In isolated blood vessels, acetylcholine releases endothelium-derived relaxing factor (EDRF). In vivo, the vasodilator action of acetylcholine may be mediated by EDRF, but prostacyclin or prejunctional inhibition of adrenergic neurotransmission may also be involved. Therefore, we investigated whether acetylcholine releases EDRF in humans in vivo and, if so, whether the response altered in essential hypertension. Acetylcholine was infused into the brachial artery, and forearm blood flow measured by venous occlusion plethysmography. In control subjects, acetylcholine (0.02-16 micrograms/min/100 ml tissue) increased flow from 2.4 +/- 5.0 to 20.6 +/- 5.2 ml/min/100 ml tissue (n = 14; p less than 0.05) and decreased forearm vascular resistance from 42.0 +/- 4.1 to 6.0 +/- 1.4 units (p less than 0.03), a response comparable to that of sodium nitroprusside (0.6 micrograms/min ml tissue). Acetylsalicylic acid (500 mg i.v.) given to block vascular prostacyclin production did not alter the response (n = 14). alpha-Adrenoceptor blockade by phentolamine (12 micrograms/min/100 ml tissue) did not prevent the increase in flow evoked by acetylcholine. In hypertensive patients, the decrease in forearm vascular resistance induced by acetylcholine but not evoked by sodium nitroprusside was reduced as compared with controls (14.5 +/- 3.1 and 6.1 +/- 1.6 units, respectively; n = 8; p less than 0.05). Thus, the vascular effects of acetylcholine in the human forearm circulation are independent of prostaglandins and adrenergic neurotransmission and therefore are most likely to be mediated by EDRF; the acetylcholine-induced release of EDRF is blunted in patients with essential hypertension.

Both the internal mammary artery and the saphenous vein are used to construct coronary-artery bypass grafts. We hypothesized that the release or production of endothelium-derived relaxing factor, which regulates blood flow and inhibits platelet function, may differ in venous and arterial grafts. We therefore studied endothelium-dependent relaxation in internal mammary arteries, internal mammary veins, and saphenous veins obtained from 58 patients undergoing coronary bypass surgery. Vascular rings with and without endothelium were suspended in organ chambers, and isometric tension was recorded. Acetylcholine (10(-8) to 10(-4) M), thrombin (1 U per milliliter), and adenosine diphosphate (10(-7) to 10(-4) M) evoked potent endothelium-dependent relaxation in the mammary artery but weak response in the saphenous vein (P less than 0.005; n = 6 to 27). In the mammary artery, relaxation was greatest in response to acetylcholine (86 +/- 4 percent reduction in norepinephrine-induced tension), followed by thrombin (44 +/- 7 percent) and adenosine diphosphate (39 +/- 8 percent). In the saphenous and mammary veins, relaxation was less than 25 percent. Relaxation was unaffected by indomethacin but was inhibited by methylene blue and hemoglobin (P less than 0.005 and 0.01, respectively), which suggests that endothelium-derived relaxing factor was the mediator. Endothelium-independent relaxation in response to sodium nitroprusside was similar in arteries and veins. We conclude that endothelium-dependent relaxation is greater in the mammary artery than in the saphenous vein. The possibility that this contributes to the higher patency rate among arterial grafts than among venous grafts will require further study.

Endothelin-1 is an endothelium-derived vasoconstrictor peptide. Its circulating levels are below those known to evoke direct vascular effects. To study whether low concentrations of endothelin-1 potentiate the effects of other vasoconstrictor hormones, we suspended isolated human internal mammary and left anterior descending coronary artery rings in organ chambers for isometric tension recording. In mammary artery rings, the contractions to norepinephrine (3 x 10-8 M) were potentiated by threshold (3 X 10-10 M) and low concentrations (10`M) of endothelin-1 (96±35% and 149±58% increase from control; p<0.01 and 0.001; n=6). The inhibitor of endothelial nitric oxide formation L-NG-monomethyl arginine did not affect the potentiating effects of the peptide. The calcium antagonist darodipine (10`M) prevented the potentiation of the response to norepinephrine evoked by endothelin-1. Similarly, contractions to serotonin (10`or 3 x 10-8 M) were amplified by endothelin-1 (3 x 10-10 M) in the mammary (30±9%o) and in the coronary arteries (59±25%). Endothelin-1 (10-9 M) further potentiated the response (57±23% in mammary and 87±26% in coronary arteries;p <0.05; n=7 and 3). The sensitivity of mammary arteries to calcium chloride was markedly enhanced in the presence of endothelini1 (3 x 1010 M; concentration shift, eightfold; pc

The endothelium not only mediates relaxation but is a source of contracting factors. Endothelium-dependent contractions are elicited by physical and chemical stimuli (i.e., hypoxia, pressure, and stretch) and autacoids, local and circulating hormones. The mechanism of endothelium-dependent contractions to hypoxia involves withdrawal of nitric oxide. The endothelial cyclooxygenase pathway can produce thromboxane A 2 , prostaglandin H 2 , and superoxide anions. The peptide endothelin is a potent contracting factor; its production is stimulated by vasopressor hormones, platelet-derived factors, coagulation products, and cytokines, whereas endothelium-derived nitric oxide, prostacyclin, and a smooth muscle cell-derived inhibitory factor reduce endothelin production. In hypertension, the release of cyclooxygenase-dependent endothelium-derived contracting factors to stretch, acetylcholine, and platelet-derived products is augmented. Vascular endothelin production in hypertension remains controversial but appears mostly normal; it is augmented in the presence of vascular disease or renal insufficiency. The endothelium-dependent inhibition of endothelin-induced contractions is reduced in hypertension while the reactivity of vascular smooth muscle may be normal, increased, or reduced. The potentiating effects of low concentrations of endothelin on contractions to norepinephrine are augmented with aging and hypertension. In atherosclerosis, the production of the cyclooxygenase-dependent endothelium-derived contracting factors and endothelin is enhanced. Thus, endothelium-derived contracting factors can profoundly affect vascular tone and counteract relaxing factors produced within the endothelium. In hypertension and atherosclerosis, the role of contracting factors appears to become more dominant, leading to an imbalance of endothelium-dependent vascular regulation. (Hypertension 1992;19:117-130) S oon after endothelium-dependent relaxations were discovered, observations were reported that the presence of endothelial cells augmented rather than inhibited contractile responses of certain blood vessels (for review, see Reference 1). In particular, contractions of canine arteries during anoxia were found to be markedly reduced after removal of the endothelium. 2 -4 This was the first indication that the endothelium might also produce contracting factors under certain conditions. In the last 10 years, an increasing number of endotheliumderived contracting factors (EDCF) have been characterized, while others remain unidentified ( Figure I). 1 -5 -7 A marked heterogeneity of these responses exists among species as well as among different vascular beds. This article reviews the physiology of EDCFs and their possible role in hypertension and atherosclerosis.

The endothelium has a strategical anatomical position between the circulating blood and vascular smooth muscle cells. It has recently been recognized that endothelial cells play an important regulatory role in the circulation. The cells metabolize or activate vasoactive hormones (ie, norepinephrine, serotonin, bradykinin, angiotensin II), produce substances involved in coagulation and can release endothelium-derived relaxing factors and contracting factors. Nitric oxide and prostacyclin are vasodilators and inhibitors of platelet function. Endothelin is the most potent vasoconstrictor substance known. Thus, the endothelium can profoundly affect platelet adhesion and aggregation, vascular smooth muscle tone and possibly also vascular smooth muscle growth. Under physiological conditions, endothelium-derived relaxing factors appear to dominate. In contrast, in hypertensive and atherosclerotic arteries the release of endothelium-derived relaxing factors and/or the responsiveness of vascular smooth muscle cells to the relaxing factors is reduced, while that of endothelium-derived contracting factors is augmented. This imbalance of endothelium-derived relaxing and contracting factors may be important in the pathogenesis of hypertension and its cardiovascular complications.