- Email: [email protected]
Respiratory modulation of sympathetic activity
Journal of the Autonomic Nervous System, 12 (1985) 77-90 Elsevier
77
JAN 00405
Respiratory modulation of sympathetic activity C.R. Bainton 1, D.W. Richter, H. Seller, D. Ballantyne a n d J.P. Klein 2 1. Physiologisches Institut der Universiti~t Heidelberg, INF 326, D -69 Heidelberg (F.R.G.), i Departments of Anesthesia and Physiolology, University of California, School of Medicine, San Francisco, CA 94143 and San Francisco Veterans Administration Medical Center, San Francisco, CA 94121 and 2 Department of Medicine, University of Texas, Medical Branch, Galveston, TX 77550 (U.S.A.) (Received June 6th, 1984) (Revised version received September 7th, 1984) (Accepted September 12th, 1984)
Key words: sympathetic activity - respiratory modulation - respiratory activity respiratory phases - brainstem organization
Abstract
Sympathetic activity recorded from cardiac and renal nerves was correlated with phrenic and internal intercostal nerve activity under normocapnea and hypercapnea. Cats were anesthetized with halothane for surgery switching to chloralose for recording. Both vagal and carotid sinus nerves were cut, animals were paralyzed and artificially ventilated. We found that sympathetic activity followed the rhythmic pattern of phrenic nerve discharge fairly closely except in two important respects: first, sympathetic activity was significantly depressed during early inspiration and second, it reached a minimum during post inspiration while phrenic activity was decaying but still active. These effects were accentuated when PA,,o, was raised. In one cat early inspiratory depression was the only manifestation-of respiratory modulation of sympathetic activity superimposed on an otherwise tonic pattern. In 4 cats sympathetic activity increased in an augmenting fashion in parallel with the augmenting discharge of expiratory alpha motoneurones. We suggest that respiratory-related, excitatory and inhibitory inputs modulate sympathetic activity at the brainstem level. Inspiratory and possibly expiratory interneurones may be the source of activation, and inhibitory inputs may derive from early inspiratory and postinspiratory interneurones. The inhibitory effects may be the only manifestation of respiratory modulation during strong tonic drive of the sympathetic activity. This paper is dedicated to Prof. Dr. H.-P. Koepchen on the occasion of his sixtieth birthday. This work has appeared in abstract form (see ref. 2). Correspondence: C. Bainton, Department of Anesthesia (129), VA Medical Center, 4150 Clement Street, San Francisco, CA 94121, U.S.A. 0165-1838/85/$03.30 ,~2 1985 Elsevier Science Publishers B.V. (Biomedical Division)
78 Introduction
Patterns of activity in sympathetic nerves tend to oscillate with a period equal to the respiratory cycle. This can be seen in animals either with intact or cut vagal and carotid sinus nerves. Although the origin and functional significance of these slow rhythms are not certain, it has been suggested that they derive from irradiation of the respiratory activity onto the brainstem system regulating sympathetic outflow (e.g. refs. 1, 5, 12, 17-19, 32, 35, 37, 38). However, the contour of integrated sympathetic activity is complex and seems to vary with experimental conditions. When compared with inspiratory activity in the phrenic nerve, it has led many investigators to postulate that sympathetic activity is activated by medullary inspiratory neurones [18-20,22,28,29], or may be activated by medullary expiratory-inspiratory phase-spanning neurones [5,9,10]. An alternative view is that sympathetic activity simply derives from an oscillator which is independent of, but entrainable to, the respiratory oscillator [4]. It is usually assumed that respiratory modulation of sympathetic activity originates solely from excitatory outputs from the respiratory network and little attention has been given to the possibility that this modulation might involve inhibitory mechanisms. Recently, Richter [30] and Ballantyne and Richter [3] stressed the importance for respiratory rhythmogenesis of at least two synaptic inhibitory events at the brainstem level. These inhibitory mechanisms act during early inspiration and during postinspiration (stage I expiration). Similar patterns of inhibition might, therefore, occur also in the brainstem system regulating sympathetic outflow. An indication for this is the 'early expiratory' inhibition which has been described for the activity of the abdominal chain [36]. Recent investigations have indicated that the respiratory rhythm consists of 3 distinct phases [30,31], i.e. inspiration, postinspiration, i.e. stage I expiration, and stage II expiration. The purpose of this work was to investigate the modulation of sympathetic activity during these respiratory phases which can be identified by phrenic and spinal expiratory nerve recordings. The experiments gave evidence for excitatory and inhibitory influences on sympathetic activity from the respiratory system.
Methods
Surgery and anesthesia We performed experiments in 5 cats (male and female, 2.3-3.9 kg). Anesthesia was induced with a mixture of halothane and 02 and maintained for the time needed for surgery. The trachea was cannulated and catheters placed in femoral artery and two veins for measurement of arterial blood pressure and administration of drugs. Carotid sinus nerves were ligated and cut bilaterally. The cats were mounted prone in a stereotaxic head holder and the spine supported with clamps at C 7 and L 5. We paralyzed cats with gallamine triethiodide and maintained ventilation with a Starling pump at a frequency of 20-36 per min and a tidal volume of 25-50 cc. Under
79
control conditions ventilation was adjusted to maintain PAco- at 25-50 mmHg. We used pure 02 to which CO 2 was added to achieve the desired ~evel of end-tidal PA,-o2" End-expiratory CO 2 was measured by a rapidly responding CO2-analyzer. An expiratory resistance (0.5-1.5 cm H 2 0 ) was used to prevent collapse of the lungs. Rectal temperature was maintained at 37-38°C using a heating blanket and an infrared lamp which were controlled by a feedback circuit. Four to five arterial blood samples were taken during the experiments and the H C O 3- concentration adjusted to 21 m e q . 1-1, based on a normal PAco.~ of 32.5 m m H g in the awake cat [11]. In all cats, 5% dextrose in Ringer solution was infused at a rate of 5 ml. kg -1. h -1. Blood pressure was maintained at a mean of 80-100 m m Hg, if necessary by infusing dopamine a n d / o r norepinephrine. The left stellate ganglion was identified by removing the costovertebral junctions and the initial portions of the ribs T 2 and T 3. The cardiac nerves were dissected free. The left external intercostal muscle was removed at T 5 and T 6. The internal intercostal nerve was identified and several filaments teased free from the internal intercostal muscle as described by Sears [33] and Sears and Stagg [34] starting with most dorsal filament available. Phrenic nerves were located bilaterally in the dorsolateral region of the neck, cut caudally and desheathed. The left flank of the animal was twisted and placed upward. Through a flank incision the left renal nerve was isolated retroperitoneally. All nerves selected for recording were cut, immersed in a pool of mineral oil, and central ends placed on bipolar platinum electrodes for recording. Both cervical vagi nerves were isolated, ligated and cut. Tests for baroreceptor denervation were done later in each cat with a bolus injection of norepinephrine ( 1 - 4 / ~ g . kg -1) into the femoral vein and in one cat by inflating a balloon catheter inserted in the descending aorta. There was no decrease in sympathetic or phrenic nerve activity in response to balloon inflation with an increase in blood pressure of 30 m m Hg or in response to norepinephrine infusion (blood pressure increase of at least 40 m m Hg). The final procedure was to make large bilateral pneumothoraces such that there was no observable chest movement during tidal volume inflation of the lung. With the conclusion of surgery, halothane was stopped and we continued anesthesia with alpha-chloralose (average cumulative dose, 30 m g . kg -1) to allow sympathetic tone to increase. Additional alpha-chloralose was given if blood pressure and heart rate increased in response to noxious stimulation of the hind paw or apneic threshold for PA..... dropped below 20 m m Hg.
Recordings We recorded activities from phrenic nerve (PN), internal intercostal nerve (Exp. Fil. Ts), renal nerve (RN) and cardiac nerve (CN). Nerve activities were filtered (80 H z - 1 0 kHz), differentially amplified and recorded on tape, a direct pen recorder and on a storage oscilloscope. Simultaneous periods in RN, CN, PN and Exp. Fil. T 5 were compared either by repeated overlap of single traces on a storage oscilloscope which was triggered at the onset or the end of the inspiratory component of phrenic nerve discharge and photographed by Polaroid, or averaged by a 4-channel signal averager. Phrenic nerve activity was used to produce two trigger signals. It was first passed through a 'leaky integrator' (time constant 0.1 s) or 'moving time averager'
80 using a bin width of 100 ms which delivered a 5 V pulse indicating both onset of inspiration (positive pulse) and cessation of inspiration, i.e. beginning of postinspiration (negative pulse). Either trigger could be used to start the simultaneous average of 4 separate neural signals for ' n ' respiratory periods of 2 or 4 s using bin widths of 2 or 4 ms. The averaged activities were then stored on a digital oscilloscope with magnetic disc memory system for later plotting on an X-Y plotter.
Data acquisition When the preparation was complete and the transition from halothane to chloralose anesthesia established, we began to record neural activities. Measurements were first made under normocapnea. PA,,o, was kept at 28-30 mm Hg for a period of 20-30 min and tape recordings were made at intervals of t0 min. The PA,~,2 was then elevated randomly by increments of 5 - 7 m m Hg up to a maximum of 53 m m Hg. Every change in PAso started from control levels and was held constant for 20-30 min. At each stable ~ O 2 level, we taped recordings from nerve activities and made repeated Polaroid photographs of storage trace overlays.
Results
Sympathetic activity was correlated to the 3 phases of the respiratory cycle [3,30,31] which were identified by simultaneous recordings from phrenic nerve and internal (expiratory) intercostal nerve filaments. Fig. 1 identifies these respiratory phases and shows the respiratory modulation of the cardiac sympathetic nerve activity. The phrenic nerve activity typically showed a ramp-like augmentation, i.e. inspiration, followed by a decrescendo of activity, i.e. postinspiration (stage I expiration), and then by a period of silence, i.e. stage II expiration. The expiratory intercostal nerve filament showed large and small spikes originating from alpha and gamma motoneurones [33]. Expiratory gamma motoneurones mostly fired continuously throughout the respiratory cycle but were modulated to a minimum during peak phrenic discharge. Expiratory alpha motoneurones on the other hand began to fire with an augmenting pattern when postinspiration ceased. They stopped promptly
EXPFIL 02mY[ ms (~) PN
o5mV[
CN
O02mVI
ls Fig. 1. Single sweep recording of activities from the internal intercostal filament (Exp. Fil. ]'s), phrenic nerve (PN) and cardiac sympathetic nerve (CN).
81 with the onset of the next burst of phrenic discharge. Thus, expiratory alpha motoneuronal activity arose only during complete silence of the phrenic nerve. Sympathetic activities can then be correlated with the (ramp) inspiratory and postinspiratory activity of the phrenic nerve or alternatively they can be observed during stage II expiration for evidence of activation or depression. Note that the cardiac sympathetic nerve is most active during peak phrenic discharge and has some but less activity during stage II expiration. The cardiac nerve is silent in early inspiration and during postinspiration. Fig. 2 shows these patterns of modulation of sympathetic activity in greater detail. A typical inspiratory augmenting discharge and a declining pattern of postinspiratory activity is seen in the phrenic nerve. Expiratory alpha motoneuronal activity is present solely during the silent period of the phrenic activity as described before. Both cardiac and renal activities exhibit respiratory modulation of their patterns with peak activity during the inspiratory period, less activity in stage II expiration and a clear lack in activity during early inspiration. Cardiac and renal patterns seem to differ in that the cardiac nerve is absolutely silent during postinspiration whereas the renal nerve continues to discharge throughout this period. In Fig. 3, multiple oscilloscopic sweeps were superimposed which were triggered by the end or the start of the inspiratory component of the phrenic nerve discharge, each showing the activities during one respiratory cycle. Although respiratory modulation is strong in both sympathetic nerves, cardiac and renal (Fig. 2) sympathetic nerves were found to be quite different. In both sympathetic nerves activation follows the contour of late inspiratory activity, i.e. sympathetic activity was strong during the late inspiratory phase. Sympathetic activity steadily increased throughout the period of stage II expiration in the cardiac nerve, but a similar augmentation of activity seems to be absent in the renal nerve. Postinspiratory depression was observed in cardiac nerves, but not so regularly in renal nerves. A typical finding, however, was a clear depression of sympathetic activity during early inspiration in both cardiac and renal nerves. This seems to be remarkable for two reasons. First, it could occur in the renal nerve despite a high level of tonic activity which may obscure other phases of respiratory modulation like a possible expiratory
CN
0.2rnV[
RN
O.OSmVE
- r " ~ ~' l ~ ~ EXP. FIL. T5
0.2mVl- ,.,,.,,,~ . , . . . .
PN
O.SmVE
.... ~ J I
,,r T
II ~
'" , ~ r r r ~
," ~,,,i,N1~,1,rf~'w . . . . .
""'~'"'!,
J ls
Fig. 2. Recordings from the cardiac sympathetic nerve (CN), renal sympathetic nerve (RN), internal intercostal filament (Exp. Fil. T5) and phrenic nerve (PN) recorded on a pen recorder.
82
activation and postinspiratory depression of activity and second, it was seen in the cardiac nerves despite the inevitable 'smearing' when activities were superimposed at a long delay after the trigger pulse had started the sweep. This 'smearing' can be seen in Fig. 3A by the 0.3 s overlap of inspiratory onset of phrenic activity and expiratory termination. It is simply a consequence of the superposition of sweeps and is caused by the spontaneous variation in the frequency of the central respiratory rhythm. The heterogeneity of sympathetic activities emphasizes the need for determining the probability of activation of sympathetic neurones more precisely, i.e. it is necessary to average the data over several respiratory cycles at different levels of PA,,," However, as seen in Fig. 3, averaging must deal with the problem of the spontaneously occurring variations in the duration of a respiratory cycle which could vary as much as one second. Any correlation of averaged data, therefore, will be accurate only close to the trigger time but will become progressively 'smeared' as the delay after the trigger increases. Averages started at the two points of respiratory phase transition, therefore, require some optimum number of sweeps to produce a faithful representation of overall contour of activity but with minimal loss of detail. Our empirical approach to this problem is seen in Fig. 4. Renal nerve activity was integrated and averaged from the end of the inspiratory phrenic burst for 15, 30 and 60 consecutive breaths. This averaging was done over the same portion of the tape recording with the results seen here. Note that the contour of averaged integrations of sympathetic activity is similar for all 3 (15, 30 and 60) sets of sweeps at the onset of triggering, i.e. the point of greatest accuracy. But note also that considerable detail is lost for averages with a sweep number greater than 15, as the averaged pattern extends to the right and thus away from the trigger time. Thus, a number of 15 sweeps was assumed to faithfully reproduce the sympathetic pattern although the resulting curves may not be as smooth as with larger sweep numbers. We, therefore, chose 15 sweeps as the optimum number for averaging our data.
A EXP FIL T5 02mV[ (~i
B EXIDFIL T5 02mV[ (I)
PN OSmV[
PN O.5mV[
EN 0.02mY[
RN 095rnV[
1s 1s Fig. 3. Activities from internal intercostal filament (Exp. Fil. T s) and phrenic nerve (PN) together with cardiac nerve (A) and renal nerve (B). 15 sweeps were superimposed on a storage oscilloscope which was triggered from peak phrenic activity (A) and the start of phrenic nerve activity (B). The sweeps were triggered at different points in phrenic activity to increase the precision of observation. (Precision is greatest close to the trigger and declines at a distance due to the spontaneous variation in inspiratory and expiratory duration.)
83
Fig. 5 reveals information as to the possible origin of this depression in sympathetic activity. Averaged cardiac nerve activities (CN~_3) of different strength from 3 cats are presented at the same amplification. The C N 3 t r a c e of cardiac nerve activity is of particular interest for two reasons: first, this cat had a very high level of tonic sympathetic drive as seen by the height of the line connecting sympathetic activity with the zero level of the integrator output and second, at normal levels of P",~o2 (28-30 mm Hg) in this cat there was no modulation of sympathetic activity with respiration. Only when the PAco2.was elevated to 53 mm Hg, as shown here, did a clear respiratory modulation of activity appear. This modulated pattern, however, did not follow late inspiratory and stage II expiratory activation, instead depression of sympathetic activity during early inspiration was the only recognizable pattern. Since CO 2 activated central respiratory activity as measured in the phrenic nerve discharge (not illustrated) and produced a clear respiratory modulation of tonically enhanced sympathetic activities, we suggest that bulbar respiratory neurones were responsible for the depression of sympathetic activity. We further speculate that it exceeds all excitatory respiratory inputs in effectiveness because it was the only manifestation of rhythmic modulation of intense tonic sympathetic activity. A depression of activity can occur during postinspiration as shown in Fig. 3. This deviation from the contour of phrenic discharge suggests that inhibition might be present. Further evidence for the possibility of a postinspiratory inhibition of sympathetic activity is presented in Fig. 6. This cat inhaled CO 2 raising PAco2 from 35 to 53 mm Hg. Elevated CO 2 markedly increased peak phrenic activity without an obvious change in postinspiratory contour (not illustrated). In both renal and cardiac nerves sympathetic activity was increased during peak phrenic discharge. More impressive, however, was the fact that the sympathetic activity contour was
RN
3o 60
I
Is Fig. 4. Averaged patterns of phrenic nerve (PN) and renal nerve (RN) activities. The averages were triggered from the end of the inspiratory component of the phrenic nerve activity for 15, 30 and 60 consecutive breaths. Repeated patterns for 15, 30 and 60 averages are offset for visual comparison and are connected by perpendicular markers to their respective baselines. Baseline indicates zero potential.
84
altered during the postinspiratory period. Sympathetic activity was reduced more sharply, and achieved a more sustained low level before the activity started to steadily increase again during the stage II expiration. We suggest that an inhibitory input to the sympathetic network might be responsible for this depression in sympathetic activity during postinspiration. A summary of sympathetic patterns compared with inspiratory, postinspiratory and expiratory activity is as follows: in 3 of 5 cats, sympathetic activity both in cardiac and renal nerves increased and decreased in responses to an increase and decrease in inhaled CO 2. Sympathetic activity roughly mimicked the late inspiratory and augmenting expiratory activity. In all cats, we observed a depression in sympathetic activity during early inspiration and there were indications for a
PN ~
.
_
~.__,,_.,,,_._~
EN~ [N2 [Na
ls Fig. 5. Averaged activity of cardiac nerve (CN) activity in 3 cats. All activity is triggered from onset of phrenic nerve discharge. Averaged activities of phrenic and cardiac nerve activity (CN 1) shown in the upper two traces are from the same cat. Averaged sympathetic activities (CN 2 ~) from the other two cats are adjusted to the same phrenic time scaling and displaced downward for ease in comparing sympathetic contours. The time base therefore refers only to the phrenic and the top cardiac nerve trace. Perpendicular lines connect each averaged sympathetic activity to its basel ne activity. PA for each cat beginning at ('°2 1 the top are 28, 30 and 53 Torr. Phrenic contours after scaling for time were essentially identical for al cats.
85
PN CN I
[N 2
RN1
RN2
I ] S Fig. 6. Averaged activities from phrenic nerve (PN), cardiac nerve (CN) and renal nerve (RN) in one cat. All activity was triggered with the end of the inspiratory discharge of the phrenic nerve. Perpendicular lines connect sympathetic activities with their respective baseline. The P^co for the traces of CNl and RN1 sympathetic patterns was 35 Torr and 53 Tort for the traces CN 2 an~ RN2. Expiratory duration estimated by the phrenic and internal intercostal nerve activities was not changed by this elevation of CO2 and there was no need for re-adjustment of time scales. Elevated CO2 markedly increased peak phrenic activity without an obvious change in postinspiratory contour.
depression d u r i n g postinspiration. I n h a l e d C O 2 depressed sympathetic activity in two cats u n d e r two circumstances while phrenic activity was enhanced. I n one cat this occurred d u r i n g early inspiration which was the only m a n i f e s t a t i o n of a respiratory coupled r h y t h m of a high tonic b a c k g r o u n d in sympathetic activity. In the other case, this appeared as a depression a n d s h a r p e n i n g of the fall of sympathetic activity d u r i n g the postinspiratory period.
Discussion This work shows that sympathetic activity is not simply related to the overall c o n t o u r of phrenic activity, b u t exhibits a p r o n o u n c e d depression both in early
86 inspiration and during postinspiration. Moreover, the averaged contour of sympathetic activity frequently follows the steadily increasing pattern of expiratory activity in alpha motoneurones. One possible implication which can be drawn from this work is that the respiratory modulation of sympathetic activity involves several components of the behaviour of propriobulbar and bulbospinal respiratory neurones. This inference raises a number of issues regarding the functional organization of the brainstem systems regulating respiratory and sympathetic activities. What evidence exists that respiratory neurones project a rhythm onto sympathetic networks at the brainstem level?
Sympathetic nerves exhibit slow modulated activity which is synchronized with the respiratory rhythm. The patterns of respiratory modulation of sympathetic activity are complex and variable suggesting that multiple inputs are involved. This makes the analysis of the relationship between respiratory and sympathetic systems difficult and tentative. Many authors have reported that peak sympathetic activity is correlated with peak phrenic activity and they have suggested that the pattern is derived from inspiratory neurones [14,17-20,23,28,29,35,36] or expiratory-inspiratory phase spanning neurones [5,9,10]. In the work of Preiss, Kirchner and Polosa [29] the activity of single filaments of the cervical sympathetic nerve is so remarkable similar to phrenic activity in its timing and profile that it seems difficult to believe that two such virtually identical activity patterns could be generated by independent systems. It seems more reasonable that the rhythmic component of their activity derives from one system (see ref. 18). Added to this is the parallel response of sympathetic and phrenic nerves to hypocapnea and hypercapnea [5,9,14,18,23,28,29], loss of the respiratory component of the sympathetic rhythm during hypocapnic apnea [20,28,29] and the recent evidence of Kubin, Trzebski and Lipski [21] and Connelly and Wurster ]6] that in the cat a midline lesion in the medulla at the level of the obex, disrupts both phrenic activity and rhythmic patterns of sympathetic activity but leaves pressor responses intact and sympathetic nerves accessible to ipsilateral stimulation of pontine parabracheal areas. There is, however, disagreement as to the imprint of respiratory neurones on the brainstem system regulating sympathetic outflow. Clearly, the activity patterns of phrenic and sympathetic nerves are not always identical. Barman and Gebber [4] raise concern that the sympathetic pattern can preceed~ be simultaneous with~ or lag behind the onset of phrenic nerve activity. Their data to show, however, that the offset of phrenic and sympathetic patterns is essentially the same. Differences in tonic sympathetic background, varying activation thresholds and varying medullary respiratory drive could account for differences in onset of activity. Barman's and Gebber's [4] evidence that sympathetic rhythms can occur with a respiratory-like frequency independent of the phrenic rhythm does not necessarily conflict with our results. Clearly, respiratory influence on sympathetic is variable. From our own data, differences in the pattern of cardiac and renal nerve activities indicate that there is a possibility for differentiated respiratory-sympathetic interaction and that a high level of tonic sympathetic background may swamp all facilitatory influences from the respiratory system. Since there is at present no direct evidence for a relationship
87 between the two systems, it would seem desirable to refine comparisons in view of the probability that sympathetic activity is indeed influenced by excitatory and inhibitory respiratory neurones at brainstem level.
Do inhibitory respiratory neurones exist which might play a role in determining sympathetic patterns? In recent publications the likely importance for normal rhythmogenesis of early inspiratory, and postinspiratory inhibition to the network interactions among respiratory cells in the medulla has been stressed [3,30,31]. This inhibition may originate in propriobulbar respiratory interneurones, two classes of which have been identified, the 'early inspiratory' [25] and the 'post inspiratory' neurons [30,31]. The potentially significant feature of both classes of neurones is their divergent inhibitory output to many types of medullary respiratory neurones. It was also shown that tonic, apparently non-respiratory, reticular neurones in the medulla are also accessible to postinspiratory inhibition [30,31]. This finding raises the possibility that these interneurones exert a generalized inhibitory influence on other systems. One candidate might be the brainstem system regulating sympathetic outflow which is known to have a respiratory modulation. Examination of sympathetic mass-activity in relation to peripheral respiratory activity constitutes an indirect way of identifying these two classes of inhibitory events in the sympathetic system. Do other authors report eoidence for inhibitory sculpturing of rhythmic sympathetic patterns? The work of Seller and Richter [36] provides evidence for postinspiratory inhibition seen on postganglionic renal sympathetic nerves in dogs. Their results showed that the level of sympathetic activity, increased synchronously with phrenic activity, was completely inhibited during postinspiration and maintained a nearly constant level during the remainder of expiration. However, these results do not exclude the possibility that total suppression of sympathetic discharge was some form of postexcitatory depression following inspiratory activation as suggested by Okada and Fox [27]. Reasoning from the work of Preiss et al. [29], mass-activity in whole sympathetic nerves is composed of fibres, 70% of which has rhythmic respiratory pattern and 30 per cent of which is tonic. The occurrence of such a clear void in all sympathetic mass-activity, normally showing considerable tonic activity, might best be explained by active inhibition of the tonically active structures in the brainstem regulating sympathetic outflow. Data from the work of J~mig et al. [14] hint that the respiratory patterns generated in sympathetic activity does not simply follow inspiratory excitation but may also be under the influence of inhibitory respiratory neurones. Their Fig. 1A-C, indicate that cutaneous vasoconstrictor sympathetic activities in the cat can have essentially tonic activity without strong respiratory modulation during normocapnea and hypocapnea. Hypercapnea brings out a well-defined sawtooth pattern, synchronous with the phrenic discharge with a maximum activity greater than, and a minimum activity less than the tonic activation seen in normocapnea and hypocapnea. The minimum activity level occurs in the immediate postinspiratory period, the maxi-
88 mum activity coincides with the maximum of phrenic nerve discharge with a linear increase from minimum to maximum. Several studies [5,9,14,18-20,23,28] clearly show that sympathetic activity, coincident with phrenic activity, increases in response to an enhanced CO 2 drive. Sympathetic activity can also decrease with hypercapnea [14]. A possible explanation for this decrease is that it results from the hypercapnic activation of respiratory inhibitory interneurones. In the same article, J~mig et al. [14] give a second example of possible sympathetic inhibition in the postinspiratory period. This is seen in their Fig. 3D-F2 in response to hypercapnea. This example is reminiscent of our experience (Fig. 6) in which a gradual decrescendo of sympathetic activity in the postinspiratory period is converted to an abrupt decrease in response to hypercapnea without a change in phrenic postinspiratory activity. There are some hints that sympathetic activities are subject to inhibition during early inspiration as well. Gootman and Cohen [9] described averaged sympathetic activity triggered from onset of phrenic discharge. Fig. 2 of their work shows a distinct depression of sympathetic activity during the early inspiratory phase. What are the implications of this work?
Our comparison of respiratory and sympathetic activities provides us with a window to the functional organization of the brainstem system regulating sympathetic outflow. This comparison, in conjunction with the work of other investigators, would indicate that sympathetic rhythmic patterns, phase-locked to respiratory patterns, do originate in elements of the bulbar respiratory network, i.e. activation by bulbar inspiratory neurones and inhibition by early inspiratory and postinspiratory interneurones which do not project to the spinal cord. There are at least two possible, non-exclusive, sorts of explanation for the relationship between sympathetic activity and stage II expiratory activity: the first is that the augmentation of sympathetic activity during stage II expiration reflects slow recovery from postinspiratory inhibition superimposed on a tonic level of activity; the second explanation is that the steadily augmenting sympathetic activity during stage II expiration derives from facilitation by expiratory neurones. If this interpretation is true it implies that bulbar interneurones with a respiratory activity can project or even belong to different systems, e.g. to the respiratory system and to the system regulating sympathetic outflow. In the parasympathetic division of the autonomic nervous system there is also evidence for various sorts of respiratory influence. Cardiac vagal motoneurones show an expiratory related discharge [7,13,15,16,22,24,26] and intracellular analysis revealed that they are activated during postinspiration, strongly inhibited during inspiration and weakly inhibited during stage II expiration [8].
Acknowledgements The authors wish to thank Ms. Annemarie Bischoff and Ms. Anita Ki~hner for their technical assistance in experiments and their help in preparing the manuscript and illustrations. This work was supported by DFG, Grant Ri. 279/7-7; SFB 90, USPHS GM-15571 and SFVAMC.
89
References 1 Adrian, E.D., Bronk, D.W. and Phillips, G., Discharges in mammalian sympathetic nerves, J. Physiol. (Lond.) 74 (1932) 115-153. 2 Bainton, C.R., Richter, D.W., Seller, H., Ballantyne, D. and Klein, J.P., Respiratory modulation of sympathetic activity, Fed. Proc., 41 (1983) 1125. 3 Ballantyne, D. and Richter, D.W., Post-synaptic inhibition of bulbar inspiratory neurons in the cat, J. Physiol. (Lond.), 348 (1984) 67-87. 4 Barman, S.M. and Gebber, G.L., Basis for synchronization of sympathetic and phrenic nerve discharges, Amer. J. Physiol., 231 (1976) 1601-1607. 5 Cohen, M.I. and Gootman, P.M., Periodicities in efferent discharge of splanchnic nerve of the cat, Amer. J. Physiol., 218 (1970) 1092-1101. 6 Connelly, C.A. and Wurster, R.D., Effects of midline medullary lesions on the respiratory modulation of sympathetic nerve activity, Fed. Proc., 43 (1984) 402. 7 Davidson, N.S., Goldner, S. and McCloskey, D.I., Respiratory modulation of baroreceptor and chemoreceptor reflexes affecting heart rate and cardiac vagal efferent nerve activity, J. Physiol. (Lond.), 259 (1976) 523-530. 8 Gilbey, M.P., Jordan, D., Richter, D.W. and Spyer, K.M., Synaptic mechanisms involved in the inspiratory modulation of vagal cardio-inhibitory neurones in the cat, J. Physiol. (Lond.) in press. 9 Gootman, P.M. and Cohen, M.I., The interrelationships between sympathetic discharge and central respiratory drive, In W. Umbach and H.-P. Koepchen (Eds.), Central-Rhythmic and Regulation, Hippokrates-Verlag, Stuttgart, 1974 pp. 195-209. 10 Gootman, P.M., Cohen, M.I., Piercey, M.P. and Wolotsky, P., A search for medullary neurons with activity patterns similar to those in sympathetic nerves, Brain Res., 87 (1975) 395-406. 11 Herbert, D. and Mitchell, R.A., Blood gas tensions and acid-base balance in awake cats, J. Appl. Physiol, 30 (1971) 434-436. 12 Hering, E., l~lber den Einfluss der Athmung auf den Kreislauf. 1. Ober Athembewegungen Gef~tsssystems, Sber. Akad. Wiss., Wien, Math. Nat. KI., II. Abtl., 60 (1869) 829-856. 13 lriuchijima, J. and Kumada, M., Activity of single vagal fibres efferent to the heart, Jap. J. Physiol., 14 (1964) 479-487. 14 J~nig, W., Ki~mmel, H. and Wiprich, L., Respiratory rhythmicities in vasoconstrictor and sudomotor neurones supplying the cats' hindlimb. In H.-P. Koepchen, S.M. Hilton and A. Trzebski (Eds.), Central Interaction Between Respiratory and Cardiovascular Control Systems, Springer-Verlag, Berlin, Heidelberg, New York, 1980, pp. 128-136. 15 Jewett, D.L., Activity of single efferent fibres in the cervical vagus nerve of the dog with special reference to possible cardioinhibitory fibres, J. Physiol. (Lond.), 175 (1964) 321-357. 16 Katona, P.G., Poitras, J.W., Barnett, G.O. and Terry, B.S., Cardiac vagal efferent activity and heart period in the carotid sinus reflex, Amer. J. Physiol., 218 (1970) 1030-1037. 17 Koepchen, H.-P., Die Bhitdruckrhythmik, D. Steinkopff Verlag, Darmstadt, 1962. 18 Koepchen, H.-P., Respiratory and cardiovascular "centres": Functional entirety or separate structures?. In M.E. SchlMke, H.-P. Koepchen and W.R. See (Eds.), Central Neurone Environment, Springer Verlag, Berlin, Heidelberg, 1983, pp. 221-237. 19 Koepchen, H.-P. Kliissendorf. D. and Sommer, D., Neurophysiological background of central neural cardiovascular-respiratory coordination: basic remarks and experimental approach, J. auton. Nerv. Syst. 3 (1981) 335-368. 20 Koizumi, K., Seller, H., Kaufman, A. and McC Brooks, C., Pattern of sympathetic discharges and their relation to baroreceptor and respiratory activities, Brain Res., 27 (1971) 281-294. 21 Kubin, L., Trzebski, A. and Lipski, J., Search for the site of the central respiratory-sympathetic synchronisation, Fed. Proc., 38 (1979) 1200. 22 Kunze, D.L., Reflex discharge patterns of cardiac vagal efferent fibres, J. Physiol. (Lond.), 222 (1972) 1-15. 23 Lipski, J., Coote, J.H. and Trzebski, A., Temporal patterns of antidromic invasion latencies of sympathetic preganglionic neurons related to central inspiratory activity and pulmonary stretch receptor reflex, Brain Res., 135 (1977) 162-166.
90 24 McAllen, R.M. and Spyer, K.M., Two types of vagal preganglionic motoneurones projecting to the heart and lungs, J. Physiol. (Lond.), 282 (1978) 353-364. 25 Merrill, E.G., Finding a respiratory function for the medullary respiratory neurons, In R. Bellairs and E.G. Gray (Eds.), Essays on the Nervous System, Clarendon Press, Oxford, 1974, pp. 451-486. 26 Neil, E. and Palmer, J.F., Effects of spontaneous respiration on the latency of reflex cardiac chronotropic responses to baroreceptor stimulation, J. Physiol. (Lond.), 247 (1975) 16p. 27 Okada, H. and Fox, l.J., Respiratory grouping of abdominal sympathetic activity in the dog, Amer. J. Physiol., 213 (1967) 48-56. 28 Polosa, C., Gerber, U. and Schondorf, R., Central mechanisms of interaction between sympathetic preganglionic neurons and the respiratory oscillator, In H.-P. Koepchen, S.M. Hilton and A. Trzebski (Eds.), Central Interaction Between Respiratory and Cardiovascular Control Systems, Springer Verlag, Berlin, Heidelberg, New York, 1980, pp. 137-143. 29 Preiss, G., Kirchner, F., and Polosa, C., Patterning of sympathetic preganglionic neuron firing by the central respiratory drive, Brain Res., 87 (1975) 363-374. 30 Richter, D.W., Generation and maintenance of the respiratory rhythm, J. exp. Biol., 100 (1982) 93-107. 31 Richter D.W. and Ballantyne, D., A three phase theory about the basic respiratory pattern generator, In M.E. Schl~fke, H.-P. Koepchen and W.R. See (Eds.), Central Neurone Environment, Springer Verlag, Berlin, Heidelberg, New York, 1983, pp. 163-174. 32 Schweitzer, A., Die Irradiation autonomer Reflexe, S. Karger Verlag, Basel, 1937. 33 Sears, T.A., Efferent discharge in alpha and fusimotor fibres of intercostal nerves of the cat, J. Physiol. (Lond.), 174 (1964) 295-315. 34 Sears, T.A. and Stagg, D., Short term synchronization of motoneurone activity, J. Physiol. (Lond.), 263 (1976) 357-381. 35 Seller, H., Langhorst, P., Richter, D. and Koepchen, H.-P., 13ber die Abhangigkeit der pressoreceptoriscben Hemmung der Sympathicus von der Atemphase und ihre Auswirkung in der Vasomotorik, Pfliagers Arch. ges. physiol., 302 (1968) 300-314. 36 Seller, H. and Richter, D.W., Some quantitative aspects of the central transmission of the baroreceptor activity. In F.F. Kao, K. Koizumi and M. Vassalle (Eds.), Research in Physiology. A liber memorialis in honor of Prof. C. McC Brooks, Aulo Gaggi, Bologna, 1971, pp. 541-549. 37 Tang, P.C., Maire, F.W. and Amassian, V.E., Respiratory influence on the vasomotor center. Amer. J Physiol., 191 (1957) 218-224. 38 Traube, D., Ober periodische T~tigkeits-~usserungen des vasomotorischen und Hemmungs Nervencentrums, Cbl. med. Wiss., 56 (1865) 881-885.
77
JAN 00405
Respiratory modulation of sympathetic activity C.R. Bainton 1, D.W. Richter, H. Seller, D. Ballantyne a n d J.P. Klein 2 1. Physiologisches Institut der Universiti~t Heidelberg, INF 326, D -69 Heidelberg (F.R.G.), i Departments of Anesthesia and Physiolology, University of California, School of Medicine, San Francisco, CA 94143 and San Francisco Veterans Administration Medical Center, San Francisco, CA 94121 and 2 Department of Medicine, University of Texas, Medical Branch, Galveston, TX 77550 (U.S.A.) (Received June 6th, 1984) (Revised version received September 7th, 1984) (Accepted September 12th, 1984)
Key words: sympathetic activity - respiratory modulation - respiratory activity respiratory phases - brainstem organization
Abstract
Sympathetic activity recorded from cardiac and renal nerves was correlated with phrenic and internal intercostal nerve activity under normocapnea and hypercapnea. Cats were anesthetized with halothane for surgery switching to chloralose for recording. Both vagal and carotid sinus nerves were cut, animals were paralyzed and artificially ventilated. We found that sympathetic activity followed the rhythmic pattern of phrenic nerve discharge fairly closely except in two important respects: first, sympathetic activity was significantly depressed during early inspiration and second, it reached a minimum during post inspiration while phrenic activity was decaying but still active. These effects were accentuated when PA,,o, was raised. In one cat early inspiratory depression was the only manifestation-of respiratory modulation of sympathetic activity superimposed on an otherwise tonic pattern. In 4 cats sympathetic activity increased in an augmenting fashion in parallel with the augmenting discharge of expiratory alpha motoneurones. We suggest that respiratory-related, excitatory and inhibitory inputs modulate sympathetic activity at the brainstem level. Inspiratory and possibly expiratory interneurones may be the source of activation, and inhibitory inputs may derive from early inspiratory and postinspiratory interneurones. The inhibitory effects may be the only manifestation of respiratory modulation during strong tonic drive of the sympathetic activity. This paper is dedicated to Prof. Dr. H.-P. Koepchen on the occasion of his sixtieth birthday. This work has appeared in abstract form (see ref. 2). Correspondence: C. Bainton, Department of Anesthesia (129), VA Medical Center, 4150 Clement Street, San Francisco, CA 94121, U.S.A. 0165-1838/85/$03.30 ,~2 1985 Elsevier Science Publishers B.V. (Biomedical Division)
78 Introduction
Patterns of activity in sympathetic nerves tend to oscillate with a period equal to the respiratory cycle. This can be seen in animals either with intact or cut vagal and carotid sinus nerves. Although the origin and functional significance of these slow rhythms are not certain, it has been suggested that they derive from irradiation of the respiratory activity onto the brainstem system regulating sympathetic outflow (e.g. refs. 1, 5, 12, 17-19, 32, 35, 37, 38). However, the contour of integrated sympathetic activity is complex and seems to vary with experimental conditions. When compared with inspiratory activity in the phrenic nerve, it has led many investigators to postulate that sympathetic activity is activated by medullary inspiratory neurones [18-20,22,28,29], or may be activated by medullary expiratory-inspiratory phase-spanning neurones [5,9,10]. An alternative view is that sympathetic activity simply derives from an oscillator which is independent of, but entrainable to, the respiratory oscillator [4]. It is usually assumed that respiratory modulation of sympathetic activity originates solely from excitatory outputs from the respiratory network and little attention has been given to the possibility that this modulation might involve inhibitory mechanisms. Recently, Richter [30] and Ballantyne and Richter [3] stressed the importance for respiratory rhythmogenesis of at least two synaptic inhibitory events at the brainstem level. These inhibitory mechanisms act during early inspiration and during postinspiration (stage I expiration). Similar patterns of inhibition might, therefore, occur also in the brainstem system regulating sympathetic outflow. An indication for this is the 'early expiratory' inhibition which has been described for the activity of the abdominal chain [36]. Recent investigations have indicated that the respiratory rhythm consists of 3 distinct phases [30,31], i.e. inspiration, postinspiration, i.e. stage I expiration, and stage II expiration. The purpose of this work was to investigate the modulation of sympathetic activity during these respiratory phases which can be identified by phrenic and spinal expiratory nerve recordings. The experiments gave evidence for excitatory and inhibitory influences on sympathetic activity from the respiratory system.
Methods
Surgery and anesthesia We performed experiments in 5 cats (male and female, 2.3-3.9 kg). Anesthesia was induced with a mixture of halothane and 02 and maintained for the time needed for surgery. The trachea was cannulated and catheters placed in femoral artery and two veins for measurement of arterial blood pressure and administration of drugs. Carotid sinus nerves were ligated and cut bilaterally. The cats were mounted prone in a stereotaxic head holder and the spine supported with clamps at C 7 and L 5. We paralyzed cats with gallamine triethiodide and maintained ventilation with a Starling pump at a frequency of 20-36 per min and a tidal volume of 25-50 cc. Under
79
control conditions ventilation was adjusted to maintain PAco- at 25-50 mmHg. We used pure 02 to which CO 2 was added to achieve the desired ~evel of end-tidal PA,-o2" End-expiratory CO 2 was measured by a rapidly responding CO2-analyzer. An expiratory resistance (0.5-1.5 cm H 2 0 ) was used to prevent collapse of the lungs. Rectal temperature was maintained at 37-38°C using a heating blanket and an infrared lamp which were controlled by a feedback circuit. Four to five arterial blood samples were taken during the experiments and the H C O 3- concentration adjusted to 21 m e q . 1-1, based on a normal PAco.~ of 32.5 m m H g in the awake cat [11]. In all cats, 5% dextrose in Ringer solution was infused at a rate of 5 ml. kg -1. h -1. Blood pressure was maintained at a mean of 80-100 m m Hg, if necessary by infusing dopamine a n d / o r norepinephrine. The left stellate ganglion was identified by removing the costovertebral junctions and the initial portions of the ribs T 2 and T 3. The cardiac nerves were dissected free. The left external intercostal muscle was removed at T 5 and T 6. The internal intercostal nerve was identified and several filaments teased free from the internal intercostal muscle as described by Sears [33] and Sears and Stagg [34] starting with most dorsal filament available. Phrenic nerves were located bilaterally in the dorsolateral region of the neck, cut caudally and desheathed. The left flank of the animal was twisted and placed upward. Through a flank incision the left renal nerve was isolated retroperitoneally. All nerves selected for recording were cut, immersed in a pool of mineral oil, and central ends placed on bipolar platinum electrodes for recording. Both cervical vagi nerves were isolated, ligated and cut. Tests for baroreceptor denervation were done later in each cat with a bolus injection of norepinephrine ( 1 - 4 / ~ g . kg -1) into the femoral vein and in one cat by inflating a balloon catheter inserted in the descending aorta. There was no decrease in sympathetic or phrenic nerve activity in response to balloon inflation with an increase in blood pressure of 30 m m Hg or in response to norepinephrine infusion (blood pressure increase of at least 40 m m Hg). The final procedure was to make large bilateral pneumothoraces such that there was no observable chest movement during tidal volume inflation of the lung. With the conclusion of surgery, halothane was stopped and we continued anesthesia with alpha-chloralose (average cumulative dose, 30 m g . kg -1) to allow sympathetic tone to increase. Additional alpha-chloralose was given if blood pressure and heart rate increased in response to noxious stimulation of the hind paw or apneic threshold for PA..... dropped below 20 m m Hg.
Recordings We recorded activities from phrenic nerve (PN), internal intercostal nerve (Exp. Fil. Ts), renal nerve (RN) and cardiac nerve (CN). Nerve activities were filtered (80 H z - 1 0 kHz), differentially amplified and recorded on tape, a direct pen recorder and on a storage oscilloscope. Simultaneous periods in RN, CN, PN and Exp. Fil. T 5 were compared either by repeated overlap of single traces on a storage oscilloscope which was triggered at the onset or the end of the inspiratory component of phrenic nerve discharge and photographed by Polaroid, or averaged by a 4-channel signal averager. Phrenic nerve activity was used to produce two trigger signals. It was first passed through a 'leaky integrator' (time constant 0.1 s) or 'moving time averager'
80 using a bin width of 100 ms which delivered a 5 V pulse indicating both onset of inspiration (positive pulse) and cessation of inspiration, i.e. beginning of postinspiration (negative pulse). Either trigger could be used to start the simultaneous average of 4 separate neural signals for ' n ' respiratory periods of 2 or 4 s using bin widths of 2 or 4 ms. The averaged activities were then stored on a digital oscilloscope with magnetic disc memory system for later plotting on an X-Y plotter.
Data acquisition When the preparation was complete and the transition from halothane to chloralose anesthesia established, we began to record neural activities. Measurements were first made under normocapnea. PA,,o, was kept at 28-30 mm Hg for a period of 20-30 min and tape recordings were made at intervals of t0 min. The PA,~,2 was then elevated randomly by increments of 5 - 7 m m Hg up to a maximum of 53 m m Hg. Every change in PAso started from control levels and was held constant for 20-30 min. At each stable ~ O 2 level, we taped recordings from nerve activities and made repeated Polaroid photographs of storage trace overlays.
Results
Sympathetic activity was correlated to the 3 phases of the respiratory cycle [3,30,31] which were identified by simultaneous recordings from phrenic nerve and internal (expiratory) intercostal nerve filaments. Fig. 1 identifies these respiratory phases and shows the respiratory modulation of the cardiac sympathetic nerve activity. The phrenic nerve activity typically showed a ramp-like augmentation, i.e. inspiration, followed by a decrescendo of activity, i.e. postinspiration (stage I expiration), and then by a period of silence, i.e. stage II expiration. The expiratory intercostal nerve filament showed large and small spikes originating from alpha and gamma motoneurones [33]. Expiratory gamma motoneurones mostly fired continuously throughout the respiratory cycle but were modulated to a minimum during peak phrenic discharge. Expiratory alpha motoneurones on the other hand began to fire with an augmenting pattern when postinspiration ceased. They stopped promptly
EXPFIL 02mY[ ms (~) PN
o5mV[
CN
O02mVI
ls Fig. 1. Single sweep recording of activities from the internal intercostal filament (Exp. Fil. ]'s), phrenic nerve (PN) and cardiac sympathetic nerve (CN).
81 with the onset of the next burst of phrenic discharge. Thus, expiratory alpha motoneuronal activity arose only during complete silence of the phrenic nerve. Sympathetic activities can then be correlated with the (ramp) inspiratory and postinspiratory activity of the phrenic nerve or alternatively they can be observed during stage II expiration for evidence of activation or depression. Note that the cardiac sympathetic nerve is most active during peak phrenic discharge and has some but less activity during stage II expiration. The cardiac nerve is silent in early inspiration and during postinspiration. Fig. 2 shows these patterns of modulation of sympathetic activity in greater detail. A typical inspiratory augmenting discharge and a declining pattern of postinspiratory activity is seen in the phrenic nerve. Expiratory alpha motoneuronal activity is present solely during the silent period of the phrenic activity as described before. Both cardiac and renal activities exhibit respiratory modulation of their patterns with peak activity during the inspiratory period, less activity in stage II expiration and a clear lack in activity during early inspiration. Cardiac and renal patterns seem to differ in that the cardiac nerve is absolutely silent during postinspiration whereas the renal nerve continues to discharge throughout this period. In Fig. 3, multiple oscilloscopic sweeps were superimposed which were triggered by the end or the start of the inspiratory component of the phrenic nerve discharge, each showing the activities during one respiratory cycle. Although respiratory modulation is strong in both sympathetic nerves, cardiac and renal (Fig. 2) sympathetic nerves were found to be quite different. In both sympathetic nerves activation follows the contour of late inspiratory activity, i.e. sympathetic activity was strong during the late inspiratory phase. Sympathetic activity steadily increased throughout the period of stage II expiration in the cardiac nerve, but a similar augmentation of activity seems to be absent in the renal nerve. Postinspiratory depression was observed in cardiac nerves, but not so regularly in renal nerves. A typical finding, however, was a clear depression of sympathetic activity during early inspiration in both cardiac and renal nerves. This seems to be remarkable for two reasons. First, it could occur in the renal nerve despite a high level of tonic activity which may obscure other phases of respiratory modulation like a possible expiratory
CN
0.2rnV[
RN
O.OSmVE
- r " ~ ~' l ~ ~ EXP. FIL. T5
0.2mVl- ,.,,.,,,~ . , . . . .
PN
O.SmVE
.... ~ J I
,,r T
II ~
'" , ~ r r r ~
," ~,,,i,N1~,1,rf~'w . . . . .
""'~'"'!,
J ls
Fig. 2. Recordings from the cardiac sympathetic nerve (CN), renal sympathetic nerve (RN), internal intercostal filament (Exp. Fil. T5) and phrenic nerve (PN) recorded on a pen recorder.
82
activation and postinspiratory depression of activity and second, it was seen in the cardiac nerves despite the inevitable 'smearing' when activities were superimposed at a long delay after the trigger pulse had started the sweep. This 'smearing' can be seen in Fig. 3A by the 0.3 s overlap of inspiratory onset of phrenic activity and expiratory termination. It is simply a consequence of the superposition of sweeps and is caused by the spontaneous variation in the frequency of the central respiratory rhythm. The heterogeneity of sympathetic activities emphasizes the need for determining the probability of activation of sympathetic neurones more precisely, i.e. it is necessary to average the data over several respiratory cycles at different levels of PA,,," However, as seen in Fig. 3, averaging must deal with the problem of the spontaneously occurring variations in the duration of a respiratory cycle which could vary as much as one second. Any correlation of averaged data, therefore, will be accurate only close to the trigger time but will become progressively 'smeared' as the delay after the trigger increases. Averages started at the two points of respiratory phase transition, therefore, require some optimum number of sweeps to produce a faithful representation of overall contour of activity but with minimal loss of detail. Our empirical approach to this problem is seen in Fig. 4. Renal nerve activity was integrated and averaged from the end of the inspiratory phrenic burst for 15, 30 and 60 consecutive breaths. This averaging was done over the same portion of the tape recording with the results seen here. Note that the contour of averaged integrations of sympathetic activity is similar for all 3 (15, 30 and 60) sets of sweeps at the onset of triggering, i.e. the point of greatest accuracy. But note also that considerable detail is lost for averages with a sweep number greater than 15, as the averaged pattern extends to the right and thus away from the trigger time. Thus, a number of 15 sweeps was assumed to faithfully reproduce the sympathetic pattern although the resulting curves may not be as smooth as with larger sweep numbers. We, therefore, chose 15 sweeps as the optimum number for averaging our data.
A EXP FIL T5 02mV[ (~i
B EXIDFIL T5 02mV[ (I)
PN OSmV[
PN O.5mV[
EN 0.02mY[
RN 095rnV[
1s 1s Fig. 3. Activities from internal intercostal filament (Exp. Fil. T s) and phrenic nerve (PN) together with cardiac nerve (A) and renal nerve (B). 15 sweeps were superimposed on a storage oscilloscope which was triggered from peak phrenic activity (A) and the start of phrenic nerve activity (B). The sweeps were triggered at different points in phrenic activity to increase the precision of observation. (Precision is greatest close to the trigger and declines at a distance due to the spontaneous variation in inspiratory and expiratory duration.)
83
Fig. 5 reveals information as to the possible origin of this depression in sympathetic activity. Averaged cardiac nerve activities (CN~_3) of different strength from 3 cats are presented at the same amplification. The C N 3 t r a c e of cardiac nerve activity is of particular interest for two reasons: first, this cat had a very high level of tonic sympathetic drive as seen by the height of the line connecting sympathetic activity with the zero level of the integrator output and second, at normal levels of P",~o2 (28-30 mm Hg) in this cat there was no modulation of sympathetic activity with respiration. Only when the PAco2.was elevated to 53 mm Hg, as shown here, did a clear respiratory modulation of activity appear. This modulated pattern, however, did not follow late inspiratory and stage II expiratory activation, instead depression of sympathetic activity during early inspiration was the only recognizable pattern. Since CO 2 activated central respiratory activity as measured in the phrenic nerve discharge (not illustrated) and produced a clear respiratory modulation of tonically enhanced sympathetic activities, we suggest that bulbar respiratory neurones were responsible for the depression of sympathetic activity. We further speculate that it exceeds all excitatory respiratory inputs in effectiveness because it was the only manifestation of rhythmic modulation of intense tonic sympathetic activity. A depression of activity can occur during postinspiration as shown in Fig. 3. This deviation from the contour of phrenic discharge suggests that inhibition might be present. Further evidence for the possibility of a postinspiratory inhibition of sympathetic activity is presented in Fig. 6. This cat inhaled CO 2 raising PAco2 from 35 to 53 mm Hg. Elevated CO 2 markedly increased peak phrenic activity without an obvious change in postinspiratory contour (not illustrated). In both renal and cardiac nerves sympathetic activity was increased during peak phrenic discharge. More impressive, however, was the fact that the sympathetic activity contour was
RN
3o 60
I
Is Fig. 4. Averaged patterns of phrenic nerve (PN) and renal nerve (RN) activities. The averages were triggered from the end of the inspiratory component of the phrenic nerve activity for 15, 30 and 60 consecutive breaths. Repeated patterns for 15, 30 and 60 averages are offset for visual comparison and are connected by perpendicular markers to their respective baselines. Baseline indicates zero potential.
84
altered during the postinspiratory period. Sympathetic activity was reduced more sharply, and achieved a more sustained low level before the activity started to steadily increase again during the stage II expiration. We suggest that an inhibitory input to the sympathetic network might be responsible for this depression in sympathetic activity during postinspiration. A summary of sympathetic patterns compared with inspiratory, postinspiratory and expiratory activity is as follows: in 3 of 5 cats, sympathetic activity both in cardiac and renal nerves increased and decreased in responses to an increase and decrease in inhaled CO 2. Sympathetic activity roughly mimicked the late inspiratory and augmenting expiratory activity. In all cats, we observed a depression in sympathetic activity during early inspiration and there were indications for a
PN ~
.
_
~.__,,_.,,,_._~
EN~ [N2 [Na
ls Fig. 5. Averaged activity of cardiac nerve (CN) activity in 3 cats. All activity is triggered from onset of phrenic nerve discharge. Averaged activities of phrenic and cardiac nerve activity (CN 1) shown in the upper two traces are from the same cat. Averaged sympathetic activities (CN 2 ~) from the other two cats are adjusted to the same phrenic time scaling and displaced downward for ease in comparing sympathetic contours. The time base therefore refers only to the phrenic and the top cardiac nerve trace. Perpendicular lines connect each averaged sympathetic activity to its basel ne activity. PA for each cat beginning at ('°2 1 the top are 28, 30 and 53 Torr. Phrenic contours after scaling for time were essentially identical for al cats.
85
PN CN I
[N 2
RN1
RN2
I ] S Fig. 6. Averaged activities from phrenic nerve (PN), cardiac nerve (CN) and renal nerve (RN) in one cat. All activity was triggered with the end of the inspiratory discharge of the phrenic nerve. Perpendicular lines connect sympathetic activities with their respective baseline. The P^co for the traces of CNl and RN1 sympathetic patterns was 35 Torr and 53 Tort for the traces CN 2 an~ RN2. Expiratory duration estimated by the phrenic and internal intercostal nerve activities was not changed by this elevation of CO2 and there was no need for re-adjustment of time scales. Elevated CO2 markedly increased peak phrenic activity without an obvious change in postinspiratory contour.
depression d u r i n g postinspiration. I n h a l e d C O 2 depressed sympathetic activity in two cats u n d e r two circumstances while phrenic activity was enhanced. I n one cat this occurred d u r i n g early inspiration which was the only m a n i f e s t a t i o n of a respiratory coupled r h y t h m of a high tonic b a c k g r o u n d in sympathetic activity. In the other case, this appeared as a depression a n d s h a r p e n i n g of the fall of sympathetic activity d u r i n g the postinspiratory period.
Discussion This work shows that sympathetic activity is not simply related to the overall c o n t o u r of phrenic activity, b u t exhibits a p r o n o u n c e d depression both in early
86 inspiration and during postinspiration. Moreover, the averaged contour of sympathetic activity frequently follows the steadily increasing pattern of expiratory activity in alpha motoneurones. One possible implication which can be drawn from this work is that the respiratory modulation of sympathetic activity involves several components of the behaviour of propriobulbar and bulbospinal respiratory neurones. This inference raises a number of issues regarding the functional organization of the brainstem systems regulating respiratory and sympathetic activities. What evidence exists that respiratory neurones project a rhythm onto sympathetic networks at the brainstem level?
Sympathetic nerves exhibit slow modulated activity which is synchronized with the respiratory rhythm. The patterns of respiratory modulation of sympathetic activity are complex and variable suggesting that multiple inputs are involved. This makes the analysis of the relationship between respiratory and sympathetic systems difficult and tentative. Many authors have reported that peak sympathetic activity is correlated with peak phrenic activity and they have suggested that the pattern is derived from inspiratory neurones [14,17-20,23,28,29,35,36] or expiratory-inspiratory phase spanning neurones [5,9,10]. In the work of Preiss, Kirchner and Polosa [29] the activity of single filaments of the cervical sympathetic nerve is so remarkable similar to phrenic activity in its timing and profile that it seems difficult to believe that two such virtually identical activity patterns could be generated by independent systems. It seems more reasonable that the rhythmic component of their activity derives from one system (see ref. 18). Added to this is the parallel response of sympathetic and phrenic nerves to hypocapnea and hypercapnea [5,9,14,18,23,28,29], loss of the respiratory component of the sympathetic rhythm during hypocapnic apnea [20,28,29] and the recent evidence of Kubin, Trzebski and Lipski [21] and Connelly and Wurster ]6] that in the cat a midline lesion in the medulla at the level of the obex, disrupts both phrenic activity and rhythmic patterns of sympathetic activity but leaves pressor responses intact and sympathetic nerves accessible to ipsilateral stimulation of pontine parabracheal areas. There is, however, disagreement as to the imprint of respiratory neurones on the brainstem system regulating sympathetic outflow. Clearly, the activity patterns of phrenic and sympathetic nerves are not always identical. Barman and Gebber [4] raise concern that the sympathetic pattern can preceed~ be simultaneous with~ or lag behind the onset of phrenic nerve activity. Their data to show, however, that the offset of phrenic and sympathetic patterns is essentially the same. Differences in tonic sympathetic background, varying activation thresholds and varying medullary respiratory drive could account for differences in onset of activity. Barman's and Gebber's [4] evidence that sympathetic rhythms can occur with a respiratory-like frequency independent of the phrenic rhythm does not necessarily conflict with our results. Clearly, respiratory influence on sympathetic is variable. From our own data, differences in the pattern of cardiac and renal nerve activities indicate that there is a possibility for differentiated respiratory-sympathetic interaction and that a high level of tonic sympathetic background may swamp all facilitatory influences from the respiratory system. Since there is at present no direct evidence for a relationship
87 between the two systems, it would seem desirable to refine comparisons in view of the probability that sympathetic activity is indeed influenced by excitatory and inhibitory respiratory neurones at brainstem level.
Do inhibitory respiratory neurones exist which might play a role in determining sympathetic patterns? In recent publications the likely importance for normal rhythmogenesis of early inspiratory, and postinspiratory inhibition to the network interactions among respiratory cells in the medulla has been stressed [3,30,31]. This inhibition may originate in propriobulbar respiratory interneurones, two classes of which have been identified, the 'early inspiratory' [25] and the 'post inspiratory' neurons [30,31]. The potentially significant feature of both classes of neurones is their divergent inhibitory output to many types of medullary respiratory neurones. It was also shown that tonic, apparently non-respiratory, reticular neurones in the medulla are also accessible to postinspiratory inhibition [30,31]. This finding raises the possibility that these interneurones exert a generalized inhibitory influence on other systems. One candidate might be the brainstem system regulating sympathetic outflow which is known to have a respiratory modulation. Examination of sympathetic mass-activity in relation to peripheral respiratory activity constitutes an indirect way of identifying these two classes of inhibitory events in the sympathetic system. Do other authors report eoidence for inhibitory sculpturing of rhythmic sympathetic patterns? The work of Seller and Richter [36] provides evidence for postinspiratory inhibition seen on postganglionic renal sympathetic nerves in dogs. Their results showed that the level of sympathetic activity, increased synchronously with phrenic activity, was completely inhibited during postinspiration and maintained a nearly constant level during the remainder of expiration. However, these results do not exclude the possibility that total suppression of sympathetic discharge was some form of postexcitatory depression following inspiratory activation as suggested by Okada and Fox [27]. Reasoning from the work of Preiss et al. [29], mass-activity in whole sympathetic nerves is composed of fibres, 70% of which has rhythmic respiratory pattern and 30 per cent of which is tonic. The occurrence of such a clear void in all sympathetic mass-activity, normally showing considerable tonic activity, might best be explained by active inhibition of the tonically active structures in the brainstem regulating sympathetic outflow. Data from the work of J~mig et al. [14] hint that the respiratory patterns generated in sympathetic activity does not simply follow inspiratory excitation but may also be under the influence of inhibitory respiratory neurones. Their Fig. 1A-C, indicate that cutaneous vasoconstrictor sympathetic activities in the cat can have essentially tonic activity without strong respiratory modulation during normocapnea and hypocapnea. Hypercapnea brings out a well-defined sawtooth pattern, synchronous with the phrenic discharge with a maximum activity greater than, and a minimum activity less than the tonic activation seen in normocapnea and hypocapnea. The minimum activity level occurs in the immediate postinspiratory period, the maxi-
88 mum activity coincides with the maximum of phrenic nerve discharge with a linear increase from minimum to maximum. Several studies [5,9,14,18-20,23,28] clearly show that sympathetic activity, coincident with phrenic activity, increases in response to an enhanced CO 2 drive. Sympathetic activity can also decrease with hypercapnea [14]. A possible explanation for this decrease is that it results from the hypercapnic activation of respiratory inhibitory interneurones. In the same article, J~mig et al. [14] give a second example of possible sympathetic inhibition in the postinspiratory period. This is seen in their Fig. 3D-F2 in response to hypercapnea. This example is reminiscent of our experience (Fig. 6) in which a gradual decrescendo of sympathetic activity in the postinspiratory period is converted to an abrupt decrease in response to hypercapnea without a change in phrenic postinspiratory activity. There are some hints that sympathetic activities are subject to inhibition during early inspiration as well. Gootman and Cohen [9] described averaged sympathetic activity triggered from onset of phrenic discharge. Fig. 2 of their work shows a distinct depression of sympathetic activity during the early inspiratory phase. What are the implications of this work?
Our comparison of respiratory and sympathetic activities provides us with a window to the functional organization of the brainstem system regulating sympathetic outflow. This comparison, in conjunction with the work of other investigators, would indicate that sympathetic rhythmic patterns, phase-locked to respiratory patterns, do originate in elements of the bulbar respiratory network, i.e. activation by bulbar inspiratory neurones and inhibition by early inspiratory and postinspiratory interneurones which do not project to the spinal cord. There are at least two possible, non-exclusive, sorts of explanation for the relationship between sympathetic activity and stage II expiratory activity: the first is that the augmentation of sympathetic activity during stage II expiration reflects slow recovery from postinspiratory inhibition superimposed on a tonic level of activity; the second explanation is that the steadily augmenting sympathetic activity during stage II expiration derives from facilitation by expiratory neurones. If this interpretation is true it implies that bulbar interneurones with a respiratory activity can project or even belong to different systems, e.g. to the respiratory system and to the system regulating sympathetic outflow. In the parasympathetic division of the autonomic nervous system there is also evidence for various sorts of respiratory influence. Cardiac vagal motoneurones show an expiratory related discharge [7,13,15,16,22,24,26] and intracellular analysis revealed that they are activated during postinspiration, strongly inhibited during inspiration and weakly inhibited during stage II expiration [8].
Acknowledgements The authors wish to thank Ms. Annemarie Bischoff and Ms. Anita Ki~hner for their technical assistance in experiments and their help in preparing the manuscript and illustrations. This work was supported by DFG, Grant Ri. 279/7-7; SFB 90, USPHS GM-15571 and SFVAMC.
89
References 1 Adrian, E.D., Bronk, D.W. and Phillips, G., Discharges in mammalian sympathetic nerves, J. Physiol. (Lond.) 74 (1932) 115-153. 2 Bainton, C.R., Richter, D.W., Seller, H., Ballantyne, D. and Klein, J.P., Respiratory modulation of sympathetic activity, Fed. Proc., 41 (1983) 1125. 3 Ballantyne, D. and Richter, D.W., Post-synaptic inhibition of bulbar inspiratory neurons in the cat, J. Physiol. (Lond.), 348 (1984) 67-87. 4 Barman, S.M. and Gebber, G.L., Basis for synchronization of sympathetic and phrenic nerve discharges, Amer. J. Physiol., 231 (1976) 1601-1607. 5 Cohen, M.I. and Gootman, P.M., Periodicities in efferent discharge of splanchnic nerve of the cat, Amer. J. Physiol., 218 (1970) 1092-1101. 6 Connelly, C.A. and Wurster, R.D., Effects of midline medullary lesions on the respiratory modulation of sympathetic nerve activity, Fed. Proc., 43 (1984) 402. 7 Davidson, N.S., Goldner, S. and McCloskey, D.I., Respiratory modulation of baroreceptor and chemoreceptor reflexes affecting heart rate and cardiac vagal efferent nerve activity, J. Physiol. (Lond.), 259 (1976) 523-530. 8 Gilbey, M.P., Jordan, D., Richter, D.W. and Spyer, K.M., Synaptic mechanisms involved in the inspiratory modulation of vagal cardio-inhibitory neurones in the cat, J. Physiol. (Lond.) in press. 9 Gootman, P.M. and Cohen, M.I., The interrelationships between sympathetic discharge and central respiratory drive, In W. Umbach and H.-P. Koepchen (Eds.), Central-Rhythmic and Regulation, Hippokrates-Verlag, Stuttgart, 1974 pp. 195-209. 10 Gootman, P.M., Cohen, M.I., Piercey, M.P. and Wolotsky, P., A search for medullary neurons with activity patterns similar to those in sympathetic nerves, Brain Res., 87 (1975) 395-406. 11 Herbert, D. and Mitchell, R.A., Blood gas tensions and acid-base balance in awake cats, J. Appl. Physiol, 30 (1971) 434-436. 12 Hering, E., l~lber den Einfluss der Athmung auf den Kreislauf. 1. Ober Athembewegungen Gef~tsssystems, Sber. Akad. Wiss., Wien, Math. Nat. KI., II. Abtl., 60 (1869) 829-856. 13 lriuchijima, J. and Kumada, M., Activity of single vagal fibres efferent to the heart, Jap. J. Physiol., 14 (1964) 479-487. 14 J~nig, W., Ki~mmel, H. and Wiprich, L., Respiratory rhythmicities in vasoconstrictor and sudomotor neurones supplying the cats' hindlimb. In H.-P. Koepchen, S.M. Hilton and A. Trzebski (Eds.), Central Interaction Between Respiratory and Cardiovascular Control Systems, Springer-Verlag, Berlin, Heidelberg, New York, 1980, pp. 128-136. 15 Jewett, D.L., Activity of single efferent fibres in the cervical vagus nerve of the dog with special reference to possible cardioinhibitory fibres, J. Physiol. (Lond.), 175 (1964) 321-357. 16 Katona, P.G., Poitras, J.W., Barnett, G.O. and Terry, B.S., Cardiac vagal efferent activity and heart period in the carotid sinus reflex, Amer. J. Physiol., 218 (1970) 1030-1037. 17 Koepchen, H.-P., Die Bhitdruckrhythmik, D. Steinkopff Verlag, Darmstadt, 1962. 18 Koepchen, H.-P., Respiratory and cardiovascular "centres": Functional entirety or separate structures?. In M.E. SchlMke, H.-P. Koepchen and W.R. See (Eds.), Central Neurone Environment, Springer Verlag, Berlin, Heidelberg, 1983, pp. 221-237. 19 Koepchen, H.-P. Kliissendorf. D. and Sommer, D., Neurophysiological background of central neural cardiovascular-respiratory coordination: basic remarks and experimental approach, J. auton. Nerv. Syst. 3 (1981) 335-368. 20 Koizumi, K., Seller, H., Kaufman, A. and McC Brooks, C., Pattern of sympathetic discharges and their relation to baroreceptor and respiratory activities, Brain Res., 27 (1971) 281-294. 21 Kubin, L., Trzebski, A. and Lipski, J., Search for the site of the central respiratory-sympathetic synchronisation, Fed. Proc., 38 (1979) 1200. 22 Kunze, D.L., Reflex discharge patterns of cardiac vagal efferent fibres, J. Physiol. (Lond.), 222 (1972) 1-15. 23 Lipski, J., Coote, J.H. and Trzebski, A., Temporal patterns of antidromic invasion latencies of sympathetic preganglionic neurons related to central inspiratory activity and pulmonary stretch receptor reflex, Brain Res., 135 (1977) 162-166.
90 24 McAllen, R.M. and Spyer, K.M., Two types of vagal preganglionic motoneurones projecting to the heart and lungs, J. Physiol. (Lond.), 282 (1978) 353-364. 25 Merrill, E.G., Finding a respiratory function for the medullary respiratory neurons, In R. Bellairs and E.G. Gray (Eds.), Essays on the Nervous System, Clarendon Press, Oxford, 1974, pp. 451-486. 26 Neil, E. and Palmer, J.F., Effects of spontaneous respiration on the latency of reflex cardiac chronotropic responses to baroreceptor stimulation, J. Physiol. (Lond.), 247 (1975) 16p. 27 Okada, H. and Fox, l.J., Respiratory grouping of abdominal sympathetic activity in the dog, Amer. J. Physiol., 213 (1967) 48-56. 28 Polosa, C., Gerber, U. and Schondorf, R., Central mechanisms of interaction between sympathetic preganglionic neurons and the respiratory oscillator, In H.-P. Koepchen, S.M. Hilton and A. Trzebski (Eds.), Central Interaction Between Respiratory and Cardiovascular Control Systems, Springer Verlag, Berlin, Heidelberg, New York, 1980, pp. 137-143. 29 Preiss, G., Kirchner, F., and Polosa, C., Patterning of sympathetic preganglionic neuron firing by the central respiratory drive, Brain Res., 87 (1975) 363-374. 30 Richter, D.W., Generation and maintenance of the respiratory rhythm, J. exp. Biol., 100 (1982) 93-107. 31 Richter D.W. and Ballantyne, D., A three phase theory about the basic respiratory pattern generator, In M.E. Schl~fke, H.-P. Koepchen and W.R. See (Eds.), Central Neurone Environment, Springer Verlag, Berlin, Heidelberg, New York, 1983, pp. 163-174. 32 Schweitzer, A., Die Irradiation autonomer Reflexe, S. Karger Verlag, Basel, 1937. 33 Sears, T.A., Efferent discharge in alpha and fusimotor fibres of intercostal nerves of the cat, J. Physiol. (Lond.), 174 (1964) 295-315. 34 Sears, T.A. and Stagg, D., Short term synchronization of motoneurone activity, J. Physiol. (Lond.), 263 (1976) 357-381. 35 Seller, H., Langhorst, P., Richter, D. and Koepchen, H.-P., 13ber die Abhangigkeit der pressoreceptoriscben Hemmung der Sympathicus von der Atemphase und ihre Auswirkung in der Vasomotorik, Pfliagers Arch. ges. physiol., 302 (1968) 300-314. 36 Seller, H. and Richter, D.W., Some quantitative aspects of the central transmission of the baroreceptor activity. In F.F. Kao, K. Koizumi and M. Vassalle (Eds.), Research in Physiology. A liber memorialis in honor of Prof. C. McC Brooks, Aulo Gaggi, Bologna, 1971, pp. 541-549. 37 Tang, P.C., Maire, F.W. and Amassian, V.E., Respiratory influence on the vasomotor center. Amer. J Physiol., 191 (1957) 218-224. 38 Traube, D., Ober periodische T~tigkeits-~usserungen des vasomotorischen und Hemmungs Nervencentrums, Cbl. med. Wiss., 56 (1865) 881-885.