Cyclic strain upregulates nitric oxide synthase in cultured bovine aortic endothelial cells

In vivo, endothelial cells (EC) are subjected to hemodynamic forces which may influence the production of nitric oxide. This study was designed to examine the effect of cyclic strain on the expression of endothelial nitric oxide synthase (eNOS) in cultured bovine aortic EC. EC were grown on flexible membranes which were subjected to deformation at 60 cycles/min with -5 or -20 kPa of vacuum. This results in an average strain of 6 and 10%, respectively, which is transmitted to the attached cells. Northern blot analysis of total cytosolic RNA demonstrated an increase in eNOS gene expression with both strain regimens but the increase with 10% average strain was greater than that at 6%. Nuclear runoff transcription assays confirmed the induction of eNOS transcripts. Western blot analysis showed an increase in eNOS level after 24 h of cyclic 10% average strain compared with controls or 6% average strain. Immunohistochemical staining of EC for eNOS was increased in the high strain periphery (7-24% strain) of membranes deformed with -20 kPa vacuum. These results demonstrate that cyclic strain upregulates the expression of eNOS transcripts and protein levels in bovine aortic EC thus emphasizing the importance of hemodynamic forces in the regulation of eNOS in vivo.

The Golgi association of endothelial nitric oxide synthase is necessary for the efficient synthesis of nitric oxide

The particulate enzyme, endothelial nitric oxide synthase (eNOS), produces nitric oxide to maintain normal vasodilator tone in blood vessels. In this study, we demonstrate that eNOS is a Golgi-associated protein in cultured endothelial cells and intact blood vessels. Using a heterologous expression system in HEK 293 cells, we show that wild-type myristoylated and palmitoylated eNOS, but not mutant, non-acylated eNOS targets to the Golgi. More importantly, HEK 293 cells expressing wild-type eNOS release substantially more NO than cells expressing the mutant, non-acylated enzyme. Thus, eNOS is a novel Golgi-associated protein, and Golgi compartmentalization is necessary for the enzyme to respond to intracellular signals and produce NO.

Functional analysis of the human endothelial nitric oxide synthase promoter. Sp1 and GATA factors are necessary for basal transcription in endothelial cells

To gain insights into the mechanisms of endothelial nitric oxide synthase (eNOS) gene expression, we have cloned the eNOS promoter and fused it to a luciferase reporter gene to map regions of the promoter important for basal transcription in bovine aortic endothelial cells (BAEC). Transfection of BAEC with F1 luciferase (LUC) (-1600 to +22 nucleotides) yielded a 35-fold increase in promoter. Progressive deletion from -1600 to -1033 (F2 and F3 LUC) did not significantly influence eNOS promoter activity. Further deletion from -1033 to -779 (F4 LUC) resulted in an approximate 40% reduction in basal promoter activity, and still further deletion from -779 to -494 (F5 LUC) did not markedly influence activity. Deletion from -494 to -166 (F6 LUC) reduced eNOS promoter activity by 40-50%. Specific mutation of the consensus GATA site (-230) in the F3 LUC construct reduced luciferase activity (by 25-30%). Gel shift analysis and antibody depletion using BAEC nuclear extracts demonstrated in vitro binding of GATA-2 to the oligonucleotide sequence containing the -230 GATA site. Next, we mutated the Sp1 site (-103) in the F3 and F6 LUC constructs and in the F3 GATA mutant construct. Expression of these Sp1 mutants in BAEC resulted in a 85-90% reduction in normalized luciferase activity. Gel shift and antibody supershift analysis using a BAEC nuclear extracts demonstrated four specific, DNA-protein complexes binding to the eNOS Sp-1 site, with the slowest migrating form composed of Sp1 and another nuclear protein. These data demonstrate that the Sp1 site is an important cis-element in the core eNOS promoter.

Cyclic strain increases endothelial nitric oxide synthase activity

BACKGROUND

Endothelial nitric oxide synthase (eNOS) is an important enzyme that controls the production of a potent vascular smooth muscle relaxing factor, nitric oxide. However, the role of hemodynamic forces (blood pressure, cyclic strain, and shear stress) on the regulation of eNOS has not been fully elucidated. Recently, we showed that cyclic strain increases eNOS gene and protein in cultured bovine aortic endothelial cells (EC). Because an increase in gene transcription and protein synthesis may not necessarily translate into an increase in functional activity, the aim of this study was to determine the effects of cyclic strain on eNOS activity.

METHODS

EC were seeded onto plates with flexible bottoms that can be deformed by vacuum and were then exposed to 60 cycles/minute of either 24% maximum strain (-20 kPa vacuum) or 10% maximum strain (-5 kPa vacuum) for 24 hours. eNOS activity was assessed, and nitric oxide production was determined (as nitrite) by the Greiss reaction.

RESULTS

Twenty-four percent strain, at 60 cycles/min, but not 10% strain significantly increases eNOS activity compared with stationary controls. Both strain regimens increased nitric oxide (as nitrite) in culture media compared with stationary controls, although nitrite in media of EC exposed to high strain were significantly increased compared with the lower strain.

CONCLUSIONS

Cyclic strain increases eNOS activity in cultured bovine aortic EC. These results may indicate the importance of hemodynamic forces in the regulation of eNOS in vivo.

Regulation of nitric oxide synthesis by proinflammatory cytokines in human umbilical vein endothelial cells. Elevations in tetrahydrobiopterin levels enhance endothelial nitric oxide synthase specific activity

We have examined cytokine regulation of nitric oxide synthase (NOS) in human umbilical vein endothelial cells (HUVEC). 24-h treatment with IFN-gamma (200 U/ml) plus TNF (200 U/ml) or IL-1 beta (5 U/ml) increased NOS activity in HUVEC lysates, measured as conversion of [14C]L-arginine to [14C]L-citrulline. Essentially, all NOS activity in these cells was calcium dependent and membrane associated. Histamine-induced nitric oxide release, measured by chemiluminescence, was greater in cytokine-treated cells than in control cells. Paradoxically, steady-state mRNA levels of endothelial NOS fell by 94 +/- 2.0% after cytokine treatment. Supplementation of HUVEC lysates with exogenous tetrahydrobiopterin (3 microM) greatly increased total NOS activity, and under these assay conditions, cytokine treatment decreased maximal NOS activity. IFN-gamma plus TNF or IL-1 beta increased endogenous tetrahydrobiopterin levels and GTP cyclohydrolase I activity, the rate-limiting enzyme of tetrahydrobiopterin synthesis. Intracellular tetrahydrobiopterin levels were higher in freshly isolated HUVEC than in cultured cells, but were still limiting. We conclude that inflammatory cytokines increase NOS activity in cultured human endothelial cells by increasing tetrahydrobiopterin levels in the face of falling total enzyme; similar regulation appears possible in vivo.

The nitric oxide synthase family of proteins

Nitric oxide (NO) is an important bio-regulatory molecule in the nervous, immune and cardiovascular systems. NO is synthesized from one of the guanidino nitrogens of L-arginine by the enzyme nitric oxide synthase (NOS). To date, several isoforms of NOS have been purified and cloned. These proteins represent a novel family of mammalian enzymes that contain both heme and cytochrome P450 reductase domains. The three prototypical forms of NOS: neuronal, cytokine-inducible and endothelial NOS, are derived from separate genes and are regulated by diverse signaling pathways. The purposes of this review are to highlight recent advances in the enzymology and molecular biology of this important family of proteins and to examine how this information pertains to the regulation of NO production.

Identification of covalently bound amino-terminal myristic acid in endothelial nitric oxide synthase

Endothelial nitric oxide synthase (eNOS) is unique among the nitric oxide synthase family of proteins due to the presence of an N-myristoylation consensus sequence elucidated from the cloning of its cDNA. Although eNOS was metabolically labeled with [3H]myristic acid and mutation of glycine 2 in the N-myristoylation consensus sequence changed the particulate localization of the enzyme to a cytosolic form, the definitive characterization of eNOS as an N-myristoylprotein has not been demonstrated. Therefore, the purpose of the present study was to determine the nature of the fatty acid incorporated into eNOS. Wild-type or G2A mutant (mutation of glycine 2, the myristic acid acceptor site, to alanine) eNOS-transfected COS cells and bovine aortic endothelial cells (BAEC) were metabolically labeled with [3H]myristic acid for 5 h. The radiolabel was primarily incorporated into membrane-associated eNOS from wild-type transfected COS cells and cultured BAEC but not into the mutant eNOS from G2A-transfected COS cells. Qualitatively similar amounts of immunoreactive protein were found in wild-type and G2A-transfected cells. In addition, linkage of the radiolabel to eNOS was insensitive to hydroxylamine treatment, and incorporation of the radiolabel into eNOS was abolished by cyclo-heximide. Chemical analysis of the fatty acid released by acid methanolysis of labeled eNOS verified the 3H-labeled fatty acid as protein-bound myristic acid. These results unequivocally demonstrate that eNOS incorporates myristic acid via an amide linkage with the amino-terminal glycine of the enzyme as a co-translational modification.

Chronic exercise in dogs increases coronary vascular nitric oxide production and endothelial cell nitric oxide synthase gene expression

Recently, we have shown that chronic exercise increases endothelium-derived relaxing factor (EDRF)/nitric oxide (NO)-mediated epicardial coronary artery dilation in response to brief occlusion and acetylcholine. This finding suggests that exercise can provide a stimulus for the enhanced production of EDRF/NO, thus possibly contributing to the beneficial effects of exercise on the cardiovascular system. Therefore, the purpose of the present study was to examine whether chronic exercise could influence the production of NO (measured as the stable degradation product, nitrite) and endothelial cell NO synthase (ECNOS) gene expression in vessels from dogs after chronic exercise. To this end, dogs were exercised by running on a treadmill (9.5 km/h for 1 hour, twice daily) for 10 days, and nitrite production in large coronary vessels and microvessels and ECNOS gene expression in aortic endothelial extracts were assessed. Acetylcholine (10(-7) to 10(-5) mol/L) dose-dependently increased the release of nitrite (inhibited by nitro-L-arginine) from coronary arteries and microvessels in control and exercised dogs. Moreover, acetylcholine-stimulated nitrite production was markedly enhanced in large coronary arteries and microvessels prepared from hearts of dogs after chronic exercise compared with hearts from control dogs. One potential mechanism that may contribute to the enhanced production of nitrite in vessels from exercised dogs may be the induction of the calcium-dependent ECNOS gene. Steady-state mRNA levels for ECNOS were significantly higher than mRNA levels for von Willebrand's factor (vWF, a specific endothelial cell marker) and glyceraldehyde-3-phosphate dehydrogenase (GAPDH, a constitutively expressed gene) in exercised dogs.(ABSTRACT TRUNCATED AT 250 WORDS)

Genomic analysis and expression patterns reveal distinct genes for endothelial and brain nitric oxide synthase

Constitutively active nitric oxide synthases (NOS) are a unique class of NADPH-dependent, calcium/calmodulin-dependent enzymes that catalyze the conversion of L-arginine to nitric oxide and L-citrulline. However, little is known about the molecular similarities or differences between the two prototypical constitutive NOS enzymes, endothelial NOS (ECNOS) and brain NOS (bNOS). The aims of this study were to begin characterizing the gene structure and tissue distribution of messenger RNAs (mRNAs) for ECNOS and bNOS and to examine the immunological resemblance of the proteins by Western blotting. Full-length complementary DNAs (cDNAs) encoding bovine ECNOS and rat bNOS hybridized, under high stringency, to different-sized fragments of endonuclease-digested bovine, rat, and human genomic DNA. In addition, more than one fragment was detected with both cDNAs, suggesting that ECNOS and bNOS genes contained multiple introns. Tissue distribution of ECNOS mRNA (4.4 kb) and bNOS mRNA (9.5 kb) in the rat was detected by Northern blotting. Patterns among tissue extracts were strikingly different, with ECNOS mRNA being most abundant in aorta, heart, lung, kidney, adrenal gland, spinal cord, and urogenital tissues and bNOS mRNA most prominent in brain regions, intestine, stomach, spinal cord, adrenal gland, and aorta. Interestingly, ECNOS cDNA detected two equally abundant RNA transcripts (4.4 and 4.0 kb) in most brain regions tested, suggesting an alternative splicing of the ECNOS pre-mRNA. Western blotting, using an ECNOS monoclonal antibody, recognized ECNOS protein from native bovine endothelial cells, cultured bovine endothelial cells, and COS cells transfected with ECNOS cDNA but did not recognize purified bNOS.(ABSTRACT TRUNCATED AT 250 WORDS)

Mutation of N-myristoylation site converts endothelial cell nitric oxide synthase from a membrane to a cytosolic protein

Endothelial cell nitric oxide synthase (ECNOS) is a membrane-associated enzyme that generates endothelium-derived relaxing factor/nitric oxide (EDRF/NO) from L-arginine. We have suggested, from the cloning of the bovine ECNOS cDNA, that the presence of an N-myristoylation consensus sequence may impart its membrane localization since cytosolic forms of NOS do not contain such domains. To test the hypothesis that N-myristoylation is necessary for particulate ECNOS, we performed site-directed mutagenesis of the myristic acid acceptor site, Gly-2, and changed the glycine codon to alanine by a single nucleotide substitution. Expression of wild-type ECNOS in COS cells resulted in greater than 95% of the enzymatic activity in crude membrane fractions (as measured by the conversion of [3H]L-arginine to [3H]L-citrulline). In contrast, expression of the Gly-2 to Ala-2 mutant (G2A) demonstrated 8% ECNOS activity in membranes and 92% in the cytosol. The back mutation (from Ala-2 to Gly-2, A2G) restored ECNOS activity to the particulate fraction as seen with the wild type. Both wild-type membrane ECNOS and cytosolic G2A ECNOS activities were dependent on NADPH and calcium and were inhibited to the same extent by NG-monomethyl L-arginine (L-NMMA) and NG-nitro-L-arginine methyl ester (L-NAME). Moreover, kinetic analysis of these enzymes revealed similar Kms for L-arginine (2-4 microM, n = 3), demonstrating that the mutation did not affect ECNOS function. Thus, N-myristoylation is necessary for the membrane localization of ECNOS and may be of special significance for the basal or flow-induced production of NO by the endothelium.

Molecular cloning and expression of a cDNA encoding endothelial cell nitric oxide synthase

Endothelium-derived relaxing factor (EDRF), identified as nitric oxide (NO), is derived from a guanidino nitrogen of L-arginine via its metabolism by nitric oxide synthase (NOS). Herein, we report the molecular cloning of a cDNA encoding the constitutive calcium-calmodulin (Ca2+/CaM)-regulated nitric oxide synthase (ECNOS). A full-length ECNOS clone was isolated by screening a bovine aortic endothelial cell cDNA library using a fragment of rat brain NOS (bNOS) cDNA. This cDNA has an open reading frame of 3615 nucleotides encoding a 1205-amino acid protein. Membranes prepared from COS cells transfected with the ECNOS cDNA demonstrated NADPH- and Ca2+/CaM- dependent conversion of L-, but not D-, arginine to NO and citrulline that was inhibited by NG-nitro-L-arginine methyl ester. Comparison of the deduced amino acid sequence of ECNOS to the bNOS and macrophage NOS (Mac-NOS) sequences revealed 57 and 50% identity, respectively. In addition, ECNOS contains a unique N-myristylation consensus sequence (not shared by bNOS or Mac-NOS) that may explain its membrane localization.

Role of intracellular thiols in release of EDRF from cultured endothelial cells

The availability of intracellular reduced thiols, such as L-cysteine or glutathione (GSH), may be critically important for the biosynthesis of endothelium-derived relaxing factor (EDRF). We have, therefore, investigated the effects of various sulfhydryl (SH) reagents, such as 1-chloro-2,4-dinitrobenzene (CDNB), diamide, 2,2'-dithiodipyridine (DTDP), or N-ethyl-maleimide (NEM), on the release of EDRF from cultured endothelial cells. None of the SH reagents tested affected the flow-induced EDRF release, but DTDP and NEM inhibited the release of EDRF stimulated by ADP, ionomycin, or poly-L-lysine. In contrast, NG-nitro-L-arginine methyl ester, an inhibitor of EDRF biosynthesis, blocked both the flow- and agonist-induced release of EDRF. Although NEM substantially potentiated the flow-induced release of prostacyclin (PGI2), probably due to a blockade of the reacylation of arachidonic acid, it inhibited the stimulated release of PGI2, whereas diamide did not significantly affect either release. Like CDNB or diamide, NEM, but not DTDP, caused a significant decrease in endothelial GSH. In contrast, both NEM and DTDP, but not CDNB or diamide, inhibited the ADP-induced mobilization of intracellular calcium, suggesting that they act on specific target proteins involved in endothelial cell calcium homeostasis rather than intracellular free SH groups. Moreover, the selective inhibition by these two SH reagents of the stimulated release of EDRF implies that a fundamental regulatory difference exists between agonist- and flow-induced EDRF biosynthesis.

High endothelin-1 immunoreactivity in Crohn's disease and ulcerative colitis

Both immunological hypersensitivity and vascular abnormalities have been implicated in the pathogenesis of inflammatory bowel disease. In an attempt to link the two hypotheses, we sought evidence of local production of endothelin-1, a potent vasoconstrictor, in patients with Crohn's disease and ulcerative colitis. An immunohistochemical method was used to detect endothelin-1 in tissue samples from sixteen Crohn's disease patients, nine ulcerative colitis patients, and thirteen controls. In the controls, positively staining cells were infrequent in both lamina propria (mean 0.9% of total cells, 95% confidence interval 0.1-1.7%) and submucosa (2.3%, 0.4-4.1%). The percentage of endothelin-immunoreactive cells was significantly higher in the two disease groups than in the controls. Among the Crohn's disease patients, there were more immunoreactive cells in the submucosa than in the lamina propria (19.1%, 15.2-22.1% vs 12.3%, 8.1-16.5%; p less than 0.001), whereas the converse was true for the ulcerative colitis group (8.6%, 1.1-16.1% vs 24.4%, 14.1-34.6%; p less than 0.001). Immunoreactive macrophage aggregates around submucosal blood vessels were common in samples from Crohn's disease patients. Endothelin concentrations, measured by radioimmunoassay, in supernatants of homogenised tissue samples were significantly higher in Crohn's disease and ulcerative colitis than in controls. We suggest that local endothelin production by inflammatory cells may contribute to vasculitis in chronic inflammatory bowel disease by inducing intestinal ischaemia through vasoconstriction.

Nitroxergic nerves mediate vagally induced relaxation in the isolated stomach of the guinea pig

Here we show that the relaxation induced by stimulation of the vagus nerve in the presence of cholinergic (muscarinic) and adrenergic blockade in the isolated stomach of the guinea pig is mediated by nitric oxide (NO). This is substantiated by inhibition of vagal relaxation by NG-monomethyl-L-arginine, an inhibitor of NO synthesis. The effect of NG-monomethyl-L-arginine was partially reversed by coincubation with L-arginine but not with D-arginine. NO activates soluble guanylate cyclase, and relaxation of the stomach induced by vagal stimulation was prevented by an inhibitor of soluble guanylate cyclase, methylene blue, further supporting our conclusions. The relaxant effect of vagal stimulation was also ablated by hexamethonium, an inhibitor of ganglionic nicotinic receptors, thereby showing that ganglionic transmission did not rely on NO, through its release from preganglionic neurons. However, hexamethonium did not inhibit the gastric relaxation brought about by increasing the intragastric pressure, which is also mediated by NO as previously described by us. The selective inhibition by hexamethonium of only the vagally mediated relaxation but not of the pressure-induced relaxation of the stomach indicates the existence of at least two separate neuronal pathways able to generate NO and bring about gastric accommodation of food or fluid.

Nitric oxide and another potent vasodilator are formed from NG-hydroxy-L-arginine by cultured endothelial cells

The hypothesis was investigated that NG-hydroxy-L-arginine (L-HOArg) is an intermediate in the biosynthesis of nitric oxide (.NO) from L-arginine (L-Arg) by the constitutive .NO synthase (NOS) present in endothelial cells (ECs). When infused through a column of bovine aortic ECs on beads, either L-HOArg or D-HOArg (1-10 microM) substantially potentiated relaxations of the bioassay tissues to .NO released from the cells by ADP or bradykinin, and this effect was abolished by coinfusions of NG-nitro-L-arginine (L-NO2Arg) methyl ester (10 microM) or NG-monomethyl-L-arginine (L-MeArg; 30 microM). Both L-HOArg and D-HOArg, irrespective of the presence of ECs, also potentiated relaxations induced by authentic .NO, but not glyceryl trinitrate. This was due to a rapid chemical reaction of either isomer with .NO, resulting in the formation of a potent and more stable vasodilator. When infusions of L-HOArg (3 microM) were consequently made in the presence of D-HOArg (10 microM), the L-isomer no longer had any effect on relaxations induced by authentic .NO, but significantly increased the stimulated release of .NO from the column of ECs. The conclusion that L-HOArg is a substrate for the constitutive NOS in cultured ECs was strongly supported by the L-NO2Arg-sensitive conversion of L-HOArg, but not D-HOArg, to .NO by NOS preparations from these cells. Interestingly, cultured ECs produced from L-HOArg (greater than or equal to 3 microM), but not D-HOArg, a stable vasodilator, the effects of which were inhibited by oxyhemoglobin (0.3-3 microM). However, the formation of this substance was not prevented by L-NO2Arg methyl ester (10 microM) or L-MeArg (10-100 microM), suggesting an enzymatic pathway different from NOS.

Inhibition of human platelet aggregation by endothelium-derived relaxing factor, sodium nitroprusside or iloprost is potentiated by captopril and reduced thiols

We have examined whether inhibitors of angiotensin-converting enzyme-containing sulfhydryl groups such as, captopril (CPT) or SQ 14,534, the nonsulfhydryl-containing angiotensin-converting enzyme inhibitors, teprotide (TPR) or enalaprilat (ENA) and other structurally unrelated sulfhydryl-containing compounds, N-2-mercaptopropionylglycine (MPG) or N-acetyl-L-cysteine (NAC), could influence platelet aggregation. Incubation of human washed platelets with CPT, SQ 14,534, TPR, ENA, MPG or NAC (0.1-0.5 mM) did not modify their aggregatory responses to thrombin. However, the antiaggregatory properties of endothelial cells cultured from bovine aorta were potentiated by CPT, SQ 14,534, MPG or NAC but not by TPR or ENA (40-100 microM). CPT (100-500 microM) or NAC (50-200 microM) but not ENA (100 and 500 microM) also potentiated the antiaggregatory effects of sodium nitroprusside (1.0 microM) or iloprost (0.2 nM). The ability of the thiol-containing compounds (CPT or NAC) to potentiate the antiaggregatory effects of sodium nitroprusside or iloprost was not associated with an elevation of platelet cyclic GMP or cyclic AMP levels, respectively. Thus, CPT and other sulfhydryl-containing compounds can synergize with antiplatelet compounds, thereby enhancing the ability of endothelial-derived autocoids to inhibit platelet aggregation. The mechanism responsible for this potentiating effect of thiols on platelet aggregation is not known, but may relate to the ability of thiol-containing compounds to act as intracellular scavengers of oxygen-derived free radicals.

Arachidonic acid elicits endothelium-dependent release from the rabbit aorta of a constrictor prostanoid resembling prostaglandin endoperoxides

This study was designed to investigate the mediator(s) of endothelium-dependent arterial constrictor responses evoked by arachidonic acid in vitro. A segment of descending rabbit thoracic aorta was isolated and perfused (1-2 ml/min) with oxygenated Krebs' bicarbonate buffer. Changes in the vascular smooth muscle-contracting activity of the aortic effluent were detected by superfusion bioassay using either strips of rabbit aorta or rings of dog saphenous vein, both denuded of endothelium and exposed to indomethacin (10 microM). Arachidonic acid (5-50 micrograms) injected into the inflow of the perfused aorta caused a dose-related increase in the vascular smooth muscle-contracting activity of the aortic effluent, whereas arachidonic acid added directly into the aortic effluent did not. The arachidonic acid-induced elevation of vascular smooth muscle-contracting activity in the aortic effluent was not apparent when indomethacin (10 microM) was added to the aortic inflow to inhibit cyclooxygenase, when the endothelium of the perfused aorta was removed by rubbing, or when the thromboxane A2/prostaglandin H2 receptors of the vascular tissues used for bioassay were blocked with an antagonist (1 microM SQ29548), and was unaffected when an inhibitor of thromboxane synthase (10 microM CGS 13080) was added to the aortic inflow. This effect of arachidonic acid was accompanied by release of prostaglandin H2 (measured as prostaglandin F2 alpha after reduction with SnCl2) in amounts sufficient to elicit contraction of the vascular tissues used for bioassay and was attenuated when a reducing agent (2 mM FeCl2) that converts prostaglandin H2 to 12-heptadecatrienoic acid was added to the aortic effluent. Collectively, these observations suggest that arachidonic acid stimulates endothelium-dependent release from the perfused aorta of a prostanoid that contracts vascular smooth muscle via interaction with thromboxane A2/prostaglandin H2 receptors. The study also suggests that the prostanoid responsible for the vascular smooth muscle-contracting activity of the aortic effluent is a prostaglandin endoperoxide(s) rather than thromboxane A2.

Involvement of nitric oxide in the reflex relaxation of the stomach to accommodate food or fluid

The fundus of the guinea-pig stomach actively dilates in response to low increases in intragastric pressure. This physiological response, now called adaptive relaxation, accommodates the intake of liquid or food. It is independent of external innervation, resistant to ganglion blockade, but reflex in origin. The nerves involved are neither adrenergic nor cholinergic in nature. Non-adrenergic, non-cholinergic (NANC) nerves have now been recognized in many parts of the gastrointestinal tract and have recently been linked with release of nitric oxide (NO) on electrical stimulation. Here we show that adaptive relaxation in isolated stomach of the guinea pig is mediated by a NANC neurotransmitter substance indistinguishable from NO derived from L-arginine. This is substantiated by inhibition of adaptive relaxation by NG-monomethyl-L-arginine or N omega-nitro-L-arginine methyl ester, both inhibitors of NO synthesis, and by methylene blue, an inhibitor of soluble guanylate cyclase. There are two distinct neuronal pathways signalling NO-dependent adaptive relaxation, as evidenced by tetrodotoxin sensitivity. The first is a local reflex arc, the afferent fibres of which sense changes in intragastric pressure. The second is stimulated by an agonist for ganglionic nicotinic receptors. Thus, the functional significance of NO release from NANC nerves in the stomach is to bring about adaptive relaxation through a reflex response to increases in intragastric pressure.

The biosynthesis of endothelin-1 by human polymorphonuclear leukocytes

Human polymorphonuclear leukocytes (PMNs) converted human big endothelin (bET; 2 microM) to an endothelin-1 (ET-1) like contractile factor, as assessed by bioassay. The generation of this ET-1 like activity was rapid (minutes), time-dependent and more pronounced in non-activated cells, suggesting a partial degradation by activated PMNs. Phosphoramidon (54 micrograms/ml) inhibited the formation of this contractile factor, whereas phenylmethylsulfonylfluoride (PMSF; 25 micrograms/ml), pepstatin A (1 microgram/ml) or epoxysuccinyl-L-leucylamido-(guanidino)butane (E-64; 10 micrograms/ml) did not. Incubations of activated PMNs with PMSF significantly potentiated the generation of ET-1 like activity and selectively inhibited the degradation of [125I]ET-1 by activated PMNs. These findings indicate that human PMNs contain and/or release neutral proteases, which can both rapidly produce and degrade ET-1, an observation which may have important (patho)physiologic implications.

Human polymorphonuclear leukocytes generate and degrade endothelin-1 by two distinct neutral proteases

Human polymorphonuclear leukocytes (PMNs, 4 x 10(6)/ml) converted human big endothelin (bET) to an endothelin-1 (ET-1)-like contractile factor, as assessed by bioassay. The formation of this ET-1-like activity from bET was partially inhibited by phosphoramidon (54 micrograms/ml), but not by pepstatin-A (1 microgram/ml), epoxysuccinyl-L-leucylamido(guanidino)butane (E-64, 10 micrograms/ml) or phenylmethylsulfonyl fluoride (PMSF, 25 micrograms/ml). In addition, nonactivated PMNs converted [125I]bET to [125I]ET-1, thus confirming the bioassay results. Incubation of ET-1 with fMLP-activated PMNs or cell-free supernatants from activated PMNs resulted in the loss of its contractile activity, and this loss of activity was paralleled by the metabolism of [125I]ET-1. The metabolism of [125I]ET-1 by PMNs or leukocyte cathepsin G (5 micrograms/ml) was prevented by PMSF (25 micrograms/ml), but not by phosphoramidon (54 micrograms/ml) or pepstatin-A (1 microgram/ml). Thus, PMNs can form ET-1 from bET via a neutral protease and degrade ET-1 via a serine protease, an observation that may have important pathophysiologic implications in disease states associated with PMN infiltration.

Characterization of the vascular thromboxane A2/prostaglandin endoperoxide receptor in rabbit aorta. Regulation by dexamethasone

Recently, we have shown that dexamethasone treatment of rabbits specifically reduces vascular smooth muscle responsiveness to agonists that interact with the vascular thromboxane A2/prostaglandin H2 (TXA2/PGH2) receptor. One potential site at which dexamethasone can influence prostanoid-mediated vasoconstriction may be at the level of the vascular TXA2/PGH2 receptor. Therefore, we characterized the vascular TXA2/PGH2 receptor in rabbit aortic membranes and examined the influence of dexamethasone treatment on vascular TXA2/PGH2 receptor affinity and number. The binding of [125I][1S-(1 alpha,2 beta(5Z),3 alpha(1E,3R)4 alpha)]-7-[3-(3- hydroxy-4-(p-iodophenoxy)-1-butenyl)-7-oxabicyclo[2.2.1] heptan-2-yl]-5-heptanoic acid ([125I]BOP), a potent TXA2/PGH2 receptor agonist, to rabbit aortic membranes was saturable, displaceable, and dependent on protein concentration. Scatchard analysis of equilibrium binding data disclosed one class of high affinity binding sites with a Kd of 0.44 +/- 0.13 nM and a Bmax of 114.4 +/- 5.2 fmol/mg protein (n = 7). Removal of the endothelium before membrane preparation did not significantly alter the affinity or number of binding sites for [125I]BOP. Kinetic analysis of the rates of [125I]BOP association/dissociation yielded a Kd of 0.62 nM. The ability of various agonists at the TXA2/PGH2 receptor to displace [125I]BOP from vascular membranes correlated well with their contractile potencies in rabbit aortic rings. Moreover, stereospecific displacement of [125I]BOP binding in aortic membranes and inhibition of U46619-mediated aortic contractions were obtained with the stereoisomers L657925(-) and L657926(+). Collectively, these data suggest that this binding site represents the functionally relevant vascular TXA2/PGH2 receptor. In functional experiments, [127I]BOP induced concentration-dependent contractions of the rabbit aorta, which were reduced by 52% in vessels from dexamethasone-treated rabbits.(ABSTRACT TRUNCATED AT 250 WORDS)

The metabolism of L-arginine and its significance for the biosynthesis of endothelium-derived relaxing factor: cultured endothelial cells recycle L-citrulline to L-arginine

We have investigated the mechanism by which cultured endothelial cells generate L-arginine (L-Arg), the substrate for the biosynthesis of endothelium-derived relaxing factor. When Arg-depleted endothelial cells were incubated in Krebs' solution for 60 min, L-Arg levels were significantly (9.7-fold) elevated. The generation of L-Arg coincided with a substantial decrease (90%) in intracellular L-glutamine (L-Gln), whereas all other amino acids were virtually unaffected. Changes in calcium, pH, or oxygen tension had no effect on L-Arg generation, which was, however, prevented when the cells were incubated in culture medium containing L-Gln. L-Arg generated by endothelial cells labeled with L-[14C]Arg was derived from an unlabeled intracellular source, for the specific activity of the intracellular L-Arg pool decreased substantially (8.8-fold) over 60 min. Arg-depleted endothelial cells did not form urea or metabolize L-ornithine but converted L-citrulline (L-Cit) to L-Arg possibly via formation of L-argininosuccinic acid. Nondepleted cells stimulated with the calcium ionophore A23187 showed only a transient accumulation of L-Cit, indicating that L-Cit is recycled to L-Arg during the biosynthesis of endothelium-derived relaxing factor. The generation of L-Arg by Arg-depleted endothelial cells was partially (45%) blocked by protease inhibitors, and various Arg-containing dipeptides were rapidly cleaved to yield L-Arg. Thus, cultured endothelial cells recycle L-Cit to L-Arg and possibly liberate peptidyl L-Arg. The Arg-Cit cycle appears to be the equivalent in the endothelial cell to the formation of urea by the liver. The biosynthesis of endothelium-derived relaxing factor may, therefore, not only produce a powerful vasodilator but also relieve the endothelial cell of excess nitrogen.

L-glutamine inhibits the release of endothelium-derived relaxing factor from the rabbit aorta

The present study investigates the effect of the amino acid L-glutamine (L-Gln) on the release of endothelium-derived relaxing factor (EDRF) from the luminally perfused rabbit aorta and on endothelium-dependent relaxations of rabbit aortic strips. L-Gln (200 microM) had no effect on the acetylcholine (Ach)-induced release of EDRF from freshly prepared aortic tissues. The concentration of L-arginine (L-Arg) in endothelial cells isolated from these aortae was approximately 4 mM, as determined by HPLC analysis. After an initial equilibration period of 2 h and two consecutive infusions of Ach (55 microM for 15 min) at 2 and 3 h, L-Arg levels fell by 62 +/- 14% (n = 4). Under these conditions, L-Gln (200 microM) but not D-Gln (200 microM) inhibited the release of EDRF by 50 +/- 4% (n = 7). This effect of L-Gln was partially reversed by infusions of L-Arg (500 microM) but not D-Arg (500 microM). L-Gln (200 microM) but not D-Gln (200 microM) potentiated the inhibitory effect of N omega-nitro-L-arginine (30 microM), an inhibitor of EDRF biosynthesis, on Ach-induced relaxations of rabbit aortic strips, whereas L-Gln alone had no effect. Thus, L-Gln inhibits the release of EDRF from intact blood vessels presumably by interfering with the generation of L-Arg by the endothelium.

Inhibition by L-glutamine of the release of endothelium-derived relaxing factor from cultured endothelial cells

L-Glutamine (0.02-2 mM) but not D-glutamine (0.2 mM and 2 mM) inhibited the release of endothelium-derived relaxing factor (EDRF) from bovine aortic endothelial cells cultured in the presence or absence of L-arginine. Inhibition was maximal at a concentration of 200 microns, and was reversed by L-arginine (50 microns) but not D-arginine (100 microns). L-Glutamic acid (2 mM) or ammonium chloride (1 mM), putative products of the metabolism of L-glutamine in endothelial cells, had no effect on EDRF release. L-Glutamine (0.2 mM and 2 mM) but not D-glutamine (2 mM), L-glutamic acid (2 mM) or ammonium chloride (1 mM) also inhibited the generation of L-arginine in endothelial cells. Thus, L-glutamine inhibits EDRF release by preventing the generation of L-arginine.

Cytochrome P450-dependent arachidonic acid metabolites, 19- and 20-hydroxyeicosatetraenoic acids, enhance sodium-potassium ATPase activity in vascular smooth muscle

Several cytochrome P450-dependent arachidonic acid metabolites have been shown to affect Na+,K(+)-ATPase activity. In the present study, we tested the effect of omega- and omega - 1-hydroxylated products, i.e., 19- and 20-hydroxyeicosatetraenoic acids (19- and 20-HETE), on the K-induced relaxation in rat aortic rings. 19-HETE and 20-HETE increased the magnitude of the potassium-induced relaxation in a dose-dependent fashion (10(-7)-10(-5) M). The inhibitory effect of ouabain on the potassium-induced relaxation was reversed by both 19- and 20-HETE. In addition, indomethacin fully inhibited the stimulatory effect of 19- and 20-HETE on relaxation induced by potassium. Vascular ouabain-sensitive 86Rb uptake was also increased by 19- and 20-HETE. These observations suggest that 19- and 20-HETE stimulate vascular Na+,K(+)-ATPase via their conversion by cyclooxygenase to prostaglandin-like material.

Release of a neutrophil-derived vasoconstrictor agent which augments platelet-induced contractions of blood vessels in vitro

1. The effects of neutrophil-derived products on vascular tone in vitro were examined by adding purified rabbit neutrophils to siliconized organ baths containing rings of rabbit thoracic aorta. 2. Neutrophil-derived products induced a concentration-dependent contraction of the blood vessels generating 3.9 +/- 0.2 g of tension at a cell concentration of 2 X 10(6) ml-1. This contractile response was not dependent on an intact endothelium and was not ameliorated by treatment of the neutrophils with inhibitors of lipoxygenase or cyclo-oxygenase enzymes, by free radical scavengers or by the use of end organ antagonists to angiotensin II, desArg9-bradykinin, histamine, catecholamines or acetylcholine. 3. Neutrophils stimulated with phorbol myristate acetate, formyl methionyl-leucyl-phenylalanine or calcium ionophore A23187 released a contractile factor into the supernatant which produced qualitatively similar contractions compared to those elicited by incubation with intact unstimulated neutrophils. 4. Ten to twenty times more platelets were required to evoke equivalent contractions to those observed with neutrophils. However, neutrophil supernates significantly augmented platelet-mediated contractions (P less than 0.001). 5. Contractions elicited by neutrophils and supernates derived from activated neutrophils were partially antagonized by 5-hydroxytryptamine (5-HT) receptor antagonists, methysergide and ketanserin. However, the inhibition of exogenous 5-HT-induced contractions on rabbit aorta and rat stomach strips by both antagonists was greater than the inhibition of contractions produced by neutrophils and neutrophil-derived supernates. 6. Extraction of the biologically active material from supernatants of activated neutrophils into acetone, but not chloroform-methanol or ethyl acetate, suggests the contractile factor may be a protein/peptide. Partial purification on a Sephadex G100 column yields a contractile factor with a molecular weight of less than 4,000 daltons. 7. This factor may augment vascular tone, either directly or by interactions with platelets, in pathophysiological states associated with neutrophil activation and accumulation.

Dexamethasone selectively attenuates prostanoid-induced vasoconstrictor responses in vitro

Glucocorticoids bind to specific vascular receptors resulting in a variety of functional consequences that may affect vascular smooth muscle behavior. We, therefore, examined in rabbits the effect of treatment with dexamethasone (2.5 mg/kg) for 6 days on vascular responses to pressor prostanoids in aortic and carotid arterial rings and in the isolated perfused kidney. Isometric tension development to prostaglandin F2 alpha and U46619, a thromboxane/prostaglandin endoperoxide mimetic, was markedly reduced in vessels from dexamethasone-treated rabbits. The inhibitory effect of dexamethasone on vascular reactivity was manifested by an increase in the concentration of agonist for threshold tension development and a reduction in the maximal response to prostaglandin F2 alpha and U46619. In contrast, reactivity to phenylephrine, potassium, histamine, or endothelin was not affected by dexamethasone treatment. In addition, pressor responses to prostaglandin F2 alpha and U46619 in Krebs'-perfused kidneys from dexamethasone-treated rabbits were also diminished. These data suggest that dexamethasone selectively interferes with the expression of receptor-mediated contractile responses to eicosanoids.

Reciprocal effects of dexamethasone on vasodilatory responses to arachidonic acid and prostanoids in the isolated perfused rabbit kidney

We reported that dexamethasone treatment of rabbits causes a reduction in renal vasoconstrictor responses to prostaglandin F2 alpha and U46619, an agonist at the thromboxane-endoperoxide receptor, but not to phenylephrine. The purpose of this study was to examine if dexamethasone treatment can affect the renal vasodilatory responses to prostacyclin (PGI2) and prostaglandin E2 (PGE2) in isolated Krebs-perfused kidneys constricted with phenylephrine. In kidneys from dexamethasone-treated rabbits, the vasodilatory response to PGI2 was reduced by 57%, whereas that to PGE2 was converted to a vasoconstrictor response. This effect of dexamethasone appears to be specific in that the renal vasodilatory responses to forskolin and to sodium nitroprusside were not affected by the steroid. Contrasting with the inhibitory effect of dexamethasone on prostanoid-induced renal vasodilation, treatment with dexamethasone augmented the renal vasodilatory response to arachidonic acid; for example, arachidonic acid, at 10 micrograms decreased perfusion pressure by 24.8 +/- 5.4 and 49.0 +/- 5.6 mm Hg in kidneys from vehicle- and dexamethasone-treated rabbits, respectively. The enhanced vasodilatory effect of arachidonic acid could not be attributed to increased renal formation of PGE2 and PGI2. In conclusion, dexamethasone interferes with prostanoid-mediated renal vasodilation, which is not associated with an impairment in renal responsiveness to direct activators of adenylate cyclase and guanylate cyclase. The reciprocal effect of dexamethasone on the renal vascular responses to arachidonic acid and vasodilatory prostanoids are indicative of a previously unrecognized influence of glucocorticoids on the renal arachidonate-prostaglandin system.

STZ-induced diabetes in SHR and renovascular hypertensive rats: dissociation between changes in arterial pressure and vascular collagen synthesis

Streptozotocin (STZ)-induced diabetes depresses the rate of vascular collagen synthesis in the spontaneously hypertensive rat (SHR), but it also reduces arterial pressure (SAP) in this strain. We investigated this phenomenon further by comparing the SHR with the renovascular hypertensive (RVH) rat, because diabetes does not affect SAP in the latter model of hypertension. Renovascular hypertension was induced by clipping the left renal artery of Wistar-Kyoto (WKY) rats; sham-operated WKY were included as normotensive controls. Collagen synthesis of arterial tissue in vitro was quantified as prolyl hydroxylase activity and the rate of radioactive proline incorporation into collagen. Arterial collagen synthesis of nondiabetic SHR and RVH animals was elevated compared to that of the nonhypertensive WKY controls. STZ-induced diabetes (8 weeks) reduced SAP of SHR, but had no effect on SAP of either RVH or normotensive WKY rats. However, diabetes significantly depressed vascular collagen synthesis of both SHR and RVH rats, and, less consistently, of the WKY. The results strongly suggest that STZ-induced diabetes in SHR impairs arterial collagen synthesis independent of associated changes in arterial pressure.

Manipulation of cytochrome P-450 dependent renal thromboxane synthase activity in spontaneously hypertensive rats

Thromboxane synthase is a cytochrome P-450-like enzyme requiring an iron-centered oxygen attack of the prostaglandin endoperoxide substrate (PGH2) for subsequent thromboxane A2 (TxA2) formation. The activity and levels of P-450 enzymes can be manipulated by decreasing heme availability. Stannous chloride (SnCl2) selectively induces renal heme oxygenase activity, depleting heme and decreasing hemoprotein synthesis. We therefore manipulated the renal cytochrome P-450 system to influence thromboxane synthase activity, as measured by the conversion of 14C-PGH2 to thromboxane B2 (TxB2) in renal cortical microsomes from spontaneously hypertensive rats (SHR). Seven-week-old SHR were treated subcutaneously with SnCl2 (1, 10 and 15 mg/100 g body weight) for 4 consecutive days, and cortical microsomal heme oxygenase activity, heme content, P-450 content, thromboxane synthase activity and systolic blood pressure were measured. Heme oxygenase activity was significantly increased from 1058 +/- 62 nmol/mg protein in controls to 3125 +/- 918, 5057 +/- 690--and 4236 +/- 581 nmol/mg protein in SHR treated with 1, 10 and 15 mg/100 g body weight SnCl2, respectively. The increase in heme oxygenase activity was associated with corresponding decreases in heme content (0.29 mumol/mg protein, for control to 0.12 mumol/mg protein for SHR treated with SnCl2, 10 mg/100 g body weight) and cytochrome P-450 content (0.18 +/- 0.1 nmol/mg protein for control to 0.06 +/- 0.01 nmol/mg protein for SHR treated with SnCl2 10 mg/100 g body weight). The reduction in heme and P-450 content was associated with a reduction in thromboxane synthase activity, i.e., decreases of 38, 35 and 47% from control levels at doses of 1, 10 and 15 mg/100 g body weight.(ABSTRACT TRUNCATED AT 250 WORDS)

Vasoactivity of 20-hydroxyeicosatetraenoic acid is dependent on metabolism by cyclooxygenase

We recently demonstrated that cortical microsomes from spontaneously hypertensive rats metabolize arachidonic acid via cytochrome P450 to omega- and omega-1 hydroxylated compounds, 19- and 20-hydroxyeicosatetraenoic acids (HETE). The vascular activities of 20-HETE and the two isomers of 19-HETE were examined in rat aortic rings. The HETEs produced concentration-dependent contractions of the aortic rings. The contraction elicited by 20-HETE was abolished partially by removal of endothelium and was inhibited completely by treatment with indomethacin and reversed to a relaxation response by treatment with the endoperoxide and thromboxane receptor antagonist SQ 29548. These data suggest that the vascular effects of 20-HETE depend on subsequent metabolism by cyclooxygenase.

Depressor effect of diabetes in spontaneously hypertensive rat: role of vascular reactivity and prolyl hydroxylase and lysyl oxidase activities

Streptozotocin (STZ)-induced diabetes (8 weeks) produced a marked depressor effect in the spontaneously hypertensive rat (SHR), confirming earlier studies, but had no effect on arterial pressure of normotensive controls (WKY). We investigated the phenomenon further by examining the effects of diabetes on the activities of aortic prolyl hydroxylase (PH) and lysyl oxidase (LO), marker enzymes for collagen biosynthesis, and on the reactivity of isolated mesenteric arteries to vasoactive agents. PH and LO activities of nondiabetic SHR were greater than those of the WKY controls. Diabetes markedly reduced PH and LO activities of SHR aortae, but had no significant effect on PH and LO activities of the WKY strain. The effects of diabetes on vascular collagen biosynthetic enzymes of SHR were not associated with reductions in mesenteric arterial responsiveness or sensitivity to norepinephrine, methoxamine, serotonin or KC1. These results suggest that the depressor effect of diabetes in SHR is associated with a reduction in vascular collagen biosynthesis but not a reduction in vascular reactivity.