Temporal Changes in the Sympathetic- Parasympathetic Interactions that Occur in the Perfused Canine Atrium

Transcription

1 835 Temporal Changes in the Sympathetic- Parasympathetic Interactions that Occur in the Perfused Canine Atrium Yasuyuki Furukawa and Matthew N. Levy From the Division of Investigative Medicine, The Mt. Sinai Medical Center and Case Western Reserve University, Cleveland, Ohio SUMMARY. In the isolated, blood-perfused, canine right atrium, stimulation of the intramural autonomic nerves evoked negative chronotropic and inotropic responses. The responses were not maintained at a constant level during tonic neural stimulation, but they tended to drift back toward their control levels. These time-dependent changes in the cardiac responses were more pronounced the higher the frequency of stimulation. After the j8-adrenergic receptors were blocked in half of the preparations, the negative cardiac responses to autonomic neural stimulation were more pronounced, but the time dependency of those responses was less. After the muscarinic receptors were blocked in the other half of the preparations, only positive responses to neural stimulation were observed. These responses faded significantly at a high stimulation frequency (30 Hz), but not at a lower frequency (5 Hz). The cardiac responses to combined autonomic neural stimulation were substantially more negative than the algebraic sum of the individual responses to sympathetic and parasympathetic stimulation. The extent of this interaction was most pronounced near the beginning of stimulation, but it became less pronounced as the stimulation progressed. Hence, the cardiac sympathetic-parasympathetic interactions change appreciably with time during a continuous train of autonomic neural stimulation. (Circ Res 55: , 1984) THE sympathetic and parasympathetic divisions of the autonomic nervous system exert antagonistic effects on the heart. Numerous studies have shown that these opposing influences are not algebraically additive, but complicated interactions exist (Levy, 1971; Levy and Martin, 1979). In previous studies, the effects of autonomic nerve stimulation on the cardiac responses have usually been examined over only a limited range of stimulation frequencies, and usually only the steady state responses have been analyzed. The inotropic and chronotropic responses to tonic sympathetic stimulation increase to a maximum value, and tend to remain at that value when the stimulation frequency is relatively low (Levy and Blattberg, 1976a). At higher stimulation frequencies, the responses tend to fade after the maximum value has been reached. The cardiac responses to tonic vagal stimulation tend to fade below the maximum value even when the stimulation frequencies are low, but the extent of fade does not vary appreciably with frequency within the low frequency range (Martin et al., 1982; Martin, 1983). The fade may be more pronounced at high frequencies, however (Jalife et al., 1980). It has not yet been determined whether such temporal changes in the cardiac effects of autonomic nervous stimulation influence the extent of the interaction between the two components of this system. The present study of isolated, blood-perfused, canine right atrium preparations was designed to determine the cardiac responses to separate and combined sympathetic and parasympathetic stimulation in order to analyze the temporal changes in the interactions between these two autonomic components. Methods Isolated atria were obtained from 14 donor dogs, and each preparation was perfused with arterial blood from a second, support dog. The details of this preparation have been described previously (Chiba et al., 1975a, 1975b). The support dogs, which weighed from 19 to 30 kg, were anesthetized with sodium pentobarbital (30 mg/kg, iv). Sodium heparin (500 USP units/kg) was administered intravenously to each dog at the beginning of the perfusion, and 200 USP units/kg were given each hour thereafter. The donor animals weighed kg, and they also were anesthetized with sodium pentobarbital. Heparin (200 USP units/kg, iv) was given, and the right atria were excised and immersed in ice-cold physiological saline. The wet weights of the isolated right atrial preparations varied from 9.5 to 15.5 g. In each preparation, the sinus node artery was cannulated via the right coronary artery, and it was perfused with blood conducted from the femoral artery of the support dog with the aid of a roller pump. A pneumatic resistance was placed in parallel with the perfusion system so that the perfusion pressure could be maintained constant at 100 mm Hg. The blood flow rate to the isolated atrium was 3-7 ml/min. The venous effluent from the preparation was led to a collecting funnel,

2 836 Circulation Research/Vol. 55, No. 6, December 1984 from which it was returned continuously to the support dog via an external jugular vein. The ventricular margin of the atrium was attached to a rigid stainless steel bar, and the preparation was placed in a glass container. The superior part of the atrium was connected to a force transducer (Grass FT03C) by silk thread. The arterial muscle was usually stretched to a resting tension of 2 g. The isometric tension and the maximum rate of tension development were recorded on a direct-writing Brush oscillograph (Mark 260). Two pairs of silver electrodes were brought into contact with the epicardial surface of the isolated atrium. The diameter of each electrode was 0.3 mm, and the distance between the electrodes in each pair was 1.5 mm. The first pair of electrodes, placed on the caval margin of the atrium, was used to stimulate the intramural nerves. The second pair of electrodes, placed on the atrial free wall, was used to record the electrogram. The atrial rate was derived from the atrial electrogram. Experimental Protocols The experiments were subdivided into two groups, each involving seven perfused atria. In group 1, the intramural autonomic nerves were stimulated before and after the administration of propranolol. In group 2, the nerves were stimulated before and after the administration of atropine. In all experiments, the intramural sympathetic and parasympathetic nerves were stimulated electrically (Grass S4) at frequencies of 5 and 30 Hz. To stimulate the intramural nerves, we arbitrarily set the duration of the electrical pulses at 1 msec, and adjusted the voltage so that it was just below the threshold for the automatic and myocardial fibers, but above the threshold for the intramural nerve fibers (Furukawa et al., 1980). The mean voltage was 5.3 V. Each train of stimulation was continued for 2 minutes, so that the responses would approach a steady state level. Sufficient recovery time (more than 4 minutes) was permitted after cessation of each stimulation train to allow control levels to be restored. In group 1, the cardiac responses to autonomic nerve stimulation at the two frequencies were first determined before autonomic blockade (period 1). Propranolol hydrochloride (1 mg/kg) was then given intravenously to the support dog. Propranolol did not reduce significantly the basal levels of heart rate or atrial contractile force. However, it did completely block the positive cardiac responses of the isolated atrium to sympathetic nerve stimulation; this is documented below. At least 10 minutes were allowed to elapse after the injection of propranolol before the nerves were again stimulated (period 2). The order of applying the two stimulation frequencies was randomized in each observation period of each experiment. The chronotropic and inotropic responses to intramural autonomic nerve stimulation were measured at 15, 30, 60, 90, and 120 seconds after the beginning of each train of stimulation. In group 2, the cardiac responses to stimulation of the intramural autonomic nerves were determined before and after the administration of atropine sulfate (300 /igag) to the support dog. This dose of atropine blocked the negative cardiac responses of the isolated atrium to parasympathetic nerve stimulation, as documented below. The atropine did not significantly affect the basal levels of heart rate or atrial contractile force. At least 10 minutes were allowed to elapse after the injection of atropine before the nerves were again stimulated. The nature of the responses to stimulation, and the alteration of those responses by atropine and propranolol, constituted the evidence that the intramural nerves were being stimulated effectively. The absence of prominent rhythm irregularities during stimulation prior to the administration of atropine or propranolol and the absence of any detectable effect of stimulation after both of these drugs had been given demonstrated that the stimuli were subthreshold for the automatic and myocardial fibers. Furthermore, atropine was given after completion of two group 1 type experiments, and propranolol was given after completion of two group 2 type experiments. Thus, at the end of these four experiments, both atropine and propranolol had been given. ulation of the intramural nerves then had no detectable affect. These results demonstrate the efficacy of the blocking drugs, and support the contention that the stimuli were below threshold for the automatic and myocardial fibers. The cardiac responses to intramural sympathetic and parasympathetic nerve stimulation in each group were analyzed by means of a four-way, mixed model analysis of variance (Sokal and Rohlf, 1969). The drug treatment, stimulation frequency, and elapsed time after the beginning of stimulation were considered to be fixed factors. The individual preparations were considered to comprise a random factor. Scheffe's test was used for comparisons of mean values (Scheffe, 1953). Results Representative Experiments Figure 1 displays the chronotropic and inotropic responses of two representative atrial preparations (A) (B) 1 Atropine FIGURE 1. The changes in the cardiac cycle length and right atrial contractile force elicited by a 2-minute train of stimulation (horizontal bars) of the intramural sympathetic and parasympathetic nerves in isolated, blood-perfused canine right atria. The stimulation characteristics were 4 V, 1 msec, and 5 Hz. Panel A shows the cardiac responses before and after propranolol in one isolated atrium, and panel B shows the cardiac responses before and after atropine in another isolated atrium.

3 Furukawa and Leuy/Autonomic Control of the Heart 837 to intramural autonomic nerve stimulation at a frequency of 5 Hz. In the absence of propranolol or atropine (left panels), the cycle length responses rapidly increased to a maximum value, and then they "faded" back during stimulation toward the prestimulation control levels. The atrial contractile force responses diminished rapidly to a minimum value, and then faded back toward (A) or beyond (B) the control values during stimulation. After cessation of stimulation, there was a transient overshoot of the chronotropic and inotropic responses. After propranolol administration (Fig. 1A, right panel), similar negative chronotropic and inotropic effects were induced by nerve stimulation, but the fade was less pronounced after propranolol than before. In the other preparation (Fig. IB), the chronotropic and inotropic responses to autonomic nerve stimulation became positive after atropine (right panel), and these responses faded only slightly during the train of nerve stimulation. Composite Data Group 1: Effects of Propranolol The negative chronotropic and inotropic responses to autonomic neural stimulation were significantly more pronounced after, than before, propranolol (Fig. 2). Both responses faded (P < 0.001) back toward their prestimulation levels during the 120-second train of stimulation. The extents of the fade were significantly less after propranolol treatment than before; i.e., the interactions between pro- CO C o Q. W <D n \j Atrial Rate rr Q O5 Hz ^ ^ % p ^ 0 5 Hz ' J fy*^\o HZ OABefore Propran. A After Propran. ii i i i Contraction a O5 Hz yp 30 Hz > Time (sec) FIGURE 2. The mean percent changes with time in atrial rate (left panel) and atrial contractile force (right panel) in response to intramural sympathetic and parasympathetic nerve stimulation at frequencies of 5 and 30 Hz, before (open symbols) and after (closed symbols) treatment with propranolol. The prestimulation control levels of atrial rate and contractile force before propranolol were 108 ± 4.6 (mean + SE> beats/min and 5.5 ± 0.57g, respectively, in seven perfused atria. Hz Hz c 20 o CO 0 Q. CO 0) C Atrial Rate Hz. o A Before Atro. A After Atro. HZ Contraction Time (sec) FIGURE 3. The mean percent changes with time in the atrial rate (left panel) and atrial contractile force (right panel) in response to intramural sympathetic and parasympathetic nerve stimulation at frequencies of 5 and 30 Hz, before (open symbols) and after (closed symbols) treatment with atropine. The prestimulation control levels of atrial rate and contractile force before atropine were 114 ± 6.0 (mean ± se) beats/min and 6.7 ± 0.47 g, respectively, in seven perfused atria. pranolol treatment and time during stimulation were highly significant (P < 0.001). This fading of the chronotropic response was more pronounced at the higher than at the lower stimulation frequency (P < 0.001). The effects of stimulation frequency on the inotropic response were less pronounced after propranolol than before (P = 0.05). However, the effects of stimulation frequency on the chronotropic response were not altered significantly by propranolol. Group 2: Effects of Atropine Before atropine, autonomic nerve stimulation evoked only negative chronotropic and inotropic responses (Fig. 3). Hence, before atropine, the inhibitory parasympathetic effects on heart rate and atrial contractility must have been preponderant over the facilitatory sympathetic effects. After atropine, however, autonomic nerve stimulation evoked only positive chronotropic and inotropic responses. The negative chronotropic and inotropic re-

4 838 sponses obtained before atropine were significantly greater at the higher than at the lower stimulation frequency (P < 0.05). Similarly, the positive chronotropic and inotropic responses that were elicited after atropine tended to be greater at the higher than at the lower stimulation frequency, but the difference for the inotropic response was not significant. The chronotropic and inotropic responses all faded significantly with time. The tendency to fade was significantly greater at the higher than at the lower stimulation frequency (Fig. 3). Also, the tendencies for the chronotropic and inotropic responses to fade were significantly less after than before atropine. Discussion Simultaneous stimulation of the intramural sympathetic and parasympathetic nerve fibers evoked negative chronotropic and inotropic responses of the isolated right atrium during the initial control period before autonomic blocking drugs had been given. Hence, the inhibitory effects of the parasympathetic stimulation were preponderant over the facilitatory effects of the concomitant sympathetic stimulation. These cardiac responses were not maintained at constant levels during the continuous trains of neural stimulation, but they tended to drift back toward their control levels (Figs. 1-3). These timedependent changes (fading) of the responses were more pronounced the higher the frequency of stimulation. Individual Responses to Sympathetic and Parasympathetic ulation After treatment with propranolol (Fig. 2), autonomic neural stimulation evoked greater reductions in heart rate and atrial contractile force than it did before propranolol. Also, the time-dependent changes in the cardiac responses were more pronounced when the stimulation frequency was increased. However, the time dependency was less pronounced after propranolol than before propranolol (Figs. 1 and 2). The negative cardiac responses to combined sympathetic and parasympathetic nerve stimulation were converted to positive responses by treatment with atropine (Figs. 1 and 3). These positive responses to neural stimulation tended to fade when the stimulation frequency was 30 Hz, but this tendency was much less prominent when the stimulation frequency was only 5 Hz. At the dosage levels used in these experiments, atropine and propranolol completely blocked the responses of the isolated atria to stimulation of the intramural parasympathetic and sympathetic nerve fibers, respectively. Furthermore, the electrical stimuli had no detectable effects on the isolated atria once both blocking drugs had been given. Hence, the responses evoked prior to the administration of both antagonists must have been ascribable entirely to the stimulation of intramural autonomic nerve fibers. Circulation Research/Vol. 55, No. 6, December 1984 The time-dependent changes in the cardiac responses may be ascribable to events occurring at prejunctional or postjunctional sites in the autonomic neuroeffector junctions in the heart. With respect to prejunctional mechanisms, a progressive reduction of neurotransmitter release during a continuous train of autonomic neural stimulation has been observed in a variety of tissue preparations. Time-dependent reductions in norepinephrine overflow have been detected in the heart during sympathetic nerve stimulation (Siegel et al., 1961; Hukovic and Muscholl, 1962; Yamaguchi et al., 1973; Levy and Blattberg, 1976a), particularly at high stimulation frequencies. Similar time-dependent changes in acetylcholine overflow from the isolated chicken heart during continuous trains of vagal stimulation have also been described (Kilbinger and Ldffelholz, 1976). The overflow of acetylcholine was maintained almost constant for 5 minutes in such preparations when the stimulation frequency was 3 Hz, but the overflow progressively diminished when the frequency was 20 Hz. With regard to postjunctional mechanisms, studies conducted with exogenous agonists have shown that fading of the cardiac responses may also be ascribable to "postjunctional" phenomena, such as desensitization of receptors and of certain post-receptor mechanisms (Galper et al., 1977; Gertjegerdes et al., 1979; Jalife et al., 1980; Chang et al., 1982; Hasuo et al., 1982). The relative importance of these various mechanisms remains to be established, however. Sympathetic-Parasympathetic Interactions If the sympathetic-parasympathetic interactions are assumed to be negligible, the cardiac responses (Rs P ) to simultaneous sympathetic and parasympathetic nerve stimulation would equal the algebraic sum of the individual responses (&> and R p ) to separate sympathetic and parasympathetic nerve stimulation; that is = R s R p. In the present study, the cardiac responses to combined autonomic neural stimulation were determined before and after treatment with autonomic blocking drugs in two groups of animals. The same experimental conditions prevailed in both groups during the initial observation period before drug treatment. Therefore, it is reasonable to expect that the cardiac responses to combined neural stimulation were similar in the two groups. Table 1 confirms that the responses during the first observation period were not significantly different in the two groups. In the above equation, therefore, we have allowed Rs P to represent the mean response of all the animals in both groups to combined autonomic neural stimulation at a given frequency of stimulation and at a given time after the beginning of stimulation. The responses to such combined stimulation were always

5 Furukawa and Levy/Autonomic Control of the Heart 839 TABLE 1 Analysis of Variance of the Temporal Changes in the ChronotTopic and Inotropic Responses to Autonomic Neural ulation at Two Frequencies* Group (G). freq. (S) Time (T) GxS G XT SxT G x SxT DF 1,12 1,12 4,12 1, 12 4,48 4,48 4,48 Chronotropism MS ,451 5, , F ratio P 0.01 <0.001 <0.001 MS 1,409 14,445 6, Inotropism F ratio P 0.05 < * In 14 isolated, perfused atrial preparations during the initial observation period before an autonomic blocking drug had been administered. DF = degree of freedom; MS = mean square; P = probability; = not significant. recorded during the first observation period, prior to drug administration. R 5 represents the mean response of the animals in group 2 to sympathetic stimulation alone (i.e., after atropine), at the given frequency of stimulation and at the given time after stimulation. Similarly, R p represents the mean response of the animals in group 1 to parasympathetic stimulation alone (i.e., after propranolol), and at the given frequency of stimulation.and at the given time after stimulation. Implicit in this formulation are the assumptions that atropine does not directly alter the responses to sympathetic stimulation and that propranolol does not directly alter the responses to parasympathetic stimulation. If the responses (R 5p ) to combined stimulation equaled the algebraic sum of the individual responses to sympathetic (R 5 ) and parasympathetic (R p ) stimulation, the difference between R sp and the sum of Rs + R p [i.e., the value of R 5p - (R s + R p )] would equal zero. By definition, a significant difference between R sp and (R s + R p ) would denote a substantial sympathetic-parasympathetic interaction. A difference greater than'zero (i.e., a positive response) would indicate that during combined autonomic neural stimulation, the sympathetic effects predominated over the parasympathetic effects. Conversely, a difference less than zero (i.e., a negative response) would signify that during combined autonomic stimulation, the parasympathetic effects predominated over the sympathetic effects. We calculated the differences between Rs P and (Rs + R p ) at various times after the beginning of stimulation, and plotted the data in Figure 4. The graphs show that, for the chronotropic and inotropic responses, the differences between Rs P and (Rs + R p ) were all negative. Hence, at both stimulation frequencies and at all times during stimulation, the parasympathetic influences predominated over the sympathetic influences. This finding of vagal predominance is consistent with previous investigations. In the dog heart in situ, the effects of parasympathetic stimulation predominated over those of sympathetic stimulation in the control of sinoatrial (SA) node pacemaker activity (Samaan, 1935; Levy + a i o. w O> o c Atrial Rate 5 Hz 30 Hz Time, (sec) FIGURE 4. The percentage differences between the cardiac responses to simultaneous stimulation of the intramural sympathetic and parasympathetic nerve fibers (R, p ) and the algebraical sum of the cardiac responses (R p + R,) to separate parasympathetic and sympathetic stimulation, respectively. The combined responses were determined before any autonomic blocking drugs were given. The separate responses were obtained after the administration of the propranolol and atropine, respectively.

6 840 and Zieske, 1969; Warner and Russell, 1969) and atrial contractile force (Stuesse et al., 1979). In isolated rat atria also, the effects of acetylcholine were predominant over those of norepinephrine on the SA node pacemaker activity (Grodner et al., 1970). The differences between Rs P and (Rs + R p ) were more pronounced at the higher than at the lower frequency of stimulation (Fig. 4). At comparable times and stimulation frequencies, the differences were greater for the inotropic responses than for the chonotropic responses. The differences between R sp and (Rs + R p ) became less pronounced with time after the beginning of stimulation. By linear regression analysis, the slopes of all of the curves in Figure 4 were substantially greater than zero, which indicates that these temporal changes were statistically significant (P < 0.01). The accentuated antagonism between the sympathetic and parasympathetic effects on the heart may be mediated at prejunctional as well as at postjunctional levels (Levy, 1971; Levy and Martin, 1979; Vanhoutte and Levy, 1980). At the prejunctional level, the acetylcholine released from vagal endings interacts with muscarinic receptors on nearby postganglionic sympathetic nerve terminals and inhibits the release of norepinephrine (Loffelholz and Muscholl, 1969, 1970; Levy and Blattberg, 1976b; Lavallee et al., 1978). At the postjunctional level, acetylcholine acts to attenuate the rise in the intracellular concentration of camp that is induced by sympathomimetic agents (Bailey et al., 1979; Watanabe et al., 1981; Watanabe, 1984). The cholinergic antagonism may also be mediated in part through a rise in the intracellular levels of cgmp (Watanabe et al., 1981; Watanabe, 1984). The accentuated antagonism was most pronounced early in the stimulation period, and then it gradually became less pronounced as the train of neural stimulation was continued (Fig. 4). These temporal changes probably depend in part on the progressive changes in the rate of neurotransmitter release from the autonomic nerve endings (Siegel et al., 1961; Hukovic and Muscholl, 1962; Yamaguchi et al., 1973; Kilbinger and Loffelholz, 1976; Levy and Blattberg, 1976a), as described above. A progressive decline in the release of acetylcholine during continuous parasympathetic stimulation, for example, would result in decreased concentrations of this transmitter in the cardiac tissues. This diminution in the tissue level of acetylcholine would tend to attenuate the sympathetic-parasympathetic interaction at prejunctional and postjunctional levels. The precise mechanisms responsible for the temporal changes in the autonomic interactions in the heart remain to be delineated, however. We wish to thank F. Walters and H. Zieske for skilled technical assistance. This work was supported by U.S. Public Health Service Grants HL and HL Circulation Research/Vol. 55, No. 6, December 1984 Dr. Furukawa was a Postdoctoral Research Fellow of the American Heart Association, Northeast Ohio Affiliate. His present address is Department of Pharmacology, Shinshu University, Matsumoto, Japan. Address for reprints: Matthew N. Levy, M.D., Division of Investigative Medicine, Mt. Sinai Medical Center, 1800 East 105th Street, Cleveland, Ohio Received June 14, 1983; received in revised form June 18, 1984; accepted for publication September 10, References Bailey JC, Watanabe AM, Besch HR Jr, Lathrop DA (1979) Acetylcholine antagonism of the electrophysiological effects of isoproterenol on canine cardiac Purkinje fibers. Circ Res 44: Chang HY, Kein RM, Kunos G (1982) Selective desensitization of cardiac beta adrenoceptors by prolonged in vivo infusion of catecholamines in rats. J Pharmacol Exp Ther 221: Chiba S, Kimura T, Hashimoto K (1975a) Muscarinic suppression of the nicotinic action of acetylcholine on the isolated, bloodperfused atrium of the dog. Naunyn Schmiedebergs Arch Pharmacol 289: Chiba S, Yabuuchi Y, Hashimoto K (1975b) Comparison of the effects of norepinephrine and acetylcholine between intraarterial and extravascular administration to the isolated, bloodperfused canine atrium. Jpn J Pharmacol 25: Furukawa Y, Kobayashi M, Chiba S (1980) A pharmacological analysis of chronotropic and inotropic responses to direct electrical stimulation of the isolated, blood-perfused canine atrium. Jpn Heart J 21: Galper JB, Klein W, Catterall WA (1977) Muscarinic acetylcholine receptors in developing chick heart. ] Biol Chem 252: Gerjegerdes W, Ravens U, Ziegler A (1979) Time course of carbachol-induced responses in guinea pig atria under the influence of ouabain, calcium, and rate of stimulation. J Cardiovasc Pharmacol 1: Grodner AS, Lahrtz HG, Pool PE, Braunwald E (1970) Neurotransmitter control of sinoatrial pacemaker frequency in isolated rat atria and in intact rabbits. Circ Res 27: Hasuo H, Tokimasa T, Koketsu K (1982) Desensitization of the muscarinic receptor controlling action potential of bullfrog atrial muscles. Jpn J Physiol 32: Hukovic S, Muscholl E (1962) Die Noradrenalin-Abgabe aus dem isolierten Kaninchenherzen bei sympathischer Nervenreizung und ihre pharmakologische Beeinflussung. Naunyn Schmiedebergs Arch Pharmacol 244: Jalife J, Hamilton AJ, Moe GK (1980) Desensitization of the chounergic receptor at the sinoatrial cell of the kitten. Am J Physiol 238: H439-H448 Kilbinger H, Loffelholz K (1976) The isolated perfused chicken heart as a tool for studying acetylcholine output in the absence of cholinesterase inhibition. J Neural Transmission 38: 9-14 Lavallee M, dechamplain J, Nadeau RA, Yamaguchi N (1978) Muscarinic inhibition of endogenous myocardial catecholamine liberation in the dog. Can J Physiol Pharmacol 56: Levy MN (1971) Sympathetic-parasympathetic interactions in the heart. Circ Res 29: Levy MN, Blattberg B (1976a) Progressive reduction in norepinephrine overflow during cardiac sympathetic nerve stimulation in the anaesthetized dog. Cardiovasc Res 10: Levy MN, Blattberg B (1976b) Effect of vagal stimulation on the overflow of norepinephrine into the coronary sinus during cardiac sympathetic nerve stimulation in the dog. Circ Res 38: Levy MN, Martin P (1979) Neural control of the heart. In Handbook of Physiology, sec 2, Cardiovascular System, vol I, Heart, edited by RM Berne, N Sperelakis. Washington, D.C., American Physiological Society, pp Levy MN, Zieske H (1969) Autonomic control of cardiac pacemaker activity and arrioventricular transmission. J Appl Physiol 27:

7 Furukawa and Lew//Autonomic Control of the Heart Loffelholz K, Muscholl E (1969) A muscarinic inhibition of the noradrenaline release evoked by postganglionic sympathetic nerve stimulation. Naunyn Schmiedebergs Arch Pharmacol 265: 1-15 Loffelholz K, Muscholl E (1970) Inhibition by parasympathetic nerve stimulation of the release of the adrenergic transmitter. Naunyn Schmiedebergs Arch Pharmacol 267: Martin P (1983) Secondary AV conduction responses during tonic vagal stimulation. Am ] Physiol 245: H584-H591 Martin P, Levy MN, Matsuda Y (1982) Fade of cardiac responses during tonic vagal stimulation. Am J Physiol 243: H219-H225 Samaan A (1935) The antagonistic cardiac nerves and heart rate. J Physiol (Lond) 83: Scheffe H (1953) A method for judging all contrasts in the analysis of variance. Biometrika 40: Siegel JH, Gilmore JP, Sarnoff SJ (1961) Myocardial extraction and production of catecholamines. Circ Res 9: Sokal RR, Rohlf PJ (1969) Biometry. San Francisco, W.H. Freeman Stuesse SL, Wallick DW, Levy MN (1979) Autonomic control of right atrial contractile strength in the dog. Am J Physiol 236: H860-H865 Vanhoutte PM, Levy MN (1980) Prejunctional cholinergic mod- 841 ulation of adrenergic neurotransmission in the cardiovascular system. Am J Physiol 238: H275-H281 Warner HR, Russell RO Jr (1969) Effect of combined sympathetic and vagal stimulation on heart rate in the dog. Circ Res 24: Watanabe AM (1984) Cellular mechanisms of muscarinic regulation of cardiac function. In Nervous Control of Cardiovascular Function, edited by WC Randall. New York, Oxford University Press, pp Watanabe AM, Lindemann JP, Jones LR, Besch HR Jr, Bailey JC (1981) Biochemical mechanisms mediating neural control of the heart. In Disturbances in Neurogenic Control of the Circulation, edited by FM Abboud, HA Fozzard, JP Gilmore, DJ Reis. Washington, D.C., American Physiological Society, pp Yamaguchi N, dechamplain J, Nadeau R (1973) Noradrenaline liberation from the dog heart in vivo. Can J Physiol Pharmacol 51: INDEX TERMS: Atropine Autonomic nervous system Chronotropic response Inotropic response Propranolol