Diabetes mellitus often leads to generalized vasculopathy. Because of the pathophysiological role of free radicals we investigated the effects of vitamin E. Twenty-eight rats were rendered diabetic by streptozotocin injection and were fed either with a diet with low (10 mg/kg of chow), medium (75 mg/kg of chow) or high amounts of vitamin E (1300 mg/kg of chow). Nine age-matched nondiabetic rats receiving 75 mg of vitamin E/kg chow served as controls. After 7 months, mesenteric microcirculation was investigated. Smooth muscle contractile function was not altered in diabetic versus nondiabetic vessels. Endothelial function was significantly reduced in diabetics; relaxation upon 1 M acetylcholine was reduced by 50% in diabetics with a medium and high vitamin E diet. In vitamin E-deprived rats, a complete loss of endothelium-dependent relaxation was observed, and instead, acetylcholine elicited vasoconstriction. L-N G -Nitro-arginine-induced vasoconstriction was reduced in small arteries in diabetics, which was not prevented by vitamin E, but was aggravated by vitamin E deprivation. In a subchronic endothelial cell culture model, cells were cultivated with 5 or 20 mM D-glucose for an entire cell culture passage (4 days) with or without vitamin E (20 mg/l versus 0.01 mg/l). Hyperglycemia led to significant reduction in basal and ATP-stimulated nitric oxide (NO)-production. Hyperglycemia-induced reduction in basal NO-release was significantly prevented by vitamin E, whereas reduction in stimulated NO-release was not influenced. NADPH-diaphorase activity was reduced by 40% by hyperglycemia, which was completely prevented by vitamin E. We conclude that 1) vitamin E has a potential to prevent partially hyperglycemia-induced endothelial dysfunction, 2) under in vivo conditions vitamin E deficiency enhanced diabetic endothelial dysfunction dramatically, and 3) positive effects of vitamin E may be attenuated with a longer disease duration.

To investigate whether ␤ 2 -adrenoceptors exist on endothelial cells and whether a ␤ 2 -adrenoceptor stimulation might prevent the development of hyperglycemia-induced endothelial dysfunction, porcine aortic endothelial cells (PAECs) were cultured and chronically exposed to either 5 mM D-glucose ("normoglycemia") or 20 mM D-glucose ("hyperglycemia"), with or without 100 nM salbutamol in absence or presence ofFor osmotic control, PAECs were exposed to 15 mM L-glucose. We measured nitric oxide release using the met-hemoglobin assay and assessed ␤-adrenoceptor density and subtypes by radioligand binding. Furthermore, we determined intracellular NADH and NADPH using high-performance liquid chromatography. High Dglucose concentrations but not L-glucose led to significantly reduced basal and stimulated nitric oxide release. Chronic salbutamol treatment significantly antagonized the impairment of the nitric oxide response, which was inhibited by ICI 118,551 but not by metoprolol. The number of giant cells was significantly increased in hyperglycemia, which could be prevented by salbutamol. Binding of the radioligand (Ϫ)-[ 125 I]iodocyanopindolol revealed a total ␤-adrenoceptor density of 29.8 Ϯ 3.7 (normoglycemic) and 30.3 Ϯ 3.6 (hyperglycemic) fmol/mg protein. Displacement by ICI 118,551 revealed ␤-adrenoceptor subtype distribution with 30.3 Ϯ 4.4 (normoglycemic) and 29.1 Ϯ 3.8% ␤ 2 -adrenoceptors. NADH production increased in hyperglycemia, which was completely prevented by salbutamol. We conclude that hyperglycemia in PAEC induces endothelial dysfunction with impaired nitric oxide release and that this can be prevented by ␤ 2 -adrenoceptor stimulation.An intriguing problem in antidiabetic therapy is the development of generalized angiopathy and concomitant hypertension in diabetes mellitus. The nature of the underlying processes is complex and not yet fully understood. Several lines of evidence suggest that vascular endothelial function is disturbed in diabetes (Fortes et al., 1983;Cagliero et al., 1991;Sank et al

Diabetes mellitus leads to the development of endothelial dysfunction which finally contributes to diabetic angiopathy. We investigated the effects of hyperglycaemia on nitric oxide (NO) liberation and a possible influence of L-arginine supplementation. Porcine endothelial aortic cells (PAEC) were cultured in Medium 199 containing 0.33 mmol/l L-arginine. During the entire third culture passage (= 4 days) cells were either exposed to 5 or 20 mmol/l D-glucose with or without additional 3 mmol/l L-arginine. For osmotic control, cells were exposed to 15 mmol/l mannitol. NO liberation was measured under basal conditions and after stimulation with 1 mmol/l ATP using the spectrophotometrical methemoglobin assay. Cells released 35 ± 8 pmol NO/1 × 10⁶ cells/10 min under basal conditions while hyperglycaemia led to a significant reduction in NO release to 16 ± 6 pmol/1 × 10⁶ cells/10 min. In osmotic control, NO release was unchanged (37 ± 10 pmol/1 × 10⁶ cells/10 min). Stimulation with 1 mmol/l ATP led to a significant increase in NO release to 103 ± 11 pmol/1 × 10⁶ cells/10 min (normoglycaemia) which was unchanged in osmotic controls. Under normoglycaemic conditions, additional L-arginine supplementation did not influence NO release from PAEC. In hyperglycaemia (0.33 mmol/l L-arginine) ATP stimulated NO release was reduced (48 ± 8 pmol/1 × 10⁶ cells/10 min, p < 0.05), which was completely prevented by 3 mmol/l L-arginine treatment (98 ± 15 pmol/ 1 × 10⁶ cells/10 min). Hyperglycaemia (but not enhanced osmotic pressure) leads to endothelial dysfunction with reduced NO release which is completely prevented by L-arginine. L-Arginine utilisation may be impaired in hyperglycaemia and L-arginine supplementation might be an interesting additional therapeutic tool in diabetic patients.

No abstract