho kinase inhibition, decreased phenylephrine-induced contractions in aortae with PVAT from HFD-fed mice. O₂^.- and H₂ O₂ were increased in PVAT from HFD-fed mice. Mitochondrial respiration analysis revealed decreased O₂ consumption rates in PVAT from HFD-fed mice. TNF-α inhibition reduced H₂ O₂ levels in PVAT from HFD-fed mice. PVAT dysfunction, i.e. increased contraction to phenylephrine in PVAT-intact aortae, was not observed in HFD-obese mice lacking TNF-α receptors. Generation of H₂ O₂ was prevented in PVAT from TNF-α receptor deficient obese mice.

Conclusion and implications: TNF-α-induced mitochondrial oxidative stress is a key and novel mechanism involved in obesity-associated PVAT dysfunction. These findings elucidate molecular mechanisms whereby oxidative stress in PVAT could affect vascular function.

Linked articles: This article is part of a themed section on Molecular Mechanisms Regulating Perivascular Adipose Tissue - Potential Pharmacological Targets? To view the other articles in this section visit http://onlinelibrary.wiley.com/doi/10.1111/bph.v174.20/issuetoc.

Figure 1

Increased superoxide anion generation contributes to the loss of the anti‐contractile effects of PVAT in HFD obesity. Concentration–effect curves to phenylephrine (PE) were performed in endothelium‐intact [(A) n = 6 for each experimental group] and endothelium‐denuded [(B) n = 7 for each experimental group] aortic rings with or without PVAT. The role of ROS on PVAT modulation of aortic smooth muscle contraction was investigated using tiron (10⁻⁴ M), an O₂ ^.‐ scavenger in endothelium‐denuded aortic rings from control [(C) n = 7 for each experimental group] and obese [(D) n = 7 for each experimental group] mice. ROS generation in PVAT was measured ( E) by DHE and (F) by lucigenin (n = 8 in all groups) in control and obese mice. Results represent the mean ± SEM. ^* P < 0.05 versus Control PVAT (−); ^# P < 0.05 versus Control PVAT (+); ^& P < 0.05 versus Obese PVAT (+); ^π P < 0.05 versus Control.

Figure 2

Mitochondria are a potential source of increased ROS in PVAT from obese mice, and mitochondria‐derived ROS alter the expression of antioxidant enzymes in PVAT. Concentration–effect curves to PE were performed in endothelium‐denuded aortic rings from control [(A and C) n = 8 for each experimental group] and obese [(B and D) n = 8 for each experimental group] mice. The role of mROS on PVAT modulation of aortic smooth muscle contraction was investigated using MnTMPyP (3 × 10⁻⁵ M), a mitochondria‐targeted superoxide scavenger and PEG‐catalase (Peg‐cat; 200 U·mL⁻¹), which dismutates mitochondria‐derived H₂O₂. Protein expression and activity of the antioxidant enzymes SOD‐Mn [(E and G) n = 5 in both groups] and catalase [(F and H) n = 5 in both groups] were determined by Western blot and SOD‐Mn and catalase activity assay kits, respectively, in PVAT from control and obese mice. Representative Western blots are shown in the upper panels, with quantitative analysis in the lower panels. Results were normalized to β‐actin expression and are expressed as relative units. Results represent the mean ± SEM. ^* P < 0.05 versus Control PVAT (+); ^# P < 0.05 versus Obese PVAT (+); ^& P < 0.05 versus Obese PVAT (−); ^π P < 0.05 versus Control.

Figure 3

HFD obesity increases mitochondria‐derived O₂ ^.‐ conversion to H₂O₂, which contributes to the loss of the anti‐contractile effects of PVAT. Mitochondrial oxygen consumption (A) and (B) H₂O₂ production (n = 9 in all groups) in PVAT from the thoracic aorta of control and obese animals in the presence of ROUTINE and increasing concentrations of CCCP, a protonophore and mitochondrial uncoupler. Concentration–effect curves to phenylephrine were performed in endothelium‐denuded aortic rings from control (C; n = 6 for each experimental group) and obese (D; n = 7 for each experimental group) mice. The role of mROS generation on PVAT modulation of aortic smooth muscle contraction was investigated using CCCP (10⁻⁶ M). Results represent the mean ± SEM. ^* P < 0.05 versus respective Control; ^≠ P < 0.05 versus respective ROUTINE; ^# P < 0.05 versus Control PVAT (+); ^& P < 0.05 versus Obese PVAT (+).

Figure 4

Increased mROS in PVAT from obese mice induces RhoA/Rho kinase activation in vascular smooth muscle cells. Concentration–effect curves to phenylephrine were performed in endothelium‐denuded aortic rings from control [(A) n = 7 for each experimental group] and obese [(B) n = 8 for each experimental group] mice. The role of RhoA/Rho kinase pathway on PVAT modulation of aortic smooth muscle contraction was investigated using Y27632 (10⁻⁴ M). Concentration–effect curves to Y27632 (C) were performed in endothelium‐denuded aortic rings with or without PVAT (n = 8 for each experimental group); Y27632‐induced relaxation was evaluated in phenylephrine (PE)‐contracted vessels. In (D), the Rho kinase activity was determined by elisa in aortas from control and obese mice (n = 5 in both groups). Representative Western blots are shown in the upper panels, with quantitative analysis in the lower panels [(E and F) n = 5 for each experimental group]. Results were normalized to β‐actin expression and are expressed as relative units. Representative images were selected from the same membrane. Results represent the mean ± SEM. ^* P < 0.05 versus Control PVAT (−); ^π P < 0.05 versus Control PVAT (+); ^# P < 0.05 versus Obese PVAT (−); ^& P < 0.05 versus Obese PVAT (+).

Figure 5

PVAT‐derived TNF‐α mediates increased mROS generation and is a critical mediator of the loss of the anti‐contractile effects of PVAT in arteries from HFD‐fed obese mice. Production of H₂O₂ by PVAT (A) was examined in the presence of vehicle, CCCP, infliximab and TNF‐α, using Amplex Red reagent (n = 7 in all groups). Concentration–effect curves to phenylephrine were performed in aortic rings from control mice incubated with TNF‐α (5 ng·mL⁻¹) [(B) n = 8 for each experimental group]. TNF‐α mRNA expression was assessed by real‐time PCR [(C) n = 6 in all groups]. Concentration–effect curves to phenylephrine (PE) were performed in aortic rings from control [(D) n = 8 for each experimental group] and obese [(E) n = 8 for each experimental group] TNF‐α receptor‐deficient mice. Production of H₂O₂ by PVAT [(F) n = 7 in all groups]. Rho kinase activity was determined by elisa in aortas from wild type and TNF receptor deficient mice [(G) n = 5 in both groups]. UCP‐1 protein expression in PVAT from HFD‐treated mice [(H) n = 6 in both groups] and PVAT from control mice incubated with TNF‐α [(I) n = 6 in both groups] was determined by Western blot. Results represent the mean ± SEM. Representative Western blots are shown in the upper panels, with quantitative analysis in the lower panels. Results were normalized to β‐actin expression and are expressed as relative units. ^* P < 0.05 versus Control; ^# P < 0.05 versus Obese; ^& P < 0.05 versus Control_TNF‐α; ^$ P < 0.05 versus Control PVAT (−); ^% P < 0,.05 versus Control PVAT (+); ^π P < 0.05 versus Obese PVAT (−); ^∞ P < 0.05 versus Obese PVAT (+).