Interaction between direct sympathetic and vagus nerve stimulation on heart rate in the isolated rabbit heart

Transcription

1 Exp Physiol 89.1 pp Experimental Physiology Interaction between direct sympathetic and vagus nerve stimulation on heart rate in the isolated rabbit heart Kieran E. Brack 1, John H. Coote 1 and G AndréNg 2 1 Department of Physiology, Division of Medical Sciences, University of Birmingham, Birmingham, UK 2 Department of Cardiovascular Sciences, University of Leicester, Leicester, UK The interaction between the effects of vagus nerve stimulation (VS) and sympathetic stimulation (SS) on intrinsic heart rate was studied in the novel innervated isolated rabbit heart preparation. The effects of background VS, at different frequencies 2 Hz (low), 5 Hz (medium), 7 Hz (high) on the chronotropic effects of different frequencies of SS 2 Hz (low), 5 Hz (medium), 10 Hz (high) were studied. The experiments were repeated in the reverse direction studying the effects of different levels of background SS on the chronotropic effects of different levels of VS. Background VS reduced the overall positive chronotropic effect of SS at steady state in a frequency dependent manner and the rate of increase in heart rate during low and medium SS (but not high SS) was slowed in the presence of background VS. These results suggest that pre- and postjunctional mechanisms may be involved in the sympatho vagal interaction on heart rate. On the other hand, the chronotropic effect of VS was enhanced in the presence of background SS. Vagal stimulation appears to play a dominant role over sympathetic stimulation in chronotropic effects on the isolated heart. The innervated isolated heart preparation is a valuable model to study the complex mechanisms underlying the interaction between sympathetic and parasympathetic stimulation on cardiac function. (Received 09 September 2003; accepted 5 November 2003) Corresponding author G. A. Ng: Cardiology Group, Department of Cardiovascular Sciences, Clinical Sciences Wing, Glenfield Hospital, Leicester LE3 9QP, UK. gan1@le.ac.uk The autonomic nervous system exerts powerful influence on the performance of the heart. In general, the sympathetic and parasympathetic nerves act in directionally opposite ways to modulate aspects of cardiac performance, achieving delicate physiological control. The interactions between the two autonomic divisions are complex. It was Samaan (1935) who discovered that the algebraic sum of the effects of stimulating the individual sets of nerves on heart rate is different from the actual effect obtained by the simultaneous stimulation of the two sets of nerves. Levy (1971) coined the term accentuated antagonism to describe the inhibition of the cardiac effects from sympathetic nerve stimulation by the presence of background vagus nerve stimulation. Early in vivo studies suggest that each of the two autonomic divisions can both inhibit and enhance the activity of the other via effects on the release of neurotransmitters at the synaptic junctions (Levy, 1971). There is also in vitro data in isolated hearts using neurotransmitter analogues to support these studies (Hoffmann et al. 1945; Grodner et al. 1970). However, the in vivo studies are complicated by the difficulty in controlling baseline conditions and by the presence of haemodynamic reflexes and circulating humoral factors. On the other hand, it is difficult to be certain that pharmacological manipulation to mimic the effects of nerve stimulation reflects the situation in vivo. The aim of this study is to investigate the nature of the interaction between direct sympathetic and parasympathetic nerve stimulation on heart rate in a newly developed in vitro isolated rabbit heart preparation with fully functional dual autonomic innervation. Preliminary results from this study have been published in abstract form (Brack et al. 2001). Methods Isolation of heart with intact autonomic innervation The isolated heart preparation with intact autonomic innervation has been previously described (Ng et al. 2001). In brief, adult male New Zealand White rabbits DOI: /expphysiol

2 Exp Physiol 89.1 pp Sympathovagal interaction on heart rate 129 ( kg, n = 6) were premedicated with Hypnorm (Janssen Pharmaceuticals LtD, Oxford, UK; 0.1 mg kg 1 S.C). General anaesthesia was induced with i.v. Hypnoval (Roche Products Ltd, Welwyn Garden City, UK; 1 mg kg 1 ) and maintained with pentobarbitone sodium (Sagital, Rhône Mérieux, Harlow, UK; 2 mg every 5 min. I.V.). The rabbit was ventilated, after tracheotomy, at 60 breaths per min using a small-animal ventilator (Harvard Apparatus Ltd, Edenbridge, Kent, UK) with an O 2 /air mixture. The vagus nerves were isolated and the blood vessels leading to and from the rib cage were ligated and dissected. The rabbit was killed with an overdose of Sagital (60 mg I.V.) together with 500 U heparin i.v. The anterior portion of the rib cage was removed and the descending aorta cannulated. The preparation extending from the neck to thorax was dissected as described before (Ng et al. 2001). The procedures were undertaken in accordance with the Animals (Scientific Procedures) Act 1986 and conformed with the Guide for the Care and Use of Laboratory Animals Published by the US National Institutes of Health (NIH Publication no , revised 1985). Langendorff perfusion The preparation was perfused via the aortic cannula in the modified Langendorff mode with Tyrode solution of the following composition (mm): Na 138, K 4.0, Ca 1.8, Mg 1.0, HCO , H 2 PO 4 0.4, Cl 121, glucose 11. The ph of the solution was maintained at 7.4 by continuously bubbling with 95% O 2 /5% CO 2 mixture and the temperature was maintained at 37 C. A constant perfusion rate of 100 ml min 1 was maintained using a Gilson Minipuls 3 peristaltic pump (Gilson Inc., Ohio, USA). Thebesian venous effluent was drained via a catheter placed at the left ventricular apex. Intraventricular pressure was monitored with a fluid-filled latex balloon connected to a pressure transducer (model MTL0380, AD Instruments Ltd, UK) and inserted into the left ventricle via the left atrium. The volume of the balloon was adjusted to give zero end-diastolic pressure. Perfusion pressure was monitored with a second pressure transducer in series with the aortic cannula. A pair of platinum electrodes (Grass Instruments, Astro-Medical Inc., Slough, UK) was inserted into the right atrial appendage for recording of atrial electrograms. Autonomic nerve stimulation Each of the two vagus nerves was supported on separate custom made bipolar silver electrodes (Advent Research Materials, UK: 0.5 mm O.D.) for individual vagus nerve stimulation (VS). For sympathetic stimulation (SS), a quadripolar catheter with 4 electrodes mounted at the end (Biosense Webmaster Inc., Diamond Bar, USA; 2 mm electrode, 10 mm spacing) was inserted into the spinal canal at the 12th thoracic vertebra and the tip of the catheter was advanced to the level of the second thoracic vertebra for stimulation of both sides of the cardiac sympathetic outflow, with the tip electrode acting as cathode and the remaining three electrodes acting as anodes. Two single channel constant voltage square pulse stimulators (SD9, Grass Instruments) were used one for VS and one for SS. At a frequency of 5 Hz and 2 ms pulse width, the stimulus output was increased (over a range from 1 to 20 V) to examine the heart rate response of sympathetic, left and right vagus nerve stimulation individually. The stimulus output producing a submaximal response, as described before (Ng et al. 2001), was used subsequently in the study. This was 4.4 ± 1.5 V for sympathetic stimulation and 8.3 ± 2.8 V for both left and right vagus nerve stimulation. Stimulation was carried out at frequencies of 2 Hz (low), 5 Hz (medium) and 7 Hz (high) with the left and right VS and at 2 Hz (low), 5 Hz (medium) and 10 Hz (high) for SS. Signal measurements and analysis All pressure and electrical signals obtained from the preparation were recorded with a PowerLab 800/s system (AD Instruments Ltd) and digitized at 1 khz using Chart software (AD Instruments Ltd) with the data stored and displayed on a Power Macintosh G3 personal computer (Apple). The stimulus signals were also registered using the same system to record the exact timing and duration of the stimulations. Instantaneous intrinsic sinus heart rate at baseline and after autonomic nerve stimulation was obtained by measuring the beat-to-beat interval of the atrial electrograms. Statistics All data are expressed as mean ± standard error. The effects of direct SS or VS on heart rate at different frequencies were analysed using repeated measures anova with Tukey s multiple comparison test for posthoc analysis. Comparison between left and right VS was made with Mann Whitney test. Two-tailed P-value of less than 0.05 was considered significant.

3 130 K. E. Brack and others Exp Physiol 89.1 pp Results Effects of background vagus nerve stimulation on the heart rate response to sympathetic stimulation Steady state heart rate changes. VS was initiated first until heart rate has reached steady state after which SS was commenced. Stimulation of the two sets of autonomic nerves were stopped after a new steady state heart rate was achieved. Figure 1A illustrates a typical experiment where left VS (5 Hz) reduced heart rate from about 150 bpm to 110 bpm. SS (2 Hz) then increased heart rate to reach a steady state level of about 140 bpm. Cessation of autonomic stimulation caused a sharp rise in heart rate followed by a gradual return to prestimulation level. Figure 1B shows the data obtained for the experiment shown in Fig. 1A. The leftmost panel shows the change in instantaneous beat-to-beat heart rate at low, medium and high frequencies of SS, in the absence of VS. The subsequent panels show the change in heart rate during different frequencies of SS in the presence of background low, medium and high frequency left VS. It is clear from this experiment that the increase in heart rate with SS was significantly reduced in the presence of increasing levels of background VS. The mean data (n = 6) for the effects of background left VS on steady state heart rate achieved with SS are summarized in Fig. 2A. Similar results for right VS are shown in Fig. 2B. SS caused a significant (P < 0.001) frequency dependent increase in heart rate in the absence and presence of VS independent of either left or right VS and of the frequency of VS. Heart rates were significantly lower for right VS compared with left (P < 0.05) at each frequency. Effects of vagal stimulation on increase in heart rate with sympathetic stimulation Figure 1B shows that the increase in heart rate with SS was reduced in the presence of increasing levels of background VS in a typical experiment. This is also suggested in the mean data for steady state heart rates shown in Figs 2A and B for both left and right VS. For quantitative analysis, the increase in heart rate with various frequencies of SS were calculated in each experiment at each level of background VS and the mean data plotted in Fig. 2C for left VS and Fig. 2D for right VS. The reduction in the positive chronotropic effect of SS in the presence of increasing levels of VS is illustrated in these two figures. The increase in heart rate with low frequency SS was reduced from 37.0 ± 3.5 bpm (no VS) to 18.0 ± 4.2 bpm with high frequency left VS (P < 0.001). The heart rate increase with high frequency SS was reduced from 87.1 ± 6.4 bpm (no VS) to 34.9 ± 6.1 bpm with high frequency left VS (P < 0.001). There was no significant difference in the increase in heart rate at the various levels of SS between left and right VS at the various stimulation frequencies. Algebraic sum of heart rate changes with sympathetic and vagal stimulation The possible interaction between the two autonomic branches was analysed by simple addition of the changes in heart rate obtained with SS and VS individually (i.e. the algebraic sum) and comparing this with the change in heart rate obtained experimentally during concomitant SS and VS. In Fig. 3A (left VS) and Fig. 3B (right VS), the open symbols represent the calculated algebraic sum for each combination of SS and VS whilst the filled symbols represent the experimentally obtained heart rate change for the corresponding combination. The experimentally obtained changes in heart rate during concomitant SS and VS at each combination of frequencies were all lower than the algebraic sum (P < 0.05 for all except low frequency SS with background low left and right VS). Effect of vagal stimulation on the delay in heart rate increases with sympathetic stimulation As reported previously (Ng et al. 2001), there is a delay or latency in the heart rate response to SS, which decreases with increase in stimulation frequency, and is thought to be related to the kinetics of the second messenger system (Levy et al. 1993). The effect of VS on this latency period was examined in this study. Latency was measured as the time period from the beginning of SS to a 1% increase in heart rate. Figure 4A shows beat-to-beat heart rate in a typical experiment during low frequency SS without and at the various levels of background VS. It can be appreciated that there was a greater delay in heart rate increase with the same level of SS at higher levels of background VS. This is clearly shown in expanded scale in Fig. 4B. Figure 4C shows the mean data for the latency period at different combinations of SS and VS. Latency was decreased significantly (P < 0.05) with increases in SS in the absence of and at each of the various levels of VS. At low frequency SS, latency period was 2.1 ± 0.2 ms with no background VS. Increasing levels of background VS prolonged the latency period with the same low frequency SS up to 4.2 ± 1.0 ms at high frequency background VS which was significant (P < 0.05). This was also true for medium frequency SS (P < 0.05). However, for high frequency SS, although there was a trend towards an increase in latency period from 1.2 ± 0.2 ms without VS to 2.0±0.4 ms with high frequency background VS, this

4 Exp Physiol 89.1 pp Sympathovagal interaction on heart rate 131 Figure 1. Heart rate change with vagal stimulation followed by sympathetic stimulation A, records showing continuous tracings of left ventricular (LV) pressure and heart rate in a typical experiment during left vagal stimulation (VS) at 5 Hz followed by concomitant sympathetic stimulation (SS) at 2 Hz. B, plot of heart rate recorded in the experiment shown in A, before and during low, medium and high frequency SS at the following levels of background VS (from left to right panels) without VS (no VS), low, medium and high frequency stimulation, respectively.

5 132 K. E. Brack and others Exp Physiol 89.1 pp Figure 2. The effect of background vagal stimulation on the heart rate response to sympathetic stimulation A and B, graphs showing the effect of different frequencies of sympathetic stimulation (SS) with various levels of background left (LVS, A) and right (RVS, B) vagus nerve stimulation on steady state heart rate. P < with comparison between different frequencies of SS for each level of VS. C and D, graphs showing the increase in heart rate (HR) compared to pre-ss values, at different frequencies of SS with various levels of background LVS (C) and RVS (D). Data are mean ± S.E.M., n = 6, P < 0.05 with comparison between groups indicated by the vertical ticks. Figure 3. Actual versus calculated changes in heart rate during sympathetic stimulation with background vagal stimulation A and B, graphs showing the calculated (algebraic sum of the individual responses) and actual overall change in heart rate (HR) with sympathetic stimulation (SS) at various levels of background left (LVS, A) and right (RVS, B) vagal stimulation. Solid lines with solid symbols represent the actual change in HR and open symbols with dotted lines represent the calculated change in HR at each combination of frequencies. Data are mean ± S.E.M., n = 6, P < 0.05.

6 Exp Physiol 89.1 pp Sympathovagal interaction on heart rate 133 change did not reach statistical significance (P = 0.069). Figure 4D shows the corresponding data for background right VS. Again, significant decrease in latency period was observed with increasing levels of SS without VS and with all levels of background VS. Similar to the results with left VS, increasing levels of right VS prolonged latency period significantly with low and medium frequency SS whilst the change in latency period from 1.3 ± 0.2 ms (no VS) to 2.0 ± 0.4 ms (high frequency VS) with high frequency SS did not reach statistical significance (P = 0.060). Effects of background sympathetic nerve stimulation on the heart rate response to vagus nerve stimulation Steady state heart rate changes. The effect of background SS on the heart rate response to VS was examined in a similar manner to the above mentioned protocol in a reciprocal fashion, with SS first, achieving a steady state heart rate followed by VS. Figure 5A illustrates a typical experiment where SS (2 Hz) increased heart rate from about 150 bpm to 190 bpm. Left VS (5 Hz) then decreased heart rate to about 140 bpm. Cessation of autonomic stimulation caused a sharp rise of heart rate followed by a gradual return to prestimulation values. Figure 5B shows the data obtained for the experiment shown in Fig. 5A. The leftmost panel shows the change in instantaneous beat-to-beat heart rate at low, medium and high frequencies of left VS, in the absence of SS. The subsequent panels show the change in heart rate during different frequencies of left VS in the presence of background low, medium and high frequency SS. It can be appreciated from this experiment that the decrease in heart rate with left VS appeared not to be reduced, but rather potentiated by the presence of increasing levels of background SS. Figure 4. The effect of background vagus nerve stimulation on the delay in heart rate increase (latency) with sympathetic stimulation A, records showing continuous tracings of heart rate during low frequency sympathetic nerve stimulation (SS) without and with various levels of background left vagal stimulation (LVS). B, expanded scale showing the same data in A expressed as percentage change in heart rate from the start of SS. Horizontal line represents the 1% change in HR used to define latency. The vertical arrowhead lines indicate the latency time during SS at each level of LVS. C and D, graphs showing averaged data for latency at different frequencies of SS with various levels of background of left (LVS, C) and right (RVS, D) vagal stimulation, respectively. Data are mean ± S.E.M., n = 6, P < 0.05 with comparison between groups indicated by the vertical ticks.

7 134 K. E. Brack and others Exp Physiol 89.1 pp Figure 5. Heart rate change with sympathetic stimulation followed by vagal stimulation A, records showing continuous tracings of left ventricular (LV) pressure and heart rate to study the effects of background sympathetic stimulation (2Hz, SS) on the chronotropic effects of left vagal stimulation (VS, 5 Hz). B, plot of heart rate recorded in the experiment shown in A, before and during low, medium and high frequency VS, at the various levels of background SS.

8 Exp Physiol 89.1 pp Sympathovagal interaction on heart rate 135 Figure 6. The effect of background sympathetic stimulation on the heart rate response to vagal stimulation A and B, graphs showing the effect of various frequencies of left (LVS, A) and right (RVS, B) vagus nerve stimulation with various frequencies of background sympathetic nerve stimulation (SS) on steady state heart rate. P < with comparison between different frequencies of VS for each level of SS. C and D, graphs showing the change in heart rate (HR) compared to prevagus nerve stimulation, with LVS and RVS at different levels of background SS. Data are mean ± S.E.M., n = 6, P < 0.05 with comparison between groups indicated by the vertical ticks. Figure 7. Actual versus calculated changes in heart rate during vagal stimulation with background sympathetic stimulation A and B, graphs showing the calculated (algebraic sum of the individual responses) and actual overall change in heart rate (HR) with various frequencies of left (LVS, A) and right (RVS, B) vagus nerve stimulation at various frequencies of background sympathetic stimulation (SS). Solid lines with solid symbols represent the actual change in HR and open symbols with dotted lines represent the calculated change in HR at each combination of frequencies. Data are mean ± S.E.M., n = 6, P < 0.05.

9 136 K. E. Brack and others Exp Physiol 89.1 pp The mean data (n = 6) for the effects of background SS on steady state heart rate achieved with left VS are summarized in Fig. 6A. Similar results for right VS are shown in Fig. 6B. Both right and left VS caused significant (P < 0.001) frequency dependent decreases in heart rate in the absence and presence of background SS independent of stimulation frequency. Heart rates achieved with low and medium frequency VS were comparable between the two vagal nerves but were significantly (P < 0.05) lower during high frequency stimulation with right vagus nerve compared to the left, in the absence of and at all frequencies of background SS. Effects of sympathetic stimulation on the decrease in heart rate with vagal stimulation Figure 5B shows that the decrease in heart rate with left VS was potentiated in the presence of increasing levels of background SS in a typical experiment. This is also suggested in the mean data for steady state heart rates shown in Figs 6A and B for both left and right VS. The decrease in heart rate with low, medium and high frequencies of VS were calculated in each experiment at the various levels of background SS and the mean data plotted in Fig. 6C (left VS) and Fig. 6D (right VS). The enhancement in the ability of VS to decrease heart rate in the presence of increasing levels of SS is illustrated in the two figures. Low frequency left VS decreased heart rate by 20.5 ± 3.7 bpm without background SS and by 43.0 ± 6.2 bpm with high frequency SS (P < 0.001). High frequency left VS decreased heart rate by 65.7 ± 7.2 bpm without background SS and by ± 11.3 bpm with high frequency SS (P < 0.001). This potentiation of heart rate slowing at increasing frequencies of background SS was evident in both left and right VS. There was a trend for the heart rate change with right SS to be greater compared to the left, with and without background SS but this difference did not reach statistical significance. Algebraic sum of heart rate changes with vagal and sympathetic stimulation In Fig. 7A (left VS) and Fig. 7B (right VS), the open symbols represent the calculated algebraic sum for each combination of VS and SS whilst the filled symbols represent the experimentally obtained heart rate change for the corresponding combination. The experimentally obtained changes in heart rate during concomitant VS and SS at each combination of frequencies were all lower than the algebraic sum (P < 0.05 for all except low frequency left and right VS with background low frequency SS). Discussion In this study, the interactions between sympathetic and vagal stimulation on intrinsic heart rate in the isolated innervated rabbit heart were investigated. The resultant effect of stimulation of both autonomic branches on heart rate was different from the algebraic sum of the effect of stimulation of the individual branch, consistent with significant sympatho vagal interaction on heart rate. It was shown that the positive chronotropic effect of sympathetic stimulation was reduced in the presence of background vagal stimulation. On the other hand, the negative chronotropic effect of vagal stimulation was enhanced in the presence of background sympathetic stimulation. These results are in agreement with previous in vivo studies (Levy & Zieske, 1969; Revington & McCloskey, 1990). Data are also presented which suggest that the vagal inhibition of the chronotropic effects of sympathetic stimulation could involve both a presynaptic mechanism with the modulation of neurotransmitter release as well as postsynpatic mechanisms involving the parasympathetic modulation of sino-atrial pacemaker currents and signalling pathways. The innervated isolated heart preparation is an ideal in vitro model which allows the investigation of the complex mechanisms underlying the effects of sympatho vagal interactions on cardiac function. Effects of background vagus nerve stimulation on sympathetic stimulation The inhibition of the chronotropic effects of sympathetic stimulation by background vagus nerve stimulation occurred in a frequency dependent manner with increased inhibition at increasing frequency of vagus nerve stimulation. This occurred in a similar manner with left or right vagus nerve stimulation. The antagonism of the chronotropic effect of the sympathetic nerves by background vagal stimulation, which is in agreement with previous work by Levy & Zieske (1969), can be the result of pre- and postsynaptic mechanisms (Levy, 1971). The levels of noradrenaline (NA) released during sympathetic stimulation are reduced by acetylcholine (Loffelholz & Muscholl, 1969) binding to M 3 receptors on presynaptic sympathetic nerve terminals (Kobayashi et al. 1987; Manabe et al. 1991). Postsynaptic mechanisms relate to the muscarinic receptor mediated inhibition of adenylate cyclase activity (Hartzell, 1988) by the inhibitory G-protein, G i. The ion currents that are phosphorylated in a camp dependent manner and consequently inhibited in this way include the

10 Exp Physiol 89.1 pp Sympathovagal interaction on heart rate 137 hyperpolarization activated current (I f ), delayed rectifier current (I K ) and the l-type calcium channel (I Ca,L ). Effects on these currents will directly or indirectly affect pacemaker action potential generation and, hence, heart rate (Irisawa et al. 1993). The inhibition of adrenergically stimulated I Ca,L is nitric oxide/cgmp-dependent (Han et al. 1994, 1995), via cgmp-stimulated phosphodiesterase II (Han et al. 1998; Sasaki et al. 2000). In addition, vagal stimulation, with the release of acetylcholine, has direct electrophysiological effects on membrane potential via the acetylcholine-activated K current [I K(ACh) ], which would affect pacemaker action potential generation (Medina et al. 2000). Not only was the overall effect on heart rate blunted with background VS, but the time taken for sympathetic stimulation to produce a 1% increase in HR was also prolonged. To our knowledge, this is the first study to show that background vagal stimulation inhibited the rate of rise in heart rate produced with sympathetic stimulation. This prolongation of latency was only significant at low and medium frequency stimulation of the sympathetic nerves. But this was not the case during high frequency SS. At high SS, there was no significant change in latency in the presence of low, medium and high frequency VS, even though the overall effect on HR was profoundly inhibited. In this instance, sympathetic nerve stimulation was able to overcome the inhibition of the vagus nerve, at least with respect to latency. This suggests that the presynaptic release of noradrenaline by SS is sufficient at high SS but its postsynaptic effect cannot overcome the inhibition of vagal stimulation since the peak effect is blunted. One possible reason for this could be that during high frequency SS, stimulation of β-adrenergic system directly links to I Ca,L via the stimulatory G-protein (G α -subunit) as suggested by Yatani & Brown (1989). Perhaps more importantly, it is possible that muscarinic stimulation is not able to inhibit this direct G s -dependent stimulation of the I Ca,L, although it will still be capable of reducing camp-dependent stimulated ion currents involved in the antagonism of the overall chronotropic effect of SS. Further studies are required to delineate these mechanistic pathways. The effect of background sympathetic nerve stimulation on vagus nerve stimulation In contrast to the results seen with the inhibition of chronotropic effects of sympathetic stimulation in the presence of background vagal stimulation, this inhibition was not seen in the reverse direction. In this study, the negative chronotropic effect of vagus nerve stimulation was not inhibited but rather enhanced in the presence of background sympathetic stimulation. These results are in agreement with others (Revington & McCloskey, 1990; Yang & Levy, 1992) who showed that the dominant direction of inhibition was the vagal inhibition of the effects of sympathetic stimulation rather than the other way round. However, Potter (1985) showed in the anaesthetized dog in vivo that sympathetic nerve stimulation reduced the chronotropic effect (Potter, 1985) and time course of the heart rate change (Yang et al. 1994) during vagal stimulation. It was subsequently shown that this antagonism was due to the concurrent release of neuropeptide Y (Warner & Levy, 1989) binding to neuropeptide Y2 receptors (Smith-White et al. 2002) on presynaptic vagal neurones, inhibiting the release of ACh (Potter, 1987). This effect was not observed in the present study but as the frequency of background SS increased, the negative chronotropic effect of vagal stimulation increased, indicating the lack of inhibition of vagus nerve stimulation with background SS. Instead the chronotropic effect of VS was augmented. We interpret this as a combination of the direct effects of muscarinic receptor stimulation of I K(ACh) and the indirect inhibition of campdependent phosphorylated ion channels mentioned above by the inhibitory G i -protein. An alternative explanation could be that the intensity of sympathetic stimulation used in the current study, which was less than that used in the study by Potter (1985, 1987) and thus was insufficient to elicit the effects from neuropeptide Y (Lundberg & Hökfelt, 1986). We have previously shown in the innervated isolated heart preparation (Ng et al. 2001) that the decrease in heart rate with vagus nerve stimulation was instantaneous, with no significantlatency period asthat seen with sympathetic stimulation. In this study, there was no appreciable effect of background sympathetic stimulation on the time course of heart rate response with vagus nerve stimulation (results not shown). The effectiveness of left and right vagus nerve stimulation and accentuated antagonism From the present study it was shown that right vagus nerve stimulation was more potent in reducing heart rate than the left vagus nerve in the absence of sympathetic stimulation, in agreement with previous studies (Loeb et al. 1981; Lang & Levy, 1989; Ng et al. 2001). It was not possible, however, to show that right VS was more potent in reducing the effects of SS than left VS. Left and right VS reduced the effect of SS to a similar extent. Additionally, the decrease in heart rate from left VS and right VS, in the

11 138 K. E. Brack and others Exp Physiol 89.1 pp presence of background SS, was not significantly different, although there was a trend for right VS to produce a larger decrease in HR. This suggests that vagal stimulation predominates over sympathetic stimulation and exerts significant inhibition over sympathetic stimulation, an effect which was more obvious at increasing frequency of sympathetic stimulation and with increasing frequency of vagal stimulation. It is common practice in studies investigating heart rate, to stimulate left or right sided sympathetic and/or vagus efferent pathways to the heart. The right vagus nerve is usually chosen as it has been shown to have a larger chronotropic effect (Shipley & Greiser, 1945; Loeb et al. 1981). This suggests that there may be greater innervation of the sinoatrial node from the right vagus nerve (Ardell & Randall, 1986). Evidence from the rat agrees with this interpretation. Vagal fibres originating from the right dorsal motor nucleus and right nucleus ambiguus innervated the sinoatrial nodal region to a greater extent than the left, whereas fibres from the left dorsal motor nucleus and left nucleus ambiguus projected to a greater extent to the atrioventricular node region than the right side (Cheng et al. 1999; Cheng & Powley, 2000). Limitations and further studies In the present study, the whole sympathetic outflow from within the spinal cord was stimulated. This would have incorporated both left and right efferent pathways to the heart, whilst left and right vagus nerves were stimulated separately. This may have implications on some of the data described above whereby isolated right and left sided sympathetic stimulation may have produced different results. There is scope to extend the current studies to study the effect of right or left sympathetic stimulation in isolation either with stellate ganglion stimulation or bilateral stimulation before and after destruction of the stellate ganglion on one side. This will also allow the different ipsilateral and contralateral combinations of interactions between sympathetic and vagus nerve stimulation to be studied. The possibility of the involvement of other autonomic mediators (Potter & Ulman, 1994), namely neuropeptide Y, vasoactive intestinal polypeptide and substance P, in the interaction between sympathetic and vagal stimulation may also be studied with the use of specific inhibitors and also specific blockers for other recepetors. These studies are difficult to be performed in vivo in the intact animal due to the difficulty with controlling loading conditions and the presence of haemodynamic reflexes and circulating humoral factors; which can be circumvented with an isolated preparation. We have previously demonstrated the selectivity of the response to nerve stimulation in the innervated isolated heart preparation in that the effects were abolished with specific adrenergic and muscarinic antagonists (Ng et al. 2001). In addition, there was no tonic autonomic activity in the absence of nerve stimulation and there was no cross-contamination of responses during either sympathetic or vagal stimulation. The innervated isolated heart preparation thus hold enormous potential as an ideal in vitro model, allowing the investigation of the complex mechanisms underlying the effects of sympatho vagal interactions on cardiac function. Conclusions Results presented in this study have provided evidence that vagal stimulation has a dominant effect over sympathetic stimulation in the isolated heart. Background vagal stimulation inhibits the chronotropic effect of sympathetic nerve stimulation whilst the reverse is not true with the effects of vagal stimulation being enhanced in the presence of background sympathetic stimulation. This novel innervated isolated heart preparation is a valuable model to study the complex pathways involved in the interaction between the sympathetic and parasympathetic nerves on cardiac function. References Ardell JL & Randall WC (1986). Selective vagal innervation of sinoatrial and atrioventricular nodes in canine heart. Am J Physiol 251, H764 H773. Brack KE, Ng GA & Coote JH (2001). Sympatho vagal interaction on sinus rate in the isolated rabbit heart with intact dual autonomic innervation. J Physiol 531 (Suppl.), 182P 183P. Cheng Z & Powley TL (2000). Nucleus ambiuus projections to cardiac ganglia of rat atria: An anterograde tracing study. J Comp Neurol 424, Cheng Z, Powley TL, Schwaber JS & Doyle FJ (1999). Projections of the dorsal motor nucleus of the vagus to cardiac ganglia of rat atria: An anterograde tracing study. J Comp Neurol 410, Grodner AS, Lahrtz HG, Pool PE & Braunwald R (1970). Neurotransmitter control of sinoatrial pacemaker frequency in isolated rat atria and intact rabbits. Circ Res 27, Han X, Kobzik L, Severson D & Shimoni Y (1998). Characteristics of nitric oxide-mediated cholinergic modulation of calcium current in rabbit sino-atrial node. J Physiol 509, Han X, Shimoni Y & Giles WR (1994). An obligatory role for nitric oxide in autonomic control of mammalian heart rate. J Physiol 476,

12 Exp Physiol 89.1 pp Sympathovagal interaction on heart rate 139 Han X, Shimoni Y & Giles WR (1995). A cellular mechanism for nitric oxide-mediated cholinergic control of mammalian heart rate. J General Physiol 106, Hartzell HC (1988). Regulation of cardiac ion channels by catacholamines, acetylcholine and second messenger systems. Prog Biophys Mol Biol 52, Hoffmann F, Hoffmann EJ, Middleton S&Talesnik J (1945). Stimulating effect of acetylcholine on the mammalian heart and the liberation of an epinephrine like substance by the isolated heart. Am J Physiol 144, Irisawa H, Brown HF & Giles W (1993). Cardiac pacemaking in the sinoatrial node. Physiol Rev 73, Kobayashi O, Nagashima H, Duncalf D, Chaudhry IA, Harsing LGJ, Foldes FF, Goldiner PL & Vizi ES (1987). Direct evidence that pancuronium and gallamine enhance the release of norepinephrine from the atrial sympathetic nerve by inhibiting prejunctional muscarinic receptors. J Auton Nerv Syst 18, Lang SA & Levy MN (1989). Effects of the vagus nerve on heart rate and ventricular contractility in chicken. Am J Physiol 256, H1295 H1302. Levy MN (1971). Sympathetic Parasympathetic interactions of the heart. Circ Res 29, LevyMN,YangT&WallickDW(1993). Assessment of beat-by-beat control of heart rate by the autonomic nervous system: molecular biology techniques are necessary, but not sufficient. J Cardiovasc Electrophysiol 4, Levy MN & Zieske H (1969). Autonomic control of cardiac pacemaker activity and atrioventricular transmission. JApp Physiol 27, Loeb JM, Dalton DP & Moran JM (1981). Sensitivity differences of SA and AV node to vagal-stimulation attenuation of vagal effects at SA node. Am J Physiol 241, H684 H690. Loffelholz K & Muscholl E (1969). A muscarinic inhibition of the noradrenaline release evoked by postganglionic sympathetic nerve stimulation. Naunyn Schmiedebergs Arch Pharmakol 265,1 15. Lundberg JM & Hökfelt T (1986). Multiple co-existence of peptides and classical transmitters in peripheral autonomic and sensory neurons: functional and pharmacological implications. Prog Brain Res 68, Manabe N, Foldes FF, Torocsik A, Nagashima H, Goldiner PL & Vizi ES (1991). Presynaptic interaction between vagal and sympathetic innervation in the heart: modulation of acetylcholine and noradrenaline release. J Auton Nerv Syst 32, Medina I, Krapivinsky G, Arnold S, Kovoor P, Krapivinsky L & Clapham DE (2000). A switch mechanism for G beta gamma activation of I (KACh). J Biol Chem 275, Ng GA, Brack KE & Coote JH (2001). Effects of direct sympathetic and vagus nerve stimulation on the physiology of the whole heart a novel model of isolated Langendorff perfused rabbit heart with intact dual autonomic innervation. Exp Physiol 86, Potter EK (1985). Prolonged non-adrenergic inhibition of cardiac vagal action following sympathetic stimulation: neuromodulation by neuropeptide Y? Neur Lett 54, Potter E (1987). Presynaptic inhibition of cardiac vagal postganglionic nerves by neuropeptide Y. Neur Lett 83, Potter EK & Ulman LG (1994). Neuropeptides in sympathetic nerves affect vagal regulation of the heart. NIPS 9, Revington ML & McCloskey DI (1990). Sympathetic parasympathetic interactions at the heart, possibly involving neuropeptide Y, in anaesthetized dogs. J Physiol 428, Samaan A (1935). Antagonistic cardiac nerves and heart rate. J Physiol 83, Sasaki S, Daitoku K, Iwasa A & Motomura S (2000). NO is involved in MCh-induced accentuated antagonism via type II PDE in the canine blood-perfused SA node. Am J Physiol 279, H2509 H2518. Shipley RE & Greiser M (1945). Cardiac response to stimulation of the stellate ganglia and cardiac nerves. Am J Physiol 143, 396. Smith-White MA, Herzog H & Potter EK (2002). Role of neuropeptide Y Y(2) receptors in modulation of cardiac parasympathetic neurotransmission. Reg Pep 103, Warner MR & Levy MN (1989). Inhibition of cardiac vagal effects by neurally released and exogenous neuropeptide Y. Circ Res 65, YangT&LevyMN(1992). Sequence of excitation as a factor in sympathetic parasympathetic interactions in the heart. Circ Res 71, Yang T, Senturia JB & Levy MN (1994). Antecedent sympathetic stimulation alters time course of chronotropic response to vagal stimulation in dogs. Am J Physiol 266, H1339 H1347. Yatani A & Brown AM (1989). Rapid beta-adrenergic modulation of cardiac calcium-channel currents by a fast g-protein pathway. Science 245, Acknowledgement This work was supported by a British Heart Foundation Project Grant.