Concurrent activity of the sympathetic and parasympathetic

Transcription

1 465 Parasympathetic Effects on in Vivo Rat Heart Can Be Regulated Through an ax-adrenergic Receptor Patricia A. McGrattan, Joan Heller Brown, and Oliver M. Brown A prejunctional mechanism involving an a,-adrenergic receptor may exert control on the release of acetylcholine from parasympathetic nerve endings in the heart. To test this hypothesis in vivo, rats were prepared for electrical stimulation of the vagus nerves. Blood pressures and heart rates were monitored, and the animals were treated with a-agonists and a-antagonists. The a,-selective agonist phenylephrine significantly attenuated vagally induced bradycardia in a dose-dependent fashion (ED S0 = 19 ij-g/kg). This is consistent with the hypothesis that there is a-adrenergic inhibition of ACh release. In contrast, the «2 -selective agonist, BHT-920, caused no change in heart rate during vagal stimulation. The effects of phenylephrine to raise heart rate and blood pressure during vagal stimulation were blocked by the a,-selective antagonist prazosin (ID S0 approximately 1 /xg/kg) but not by the a2-selective antagonists yohimbine and rauwolscine. This further supports an a x assignment to the prejunctional adrenergic receptor mechanism, which can regulate the release of acetylcholine from cardiac parasympathetic neurons. (Circulation Research 1987;60: ) Concurrent activity of the sympathetic and parasympathetic nervous systems results in complex interactions in the autonomic control of cardiac function.' Several mechanisms may be responsible for these interactions, including local control of the release of the autonomic neurotransmitters, acetylcholine (ACh), and norepinephrine. Evidence has been presented not only for the autoreceptor control of ACh and norepinephrine release from autonomic neurons but also for heteroreceptor control mechanisms. 2 Thus, vagal stimulation or applied ACh have been shown to decrease the release of norepinephrine from sympathetic nerve endings in dog hearts 3 and isolated rabbit atria, 4 respectively. Such a prejunctional mechanism for the control of ACh release by norepinephrine has also been suggested on the basis of in vitro studies of ACh release from isolated guinea pig right atrium 5 and isolated rat right atrium. 6 " 8 The present report supports the hypothesis that there is adrenergic control of ACh release with evidence from an in vivo preparation. Preliminary reports of some of these results have been presented. 910 From the Department of Pharmacology, State University of New York, Health Science Center, Syracuse, N.Y. (P.A.M. and O.M.B.) and Division of Pharmacology, Department of Medicine, University of California, San Diego, La Jolla, Calif. (J.H.B.) This work was supported by a Grant-in-Aid from the American Heart Association with funds contributed in pan by the Broome County, New York, Chapter of the American Heart Association. J.H.B. is an Established Investigator of the American Heart Association. Ms. McGrattan's present address: The Perkin-Elmer Corporation, Norwalk, CT Address for reprints: Oliver M. Brown, Pharmacology Department, SUNY/Health Science Center, 766 Irving Avenue, Syracuse, NY Received October 3, 1986; accepted December 31, Materials and Methods Animal Preparation Male Sprague-Dawley rats (Taconic Farms, Germantown, N.Y.) weighing g were used in all experiments. Rats were anesthetized with a combination of 300 mg/kg chloral hydrate (Merck and Co. Inc., West Point, Penn.) and 30 mg/kg sodium pentobarbital (Abbott Laboratories, North Chicago, 111.). Following a midline cervical incision, the trachea was intubated, and the right jugular vein and the left carotid artery were cannulated. Both the right and left vagus nerves were exposed and suspended over stainless steel bipolar electrodes and cut rostral to the electrodes to prevent afferent and efferent vagal reflexes. The electrodes were connected to a stimulus isolation unit (Grass, Quincy, Mass., model S1U5) driven by a Grass (model S88) stimulator. The vagi were given electrical pulse train stimulation: twin 5 volt pulses 25 milliseconds apart and 0.5 milliseconds in duration were administered at a pulse rate (4-5 Hz) chosen to entrain the heart to a rate that was approximately 65-70% of the control (prestimulation) rate. With pulse train stimulation in this range of frequencies, sinoatrial nodal rhythm synchronized with the periodic bursts of vagal activity," 12 allowing reproduction of the desired percent decrease in heart rate for each animal regardless of exact individual resting rate. Blood pressure was monitored from the carotid artery cannula through a pressure transducer (Statham model P23AC, Hato Rey, Puerto Rico) and recorded throughout the experiment on a Grass (model 7D) polygraph. A lead II surface electrocardiogram was recorded continuously on the Grass polygraph, and heart rate was determined from the interval between successive P waves.

2 466 Circulation Research Vol 60, No 4, April 1987 Drug Protocols The a,-selective agonist phenylephrine (PE) and the a 2 -selective agonist BHT-920 were tested for their effect on heart rate and blood pressure when administered alone or during bilateral vagus nerve stimulation. Vagal stimulation was initiated 1 minute before the i.v. injection of agonist. PE was administered at several dose levels from /ng/kg, and BHT-920 was given in doses of 10 and 100 fig/kg. Three or four trials of vagal stimulation alone, agonist alone, and the combined treatments were made on each animal with a washout (recovery) period of at least 5 minutes between treatments. The effects of PE on heart rate and blood pressure were challenged with the a,-selective antagonist prazosin and the a 2 -selective antagonists yohimbine and rauwolscine. The antagonists were administered 2 minutes before the initiation of vagal stimulation (thus, 3 minutes before the injection of PE). In some animals, bradycardia was produced by administration of the muscarinic agonist bethanechol (2.5 mg/kg, i.p.) in place of vagal stimulation. In another group of animals, blood pressure was increased by the temporary occlusion of the abdominal aorta in place of PE administration. Drugs BHT-920 was generously provided by Boehringer Ingelheim KG (Ingelheim am Rhein, Germany), prazosin HC1 was the kind gift of Pfizer Inc. (Groton, Conn.), and phenylephrine HC1 (PE) was provided by Winthrop Labs (New York, N.Y.). Yohimbine HC1 was purchased from Sigma (St. Louis, Mo.), rauwolscine HC1 from Atomergic Chemetals (Plainview, N.Y.), and bethanechol chloride from Merck Sharp & Dohme (West Point, Penn.). All drugs were dissolved in physiologic saline solution except prazosin HC1, which was dissolved in dimethylsulfoxide and diluted with saline to a final concentration of dimethylsulfoxide not greater than 10%; this vehicle solution had no effect on the measured properties. All drugs were administered by i.v. injection unless otherwise noted. Calculations and Statistics Heart rate and mean arterial blood pressure were determined during the following treatment periods: 1) control conditions, 2) agonist administration alone, 3) vagal stimulation alone, and 4) agonist administration during vagal stimulation (these periods were repeated in the case of pretreatment with antagonist). The values from 3 or 4 trials of each treatment period were averaged for each animal. The data reported here are the maximal treatment-induced changes from control values, which were seen at 10 seconds after agonist administration for blood pressure (mm Hg, mean ± SEM) and at 60 seconds for heart rate (bpm, mean ± SEM). Where appropriate, blood pressure and heart rate data from each treatment group were compared by two-way analysis of variance (ANOVA) and orthogonal comparisons. 13 Results The resting heart rate for the rats in these studies was 433 ± 5 beats per minute. This rate is elevated compared to that in control rats because cutting the cervical vagus nerves releases the heart from central parasympathetic control. Electric stimulation of the vagus nerves was adjusted to effect a 35-40% decrease in heart rate (a decrease of bpm). This bradycardia was maintained for 150 seconds, the entire period of vagal stimulation. A slight decrease in blood pressure accompanied the bradycardia. Figure 1 shows the effects of vagal stimulation on heart rate (lower panel) and blood pressure (upper panel). Also shown in Figure 1 are the heart rate and blood pressure responses to the a,-selective agonist PE. Note that since afferent and efferent vagal fibers were severed, reflex responses to vagal stimulation or PE injection were not observed for blood pressure or heart rate, respectively. When injected alone, PE (25 Mg/kg) increased blood pressure dramatically, from 80 to 148 mm Hg, but had no effect on heart rate. However, when PE was administered during vagal stimulation, not only did blood pressure increase, but there was also a significant attenuation of the vagally induced bradycardia (the maximum increase in heart rate being 47 bpm). This observation is consistent with the hypothesis that activation of a-receptors decreases the release of ACh from parasympathetic neurons. The maximum increase in blood pressure occurred at 10 seconds after PE administration, and the maximum change in heart! - so , a E -20- CL I < PE J,vs Time (seconds) FIGURE 1. Time course for the effects of phenylephrine (25 fig/kg) and electric stimulation of the vagus nerves (6 rats/ point). Upper panel: Change in blood pressure from resting conditions. Lower panel: Change in heart rate from resting conditions. Means ± SEM are plotted; the vertical bars indicate the SEM when these are greater than the size of the data point symbols. Arrows indicate the application of vagal stimulation (VS) or the injection ofpe (PE). o o Vagal stimulation only; A A PE administration only; and PE administration during vagal stimulation. 120

3 McGrattan et al Parasympathetic Effects Regulated Through «r Receptor rate was observed at 60 seconds. ANOVA treatment of the heart rate data from seconds in Figure 1 indicated that the heart rate resulting from PE administration during vagal stimulation was significantly higher than from vagal stimulation alone (p<0.05, F [1, 50] = 6.94). The effect of PE to increase blood pressure was also highly significant ( p < < , F [1,50]= 176). Both blood pressure and heart rate responses to PE were dose dependent, as shown in Figure 2. Curves were fitted by eye to the data points, which represent the changes in blood pressure (upper panel) and heart rate (lower panel) with increasing doses of PE during vagal stimulation. The blood pressure changes were plotted for 10 seconds post-pe injection, and the heart rate changes were plotted for 60 seconds post-pe the maximum response times for each (see Figure 1). The ED50 for the increase in blood pressure was 8 /xg/kg, and the ED50 for the increase in heart rate was 19 Mg/kgSeveral a-antagonists were used to characterize the heart rate and blood pressure effects of PE during stimulation of the vagus nerves. Figure 3 shows the results of treatment with prazosin, an a,-selective antagonist, and rauwolscine, an a2-selective antagonist. When injected alone, 50 tg/kg of prazosin had little effect on heart rate but decreased blood pressure as expected. Initiation ofvagal stimulation in the presence of prazosin still produced a bradycardia. However, injection of PE subsequent to prazosin resulted in no change in blood pressure or heart rate. Rauwolscine injected alone (50 /Ag/kg) had no effect on heart rate and caused little decrease in blood pressure. Vagal stimulation following rauwolscine administration resulted in the characteristic pronounced decrease in heart rate and a 80-, j>~ %\ o t o m , BO Dose of Phenylephrine 100 FIGURE 2. Dose-response curves for the effects ofpe on heart rate and blood pressure (4-6 animals/point). Upper panel: Change in blood pressure from vagal stimulation level. Lower panel: Change in heart rate from rate during vagal stimulation alone. (Means ± SEM, see Figure I legend). 467 J.ANT VSj PEJ, II so <" «- 1 " 20 H * 5 X Time (seconds) 240 FIGURE 3. Time course for the effects of PE (25 iiglkg) on heart rate and blood pressure in the presence of a-antagonists (5 animals/point). Upper panel: Change from resting blood pressure. Lower panel: Change from resting heart rate. (Means ± SEM, see Figure 1 legend.) Arrows indicate the application ofvagal stimulation (VS) or the injection of antagonist (ANT) or phenylephrine (PE). o O Vagal stimulation alone; O a PE administration during vagal stimulation; prazosin (50 figlkg, injected at ANT) with PE administration during vagal stimulation; rauwolscine (50 /xg/kg injected at ANT) with PE administration during vagal stimulation. small decrease in blood pressure. Injection of PE subsequent to rauwolscine resulted in an increase in blood pressure and a marked attenuation of the vagal bradycardia, as seen in the absence of antagonist. The observation that the ability of PE to raise blood pressure and to attenuate vagal bradycardia is blocked by prazosin, but not by rauwolscine, suggests that both responses are mediated through a,-adrenergic receptors. The heart rate data in Figure 3, following PE injection, were compared by ANOVA and orthogonal comparisons. The heart rate changes induced by PE with prazosin during vagal stimulation were not different from those induced by vagal stimulation alone (p 0.05). Further, PE with rauwolscine during vagal stimulation had the same effect on heart rate as did PE during vagal stimulation (/? 0.05). The dose-response relations for prazosin, rauwolscine, and another a2-selective antagonist, yohimbine, are compared in Figure 4 (curves were fitted to the data points by eye). At low doses ( /i.g/kg), yohimbine and rauwolscine had no pronounced effect on heart rate or blood pressure and did not block the effects of PE during vagal stimulation. At much higher doses (1-2 mg/kg), these antagonists produced decreases in both heart rate and blood pressure and decreased the effects of PE on blood pressure and vagal bradycardia, as shown in Figure 4. Prazosin (ID50 = 0.8 /u.g/kg), however, was about 1,000 times more potent than yohimbine (ID50 = 600 /Ag/kg) or

4 Circulation Research 46S 70-, finding supports the argument that a2-adrenoceptors are present in vascular smooth muscle where they mediate vasoconstriction.14 Analysis of the heart rate data in Figure 5 indicated no significant difference between treatment with BHT-920 during vagal stimulation and vagal stimulation alone ( p > ). In some animals,the same degree of bradycardia that resulted from electric stimulation of the vagus nerves was produced and sustained by administration of the muscarinic agonist bethanechol (2.5 mg/kg i.p.). When PE was injected during bethanechol induced bradycardia, no significant change in heart rate was seen (Table 1). In another experiment, the abdominal aorta was temporarily clamped to produce a sharp rise in blood pressure comparable to that resulting from PE administration. When this maneuver was performed during vagally induced bradycardia no further change in heart rate was seen (Table 1). 60 X I 50 o-j 60-, ~ 50 E a: Vol 60, No 4, April o -I Log Dose Antagonist (mole/kg) 6 FIGURE 4. Log dose-response curve for antagonist inhibition of the effects of PE on heart rate and blood pressure during vagal stimulation (4-6 animals/point). Upper panel: Change in blood pressure from vagal stimulation level. Lower panel: Change in heart rate from rate during vagal stimulation. (Means ± SEM, see Figure 1 legend.) O o Prazosin (0.5, 5.0, and50 ng/kg) (1.2 x 10 ~9, 1.2 x 10'8, and 1.2 x 10~7 mollkg); a ao yohimbine (0.1, 0.5, and 2.0 mg/kg) (2.6 x 10~7, 1.3 x 10~6 and 5.1 x 10~6 mollkg); A A rauwolscine(0.05,0.5,and2.5mg/kg)(1.3x10~7,1.3 x 10~6,and 6.5 x 10~6 mollkg). Discussion In the studies presented here, blood pressure and heart rate were monitored from rats that received electrical stimulation of vagus nerves, drug injections, or both. Pulse train electric stimulation of the cardiac ends of the transected right and left vagus nerves resulted in a maintained bradycardia (Figure 1). The injection of PE alone caused an increase in blood pressure with no change in heart rate (Figure 1). However, when PE was injected during vagal stimulation, not only was the expected increase in blood pressure observed but also an increase in heart rate (Figure 1). The increases in heart rate and blood pressure resulting from PE administration during vagal stimulation 20^ j VS BHT-920 ^ o rauwolscine (ID 50 = 1,300 ^ig/kg) in inhibiting the heart rate response to PE during vagal stimulation at a dose of 5 /Ag/kg. Prazosin (ID50 = 1. 2 /ng/kg) was also considerably more potent than yohimbine (ID50 = 220 /*.g/kg) or rauwolscine (ID50 = 1,500 /u.g/kg) in blocking the effect of PE on blood pressure. The role of an a,-adrenergic receptor in mediating the effect of PE on vagal bradycardia was further supported by studies with the a2-selective agonist BHT920. In contrast to the effect of PE, BHT-920 did not attenuate the bradycardia resulting from vagal stimulation. Figure 5 shows that when BHT-920 was injected alone, it increased blood pressure (upper panel) and decreased heart rate only slightly (lower panel). When administered during stimulation of the vagus nerves, BHT-920 caused a slight increase in blood pressure but did not increase heart rate. Statistical analysis of the blood pressure data in Figure 5 revealed that blood pressure was significantly higher with BHT-920 alone ( p < < ) or with BHT-920 treatment during vagal stimulation (p< 0.05) when compared to vagal stimulation alone. This -20- s < k. a Time (seconds) 120 FIGURE 5. Time course for the effects of BHT-920 (100 Hg/kg) and electrical stimulation of the vagus nerves (6 animals/point). Upper panel: Change in blood pressure from resting conditions. Lower panel: Change in heart rate from resting conditions. (Means ± SEM, see Figure 1 legend.) Arrows indicate the application ofvagal stimulation (VS) or the injection of BHT-920 (BHT-920). o o Vagal stimulation alone, A A BHT-920 administration alone; O D BHT-920 administration during vagal stimulation.

5 McGrattan et al Parasympathetic Effects Regulated Through a,-receptor 469 Table 1. Changes in Rat Blood Pressure and Heart Rate With Change in blood pressure (mm Hg)t Treatments (A followed by B) A B Vagal stimulation (A) with PE injection (B)t Vagal stimulation (A) with aortic clamp (B) Bethanechol injection (A) with PE injection (B) ± ± ±6.3 Various Treatments* Change in heart rate (bpm)t A B n -169±12-122± ± ± Protocol similar to that used in Figure 1. (Treatment A was ongoing for 1 minute (vagal stimulation) or 5 minutes (bethanechol injection, 2.5 mg/kg i.p.) before initiation of treatment B (injection of PE, 25 ig/kg i. v.; or clamping of abdominal aorta). Change in blood pressure and heart rate from pretreatment (control) value was recorded for treatment A prior to initiation of treatment B, and again at 10 seconds after treatment B for blood pressure and 60 seconds after treatment B for heart rate (means ± SEM). JThese data are from Figure 1. p<0.05 when A value compared to B value with Student's t test. both occurred over the same dosage range, as shown in Figure 2. To rule out the possibility that the increase in heart rate was secondary to the effect of PE to increase blood pressure, an additional experiment was performed in which the abdominal aorta was temporarily clamped to produce a sharp rise in blood pressure comparable to that resulting from PE administration. When this maneuver was performed during vagally mediated bradycardia, no change in heart rate was seen (Table 1). We also considered the possibility that the effect of PE was a rate related event, attributed to the decrease in heart rate resulting from vagal stimulation. To test this, the same degree of bradycardia that resulted from electric stimulation of the vagus nerves was produced by administration of the muscarinic agonist bethanechol. When PE was injected during bethanechol induced bradycardia, there was no significant change in heart rate (Table 1). Thus, the increase in heart rate caused by PE administration during vagal stimulation does not appear to be a pressure- or rate-related phenomenon. The bethanechol experiment also affords strong evidence for a prejunctional rather than a postjunctional site for the a mechanism described here. If the a-receptor stimulated increase in heart rate was mediated through a postjunctional adrenoceptor, PE should be equally effective at increasing heart rate when administered during vagally induced bradycardia (ACh as the muscarinic agonist) or during bethanechol induced bradycardia. Since PE had no effect on heart rate in the presence of bethanechol, a prejunctional mechanism, inhibition of ACh release is supported. That the prejunctional a-adrenergic receptor studied here is of the a,-subtype is suggested by the differential effects of the a,-selective agonist PE and the a 2 - selective agonist BHT-920 on heart rate during vagal stimulation (Figures 1 and 5). An a, assignment is further supported by studies with specific a-adrenoceptor antagonists. The effects of PE on heart rate and blood pressure during vagal stimulation were uniquely blocked by the a,-antagonist prazosin over a dosage range consistent with other characterizations of a- adrenoceptor binding or physiology.' 516 In contrast, blockade by the a 2 -antagonists yohimbine and rauwolscine was evident only in much higher doses (Figures 3 and 4), well above those generally needed for a 2 - blockade. The similarity between the antagonist doseresponse curves shown here for in vivo heart rate (Figure 4) and that of Wetzel et al 8 and McDonough et al 6 for in vitro ACh release is strong evidence that both parameters are regulated through the same (a,-adrenergic) receptor. The positive chronotropic and positive inotropic effects of catecholamines in the heart are largely mediated by /3-adrenergic receptors. There is some evidence that a-adrenergic receptors may contribute to the positive inotropic response to sympathetic activity, 1718 but a direct role for a-adrenergic receptors in mediating chronotropic responses is unlikely. l9 Our data are consistent with this latter point; no changes in heart rate were seen in our preparation with the administration of either the a-agonist PE (Figure 1) or with the administration of a-antagonists (Figure 3). A number of studies suggest that prejunctional a- adenergic receptors do serve an important role in controlling the release of norepinephrine from sympathetic nerve endings in the heart. 520 Starke 5 reported that several a-agonists decreased, while a-antagonists increased, the outflow of norepinephrine from isolated rabbit hearts subjected to electric sympathetic nerve stimulation. That this prejunctional a-adrenergic inhibitory feedback mechanism is subserved by a 2 - adrenergic receptors is supported by much evidence and has been reviewed. 21 However, several studies suggest that prejunctional receptors subtyped as a,- adrenergic may also play a role in this function. Prejunctional a-adrenoceptor activity was determined by Kobinger and Pichler 20 by monitoring changes in tachycardia elicited by electric stimulation of spinal sympathetic roots in the pithed rat preparation. This tachycardia was inhibited by several agonists (including the a -selective agonist methoxamine), an effect that was selectively blocked by prazosin. Quantitative comparison of their agonist and antagonist data suggests that most of the adrenoreceptors at this site are of the a 2 -subtype but supports the existence of a popula-

6 470 Circulation Research Vol 60, No 4, April 1987 tion of a,-prejunctional receptors on cardiac sympathetic neurons. Somewhat similar experiments in dog 22 and studies with isolated guinea pig atria 23 have also provided evidence that prejunctional a,-adrenoceptors, as well as a? -adrenoceptors, modulate norepinephrine release from cardiac sympathetic nerves. An analogous regulatory mechanism has been described in the parasympathetic system. Thus, prejunctional muscarinic receptor activation has been shown to inhibit ACh release from various preparations, for example chicken heart 24 and rat right atrium. 7 In addition to prejunctional autoreceptor regulation of neurotransmitter release, heteroreceptor inhibition of transmitter release has been proposed as a mechanism contributing to the complicated interactions between the sympathetic and parasympathetic innervations of several organs, including the heart.'" 3 Ldffelholz and Muscholl 425 demonstrated that the addition of ACh to isolated perfused rabbit atria reduced the quantity of norepinephrine released in response to sympathetic nerve stimulation. In an open-chest dog preparation, Levy and Blattberg 3 found that stimulation of the left cardiac sympathetic nerve produced an increase in ventricular contractile force and an increase in norepinephrine overflow into the coronary sinus. Concurrent vagal stimulation caused a reduction in contractile force and a decrease in norepinephrine overflow. Levy and Blattberg 13 concluded that ACh released from vagus nerves can, through a prejunctional mechanism, exert a potent inhibitory effect on the release of norepinephrine from sympathetic nerve endings. Several studies on gut preparations suggest that the converse of this heteroreceptor inhibitory mechanism, adrenergic inhibition of ACh release from parasympathetic neurons, also occurs. 26 Starke? first reported that a-agonists inhibit the vagally induced bradycardia in isolated rabbit heart by decreasing the release of ACh. This hypothesis was not supported by Lew and Angus, 27 who studied the electric responses of isolated guinea pig right atria to field stimulation. However, more recent reports from Wetzel and cowprkers 6 " 8 and from Langer's group 28 provide strong support for the a prejunctional control of ACh release. In the superfused rat right atrium, prelabelled with [ 3 H]-choline, both groups demonstrated that a-agonists inhibit the overflow of [ 3 H]-acetylcholine from the atrium in response to potassium stimulation or field stimulation. Unlike the prejunctional a-adrenergic receptors described in most systems, which are of the a 2 -subtype, the receptor described by Wetzel et al 8 appears to be of the a,- subtype. They found that the overflow of [ 3 H]-acetylcholine was inhibited by norepinephrine and by the a,-selective adrenergic agonist methoxamine, 8 as well as by phenylephrine. 6 The adrenergic inhibition of ACh release was blocked by the a,-selective adrenergic antagonists prazosin and YM but not by yohimbine or rauwolscine. The a, characterization of this receptor was very recently confirmed by Benkirane et al 28 in similar studies on the effects of several other selective a-agonists and antagonists on the release of [ 3 H]-acetylcholine from superfused rat atrial slices. Therefore, the study presented here provides physiologic evidence to support the hypothesis of Wetzel et al 6 " 8 that prejunctional a,-adrenergic receptors participate in the control of ACh release from parasympathetic nerve endings in the rat heart. The studies of Wetzel and coworkers showed that norepinephrine and epinephrine were very effective at inhibiting ACh release from isolated rat atria, suggesting that increased sympathetic tone might attenuate parasympathetic responses under certain physiologic conditions. In the in vivo preparation, we were unable to use norepinephrine or epinephrine because the dose of propranolol necessary to completely block the /3 effects of these catecholamines was toxic to our animals. Nonetheless, our series of observations made with PE support the proposed inhibition of ACh release from parasympathetic varicosities by activation of prejunctional a,- adrenergic receptors. 68 The mechanism demonstrated here may contribute to the interactions between the sympathetic and parasympathetic branches in the autonomic control of cardiac function. This mechanism may also underlie the lack of significant tachycardiac side effects in the treatment of hypertension with prazosin. Because of its a,-selectivity, prazosin would not disrupt activation of prejunctional a 2 -receptors by norepinephrine, and thus normal sympathetic feedback inhibition would occur. 29 At the same time, this selective antagonism would interfere with the proposed a,- adrenoceptor inhibition of ACh release. The combination of decreased norepinephrine release and increased ACh release may explain the low incidence of tachycardia in prazosin therapy. Acknowledgments Lynne A. Graziani provided excellent technical assistance. We thank Dr. Richard P. Oates (Preventive Medicine, State University of New York/Health Science Center) for his helpful counsel on the statistical treatment of our data. References 1. Levy MN: Cardiac sympathetic-parasympathetic interactions. FedProc 1984;43: Laduron PM: Presynaptic heteroreceptors in regulation of neuronal transmission. Biochem Pharmacol 1985;34: Levy MN, Blattberg B: Effect of vagus stimulation on the overflow of norepinephrine into the coronary sinus during cardiac sympathetic nerve stimulation in the dog. Circ Res 1976;38: Ldffelholz K, Muscholl E: A muscarinic inhibition of the noradrenaline release evoked by postganglionic sympathetic nerve stimulation. Naunyn-Schmiedebergs Arch Pharmacol 1969; 265: Starke K: Alpha sympathomimetic inhibition of adrenergic and cholinergic transmission in the rabbit heart. Naunyn-Schmiedebergs Arch Pharmacol 1972;274: McDonough PM, Wetzel GT, Brown JH: Further characterization of the presynaptic alpha-1 receptor modulating [ 3 H]ACh release from rat atria. J Pharmacol Exp Ther 1986;238: Wetzel GT, Brown JH: Presynaptic modulation of acetylcholine release from cardiac parasympathetic neurons. Am J Physiol 1985;248:H33-H39

7 McCrattan et al Parasympathetic Effects Regulated Through «i-receptor Wetzel GT, Goldstein D, Brown JH: Cardiac acetylcholine release can be regulated through an alpha r adrenergic receptor. Circ Res 1985;56: Brown OM, McGrattan PA: Parasympathetic control of the heart: Modification by alpha adrenergic agents. Circulation 1985;72(suppl 3): McGrattan PA, Brown OM: Sympathetic modulation of parasympathetic effects on the heart. Fed Proc 1985;44: Levy MN, Martin PJ, Iano T, Zieske H: Paradoxical effect of vagus nerve stimulation on heart rate in dogs. Circ Res 1969;25: Levy MN, Iano T, Zieske H: Effects of repetitive bursts of vagal activity on heart rate. Circ Res 1972;30: Steele RGD, Torrie JH: Principles and Procedures of Statistics. New York, McGraw-Hill Book Company, Langer SZ, Hicks PE: Alpha-adrenoreceptor subtypes in blood vessels: Physiology and pharmacology. J Cardiovas Pharmacol 1984;6:S547-S Kobinger W, Pichler L: a,- and a 2 -a drenoceptor subtypes: Selectivity of various agonists and relative distribution of receptors as determined in rats. Eur J Pharmacol 1981 ;73: Hoffman BB, Lefkowitz RJ: Agonist interactions with a- adrenergic receptors. J Cardiovas Pharmacol 1982;4:S14- S Govier WC, Mosal MC, Whittington P, Broom AH: Myocardial a and /3 adrenergic receptors as demonstrated by atrial functional-refractory period changes. J Pharmacol Exp Ther 1966; 154: Aass H, Skomedal T, Osnes JB: Demonstration of an a adrenoceptor-mediated inotropic effect of norepinephrine in rabbit papillary muscle. J Pharmacol Exp Ther 1983;266: Benfy BG: Function of myocardial a-adrenoceptors. Life Sci 1982;31: Kobinger W, Pichler L: Investigation into different types of post- and presynaptic a-adrenoceptors at cardiovascular sites in rats. Eur J Pharmacol 1980;65: Langer SZ: Presynaptic regulation of the release of catecholamines. Pharmacol Rev 1981;32: Uchida W, Kimura T, Satoh S: Presence of presynaptic inhibitory a,-adrenoceptors in the cardiac sympathetic nerves of the dog: Effects of prazosin and yohimbine on sympathetic neurotransmission to the heart. Eur J Pharmacol 1984;103: Ledda F, Mantelli L: Differences between the prejunctional effects of phenylephrine and clonidine in guinea-pig isolated atria. Br J Pharmacol 1984;81: Lindmar R, Loffelholz K, Weide W, Witzke J: Neuronal uptake of choline following release of acetylcholine in the perfused heart. J Pharmacol Exp Ther 1980;215: Loffelholz K, Muscholl E: Inhibition by a parasympathetic nerve stimulation of the release of the adrenergic transmitter. Naunyn-Schmiedebergs Arch Pharmacol 1970;267: Vizi ES: Presynaptic modulation of neurochemical transmission. Prog Neurobiol 1979; 12: Lew MJ, Angus JA: Clonidine and noradrenaline fail to inhibit vagal induced bradycardia: Evidence against prejunctional a- adrenoceptors on vagal varicosities in guinea pig right atria. Naunyn-Schmiedebergs Arch Pharmacol 1983;323: Benkirane S, Arabilla S, Langer SZ: Phenylethylamine-induced release of noradrenaline fails to stimulate a r adrenoceptors modulating [ 3 H]-acetylcholine release in rat atria, but activates a 2 -adrenoceptors modulating [ 3 H]-serotonin release in the hippocampus. Naunyn-Schmiedebergs Arch Pharmacol 1986;334: Rudd P, Blashke TF: Antihypertensive agents and the drug therapy of hypertension, in Gilman AG, Goodman LS, Rail TW, Murad F (eds): The Pharmacological Basis of Therapeutics. New York, Macmillan Publishing Co.,1985, pp KEY WORDS autonomic control parasympathetic control autonomic interactions adrenergic control