Effects of arachidonic acid hydroperoxides on vascular and non-vascular smooth muscle

Dexmedetomidine-induced contraction involves c-Jun NH2 -terminal kinase phosphorylation through activation of the 5-lipoxygenase pathway in the isolated endothelium-denuded rat aorta

Vasoconstriction induced by dexmedetomidine, a highly selective alpha-2 adrenoceptor agonist, mainly involves c-Jun NH2 -terminal kinase (JNK) phosphorylation in the isolated endothelium-denuded aorta. We carried out an in vitro study to determine the main arachidonic acid metabolic pathway that is involved in dexmedetomidine-induced JNK activation. Cumulative dexmedetomidine concentration-contractile response curves were generated in the endothelium-denuded rat aorta in the presence or absence of the following inhibitors: the JNK inhibitor SP600125, the phospholipase A2 inhibitor quinacrine dihydrochloride, the non-specific lipoxygenase (LOX) inhibitor nordihydroguaiaretic acid, the 5-LOX inhibitor AA-861, the dual 5-LOX and cyclooxygenase (COX) inhibitor phenidone, the non-specific COX inhibitor indomethacin, the cytochrome p450 epoxygenase inhibitor fluconazole, the COX-1 inhibitor SC-560, and the COX-2 inhibitor NS-398. The effect of the alpha-2 adrenoceptor inhibitor rauwolscine and other inhibitors, such as quinacrine dihydrochloride, nordihydroguaiaretic acid, AA-861, phenidone, indomethacin and the protein kinase C inhibitor GF 109203X, on dexmedetomidine-induced JNK phosphorylation was investigated in rat aortic vascular smooth muscle cells with western blotting. The effect of dexmedetomidine on 5-LOX and COX-2 expression was investigated in vascular smooth muscle cells. SP600125, quinacrine dihydrochloride, nordihydroguaiaretic acid, AA-861, phenidone, rauwolscine and chelerythrine attenuated dexmedetomidine-induced contraction. Indomethacin slightly attenuated dexmedetomidine-induced contraction. Fluconazole and SC-560 had no effect on dexmedetomidine-induced contraction, whereas NS-398 attenuated contraction. SP600125, rauwolscine, quinacrine dihydrochloride, nordihydroguaiaretic acid, AA-861, phenidone and GF 109203X attenuated dexmedetomidine-induced JNK phosphorylation. 5-LOX and COX-2 were upregulated by dexmedetomidine. Thus, dexmedetomidine-induced alpha-2 adrenoceptor-mediated contraction is mediated mainly by 5-LOX and partially by COX-2, which leads to JNK phosphorylation.

The direct effect of levobupivacaine in isolated rat aorta involves lipoxygenase pathway activation and endothelial nitric oxide release

BACKGROUND

Levobupivacaine is a long-acting local anesthetic with a clinical profile similar to that of racemic bupivacaine but with a greater margin of safety. Levobupivacaine produces dose-dependent vasoconstriction in vivo. Our goal in this in vitro study was to investigate the role of pathways involved in arachidonic acid metabolism in the levobupivacaine-induced contraction of isolated rat aorta and to determine which endothelium-derived vasodilators are involved in the modulation of levobupivacaine-induced contraction.

METHODS

Rat thoracic aortic rings were isolated and suspended for isometric tension recording. Cumulative levobupivacaine dose-response curves over a range of 10(-6) to 3 x 10(-4) M were constructed in 1) aortic rings with no drug pretreatment; 2) endothelium-denuded rings pretreated with quinacrine dihydrochloride (nonspecific phospholipase A(2) inhibitor: 2 x 10(-5), 4 x 10(-5) M), nordihydroguaiaretic acid (NDGA) (lipoxygenase inhibitor: 10(-5), 3 x 10(-5) M), indomethacin (nonspecific cyclooxygenase inhibitor: 10(-5) M), AA-861 (5-lipoxygenase inhibitor: 10(-5), 5 x 10(-5) M), fluconazole (cytochrome P450 epoxygenase inhibitor: 10(-5) M), verapamil (10(-5) M), or calcium-free solution; and 3) endothelium-intact rings pretreated with N(omega)-nitro-L-arginine methyl ester (L-NAME) (nitric oxide synthase inhibitor: 5 x 10(-5) M), indomethacin, or fluconazole. Levobupivacaine-induced contractile response at each concentration (10(-4), 3 x 10(-4) M) was assessed in endothelium-denuded rings. Dose-response curves for potassium chloride in endothelium-denuded rings were generated in the presence or absence of NDGA and AA-861. Intracellular Ca(2+) levels were monitored by Ca(2+) image analysis using Fluo-4 fluorescence in vascular smooth muscle cells treated with levobupivacaine alone or AA-861 plus levobupivacaine.

RESULTS

Levobupivacaine produced a tonic contraction in isolated rat aorta rings; this response was maximal at 10(-4) M levobupivacaine and gradually attenuated at 3 x 10(-4) M levobupivacaine. Levobupivacaine-induced contractions of endothelium-denuded rings were larger than those of endothelium-intact rings. Levobupivacaine-induced contraction of endothelium-denuded rings was attenuated by quinacrine dihydrochloride, NDGA, AA-861, verapamil, and calcium-free solution and, to a lesser extent, by indomethacin. L-NAME enhanced levobupivacaine-induced contraction of endothelium-intact rings and indomethacin slightly attenuated this contraction. NDGA and AA-861 attenuated the potassium chloride-induced contraction. AA-861 attenuated the levobupivacaine-induced intracellular calcium increase in vascular smooth muscle cells.

CONCLUSIONS

Our data indicate that levobupivacaine-induced contraction of rat aortic smooth muscle is mediated mainly by activation of the lipoxygenase pathway and in part by activation of the cyclooxygenase pathway. In addition, activation of the lipoxygenase pathway seems to facilitate calcium influx via L-type calcium channels. Endothelial nitric oxide attenuates levobupivacaine-induced contraction.

The arachidonate 12/15 lipoxygenases. A review of tissue expression and biologic function

12/15-Lipoxygenase is a highly regulated lipid-peroxidating enzyme whose expression and arachidonic acid metabolites are implicated in several important inflammatory conditions including airway and glomerular inflammation as well as atherosclerosis. Tissue expression of the original 12/15-lipoxygenase is well characterized in reticulocytes, eosinophils, airway epithelial cells, and monocytes/macrophages and is likely in other cell systems and tissues under specific conditions. The physiologic role of this family of enzymes is dependent on the context in which it is expressed. In general, the arachidonic acid metabolites antagonize inflammatory responses and counteract the proinflammatory effects of the 5-lipoxygenase pathway. However, certain diHETEs are associaled with pro-inflammatory effects, specifically neutrophilic and eosiniphilic chemotaxis. The direct action of these enzymes on complex lipids and cellular membranes also links them to such significant process as reticulocyte maturation, LDL oxidation in atherosclerosis and pulmonary host defenses. The availability of new specific inhibitors and murine lines that lack expression of the homologous 12-lipoxygenase will allow confirmation of many of these effects with in vivo models of inflammation.

Influence of hypercholesterolaemia on the reactivity of isolated rabbit arteries to 15-lipoxygenase metabolites of arachidonic acid: comparison with platelet-derived agents and vasodilators

The lipoxygenase product 15-hydroxyeicosatetraenoic acid (15-HETE) was shown to be the most important eicosanoid formed in the atherosclerotic rabbit aorta. The aim of the present study was to compare the effects of 15-HETE and its hydroperoxy precursor 15-HpETE with those of other vasoconstrictor and vasodilator agents in arteries from rabbits fed either a control or a cholesterol-rich diet for 16 and 30 weeks. 5-Hydroxytryptamine (5-HT) aggregated platelets and thrombin caused contractions of isolated rabbit aortas. The contractile responses elicited by platelets from control animals were similar to those evoked by platelets from atherosclerotic rabbits. After 16 weeks of hypercholesterolemia, the contractile responses were either augmented (5-HT), unchanged (platelets) or reduced (thrombin). After 30 weeks of hypercholesterolemia, the responses to all contractile agents used had decreased. In both aortas and pulmonary arteries the endothelium-dependent relaxations to the calcium ionophore, A23167, and to acetylcholine were progressively lost and the endothelium-independent relaxations to nitroglycerin were reduced by the progressing hypercholesterolemia. The 15-lipoxygenase metabolites contracted the isolated thoracic aorta and pulmonary artery from control rabbits and to a lesser extent those of the cholesterol-fed rabbits. After raising the tone in these vessels with prostaglandin F2 alpha PGF2 alpha) or noradrenaline, 15-HpETE induced relaxations which were not significantly influenced by the development of fatty streaks. Our data illustrate that the contractions of the blood vessel wall to 15-HETE, like those to other vasoconstrictors, are markedly reduced by developing atherosclerosis. In contrast, the relaxations to 15-HpETE in the rabbit arteries remain unaltered after 16 to 30 weeks of hypercholesterolemia. This is unlike the reactions to other vasodilators, which are markedly reduced.

Histidine attenuates cerebral vasospasm in a rabbit model of subarachnoid hemorrhage

BACKGROUND

Free radical generation following hemolysis of a subarachnoid blood clot is believed to be a key component in the development of cerebral vasospasm. Histidine, an essential amino acid with free radical scavenging characteristics, was examined for its effects on cerebral vasospasm.

METHODS

An experimental rabbit model of subarachnoid hemorrhage-induced vasospasm was used in which autologous arterial blood was injected into the cisterna magna. Basilar arteries were removed following perfusion-fixation two days after the injection of blood, and their cross-sectional luminal areas were measured using computerized image analysis. Rabbits received intravenous injections of L-histidine or vehicle starting 30 min prior to induction of subarachnoid hemorrhage (SAH), with additional injections given four times per day for the next 2 days.

RESULTS

The luminal area of arteries from animals treated with histidine (50 mg/kg/dose or 100 mg/kg/dose) were significantly larger than those from vehicle-treated animals. Relative to the SAH-only groups (mean cross-sectional area = 106.8 x 10(3) microns 2), vasoconstriction was attenuated by 31% in the low dose treatment group (180.0 x 10(3) microns 2) and by 52% in the high dose treatment group (227.4 x 10(3) microns 2). Mean luminal area of control basilar arteries was 340.5 x 10(3) microns 2.

CONCLUSIONS

These findings demonstrate that histidine reduces the amount of cerebral vasospasm occurring subsequent to experimental SAH. It is suggested that the free radical scavenging characteristics of histidine, particularly its ability to scavenge singlet oxygen, may be responsible for the reduction in vasospasm.

Vascular interaction between 5-hydroxytryptamine and 15-lipoxygenase metabolites of arachidonic acid

In isolated canine saphenous veins, the contractions elicited by the 15-lipoxygenase metabolites 15-HETE and 15-HPETE were augmented by 5-hydroxytryptamine (5-HT) in a concentration-dependent way. This potentiation was not mediated by the endothelium nor was it influenced by the 5-HT2-antagonist ketanserin. Phentolamine, however, reduced both the contractions and the potentiation by 5-HT. These data provide evidence for a receptor-mediated potentiation by 5-HT which occurs independently of 5-HT2-receptors. The interaction between 5-HT or aggregating platelets and 15-HPETE was studied in isolated rabbit brachiocephalic arteries. Threshold concentrations of 5-HT and platelets markedly potentiated the contractions elicited by 15-HPETE. In brachiocephalic arteries obtained from cholesterol-fed rabbits, 15-HPETE, 5-HT and platelets caused contractions similar to those obtained in control rabbits. The potentiating effect of 5-HT and platelets on the 15-HPETE-induced contractions was also comparable to that observed in control rabbits. Moreover, no difference was found between control platelets and platelets obtained from hypercholesterolemic rabbits. Our findings demonstrate a positive interaction between 5-HT and 15-lipoxygenase metabolites of arachidonic acid in arteries and veins. This interaction persists in atherosclerotic arteries and could indicate that this mechanism is involved in the genesis of vasospasm.

Vascular relaxation to omega-3 fatty acids: comparison to sodium nitroprusside, nitroglycerin, papaverine, and D600

The vasorelaxant activity of the omega-3 fatty acids--docosahexaenoic (DHA) and eicosapentaenoic (EPA) acids--in comparison with other known vasodilators--sodium nitroprusside, nitroglycerin, papaverine, and D600--were studied in the isolated rat aorta. The relaxant responses of these vasodilators and fatty acids at concentrations of 1-100 microM were assessed in aortic rings contracted with norepinephrine (NE 10(-6) M) or with KCl (30 mM). Cyclic nucleotide enhancers (sodium nitroprusside, nitroglycerin, papaverine) were more effective in producing relaxation, regardless of the contractile mechanism, i.e., alpha-adrenoceptor stimulation or depolarization. In contrast, the omega-3 fatty acids produced augmented relaxation in NE-contracted vessels. Relaxations produced by DHA (15 +/- 2% to 45 +/- 10%) were similar to D600 (16 +/- 2% to 60 +/- 7%) in NE-contracted rings, but not in KCl contracted rings. The responses to D600 and DHA in KCl-contracted vessels were 79 +/- 2% to 104 +/- 3% and 5 +/- 1% to 21 +/- 3%, respectively. In another set of experiments, the effects of omega-3 fatty acids in the presence of albumin were examined; no significant differences in the induced relaxant responses were noted. These results suggest that the mechanisms of vascular relaxation, such as cyclic nucleotide elevation and calcium antagonism of potential-operated channels, are different from those induced by the omega-3 fatty acids.

The role of endothelial cells in the relaxations induced by 13-hydroxy- and 13-hydroperoxylinoleic acid in canine arteries

1. One of the major fatty acids in the arterial wall is linoleic acid. It has been shown that its 13-hydroxy metabolite (13-HODE) is generated in significant amounts by cultured endothelial cells. The aim of the present study was to investigate the relaxations to 13-HODE and its hydroperoxyprecursor (13-HPODE) and to examine the role of the endothelial cells. 2. Ring segments of canine circumflex and splenic artery were mounted in organ chambers for isometric tension recording. During contractions induced by prostaglandin F2 alpha or noradrenaline, 13-HODE and 13-HPODE evoked dose-dependent relaxations. Removal of the endothelial cells reduced the relaxations to 13-HODE, but had no effect on those elicited by 13-HPODE. 3. Indomethacin and meclofenamate (0.3 microM to 30 microM) blocked the relaxations evoked by 13-HODE and 13-HPODE in endothelium-denuded rings. In segments with endothelium, both cyclo-oxygenase inhibitors again abolished the relaxations to 13-HODE, but only diminished those to 13-HPODE. 4. Prostacyclin biosynthesis, as measured by radioimmunoassay, increased upon incubation with 13-HODE and 13-HPODE (10 microM). Bioassay of the release of nitric oxide (NO) indicated that NO was not involved in the relaxations elicited by either metabolite. Moreover, L-NG-nitroarginine (100 microM), a specific inhibitor of NO synthesis, did not influence the relaxations to 13-HODE and 13-HPODE. The responses to 13-HPODE were also not altered by superoxide dismutase. 5. In the splenic artery 13-HPODE and 13-HODE induced contractions above 3 microM which were blocked by the thromboxane receptor antagonist, daltroban.In the circumflex artery contractile responses to high concentrations of 13-HODE could be observed only after inhibition of cyclo-oxygenase.6. We conclude that the vasodilatation induced by 13-HODE and 13-HPODE was due to stimulation of prostacyclin biosynthesis both in the endothelium and smooth muscle cells or other subendothelial structures. An additional, unidentified intermediate, which was neither NO nor a cyclo-oxygenase product nor superoxide anion, contributed to the relaxations to 13-HPODE in arteries with endothelium.

Vascular responses to leukocyte products in atherosclerotic primates

Little is known about the possible role of leukocytes in the pathogenesis of vasospasm. We hypothesized that vasoactive products released by leukocytes might produce constriction of atherosclerotic arteries. To test this hypothesis, we infused fmet-leu-phe (fMLP), a peptide that activates leukocytes to release their vasoactive products, into the perfused hind limb of normal and atherosclerotic cynomolgus monkeys. Infusion of fMLP did not change resistance of large arteries in normal monkeys. In contrast, fMLP produced pronounced constriction of large arteries in atherosclerotic monkeys. To determine whether leukotrienes, platelet-activating factor, or prostaglandin E2 (PGE2), which are released by leukocytes, may contribute to leukocyte-induced vasoconstriction in atherosclerotic monkeys, we injected leukotriene D4, platelet-activating factor, and PGE2 intra-arterially into the perfused hind limb. Leukotriene D4 and platelet-activating factor had minimal effects on large arteries in both normal and atherosclerotic monkeys. PGE2 produced marked constriction of large arteries in atherosclerotic, but not normal, monkeys. Thus, pronounced constriction in atherosclerotic, but not normal, arteries during infusion of fMLP suggests that products released by leukocytes may mediate vasoconstriction in atherosclerotic vessels. Vasoconstrictor responses to PGE2 are profoundly potentiated by atherosclerosis, which suggests that PGE2 may contribute to leukocyte-induced vasoconstriction.

Histamine stimulation of prostaglandin and HETE synthesis in human endothelial cells

Endothelial cells (EC) cultured from human umbilical artery (UA) and vein (UV) metabolized [14C]arachidonic acid to prostaglandins (PGs), monohydroxyeicosatetraenoic acids (HETEs), and epoxyeicosatrienoic acids (EETs). Major radioactive products were identified as 6-keto-PGF1 alpha, PGE2, PGF2 alpha, 12-hydroxy heptadecatrienoic acid, 15-HETE, and 11-HETE. In addition, extracts from UV ECs contained 12-HETE, 5-HETE, 14,15-EET, and 5,6-EET as minor products, whereas extracts from UA ECs contained only 12-HETE as a minor product. UA ECs also produced metabolites comigrating with 14,15-EET, 11,12-EET, 8,9-EET, and 5,6-EET. Histamine increased the release of [14C]PGs and [14C]HETEs from [14C]arachidonic acid-labeled ECs. Indomethacin, aspirin, and nordihydroguauretic acid completely inhibited synthesis of both [14C]PGs and [14C]HETEs from exogenous [14C]arachidonic acid in these cells. Microsomes metabolized [14C]arachidonic acid to the same [14C]PGs and [14C]HETEs as intact cells. Pretreatment of microsomes with indomethacin completely inhibited formation of these products. These data indicate that UA ECs and UV ECs metabolize endogenous and exogenous arachidonic acid to both PGs and HETEs. Also 15-HETE and 11-HETE appear to be synthesized by a microsomal enzyme with the properties of cyclooxygenase.

Free radical reaction and biological defense mechanism in the pathogenesis of prolonged vasospasm in experimental subarachnoid hemorrhage

The relationship between free radical reactions and the defense mechanisms against them was investigated in the pathogenesis of prolonged vasospasm following experimental subarachnoid hemorrhage (SAH) in dogs. The concentration of lipid peroxides in the cerebro spinal fluid (CSF) increased markedly up to the eighth day following SAH; the concentrations also rose in the arterial wall (p less than 0.01) and the gray matter of the temporal lobe where the subarachnoid blood clots were (p less than 0.01). On the other hand, the activity of superoxide dismutase (SOD) decreased significantly up to the eighth day after SAH (p less than 0.01), and there was a gradual increase of glutathione peroxidase (GSH-px) in the CSF. In the arterial wall, there was a slight decrease in the activity of SOD, a significant decrease in the activity of GSH-px (p less than 0.01), and also a significant decrease in the concentration of glutathione (p less than 0.01) up to the eighth day following SAH. In conclusion, lipid peroxidation with insufficient biological defense mechanisms against it in the arterial wall, concomitant with that in the CSF, might take part in the genesis of prolonged vasospasm following SAH.

Cultured bovine coronary arterial endothelial cells synthesize HETEs and prostacyclin

Arachidonic acid metabolism was examined in endothelial cells cultured from bovine coronary arteries. In culture, these cells exhibit specific characteristics of endothelial cells. They form a contact-inhibited monolayer with a cobblestone appearance, contain immunoreactive von Willebrand's factor antigen, and have angiotensin I converting enzyme activity. Prostacyclin was the major prostaglandin synthesized from exogenous and endogenous arachidonic acid in these cells. In addition, exogenous arachidonic acid was metabolized to small amounts of prostaglandin E2 (PGE2) and several relatively nonpolar metabolites including 12-, 15-, and 11-hydroxyeicosatetraenoic acids (12-, 15-, and 11-HETE). Histamine, bradykinin, and thrombin increased PGI2 synthesis in these bovine coronary endothelial cells. Of these agonists, bradykinin was the most potent, increasing basal PGI2 release by fourfold. More vigorous stimulation of the cells with mechanical disruption of the cell monolayer, melittin, or A23187 resulted in release of both PGI2 and PGE2. Pretreatment of cells with exogenous arachidonic acid (10(-5) M) abolished their responsiveness to subsequent stimulation by arachidonic acid or vasoactive agents, but not PGH2. Furthermore, treatment of cells with 15-HPETE (10(-7)-10(-4) M), but not 15-HETE, specifically inhibited basal as well as A23187-stimulated PGI2 release. PGE2 release was increased slightly after 15-HPETE treatment. These studies indicate that bovine coronary endothelial cells can metabolize arachidonic acid to several biologically active products and that PGI2 synthesis by these cells is specifically related to the type of vasoactive agent employed. Both the qualitative pattern and quantity of eicosanoids synthesized by bovine coronary endothelial cells differ substantially from endothelial cells isolated from noncardiac vascular beds.

Prostaglandin synthesis in rat embryo tissue: the effect of non-steroidal anti-inflammatory drugs in vivo and ex vivo

Aspirin and salicylate are well-known but poorly understood teratogens in laboratory animals. Because aspirin inhibits PG synthesis, we systematically examined PG synthesis in rat embryo homogenates, the inhibition of PG synthesis in vivo and ex vivo by various non-steroidal anti-inflammatory drugs, and tested the hypothesis that the inhibition of PG synthesis is responsible for aspirin-induced limb defects in rats. We report that embryonic rat homogenates synthesize 6-keto-PGF1alpha, PGE, and PGF in large amounts from endogenous substrate, that aspirin and other non-steroidal anti-inflammatory drugs inhibit PG synthesis in vitro but not necessarily in vivo, and that contrary to our original hypothesis, the inhibition of PG synthesis is likely not responsible for aspirin-induced limb defects in rats.

Effect of thyrotropin-releasing hormone on lipoxygenase-induced hypotension in the unanesthetized Guinea pig

Soybean lipoxygenase, an enzyme which catalyzes the formation of the vasoactive lipid 15-hydroperoxy eicosatetraenoic acid (15-HPETE) from arachidonic acid, was administered to unanesthetized guinea pigs previously prepared with indwelling vascular cannulae for continuous cardiovascular monitoring. Administration of this enzyme (150 mg/kg IV) resulted in profound hypotension in this model, but no cardiovascular change was observed after administration of equal weight or equimolar amounts of another protein (ovalbumin). The lipoxygenase-induced hypotension, moreover, was promptly reversed by the peptide thyrotropin-releasing hormone (TRH) (2 mg/kg IV) but not by the opiate receptor antagonist naloxone (2 mg/kg IV). This TRH-naloxone dissociation was comparable to that previously observed in hypotension produced by leukotriene D4 (LTD4), platelet-activating factor (PAF), or antigen-induced anaphylaxis in the same species. Thus, although its properties as a "physiologic" opiate antagonist led to the early trials of TRH in endotoxic, hypovolemic and spinal shock, it is now apparent that TRH reverses several other forms of experimental shock, including that caused by lipoxygenase, through non-endorphin-related mechanism.

Vessel wall arachidonate metabolism after angioplasty: possible mediators of postangioplasty vasospasm

The mechanism of postangioplasty vasospasm is uncertain. It was postulated that a reduction in vasodilator prostaglandin I2 (PGI2) or prostaglandin E2 (PGE2) or an increase in vasoconstrictor hydroperoxy acids might contribute to spasm of a dilated artery. Twelve mongrel dogs were anesthetized, intubated, and mechanically ventilated. Heart rate and aortic pressures were continuously monitored and arterial blood gases maintained within physiologic limits. A single carotid artery was dilated in each animal (4 atm X 1 minute X 3) using nonexpandable polyethylene balloon catheters with inflated balloon diameters 50 to 100% larger than the internal arterial lumen. The opposite carotid artery served as a control. In 4 animals, aspirin (10 mg/kg, intravenously) was injected 30 minutes before dilation. Sixty minutes after dilation, animals were heparinized and the carotid arteries carefully removed. The in vitro conversion of carbon-14(14C)-arachidonic acid (AA) to 6-keto PGF1 alpha (PGI2), PGE2, and 12L-hydroxy-5,8,10,14-eicosatetraenoic acid (HETE) was determined using thin-layer radiochromatography. Angioplasty caused a 70% decrease in vessel wall PGI2 production and a 44% decrease in PGE2 production (n = 4, p less than 0.05). Reductions in in vitro conversion of 14C-AA to PGI2 and PGE2 induced by angioplasty were comparable to that produced by cyclooxygenase inhibition with aspirin. Angioplasty, in addition, caused a 104% increase in vessel wall HETE production (n = 4, p less than 0.05). Therefore, angioplasty results in a local derangement of AA metabolism characterized by decreases in vasodilator prostaglandins and increases in vasoconstrictor hydroperoxy acids. These local changes may contribute, in part, to sudden arterial occlusion after angioplasty.