Role of peroxisome proliferators-activated receptor-gamma in advanced glycation end product-mediated functional loss of voltage-gated potassium channel in rat coronary arteries

Male Sprague-Dawley rats (8-week-old, 180-200 g weight) were provided by Vital River Laboratory Animal Technology (Beijing, China). They were housed under specific pathogen-free conditions and given free access to food and water. All animal experiments were performed based on the National Institutes of Health guidelines for care and use of laboratory animals, and were approved by the Animal Care and Use Committee of Capital Medical University.

Rats were anesthetized with sodium pentobarbital (60 mg/kg i.p.) and were sacrificed when the heart was removed. RSCAs (internal diameter 150-200 μm) were carefully dissected from surrounding myocardium and immediately placed in ice-cold Hanks solution. The arteries were cleared of fat and connective tissue and were cut into 2-mm long rings. The endothelium was denuded by dry air, and the denudation was verified by failure to dilate in response to 1 μM (mol/L) acetylcholine. For isometric force measurement, arteries were directly mounted in the myograph chamber containing HEPES-buffered solution. For protein level’s detection, arteries were incubated in Dulbecco’s modified Eagle’s medium (DMEM) containing 1 g/L glucose, 10% fetal bovine serum (FBS), 100 U/mL penicillin, and 100 mg/mL streptomycin at 37 °C for 24 h.

The effects of ALA (AGE cross-linking breaker), PIO (PPAR-γ activator), and GW9662 (PPAR-γ inhibitor) on the AGE-induced impairment of Kv channel in RSCAs were evaluated. The arteries were divided into the following treatment groups: vehicle (DMEM group), 200 μg/mL BSA (BSA group), 200 μg/mL AGE-BSA (AGE group), 200 μg/mL AGE-BSA + 1 mM ALA (AGE + ALA group), 200 μg/mL AGE-BSA + 0.1 mM PIO (AGE + PIO group), and 200 μg/mL AGE-BSA + 0.1 mM PIO + 0.1 mM GW9662 (AGE + PIO + GW9662 group). The concentrations of AGE and other reagents were based on previous published studies [14, 15, 21].

Freshly isolated arterial rings of RSCAs were threaded onto two stainless steel wires (40 μm in diameter) and were mounted in the 5 mL chambers of a multi-myograph system (model 610 M, DMT, Aarhus, Denmark) containing HEPES-buffered solution (in g/L: 8.415 NaCl, 0.432 KCl, 0.244 MgCl₂, 0.277 CaCl₂, 2 glucose, 1.1915 HEPES) continuously gassed with a mixture of 95% O₂ and 5% CO₂ at 37 °C for isometric force measurement (Supplementary Fig. 1). Tension signals were recorded by the PowerLab and saved to LabChart (ADInstruments, Ltd., Aarhus, Denmark). The standardization procedure was performed as previously described by McPherson [22], and the transmural pressure was set to a baseline value of 100 mmHg. Then, the arteries were equilibrated for 2 h in HEPES-buffered solution containing vehicle, BSA (200 μg/mL), AGE-BSA (200 μg/mL), AGE-BSA + ALA (1 mM), AGE-BSA + PIO (0.1 mM), or AGE-BSA + PIO + GW9662 (0.1 mM). After incubation, vessel viability was checked twice by administration of 60 mM KCl. When tension comes to baseline after washout, the arteries were precontracted with 1 μM U46619 (a vasoconstrictor analog of thromboxaneA₂), and Forskolin (cyclic adenosine monophosphate activator) with a concentration from 10^− 10 M to 10^− 5 M was used to induce an endothelium-independent vasodilation (Supplementary Fig. 2). After further washout and equilibration, the vessels were treated with 3 mM 4-aminopyridine (4-AP, a broad-spectrum Kv channel blocker) for 30 min and the Forskolin-induced vasodilation was repeated after precontraction with 1 μM U46619. In this study, Kv channel-mediated relaxation refers to the relaxation which can be blocked by 4-AP, that is the difference between maximum dilation measured before and after incubation with 4-AP.

Total protein was extracted from RSCAs or the supernatant of cell lysate of CSMCs after incubation in vitro. Protein concentration was determined using a BCA protein assay kit. Equal amounts of protein were separated by 10% SDS-PAGE, transferred to PVDF membranes, and stained with the following primary antibodies (all from Abcam, Cambridge, UK): polyclonal rabbit anti-RAGE (1:500), monoclonal mouse anti-Kv1.2 (1:1000), monoclonal mouse anti-Kv1.5 (1:1000), monoclonal rabbit anti-PPAR-γ (1:1000), monoclonal rabbit anti-NOX-2 (1:1000), and anti-β-actin (1:10000), followed by incubation with HRP-conjugated goat anti-mouse IgG antibody (1:20000, Proteintech Group Inc., IL, USA). Signals were visualized and quantified by a Western Blotting Imaging System (Clinx Science Instruments Co., Shanghai, China). Original Western blot images of target proteins are shown in another supplementary material (Supplementary Figs. 1 to 2 for original images in RSCAs, Supplementary Figs. 3 to 5 in CSMCs, Supplementary Figs. 6 to 7 in cells with RAGE siRNA transfection).

The incubated RSCAs were fixed with 4% paraformaldehyde and cross-sectioned. HE and Masson staining were performed. Standard immunohistochemistry protocols were applied. The sections were probed with primary antibody against RAGE (1:200; Abcam) and the biotinylated goat anti-rabbit IgG was used as the secondary antibody. Sections were visualized with diaminobenzidine and counterstained with HE. Images were captured digitally and analyzed using the IMS imaging processing system (Jierdun Biotech, Shanghai, China). Positively stained regions were identified and analyzed. Cardiomyocytes were excluded.

The RSCAs were obtained with endothelium carefully denuded and adventitia dissected. Primary CSMCs were isolated enzymatically as previous described [8]. Cells were cultured in DMEM for 24 h at 37 °C. The culture medium was changed when the cells were fully adherent, and the cell passages were performed when cell confluency reached 80%. The morphology and growth characteristics of the cells were typical of SMCs and were identified as SMCs by α-smooth muscle actin staining using immunofluorescence (Supplementary Fig. 3). The cells were then used for drug exposure experiments. Similar with the arteries, CSMCs were treated with vehicle (DMEM group), 100 μg/mL BSA (BSA group), 100 μg/mL AGE-BSA (AGE group), 100 μg/mL AGE-BSA + 10 μM ALA (AGE + ALA group), 100 μg/mL AGE-BSA + 1 μM PIO (AGE + PIO group), and 100 μg/mL AGE-BSA + 1 μM PIO + 1 μM GW9662 (AGE + PIO + GW9662 group) at 37 °C for 24 h. The concentrations of the above agents were adopted according to previous published studies [13, 18, 23].

Silencer Select Pre-Designed Small interfering RNAs (siRNAs) against the AGER gene of Rattus norvegicus were ordered from Invitrogen (CA, USA). The CSMCs were transfected with the siRNA in vitro using lipid-mediated Lipofectamine 3000 (Invitrogen) according to the manufacturer’s protocols. The CSMCs were plated in 6-well plates at 24 h before transfection and were transfected with 75 pmol of siRNA per 6-well plate and used at 48 h after transfection. A scrambled siRNA (Invitrogen) was used as a non-targeting negative control.

AGE-BSA was purchased from Abcam (Cambridge, UK). Endotoxin levels were found to be less than 0.8 EU/mg protein with the Limulus amebocyte assay (E-Toxate kit, Sigma, MO, USA). ALA was obtained from Shanghai Biopharmaleader (Shanghai, China). All the other chemicals were purchased from Sigma (MO, USA).

Isometric force of RSCAs in response to vasoactive agents was measured. Representative records (Fig. 1a) illustrate that 1 μM U46619 induced a constant maximal contraction, while Forskolin from 10^− 10 to 10^− 5 M elicited a concentration-dependent vasodilation. A high dose of acetylcholine failed to dilate the arteries after endothelial denudation. The dilation was largely depressed by 4-AP, indicating the critical role of Kv channel in the Forskolin-induced dilation.

We investigated the effect of AGE on Kv channel-mediated dilation in RSCAs. As shown in Fig. 1b, similar dilation was observed in control and BSA groups, suggesting that BSA did not affect the vasodilator response to Forskolin. By contrast, coronary relaxation was markedly reduced in the arteries exposed to AGE. The differences in maximal Forskolin-induced dilation between BSA and AGE groups were significant when the concentration of Forskolin was 10^− 7 M and 10^− 6 M (Supplementary Table 1). The impaired relaxation, however, could be reversed by ALA. After treatment with 4-AP, new dose-response curves were repeated to ascertain the contribution of Kv channel to Forskolin-elicited dilation (Fig. 1c). There were no significant differences between control and BSA groups. Of note, AGE remarkably attenuated the Kv channel-mediated dilation, while ALA largely restored the AGE-induced functional loss of Kv channel. These findings demonstrate that the impaired relaxation of RSCAs induced by AGE was associated with the vasodilator dysfunction of Kv channel.

Given excessive formation of AGE can directly enhance oxidative stress in arterial wall, we further explored the role of a redox-sensitive transcription factor, PPAR-γ, in the Kv channel dysfunction induced by AGE. PIO with or without GW9662 were used to treat arteries in the presence of AGE. As shown in Fig. 1d, the Forskolin-induced dilation was largely restored in arteries when co-treated with AGE and PIO as compared with the AGE group. However, this ameliorative effect brought by PIO was offset by additional treatment with GW9662. The effect of PPAR-γ modulation on Kv channel dysfunction was also confirmed (Fig. 1e). We found that PIO significantly reversed the AGE-induced impairment of Kv channel-mediated dilation whereas GW9662 attenuated this reversal. These data indicate that inhibition of PPAR-γ may contribute to the functional loss of Kv channel in RSCAs exposed to AGE.

The Kv1.2 and K1.5 channel are two major subunits of “Shaker-type” Kv1 channel family. They are abundantly expressed in coronary smooth muscles and their activities are susceptible to oxidative stress level in arteries. Therefore, we detected the protein expression of Kv1.2/1.5, RAGE, PPAR-γ, and the critical NADPH oxidase subunit (NOX-2) in both RSCAs and CSMCs.

Western blot analyses (Fig. 2) of RSCAs showed that Kv1.2/1.5 expression was similar in control and BSA group, while AGE significantly reduced Kv1.2/1.5 protein levels as compared with BSA. Meanwhile, AGE markedly increased RAGE and NOX-2 expression and reduced PPAR-γ expression. The above effects induced by AGE could be largely reversed by additional treatment with ALA or PIO (Fig. 2). No obvious arterial structural abnormalities or fibrosis in media layer were observed in sections of RSCAs (Supplementary Figs. 4 & 5). The immunohistochemical staining (Fig. 3) revealed that AGE increased the expression of RAGE, and this AGE-mediated RAGE overexpression was reversed by ALA or PIO.

In cultured CSMCs, AGE significantly reduced Kv1.2/Kv1.5 expression, increased the expression of RAGE and NOX-2, and reduced PPAR-γ expression; ALA or PIO reversed these AGE-mediated effects; and GW9662 partly reversed the effects of PIO (Fig. 4). The interaction between AGE/RAGE and Kv channel impairment was further elucidated by gene transfection with RAGE siRNA in CSMCs. Western blot analyses showed that RAGE siRNA reduced the RAGE expression by more than 50%, confirming the desired efficiency of RAGE silencing (Fig. 5a-b). After being transfected for 48 h and then incubated with or without AGE for 24 h, the protein expression of Kv1.2 and Kv1.5 in CSMCs was significantly higher when co-treated with RAGE siRNA and AGE as compared with the scrambled siRNA (negative control) + AGE group (Fig. 5c-f), suggesting that silencing of RAGE alleviated the AGE-induced Kv1.2/1.5 channel impairment.