- Email: [email protected]
The effects of hypercapnia and cooling the ventral medullary surface on capsaicin induced respiratory reflexes
Respiration Physiology (1985) 60, 377-385 Elsevier
377
THE EFFECTS OF HYPERCAPNIA AND COOLING THE VENTRAL MEDULLARY SURFACE ON CAPSAICIN INDUCED RESPIRATORY REFLEXES
J. M I T R A , N . R . P R A B H A K A R , M . A . H A X H I U
and N.S. CHERNIACK
Department of Medicine, Case Western Reserve University, Cleveland, 0H44106, U.S.A.
Abstract. The effect of right atrial (RA) injection of 3/~g/kg capsaicin on phrenic, hypoglossal and recurrent
laryngeal activities was studied in chloralose anesthetized, paralyzed and artificially ventilated cats. Within 2 sec following capsaicin injection, the phrenic and hypoglossal activities completely disappeared (apnea), while the recurrent laryngeal activity markedly increased. Similar responses were also obtained with RA injection of phenyldiguinide (PDG), suggesting that the respiratory responses of both drugs are essentially similar. Sino-aortic denervation did not affect ~ e capsaicin induced respiratory responses. Bilateral vagotomy abolished the responses, suggesting that vasal sensory receptors are responsible for the reflex effects. Hyperoxic hypercapnia (3 and 7% CO2 in 02) reduced the apneic duration of phrenic and hypoglossal nerves. The magnitude of the Jre$~4~'r~fltlaryngeal excitation was decreased during CO 2 breathing. Graded focal cooling of the intermediate area (I s area) of the ventral medullary surface (to inhibit central chemoreceptor activity) significantly prolonged capsaicin induced apneic duration of hypoglossal nerves more than the phrenic, The recurrent laryngeal responses, however, were unaffected by cooling of the ventral medullary surface. The results show that capsaicin and PDG, presumably by stimulating C fibers, affect cranial nerves as well as the phrenic. The reflex responses to C fiber stimulation seem to be altered by intervention which stimulate (hypercapnia) or depress (I s cooling) 'central chemoreceptors'. Capsaicin Cat
Hypoglossal nerve Phrenic nerve
Pulmonary C fibers Recurrent laryngeal nerves
In recent years a considerable a m o u n t o f evidence has a c c u m u l a t e d suggesting that the central chemosensitive drive for ventilation is l o c a t e d near the ventral surface of the m e d u l l a o b l o n g a t a ( V M S ) (Schlaefke ~t al., 1970). Reversible focal cooling of the ventral medullary surface ( V M S ) not only blocks the C O 2 i n d u c e d ventilatory drive but also affects certain reflexes arising in the lung. F o r example, m o d e r a t e cooling o f the V M S to 20 °C attenuates the tachypnic r e s p o n s e resulting from deflation o f the lungs, p r e s u m a b l y due to the stimulation of the rapidly a d a p t i n g p u l m o n a r y stretch receptors, while the B r e u e r - H e r i n g reflex from slowly adapting p u l m o n a r y stretch receptors is Accepted for publication 16 March 1985
0034-5687/85/$03.30 © 1985 Elsevier Science Publishers B.V. (Biomedical Division)
378
J. MITRA etal.
unaffected (Cherniack et al., 1979a,b). Besides the slowly and rapidly adapting pulmonary stretch receptors, tracheopulmonary 'C' fibers are important in the control of breathing (Coleridge and Coleridge, 1984). In the present study we investigated the effects of hypercapnia and cooling of the ventral medullary surface (VMS) on the respiratory responses to pulmonary 'C' fiber (J receptor) stimulation with right atrial injection of capsaicin and phenyldiguinide (PDG). Though the responses of the phrenic activity to pulmonary 'C' fiber excitation is fairly well established (Coleridge and Coleridge, 1984), its effect on the motor output of cranial nerves innervating the upper airways, however, has not yet been examined. Previous studies from this laboratory have demonstrated that the activity of the upper airway muscles increases more than the diaphragm with chemical stimulation and with various other reflexes which affect ventilation (Weiner et al., 1982; Cherniack et al., 1984). In order to obtain more complete information about the respiratory responses to 'J' receptor stimulation, we also measured the activities of the hypoglossal and recurrent laryngeal nerves along with phrenic activity. Methods
Experiments were performed on eleven cats of both sexes (1.5-4.5 kg body weight) which were anesthetized with an initial intraperitoneal injection of c~-chloralose (40-50/~g/kg) followed by an intravenous injection of 10 mg of pentobarbital. Body temperature was maintained at 38 _+ 1 °C using an electric thermal blanket. The saphenous vein of one forelimb was cannulated for infusion of solutions, and the femoral artery to monitor systemic blood pressure. The trachea was intubated for artificial ventilation and to allow the insertion of a capnograph (Godart-Statham) probe to monitor end-tidal CO 2. Since our aim is to study the interaction of different levels of CO2 stimulation with respiratory responses induced by 'J' receptor stimulation without the complications produced by altered lung mechanics on ventilation, all experiments were conducted on paralyzed and artificially ventilated animals. The animals were paralyzed with intravenous administration ofgallamine triethiodide (5 mg/kg; flaxedil). A catheter was inserted through the right external jugular vein so that the tip lay in the right atrium and was used for injection ofcapsaicin. The dead space of the catheter was 0.1 ml. Through a midline incision extending from the symphysis of the lower jaw to the manubrium, the medial branch of the hypoglossal nerve was dissected free from the surrounding tissue. The medial branch was chosen because it exhibits mainly inspiratory activity (Mitra and Cherniack, 1983). The fifth root of the phrenic nerve and the right recurrent laryngeal nerve were also isolated. The cut central ends of these nerves were desheathed and were used for recording the electrical activity using the conventional recording techniques. The electrical activities of the nerves were integrated using a time constant of 100 or 200 msec. The carotid sinus, aortic, and vagus nerves were isolated and were surgically denervated during the course of the experiment. The details of the exposure of ventral medullary surface have been described
VENTRAL MEDULLARY SURFACE & PULMONARY C FIBER REFLEXES
379
elsewhere (Bruce et aL, 1982). In brief, to gain access to the ventral surface of the medulla, the trachea and the esophagus were freed from the tissues, severed, and then reflected rostrally. Subsequently, the underlying muscles were removed and then craniotomy was performed to expose the VMS. The dura was then opened and retracted by ligatures. After opening of the dura, the exposed medullary surface was covered with warm artificial CSF. For the purpose of ventral medullary surface cooling, a specially built small two footed (each 2 mm z) thermode mounted on a micromanipulator was placed on the medullary surface. Cooling was performed by circulating antifreeze of different temperatures through the thermode. The temperature of the cooling probe was continuously monitored with a thermocouple embedded in an area of the foot of the probe in contact with the VMS. The area of the ventral medullary cooled extended between the 7th cranial nerve rostrally (R M area) to the region medial to the hypoglossal roots caudally (CL). Maximum depression of phrenic activity was obtained from an area 1 mm rostral of the hypoglossal roots and 2.5 mm lateral to the midline. The area approximately corresponds to intermediate area (Is) described by Schlaefke etal. (1970). A solution ofcapsaicin (Sigma) was prepared according to the method described by Coleridge et al. (1964). Phenyldiguinide (PDG) was dissolved in saline (0.9~o NaC1). The doses of capsaicin and PDG used in this study were 3 #g and 30 #g/kg, respectively. The total volume of the solution used per injection was 0.5 ml given into the right atrium over a period of 2 sec. During steady-state CO 2 breathing, the inspiratory port of the ventilator was connected to a 3 L bag which contained either 3 or 7 ~/o CO2 in 02. The animals breathed CO2 gases for at least 10 min before any intervention. Analysis. The duration of apnea in phrenic and hypoglossal nerves after the injection of capsaicin was used as an index of respiratory inhibition. During complete phrenic inhibition the increase in recurrent laryngeal activity from its baseline at end inspiration was used as a measure of recurrent laryngeal nerve excitation during the same period. Statistical analysis was performed by using the paired t-test. The criteria for statistical significance was P < 0.05.
Results
A typical response to right atrial (RA) injection of capsaicin (3 #g/kg) in a paralyzed, artificially ventilated cat is shown in fig. 1. Within 2.0 sec following capsaicin injection the activity of the phrenic and hypoglossal nerves completely disappeared almost at the same time, and in this experiment lasted about 7 sec. The change in phrenic activity was associated with a concomitant increase in the recurrent laryngeal activity. The respiratory responses were accompanied with a small biphasic change in arterial pressure, namely an initial slight increase followed by a slight decrease, without a change in heart
J. MITRA etal.
380
Fi CO2 r.o%
,..,.::2j 5 m ,,
sec
.i
Fig. 1. Responses of the phrenic, hypoglossal and recurrent laryngeal nerves to right atrial injection of capsaicin (3 #g/kg) in an anesthetized, paralyzed, and artificially ventilated cat. Following the capsaicin injection within 2 sec the integrated activity of the phrenic (Ph) and hypoglossal(Hy) nerves disappeared and the recurrent laryngeal activity(RL) increased. After capsaicin, recurrent laryngeal activitywas locked with the ventilator. It took 6-8 rain for both neural activities to return to control amplitudes. The arrow indicates capsaicin injection in this and in the next figure. rate. Following the apnea, the reinitiation of the phrenic and hypoglossal nerves occurred at the same time. On an average it took 8-10 min for both the nerve activities to return to the control amplitudes (not shown in fig. 1). Before capsaicin injection, the recurrent laryngeal showed inspiratory activity synchronized with phrenic activity and usually displayed a small amount of expiratory activity. However, during capsaicin induced apnea (reflected in the phrenic and hypoglossal activities) the recurrent laryngeal nerve showed increased phasic activity which was locked to the rhythm of the ventilator (activity was less during lung inflation). As phrenic activity returned, the recurrent laryngeal nerve displayed activity during both expiration and inspiration. Expiratory activity was more exaggerated than before the injection of caps aicin. Over a period of several minutes, the expiratory activity gradually decreased and the inspiratory activity increased. To eliminate the phasic feedback from vagal stretch receptors, the ventilator was stopped briefly and capsaicin was injected. As shown in fig. 2, the phasic activity in the recurrent laryngeal nerve was replaced by tonic activity. This sequence of events was the same in intact and sino-aortic denervated animals. Often a small amount of residual tonic activity in the phrenic nerve was observed if capsaicin was injected while the respirator was stopped. After bilateral vagotomy, capsaicin-induced apnea in phrenic, hypoglossal and excitation of recurrent laryngeal nerves were abolished.
Comparison of PDG and capsaicin. P D G with a dose of 30 #g/kg stimulates J receptors nearly maximally (Anand and Paintal, 1980). In three experiments in paralyzed animals,
VENTRAL MEDULLARY SURFACE & PULMONARY C FIBER REFLEXES
381
Fi CO 2 7 . 0 %
/ Ph.
/ R.L.
150,=
!
5 see
¢
Fig. 2. Effectof capsaicin injectionon the integrated activitiesof phrenic(Ph) and recurrent laryngeal(RL) during the stopping of ventilator in an anesthetizedparalyzedcat. Followingthe capsalein administration, the phasic recurrent laryngealactivitywas replaced by tonic activitywhich is associated with absence of phasic phrenic activity.The first three breaths followingthe tonic activity are expiratory phasic activity, subsequently switched to inspiratory phasic. 30 #g/kg of P D G was given right atrially, In all the three animals, the phrenic nerve activity disappeared on an average for 7-9 see. Recurrent laryngeal also showed an increased activity similar to capsaicin (3 #g/kg). The fall in the arterial pressure and bradycardia, however, was more marked.
Effect of C02 on capsaicin-induced reflex responses. We investigated the effect of C O 2 on the duration of apnea in the phrenic and hypogiossal nerves in six animals. The animals breathed two levels of CO2 (3 and 7% CO2) in 02 mixtures. Capsalcin was injected after the animals inspired CO2 reached a steady-state level. The results are summarized in fig. 3. At each level of CO2 the duration of apnea in the hypoglossal and phrenic nerves was identical. With increasing CO2 the duration of apnea in both phrenic and hypoglossal nerves decreased significantly from 9 to 6.5 see (P < 0.01). The reduction in the duration of hypoglossal and phrenic nerve apnea by COz was accompanied by a decrease in the amplitude of the recurrent laryngeal activity. As inspired CO 2 was increased from 3 to 7 % the peak amplitude of the integrated recurrent laryngeal activity after capsaicin injection on average fell by 80% (P < 0.01).
382
J. MITRA et al.
10 -
[]
Phrenic
[]
Hypoglossal
T I::::::::::
v
!iiiiiiiil
~8-
::::::::::
t~
~7
-
ii~iiiiii: :1111:::1:
g ~6
-
iiiiiiiil !i....iiil
3
El 5 -
3
7
3
7
%F I C 0 2
Fig. 3. The effect of elevated C O 2 o n the apneic duration of phrenic (Ph) and hypoglossal (Hy) nerve activities. The data presented are mean values with standard error of mean. P < 0.01 (paired t-test)•
Effect of ventralmedullarysurface cooling. Since elevated CO2 reduced the duration of capsaicin induced apnea, focal cooling of the central chemoceptive structures located on the ventral medullary surface might be expected to have an opposite effect. In another group of five animals during breathing 7~o CO2 in 02, rostral (RM), intermediate (Is) and caudal (CL) areas were subjected to graded focal cooling. Once the phrenic amplitude reached the steady-state level during cooling, capsaicin injections were repeated. Cooling the intermediate area (Is) to 20 ° C markedly prolonged the capsaicin induced apnea in phrenic and hypoglossal nerves. On the other hand, cooling the caudal (CL) had less effect, while rostral area (RM) had the least effect. Since the responses obtained with cooling of the I s area are more pronounced, the results were analyzed only with cooling of Is area. The temperature of the I s area was lowered in a graded fashion from 37 to 20 °C and then to 15 °C. Cooling dissociated the apneic duration of phrenic and hypoglossal nerves. Lowering the temperature to 20 °C increased capsaicin induced apnea in the hypoglossal nerve to 90 see, whereas in the phrenic it was only 18 sec. Further lowering of temperature to 15 ° C eliminated completely the hypoglossal activity, but it prolonged the phrenic apnea to 80 sec. In all the experiments, lowering the temperature of the 1s area to 15 °C completely eliminated the hypoglossal activity. Therefore, for quantitative analysis, the effect of capsaicin with I s area temperature of 37 ° C was compared with results obtained with cooling to 20 °C. The results are summarized in fig. 4. At 37 °C the duration of apnea for both phrenic and hypoglossal was similar. When I s temperature was lowered to 20 ° C, the duration of apnea for the hypoglossal nerve was increased more than the phrenic, which is statistically significant (P < 0.01). Although the peak amplitude of the integrated recurrent laryngeal activity was somewhat higher at 20 °C (35.3 _+ 9.1 A. U) than at 37 °C (28.9 + 4.5 A. U), the difference was not statistically significant.
VENTRAL MEDULLARY SURFACE & PULMONARY C FIBER REFLEXES 90
383
FI C027% Ph H
8O
,qr'
Hy 0 . . . .
0
7O I 6O !
6
/i
,o
"~
•
40 !
I c~
I
/ / I,
I
! J=
3o
I 20
I
I
lO
I
I
37
20 T is ( c ' )
Fig. 4. Comparison of the capsaicin-induced apneic duration in hypoglossal (Hy) and phrenic nerve during the cooling of Is area to 20 °C while the animals were ventilated with 7% CO2 in 02,
Discussion
The results of the present study show that fight atrial injection of capsaicin induces apnea not only in the phrenic but also in the upper airway nerves like the hypoglossal. They also suggest that changes in CO 2 drive affect the intensity of the reflex response. Cooling prolonged the reflex apnea caused by capsaicin injection in phrenic and hypoglossal nerves, while hypercapnia shortened it. The data obtained with hypercapnia were different from the observations made by Anand and Paintal (1980). They found that hypoxia, but not hypercapnia, attenuated the reflex apnea caused by stimulation of J receptors with phenyldiguinide. The difference could be due to the fact that their animals were spontaneously breathing 4Yo CO2 in room air, while our animals were paralyzed and ventilated with 3 and 7 YoCO2 in 02. According to Palntal (1973) if apnea appears within 2.5 sec in the phrenic following the right atrial injection of PDG, it can be regarded as due to stimulation of J receptors. In all our experiments, within 2 sec after right atrial injection of capsalcin and PDG, apnea was consistently observed in phrenic and also in hypoglossal nerves, suggesting that the effect probably resulted from J receptor stimulation, Furthermore, the pattern of the respiratory responses was essentially the same for both capsaicin and PDG. Although the duration of apnea was similar for both phrenic and hypoglossal nerves when the I s area was at 37 °C, lowering of the I s temperature to 20 °C increased the duration of apnea for hypoglossal nerve more than the phrenic. Thus it is conceivable that the increase in the duration of absent hypoglossal activity could be due to cooling the XII rootlets, which are anatomically located close to I s area. However, unilateral
384
J. M I T R A etal.
cooling of I s area to 20 °C eliminates the activity in the contralateral hypoglossal nerve indicating that the observed effects are not caused by rootlet cooling. Cooling the I s area might decrease central chemoreceptor afferent input. It is possible that the low CO2 drive in itself dissociates hypoglossal and phrenic response to capsaicin. Previously we reported that hypoglossal nerve activity was more sensitive to chemical manipulation of the VMS than phrenic nerve activity (Weiner etal., 1982). Also, during hyperventilation hypoglossal activity disappears first relative to phrenic activity. To test this hypothesis we hyperventilated two animals with 100~o 02 to reduce the hypoglossal phasic activity approximately to the level observed during I s area cooling to 20 °C and then capsaicin was injected. In these two experiments the apneic duration of both phrenic and hypoglossal nerves was similar. Hence, it seems that I s area cooling has a different effect from hypocapnia on capsaicin-induced apnea. Cooling of the VMS increases the inhibitory response of capsaicin injection on the hypoglossal more than the phrenic while hypocapnia does not. The significance of this is not yet clear, but suggests that cooling may not affect exactly the same structures as low CO 2. The effects of H + and CO2 on cerebral vasculature and blood flow are known (Heistad and Kontos, 1983). However, we are not aware of studies dealing with the effects of focal cooling on cerebral vasculature and microcirculation. Recurrent laryngeal nerve activity increased during reflex inhibition ofphrenic activity by capsaicin or PDG. Stransky etal. (1973) reported that in non-paralyzed cat intravenous injection of phenyldiguinide caused apnea and complete closure of the larynx. Our observation of increased recurrent laryngeal nerve activity agrees well with their observation and also suggests that the responses induced by capsaicin are identical with PDG. In our study, recurrent laryngeal nerve in the paralyzed animal had phasic activity during reflex apnea which was locked to ventilatory frequency. When we stopped the ventilator for a few seconds during apnea, the phasic recurrent laryngeal activity was replaced by tonic activity, which suggests that the phasic activity was caused by stimulation of vagal sensory tracheopulmonary receptors. As phrenic nerve activity returned, the recurrent laryngeal nerve began to show more expiratory than inspiratory activity, which gradually decreased over a period of several minutes to pre-injection level. Koepchen etal. (1977) have shown that stimulation of J receptors leads to a marked increase of expiratory modulated neuronal activity which lasted for several minutes after P D G injection. The long-lasting recurrent laryngeal expiratory activity after capsaicin could be due to long-lasting stimulation of these expiratory neurons. Whether in a spontaneously breathing animal a similar increase in the expiratory activity of expiratory muscles occurs following capsaiein-induced apnea is yet to be determined. We also found that the amplitude of the recurrent laryngeal nerve activity during phrenic inhibition is inversely related to CO 2 drive; i.e., the lower the CO2 level, the bigger the amplitude. However, when the effect ofcapsaicin on recurrent laryngeal nerve activity at 20 °C I s area temperature was compared with 37 °C I s area temperature, activity was greater but there was no statistically significant difference. It may be that strong inhibition of inspiratory motoneuron pool by vagal 'C' fibers at low CO 2 level
VENTRAL MEDULLARY SURFACE & PULMONARY C FIBER REFLEXES
385
causes a release of expiratory activity from the recurrent laryngeal motoneuron pool. Nonetheless, these experiments also suggest a quantitative if not qualitative difference between VMS cooling and hypocapnia. Another observation made in the present study is that in paralyzed animals capsaicin causes negligible fall in blood pressure and lacks bradycardia. These results are in variance with responses reported in spontaneous breathing animals, where stimulation of pulmonary C fibers causes bradycardia and hypotension (Coleridge et al., 1964). In paralyzed animals, it seems that the responses of pulmonary chemoreflex include apnea while hypotension, bradycardia and tachypnea are either absent or reduced greatly. References Anand, A. and A. S. Paintal (1980). Reflex effects following selective stimulation of J receptors in the cat. d. Physiol. (London) 299: 553-572. Bruce, E. N., J. Mitra and N. S. Cherniack (1982). Central and peripheral chemoreceptor inputs to phrenic and hypoglossal motoneurons. J. AppL Physiol. 53:1504-1511. Cherniack, N. S., C. yon Euler, I. Homma and F.F. Kao (1979a). Graded changes in central chemoreceptor input by local temperature changes on the ventral surface of medulla. J. Physiol. (London) 287: 191-209. Cherniack, N. S., C. von Euler, I. Homma and F.F. Kao (1979b). Interactions between a central chemoceptive system and other ventilatory drives. In: Wenner-Gren Center International Symposium Series. Vol. 32, Central Nervous Control Mechanisms in Breathing, edited by C. yon Euler and H. Lagercrantz. Oxford, Pergamon Press, pp. 35-41. Cherniack, N. S., M. S. Haxhiu, J. Mitra, K. Strohl and E. van Lunteren (1984). Responses of upper airway, intercostal and diaphragm muscle activity to stimulation of oesophageal afferents in dogs. J. Physiol. (London) 349: 15-25. Coleridge, H. M., J. C. G. Coleridge and C. Kidd (1964). Role of the pulmonary arterial baroreceptors in the effects produced by capsaicin in the dog. J. Physiol. (London) 170: 272-285. Coleridge, J. C. G. and H.M. Coleridge (1984). Afferent vagal C fibre innervation of the lungs and airways and its functional significance. Rev. Physiol. Biochem. Pharmacol. 99: 1-110. Heistad, D.D. and H.A. Kontos (1983). Cerebral circulation. In: Handbook of Physiology. Sec. 2. The Cardiovascular System. Vol. III. Peripheral Circulation. Part I, edited by J.T. Shepherd and F.M. Abboud. Bethesda, MD, American Physiological Society, pp. 137-182. Koepchen, H.P., M. Kalia, D. Sommer and D. Kltissendorf (1977). Action of type J afferents on the discharge pattern of medullary respiratory neurons. In: Respiratory Adaptations, Capillary Exchange and Reflex Mechanisms, edited by A.S. Paintal and P. Gill-Kumar. Delhi, V.P. Chest Institute, pp. 407-425. Mitra, J. and N.S. Cherniack (1983). The effects of hypercapnia and hypoxia on single hypoglossal nerve fiber activity. Respir. Physiol. 54: 55-66. Paintal, A.S. (1973). Vagal sensory receptors and their reflex effects. Physiol. Rev. 53: 159-227. Schlaefke, M.E., W. See and H.H. Loeschcke (1970). Ventilatory response to alterations of H + ion concentration in small areas of the ventral medullary surface. Respir. Physiol. 10: 198-212. Stransky, A., M. Szereda-Prazestaszewska and J.G. Widdicombe (1973). The effects of lung reflexes on laryngeal resistance and motoneurone discharge. J. Physiol. (London) 231: 417-438. Weiner, D., J. Mitra, J. Salamone and N. S. Cherniack (1982). Effect of chemical stimuli on nerves supplying upper airway muscles. J. Appl. PhysioL 52: 530-536.
377
THE EFFECTS OF HYPERCAPNIA AND COOLING THE VENTRAL MEDULLARY SURFACE ON CAPSAICIN INDUCED RESPIRATORY REFLEXES
J. M I T R A , N . R . P R A B H A K A R , M . A . H A X H I U
and N.S. CHERNIACK
Department of Medicine, Case Western Reserve University, Cleveland, 0H44106, U.S.A.
Abstract. The effect of right atrial (RA) injection of 3/~g/kg capsaicin on phrenic, hypoglossal and recurrent
laryngeal activities was studied in chloralose anesthetized, paralyzed and artificially ventilated cats. Within 2 sec following capsaicin injection, the phrenic and hypoglossal activities completely disappeared (apnea), while the recurrent laryngeal activity markedly increased. Similar responses were also obtained with RA injection of phenyldiguinide (PDG), suggesting that the respiratory responses of both drugs are essentially similar. Sino-aortic denervation did not affect ~ e capsaicin induced respiratory responses. Bilateral vagotomy abolished the responses, suggesting that vasal sensory receptors are responsible for the reflex effects. Hyperoxic hypercapnia (3 and 7% CO2 in 02) reduced the apneic duration of phrenic and hypoglossal nerves. The magnitude of the Jre$~4~'r~fltlaryngeal excitation was decreased during CO 2 breathing. Graded focal cooling of the intermediate area (I s area) of the ventral medullary surface (to inhibit central chemoreceptor activity) significantly prolonged capsaicin induced apneic duration of hypoglossal nerves more than the phrenic, The recurrent laryngeal responses, however, were unaffected by cooling of the ventral medullary surface. The results show that capsaicin and PDG, presumably by stimulating C fibers, affect cranial nerves as well as the phrenic. The reflex responses to C fiber stimulation seem to be altered by intervention which stimulate (hypercapnia) or depress (I s cooling) 'central chemoreceptors'. Capsaicin Cat
Hypoglossal nerve Phrenic nerve
Pulmonary C fibers Recurrent laryngeal nerves
In recent years a considerable a m o u n t o f evidence has a c c u m u l a t e d suggesting that the central chemosensitive drive for ventilation is l o c a t e d near the ventral surface of the m e d u l l a o b l o n g a t a ( V M S ) (Schlaefke ~t al., 1970). Reversible focal cooling of the ventral medullary surface ( V M S ) not only blocks the C O 2 i n d u c e d ventilatory drive but also affects certain reflexes arising in the lung. F o r example, m o d e r a t e cooling o f the V M S to 20 °C attenuates the tachypnic r e s p o n s e resulting from deflation o f the lungs, p r e s u m a b l y due to the stimulation of the rapidly a d a p t i n g p u l m o n a r y stretch receptors, while the B r e u e r - H e r i n g reflex from slowly adapting p u l m o n a r y stretch receptors is Accepted for publication 16 March 1985
0034-5687/85/$03.30 © 1985 Elsevier Science Publishers B.V. (Biomedical Division)
378
J. MITRA etal.
unaffected (Cherniack et al., 1979a,b). Besides the slowly and rapidly adapting pulmonary stretch receptors, tracheopulmonary 'C' fibers are important in the control of breathing (Coleridge and Coleridge, 1984). In the present study we investigated the effects of hypercapnia and cooling of the ventral medullary surface (VMS) on the respiratory responses to pulmonary 'C' fiber (J receptor) stimulation with right atrial injection of capsaicin and phenyldiguinide (PDG). Though the responses of the phrenic activity to pulmonary 'C' fiber excitation is fairly well established (Coleridge and Coleridge, 1984), its effect on the motor output of cranial nerves innervating the upper airways, however, has not yet been examined. Previous studies from this laboratory have demonstrated that the activity of the upper airway muscles increases more than the diaphragm with chemical stimulation and with various other reflexes which affect ventilation (Weiner et al., 1982; Cherniack et al., 1984). In order to obtain more complete information about the respiratory responses to 'J' receptor stimulation, we also measured the activities of the hypoglossal and recurrent laryngeal nerves along with phrenic activity. Methods
Experiments were performed on eleven cats of both sexes (1.5-4.5 kg body weight) which were anesthetized with an initial intraperitoneal injection of c~-chloralose (40-50/~g/kg) followed by an intravenous injection of 10 mg of pentobarbital. Body temperature was maintained at 38 _+ 1 °C using an electric thermal blanket. The saphenous vein of one forelimb was cannulated for infusion of solutions, and the femoral artery to monitor systemic blood pressure. The trachea was intubated for artificial ventilation and to allow the insertion of a capnograph (Godart-Statham) probe to monitor end-tidal CO 2. Since our aim is to study the interaction of different levels of CO2 stimulation with respiratory responses induced by 'J' receptor stimulation without the complications produced by altered lung mechanics on ventilation, all experiments were conducted on paralyzed and artificially ventilated animals. The animals were paralyzed with intravenous administration ofgallamine triethiodide (5 mg/kg; flaxedil). A catheter was inserted through the right external jugular vein so that the tip lay in the right atrium and was used for injection ofcapsaicin. The dead space of the catheter was 0.1 ml. Through a midline incision extending from the symphysis of the lower jaw to the manubrium, the medial branch of the hypoglossal nerve was dissected free from the surrounding tissue. The medial branch was chosen because it exhibits mainly inspiratory activity (Mitra and Cherniack, 1983). The fifth root of the phrenic nerve and the right recurrent laryngeal nerve were also isolated. The cut central ends of these nerves were desheathed and were used for recording the electrical activity using the conventional recording techniques. The electrical activities of the nerves were integrated using a time constant of 100 or 200 msec. The carotid sinus, aortic, and vagus nerves were isolated and were surgically denervated during the course of the experiment. The details of the exposure of ventral medullary surface have been described
VENTRAL MEDULLARY SURFACE & PULMONARY C FIBER REFLEXES
379
elsewhere (Bruce et aL, 1982). In brief, to gain access to the ventral surface of the medulla, the trachea and the esophagus were freed from the tissues, severed, and then reflected rostrally. Subsequently, the underlying muscles were removed and then craniotomy was performed to expose the VMS. The dura was then opened and retracted by ligatures. After opening of the dura, the exposed medullary surface was covered with warm artificial CSF. For the purpose of ventral medullary surface cooling, a specially built small two footed (each 2 mm z) thermode mounted on a micromanipulator was placed on the medullary surface. Cooling was performed by circulating antifreeze of different temperatures through the thermode. The temperature of the cooling probe was continuously monitored with a thermocouple embedded in an area of the foot of the probe in contact with the VMS. The area of the ventral medullary cooled extended between the 7th cranial nerve rostrally (R M area) to the region medial to the hypoglossal roots caudally (CL). Maximum depression of phrenic activity was obtained from an area 1 mm rostral of the hypoglossal roots and 2.5 mm lateral to the midline. The area approximately corresponds to intermediate area (Is) described by Schlaefke etal. (1970). A solution ofcapsaicin (Sigma) was prepared according to the method described by Coleridge et al. (1964). Phenyldiguinide (PDG) was dissolved in saline (0.9~o NaC1). The doses of capsaicin and PDG used in this study were 3 #g and 30 #g/kg, respectively. The total volume of the solution used per injection was 0.5 ml given into the right atrium over a period of 2 sec. During steady-state CO 2 breathing, the inspiratory port of the ventilator was connected to a 3 L bag which contained either 3 or 7 ~/o CO2 in 02. The animals breathed CO2 gases for at least 10 min before any intervention. Analysis. The duration of apnea in phrenic and hypoglossal nerves after the injection of capsaicin was used as an index of respiratory inhibition. During complete phrenic inhibition the increase in recurrent laryngeal activity from its baseline at end inspiration was used as a measure of recurrent laryngeal nerve excitation during the same period. Statistical analysis was performed by using the paired t-test. The criteria for statistical significance was P < 0.05.
Results
A typical response to right atrial (RA) injection of capsaicin (3 #g/kg) in a paralyzed, artificially ventilated cat is shown in fig. 1. Within 2.0 sec following capsaicin injection the activity of the phrenic and hypoglossal nerves completely disappeared almost at the same time, and in this experiment lasted about 7 sec. The change in phrenic activity was associated with a concomitant increase in the recurrent laryngeal activity. The respiratory responses were accompanied with a small biphasic change in arterial pressure, namely an initial slight increase followed by a slight decrease, without a change in heart
J. MITRA etal.
380
Fi CO2 r.o%
,..,.::2j 5 m ,,
sec
.i
Fig. 1. Responses of the phrenic, hypoglossal and recurrent laryngeal nerves to right atrial injection of capsaicin (3 #g/kg) in an anesthetized, paralyzed, and artificially ventilated cat. Following the capsaicin injection within 2 sec the integrated activity of the phrenic (Ph) and hypoglossal(Hy) nerves disappeared and the recurrent laryngeal activity(RL) increased. After capsaicin, recurrent laryngeal activitywas locked with the ventilator. It took 6-8 rain for both neural activities to return to control amplitudes. The arrow indicates capsaicin injection in this and in the next figure. rate. Following the apnea, the reinitiation of the phrenic and hypoglossal nerves occurred at the same time. On an average it took 8-10 min for both the nerve activities to return to the control amplitudes (not shown in fig. 1). Before capsaicin injection, the recurrent laryngeal showed inspiratory activity synchronized with phrenic activity and usually displayed a small amount of expiratory activity. However, during capsaicin induced apnea (reflected in the phrenic and hypoglossal activities) the recurrent laryngeal nerve showed increased phasic activity which was locked to the rhythm of the ventilator (activity was less during lung inflation). As phrenic activity returned, the recurrent laryngeal nerve displayed activity during both expiration and inspiration. Expiratory activity was more exaggerated than before the injection of caps aicin. Over a period of several minutes, the expiratory activity gradually decreased and the inspiratory activity increased. To eliminate the phasic feedback from vagal stretch receptors, the ventilator was stopped briefly and capsaicin was injected. As shown in fig. 2, the phasic activity in the recurrent laryngeal nerve was replaced by tonic activity. This sequence of events was the same in intact and sino-aortic denervated animals. Often a small amount of residual tonic activity in the phrenic nerve was observed if capsaicin was injected while the respirator was stopped. After bilateral vagotomy, capsaicin-induced apnea in phrenic, hypoglossal and excitation of recurrent laryngeal nerves were abolished.
Comparison of PDG and capsaicin. P D G with a dose of 30 #g/kg stimulates J receptors nearly maximally (Anand and Paintal, 1980). In three experiments in paralyzed animals,
VENTRAL MEDULLARY SURFACE & PULMONARY C FIBER REFLEXES
381
Fi CO 2 7 . 0 %
/ Ph.
/ R.L.
150,=
!
5 see
¢
Fig. 2. Effectof capsaicin injectionon the integrated activitiesof phrenic(Ph) and recurrent laryngeal(RL) during the stopping of ventilator in an anesthetizedparalyzedcat. Followingthe capsalein administration, the phasic recurrent laryngealactivitywas replaced by tonic activitywhich is associated with absence of phasic phrenic activity.The first three breaths followingthe tonic activity are expiratory phasic activity, subsequently switched to inspiratory phasic. 30 #g/kg of P D G was given right atrially, In all the three animals, the phrenic nerve activity disappeared on an average for 7-9 see. Recurrent laryngeal also showed an increased activity similar to capsaicin (3 #g/kg). The fall in the arterial pressure and bradycardia, however, was more marked.
Effect of C02 on capsaicin-induced reflex responses. We investigated the effect of C O 2 on the duration of apnea in the phrenic and hypogiossal nerves in six animals. The animals breathed two levels of CO2 (3 and 7% CO2) in 02 mixtures. Capsalcin was injected after the animals inspired CO2 reached a steady-state level. The results are summarized in fig. 3. At each level of CO2 the duration of apnea in the hypoglossal and phrenic nerves was identical. With increasing CO2 the duration of apnea in both phrenic and hypoglossal nerves decreased significantly from 9 to 6.5 see (P < 0.01). The reduction in the duration of hypoglossal and phrenic nerve apnea by COz was accompanied by a decrease in the amplitude of the recurrent laryngeal activity. As inspired CO 2 was increased from 3 to 7 % the peak amplitude of the integrated recurrent laryngeal activity after capsaicin injection on average fell by 80% (P < 0.01).
382
J. MITRA et al.
10 -
[]
Phrenic
[]
Hypoglossal
T I::::::::::
v
!iiiiiiiil
~8-
::::::::::
t~
~7
-
ii~iiiiii: :1111:::1:
g ~6
-
iiiiiiiil !i....iiil
3
El 5 -
3
7
3
7
%F I C 0 2
Fig. 3. The effect of elevated C O 2 o n the apneic duration of phrenic (Ph) and hypoglossal (Hy) nerve activities. The data presented are mean values with standard error of mean. P < 0.01 (paired t-test)•
Effect of ventralmedullarysurface cooling. Since elevated CO2 reduced the duration of capsaicin induced apnea, focal cooling of the central chemoceptive structures located on the ventral medullary surface might be expected to have an opposite effect. In another group of five animals during breathing 7~o CO2 in 02, rostral (RM), intermediate (Is) and caudal (CL) areas were subjected to graded focal cooling. Once the phrenic amplitude reached the steady-state level during cooling, capsaicin injections were repeated. Cooling the intermediate area (Is) to 20 ° C markedly prolonged the capsaicin induced apnea in phrenic and hypoglossal nerves. On the other hand, cooling the caudal (CL) had less effect, while rostral area (RM) had the least effect. Since the responses obtained with cooling of the I s area are more pronounced, the results were analyzed only with cooling of Is area. The temperature of the I s area was lowered in a graded fashion from 37 to 20 °C and then to 15 °C. Cooling dissociated the apneic duration of phrenic and hypoglossal nerves. Lowering the temperature to 20 °C increased capsaicin induced apnea in the hypoglossal nerve to 90 see, whereas in the phrenic it was only 18 sec. Further lowering of temperature to 15 ° C eliminated completely the hypoglossal activity, but it prolonged the phrenic apnea to 80 sec. In all the experiments, lowering the temperature of the 1s area to 15 °C completely eliminated the hypoglossal activity. Therefore, for quantitative analysis, the effect of capsaicin with I s area temperature of 37 ° C was compared with results obtained with cooling to 20 °C. The results are summarized in fig. 4. At 37 °C the duration of apnea for both phrenic and hypoglossal was similar. When I s temperature was lowered to 20 ° C, the duration of apnea for the hypoglossal nerve was increased more than the phrenic, which is statistically significant (P < 0.01). Although the peak amplitude of the integrated recurrent laryngeal activity was somewhat higher at 20 °C (35.3 _+ 9.1 A. U) than at 37 °C (28.9 + 4.5 A. U), the difference was not statistically significant.
VENTRAL MEDULLARY SURFACE & PULMONARY C FIBER REFLEXES 90
383
FI C027% Ph H
8O
,qr'
Hy 0 . . . .
0
7O I 6O !
6
/i
,o
"~
•
40 !
I c~
I
/ / I,
I
! J=
3o
I 20
I
I
lO
I
I
37
20 T is ( c ' )
Fig. 4. Comparison of the capsaicin-induced apneic duration in hypoglossal (Hy) and phrenic nerve during the cooling of Is area to 20 °C while the animals were ventilated with 7% CO2 in 02,
Discussion
The results of the present study show that fight atrial injection of capsaicin induces apnea not only in the phrenic but also in the upper airway nerves like the hypoglossal. They also suggest that changes in CO 2 drive affect the intensity of the reflex response. Cooling prolonged the reflex apnea caused by capsaicin injection in phrenic and hypoglossal nerves, while hypercapnia shortened it. The data obtained with hypercapnia were different from the observations made by Anand and Paintal (1980). They found that hypoxia, but not hypercapnia, attenuated the reflex apnea caused by stimulation of J receptors with phenyldiguinide. The difference could be due to the fact that their animals were spontaneously breathing 4Yo CO2 in room air, while our animals were paralyzed and ventilated with 3 and 7 YoCO2 in 02. According to Palntal (1973) if apnea appears within 2.5 sec in the phrenic following the right atrial injection of PDG, it can be regarded as due to stimulation of J receptors. In all our experiments, within 2 sec after right atrial injection of capsalcin and PDG, apnea was consistently observed in phrenic and also in hypoglossal nerves, suggesting that the effect probably resulted from J receptor stimulation, Furthermore, the pattern of the respiratory responses was essentially the same for both capsaicin and PDG. Although the duration of apnea was similar for both phrenic and hypoglossal nerves when the I s area was at 37 °C, lowering of the I s temperature to 20 °C increased the duration of apnea for hypoglossal nerve more than the phrenic. Thus it is conceivable that the increase in the duration of absent hypoglossal activity could be due to cooling the XII rootlets, which are anatomically located close to I s area. However, unilateral
384
J. M I T R A etal.
cooling of I s area to 20 °C eliminates the activity in the contralateral hypoglossal nerve indicating that the observed effects are not caused by rootlet cooling. Cooling the I s area might decrease central chemoreceptor afferent input. It is possible that the low CO2 drive in itself dissociates hypoglossal and phrenic response to capsaicin. Previously we reported that hypoglossal nerve activity was more sensitive to chemical manipulation of the VMS than phrenic nerve activity (Weiner etal., 1982). Also, during hyperventilation hypoglossal activity disappears first relative to phrenic activity. To test this hypothesis we hyperventilated two animals with 100~o 02 to reduce the hypoglossal phasic activity approximately to the level observed during I s area cooling to 20 °C and then capsaicin was injected. In these two experiments the apneic duration of both phrenic and hypoglossal nerves was similar. Hence, it seems that I s area cooling has a different effect from hypocapnia on capsaicin-induced apnea. Cooling of the VMS increases the inhibitory response of capsaicin injection on the hypoglossal more than the phrenic while hypocapnia does not. The significance of this is not yet clear, but suggests that cooling may not affect exactly the same structures as low CO 2. The effects of H + and CO2 on cerebral vasculature and blood flow are known (Heistad and Kontos, 1983). However, we are not aware of studies dealing with the effects of focal cooling on cerebral vasculature and microcirculation. Recurrent laryngeal nerve activity increased during reflex inhibition ofphrenic activity by capsaicin or PDG. Stransky etal. (1973) reported that in non-paralyzed cat intravenous injection of phenyldiguinide caused apnea and complete closure of the larynx. Our observation of increased recurrent laryngeal nerve activity agrees well with their observation and also suggests that the responses induced by capsaicin are identical with PDG. In our study, recurrent laryngeal nerve in the paralyzed animal had phasic activity during reflex apnea which was locked to ventilatory frequency. When we stopped the ventilator for a few seconds during apnea, the phasic recurrent laryngeal activity was replaced by tonic activity, which suggests that the phasic activity was caused by stimulation of vagal sensory tracheopulmonary receptors. As phrenic nerve activity returned, the recurrent laryngeal nerve began to show more expiratory than inspiratory activity, which gradually decreased over a period of several minutes to pre-injection level. Koepchen etal. (1977) have shown that stimulation of J receptors leads to a marked increase of expiratory modulated neuronal activity which lasted for several minutes after P D G injection. The long-lasting recurrent laryngeal expiratory activity after capsaicin could be due to long-lasting stimulation of these expiratory neurons. Whether in a spontaneously breathing animal a similar increase in the expiratory activity of expiratory muscles occurs following capsaiein-induced apnea is yet to be determined. We also found that the amplitude of the recurrent laryngeal nerve activity during phrenic inhibition is inversely related to CO 2 drive; i.e., the lower the CO2 level, the bigger the amplitude. However, when the effect ofcapsaicin on recurrent laryngeal nerve activity at 20 °C I s area temperature was compared with 37 °C I s area temperature, activity was greater but there was no statistically significant difference. It may be that strong inhibition of inspiratory motoneuron pool by vagal 'C' fibers at low CO 2 level
VENTRAL MEDULLARY SURFACE & PULMONARY C FIBER REFLEXES
385
causes a release of expiratory activity from the recurrent laryngeal motoneuron pool. Nonetheless, these experiments also suggest a quantitative if not qualitative difference between VMS cooling and hypocapnia. Another observation made in the present study is that in paralyzed animals capsaicin causes negligible fall in blood pressure and lacks bradycardia. These results are in variance with responses reported in spontaneous breathing animals, where stimulation of pulmonary C fibers causes bradycardia and hypotension (Coleridge et al., 1964). In paralyzed animals, it seems that the responses of pulmonary chemoreflex include apnea while hypotension, bradycardia and tachypnea are either absent or reduced greatly. References Anand, A. and A. S. Paintal (1980). Reflex effects following selective stimulation of J receptors in the cat. d. Physiol. (London) 299: 553-572. Bruce, E. N., J. Mitra and N. S. Cherniack (1982). Central and peripheral chemoreceptor inputs to phrenic and hypoglossal motoneurons. J. AppL Physiol. 53:1504-1511. Cherniack, N. S., C. yon Euler, I. Homma and F.F. Kao (1979a). Graded changes in central chemoreceptor input by local temperature changes on the ventral surface of medulla. J. Physiol. (London) 287: 191-209. Cherniack, N. S., C. von Euler, I. Homma and F.F. Kao (1979b). Interactions between a central chemoceptive system and other ventilatory drives. In: Wenner-Gren Center International Symposium Series. Vol. 32, Central Nervous Control Mechanisms in Breathing, edited by C. yon Euler and H. Lagercrantz. Oxford, Pergamon Press, pp. 35-41. Cherniack, N. S., M. S. Haxhiu, J. Mitra, K. Strohl and E. van Lunteren (1984). Responses of upper airway, intercostal and diaphragm muscle activity to stimulation of oesophageal afferents in dogs. J. Physiol. (London) 349: 15-25. Coleridge, H. M., J. C. G. Coleridge and C. Kidd (1964). Role of the pulmonary arterial baroreceptors in the effects produced by capsaicin in the dog. J. Physiol. (London) 170: 272-285. Coleridge, J. C. G. and H.M. Coleridge (1984). Afferent vagal C fibre innervation of the lungs and airways and its functional significance. Rev. Physiol. Biochem. Pharmacol. 99: 1-110. Heistad, D.D. and H.A. Kontos (1983). Cerebral circulation. In: Handbook of Physiology. Sec. 2. The Cardiovascular System. Vol. III. Peripheral Circulation. Part I, edited by J.T. Shepherd and F.M. Abboud. Bethesda, MD, American Physiological Society, pp. 137-182. Koepchen, H.P., M. Kalia, D. Sommer and D. Kltissendorf (1977). Action of type J afferents on the discharge pattern of medullary respiratory neurons. In: Respiratory Adaptations, Capillary Exchange and Reflex Mechanisms, edited by A.S. Paintal and P. Gill-Kumar. Delhi, V.P. Chest Institute, pp. 407-425. Mitra, J. and N.S. Cherniack (1983). The effects of hypercapnia and hypoxia on single hypoglossal nerve fiber activity. Respir. Physiol. 54: 55-66. Paintal, A.S. (1973). Vagal sensory receptors and their reflex effects. Physiol. Rev. 53: 159-227. Schlaefke, M.E., W. See and H.H. Loeschcke (1970). Ventilatory response to alterations of H + ion concentration in small areas of the ventral medullary surface. Respir. Physiol. 10: 198-212. Stransky, A., M. Szereda-Prazestaszewska and J.G. Widdicombe (1973). The effects of lung reflexes on laryngeal resistance and motoneurone discharge. J. Physiol. (London) 231: 417-438. Weiner, D., J. Mitra, J. Salamone and N. S. Cherniack (1982). Effect of chemical stimuli on nerves supplying upper airway muscles. J. Appl. PhysioL 52: 530-536.