Invitro.umassmed.edu

Possible Role of Nitric Oxide and Arachidonic Acid Pathways in
Hypoxia-Induced Contraction of Rabbit Coronary Artery Rings
C. Barbé1, V. De Crescenzo2, F. Diemont2 and P. Bonnet2 1Lab. de Neurophysiologie, Équipe du Préconditionnement du Myocarde, Université d'Angers, Cedex, France;2Lab. de Physiologie des Cellules Cardiaques et Vasculaires, CNRS UMR 6542, Faculté de Médecine, Tours, France Abstract
In isolated coronary arteries, hypoxia induces an increase in nary arteries of dogs (Borda et al., 1980; Vanhoutte, 1988), tone by releasing an unidentified endothelium-derived con- pigs (Rubanyi & Paul, 1985) and sheep (Kwan et al., 1989), tracting factor (EDCF). Isometric force was measured in it has been suggested that vasospasm could be triggered by an isolated rabbit coronary artery ring at 37°C in control a decrease in oxygen partial pressure. The vascular endothe- and high K+ (40 mM) pre-contracted conditions. Hypoxia lium modulates the tone of the underlying smooth muscle by (15 mmHg pO2) induced by equilibrating the perfusate with producing relaxing (Furchgott, 1983; Bolton & Clapp, 1986) nitrogen. Hypoxia did not affect the resting tone but induced and contracting factors (Vanhoutte, 1987). A modulation of an endothelium-dependent contraction on pre-contracted tone, secondary to the release of diffusible endothelium- rings. Inhibitors of nitric oxide (NO) were tested, L-NAME derived contracting factors (EDCF), can play a major role (10-4 M) totally and L-NMMA (10-4 M) partially convert the in coronary artery vasospasm and myocardial ischemia hypoxic contraction to an hypoxic relaxation. The addition of (Rubanyi & Vanhoutte, 1985; Pearson et al., 1991). EDCF L-arginine (10-4 or 10-3 M) did not restore the response.
has not yet been identified (Rubanyi & Vanhoutte, 1985) nor Methylene blue (10-5 M) and ODQ (1 H-[1,2,4] oxadiazolo- quantified by bioassay. Previous studies have excluded cate- [4,3-a] quinoxalin-1-one, 10-5 M), both inhibitors of guany- cholamines, serotonin, histamine, adenine nucleotides and late cyclase, also changed the hypoxic contraction into a cyclooxygenase products as EDCF (De Mey & Vanhoutte, hypoxic relaxation. Catalase (1200 U/ml), which decom- 1983; Rubanyi & Vanhoutte, 1985; Katusic & Vanhoutte, poses hydrogen peroxide (H2O2), and superoxide dismutase 1986). The vasoconstrictive peptide, endothelin is an unlikely (150 U/ml, SOD), a free radical scavenger, did not change the candidate for EDCF because the hypoxic contraction is hypoxic response but quinacrine (50 mM), an inhibitor of observed quickly after the onset of the hypoxic exposure and phospholipase A2, significantly decreased it. Inhibitors of quickly reverses (Vanhoutte et al., 1989), whereas endothe- arachidonic acid metabolism (indomethacin, diethylcarba- lin needs time to be synthesised before being released. In mazine, miconazole) however did not affect the hypoxic human coronary arteries, indomethacin and diethylcarba- response. We conclude that in K+ pre-contracted rabbit coro- mazine suppress hypoxic contraction whereas indomethacin nary artery rings, hypoxia induces a contraction which is does not prevent the hypoxic contraction in dog coronary nitric oxide and arachidonic acid dependent.
arteries. Three different mechanisms have been proposed toaccount for hypoxic contraction. There are the inhibition of Keywords: hypoxic contraction, NO, arachidonic acid, coro-
an endothelium-derived relaxing factor (EDRF), for example NO (Rees et al., 1989; Yang et al., 1994), the synthesis or therelease of contractile superoxide anions (Pearson et al., 1991;Lin et al., 1991) or the participation of an arachidonic acid Introduction
(AA) metabolism product (Katusic & Vanhoutte, 1986; Rees Vasospasm, one cause of myocardial infarction can be et al., 1989; Graff & Gozal, 1999). In this study, we con- defined as a sustained coronary artery contraction. As firmed the role of the endothelium in the hypoxic contraction hypoxia induces contraction in isolated rings of large coro- and tested each of these three hypotheses by using nitric Address correspondence to: Dr Bonnet, Lab. de Physiologie des Cellules Cardiaques et Vasculaires, CNRS UMR 6542, Faculté deMédecine, 2 bis bd Tonnellé, 37032 Tours, France. Tel.: +33 2 47 36 60 91; Fax: +33 2 47 36 60 64; E-mail: bonnet@med.univ-tours.fr oxide synthase blockers, superoxide anion scavengers and increasing doses of acetylcholine (ACh) in rings pre- different blockers of AA metabolism. The goal of our study contracted by histamine (His). This test was done at the end was to focus on the pharmaco-physiological events explain- of the 120-min equilibration period before the first hypoxic ing the hypoxic contraction and the factors that determine challenge and again at the end of the experiment. Relaxation is known to occur only in preparations with intact endothe-lium (Furchgott & Zawadzki, 1980). This relaxation was usedas a positive test. Rings were sequentially superfused with Material and methods
normoxic high K+ solution (40 mM) for 15 min, followed bysuperfusion with hypoxic high K+ solution until we obtained Animal preparation
a maximal hypoxic contraction. We considered that the Rabbits weighing 1.5 to 2 kg were heparinized (3500 UI) by maximal hypoxic effect was reached when the initial increase intraperitoneal injection and killed by cervical dislocation.
in tension remained stable for approximately five minutes.
After thoracotomy, the heart was quickly excised and placed Rings were then superfused again with normoxic high K+ in a cold (4°C), calcium-free physiological saline solution solution for 15 minutes. After this control hypoxic protocol, with the following composition (in mmol/l): NaCl 138.6; KCl the coronary artery rings were returned to baseline tension 5.4; MgCl2 1.2; NaH2PO4 0.33; HEPES 10; Glucose 11. The by superfusion with normoxic PSS solution. After 60 min, the pH was adjusted to 7.4 with NaOH. The epicardial part of the same hypoxic protocol was repeated but in the presence of a left circumflex coronary artery was removed, cleaned of test drug dissolved in high K+ solution. The reproducibility surrounding tissues taking care not to damage the luminal of the hypoxic response was tested in control arterial rings surface and prepared as segments (4-5 mm in length and 500- by exposure to three successive hypoxic periods separated by 1000 mm in external diameter). In some rings the endothelium 20-min normoxia. We observed that repeated exposures to was mechanically removed by rubbing the intima. Each hypoxia neither potentiated nor inhibited of the responses segment was suspended by two tungsten wires of 15 mm of isolated coronary artery rings. A single drug was tested of diameter passed through its lumen (internal diameter 400- 800 mm). One wire was anchored to the bottom of the organchamber and the other was connected to a force transducer (Kent TRN 001-220) for measurement of isometric force.
Segments of coronary artery, with or without endothelium, Acetylcholine (ACh), Histamine (His), catalase, superoxide from the same animal were suspended in four organ chambers dismutase (SOD), quinacrine, diethylcarbamazine (DEC), (6 ml volume) containing normoxic physiological saline solu- miconazole, Nw-nitro-L-arginine methyl ester (L-NAME), tion (PSS) which had the following composition (in mmol/l): Nw-monomethyl L-arginine (L-NMMA), indomethacin, NaCl 138.6; KCl 5.4; MgCl2 1.2; NaH2PO4 0.33; CaCl2 1.8; methylene blue (MB) and 1H-[1,2,4]oxadiazolo[4,3- HEPES 10; glucose 11. The pH was adjusted to 7.4 with a]quinoxalin-1-one (ODQ) were obtained from SIGMA.
NaOH and the temperature held at 37°C. Before experiments,rings were preloaded (2 g) and allowed to equilibrate for 120 Statistical analysis
minutes at their optimal resting tension. Normoxic solutionswere equilibrated with 20% O2-80% N2 (pO2 150 mmHg).
The tension developed under hypoxia before and after the Hypoxia was induced by switching to solution equilibrated addition of each drug was compared. Contraction and relax- with 100% N2. The pO2 of the hypoxic solution measured with ation are expressed as % of the contraction induced by nor- To test whether the ring preparation had an intact endothe- All the results were expressed as mean ± standard error lium, sections were taken from the coronary artery. Tissues of the mean. The n value quoted is the number of rings used.
were fixed for a minimum of 48 h in 10% formalin and Statistical evaluation of data in the same groups was per- embedded in parafin wax. Sections of 4-5 mm were cut and formed by paired Student's t test analysis. Differences were stained with haematoxylin and eosine. Using argentafin col- considered significant when P was less than 0.05.
orimetry we demonstrated that the technical procedure main-tained an intact endothelial layer.
All animals received care in compliance with the 'Guide for the Care and Use of Laboratory Animals' published by Effects of hypoxia on ring tone
the National Institute of Health (NIH publication n°85-23,revised 1985).
Preliminary experiments determined the hypoxic response ofisolated rabbit coronary artery rings. In the first series ofexperiments, rings were superfused for 15 min with normoxic Protocol
PSS followed by 15 min with hypoxic PSS. Hypoxia did not The presence or the absence of a functional endothelium able affect the resting tone (n = 11, Fig. 1A). The concentration to release NO was tested by determining the response to of K+ required to pre-constrict arterial rings, in order to Isometric tension recording of an isolated rabbit coronary artery ring. A: Ring was exposed to hypoxic PSS solution. No effect was observed. B: Ring was pre-contracted with 40 mM K+ solution (K+ 40 mM) before exposure to hypoxia. Hypoxic K+ solution induced anadditonal contraction of about 3 mN.
obtain an optimal reaction to hypoxia, was determined by arginine (L-NMMA, 10-4 M), two inhibitors of NO synthase.
successive 15-min periods of superfusion of normoxic, Hypoxia induced a contraction of 41.3 ± 12.3% (n = 8) hypoxic and normoxic with different high K+ solutions but in the presence of L-NAME, hypoxia induced a relax- obtained by equimolar substitution of NaCl by KCl. In all ation (43.5 ± 6.2% (n = 8)). The L-NAME mediated cases (n = 6) hypoxia induced a further contraction of arte- conversion of an hypoxic contraction into a relaxation was rial rings pre-contracted by 40 mM K+ solution (Fig. 1B). In not reversed by the addition of L-arginine at two concentra- each case the effect of hypoxia was fully reversible and the tions (10-4 and 10-3 M, n = 6 for each) (Fig. 3). The level of contraction induced by high K+ solution post-hypoxia amplitude of the contraction induced by high K+ solution was was the same as that induced by high K+ pre-hypoxia (Fig.
significantly increased by L-NAME (4.9 ± 2.9 mN vs 23 ± 3.6 mN). This increase in high K+ contraction was partially The effect of the endothelium was examined in 6 coro- reversed by L-arginine since the magnitude of the K+ con- nary artery rings where the endothelium had been removed traction returned to 10.2 ± 3.1 mN. Experiments performed mechanically and its absence verified by the failure of ACh with L-NMMA (5 ¥ 10-5 and 10-4 M) modified the hypoxic to induce relaxation (Fig. 2). The magnitude of the high K+ response: upon hypoxia a contraction was observed followed contraction was the same (4.2 ± 2.1 mN vs 4.4 ± 1.9 mN in by a relaxation (Fig. 4). The addition of L-arginine at 10-4 endothelium and endothelium-denuded rings, respectively) and 10-3 M (n = 7) did not reverse the original hypoxic suggesting that our technique to remove the intimal layer pre- response (Fig. 4). The hypoxic contraction decreased from served smooth muscle contraction capabilities. Hypoxia 4.2 ± 1.6 mN to 2.5 ± 0.9 mN (p < 0.05) in the presence of induced a contraction of 34.8 ± 8.3% in rings with endothe- L-NMMA. In the presence of L-NMMA with 10-3 M L- lium but a relaxation of 37 ± 4.3% in rings without endo- arginine the hypoxic contraction of 2.2 ± 0.9 mN was not thelium (Fig. 2). In differents experiments rings were different from the hypoxic contraction observed without L- precontracted by histamine (10-5 M) (in place of high K+ arginine (2.5 ± 0.9 mN). L-NMMA increased the K+- solution) with intact endothelium hypoxia still induced a con- induced contraction (6.1 ± 2.8 mN control, 18.5 ± 6.2 mN traction of 48 ± 7% and a relaxation of 26 ± 5% without with L-NMMA) and this effect was partially reversed by the addition of L-arginine (18.5 ± 6.2 mN to 7.8 ± 4.1 mN) sug-gesting that L-NMMA induced a blockade of the basal NO Role of nitric oxide metabolism
Since NO is known to increase cGMP concentration in The endothelium-dependent hypoxia-induced contraction smooth muscle, we blocked its synthesis with methylene blue has been suggested by Yang and co-workers (1994) to be the (10-5 M), a potent inhibitor of guanylate cyclase. In the pres- result of an inhibition in basal EDRF release. To test this ence of methylene blue the magnitude of the high K+ con- hypothesis, we blocked NO synthesis by Nw-nitro-L-arginine traction did not change (6.8 ± 1 mN in control vs 9.6 ± methyl ester (L-NAME, 10-4 M) and Nw-monomethyl L- 1.3 mN with the drug). Methylene blue changed the hypoxic Isometric tension recording of isolated rabbit coronary artery rings with (A) or without (B) endothelium. Rings were pre-contracted with either 10-5 M histamine (HIS) (left panel) or with 40 mM K+ solution (K+ 40 mM) (right panel). The presence of endothelium was testedon these rings by the response to increasing doses (a, b, c, d, respectively 10-7, 10-6, 10-5 and 10-4 M) of acetylcholine (left panel). Hypoxiainduced a contraction if the endothelium was intact (A) and a relaxation in a ring without endothelium (B).
Isometric tension recording of an isolated rabbit coronary artery ring exposed to 40 mM hypoxic K+ solution. A: In control condi- tions, hypoxia induced a contraction. B: In the presence of L-NAME (10-4 M) hypoxia induced a relaxation. C: The addition L-arginine (10-3 M) did not restore the hypoxic contraction.
contraction into a hypoxic relaxation (n = 7) (Table 1). After tion into a hypoxic relaxation without any change in the high washout, the hypoxic contraction was restored. We also K+ contraction (Table 1). We conclude that hypoxic EDCF tested ODQ (10-5 M) which is a more specific blocker of release is dependent on the NO synthase pathway. Contrary guanylate cyclase. ODQ also changed the hypoxic contrac- to NO inhibitors, guanylate cyclase blockers did not enhance Isometric tension recording of an isolated coronary artery ring pre-contracted by 40 mM K+ solution. A: In control conditions. B: Exposure to L-NMMA (50 mM) (1) induced an additional contraction. The increase in L-NMMA concentration to 10-4 M (2) was withoutfurther effect on tension. Hypoxia induced a small hypoxic contraction followed by a relaxation. C: The application of L-NMMA (10-4 M)and L-arginine (10-3 M) (3) did not potentiate the potassium contraction as seen in B but hypoxia still induced the same response.
Effect of Methylene blue and ODQ on the hypoxic Role of arachidonic acid metabolism
response express in percentage of the maximal amplitude of the A role for arachidonic acid metabolism particularly cyclooxygenase products has been proposed by severalauthors to explain the hypoxic contraction (Rubanyi & Vanhoutte, 1985). The blockade of the endogenous synthesis of arachidonic acid by 50 mM quinacrine an inhibitor of phos-pholipase A2 induced a significant decrease in the hypoxiccontraction (31 ± 6.5% vs 2.9 ± 1.7%, n = 6, p < 0.05)without any effect on high K+-induced contraction. After wash-out, the hypoxic contraction reached the same value as K+ contraction by 31 ± 15% and hypoxic contraction from9.5 ± 2.1 mN to 3.1 ± 1.9 mN, (p < 0.05). Since coronary the K+ pre-contraction. According to the hypothesis of Yang endothelial cells and coronary arteries metabolise arachi- et al. (1994), we show that the NO pathway is implicated in donic acid via the cyclooxygenase, lipooxygenase, and cytochrome P-450 pathway, we blocked arachidonic acidmetabolism by using indomethacin (an inhibitor of cyclooxy-genase), diethylcarbamazine (an inhibitor of lipooxygenase) Role of oxidative metabolism in hypoxic response
and miconazole (an inhibitor of cytochrome P-450). None of Pearson et al. (1991) suggested that free radicals like the these inhibitors modified the hypoxic contraction (Table 2).
superoxide anion (O-2) could react with NO and form We conclude that hypoxic contraction is dependent upon peroxynitrite (ONOO-). To test this hypothesis, we examined arachidonic acid but not its metabolites.
the effects of superoxide dismutase (150 U/ml, SOD), known to scavenge O-2, on the hypoxic contraction. We also used catalase (1200 U/ml) which decomposes hydrogen Discussion
peroxide (H2O2). Neither SOD nor catalase significantlymodified the hypoxic contraction: 19.7 ± 2% vs 18.5 ± 2% This study confirmed that rabbit coronary artery rings need (n = 9); 49 ± 9.4% vs 45 ± 11% (n = 8), respectively to be pre-contracted in order to undergo contraction by (p < 0.05). We conclude that oxidative metabolism is not hypoxia, as reported previously (De Mey & Vanhoutte, 1983; involved in the mechanical response of coronary artery ring Rubanyi & Vanhoutte, 1985). We also showed that the response to hypoxia was independent of whether pre- Effect of hypoxia on rabbit coronary arteries in the pres- never observed this difference in reversibility, even with ence of blockers of arachidonic acid metabolism. The baseline con- high dose of L-arginine. This adds to the evidence that traction was obtained with 40 mM potassium.
hypoxia might act on the binding sites between NOS and L-arginine. This effect seems to occur only during hypoxia and not during normoxia as suggested by the reversibility of NOS inhibitors on the K+ contraction before hypoxic We demonstrated that MB and ODQ converted hypoxia- induced endothelium-dependent contraction to relaxation without impairment of K+ induced contraction. It is well Data are expressed as means ± SEM. DEC: diethylcarbazemine.
known that MB, like ODQ inhibits soluble guanylate cyclaseand thus depletes cGMP in vascular smooth muscle. Deple-tion of cGMP should lead to contraction (Murad et al., 1978; contraction was evoked by superfusion with high K+ solution Ignarro & Kadowitz, 1985). Gräser and Vanhoutte (1991) demonstrated that cGMP was involved in the endothelium- In agreement with previous studies we demonstrated that dependent increase in tension caused by hypoxia and con- hypoxic contraction of pre-contracted arteries depended on cluded that a moderate increase in the cGMP level in the the presence of the endothelium (Borda et al., 1980; vascular smooth muscle cell is necessary for the hypoxic con- Furchgott & Zawadski, 1980; De Mey & Vanhoutte, 1983; traction in canine coronary artery. We suggest that hypoxia Rubanyi & Vanhoutte, 1985; Kwan et al., 1989). This acts on NOS to produce NO in a very specific concentration hypoxic contraction might be due to a 'down-regulation' of range that induces a moderate cGMP increase.
basal EDRF release which would lead to contraction (Rees Pearson et al. (1991) tested the idea that NO could et al., 1989). This hypothesis is supported by studies demon- combine with other compounds, and found that superoxide strating that hypoxia impairs the release of EDRF in canine anion (O-2) formed in hypoxic endothelial cells combined femoral artery (De Mey & Vanhoutte, 1983) and in canine with NO to form the peroxynitrite anion (ONOO-) which was coronary artery (Muramatsu et al., 1992). Our results show then metabolized to an endothelium-derived contracting an increase in the high K+ contraction secondary to L-NAME factor or induced the synthesis of such a factor. However, or L-NMMA exposure. This effect was not totally reversed peroxynitrite anion itself could not reproduce the hypoxic by L-arginine when using L-NAME but totally reversed with contraction, and several authors (Liu et al., 1994; Wu et al., L-NMMA. This increase in high K+ contraction, secondary 1994) found that this compound had a vasodilator activity.
to NOS block exposure confirms a basal release of NO and Peroxynitrite anion is a powerful oxidant which can react strongly suggests that hypoxic contraction cannot be sec- with a wide variety of compounds leading to the formation ondary to the inhibition of basal EDRF release. Similar of the endothelium derived-contracting factor. Our results results were observed by other authors with different tissues clearly demonstrated that free radicals were not involved in and species (Pearson et al., 1991; Muramatsu et al., 1992; the hypoxic contraction of rabbit coronary arteries.
Auch-Schwelk et al., 1992). This implicates the secretion Cytochrome P-450 metabolites could play a role in the of a contracting factor by hypoxia. We found a differential hypoxic response (Campbell et al., 1996). However, since we effect between L-NAME and L-NMMA. Our results clearly did not observe any effect of miconazole, it seems unlikely demonstrate that L-NAME reverses the hypoxic effect: a that structural modification of cytochrome P-450 metabolites hypoxic relaxation instead of a hypoxic contraction. In the induced by hypoxia might explain the hypoxic contraction.
presence of L-NMMA we observed a transient hypoxic con- Zelenkov et al. (1993) showed that indomethacin made no traction followed by a relaxation. These two effects clearly significant difference in hypoxic contraction of rat aorta, indicate the obligatory role of NO in hypoxic contraction.
ruling out a prostaglandin mediator. The lack of effect of As we found a differential effect with NOS inhibitors, we inhibitors of the arachidonic acid metabolism pathway in our propose that the blocking effect of L-NAME could be more study leads to the same conclusion for rabbit coronary arter- efficient under hypoxia compared to the blocking effect of L- ies. On the contrary, quinacrine, an inhibitor of phospholi- NMMA. This hypothesis is supported by a study that clearly pase A2, inhibited the hypoxic contraction. Quinacrine is demonstrated that relaxation induced by NO on vessels is suspected to inhibit calcium channels (Nagano et al., 1996) dose-dependent (Lincoln et al., 1996). Since both inhibitors which could explain the decrease in the K+ pre-contraction have the same blocking effect on the basal NO release under with 10-3 M quinacrine. However, this phenomenon was not normoxic conditions (Knowles & Moncada, 1994), we would observed with 50 mM quinacrine which still inhibited the suggest that hypoxia acts on the link between NOS and L- hypoxic contraction. This strongly suggests that arachidonic arginine or its competitive inhibitor (L-NAME and L- acid itself and not its metabolites might be involved in the NMMA). L-NAME, unlike L-NMMA, shows a progressive and irreversible or slowly reversible inhibition of NOS The NO and eicosanoid-producing metabolic pathways (Knowles & Moncada, 1994). Under hypoxic conditions we have been studied extensively. NO is an oxidizing radical that has been reported to induce transcription factors (Herdegen Gräser T, Vanhoutte PM (1991): Hypoxic facilitation of canine et al., 1994) and interact with heme-containing proteins of coronary arteries: role of the endothelium and cGMP. Am guanylate cyclase (Lei et al., 1992). Guanylate cyclase acti- J Physiol 261: H1769-H1777.
vation leads to an increase in cGMP-stimulated protein Herdegen T, Rüdiger S, Mayer B, Bravo R, Zimmermann M kinase G. Arachidonic acid is a contracting agent, which (1994): Expression of nitric oxide synthase and colocalisa- inhibits myosin light chain phosphatase (Parsons et al., tion with Jun, Fos and Krox transcription factors in spinal 1996) but it can also be a cofactor of protein kinase C like cord neurons following noxious stimulation of the rat hind- pare. Molec Brain Res 22: 245-258.
Our results implicate NO and eicosanoid-producing meta- Ignarro LJ, Kadowitz PJ (1985): The pharmacological and phy- bolic pathways in the hypoxic contraction of rabbit coronary siological role of cyclic GMP in vascular smooth muscle artery rings. We suggest that hypoxia induces a release of NO relaxation. Ann Rev Pharmacol Toxicol 25: 171-191.
leading to a sufficient cGMP level necessary to be implicated Katusic ZS, Vanhoutte PM (1986): Anoxic contractions in iso- in the production of the EDCF, issued from the phospholi- lated canine cerebral arteries: contribution of endothelium- derived factors, metabolites of arachidonic acid, and We conclude that hypoxic contraction observed in isolated calcium entry. J Cardiovasc Pharmacol 8: S97-S101.
rabbit coronary artery rings depends upon arachidonic acid Knowles RG, Moncada S (1994): Nitric oxide synthases in and nitric oxide, and involves a cGMP-dependent intracellu- mammals. Biochem J 298: 249-258.
lar mechanism. This mechanism may be implicated in coro- Kwan YW, Wadsworth RM, Kane KA (1989): Effects of hypoxia on the pharmacological responsiveness of isolated coronaryrings from the sheep. Br J Pharmacol 96: 849-856.
Lei SZ, Pan ZH, Aggarwal SK, Chen HSV, Hartman J, Sucher NJ, Lipton SA (1992): Effect of nitric oxide production on Acknowledgements
the redox modulatory site of the NMDA receptor-channel This study supported by operating grants from the Fondation complex. Neuron 8: 1087-1099.
de France. The authors thank May Fuentes for her help Lin PJ, Pearson PJ, Cartier R, Schaff HV (1991): Superoxide to type the manuscript and Manuel Rebocho for his techni- anion mediates the endothelium-dependent contractions to cal assistance. We thank Dr Kumar for checking the serotonine by regenerated endothelium. J Thorac Cardio- Lincoln TM, Cornwell TL, Komalavilas P, Macmillan-Crow LA, Boerth A (1996): The nitric oxide-cyclic GMP signaling References
system of biochemistry of smooth muscle contractionextract. In: Barany M, ed., Biochemistry of Smooth Muscle Auch-Schwelk W, Katusic ZS, Vanhoutte PM (1992): Nitric Contraction Extract. London, Academic Press, pp.
oxide inactivates endothelium derived- relaxing factor in the rat aorta. Hypertension 19: 442-445.
Liu S, Beckman JS, Ku DD (1994): Peroxynitrite a product of Bolton TB, Clapp LH (1986): Endothelial-dependent relaxant superoxide and nitric oxide, produces coronary actions of carbachol and substance P in arterial smooth vasorelaxation in dogs. J Pharmacol Exp Ther 268: 1114- muscle. Br J Pharmacol 87: 713-723. Borda LJ, Shuchleib R, Henry PD (1980): Hypoxic contraction Murad FC, Mittad CK, Arnold WP, Katsuki S, Kimura H (1978): of isolated canine coronary artery. Mediation by potassium- Guanylate cyclase: activation by azide, nitro compounds, dependent exocytosis of norepinephrine. Circ Res 46: nitric oxide, and hydroxyl radical and inhibition by hemo- globin and myoglobin. Adr Cyclic Nucl Res 9: 145-158.
Campbell WB, Gebremedhin D, Pratt PF, Harder DR (1996): Muramatsu M, Iwama Y, Shimizu K, Asano H, Toki Y, Miyasaki Identification of epoxyeicosatrienoic acids as endothelium- Y, Okumura K, Hashimoto H, Ito T (1992): Hypoxia- derived hyperpolarizing factors. Circ Res 78: 415-423.
elicited contraction of aorta and coronary artery via De Mey JG, Vanhoutte PM (1983): Anoxia and endothelium- removal of endothelium-derived nitric oxide. Am J Physiol dependent reactivity of the canine femoral artery. J Physiol Nagano N, Imaizumi Y, Watanabe M (1996): Novel blockade of Furchgott RF (1983): Role of endothelium in responses of vas- Ca2+ current by quinacrine in smooth muscle cells of the g- cular smooth muscle. Circ Res 53: 557-573.
uinea pig. Jpn J Pharmacol 71: 51-60.
Furchgott RF, Zawadzki JV (1980): The obligatory role of Parsons SJW, Sumner MJ, Garland CJ (1996): Phospholipase A2 endothelial cells in the relaxation of arterial smooth muscle and protein kinase C contribute to myofilament sensitiza- by acetylcholine. Nature 299: 373-376.
tion to 5-HT in the rabbit mesenteric artery. J Physiol 491: Graff GR, Gozal D (1999): Cardiorespiratory responses to interleukin-1b in adult rats: role of nitric oxide, eicona- Pearson PJ, Lin PJ, Schaff HV (1991): Production of soids, and glucocorticoids. Arch Physiol Biochem 107: 97- endothelium-derived contracting factor is enhanced after coronary reperfusion. Ann Thorac Surg 51: 788-793.
Rees DD, Palmer RMJ, Moncada S (1989): Role of endothelium-derived nitric oxide in the regulation of bloodpressure. Proc Nat Acad Sci USA 86: 3375-3378.
Rubanyi GM, Paul RJ (1985): Two distinct effects of oxygen on vascular tone in isolated porcine coronary arteries. Circ Res56: 1-10.
Rubanyi GM, Vanhoutte PM (1985): Hypoxia releases a vaso- constriction substance from the canine vascular endothe-lium. J Physiol 364: 45-56.
Vanhoutte PM (1987): Endothelium-dependent contractions in arteries and veins. Blood Vessels 24: 141-144.
Vanhoutte PM (1988): The endothelium modulator of vascular smooth muscle tone. N Engl J Med 319: 512-513.
Vanhoutte PM, Auch-Schwelk W, Boulanger C, Janssen PA, Katusic ZS, Komori K, Miller VM, Schini VB Vidal M(1989): Does endothelin-1 mediate endothelium-dependentcontractions during anoxia? J Cardiovasc Pharmacol 13:S124-S128.
Wu M, Pritchard KAJR, Kamiznski PM, Fayngersh RP, Hintze TH, Wolin MS (1994): Involvement of nitric oxide andnitrosothiols in relaxation of pulmonary arteries to peroxy-nitrite. Am J Physiol 266: H2108-H2113.
Yang XP, Wu ZJ, Xu YF, Dong W, Yang W, Fu SX, Li YS (1994): Effects of inhibitor of endothelium-derived relaxing factoron hypoxic contraction of isolated pig coronary artery. ActaPharmacol Sin 15: 323-326.
Zelenkov P, McLoughlin T, Johns RA (1993): Endotoxin enhances hypoxic contraction of rat aorta and pulmonaryartery through induction of EDRF/NO synthase. Am JPhysiol 265: L346-L354.

Source: http://invitro.umassmed.edu/~vmd/Barbe_et_al.pdf