Glimepiride induces nitric oxide production in human coronary artery endothelial cells via a PI3-kinase-Akt dependent pathway

Atherosclerosis 183 (2005) 35–39

Glimepiride induces nitric oxide production in human coronary artery endothelial cells via a PI3-kinase-Akt dependent pathway Hiroto Ueba, Masatoshi Kuroki, Shigemasa Hashimoto, Tomio Umemoto, Takanori Yasu, San-e Ishikawa, Muneyasu Saito, Masanobu Kawakami ∗ Department of Internal Medicine, Omiya Medical Center, Jichi Medical School, Amanuma-Cho 1-847, Saitama City 330-8503, Japan Received 4 January 2004; received in revised form 10 January 2005; accepted 12 January 2005 Available online 18 April 2005

Abstract Diabetes mellitus is one of the major risk factors for coronary artery disease (CAD). A recent study reported that glimepiride, a new third-generation sulfonylurea, inhibited the formation of atheromatous plaques in high-cholesterol fed rabbits. However, the mechanism by which glimepiride induces atheroprotection remains unknown. In the present study, we tested the hypothesis that glimepiride may stimulate NO production in vascular endothelial cells. Human coronary artery endothelial cells (HCAECs) were treated with glimepiride, glibenclamide or vehicle, and NO release was measured. Akt phosphorylation was evaluated by Western blot. The effects of LY294002, a specific PI3-kinase inhibitor, and antisense oligonucleotides directed to Akt, on glimepiride-induced NO production were examined. Glimepiride (0.1–10 ␮M), but not glibenclamide, induced NO production, significantly increasing it by 1.8-fold (n = 6, p < 0.05). LY294002 inhibited glimepiride-induced NO production by 68%. Akt was rapidly phosphorylated by glimepiride and antisense oligonucleotides directed to Akt completely inhibited glimepiride-induced NO production. These data demonstrate that glimepiride induces NO production in HCAECs by activating PI3-kinase and Akt, and also suggest that use of glimepiride in type 2 diabetes may show promise for preventing CAD in addition to lowering glucose levels. © 2005 Elsevier Ireland Ltd. All rights reserved. Keywords: Sulfonylurea; Nitric oxide; Endothelium; Signal transduction; Atherosclerosis

1. Introduction Diabetic macroangiopathy such as coronary artery disease (CAD) is a serious problem that negatively impacts the prognosis of diabetic patients [1]. The incidence of CAD is two to four times higher in diabetic patients than in non-diabetic patients and mortality in diabetic patients over the 30 days following the onset of acute myocardial infarction in diabetic patients is two-fold higher [2,3]. Because CAD is becoming the leading cause of mortality in developed countries, the prevention of diabetic macroangiopathy is critical for improving the prognosis of diabetic patients. Sulfonylureas are widely used for the treatment of type 2 diabetes, but their possible deleterious effects on CAD are ∗

Corresponding author. Tel.: +81 48 647 2111; fax: +81 48 648 5188. E-mail address: [email protected] (M. Kawakami).

0021-9150/$ – see front matter © 2005 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.atherosclerosis.2005.01.055

still debated. Treatment of diabetic patients with conventional sulfonylureas such as glibenclamide has been reported to be associated with adverse cardiovascular effects and a high incidence of cardiovascular death [4]. Previous studies have shown that glimepiride, a new third-generation sulfonylurea, has less adverse cardiovascular activity than conventional sulfonylureas and does not appear to have deleterious effects on ischemic preconditioning [5,6]. Recently, glimepiride has been demonstrated to inhibit the formation of atheromatous plaques in thoracic aortae of high-cholesterol fed rabbits [7]. However, the mechanism by which glimepiride induces atheroprotective effects remains to be elucidated. Glimepiride also has been reported to show extrapancreatic actions including phosphorylation of insulin receptor substrate-1 (IRS-1), an upstream regulator of phosphatidylinositol 3-kinase (PI3-kinase), in adipocytes [8]. PI3-kinase has been shown to be involved in

36

H. Ueba et al. / Atherosclerosis 183 (2005) 35–39

agonist-induced nitric oxide (NO) production in vascular endothelial cells (ECs) [9]. Taken together, these findings suggest that glimepiride may stimulate NO production in ECs through a PI3-kinase dependent pathway, leading to the induction of atheroprotective effects. In the present study, we investigated the effect of glimepiride on NO production in ECs and its associated signaling pathway using human coronary artery endothelial cells (HCAECs). We demonstrated that glimepiride induces NO production in HCAECs in a concentration-dependent manner and that PI3-kinase and Akt play important roles in glimepiri de-induced NO production.

Griess reagent and the reaction products were measured by a flow-through spectrophotometer at 540 nm. Because Krebs’ buffer does not contain metallic ions to generate hydroxyl radical by which any nitrite present would be oxidized to nitrate, released NO was measured as the difference of values of nitrite before and after the addition of the test compounds. NaNO2 was used as a standard. For the experiments in which antisense oligonucleotides were used, antisense Akt or control ODN were added to HCAECs 12 h prior to the experiments. Then cells were incubated with glimepiride and NO was measured as described above. Transfection efficiency was evaluated to be more than 90% using FITC-tagged oligonucleotides by fluorescent microscopic examination.

2. Materials and methods

2.4. Western blot analysis

2.1. Chemicals

After treatment, HCAECs were washed in ice-cold PBS (137 mM NaCl, 2.7 mM KCl, 4.3 mM Na2 HPO4 , 1.4 mM KH2 PO4 , pH 7.3), and cell lysates were prepared by flashfreezing and thawing in lysis buffer (50 mM sodium pyrophosphate, 50 mM NaF, 50 mM NaCl, 5 mM EDTA, 5 mM EGTA, 100 mM Na3 VO4 , 10 mM HEPES, pH 7.4, 500 ␮M PMSF, 10 ␮M leupeptin) with 1% Triton X-100 and 0.05% SDS. Cell lysates were subjected to SDS-PAGE, and were transferred to nitrocellulose membranes (Amersham Biosciences, Piscataway, NJ). The membrane was blocked for 1 h at room temperature with PBS containing 2% BSA and 0.05% Tween 20. The blots were then incubated overnight at 4 ◦ C or for 4 h at room temperature with anti-phosphoAkt antibody, anti-Akt antibody or anti-ERK2 antibody (Santa Cruz Biotechnology Inc., Santa Cruz, CA, USA), followed by incubation for 1 h with a secondary, horseradish peroxidaseconjugated antibody. Immunoreactive bands were visualized using ECL (Amersham Biosciences). All experiments were performed at least three times.

Glimepiride and glibenclamide were provided by Aventis Pharma Ltd. (Strasbourg, France). Recombinant human hepatocyte growth factor (HGF) was obtained from R&D Systems Inc. (Minneapolis, MN). LY294002 was purchased from Calbiochem (San Diego, CA). Antisense oligodeoxynucleotides directed to Akt (antisense Akt) and control oligodeoxynucleotides (control ODN), were obtained from Biognostik (G¨ottingen, Germany). All other materials were from Sigma–Aldrich Co. (St. Louis, MO) except where indicated. 2.2. Cell culture HCAECs and cell culture kits were purchased from Asahi Techno Glass Co. (Tokyo, Japan). Cells were grown in medium EBM supplemented with 5% fetal bovine serum, 12 ␮g/ml bovine brain extract, 10 ng/ml epidermal growth factor, 1 ␮g/ml hydrocortisone, 50 ␮g/ml gentamicin and 50 ng/ml amphotericin B, at 37 ◦ C under air containing 5% CO2 in a humidified incubator. Cells were subcultured in the same medium and cells from passages four to five were used for the experiments. 2.3. NO assay HCAECs were incubated with glimepiride, glibenclamide, HGF (positive control) or vehicle in warmed Krebs’ buffer (118 mM NaCl, 4.7 mM KCl, 2.5 mM CaCl2 , 1.2 mM MgSO4 , 1.2 mM KH2 PO4 , 11 mM dextrose, 25 mM NaHCO3 , 20 mM HEPES, pH 7.4) maintained at 37 ◦ C and supernatants were withdrawn at the indicated times from the culture dishes and stored at 4 ◦ C. Released NO was measured using a NOx analyzing high-performance liquid chromatography (HPLC) system (ENO-20, EICOM, Kyoto, Japan) within 24 h after each sample collection as previously described [10]. Briefly, nitrate was separated by HPLC with a reversed-phase column, and was reduced to nitrite by passage through a cadmium column. Samples were then mixed with

2.5. Statistical analysis Data are shown as mean ± S.D. Differences were analyzed with paired t-tests between two groups, and one-way analysis of variance (ANOVA) among more than three groups. Values of p < 0.05 were considered statistically significant.

3. Results 3.1. Glimepiride induces NO production by HCAECs To determine whether glimepiride induces NO production by ECs, we measured NO in cultured HCAECs. Glimepiride, but not glibenclamide, induced NO production by HCAECs (Fig. 1). Glimepiride-induced NO production increased significantly in a concentration-dependent manner (n = 6, p < 0.05; Fig. 2A) and this increase in NO production began within 1 min following addition of the drug, and reached 1.8 ± 0.3-fold at 30 min (n = 6, p < 0.05; Fig. 2B).

H. Ueba et al. / Atherosclerosis 183 (2005) 35–39

37

Fig. 1. Effects of glimepiride and glibenclamide on NO production in HCAECs. Cells were incubated with glimepiride (10 ␮M), glibenclamide (10 ␮M), HGF (20 ng/ml, positive control) or vehicle for 30 min. Released NO was measured as the difference in NO2 determined before and after incubation using a NOx analyzing HPLC system as described in Section 2. Results are mean ± S.D. * p < 0.05 vs. glibenclamide or vehicle; n = 5–6.

3.2. Requirement of PI3-kinase and Akt in glimepiride-induced NO production We examined the involvement of PI3-kinase in glimepiride-induced NO production in HCAECs using LY294002, a specific PI3-kinase inhibitor. LY294002 (20 ␮M) significantly inhibited glimepiride-induced NO production by 68% (n = 3, p < 0.05; Fig. 3). To evaluate the effect of glimepiride on activation of Akt, a downstream effector of PI3-kinase, we performed Western blot using an anti-phosphoAkt antibody. Akt was rapidly phosphorylated by glimepiride treatment (Fig. 4A) and antisense Akt completely inhibited glimepiride-induced NO production (n = 5, p < 0.05; Fig. 4B). Fig. 4C shows the significant downregulation of Akt at the protein level by antisense Akt.

4. Discussion This study shows a novel action of glimepiride, the induction of NO production in ECs, which requires PI3-kinase and Akt. This conclusion is based on the following findings. First, glimepiride, but not glibenclamide, induces NO production in HCAECs in a concentration-dependent manner. Second, LY294002 significantly inhibits glimepirideinduced NO production. Third, antisense Akt completely inhibit NO production induced by glimepiride. These data suggest that use of glimepiride in type 2 diabetes may show promise for preventing CAD in addition to lowering glucose levels.

Fig. 2. (A) Dose–response of glimepiride-induced NO production in HCAECs. Cells were incubated for 30 min with the indicated concentrations of glimepiride. (B) Time course of glimepiride-induced NO production in HCAECs. Cells were incubated for the indicated times with 10 ␮M of glimepiride. Released NO was measured as described above. Results are mean ± S.D. * p < 0.05 vs. 0 and 0.01 ␮M (A), * p < 0.05 vs. 0 min (B). # p < 0.05 vs. 0.1 and 1 ␮M (A), # p < 0.05 vs. 1, 5 and 15 min (B); n = 6.

It is generally accepted that NO plays an important role in regulating normal vascular function and confers protection against the development and progression of atherosclerosis. NO production by endothelial NO synthase in ECs has been shown to be regulated by several factors including shear stress, vascular endothelial growth factor and HGF [9,11,12]. Recently, 3-hydroxy-3-methyl-glutaryl-CoA (HMG-CoA) reductase inhibitors (statins) have been also reported to induce NO production in ECs and to show atheroprotective effects independently of their lipid lowering effects [13]. In the present study, we have clearly demonstrated that glimepiride induces NO production in HCAECs, a mechanism that can account for the inhibition of atheromatous plaque formation

38

H. Ueba et al. / Atherosclerosis 183 (2005) 35–39

Fig. 3. Effect of a specific PI3-kinase inhibitor, LY294002, on glimepirideinduced NO production. HCAECs were preincubated with vehicle or LY294002 (20 ␮M) for 30 min, then cells were incubated with glimepiride (10 ␮M) or vehicle for 30 min and released NO was measured as in Fig. 1. Results are mean ± S.D. * p < 0.05 vs. glimepiride; n = 3.

in the thoracic aortae in high-cholesterol fed rabbits shown by Shakuto et al. [7]. Considering these findings and the established anti-inflammatory effects of NO, glimepiride may have pleiotropic beneficial vascular effects just like statins. We and others previously showed that PI3-kinase and Akt are involved in shear stress- and agonist-induced NO production in ECs [9,12,14]. Recent work has shown that statininduced NO production also requires activation of PI3-kinase and Akt [15]. In the present study, we have demonstrated that glimepiride induces NO production through a PI3-kinaseAkt dependent pathway. There was a correlation between the effects of glimepiride/glibenclamide on NO production and activation of the PI3-kinase/Akt pathway because the extent of Akt phosphorylation level by glibenclamide was significantly lower than that by glimepiride (data not shown). Taken together, PI3-kinase and Akt may play a critical role in druginduced NO production in ECs as well as shear stress- and agonist-induced NO production. An upstream pathway of PI3-kinase including a receptor involved in glimepiride-induced NO production remains to be identified. However, sulfonylurea receptor 2 (SUR2) is unlikely to be involved in this pathway as shown by our finding that glibenclamide, whose affinity to SUR2 is close to that of glimepiride [16], did not induce NO production in ECs. Muller et al. reported that interaction of phosphatidylinositolglycan with plasma membrane lipid rafts induces activation of the nonreceptor tyrosine kinase, pp59 (Lyn) and subsequent tyrosine phosphorylation of IRS-1/2 in adipocytes, and that stimulation of glucose transport in adipocytes by glimepiride is also based on IRS-1/2 tyrosine phosphory-

Fig. 4. (A) Effect of glimepiride on phosphorylation of Akt. HCAECs were incubated with glimepiride (10 ␮M) for the indicated times and cell lysates were prepared as described in Section 2. Activation of Akt was assessed by Western blot using an anti-phosphoAkt antibody. Results shown are representative of three separate experiments. (B) Effect of antisense Akt oligonucleotides (antisense Akt) on glimepiride-induced NO production. Antisense Akt or control oligonucleotides (control ODN) were added to HCAECs 12 h prior to experiments. Cells were incubated with glimepiride (10 ␮M) for 30 min and released NO was measured as in Fig. 1. Results are mean ± S.D. * p < 0.05; n = 5. (C) Effect of antisense Akt on Akt down-regulation at the protein level. HCAECs were incubated with antisense Akt or control ODN for 12 h, and then cell lysates were prepared as described in Section 2. The extent of Akt down-regulation was assessed by Western blot. ERK2 was used as a control to show equal loading of protein samples. Results shown are representative of three separate experiments.

lation and downstream insulin-mimetic signaling involving activation of pp59 (Lyn) [8,17]. Moreover, we previously demonstrated that cultured bovine aortic endothelial cells have much amount of rafts/caveolae in the plasma membrane [18], and it is generally well-known that endothelial cells harbor plasma membrane lipid rafts in high number. Based on these findings and our observations supporting the role of PI3-kinase and Akt in glimepiride-induced NO production, glimepiride may interact with plasma membrane lipid rafts in the same manner as phosphatidylinositolglycan, and phosphorylates IRS-1/2, leading to activation of a PI3-kinase-Akt dependent pathway necessary for NO production. Maximal blood concentration of glimepiride used in a clinical setting is reported to approach up to 1.0 ␮M (data from Aventis Pharma Ltd.); however, more than 99% of the drug

H. Ueba et al. / Atherosclerosis 183 (2005) 35–39

remain bound to serum albumin, resulting in about 10 nM free drug. Because HCAECs were incubated in serum-free medium in the present experiments, the effective concentrations reflect free drug with 100 nM being the lowest one (Fig. 2A). This 10-fold discrepancy between effective concentrations in vivo and in vitro may be explained by timedependent accumulation of lipophilic glimepiride at lipid rafts rather than a typical receptor–ligand interaction. As shown in Fig. 2B, the increase rate of NO production (difference of NO2 fold increase between sampling times) is highest between 0 and 1 min after addition of glimepiride, and then slows down. This slow-down of NO increase rate is presumably due to a reduction in the effective concentrations caused by time-dependent accumulation of glimepiride at lipid rafts. These findings suggest that glimepiride may produce NO in the effective concentrations in vivo and confer protection against the formation of atheromatous plaque. Acknowledgments This study was supported by grants-in-aid for scientific research 14570690 from the Ministry of Education, Culture, Sports, Science and Technology of Japan, and by research funding from Aventis Pharma Ltd. We thank Kazuko Futaka and Harue Fukaya for their technical assistance in the present study.

References [1] Bierman E. George Lyman Duff Memorial Lecture. Atherogenesis in diabetes. Arterioscler Thromb 1992;12(6):647–56. [2] Laakso M, Lehto S. Epidemiology of risk factors for cardiovascular disease in diabetes and impaired glucose tolerance. Atherosclerosis 1998;137(Suppl.):65–73. [3] Woodfield S, Lundergan C, Reiner J, et al. Angiographic findings and outcome in diabetic patients treated with thrombolytic therapy for acute myocardial infarction: the GUSTO-I experience. J Am Coll Cardiol 1996;28(7):1661–9. [4] del Valle H, Lascano E, Negroni J, Crottogini AJ. Glibenclamide effects on reperfusion-induced malignant arrhythmias and left ventricular mechanical recovery from stunning in conscious sheep. Cardiovasc Res 2001;50(3):474–85.

39

[5] Geisen K, Vegh A, Krause E, Papp JG. Cardiovascular effects of conventional sulfonylureas and glimepiride. Horm Metab Res 1996;28(9):496–507. [6] Riveline J, Danchin N, Ledru F, Varroud-Vial M, Charpentier G. Sulfonylureas and cardiovascular effects: from experimental data to clinical use. Available data in humans and clinical applications. Diab Metab 2003;29(3):207–22. [7] Shakuto S, Sato Y, Ohshima K, Yaguchi M. Atheroprotective effects of a new sulfonylurea of the third generation, glimepiride. Diabetic Complications 2001;15(Suppl. 1):68. [8] Muller G, Jung C, Wied S, Welte S, Frick W. Insulin-mimetic signaling by the sulfonylurea glimepiride and phosphoinositolglycans involves distinct mechanisms for redistribution of lipidraft components. Biochemistry 2001;40(48):14603–20. [9] Maejima Y, Ueba H, Kuroki M, et al. Src family kinases and nitric oxide production are required for hepatocyte growth factor-stimulated endothelial cell growth. Atherosclerosis 2003;167(1):89–95. [10] Hiraoka A, Masuya Y, Tominaga I. Simultaneous determination by high-performance liquid chromatography of nitrite and nitrate in cerebrospinal fluid and serum from patients with neurological disorders. Chromatography 1999;20(1):45–9. [11] Ueba H, Poppa V, Suero J, Okuda M, Berk BC. c-Src is required for flow-stimulated NO production in bovine aortic endothelial cells. Circulation 1998;98(17):I-312. [12] Gallis BCG, Goodlett DR, Ueba H, et al. Identification of flowdependent endothelial nitric-oxide synthase phosphorylation sites by mass spectrometry and regulation of phosphorylation and nitric oxide production by the phosphatidylinositol 3-kinase inhibitor LY294002. J Biol Chem 1999;274(42):30101–8. [13] Aengevaeren W. Beyond lipids—the role of the endothelium in coronary artery disease. Atherosclerosis 1999;147(Suppl. 1):11– 6. [14] Fulton D, Gratton J, McCabe TJ, et al. Regulation of endotheliumderived nitric oxide production by the protein kinase Akt. Nature 1999;399(6736):597–601. [15] Skaletz-Rorowski A, Lutchman M, Kureishi Y, Lefer D, Faust J, Walsh K. HMG-CoA reductase inhibitors promote cholesteroldependent Akt/PKB translocation to membrane domains in endothelial cells. Cardiovasc Res 2003;57(1):253–64. [16] Song DK, Ashcroft FM. Glimepiride block of cloned beta-cell, cardiac and smooth muscle K (ATP) channels. Br J Pharmacol 2001;133(1):193–9. [17] Muller G, Jung C, Frick W, Bandlow W, Kramer W. Interaction of phosphatidylinositolglycan(-peptides) with plasma membrane lipid rafts triggers insulin-mimetic signaling in rat adipocytes. Arch Biochem Biophys 2002;408(1):7–16. [18] Peterson TE, Poppa V, Ueba H, Wu A, Yan C, Berk BC. Opposing effects of reactive oxygen species and cholesterol on endothelial NO synthase and endothelial cell caveolae. Circ Res 1999;85:29– 37.