INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

SYMPATHETIC AND PARASYMPATHETIC divisions of the autonomic nervous system play antagonistic roles in control of cardiac activity (29). Stimulation of the sympathetic nervous system increases heart rate (sinoatrial node automaticity), atrioventricular conduction, and ventricular contractility, and stimulation of the parasympathetic nervous system reduces heart rate and atrioventricular conduction. While stimulation of the parasympathetic (vagus) nervous system has minimal effects on ventricular contractility under basal conditions, it reduces contractility during sympathetic stimulation or administration of -adrenergic agonists. This adrenergic-cholinergic interaction was first suggested in the mid-1960s by Hollenberg et al. (21), who observed that intracoronary infusions of the parasympathetic neurotransmitter ACh in dogs had minimal effects on myocardial contractility except during sympathetic stimulation or infusions of norepinephrine, when they produced severe cardiac depression. This phenomenon has subsequently been confirmed in various animal models (19, 20, 26, 38) and in human subjects (25).

Traditionally, the ACh-induced reductions in myocardial contractility have been attributed to stimulation of muscarinic receptors on the cardiomyocytes, leading to a G protein-mediated decrease in adenylate cyclase activity and a resultant fall in cAMP levels (11, 13, 20). However, the ability of ACh to release nitric oxide (NO) from the coronary vascular endothelium (4) and in vitro studies demonstrating NO-induced cardiac depression (5) have provided a theoretical basis for a role for the NO-cGMP pathway in the negative inotropic effects of ACh. According to this hypothesis, ACh-induced stimulation of the muscarinic receptors causes release of NO, which diffuses to the underlying myocytes where it stimulates production of cGMP, whose negative inotropic action offsets the effect of cAMP. Despite intensive investigation, the role of the NO-cGMP pathway in the ACh-induced reductions in myocardial contractility remains unresolved (10, 12, 16, 18, 19, 30, 31, 37, 39).

The present study employed a canine model, permitting tight control of hemodynamic conditions, to investigate ACh-induced effects on myocardial contractility in vivo. In our initial studies, we wanted to identify the conditions necessary for these effects. As anticipated, ACh had a prominent negative inotropic effect only when it was infused in the presence of the -adrenergic receptor agonist dobutamine. We then evaluated the role of NO in this effect by attempting to attenuate the effect with the NO synthase inhibitor N^G-nitro-L-arginine methyl ester (L-NAME) and to mimic it with bradykinin, a nonmuscarinic releaser of NO from the coronary vascular endothelium (34), or sodium nitroprusside (SNP), an exogenous NO donor (32). We then assessed the effect of ACh on the increases in myocardial contractility caused by the cAMP-phosphodiesterase inhibitor amrinone (7). The latter studies were performed to gain insight as to where in the -adrenergic receptor-cAMP pathway ACh was producing an inhibitory influence on myocardial contractility. Finally, we evaluated the role of NO in physiological modulation of myocardial contractility by ACh. The ability of vagal stimulation to attenuate the myocardial contractile response to dobutamine was determined before and after L-NAME.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Canine Preparation

1

1

1

1

1

1

Polyethylene cannulas were inserted into the thoracic aorta via the right carotid artery for measuring arterial blood pressure, the left atrium for measuring left atrial pressure, the right femoral artery for collecting samples of arterial blood for gas analysis, and the right femoral vein for administration of heparin sodium (400 U/kg with supplementation) for anticoagulation and of supplementary anesthetic.

A micromanometer-tip pressure transducer (Millar Instruments; Houston, TX) was inserted into the left ventricle via the left atrium and mitral valve to measure left ventricular pressure. The maximum rate of rise of left ventricular systolic pressure (dP/dt_max) was obtained from the left ventricular pressure pulse with an electronic differentiator. The left ventricular pressure signal was used to drive a cardiotachometer. Arterial pressure, left atrial pressure, and coronary perfusion pressure were measured with Statham pressures transducers (model P23ID, Gould; Cleveland, OH). A continuous record of blood pressures, left ventricular dP/dt_max, heart rate, coronary blood flow, and segmental length were obtained on a physiological recorder (model 2800S, Gould).

The left anterior descending (LAD) coronary artery was isolated ~2 cm from its origin for cannulation. A thin-wall stainless steel cannula (2.5-mm inside diameter) was introduced into the isolated segment of the LAD, so that the artery could be perfused selectively via an extracorporeal system (6). Briefly, this system consisted of a reservoir (pressurized with compressed gas) that was supplied by a peristaltic pump with blood from the left femoral artery. The tubing connecting the reservoir to the LAD was equipped with the following: 1) a heat exchanger to maintain temperature of the arterial blood perfusate at 38°C, 2) an ultrasonic, transient-time flow transducer (Transonic Systems; Ithaca, NY) to measure coronary blood flow, 3) ports for collecting blood samples and for infusing drugs, and 4) a mixing chamber for drugs administered into the perfusion tubing. Coronary perfusion pressure was measured through a small-diameter tube positioned at the orifice of the perfusion cannula. To avoid hypovolemia in the experimental animal, lactated Ringer solution was infused intravenously during priming of the perfusion system.

At the termination of each experiment, Evans blue dye was injected into the LAD to identify its perfusion territory. After the heart was stopped with KCl, it was removed and trimmed. The dyed tissue was excised and weighed so that coronary blood flow could be expressed on a per-100-g basis. The LAD perfusion territory weighed 33 ± 1 g.

Experimental Measurements

Myocardial segment shortening. Changes in myocardial contractility in the LAD perfusion field were evaluated from measurements of segmental shortening (SS) obtained using the ultrasonic crystal technique (6). A pair of crystals was implanted to a subendocardial depth in the LAD perfusion field. This location was verified by segmental lengthening during the brief (<2 min) period of flow stoppage required for cannulation of the LAD. The crystals were oriented so that they were parallel with the anticipated direction of myocardial fibers in the left ventricular subendocardium (36). Changes in distance between the crystals were recorded from measurements of the ultrasonic transit time between the crystals (Triton Technology; San Diego, CA). The end-diastolic length (EDL) and end-systolic length (ESL) were identified by the beginning of the rapid increase in left ventricular pressure just before isovolumetric contraction and dP/dt_min, respectively. SS (in percent) was calculated from the formula SS = [(EDL  ESL)/EDL] × 100.

Experimental Protocols

Series 1: unpaced hearts. At least 45 min was allowed for recovery from surgical preparation before experimental runs were initiated. A total of 1-5 experimental trials were performed in each animal. Values for SS and coronary hemodynamic parameters were obtained 2-3 min after initiating a drug infusion (at a time that steady-state conditions prevailed). At least 15 min was allowed for recovery after each experimental trial. In our initial studies, the changes in SS were evaluated during intracoronary infusions of ACh alone (n = 10; 20 µg/min) and during CaCl₂ (n = 6) or dobutamine (n = 10) administered at 15 mg/min and 10 µg/min ic, respectively. We chose our dose for ACh because it was the highest that could be used without causing aortic hypotension secondary to recirculation into the systemic circulation (6), and those for dobutamine and CaCl₂ because they caused the maximal increases in SS possible without systemic hemodynamic effects (7-9). Baseline measurements were initially obtained with coronary perfusion pressure set at 80 mmHg. Coronary perfusion pressure was held constant during the intracoronary infusions of dobutamine and CaCl₂, which permitted the normal coronary hyperemia accompanying augmented contractile activity. Coronary blood flow was held constant during the intracoronary infusions of ACh (both in the absence and presence of dobutamine and CaCl₂) to avoid masking a potential negative inotropic effect by Gregg's phenomenon (15) and to avoid diluting the inotropic agents in the coronary blood supply. This was accomplished by reducing coronary perfusion pressure manually, as necessary to offset the vasodilating effect of ACh. An analogous protocol was used in subsequent studies in series 1. The effect of ACh on SS was evaluated during dobutamine after treatment with L-NAME (n = 6; 300 µg/min ic, for 15 min), and the effects of bradykinin (n = 7; 2.5 µg/min ic) and SNP (n = 9; 80 µg/min ic) were assessed during dobutamine. Additional studies were conducted to clarify the effects of bradykinin alone (n = 7) on SS. Our dose for L-NAME was adopted from previous studies (8), which showed that it produced a 70-80% attenuation of the ACh-induced increases in coronary blood flow without effects on systemic circulatory variables or on the increases in coronary blood flow caused by SNP. The doses for SNP and bradykinin were selected on the basis of preliminary studies indicating that they caused a decrease in coronary perfusion pressure, i.e., vascular relaxation, which was comparable to that during ACh. In our final studies in series 1, we assessed the effect of ACh during amrinone at 300 µg/min ic, which was sufficient to increase SS to an extent similar to that caused by dobutamine. All drugs infused into the LAD perfusion line were dissolved in isotonic saline to achieve concentrations that permitted intracoronary infusions at 1.0 ml/min. Preliminary studies demonstrated that infusions of the saline vehicle at this low rate had no effect on SS or coronary blood flow.

Series 2: paced hearts. Series 2 dogs were prepared as described above under series 1 with several additions. The cervical vagus nerves were identified through a midline incision, double ligated, and transected between the two ligatures. Shielded electrodes were applied to the cardiac segments of each vagus nerve. The electrodes were connected to an electronic stimulator with stimulator isolation units (Grass Instruments; Quincy, MA). Our plan was to stimulate the vagus nerves at frequencies that would alter heart rate and, in turn, ventricular function in animals with intact atrioventricular (AV) conduction. Therefore, in each dog in series 2, we produced a complete AV block and paced the heart. The technique described by Steiner and Kovalik (35) was used to produce the AV block. In brief, a right thoracotomy was performed and <1.0 ml of 40% formaldehyde was injected into the AV junction. After a complete AV block was confirmed with the use of the electrocardium, the right thoracotomy was closed, the dog was turned on its opposite side, and a left thoracotomy was performed. The heart was then instrumented as described above. Heart rate was maintained at 100 beats/min by pacing the ventricles through a bipolar electrode attached to the anterior surface of the right ventricle free wall.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Series 1: unpaced hearts. ACh had no effect on SS under baseline conditions or during CaCl₂, but it reduced SS during dobutamine sufficiently to return SS to the control level (Table 1). ACh also caused pronounced decreases in coronary perfusion pressure under all conditions, which, under constant-flow conditions, reflected proportional reductions in coronary vascular resistance (Table 1).

View larger version (87K): [in this window] [in a new window]   Fig. 1.   Original tracing showing effects of ACh in myocardium stimulated with dobutamine. At trace A, an intracoronary infusion of dobutamine was initiated, and at trace B an intracoronary infusion of ACh was added. At trace C, the intracoronary infusions were stopped. ACh reversed the increase in segmental shortening caused by dobutamine and caused a decrease in coronary perfusion pressure. dP/dt, rate of rise of left ventricular systolic pressure.

View larger version (14K): [in this window] [in a new window]   Fig. 2.   Normalized decreases in segmental shortening relative to the dobutamine-induced values caused by an intracoronary infusion of ACh or vagal stimulation (5 Hz) before and after NG-nitro-L-arginine methyl ester (L-NAME). L-NAME did not alter the ACh- or vagal-induced decreases in segmental shortening. *P < 0.05 vs. dobutamine-induced value. Values are means ± SE.

View larger version (89K): [in this window] [in a new window]   Fig. 3.   Original tracing showing effects of bradykinin in myocardium stimulated with dobutamine. At trace A, an intracoronary infusion of dobutamine was initiated, and at trace B an intracoronary infusion of bradykinin was added. At trace C, the intracoronary infusions were stopped. Bradykinin, in contrast to ACh, did not alter the increase in segmental shortening caused by dobutamine, although, like ACh, it caused a decrease in coronary perfusion pressure.

Series 2: paced hearts. Cardiac effects of vagal stimulation during dobutamine in series 2 were comparable to those of ACh in series 1 (Table 1 vs. Table 4; Fig. 2). Vagal stimulation caused reductions in both SS and coronary perfusion pressure; while the reductions in SS were unaffected by L-NAME, those in coronary perfusion pressure were attenuated. Vagal stimulation had no effect on systemic hemodynamic parameters (Table 4).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Critique of Methods

The ability of changes in SS to reflect changes in myocardial contractility is limited by variations in heart rate and in the loading conditions of the heart (2). The constant values for heart rate and in indices of afterload (arterial pressure) and preload (left atrial pressure) during the selective intracoronary infusions of drugs and the vagal stimulations suggested that this limitation was not a factor in the present study. The use of such infusions ensured that recovery times were short. This permitted the evaluation of multiple experimental conditions in the same preparation, which facilitated the intertreatment comparisons.

Infusion of crystalloid during priming of the extracorporeal perfusion system produced modest reductions in hematocrit. The animals were ventilated with room air enriched with oxygen to increase arterial PO₂ and thus minimize the concomitant decreases in arterial oxygen content.

Effects of ACh on Myocardial Contractility

It is well recognized that the sympathetic and parasympathetic divisions of the autonomic nervous system exert antagonistic influences on various aspects of cardiac performance (29). However, these opposing effects are not algebraically related; complicated interactions are evident. The ability of sympathetic stimulation to enhance the negative inotropic effects of vagal stimulation, termed "accentuated antagonism" (26), has been appreciated for many years (21). In the present study, intracoronary infusions of ACh and the -adrenergic receptor agonist dobutamine were used to simulate this effect.

Accentuated antagonism was demonstrated dramatically in the current study; intracoronary infusions of ACh had no effect on contractility in unstimulated myocardium, whereas they inhibited contractility markedly in dobutamine-stimulated myocardium. The failure of ACh to affect myocardial contractility during inotropic stimulation with CaCl₂ ruled out the possibility that the findings during dobutamine were simply the result of an augmented baseline contractile activity.

The subcellular pathway underlying the -adrenergic-induced augmentation in myocardial contractility has been described in detail. It involves stimulation of adenylate cyclase, which causes an increased conversion of ATP to cAMP within the cardiomyocytes, leading to phosphorylation of the voltage-sensitive Ca²⁺ channels and Ca²⁺ influx. It is known that muscarinic receptors are coupled to adenylate cyclase by an inhibitory G protein (11, 13). An activation of this G protein has been longed considered integral to the ACh-induced reductions in myocardial contractility during activation of the -adrenergic receptors (20). It has been suggested that an inhibition of norepinephrine release from the sympathetic nerve terminals may also contribute (27).

Previous studies (10, 12, 16, 18, 19, 30, 31, 37, 39), primarily performed in vitro, have considered a role for an additional mechanism, i.e., the NO-cGMP pathway, in cholinergic inhibition of myocardial contractility. The results of these studies have been inconsistent. For example, on behalf of a role for this pathway, George et al. (12) showed that ACh caused increases in cGMP levels that were associated with negative inotropic effects in isolated unstimulated rat hearts. Watanabe and Besch (39) showed that ACh-induced increases in cGMP levels in guinea pig ventricles produced negative inotropic effects when they were combined with a cAMP-increasing drug, such as isoproterenol. Han et al. (18) recently reported that targeted disruption of the constitutive endothelial NOS gene completely abolished the decreases in calcium channel current and contractile amplitude caused by the muscarinic agonist carbachol in isoproterenol-stimulated mouse ventricular myocytes. On the other hand, Groschner et al. (16) reported that ACh reduced contractility in isolated guinea pig atria, although it had no effect on cGMP concentration. Endo and Shimizu (10) demonstrated a failure of 8-bromo-cGMP to mimic the antagonistic action of carbachol on contractility of isolated canine right ventricular myocardium exposed to isoproterenol. MacDonell et al. (30) found that ACh and carbachol both inhibited the positive inotropic effect of isoproterenol in rat ventricular myocytes, while SNP had no effect, even though the three agents caused similar increases in cGMP concentration. Méry et al. (31) showed that neither L-arginine (the substrate of NO) nor two NOS inhibitors affected the ability of ACh to reduce contractility of isolated frog ventricular fibers stimulated with isoproterenol. Finally, in apparent direct contradiction to the work of Han et al. (18) alluded to above, Vandecasteele et al. (37) demonstrated that papillary muscles obtained from genetically engineered mice without endothelial constitutive NOS retained normal negative inotropic responses to carbachol during -adrenergic stimulation. These widely divergent findings are likely attributable to the peculiarities of the various in vitro models utilized, and they point out the difficulty in extrapolating findings from such models to the intact heart in vivo.

Our findings during vagal stimulation are in apparent conflict with those of Hare et al. (19), which demonstrated that NOS inhibition blunted the reduction in myocardial contractility caused by vagal stimulation during an intracoronary infusion of dobutamine in anesthetized dogs, thus implying a role for the NO-cGMP pathway. The cause for this discrepancy is uncertain, but it may be related to methodological differences, including the use of a different anesthetic (-chloralose) and a higher heart pacing rate (170-180 beats/min) by Hare et al. (19). Another factor may be that Hare et al. (19) allowed coronary blood flow to vary naturally. The increases in coronary blood flow during vagal stimulation in the study of Hare et al. (19) would have diluted the dobutamine infused into the coronary artery, thus potentially contributing to the observed reductions in contractility. This factor would have been attenuated after NOS inhibition. It is possible that Hare et al. (19) mistook an accentuated dobutamine-induced positive inotropic effect following NOS inhibition for a blunted negative inotropic effect via the NO-cGMP pathway. Because we held coronary blood flow constant during the vagal stimulations, varying coronary arterial levels for dobutamine did not complicate our findings.

An additional new finding from the present study was that ACh had a negative inotropic effect during administration of the cAMP phosphodiesterase inhibitor amrinone. This finding implies that ACh interferes with a cAMP-related mechanism downstream from adenylate cyclase. Previous studies conducted in in vitro cardiac models suggest possible targets for this effect. ACh may be stimulating cAMP phosphodiesterase (thus directly offsetting the influence of amrinone) (24), and/or it may be increasing phosphatase activity, which would attenuate phosphorylation of phospholamban (1, 28). Both these mechanisms would theoretically result in a return of Ca²⁺ influx and myocardial contractility towards preamrinone levels.

Like ACh (Table 1) and SNP in our previous study (6), bradykinin, when infused directly into the coronary artery, had no effect on contractility in unstimulated myocardium. These findings are in keeping with those from previous canine studies (22, 33) suggesting that physiologically relevant concentrations of bradykinin have no direct effect on myocardial contractility in vivo. Although Munch and Longhurst (33) demonstrated positive inotropic responses (accompanied by increases in coronary blood flow) during intracoronary injections of bradykinin in open-chest dogs, these responses were eliminated when flow was first maximally increased with the vasodilator adenosine. This suggested that the initial bradykinin-induced increases in myocardial contractility were not due to a direct effect on the myocardium but secondary to Gregg's phenomenon (15).

Recent studies (3) have shown that constitutive endothelial NOS is also expressed within the cardiomyocytes. We have no way of ascertaining whether our administrations of L-NAME blocked this potential pathway for increased tissue cGMP levels. However, because the amount of NOS in the cardiomyocytes is normally very small (compared to that in the coronary vascular endothelium), this factor should not have influenced our results substantively.

Taken together, our findings in unstimulated and inotropically stimulated myocardium, provide a compelling argument against NO as an important modulator of contractile function in the normal, in situ heart. The extensive coronary vasodilation observed during the administrations of ACh, bradykinin, and SNP, and stimulation of the vagus nerves suggests that these interventions caused significant increases in NO concentration within the coronary vascular smooth muscle. The absence of apparent NO-mediated decreases in myocardial contractility suggests a relative lack of responsiveness of the myocytes (compared to vascular smooth muscle) to NO, a restricted access of NO to the myocytes, or a combination of these factors. It is possible that myoglobin, a known scavenger of NO (23), acts as a barrier for movement of NO from the vascular endothelium to the surrounding parenchyma.

In summary, the present findings indicate that ACh has an negative inotropic effect in vivo, which requires a concurrent increase in tissue cAMP level and which is independent of NO. The increases in cAMP level need not be associated with activation of -adrenergic receptors or of adenylate cyclase. Our findings add to the accumulating evidence that NO, in physiologically relevant concentrations, is not an important modulator of ventricular contractile function in vivo.