Studies using different animal models of diabetes in several different vascular beds (46-49) suggest that there is a decrease in endothelium-dependent vasodilation in the diabetic state. In two such animal models of type 1 diabetes, rats are made diabetic with streptozocin or rabbits made diabetic with alloxan, pancreatic P-cells are destroyed, with a corresponding decrease in insulin secretion. Studies evaluating endothelium-depen-dent vasodilation in these animal models have demonstrated a decreased response to endothelial stimulators such as ADP, acetylcholine, or its analogue methacholine (47).
Similarly, in an animal model of type 2 diabetes, the Zucker rat, which is characterized by hyperglycemia because of insulin resistance, abnormal endothelium-dependent va-sodilation is also seen (46). The early vascular dysfunction that occurs in type 1 diabetic animal models can be prevented by insulin therapy (50,51). The abnormal endothelial cell function that develops appears to be as a result of hyperglycemia rather than any other metabolic disturbance. This has been demonstrated by in vitro incubation experiments in which isolated arteries exposed to elevated glucose concentrations have similar decreases in endothelium-dependent vasodilation (52,53). This effect does not seem to be as a result of the hyperosmolarity because similar concentrations of mannitol have no effect on endothelium-dependent relaxation (52). The decreased endothelium-dependent vasodilation that occurs may be as a result of decreased synthesis or release of NO, decreased responsiveness of the smooth muscle to NO, the inactivation of NO by superoxide radicals, or the generation of endothelial vasoconstrictive factors. This will be discussed in greater detail later in this chapter.
Early in the course of experimental diabetes, there is a selective decrease in the response to those endothelium-dependent vasodilators that are mediated by endothelial cell receptors. The responsiveness of the endothelium to the direct endothelial vasodilator A23187, or the smooth muscle to nitrovasodilators, is preserved. Using a diabetic rabbit model, abnormal endothelium-dependent relaxation was also found (54) within 6 weeks of initiating the diabetic state. This may be explained by a decrease in the number of receptors, or in their function. These changes are specific to the diabetic state because these abnormal responses do not occur within 2 weeks after initiating the diabetic state and are not found in rabbits not made diabetic after alloxan treatment (55). Yet after a longer duration of diabetes, several groups have demonstrated a decrease in smooth muscle cGMP, suggesting a decrease in NO release or action over time (46,56).
Endothelial cell dysfunction in diabetes may be explained in part not only to perturbations in NO activity or levels but the effect of vasoconstrictor prostanoids. There is increased expression of cyclooxygenase-2 mRNA and proteins levels with hyperglyce-mia in cultured human aortic endothelial cells but not cyclooxygenase-1. Cohen's group noted that endothelium-dependent relaxation in arteries of diabetic animals could be restored by the administration of cyclo-oxygenase inhibitors or thromboxane A2 receptor antagonists, suggesting the presence of vasoconstrictor prostanoids (48,53). The responsiveness of smooth muscle to direct smooth muscle vasodilators is similar in both diabetic and normal animal models, suggesting that decreased responsiveness to NO is not affected (47,48).
This is an increase in oxygen-derived free radicals (57), either because of an increase in free radical production or because of a decrease in the free radical scavenger system. Furthermore, free radical scavengers have been shown to improve the abnormal endot-helium-dependent vasodilation (58,59), implying that such free radicals may contribute to the abnormal endothelium-dependent relaxations.
Was this article helpful?