Paracrine Systems in the Cardioprotective Effect of Angiotensin-Converting Enzyme Inhibitors on Myocardial Ischemia/Reperfusion Injury in Rats
Abstract After transient episodes of ischemia, benefits of thrombolytic or angioplastic therapy may be limited by reperfusion injury. Angiotensin-converting enzyme inhibitors protect the heart against ischemia/reperfusion injury, an effect mediated by kinins. We examined whether the protective effect of the angiotensin-converting enzyme inhibitor ramiprilat on myocardial ischemia/reperfusion is due to kinin stimulation of prostaglandin and/or nitric oxide release. The left anterior descending coronary artery of Lewis inbred rats was occluded for 30 minutes, followed by 120 minutes of reperfusion. Immediately before reperfusion rats were treated with vehicle, ramiprilat, or the angiotensin II type 1 receptor antagonist losartan. We tested whether pretreatment with the kinin receptor antagonist Hoe 140, the nitric oxide synthase inhibitor NG-nitro-l-arginine methyl ester, or the cyclooxygenase inhibitor indomethacin blocked the effect of ramiprilat on infarct size and reperfusion arrhythmias. In controls, infarct size as a percentage of the area at risk was 79±3%; ramiprilat reduced this to 49±4% (P<.001), but losartan had little effect (74±6%, P=NS). Pretreatment with Hoe 140, NG-nitro-l-arginine methyl ester, or indomethacin abolished the beneficial effect of ramiprilat. Compared with the 30-minute ischemia/120-minute reperfusion group, nonreperfused hearts with 30 minutes of ischemia had significantly smaller infarct size as a percentage of the area at risk, whereas in the 150-minute ischemia group it was significantly larger. This suggests that reperfusion caused a significant part of the myocardial injury, but it also suggests that compared with prolonged ischemia, reperfusion salvaged some of the myocardium. Ventricular arrhythmias mirrored the changes in infarct size. Thus, angiotensin-converting enzyme inhibitors protect the myocardium against ischemia/reperfusion injury and arrhythmias; these beneficial effects are mediated primarily by a kinin–prostaglandin–nitric oxide pathway, not inhibition of angiotensin II formation.
- angiotensin-converting enzyme inhibitors
- myocardial reperfusion injury
- myocardial infarction
- nitric oxide
Myocardial ischemia due to coronary occlusion results in myocardial cell necrosis. After transient episodes of myocardial ischemia, the benefits of thrombolytic or angioplastic therapy in restoring blood flow may be limited by reperfusion injury. Although reperfusion can salvage the myocardium, the damaging effect of reperfusion injury per se may accelerate tissue necrosis with reversible myocardial injury. The mechanism and prevention of reperfusion injury have been the focus of intensive investigation (for review, see References 1 through 3). Many clinical and experimental studies have established the therapeutic benefits of ACE inhibitors, not only in treating hypertension and congestive heart failure but also in reducing reinfarction, limiting infarct size, and reducing reperfusion arrhythmias.4 5 6 7 ACE inhibitors block not only formation of Ang II but also degradation of kinins. Martorana et al8 reported that in acute dog studies intracoronary bradykinin reduced infarct size to the same extent as the ACE inhibitor ramiprilat, whereas the bradykinin B2 receptor antagonist Hoe 140 (d-Arg,[Hyp2,Thi5,8,d-Phe7]-bradykinin) abolished the effect of ramiprilat. Recently, Hartman et al9 10 demonstrated that in rabbits ramiprilat reduced infarct size after myocardial ischemia/reperfusion, whereas the Ang II type 1 antagonist losartan did not, and that this cardioprotective effect was mediated by kinins.
In the present study we tested the hypothesis that the decrease in myocardial infarct size produced by ACE inhibitors during ischemia/reperfusion injury is due to inhibition of kinin degradation, which in turn stimulates release of endothelial prostaglandins and NO. Our study focused on (1) whether the reduction of myocardial infarct size caused by ramiprilat is also observed in rats (as it is in rabbits and dogs11 ); (2) whether this protective effect is mediated by prostaglandins, as determined by blockade with the cyclooxygenase inhibitor indomethacin; (3) whether the effect is mediated by NO, as determined by administration of the NO synthase inhibitor L-NAME; and (4) whether attenuation of reperfusion arrhythmias occurs in rats in vivo, since ACE inhibitors reduced reperfusion arrhythmias in the isolated rat heart. We used Lewis inbred rats instead of the Sprague-Dawley strain because (1) our preliminary studies showed that the branching pattern of their left coronary arteries is very similar; (2) there is less variation in infarct size created by ligation of the same coronary arterial site; and (3) even though infarcts were larger, mortality rates were lower compared with Sprague-Dawley rats.
We induced ischemia and immediately before reperfusion treated rats with either vehicle, ramiprilat, or the Ang II type 1 receptor antagonist losartan. Thus any protective effect could only occur during the reperfusion period. We tested whether pretreatment with the kinin receptor antagonist Hoe 140, the NO synthase inhibitor L-NAME, or the cyclooxygenase inhibitor indomethacin could negate the beneficial effects of the ACE inhibitor.
Male Lewis inbred rats (Charles River Laboratories, Wilmington, Mass) weighing 280 to 330 g were housed in an air-conditioned room with a 12-hour light/dark cycle; they received standard laboratory rat chow (0.4% sodium) and drank tap water. Rats were given 5 to 7 days to adjust to their new environment. The study was approved by the Henry Ford Hospital Care of Experimental Animals Committee. All surgical procedures were performed with rats under sodium pentobarbital anesthesia (50 mg/kg IP).
Hemodynamics and Electrocardiography
A polyethylene catheter (PE-10 fused to PE-50) was inserted into the abdominal aorta via the femoral artery for measurement of arterial pressure and a second catheter into the vena cava via the femoral vein for administration of drugs. Mean blood pressure and heart rate were measured with a pressure transducer connected to a processor (Micron MP15) and monitored on a two-channel recorder (Brush 220, Gould Instruments). Electrocardiograms were obtained by subcutaneously inserting needle electrodes into the limbs and displaying the signal on an oscilloscope. An analog tape recorder (3968 A, Hewlett-Packard) was used for recordings of the electrocardiographic lead II signal, beginning 1 minute before reperfusion and continuing for 30 minutes. The duration of ventricular tachycardia and incidence of ventricular premature beats were analyzed. Ventricular tachycardia was defined as a run of four or more consecutive ventricular premature beats. Blood pressure and the electrocardiogram were monitored throughout the experiment and recorded during the control period; 15 and 30 minutes of occlusion; and 15, 30, 60, and 120 minutes of reperfusion.
Coronary Occlusion/Reperfusion and Measurements of Myocardial Infarct Size and Area at Risk
Rats were intubated and ventilated with room air. A thoracotomy was performed in the fourth intercostal space, and the lungs were retracted to expose the heart. After an incision was made in the pericardium, a 6-0 silk suture was passed loosely around the left anterior descending coronary artery. Both ends were threaded through a vinyl tube to form a snare, which was then fixed by clamping it with a hemostat to facilitate coronary occlusion and reperfusion. Once hemodynamics had stabilized, heparin (100 U) was given and coronary occlusion performed by tightening the suture loop for 30 minutes. Occlusion was verified by regional cyanosis of the myocardial surface distal to the suture, accompanied by changes in arterial pressure and elevation of the ST segment on the electrocardiogram. After 30 minutes the loop was loosened and reperfused as verified by return of the original color. The chest was closed in layers and the rat placed on a heating pad for recovery. After 120 minutes of reperfusion the rats were reanesthetized, and an additional 100 U heparin was administered. The loop around the left anterior descending coronary artery was retightened, and 5% Evans blue was rapidly injected into the LV to distinguish the nonischemic area from the area at risk. The heart was then excised, and the atria, great vessels, and right ventricle were dissected. The LV was cut into four slices perpendicular to the base-apex axis. The slices were incubated in a phosphate-buffered solution of triphenyltetrazolium chloride in saline for 15 to 30 minutes. Noninfarcted myocardium stained brick red in the presence of dehydrogenase enzymes, whereas necrotic tissue (infarct) remained unstained because of lack of enzymes. Each slice was photographed, magnified, and projected onto a screen. A planimeter was used for measurement of infarcted area (uncolored), area at risk (uncolored plus brick red), and nonoccluded area (blue-stained). The following parameters were averaged across the four slices for each heart: infarct size expressed as a percentage of the area at risk; infarct size expressed as a percentage of the total area of the slice (infarct size/LV); and area at risk, expressed as a percentage of the total area of the slice (risk area/LV).
In seven rats reperfusion after release of coronary ligation was confirmed by infusion of radioactive microspheres into the LV during the ischemic period and reperfusion. Microspheres (DuPont–New England Nuclear) with a diameter of 15±1.5 μm labeled with either 141Ce or 85Sr were suspended in 3.5 mol/L glucose at a concentration of 400 000/mL with the use of 0.01% Tween-80 as an antiaggregant and were agitated ultrasonically for approximately 15 minutes. After coronary ligation a volume of 0.2 mL of the suspension, corresponding to approximately 80 000 microspheres, was infused into the LV over 20 seconds via a 26-gauge needle. Two minutes after reperfusion was started, the second set of microspheres was infused. Five minutes later the coronary artery was again ligated and Evans blue solution infused into the LV for clear demarcation of the ischemic and nonischemic regions of the heart. These two sections were separated from each other and radioactivity counted with a Packard gamma counter at dual window settings of 50 to 250 and 400 to 700 MeV and a sample level of 0.5 cm. Results were expressed as a percentage of total radioactivity in the heart normalized per milligram tissue. During coronary occlusion only 7.4±4.2% radioactivity was in the ischemic region and 92.6±4.2% in the nonischemic region. In contrast, during reperfusion the distribution was 60.2±2.1% and 39.8±2.1%, respectively, clearly indicating that reperfusion had occurred.
Rats were randomly divided into nine groups: (1) treated with vehicle (saline, 0.3 mL) before reperfusion, (2) treated with ramiprilat (50 μg/kg) before reperfusion, (3) treated with losartan (10 mg/kg) before reperfusion, (4) pretreated with Hoe 140 (1 μg/kg) 15 minutes before coronary occlusion and then with vehicle before reperfusion, (5) pretreated with Hoe 140 before coronary occlusion and then ramiprilat before reperfusion, (6) pretreated with indomethacin (10 mg/kg) and then with vehicle before reperfusion, (7) pretreated with indomethacin and then with ramiprilat before reperfusion, (8) pretreated with L-NAME (10 mg/kg) before coronary occlusion and then with vehicle, and (9) pretreated with L-NAME and then with ramiprilat before reperfusion. All test agents were dissolved in saline. Hoe 140, L-NAME, or indomethacin was always given as an intravenous bolus injection 15 minutes before coronary occlusion to allow adequate time for inhibition. Vehicle, ramiprilat, or losartan was administered just before reperfusion as a slow bolus; thus their effect occurred only during reperfusion. L-NAME and indomethacin were purchased from Sigma Chemical Co. Ramiprilat was donated by Upjohn Co, Hoe 140 by Hoechst, and losartan by DuPont-Merck.
We used 16 additional rats to study the effect of ischemia without reperfusion on myocardial infarct size. Half of the rats underwent coronary ligation for 30 minutes and the other half for 150 minutes. Evans blue was rapidly injected into the LV, and the heart slices were treated with a 3% sucrose solution in 0.2 phosphate buffer (pH 7.4) and shaken in a rotator (Baxter) for 30 minutes for removal of the dehydrogenase enzyme from the necrotic myocytes. Then the slices were incubated with triphenyltetrazolium chloride for another 30 minutes, and infarct size was determined as in the ischemia/reperfusion study.
We used five rats to test whether the dose of losartan (10 mg/kg IV) could block the mean blood pressure response to exogenous Ang II (25, 50, and 100 ng). Dose-response curves were obtained before and 15 minutes and 2 hours after losartan administration. Ang II induced a dose-dependent increase in blood pressure that was blocked by losartan at both 15 and 120 minutes (P<.025) (Fig 1a⇓). We used another five rats to test the effect of ramiprilat (50 μg/kg) with and without the kinin antagonist Hoe 140 (1 μg/kg IV) on the blood pressure response to bradykinin (25, 50, and 100 ng). Ramiprilat potentiated the blood pressure response to exogenous bradykinin (P<.01) (Fig 1b⇓); this potentiation was attenuated 15 minutes after Hoe 140 treatment but did not attain statistical significance (Fig 1b⇓) and was not present after 2 hours (Fig 1c⇓). All drugs were given intravenously. The experiments involved closed-chest preparations in which myocardial ischemia was not produced.
Data are shown as mean±SE. For drug action and hemodynamic studies, univariate repeated-measures ANOVA with the Greenhouse-Geisser sphericity correction was used for comparison of the blood pressure response to drugs across the three time points. For myocardial infarct size and reperfusion arrhythmias, Student’s one-sided two-sample t test was used for assessment of differences between groups. Welch’s correction was used when intergroup variances were not equal. The overall α level was .05 for each variable of interest. Because two groups were compared with the controls and three groups were compared with the ramiprilat group, the significance criterion for a given test was set at values of P=.025 for two groups or P=.017 for three (Bonferroni’s correction).
Of the 98 rats studied, 5 died of ventricular fibrillation or cardiac shock during coronary occlusion; they were not included in the study because they were neither treated nor reperfused. Three others were rejected because reperfusion was not observed. All data from these 8 rats were excluded. The remaining 82 rats were divided into nine groups as follows: (1) controls (n=11), (2) ramiprilat (n=14), (3) Hoe 140 (n=8), (4) Hoe 140 plus ramiprilat (n=11), (5) losartan (n=8), (6) L-NAME (n=7), (7) L-NAME plus ramiprilat (n=8), (8) indomethacin (n=7), and (9) indomethacin plus ramiprilat (n=8).
Blood pressure was decreased during occlusion and remained reduced during the reperfusion period in all groups compared with baseline (before coronary occlusion). There were no significant differences between saline and drug groups, except for the rats given L-NAME or L-NAME plus ramiprilat, in which blood pressure was significantly higher during the occlusion and reperfusion periods (P<.01) (Table 1⇓). In all groups heart rate remained essentially unchanged throughout the experiment (Table 2⇓).
Myocardial Infarct Size in Ischemia With Reperfusion
Fig 2⇓ shows the effect of ramiprilat and losartan on myocardial infarct size. In vehicle-treated rats infarct size/risk area was 79±3% and infarct size/LV was 41±2%. Ramiprilat significantly reduced infarct size/risk area to 49±4% and infarct size/LV to 22±2% (P<.001). Losartan did not reduce either parameter (74±6% and 37±4%, respectively). The protective effect of ramiprilat was blocked by Hoe 140 (Fig 3⇓). In the Hoe 140 plus ramiprilat group, infarct size/risk area was 77±3% and infarct size/LV was 39±2%, significantly larger than in the ramiprilat group (P<.001) but no different from values in controls.
The protective effect of ramiprilat was also blocked by indomethacin or L-NAME (Fig 4⇓). In the indomethacin plus ramiprilat group, infarct size/risk area was 78±3% (P<.001 versus ramiprilat alone) and infarct size/LV was 41±4% (P<.001 versus ramiprilat alone). In the L-NAME plus ramiprilat group, infarct size/risk area was 81±3% (P<.001 versus ramiprilat alone) and infarct size/LV was 41±2% (P<.001 versus ramiprilat alone). Hoe 140, indomethacin, or L-NAME alone had no effect (Fig 5⇓). Risk area/LV was identical in all experimental groups compared with controls (Fig 6⇓).
Myocardial infarct size in rats with 30 minutes of ligation without reperfusion was 41±3% of the area at risk, significantly smaller than in rats with 30 minutes of ligation and 120 minutes of reperfusion (79±3%, P<.01). Infarct size in rats with 150 minutes of ligation without reperfusion was 91±2% of the area at risk, significantly larger than in rats with reperfusion (P<.05) (Fig 7⇓). There were no significant differences in risk area of the LV in these three groups (44±2%, 49±2%, and 44±2%), indicating that reperfusion caused myocardial damage but in the meantime salvaged some of the ischemic myocardium.
Fig 8⇓ shows the effect of ramiprilat and losartan on the mean duration of sustained ventricular tachycardia and the incidence of ventricular premature beats. In saline controls the average duration of ventricular tachycardia was 18.63±1.25 seconds, and the total number of ventricular premature beats was 148±10. Ramiprilat significantly reduced the duration to 10.3±2.7 seconds (P<.025) and the number of ventricular premature beats to 87±19 (P<.05). Losartan increased the duration of ventricular tachycardia to 26.25±3.02 seconds and number of ventricular premature beats to 223±20 (P<.01). The effect of ramiprilat on arrhythmias was attenuated by pretreatment with Hoe 140 (22.0±4.61 seconds, P<.05; and 190±36, P<.025 versus ramiprilat alone), indomethacin (28.25±4.38 seconds, P<.01; and 191±24, P<.05), and L-NAME (19.13±2.79 seconds, P<.05 versus ramiprilat alone; and 204±49, P=.06) (Fig 9⇓).
Our studies showed that the ACE inhibitor ramiprilat reduces myocardial infarct size and reperfusion arrhythmias after acute coronary occlusion/reperfusion in vivo in rats and that these cardioprotective effects are blocked by the kinin receptor antagonist Hoe 140. Losartan had no effect on infarct size but tended to increase reperfusion arrhythmias. The effects of ramiprilat were also blocked by the cyclooxygenase inhibitor indomethacin and the NO synthase inhibitor L-NAME. Taken together, these results suggest that the cardioprotective effect of ACE inhibitors is mediated by kinins via stimulation of prostaglandin and NO release. These observations support the involvement of paracrine/autocrine hormones such as kinins, prostaglandins, and NO in the cardioprotective effect of ACE inhibitors.12 13 14
Recent studies have explored the question of whether reduction of infarct size is independent of Ang II synthesis inhibition. Noda et al15 reported that in anesthetized and bilaterally nephrectomized dogs, the ACE inhibitor captopril reduced infarct size after 90 minutes of occlusion of the left anterior descending coronary artery. They linked reduction of ischemic damage to increased kinin levels in the local anterior interventricular vein. Hartman et al9 10 found that in rabbits with coronary occlusion/reperfusion the cardioprotective effect of ramiprilat was abolished by Hoe 140, whereas the Ang II type 1 receptor antagonist losartan had no protective effect. Although the precise mechanisms by which ACE inhibitors protect the myocardium from ischemic injury are not fully understood, the findings that ramiprilat reduced infarct size and Hoe 140 reversed this effect indicate that the cardioprotective effect of ramiprilat is mediated by kinins. Kinins may be generated by kallikrein originating from the arterial wall and/or myocardium. We have found that kallikrein mRNA is present in vascular tissue and myocardium and that kallikrein is synthesized and released from these regions. In addition, kininogen, the kallikrein substrate, has been found in both vascular tissue and myocardium.16 17 Thus kinins may be generated locally by the tissue kallikrein-kinin system. Another possibility is that ischemia induces endothelial damage. The damaged vascular tissue may activate plasma prekallikrein, which in turn generates kinins from circulating high molecular weight kininogen or tissue-bound kininogen.13 (Our study did not permit us to differentiate whether kinins were generated by plasma or tissue kallikrein.) Kinins are known to be potent stimulators of the release of prostaglandins and NO from the endothelium.18 19 These autacoids may protect the myocardium via their effects on vasodilation, which in turn causes myocardial blood flow to increase; however, this is unlikely, because the rat coronary collateral circulation is not very effective. Furthermore, in our study the protective effect of the ACE inhibitor occurred during reperfusion, and our microsphere experiments suggest that during reperfusion blood flow to the ischemic region is significantly increased. Another possibility is that NO exerts an inhibitory effect on polymorphonuclear neutrophils, which are known to play an important role in myocardial ischemia/reperfusion injury, platelet aggregation, or both.3 20
It has been shown that administration of an NO donor or l-arginine (a substrate of NO synthase)21 22 or prostaglandin I2 and its analogues23 24 is effective in protecting the myocardium against ischemia/reperfusion injury. In in vitro experiments with isolated rat aortas, vascular prostaglandin I2 production was stimulated by ACE inhibitors and could be blocked completely by a kinin antagonist.11 25 However, we know of no in vivo data indicating that potentiation of endogenous prostaglandins and/or NO is involved in the cardioprotective effect of ACE inhibitors. Our in vivo study showed (we believe for the first time) that inhibition of endogenous prostaglandins and NO with indomethacin or L-NAME can block the beneficial effect of an ACE inhibitor, indicating that the cardioprotective effects of ACE inhibitors are mediated by prostaglandins and NO. Since formation of oxygen-derived free radicals such as superoxide is increased during myocardial reperfusion,26 it is likely that kinin- and prostaglandin I2–stimulated NO release would potentiate scavenging of superoxide, which might contribute to the myocardial protective effects of ACE inhibitors.26 27 28 29
Myocardial infarct size at 150 minutes of ischemia without reperfusion was larger than in ischemia with reperfusion, indicating that reperfusion salvaged some of the ischemic myocardium. However, infarct size in rats with 30 minutes of ligation but no reperfusion was significantly smaller than in those with reperfusion, indicating that reperfusion also injured the myocardium. The final infarct size may depend on a balance between beneficial and injurious effects of reperfusion. Drug intervention was carried out 5 minutes before reperfusion, suggesting that the ACE inhibitor protected the myocardium against reperfusion injury.
In addition to these beneficial effects on myocardial infarction, ACE inhibitors have also been found to exert an antiarrhythmic action on the isolated rat heart.29 30 31 Several investigators have demonstrated significant reduction of reperfusion arrhythmias.32 33 Our in vivo data showed that ramiprilat reduced the mean duration of ventricular tachycardia and the total number of ventricular premature beats during reperfusion and that this effect was blocked by Hoe 140, L-NAME, or indomethacin (although the tendency of Hoe 140 and L-NAME to reverse the action of ramiprilat was marginal by Bonferroni’s correction). These data suggest that kinins, prostaglandins, and NO also play an important role in the antiarrhythmic effect of ACE inhibitors. Although the mechanism is not clear, it may be related to reduction of infarct size. Also, kinins are known to stimulate glucose uptake by myocytes, which may reduce potassium loss and maintain the normal action potential duration34 as well as stimulate the release of prostacyclin and NO, thereby lessening the release of catecholamines. Our results also showed that this antiarrhythmic effect was not related to the formation of Ang II, because the mean duration of ventricular tachycardia and the total number of ventricular premature beats were even higher in the losartan-treated group. We have no immediate explanation for this.
In our study L-NAME by itself did not increase infarct size. This may be related to the fact that although this inhibitor blocks NO synthesis, it also blocks O2− formation.35 It could be that normally there is a balance between the protective effect of NO and the harmful effect of the free radical. The dose of Hoe 140 we used was the same as that used by Hartman et al9 10 in their rabbit model. Blockade of the blood pressure response to exogenous bradykinin at 15 minutes was small and was not present 2 hours after Hoe 140 treatment. It is possible that Hoe 140 blocks kinin receptors at the tissue level and endogenous kinins are released very slowly. This situation may be completely different from the bolus administration of exogenous bradykinin, in which kinin concentrations in the circulation suddenly increase; nevertheless, this finding was unexpected.
In summary, we have demonstrated that the ACE inhibitor ramiprilat is able to reduce myocardial infarct size and attenuate reperfusion arrhythmias in vivo in a rat ischemia/reperfusion model. These effects can be reversed by the kinin antagonist Hoe 140, the cyclooxygenase inhibitor indomethacin, or the NO synthase inhibitor L-NAME. This effect of ramiprilat is not related to an Ang II–dependent mechanism. We speculate that ACE inhibitors block kinin degradation and kinins stimulate synthesis of prostaglandins and NO, which act synergistically to protect the myocardium during ischemia.
Selected Abbreviations and Acronyms
|Ang II||=||angiotensin II|
|L-NAME||=||NG-nitro-l-arginine methyl ester|
This work was supported by National Institutes of Health grant HL-28982-12 and by a grant from Fujimoto Pharmaceutical Corp. We are grateful to Upjohn Co for donating ramiprilat, Hoechst for donating Hoe 140, and DuPont-Merck for providing losartan.
- Received February 7, 1995.
- Revision received March 16, 1995.
- Accepted September 27, 1995.
Parratt JR. Cardioprotection by angiotensin converting enzyme inhibitors: the experimental evidence. Cardiovasc Res. 1994;28:183-189.
Pfeffer MA, Braunwald E, Moyé LA, Basta L, Brown EJ Jr, Cuddy TE, Davis BR, Geltman EM, Goldman S, Flaker GC, Klein M, Lamas GA, Packer M, Rouleau J, Rouleau JL, Rutherford J, Wertheimer JH, Hawkins CM, on behalf of the SAVE Investigators. Effect of captopril on mortality and morbidity in patients with left ventricular dysfunction after myocardial infarction: results of the survival and ventricular enlargement trial. N Engl J Med. 1992;327:669-677.
Ertl G, Kloner RA, Alexander RW, Braunwald E. Limitation of experimental infarct size by an angiotensin-converting enzyme inhibitor. Circulation. 1982;65:40-48.
Linz W, Schölkens BA, Han Y-F. Beneficial effects of the converting enzyme inhibitor, ramipril, in ischemic rat hearts. J Cardiovasc Pharmacol. 1986;8(suppl 10):S91-S99.
Carretero OA, Scicli AG. Local hormonal factors (intracrine, autocrine and paracrine) in hypertension. Hypertension. 1991;18(suppl I):I-58-I-69.
Scicli AG. Increases in cardiac kinins as a new mechanism to protect the heart. Hypertension. 1994;23:419-421. Editorial comment.
van Gilst WH, Tio RA, van Wijngaarden J, de Graeff PA, Wesseling H. Effects of converting enzyme inhibitors on coronary flow and myocardial ischemia. J Cardiovasc Pharmacol. 1992;19(suppl 5):S134-S139.
Noda K, Sasaguri M, Ideishi M, Ikeda M, Arakawa K. Role of locally formed angiotensin II and bradykinin in the reduction of myocardial infarct size in dogs. Cardiovasc Res. 1993;27:334-340.
Nolly H, Carretero OA, Scicli AG. Kallikrein release by vascular tissue. Am J Physiol. 1993;265:H1209-H1214.
van Wijngaarden J, Tio RA, van Gilst WH, de Graeff PA, de Langen CD, Wesseling H. Basic pharmacology of ACE-inhibitors with respect to ischaemic heart disease: prostaglandins and bradykinin. Eur Heart J. 1990;11(suppl B):84-93.
Wiemer G, Schölkens BA, Becker RHA, Busse R. Ramiprilat enhances endothelial autacoid formation by inhibiting breakdown of endothelium-derived bradykinin. Hypertension. 1991;18:558-563.
Entman ML, Smith CW. Postreperfusion inflammation: a model for reaction to injury in cardiovascular disease. Cardiovasc Res. 1994;28:1301-1311.
Johnson G III, Tsao PS, Mulloy D, Lefer AM. Cardioprotective effects of acidified sodium nitrite in myocardial ischemia with reperfusion. J Pharmacol Exp Ther. 1994;252:35-41.
Nakanishi K, Vinten-Johansen J, Lefer DJ, Zhao Z, Fowler WC III, McGee DS, Johnston WE. Intracoronary L-arginine during reperfusion improves endothelial function and reduces infarct size. Am J Physiol. 1992;263:H1650-H1658.
Kanayama T, Kimura Y, Mizogami S. A new prostacyclin analog, KP-10614, inhibits platelet-polymorphonuclear leukocyte interaction and limits experimental infarct size in rat heart. J Pharmacol Exp Ther. 1993;266:344-349.
Schrör K. Role of prostaglandins in the cardiovascular effects of bradykinin and angiotensin-converting enzyme inhibitors. J Cardiovasc Pharmacol. 1992;20(suppl 9):S68-S73.
Hoffmann G, Düsing R. ACE-inhibition, kinins, and vascular PGI2 synthesis. Eicosanoids. 1992;5(suppl):S60-S62.
Hori M, Gotoh K, Kitakaze M, Iwai K, Iwakura K, Sato H, Koretsune Y, Inoue M, Kitabatake A, Kamada T. Role of oxygen-derived free radicals in myocardial edema and ischemia in coronary microvascular circulation. Circulation. 1991;84:828-840.
Parratt J. Endogenous myocardial protective (antiarrhythmic) substances. Cardiovasc Res. 1993;27:693-702.
Schölkens BA, Linz W, König W. Effects of the angiotensin converting enzyme inhibitor, ramipril, in isolated ischaemic rat heart are abolished by a bradykinin antagonist. J Hypertens. 1988;6(suppl 4):S25-S28.
deGraeff PA, de Langen CDJ, van Gilst WH, Bel K, Scholtens E, Kingma JH, Wesseling H. Protective effects of captopril against ischemia/reperfusion-induced ventricular arrhythmias in vitro and in vivo. Am J Med. 1988;84(suppl 3A):67-74.
Bernier M, Hearse DJ. Reperfusion-induced arrhythmias: mechanisms of protection by glucose and mannitol. Am J Physiol. 1988;254:H862-H870.
Pou S, Pou WS, Bredt DS, Snyder SH, Rosen GM. Generation of superoxide by purified brain nitric oxide synthase. J Biol Chem. 1992;267:24173-24176.