Investigation of the Mechanisms Underlying H2O2-Evoked Contraction in the Isolated Rat Aorta

ISSN 0306-3623/98 $19.00 1 .00 PII S0306-3623(97)00392-3 All rights reserved

Gen. Pharmac. Vol. 31, No. 1, pp. 115–119, 1998 Copyright  1998 Elsevier Science Inc. Printed in the USA.

Investigation of the Mechanisms Underlying H2O2-Evoked Contraction in the Isolated Rat Aorta Ruzˇena Sotnı´kova´* Institute of Experimental Pharmacology, Slovak Academy of Sciences, Bratislava, Slovakia ABSTRACT. 1. H2O2 (20–200 mM) elicited concentration-dependent contraction of rat isolated aortic rings. Pretreatment of the preparations with N-nitro-L-arginine (L-NNA) or removal of the endothelium augmented significantly the contractile response of rings to H2O2. The increase of the extracellular potassium concentration ([K1]o) resulted in a K1-dependent increase of the contractile response. 2. In Ca21-free physiological salt solution (PSS), responses to H2O2 were decreased approximately to 4% of the 100 mM KCl-induced response. 3. In addition, calcium channel blockers (nisoldipine, nifedipine, verapamil, flunarizine) did not completely inhibit H2O2-induced contractions. 4. H2O2-evoked contractions were abolished after repeated applications of caffeine or noradrenaline in Ca21-free PSS. 5. Both extracellular Ca21entry through voltage-dependent Ca21channels and intracellular Ca21 from caffeine- and noradrenaline-sensitive stores appear to participate in H2O2-induced contractions of rat aortic rings. gen pharmac 31;1:115–119, 1998.  1998 Elsevier Science Inc. KEY WORDS. H2O2, isolated rat aorta, isometric contraction

INTRODUCTION Oxygen free radicals are highly reactive metabolites of molecular oxygen that are now believed to contribute to the etiology of a variety of cardiovascular disorders, e.g. post-ischemic reperfusion injury (McCord, 1985). Oxygen radicals include: the superoxide anion (O2.-), hydrogen peroxide (H2O2), hydroxyl radical (OH.), and singlet oxygen (1O2). These reactive metabolites could be involved in the control of vascular tone and possibly organ blood supply. Under normal physiological conditions, H2O2 seldom accumulates because it is reduced to water by catalase or glutathion peroxidase. Under some pathological states, however, the tissue supplies of these enzymes are depleted and concentration of H2O2 can increase up to mM levels (Schraufstatter and Cochrane, 1991). In intact arteries, the primary target of exogenously formed H2O2 appears to be the vascular endothelium and the majority of work to date has concentrated on the relaxant action of H2O2 (Burke and Wolin, 1987; Burke-Wolin and Wolin, 1990; Rubanyi and Vanhoutte, 1986a, 1986b; Wei and Kontos, 1990; Wolin, 1990). Other authors, however, found H2O2 or organic peroxides to increase pulmonary perfusion pressure (Gurtner et al., 1983; Farrukh et al., 1985; 1987; Gurtner et al., 1983; Olafsdo¨ttir et al., 1991) or to induce contraction of isolated arteries (Katusˇic et al., 1993; Sheenan et al., 1993). In this study, we examined the responses of isolated rat aortic rings to H2O2. In addition, we examined the role of the vascular endothelium and of intracellular and extracellular Ca21 in modulating the effects of H2O2 in this vessel. MATERIAL AND METHODS Male Wistar rats (250–300 g) were killed by decapitation, and the thoracic aorta was removed and immersed in physiological salt solu*To whom correspondence should be addressed. Received 11 March 1997; accepted 22 August 1997.

tion (PSS). After careful removal of adherent tissues, four to ten 2-mm long rings were cut. Care was taken not to damage the inner surface of the blood vessel. The rings were mounted between two platinum hooks in organ bath chambers containing PSS maintained at 378C and bubbled with a mixture of 95% O2 and 5% CO2. The rings were stretched passively to the resting tension of 20 mN. They were allowed to equilibrate for 1 hr, and during this period of time PSS was replaced and the tension was readjusted to 20 mN. Responses were measured using an isometric transducer coupled to a potentiometric pen recorder. After an equilibration period, the arteries were contracted maximally in a depolarizing salt solution containing 100 mM KCl (DSS). At the plateau of the contraction, the functional presence of the endothelium was checked by applying 1 mM of acetylcholine. In experiments testing the role of the endothelium, it was destroyed by gentle rubbing of the preparations with a forceps, or NO production was abolished by 30-min pretreatment of the rings with 0.1 mM L-NNA. In the group of experiments testing the influence of Ca21 channel blockers on the H2O2-induced contraction, the vessels were incubated for 90 min in PSS with or without various concentrations of Ca21 channel blockers. H2O2 was added cumulatively. From one aorta, two preparations served as a control and others were treated in pairs with different concentrations of the antagonist. The influence of extracellular Ca21 on the H2O2-induced response was examined in PSS without Ca21 and with 0.2 mM EGTA, which was in contact with the tissue for 1 hr. The role of intracellular Ca21 was tested in rings incubated in Ca21-free PSS with 1 mM EGTA for 30 min, then 25 mM caffeine or 1 mM noradrenaline was applied in three consecutive doses followed by three washouts with Ca21-free PSS. Under these conditions, the response to H2O2 was tested. At the end of each experiment the response to DSS was tested. The composition of PSS was (in mM): NaCl (112.0), KCl (5.0),

116

R. Sotnı´kova´

FIGURE 1. Concentration–response curves showing the effect of repeated application on H2O2-induced contraction in rat isolated aortic rings. Experiments were carried out in the presence (n,m) or absence (s,d) of 0.1 mM L-NNA (closed signs: the first concentration–response curves; open signs: the second concentration–response curves). Contractions are expressed as percentage of the maximal tissue response to DSS. All data are the mean of eight determinations and SEM is shown in vertical bars.

KH2PO4 (1.0), MgSO4 (1.2), CaCl2 (2.5), NaHCO3 (25.0) and glucose (11.5). The concentration of KCl was changed with equimolar substitution of NaCl to achieve 10, 15, or 100 mM concentration of KCl in PSS. ANALYSIS OF DATA Responses to H2O2 are expressed as percentages of the DSS-induced response. The data are given as means6SEM. Student’s t-test for paired or unpaired values was used, P,0.05 was considered significant. RESULTS H2O2 evoked concentration-dependent contractions of rat isolated aortic rings, which were abolished by catalase (1,000 U. ml21; n55). The responses to H2O2 were tachyphylactic, i.e., it was not possible to obtain two equal subsequent responses in the same aortic preparation (Fig. 1). Pretreatment with L-NNA (100 mM; n58), indomethacin (1 mM; n55), 3-amino-1,2,4-triazole (100 mM; n56), methylene blue (10 mM; n55), or mechanical denudation of the endothelium (n55) did not prevent tachyphylaxis. As shown in Figure 2A, pretreatment of the preparations with L-NNA or removal of the endothelium augmented significantly the contractile response of rings to H2O2, with removal of the endothelium being more efficient than the blockade of NO production. In the presence of raised extracellular potassium concentration [K1]o, contractions induced by H2O2 were significantly potentiated (Fig. 2B). In normal PSS containing 5 mM [K1]o, the maximum response to H2O2 was 10.9162.32% of the maximal contraction induced by DSS. In the presence of 10 and 15 mM [K1]o this was increased to 30.3262.42 and 48.9863.71 % of the DSS-induced response, respectively. Similar results were obtained in tissues preincubated with L-NNA as shown in Figure 3. Responses to H2O2 were increased compared with those recorded in preparations with functionally intact endothelium. The maximal responses to H2O2 of the rings incubated in PSS containing 5, 10, and 15 mM [K1]o reached

FIGURE 2. Effect of L-NNA and raised [K1]o on concentration– response curves for H2O2-evoked contraction of rat isolated aortic rings. In the presence of 0.1 mM L-NNA (A) and in endotheliumdenuded rings (B). In the presence of 5 (s), 10 (d), and 15 mM [K1]o (n). All contractions are expressed as percentage of the maximal contraction induced by DSS. Data are means of eight to seventeen determinations and SEM is shown in vertical bars. *P,0.05, **P,0.01 compared with (s). values of 24.4764.03, 44.3064.54, and 65.8667.23% of the DSSinduced response, respectively. When the preparations were contracted with 100 mM K1-solution, a slight additional increase in force was observed for H2O2 concentrations from 20 to 80 mM, but concentrations of 200 mM or higher evoked a decrease of the tone of the smooth muscle. These responses were not influenced by pretreatment of the aortas with L-NNA (n55). In the presence of a subthreshold concentration of noradrenaline (1 nM), contractions induced by H2O2 were unaltered. However, in the presence of 0.05 and 1 mM noradrenaline, which evoked contractions of 27.4466.12% and 85.5966.02% of the response to DSS, respectively, contractions to H2O2 were potentiated, although to a lesser extent than observed with raised [K1]o (Fig. 4). In the absence of extracellular Ca21, H2O2-evoked contractions, in either the presence or absence of L-NNA, were significantly inhibited and contraction was observed only at very high concentrations (Fig. 5). Similar results were obtained after depletion of intracellular Ca21 stores with repeated applications of either caffeine or noradrenaline (n57).

H2O2-Evoked Contraction in Rat Aorta

117

FIGURE 3. Concentration-response curves of H2O2 in rings of the rat aorta incubated in PSS containing 5 (s), 10 (d), or 15 mM [K1]o (n). Rings were pretreated with 0.1 mM L-NNA. All contractions are expressed as percentage of the maximal contraction induced by DSS. Data are means of eight to seventeen determinations and SEM are shown in vertical bars. *P,0.05, **P,0.01 compared to (s).

FIGURE 5. Influence of removal of extracellular Ca21 concentration–response curves for H2O2-evoked contraction of rat isolated aortic rings. Tissues were incubated 1 hr in Ca21-free PSS containing 0.2 mM EGTA with (s) or without 0.1 mM L-NNA (d). Contractions are expressed as percentage of the maximal tissue response to DSS. Data are the mean of six to eight determinations and SEM is shown in vertical bars.

The effects of preincubation of tissues with the Ca21 channel blockers verapamil, flunarizine, nifedipine, and nisoldipine (1 nM and 1 mM), in the presence of 10 mM [K1]o, and the presence or absence of L-NNA, are shown in Figure 6. At the concentration of 1 nM, nisoldipine was the most effective inhibitor against H2O2evoked contraction in both the presence and absence of L-NNA, whereas verapamil had little effect. In contrast, at a concentration of 1 mM, verapamil was the most effective inhibitor, reducing H2O2evoked contractions. Up to the concentration of 200 mM, H2O2 did not influence the contractile properties of the aortic preparations as the response to

DSS was unchanged at the end of experiment (data not shown). The acetylcholine-induced relaxation (82.9461.79% of the DSS contraction) was enhanced to 75.6312.74% of the DSS-contraction (n517) at the end of the experiment.

FIGURE 4. Effect of noradrenaline on concentration-response curves for H2O2-evoked contraction of rat isolated aortic rings. Experiments were carried out in the absence (s) and presence of 50 nM (d) and 1 mM (n) noradrenaline. Noradrenaline 50 nM and 1 mM induced contraction of preparations that represented 27.4466.12 and 85.5966.02% of DSS-induced contraction. Contractions are expressed as percentage of the maximal tissue response to DSS. All data are the mean of six to eight determinations and SEM is shown in vertical bars.

DISCUSSION The results of this study show that H2O2 (20–200 mM) caused contraction of rat isolated aortic rings. Similarly, H2O2 in the concentrations ranging from 100–300 mM contracted rabbit pulmonary arteries (Sheenan et al., 1993) and in the concentrations from 10–100 mM isolated canine basilar arteries (Katus˘ic et al., 1993). When the rat aortas were precontracted with 100 mM KCl, the preparations responded to H2O2 with a small contraction and in its presence in high concentrations relaxation was apparent. Rubanyi and Vanhoutte (1986a) and Burke and Wolin (1987) reported relaxation induced by 1–100 mM H2O2 in canine coronary and bovine intrapulmonary arteries precontracted either with prostaglandin F2a or by serotonin. Thus responses of arterial preparations to H2O2 seem to depend on the initial tension of the preparations and on the arterial preparation used. In our experiments, it was not possible to obtain two equal consecutive responses to H2O2 in one preparation. This was not due to impairment of contractile function as K1- and noradrenalineevoked contractions were not altered after application of H2O2. H2O2 did not influence responses to noradrenaline, so that depletion of intracellular Ca21, and/or damage of Ca21 entry through voltagedependent or receptor-operated Ca21 channels could not be involved in the mechanism of tachyphylaxis. Not even relaxant responses to acetylcholine were decreased; on the contrary, they were potentiated after H2O2 in the concentrations used. Activation of soluble guanylate cyclase by H2O2 and, as a consequence, an increase of tissue levels of c-GMP (Burke and Wolin, 1987; Wolin and Burke, 1987) may take place. Activation of soluble guanylate cyclase by H2O2 might potentiate the relaxant component of its effect and by this mechanism depress the second contractile response of preparations to H2O2. However, inhibition of guanylate cyclase by methylene blue, inhibition of NOS by L-NNA, or denudation of

118

R. Sotnı´kova´

FIGURE 6. Effect of Ca21channel blockers on concentration–response curves for H2O2-evoked contraction of rat isolated aortic rings. All experiments were carried out in the presence of 10 mM [K1]o and in the absence (A, B), or presence of 0.1 mM L-NNA (C, D). Symbols represent nisoldipine (m), nifedipine (n), verapamil (s), or flunarizine (d) at concentrations of 1 nM (A, C) and 1 mM (B, D). Controls (h). Data are means 6 SEM of at least six determinations.

rings did not prevent tachyphylactic response to H2O2. Neither activation of peroxidase nor participation of prostaglandins seems to be involved in the tachyphylaxis, because pretreatment with 3-amino1,2,4-triazole, as well as premedication with indomethacin was without effect. Further studies are needed to clarify the mechanisms responsible for the absence of reproducibility of the H2O2-induced contraction. Jin et al. (1991) reported that H2O2 generated by the system glucose and glucose oxidase elicited contraction of the rat pulmonary artery, which was present also in Ca21-free media. In our experiments, responses of rat aortas to H2O2 were decreased in Ca21-free PSS. However, a small residue of the contraction still remained. Moreover, Ca21 channel blockers in micromolar concentrations that completly blocked the K1-induced contraction (not shown) did not fully inhibit H2O2-induced contractions. Furthermore, H2O2 elicited concentration-dependent contractions of the rat aorta even in the course of depolarization of the smooth muscle membrane induced by 100 mM KCl. It seems that a substantial part of the H2O2-

induced contraction is dependent on the entry of extracellular Ca21 through voltage-dependent Ca21 channels, but a small component of it is mediated by intracellular Ca21. It has been reported that caffeine and noradrenaline release Ca21 from intracellular stores of vascular smooth muscle through different mechanisms and thus induce contractions in the absence of extracellular Ca21 (Itoh et al., 1983). Our findings show that in Ca21-depleted aortic rings, after three repeated applications of caffeine or noradrenaline in Ca21-free PSS, responses to H2O2 in the concentrations of 20–200 mM were blocked. These results support an assumption that contraction induced by H2O2 is dependent on both extracellular and intracellular Ca21. To study the role of different Ca21sources, H2O2 was also applied under conditions when the smooth muscle membrane had been slightly depolarized or the muscle tension elevated. It was recently described by Cattaneo et al. (1991) that [K1]o 10 and 15 mM in the bathing solution induced partial depolarization of the smooth muscle membrane accompanied with an increase of Ca21 influx without influencing the tension of rat aortic rings. On the contrary, nor-

H2O2-Evoked Contraction in Rat Aorta adrenaline evoked contraction of the vessels without changes in membrane potential (Casteels, 1980). The contraction induced by a-adrenoceptor stimulation mobilizes two Ca21 pools: intracellular and extracellular (Godfraind, 1981; Godfraind and Kaba, 1972). In our experiments, increase of [K1]o in PSS from 5 to 10 and 15 mM resulted in increasing amplitudes of the contractions induced by H2O2. Noradrenaline, in concentrations of 50 nM and 1 mM, also enhanced responses to H2O2. In low concentrations (1 nM), noradrenaline induced neither contraction, nor did it affect responses to H2O2. These results suggest that enhancement of the responses of the rat aorta to H2O2 is due to the increased level of free cytosolic Ca21 elicited by [K1]o and noradrenaline, independently of whether it results from influx through voltage-operated Ca21 channels (in the case of increased [K1]o in PSS), through receptor-operated Ca21 channels, or by mobilization of intracellular Ca21 (in the case of noradrenaline). Pretreatment with L-NNA, an inhibitor of nitric oxide biosynthesis (Moore et al., 1990), increased the amplitude of the H2O2induced contraction. Similarly denudation of preparations resulted in an enhancement of H2O2-induced contraction of the rat aorta. These changes are probably due to the ability of H2O2 to facilitate the release of EDRF, as observed by Rubanyi and Vanhoutte (1986a, 1986b) in canine arteries. Their suggestion is supported also by our observation that H2O2 enhanced relaxation of rat aortas elicited by acetylcholine. We suppose that the response of the rat aorta to H2O2 may involve two components—contractile and relaxant. Thus, increased amplitudes of contraction in the rat aorta with mechanically or functionally damaged endothelium resulted probably from the absence of the endothelium-dependent relaxant component. As the responses of the denuded aortas to H2O2 were higher than those of the preparations with endothelium pretreated with L-NNA, some other mechanisms (such as EDHF) of potentiation of the H2O2induced contraction might be involved as well. Because L-NNA did not block the relaxation induced by high concentrations of H2O2, this component of its action is due conceivably to changes in smooth muscle cells themselves. One of these might be the activation of guanylate cyclase, as found by Burke-Wolin and Wolin (1990). Our results suggest that H2O2-induced changes in the muscle tension of the rat aorta involve direct effects on the smooth muscle and on the endothelium. H2O2-induced contraction of rat aortic rings is mediated by extracellular Ca21 entry into cells through voltage dependent Ca21 channels and by intracellular Ca21 from caffeine- and noradrenaline-sensitive stores. This work was conducted in the Pharmacological Laboratory (Head Prof. Godfraind), Faculty of Medicine, Universite´ Catholique de Louvain, Brussels, Belgium, and it was supported in part by the Grant Agency for Science (grant no. 1017).

References Burke T. M. and Wolin M. S. (1987) Hydrogen peroxide elicits pulmonary arterial relaxation and guanylate cyclase activation. Am. J. Physiol. 252, H721–H732.

119 Burke-Wolin T. M. and Wolin M. S. (1990) Inhibition of c-GMP associated pulmonary arterial relaxation to H2O2 and O-2 by ethanol. Am. J. Physiol. 258, H1267–H1273. Casteels R. (1980) Electro- and pharmacomechanical coupling in vascular smooth muscle. Chest 78, 150–156. Cattaneo E. A., Gende O. A., Cingolani H. E. and Venosa R. A. (1991) Inward movement of Ca21 in K1-depolarized rat vascular smooth muscle. Arch. Inter. Physiol. Biochim. Biophys. 99, 227–235. Farrukh I. S., Michael J. R., Summer W. R., Adkinson N. F. and Gurtner G. H. (1985) Tromboxane-induced pulmonary vasoconstriction: involvement of Ca21. J. Appl. Physiol. 58, 34–44. Farrukh I. S., Gurtner G. H. and Michael J. R. (1987) Pharmacological modification of pulmonary vascular injury: possible role of cAMP. J. Appl. Physiol. 62, 47–54. Godfraind T. (1981) Calcium influx and receptor-response coupling. In: Calcium Antagonists. pp. 95–107. American Physiological Society, USA. Godfraind T. and Kaba A. (1972) The role of calcium in the action of drugs on vascular smooth muscle. Arch. Int. Pharmacodyn. The´r. 196, 35–49. Gurtner G. H., Knoblauch A., Smith P. L., Sies H. and Adkinson N. F. (1983) Oxidant- and lipid-induced pulmonary vasoconstriction mediated by arachidonic acid metabolites. J. Appl. Physiol. 55, 949–954. Itoh T., Kuriyama H. and Suzuki H. (1983) Differences and similarities in the noradrenaline- and caffeine-induced mechanical responses in the rabbit mesenteric artery. J. Physiol. 337, 609–629. Jin N., Packer C. S. and Rhoades R. A. (1991) Reactive oxygen-mediated contraction in pulmonary arterial smooth muscle: cellular mechanisms. Can. J. Physiol. Pharmacol. 69, 383–388. Katusˇic Z. S., Schugel J., Cosentino F. and Vanhoutte P. M. (1993) Endothelium-dependent contractions to oxygen-derived radicals in the canine basilar artery. Am. J. Physiol. 264, H859–H864. McCord J. M. (1985) Oxygen-derived free radicals in post-ischemic tissue injury. N. Engl. J. Med. 312, 159–163. Moore P. K., Al-Swaye O. A., Chong N. W. S., Evans R. A. and Gibson A. (1990) L-NG nitroarginine (L-NOARG), a novel, L-arginine-reversible inhibitor of endothelium-dependent vasodilatation in vitro. Br. J. Pharmacol. 99, 408–412. Olafsdo¨ttir K., Atrori L., Ryrfeldt A., Berggren M., Kumlin M. and Molde´us P. (1991) Mechanisms of hydroperoxide-induced broncho- and vasoconstriction in isolated and perfused rat lung. Pharmacol. Toxicol. 68, 181–186. Rubanyi G. M. and Vanhoutte P. M. (1986a) Oxygen-derived free radicals, endothelium, and responsiveness of vascular smooth muscle. Am. J. Physiol. 19, H815–H821. Rubanyi G. M. and Vanhoutte P. M. (1986b) Superoxide anions and hyperoxia inactivate endothelium-derived relaxing factor. Am. J. Physiol. 19, H822–H827. Schraufstatter M. U. and Cochrane C. G. (1991) Oxidants: types, sources, and mechanisms of injury. In: The Lung: Scientific Foundations (Edited by Crystal R. G., West J. B., Barnes P. J., Cherniack N. S. and Weibel E. R.) pp. 1803–1810. Raven, New York. Sheenan D. W., Giese E. C., Gugine S. F. and Russell J. A. (1993) Characterization and mechanisms of H2O2-induced contractions of pulmonary arteries. Am. J. Physiol. 264, H1542–H1547. Wei E. P. and Kontos A. (1990) H2O2 and endothelium-dependent cerebral arterioral dilation. Implications for the identity of endothelium-derived relaxing factor generated by acetylcholine. Hypertension 16, 162–169. Wolin M. S. (1990) Similarities in the pharmacological modulation of reactive hyperemia and vasodilation to hydrogen peroxide in rat skeletal muscle arterioles: effects of probes for endothelium-derived mediators. J. Pharmacol. Exp. Ther. 253, 508–512. Wolin M. S. and Burke T. M. (1987) Hydrogen peroxide elicits activation of bovine arterial soluble guanylate cyclase by a mechanism associated with its metabolism by catalase. Biochem. Biophys. Res. Commun. 143, 20–25.