The effects of stimulation of trigeminal sensory afferents upon caudate units in cats

Brain Resecrrch Bulletin,

Vol. 4, pp. sC14.

Printed in the U.S.A.

The Effects of Stimulation of Trigeminal Sensory Afferents upon Caudate Units in Cats T. I. LIDSKY,

T. LABUSZEWSKI,

Department

M. J. AVITABLE

AND J. H. ROBINSON

of Psychology, State University of New York Stony Brook, New York I I794 (Received 7 September

1978)

LIDSKY, T. I., T. LABUSZEWSKI, M. J. AVITABLE AND J. H. ROBINSON. The effecrs ofstimulation offrigeminal sen.sor~ ufferenrs upon cuudute units in cafs. BRAIN RES. BULL. 4(l) 9-14, 1979.-This investigation assessed the influences of trigeminal primary sensory afferents upon caudate neuronal activity in locally anesthetized and chloralose anesthetized cats. Afferents from jaw elevator stretch receptors were stimulated via electrodes in the trigerninal mesencephalic nucleus (Mes 5). Afferents from dental and periodontal receptors were stimulated via electrodes in the inferior dental nerve (IDN). Low intensity electrical stimulation of either locus evoked caudate neuronal responses with Mes 5 being more effective. Higher intensity stimulation of IDN in chloralose anesthetized cats was used to determine if thresholds of trigeminal-evoked caudate responses corresponded to thresholds of particular fiber groups in the sensory afferent. In all tested units, neuronal responses were only evoked when stimulation was suprathreshold for both A/3 and A6 fibers. These data were discussed in relation to processing of oropharyngeal sensory information within the basal ganglia. Possible implications for bucco-lingual dyskinesias were noted. Trigeminal system

Caudate nucleus

Sensory-evoked

THE PROFOUND masticatory movement disorders which follow globus pallidus and substantia nigra damage [18, 22, 231 are indicative of an important basal ganglionic role in trigeminal processes. One obstacle to clarification of the functional nature of this role is a lack of data concerning parameters of oropharyngeal sensory input to the basal ganglia. The purpose of the present research was to provide some of this information by describing caudate unit responses to activation of several trigeminal primary sensory afferents. In this initial investigation, we focussed on trigeminal sensory inputs which are demonstrably important in the regulation of jaw movements. Electrical stimulation was applied to the inferior dental nerve (IDN) whose fibers (AP, AS, C) [14] carry innocuous and nociceptive information from periodontal, gingival and toothpulp receptors [ 1,131. This stimulation evokes reflex jaw opening (digastric reflex) with concurrent inhibition of jaw elevator motomeurons [ 131. Electrical stimulation was also applied to the trigeminal mesencephalic nucleus (Mes 5). Mes 5 stimulation activates 1A afferents from jaw elevator spindle receptors and elicits reflex jaw closure (masseteric reflex) [ 131 METHOD

Surgery Locally anesthetized, paralyzed cats were used to assess the effects of low intensity Mes 5 and IDN stimulation. High current levels were not used in order to avoid potentially aversive effects from activation of high threshold fibers in the IDN. Influences of these high threshold fibers were ex-

Copyright

0 1979 ANKHO

International

unit responses

plored in a second series of animals which were anesthetized with chloralose (70 mg/kg). Preliminary work showed that even light barbiturate anesthesia precluded observation of trigeminal-evoked responses in the striatum. All animals were under general anesthesia (either sodium thiamylal or ketamine) during surgery. The trachea was cannulated for artificial ventilation, and the femoral vein was cannulated for introduction of drugs. The IDN was visualized through burr holes in the anterior portion of the mandible and bipolar stimulating electrodes were inserted directly into the nerve. In chloralose anesthetized cats, a bipolar stainless steel recording electrode was also implanted through burr holes in the mandible into the nerve proximal to the stimulating electrodes (20-23 mm) in order to record the IDN compound action potential. A wider separation between stimulating and recording electrodes would have been desirable. However, in order to accomplish this, the IDN would have to be exposed between the mandibular foramen at the posterior end of the jawbone and the foramen ovale in the base of the skull. Such a procedure is precluded since it would entail extensive dissection and the attendant tissue destruction would result in afferent inhibition that could alter or obliterate sensory responsiveness. Prior to paralyzing cats (galamine triethiodide), the threshold for the digastric reflex was determined and this current level (40-60 pa with 0.5 msec pulses) was not appreciably exceeded during the subsequent unit recording session. With the shorter duration pulses (0.01-0.02 msec) used in the compound action potential experiments, threshold currents were about 200 pa. A bipolar stainless steel stimulating electrode was implanted in the rostra1 portion of Mes 5 and its final position adjusted so that minimal current (120-180 pa) evoked the masseteric re-

Inc.-0361-9230/79/010009-06$01.10/O

10

LIDSKY

E7‘ AL.

IO MSEC

CONTRA MES 5 100

MSEC

IPSI MES 5 -

\

,-

100

MSEC

FIG. 1. A shows the level at which caudate units were recorded. A typical electrode track is illustrated. Fitted circles denote loci of units responsive to Mes 5, horizontal lines denote IDN responsive units and open circles unresponsive units. Abbreviations: Sept.-septum, Caud.-caudate, A.C.-anterior commissure. B shows spontaneous thing of a caudate unit with an A-B break in the second action potential (indicated by arrow). C and D show quantitatively different responses evoked by threshold Mes 5 and IDN stimuli in the same unit. Arrow in C indicates stimulus pulse, horizontal line in D indicates stimulus train. E and F show inhibitory responses evoked by contralateral (E) and ipsilateral (F) Mes 5 stimulus trams in the same caudate unit. Each trace records 20 superimposed responses. stimulation indicated as in D. Units in B-F are from locally anesthetized cats.

TABLE

Units Sampled 1DN (chloralose) IDN (local) Mes 5 (local) Both (local) No response (local)

1

Units Responsive

4% 5% 5% 5% 5%

33 (69W 11 (1%) 30 (52%,) 8 (14%) 25 (43%)

Median Latency

Latency Range

49 50 44

1Z-400 4G150 24-160

flex. The skull and dura were removed over the head of the caudate for later introduction of recording micropipettes. In the locally anesthetized cats, lidocaine was injected into all wound margins and then the tracheal cannula was rigidly fixed in position. The head was immobilized in a non-aversive headholder [ 191 and then stereotaxic earbars and eyepieces were removed thereby eliminating potential sources of noxious pressure. After these procedures, general anethesia was allowed to dissipate. Local anesthetic was reapplied periodically throughout the recording session. All animals were paralyzed during recording sessions.

scope and recorded on FM tape for later analysis with a laboratory computer (PDP 12-Digital Corp.). Except in cases in which unit responses were clearly time-locked to consecutive stimulus presentations, validity of responses was determined by comp~ing post-stimulus histograms to histograms const~cted from control (non-stimulation) periods. Differences of at least 3 standard deviations from the mean control rate were defined as responses. Increases in rate were termed “excitatory” and decreases were termed “inhibitory” for ease of description with no implication of underlying synaptic mechanisms.

Stimulation was square wave, monophasic pulses (CO.5 msec) delivered either singly or in short trains (5 pulses with 10 msec interpulse interval). Units were extracellularly recorded with glass micropipettes (electrolyte 1.6 M potassium citrate). Unit potentials were displayed on a storage oscillo-

After completion of recording, the tip of the micropipette was broken off and the larger diameter shaft was driven into the brain to make a visible track. The recording electrode was then replaced with a stainless steel electrode and marking lesions were made at depths determined from previous

TRIGEMINALLY-EVOKED

CAUDATE

11

RESPONSES

microdrive settings. The animal was perfused with saline followed by 10% Formalin. Brains were removed and, one week later, sectioned and stained with cresyl violet. RESULTS

Cells were recorded in the head of the caudate (AP:1516.5) [26] (Fig. 1A). Although there were no obvious differences in response as a function of location within this area, the limited sampling allows no statement regarding possible topographic organization. Recordings were probably not from fibers of passage since rapid firing caused by microelectrode movement typically revealed the A-B break that is associated with recordings made in the vicinity of neural soma [lo] (Fig. 1B).

Fifty-eight caudate units were tested and a large proportion (57%) responded to trigeminal stimulation (Table 1). Excitatory responses (Fig. lC, D) were observed as frequently as were inhibitory responses (Fig. lE, F). HOWever, due to the very low spontaneous firing rate of caudate neurons (
inputs (i.e., either excitation or inhibition to both) in 6 of these units. Quantitative differences (e.g., latency, magnitude) (Fig. lC, D) served to differentiate stimulation loci. Chlordose

Anesthetized

Cuts

Caudate units in chloralose anesthetized preparations showed extremely slow rates of spontaneous firing (co.5 spikeslsec). Presence of units was first detected by increases in background activity corresponding to synaptic noise and subsequently confirmed either when IDN stimulation evoked a response or when micropipette movement caused sufficient depolarization to yield “spontaneous” activity. Under these conditions, inhibitory responses were difficult to demonstrate and many might have gone unnoticed. High intensity stimulation of the IDN (ztwice the threshold for reflex jaw opening) was much more effective than was lower current levels. Although no units were affected by stimulation at threshold for the jaw reflex, more intense stimuli drove 6% (Table 1). In contrast to locally anesthetized cats, some response latencies were relatively short. Comparison of IDN evoked response latencies for the two types of preparations are illustrated in Fig. 2A (right panel). A few responses of extremety long latency (>lOO msec) were noted with both Mes-5 (1 unit-latency 160 msec) and IDN (2 units-iatencies 150 and 400 msec) stimulation. The origins of such long latency responses are quite obscure. However, in these 3 units, spontaneous firing was totally absent. The possibility exists that these very late responses represented rebound firing after an undetected IPSP. Thresholds for activation of the various fiber groups which compose the IDN were compared to caudate unit response thresholds in 20 IDN responsive units. Such an experiment is illustrated in Fig. 2B and C. Although the separation between IDN stimulation and recording electrodes was not sufficient to allow precise conduction velocity measurements of the IDN compound action potential (see Method), responses corresponding to A@ and A6 groups were clearly distinguishable. At threshold stimulation (Fig. 2B, left panel), a deflection with 0.3 msec latency is seen. Given the 20 mm stimulating/recording separation, the fibers mediating this response have a conduction velocity of 67 misec and are therefore in the Ap range. At twice the A@ threshold, responses in the low end of A6 conduction velocity (4.5 m/set) are first detectable (Fig. 2B, middle panel; note slower time base of middle and right vs. left panel). These responses become more pronounced at higher current levels (Fig. 2B, right panel). Perhaps the faster A8 responses were not detected due to the minimal interelectrode distance but this is not germane to the present discussion since no caudate unit responses were evoked at less than twice the AJ~ threshold (see below). in accord with other eiectrophysiological studies of dental nerves [25], no clear C fiber compound was observed. Minimum threshold for activation of caudate units was equal to that intensity which evoked Afi responses in the IDN. Two of 20 responsive units fired when A6 responses were just discernible (twice AP threshold) (Fig. 2C) and the remaining I8 units fired at 3 to 5 times the AP threshold. An observation common to both locally anesthetized and chloralose anesthetized cats was that single pulse stimulation of trigeminal afferents was quite ineffective in evoking caudate neuronal responses. Short trains of 3-5 pulses were necessary with the majority of units. This was typical of both

LIDSKY

MES

II-20

21-30

IDN

5

31-40

41-50

51-60

>60

LATENCY

(MSECI

IDN COMPOUND ACTION

CAUDATE L

L-7‘ AL,.

UNIT _

2L

POTENTIALS

RESPONSES I. 4-k

3L L-

FIG. 2. A: Latency dist~butions of Mes 5 (left panel) and IDN (right panel)-evoked unit responses. IDN distribution Gstrates data from locally anesthetized cats with low intensity stimulation (open columns) and chloralose anesthetized cats with high intensity stimulation (columns with oblique lines). B: Compound action potentials from the IDN with stimulation-recording distance 20 mm. Each pdnei shows 4 successive responses. Note differences in time base between left vs. middle and right panels. Pulse duration 0.017 msec. At threshold (L) for the AP response (left panel), current intensity was 200 ~a. Stimulus pulse indicated by arrow at bottom of each panel. Middle panel shows responses at twice A@ threshold. A series of small response components in the A6 latency range can be seen (start of series indicated by oblique double arrow at top panel). The A/3 response can no longer be seen since its amplitude was much greater than a full screen deflection and also because, at this slower time base, it is merged with the stimulus artifact. At still higher stimulation intensity, the A6 responses (indicated by oblique double arrow) are more obvious (right panel). C: Responses of a caudate unit to repetitive stimulation at current levels corresponding to one, twice and three times the threshold of the AB IDN response. Each trace shows 2 successive responses. In middle panel, unit response threshold is reached (unit responded 3 of 8 times). In right panel, unit responded 7 of 8 times.

Mes 5 tion

(trains

(local

necessary

anesthetic:

in 29 of 30 units) and IDN stimula11 of 11 units: chloralose anesthetic:

27 of 33 units).

DISCUSSION

we found that the entoIn previous experiments, peduncular nucleus 1201 and substantia nigra [ 121 receive trigeminal input. Since the caudate is considered to be the primary source of afferents to the aforementioned structures [24], it was not particularly surprising to find many trigeminally driven units during the course of the present experi-

ment. There are, however, a number of notable differences between the response properties of caudate units when compared to entopeduncular and substantia nigra unit responses. First, trains of repetitive stimulation had to be delivered to trigeminal afferents in order to evoke caudate unit responses. In contrast, single pulse stimulation of the same loci had potent effects upon both entopeduncular 1201 and substantia nigra units [ 121. Second, the latencies of caudate unit responses exceeded the latencies of many unit responses recorded in the entopeduncular nucleus 1201 and substantia nigra [12) even though stimulation locations were similar. Third, very low threshold mechanosensory inputs (as-

TRIGEMINALLY-EVOKED

CAUDATE

RESPONSES

13

sociated with large diameter fibers [ 1l]), were quite effective in altering both entopeduncular [ 191and substantia nigra [ 121 unit activity. Responsive caudate units were only affected by IDN stimulation when current levels were suprathreshold for A6 fibers. Taken together, these differences of response characteristics suggest that the entopeduncular nucleus and substantia nigra receive some oropharyngeal sensory information from sources external to the portion of striatum sampled in the present experiment. Whether this input originates in another part of the striatum or from areas outside the basal ganglia (e.g. [3,21]) remains to be determined. Recent data favor the latter alternative, however. Substantia nigra unit responses to limb shock persist after striatonigral connections are severed [9]. The source of trigeminal inputs to the striatum also is unknown. Cortex and intralaminar thalamus are reasonable candidates in view of their direct connections with the striatum [24]. Findings concerning somatosensory input from the forelimb to the caudate are supportive of the thalamic pathway [7,8]. Lesions in nucleus centralis medialis eliminate caudate field potentials evoked by radial nerve stimulation [8]. The substantia nigra may also relay some trigeminal input into the caudate. Axons of the nigrostriatal bundle are of sufficiently finer diameter to give rise to many of the late responses reported here. Indeed, the responses of most substantia nigra units to trigeminal stimuli were of shorter latency than similarly-evoked caudate unit responses ]121. The functional significance of trigeminal afference to the basal ganglia is as yet uncertain. However, hypotheses advanced concerning the basal ganglia’s role in the control of limb movement [6,15] might also prove fruitful in regard to oropharyngeal activity. A basic tenet of these hypotheses is that the basal ganglia modulate ongoing movement sequences on the basis of sensory feedback. That the basal ganglia can exert modulatory influences on oropharyngeal movement has been demonstrated; electrical stimulation of the caudate has effects upon both sensory [16] and motor [29] components of the trigeminal system. The relatively long response latencies of caudate neurons do not rule out this type of modulatory function. Oropharyngeal movements are typically rhythmic and of quite long duration.

Indirect support for a notion of sensory-related basal ganglia modulation of oropharyngeal movement can be adduced from consideration of drug-induced bucco-lingual dyskinesias. These disorders (tardive dyskinesias) are characterized by masticatory difftculties, lip smacking and involuntary movements of jaw and tongue [2]. Basal ganglia damage is thought to underlie this movement disorder because chronic blockade of dopamine receptors has been implicated in the disease’s etiology [2] and post-mortem studies of dyskinetic patients have revealed abnormalities in striatum or substantia nigra [4,11]. Recent findings have emphasized the importance of sensory input for the manifestation of dyskinetic symptoms. Specifically, it was shown that mechano-sensory stimulation of perioral tissue evokes abnormal tongue and jaw movements in patients with tardive dyskinesia [27]. One interpretation of this finding is that the basal ganglia normally modulate buccolingual motor activity on the basis of sensory feedback from the perioral area. Drug induced damage results in abnormal modulatory ouput and, therefore, disordered movement patterns. One objection to a sensory feedback based modulatory function is the possibility that the afferent responses reported here are relatively nonspecific. The qualitative similarity of responses evoked by IDN and Mes 5 stimulation as well as the high proportion of responsive units are at least suggestive of such non-specificity. This issue can best be resolved by experiments which analyze the parameters of information encoded within sensory inputs to the basal ganglia. For this work, natural stimulation of peripheral receptive fields is necessary in awake animals. Such research is currently in progress in this laboratory.

ACKNOWLEDGEMENTS

The helpful comments of M. S. Levine are gratefully acknowledged. This investigation was supported by U.S.P.H.S. Biomedical Sciences Support Grant 5 SO5 RR07067-10 to State University of New York at Stony Brook and a SUNY University Awards Committee Grant.

REFERENCES 1. Anderson, D. J., A. G. Hannam and B. Matthews. Sensory mechanisms in mammalian teeth and their supporting structures. Ph\lsio/. Rev. 50: 171-195, 1970. 2. Baldessahni, R. J. and D. Tarsy. Mechanisms underlying tardive dyskinesia. In: The Bad Gungliu, edited by M. D. Yahr. New York: Raven Press, 1976, pp. 433-446. 3. Bobillier, P., S. Seguin, F. Petitjean, D. Salvert, M. Touret and M. Jouvet. The raphe nuclei of the cat brainstem: a topographic atlas of their efferent projections as revealed by autoradiography. Bruin Res. 113: 449-486, 1976. 4. Christensen, E., J. E. Moller and A. Faurbye. Neuropathological investigation of 28 brains from patients with dyskinesia. Actu Psvchiut.

Neural.

scud.

46: 1423,

1970.

5. Couch, J. R. and D. G. Goldstein. Projections of locus coeruleus and raphe nuclei to neostriatum. Sot. Neurosci. 7: 36, 1977. 6. DeLong, M. R. and P. L. Strick. Relation of basal ganglia, cerebellum and motorcortex units to ramp and ballistic limb movements. Bruin Res. 71: 327-335, 1974.

7. Diez-Martinez, O., M. Garcia Munoz, G. Prieto, J. A. Roig and H. Brust-Carmona. Sensory afferents to the caudate nucleus. Physiol.

Behuv.

16: 727-733,

1976.

8. Diez-Martinez, O., J. A. Roig, J. E. Vargas, J. Sepulveda and H. Brust-Carmona. Neural pathway mediating somatic evoked responses in the caudate nucleus of cats. Ph~*sio/. Behuv. 19: 269-276, 1977. 9. Feger, J., J. Jacquemin and C. Ohye. Peripheral excitatory input to substantia nigra. Expl Neuroi. 59: 351-360, 1978. _ 10. Frank, K. and M. G. F. Fuortes. Potentials recorded from the spinal cord with microelectrodes. J. PhJ)sio/. 130: 625-654, 1955. 11. Gross, H. and E. Klatenbach. Neuropathological findings in persistent dyskinesia after neuroleptic long-term therapy. In: The Present Status of Psychotropic Drugs. edited by A. Cerletti and F. J. Bove. Amsterdam: Excerpta Medica, I%& pp. 474476.

12. Harper, J. A. and T. I. Lidsky. Trigeminal influences on caudate and substantia nigra units. Sot. Neurosci. 7: 38, 1977.

LIDSKY

14

13. Kidokoro. Y., K. Kubota, S. Shuto and R. Sumino. Reflex organization of cat masticatory muscles. J. No~roph.rsio/. 31: 695-708, 1968. 14. Kizior. J. E.. J. W. Cuozzo and D. C. Bowman. Functional and histologic assessment of the sensory innervation of the periodontal ligament of the cat. J. tle~rt. Rcs. 47: 59-64, 1968. 15. Kornhuber, H. H. Cerebral cortex, cerebellum and basal ganglia: an introduction to the motor functions. In: The Nc~umrci~IICES: Third StrrcIy Propnrm. edited by F. 0. Schmidt and F. G. Worden. Cambridge, Mass.: MIT Press, 1974, pp. 267-280. 16. Labuszewski, T. and T. I. Lidsky. Effects of basal ganglia stimulation on brainstem trigeminal neurons. Sot. Nrr/ro.sci. 7: 40, 1977. 17. Landau, W. and G. H. Bishop. Pain from dermal, periosteal and fascial endings and from inflammation. Archs Nerwd. PsFchitrt. 69: 490-504, 1958. 18. Levine, M. S. and J. S. Schwartzbaum. Sensorimotor functions of the striatopallidal system and lateral hypothalamus and consummatory behavior in rats. J. co/flp. physiol. P.syho/. 85: 615-635, 1973. 19. Lidsky, T. I., N. A. Buchwald, C. D. Hull and M. S. Levine. Pallidal and entopeduncular single unit activity in cats during drinking. Elecf~ourrceph. c/i/r. Nerr,ophy.sio/. 39: 79-84, 1975. 20. Lidsky, T. I., J. H. Robinson, F. J. Denaro and P. M. Weinhold. Trigeminal influences on entopeduncular units. &t/i/r Rev. 141: 227-234, 1978. 21. Locke, S. Subcortical projection of the magnocellular medial geniculate body of monkey. J. corrip. Nrrrrol. 138: 321-328. 1970.

ET A,!,.

22. Marshall, J. F., J. S. Richardson and P. Teitelbaum. Nigrostriatal damage and the lateral hypothalamic syndrome: J. camp. ph~siol. Psycho/. 87: 808-830, 1974. 23 Morgan, L. 0. The corpus striatum. A study of secondary degenerations following lesions in man and of symptoms and acute degenerations following experimental lesions in cats. Arch.s Ner~rol. Ps~chirrt.

18: 495-549,

1927.

24 Nauta, W. J. H. and W. R. Mehler. Projections of the lentiform nucleus in the monkey. Brtrin Re.,. 1: 3-42, 1%6. 25 Pfaffman, C. Afferent impulses from the teeth due to pressure and noxious stimulations. J. PIi.v.tio/. 97: 207-219, 1939. 26 Snider, R. A. and W. T. Niemer. A Star-eottr.vic Atlas of‘thr Ctrt Burirr. Chicago: University of Chicago Press, 1961. 27. Thach. B. T. and J. M. Weiffenbach. The influence of controlled oral stimulation on tongue movements in patients with oral facial dyskinesia: elicitation of lateral tongue movements. In: For~rth Syulposirrttr WI Ortrl Seustrtim trd Pcrceptim. fk~vlopuer~t i/r the Fetus c/m/ Iufirut, edited by J. F. Bosma. Wash-

ington, D.C.: U.S. Government Printing Office, 1973, pp. 370383. 28. Valenstein, E. S. and B. Beer. Unipolar and bipolar electrodes in self-stimulation experiments. Am. J. Ph~.rio/. 201: 1181-1186, 1961. 29. Weinhold, P. M.. J. W. Gustafson and T. I. Lidsky. The effect of caudate nucleus stimulation on jaw movements. Sot. Nerrrosci. 6: 70, 1976.