- Email: [email protected]
The interaction between atrial natriuretic peptide and cardiac parasympathetic function
Journal of the Autonomic Nervous System, 42 (1993) 81-88
81
© 1993 Elsevier Science Publishers B.V. All rights reserved 0165-1838/93/$06.00 JANS 01348
The interaction between atrial natriuretic peptide and cardiac parasympathetic function D.J. Atchison and U. A c k e r m a n n Department of Physiology, University of Toronto, Toronto, Canada (Received 5 May 1992) (Revision received 24 August 1992) (Accepted 27 August 1992)
Key words: ANP; Heart rate regulation; avAdrenoceptor; Prazosin; Phenylephrine; Vagus nerve Abstract We have demonstrated previously that atrial natriuretic peptide (ANP) inhibits hypotension-induced reflex tachycardia via a parasympathetic mechanism. The present study further defines that parasympathetic mechanism. We tested the hypothesis that ANP, during vagus nerve stimulation, acts as a physiological antagonist to interfere with al-adrenoceptor modulation of efferent cardiac vagal action. Sprague Dawley rats were divided into five groups, each group receiving a different infusion. Infusates included one of vehicle (Ringer's solution; RS), an cq-adrenoceptor agonist (phenylephrine; PE), a combination of agonist and either a known al-adrenoceptor antagonist (prazosin; PE + PRZ) or the putative physiologic antagonist, ANP (PE + ANP). The fifth group received all three drugs, PE + PRZ + ANP. Under Inactin anesthesia (100 mg/kg i.p.), efferent autonomic input to the heart was surgically interrupted. Animals were also adrenalectomized to limit the effects of circulating catecholamines. We then monitored each group for the change in heart rate (AHR) in response to efferent vagus nerve stimulation at various frequencies (2 Hz, 5 Hz, 10 Hz). Infusion of PE significantly (P < 0.01 by ANOVA) attenuated the magnitude of AHR when compared to the RS group. This attenuation of vagally-induced bradycardia was eliminated by the addition of the al-adrenoceptor antagonist, prazosin (PE + PRZ group). The PE + ANP group responded with results similar to those of the PE + PRZ group. There was no difference between AHR responses of the PE + PRZ + ANP group and the PE + PRZ group. The results show that, in a setting of experimentally manipulated autonomic input to the heart, ANP produces a response to vagus nerve stimulation that is equivalent to that seen with the al-adrenoceptor antagonist prazosin. Furthermore, in the presence of prazosin, ANP had no additional chronotropic effects. These results are consistent with our hypothesis that ANP influences parasympathetic control of heart rate by modifying the consequences of al-adrenergic receptor activation.
Introduction
An earlier report from this laboratory determined that ANP enhances the negative chronotropic effect of cardiac parasympathetic (vagus)
Correspondence to: U. Ackermann, Department of Physiology, Medical Sciences Building, University of Toronto, Toronto, Canada M5S 1A8.
nerve stimulation, but has no effect either on cardiac sympathetic chronotropic actions or directly on sino-atrial (SA) nodal activity [3]. Our earlier observations are particularly intriguing because ANP receptors have yet to be demonstrated on cardiac autonomic nerve terminals. Although ANP receptors have been localized to the sympathetic stellate ganglion [11] and postsynaptically on atrial [2] and ventricular myocytes [22] our previous experiments demonstrated these
82 ANP receptors were not directly involved in the modulation of heart rate [3]. The purpose of the present study was to characterize further the interaction between ANP and cardiac efferent parasympathetic function. It has been documented that ANP inhibits sympathetic nerve activity [15,23] and attenuates the release of norepinephrine [7,20]. However, sympathetic nerve activity and the release of norepinephrine are not the sole modulators of heart rate. Among the chronotropic mechanisms, consideration must be granted, not only to the role of parasympathetic nerve activity, but also to the interaction between sympathetic and parasympathetic systems [17]. Any investigation into the regulation of heart rate should acknowledge and account for the possibility that an autonomic interaction may influence the results. In this whole-animal study we have used a cardiac-denervated preparation that allowed selective aj-adrenoceptor stimulation or blockade on vagal presynpatic terminals during experimental manipulation of vagus nerve activity. The preparation was chosen because three distinct observations suggested an hypothesis: (1) ANP is capable of directly inhibiting sympathetic nerve action [7,15]; (2) activation of presynaptic aladrenergic receptors can modulate acetylcholine release [26] and attenuate the negative chronotropic response during vagus nerve excitation [19,27]; (3) ANP and the specific al-adrenergic antagonist, prazosin, exhibit similarities in their cardiovascular effects [12]. Our results suggest that ANP interferes with the inhibitory action of presynaptic al-adrenoceptors and thereby enhances the negative chronotropic effect of cardiac efferent parasympathetic nerve activation. The interference is not likely to arise from ANP binding to the a~-adrenoceptor site, but is more likely to result from an as yet unspecified physiological antagonism.
5-sec-butyl-5-ethyl-2-thiobarbituric acid (Inactin; Byk) 100 m g / k g i.p.. If necessary, one supplemental dose (25-33% of initial dose) was given. When the animal was fully anesthetized, the following surgical procedures were performed: bilateral adrenalectomy, tracheotomy (PE-390 tubing heated and pulled to an appropriate diameter), bilateral vagotomy in the mid-neck region, and bilateral femoral vein catheterization (PE-50 tubing) for the infusion of drugs or vehicle. The left femoral artery was catheterized (PE-50 tubing) to monitor arterial blood pressure (Electromedics Inc. MS20 transducer). Heart rate was monitored using a cardiotachometer (Sensormedics type 9857B) triggered by arterial pressure pulses. A chart recording (Beckman dynograph Type RP) of blood pressure and heart rate was obtained during control and experimental periods. Temperature was maintained between 37-38°C using a warming pad. Animals were ventilated (Harvard Apparatus Rodent Respirator Model 680) with 100% 0 2 at a rate and tidal volume that maintained normal acid-base balance as determined by arterial blood pH. Via a median sternotomy we exposed, isolated and transected nerves leading to the heart from both the right and left stellate ganglia. We have demonstrated that this procedure effectively eliminated sympathetic input to the heart [3]. The removal of autonomic innervation by these procedures (vagotomy and sympathectomy) decentralized efferent cardiac nerve supply, creating, in effect, an in vivo denervated heart model. Efficacy of denervation was ascertained by absence of a change in heart rate during transient bilateral carotid artery occlusion (baroreflex activation). After completion of surgical interventions, a section of the right vagus nerve, 10-15 mm caudal to the transected end, was carefully dissected using glass dissecting rods. The nerve was gently impaled upon a bipolar stainless steel hook electrode.
Materials and Methods
Drugs
Surgical preparation Male Sprague Dawley rats (240-341 g; Harlan Sprague Dawley, Inc.) were anesthetized with
Ringer's solution (RS; vehicle) was obtained from Abbott Laboratories Limited(catalogue no. 1582), L-phenylephrine hydrochloride (PE) and prazosin hydrochloride (PRZ) from Sigma (cata-
83 logue nos. P-6126 and P-7791, respectively), and atrial natriuretic peptide (ANP; rat 28-amino acid) from Peninsula Laboratories (catalogue no. 9103). PE and A N P were dissolved in Ringer's solution. P R Z was dissolved initially in propylene glycol (Fisher Scientific; catalogue no. P-354) then in Ringer's solution to a final propylene glycol concentration of < 5% vol/vol. To demonstrate physiologic antagonism in our denervated preparation it was necessary to choose a dose of a l - a d r e n o c e p t o r agonist that would measurably diminish the chronotropic effect of vagus nerve stimulation at physiologically relevent frequencies. On that basis the phenylephrine dose was chosen such that its associated blood pressure increase was less than 20 mmHg, but its chronotropic effect was sufficient to diminish measurably the heart rate responses to vagal stimulation in the frequency range 0 - 1 0 Hz. These objectives were accomplished with a PE infusion rate of 2.0 t z g / k g min -1. A prazosin dose was chosen that would block completely the observed chronotropic effect of PE. This dose was 5 . 0 / x g / k g min-~ and the completeness of its blocking effectiveness was indicated by the observation that it prevented phenylephrine-induced increases in mean arterial pressure of up to 40 mmHg. A N P was infused at 0.28 / z g / k g min -~ because that dose produced hemodynamic effects comparable to those of prazosin (at 5 p . g / k g min -1) in intact, anesthetized animals. It was necessary to choose the dose on the basis of pilot experiments with fully innervated rats because vagotomy eliminates ANP-induced hypotension [25]. The cumulative infusion volume was 0.051 ml/min.
Protocol Animals were randomly assigned to one of five groups. Each group was distinguished by a different drug infusion or combination of drug infusions: RS (n = 9), PE (n = 8), PE + P R Z (n = 9), P E + A N P ( n = 8 ) or P E + P R Z + A N P (n=8). Thirty minutes were allowed, after the start of each infusion, for stabilization and equilibration. After this interval, we began with the collection of baseline data (heart rate and blood pressure in the absence of cardiac autonomic nerve activity)
followed by brief periods of vagus nerve stimulation. D a t a were collected for 2 min at each stimulation frequency. We observed the change in heart rate ( A H R ) in response to vagus nerve stimulation at randomized frequencies (2 Hz, 5 Hz, 10 Hz) that had previously been shown to produce heart rate changes within a physiological range [3]. Electrical nerve stimulation (Grass SD9 Stimulator) lasted 2 min, with a 5 min rest period between stimulations to allow heart rate to return to baseline. Pulse duration was 1 ms. The necessary stimulus voltage was determined for each animal prior to beginning an infusion and was set at the minimum voltage required to achieve a predetermined change in heart rate ( > 12, but < 20 beats per minute) at the lowest frequency (2 Hz). The range of voltage used was 1.4-3.3 volt, the mean voltage ( + S . E . M . ) w a s 2 . 2 + 0 . 1 V. Animals underwent only one 2 min period of nerve stimulation at each frequency. Heart rate data were collected at time 0 and continuously thereafter for 2 min. Group comparisons were based on steady-state response to stimulation; this has been shown to occur by the 90-s time point [3].
Analysis Data analysis focussed on the relationship between A H R and frequency of vagus nerve stimulation. The change in heart rate during stimulation was calculated by subtracting the heart rate obtained at time 0 from the rate measured at 90 s of stimulation. G r o u p differences were analyzed by an analysis of variance (ANOVA). An A N O V A first compared the PE group with the RS group to determine that PE, at the concentration used here, influenced A H R in the same direction as shown by Pardini et al. [21] and McGrattan et al. [19]. All other comparisons were made relative to the PE group (activated a~-receptor). The level of statistical significance was set at P < 0.05.
Results Baseline mean arterial pressure, as measured after stabilization following drug infusion, ( + S.E.M.) was 93 + 5 m m H g for the RS group,
84
85 + 4 m m H g for PE + ANP, 60_+ 3 m m H g for PE + PRZ, and 54 + 6 m m H g for PE + P R Z + ANP. The lower baseline arterial pressure in the groups receiving prazosin was a consequence of combined denervation and adrenalectomy and, therefore, unavoidable. A lesser dose of prazosin would not have blocked the phenylephrine effect on vagus nerve chronotropy. For the PE group, the mean arterial pressure rose to 101 _+ 9 m m H g within 10 min of beginning the infusion, then stabilized at 95_+ 9 m m H g by the end of the 30-min equilibration period. Baseline heart rate (absolute values; mean _+ S.E.M.) for each group was also obtained at the end of the 30-min equilibration period (Fig. 1). These values were not significantly different from one another ( P = 0.27 by ANOVA). Thus it is unlikely that the magnitude of A H R during parasympathetic activity in each of the groups was influenced by the initial heart rate. The / t H R in response to stimulation at the various frequencies and for each infusion group is summarized in Figs. 2-4.
400 HEART
375
(BPM)
350 r
325 [
30O
RS
PE
PE
PE
+
+
PE +
PRZ
ANP
PRZ +
ANP Fig. 1. Baseline heart rates (beats per min, bpm) at the end of a 30 min infusion-equilibration period (mean±S.E.M.). Values were obtained at t = 9 0 s of the baseline (0 Hz) data collection period in the absence of autonomic nervous input. An ANOVA including all groups showed differences were not statistically significant. RS = Ringer's solution; PE = phenylephrine; PRZ = prazosin; ANP = atrial natriuretic peptide.
10 0 CHANGE IN
-10
-20
HEART RATE (BPM)
-30 -40
RS
-50 -60
i
i
i
0
5
10
FREQUENCY OF VAGAL STIMULATION (HZ) Fig. 2. Average change in heart rate (±S.E.M.; bpm) in response to right vagus nerve stimulation at selected frequencies during the infusion of phenylephrine (PE; e) or Ringer's solution (RS; ©). P < 0.01 between groups by ANOVA.
Figure 2 illustrates that heart rate for the RS group was reduced decrementally at increasing frequencies of vagus nerve stimulation. We found that the bradycardic response to vagus nerve stimulation for the PE group was less than for the RS group (Fig. 2; P = 0.006 by ANOVA). The ability of PE to attenuate the magnitude of heart rate response to vagus nerve stimulation was independent of any interaction or interdependence between the infusion of PE and frequency of stimulation ( P = 0.74 by ANOVA). Phenylephrine-induced reduction in bradycardia at any given level of vagal stimulation was eliminated both in the presence of the known eel-antagonist prazosin (Fig. 3) as well as in the presence of A N P ( P = 0.0003 and P = 0.0001, respectively). In each comparison, a H R for the PE + P R Z and PE + ANP groups was greater than the A H R achieved in the PE group. The frequency/infusion interaction term provided by the A N O V A was not significant in both the PE + P R Z and PE + ANP groups ( P = 0.06 for each group). The magnitude of A H R during vagal stimulation responses observed in the PE + ANP and PE + P R Z groups, were remarkably similar (Fig. 4). Furthermore, A N O V A of the results obtained in the PE + P R Z + ANP group showed no significant difference when compared to the
85 10 0 CHANGE IN
-10 -20
HEART RATE (BPM)
-30 -40 "
~
~ PE + PRZ
-50 -60
I
I
I
0
§
10
FREQUENCY OF VAGAL STIMULATION (HZ)
Fig. 3. Average change in heart rate (+S.E.M.; bpm) in response to right vagus nerve stimulation during the infusion of phenylephrine (PE; e) or the simultaneous infusion of phenylephrine and prazosin (PE+PRZ; a). P < 0.01 between groups by ANOVA. 10 o
CHANGE IN
-10
.2o
|~,
HEART RATE (BPM)
-40
~ PE + PRZ + ANP J. ~PE + PRZ ±PE + ANP
-50 -50
0'
5'
I 'o
FREQUENCY OF VAGAL STIMULATION (HZ)
Fig. 4. Average change in heart rate (+ S.E.M.; bpm) resulting from right vagus nerve stimulation. Values were obtained during the infusions of phenylephrine and prazosin (PE+ PRZ; zx), phenylephrine and ANP (PE+ANP; v), or phenylephrine, prazosin and ANP ( P E + P R Z + A N P ; II). ANOVA found no significant differences among the groups.
PE + P R Z and PE + ANP groups ( P = 0.71; Fig. 4).
Discussion As a follow-up to previous work, the present study attempted to characterize more fully the ability of ANP to enhance the negative chronotropic response to vagus nerve stimulation. The
results presented here suggest that ANP acts as a physiological antagonist of a]-adrenoceptors in the parasympathetic regulation of heart rate. This antagonism most likely does not involve direct competition for the receptor binding site. Hence the criteria used to select drug dosages did not include dose-response curves. Instead, doses were chosen on the basis of their effect on the chronotropic action of vagus nerve stimulation. Confirmation of the involvement of aj-adrenoceptors in vagally mediated heart rate effects required that sufficient phenylephrine be given to cause a significant diminution of vagal bradycardia at the chosen stimulus frequencies. Similarly, sufficient prazosin needed to be given to block the phenylephrine effect. Many investigators have administered prazosin as a bolus or as a brief ( < 30 min) infusion. In our experimental preparation, prazosin, administered continuously, caused profound hypotension even at doses much lower than others [e.g. 21] have used. The need to maintain blood pressure near normal governed the upper limit of an acceptable prazosin dose. Its lower limit was set by the need to elicit observable antagonism to the chronotropic effects of PE. Prazosin at 5 ~ g / k g m i n - I lowers mean arterial blood pressure by about 15 mmHg ([4] and unpublished observation) and completely blocked the influence of PE in our experiments. We administered ANP at a dose that produced hemodynamic effects comparable to those of the selected prazosin dose. The surgical technique used here effectively interrupts cardiac efferent nerve supply from the central nervous system [3]. We added bilateral adrenalectomy in order to diminish effects arising from changes in humoral catecholamine levels. Though cardiac sympathectomy combined with adrenalectomy may not have removed all endogenous sources of catecholamines [5], the observation that combined administration of phenylephrine and prazosin (PE + PRZ; Fig. 3) yielded results similar to those in the control infusion group (RS; Fig. 2) confirmed our belief that unaccounted sources of catecholamines (e.g. norepinephrine released from systemic sympathetic terminals) or involvement of other adrenoceptors did not play a significant role in influencing zlHR.
86
Our success at manipulating a l - a d r e n o c e p t o r activity is indicated by the observation that combined infusion of phenylephrine and prazosin (PE + P R Z group; Fig. 3) prevented the attenuated chronotropic response to nerve stimulation seen in the group receiving phenylephrine alone (PE; Fig. 3), yielding heart rate changes that were comparable to those seen in animals receiving vehicle infusion alone (RS group). Sympathetic and parasympathetic nervous interactions within the heart have been localized to the presynaptic level, the parasympathetic nervous system modulating release of sympathetic neurotransmitters [16] and the sympathetic nervous system modulating the release of acetylcholine [19]. Although adrenergic receptors located presynaptically are considered to be predominantly ~2-adrenoceptors, there have been suggestions that the a,-adrenergic receptor subtype can also be located presynapticatly [6,14]. The reputed function of such presynaptic a 1adrenoceptors is inhibition of neurotransmitter release [18,26,27]. Presynaptic adrenergic inhibition of acetylcholine release may explain, in part, the interaction of sympathetic and parasympathetic nerve activity that has been demonstrated in the autonomic control of heart rate [1,17]. Furthermore, the documented effects of ANP to interfere with sympathetic action [7,15,24] together with its pronounced ability to augment cardiac parasympathetic effects [1,3,28] are consistent with an ANP action to interfere with presynaptic a~-adrenergic mechanisms that influence Ach release. The specific a~-adrenoceptor antagonist, prazosin [12], as well as ANP [1,13], cause hypotension that is not associated with a compensatory reflex tachycardia. The similarity of their effects on blood pressure and heart rate reinforces our speculation that the chronotropic effects of ANP may be mediated by interference with a~adrenoreceptor mechanisms, though we do not expect this interaction to occur directly with the adrenoceptor. The possibility that prazosin acted directly upon the SA node [8] is contradicted by the observation that baseline heart rate for the PE + P R Z group, in the absence of nerve stimulation, was not statistically different when corn-
pared to baseline heart rate in the control (RS) group (Fig. 1). Thus it is unlikely that our results are due to an effect of prazosin directly upon the SA node. Nor is it likely that our results are a direct consequence of ANP acting upon the SA node because we have previously found that ANP did not affect SA node activity in the absence of autonomic nerve input [3]. We expected that administration of ANP in combination with e~-blockade by prazosin would allow us to reveal separate sites of action for ANP and prazosin. If the observed effects of ANP did not arise from an influence on a~adrenoceptors, then we expected that administration of ANP in combination with prazosin would produce chronotropic results different from those already obtained during the separate infusion of each drug. This was not the case (Fig. 4). When P R Z and ANP were used in combination (PE + P R Z + ANP group) the magnitude of A H R was similar to that obtained during the infusion of either PE + P R Z or PE + ANP (Fig. 4). We believe these data are evidence for an inhibitory influence of ANP on c~radrenoceptors and suggests that both P R Z and ANP interfered with a~-adrenoceptor activation. The lack of an additive effect during the simultaneous administration of ANP and P R Z is consistent with a final common pathway of action of ANP and PRZ. Several investigators [18,19,26] have suggested that this site of inhibition is presynaptic. We demonstrated, in a former study, that ANP alone did not influence heart rate response to sympathetic nerve (stellate ganglion) stimulation, but did enhance the magnitude of heart rate response to vagus nerve stimulation [3]. The results of the present study indicate that ANP did not show effects additional to those which could be demonstrated via inhibition of presynaptic al-adrenoreceptors on vagal nerve endings. The suggestion that ANP influences o~-adrenoceptor activity in the autonomic control of heart rate is consistent with a study of ANP influence in the peripheral vasculature. Faber et al. [9] reported that, in denervated cremaster muscle, ANP had no effect on vessels pre-contracted with an ax-adrenergic agonist. Though no mechanism of action was proposed, they demon-
87
strated that ANP was effective in reversing the effects of al-agonist-induced vasoconstriction by phenylephrine. This demonstration of selective al-adrenoceptor-inhibitory effects of ANP in the peripheral vasculature supports our interpretation that al-adrenergic-dependent mechanisms are responsible for the chronotropic effects of ANP even though localization of the receptor and the nature of the mechanism are still undetermined. In conclusion, we have determined that ANP inhibits the adrenergic regulation of parasympathetic nerve activity in the autonomic control of heart rate. We speculate that ANP may act indirectly through ANP-receptor-activated G proteins. Moreover it has also been hypothesized that ANP and ANP-like peptides within the autonomic nervous system act as neurotransmitters [10]. Though the extent and significance of this role are still undetermined, the results we have obtained may be due to a neurotransmitter function of ANP within the autonomic nervous system at the level of synaptic terminals.
6
7
8
9
10
11
12
13
Acknowledgements The authors thank Dr. Evert Vos and CibaGeigy, Canada for financial support and Dr. Peter Pennefather for constructive criticism of the manuscript.
14
15
References 1 Ackermann, U., Khanna, J. and Irizawa, T., Atrial natriuretic factor alters autonomic interactions in the control of heart rate in conscious rats, Can. J. Physiol. Pharmacol., 66 (1988) 930-936. 2 Anand-Srivastava, M.B. and Cantin, M., Atrial natriuretic factor receptors are negatively coupled to adenylate cyclase in cultured atrial and ventricular cardiocytes, Biochem. Biophys. Res. Commun., 138 (1986) 427-436. 3 Atchison, D.J. and Ackermann, U., Influence of atrial natriuretic factor on autonomic control of heart rate. Am. J. Physiol., 258 (1990) H718-H723. 4 Cavero, I., Effects of prazosin on reflex changes in heart rate evoked by vasopressor and vasodepressor stimuli in conscious rabbits. J. Cardiovasc. Pharmacol., 4 (1982) S108-Sl13. 5 Darlington, D.N., Chew, G., Ha, T., Keil, L.C. and Dall-
16 17 18
19
20
man, M.F., Corticosterone, but not glucose, treatment enables fasted adrenalectomized rats to survive moderate hemorrhage, Endocrinology, 127 (1990) 766-772. Docherty, J.R., An investigation of presynaptic a-adrenoceptor subtypes in the pithed rat heart and in the rat isolated vas deferens, Br. J. Pharmacol., 82 (1984) 15-23. Drewett, J.G., Trachte, G.J. and Marchand G.R, Atrial natriuretic factor inhibits adrenergic and purinergic neurotransmission in the rabbit isolated vas deferens, J. Pharmacol. Exp. Ther., 248 (1989) 135-142. Dukes, I.D. and Vaughan Williams, E.M., Electrophysiological effects of a-adrenoceptor antagonists in rabbit sino-atrial node, cardiac Purkinje cells and papillary muscles, Br. J. Pharmacol., 83 (1984) 419-426. Faber, J.E., Gettes, D.R. and Gianturco, D.P., Microvascular effects of atrial natriuretic factor: interaction with a 1- and a2-adrenoceptors, Circ. Res., 63 (1988) 415-428. Forssmann, W.G., Magert, H.J., Meyer, M. and SchulzKnappe, P., Cardiac peptides are also regulatory peptides in the autonomic nervous system and in paraneuronal systems, J. Auton. Nerv. Syst., 33 (1991) 117-118. Gutkind, J.S., Kurihara, M., Castren, E. and Saavedra, J.M., Atrial natriuretic peptide receptors in sympathetic ganglia: Biochemical response and alterations in genetically hypertensive rats, Biochem. Biophys. Res. Commun., 149 (1987) 65-72. Hardey, D.W. and Lokhandwala, M.F., Influence of prazosin on cardiac reflex function; Eur. J. Pharmacol., 57 (1979) 251-254. Hirata, Y., Ishii, M., Sugimoto, T., Matsuoka, H., Kangawa, K. and Matsuo, H., The effects of human atrial 28-amino acid peptide on systemic and renal hemodynamics in anesthetized rats, Circ. Res., 57 (1985) 634-639. Kobinger, W. and Pichler, L., Investigation into different types of post- and presynaptic a-adrenoceptors at cardiovascular sites in rats, Eur. J. Pharmacol., 65 (1980) 393-402. Kuchel, O., Debinski, W., Racz, K., Buu, N.T., Garcia, R., Cusson, J.R., Larochelle, P., Cantin, M. and Genest, J., An emerging relationship between peripheral sympathetic nervous activity and atrial natriuretic factor, Life Sciences, 40 (1987) 1545-1551. Langer, S.Z., Presynaptic regulation of the release of catecholamines, Pharmacol. Rev., 32 (1981) 337-362. Levy, M.N., Sympathetic-parasympathetic interactions in the heart, Circ. Res., XXIX (1971) 437-443. McDonough, P.M., Wetzel, G.T. and Brown, J.H., Further characterization of the presynaptic Alpha-1 receptor modulating [3H]Ach release from rat atria, J. Pharmacol. Exp. Ther., 238 (1986) 612-617. McGrattan, P.A., Brown, J.H. and Brown, O.M., Parasympathetic effects on in vivo rat heart can be regulated through an aa-adrenergic receptor, Circ. Res., 60 (1987) 465-471. Nakamaru, M. and Inagami, T., Atrial natriuretic factor inhibits norepinephrine release evoked by sympathetic nerve stimulation in isolated perfused rat mesenteric arteries, Eur. J. Pharmacol., 123 (1986) 459-461.
88 21 Pardini, B.J., Lund, D.D. and Schmid, P.G., Contrasting preganglionic and postganglionic effects of phenylephrine on parasympathetic control of heart, Am. J. Physiol., 260 (1991) H118-HI22. 22 Rugg, E.L., Alton, J.F. and Cramb, G., Atrial natriuretic peptide receptors and activation of guanylate cyclase in rat cardiac sarcolemma, Biochem. Biophys. Res. Commun., 162 (1989) 1339-1345. 23 Thoren, P., Mark, A.L., Morgan, D.T., O'Neill, P., Needleman, P. and Brody, M.J., Activation of vagal depressor reflexes by atriopeptins inhibits renal sympathetic nerve activity, Am. J. Physiol., 251 (1986) H1252-H1259. 24 Tsuda, K., Shima, H., Nishio, I. and Masuyama, Y., Inhibitory action of alpha-human atrial natriuretic polypeptide on vascular adrenergic neurotransmission is attenuated in spontaneously hypertensive rats, Jpn. Circ. J.. 51 (1987) 589-593.
25 Volpe, M., Cuocolo, A., Vecchione, F., Condorelli, M. and Trimarco, B., Hemodynamic mechanisms of the blood pressure response to atrial natriuretic factor in the dog: Effects of vagotomy, Ann. N.Y. Acad. Sci., 488 (1986) 553 554. 26 Wetzel, G.T. and Brown, J.H., Presynaptic modulation of acetylcholine release from cardiac parasympathetic neurons, Am. J. Physiol., 248 (1985) H33-H39. 27 Wetzel, G.T., Goldstein, D. and Brown, J.H., Acetylcholine release from rat atria can be regulated through an c~l-adrenergic receptor, Circ. Res., 56 (1985) 763-766. 28 Zeuzem, S., Olbrich, H.G.. Haak, T. and Jungmann, E., In vivo evidence that human atria natriuretic factor-(99-126) (hANF) stimulates parasympathetic activity in man, Eur. J. Clin. Pharmacol. 39 (1990) 77-79.
81
© 1993 Elsevier Science Publishers B.V. All rights reserved 0165-1838/93/$06.00 JANS 01348
The interaction between atrial natriuretic peptide and cardiac parasympathetic function D.J. Atchison and U. A c k e r m a n n Department of Physiology, University of Toronto, Toronto, Canada (Received 5 May 1992) (Revision received 24 August 1992) (Accepted 27 August 1992)
Key words: ANP; Heart rate regulation; avAdrenoceptor; Prazosin; Phenylephrine; Vagus nerve Abstract We have demonstrated previously that atrial natriuretic peptide (ANP) inhibits hypotension-induced reflex tachycardia via a parasympathetic mechanism. The present study further defines that parasympathetic mechanism. We tested the hypothesis that ANP, during vagus nerve stimulation, acts as a physiological antagonist to interfere with al-adrenoceptor modulation of efferent cardiac vagal action. Sprague Dawley rats were divided into five groups, each group receiving a different infusion. Infusates included one of vehicle (Ringer's solution; RS), an cq-adrenoceptor agonist (phenylephrine; PE), a combination of agonist and either a known al-adrenoceptor antagonist (prazosin; PE + PRZ) or the putative physiologic antagonist, ANP (PE + ANP). The fifth group received all three drugs, PE + PRZ + ANP. Under Inactin anesthesia (100 mg/kg i.p.), efferent autonomic input to the heart was surgically interrupted. Animals were also adrenalectomized to limit the effects of circulating catecholamines. We then monitored each group for the change in heart rate (AHR) in response to efferent vagus nerve stimulation at various frequencies (2 Hz, 5 Hz, 10 Hz). Infusion of PE significantly (P < 0.01 by ANOVA) attenuated the magnitude of AHR when compared to the RS group. This attenuation of vagally-induced bradycardia was eliminated by the addition of the al-adrenoceptor antagonist, prazosin (PE + PRZ group). The PE + ANP group responded with results similar to those of the PE + PRZ group. There was no difference between AHR responses of the PE + PRZ + ANP group and the PE + PRZ group. The results show that, in a setting of experimentally manipulated autonomic input to the heart, ANP produces a response to vagus nerve stimulation that is equivalent to that seen with the al-adrenoceptor antagonist prazosin. Furthermore, in the presence of prazosin, ANP had no additional chronotropic effects. These results are consistent with our hypothesis that ANP influences parasympathetic control of heart rate by modifying the consequences of al-adrenergic receptor activation.
Introduction
An earlier report from this laboratory determined that ANP enhances the negative chronotropic effect of cardiac parasympathetic (vagus)
Correspondence to: U. Ackermann, Department of Physiology, Medical Sciences Building, University of Toronto, Toronto, Canada M5S 1A8.
nerve stimulation, but has no effect either on cardiac sympathetic chronotropic actions or directly on sino-atrial (SA) nodal activity [3]. Our earlier observations are particularly intriguing because ANP receptors have yet to be demonstrated on cardiac autonomic nerve terminals. Although ANP receptors have been localized to the sympathetic stellate ganglion [11] and postsynaptically on atrial [2] and ventricular myocytes [22] our previous experiments demonstrated these
82 ANP receptors were not directly involved in the modulation of heart rate [3]. The purpose of the present study was to characterize further the interaction between ANP and cardiac efferent parasympathetic function. It has been documented that ANP inhibits sympathetic nerve activity [15,23] and attenuates the release of norepinephrine [7,20]. However, sympathetic nerve activity and the release of norepinephrine are not the sole modulators of heart rate. Among the chronotropic mechanisms, consideration must be granted, not only to the role of parasympathetic nerve activity, but also to the interaction between sympathetic and parasympathetic systems [17]. Any investigation into the regulation of heart rate should acknowledge and account for the possibility that an autonomic interaction may influence the results. In this whole-animal study we have used a cardiac-denervated preparation that allowed selective aj-adrenoceptor stimulation or blockade on vagal presynpatic terminals during experimental manipulation of vagus nerve activity. The preparation was chosen because three distinct observations suggested an hypothesis: (1) ANP is capable of directly inhibiting sympathetic nerve action [7,15]; (2) activation of presynaptic aladrenergic receptors can modulate acetylcholine release [26] and attenuate the negative chronotropic response during vagus nerve excitation [19,27]; (3) ANP and the specific al-adrenergic antagonist, prazosin, exhibit similarities in their cardiovascular effects [12]. Our results suggest that ANP interferes with the inhibitory action of presynaptic al-adrenoceptors and thereby enhances the negative chronotropic effect of cardiac efferent parasympathetic nerve activation. The interference is not likely to arise from ANP binding to the a~-adrenoceptor site, but is more likely to result from an as yet unspecified physiological antagonism.
5-sec-butyl-5-ethyl-2-thiobarbituric acid (Inactin; Byk) 100 m g / k g i.p.. If necessary, one supplemental dose (25-33% of initial dose) was given. When the animal was fully anesthetized, the following surgical procedures were performed: bilateral adrenalectomy, tracheotomy (PE-390 tubing heated and pulled to an appropriate diameter), bilateral vagotomy in the mid-neck region, and bilateral femoral vein catheterization (PE-50 tubing) for the infusion of drugs or vehicle. The left femoral artery was catheterized (PE-50 tubing) to monitor arterial blood pressure (Electromedics Inc. MS20 transducer). Heart rate was monitored using a cardiotachometer (Sensormedics type 9857B) triggered by arterial pressure pulses. A chart recording (Beckman dynograph Type RP) of blood pressure and heart rate was obtained during control and experimental periods. Temperature was maintained between 37-38°C using a warming pad. Animals were ventilated (Harvard Apparatus Rodent Respirator Model 680) with 100% 0 2 at a rate and tidal volume that maintained normal acid-base balance as determined by arterial blood pH. Via a median sternotomy we exposed, isolated and transected nerves leading to the heart from both the right and left stellate ganglia. We have demonstrated that this procedure effectively eliminated sympathetic input to the heart [3]. The removal of autonomic innervation by these procedures (vagotomy and sympathectomy) decentralized efferent cardiac nerve supply, creating, in effect, an in vivo denervated heart model. Efficacy of denervation was ascertained by absence of a change in heart rate during transient bilateral carotid artery occlusion (baroreflex activation). After completion of surgical interventions, a section of the right vagus nerve, 10-15 mm caudal to the transected end, was carefully dissected using glass dissecting rods. The nerve was gently impaled upon a bipolar stainless steel hook electrode.
Materials and Methods
Drugs
Surgical preparation Male Sprague Dawley rats (240-341 g; Harlan Sprague Dawley, Inc.) were anesthetized with
Ringer's solution (RS; vehicle) was obtained from Abbott Laboratories Limited(catalogue no. 1582), L-phenylephrine hydrochloride (PE) and prazosin hydrochloride (PRZ) from Sigma (cata-
83 logue nos. P-6126 and P-7791, respectively), and atrial natriuretic peptide (ANP; rat 28-amino acid) from Peninsula Laboratories (catalogue no. 9103). PE and A N P were dissolved in Ringer's solution. P R Z was dissolved initially in propylene glycol (Fisher Scientific; catalogue no. P-354) then in Ringer's solution to a final propylene glycol concentration of < 5% vol/vol. To demonstrate physiologic antagonism in our denervated preparation it was necessary to choose a dose of a l - a d r e n o c e p t o r agonist that would measurably diminish the chronotropic effect of vagus nerve stimulation at physiologically relevent frequencies. On that basis the phenylephrine dose was chosen such that its associated blood pressure increase was less than 20 mmHg, but its chronotropic effect was sufficient to diminish measurably the heart rate responses to vagal stimulation in the frequency range 0 - 1 0 Hz. These objectives were accomplished with a PE infusion rate of 2.0 t z g / k g min -1. A prazosin dose was chosen that would block completely the observed chronotropic effect of PE. This dose was 5 . 0 / x g / k g min-~ and the completeness of its blocking effectiveness was indicated by the observation that it prevented phenylephrine-induced increases in mean arterial pressure of up to 40 mmHg. A N P was infused at 0.28 / z g / k g min -~ because that dose produced hemodynamic effects comparable to those of prazosin (at 5 p . g / k g min -1) in intact, anesthetized animals. It was necessary to choose the dose on the basis of pilot experiments with fully innervated rats because vagotomy eliminates ANP-induced hypotension [25]. The cumulative infusion volume was 0.051 ml/min.
Protocol Animals were randomly assigned to one of five groups. Each group was distinguished by a different drug infusion or combination of drug infusions: RS (n = 9), PE (n = 8), PE + P R Z (n = 9), P E + A N P ( n = 8 ) or P E + P R Z + A N P (n=8). Thirty minutes were allowed, after the start of each infusion, for stabilization and equilibration. After this interval, we began with the collection of baseline data (heart rate and blood pressure in the absence of cardiac autonomic nerve activity)
followed by brief periods of vagus nerve stimulation. D a t a were collected for 2 min at each stimulation frequency. We observed the change in heart rate ( A H R ) in response to vagus nerve stimulation at randomized frequencies (2 Hz, 5 Hz, 10 Hz) that had previously been shown to produce heart rate changes within a physiological range [3]. Electrical nerve stimulation (Grass SD9 Stimulator) lasted 2 min, with a 5 min rest period between stimulations to allow heart rate to return to baseline. Pulse duration was 1 ms. The necessary stimulus voltage was determined for each animal prior to beginning an infusion and was set at the minimum voltage required to achieve a predetermined change in heart rate ( > 12, but < 20 beats per minute) at the lowest frequency (2 Hz). The range of voltage used was 1.4-3.3 volt, the mean voltage ( + S . E . M . ) w a s 2 . 2 + 0 . 1 V. Animals underwent only one 2 min period of nerve stimulation at each frequency. Heart rate data were collected at time 0 and continuously thereafter for 2 min. Group comparisons were based on steady-state response to stimulation; this has been shown to occur by the 90-s time point [3].
Analysis Data analysis focussed on the relationship between A H R and frequency of vagus nerve stimulation. The change in heart rate during stimulation was calculated by subtracting the heart rate obtained at time 0 from the rate measured at 90 s of stimulation. G r o u p differences were analyzed by an analysis of variance (ANOVA). An A N O V A first compared the PE group with the RS group to determine that PE, at the concentration used here, influenced A H R in the same direction as shown by Pardini et al. [21] and McGrattan et al. [19]. All other comparisons were made relative to the PE group (activated a~-receptor). The level of statistical significance was set at P < 0.05.
Results Baseline mean arterial pressure, as measured after stabilization following drug infusion, ( + S.E.M.) was 93 + 5 m m H g for the RS group,
84
85 + 4 m m H g for PE + ANP, 60_+ 3 m m H g for PE + PRZ, and 54 + 6 m m H g for PE + P R Z + ANP. The lower baseline arterial pressure in the groups receiving prazosin was a consequence of combined denervation and adrenalectomy and, therefore, unavoidable. A lesser dose of prazosin would not have blocked the phenylephrine effect on vagus nerve chronotropy. For the PE group, the mean arterial pressure rose to 101 _+ 9 m m H g within 10 min of beginning the infusion, then stabilized at 95_+ 9 m m H g by the end of the 30-min equilibration period. Baseline heart rate (absolute values; mean _+ S.E.M.) for each group was also obtained at the end of the 30-min equilibration period (Fig. 1). These values were not significantly different from one another ( P = 0.27 by ANOVA). Thus it is unlikely that the magnitude of A H R during parasympathetic activity in each of the groups was influenced by the initial heart rate. The / t H R in response to stimulation at the various frequencies and for each infusion group is summarized in Figs. 2-4.
400 HEART
375
(BPM)
350 r
325 [
30O
RS
PE
PE
PE
+
+
PE +
PRZ
ANP
PRZ +
ANP Fig. 1. Baseline heart rates (beats per min, bpm) at the end of a 30 min infusion-equilibration period (mean±S.E.M.). Values were obtained at t = 9 0 s of the baseline (0 Hz) data collection period in the absence of autonomic nervous input. An ANOVA including all groups showed differences were not statistically significant. RS = Ringer's solution; PE = phenylephrine; PRZ = prazosin; ANP = atrial natriuretic peptide.
10 0 CHANGE IN
-10
-20
HEART RATE (BPM)
-30 -40
RS
-50 -60
i
i
i
0
5
10
FREQUENCY OF VAGAL STIMULATION (HZ) Fig. 2. Average change in heart rate (±S.E.M.; bpm) in response to right vagus nerve stimulation at selected frequencies during the infusion of phenylephrine (PE; e) or Ringer's solution (RS; ©). P < 0.01 between groups by ANOVA.
Figure 2 illustrates that heart rate for the RS group was reduced decrementally at increasing frequencies of vagus nerve stimulation. We found that the bradycardic response to vagus nerve stimulation for the PE group was less than for the RS group (Fig. 2; P = 0.006 by ANOVA). The ability of PE to attenuate the magnitude of heart rate response to vagus nerve stimulation was independent of any interaction or interdependence between the infusion of PE and frequency of stimulation ( P = 0.74 by ANOVA). Phenylephrine-induced reduction in bradycardia at any given level of vagal stimulation was eliminated both in the presence of the known eel-antagonist prazosin (Fig. 3) as well as in the presence of A N P ( P = 0.0003 and P = 0.0001, respectively). In each comparison, a H R for the PE + P R Z and PE + ANP groups was greater than the A H R achieved in the PE group. The frequency/infusion interaction term provided by the A N O V A was not significant in both the PE + P R Z and PE + ANP groups ( P = 0.06 for each group). The magnitude of A H R during vagal stimulation responses observed in the PE + ANP and PE + P R Z groups, were remarkably similar (Fig. 4). Furthermore, A N O V A of the results obtained in the PE + P R Z + ANP group showed no significant difference when compared to the
85 10 0 CHANGE IN
-10 -20
HEART RATE (BPM)
-30 -40 "
~
~ PE + PRZ
-50 -60
I
I
I
0
§
10
FREQUENCY OF VAGAL STIMULATION (HZ)
Fig. 3. Average change in heart rate (+S.E.M.; bpm) in response to right vagus nerve stimulation during the infusion of phenylephrine (PE; e) or the simultaneous infusion of phenylephrine and prazosin (PE+PRZ; a). P < 0.01 between groups by ANOVA. 10 o
CHANGE IN
-10
.2o
|~,
HEART RATE (BPM)
-40
~ PE + PRZ + ANP J. ~PE + PRZ ±PE + ANP
-50 -50
0'
5'
I 'o
FREQUENCY OF VAGAL STIMULATION (HZ)
Fig. 4. Average change in heart rate (+ S.E.M.; bpm) resulting from right vagus nerve stimulation. Values were obtained during the infusions of phenylephrine and prazosin (PE+ PRZ; zx), phenylephrine and ANP (PE+ANP; v), or phenylephrine, prazosin and ANP ( P E + P R Z + A N P ; II). ANOVA found no significant differences among the groups.
PE + P R Z and PE + ANP groups ( P = 0.71; Fig. 4).
Discussion As a follow-up to previous work, the present study attempted to characterize more fully the ability of ANP to enhance the negative chronotropic response to vagus nerve stimulation. The
results presented here suggest that ANP acts as a physiological antagonist of a]-adrenoceptors in the parasympathetic regulation of heart rate. This antagonism most likely does not involve direct competition for the receptor binding site. Hence the criteria used to select drug dosages did not include dose-response curves. Instead, doses were chosen on the basis of their effect on the chronotropic action of vagus nerve stimulation. Confirmation of the involvement of aj-adrenoceptors in vagally mediated heart rate effects required that sufficient phenylephrine be given to cause a significant diminution of vagal bradycardia at the chosen stimulus frequencies. Similarly, sufficient prazosin needed to be given to block the phenylephrine effect. Many investigators have administered prazosin as a bolus or as a brief ( < 30 min) infusion. In our experimental preparation, prazosin, administered continuously, caused profound hypotension even at doses much lower than others [e.g. 21] have used. The need to maintain blood pressure near normal governed the upper limit of an acceptable prazosin dose. Its lower limit was set by the need to elicit observable antagonism to the chronotropic effects of PE. Prazosin at 5 ~ g / k g m i n - I lowers mean arterial blood pressure by about 15 mmHg ([4] and unpublished observation) and completely blocked the influence of PE in our experiments. We administered ANP at a dose that produced hemodynamic effects comparable to those of the selected prazosin dose. The surgical technique used here effectively interrupts cardiac efferent nerve supply from the central nervous system [3]. We added bilateral adrenalectomy in order to diminish effects arising from changes in humoral catecholamine levels. Though cardiac sympathectomy combined with adrenalectomy may not have removed all endogenous sources of catecholamines [5], the observation that combined administration of phenylephrine and prazosin (PE + PRZ; Fig. 3) yielded results similar to those in the control infusion group (RS; Fig. 2) confirmed our belief that unaccounted sources of catecholamines (e.g. norepinephrine released from systemic sympathetic terminals) or involvement of other adrenoceptors did not play a significant role in influencing zlHR.
86
Our success at manipulating a l - a d r e n o c e p t o r activity is indicated by the observation that combined infusion of phenylephrine and prazosin (PE + P R Z group; Fig. 3) prevented the attenuated chronotropic response to nerve stimulation seen in the group receiving phenylephrine alone (PE; Fig. 3), yielding heart rate changes that were comparable to those seen in animals receiving vehicle infusion alone (RS group). Sympathetic and parasympathetic nervous interactions within the heart have been localized to the presynaptic level, the parasympathetic nervous system modulating release of sympathetic neurotransmitters [16] and the sympathetic nervous system modulating the release of acetylcholine [19]. Although adrenergic receptors located presynaptically are considered to be predominantly ~2-adrenoceptors, there have been suggestions that the a,-adrenergic receptor subtype can also be located presynapticatly [6,14]. The reputed function of such presynaptic a 1adrenoceptors is inhibition of neurotransmitter release [18,26,27]. Presynaptic adrenergic inhibition of acetylcholine release may explain, in part, the interaction of sympathetic and parasympathetic nerve activity that has been demonstrated in the autonomic control of heart rate [1,17]. Furthermore, the documented effects of ANP to interfere with sympathetic action [7,15,24] together with its pronounced ability to augment cardiac parasympathetic effects [1,3,28] are consistent with an ANP action to interfere with presynaptic a~-adrenergic mechanisms that influence Ach release. The specific a~-adrenoceptor antagonist, prazosin [12], as well as ANP [1,13], cause hypotension that is not associated with a compensatory reflex tachycardia. The similarity of their effects on blood pressure and heart rate reinforces our speculation that the chronotropic effects of ANP may be mediated by interference with a~adrenoreceptor mechanisms, though we do not expect this interaction to occur directly with the adrenoceptor. The possibility that prazosin acted directly upon the SA node [8] is contradicted by the observation that baseline heart rate for the PE + P R Z group, in the absence of nerve stimulation, was not statistically different when corn-
pared to baseline heart rate in the control (RS) group (Fig. 1). Thus it is unlikely that our results are due to an effect of prazosin directly upon the SA node. Nor is it likely that our results are a direct consequence of ANP acting upon the SA node because we have previously found that ANP did not affect SA node activity in the absence of autonomic nerve input [3]. We expected that administration of ANP in combination with e~-blockade by prazosin would allow us to reveal separate sites of action for ANP and prazosin. If the observed effects of ANP did not arise from an influence on a~adrenoceptors, then we expected that administration of ANP in combination with prazosin would produce chronotropic results different from those already obtained during the separate infusion of each drug. This was not the case (Fig. 4). When P R Z and ANP were used in combination (PE + P R Z + ANP group) the magnitude of A H R was similar to that obtained during the infusion of either PE + P R Z or PE + ANP (Fig. 4). We believe these data are evidence for an inhibitory influence of ANP on c~radrenoceptors and suggests that both P R Z and ANP interfered with a~-adrenoceptor activation. The lack of an additive effect during the simultaneous administration of ANP and P R Z is consistent with a final common pathway of action of ANP and PRZ. Several investigators [18,19,26] have suggested that this site of inhibition is presynaptic. We demonstrated, in a former study, that ANP alone did not influence heart rate response to sympathetic nerve (stellate ganglion) stimulation, but did enhance the magnitude of heart rate response to vagus nerve stimulation [3]. The results of the present study indicate that ANP did not show effects additional to those which could be demonstrated via inhibition of presynaptic al-adrenoreceptors on vagal nerve endings. The suggestion that ANP influences o~-adrenoceptor activity in the autonomic control of heart rate is consistent with a study of ANP influence in the peripheral vasculature. Faber et al. [9] reported that, in denervated cremaster muscle, ANP had no effect on vessels pre-contracted with an ax-adrenergic agonist. Though no mechanism of action was proposed, they demon-
87
strated that ANP was effective in reversing the effects of al-agonist-induced vasoconstriction by phenylephrine. This demonstration of selective al-adrenoceptor-inhibitory effects of ANP in the peripheral vasculature supports our interpretation that al-adrenergic-dependent mechanisms are responsible for the chronotropic effects of ANP even though localization of the receptor and the nature of the mechanism are still undetermined. In conclusion, we have determined that ANP inhibits the adrenergic regulation of parasympathetic nerve activity in the autonomic control of heart rate. We speculate that ANP may act indirectly through ANP-receptor-activated G proteins. Moreover it has also been hypothesized that ANP and ANP-like peptides within the autonomic nervous system act as neurotransmitters [10]. Though the extent and significance of this role are still undetermined, the results we have obtained may be due to a neurotransmitter function of ANP within the autonomic nervous system at the level of synaptic terminals.
6
7
8
9
10
11
12
13
Acknowledgements The authors thank Dr. Evert Vos and CibaGeigy, Canada for financial support and Dr. Peter Pennefather for constructive criticism of the manuscript.
14
15
References 1 Ackermann, U., Khanna, J. and Irizawa, T., Atrial natriuretic factor alters autonomic interactions in the control of heart rate in conscious rats, Can. J. Physiol. Pharmacol., 66 (1988) 930-936. 2 Anand-Srivastava, M.B. and Cantin, M., Atrial natriuretic factor receptors are negatively coupled to adenylate cyclase in cultured atrial and ventricular cardiocytes, Biochem. Biophys. Res. Commun., 138 (1986) 427-436. 3 Atchison, D.J. and Ackermann, U., Influence of atrial natriuretic factor on autonomic control of heart rate. Am. J. Physiol., 258 (1990) H718-H723. 4 Cavero, I., Effects of prazosin on reflex changes in heart rate evoked by vasopressor and vasodepressor stimuli in conscious rabbits. J. Cardiovasc. Pharmacol., 4 (1982) S108-Sl13. 5 Darlington, D.N., Chew, G., Ha, T., Keil, L.C. and Dall-
16 17 18
19
20
man, M.F., Corticosterone, but not glucose, treatment enables fasted adrenalectomized rats to survive moderate hemorrhage, Endocrinology, 127 (1990) 766-772. Docherty, J.R., An investigation of presynaptic a-adrenoceptor subtypes in the pithed rat heart and in the rat isolated vas deferens, Br. J. Pharmacol., 82 (1984) 15-23. Drewett, J.G., Trachte, G.J. and Marchand G.R, Atrial natriuretic factor inhibits adrenergic and purinergic neurotransmission in the rabbit isolated vas deferens, J. Pharmacol. Exp. Ther., 248 (1989) 135-142. Dukes, I.D. and Vaughan Williams, E.M., Electrophysiological effects of a-adrenoceptor antagonists in rabbit sino-atrial node, cardiac Purkinje cells and papillary muscles, Br. J. Pharmacol., 83 (1984) 419-426. Faber, J.E., Gettes, D.R. and Gianturco, D.P., Microvascular effects of atrial natriuretic factor: interaction with a 1- and a2-adrenoceptors, Circ. Res., 63 (1988) 415-428. Forssmann, W.G., Magert, H.J., Meyer, M. and SchulzKnappe, P., Cardiac peptides are also regulatory peptides in the autonomic nervous system and in paraneuronal systems, J. Auton. Nerv. Syst., 33 (1991) 117-118. Gutkind, J.S., Kurihara, M., Castren, E. and Saavedra, J.M., Atrial natriuretic peptide receptors in sympathetic ganglia: Biochemical response and alterations in genetically hypertensive rats, Biochem. Biophys. Res. Commun., 149 (1987) 65-72. Hardey, D.W. and Lokhandwala, M.F., Influence of prazosin on cardiac reflex function; Eur. J. Pharmacol., 57 (1979) 251-254. Hirata, Y., Ishii, M., Sugimoto, T., Matsuoka, H., Kangawa, K. and Matsuo, H., The effects of human atrial 28-amino acid peptide on systemic and renal hemodynamics in anesthetized rats, Circ. Res., 57 (1985) 634-639. Kobinger, W. and Pichler, L., Investigation into different types of post- and presynaptic a-adrenoceptors at cardiovascular sites in rats, Eur. J. Pharmacol., 65 (1980) 393-402. Kuchel, O., Debinski, W., Racz, K., Buu, N.T., Garcia, R., Cusson, J.R., Larochelle, P., Cantin, M. and Genest, J., An emerging relationship between peripheral sympathetic nervous activity and atrial natriuretic factor, Life Sciences, 40 (1987) 1545-1551. Langer, S.Z., Presynaptic regulation of the release of catecholamines, Pharmacol. Rev., 32 (1981) 337-362. Levy, M.N., Sympathetic-parasympathetic interactions in the heart, Circ. Res., XXIX (1971) 437-443. McDonough, P.M., Wetzel, G.T. and Brown, J.H., Further characterization of the presynaptic Alpha-1 receptor modulating [3H]Ach release from rat atria, J. Pharmacol. Exp. Ther., 238 (1986) 612-617. McGrattan, P.A., Brown, J.H. and Brown, O.M., Parasympathetic effects on in vivo rat heart can be regulated through an aa-adrenergic receptor, Circ. Res., 60 (1987) 465-471. Nakamaru, M. and Inagami, T., Atrial natriuretic factor inhibits norepinephrine release evoked by sympathetic nerve stimulation in isolated perfused rat mesenteric arteries, Eur. J. Pharmacol., 123 (1986) 459-461.
88 21 Pardini, B.J., Lund, D.D. and Schmid, P.G., Contrasting preganglionic and postganglionic effects of phenylephrine on parasympathetic control of heart, Am. J. Physiol., 260 (1991) H118-HI22. 22 Rugg, E.L., Alton, J.F. and Cramb, G., Atrial natriuretic peptide receptors and activation of guanylate cyclase in rat cardiac sarcolemma, Biochem. Biophys. Res. Commun., 162 (1989) 1339-1345. 23 Thoren, P., Mark, A.L., Morgan, D.T., O'Neill, P., Needleman, P. and Brody, M.J., Activation of vagal depressor reflexes by atriopeptins inhibits renal sympathetic nerve activity, Am. J. Physiol., 251 (1986) H1252-H1259. 24 Tsuda, K., Shima, H., Nishio, I. and Masuyama, Y., Inhibitory action of alpha-human atrial natriuretic polypeptide on vascular adrenergic neurotransmission is attenuated in spontaneously hypertensive rats, Jpn. Circ. J.. 51 (1987) 589-593.
25 Volpe, M., Cuocolo, A., Vecchione, F., Condorelli, M. and Trimarco, B., Hemodynamic mechanisms of the blood pressure response to atrial natriuretic factor in the dog: Effects of vagotomy, Ann. N.Y. Acad. Sci., 488 (1986) 553 554. 26 Wetzel, G.T. and Brown, J.H., Presynaptic modulation of acetylcholine release from cardiac parasympathetic neurons, Am. J. Physiol., 248 (1985) H33-H39. 27 Wetzel, G.T., Goldstein, D. and Brown, J.H., Acetylcholine release from rat atria can be regulated through an c~l-adrenergic receptor, Circ. Res., 56 (1985) 763-766. 28 Zeuzem, S., Olbrich, H.G.. Haak, T. and Jungmann, E., In vivo evidence that human atria natriuretic factor-(99-126) (hANF) stimulates parasympathetic activity in man, Eur. J. Clin. Pharmacol. 39 (1990) 77-79.