- Email: [email protected]
Direct pathway from neurons in the ventrolateral medulla relaying cardiovascular afferent information to the supraoptic nucleus in the cat
Brain Research, 292 (1984) 221-228 Elsevier
221
Direct Pathway From Neurons in the Ventrolateral Medulla Relaying Cardiovascular Afferent Information to the Supraoptic Nucleus in the Cat JOHN CIRIELLO and MONICA M. CAVERSON Department of Physiology, Health Sciences Centre, University of Western Ontario, London, N6A 5C1 (Canada) (Accepted June 7th, 1983) Key words: supraoptic nucleus - - ventrolateral medulla - - aortic depressor nerve - - carotid sinus nerve - ascending cardiovascular pathways
In chloralose anesthetized cats experiments were done to electrophysiologically identify neurons in the ventrolateral medulla (VLM) which relay cardiovascular afferent information directly to the supraoptic nucleus (SON). Action potentials elicited antidromically by electrical stimulation of the SON were recorded from 69 histologically verified single units in the VLM. Single units responded with latencies corresponding to conduction velocities of 7.8 + 0.6 m/s. Of these units 26 were excited orthodromically by stimulation of the buffer nerves; 12 responded to stimulation of only the carotid sinus nerve, 7 responded to stimulation of only the aortic depressor nerve, and 7 responded to both buffer nerves. The axons of VLM units that responded to buffer nerves conducted at a significantly slower velocity than those of non-responsive units (5.7 +_0.4 and 9.1 + 0.8 m/s, respectively). These data provide electrophysiological evidence of two different populations of VLM neurons which project directly to the SON, and suggest that the direct pathway from the VLM to the SON is involved in the release of vasopressin by SON neurons during activation of baroreceptor and chemoreceptor afferent fibers. INTRODUCTION It is well established that activation of baroreceptor and c h e m o r e c e p t o r afferent fibers in the carotid sinus (CSN) and aortic d e p r e s s o r ( A D N ) nerves, and of the nucleus of the solitary tract, the primary site of termination of cardiovascular afferent fibers 12-14, alter both the rate of discharge of neurons in the paraventricular and supraoptic (SON) nuclei 5,6.17,1s,35, and the release of vasopressin 31,32. A l t h o u g h a direct cardiovascular pathway from the nucleus of the solitary tract to the parvocellular division of the paraventricular nucleuslO,U, and possibly a disynaptic pathway through the parabrachial nucleusa6, 21,28, have been d e m o n s t r a t e d , the brainstem pathways which mediate cardiovascular afferent information to either the magnocellular division of the paraventricular nucleus or the S O N are not known. Recent anatomical studies have described a projection from neurons in the nucleus of the solitary tract to the region of the ventrolateral medulla (VLM2 O, which then in turn projects directly to the 0006-8993/84/$03.00 © 1984 Elsevier Science Publishers B.V.
magnoceilular n e u r o s e c r e t o r y portions of the paraventricular nucleus and the S O N 22,29. This evidence suggests that the V L M m a y be a site of origin of an ascending pathway through which afferent information from b a r o r e c e p t o r s and c h e m o r e c e p t o r s reach the SON. This suggestion is further s u p p o r t e d by the demonstration that single units in the region of the V L M alter their rate of discharge during electrical stimulation of the CSN and A D N , and selective activation of b a r o r e c e p t o r s and chemoreceptors7, s. In addition, lesions of the V L M have been shown to attenuate baroreceptor23,24 and chemoreceptor8 reflex responses. F u r t h e r m o r e , destruction of n o r a d r e n e r gic neurons in the V L M results in an elevation of plasma vasopressin 4, possibly due to the removal of a tonic inhibition on vasopressinergic neurons exerted by cardiovascular afferent reflex mechanisms. The present study was done to provide electrophysiological evidence of a m o n o s y n a p t i c p a t h w a y mediating cardiovascular afferent information from the V L M to the SON. The V L M was systematically explored for single units antidromically excited by elec-
222 trical stimulation of the SON. These units were then tested for their responses to stimulation of the CSN and A D N . METHODS Experiments were done in 8 adult cats of either sex anesthetized with a-chloralose (60 mg/kg, i.v., initially, supplemented by additional doses of 30 mg/kg at 8-10 h intervals) after ethyl chloride and ether induction. The trachea was cannulated, the animals were paralyzed with d e c a m e t h o n i u m b r o m i d e (0.5 mg/kg, i.v., initially and additional doses when necessary), and then artificially ventilated. Polyethylene catheters were inserted in the femoral artery and vein for the recording of arterial pressure and the administration of drugs, respectively. The arterial pressure was recorded through a Statham P23Dd transducer, and heart rate was m o n i t o r e d with a 7P44B Grass tachograph triggered by the arterial pressure pulse, both of which were continuously recorded on a Grass 7 polygraph. Rectal t e m p e r a t u r e was mainrained at 37 + 0.2 °C by a heating pad controlled by a Yellow Springs 73 t e m p e r a t u r e controller. The head of the animal was fixed in a K o p f stereotaxic instrument, and access to the VLM was obtained by removing the occipital bone and exposing the floor of the fourth ventricle after removing the cerebellum by suction. Access to the SON was obtained by a unilateral parietal craniotomy. Exposed nervous tissue was covered with warm 360 medical fluid (Dow Corning, Midland, MI) to prevent drying. The left CSN was approached through a lateral incision at the level of the tympanic bulla. The sternocleidomastoid and digastric muscles overlying the sinus region were reflected; a large lymph node and a 2 cm segment of the hypoglossal nerve were removed. The CSN was isolated from surrounding tissue and crushed as far away as possible from its junction with the glossopharyngeal nerve. The left A D N was identified near the junction of the superior laryngeal and vagus nerves through a midline incision in the neck at the level of the larynx, dissected free for a few cm from the cervical vagus, and crushed distally. The central ends of the isolated nerves were placed on bipolar stainless steel electrodes and covered with cotton pellets soaked in warm 36{) medical fluid to prevent drying.
Recording single unit activity in the VLM The region of the VLM was systematically explored; it extended from 1.0 mm caudal to 4.0 mm rostral to the obex, from 3 to 5 mm lateral to the midline and from 2 mm below the dorsal surface to the ventral surface. The reference point for positioning the electrode was the obex and penetrations were made on a grid with points 500/~m apart. Stainless steel microelectrodes 15 with tip diameters of 1-3/~m and a resistance in saline of 0.8-2 Mr2 were used for extracellular recording. The indifferent electrode was a hypodermic needle inserted into the brain tissue through a small hole in the right parietal bone. Unit activity was amplified through a Grass P15 differential preamplifier, and displayed on a Tektronix R5103N oscilloscope for observation and photography. Spontaneously active and silent single units in the VLM encountered during an electrode penetration were tested for their response to electrical stimulation of the ipsilateral SON with 0.1-0.2 ms rectangular pulses at (I.5 Hz and at current intensities of 0.25-1.5 m A delivered through the central electrode of bipolar stainless steel electrodes (SNEX-100, David Kopf, Tujunga, C A ; 0.25 mm tip diameter; 50-90 k ~ initial DC resistance in saline) from a Grass $88 stimulator through a Grass PSIU6B stimulus isolation and constant current unit. The indifferent electrode was an alligator clip attached to exposed scalp muscle. Single units in the V L M that responded to electrical stimulation of the SON were assessed for antidromic activation using previously established criteria (for a review see ref. 20): (a) constant latency of the evoked spike: (b) high following frequency of the evoked spike; (c) occurrence of a single evoked spike at threshold and suprathreshold stimulus intensities, and (d) collision of evoked and spontaneous spikes. Not all single units classified as antidromically activated were evaluated for all criteria, as 80f/~ of the units were not spontaneously active and the activity could not be tested by the collision method. All spontaneously active units showed collision of the evoked and spontaneous spikes (Fig. 1). Units antidromically excited by stimulation of the SON were subsequently tested for responses to stimulation of the CSN and A D N with pulses of 0.3 ms duration at 0.5 Hz and current intensities up to 5 times the current required to elicit a threshold decrease in heart rate when using a 15 s stimulus train at 25 Hz and a
223 pulse duration of 0.3 ms. The ' t h r e s h o l d ' stimulus was defined as the current required to elicit a 10-15 beat/rain decrease in heart rate. Stimulus intensities at 5 times threshold have previously been shown to elicit maximal reflex b r a d y c a r d i a during electrical stimulation of either buffer nerve 9.
"14+ B
Histological localization of recording and stimulation sites Sites of recording and stimulation were identified by depositing iron from the electrode tip (20/~A for 15 s, tip positive). A n i m a l s were perfused with 0.9% saline followed by a 1% potassium ferrocyanide in 10% formalin-saline solution to reveal the m a r k e d sites by the Prussian blue reaction. The brains were fixed in 10% formalin for at least 3 days and transverse 50/~m frozen sections of the h y p o t h a l a m u s and medulla were cut and stained with thionin. Stimulation and recording sites were m a p p e d on transverse sections of the h y p o t h a l a m u s modified from a stereotaxic atlas I and from projection drawings of the medulla. The conduction distance for each unit was calculated from these maps and ranged between 21.5 and 27.8 mm.
Data analysis Statistical analysis of means + S.E. of conduction velocities was done by using a Student's t-test. P < 0.01 was considered to be statistically significant.
A
RESULTS A total of 240 electrode p e n e t r a t i o n s were m a d e through the region of the VLM. Sixty-nine histologically verified single units in the V L M were antidromically activated by electrical stimulation of the SON. Of these units, 14 were discharging spontaneously (mean discharge rate, 3.5 + 1.2 spikes/s) and 55 were silent. Units r e s p o n d e d with a mean latency of 5.0 + 0.4 ms (range, 1.7-17 ms) which was constant for any one unit and followed rates of stimulation of 100-500 Hz. The latencies of the antidromic responses corres p o n d e d to conduction velocities of these ascending fibers of 7.8 + 0.6 m/s (Fig. 2). A s seen in Fig. 2, there a p p e a r to be two distinct populations of units in the V L M projecting to the S O N identified on the basis of their conduction velocities; the first group contains units whose axons conduct in the range of 1-7
t
v
I
Fig. 1. Response of spontaneously active single unit in VLM to stimulation of SON. A: spontaneous discharge rate of VLM unit. Single sweep. Calibrations, 5 s and 200~V. B: antidromic response of VLM unit to stimulation of SON at 0.5 Hz. Arrow indicates time of stimulation; 5 superimposed sweeps. Calibrations, 2 ms and 200 ~V. C: cancellation of antidromic spike by spontaneous activity in VLM unit; 6 successive sweeps; stimulus delivered at arrow. Asterisks mark sweeps in which cancellation occurred. Calibrations, 2 ms and 200 ~V.
m/s (mean, 4.4 + 0.3 m/s; n = 36), and the second group contains units which r e s p o n d with latencies corresponding to conduction velocities of greater
224 than 7 m/s (mean, 11.5 + 0.6 m/s; range, 7.2-27 m/s; n
I0
= 33). The threshold current to evoke these antidromic spikes ranged from 0.25 to 1.5 m A (mean, 0.75 +
n=69
0.04 m A when using a pulse duration of 0.1-0.2 ms). The threshold current was found to be always less than 0.5 m A when the tip of the stimulating electrode was located within the SON and usually greater when
Number of
it was located medial, lateral or just dorsal to the SON. All units responded with a single spike (duration of spike, 0.8-3.0 ms) at threshold and supra-
Units
4
threshold stimulus intensities. All antidromically ac-
2
tivated units showed two-component spikes (initial segment and somatodendritic components) of which, 0
, 0
,
, 2
i
,
4
|
,
,
,
6 8 Conduction
i
,
w
,
I0 12 Velocity
n
i
w
,
14 16 (m/s)
26
2B
Fig. 2. Histogram of conduction velocities of 69 VLM single units antidromically activated by electrical stimulation of SON. Hatched area represents the conduction velocities of antidromically activated units which were also excited by stimulation of the buffer nerves.
A
in some cases, the somatodendritic c o m p o n e n t could be induced to fail by high frequency of stimulation (Fig. 3C). Characteristic antidromic potentials recorded from neurons in the VLM during stimulation of the SON are shown in Figs. 1 and 3. Of the 69 antidromically identified units, 26 responded orthodromically to stimulation of the buffer
B
t
t/
/
tt
t
Fig. 3. Response of an antidromically activated unit in VLM to different frequencies of stimulation of SON (A-C) and to stimulation of the CSN (D). Each record is 5 superimposed sweeps, and the stimuli were delivered at arrows. A: note the constant latency and 2 component (IS-SD) spike in record at 0.5 Hz. B: action potentials evoked with two stimuli applied at 116 Hz. C: note the separation of the IS-SD components (*) and the occasional failure of the SD component of the spike after the second stimulus at 500 Hz. D: orthodromic response of same unit to stimulation of the CSN. Calibrations, 2 ms and 100/~V(A-C), and 5 ms and 100~V (D).
225
Obex
+5.0
|
I
2mm Fig. 4. Location of units antidromically excited by stimulation of the SON and orthodromicallyexcited by stimulation of the buffer nerves plotted on representative transverse sections of the VLM of the cat (obex to 4 mm rostral to obex). O, units excited only by CSN stimulation; A, units excited only by ADN stimulation; *, units excited by stimulation of both the CSN and ADN; O, units not responding to stimulation of buffer nerves. 5SP, spinal trigeminal nucleus; 5ST, spinal trigeminal tract; 12N, hypoglossal nerve; ION, inferior olivary nucleus; LRN, lateral reticular nucleus; P, pyramidal tract; PPR, postpyramidal nucleus of the raphe.
response. Fig. 3D shows a characteristic response of an antidromically identified unit orthodromically excited by stimulation of the CSN. The remaining 43 units did not respond to stimulation of either buffer nerve. Single units orthodromically excited by stimulation of the buffer nerves responded antidromically to stimulation of the SON with latencies corresponding to a mean conduction velocity of 5.7 + 0.4 m/s (range, 1.7-9.5 m/s), whereas the mean conduction velocity of the units non-responsive to the buffer nerves was 9.1 + 0.8 m/s (range, 1.4-27 m/s; Fig. 2). Although the ranges of conduction velocities of the responsive and non-responsive units overlap, the mean conduction velocities for the two groups were significantly different (P < 0.005). The anatomical distribution of units activated antidromically by stimulation of the SON and orthodromically excited by stimulation of the buffer nerves is shown in Fig. 4. These units were located primarily in a region ventromedial to the lateral reticular nucleus and lateral to the intramedullary rootlets of the hypoglossal nerve. This group of neurons extended into the region just dorsal to the lateral reticular nucleus. On the other hand, the non-responsive antidromically activated units occupied the region immediately dorsolateral and ventrolateral to that containing units responsive to the buffer nerves. These latter units were scattered mostly near the periphery of the lateral reticular nucleus, as defined cytoarchitectonically, and in the reticular formation in a region ventrolateral and dorsolateral to this nucleus. The majority of units (60/69) were found in the caudal ventrolateral medulla extending from 0.5 mm caudal to 2.0 mm rostral to the obex. DISCUSSION
nerves; 12 were excited by stimulation of only the CSN with a mean latency of 11.7 + 1.9 ms, 7 were excited by stimulation of only the A D N with a mean latency of 9.9 + 1.1 ms, and 7 were excited by both the CSN and A D N (mean latencies, 8.9 + 0.8 and 11.0 + 1.1 ms, respectively). The evoked orthodromic response of these units to stimulation of the buffer nerves consisted of either a single spike or a burst of 2-3 spikes. Increasing the intensity of the stimulus usually decreased the latency of the response and caused an increase in the number of spikes during a
These data provide the first electrophysiological evidence of a monosynaptic pathway from neurons in the VLM mediating cardiovascular afferent information to the SON. These findings are supported by the recent anatomical demonstration in the rat and rabbit of direct connections between neurons in the VLM and the SON using both the autoradiographic technique 3.22.29 and the retrograde transport of horseradish peroxidase and fluorescent dyes 3.29. Neurons in the VLM have been shown to innervate
226 that portion of the SON which contains primarily vasopressinergic cell bodies 29, and it has been suggested that these VLM neurons are noradrenergic 3.22,29. The anatomical distribution of units activated antidromically by stimulation of the SON and orthodromically by stimulation of the buffer nerves corresponds to that of catecholamine-containing neurons in the VLM of the cat 2.19,25 and to that described for DBH-stained neurons in the A1 cell group of the VLM in the rat 29. This suggests the possibility that some of the neurons relaying cardiovascular afferent information to the SON were noradrenergic. This suggestion is supported by the observation that destruction of the A I noradrenergic neurons in the VLM results in an increase in the release of vasopressin due to the removal of an inhibitory input into magnocellular neurosecretory neurons 4. Since baroreceptor reflex mechanisms are known to maintain a tonic inhibition on vasopressinergic neurons 29,3°, this further suggests that some of the neurons recorded in this study in the region of the VLM are likely to be noradrenergic and involved in the release of vasopressin. However, the possibility cannot be excluded that neurons which do not receive buffer nerve information are also catecholaminergic nor that non-catecholaminergic neurons in the VLM also project to magnocellular neurosecretory SON neurons. An unexpected finding was that axons of neurons in the VLM relaying cardiovascular afferent information conducted at significantly slower velocities than axons of neurons which were non-responsive to buffer nerve stimulation. The conduction velocities of these ascending cardiovascular fibers is in agreement with an earlier estimate6. These data suggest that there are two different populations of axons that project directly to the SON originating from functionally different neurons in the VLM. The finding that neurons in the VLM projecting to the SON responded to stimulation of the CSN and A D N is in agreement with an earlier study showing that single units in the SON alter their rate of discharge during stimulation of the buffer nerves 6. Furthermore, the finding that the majority of VLM neurons responding to buffer nerve stimulation were excited by stimulation of the CSN is in agreement with the previous demonstration that most of the units responding to stimulation of buffer nerves in the SON were excited by stimulation of the CSN 6. In addition,
the observation that some VLM units respond to both the CSN and A D N suggests that these neurons process functionally equivalent afferent information originating in either baroreceptors and chemoreceptors from the two major sites of cardiovascular receptors. The finding of neurons in the VLM orthodromically activated by stimulation of the A D N and CSN was not unexpected as buffer nerve responses have previously been recorded in this area7, s. The likely pathway by which buffer nerve information is relayed to the VLM is from the nucleus of the solitary tract to the VLM, as it has been demonstrated that fibers from the nucleus of the solitary tract project directly to the region of the VLM 21.2a,26. This suggestion is supported by the finding that electrical stimulation of the nucleus of the solitary tract alters the activity of neurons in the ventrolateral medulla which project directly to the SON (Ciriello and Caverson, unpublished observations, 1983). In view of the previous demonstration that selective activation of baroreceptots and chemoreceptors alter the discharge rate of vasopressinergic neurons in the SON 18,35 and that lesions of the VLM result in an increase in the plasma level of vasopressin 4, this direct pathway to the SON from VLM neurons is suggested to be involved in the regulation of vasopressin release by magnocellular neurosecretory SON neurons. It is not surprising to find that some of the neurons in the VLM antidromically activated by stimulation of the SON were not responsive to the buffer nerves, as this area has been shown also to relay visceral sensory information unrelated to the cardiovascular system (for review see ref. 29). In addition, some of these units were found close to the ventral surface of the brainstem in an area known to be sensitive to [H +] and p C O : (for review see ref. 30). Activation of central CO 2 chemoreceptors has been shown to alter the metabolic activity of the V L M and SON 27 and influence the release of vasopressin 33, suggesting that some of these neurons are involved in mediating central chemoreceptor information to the SON. In summary, these data have provided electrophysiological evidence of a direct cardiovascular pathway from neurons in the VLM to the SON, and suggest that this pathway is involved in the regulation of vasopressin release from magnoceilular neurosecretory neurons in the SON during activation of cardiovascular afferent fibers.
227 S c h o l a r . T h e a u t h o r s w i s h t o t h a n k D r s . G . J. M o -
ACKNOWLEDGEMENTS
g e n s o n a n d D. L. J o n e s f o r t h e i r c o m m e n t s o f t h e This work was supported by the Ontario Heart F o u n d a t i o n . J . C . is a C a n a d i a n
Heart Foundation
REFERENCES 1 Bleier, R., The Hypothalamus of the Cat, Johns Hopkins Press, Baltimore, MD, 1961. 2 Blessing, W. W., Frost, P. and Furness, J. B., Catecholamine cell groups of the cat medulla oblongata, Brain Research, 192 (1980) 69-75. 3 Blessing, W. W., Jaeger, C. B., Ruggiero, D. A. and Reis, D. J., Hypothalamic projections of medullary catecholamine neurons in the rabbit: a combined catecholamine fluorescence and HRP transport study, Brain Res. Bull., 9 (1982) 279-286. 4 Blessing, W. W., Sved, A. F. and Reis, D. J., Destruction of noradrenergic neurons in rabbit brainstem elevates plasma vasopressin, causing hypertension, Science, 217 (1982) 661-663. 5 Calaresu, F. R. and Ciriello, J., Electrophysiology of the hypothalamus in relation to central regulation of the cardiovascular system. In P. Meyer and H. Schmitt (Eds.), Nervous System and Hypertension, John Wiley, New York, 1979, pp. 129-136. 6 Calaresu, F. R. and Ciriello, J., Projections to the hypothalamus from buffer nerves and n~.cleus tractus solitarius in the cat, Amer. J. Physiol., 239 (1980) R130-R136. 7 Calaresu, F. R. and Ciriello, J., Projections of buffer nerves to medulla and hypothalamus. In P. S. Sleight (Ed.), Baroreceptors and Hypertension, Oxford University Press, Oxford. 1980, pp. 252-260. 8 Ciriello, J. and Calaresu, F. R., Lateral reticular nucleus: a site of somatic and cardiovascular integration in the cat, Amer. J. Physiol., 233 (1977) R100-R109. 9 Ciriello J. and Calaresu, F. R., Separate medullary pathways mediating reflex vagal bradycardia to stimulation of buffer nerves in the cat, J. Auton. Nerv. Syst., 1 (1979) 13-32. 10 Ciriello J. and Calaresu, F. R., Monosynaptic pathway from cardiovascular neurons in the nucleus tractus solitarii to the paraventricular nucleus in the cat, Brain Research, 193 (1980) 529-533. 11 Ciriello J. and Calaresu, F. R., Autoradiographic study of ascending projections from cardiovascular sites in the nucleus tractus solitarii in the cat, Brain Research, 186 (1980) 448-453. 12 Ciriello J. and Calaresu, F. R., Projections from buffer nerves to the nucleus of the solitary tract: an anatomical and electrophysiological study in the cat, J. Auton. Nerv. Syst., 3 (1981) 299-310. 13 Ciriello, J., Hrycyshyn, A. W. and Calaresu, F. R., Horseradish peroxidase study of brain stem projections of carotid sinus and aortic depressor nerves in the cat, J. Auton. Nerv. Syst., 4 (1981) 43-61. 14 Crill, N. E. and Reis, D. J., Distribution of carotid sinus and depressor nerves in cat brain stem, Amer. J. Physiol., 214 (1968) 269-276. 15 Green, J. D., A simple microelectrode for recording from central nervous system, Nature (Lond.), 182 (1958) 962.
m a n u s c r i p t a n d Miss G . M c C r a e f o r t e c h n i c a l assistance.
16 Hamilton, R. B., Ellenberger, H., Liskowsky, D. and Schneiderman, N., Parabrachial area as mediator of bradycardia in rabbits, J. Auton. Nerv. Syst., 4 (1981) 261-281. 17 Kannan, H. and Koizumi, K., Pathways between the nucleus tractus solitarius and neurosecretory neurons of the supraoptic nucleus: electrophysiological studies, Brain Research, 213 (1981) 17-28. 18 Kannan, H. and Yagi, K., Supraoptic neurosecretory neurons: evidence for the existence of converging inputs both from carotid baroreceptors and osmoreceptors, Brain Research, 145 (1978) 385-390. 19 Lackner, K. J., Mapping of monoamine neurones and fibres in the cat lower brainstem and spinal cord, Anat. Embryol., 161 (1980) 169-195. 20 Lipski, J., Antidromic activation of neurones as an analytic tool in the study of the central nervous system, J. Neurosci. Meth., 4 (1981) 1-32. 21 Loewy, A. D. and Burton, H., Nuclei of the solitary tract: efferent projections to the lower brain stem and spinal cord of the cat, J. comp. Neurol., 181 (1978)421--450. 22 Loewy, A. D., Wallach, J. H. and McKellar, S., Efferent connections of the ventral medulla oblongata in the rat, Brain Res. Rev., 3 (1981) 63-80. 23 McAIlen, R. M., Neil, J. J. and Loewy, A. D., Effects of kainic acid applied to the ventral surface of the medulla oblongata on vasomotor tone, the baroreceptor reflex and hypothalamic autonomic responses, Brain Research, 238 (1982) 65-76. 24 Norgren, R., Projections from the nucleus of the solitary tract in the rat, Neuroscience, 3 (1978) 207-218. 25 Poitras, D. and Parent, A., Atlas of the distribution of monoamine-containing nerve cell bodies in the brain stem of the cat, J. comp. Neurol., 179 (1978) 699-718. 26 Ricardo, J. A. and Koh, E. J., Anatomical evidence of direct projections from the nucleus of the solitary tract to the hypothalamus, amygdala, and other forebrain structures in the rat, Brain Research, 153 (1978) 1-26. 27 Rohlicek, C. V., Ciriello, J. and Polosa, C., Deoxyglucose mapping of central nervous system structures during central chemoreceptor mediated hypercapnic responses in the rat, Soc. Neurosci. Abstr., 7 (1981) 116. 28 Saper, C. B. and Loewy, A. D., Efferent connections of the parabrachial nucleus in the rat, Brain Research, 197 (1980) 291-317. 29 Sawchenko, P. E. and Swanson, L. W., The organization of noradrenergic pathways from the brainstem to the paraventricular and supraoptic nuclei in the rat, Brain Res. Rev., 4 (1982) 275-325. 30 Schlaefke, M. E., Central chemosensitivity: a respiratory drive, Rev. Physiol. Biochem. Pharmacol., 90 (1981) 172-244. 31 Share, L. and Levy, M. N., Cardiovascular receptors and blood titer of antidiuretic hormone, Amer. J. Physiol., 203 (1962) 425~128. 32 Share, L. and Levy, M. N., Effect of carotid chemoreceptor stimulation on plasma antidiuretic hormone titer, Amer.
228
J. Physiol., 210 (1966) 157-161. 33 Tenney, S. M., The effect of CO 2 on neurohumoral and endocrine mechanisms, Anesthesiology, 21 (1960) 674-685. 34 West, M. J., Blessing, W. W. and Chalmers, J., Arterial baroreceptor reflex function in the conscious rabbit after
brain stem lesions coinciding with A1 group of catecholamine neurons, Circular. Res., 49 (1981) 959-970. 35 Yamashita, H., Effect of baro- and chemoreceptor activation on supraoptic nuclei neurons in the hypothalamus, Brain Research, 126 (1977) 551-556.
221
Direct Pathway From Neurons in the Ventrolateral Medulla Relaying Cardiovascular Afferent Information to the Supraoptic Nucleus in the Cat JOHN CIRIELLO and MONICA M. CAVERSON Department of Physiology, Health Sciences Centre, University of Western Ontario, London, N6A 5C1 (Canada) (Accepted June 7th, 1983) Key words: supraoptic nucleus - - ventrolateral medulla - - aortic depressor nerve - - carotid sinus nerve - ascending cardiovascular pathways
In chloralose anesthetized cats experiments were done to electrophysiologically identify neurons in the ventrolateral medulla (VLM) which relay cardiovascular afferent information directly to the supraoptic nucleus (SON). Action potentials elicited antidromically by electrical stimulation of the SON were recorded from 69 histologically verified single units in the VLM. Single units responded with latencies corresponding to conduction velocities of 7.8 + 0.6 m/s. Of these units 26 were excited orthodromically by stimulation of the buffer nerves; 12 responded to stimulation of only the carotid sinus nerve, 7 responded to stimulation of only the aortic depressor nerve, and 7 responded to both buffer nerves. The axons of VLM units that responded to buffer nerves conducted at a significantly slower velocity than those of non-responsive units (5.7 +_0.4 and 9.1 + 0.8 m/s, respectively). These data provide electrophysiological evidence of two different populations of VLM neurons which project directly to the SON, and suggest that the direct pathway from the VLM to the SON is involved in the release of vasopressin by SON neurons during activation of baroreceptor and chemoreceptor afferent fibers. INTRODUCTION It is well established that activation of baroreceptor and c h e m o r e c e p t o r afferent fibers in the carotid sinus (CSN) and aortic d e p r e s s o r ( A D N ) nerves, and of the nucleus of the solitary tract, the primary site of termination of cardiovascular afferent fibers 12-14, alter both the rate of discharge of neurons in the paraventricular and supraoptic (SON) nuclei 5,6.17,1s,35, and the release of vasopressin 31,32. A l t h o u g h a direct cardiovascular pathway from the nucleus of the solitary tract to the parvocellular division of the paraventricular nucleuslO,U, and possibly a disynaptic pathway through the parabrachial nucleusa6, 21,28, have been d e m o n s t r a t e d , the brainstem pathways which mediate cardiovascular afferent information to either the magnocellular division of the paraventricular nucleus or the S O N are not known. Recent anatomical studies have described a projection from neurons in the nucleus of the solitary tract to the region of the ventrolateral medulla (VLM2 O, which then in turn projects directly to the 0006-8993/84/$03.00 © 1984 Elsevier Science Publishers B.V.
magnoceilular n e u r o s e c r e t o r y portions of the paraventricular nucleus and the S O N 22,29. This evidence suggests that the V L M m a y be a site of origin of an ascending pathway through which afferent information from b a r o r e c e p t o r s and c h e m o r e c e p t o r s reach the SON. This suggestion is further s u p p o r t e d by the demonstration that single units in the region of the V L M alter their rate of discharge during electrical stimulation of the CSN and A D N , and selective activation of b a r o r e c e p t o r s and chemoreceptors7, s. In addition, lesions of the V L M have been shown to attenuate baroreceptor23,24 and chemoreceptor8 reflex responses. F u r t h e r m o r e , destruction of n o r a d r e n e r gic neurons in the V L M results in an elevation of plasma vasopressin 4, possibly due to the removal of a tonic inhibition on vasopressinergic neurons exerted by cardiovascular afferent reflex mechanisms. The present study was done to provide electrophysiological evidence of a m o n o s y n a p t i c p a t h w a y mediating cardiovascular afferent information from the V L M to the SON. The V L M was systematically explored for single units antidromically excited by elec-
222 trical stimulation of the SON. These units were then tested for their responses to stimulation of the CSN and A D N . METHODS Experiments were done in 8 adult cats of either sex anesthetized with a-chloralose (60 mg/kg, i.v., initially, supplemented by additional doses of 30 mg/kg at 8-10 h intervals) after ethyl chloride and ether induction. The trachea was cannulated, the animals were paralyzed with d e c a m e t h o n i u m b r o m i d e (0.5 mg/kg, i.v., initially and additional doses when necessary), and then artificially ventilated. Polyethylene catheters were inserted in the femoral artery and vein for the recording of arterial pressure and the administration of drugs, respectively. The arterial pressure was recorded through a Statham P23Dd transducer, and heart rate was m o n i t o r e d with a 7P44B Grass tachograph triggered by the arterial pressure pulse, both of which were continuously recorded on a Grass 7 polygraph. Rectal t e m p e r a t u r e was mainrained at 37 + 0.2 °C by a heating pad controlled by a Yellow Springs 73 t e m p e r a t u r e controller. The head of the animal was fixed in a K o p f stereotaxic instrument, and access to the VLM was obtained by removing the occipital bone and exposing the floor of the fourth ventricle after removing the cerebellum by suction. Access to the SON was obtained by a unilateral parietal craniotomy. Exposed nervous tissue was covered with warm 360 medical fluid (Dow Corning, Midland, MI) to prevent drying. The left CSN was approached through a lateral incision at the level of the tympanic bulla. The sternocleidomastoid and digastric muscles overlying the sinus region were reflected; a large lymph node and a 2 cm segment of the hypoglossal nerve were removed. The CSN was isolated from surrounding tissue and crushed as far away as possible from its junction with the glossopharyngeal nerve. The left A D N was identified near the junction of the superior laryngeal and vagus nerves through a midline incision in the neck at the level of the larynx, dissected free for a few cm from the cervical vagus, and crushed distally. The central ends of the isolated nerves were placed on bipolar stainless steel electrodes and covered with cotton pellets soaked in warm 36{) medical fluid to prevent drying.
Recording single unit activity in the VLM The region of the VLM was systematically explored; it extended from 1.0 mm caudal to 4.0 mm rostral to the obex, from 3 to 5 mm lateral to the midline and from 2 mm below the dorsal surface to the ventral surface. The reference point for positioning the electrode was the obex and penetrations were made on a grid with points 500/~m apart. Stainless steel microelectrodes 15 with tip diameters of 1-3/~m and a resistance in saline of 0.8-2 Mr2 were used for extracellular recording. The indifferent electrode was a hypodermic needle inserted into the brain tissue through a small hole in the right parietal bone. Unit activity was amplified through a Grass P15 differential preamplifier, and displayed on a Tektronix R5103N oscilloscope for observation and photography. Spontaneously active and silent single units in the VLM encountered during an electrode penetration were tested for their response to electrical stimulation of the ipsilateral SON with 0.1-0.2 ms rectangular pulses at (I.5 Hz and at current intensities of 0.25-1.5 m A delivered through the central electrode of bipolar stainless steel electrodes (SNEX-100, David Kopf, Tujunga, C A ; 0.25 mm tip diameter; 50-90 k ~ initial DC resistance in saline) from a Grass $88 stimulator through a Grass PSIU6B stimulus isolation and constant current unit. The indifferent electrode was an alligator clip attached to exposed scalp muscle. Single units in the V L M that responded to electrical stimulation of the SON were assessed for antidromic activation using previously established criteria (for a review see ref. 20): (a) constant latency of the evoked spike: (b) high following frequency of the evoked spike; (c) occurrence of a single evoked spike at threshold and suprathreshold stimulus intensities, and (d) collision of evoked and spontaneous spikes. Not all single units classified as antidromically activated were evaluated for all criteria, as 80f/~ of the units were not spontaneously active and the activity could not be tested by the collision method. All spontaneously active units showed collision of the evoked and spontaneous spikes (Fig. 1). Units antidromically excited by stimulation of the SON were subsequently tested for responses to stimulation of the CSN and A D N with pulses of 0.3 ms duration at 0.5 Hz and current intensities up to 5 times the current required to elicit a threshold decrease in heart rate when using a 15 s stimulus train at 25 Hz and a
223 pulse duration of 0.3 ms. The ' t h r e s h o l d ' stimulus was defined as the current required to elicit a 10-15 beat/rain decrease in heart rate. Stimulus intensities at 5 times threshold have previously been shown to elicit maximal reflex b r a d y c a r d i a during electrical stimulation of either buffer nerve 9.
"14+ B
Histological localization of recording and stimulation sites Sites of recording and stimulation were identified by depositing iron from the electrode tip (20/~A for 15 s, tip positive). A n i m a l s were perfused with 0.9% saline followed by a 1% potassium ferrocyanide in 10% formalin-saline solution to reveal the m a r k e d sites by the Prussian blue reaction. The brains were fixed in 10% formalin for at least 3 days and transverse 50/~m frozen sections of the h y p o t h a l a m u s and medulla were cut and stained with thionin. Stimulation and recording sites were m a p p e d on transverse sections of the h y p o t h a l a m u s modified from a stereotaxic atlas I and from projection drawings of the medulla. The conduction distance for each unit was calculated from these maps and ranged between 21.5 and 27.8 mm.
Data analysis Statistical analysis of means + S.E. of conduction velocities was done by using a Student's t-test. P < 0.01 was considered to be statistically significant.
A
RESULTS A total of 240 electrode p e n e t r a t i o n s were m a d e through the region of the VLM. Sixty-nine histologically verified single units in the V L M were antidromically activated by electrical stimulation of the SON. Of these units, 14 were discharging spontaneously (mean discharge rate, 3.5 + 1.2 spikes/s) and 55 were silent. Units r e s p o n d e d with a mean latency of 5.0 + 0.4 ms (range, 1.7-17 ms) which was constant for any one unit and followed rates of stimulation of 100-500 Hz. The latencies of the antidromic responses corres p o n d e d to conduction velocities of these ascending fibers of 7.8 + 0.6 m/s (Fig. 2). A s seen in Fig. 2, there a p p e a r to be two distinct populations of units in the V L M projecting to the S O N identified on the basis of their conduction velocities; the first group contains units whose axons conduct in the range of 1-7
t
v
I
Fig. 1. Response of spontaneously active single unit in VLM to stimulation of SON. A: spontaneous discharge rate of VLM unit. Single sweep. Calibrations, 5 s and 200~V. B: antidromic response of VLM unit to stimulation of SON at 0.5 Hz. Arrow indicates time of stimulation; 5 superimposed sweeps. Calibrations, 2 ms and 200 ~V. C: cancellation of antidromic spike by spontaneous activity in VLM unit; 6 successive sweeps; stimulus delivered at arrow. Asterisks mark sweeps in which cancellation occurred. Calibrations, 2 ms and 200 ~V.
m/s (mean, 4.4 + 0.3 m/s; n = 36), and the second group contains units which r e s p o n d with latencies corresponding to conduction velocities of greater
224 than 7 m/s (mean, 11.5 + 0.6 m/s; range, 7.2-27 m/s; n
I0
= 33). The threshold current to evoke these antidromic spikes ranged from 0.25 to 1.5 m A (mean, 0.75 +
n=69
0.04 m A when using a pulse duration of 0.1-0.2 ms). The threshold current was found to be always less than 0.5 m A when the tip of the stimulating electrode was located within the SON and usually greater when
Number of
it was located medial, lateral or just dorsal to the SON. All units responded with a single spike (duration of spike, 0.8-3.0 ms) at threshold and supra-
Units
4
threshold stimulus intensities. All antidromically ac-
2
tivated units showed two-component spikes (initial segment and somatodendritic components) of which, 0
, 0
,
, 2
i
,
4
|
,
,
,
6 8 Conduction
i
,
w
,
I0 12 Velocity
n
i
w
,
14 16 (m/s)
26
2B
Fig. 2. Histogram of conduction velocities of 69 VLM single units antidromically activated by electrical stimulation of SON. Hatched area represents the conduction velocities of antidromically activated units which were also excited by stimulation of the buffer nerves.
A
in some cases, the somatodendritic c o m p o n e n t could be induced to fail by high frequency of stimulation (Fig. 3C). Characteristic antidromic potentials recorded from neurons in the VLM during stimulation of the SON are shown in Figs. 1 and 3. Of the 69 antidromically identified units, 26 responded orthodromically to stimulation of the buffer
B
t
t/
/
tt
t
Fig. 3. Response of an antidromically activated unit in VLM to different frequencies of stimulation of SON (A-C) and to stimulation of the CSN (D). Each record is 5 superimposed sweeps, and the stimuli were delivered at arrows. A: note the constant latency and 2 component (IS-SD) spike in record at 0.5 Hz. B: action potentials evoked with two stimuli applied at 116 Hz. C: note the separation of the IS-SD components (*) and the occasional failure of the SD component of the spike after the second stimulus at 500 Hz. D: orthodromic response of same unit to stimulation of the CSN. Calibrations, 2 ms and 100/~V(A-C), and 5 ms and 100~V (D).
225
Obex
+5.0
|
I
2mm Fig. 4. Location of units antidromically excited by stimulation of the SON and orthodromicallyexcited by stimulation of the buffer nerves plotted on representative transverse sections of the VLM of the cat (obex to 4 mm rostral to obex). O, units excited only by CSN stimulation; A, units excited only by ADN stimulation; *, units excited by stimulation of both the CSN and ADN; O, units not responding to stimulation of buffer nerves. 5SP, spinal trigeminal nucleus; 5ST, spinal trigeminal tract; 12N, hypoglossal nerve; ION, inferior olivary nucleus; LRN, lateral reticular nucleus; P, pyramidal tract; PPR, postpyramidal nucleus of the raphe.
response. Fig. 3D shows a characteristic response of an antidromically identified unit orthodromically excited by stimulation of the CSN. The remaining 43 units did not respond to stimulation of either buffer nerve. Single units orthodromically excited by stimulation of the buffer nerves responded antidromically to stimulation of the SON with latencies corresponding to a mean conduction velocity of 5.7 + 0.4 m/s (range, 1.7-9.5 m/s), whereas the mean conduction velocity of the units non-responsive to the buffer nerves was 9.1 + 0.8 m/s (range, 1.4-27 m/s; Fig. 2). Although the ranges of conduction velocities of the responsive and non-responsive units overlap, the mean conduction velocities for the two groups were significantly different (P < 0.005). The anatomical distribution of units activated antidromically by stimulation of the SON and orthodromically excited by stimulation of the buffer nerves is shown in Fig. 4. These units were located primarily in a region ventromedial to the lateral reticular nucleus and lateral to the intramedullary rootlets of the hypoglossal nerve. This group of neurons extended into the region just dorsal to the lateral reticular nucleus. On the other hand, the non-responsive antidromically activated units occupied the region immediately dorsolateral and ventrolateral to that containing units responsive to the buffer nerves. These latter units were scattered mostly near the periphery of the lateral reticular nucleus, as defined cytoarchitectonically, and in the reticular formation in a region ventrolateral and dorsolateral to this nucleus. The majority of units (60/69) were found in the caudal ventrolateral medulla extending from 0.5 mm caudal to 2.0 mm rostral to the obex. DISCUSSION
nerves; 12 were excited by stimulation of only the CSN with a mean latency of 11.7 + 1.9 ms, 7 were excited by stimulation of only the A D N with a mean latency of 9.9 + 1.1 ms, and 7 were excited by both the CSN and A D N (mean latencies, 8.9 + 0.8 and 11.0 + 1.1 ms, respectively). The evoked orthodromic response of these units to stimulation of the buffer nerves consisted of either a single spike or a burst of 2-3 spikes. Increasing the intensity of the stimulus usually decreased the latency of the response and caused an increase in the number of spikes during a
These data provide the first electrophysiological evidence of a monosynaptic pathway from neurons in the VLM mediating cardiovascular afferent information to the SON. These findings are supported by the recent anatomical demonstration in the rat and rabbit of direct connections between neurons in the VLM and the SON using both the autoradiographic technique 3.22.29 and the retrograde transport of horseradish peroxidase and fluorescent dyes 3.29. Neurons in the VLM have been shown to innervate
226 that portion of the SON which contains primarily vasopressinergic cell bodies 29, and it has been suggested that these VLM neurons are noradrenergic 3.22,29. The anatomical distribution of units activated antidromically by stimulation of the SON and orthodromically by stimulation of the buffer nerves corresponds to that of catecholamine-containing neurons in the VLM of the cat 2.19,25 and to that described for DBH-stained neurons in the A1 cell group of the VLM in the rat 29. This suggests the possibility that some of the neurons relaying cardiovascular afferent information to the SON were noradrenergic. This suggestion is supported by the observation that destruction of the A I noradrenergic neurons in the VLM results in an increase in the release of vasopressin due to the removal of an inhibitory input into magnocellular neurosecretory neurons 4. Since baroreceptor reflex mechanisms are known to maintain a tonic inhibition on vasopressinergic neurons 29,3°, this further suggests that some of the neurons recorded in this study in the region of the VLM are likely to be noradrenergic and involved in the release of vasopressin. However, the possibility cannot be excluded that neurons which do not receive buffer nerve information are also catecholaminergic nor that non-catecholaminergic neurons in the VLM also project to magnocellular neurosecretory SON neurons. An unexpected finding was that axons of neurons in the VLM relaying cardiovascular afferent information conducted at significantly slower velocities than axons of neurons which were non-responsive to buffer nerve stimulation. The conduction velocities of these ascending cardiovascular fibers is in agreement with an earlier estimate6. These data suggest that there are two different populations of axons that project directly to the SON originating from functionally different neurons in the VLM. The finding that neurons in the VLM projecting to the SON responded to stimulation of the CSN and A D N is in agreement with an earlier study showing that single units in the SON alter their rate of discharge during stimulation of the buffer nerves 6. Furthermore, the finding that the majority of VLM neurons responding to buffer nerve stimulation were excited by stimulation of the CSN is in agreement with the previous demonstration that most of the units responding to stimulation of buffer nerves in the SON were excited by stimulation of the CSN 6. In addition,
the observation that some VLM units respond to both the CSN and A D N suggests that these neurons process functionally equivalent afferent information originating in either baroreceptors and chemoreceptors from the two major sites of cardiovascular receptors. The finding of neurons in the VLM orthodromically activated by stimulation of the A D N and CSN was not unexpected as buffer nerve responses have previously been recorded in this area7, s. The likely pathway by which buffer nerve information is relayed to the VLM is from the nucleus of the solitary tract to the VLM, as it has been demonstrated that fibers from the nucleus of the solitary tract project directly to the region of the VLM 21.2a,26. This suggestion is supported by the finding that electrical stimulation of the nucleus of the solitary tract alters the activity of neurons in the ventrolateral medulla which project directly to the SON (Ciriello and Caverson, unpublished observations, 1983). In view of the previous demonstration that selective activation of baroreceptots and chemoreceptors alter the discharge rate of vasopressinergic neurons in the SON 18,35 and that lesions of the VLM result in an increase in the plasma level of vasopressin 4, this direct pathway to the SON from VLM neurons is suggested to be involved in the regulation of vasopressin release by magnocellular neurosecretory SON neurons. It is not surprising to find that some of the neurons in the VLM antidromically activated by stimulation of the SON were not responsive to the buffer nerves, as this area has been shown also to relay visceral sensory information unrelated to the cardiovascular system (for review see ref. 29). In addition, some of these units were found close to the ventral surface of the brainstem in an area known to be sensitive to [H +] and p C O : (for review see ref. 30). Activation of central CO 2 chemoreceptors has been shown to alter the metabolic activity of the V L M and SON 27 and influence the release of vasopressin 33, suggesting that some of these neurons are involved in mediating central chemoreceptor information to the SON. In summary, these data have provided electrophysiological evidence of a direct cardiovascular pathway from neurons in the VLM to the SON, and suggest that this pathway is involved in the regulation of vasopressin release from magnoceilular neurosecretory neurons in the SON during activation of cardiovascular afferent fibers.
227 S c h o l a r . T h e a u t h o r s w i s h t o t h a n k D r s . G . J. M o -
ACKNOWLEDGEMENTS
g e n s o n a n d D. L. J o n e s f o r t h e i r c o m m e n t s o f t h e This work was supported by the Ontario Heart F o u n d a t i o n . J . C . is a C a n a d i a n
Heart Foundation
REFERENCES 1 Bleier, R., The Hypothalamus of the Cat, Johns Hopkins Press, Baltimore, MD, 1961. 2 Blessing, W. W., Frost, P. and Furness, J. B., Catecholamine cell groups of the cat medulla oblongata, Brain Research, 192 (1980) 69-75. 3 Blessing, W. W., Jaeger, C. B., Ruggiero, D. A. and Reis, D. J., Hypothalamic projections of medullary catecholamine neurons in the rabbit: a combined catecholamine fluorescence and HRP transport study, Brain Res. Bull., 9 (1982) 279-286. 4 Blessing, W. W., Sved, A. F. and Reis, D. J., Destruction of noradrenergic neurons in rabbit brainstem elevates plasma vasopressin, causing hypertension, Science, 217 (1982) 661-663. 5 Calaresu, F. R. and Ciriello, J., Electrophysiology of the hypothalamus in relation to central regulation of the cardiovascular system. In P. Meyer and H. Schmitt (Eds.), Nervous System and Hypertension, John Wiley, New York, 1979, pp. 129-136. 6 Calaresu, F. R. and Ciriello, J., Projections to the hypothalamus from buffer nerves and n~.cleus tractus solitarius in the cat, Amer. J. Physiol., 239 (1980) R130-R136. 7 Calaresu, F. R. and Ciriello, J., Projections of buffer nerves to medulla and hypothalamus. In P. S. Sleight (Ed.), Baroreceptors and Hypertension, Oxford University Press, Oxford. 1980, pp. 252-260. 8 Ciriello, J. and Calaresu, F. R., Lateral reticular nucleus: a site of somatic and cardiovascular integration in the cat, Amer. J. Physiol., 233 (1977) R100-R109. 9 Ciriello J. and Calaresu, F. R., Separate medullary pathways mediating reflex vagal bradycardia to stimulation of buffer nerves in the cat, J. Auton. Nerv. Syst., 1 (1979) 13-32. 10 Ciriello J. and Calaresu, F. R., Monosynaptic pathway from cardiovascular neurons in the nucleus tractus solitarii to the paraventricular nucleus in the cat, Brain Research, 193 (1980) 529-533. 11 Ciriello J. and Calaresu, F. R., Autoradiographic study of ascending projections from cardiovascular sites in the nucleus tractus solitarii in the cat, Brain Research, 186 (1980) 448-453. 12 Ciriello J. and Calaresu, F. R., Projections from buffer nerves to the nucleus of the solitary tract: an anatomical and electrophysiological study in the cat, J. Auton. Nerv. Syst., 3 (1981) 299-310. 13 Ciriello, J., Hrycyshyn, A. W. and Calaresu, F. R., Horseradish peroxidase study of brain stem projections of carotid sinus and aortic depressor nerves in the cat, J. Auton. Nerv. Syst., 4 (1981) 43-61. 14 Crill, N. E. and Reis, D. J., Distribution of carotid sinus and depressor nerves in cat brain stem, Amer. J. Physiol., 214 (1968) 269-276. 15 Green, J. D., A simple microelectrode for recording from central nervous system, Nature (Lond.), 182 (1958) 962.
m a n u s c r i p t a n d Miss G . M c C r a e f o r t e c h n i c a l assistance.
16 Hamilton, R. B., Ellenberger, H., Liskowsky, D. and Schneiderman, N., Parabrachial area as mediator of bradycardia in rabbits, J. Auton. Nerv. Syst., 4 (1981) 261-281. 17 Kannan, H. and Koizumi, K., Pathways between the nucleus tractus solitarius and neurosecretory neurons of the supraoptic nucleus: electrophysiological studies, Brain Research, 213 (1981) 17-28. 18 Kannan, H. and Yagi, K., Supraoptic neurosecretory neurons: evidence for the existence of converging inputs both from carotid baroreceptors and osmoreceptors, Brain Research, 145 (1978) 385-390. 19 Lackner, K. J., Mapping of monoamine neurones and fibres in the cat lower brainstem and spinal cord, Anat. Embryol., 161 (1980) 169-195. 20 Lipski, J., Antidromic activation of neurones as an analytic tool in the study of the central nervous system, J. Neurosci. Meth., 4 (1981) 1-32. 21 Loewy, A. D. and Burton, H., Nuclei of the solitary tract: efferent projections to the lower brain stem and spinal cord of the cat, J. comp. Neurol., 181 (1978)421--450. 22 Loewy, A. D., Wallach, J. H. and McKellar, S., Efferent connections of the ventral medulla oblongata in the rat, Brain Res. Rev., 3 (1981) 63-80. 23 McAIlen, R. M., Neil, J. J. and Loewy, A. D., Effects of kainic acid applied to the ventral surface of the medulla oblongata on vasomotor tone, the baroreceptor reflex and hypothalamic autonomic responses, Brain Research, 238 (1982) 65-76. 24 Norgren, R., Projections from the nucleus of the solitary tract in the rat, Neuroscience, 3 (1978) 207-218. 25 Poitras, D. and Parent, A., Atlas of the distribution of monoamine-containing nerve cell bodies in the brain stem of the cat, J. comp. Neurol., 179 (1978) 699-718. 26 Ricardo, J. A. and Koh, E. J., Anatomical evidence of direct projections from the nucleus of the solitary tract to the hypothalamus, amygdala, and other forebrain structures in the rat, Brain Research, 153 (1978) 1-26. 27 Rohlicek, C. V., Ciriello, J. and Polosa, C., Deoxyglucose mapping of central nervous system structures during central chemoreceptor mediated hypercapnic responses in the rat, Soc. Neurosci. Abstr., 7 (1981) 116. 28 Saper, C. B. and Loewy, A. D., Efferent connections of the parabrachial nucleus in the rat, Brain Research, 197 (1980) 291-317. 29 Sawchenko, P. E. and Swanson, L. W., The organization of noradrenergic pathways from the brainstem to the paraventricular and supraoptic nuclei in the rat, Brain Res. Rev., 4 (1982) 275-325. 30 Schlaefke, M. E., Central chemosensitivity: a respiratory drive, Rev. Physiol. Biochem. Pharmacol., 90 (1981) 172-244. 31 Share, L. and Levy, M. N., Cardiovascular receptors and blood titer of antidiuretic hormone, Amer. J. Physiol., 203 (1962) 425~128. 32 Share, L. and Levy, M. N., Effect of carotid chemoreceptor stimulation on plasma antidiuretic hormone titer, Amer.
228
J. Physiol., 210 (1966) 157-161. 33 Tenney, S. M., The effect of CO 2 on neurohumoral and endocrine mechanisms, Anesthesiology, 21 (1960) 674-685. 34 West, M. J., Blessing, W. W. and Chalmers, J., Arterial baroreceptor reflex function in the conscious rabbit after
brain stem lesions coinciding with A1 group of catecholamine neurons, Circular. Res., 49 (1981) 959-970. 35 Yamashita, H., Effect of baro- and chemoreceptor activation on supraoptic nuclei neurons in the hypothalamus, Brain Research, 126 (1977) 551-556.