Beneficial Effects of Inhibition of Angiotensin-Converting Enzyme on Ischemic Myocardium During Coronary Hypoperfusion in Dogs
Background Angiotensin-converting enzyme (ACE) produces angiotensin II, causing vasoconstriction of coronary arteries and reduction of coronary blood flow. The present study was undertaken to test the hypothesis that an ACE inhibitor increases coronary blood flow and improves myocardial metabolic and contractile functions of ischemic myocardium.
Methods and Results In 65 open-chest dogs, the left anterior descending coronary artery was perfused through an extracorporeal bypass tube from the left carotid artery. When cilazaprilat (3 μg/kg per minute) was infused into the bypass tube for 10 minutes after reduction of coronary blood flow due to partial occlusion of the bypass tube, coronary blood flow increased from 30±1 to 43±2 mL/100 g per minute despite there being no changes in coronary perfusion pressure (43±1 mm Hg). The ratio of myocardial endocardial flow to epicardial flow increased during an infusion of cilazaprilat. Both fractional shortening and lactate extraction ratio of the perfused area were increased (fractional shortening: 4.1±0.6% to 8.9±0.6%, P<.001; lactate extraction ratio: −55.7±3.3% to −36.7±3.9%, P<.001). During an infusion of cilazaprilat, the bradykinin concentration of coronary venous blood was markedly increased. The increased coronary blood flow due to cilazaprilat was attenuated by HOE-140 (an inhibitor of bradykinin receptors; coronary blood flow: 35±2 mL/100 g per minute), and by Nω-nitro-l-arginine methyl ester (an inhibitor of nitric oxide synthase; coronary blood flow: 34±2 mL/100 g per minute). Intracoronary administration of bradykinin mimicked the beneficial effects of cilazaprilat. Cyclic GMP content of the coronary artery was increased by cilazaprilat compared with the untreated condition in the ischemic myocardium. In the denervated hearts, the increased coronary blood flow due to cilazaprilat was not attenuated. On the other hand, CV11974, an inhibitor of angiotensin II receptors, slightly increased coronary blood flow to 34±2 from 30±1 mL/100 g per minute.
Conclusions We conclude that an inhibitor of ACE can increase coronary blood flow and ameliorate myocardial ischemia, primarily due to accumulation of bradykinin and production of nitric oxide from the ischemic myocardium. Inhibition of angiotensin II production due to inhibition of ACE partially contributes to coronary vasodilation in the ischemic myocardium.
Angiotensin-converting enzyme (ACE) is known to produce angiotensin II,1 which may cause potent coronary vasoconstriction. In ischemic hearts, ACE inhibitors may increase coronary blood flow (CBF) and attenuate the extent of myocardial ischemia through the inhibition of receptors of angiotensin II.2 Furthermore, ACE inhibitors are reported to inhibit the degradation of bradykinin by inhibiting kininase II.3 4 Because bradykinin mediates the generation of nitric oxide (NO) through B2 receptor activation5 6 7 8 and bradykinin directly causes coronary vasorelaxation via B1 receptor activation,9 ACE inhibitors may further increase CBF.10 According to one report,11 in the renal artery an ACE inhibitor mediates renal vasodilation through both inhibition of angiotensin II production and activation of bradykinin B2 receptors. However, there is no clear consensus as to whether and how ACE inhibitors modify CBF and affect the myocardial contractile and metabolic functions in ischemic hearts.
Thus, in the present study, to test the effect of an ACE inhibitor on myocardial ischemia, we infused cilazaprilat into the coronary artery during coronary hypoperfusion and measured CBF and regional myocardial contractile and metabolic functions. Furthermore, to examine the possibility that this beneficial effect of cilazaprilat is attributable to increases in NO release through accumulation of bradykinin, we observed the changes in CBF and the regional contractile and metabolic functions during HOE-14012 and Nω-nitro-l-arginine methyl ester (L-NAME) treatments.13 Finally, to investigate the cellular mechanism of this phenomenon, we measured cyclic GMP of the coronary smooth muscles with and without the administration of cilazaprilat during myocardial ischemia.14
Sixty-five mongrel dogs weighing 14 to 23 kg were used. In six dogs, 5 days before the experimental instrumentation, systemic chemical sympathectomy was performed by intravenous injection of 50 mg/kg 6-hydroxydopamine.15 16 17 Deleterious side effects of 6-hydroxydopamine were prevented by intravenous injections of propranolol (1 mg/kg) and phentolamine (1 mg/kg). Three doses of 6-hydroxydopamine (10, 20, and 20 mg/kg) were administered separately over a period of 24 hours.16 17
These dogs were anesthetized with pentobarbital sodium (30 mg/kg IV). The trachea was intubated, and the animal was ventilated with room air mixed with oxygen (1 L/min). The chest was opened through the left fifth intercostal space, and the heart was suspended in a pericardial cradle. The left anterior descending coronary artery (LAD) was cannulated and perfused with blood via the left carotid artery through an extracorporeal bypass tube. Coronary perfusion pressure (CPP) was monitored at the tip of the coronary arterial cannula, and CBF of the perfused area was measured with an electromagnetic flow probe attached at the bypass tube. A small, short collecting tube (1-mm diameter and 7-cm length) was inserted into a small coronary vein near the center of the perfused area to sample coronary venous blood. The drained venous blood was collected in the reservoir placed at the level of the left atrium and was returned to the jugular vein. High-fidelity left ventricular (LV) pressure was measured with a micromanometer (Konigsberg P-5) placed in the LV cavity through the apex. A pair of ultrasonic crystals were placed in the inner third of the myocardium ≈1 cm apart to measure myocardial segment length with an ultrasonic dimension gauge (5 MHz, 2-mm diameter; Schuessler). In five dogs, an electromagnetic flow probe was attached at the root of the ascending aorta to measure cardiac output.
Protocol 1: Effects of Cilazaprilat on Myocardial Ischemia Produced by Coronary Hypoperfusion
Forty-one dogs were used in this protocol. First, in 13 dogs, after hemodynamic stabilization, coronary arterial and venous blood samples were taken for blood gas analysis and determination of lactate, norepinephrine, and bradykinin concentrations and plasma ACE activity. Hemodynamic parameters, ie, LV pressure (LVP), dP/dt, and segment length of the perfused area, were measured. End-diastolic length (EDL) was determined at the R wave of the ECG, and end-systolic length (ESL) was determined at the minimal dP/dt.18 Fractional shortening (FS) was calculated by (EDL−ESL)/EDL as an index of myocardial contractility of the perfused area. In 5 of 13 dogs, cardiac output was measured. With an occluder attached at the extracorporeal bypass tube, CPP was reduced so that CBF decreased to one third of the control CBF. After a low CPP was determined, the occluder was adjusted exactly to keep CPP constant at the low level. All hemodynamic parameters were measured 3, 5, 7, and 10 minutes after the onset of hypoperfusion, and both coronary arterial and venous blood for the metabolic parameters were sampled at 10 minutes. After these measurements, cilazaprilat (3 μg/kg per minute; Nippon Roche KK) was infused into the LAD, and all hemodynamic and metabolic parameters were measured again. Ten minutes later, cilazaprilat infusion was discontinued, and the hemodynamic and metabolic parameters were obtained when hemodynamic parameters were stabilized (n=13). For assessment of the ratio of endocardial flow to epicardial flow (endo/epi flow ratio), microspheres were injected before and 10 (ischemia without cilazaprilat), 20 (ischemia with cilazaprilat), and 30 minutes (ischemia after withdrawal of cilazaprilat) after the onset of coronary hypoperfusion. In the preliminary study, we tested four dosages of cilazaprilat (0.33, 1, 3, and 9 μg/kg per minute IC) during coronary hypoperfusion (n=3 in each dosage in 3 dogs). In the groups receiving cilazaprilat 0.33 to 3 μg/kg per minute, CBF increased to 31±1, 37±2, and 42±2 from 29±2 mL/100 g per minute during coronary hypoperfusion, and 9 μg/kg per minute cilazaprilat increased CBF to 41±2 mL/100 g per minute. Therefore, we decided to perform the experiments using 3 μg/kg per minute cilazaprilat.
Second, to test that the effect of cilazaprilat is related to activation of the B2 receptors of bradykinin (n=6), we infused cilazaprilat into the LAD during hypoperfusion during an intracoronary infusion of HOE-140 (0.5 ng/kg per minute), a selective antagonist of bradykinin B2 receptors. The administration of HOE-140 was initiated 10 minutes before the coronary hypoperfusion.
Third, to test the direct effect of bradykinin on ischemic myocardium, we infused bradykinin instead of cilazaprilat into the LAD during hypoperfusion. During coronary hypoperfusion, all hemodynamic parameters were measured 3, 5, 7, and 10 minutes after the onset of hypoperfusion, and both coronary arterial and venous blood for the metabolic parameters were sampled at 10 minutes. After these measurements, bradykinin (20 ng/kg per minute) was infused into LAD, and all hemodynamic and metabolic parameters were measured 10 minutes later. After the measurements, bradykinin infusion was discontinued, and the hemodynamic and metabolic parameters were obtained when hemodynamic parameters were stabilized (n=5).
Fourth, since bradykinin is reported to increase the release of NO, we tested the idea that cilazaprilat increases NO and thus CBF via the activation of bradykinin receptors (n=8). Cilazaprilat was infused into the LAD during hypoperfusion during intracoronary infusion of L-NAME (3 μg/kg per minute), an inhibitor of NO synthase, to inhibit release of NO. Administration of L-NAME was initiated 10 minutes before coronary hypoperfusion.
Fifth, in six other denervated dogs, the identical procedures and measurements of all variables were performed in experiments with cilazaprilat. We confirmed that the norepinephrine concentrations in the myocardium in systemically denervated (n=5) and innervated control (n=5) dogs were 21±6 and 398±21 pg/mg (P<.001), respectively.
Protocol 2: Effects of an Antagonist of Angiotensin II Receptors on Myocardial Ischemia Produced by Coronary Hypoperfusion
Nine dogs were used in this protocol. After hemodynamic stabilization, coronary arterial and venous blood were sampled for blood gas analysis and determination of lactate and norepinephrine concentrations. Hemodynamic functions were measured as in protocol 1. With an occluder attached at the extracorporeal bypass tube, CPP was reduced so that CBF decreased to one third of the control CBF. After a low CPP was determined, the occluder was adjusted exactly to keep CPP constant at the low level. All hemodynamic parameters were measured 3, 5, 7, and 10 minutes after the onset of hypoperfusion. After these measurements, CV11974 (10 μg/kg per minute; Takeda Pharmaceutical Co) was infused into the LAD, and all hemodynamic parameters were measured again (n=6). Ten minutes later, CV11974 infusion was discontinued, and hemodynamic parameters were obtained when they were stabilized. In the preliminary study, we tested the effects of 3.3, 10, and 33 μg/kg per minute of CV11974 on CBF during coronary hypoperfusion. Three dosages of CV11974 increased CBF to 31±1, 35±1, and 34±2 from 29±1 mL/100 g per minute (n=3 for each dosage) with constant CPP (CPP=42±1 mm Hg). Thus, we decided to use 10 μg/kg per minute of CV11974 in this protocol.
Protocol 3: Effects of Cilazaprilat on Coronary Circulation in the Nonischemic Myocardium
To examine the effects of cilazaprilat on CBF in the normoxic condition, we infused three dosages of cilazaprilat (1, 3, and 9 μg/kg per minute) into five dogs. After the hemodynamic stabilization, CBF was measured before and during infusion of each dose of cilazaprilat. Coronary arterial and venous blood were sampled for the measurement of bradykinin concentration.
Protocol 4: Effects of Cilazaprilat on Cyclic GMP Content of Epicardial Coronary Artery in Ischemic Hearts
We tested whether cilazaprilat increases cyclic GMP content of the coronary artery in the ischemic myocardium. With an occluder attached at the extracorporeal bypass tube, CPP was reduced so that CBF decreased to one third of the control CBF. After a low CPP was determined, the occluder was adjusted to keep CPP constant at the low level. After the low CPP was maintained for 10 minutes, cilazaprilat (3 μg/kg per minute) was infused into the LAD for 10 minutes, and we rapidly removed the epicardial LAD (ischemic region) and left circumflex coronary artery (nonischemic control region) (n=5) with the use of precooled stainless steel scissors and tongs. We rapidly stored samples in liquid nitrogen. In five other dogs, CPP was reduced so that CBF decreased to one third of the control CBF for 20 minutes, and the epicardial LAD (ischemic region) and left circumflex coronary artery (nonischemic control region) were rapidly removed and stored in liquid nitrogen.
Measurements of Regional Coronary Blood Flow
Regional myocardial blood flow was determined by the microsphere technique as previously reported19 with the use of nonradioactive microspheres (Sekisui Plastic Co) made of inert plastic labeled with different types of stable heavy elements as described in detail previously.16 In the present study, microspheres labeled with Nb, Br, Zr, and I were used. The mean diameter was 15 μm, and specific gravity was 1.32 for Nb, 1.34 for Br, 1.36 for Zr, and 1.60 for I. Microspheres were suspended in isotonic saline with 0.01% Tween 80 to prevent aggregation. The microspheres were ultrasonicated for 5 minutes followed by 5 minutes of voltexing immediately before injection. Approximately 1 mL of the microsphere suspension (2 to 4×105 spheres) was injected into the left atrium followed by several warm (37°C) saline flushes (5 mL).
The x-ray fluorescence activity of stable heavy elements was measured by a wavelength dispersive spectrometer (PW 1480, Phillips Co). Specification of this x-ray fluorescence spectrometer has been described previously. In brief, when the microspheres are irradiated by the primary x-ray beam, the electrons fall back to a lower orbit and emit measurable energy. This energy level of the x-ray fluorescence depends on the characteristics for each element. Therefore, it was possible to qualify the x-ray fluorescence of several differently labeled microspheres in the mixture. The endo/epi flow ratio was calculated as the ratio of each tissue content normalized with the wet weight of the sampled myocardium.
MV̇o2 (mL/100 g per minute) was calculated as: CBF (mL/100 g per minute)×coronary arterial and venous blood oxygen difference (mL/dL). Lactate was assessed by the enzymatic assay, and lactate extraction ratio (LER) was obtained by multiplying the coronary arteriovenous difference in lactate concentration by 100 and dividing by arterial lactate concentration.
The method of norepinephrine measurement has been described previously.20 Five milliliters of coronary arterial and venous blood taken into a tube containing EDTA was immediately placed in ice water and centrifuged for 20 minutes. The plasma was kept at −80°C. Within 2 weeks, plasma norepinephrine was adsorbed on alumina and separated by high- performance liquid chromatography (HPLC) (pump, LC-3A; column Zpax-SCX; Shimazu Seisakusho Co). Plasma norepinephrine was determined spectrofluorometrically by the trihydoxyindole (THI) method (spectrofluorophotometer RF-500LCA, Shimazu). In this system, sensitivity of the assay is 10 pg/mL plasma and the intra-assay coefficient of variation is 6.8%.20
The method of bradykinin measurement has been described previously.21 One milliliter of blood withdrawn from the sampling tube was squirted rapidly into siliconized polyethylene tubes containing 4 mL of 96% ethanol, which was centrifuged at 2500g at 4°C for 15 minutes. The supernatant was decanted into a siliconized 250-mL round-bottomed flask. The precipitate was resuspended in 20 mL 75% ethanol and recentrifuged. This supernatant was combined with the previous supernatant. After 0.5 mL of octanol was added to prevent frothing, the ethanol was removed, and the volume was reduced to ≈2 mL through evaporation at 60°C under the reduced pressure. The residual solution was acidified with 5 mL of 0.01 mol/L HCl and partitioned twice with 20 mL of diethyl ether. This procedure was performed in the same flask as the original evaporation, with the ether supernatant being removed by suction after each partitioning. The aqueous phase remaining in the flask after the ether extractions was subsequently reduced to dryness using a rotary evaporator. These dried samples were stored at −80°C before assay. The dried samples were redissolved in 2.5 mL of 0.1 mol/L Tris-HCl buffer containing 0.2% gelatin, 0.1% neomycin, and 0.01 EDTA, adjusted to pH 7.4. The incubation mixture for radioimmunoassay consisted of 0.1 mL of 0.01 mol/L 1,10-phenanthroline HCl, diluent buffer of 0.5 mL containing the unknown or standard bradykinin, 0.1 mL antiserum diluted 1:600 with diluent buffer, and 0.1 mL (125I-Try8)-bradykinin (approximately 8000 cpm) dissolved in normal saline. It was incubated in a polyethylene tube at 4°C for 24 hours, and dextran-coated charcoal was used to separate the free labeled antigen from that bound to antibody. Three replicate tubes containing only buffer, phenanthroline, and (125I-Try8)-bradykinin were incubated and treated with coated charcoal to determine the amount of labeled antigen that remained in the supernatant in the absence of antibody. The mean value of this measurement was subtracted from supernatant radioactivity after centrifugation of the antibody-containing tubes, and the resultant value was used to calculate the proportion of label bound to antibody.
Cyclic GMP Measurement
The method of cyclic GMP measurement in tissue has been previously described.22 After removal of the adventitial connective tissues in the coronary arteries (20 to 40 mg), the frozen tissue was powdered, homogenized at 4°C in 1 mL of ice-cold 6% trichloroacetic acid, and centrifuged at 2500g for 20 minutes. The supernatant fluid was removed, extracted three times with 3 mL of diethyl ether saturated with water, and stored in the freezer (−80°C). Cyclic GMP concentration in the supernatant fluid was measured by the radioimmunoassay method22 within 7 days. Briefly, we used 100 μL of dioxane-triethylamine mixture containing succinic acid anhydride succinylated cyclic GMP in the supernatant (100 μL). After a 10-minute incubation, the reaction mixture was added to 800 μL of 0.3 mol/L imidazole buffer (pH 6.5). One hundred microliters of succinyl cyclic GMP tyrosine methyl ester iodinated with 125I (15 000 to 20 000 cpm in <10−14 mol/L) was added to the assay mixture containing 100 μL of the supernatant and 100 μL of diluted antisera in the presence of chloramine T22 ; the mixture was kept at 4°C for 24 hours. A cold solution of dextran-coated charcoal (500 μL) was added to the mixture in an ice-cold water bath. The charcoal was spun down, and 0.5 mL of the supernatant was counted for radioactivity in a gamma spectrometer. The amount of cyclic GMP was normalized by protein content of coronary artery assayed by the Lowry method.23
Measurement of Plasma ACE Activity
The method of measurement of the plasma ACE activity has been described previously.24 Three milliliters of coronary arterial and venous blood taken into a polyethylene tube was immediately placed in ice water and centrifuged for 20 minutes. The plasma was kept at −80°C. Within 7 days, plasma ACE activity was assessed by measuring the production of hippuric acid from the substrate (Hip-His-Leu). The concentration of hippuric acid was measured by colorimetric methods24 using HPLC with a Model Tri-Rotor (Japan Spectroscopic Co).
Table 1⇓ depicts systolic and diastolic blood pressures and heart rate for all groups. Before and during coronary hypoperfusion with and without pharmacological interventions did not affect these systemic hemodynamic parameters. In the denervated dogs, although systolic and diastolic blood pressures were not different, baseline heart rate was reduced compared with the cilazaprilat group.
Effects of Inhibition of ACE on Myocardial Ischemia
Table 2⇓ shows the changes in CPP, CBF, and the bradykinin concentration of coronary arterial and venous blood during infusions of cilazaprilat (0.33, 1, 3, and 9 μg/kg per minute) in the nonischemic condition. Intracoronary infusions of 0.33, 1, and 3 μg/kg per minute of cilazaprilat did not change CBF but slightly increased the bradykinin concentration of coronary venous blood at 3 μg/kg per minute of cilazaprilat. An intracoronary infusion of 9 μg/kg per minute of cilazaprilat slightly increased CBF with a further increment of the bradykinin concentration.
Table 3⇓ shows the coronary hemodynamic and metabolic parameters before the reduction of CPP. There were no significant differences in the baseline coronary hemodynamic and metabolic parameters of the nonischemic condition. Neither L-NAME nor HOE-140 altered the coronary hemodynamic and metabolic parameters. In the denervated myocardium, MV̇o2 in the baseline condition was significantly reduced compared with the baseline condition in the cilazaprilat group. Fig 1⇓ shows the effect of cilazaprilat on CBF and FS during coronary hypoperfusion. After an abrupt reduction of CPP, CBF decreased from 91±2 to 30±1 mL/100 g per minute, and FS decreased within 1 minute keeping a steady state for 10 minutes (5.1±0.6%). Within 1 minute, once CPP was set, CPP was kept constant at 43±1 mm Hg. After the initiation of an intracoronary infusion of cilazaprilat, both CBF and FS gradually increased. Ten minutes after the onset of the cilazaprilat infusion, both CBF and FS increased to 43±2 mL/100 g per minute and 8.9±0.6%, respectively, indicating that an intracoronary infusion of cilazaprilat increases both CBF and FS of the ischemic myocardium. Table 4⇓ depicts cardiac output before and during coronary hypoperfusion. Cardiac output was reduced due to coronary hypoperfusion, and cilazaprilat administration increased cardiac output during coronary hypoperfusion. Systemic vascular resistance (SVR) was decreased due to coronary hypoperfusion; however, SVR did not change regardless of whether cilazaprilat was administered during coronary hypoperfusion. Fig 2⇓ shows the endo/epi flow ratio during coronary hypoperfusion with and without administration of cilazaprilat. Before the reduction in CPP, endo/epi flow ratio was 1.17±0.07. The endo/epi flow ratio decreased to 0.74±0.03, and administration of cilazaprilat significantly increased endo/epi flow ratio to 0.81±0.01. This result indicates that cilazaprilat increases CBF in the endocardium more than the epicardium. This beneficial effect of cilazaprilat was demonstrated from the aspect of myocardial metabolism. Fig 3⇓ shows the changes in LER, MV̇o2, and coronary arterial and venous concentrations of norepinephrine during ischemia with and without cilazaprilat. Cilazaprilat increased LER, indicating that myocardial anaerobic metabolism was improved by the cilazaprilat infusion, and thus MV̇o2 increased. There were no significant differences in norepinephrine concentrations in the coronary venous blood. Norepinephrine release (coronary arteriovenous differences in norepinephrine multiplied by CBF [ng/100 g per minute]) in the control condition and in the ischemic myocardium before, during, and after administration of cilazaprilat were 2.44±0.83, 1.12±0.75, 0.71±1.45, and 1.07±1.03 ng/100 g per minute, respectively. There were no significant differences in norepinephrine release among these four conditions. pH in coronary venous blood was 7.38±0.01 at baseline and decreased (P<.001) to 7.22±0.02 during coronary hypoperfusion. However, the cilazaprilat infusion improved (P<.005) pH in coronary venous blood (7.31±0.01). pH of coronary arterial blood did not change throughout this study (7.40±0.02). There were no significant differences in plasma ACE activity in the coronary arterial blood in the control condition and in the ischemic myocardium before, during, and after administration of cilazaprilat (5.6±0.9, 5.8±0.8, 5.6±0.9, and 5.4±0.8 IU/L, respectively). Plasma ACE activities in the coronary venous blood in the control and ischemic myocardium were 5.5±0.9 and 5.9±0.5 IU/L, respectively. Cilazaprilat administration markedly reduced plasma ACE activity in the coronary venous blood to 0.2±0.1 IU/L during coronary hypoperfusion (P<.001), and plasma ACE activity returned to 5.0±1.0 IU/L 10 minutes after withdrawal of cilazaprilat infusion.
Mechanisms by Which Cilazaprilat Increases CBF and Improves Myocardial Ischemia
Fig 4⇓ shows the coronary arteriovenous differences in the concentration of bradykinin during coronary hypoperfusion with and without cilazaprilat administration. In the nonischemic condition, the coronary arteriovenous difference in the bradykinin concentration was −2.1±2.5 pg/mL. In the ischemic condition, the coronary arteriovenous difference in the bradykinin concentration increased, and administration of cilazaprilat further increased the coronary arteriovenous difference in the bradykinin concentration. Fig 5⇓ shows the effect of bradykinin on CBF and FS during coronary hypoperfusion. After an abrupt reduction of CPP associated with CBF, FS decreased within 1 minute, keeping a steady state for 10 minutes. Within 1 minute, once CPP was set, it was kept constant. Ten minutes after the onset of the bradykinin infusion, both CBF and FS increased to 46±4 mL/100 g per minute and 9.7±1.0%, respectively, indicating that an intracoronary infusion of bradykinin can increase both CBF and FS of the ischemic myocardium. Fig 6⇓ shows the changes in LER, MV̇o2, and coronary arterial and venous concentrations of norepinephrine during ischemia with and without bradykinin administration. Bradykinin increased LER, indicating that myocardial anaerobic metabolism was improved by the bradykinin infusion, and thus MV̇o2 increased. There were no significant differences in norepinephrine concentrations in the coronary venous blood. There also were no differences between norepinephrine release in the control condition and in the ischemic myocardium before, during, and after administration of bradykinin (2.10±0.98, 1.40±0.93, 1.73±1.39, and 1.38±0.98 ng/100 g per minute, respectively).
Fig 7⇓ shows that these beneficial effects of cilazaprilat are antagonized by concomitant treatment with HOE-140. Intracoronary infusion of cilazaprilat during administration of HOE-140 slightly increased both CBF and FS during coronary hypoperfusion with the constant low CPP. The increases in CBF (12.8±1.1 versus 2.8±0.9 mL/100 g per minute) and FS (4.8±0.2% versus 0.7±0.2%) were attenuated by 78% and 85% relative to the group of the cilazaprilat administration (Fig 1⇑). Fig 8⇓ shows the changes in LER, MV̇o2, and coronary arterial and venous concentrations of norepinephrine during ischemia with and without cilazaprilat during administration of HOE-140. Cilazaprilat slightly increased LER. There were no significant differences in MV̇o2 and norepinephrine concentrations in the coronary arterial and venous blood. There also were no differences between norepinephrine release in the control condition with and without HOE-140 administration (1.63±1.13 and 1.95±0.69 ng/100 g per minute, respectively) and in the ischemic myocardium before, during, and after administration of cilazaprilat (0.79±0.41, 0.25±0.47, and 1.33±0.25 ng/100 g per minute, respectively). pH in the coronary venous blood was 7.37±0.01 at baseline and decreased to 7.22±0.01 during coronary hypoperfusion. However, the cilazaprilat infusion did not improve pH in coronary venous blood (7.24±0.01). These results indicate that HOE-140 attenuates the beneficial effects of cilazaprilat on myocardial ischemia by ≈80%, indicating that the beneficial effect of cilazaprilat is mainly attributable to the accumulation of bradykinin in the ischemic myocardium.
Because bradykinin is reported to increase NO production in endothelial cells, we tested whether L-NAME attenuates the beneficial effects of cilazaprilat in the ischemic myocardium. Fig 9⇓ shows that L-NAME treatment attenuates the cilazaprilat-induced increases in CBF (12.8±1.4 versus 4.0±0.7 mL/100 g per minute) and FS (4.8±0.2% versus 1.2±0.4%) during the constant low CPP by 69% and 75%, respectively. Fig 10⇓ shows the changes in LER, MV̇o2, and coronary arterial and venous concentrations of norepinephrine during coronary hypoperfusion with and without cilazaprilat during treatment with L-NAME. Cilazaprilat slightly increased LER and MV̇o2. There were no significant differences in norepinephrine concentrations in the coronary arterial and venous blood. There also were no differences between norepinephrine release in the control condition with and without L-NAME administration (3.01±1.99 and 2.97±1.03 ng/100 g per minute, respectively) and in the ischemic myocardium before, during, and after administration of cilazaprilat (1.75±0.62, 2.63±0.69, and 1.55±0.72 ng/100 g per minute, respectively). pH in coronary venous blood was 7.38±0.01 at baseline and decreased to 7.21±0.01 during coronary hypoperfusion. However, the cilazaprilat infusion did not improve the pH in coronary venous blood (7.25±0.02). These results indicate that the beneficial effect of cilazaprilat is mostly attributable to augmentation of NO release due to the bradykinin accumulation. To examine whether NO increases CBF of the ischemic myocardium by its direct coronary vasodilatory action or by attenuation of norepinephrine release, we tested the effects of cilazaprilat in the denervated ischemic myocardium. Fig 11⇓ shows the effect of cilazaprilat on CBF and FS during coronary hypoperfusion in the denervated myocardium. After an abrupt reduction of CPP associated with CBF, FS decreased within 1 minute keeping a steady state for 10 minutes. Ten minutes after the onset of the cilazaprilat infusion, both CBF and FS increased to the levels in the cilazaprilat group in the innervated control myocardium (Fig 1⇑). Fig 12⇓ shows the changes in LER, MV̇o2, and coronary arterial and venous concentrations of norepinephrine during ischemia with and without cilazaprilat in the denervated myocardium. Cilazaprilat increased LER, indicating that myocardial anaerobic metabolism was also improved in the denervated myocardium, and thus MV̇o2 increased. There were no significant differences in norepinephrine concentration in the coronary venous blood. There also were no differences between norepinephrine release in the control condition and in the ischemic myocardium before, during, and after administration of cilazaprilat (1.50±2.14, 0.49±0.33, 1.61±2.63, and 1.41±0.52 ng/100 g per minute, respectively).
The remaining 20% to 30% of the beneficial effects of cilazaprilat may be attributable to the inhibition of angiotensin II receptors. Therefore, we tested whether inhibition of angiotensin II receptors can account for the remaining 30% of the beneficial effect of cilazaprilat. Fig 13⇓ shows that CV11974 increases CBF and FS by 4.0±0.5 mL/100 g per minute and 1.7±0.3%, respectively, despite the constant low CPP. CV11974 increased LER from −52.5±4% to −43.2±2.2% (P<.05) but did not change MV̇o2 (1.6±0.1 to 1.7±0.1 mL/100 g per minute). CV11974 did not alter the concentrations of norepinephrine in the coronary arterial (from 393±21 to 404±20 pg/mL) and venous blood (from 417±11 to 411±16 pg/mL). There also were no differences between norepinephrine release in the control condition and in the ischemic myocardium before, during, and after administration of CV11974 (2.79±2.05, 0.80±1.05, 2.84±1.00, and 0.71±0.83 ng/100 g per minute, respectively).
Table 5⇓ represents the cellular basis of the interaction between cilazaprilat and NO. Without cilazaprilat treatment, myocardial ischemia (CPP: 105±4 to 42±2 mm Hg, CBF: 82±3 to 27±2 mL/100 g per minute) increased cyclic GMP content of the coronary artery from 69±5 to 126±7 fmol/mg protein (P<.01). However, treatment with cilazaprilat during myocardial ischemia further increased (P<.01) cyclic GMP content of the involved coronary artery to 272±16 fmol/mg protein.
In the present study, we showed that an inhibition of ACE causes coronary vasodilation and improves contractile and metabolic functions in the ischemic heart, mainly through accumulation of bradykinin and NO release and partially through inhibition of activation of angiotensin II receptors. However, before reaching a conclusion, we should consider several possibilities of the effect of an inhibitor of ACE.27
Validity of the Experimental Model in the Present Study
The major assumption in all of the experimental protocols in the present study was that intracoronary infusion of chemicals, such as cilazaprilat, L-NAME, HOE-140, and CV11974, does not have any effects on peripheral vessels, and the observed changes in the LAD area were due only to local effects on the coronary vasculature. If pharmacological interventions to the LAD area also affect systemic hemodynamics, the beneficial effects of cilazaprilat may be secondary to the systemic vascular effects such as afterload reduction. However, in the present study, systolic and diastolic blood pressures and heart rate did not change during any pharmacological interventions (Table 1⇑), suggesting that pharmacological interventions to the LAD area minimally affect the systemic hemodynamic parameters in the present study. Furthermore, when intracoronary administration of 3 μg/kg per minute cilazaprilat markedly reduced plasma ACE activity in the coronary venous blood, this administration of cilazaprilat did not affect the plasma ACE activity in the coronary arterial blood. Indeed, intracoronary cilazaprilat administration did not affect SVR (Table 4⇑). These observations strengthen the idea that the beneficial effects of cilazaprilat are attributable to the local coronary vascular and myocardial changes rather than the systemic hemodynamic changes.
Coronary Vasodilation due to Inhibition of ACE in the Ischemic Hearts
ACE inhibitors inhibit the accumulation of angiotensin II and accumulate bradykinin in the myocardium. First, because angiotensin II is reported to promote the release of norepinephrine from the presynaptic vesicles, ACE inhibitors may decrease the release of norepinephrine from the presynaptic vesicles,28 and the subsequent withdrawal of α-adrenoceptor activation in the ischemic myocardium would cause coronary vasodilation. However, in our experiment there is evidence that cilazaprilat during coronary hypoperfusion does not alter the norepinephrine concentration in the coronary venous blood (Fig 2⇑), and the beneficial effects of cilazaprilat were not blunted in the denervated ischemic myocardium (Figs 11⇑ and 12⇑). These observations suggest that withdrawal of sympathetic nerve activity is not likely for the mechanisms of the coronary vasodilation in the ischemic myocardium.
Second, angiotensin II also directly constricts the coronary smooth muscles,2 which may constitute the mechanisms for coronary vasodilation in the ischemic heart. However, CV11974, an inhibitor of angiotensin II receptors, increased CBF only by 20% to 30% of cilazaprilat-induced increases in CBF, suggesting that inhibition of angiotensin II accumulation is not the major factor.
Third, because our results revealed that bradykinin accumulation due to cilazaprilat administration is a major factor for coronary vasodilation and improvement of myocardial contractile and metabolic functions in the ischemic myocardium, the direct coronary vasodilation due to bradykinin may be a primary factor. Bradykinin is reported to relax vascular smooth muscles,9 and our data revealed that bradykinin increases CBF during coronary hypoperfusion and improves ischemic myocardium.
Fourth, bradykinin-induced coronary vasodilation may be involved in increases in NO production5 6 7 8 or accumulation of prostacyclin.29 The present study revealed that L-NAME, an inhibitor of NO synthase, attenuates the cilazaprilat-induced coronary vasodilation to the same extent as HOE-140, indicating that the coronary vasodilation due to bradykinin during administration of cilazaprilat is attributable to NO accumulation. Because NO is known to be a potent coronary vasodilator,30 31 32 33 it is likely that NO-induced coronary vasodilation is a major cause for the cilazaprilat-induced coronary vasodilation in the ischemic myocardium. Furthermore, coronary arterial content of cyclic GMP, which is increased by NO via guanylate cyclase activation, was increased by administration of cilazaprilat during coronary hypoperfusion.
Aside from the NO-induced coronary vasodilation, NO can inhibit platelet aggregation and neutrophil adherence to coronary vasculature.34 35 36 If this is the case, increased CBF may be attributable to the inhibition of progressive platelet aggregation and neutrophil adherence. However, this may be unlikely, because even during administration of L-NAME, both CBF and FS remained constant at low levels during coronary hypoperfusion without progressive decreases, suggesting that progressive platelet aggregation and adherence of neutrophils may not occur in the ischemic hearts even without cilazaprilat in the present experimental protocols. NO accumulation due to an intracoronary infusion of cilazaprilat also may open functional collateral vessels to the ischemic area, which may reduce the severity of ischemia. We cannot negate this possibility; however, increases in CBF measured at the bypass tube suggest that forward flow into the coronary artery is essentially increased due to cilazaprilat. Taken together, cilazaprilat increases CBF in the ischemic heart through NO production via augmentation of accumulation of bradykinin.
Role of Bradykinin for Coronary Vasodilation in the Ischemic Heart
Results of the present study also indicate that endo/epi flow ratio during myocardial ischemia is increased due to administration of cilazaprilat, suggesting that bradykinin accumulation or NO production due to bradykinin is more prominent in the endocardium than the epicardium. In the kidney, the formation of kinins is reported to be H+ dependent,37 suggesting that H+ may increase bradykinin concentration in the tissues. If this is the case in the ischemic heart, endocardium may produce H+ more than epicardial myocardium because epicardial flow is slightly less than the endocardium, although myocardial oxygen consumption in the endocardium is more than in the epicardium. Bradykinin accumulation measured in the coronary venous effluent in the ischemic myocardium was more than twice that in the nonischemic myocardium when cilazaprilat was administered in the coronary artery. This difference in the bradykinin accumulation may be attributable to the differences in H+ accumulation in the myocardium. Furthermore, it has been reported that Ca2+ concentration in the endocardium is higher than that in the epicardium, and ischemia elevates myocardial Ca2+ concentrations.38 The rise in Ca2+ may affect release of bradykinin and activity of NO synthase, which may account for the higher endocardial flow and higher accumulation of bradykinin in the ischemic myocardium. In addition, activation of α1-adrenoceptor activation produces a >40-fold increase in the rate of kallikrein secretion, and the kinin output in the venous effluent from the submandibular gland becomes 700-fold,39 suggesting that α1-adrenoceptor stimulation, which occurs during myocardial ischemia, augments the increases in bradykinin accumulation.
Although our results indicate that coronary arteriovenous differences in bradykinin concentrations are increased (Table 1⇑ and Fig 4⇑), this extent of increases in bradykinin may minimally affect hemodynamic parameters, because it is reported that several nanograms per milliliter of bradykinin are necessary to affect the hemodynamic parameters.40 However, bradykinin is very fragile and degrades rapidly in the blood, suggesting that bradykinin concentration near the vessel wall may be very high to cause substantial vasodilation. During administration of cilazaprilat in the nonischemic myocardium (Table 1⇑), although release of bradykinin was increased, CBF did not increase at 3 μg/kg per minute cilazaprilat. When CPP varies in the range of coronary flow autoregulation, coronary vasodilatory or constrictive response may be overwhelmed by other vasoactive substances. Another possible explanation for the discrepancy between the concentrations of bradykinin and CBF in the nonischemic myocardium may be the shape of their dose-response relation.
It has been reported that ACE inhibitors are effective for attenuation of infarct size after myocardial ischemia and reperfusion.41 42 Martorana et al41 42 reported that ACE inhibitors attenuate infarct size from 55% to 25%. Furthermore, ACE inhibitors are effective for promoting remodeling of ventricle after the acute myocardial infarction, which prevents enlargement of the ventricle.43 The present study helps to clarify the role of bradykinin in the beneficial effects of ACE inhibitors in the canine experimental model of ischemic heart disease, although further basic and clinical research must be performed to investigate the possible use of ACE inhibitor in the treatment of effort angina.
This work was supported by Scientific Research Grant-in-Aid 03670449 from the Ministry of Education, Science, and Culture, Japan. The authors gratefully acknowledge Noriko Tamai, Yoshitomo Edahiro, and Shinya Suzuki for their technical assistance.
- Received January 4, 1995.
- Accepted February 7, 1995.
- Copyright © 1995 by American Heart Association
Victor JD, Richard EP. Renin-angiotensin system. In: Fozzard HA, ed. The Heart and Cardiovascular System. New York, NY: Raven Press; 1992:1817-1849.
Erdos EG. Angiotensin I converting enzyme and changes in our concepts through the years. Hypertension. 1990;16:363-370.
Schini VB, Boulanger C, Regoli D, Vanhoutte PM. Bradykinin stimulates the production of cyclic GMP via activation of B2 kinin receptors in cultured porcine aortic endothelial cells. J Pharmacol Exp Ther. 1990;252:581-585.
Arnold WP, Mittal CK, Katsuki S, Murad F. Nitric oxide activates guanylate cyclase and increased guanosine 3:5′-cyclic monophosphate levels in various tissue preparation. Proc Natl Acad Sci U S A. 1977;74:3203-3207.
Pelc LR, Garrett JG, David CW. Mechanism of coronary vasodilation produced by bradykinin. Circulation. 1991;83:2048-2056.
Creager MA, Halperin JL, Bernard DB, Faxon DP, Melidossina CD, Bavras H, Ryan TJ. Acute regional circulatory and renal hemodynamic effects of converting-enzyme inhibition in patients with congestive heart failure. Circulation. 1981;64:483-489.
McMahon TJ, Hood JS, Kadowitz PJ. Pulmonary vasodilator response to vagal stimulation is blocked by Nω-nitro-l-arginine methyl ester in the cat. Circ Res. 1992;70:364-369.
Johns RA, Peach MJ, Linden J, Tichotsky A. NG-Monomethyl l-arginine inhibits endothelium-derived relaxing factor-stimulated cyclic GMP accumulation in cocultures of endothelial and vascular smooth muscle cells by an action specific to the endothelial cell. Circ Res. 1990;67:979-985.
Gauthier P, Nadeau R, Champlain J. Acute and chronic cardiovascular effects of 6-hydroxydopamine in dogs. Circ Res. 1972;31:207-217.
Acad BA, Joselevitz-Goldman J, Scholz PM, Weiss HR. Improved distribution of regional oxygenation in denervated ischemic dog myocardium. Circ Res. 1988;62:1041-1048.
Kitakaze M, Hori M, Gotoh K, Sato H, Iwakura K, Kitabatake A, Inoue M, Kamada T. Beneficial effects of α2-adrenoceptor activity on ischemic myocardium during coronary hypoperfusion in dogs. Circ Res. 1989;65:1632-1645.
Abel FL. Maximal negative dP/dt as an indicator of end of systole. Am J Physiol. 1981;240:H676-H679.
Mori H, Haruyama S, Shinozaki Y, Okino H, Iida A, Takanashi R, Sakura I, Husseini W, Payne B, Hoffman JIE. New nonradioactive microspheres and more sensitive x-ray fluorescence to measure regional blood flow. Am J Physiol. 1992;263:H1946-H1957.
Sato H, Inoue M, Matsuyama T, Ozaki H, Shimazu T, Takeda H, Ishida Y, Kamada T. Hemodynamic effects of the β1-adrenoceptor partial agonist xamoterol in relation to plasma norepinephrine levels during exercise in patients with left ventricular dysfunction. Circulation. 1987;75:213-220.
Lowry OH, Rosebrough NJ, Farr AL, Randall RJ. Protein measurement with folin phenol reagent. J Biol Chem. 1951;193:265-275.
Kasahara Y, Ashihara Y. Colorimetry of angiotensin-1 converting enzyme activity in serum. Clin Chem. 1981;27:1922-1926.
Winer BJ. Statistical Principles In Experimental Design. New York, NY: McGraw-Hill Inc; 1982:1-907.
Snedecor GW, Cochran WG. Statistical Methods. Ames, IA: Iowa State University Press; 1972:258-298.
James RP. Cardioprotection by angiotensin converting enzyme inhibitors: the experimental evidence. Cardiovasc Res. 1994;28:183-189.
de Jonge A, Thoolen MJMC, Timmermanns PBMWM, van Zweiten PA. Interaction of angiotensin converting enzyme inhibitors with the sympathetic nervous system. Prog Pharmacol. 1984;5:25-38.
van Gilst WH, van Wijngaarden J, Scholtens E, de Graeff PA, de Langen CDJ, Wesseling H. Captopril-induced increase in coronary flow: an SH-dependent effect on arachidonic acid metabolism? J Cardiovasc Pharmacol. 1987;9(suppl 2):S31-S36.
Ignarro LJ, Buga GM, Wood KS, Byrns RE, Chaudhuri G. Endothelium-derived relaxing factor produced and released from artery and vein is nitric oxide. Proc Natl Acad Sci U S A. 1987;84:9265-9269.
Dirk JD, Robert JB. Inhibition of nitric oxide production aggravates myocardial hypoperfusion during exercise in the presence of a coronary artery stenosis. Circ Res. 1994;74:629-640.
Sneddon JM, Vane JR. Endothelium-derived relaxing factor reduces platelet adhesion to bovine endothelial cells. Proc Natl Acad Sci U S A. 1988;85:2800-2804.
Guillermo AS, Mario AD, Oscar AC. Effect of pH and amiloride on the intrarenal formation of kinins. Am J Physiol. 1983;245:F198-F203.
Vincent MF, Rolf B, Michael WW, Barry MM, Albert SC. Endocardial versus epicardial differences of intracellular free calcium under normal and ischemic conditions in perfused rat hearts. Circ Res. 1993;72:1082-1090.
Rabito SF, Orstavik TB, Scicli AG, Schork A, Carretero OA. Role of the autonomic nervous system in the release of rat submandibular gland kallikrein into the circulation. Circ Res. 1983;52:635-641.
Salgado MCO, Rabito SF, Carretero OA. Blood kinin in one-kidney, one clip hypertensive rats. Hypertension. 1986;8(suppl I):I-110-I-113.