Nitric oxide production is maintained in exercising swine with chronic left ventricular dysfunction

David B. Haitsma, Daphne Merkus, Jefrey Vermeulen, Pieter D. Verdouw, Dirk J. Duncker

Abstract

Left ventricular (LV) dysfunction caused by myocardial infarction (MI) is accompanied by endothelial dysfunction, most notably a loss of nitric oxide (NO) availability. We tested the hypothesis that endothelial dysfunction contributes to impaired tissue perfusion during increased metabolic demands as produced by exercise, and we determined the contribution of NO to regulation of regional systemic, pulmonary, and coronary vasomotor tone in exercising swine with LV dysfunction produced by a 2- to 3-wk-old MI. LV dysfunction resulted in blunted systemic and coronary vasodilator responses to ATP, whereas the responses to nitroprusside were maintained. Exercise resulted in blunted systemic and pulmonary vasodilator responses in MI that resembled the vasodilator responses in normal (N) swine following blockade of NO synthase withN ω-nitro-l-arginine (l-NNA, 20 mg/kg iv). However, l-NNA resulted in similar decreases in systemic (43 ± 3% in N swine and 49 ± 4% in MI swine), pulmonary (45 ± 5% in N swine and 49 ± 4% in MI swine), and coronary (28 ± 4% in N and 35 ± 3% in MI) vascular conductances in N and MI swine under resting conditions; similar effects were observed during treadmill exercise. Selective inhibition of inducible NO synthase with aminoguanidine (20 mg/kg iv) had no effect on vascular tone in MI. These findings indicate that while agonist-induced vasodilation is already blunted early after myocardial infarction, the contribution of endothelial NO synthase-derived NO to regulation of vascular tone under basal conditions and during exercise is maintained.

  • coronary circulation
  • myocardial infarction
  • pulmonary circulation
  • regional blood flows
  • systemic circulation

heart failure is accompanied by increased production of neurohormones such as norepinephrine, angiotensin II, and endothelin (19, 50), which serves to maintain arterial pressure by partially restoring cardiac output and increasing peripheral vascular resistance, to maintain perfusion of essential tissues like the brain and heart at the expense of renal, pancreatic, and intestinal blood flow. Several studies have indicated that also endothelial dysfunction, in particular a decreased production of nitric oxide (NO), contributes to an increased peripheral resistance in congestive heart failure (18,25, 29, 32, 58). A loss of NO availability could aggravate left ventricular (LV) dysfunction due to the peripheral vasoconstriction-induced increase in LV afterload, coronary vasoconstriction, and increased myocardial O2 consumption (15, 34, 47, 58). In support of this concept, a decreased NO production coincides with the transition of LV dysfunction to overt congestive heart failure in models of pacing-induced heart failure (47, 58).

Clinically, the most common cause for heart failure is myocardial infarction (MI). Studies on the role of NO in the regulation of vasomotor tone in MI-induced LV dysfunction in rats indicate that between 4 and 16 wk after infarction, acetylcholine- or ADP-induced NO-mediated relaxation of systemic conductance arteries (4, 40,43, 53, 57) and resistance vessels (13) is reduced, although this is not an unequivocal finding (2, 6). In contrast, basal NO production is maintained in systemic conductance arteries (2, 43) and in resistance vessels in various regional vascular beds in vivo (6, 10, 13, 21). These findings could be interpreted to suggest that early after infarction, a reduced NO-mediated vasodilator capacity could contribute to a blunted vasodilator response, and hence tissue hypoperfusion during exercise (9) at a time when basal NO production is still intact. Recently, Duncker et al. (15) reported that in resting and exercising normal swine, inhibition of NO synthase withN ω-nitro-l-arginine (l-NNA) decreased systemic and pulmonary vascular conductance and decreased flow to the kidney, pancreas, and part of the gut, whereas flow to the skeletal muscle was unaffected. A similar pattern was found in exercising swine with a 3-wk-old MI (24), suggesting that in this model of moderate LV dysfunction, NO availability is already impaired, in particular during exercise. Consequently, we tested in the present study the hypothesis that loss of NO-mediated vasodilation contributes to abnormalities in vascular tone control in regional systemic, pulmonary, and coronary beds in exercising swine with LV dysfunction after MI. Because our results indicated that the contribution of NO to exercise-induced vasodilation was maintained, and because inducible NO synthase (iNOS) may be upregulated after infarction (12, 52), we further tested whether upregulation of iNOS compensated for the potential loss of endothelial NO synthase (eNOS) activity.

MATERIALS AND METHODS

Studies were performed in accordance with the Guide for the Care and Use of Laboratory Animals (NIH Publication 86-23, Revised 1985) and with approval of the Animal Care Committee of the Erasmus University Rotterdam. Thirty-six 2- to 3-mo-old Yorkshire × Landrace swine (22 ± 1 kg at the time of surgery) of either sex entered the study; 22 swine were designated to the MI group and 14 to the normal group (N). Results from 11 of 14 N have been previously reported (15). Daily adaptation of animals to laboratory conditions started 1 wk before surgery.

Surgical Procedures

Swine were sedated (ketamine, 20 mg/kg im), anesthetized (thiopental, 10 mg/kg iv), intubated, and ventilated with O2 and N2O to which 0.2%-1% (vol/vol) isoflurane was added (14, 16, 17, 56). Anesthesia was maintained with midazolam (2 mg/kg followed by 1 mg · kg−1 · h−1 iv, for 2 h) and fentanyl (5–10 μg · kg−1 · h−1 iv). Under sterile conditions, the chest was opened via the fourth left intercostal space, and a fluid-filled polyvinylchloride catheter was inserted into the aortic arch for aortic blood pressure measurement (Combitrans pressure transducers, Braun) and blood sampling for determination of blood gases (Acid-Base Laboratory model 505, Radiometer), O2 saturation, and hemoglobin concentration (OSM2, Radiometer), and computation of O2 content, O2 supply, and O2 consumption (16,17). An electromagnetic flow probe (14–15 mm, Skalar) was positioned around the ascending aorta for measurement of cardiac output. A microtipped pressure-transducer (P4.5, Konigsberg Instruments) was inserted into the LV via the apex. Polyvinylchloride catheters were inserted into the LV to calibrate the Konigsberg transducer LV pressure signal (16, 17) and into the left atrium to measure pressure and inject radioactive microspheres to determine regional blood flows (14). Catheters were inserted into the pulmonary artery to measure pressure, administer drugs, and collect mixed venous samples, while an angiocatheter was inserted into the anterior interventricular vein for blood sampling, and a Doppler flow probe (2.0–3.0 mm, Crystal Biotech) was placed around the left anterior descending coronary artery (16,17). The proximal left circumflex coronary artery (LCx) was permanently occluded in 22 MI swine (56), which were monitored for 1 h and if needed internally defibrillated (10–30 J). Six MI swine died due to recurrent fibrillation. Catheters were tunneled to the back, and animals were allowed to recover, receiving analgesia (0.3 mg buprenorphine im) for 2 days and antibiotic prophylaxis (25 mg/kg amoxicillin and 5 mg/kg gentamycin iv) for 5 days (16, 17). Four MI swine died overnight during the first week after surgery.

Experimental Protocols

Agonist-induced vasodilation.

The hemodynamic responses to the vasodilator ATP (50–200 μg · kg−1 · min−1 iv) were determined in five resting MI swine and compared with six N swine. ATP was employed instead of acetylcholine, because acetylcholine is less effective as an endothelium-dependent vasodilator in the porcine coronary circulation. We have previously shown in N swine (15) that vasodilation produced by these doses of ATP is virtually abolished by pretreatment with l-NNA (20 mg/kg iv), indicating that at these doses ATP produces vasodilation principally via NO. To study the responsiveness of the vascular beds to NO, we determined in the same animals the hemodynamic responses to the endothelium-independent NO donor sodium nitroprusside (SNP, 0.5–5 μg · kg−1 · min−1 iv).

Exercise.

Central and regional systemic, pulmonary and coronary hemodynamic responses to exercise were studied in 14 N (29 ± 1 kg) and 12 MI (28 ± 1 kg) swine ∼2–3 wk after surgery. After baseline measurements (lying, 0L, and standing, 0S) were obtained, a treadmill exercise protocol was begun (1–4 km/h); data were collected during the last 30 s of each 3-min exercise stage (16, 17). After completion of the exercise protocol the swine were allowed to rest for 90 min, and then l-NNA (20 mg/kg iv) was administered and the exercise protocol repeated (15).

On another day (separated by at least 4 days from the l-NNA protocol and performed in random order with the l-NNA protocol), the reproducibility of the responses to exercise was studied. Animals underwent a second control exercise trial 90 min after the first trial, as previously described (15-17).

On a different day (separated by at least 4 days from thel-NNA protocol), the exercise protocol was repeated in 4 MI swine, but before the second exercise protocol the iNOS inhibitor aminoguanidine [10- to 50-fold selectivity iNOS/eNOS (5,61)] was administered in a dose of 20 mg/kg iv (infused in 15 min starting 30 min before exercise), which is in the dose range (10–30 mg/kg) that has been reported to produce a high degree of iNOS inhibition in rats (62), rabbits (59), and swine (64).

Regional Blood Flows

In seven N and seven MI swine, regional blood flows were determined on a separate day using the radioactive microsphere technique (14, 15). In N swine, radioactive microspheres were injected at rest (lying, 0L) and during exercise at 5 km/h, whereas in MI swine, microspheres were injected at rest (lying) and during exercise at 4 km/h [maximum treadmill speed for most MI (24)].

Data Analysis

Digital recording and off-line analysis of hemodynamics and regional blood flow have been described previously (14, 16,17). Statistical analysis was performed using two-way (exercise level and infarction) analysis of variance (ANOVA) for repeated measures, followed by Dunnett's test (exercise effect) or unpairedt-test (MI vs. N). Analysis of covariance (ANCOVA) for repeated measures was used to detect statistically significant differences of relations between hemodynamic variables and body or myocardial O2 consumption (V˙o 2) (withV˙o 2 as a covariate) in MI versus N swine. Significance was accepted when P < 0.05. Data are means ± SE.

RESULTS

Cardiac Anatomic Data

Despite the loss of viable LV myocardial tissue due to infarction, the LV weight-to-body weight ratio was 15% higher than in normal swine (3.69 ± 0.16 vs. 3.18 ± 0.17 g/kg, P < 0.05). Right ventricular weight-to-body weight ratio was also higher in MI swine compared with N swine (1.71 ± 0.23 vs. 1.14 ± 0.13 g/kg, P < 0.05), which correlated well with the elevated pulmonary artery pressure in MI swine (r 2 = 0.63; P < 0.05).

Agonist-Induced Vasodilation

The NO-dependent vasodilator ATP caused dose-dependent dilation of the systemic and coronary circulation (Fig.1), but did not result in significant pulmonary vasodilation (not shown). At each dose of ATP, dilation was less in MI swine than in N swine, indicating either a reduced NO production, an increased NO degradation, or reduced vascular smooth muscle responsiveness to NO after MI. The vasodilator response of systemic and coronary vascular beds to the NO donor SNP were comparable in MI and N swine, suggesting that the blunted vasodilator response to ATP was not due to a reduced bioavailability or vascular responsiveness to NO.

Fig. 1.

Increases in systemic vascular conductance (SVC = cardiac output/mean arterial pressure) and coronary vascular conductance [CVC = coronary blood flow/(mean arterial pressure − mean left atrial pressure)] in myocardial infarction (MI) and normal (N) swine. Changes in CVC (ΔCVC) have been presented as % changes because of differences in baseline flow (see Table 1) due to the presence of left ventricular (LV) hypertrophy in MI (56). Note that ATP-induced vasodilation was blunted (P < 0.05 by ANOVA), whereas nitroprusside-induced vasodilation was maintained. Data are means ± SE.

Role of NO at Rest and During Exercise

Systemic circulation.

Under resting conditions, l-NNA- induced inhibition of NO production caused peripheral vasoconstriction in both MI and N swine, as indicated by the decrease in systemic vascular conductance (SVC, Fig. 2). The systemic vasoconstriction markedly increased aortic blood pressure despite concomitant decrease in cardiac output. The latter was due to a decrease in heart rate, probably mediated by the baroreceptor reflex (Table 1). The decrease in cardiac output was accompanied by an increase in body O2 extraction so that whole bodyV˙o 2 was maintained, resulting in a decrease in mixed venous Po 2 (Fig. 2). These responses to l-NNA were similar in MI and N groups, indicating that the contribution of NO to the regulation of vasomotor tone in the systemic vascular bed was maintained 2–3 wk after infarction.

Fig. 2.

Effect of inhibition of nitric oxide (NO) synthase byN ω-nitro-l-arginine (l-NNA, 20 mg/kg iv) on SVC at rest (0L, lying; 0S, standing) and during treadmill exercise (1–4 km/h;A) and on the relation between mixed venous Po 2 at rest (lying) and during treadmill exercise (1–4 km/h; B) in MI and N swine. Data are means ± SE. V˙o 2, O2consumption. Dot inside a symbol denotes P < 0.05 vs. 0L; †P < 0.05 l-NNA vs. corresponding control; there were no significant differences in the responses to l-NNA between MI and N swine.

View this table:
Table 1.

Effects of NO synthase inhibition on hemodynamics in MI and N swine

In N swine, l-NNA blunted the exercise-induced increase in SVC (ΔSVC from 0L to 4 km/h) from 46 ± 2 ml · min−1 · mmHg1 during control exercise to 32 ± 2 ml · min−1 · mmHg−1 during exercise in the presence of l-NNA (P < 0.05, Fig. 2). In MI, the exercise-induced increase in SVC (35 ± 3 ml · min−1 · mmHg−1,P < 0.05 vs. control run in N swine) resembled that in N swine after l-NNA. This should not be interpreted to suggest that a loss of NO production contributed to the blunted exercise-induced increase of SVC in MI, as l-NNA blunted the exercise-induced increase in SVC in MI swine (21 ± 2 ml · min−1 · mmHg−1) to the same extent as in N swine. Similarly, the effects of l-NNA on whole body O2 extraction and mixed venous Po 2 were virtually identical in N and MI swine (Fig. 2), indicating that the contribution of NO to the exercise-induced vasodilation of the systemic vascular bed was unperturbed 2–3 wk after infarction.

Regional systemic vascular beds.

l-NNA did not affect skeletal muscle blood flow in either N or MI group, either at rest or during exercise (Fig.3). The exercise-induced decrease in blood flow to various abdominal organs was more pronounced in the MI swine than the N swine. However, the l-NNA-induced decrease in flow to these organs was not different between MI and N groups both at rest and during exercise.

Fig. 3.

Effect of inhibition of NO synthase by l-NNA (20 mg/kg iv) on regional blood flows at rest (lying) and during treadmill exercise at 4 km/h (MI) or at 5 km/h (N). Regional blood flows are plotted as a function of whole body O2consumption (BV˙o 2) to facilitate comparison of the exercise data obtained in MI (4 km/h) and N (5 km/h) swine. Data are means ± SE. Dot inside a symbol denotesP < 0.05 vs. 0L; †P < 0.05 l-NNA vs. corresponding control. There were no differences in the responses to l-NNA between MI and N swine either at rest or during exercise.

Coronary circulation.

l-NNA decreased coronary vascular conductance (CVC) at rest and during exercise to a similar extent in MI and N swine (Table 1). The decrease in CVC was not simply the result of myogenic vasoconstriction secondary to the increase in aortic blood pressure, because it limited myocardial O2 supply thereby necessitating an increase in myocardial O2 extraction and resulting in a decrease of coronary venous Po 2(Fig. 4). Importantly, the effects ofl-NNA on CVC, O2 extraction, and coronary venous Po 2 were not different in the MI and N group, either at rest or during exercise, indicating that the contribution of NO to the regulation of vasomotor tone in the coronary resistance vessels was maintained in MI swine both at rest and during exercise.

Fig. 4.

Effect of inhibition of NO synthase by l-NNA (20 mg/kg iv) on myocardial O2 extraction and coronary venous Po 2 at rest (lying) and during treadmill exercise (1–4 km/h) in MI and N swine. MV˙o 2, myocardial O2consumption. Data are means ± SE. Dot inside a symbol denotesP < 0.05 vs. 0L; †P < 0.05 l-NNA vs. control; there were no significant differences in the responses to l-NNA between MI and N swine either at rest or during exercise.

Pulmonary circulation.

l-NNA produced similar increases in mean pulmonary arterial pressure (MPAP) and decreases in pulmonary vascular conductance (PVC) in MI and N groups under resting conditions (Table 1, Fig.5). In the N group, l-NNA blunted the exercise-induced increase in PVC (ΔPVC from 0L to 4 km/h) from 119 ± 40 ml · min−1 · mmHg−1 during control exercise (P < 0.05) to 53 ± 33 ml · min−1 · mmHg−1 during exercise in the presence of l-NNA (Fig. 5). In MI swine, the exercise-induced increase in PVC (48 ± 22 ml · min−1 · mmHg−1) resembled that in the N swine after l-NNA, suggesting that a loss of NO production contributed to the blunted exercise-induced increase in PVC in MI swine. However, the effects of l-NNA on PVC during exercise were similar in MI and N swine. These findings indicate that the contribution of NO to basal tone and the exercise-induced vasodilation of the pulmonary resistance vessels is unperturbed 2–3 wk after infarction.

Fig. 5.

Effect of inhibition of NO synthase by l-NNA (20 mg/kg iv) on pulmonary vascular conductance [cardiac output/(mean pulmonary artery pressure − mean left atrial pressure)] at rest (0L, lying; 0S, standing) and during treadmill exercise (1–4 km/h) in MI and N swine. Data are means ± SE. Dot inside a symbol denotes P < 0.05 vs. 0L; †P < 0.05 l-NNA vs. Control; there were no significant differences in the responses tol-NNA between MI and N swine.

Contribution of iNOS

Because l-NNA in the dose used inhibits both eNOS and iNOS, we used the specific iNOS inhibitor aminoguanidine to determine whether increased levels of iNOS-derived NO acted to compensate for a loss of eNOS-derived NO. Aminoguanidine did not have any effect on systemic and coronary vascular conductance or on whole body and myocardial O2 extraction either at rest or during exercise (Table 2), indicating that NO production by iNOS influenced neither basal tone nor exercise-induced vasodilation 2–3 wk after infarction.

View this table:
Table 2.

Effects of inducible NO synthase inhibition on hemodynamics in MI swine

Reproducibility of the Responses to Exercise

Ninety minutes after the first control exercise period, at a time when all hemodynamic variables had returned to baseline resting values, a second period of exercise resulted in virtually identical hemodynamic responses to exercise (Table 3).

View this table:
Table 3.

Reproducibility of systemic hemodynamic responses to exercise in MI and N swine

DISCUSSION

The major findings of the present study are that in swine with LV dysfunction due to a 2- to 3-wk-old MI: 1) agonist-induced receptor-mediated NO production was blunted compared with normal swine, whereas the vasodilator response to the endothelium-independent but NO-mediated vasodilation by nitroprusside was maintained; 2) inhibition of NO production by l-NNA resulted in similar increments in vascular tone compared with normal swine in the (regional) systemic, pulmonary, and coronary beds under basal resting conditions as well as during exercise up to 85% of maximum heart rate; and 3) inhibition of NO production by iNOS had no effect on vascular tone in the systemic, pulmonary, and coronary beds either at rest or during exercise.

Endothelial Dysfunction and Heart Failure

Agonist-induced vasodilation.

Advanced heart failure is associated with endothelial dysfunction, in particular, a reduced biological availability of NO. Thus clinical studies have shown that chronic heart failure is accompanied by blunted vasodilator responses to endothelium-dependent, receptor-mediated vasodilators (particularly acetylcholine) in the microcirculation of the LV myocardium (55), leg (25, 30), and forearm (3, 11, 28, 33, 41). In the canine model of pacing-induced end-stage congestive heart failure, attenuated vasodilator responses of resistance vessels to acetylcholine in vivo have also been observed in the microvasculature of the hindleg circulation (13, 35) and the coronary circulation (58). Also, in rats with MI-induced LV dysfunction, vasodilator responses in the microcirculation of the hindlimb to acetylcholine were blunted 10 wk after infarction (13), although this was not confirmed in another study at 1, 2, 3, 5, and 13 wk after infarction in rats (58). An explanation for the discrepancy between the two studies is not readily found but may be related to the use of blood (13), a scavenger of NO, versus buffered solution (58) for perfusion of the hindlimb. In the present study we also observed reduced vasodilator responses in the systemic and coronary microvasculature to ATP in doses that we have previously shown to be completely abolished by pretreatment with l-NNA (15). These findings of a blunted ATP-induced vasodilation are in agreement with the hypothesis that agonist-induced NO synthase-mediated NO production is blunted 2–3 wk after MI, but could also be explained by a decreased biological availability of NO or a reduced vascular smooth muscle responsiveness to NO. Thus, whereas several clinical studies reported that the responses to nitroglycerine or nitroprusside were maintained in patients with chronic heart failure (11, 25, 28,39), other studies reported a blunted vasodilator response to these NO donors (30, 33, 41). The severity of heart failure may underlie these equivocal findings because Bank et al. (3) observed blunted vasodilator responses to methacholine in the forearm of patients with mild or severe heart failure, whereas nitroprusside-induced vasodilation was blunted only in patients with severe heart failure. In agreement with this notion, the vasodilation by nitroglycerine or nitroprusside was maintained in LV dysfunction produced by MI (10, 13, present study), whereas vasodilation by nitroglycerine was blunted in the model of pacing-induced end-stage heart failure (58). This reduction in vasodilator response to NO donors could result from a decreased vascular smooth muscle responsiveness to NO or from a reduced half-life of NO due to increased levels of NO scavenging substances such as O 2 , which may be increased in chronic severe heart failure in part due to an increased production of tumor necrosis factor-α (22).

The preserved nitroprusside-induced vasodilation in the present study indicates that NO degradation and vascular smooth muscle responsiveness to NO were normal in MI. Therefore, the blunted vasodilator response to ATP is in agreement with the hypothesis that endothelial NO production in response to ATP was attenuated. However, it could be argued that in MI, accelerated ATP breakdown contributed to the attenuated vasodilation that was observed during ATP infusion in the MI group compared with the N group. Indeed, one cannot exclude that following administration of endothelium-dependent vasodilators such as acetylcholine, substance P, bradykinin, and ADP (endogenous compounds that are usually employed to assess endothelial function in heart failure), increased breakdown of these compounds contributes to the blunted vasodilator response. In previous experiments we observed in normal animals that the vasodilation in response to intravenous infusion of ATP in doses of 50–200 μg · kg−1 · min−1 was abolished by l-NNA both in the systemic and coronary vascular beds, indicating that at these doses ATP produces vasodilation exclusively via NO synthase-mediated NO production (15). Part of this vasodilation could be due to the production of ADP and adenosine (37). ADP has been shown to produce endothelium-dependent NO-mediated vasodilation via activation of the purinergic P2Y receptor, which is at least equipotent to ATP (48, 60). Adenosine, which is also equipotent to ATP (37), produces vasodilation that is primarily endothelium dependent at lower doses, whereas principally endothelium independent at higher doses (45, 65). This may explain our previous findings in normal swine that the vasodilation produced by ATP in doses of 300–500 μg · kg−1 · min−1 was only partially attenuated by nitro-l-arginine (15). The finding that the vasodilator response to nitroprusside was maintained suggests that vascular smooth muscle responsiveness is unaltered and hence that the endothelium-independent component of adenosine-induced vasodilation would likely be unperturbed. Thus, if ATP infusion in MI would have resulted in higher adenosine levels, this would have led to a greater degree of endothelium-independent vasodilation, which could then actually have contributed to an underestimation of the degree of endothelial dysfunction. Although we cannot entirely exclude that ATP breakdown to ADP and adenosine was enhanced in MI, the finding that the vasodilator response to ATP infusion was blunted in the MI group compared with the N group is in agreement with the hypothesis that agonist-induced (ATP, ADP, and adenosine) endothelium-dependent NO-mediated vasodilation is blunted in swine with LV dysfunction 2–3 wk after infarction.

Basal NO biological availability.

A loss of NO-mediated vasodilation in heart failure could contribute to reduced tissue perfusion. In addition, the peripheral vasoconstriction can increase the work load of the heart, which together with an increased coronary vasomotor tone, may lead to myocardial ischemia thereby aggravating LV dysfunction and enhancing the progression of LV dysfunction to heart failure. This is supported by studies in dogs with pacing-induced dilated cardiomyopathy, in which the loss of basal NO production in the LV myocardium coincides with the progression from LV dysfunction to overt heart failure (47,58). In contrast, we found no evidence of a reduced vasodilator influence of endogenous NO represented by the similar decreases in (regional) systemic, pulmonary, and coronary conductances produced byl-NNA (blocking both eNOS and iNOS) in resting swine, 2–3 wk after MI. Similar observations were made in rats in which up to 13 wk after a large MI (40% of the left ventricle), constrictor responses to the NO synthase inhibitorN ω-nitro-monomethyl-l-arginine (l-NMMA) were not different from those in normal rats in the total systemic bed (21) and resistance vessels in the renal, cutaneous, mesenteric, cerebral, and hindlimb microcirculation (6, 10, 13). Only in the coronary bed was vasoconstriction by l-NMMA less at 8 wk after MI compared with normal rats (10). In humans with chronic heart failure, studies on the contribution of NO to basal microvascular tone in the forearm, leg, or total systemic bed have yielded equivocal results with responses to NO synthase inhibition varying from decreased (25, 31, 32,41), maintained (38), to enhanced (11,23) vasoconstriction. It is possible that a maintained or increased production of NO that was observed in some studies was the result of increased expression of iNOS (12, 44) as part of a generalized inflammatory response in end-stage heart failure, which occurred in the presence of either a decreased (12, 49) or increased (20, 27) expression of eNOS. Interestingly, iNOS upregulation has been suggested to be of greater importance in idiopathic dilated cardiomyopathy than in ischemic cardiomyopathy (7), although this could not be confirmed by others (20, 26, 51). An explanation for the equivocal findings in clinical studies could lie in differences in severity and etiology of heart failure. However, most clinical studies used patient populations that consisted of mixed etiology and a range of severities, which makes interpretation and comparison of these studies difficult.

The inflammatory reaction that occurs after MI not only causes an increase in reactive oxygen species (22), but could also upregulate iNOS (12, 44), particularly in the myocardium (52). Consequently, in the present study, iNOS-mediated NO production could have increased and thereby have masked a possible reduction in eNOS activity. Because l-NNA, in the concentration used, blocks all three isoforms of NOS (5), we performed additional experiments in which we blocked iNOS with aminoguanidine in a dose of 20 mg/kg iv. This dose was based on studies in rabbits, in which a dose of 10 mg/kg aminoguanidine reduced cGMP levels and blocked iNOS activity by over 80% (59), and in rats, in which a dose of 15 mg/kg iv prevented and reversed endotoxic shock-induced systemic hypotension (62). More recently, a study was published that reported that 30 mg/kg, administered subcutaneously, completely abolished the induction of iNOS activation and cardioprotection by monophosphoryl lipid A in swine during a 24-h period (64). Hence, the dose of aminoguanidine of 20 mg/kg given intravenously 10 min before the exercise trial will produce effective blockade of iNOS. The observation that iNOS blockade had no effect on either basal systemic, pulmonary, or coronary vascular tone is in agreement with the hypothesis that NO production via iNOS does not contribute significantly to vascular tone in MI swine, and consequently that basal endothelial NO production is maintained early after MI despite the presence of LV dysfunction.

It could also be argued that endothelial NO production was already maximal so that ATP could not result in any further stimulation of NO production. However, this is unlikely because ATP did produce, albeit blunted, vasodilation in MI. Finally, ATP may activate eNOS through a different mechanism than shear stress; the latter is present in blood vessels in vivo, even under resting conditions. Indeed, ATP-induced activation of eNOS is mediated through a calcium-calmodulin-dependent pathway (42), whereas shear stress activates eNOS through Akt-mediated phosphorylation (8), resulting in calcium-independent activation of eNOS. Hence, it is possible that perturbations in the calcium homeostasis of endothelial cells contributed to the selective impairment of ATP-induced vasodilation in swine with MI.

NO Availability During Treadmill Exercise

A loss of endothelial NO-mediated vasodilator influence in heart failure could contribute to impaired tissue perfusion, when myocardial O2 demand is increased such as during exercise. The exercise-induced vasodilation in the systemic and pulmonary beds of the MI group during control was similar to the vasodilation observed in the normal group during exercise in the presence of l-NNA, which might suggest that a loss of NO production contributed to the blunted systemic and pulmonary vasodilation in MI. However, the effects of l-NNA were similar in MI and N swine, both at rest and during exercise. This observation implies that while agonist-induced receptor-mediated NO production was reduced, the exercise-induced NO production was maintained. Several possibilities might explain the difference between agonist- and exercise-induced NO production. First, Traverse et al. (54) have shown that the amount of NO produced after stimulation with an agonist is larger than the amount of NO produced with moderate exercise (60% increase in heart rate). Hence, the maximal capacity of NO production may be reduced, whereas the capacity of eNOS is sufficient to maintain basal and exercise-induced NO production. This explanation is unlikely because agonist-induced dilation is already affected at the lowest dose of ATP administered (which probably releases less NO than strenuous exercise), and higher doses of ATP still produce more dilation. Second, as stated above, ATP may activate eNOS through a different mechanism than shear stress (8, 42), so that calcium homeostasis of endothelial cells is selectively affected by MI.

Similar to the systemic bed as a whole, we also observed similar responses to l-NNA in regional vascular beds in MI and N swine. In patients with chronic heart failure [ejection fraction 22%, NYHA class II-III; (32)] inhibition of NO production byl-NMMA had no effect on forearm blood flow either at rest or during exercise. In contrast, l-NMMA decreased forearm blood flow and vascular conductance in normal subjects both at rest and during handgrip exercise. Interpretation of that study is difficult, however, because forearm blood flow was measured with venous occlusion plethysmography, which requires interruption of exercise and the measurements reflect blood flow during the early recovery from exercise. Indeed, Radegran and Saltin (46) have recently shown that inhibition of NO hastens the recovery of blood flow following exercise but does not reduce skeletal muscle blood flow during exercise. In rats with heart failure produced by a 6-wk-old MI, inhibition of NO production elicited smaller decreases in slow oxidative (i.e., red) skeletal muscle blood flow and vascular conductance, compared with normal rats (27a). In contrast, in the present study l-NNA had no effect on skeletal muscle blood flow in either normal or MI swine, whereasl-NNA-induced decrease in skeletal muscle vascular conductance was similar in MI and N swine in both red and white muscle groups.

Possible Causes for Increased Vascular Tone After MI

The present findings indicate that a reduction in NO production does not contribute to the increased tone of the systemic, pulmonary, and coronary resistance vessels at rest and the reduced dilatory response in these beds to exercise in 2–3 wk after MI. We (24) previously found that early after infarction, the neurohumoral status has changed so that circulating levels of endothelin, norepinephrine, epinephrine, and angiotensin II increased more during exercise in MI than in N groups, which may have contributed to the decreased dilator response to exercise. Also, other vasodilator systems, such as prostacyclin, exert a tonic vasodilator influence in pigs (1). Because prostacyclin has been shown to contribute to shear stress-induced vasodilation (36) and its production from endogenously administered arachidonic acid is reduced in the large coronary arteries of dogs with pacing-induced heart failure (63), a reduced prostacyclin production may also have contributed to the increased peripheral resistance at rest and during exercise.

In conclusion, although agonist-induced vasodilation is already blunted, basal and exercise-induced production of NO are maintained in swine with moderate LV dysfunction 2–3 wk after MI.

Acknowledgments

The authors gratefully acknowledge Rob H. van Bremen and RenéStubenitsky for technical assistance.

Footnotes

  • D. J. Duncker is the recipient of an Established Investigator stipend from the Netherlands Heart Foundation (2000D038). D. Merkus is supported by a postdoc stipend from The Netherlands Heart Foundation (2000D042).

  • Address for reprint requests and other correspondence: DJ Duncker, Experimental Cardiology, Thoraxcenter, Erasmus Univ. Rotterdam, PO Box 1738, 3000 DR Rotterdam, The Netherlands (E-mail:duncker{at}tch.fgg.eur.nl).

  • The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

  • First published February 21, 2002;10.1152/ajpheart.00834.2001

REFERENCES

  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
  11. 11.
  12. 12.
  13. 13.
  14. 14.
  15. 15.
  16. 16.
  17. 17.
  18. 18.
  19. 19.
  20. 20.
  21. 21.
  22. 22.
  23. 23.
  24. 24.
  25. 25.
  26. 26.
  27. 27.
  28. 27a.
  29. 28.
  30. 29.
  31. 30.
  32. 31.
  33. 32.
  34. 33.
  35. 34.
  36. 35.
  37. 36.
  38. 37.
  39. 38.
  40. 39.
  41. 40.
  42. 41.
  43. 42.
  44. 43.
  45. 44.
  46. 45.
  47. 46.
  48. 47.
  49. 48.
  50. 49.
  51. 50.
  52. 51.
  53. 52.
  54. 53.
  55. 54.
  56. 55.
  57. 56.
  58. 57.
  59. 58.
  60. 59.
  61. 60.
  62. 61.
  63. 62.
  64. 63.
  65. 64.
  66. 65.
View Abstract