INTRODUCTION

Top Abstract Introduction Methods Results Discussion References

IN ADDITION to the well-known negative inotropism induced by ACh (26), a biphasic inotropic response to exogenous ACh has been sporadically reported in both amphibian and mammalian hearts (6, 24). The cellular and subcellular mechanisms governing this biphasic response are poorly understood. More recent studies showed that the ACh-stimulated isolated rabbit heart releases nitric oxide (NO) (2) and that myocytes themselves contain a constitutive NO synthase (30). Subsequently, studies with diverse kinds of mammalian cultured endothelial and myocardial cells suggested that ACh affects myocardial contractility via an NO signal transduction pathway. Thus NO appears to modulate intramyocardial cGMP levels and hence contractility (3, 19), thereby acting as a short-distance bidirectional messenger between endocardial endothelium (EE) cells and myocytes (5). However, because of a rapid loss of ACh receptors in some culture conditions, the results may not reflect in vivo conditions (7, 23). In the intact working heart the EE membrane is subjected to more distension and pressure than in cultured cell systems, and mechanical stresses induce the release of several factors that affect cardiac performance (31, 35). Thus studies of the response of the EE to ACh conducted in the whole working cardiac pump might provide more reliable information on the biphasic inotropic response sporadically found in amphibian and mammalian hearts.

In the coronary-supplied heart of homeotherms, not only does the coronary vascular endothelium constitute an almost contiguous stretch of tissue with the EE, but it seems to affect the contractile behavior of the adjacent myocardium in a manner similar to that of the EE so that the effects of both endothelial tissues appear additive and complementary (5, 29). Therefore, in whole heart studies of homeotherms it is difficult to dissect out an EE-mediated intracavitary autoregulation of cardiac performance via a receptor-mediated mechanism such as that involving an ACh-NO pathway.

The avascular heart of the frog, isolated and working at physiological loads, is an ideal system with which to explore the specific autocrine role of the EE as a source of NO without the confounding effects of the vascular endothelium (33). In fact, the frog EE is unique in being the only endothelial barrier between the superfusing blood and the myocardial microenvironment; only substances from the blood cross this barrier and interact with the myocardium (32).

The aim of this study was to analyze the role of the EE in the ACh-stimulated isolated and working frog heart. We demonstrate that in this heart preparation of Rana esculenta, an intact EE is necessary for the generation of the biphasic inotropic action of ACh and that this action involves an NO-cGMP signal transduction mechanism. Some preliminary results of this study have appeared in abstract form (11).

	METHODS

Top Abstract Introduction Methods Results Discussion References

Isolated and Perfused Working Heart Preparation

Measurements and Calculations

Ventricular stroke work, an index of systolic functionality, was calculated as (afterload  preload) × stroke volume/ventricle weight (mJ/g). The duration of the systolic phase and the height of peak pressure were calculated from recording traces.

Experimental Protocols

Basal conditions. In all experiments the diastolic afterload pressure was set at 3.92 kPa (40 cmH₂O), and the input pressure was regulated to obtain a cardiac output of ~110 ml · min¹ · kg wet body wt¹. These values are within the physiological range (33). The heart generated its own rhythm. Cardiac output, heart rate, and aortic pressure were measured simultaneously during the experiments. Hearts that did not stabilize within 10 min from the onset of perfusion were discarded. The basal condition parameters of cardiac performance were measured after a 20-min perfusion. These parameters are stable for >1 h (1, 33).

Functional impairment of EE. Ten to fifteen minutes after the onset of perfusion, when the heart was stabilized at basal conditions, 0.1 ml of Triton X-100 at a concentration of 0.05% was introduced through the aortic trunk, to avoid damage to the atrium, as follows. The inflow was closed, the afterload was simultaneously increased to ~7 kPa, and Triton X-100 was injected through a needle inserted in the output cannula so that the ventricle filled retrogradely because of a temporary incompetence of the valve. After three or four isovolumetric systoles, the inflow was opened and the outflow was adjusted to the control value, the perfusion being continued with the saline. Preliminary experiments with Evans blue dye showed that no backflow into the atrium occurs with this procedure. Parameters of cardiac performance were measured after 20 min of perfusion with the saline. In the hearts pretreated with detergent that were to be tested for the effects of ACh, this 20-min period was followed by a 20-min perfusion with the ACh-enriched saline, after which the performance parameters were measured.

Statistics

Drugs and Chemicals

	RESULTS

Top Abstract Introduction Methods Results Discussion References

Isolated Working Heart Preparations

8

View larger version (25K): [in this window] [in a new window]   Fig. 1.   Top panels: 2 typical pressure recording traces under control conditions (left) and after addition of isoproterenol (right). Dotted lines indicate systolic phase and peak pressure. Bottom panel: average time course of stroke volume (SV; ml/kg) and heart rate (HR; beats/min) in control conditions (n = 5 experiments). Before averaging (means ± SE), data were expressed as percentage of value at 5 min of perfusion, i.e., after stabilization. These baseline values were 2.3 ± 0.12 for SV and 51 ± 2.8 for HR.

ACh-Stimulated Preparations

Doses of ACh were from 5 × 10¹² to 10⁷ M. At higher concentrations of ACh all hearts were irreversibly blocked. Doses of 10¹⁰ M and higher caused a significant dose-dependent negative chronotropic effect (Fig. 2, top) paralleled by a remarkable decrease of cardiac output (data not shown). ACh exerted a biphasic effect on stroke volume, i.e., a gradual increase up to 10⁸ M, followed by a dramatic decrease at immediately higher concentrations (Fig. 2, bottom). A dose-dependent curve was generated with electrically paced preparations to analyze this effect free from the chronotropic action of the substance. In preliminary control experiments, the hemodynamic parameters in the paced hearts were stable up to 1 h (data not shown). ACh clearly exerted a direct biphasic action on myocardial performance; a maximal positive effect on stroke volume and stroke work (Fig. 3) occurred at 10⁸ M, followed by a sharp negative effect at 10⁷ M. The effect on stroke volume and stroke work was stable and was readily reversed by perfusing with ACh-free Ringer solution. The duration of systolic phase was increased at lower concentrations (significant only at 10⁸ M) and reduced at the higher dose (10⁷ M) of ACh (Table 1).

View larger version (27K): [in this window] [in a new window]   View larger version (26K): [in this window] [in a new window]   Fig. 2.   Dose-response curve for ACh on HR and SV in isolated, perfused nonpaced frog hearts. Data are means ± SE of 3 [5 × 1012 M ACh concentration ([ACh])], 4 (1011 M), 4 (1010 M), 3 (109 M), 6 (108 M), 3 (1.25 × 108 M), 4 (3.75 × 108 M), and 5 (107 M) experiments. Line graph, percent change; bars, actual mean values ± SE. Percent changes from control values from 5 × 1012 to 107 M were 2.02 ± 1.01, 2.53 ± 0.89, 10.31 ± 3.7, 5.79 ± 0.51, 9.58 ± 3.45, 9.16 ± 1.04, 11.67 ± 2.98, and 28.44 ± 4 for HR and +2.1 ± 1.05, +3.42 ± 1.32, +12.56 ± 4.75, +8 ± 2.66, +16.17 ± 2.66, +2.36 ± 5.18, 17.1 ± 6.25, and 31 ± 7.5 for SV. * P < 0.05 vs. control value.

View larger version (27K): [in this window] [in a new window]   Fig. 3.   Dose-response curve for ACh on SV and stroke work in isolated and perfused paced frog hearts. Data are means ± SE of 3 (1010 M [ACh]), 4 (109 M), 6 (108 M), and 5 (107 M) experiments. Line graph, percent change; bars, actual mean values ± SE. Percent changes from control values from 1010 to 107 M were +5.32 ± 0.13, +4.65 ± 1.27, +8.6 ± 1.4, and 31.7 ± 3.79 for SV and +3.6 ± 0.92, +3.56 ± 0.86, +6.58 ± 1.71, and 33 ± 4 for stroke work. * P < 0.05 vs. control value.

Both the positive and the negative effect of ACh on stroke volume were mediated by muscarinic receptors. This clearly emerges from experiments in which pretreatment with atropine (10⁶ M) or pirenzepine (10⁸ M) abolished the ACh-mediated changes of stroke volume (Fig. 4). Neither heart rate nor cardiac output was affected by atropine or pirenzepine alone (Table 2). It is noteworthy that the minimal concentrations of antagonists that abolished both these effects were very different, pirenzepine being more potent than atropine.

View larger version (36K): [in this window] [in a new window]   Fig. 4.   Effects of ACh (108 and 107 M) after pretreatment with atropine (106 M) and pirenzepine (108 M) and after pretreatment with propanolol (108 M) and phentolamine (108 M) on SV in isolated and perfused paced frog hearts. Data are means ± SE of 4 experiments in each group. Line graphs, percent change; bars, actual mean values ± SE. Although treatment with muscarinic antagonists blocked both positive and negative cholinergic inotropic effect, treatment with adrenergic antagonists (propanolol and phentolamine) abolished positive inotropic effects and reduced negative inotropism of ACh from 31.7 ± 8.79 to 7.54 ± 1.66 for propanolol and to 22 ± 7.29 for phentolamine. MR, medicated Ringer. Neither atropine nor pirenzepine alone had any effect on cardiac parameters. Both propanolol and phentolamine alone significantly decreased SV (9.4 ± 3.4 and 5.3 ± 1.7%, respectively) and SW (10.0 ± 3.4 and 6.1 ± 2.6%, respectively). * P < 0.05 vs. control value.

Negative inotropism by ACh and other muscarinic agonists is a well-recognized effect in the presence of cAMP-elevating agents. Pretreatment with - and -adrenergic antagonists (phentolamine and propanolol) blocked the positive and reduced the negative inotropic effect of ACh (Fig. 4). Interestingly, although pretreatment with the Ca²⁺ antagonist diltiazem abolished the negative effect of ACh (10⁷ M), it reverted the positive effect of ACh (10⁸ M) (Fig. 5). Diltiazem at a concentration of 10⁸ M did not affect cardiac parameters, whereas at a higher concentration (10⁷ M) it decreased both CO and HR.

View larger version (11K): [in this window] [in a new window]   Fig. 5.   Effects of ACh (108 and 107 M) after pretreatment with diltiazem (108 M) on SV in isolated and perfused paced frog hearts. Results are expressed as percent changes (%) from control value. Data are means ± SE of 4 experiments in each group. Treatment with calcium antagonist induced a negative inotropism at a low concentration (108 M) (6.52 ± 2.24) and abolished negative inotropic effect at a high concentration (107 M). Diltiazem (108 M) alone had no effect on cardiac parameters. * P < 0.05 vs. control value.

Effects of Dysfunctional Ventricular Endocardial Endothelium

After detergent pretreatment there was a significant reduction of heart rate and a mild increase of stroke volume, which persisted in the electrically paced hearts, whereas the endurance of the preparation remained unchanged for >1 h (33).

A dose-response curve of the effect of ACh on the detergent-treated hearts showed that both the positive and negative changes in stroke volume were abolished (Fig. 6). In contrast, the isoproterenol response in the detergent-treated hearts was unchanged (data not shown).

View larger version (17K): [in this window] [in a new window]   Fig. 6.   Dose-response curve for ACh on SV after pretreatment with Triton X-100 (0.05%) in isolated and perfused paced frog hearts. Data are means ± SE of 3 experiments in each group. Line graph, percent change; bars, actual mean values ± SE. There were no significant changes (paired Student's t-test: P = 1.0).

Involvement of NO Signaling System

4

5

4

6

5

6

5

View larger version (13K): [in this window] [in a new window]   Fig. 7.   Effects of ACh (108 and 107 M) after pretreatment with NG-nitro-L-arginine (L-NNA, 105 M), NG-nitro-L-arginine methyl ester (L-NAME, 104 M), NG-monomethyl-L-arginine (L-NMMA, 104 M), methylene blue (MB, 106 M), 1H-[1,2,4]oxadiazolo-[4,3-a]quinoxalin-1-one (ODQ, 105 M), and 8-bromoguanosine 3',5'-cyclic monophosphate (8-BrcGMP, 106 M) on SV and stroke work in isolated, perfused paced frog hearts. Results are expressed as percent change from control value. Data are means ± SE of 3-4 experiments in each group. Although treatment with L-NNA, L-NAME, L-NMMA, MB, and ODQ abolished cholinergic effects on both SV and stroke work, treatment with 8-BrcGMP induced a negative response at a low concentration (108 M) (9.7 ± 0.8% for SV and 9.79 ± 1.5% for stroke work) and did not modify negative effects at a high dose (107 M). L-NNA, L-NAME, L-NMMA, MB, and ODQ, individually tested, induced significant increases of both SV (+3.84 ± 1.35, + 3.18 ± 1.02, +6.47 ± 1.58, +6.44 ± 2.11, and +5.74 ± 1.91%, respectively) and SW (+3.84 ± 1.35, +3.3 ± 1.06, +6.78 ± 2.19, +6.86 ± 2.55, and +5.15 ± 2.18%, respectively). 8-BrcGMP alone decreased both SV (6.55 ± 2.94%) and SW (7.21 ± 2.78%). * P < 0.05 vs. control value.

View larger version (24K): [in this window] [in a new window]   Fig. 8.   Effects of ACh (108 and 107 M) after pretreatment with milrinone (105 M) on SV and stroke work in isolated, perfused paced frog hearts. Data are means ± SE of 4 experiments in each group. Line graph, percent change; bars, actual means values ± SE. Treatment with specific inhibitor of cGi-PDE abolished positive inotropic effect without affecting negative effect. * P < 0.05 vs. control value.

View larger version (36K): [in this window] [in a new window]   Fig. 9.   Dose-response curve for 3-morpholinosydnonimine (SIN-1) on SV and stroke work in isolated, perfused nonpaced frog hearts. Data are means ± SE of 3 [1010 M SIN-1 concentration ([SIN-1])], 4 (109 M), 4 (108 M), 5 (5 × 108 M), and 5 (107 M) experiments. Line graph, percent change; bars, actual mean values ± SE. Percent changes from control values from 1010 to 107 M were 4.09 ± 1.7, +4.97 ± 1.15, +8.17 ± 1.91, 12.17 ± 3.77, and 7.83 ± 3.21 for SV and 5.1 ± 2.21, +4.97 ± 1.66, +5.63 ± 2.34, 13.28 ± 3.33, and 7.45 ± 2.76 for stroke work. * P < 0.05 vs. control value.

	DISCUSSION

Top Abstract Introduction Methods Results Discussion References

The response to hemodynamic stimuli of the in vitro heart preparation used in this study mimics the response of the in vivo heart. When the isolated, perfused working heart preparation of R. esculenta, set up and standardized by us (1, 33), was perfused at constant pressure, it generated physiologically comparable values of output pressure, cardiac output, ventricle work, and power and showed the typical "hypodynamic state" after a relatively constant time from the onset of the perfusion. With this preparation, we demonstrate that exogenous ACh induces a biphasic inotropic response consequent to the muscarinic receptor-mediated stimulation of the EE. Exposure of each cardiac preparation to a single concentration of ACh only avoids the confounding effects of desensitization (36).

The parasympathetic innervation of the frog heart is well documented; the ventricle exhibits a higher cholinergic nerve density than mammalian hearts (20), and the cholinergic varicosities are densely distributed on the surface of the myocytes, without synaptic specialization, and in the space between them (13), whereas the muscarinic receptors are randomly distributed over the entire cellular surface of the myocytes and also of the nonmyocyte components (13). These features suggested that ACh "bathes" the muscle, so that some aspects of the vagal effects on the whole working heart are obtained by perfusing heart preparations with ACh, rather than by "focal" (e.g., iontophoretic) application (20).

In addition to its classic negative inotropism, exogenous ACh, at concentrations of 10⁵ M or higher, exerts a positive inotropic effect distinct from the well-known catecholamine-dependent positive inotropism of exogenous ACh (20) in mammalian (14, 6), avian (4), and amphibian (24) heart ventricle preparations. The mechanism of this positive inotropism was poorly understood until the recognition that stimulation of low-affinity myocardial muscarinic receptors produced a positive inotropic effect parallel to a rise in intracellular Na⁺ activity (17) and leading to an increase in free intracellular Ca²⁺ concentration (18). Present evidence obtained in myocardial preparations indicates that these cholinergic-mediated inotropic effects result from the activation by ACh of cardiac muscarinic receptors coupled with G proteins, which in turn modifies the activity of second messenger pathways, Ca²⁺ homeostasis, ionic channels, and contractile proteins (8).

The putative role of EE cells in mediating such cholinergic stimuli in the intact heart has been ignored. Whereas in isolated myocytes the concentrations of ACh or carbachol were higher than those associated with the negative inotropism of ACh, thereby supporting the conclusion that muscarinic agonists do not increase the force of contraction under physiological conditions (16), the opposite is true in our whole cardiac preparation. This illustrates how the effects of the various mechanisms responsible for the inotropic action of ACh can vary according to the experimental design. For example, whereas only mRNA for the M₂ receptor has been detected in cultured endothelial cells, mRNA for M₁-, M₂-, and M₃-receptor subtypes have been identified in freshly isolated endothelial cells (34). A critical reevaluation of previous studies using multicellular or isolated cardiomyocytes, in which the EE might have been mechanically impaired, is clearly needed.

Because of the relatively long exposure of the perfused cardiac preparation to ACh under our experimental conditions, the drug might act via various second-messenger signaling systems linked to the different muscarinic receptor subtypes, which, in addition to the predominant myocardial subtype M₂, include those of the EE, the first barrier involved in luminal stimulus-secretion coupling. Although atropine blockade cannot distinguish among muscarinic receptor subtypes, low concentrations of pirenzepine, an "M₁-selective" muscarinic ACh receptor (mAChR) antagonist, block only M₁ sites (15). Therefore, the finding that pirenzepine completely inhibited both the positive and the negative effects of ACh at concentrations two orders of magnitude lower than those of atropine indicates that M₁-type muscarinic receptors play an important role in the transduction of cholinergic signals in the EE of the frog heart ventricle.

In the isolated ventricular myocytes of R. esculenta, atropine and other "M₂-selective" antagonists exert an intrinsic negative effect on mAChR, i.e., they reduce the interaction between the receptor and the G (G_i and G_k) proteins, and more importantly, they stimulate the L-type calcium current previously stimulated by isoprenaline and inhibit the mAChR-activated K⁺ current (12). Consequently, we tested the effects on our system of different periods of perfusion of atropine or of pirenzepine. As shown in Table 2, neither of these antagonists affected the basal hemodynamic parameters of the preparations.

ACh, via M₂ muscarinic receptors coupled to G proteins, directly activates I_K,ACh (the pathway probably responsible for the basal increase in K⁺ current elicited by ACh) (21). However, in a variety of cell tissues, ACh also acts via M₁- and M₃-receptor subtypes coupled to G proteins to stimulate phospholipase C and the hydrolysis of phosphatidylinositol (PI) (28). The hydrolysis of PI elicits the production of second messengers, which may modulate channel function. Very recently, molecular cloning and ectopic expression of muscarinic receptor subtypes have provided further evidence that although M₂ and M₄ subtypes are principally coupled to adenylate cyclase inhibition, M₁-, M₃-, and M₅-receptor subtypes are functionally coupled to mobilization of intracellular Ca²⁺ (8, 15). It is thus of interest that the positive inotropism of ACh is reversed into negative inotropism when the frog heart is pretreated with the calcium antagonist diltiazem.

In studies on intact ventricular preparations, the negative inotropic effects of muscarinic agonists were most evident under conditions of elevated cAMP (3, 22, 26, 27). The experiments with - and -adrenergic antagonists revealed that adrenergic (intrinsic) tone is involved in the effects exerted by ACh. In fact, pretreatment with the -blocker propanolol and the -blocker phentolamine abolished the positive and reduced the negative cholinergic inotropism.

Both cholinergic antagonist pretreatment and functional impairment of the ventricular EE by Triton X-100 abolished the biphasic inotropic response of ACh. Under our experimental conditions, 0.1 ml of 0.05% Triton X-100 was injected in the aortic trunk to perfuse the ventricle, but not the atrium, through the temporarily incompetent valve. This concentration does not cause structural or ultrastructural changes in EE cells (33). It is noteworthy that when only the ventricular luminal surface was treated with detergent, pacemaker activity and atrial myocardial mechanical and endothelial secretory performance remained intact (33). This mild treatment with Triton X-100 blocked cholinergic stimulation as effectively as did antagonist pretreatment, a result that illustrates the high sensitivity of muscarinic receptors to their microenvironment, which is also well documented by the loss of antagonist selectivity after detergent solubilization (15). Therefore, the functional integrity of the EE appears to be a prerequisite for the transduction of blood-borne cholinergic signals to the myocardium.

We also demonstrate that cholinergic-mediated stimulation of the EE involves an NO signaling system by which EE cells are triggered to release NO or a nitroso compound that in turn might affect guanylate cyclase activity. In fact, pretreatment with either specific NOS inhibitors (L-NAME, L-NNA, and L-NMMA) or aspecific (methylene blue) or specific (ODQ) inhibitors of guanylate cyclase abolishes the biphasic inotropic action of ACh, as does cholinergic blockade.

There is controversy as to the inotropic effects of NO donors. Negative inotropism of sodium nitroprusside has been observed in some cardiac preparations but not by others (22). To our knowledge, ours is the only study to show the biphasic inotropism of SIN-1 on an intact cardiac preparation. The finding that SIN-1 exerts an inotropic biphasic dose-response curve identical with that induced by ACh coincides with the hypothesis that both these cGMP-elevating agents (22, 27) act via an NO-dependent mechanism.

Taken together, these results indicate that the EE of the frog heart ventricle is an important cellular source of the NO signal under basal conditions (33), but when stimulated it exerts a paracrine effect on the subjacent myocytes, possibly through a modulation of intracellular messengers such as cGMP. This possibility is supported by the finding that with 8-BrcGMP preincubation the negative inotropic response occurred at lower doses of ACh (i.e., 10⁸ M).

Such functional NO-mediated responses could result from the integrated activation of a variety of signal transduction pathways, involving, for example, a rise in intracellular calcium, stimulation of guanylate cyclase, and increase of cGMP levels, with in turn feedback modulation of intracellular calcium and cAMP levels via the cGMP-dependent cAMP phosphodiesterases. In isolated rat ventricular myocytes, both muscarinic cholinergic and -adrenergic stimulations are mediated, at least in part, by NO (3). In their study on the isolated ventricular myocytes of the frog (R. esculenta), Méry et al. (27) demonstrated a biphasic transsarcolemmal calcium current (I_Ca) response to NO donors, which was excitatory or inhibitory depending on the nanomolar or micromolar ranges of concentration of the NO donor, respectively. Both of these stimulatory and inhibitory effects appeared to be mediated by NO and by the consequent accumulation of cGMP not only via the "soluble" NO-sensitive guanylyl cyclase but possibly via the membrane-bound isoform of the enzyme. In the light of these results and previous data on a cG_i-PDE or PDE3 (cGMP-inhibited cAMP phosphodiesterase) functionally coupled to the L-type calcium channel in the frog heart (9, 25), Méry et al. (27) suggested that guanylyl cyclase and cGMP can play a role in the fine tuning of cardiac cAMP concentration by positive and negative controls via inhibition of the cG_i-PDE and stimulation of the cG_s-PDE, respectively. In agreement with the aforementioned data, our finding that milrinone, a specific inhibitor of PDE3, blocks the positive inotropism of ACh without affecting the negative one suggests a role for cG_i-PDE in our system.

In conclusion, our results document in an isolated working frog heart preparation that a functionally intact EE is necessary to activate the signal transduction pathway interposed between the blood-borne cholinergic stimuli acting on the muscarinic receptors of the luminal side of the endothelial cells and the contractile machinery of the subjacent ventricular myocytes. In fact, the selective damage of the ventricular EE with Triton X-100 completely abolished the biphasic cholinergic response in the same way that atropine and pirenzepine block muscarinic receptors. This pathway includes an NO signaling system present in the EE cells that is involved in both the positive and negative inotropic responses to ACh.