Atrial Natriuretic Peptide Protects Against Ischemia-Reperfusion Injury in the Isolated Rat Heart

Transcription

1 Atrial Natriuretic Peptide Protects Against Ischemia-Reperfusion Injury in the Isolated Rat Heart Kenji Sangawa, MD, Koji Nakanishi, MD, Kozo Ishino, MD, Masahiro Inoue, MD, Masaaki Kawada, MD, and Shunji Sano, MD Department of Cardiovascular Surgery, Okayama University Graduate School of Medicine and Dentistry, Okayama, Japan Background. Atrial natriuretic peptide (ANP), a stimulator of particulate guanylate cyclase, has been found to protect against reoxygenation-induced hypercontracture in isolated cardiomyocytes by increasing cyclic guanosine monophosphate synthesis. The purpose of this study was to investigate the cardioprotective effects of ANP against ischemia-reperfusion injury in isolated rat hearts. Methods. Twenty-four hearts were perfused with ANP at 0.01, 0.1, and 1 mol/l or without ANP (n 6 each) in normoxic conditions. Because 0.1 mol/l ANP induced a threefold increase in cyclic guanosine monophosphate release into the coronary effluent without any effect on cardiac function, we used the 0.1 mol/l ANP dose for ischemia-reperfusion studies. Eighteen hearts were subjected to 15 minutes of normothermic global ischemia followed by 15 minutes of reperfusion. The hearts were divided into three groups (n 6 each). In group 1, ANP was added before ischemia. In group 2, ANP was added to the reperfusate. Hearts were untreated in the control group. Results. In group 1, the postischemic recovery of cardiac output, coronary flow, and cyclic guanosine monophosphate release was similar to the control group. In group 2, the recovery of cardiac output was significantly better than the control group (82.1% 9.8% vs 61.8% 6.8%, respectively, p < 0.01) with a similar trend to recovery of coronary flow (90.7% 8.5% vs 79.3% 11.8%, respectively). The improved cardiac function was closely related to a significant increase in postischemic cyclic guanosine monophosphate release. Conclusions. Administration of ANP at the time of reperfusion protects the myocardium from ischemiareperfusion injury. The concentrations of administration must not only increase the release of cyclic guanosine monophosphate release, but also lack negative inotropic effects. (Ann Thorac Surg 2004;77:233 7) 2004 by The Society of Thoracic Surgeons Accepted for publication July 30, Address reprint requests to Dr Ishino, Department of Cardiovascular Surgery, Okayama University Graduate School of Medicine and Dentistry, Shikata-cho, Okayama-City , Japan; ishino@tb3.so-net.ne.jp. Ischemia-reperfusion injury after regional or global ischemia involves damage to cardiomyocytes, vascular smooth muscle, and endothelial cells. When cardiomyocytes are reoxygenated after a prolonged period of energy depletion, severe cytosolic calcium overload and reactivation of energy production cause deleterious hypercontracture, which leads to cell disruption in tissue ( oxygen paradox [1]). Recent experimental studies have demonstrated that reoxygenation-induced hypercontracture can be prevented if the contractile apparatus is temporarily blocked during the initial phase of reoxygenation to reestablish normal cytosolic calcium control [2]. Myocardial guanosine 3, 5 -cyclic monophosphate (cgmp), which reduces calcium sensitivity of myofilaments [3], is severely depleted during reperfusion after prolonged ischemia [4, 5]. Increasing cgmp synthesis by atrial natriuretic peptide (ANP), a stimulator of the particulate guanylate cyclase, has been found to protect against reoxygenation-induced hypercontracture in isolated cardiomyocytes [6]. However, it has been suggested that too high concentration of cgmp could be detrimental to reperfused myocardium because of its contractile inhibitory action [5], subsequently leading to apoptotic cell death [7]. In the present study, we initially determined the optimal concentration of ANP to induce a sufficient increase in cgmp synthesis without producing negative inotropic effects in normoxic perfused hearts. We then investigated the cardioprotective effects of ANP against ischemiareperfusion injury in isolated rat hearts. Material and Methods Isolated Perfused Rat Heart We modified and used an isolated, perfused, rat heart apparatus that was previously described by Hearse and associates [8]. This circuit was designed to work in two interconvertible perfusion conditions, which were the Langendorff and working modes. In the former mode, 2004 by The Society of Thoracic Surgeons /04/$30.00 Published by Elsevier Inc doi: /s (03)

2 234 SANGAWA ET AL Ann Thorac Surg MYOCARDIAL PROTECTION BY ANP 2004;77:233 7 The pulmonary artery was incised to facilitate coronary drainage. The heart was then perfused in a retrograde manner under Langendorff mode perfusion conditions at 37.0 C for 10 minutes. Fig 1. Experimental protocol for normoxic study. Atrial natriuretic peptide (0.01, 0.1, 1 mol/l) was added to the perfusate for 15 minutes of working perfusion (W2). (cgmp cyclic guanosine monophosphate; L Langendorff perfusion; W working perfusion.) hearts were perfused through the aorta at a pressure of 80 cm H 2 O and continued to beat without external work. In the latter mode, hearts were perfused through the left atrium at an atrial pressure of 20 cm H 2 O. The left ventricle spontaneously ejected against a hydrostatic pressure of 100 cm H 2 O. The hearts were not paced. Only aortic flow was pooled and recirculated, but coronary effluent was discarded. The perfusate was a modified Krebs-Henseleit bicarbonate buffer (KHB) that was composed of the following (in mmol/l): NaCl, 118; NaHCO 3, 25; KCl, 4.7; CaCl 2, 2.5; MgSO 4, 1.2; KH 2 PO 4, 1.2; and glucose, 11. The KHB was bubbled with 95% oxygen and 5% carbon dioxide gas at 37.0 C to maintain the aortic partial pressure of oxygen higher than 500 mm Hg. It was then filtered through a cellulose acetate membrane (pore size 0.45 m) to remove any particulate contaminants [9]. Male Wister rats weighing 280 to 380 g were used in all experiments. Forty-two hearts were allocated to either normoxic or ischemic-reperfusion studies. Six hearts were investigated in each group. All animals received humane care in compliance with the Guide for the Care and Use of Laboratory Animals published by the National Institutes of Health (National Institutes of Health publication 85-23, revised 1985). The animals were anesthetized with diethyl ether. The femoral vein was exposed, and heparin (200 IU) was injected. The abdominal cavity was opened through a transverse incision, the diaphragm was transected, and lateral incisions were made along both sides of the rib cage. The anterior chest wall was folded back. The heart was excised and immediately placed into cold (4 C) KHB. The aorta and left atrium were cannulated within 1 minute after excision. Normoxic Study We determined the dose of ANP ( -human atrial natriuretic peptide; Suntory Co, Tokyo, Japan) that would induce the maximum increase in myocardial cgmp concentration without negative inotropic effects. Twentyfour hearts were perfused under normoxic conditions. The experimental protocol is depicted in Figure 1. After 10 minutes of Langendorff perfusion, conditions were switched to the working mode. Heart rate, coronary flow, and cardiac output were measured every 5 minutes, and the values were averaged to assess cardiac function. Heart rate was obtained from the aortic pressure tracing. Aortic flow was measured in the aortic column by an electromagnetic flowmeter (MFV-1100; Nihon Koden Co, Tokyo, Japan). The coronary flow was measured by timed volumetric collection from the right side of the heart. Cardiac output was calculated by summing the aortic and coronary flows. The coronary effluent was collected for 1 minute to measure the release of cgmp. After the second Langendorff perfusion, hearts were again perfused in the working mode with medium containing 0.01, 0.1, and 1 mol/l ANP (n 6 in each concentration). The coronary effluent was discarded to keep ANP concentration at a constant level in the circuit. Six hearts were perfused without ANP (control). After the second working perfusion, assessment of cardiac function was repeated, and the coronary effluent was collected for the measurement of cgmp release. Hearts were removed from the apparatus at the end of each experiment and heated to 70 C for 14 days. Hearts were then reweighed to determine the dry weight of the ventricular myocardium. The concentration of cgmp was determined by radioimmunoassay as previously described [10] and was expressed in picomoles per grams dry weight per minute. Results are presented in Table 1. Cardiac function after the second working perfusion is expressed as a percentage of the value obtained during the initial working perfusion without ANP. Heart rate and coronary flow were not affected by any concentration of ANP. The Table 1. Percent Changes in Heart Rate, Coronary Flow, Cardiac Output, and Cyclic Guanosine Monophosphate Leakage With Administration of Atrial Natriuretic Peptide Variable Control Atrial Natriuretic Peptide Heart rate (%) Coronary flow (%) Cardiac output (%) a Cyclic GMP (%) a a a p 0.01 versus control. GMP guanosine monophosphate.

3 Ann Thorac Surg SANGAWA ET AL 2004;77:233 7 MYOCARDIAL PROTECTION BY ANP Fig 2. Experimental protocol for ischemia-reperfusion study. (cgmp cyclic guanosine monophosphate; L Langendorff perfusion; W working perfusion.) Table 2. Heart Rate, Coronary Flow, and Cardiac Output During Working Mode Perfusion Before Ischemia Variable Control Group 1 Group 2 Heart rate (beats/min) Coronary flow (ml/min) Cardiac output (ml/min) Cyclic GMP (pmol dry weight 1 min 1 ) cardiac output of hearts perfused with 0.01 and 0.1 mol/l ANP did not change. However, the cardiac output of hearts perfused with 1 mol/l ANP was significantly lower than control hearts. Both 0.1 and 1 mol/l ANP induced a threefold increase in cgmp release (mean cgmp release; 19.2 to 65.5 pmol/grams dry weight per minute by 0.1 mol/l, 32.8 to 84.6 pmol/grams dry weight per minute by 1 mol/l). Ischemia-Reperfusion Study Because addition of 0.1 mol/l ANP to the perfusate significantly increased cgmp leakage and did not change cardiac function in the normoxic study, we used a concentration of ANP at 0.1 mol/l in the ischemiareperfusion study. Another reservoir was filled with the KHB containing 0.1 mol/l ANP and was connected to the aortic cannula. Thus hearts could be separately perfused with the KHB only or the KHB containing ANP. The experimental protocol is shown in Figure 2. After an initial washout period, hearts were perfused in the working mode for 20 minutes. During this phase, heart rate, coronary flow, and cardiac output were measured every 5 minutes, and the averages were considered as the preischemic indices. After 10 minutes of Langendorff perfusion, hearts were subjected to global ischemia by clamping the aortic and atrial inflow lines. The hearts were then maintained at 36.5 to 37.0 C for 15 minutes in the sealed chamber. Hearts were reperfused for 15 minutes in the Langendorff mode, followed by perfusion in the working mode for the next 20 minutes. The postischemic cardiac function was evaluated at the end of this phase. Eighteen hearts were allocated to one of the following three groups (n 6 in each): group 1, hearts were perfused with ANP during 10 minutes of Langendorff perfusion immediately before ischemia; group 2, hearts were perfused with ANP during 15 minutes of Langendorff perfusion immediately after ischemia; control group, hearts were untreated. Inasmuch as coronary effluent was discarded, the circuit was not contaminated by ANP. In all groups, the release of cgmp into the coronary effluent was measured at the end of the following perfusion phases; preischemic working perfusion (W1), preischemic Langendorff perfusion (L2), postischemic Langendorff perfusion (L3), and postischemic working perfusion (W3; Fig 2). GMP guanosine monophosphate. Criteria for Exclusion At the first hemodynamic evaluation, hearts presenting with a heart rate of less than 250 beats/min or aortic flow of less than 60 ml/min were considered to have suffered from myocardial damage during the preparation. These hearts were excluded from the study. Statistical Analysis All data were expressed as mean standard deviation. Differences among groups were first determined by the one-way-analysis of variance. Intergroup analysis was completed by multiple comparisons using the Dunnett test, when significance was indicated. A p value of less than 0.05 was regarded as statistically significant. Results Heart rate, coronary flow, cardiac output, and cgmp concentration in the coronary effluent determined during the preischemic working perfusion were not different among the groups (Table 2). The postischemic recoveries of heart rate, coronary flow, and cardiac output are expressed as a percentage of the preischemic value and are listed in Table 3. Changes in cgmp release are shown in Figure 3. There was no difference in the extent of functional recovery between group 1 hearts and control hearts. The cgmp release at the end of the postischemic Langendorff perfusion in group 1 was similar to the initial preischemic value, despite a significant increase at the end of the preischemic Langendorff perfusion. In group 2, the postischemic recovery of cardiac output was significantly better than the control group with a similar trend to recovery of Table 3. Postischemic Recovery of Heart Rate, Coronary Flow, and Cardiac Output Variable Control Group 1 Group 2 Heart rate (%) Coronary flow (%) Cardiac output (%) a a p 0.01 versus control.

4 236 SANGAWA ET AL Ann Thorac Surg MYOCARDIAL PROTECTION BY ANP 2004;77:233 7 Fig 3. Changes in cyclic guanosine monophosphate (cgmp) leakage in ischemia-reperfusion study. Control group (white bars) is compared with group 1 (striped bars) and group 2 (black bars). Error bar represents standard deviation. *p 0.01 versus control. (L2, L3 Langendorff perfusion periods 2, 3; W1, W3 working perfusion periods 1, 3.) coronary flow, although neither reached preischemic values. The improved cardiac function in group 2 was closely related to a significant increase in cgmp release at the end of the postischemic Langendorff perfusion. Comment The present study has shown that the administration of ANP elicited a significant improvement in the postischemic recovery of cardiac function in hearts that underwent 15 minutes of normothermic global ischemia. Resultant increases in cgmp release after the initial 15 minutes of reperfusion indicated that the protective effects of ANP against ischemia-reperfusion injury would be mediated, at least in part, by a direct effect of cgmp on cardiomyocytes. The timing of treatment application to increase cgmp for protection against hypoxia-reoxygenation or ischemia-reperfusion injury has been controversial. In hypoxia-reoxygenation studies, the temporary presence of cgmp-raising agents, such as ANP [6] and nitric oxide donors [11], is required from the last minutes of anoxia to the initial phase of reoxygenation to prevent reoxygenation-induced hypercontracture in isolated cardiomyocytes. Agullo and colleagues [12] reported that l-arginine supplementation before hypoxia increases cgmp release during reoxygenation and improves functional recovery in isolated rat hearts submitted to 40 minutes of hypoxia. Recent ischemia-reperfusion studies [4, 5] have demonstrated that urodilatin, a member of the natriuretic peptide family, improves functional recovery when administered at the time of reperfusion. It also limits myocardial necrosis in isolated rat hearts submitted to 60 minutes of ischemia. In the present study, administration of ANP during the initial reperfusion improved postischemic recovery, but no improvement was observed when added to the perfusate before ischemia. Reperfusion-induced hypercontracture occurs only after intracellular acidosis is corrected on restoration of myocardial oxygen supply. Hence, this time window could allow treatment applications of ANP at reperfusion to exert protective effects that increase the concentration of cgmp. The relation between cgmp concentration in reperfused myocardium and the extent of protective effects against reperfusion injury is not known. It has been suggested that overstimulation of cgmp synthesis could be detrimental to reperfused myocardium and may result in enhanced apoptotic cell death [7]. Padilla and associates [5] clearly demonstrated this negative effect of too high levels of cgmp in isolated rat hearts that received large doses of urodilatin after 40 minutes of ischemia. Therefore, it is important to determine the doses of agents necessary to obtain the targeted increase of cgmp in reperfused myocardium. The actual cellular cgmp concentrations were not measured in this study. However, changes in cgmp release into the coronary effluent have been shown to be a reliable index of changes in myocardial cgmp synthesis in previous studies [4]. In the present study, 0.1 mol/l ANP, which induced an increase in cgmp release but lacked negative hemodynamic effects in normoxic hearts, showed cardioprotective action in the ischemia-reperfused hearts. However, the response of myocardial cgmp synthesis to ANP might be different between normoxic and ischemic conditions. Previous studies have demonstrated that ANP directly causes cytosolic acidification by means of cgmp signaling by disabling sarcolemmal sodium hydrogen exchange [13, 14]. Inasmuch as cytosolic acidification depresses myofilament sensitivity to calcium and causes reductions in tension development and contraction amplitude [15], negative inotropic effects of ANP are mediated in part by a desensitizing effect of cgmp on myofilaments. Although the effects of ANP on intracellular calcium kinetics remain unclear, it has been reported that ANP decreases calcium current through the activation of cgmp-dependent protein kinase [16]. Furthermore, cgmp is thought to modulate calcium efflux by stimulating the sarcolemmal sodium calcium exchanger [17]. We believe that, in the present study, either enhanced or prolonged acidosis, or both, or improved calcium kinetics could explain the beneficial effect of ANP at 0.1 mol/l on contractile recovery after 15 minutes of normothermic global ischemia. An experimental study has presented evidence for the existence of an independently functioning local reninangiotensin system in the heart [18]. In the isolated perfused rat heart, angiotensin II exacerbated ischemiainduced ventricular fibrillation and impaired cardiodynamics, whereas these deleterious effects were abolished by ANP [19]. Morales and coworkers [20] reported that a direct blockade of the local renin-angiotensin system with angiotensin II receptor antagonists ameliorates myocardial stunning after global ischemia. Hence, functional antagonism of angiotensin II may underlie the protective effect of ANP against ischemia-reperfusion injury.

5 Ann Thorac Surg SANGAWA ET AL 2004;77:233 7 MYOCARDIAL PROTECTION BY ANP There are several limitations in our study. First, exact effects of ANP on postischemic recovery of left ventricular contractility were not evaluated, as we measured cardiac output instead of the maximum rate of increase of left ventricular pressure and end-diastolic pressure as our indicator of left ventricular function. Second, we did not confirm a cause and effect relationship between cgmp and postischemic recovery by using cgmp analogs and antagonists. However, previous studies using isolated rat heart models demonstrated that improved functional recovery by administration of the natriuretic peptide urodilatin during initial reperfusion after 40 minutes of ischemia was reproduced by the cgmp analog 8-bromo-cGMP [4], and that reduction in lactate dehydrogenase release by urodilatin after 60 minutes of ischemia was abolished by adding the ANP receptor antagonist isatin [5]. Third, we used an isolated perfused preparation. Although the preparation was denervated, direct cardiac responses can be studied independent of the systemic effects of ANP. Fourth, we used a crystalloid solution in the perfusion circuit. Blood perfusion may have induced different results from those of crystalloid perfusion [21]. Because each blood component serves different roles during ischemia and reperfusion and may confuse our results, we used a simple crystalloid solution in this initial study. Our data indicate that ANP could protect the myocardium from ischemia-reperfusion injury when applied at the beginning of reperfusion. Furthermore, we used concentrations that maximally increased cgmp content without inducing negative inotropic actions in normoxic myocardium. If proper doses of ANP can be determined in patients undergoing cardiopulmonary bypass, controlled reperfusion with ANP could be a beneficial adjunctive for myocardial protection during open heart surgery. This study was supported in part by the Grant-in-Aid for Scientific Research (C) from the Ministry of Education, Science, and Culture in Japan (No ). We thank Zeria Pharmaceutical Corporation, Tokyo, Japan, for providing ANP as a generous gift. References 1. Hearse DJ, Humphrey SM, Chain EB. Abrupt reoxygenation of the anoxic potassium-arrested perfused rat heart: a study of myocardial enzyme release. J Mol Cell Cardiol 1973;5: Siegmund B, Klietz T, Schwartz P, Piper HM. Temporary contractile blockade prevents hypercontracture in anoxicreoxygenated cardiomyocytes. Am J Physiol 1991;260:H Shah AM, Spurgeon HA, Sollot SJ, Talo A, Lakatta EG. 8-Bromo-cGMP reduces the myofilaments response to Ca 2 in intact cardiac myocytes. Circ Res 1994;74: Inserte J, Garcia-Dorado D, Agullo L, Paniagua A, Soler- Soler J. Urodilatin limits acute reperfusion injury in the isolated rat heart. Cardiovasc Res 2000;45: Padilla F, Garcia-Dorado D, Agullo L, et al. Intravenous administration of the natriuretic peptide urodilatin at low doses during coronary reperfusion limits infarct size in anesthetized pigs. Cardiovasc Res 2001;51: Hempel A, Friedrich M, Schluter KD, Forssmann WG, Kuhn M, Piper HM. ANP protects against reoxygenation-induced hypercontracture in adult cardiomyocytes. Am J Physiol 1997;273:H Taimor G, Hofstaetter B, Piper HM. Apoptosis induction by nitric oxide in adult cardiomyocytes via cgmp-signaling and its impairment after simulated ischemia. Cardiovasc Res 2000;45: Hearse DJ, Stewart DA, Braimbridge MV. Hypothermic arrest and potassium arrest. Circ Res 1975;36: Nakanishi K, Inoue M, Sugawara E, Sano S. Ischemic and reperfusion injury of cyanotic myocardium in chronic hypoxic rat model: changes in cyanotic myocardial antioxidant system. J Thorac Cardiovasc Surg 1997;114: Steiner AL, Parker CW, Kipnis DM. Radioimmunoassay for cyclic nucleotides 1. Preparation of antibodies and iodinated cyclic nucleotides. J Biol Chem 1972;247: Schluter KD, Weber M, Schraven E, Piper HM. NO donor SIN-1 protects against reoxygenation-induced cardiomyocyte injury by a dual action. Am J Physiol 1994;267:H Agullo L, Garcia-Dorado D, Inserte J, et al. l-arginine limits myocardial cell death secondary to hypoxia-reoxygenation by a cgmp-dependent mechanism. Am J Physiol 1999;276: H Ito N, Bartunek J, Spitzer KW, Lorell BH. Effects of the nitric oxide donor sodium nitroprusside on intracellular ph and contraction in hypertrophied myocytes. Circulation 1997;95: Tajima M, Bartunek J, Weinberg EO, Ito N, Lorell BH. Atrial natriuretic peptide has different effects on contractility and intracellular ph in normal and hypertrophied myocytes from pressure-overload hearts. Circulation 1998;98: Spitzer KW, Bridge JH. Relationship between intracellular ph and tension development in resting ventricular muscle and myocytes. Am J Physiol 1992;262:C Tohse N, Kanno M. Human atrial natriuretic peptide decreases cardiac Ca current through activation of cgmpdependent protein kinase [Abstract]. Jpn J Pharmacol 1993; 61(Suppl 1): Furukawa K, Tawada Y, Shigekawa M. Regulation of the plasma membrane Ca 2 pump by cyclic nucleotides in cultured vascular smooth muscle cells. J Biol Chem 1988;263: Lindpaintner K, Wilhelm MJ, Jin M, et al. Tissue reninangiotensin systems: focus on the heart. J Hypertension 1987;5(Suppl 2):S Linz W, Scholkens BA, Albus U, Petry P, Breipohl G, Knolle J. Atrial natriuretic factor protects the isolated working ischaemic rat heart against the action of angiotensin II. J Hypertension 1988;6(Suppl 4):S Morales C, Rodriguez M, Scapin O, Gelpi RJ. Comparison of the effects of ACE inhibition with those of angiotensin II receptor antagonism on systolic and diastolic myocardial stunning in isolated rabbit heart. Mol Cell Biochem 1998;186: Walters HL III, Digerness SB, Naftel DC, et al. The response to ischemia in blood perfused versus crystalloid perfused isolated rat heart preparations. J Mol Cell Cardiol 1992;24: