Chorda tympani and vagus nerve convergence onto caudal brain stem neurons in the rat

Brr& Resrurch Butfprir,,Vol. 7, pp. 261-266, 1981.Printed in the U.S.A.

Chorda Tympani and Vagus Nerve Convergency onto Caudal Brain Stem Neurons in the Rat D. A. BEREITER,’ Lahoratoires

H.-R. BERTHOUD

AND B. JEANRENAUD

de Recherches MPtaholiques et Department de Medecine UniversitP de GenPve, 1205 Geneva, Swaitzerland

Received

2 June 1981

BEREITER, D. A., H.-R. BERTHOUD AND B. JEANRENAUD. Chords tympcrni crnd vagus ncwe wnwrpvzce onfo c~rrudcclhmin .sicm neuro~~.sin the rut. BRAIN RES. BULL. 7(3) 261-266, 1981.-Neurons responsive to chorda tympani (CT) and cervical vagus (CV) nerve stimulation were identified electrophysioiogicaily in the caudal brain stem of the anesthetized rat. Ail identified dualiv responsive (CT/W) neurons included in this study were orthodromicailv activated by both CT and CV stimulation (n=80j. These cells were iocated mainly in the lateral nucleus tractus solitariusW(nts) or more ventrally in the region of nucleus ambiguus famb). No CTKV cells were found to lie clearly within the dorsal motor nucleus of the vagus nerve. Among CTiCV neurons in nts, there was no correlation of response latency to CT stimulation versus response latency to CV stimulation, while CTKV neurons found in the region of amb demonstrated a good correlation between the two stimuli and suggested that sensory convergence occurred prior to the level of amb. Conditioning pulses applied to CV were able to alter the neural response characteristics to CT test pulses in a majority of the units tested. It is concluded that oropharyngeai afferents converge on brain stem neurons that are also responsive to vagal afferent input. These dually responsive CTiCV neurons are implicated in the integration of sensory information relevant for cephalic phase reflexes. Vagus nerve

Chorda tympani

Caudai brain stem neurons

OROPHARYNGEAL sensory input evokes a variety of vagal-dependent physiological 1201 and hormonal 14, 9, 2 I, 271 responses, i.e., the so-called cephalic phase responses of food ingestion. The underlying neural substrate for these responses is not well known, but behavioral [12,13] and physiological [ 1S] studies with decerebrate preparations suggest that a certain amount of reflex activity relevant for food ingestion may be integrated directly at the brain stem level. Although a considerable number of studies have focused separately on the anatomical details of gustatory or oropharyngeal brain stem input [5,15], vagal afferent input [ 1, 8, 14, 16. 231 or the origin of vagal efferents innervating the abdominal viscera [7, 19, ZS], the relationship between oropharyngeal input and the vagal system has not been specifically investigated. The purpose of this study was to explore the caudal brain stem of the rat using extracellular recording techniques to examine the location and response characteristics of single neurons sensitive to chorda tympani and vagal stimulation. METHOD Gene4

and Surgicut

Procedures

A totai of 41 male rats (Wistar-derived) weighing 30~400 g were used in these experiments. Animals were housed in ‘Current address: Rhode Island Hospital, Endocrinology

Copyright

@ 1981 ANKHO

Laboratory,

International

groups of 6-10, given free access to food and water, and maintained on a 12 hr iighting schedule prior to the day of the experiment. All animals were initially anesthetized with thiopentothal sodium (60 mg/kg, IP) during the surgical procedures, but maintained on an cu-chloralose/urethane (30/300 mg/kg, IV) mixture for the duration of the recording experiment, lasting up to 12 hr. Tracheostomy was routinely performed to assist respiration and body temperature was maintained at 37°C with a thermal probe-heating blanket system. The right cervical vagal (0 trunk was exposed distal to the entrance of the superior laryngeal nerve for electrica stimulation. The stimulating electrode consisted of a short length of silastic tubing slit longitudinally and traversed cross-wise by two AgCl wires separated by about 0.5 mm. The tubing could be opened with a pair of forceps and slipped under the nerve then allowed to close itself around the entire vagal trunk. The entire nerve-electrode assembly was bathed in warm mineral oil to reduce the possibility of nonspecific current spread to adjacent structures. The right chorda tympani (CT) nerve was exposed rostra1 to the bulla tympani and fitted with a bipolar stainless steel electrode as previously described 121. The animal was then placed in a stereotaxic frame (KopR using non-traumatic earbars in a “nose-down” position of approximately 20” off Providence,

Inc.-

RI 02902.

0361-9230/81/090261-~$Ol.lO/O

BEREITER,

262

BERTHOUD

AND JEANRENAUD

horizontal. The dorsal surface of the brain stem was exposed from the obex level to approximately 2.5 mm rostra1 to the obex, ipsilateral to the CV and CT stimulating electrodes as previously described [2]. Recording

Techniques

Unit activity was recorded with glass microelectrodes (resistance 2-10 Ma) filled with saturated fast green dye in 2 M NaCI, amplified and displayed by conventional techniques, and stored on magnetic tape for later analyses. The brain stem area explored extended from +0.5 mm to ~2.4 mm rostral to the obex, medially from 0.5 to 2.0 mm from the midline and ventrally from the dorsal brain stem surface to 3.0 mm below the surface. All electrode punctures were ipsilateral (right side of the brain stem) to the CT and CV stimulating electrodes. Stimulation

1.7

Techniques

The stimulation parameters consisted of biphasic pulses Xl-908 PA, 0.2 msec duration, delivered as single shocks at rates of
RESULTS

Unit Locations

in the Brain Stem

Electrical stimulation of the intact chorda tympani (CT) nerve was used to identify individual neurons recorded ipsilaterally in the caudal brain stem of the male rat. A total of 80 CT-responsive neurons were identified that also responded to electrical stimulation of the intact cervical vagus

1.0

FIG. 1. Brain stem location of dually responsive (CTKV) neurons. Abbreviations: amb=nucleus ambiguus, io=inferior olivary nucleus, nrv=nucleus reticularis pars ventralis, nts=nucleus tractus solitarius, ntV=nucleus tractus spinalis trigemini, nX=nucleus dorhypoglossi, P=corticospinal tract, salis vagi, nXII=nucleus rgi=nucleus reticularis gigantocellularis, rl=nucleus reticularis lateralis, X=vagus nerve. The numbers adjacent to each outline refer to the approximate distance rostra1 to the obex level.

(CV) nerve. The experimental paradigm was to explore the caudal brain stem rostral to the obex while stimulating CT. Stimulation of CV was appiied only after having identifmd a neuron as being CT responsive, thus an additional population of CV-only sensitive neurons was not specifically investigated. Dually responsive CT/CV neurons were particularly sought in this study and many additional CT-only neurons were encountered but are not discussed here. The brain stem locations of the 80 CT/CV neurons are shown in Fig. 1. The majority of these units were recorded at AP levels of + 1.5 to +2.4 mm rostral to the obex and were found to lie either in the extreme ventrolateral edge of nucleus tractus solitarius (nts) or more ventrally in the region of nucleus ambiguus (amb). Although the anatomical separation between CT/CV units lying in the nts region versus the amb region was not complete, all but 9 of these units were assigned to the nts or amb group for additional statistical analyses.

CHORDA TYMPANI

263

AND VAGAL CONVERGENCE

I

10

20

30

40

50

60

70

80

CV Latency (msec ) FIG. 2. Scatter-plot distribution of the minimum latency of dually responsive neurons tympani (ordinate) and cervical vagus (abscissa) stimulation. Only neurons located in the amb (V) region are included. The calculated linear regression lines for nts neurons (solid and for amb neurons (dashed line, n=26) differ significantly from each other (p~O.01, variance) where p=the linear regression coefficient for each neuronal group.

The location of 71 additional neurons responsive to only CT stimulation overlapped that of the CT/CV dually responsive neurons, with an apparent tendency to be more widely distributed; CT-only neurons were also found in medial nts and throughout the medial reticular formation, i.e., nucleus reticularis gigantocellularis (data not shown). Response

to chorda nts (0) and line, n=45) analysis of

Also shown in Fig. 2 are the regression lines calculated from the individual neuronal minimum latencies to CT and CV stimulation for units located in the nts and amb regions. Statistical analyses of these regression lines revealed no correlation between CT and CV latency for nts neurons, but amb neurons showed a good correlation between CT and CV latency @<0.02).

Characteristics

Among the 80 CTICV neurons, 47 cells were spontaneously active with low rates of <2/set. Stimulation of CT or CV alone evoked an excitatory response in each of these 80 neurons. Several additional spontaneously active units were inhibited by CT stimulation alone (n=4) or by both CT or CV stimulation (n=l). All CTKV units were considered to be orthodromically driven by both stimuli as the latency to repeated stimulus presentations was not stable nor were these neurons able to follow stimulation frequencies in excess of 25/set. In particular, the response to CV stimulation was more labile than that seen following CT stimulation with response latencies being more variable and frequency following rarely exceeded lO/sec. The minimum latency to the first spike following CT and CV stimulation is shown in Fig. 2 as a scatter plot for CTKV neurons falling in the nts region (n=45) and those located in the amb region (n=26). The mean latency (*SD) for the CT response was 9.6Ok6.01 msec and 14.81 -t-15.41 msec for the nts and amb groups, respectively. The mean minimum latency (-cSD) to CV stimulation was 15.33~12.42 msec for nts neurons and 41.35k26.52 msec for amb neurons. The response latencies among nts and amb neurons responsive only to CT stimulation were not significantly different than those seen for CT/CV dually responsive cells.

Conditioning

Experiments

Forty-seven of the 80 CT/CV neurons were tested for conditioning effects of CV stimulation on the excitatory CT response by presenting the CV stimulus as a single pulse at varying time intervals before the CT test pulse. Of the 47 units tested, the majority (n=25) showed a decreased CT response following the CV conditioning pulse (see Method). An additional 9 units demonstrated a CV-induced facilitation of the CT response either by an increased number of spikes for a given number of stimulus trials or by a reduced latency to CT stimulation. Thirteen of the 47 units tested showed no apparent effect of CV conditioning pulses on the excitatory CT-evoked response. An example of CV-induced facilitation of the excitatory CT response is shown for an nts neuron in Fig. 3. In this example, stimulation of CT alone evoked a burst of 2-3 spikes with a mean latency (10.3k2.31 msec, Fig. 3, top panels) that did not differ from that seen when CT and CV stimulation was presented simultaneously (10.19k3.77 msec, Fig. 3, bottom left panel). However, when CV stimulation preceded CT stimulation by 30 msec, (Fig. 3 right, middle panel) there was both an increase in the number of spikes per 20 trials and a decrease in the mean latency (7.3622.59 msec) that was significantly different from that following CT stimu-

BEREITER,

264

CT r!:hC

30

20

r .&

10

m.

BERTHOUD

AND JEANRENAlJD

CT

LL

CV**

n:fio

cv-CT** 30msec

n.52

n:74

I

CV-CT 0 msec n=72

Response

CV-CT

Latency

(msec)

FIG. 3. Frequency histograms of response latencies for a neuron showing facilitation of the excitatory CT response following CV conditioning. Left: distribution of spike latenties following CT (top), CV (middle) and simultaneous CTKV (bottom) stimulation. Right: distribution of CT-evoked spike latencies following CT (top), CT preceded 30 msec by CV (middle) and CT preceded 50 msec by CV (bottom) stimulation. Each histogram derives from 20 stimulus trials. The number of spikes per 20 trials is in the upper right comer of each histogram. See text for additional description. test, versus CT stimulation alone.

lation alone or from that following simultaneous CTKV stimulation (p0.05). The time course for maximum facilitation was 30-50 msec for the 9 CT/CV neurons demonstrating CV-induced facilitation of the CT response, however, 1 neuron was facilitated when CV stimulation preceded CT stimulation by 300 msec as shown in Fig. 4. Among the CTICV neurons showing a reduced CT response following a CV conditioning pulse (n=25) maximum inhibition was generally seen at SO-100 msec. Figure 4 demonstrates the time course of the response of an amb interneuron to varying conditioning-test pulse intervals based on the number of spikes per 35 stimulus trials. The number of spikes following CT stimulation alone was taken as 100 percent. Maximum inhibition of the CT response was seen at a CV-CT interpulse interval of 70 msec. However, at an interpulse interval of 300 msec there was additionally a pronounced facilitation of the CT response.

**p
DISCUSSION

These results demonstrate the occurrence and distribution of intemeurons in the caudal brain stem of the rat that respond to both chorda tympani (CT) and cervical vagus (CV) afferent neural input. The majority of these dually responsive CTKV neurons were located in the parasolitarius area-lateral nucleus tractus solitarius (nts) or in the region of nucleus ambiguus (amb). Previous electrophysiological studies have shown that nts 1251and amb [22] neurons receive convergent sensory information from several peripheral sources, however, chorda tympani nerve stimulation was not included. Although taste stimuli were not applied to the tongue in the present experiments, presumably some of these CT-evoked responses represent gustatory input, based on the response latencies to CT stimulation as we have previously reported [2]. There is general agreement between the present data and that of others that the lateral nts region receives gustatory and oropharyngeal sensory input [5,15] as well as input from cervical vagal [8,23] and abdominal vagal afferents 114,161. Others have also shown that the amb region receives CT [5]

CHORDA TYMPANI

I,.. 020406080

265

AND VAGAL CONVERGENCE

loo

Conditioning -Test

200

300

350

Interval (msec)

FIG. 4. Effect of varying conditioning-test pulse intervals on the excitatory response to chorda tympani nerve stimulation. The response to chorda tympani stimulation alone was considered as the 100 percent response, based on the number of spikes per 35 stimulus trials for the first 30 msec following the CT test pulse.

and vagal 122,231 orthodromic input as we report here, but no previous description of CT-CV convergence onto amb (or nts) neurons has been made. The underlying neural substrate for gustatory-evoked cephalic phase reflexes is not known, but must involve neurons which contribute to the vagal efferent system. Dually responsive CT/CV nts and amb cells may be particularly involved in the integration of sensory information relevant to

cephalic phase reflexes as well as for other more general brain stem ingestive reflexes [22]. Cell bodies in the amb region have been shown to contribute vagal efferent fibers to the subdiaphragmatic vagus [7] and to the direct vagal innervation of the pancreas [ 19,281. Also, direct stimulation of the amb region led to a prompt rise in plasma insulin levels comparable in time-course and magnitude to that seen following oral stimulation [3]. The relatively long latencies observed among CT/CV neurons to either stimulus suggested a multisynaptic pathway that could involve higher brain centers. Hypothalamic stimulation can modulate nts neuronal responses to CT stimulation [2]. The significant correlation among CT/CV response latencies among amb neurons, but not among nts cells, suggested that the sensory convergence from these two peripheral nerve sources occurred prior to the level of amb. Thus, regardless of the pathways involved, information reaching the amb region may represent a portion of a final common output for integrated CTlCV afferent sensory information. The majority (34/47) of CTICV neurons tested showed a conditioning effect of CV stimulation on the CT-evoked neuronal response. Such modulatory effects of vagal afferents on CT-evoked responses may account for the changes in gustatory-evoked neural activity seen during food ingestion [6], increased blood glucose levels [lo] or during stomach distension [ 11,171, all of which presumably involve vagal afferent nerves. The physiological significance of these CT/CV neurons is not known, however in conjunction with previous anatomical [7, 19, 281 and physiological [3] studies, these results implicate caudal brain stem CT/CV dually responsive neurons in the integration of afferent sensory information relevant for cephalic phase reflexes. ACKNOWLEDGEMENTS

The authors wish to thank Mrs. Maria Becker (Geneva, Switzerland) for assistance with the histology and Mrs. Hannah Brewer (Providence, RI) for preparation of the manuscript. This study was supported by grant No. 3.637.80 SR of the Swiss National Science Foundation (Berne, Switzerland), by grant No. I ROI AM 25220-01 of the National Institutes of Health (Bethesda, MD), and by a grantin-aid of Nestle S.A. (Vevey, Switzerland).

REFERENCES I. Beckstead, R. M. and R. Norgren. An autoradiographic

2. 3. 4.

5.

6.

examination of the central distribution of the trigeminal, facial, glossopharyngeal and vagus nerves in the monkey. J. camp. Neural. 184: 455-472, 1979. Bereiter, D., H. R. Berthoud and B. Jeanrenaud. Hypothalamic input to brain stem neurons responsive to oropharyngeal stimulation. Expl Brain Rrs. 39: 3x39, 1980. Bereiter, D. A., M. Brunsmann, H. R. Berthoud and B. Jeanrenaud. Nucleus ambiguus stimulation increases plasma insulin levels in the rat. Am. J. Physiol., in press, 1981. Berthoud, H. R., E. R. Trimble, E. G. Siegel, D. A. Bereiter and B. Jeanrenaud. Cephalic-phase insulin secretion in normal and pancreatic islet-transplanted rats. Am. J. Physiol. 238: E336E340. 1980. Blomquist, A. J. and A. Antem. Localization of the terminals of the tongue afferents in the nucleus of the solitary tract. J. camp. Neural. 124: 127-130, 1965. Burton, M. J., E. T. Rolls and F. Mora. Effects of hunger on the responses of neurons in the lateral hypothalamus to the sight and taste of food. Expl Neural. 51: 668-677, 1976.

7. Coil, J. D. and R. Norgren. Cells of origin of motor axons in the subdiaphragmatic vagus of the rat. J. Aufonomic~ New. Syst. 1: 203-210, 1979. 8. Cottle, M. K. Degeneration studies of primary afferents of IXth and Xth cranial nerves in the cat. J. camp. Nrurol. 122: 329343, 1964. 9. Dockray, Cl. J. and H. J. Tracy. Atropine does not abolish cephalic vagal stimulation of gastrin release in dogs. J. Physiol. 306: 473-480, 1980. 10. Giza, B. K. and T. R. Scott. Blood sugar levels affect taste responses in the rat nucleus tractus solitarius. SW. Nrurosci. Absfr. 6: 244, 1980. 1I. Glenn, J. F. and R. P. Erickson. Gastric modulation of gustatory afferent activity. Physid. Bahnv. 16: 561-568, 1976. 12. Grill, H. J. and R. R. Miselis. Lack of ingestive compensation to osmotic stimuli in chronic decerebrate rats. Am. J. Ph~siol. 240: R81-R86, 1981. 13. Grill, H. J. and R. Norgren. The taste reactivity test. II. Mimetic responses to gustatory stimuli in chronic thalamic and chronic decerebrate rats. Brain Rrs. 143: 281-298. 1978.

BEREITER,

266

14. Gwyn, D. G., R. A. Leslie and D. A. Hopkins. Gastric tierents to the nucleus of the solitary tract in the cat. Neurosci. Left. 14: 13-17, 1979. 15. Halpern, B. P. and L. M. Nelson. Bulbargustatory responses to anterior and to posterior tongue stimulation in the rat. Am. J. Physiol. 209: 105-110, 1%5. 16. Harding, R. and B. F. Leek. Central projections of gastric afferent vagal inputs. J. Physiol. 228: 73-90, 1973. 17. Hellekant, G. The effect of stomach distension on the efferent activity in the chorda tympani nerve of the rat. Acta physiol. stand. 83: 527-53 1, 1971. 18. Kawamura, Y. and T. Yamamoto. Studies on neural mechanisms of the gustatory-salivary reflex in rabbits. J. Physiol. 285: 35-48, 1978. 19. Laughton, W. and T. L. Powley. Four central nervous system sites project to the pancreas. Sot. Neurosci. Abstr. 5: 46, 1979. 20. Lin, T. M. and R. S. Alphin. Cephalic phase of gastric secretion in the rat. Am. J. Physiol. 192: 23-26, 1958. 21. Louis-Sylvestre, J. Preabsorptive insulin release and hypoglycemia in rats. Am. J. Physiol. Uo: 5660, 1976. 22. Lucier, G. E., J. Daynes and B. J. Sessle. Laryngeal reflex regulation: peripheral and central neural analyses. Expl Neural. 62: 20&213.

1978.

BERTHOUD

AND JEANRENAUD

23. No&a, S., T. Kamaike and K. Yasunaga. Central vagal organization in rats: an electrophysiological study. Expl Ncrrrol. 60: 405-419, 1978. 24. Palkovits, M. and D. M. Jacobowitz. Topographic atlas of catecholamine and acetylcholinesterase-containing neurons in the rat brain. II. Hindbrain (mesencephalon, rhombencephalon). .I. camp. Neural. 157: 29-42, 1974. 25. Sessle, B. J. Excitatory and inhibitory inputs to single neurones

in the solitary tract nucleus and adjacent reticular Formation. Brain Res. 53: 319-331, 1973. 26. Siegel, S. Nonparametric Statistics for the Behuviorul Sknces.

New York: MiGraw-Hill, 1956, pp. 127-136. 27. Tavlor. I. L.. M. Imnicciatore. D. C. Carter and J. H. Walsh. Effect of atropine anb vagotomy on pancreatic polypeptide response to a meal in dogs. Am. J. Physiol. 235: E443-E447. 1978. 28. Weaver, FR. C. Localization of parasympathetic preganglionic cells bodies innervating the pancreas within the vagal nucleus and nucleus ambiguus of the rat brain stem: Evidence of dual innervation based on the retrograde axonal transport of horseradish peroxidase. J. Autonomic Neril. S~~st. 2: 61-70, 1980. 29. Winer, B. J. Statistical Principles in Experimental L)esiw. 2nd Ed. New York: McGraw-Hill. 1971, pp. 309-430.