ORIGINAL ARTICLE Influence of Diabetes Mellitus, Hypercholesterolemia, and Their Combination on EDHF-Mediated Responses in Mice Keiko Morikawa, MD,* Tetsuya Matoba, MD, PhD,* Hiroshi Kubota, MS,* Makoto Hatanaka, MS,* Takako Fujiki, MD,* Shosuke Takahashi, MD, PhD,† Akira Takeshita, MD, PhD,* and Hiroaki Shimokawa, MD, PhD*‡ Abstract: The endothelium synthesizes and releases several vaso- dilator substances, including vasodilator prostaglandins, NO, and EDHF. NO-mediated relaxations are reduced by various risk factors, such as diabetes mellitus and hypercholesterolemia. However, it remains to be elucidated whether EDHF-mediated relaxations also are reduced by those factors and their combination. In this study, we addressed this point in mice. We used small mesenteric arteries from control, diabetic (streptozotocin-induced), apolipoprotein-E-deficient (ApoE 2/2 ), and diabetic ApoE 2/2 mice. In control mice, endothe- lium-dependent relaxations to acetylcholine were largely mediated by EDHF. This EDHF-mediated component was slightly reduced in diabetic mice, preserved in ApoE 2/2 mice, and markedly reduced in diabetic ApoE 2/2 mice with an increase in NO-mediated component and a negative contribution of indomethacin-sensitive endothelium- derived contracting factor (EDCF). Endothelium-independent relax- ations to sodium nitroprusside or NS1619, a direct opener of calcium- activated K channels, were attenuated in ApoE 2/2 and diabetic ApoE 2/2 mice. Endothelium-dependent hyperpolarizations were significantly reduced in diabetic mice, preserved in ApoE 2/2 mice, and again markedly reduced in diabetic ApoE 2/2 mice. These results indicate that hypercholesterolemia alone minimally affects the EDHF- mediated relaxations, and diabetes mellitus significantly attenuated the responses, whereas their combination markedly attenuates the responses with a compensatory involvement of NO and a negative contribution of EDCF. Key Words: endothelium, endothelium-derived hyperpolarizing factor, diabetes mellitus, hypercholesterolemia, membrane potential (J Cardiovasc Pharmacol TM 2005;45:485–490) T he endothelium plays an important role in maintaining vascular homeostasis by synthesizing and releasing several vasodilating factors, including vasodilator prostaglandins, nitric oxide (NO), and still unidentified endothelium-derived hyperpolarizing factor (EDHF). 1 Several candidates for EDHF have been proposed, including cytochrome P450 metabolites (EETs), 2,3 potassium ions, 4,5 gap junctions, 6,7 and, as we have recently demonstrated, hydrogen peroxide. 8–11 EDHF plays an important role in modulating vascular tone, especially in microvessels. 1,12–14 It is reported that various risk factors, such as diabetes mellitus, hypercholesterolemia, and hypertension, cause endo- thelial dysfunction, leading to atherosclerosis and microvas- cular disorders. 1,15,16 The effects of these risk factors on NO- mediated responses have been extensively studied. It has also been reported that diabetes mellitus, hypertension, and hyper- cholesterolemia affect EDHF-mediated relaxations. 17 However, it remains to be examined how those risk factors affect EDHF-mediated responses when combined. In this study, we examined whether diabetes mellitus, hypercholesterolemia, and their combination affect EDHF- mediated responses, using apolipoprotein E-deficient mice (ApoE 2/2 ), streptozotocin-induced diabetic mice, and diabetic ApoE 2/2 mice. METHODS Animals and Tissue Preparation This study was approved by the Research Committee of the Kyushu University Graduate School of Medical Sciences. We used control C57BL/6N and ApoE 2/2 mice. At 8 weeks of age, animals were made diabetic by intraperitoneal injection of streptozotocin (100 mg/kg/day on 3 consecutive days) that was dissolved in 0.1 mol/L citrate buffer. Control mice received the citrate buffer only. Nonfasting blood glucose levels higher than 300 mg/dL were defined as diabetic. 18 There were 4 experimental groups: control, diabetic, ApoE 2/2 , and diabetic ApoE 2/2 mice. Animals with an intraperitoneal injection of pentobarbital sodium (70 mg/kg) were killed 10 weeks after the injection of streptozotocin. 19 Small mesenteric arteries (200–250 mm in diameter) were excised and carefully cleaned of adherent perivascular connective tissue under a micro- scope. 8,20 The vessels were cut into rings for organ chamber Received for publication November 16, 2004; accepted February 1, 2005. From the *Department of Cardiovascular Medicine, Kyushu University Graduate School of Medical Sciences, Fukuoka, Japan; †Department of Anesthesiology, Kyushu University Graduate School of Medical Sciences, Fukuoka, Japan; and the ‡Kyushu University COE Program on Lifestyle-Related Diseases, Kyushu University Graduate School of Medical Sciences, Fukuoka, Japan. This work was supported in part by the grant for the 21st Century COE Program and by grants-in-aid (Nos. 13307024, 13557068, 15256003, 16209027) from the Japanese Ministry of Education, Science, Sports, and Culture (Tokyo, Japan). Reprints: Hiroaki Shimokawa, MD, PhD, Department of Cardiovascular Medicine, Kyushu University Graduate School of Medical Sciences, Fukuoka 812-8582, Japan (e-mail: shimo@cardiol.med.kyushu-u.ac.jp). Copyright Ó 2005 by Lippincott Williams & Wilkins J Cardiovasc Pharmacol ä  Volume 45, Number 5, May 2005 485