Cardioprotective Mechanism of Ischemic Preconditioning Is Impaired by Postinfarct Ventricular Remodeling Through Angiotensin II Type 1 Receptor Activation
Background—Activation of protein kinase C–linked receptors and subsequent opening of the mitochondrial ATP-sensitive K+ (mitoKATP) channel are crucial in preconditioning (PC). This study examined whether postinfarct ventricular remodeling interferes with the PC mechanism.
Methods and Results—Two weeks before isolation of hearts, rabbits underwent a sham operation or coronary ligation (COL) to induce remodeling. Isolated buffer-perfused hearts were subjected to 30-minute global ischemia/2-hour reperfusion, and infarct size was expressed as a percentage of the left ventricle (%I/LV), from which the scarred infarct by COL was excluded. Although %I/LV was similar in sham-operated and remodeled hearts (52.9±6.5% versus 45.8±5.2%), PC with 2 episodes of 5-minute ischemia protected sham-operated but not remodeled hearts (%I/LV=18.1±2.5% versus 54.8±2.9%, P<0.05). Infusion of valsartan (10 mg · kg−1 · d−1), an angiotensin II type 1 (AT1) receptor blocker, for 2 weeks after COL prevented the ventricular remodeling and preserved the response to PC (%I/LV=27.4±3.8%), although valsartan alone did not change %I/LV. Diazoxide, a mitoKATP channel opener, protected both sham-operated and remodeled hearts (%I/LV=14.1±3.1% and 8.3±3.6%).
Conclusions—The myocardium remodeled after infarction is refractory to PC, which is probably due to interruption of cellular signaling by PC upstream of mitoKATP channels. An AT1 receptor blocker is beneficial not only for suppression of ventricular remodeling but also for preservation of the PC mechanism.
Left ventricular volumes and function have been recognized to be major determinants of survival after acute myocardial infarction.1 When infarct size is larger than ≈30% of the ventricle, dilation of the left ventricle and hypertrophy of the residual myocardium progress over months, or even years, after myocardial infarction.2 3 This process of ventricular remodeling substantially influences the prognosis of patients who have survived the acute phase of myocardial infarction.1 3 However, whether ventricular remodeling results in any alteration in the myocardial response to ischemia has not been fully established. In the normal myocardium, anti-infarct tolerance is markedly enhanced by exposure to brief transient ischemia, which is called ischemic preconditioning (PC).4 However, it remains unknown whether the remodeled myocardium can be effectively preconditioned. The mechanism of PC is triggered by activation of several classes of Gq protein–coupled receptors.5 6 7 8 9 The signals from these receptors appear to be transmitted through protein kinase C (PKC)7 8 10 and mitogen-activated protein kinase (MAPK)11 12 to mitochondrial ATP-sensitive K+ channels (mitoKATP channels), the opening of which enhances anti-infarct tolerance.11 13 14 Because activation of PKC and MAPK15 16 by angiotensin (Ang) II type 1 (AT1)17 and endothelin18 receptor stimulation is involved in hypertrophy of the residual myocytes after infarction, it is possible that signals relevant to the remodeling process may interfere with the mechanism of PC.
Accordingly, the present study examined (1) whether the alteration of anti-infarct tolerance and/or response to PC in the remodeled myocardium, if any, is prevented by the suppression of ventricular remodeling by an AT1 receptor antagonist (valsartan) and (2) whether the PC mechanism downstream of the mitoKATP channel, a tentative effector of PC, remains intact in the remodeled myocardium.
This study was conducted in accordance with the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health (NIH publication No. 85-23, revised 1996).
Experiment 1: Effects of Ventricular Remodeling on Infarct Size and PC
Surgery for Preparing Ventricular Remodeling
Male rabbits (Japanese White) were anesthetized with intravenous sodium pentobarbital (30 mg/kg) and ventilated with a Harvard respirator (model 683, Harvard Apparatus) with room air and oxygen supplement. After a left thoracotomy, a 4-0 silk thread was passed around a marginal branch of the left coronary artery. Rabbits were then divided into 3 groups: the Sham, coronary ligation (COL), and COL+valsartan (Val) groups. In the Sham group, the coronary artery was not ligated. In the COL and COL+Val groups, the coronary branch was permanently occluded by the ligature, and an osmotic minipump (Alzet model 2 ML4; Alza Co) was subcutaneously implanted in rabbits in the COL+Val group. The osmotic minipump was filled and adjusted to deliver valsartan at 10 mg · kg−1 · d−1 for 2 weeks. We confirmed in pilot experiments that infusion of this dose of valsartan inhibits pressor response to Ang II (0.5 μg/kg IV) by 90%, which is comparable to the effect of acute intravenous injection of 1 mg/kg valsartan. The surgical wounds were repaired, and the rabbits were returned to their cages for recovery. These surgical procedures were performed under sterile conditions, and 100 mg ampicillin and 100 mg cloxacillin were injected intramuscularly for prophylaxis of infection.
Two weeks after surgery, each rabbit was brought into the laboratory and reanesthetized and ventilated. A catheter was placed in the carotid artery and connected to a Nihon-Kohden SCK-580 transducer, and bipolar electrodes were placed across the chest. After blood pressure and heart rate had been measured, arterial blood was sampled for catecholamines and Ang II assay. Each rabbit received 2000 U heparin, and the heart was quickly excised for isolated heart preparation.
Isolated Rabbit Heart Preparation
The excised heart was mounted onto a Langendorff apparatus with a water jacket and perfused at 75 mm Hg pressure with noncirculating Krebs-Henseleit buffer (mmol/L: NaCl 118.5, KCl 4.7, MgSO4 1.2, KH2PO4 1.2, NaHCO3 24.8, CaCl2 2.5, and glucose 10). The buffer was gassed with 95% O2/5% CO2, resulting in a pH of 7.4 to 7.5, and the temperature of the perfusate was maintained at 38°C. A fluid-filled latex balloon with a PE-160 tube was inserted into the left ventricle and was connected to an SCK-580 transducer. Baseline left ventricular end-diastolic pressure was adjusted to <5 mm Hg. Atrial pacing was performed at 210 bpm if the spontaneous rate was lower. Coronary flow was measured by timed collection of perfusate dripping from the heart.
Infarct Size Experiment In Vitro
After 20 minutes of equilibration, the heart was subjected to no pretreatment (No-Tx); PC with 2 cycles (PC×2) or 4 cycles (PC×4) of 5-minute global ischemia/5-minute reperfusion; or 10-minute infusion of diazoxide (100 μmol/L), a mitoKATP channel opener. Myocardial infarction was then induced by 30-minute global ischemia and 2-hour reperfusion. Global ischemia was achieved by complete interruption of coronary perfusion.
Measurement of Infarct Size
After 2 hours of reperfusion, hearts were weighed, frozen, and cut into 2-mm-thick sections from apex to base. Infarcts in the heart slices were visualized by tetrazolium staining as previously reported.8 19 The fresh infarcts, scar areas due to coronary ligation, and outlines of the ventricle were traced on a clear acetate sheet. The scarred infarct due to coronary ligation was transmural and clearly distinguished from other areas of the left ventricle (Figure 1⇓). The sizes of the infarct, scar regions, and left ventricle were measured by computer-assisted planimetry.19 Their volumes were obtained by multiplication of each area by 2 mm, ie, the thickness of the heart slice. In hearts that had received coronary ligation (ie, COL and COL+Val groups), the scar area was excluded from infarct size measurement.
Determination of Plasma Catecholamine and Ang II
Each blood sample was immediately heparinized and centrifuged at 2000 rpm at 4°C for 30 minutes to separate plasma. Levels of catecholamines were determined by HLC-8030 (Tohso Co Inc), an automated high-performance liquid chromatograph. Plasma Ang II level was determined by radioimmunoassay, as in our previous study.19
Experiment 2: Pharmacological PC With Ang II in Remodeled Hearts
Surgery and Isolated Rabbit Heart Preparation
As in experiment 1, rabbits were divided into Sham and COL groups. Rabbits in the former group received a sham operation, and the coronary artery was ligated in the latter group. Hearts were isolated 2 weeks later and perfused in Langendorff mode. In addition to coronary flow and left ventricular pressure (LVP), the first derivative of the LVP (LV dP/dt) was measured by Nihon-Kohden EQ-601G, an electrical differentiation unit.
Protocol 1: hemodynamic responses to Ang II in remodeled hearts. To assess possible alteration in AT1 receptors by ventricular remodeling, effects of 100 nmol/L and 1 μmol/L Ang II on coronary flow, LVP, and LV dP/dt were measured in sham-operated and remodeled hearts.
Protocol 2: effect of pharmacological PC with Ang II on infarct size. Infarction was induced by 30-minute global ischemia/2-hour reperfusion, and infarct size was determined as in experiment 1. Hearts received no drug or infusion of 100 nmol/L Ang II for 10 minutes commencing at 15 minutes before the onset of ischemia.
Diazoxide and Ang II were obtained from Sigma Chemical Co. Valsartan was kindly provided by Novartis Pharma AG.
All data are presented as mean±SEM. Difference in mortality rate was tested by the χ2 test. One-way ANOVA combined with the Student-Newman-Keuls post hoc test was used to test for differences in plasma catecholamine and Ang II levels and infarct size between groups. Repeated-measures ANOVA was used to test for differences in hemodynamics in any given group. The difference was considered significant at a level of P<0.05.
Mortality After Coronary Ligation
Seventy rabbits were initially entered into this experiment. Two rabbits each in the COL and COL+Val groups died of ventricular fibrillation during surgery. One rabbit in the Sham group and 4 rabbits each in the COL and COL+Val groups died in their cages after surgery, probably due to atelectasis in the lungs and/or heart failure. Although the valsartan treatment lowered mean blood pressure by ≈20 mm Hg without alteration of heart rate, there was no statistically significant difference in mortality rates among groups. Therefore, 57 surviving rabbits contributed to the following analysis.
Table 1⇓ summarizes hemodynamic parameters. There were no significant differences in baseline heart rate, left ventricular developed pressure (LVDP), or coronary flow among the groups, although LVDP in the COL group was slightly lower than that in the Sham group. PC with 2 and 4 cycles of 5-minute global ischemia/5-minute reperfusion reduced LVDP and tended to increase coronary flow before 30-minute global ischemia. Administration of diazoxide had little effect on heart rate and LVDP but increased coronary flow. LVDP and coronary flow after reperfusion were decreased in all groups, without significant intergroup differences.
Infarct Size Data
As shown in Table 2⇓, heart weight and heart/body weight were significantly larger in the COL group than in the Sham group. Increases in heart weight and heart/body weight were not observed in the COL+Val group, indicating that valsartan prevented ventricular remodeling. There was no significant difference in the sizes of scarred infarct in the COL and COL+Val groups, ranging from 20% to 30% of the left ventricle. Within each of the Sham, COL, and COL+Val groups, risk area size was comparable in the hearts subjected to No-Tx, PC×2, PC×4, and diazoxide. Infarct size as a percentage of the left ventricle (%I/LV) in each heart is presented in Figure 2⇓. In the Sham group, PC×2 significantly reduced %I/LV from 52.9±6.5% to 18.1±2.5%. This infarct size–limiting effect of PC×2 was mimicked by diazoxide (%I/LV=14.1±3.1%). In the COL group, %I/LV with No-Tx (45.8±5.2%) was comparable to %I/LV with No-Tx in the Sham group, but PC failed to limit infarct size: %I/LV was 54.8±2.9% in PC×2 and 55.1±8.6% in PC×4. In contrast, diazoxide significantly reduced %I/LV in the COL group as well (%I/LV=8.3±3.6%). In the COL+Val group, PC×2 reduced %I/LV from 56.4±5.0% in the hearts with No-Tx to 27.4±3.8%. This value is slightly larger than the %I/LV in the preconditioned hearts in the Sham group, but the difference did not reach statistical significance. To exclude the possibility that the preserved PC effect in the COL+Val group is due to AT1 receptor blockade shortly before the ischemic insult, we examined the effect of valsartan (1 mg/kg IV) administered 10 minutes before isolation of the heart in 3 rabbits in the COL group. In these hearts, PC×2 failed to reduce infarct size (%I/LV=51.8±11.0%).
Plasma Catecholamine and Ang II Levels
There was a clear trend toward higher plasma epinephrine and norepinephrine levels in the COL group (54.4±20.7 and 354.9±67.1 pg/mL, n=12) than in the Sham group (17.5±4.3 and 207.2±19.0 pg/mL, n=11), although the differences did not reach statistical significance, presumably because of the small number of animals. In the COL+Val group, the plasma Ang II level (181.9±25.4 pg/mL, n=10) was 6-fold higher than the levels in the Sham and COL groups (27.1±6.1 and 29.7±6.5 pg/mL, respectively), reflecting chronic blockade of the AT1 receptor in juxtaglomerular cells.20
In the Sham group, 100 nmol/L and 1 μmol/L Ang II increased LVDP and maximum LV dP/dt by ≈10% and 20%, respectively (Table 3⇓). A similar inotropic response to Ang II was observed in the COL group, in which the size of scarred infarct was 25.2±4.8% of the left ventricle. Coronary flow was reduced by 15% in both study groups.
Baseline heart rate and coronary flow were comparable in hearts from the Sham and COL groups, whereas LVDP was 20% lower in the COL group (data not shown). As in protocol 1, infusion of Ang II (100 nmol/L) decreased coronary flow by ≈15% and increased LVDP by 10%, and these parameters returned almost to their baseline levels after a 5-minute washout period (data not shown). As shown in Table 4⇓, %I/LV was significantly limited by pretreatment with Ang II in the sham-operated hearts but not in the remodeled hearts.
The present study showed that ventricular remodeling after infarction makes the myocardium refractory to PC against infarction. This loss of myocardial response to PC is probably due to failure of cellular signaling by PC to reach mitoKATP channels, because diazoxide protected the heart from infarction in the remodeled myocardium as in the nonremodeled hearts. Furthermore, chronic treatment with valsartan prevented ventricular remodeling after infarction and preserved PC. These results suggest that AT1 receptor activation plays a crucial role in the loss of myocardial response to PC during ventricular remodeling.
The present study showed for the first time that the remodeled myocardium after infarction is refractory to PC. This change in myocardial response to PC is unlikely to be explained by increased threshold for PC, because an increase in PC episodes from 2 to 4 cycles of ischemia/reperfusion failed to induce cardioprotection (Table 2⇑). It should also be noted that this change in response to PC occurred early (ie, 2 weeks) after infarction in the modestly hypertrophied myocytes. Furthermore, none of the rabbits showed clear physical signs of heart failure. These findings suggest that the PC mechanism is impaired at the early stage of ventricular remodeling even without severe heart failure.
The effects of valsartan on heart weight (Table 2⇑) and on PC (Figure 2⇑) strongly indicate that the loss of response to PC in the remodeled myocardium is AT1 receptor–mediated. Valsartan, a noncompetitive AT1 receptor antagonist, suppressed the ventricular hypertrophy and preserved ≈80% of the infarct size–limiting effect of PC in the heart 2 weeks after infarction. Because the dissociation of valsartan from its receptor is not rapid (t1/2=56 minutes),21 AT1 receptors might be blocked during PC ischemia and/or subsequent global ischemia in an isolated condition. However, AT1 receptor blockade during these periods cannot explain the effect of valsartan on PC. First, an acute injection of valsartan shortly before the heart was isolated failed to restore the PC effect in the remodeled heart. Second, a previous study from this laboratory showed that blockade of AT1 receptors at the time of PC partly blocks cardioprotection of PC in vivo,19 presumably through attenuation of PKC stimulation. Therefore, the preserved myocardial response to PC in the valsartan-treated rabbits is most likely to have been the result of suppressed remodeling and not the direct effect of valsartan on PC.
It is not clear how the PC mechanism is impaired during postinfarct ventricular remodeling, but some speculation is possible. The mechanism of PC is triggered by the stimulation of adenosine A1/A3,5 6 bradykinin B2,7 opioid,8 9 and AT119 22 receptors and by the generation of free radicals.23 These G protein–coupled receptors and free radicals activate signal cascades, in which PKC,7 8 10 tyrosine kinase,24 and MAPK11 12 are involved, to the mitoKATP channel, a probable effector of PC.11 13 14 Because the remodeled heart is protected from infarction by diazoxide, the loss of the PC effect due to remodeling must be somewhere upstream of the mitoKATP channel. Therefore, theoretical possibilities are (1) insufficient generation of receptor agonists or free radicals triggering PC, (2) downregulation of PC-relevant receptors, and (3) downregulation of PKC and/or other kinases transmitting signals to the mitoKATP channel. In contrast to PC in the in situ heart, PC in the buffer-perfused isolated heart is not triggered by endogenous bradykinin7 and Ang II,22 because the kininogen and angiotensinogen content in the cardiac interstitium is limited. It is controversial whether the opioid store in the heart contributes to PC in the isolated heart.8 9 Conversely, the contribution of adenosine and free radicals has been confirmed not only in hearts in situ but also in buffer-perfused isolated hearts.5 6 23 However, production of these triggers is unlikely to be reduced in the remodeled myocardium. The myocardial content of ATP and its metabolism, except for reduced creatine kinase reaction, are generally preserved in the remodeled myocardium,25 and the density of adenosine A1 receptors remains normal.26 It has been reported that activities of superoxide dismutase and glutathione peroxidase are progressively decreased after myocardial infarction with development of heart failure,27 which would allow an increase in free radical production in PC in the remodeled myocardium.
The results in experiment 2 support the possibility that PKC and/or other kinases transmitting signals to mitoKATP channels are downregulated. Inotropic responses to Ang II were similar in the sham-operated and remodeled hearts, indicating that the AT1 receptor was not substantially downregulated after 2 weeks of the remodeling process in the rabbit. However, pharmacological PC with Ang II failed to protect remodeled hearts against infarction. These findings suggest that the AT1 receptor did not activate kinase cascades responsible for the PC effect. Because either 1 of 2 second messengers of the AT1 receptor, ie, inositol 1,4,5-trisphosphate and PKC, potentially mediates the Ang II–induced positive inotropic response,28 the present results do not give an insight into PKC in the remodeled heart. Nevertheless, it has been established that activation of Ang II17 and of endothelin receptors18 contributes to myocardial hypertrophy in the remodeling heart. Persistent activation of these PKC- and MAPK-linked pathways during remodeling might downregulate these kinases, which play roles in PC-induced cardioprotection as well.10 11 12
There is now substantial circumstantial evidence for the existence of PC in the human myocardium. It has been shown that cultured human cardiomyocytes can be preconditioned and protected from infarction, as the rabbit cardiomyocytes were.29 Clinical studies30 31 suggest that preinfarct angina within 24 hours before the onset of acute myocardial infarction significantly improves the cardiac function and prognosis of patients. Although it is unclear whether the loss of myocardial response to PC is involved in the poor prognosis of patients with postinfarct ventricular remodeling, the present results suggest that AT1 receptor antagonists and probably ACE inhibitors are beneficial not only for suppressing ventricular remodeling17 but also for preserving the PC mechanism.
This study was supported by grants-in-aid for research from the Ministry of Education, Science, Culture, and Sports of Japan (08670812 to Dr Miura).
- Received July 9, 1999.
- Revision received February 14, 2000.
- Accepted February 29, 2000.
- Copyright © 2000 by American Heart Association
White HD, Norris RM, Brown MA, et al. Left ventricular end-systolic volume as the major determinant of survival after recovery from myocardial infarction. Circulation. 1987;76:44–51.
Pfeffer JM, Pfeffer MA, Fletcher PJ, et al. Progressive ventricular remodeling in rat with myocardial infarction. Am J Physiol. 1991;260:H1406–H1414.
Murry CE, Jennings RB, Reimer KA. Preconditioning with ischemia: a delay of lethal cell injury in ischemic myocardium. Circulation. 1986;74:1124–1136.
Liu GS, Thornton J, Van Winkle DM, et al. Protection against infarction afforded by preconditioning is mediated by A1 adenosine receptors in rabbit heart. Circulation. 1991;84:350–356.
Liu GS, Richards SC, Olsson RA, et al. Evidence that adenosine A3 receptor may mediate the protection afforded by preconditioning in the isolated rabbit heart. Cardiovasc Res. 1994;28:1057–1061.
Goto M, Liu Y, Yang X-M, et al. Role of bradykinin in protection of ischemic preconditioning in rabbit hearts. Circ Res. 1995;77:611–621.
Ytrehus K, Liu Y, Downey JM. Preconditioning protects ischemic rabbit heart by protein kinase C activation. Am J Physiol. 1994;266:H1145–H1152.
Baines CP, Liu GS, Birincioglu M, et al. Ischemic preconditioning depends on interaction between mitochondrial KATP channels and actin cytoskeleton. Am J Physiol. 1999;276:H1361–H1368.
Ping P, Zhang J, Cao X, et al. PKC-dependent activation of p44/p42 MAPKs during myocardial ischemia-reperfusion in conscious rabbits. Am J Physiol. 1999;276:H1468–H1481.
Garlid KD, Paucek P, Yarov-Yarovoy V, et al. Cardioprotective effect of diazoxide and its interaction with mitochondrial ATP-sensitive K+ channels: possible mechanism of cardioprotection. Circ Res. 1997;81:1072–1082.
Sato T, O’Rourke B, Marbán E. Modulation of mitochondrial ATP-dependent K+ channels by protein kinase C. Circ Res. 1998;83:110–114.
Clerk A, Michael A, Sugden PH. Stimulation of the p38 mitogen-activated protein kinase pathway in neonatal rat ventricular myocytes by the G protein-coupled receptor agonists, endothelin-1 and phenylephrine: a role in cardiac myocyte hypertrophy? J Cell Biol. 1998;142:523–535.
Bogoyevitch MA, Andersson MB, Gillespie-Brown J, et al. Adrenergic receptor stimulation of the mitogen-activated protein kinase cascade and cardiac hypertrophy. Biochem J. 1996;314:115–121.
Schieffer B, Wirger A, Meybrunn M, et al. Comparative effects of chronic angiotensin-converting enzyme inhibition and angiotensin II type 1 receptor blockade on cardiac remodeling after myocardial infarction in the rat. Circulation. 1994;89:2273–2282.
Pals DT, Couch SJ. Renin release induced by losartan (DuP 753), an angiotensin II receptor antagonist. Clin Exp Hypertens. 1993;15:1–13.
Neubauer S, Horn M, Naumann A, et al. Impairment of energy metabolism in intact residual myocardium of rat hearts with chronic myocardial infarction. J Clin Invest. 1995;95:1092–1100.
Böhm M, Gierschik P, Jakobs K-H, et al. Increase of Giα in human hearts with dilated but not ischemic cardiomyopathy. Circulation. 1990;82:1249–1265.
Mattiazzi A. Positive inotropic effect of angiotensin II increases in intracellular Ca2+ or changes in myofilament Ca2+ responsiveness? J Pharmacol Toxicol. 1997;37:205–214.
Ikonomidis JS, Tumiati LC, Weisel RD, et al. Preconditioning human ventricular cardiomyocytes with brief periods of simulated ischaemia. Cardiovasc Res. 1994;28:1285–1291.
Kloner RA, Shook T, Antman EM, et al, and the TIMI-9B investigators. Prospective temporal analysis of the onset of preinfarction angina versus outcome: an ancillary study in TIMI-9B. Circulation. 1998;97:1042–1045.