INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

THE MAJORITY OF STUDIES OF ischemia and reperfusion on coronary reactivity have focused on the acute phase either during ischemia or immediately after reperfusion (16, 22). Despite its importance from a physiological and clinical standpoint, very few studies have examined the effects of -adrenergic stimulation on coronary vascular responses in the presence of stable stunned myocardium (10, 29). Interestingly, even fewer studies have examined adenylyl cyclase-induced coronary reactivity in the stunned myocardium (29, 30), since most prior studies have focused on -adrenergic or adenylyl cyclase regulation of myocardial function (2, 5, 13, 25, 28), neglecting a potentially important direct role on the coronary vasculature. Because adenylyl cyclase is a key component in the -adrenergic and other signaling pathways, this would be important to understand, particularly since adenylyl cyclase stimulation is currently being considered for inotropic support in patients (21, 24).

The current study was designed to examine the effects of cAMP-mediated coronary vascular responsiveness in stunned myocardium. To address this goal, we utilized a model of stable myocardial stunning in conscious pigs induced by 1.5 h of coronary stenosis (CS) and 12 h of full coronary artery reperfusion (CAR). The in vivo model of low-flow ischemia induced by CS used in the current study has particular relevance to the clinical occurrence of ischemia due to coronary artery disease. The 12-h reperfusion period provided a consistent, stable level of myocardial stunning with normal resting coronary blood flow (CBF). The first goal of the study was to determine the effects of forskolin, a direct activator of adenylyl cyclase, in this model. Once it was determined that forskolin elicited enhanced coronary vasodilation in the postischemic, stunned myocardium, we determined its mechanism. We repeated experiments in the presence of nitric oxide (NO) synthase inhibition by N-nitro-L-arginine (L-NNA) and examined other interventions that elicit coronary vasodilation, which are mediated in part by NO (bradykinin and reactive hyperemia) or are endothelial independent (sodium nitroprusside). The third goal was to determine if the enhanced cAMP-induced coronary vasodilation in stunned myocardium was primary or secondary to changes in myocardial metabolic demand, determining regional left ventricular (LV) function and measuring coronary venous oxygen content and oxygen consumption. Because it is not well recognized that cAMP can elicit NO-mediated vasodilation, we determined whether cAMP could affect NO production in vitro in isolated coronary microvessels in the absence of myocardial effects by administering direct activators of adenylyl cyclase and protein kinase A, e.g., forskolin (12, 20), NKH-477 (12, 21, 25), and 8-bromo-cAMP (30).

The pig was selected for this study because its coronary anatomy is similar to that of the human heart. Also, the lack of preformed collateral vessels in this species allows a more consistent, reproducible reduction in tissue blood flow distal to the CS, since in the presence of preformed collaterals blood flow constantly changes in the myocardium distal to the stenosis. The conscious animal was selected to avoid complicating factors such as anesthesia and recent surgery, which could affect coronary vascular and myocardial responses. In addition, this permitted the utilization of paired experiments, i.e., in the presence and absence of NO synthase inhibition.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Five domestic swine, weighing 20.9 ± 0.2 kg, were sedated with 5 mg/kg im telazol and 0.05 mg/kg im atropine. General anesthesia was maintained with isoflurane (0.5-1.5 vol%) after tracheal intubation. With the use of sterile surgical technique, a left thoracotomy was performed at the fourth intercostal space. Tygon catheters (Norton Plastics, Akron, OH) were implanted in the descending aorta and in the left atrium for measurement of pressures and radioactive microsphere injection. A Silastic catheter was placed retrograde into the coronary vein as close to the left anterior descending coronary artery as anatomically possible. A solid-state miniature pressure gauge was implanted in the LV cavity to obtain LV pressure and the derivative of pressure in relation to time (dP/dt). The left anterior descending coronary artery was isolated, and a flow transducer (Transonics Systems, Ithaca, NY) and a hydraulic occluder made of Tygon tubing were implanted. Additionally, two pairs of ultrasonic crystals were implanted transmurally across the LV free wall, in the anterior and in the posterior regions, for measurement of regional wall thickness. The subendocardial crystal was introduced obliquely, so that injury or fibrosis would not impair the myocardium between the two crystals. Proper alignment of the epicardial and endocardial crystals was achieved during surgical implantation by positioning the crystals to obtain a received signal on the oscilloscope with the greatest amplitude and shortest transit time. The crystals distal to the occluder were implanted in the central ischemic zone as defined by a test coronary artery occlusion at the time of operation. The correct placement of the crystals was also confirmed at necropsy. The wires and catheters were externalized between the scapulae, the incision was closed in layers, and the chest was evacuated. Each pig was treated with 1 g iv of cefazolin before surgery and with cephalexin (15 mg/kg, 2 times/day) orally for 1 wk after surgery. Animals used in this study were maintained in accordance with the Guide for the Care and Use of Laboratory Animals (National Research Council, Revised 1996).

Hemodynamics were recorded on a digital recorder (PC216Ax; Sony Precision Technology) and on a multiple-channel thermal strip chart (Astro-Med). Aortic and left atrial pressures were measured with a strain-gauge manometer (Datex Ohmeda, Madison, WI) connected to the respective fluid-filled catheters. The solid-state LV pressure gauge was cross-calibrated with measurements of systolic aortic and left atrial pressures. LV dP/dt was calculated with an operational amplifier connected as a differentiator, which has a frequency response of 700 Hz. Mean arterial pressure was determined using a resistance-capacitance filter having a 2-s time constant. The tracings corresponding to the CBF responses before and after 15 s of coronary artery occlusion were digitized using a scanner interfaced to a computer. The area under the curve representing the volume and duration of the CBF deficit, i.e., the flow debt, and the excess of CBF that followed the release of the coronary artery occlusion, i.e., flow repayment, were quantified with image analysis software. Coronary vascular resistance was calculated by dividing the mean aortic pressure by the mean CBF. During the peak CBF response, coronary vascular resistance was calculated using the value of mean aortic pressure at that time point. Regional myocardial function was measured with an ultrasonic transit-time dimension gauge. This instrument measures the transit time of acoustic signals traveling at a sonic velocity of 1.58 × 10⁶ mm/s between the myocardial crystal pairs. The drift of this instrument, although minimal, was effectively compensated for by repeated calibrations. A cardiotachometer triggered by the LV pressure pulse provided instantaneous and continuous records of heart rate.

All pigs were introduced to a sling for training 1-2 h daily over a 1- to 2-wk period before surgery, and training was resumed after 1 wk of postoperative recovery. The experiments in conscious pigs were initiated 10-14 days after surgery. Five pigs were used for two 1.5-h CS protocols, i.e., before and during NO synthase inhibition. The two 1.5-h CS protocols were performed approximately 1 wk apart. Valium was administered at 0.5-1.0 mg/kg for tranquilization before initiation of the experimental protocol and additionally as required, i.e., if the pigs became agitated transiently. After global and regional baseline hemodynamic data were recorded, CS was induced by introducing saline into the hydraulic occluder to reduce CBF by ~40%. The degree of CBF reduction was then continuously monitored and sustained for 1.5 h.

Before CS and after a 12-h reperfusion period, coronary reactivity was investigated by reactive hyperemia induced with a 15-s coronary occlusion bolus infusion of bradykinin (0.05 µg/kg), a bolus infusion of sodium nitroprusside (2.5 µg/kg) into the left atrium, and by intravenous forskolin infusion (20 nmol · kg¹ · min¹). The dose of forskolin was chosen to provide a consistent 25-50% increase in baseline CBF in a control animal without inducing excitement. After a 4-day recovery period, systemic NO synthase inhibition (L-NNA, 30 mg · kg¹ · day¹ for 3 days) was initiated, and a second 1.5-h CS was induced to match the CBF reduction induced during the first CS. Also, similar to the prior CS, coronary reactivity was again examined before CS and after 12 h of reperfusion. Blood samples were taken at baseline in four animals to assess myocardial oxygen consumption in the absence and presence of L-NNA.

Six million microspheres (15 ± 1 µm) labeled with ⁹⁵Nb, ⁸⁵Sr, ¹⁴¹Ce, ⁴⁶Sc, ¹¹³Sn, ⁵¹Cr, ¹¹⁴In, or ¹⁰³Ru were suspended in 0.01% Tween 80 solution (10% dextran) and were placed in an ultrasonic bath for 30-60 min. Before the first injection of microspheres, 1 ml of Tween 80 solution was injected to test for potential adverse cardiovascular effects. Microspheres were injected and flushed with saline over a 20-s period via the left atrial catheter. Arterial blood reference samples were withdrawn at a rate of 7.75 ml/min for a total of 120 s. Radioactive microspheres were administered at baseline, during forskolin infusion, and one time (90 min) during each CS.

Three days after the final CS, the animals were anesthetized with pentobarbital sodium (30-50 mg/kg iv) and heparinized (600 USP U/kg). The heart was excised, and samples were taken from the area at risk and normal zone for radioactive microsphere tissue blood flow determinations. The samples were counted in a gamma-counter (Searle Analytical) with appropriately selected energy windows. After correction of counts for background and crossover, regional myocardial blood flow was obtained and expressed as milliliters per minute per gram of tissue.

Isolated microvessel preparations. Eight additional pigs were used to determine whether interventions that activate cAMP can effect endothelial formation of NO in porcine coronary microvessels. Hearts were obtained from pentobarbital sodium-anesthetized pigs and were kept in ice-cold PBS containing 0.1% BSA at pH 7.4. The isolation of coronary microvessels from the left ventricle of the pig heart was performed by using the method originally developed by Gerritsen and Printz (8, 27, 32). Coronary microvessels were obtained free of both large arteries and veins and also of myocytes by a series of steps involving sequential dissection, homogenation, sieving, and glass bead purification. Microvessels were placed in a small package of 80-µm nylon mesh, transferred to a tissue bath containing PBS, and oxygenated with 95% oxygen and 5% carbon dioxide for 30 min. Approximately 20 mg (wet weight) of tissue were placed in the 5-ml plastic tubes that contained 500 µl PBS as control or 450 µl PBS and a 50-µl aliquot of drug used to stimulate (e.g., forskolin) or inhibit [e.g., N-nitro-L-arginine methyl ester (L-NAME)] NO formation. All drugs were incubated with tissue for 20 min. At the end of the incubation period, the tubes were removed from the bath, the microvessels were removed from the tubes, and sulfanilamide (450 µl of 1%) and N-(1-naphthyl)ethylene diamine (NEDA; 50 µl of 0.2%) were added to each tube for diazotization of sulfanilic acid by NO (Griess reaction), resulting in a pink color that is proportional to the amount of nitrite present. After 5-10 min incubation at room temperature, the supernatant was removed from each tube. Formation of NO was measured as nitrite, which is the major metabolite of NO in aqueous solution. Formation of NO was measured using a spectrophotometer (Uvikon 930 spectrophotometer; Kontron Instruments) to determine the increased absorbance at 540 nm in relation to a standard curve generated by known concentrations of nitrite. To construct a standard curve for nitrite, a stock solution of NaNO₂ (10⁵ mol/l) was prepared and diluted each day. Sulfanilamide (450 µl of 1%) and NEDA (50 µl of 0.2%) were added to each tube and mixed well. The tubes were allowed to stand at room temperature for 5-10 min for full color (pink) development, and absorbance of nitrite was measured at 540 nm. Absorbance was computed and converted to a straight line using a regression analysis (y = a + bx, r > 0.99). Nitrite absorbance produced by microvessels from pig heart was measured using the linear regression formula, and resulting values were computed.

10

4

8

4

8

4

10

4

8

4

8

4

3

2

4

4

4

4

Data/statistics. All data are reported as means ± SE. Paired data were examined statistically for all comparisons. Myocardial function during CS was recorded continuously but was analyzed at every 30 min during the 1.5-h CS. These values were averaged and reported as average CS. Hemodynamics and regional myocardial blood flows during CS and reperfusion were analyzed using repeated-measures ANOVA between groups with a Student-Newman-Keuls test for post hoc comparison of means. P < 0.05 was taken as the level for significance. Differences of nitrite production from control were determined using ANOVA. A value of P < 0.05 was considered statistically significant.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Effects of CS. During 1.5 h of CS, CBF was reduced by an average of 38 ± 1% from a baseline of 28 ± 2 ml/min, and ischemic zone wall thickening decreased by an average of 58 ± 4% from 2.3 ± 0.4 mm. After 12 h of reperfusion, CBF had returned to baseline levels, but ischemic zone wall thickening remained depressed (P < 0.05) by 25 ± 5% from baseline. During CS, there was an increase (P < 0.05) in LV end-diastolic pressure (+45 ± 9% from 9 ± 1 mmHg) that remained significantly elevated from baseline (+15 ± 7%) at 12 h CAR. Other hemodynamic variables did not change during CS.

Effects of NO synthase inhibition. After 3 days of L-NNA (30 mg · kg¹ · day¹), peak systolic LV pressure had increased by an average of 27 ± 8% (P < 0.05) to 142 ± 8 mmHg, but baseline heart rate, CBF, and LV dP/dt remained unchanged from baseline values.

Effects of CS with systemic NO synthase inhibition. Although CBF was reduced with the CS during systemic NO synthase inhibition (40 ± 2%) similar to the control CS (38 ± 1%) by experimental design, ischemic zone wall thickening decreased significantly more (P < 0.05) during CS, by an average of 86 ± 8% from 2.0 ± 0.4 mm. After 12 h of reperfusion, CBF was similar to baseline, but post-CS regional wall thickening in the ischemic zone was more severely depressed (P < 0.05, by 66 ± 7%) compared with the control response in the absence of NO synthase inhibition (25 ± 5%). There was a moderate, but significant, change in LV end-diastolic pressure during CS that had returned to baseline by 12 h CAR. Other hemodynamic variables did not change significantly with CS.

Coronary reactivity: endothelium-dependent and -independent responses. Coronary vascular response data are illustrated as a change from baseline (ml/min) in Fig. 1. Baseline CBF was similar during each intervention (Table 1). Reactive hyperemia repayment of the flow deficit ratio had increased (+22 ± 4%) after 12 h of CAR compared with pre-CS. Also at 12 h after CAR, enhanced CBF responses compared with pre-CS were observed for bradykinin (+55 ± 5%) but not for sodium nitroprusside. With the use of a similar protocol during systemic NO synthase inhibition, the enhanced CBF responses with reactive hyperemia and bradykinin were not observed after CS. Baseline coronary vascular resistance was higher (P < 0.05) after NO synthase inhibition (4.54 ± 0.40 mmHg · ml¹ · min¹) compared with before NO synthase inhibition (3.62 ± 0.18 mmHg · ml¹ · min¹). The decreases, however, in coronary vascular resistance during each intervention were consistent with the blood flow response data. Decreases in coronary vascular resistance in response to bradykinin were significantly greater (P < 0.05) after CS compared with before CS (1.78 ± 0.14 vs. 1.40 ± 0.16 mmHg · ml¹ · min) in the absence of NO synthase inhibition but were unchanged in the presence of NO synthase inhibition (1.59 ± 0.29 vs. 1.47 ± 0.39 mmHg · ml¹ · min). Decreases in coronary vascular resistance in response to sodium nitroprusside were unchanged after CS compared with before CS in the absence (1.49 ± 0.18 vs. 1.39 ± 0.22 mmHg · ml¹ · min) and presence (1.82 ± 0.31 vs. 1.67 ± 0.39 mmHg · ml¹ · min) of NO synthase inhibition.

View larger version (35K): [in this window] [in a new window]   Fig. 1.   Repayment of flow deficit ratio in response to reactive hyperemia (RH) and increase in coronary blood flow (CBF, ml/min) in response to bolus bradykinin (BK) and sodium nitroprusside (SNP) before (open bars) and 12 h after (hatched bars) coronary stenosis (CS) in 5 pigs in the absence () and presence (+) of N-nitro-L-arginine (L-NNA). RH and BK responses were augmented at 12 h post-CS in the absence of L-NNA but not in the presence of L-NNA. *P < 0.05.

Coronary reactivity: cAMP-mediated response. Representative waveforms demonstrating responses to intravenous forskolin infusion before and after 1.5 h CS and 12 h reperfusion are shown in Fig. 2. Hemodynamic and coronary vascular responses to forskolin before and 12 h after reperfusion in the absence and presence of NO synthase inhibition are summarized in Tables 1 and 2, respectively. The increase in CBF was significantly greater (P < 0.05) 12 h after reperfusion for 1.5 h CS (62 ± 9%) compared with before CS (37 ± 3%). Similarly, myocardial blood flow responses assessed by radioactive microspheres (Table 3) also demonstrated enhanced responses to forskolin (P < 0.05) in the previously ischemic zone after 12 h of reperfusion (73 ± 14%) compared with before CS (24 ± 6%). Tissue flow responses to forskolin in the nonischemic zone were similar after 12 h of reperfusion (32 ± 10%) compared with before CS (29 ± 10%). Interestingly, this enhanced flow response was elicited concurrent with a significant reduction in the functional response to forskolin in the previously ischemic zone after 12 h of reperfusion (+0.09 ± 0.03 mm) compared with before CS (+0.56 ± 0.08 mm). Coronary vascular responses were repeated during recovery from 1.5 h CS in four animals, and enhanced coronary vascular responses to forskolin were still present at 48 h after reperfusion. However, this enhanced vascular response subsided and was no longer present 4 days after CS.

View larger version (28K): [in this window] [in a new window]   Fig. 2.   Representative waveforms of left ventricular pressure (LVP), LV derivative of pressure in relation to time (dP/dt), aortic (Ao) pressure, and phasic and mean coronary blood flow (CBF) at baseline and during forskolin infusion (20 nmol · kg1 · min1) before CS (pre-CS) and at 12 h reperfusion after a 1.5-h CS (post-CS). Note the enhanced flow response during forskolin infusion after CS and reperfusion.

Myocardial oxygen consumption. Calculated arterial-venous differences in volume percent blood oxygen content at baseline were not significantly different before and after 1.5 h CS and 12 h reperfusion either in the absence (9.1 ± 0.7 vs. 7.9 ± 0.6 vol% oxygen) or presence (9.7 ± 1.3 vs. 9.2 ± 1.1 vol% oxygen) of systemic NO synthase inhibition. However, there was a significant decrease in the arterial-venous difference in oxygen content during forskolin infusion 12 h after 1.5 h CS (6.4 ± 0.4 vol% oxygen) compared with before CS (9.1 ± 0.4 vol% oxygen). Similarly, forskolin increased myocardial oxygen consumption less after 1.5 h CS (+0.59 ± 0.03 from 2.22 ± 0.28 ml/min) compared with before CS (+0.94 ± 0.11 from 2.56 ± 0.37 ml/min). This resulted in a significant shift (P < 0.05) in the relationship between CBF and myocardial oxygen consumption, indicating primary coronary vasodilation (Fig. 3). This shift was not observed after NO synthase inhibition (Fig. 3).

View larger version (15K): [in this window] [in a new window]   Fig. 3.   Mean ± SE increase in CBF from baseline vs. increase of myocardial oxygen consumption (MO2) in response to forskolin infusion pre-CS () and 12 h post-CS () in the absence of L-NNA and pre-CS () and post-CS () in the presence of L-NNA. Note the significant (P < 0.05) left shift in the post-CS regression line compared with pre-CS (n = 5) in the absence of L-NNA, indicating primary coronary vasodilation.

Coronary microvessel preparations. Forskolin (10¹⁰ to 10⁴ mol/l) increased nitrite production in a concentration-related manner (Fig. 4). L-NAME and Rp-cAMP significantly (P < 0.05) decreased nitrite production by 91 and 85%, respectively, to the highest dose of forskolin (Fig. 5). There was no inhibitory effect of the bradykinin receptor antagonist HOE-140 (not shown).

View larger version (14K): [in this window] [in a new window]   Fig. 4.   Increase in nitrite production (pmol/mg) in response to forskolin, NKH-477 (NKH), and 8-bromo-cAMP with microvessels isolated from pig hearts (n = 8). Each intervention resulted in significant dose-dependent increases in nitric oxide (NO) production.

View larger version (14K): [in this window] [in a new window]   Fig. 5.   NO synthase inhibition with N-nitro-L-arginine methyl ester (L-NAME) or protein kinase A [Rp stereoisomer of cAMP (Rp-cAMP)] inhibition prevented the enhanced nitrite production (pmol/mg) in response to the highest dose of forskolin (104; A), NKH (102; B), and 8-bromo-cAMP (104; C). Data indicate that nitrite reflects NO production and suggest a role for cAMP-dependent production (n = 8).

8

4

8

4

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

The major finding of the current study was that stimulation of adenylyl cyclase evokes cAMP-mediated coronary vasodilation, which is enhanced in the presence of stable myocardial stunning in the conscious pig, with a mechanism involving NO. This finding is of clinical and physiological importance with regard to the effects of ischemia and reperfusion, as this model may more closely parallel the type of ischemia that most commonly occurs in humans with coronary artery disease. Importantly, the ischemia in this model is totally reversible and does not result in infarction (18, 26). This point may reconcile the difference between the current study and prior results. For example, Tofukuji et al. (29, 30) reported opposite results, i.e., that coronary microvascular relaxation to forskolin and 8-bromo-cAMP in anesthetized open-chest pigs was significantly reduced after 30 min of coronary occlusion and 60 min of reperfusion and in arrested hearts after 1 h with cardioplegia. However, 30 min of coronary artery occlusion in dogs (14) and pigs (7) results in extensive infarction, which would obviously impair the coronary vasoactivity. In addition, many previous studies have examined -adrenergic receptor-stimulated responses in stunned myocardium, but relatively few have examined direct stimulation of adenylyl cyclase. As noted above, those studies may have been complicated by the presence of infarction (29, 30).

The second goal of the current study was to determine the mechanism of this enhanced cAMP-induced vasodilation in stunned myocardium. A recent study from our laboratory observed delayed, enhanced NO-dependent vasodilation in stunned myocardium after 10 min of coronary occlusion in the conscious dog (15). However, enhanced NO-dependent, cAMP-mediated coronary vasodilation is not well recognized. Thus we examined the effects of forskolin in the absence and presence of NO synthase inhibition. In the current study, the enhanced cAMP-induced coronary vasodilation in stunned myocardium was prevented by NO synthase inhibition. This appears to be the first demonstration that direct adenylyl cyclase stimulation can induce enhanced NO-mediated coronary vasodilation in the stunned myocardium. This is particularly important in view of the potential role of NO mediating ischemic cardioprotection (1, 9, 11, 18, 19, 23). One might argue that the sequence of the experiments dictated the results, i.e., that experiments after NO synthase inhibition generally followed the initial control experiments. To determine whether prior CS in the absence of NO synthase inhibition had an effect on the responses of a second CS in the presence of NO synthase inhibition, we evaluated the forskolin response before and 12 h after reperfusion following CS in the presence of NO synthase inhibition as the only protocol (n = 2). We demonstrated that enhanced coronary vasodilation to forskolin in stunned myocardium was abolished even in the absence of a prior period of CS and reperfusion.

We recently observed that basal and forskolin-stimulated adenylyl cyclase activity decreased in the conscious pig after 10 min of coronary occlusion and 30 min of reperfusion, as did myocardial inotropic responses to forskolin stimulation (25, 31). These findings are consistent with those of the current study during stunning after 90 min of CS and 12 h of reperfusion, which show depressed functional responses to forskolin stimulation in vivo. Thus the depressed inotropic response observed earlier and also in this study cannot account for the markedly enhanced coronary vascular responses to adenylyl cyclase stimulation that are described for the first time in the current investigation. Indeed, it actually exaggerates the importance of the current findings, since the decreased contraction should require less oxygen and less CBF. This was verified by the calculation of myocardial oxygen consumption, which was reduced with forskolin after CS.

Prior studies have shown that forskolin can induce NO release from porcine aortic endothelial cells (17). We wished to determine if this mechanism could be demonstrated in vitro in a preparation of isolated porcine coronary microvessels (8, 27, 32). We observed that NO production is enhanced by cAMP stimulation with three separate agonists of adenylyl cyclase and protein kinase A (forskolin, NKH-477, and 8-bromo-cAMP) in these isolated microvessels and that these actions were blocked by L-NAME. These experiments in isolated microvessels from normal pigs demonstrate that an NO-mediated cAMP-dependent coronary vascular response mechanism is available. It remains to be determined if this mechanism is augmented under in vitro conditions after recovery from ischemia, as it is in vivo, which was the major conclusion of the current investigation.

These data taken together point out a potentially new pathway for the regulation of coronary vascular tone in the stunned myocardium, which may also play a role in the mechanism of ischemic preconditioning. It is interesting to speculate that the mechanism in part may involve phosphorylation of endothelial NO synthase by cAMP and regulation of NO by protein kinase Akt (4, 6), since AMP-activated kinase can phosphorylate human endothelial NO synthase (3). Although forskolin activates -adrenergic signaling distal to the -adrenergic receptor, these findings of regulation of coronary microvessels by cAMP and its enhanced action in stunned myocardium may be clinically relevant with regard to naturally induced sympathetic stimulation and stimulation of adenylyl cyclase both by sympathomimetic amines and noncatecholamine agonists in patients with stunned myocardium induced naturally or by mechanical or surgical interventions.