- Email: [email protected]
Nociceptive responses in medial thalamus of the normal and arthritic rat
93
Pain, 40 (1990) 93-104 Elsevier
PAIN 01534
Nociceptive responses in medial thalamus of the normal and arthritic rat Jonathan 0. Dostrovsky a and Gisele Guilbaud b uDepr.
of Physiologv,
University of Toronto, Toronto. Ont. MSS IA8 (Canada), and h Oni& 161, INSERM, 75014 (France) (Received
6 April 1989, revision received 7 July 1989, accepted
25 August
2 rue difi&sisia, Puris
1989)
Recordings were obtained from 773 neurons located in the medial thalamus of rats. 23 of the 46 rats studied had S~rn~ been rendered arthritic by prior inoculation with Freund’s adjuvant. 262 of the neurons could be activated by peripheral stimulation. In all cases but one, only stimuli considered to be nociceptive were effective in producing responses. Most of the responses were excitatory. The majority of the responsive neurons were located in the submedius (SM), mediodorsal (MD), centrolateral. paracentral. ventromedial (VM) nuclei and medial parts of the ventrolateral (VL) nucleus. A few nociceptive neurons were also recorded in anteromedial (AM), reuniens and a few other nearby regions of thalamus. Most neurons could be activated by stimuli applied bilaterally and frequently to large regions of the body. In almost all cases the responses were maintained for the entire duration of the 15 set stimuli used and in some cases continued after cessation of the stimuli. No marked differences in incidence of responsive neurons were found between the normal and arthritic rats or between different regions. There were also no marked differences in the spontaneous rates, magnitudes of responses, or incidence of after-discharges of neurons in the various regions of medial thalamus. These findings indicate the existence of neurons responding to nociceptive stimuli in MD, AM, VM, and VL in addition to the intralaminar nuclei and SM and suggest that all these regions may be involved in mediating various aspects of nociception. Key words:
Thalamus;
Mediodorsal;
Intralaminar
nuclei;
Nociception
Introduction The role of the thalamus in mediating nociception is far from clear at the present time. It is generally assumed that different regions of the thalamus are involved in mediating various aspects of nociception. In particular, it has been proposed [29] and is generally assumed [17,39] that the ventrobasal complex is involved in mediating the sensory discriminative aspects of pain, whereas structures in medial thalamus are concerned with the affective and motivational aspects of pain. The
Correspondence to: Dr. J.O. Dostrovsky, Dept. of Physiology, University of Toronto, Toronto, Ont. M5S 1A8, Canada. ~3~-3959/~/~3.50
0 1990 Elsevier Science Publishers
intralaminar nuclei consisting of centromedial (CM), centrolateral (CL), paracentral (PC), centre median, and the parafascicular (PF) region have been implicated repeatedly in nociception [17,39], and a number of investigators have reported the existence of nociceptive neurons in these regions in the monkey, cat and rat [1,3,7,12,14,34,37]. With the advent of more sensitive anterograde tracing techniques in the past few years, it has been possible to show that the spinothalamic tract (SIT) and the functionally equivalent part of the trigeminothalamic tract have terminations in many regions of thalamus. In addition to the well known and documented termination sites in ventrobasal complex and intralaminar nuclei [2,4,5,21,28], the more recent studies have also indicated termina-
B.V. (Biomedical
Division)
94
lions in other structures such as submedius (SM), mediodorsal (MD), ventromedial (VM), ventrolateral (VL) and reuniens (Re) [9,10,25,26,35]. Since the STT is the major nociceptive ascending pathway [39], sites receiving SIT input may be involved in nociception. One particular site that has attracted a great deal of interest over the past few years is the nucleus submedius which receives a dense input from the SIT [9,25]. We recently undertook a study aimed at determining the functional characteristics of SM neurons in the rat (131. In the course of that study we had the opportunity to record from other neurons in medial thalamus. Previous studies of nociceptive inputs in the rat have tended to explore the more caudal parts of medial thalamus [3,27,37]. In this paper we present our findings on the incidence and characteristics of nociceptive neurons in rat medial thatamus. Since previous studies have suggested that it is much easier to find thalamic neurons responding to noxious stimuli in the arthritic rat [15,20,22,30], initial studies were performed on such animals and the results compared with those obtained later in normal rats. The details of the recordings from SM have been pubhshed in a separate paper ]13].
Methods The methods utilized in this study are identical to those described in our report on the response characteristics of neurons in SM [13] and therefore will be described only briefly. Experiments were performed on normal and arthritic Sprague-Dawley rats, of the same age (8-9 weeks old), weighing 250-290 g and 170-240 g respectively. Arthritic rats were purchased from Charles River (France) and housed so as to minimize animal discomfort [see 131. Animals were anesthetized with halothane in a mixture of l/3 O,, 2/3 N,O, given atropine, paralyzed with an i.v. infusion of gallamine triethiodide (Flaxedil) and artificially ventilated. The animals were maintained under deep anesthesia during the surgical procedure (2-2.5% halothane). The percentage of halothane was then reduced to OS-0.6% and maintains at this level during the
recording period which began when anesthesia reached a stable level, typically 1 h after reduction of the halothane concentration. The level of anesthesia was checked throughout the experiment [see 131. Body temperature was maintained at 37m-37S”C. and the heart rate was monitored continuously. A craniotomy was performed over the sagittal suture in the region overlying thalamus, and the dura mater was removed. The recording electrodes were glass micropipettes filled with a mixture of 5% NaCl and Pontamine Sky Blue. Usually only 2 tracks per animal. one on each side, were made. Standard extracellular single unit recording techniques were employed. The firing rate of each unit under study was displayed continuously on a chart recorder. Receptive fields were roughly determined by stimulating the paws and sometimes the tail, ears and lips. The unit’s responses to various stimuli were characterized by brushing, rubbing, pinching, stroking, tapping, moving joints, and in some cases noxious thermal stimulation. The number of sites from which responses could be evoked (paws, tail, ears, lips), the type of response (slowly adapting, rapidly adapting, degree of after-discharge), the spontaneous activity (low: < 2 Hz, medium: 2-10 Hz, high: > 10 Hz, variable, bursting), the magnitude of the response relative to the spontaneous activity (weak: < 1.S times spontaneous; medium: 1.5-3.0 x spontaneous, high: > 3 x spontaneous), and the variability of the response or after-discharge were noted. All recording sites were reconstructed from histological sections containing electrode tracks and dye marks as described previously [ 131. Only a few sites were in the medial part of mediodorsal (MD) nucleus and no attempt was made to distinguish this region termed intermediodorsal by Paxinos and Watson [33] from MD. At the levels sampled there was no clear demarcation between the anteromedial (AM) and the interanteromedial thalamic nucleus, and neurons in these regions have not been categorized separately. Results Results were obtained normal and 23 arthritic
from 773 neurons in 23 rats. Of these, 262 re-
95
sponded to stimulation of the periphery; 246 of these neurons were excited and the remaining 16 inhibited. With a single exception none of the neurons in normal rats responded to low threshold innocuous tactile stimulation of the skin. In the arthritic rats many neurons responded to joint
movement or pressure applied to the inflamed joints. However, such stimuli in the arthritic rat have been shown to activate nociceptors [16] and to evoke nociceptive responses [6,19,30,38], and we have considered these responses to be nociceptive responses.
A 7.2
oN0 RESPONSE
*PINCH
-INHIBITION
Fig. 1. Reconstruction of all the recording sites in medial thalamus of normal animals. The symbols indicate whether the neuron at each site was excited (closed circle), inhibited (-) or unaffected (open circle) by somatosensory stimulation. The sites have been plotted on slightly modified and simplified outlines taken from the atlas of Paxinos and Watson [33]. All the same nomenclature has been used except that nucleus submedius has been designated SM instead of gelatinosus (G) as used in the atlas. Abbreviations: AD, anterodorsal; AM, anteromedial; AV, anteroventral; CL, centrolateral; CM, centromedial; LD, laterodorsal; LHb, lateral habenular; MD, mediodorsal; MDL, lateral mediodorsal; MHb, medial habenular; mt, mammillothalamic tract; PC, paracentral; PO, posterior group; PV, paraventricular; PVP, posterior paraventricular; PT, paratenial; Re, reuniens; Rh, rhomboid; RT, reticular thalamic; SM, submedius; SPF, subparafascicular; VL, ventrolateral; VM, ventromedial; 21, zona incerta; 3V, third ventricle.
96
The distributions of the locations of responsive and unresponsive neurons are shown in Figs. 1 and 2. The figures reveal that only a small number of regions in medial thalamus failed to contain neurons responding to noxious stimulation. Of those where a reasonable number of neurons were sampled, only anteroventral (AV) failed to contain responsive neurons. The number of neurons sampled in anterodorsal (AD), laterodorsal (LD), lateral posterior (LP), lateral and medial habenula (LHb, MHb), paraventricular (PV), posterior nucleus (PO), subparafascicular (SPF) and ventroposterior medial (VPM) were too small to allow any reasonable conclusions to be made regarding incidence of responding neurons. The incidence of responsive neurons in those sites where more than 20 neurons were sampled, anteromedial (AM), centrolateral (CL), centromedial (CM), mediodorsal (MD), lateral mediodorsal (MDL), paracentrai (PC), reuniens (Re), rhomboid (Rh). submedius (SM), ventrolateral (VL) and ventromedial (VM). A 5.7
,
A 6.7
oN0
RESPONSE
A 7.2
*JOINT
PRESS/Mvt
-INHIBITION
Fig. 2. Reconstruction of all the recording sites in medial thalamus of arthritic animals. Symbols and abbreviations as in Fig. 1.
TABLE
I
INCIDENCE
OF RESPONSIVE
Numbers in parentheses: ilow as in Fig. 1. Nucleus
I______5%resp norm
AM
36(14)
AV
0 (6)
<‘I. CM
PC Rk RH
26 (23) 49 (35) 39 (72) 24(17) 49 (35) 14 (14) 3x (8)
SM VI. VM
49 (66) 32 (19) 61 (23)
MD MDL.
NEURONS
total number
of neurons.
Abbrevia-
% resp arthritic ___._..___~~_. 17 (18) If (14) 11 (9) 2x (47) 49 (R3) 29 (17) 36 (251 0 (13) 33 (12) 41 (81) 23 (22) 36 (36) --_-._---I.
varied from a low of 7% for Re to a high of 46% for VM. However, these differences in incidence of nociceptive neurons in the various nuclei were not statistically significant except between SM (44%; total n = 147) and CL (22%; n = 32). The incidence of inhibitory responses was small for all regions and no statistically significant differences between regions were noted. In all regions sampled except for VM and CM the incidence of responsive neurons was not significantly different in normal compared to arthritic animals. Table I shows the data obtained from both groups of animals for those structures where reasonable numbers of units were recorded. In the case of VM and CM the incidence of responsive neurons was significantly higher in the normal series compared to the arthritic. A total of 190 neurons of which 71 responded to somatosenso~ stimuli were recorded in the region designated as MD and MDL in the atlas of Paxinos and Watson [33). The boundary between these two regions is not always clear in all sections and some of our sites may have been us-classified. Most of the tracks passed through the larger main portion. Only one neuron was tested with noxious heat and it responded. Fig. 3 shows an example from a normal rat of the responses of a neuron in MD to noxious mechanical stimulation of the ipsilateral and contra~atera~ hind paw, ear and tail and noxi-
p.i
t Ear
I
48’p.c.
t I Pinch tail
Fig. 3. An example of the responses of a neuron located in MD of a normal rat. The diagram at the top tight side of figure shows the reconstruction of the recording site. The 5 segments of recordings show the responses of this neuron to noxious pressure applied to the ipsilaterai and contralateral hind paws (pi. and p.c. respectively), the ear and the tail. The response to immersion of the contralateral hind paw in water at 48“C is also shown. In each case the activity of the neuron is represented as frequency of discharge histograms (2 set bin width). Abbreviations as in Fig. 1.
ous thermal stimulation by immersion of the contraiateral paw into water at 48 o C. Fig. 4 shows an example of a neuron in MD in an arthritic rat that responded to pressure applied to the limbs and to flexion of the contralateral leg. A total of 174 neurons of which 63 responded to noxious stimuli were recorded in the region comprising CL, PC and CM. The boundaries between PC and CM and PC and CL are indistinct and thus the breakdown into these groups (Table I) is only approximate. Fig. 5 shows responses of a neuron located on the border between CM and PC in a normal rat. The neuron responded with Iittle or no after-discharge to pinch stimuli applied to the ipsilateral and contralateral hind paws and tail (other regions not tested for this neuron). This neuron also responded to noxious (48” C) thermal stimulation. Two of 3 neurons located in CM and all 6 neurons in PC of normal rats that responded
to noxious mechanical stimulation and that were tested for activation by noxious thermal stimulation were excited by the thermal stimuli. A total of 59 recording sites were found to be within the region denoted by VM [33]. Fig. 6 shows an example of the responses of a neuron in Vm in a normal animal. This neuron responded to pinch stimuh applied to the tail and ipsilateral and contralateral limbs. The responses to ipsilateral hind paw pinch and tail pinch were delayed; delayed responses were only rarely encountered in medial thalamus (3%). Fig. 7 shows examples of the responses of two neurons in arthritic rats to mechanical and thermal stimulation. Note the long after-discharges of the neuron in part B to pressure, flexion and thermal stimulation. The neuron in part A, in contrast, had very little after-discharge even to thermal stimulation. Note also that this neuron did not respond to pressure applied to
t1
tt
FLEX. p.c.
PRESS. p.i.
QL!. a i
QLf. a.c.
Fig. 4. An additional example of a neuron in MD. In tkis case the recordings were obtained from an arthritic rat. Tke focation of the recording site is shown in the reconstruction. The continuous rate-meter record (2 set bin width) shows the responses to pressure applied to the inflamed ipsilateral posterior (p.i.) and anterior (ai.) paws and contralateral anterior paw (a.c.) as well as to flexion of the contralateral posterior anklejoint. Abbreviations as in Fig. 1.
I
lair
t
t
Pinch p.i
t
t
Pinch p.c
t
c
pinch Tail
t t 48 pi.
Fig. 5. Tkis figure consists of 4 segments from a rate-meter recording that show the responses uf a ttenron locsted on the border between CM and PC of a normal rat (see location of recording site on reconstruction). The first 3 responses are to pinch stimuli applied to the ipsilateral and contralateral kind paws (p.i. and p.c. respectively) and to the tail. The last segment shows the response to immersing the ipsilateral kind paw into water at 48OC. Abbreviations as in Fig. 1.
t ai.
1
lOti.?
30s
&_&IL
Pinch L
tp.c.’
t p.i
I
‘tail ’
Fig. 6. Examples of responses to noxious mechanical stimuli of a neuron located in medial VM in a normal rat. The location of the recording site is shown on the diagram at the right. The rate-meter segments show the responses of the neuron to noxious stimulation of anterior contralateral and ipsilateral paw (a.c. and a.i. respectively). The third segment on the top row shows lack of clear response to pinching the contralateral ear. On the bottom row are shown the responses to pinching the contralateral and ipsilateral posterior paws (p.c. and p.i. respectively) and the tail. Note the delayed response to stimuli delivered to the ipsilateral hind paw and tail. Abbreviations
the inflamed forelimbs. The response to thermal stimulation at 50 o C was much higher than that to 48 o C stimulation. Three of 4 neurons located in VM of arthritic rats and 1 of 3 in normals that responded to noxious mechanical stimulation and that were tested for activation by noxious thermal stimulation were excited by the thermal stimuli. Most (85%) of the neurons that responded could be activated by stimuli applied bilaterally and frequently to all body parts tested. However, it is important to note that 33% of the neurons responding to noxious stimuli that were tested for inputs from at least 3 different sites could not be activated from at least one of the sites tested, since this indicates that the observed effects are probably due to specific sensory inputs rather than generalized changes in excitability elicited by any noxious stimulus. There were no significant dif-
as in Fig. 1.
ferences in incidence of neurons with bilateral receptive fields or neurons that could not be activated from all sites on the body depending on their locations in thalamus. Most (86%) of the neurons responded for the entire duration of the 15 set stimulus. Many (51%) displayed after-discharges ranging from 15 set (32%) to over 1 min. The responses and degree of after-discharge of many of the neurons varied from stimulus to stimulus and/or from site to site (e.g., see Figs. 5 and 6). Less than 10% of the neurons showed weak responses whereas the responses of 37% of the neurons were classified as high. No significant differences in types of responses or their magnitude relative to the locations of the neurons in thalamus were noted. Most of the neurons were spontaneously active. 47% had low rates of firing and 13% high. The
A
t I
t
FLEX
I
PRESS.
t
I
p.0.
48-p.i.
50’p.i
Fig. 7. This figure illustrates the responses of 2 VM neurons in 2 arthritic rats. The neuron in A, whose location is indicated on the reconstruction, responded with long after-discharges to extension and flexion of the ipsilateral and contralateral ankles of the hind paw (p.i. and p.c. respectively) as well as to immersion of the ipsilateral hind paw (p.i.) into water at 4X and 50°C. Note that this neuron did not respond to pressure applied to the inflamed ankle joint of the contralateral and ipsilateral anterior paws (a.c.). In B (next page) are illustrated the responses of the other neuron to extension and flexion of the ankle joint of the ipsilateral and contralateral hind paws (p.i. and p.c.) and to immersion of the ipsilateral hind paw in water at 50 o C. This neuron also responded to pressure delivered to the anterior contralateral paw. Abbreviations as in Fig. 1.
firing rate of 50% of the neurons was variable. 25% of the neurons tended to fire in bursts. There were no significant differences in the incidences of neurons in the various regions of medial thalamus with respect to spontaneous rate (low, medium and high), bursting type of activity or variable rates of firing.
Discussion The results of the present study clearly indicate that neurons excited by noxious stimuli exist over a large part of the medial thalamus and are not
limited to the intralaminar nuclei and SM. The existence of such neurons in CM, PC and CL is not surprising since the intralaminar nuclei have been implicated in nociception and previous studies have reported the existence of nociceptive neurons in these regions in a variety of species including the rat [1,3,7,12,14,17,34,37,39]. The finding of nociceptive neurons in regions surrounding the intralaminar nuclei is perhaps more surprising. The existence of nociceptive neurons in SM is consistent with recent findings of a dense spinothalamic and trigeminothalamic input to this region [9,25,35]. Perhaps the most surprising finding was the clear existence of nociceptive neurons
101
I
10Hz
t
I 5O'p.i.
t I PRESSa.c. Fig. JB.
in MD. The responses of neurons spanning large regions of medial thalamus including VM and VL are consistent with recent findings of spinothalamic inputs to many of these regions [9,10,25,26, 351, although obviously less direct polysynaptic pathways may be involved in providing the nociceptive afferent information to these regions. Most earlier studies that examined the effects of nociceptive stimuli on responses of neurons in medial thalamus were conducted on cats or monkeys. Most of the studies concentrated on the more caudal parts of medial thalamus and reported the existence of neurons responsive to noxious inputs in centrolateral and parafascicular regions [1,12,34]. More recently there have been a number of studies in rats reporting nociceptive neurons in medial thalamus [3,8,27,37]. Again most
of these have concentrated on the caudal parafascicular region or on centrolateral nucleus. Thus, the present study constitutes the first extensive exploration of the more rostra1 and medial parts of the rat medial thalamus. All the responsive neurons in this study had large, usually bilateral, receptive fields. These characteristics are in many ways similar to those of the nociceptive neurons in the centre median-parafascicular region and centrolateral nucleus reported in the previous studies [1,3,8,12, 27,34,37]. However, the incidence of neurons responding to nonnoxious stimuli was practically zero in this study whereas in some of the other studies there was a significant proportion of neurons activated by taps or low threshold mechanical stimuli [7,12,37]. The characteristics of the
102
neurons recorded in this study are consistent with the possible involvement cf these nuclei in mediating the affective-motivational aspects of pain. Some marked or reconstructed sites occurred on or adjacent to a cytoarchitectonic border. In other cases the cytoarchitectural borders were difficult to delineate. Thus there may be errors in the placement of some of the reconstructed recording within the indicated boundaries of the nuclei in Figs. 1 and 2. However. in all of the major regions studied at least some of the marked recording sites were found to be well within the boundaries of the nucleus. Thus there is no doubt that there are at least some nociceptive neurons within the boundaries of CL/PC/CM, MD/MDL. SM, AM. VM and VL. It should be pointed out that all the VL neurons recorded in the present study (see Fig. 1. AS.7) were located near the medial boundary of the nucleus and no more lateral than many of the neurons that were located in VM. The neurons recorded in AM generally were located close to the rostra1 border of CM. T‘hc anatomical studies have shown that SM is quite distinct from the other surrounding regions in terms of its anatomical connections in that it receives a dense input from the spinal cord and trigeminal nucleus caudalis [9,25,35] and has reciprocal connections with a circumscribed region of frontal cortex [9,11.23]. These anatomical differences suggest functional differences and it was thus surprising that no marked differences were noted in the response characteristics or receptive fields of the SM neurons in comparison with those of surrounding regions. This study provides the first report of a substantial number of nociceptive neurons in MD. Casey [7] reported the existence of a nociceptive neuron in monkey MD. In the cat it has been reported that tooth pulp stimulation excites some neurons in MD [24.32]. MD has been reported to receive a sparse input from the SIT in monkey and cat [25,26]. It has also been shown to receive inputs from the reticular formation, possibly as part of the spino-reticula-thalamic tract [36] as well as from the hypothalamus [31]. The MD is well known to have extensive projections to frontal cortex and receive afferent inputs from limbic structures [lg], connections consistent with a pos-
sible involvement of this region in motivational and affective components of pain. Although some previous studies of nociceptive responsive neurons in thalamus have noted an increase in incidence of nociceptive neurons in the arthritic rat [15,20.22.30], in the present study no significant differences in incidence were noted between the two groups of animals. Presumably this is due to more medial and rostra1 recording sites in the present study and suggests a difference between the functional properties of the neurons sampled. Nonetheless, it is obvious that afferent inputs affected by the arthritic condition were reaching these parts of thalamus since in the arthritic animals all of the responsive neurons in these regions responded to low threshold stimuli applied to the inflamed tissue and/or to joint movements, stimuli which in the normal animals were ineffective in eliciting responses. The responses elicited in the arthritic animals by mechanical stimulation and movements of the inflamed joints were assumed to result from activation of nociceptive afferents. Such stimuli have been shown to result in nociceptive behavior and to activate small diameter fibers [6,16.19,38]. Moreover, such stimuli are ineffective in exciting neurons in ventrobasal thalamus and medial thalamus in the normal animals [15,20,37]. The receptive fields of most of the neurons were large and in many cases included the whole body. Due to the fact that nociceptive stimuli had to be used, it was not feasible to examine the whole body and thus we are only assuming that some of the neurons could be excited from the whole body since all sites (4 to a maximum of 7) tested were effective. It is, however, noteworthy that in many cases although the receptive fields are large they did not include the whole body since stimulations at some sites were ineffective. This observation indicates that the responses observed, at least in these neurons, were not due to a general non-specific arousal type response but rather were related in some manner to nociception. This conclusion is also consistent with the fact that the stimuli used in these experiments did not cause changes in the electroencephalogram. The large size of the receptive fields of the neurons in our sample indicates that these neurons
103
are unlikely to be involved in those aspects of nociception that are related to localization of the stimulus, although they do not rule out a role in intensity discrimination. This conclusion is consistent with the proposed role of the medial structures in mediating affective and motivational rather than sensory-discriminative aspects of pain [17,29,39].
9
10
11
Acknowledgements 12
The authors wish to thank Michele Gautron for her excellent help in preparing the animals, data collection and preparing figures and to acknowledge the excellent help of Jacqueline Carroue in preparing the histological sections and of Mary Teofilo in preparing some of the figures. This study was supported by the French INSERM, a joint Canadian MRC-INSERM travel grant to J.O.D. and a U.S. NIH Grant DE05404 awarded to J.O.D.
13
14
15
16
References 17 1 Albe-Fessard, D. and Kruger, L., Duality of unit discharges from cat centrum medianum in response to natural and electrical stimulation, J. Neurophysiol., 25 (1962) 3-20. 2 Applebaum, A.E., Leonard, R.B., Kenshalo, Jr., D.R., Martin, R.F. and Willis, W.D., Nuclei in which functionally identified spinothalamic tract neurons terminate, J. Comp. Neural., 188 (1979) 575-586. 3 Benabid, A.L., Hemiksen, S.J., McGinty, J.F. and Bloom, F.E., Thalamic nucleus ventro-postero-lateralis inhibits nucleus parafascicularis response to noxious stimuli through a non-opioid pathway, Brain Res., 280 (1983) 217-231. 4 Boivie, J., The termination of the spinothalamic tract in the cat. An experimental study with silver impregnation methods, Exp. Brain Res., 12 (1971) 331-353. 5 Bowsher, D., Termination of the central pain pathway in man. The conscious appreciation of pain, Brain, 80 (1957) 606-622. 6 Calvino, B., Crepon-Bernard, M.-O. and Le Bars, D., Parallel clinical and behavioural studies of adjuvant-induced arthritis in the rat: possible relationship with ‘chronic pain,’ Behav. Brain Res., 24 (1987) 11-29. 7 Casey, K.L., Unit analysis of nociceptive mechanisms in the thalamus of the awake squirrel monkey, J. Neurophysiol., 29 (1966) 727-750. 8 Condes-Lara, M. and Zapata, LO., Suppression of noxious
18 19
20
21
22
23
24
thermal evoked responses in thalamic central lateral nucleus by cortical spreading depression, Pain, 35 (1988) 199-204. Craig, Jr., A.D. and Burton, H.. Spinal and medullary lamina I projection to nucleus submedius in medial thalamus: a possible pain center, J. Neurophysiol., 45 (1981) 443-465. Craig, Jr., A.D. and Burton, H., The distribution and topographical organization in the thalamus of anterogradely transported horseradish peroxidase after spinal injections in cat and raccoon, Exp. Brain Res.. 58 (1985) 227-254. Craig, Jr., A.D., Wiegand, S.J. and Price, J.L.. The thalamo-cortical projection of the nucleus submedius in the cat. J. Comp. Neurol., 206 (1982) 28-48. Dong, W.K.. Ryu, H. and Wagman. LH., Nociceptive responses of neurons in medial thalamus and their relationship to spinothalamic pathways, J. Neurophysiol., 41 (1978) 1592-1613. Dostrovsky, J.O. and Guilbaud, G.. Noxious stimuli excite neurons in nucleus submedius of the normal and arthritic rat, Brain Res., 460 (1988) 269-280. Duncan, G.H., Bushnell, M.C. and Feine. J.S., Single-unit responses in monkey medial thalamus during nociceptive and nonnociceptive discrimination tasks, Sot. Neurosci. Abst., 14 (1988) 581. Gautron, M. and Guilbaud, G., Somatic responses of ventrobasal thalamic neurones in polyarthritic rats. Brain Res., 237 (1982) 459-471. Guilbaud, G.. Iggo, A. and Tegner, R., Sensory receptors in ankle joint capsules of normal and arthritic rats, Exp. Brain Res., 58 (1985) 29-40. Guilbaud, G., Peschanski, M. and Besson, J.-M., Experimental data related to nociception and pain at the supraspinal level. In: P.D. Wall and R. Melzack (Eds.), Textbook of Pain, Edn. 2, Churchill Livingstone. Edinburgh. 1989, pp. 141-153. Jones, E.G., The Thalamus, Plenum Press, New York, 1985. Kayser, V. and Guilbaud, G., The analgesic effects of morphine, but not those of the enkephalinase inhibitor thiorphan are enhanced in arthritic rats, Brain Res., 267 (1983) 131-138. Kayser, V. and Guilbaud, G.. Further evidence for changes in the responsiveness of somatosensory neurons in arthritic rats: a study of the posterior intralaminar region of the thalamus, Brain Res., 323 (1984) 144147. Kerr, F.W.L., The ventral spinothalamic tract and other ascending systems of the ventral funiculus of the spinal cord, J. Comp. Neurol., 159 (1975) 335-356. Lamour, Y., Guilbaud, G. and Wilier, J.C., Altered properties and laminar distribution of neuronal responses to peripheral stimulation in the SmI cortex of the arthritic rat, Brain Res., 273 (1983) 183-187. Ma. W., Peschanski, M. and Ohara, P.T., Fine structure of the dorsal part of the nucleus submedius of the rat thalamus: an anatomical study with reference to possible pain pathways, Neuroscience, 26 (1988) 147-159. Mancia, M.. Mariotti, M.. Caraceni. A., Formenti, A.,
104
25 26
27
28
29
30
31
Imeri, L. and Palestini. M., Center median-parafascicular thalamic complex and mediodorsal nucleus unitary responses to noxious stimuli and their conditioning by limbic and mesencephalic stimulations. In: M. Tiengo et al. (Eds.), Advances in Pain Research and Therapy, Vol. 10. Raven Press, New York, 1987, pp. 17-29. Mantyh. P.W., The terminations of the spinothalamic tract in the cat, Neurosci. Lett., 38 (1983) 119-124. Mantyh, P.W.. The spinothalamic tract in the primate: a re-examination using wheatgerm agglutinin ConJugated to horseradish peroxidase, Neuroscience, 9 (1983) 847-862. McClung, R.E. and Dafny. N., The parafascicular nucleus of thalamus exhibits convergence input from the dorsal raphe and the spinal tract of the trigeminal nerve, Brain Res., 197 (1980) 525-531. Mehler, W.R., Feferman, M.E. and Nauta, W.J.H. Ascending axon degeneration following anterolateral cordotomy. An experimental study in the monkey. Brain, 83 (1960) 7188751. Melzack, R. and Casey, K.L., Sensory, motivational, and central control determinants of pain: a new conceptual model. In: D. Kenshalo (Ed.), The Skin Senses, Thomas, Springfield, IL, 1968, pp. 423-443. Menttrey. D. and Besson, J.-M., Electrophysiological characteristics of dorsal horn cells in rats with cutaneous inflammation resulting from chronic arthritis. Pain. 13 (1982) 343-364. Gno. K. and Niimi, K., Direct projections of the hypo-
32
33 34
35 36
37
38
39
thalamic nuclei to the thalamic mediodorsal nucleus in the cat. Neurosci. Lett., 57 (1985) 283-287. Palestini, M., Mariotti, M., Velasco, J.M., Formenti, A. and Mancia, M., Medialis dorsalis thalamic unitary response to tooth pulp stimulation and its conditioning by brainstem and limbic activation, Neurosci. Lett., 78 (1987) 161-165. Paxinos, G. and Watson, C., The Rat Brain in Stereotaxic Coordinates, Academic Press, Sydney, 1982. Pert. E.R. and Whitlock, D.G.. Somatic stimuli excitmg spinothalamic projections to thalamic neurons in cat and monkey, Exp. Neural.. 3 (1961) 256-296. Peschanski. M., Trigeminal afferents to the diencephalon in the rat, Neuroscience, 12 (1984) 4655487. Peschanski. M. and Besson, J.-M., A spino-reticula-thalamic pathway in the rat: an anatomical study with reference to pain transmission, Neuroscience, 12 (1984) 165-178. Peschanski, M., Guilbaud, G. and Gautron, M., Posterior intralaminar region in rat: neuronal responses to noxious and nonnoxious cutaneous stimuli, Exp. Neural.. 72 (1981) 2266238. Pircio, A.W., Fedele, C.T. and Bierwagen, M.E., A new method for the evaluation of analgesic activity using adJuvant-induced arthritis in the rat, Eur. J. Pharmacol.. 31 (1975) 2077215. Willis, Jr.. W.D.. The pam system. The neural basis of nociceptive transmission in the mammalian nervous system. In: P.L. Gildenberg (Ed.), Pain and Headache, Karger. Basel. 1985. pp. l-346.
Pain, 40 (1990) 93-104 Elsevier
PAIN 01534
Nociceptive responses in medial thalamus of the normal and arthritic rat Jonathan 0. Dostrovsky a and Gisele Guilbaud b uDepr.
of Physiologv,
University of Toronto, Toronto. Ont. MSS IA8 (Canada), and h Oni& 161, INSERM, 75014 (France) (Received
6 April 1989, revision received 7 July 1989, accepted
25 August
2 rue difi&sisia, Puris
1989)
Recordings were obtained from 773 neurons located in the medial thalamus of rats. 23 of the 46 rats studied had S~rn~ been rendered arthritic by prior inoculation with Freund’s adjuvant. 262 of the neurons could be activated by peripheral stimulation. In all cases but one, only stimuli considered to be nociceptive were effective in producing responses. Most of the responses were excitatory. The majority of the responsive neurons were located in the submedius (SM), mediodorsal (MD), centrolateral. paracentral. ventromedial (VM) nuclei and medial parts of the ventrolateral (VL) nucleus. A few nociceptive neurons were also recorded in anteromedial (AM), reuniens and a few other nearby regions of thalamus. Most neurons could be activated by stimuli applied bilaterally and frequently to large regions of the body. In almost all cases the responses were maintained for the entire duration of the 15 set stimuli used and in some cases continued after cessation of the stimuli. No marked differences in incidence of responsive neurons were found between the normal and arthritic rats or between different regions. There were also no marked differences in the spontaneous rates, magnitudes of responses, or incidence of after-discharges of neurons in the various regions of medial thalamus. These findings indicate the existence of neurons responding to nociceptive stimuli in MD, AM, VM, and VL in addition to the intralaminar nuclei and SM and suggest that all these regions may be involved in mediating various aspects of nociception. Key words:
Thalamus;
Mediodorsal;
Intralaminar
nuclei;
Nociception
Introduction The role of the thalamus in mediating nociception is far from clear at the present time. It is generally assumed that different regions of the thalamus are involved in mediating various aspects of nociception. In particular, it has been proposed [29] and is generally assumed [17,39] that the ventrobasal complex is involved in mediating the sensory discriminative aspects of pain, whereas structures in medial thalamus are concerned with the affective and motivational aspects of pain. The
Correspondence to: Dr. J.O. Dostrovsky, Dept. of Physiology, University of Toronto, Toronto, Ont. M5S 1A8, Canada. ~3~-3959/~/~3.50
0 1990 Elsevier Science Publishers
intralaminar nuclei consisting of centromedial (CM), centrolateral (CL), paracentral (PC), centre median, and the parafascicular (PF) region have been implicated repeatedly in nociception [17,39], and a number of investigators have reported the existence of nociceptive neurons in these regions in the monkey, cat and rat [1,3,7,12,14,34,37]. With the advent of more sensitive anterograde tracing techniques in the past few years, it has been possible to show that the spinothalamic tract (SIT) and the functionally equivalent part of the trigeminothalamic tract have terminations in many regions of thalamus. In addition to the well known and documented termination sites in ventrobasal complex and intralaminar nuclei [2,4,5,21,28], the more recent studies have also indicated termina-
B.V. (Biomedical
Division)
94
lions in other structures such as submedius (SM), mediodorsal (MD), ventromedial (VM), ventrolateral (VL) and reuniens (Re) [9,10,25,26,35]. Since the STT is the major nociceptive ascending pathway [39], sites receiving SIT input may be involved in nociception. One particular site that has attracted a great deal of interest over the past few years is the nucleus submedius which receives a dense input from the SIT [9,25]. We recently undertook a study aimed at determining the functional characteristics of SM neurons in the rat (131. In the course of that study we had the opportunity to record from other neurons in medial thalamus. Previous studies of nociceptive inputs in the rat have tended to explore the more caudal parts of medial thalamus [3,27,37]. In this paper we present our findings on the incidence and characteristics of nociceptive neurons in rat medial thatamus. Since previous studies have suggested that it is much easier to find thalamic neurons responding to noxious stimuli in the arthritic rat [15,20,22,30], initial studies were performed on such animals and the results compared with those obtained later in normal rats. The details of the recordings from SM have been pubhshed in a separate paper ]13].
Methods The methods utilized in this study are identical to those described in our report on the response characteristics of neurons in SM [13] and therefore will be described only briefly. Experiments were performed on normal and arthritic Sprague-Dawley rats, of the same age (8-9 weeks old), weighing 250-290 g and 170-240 g respectively. Arthritic rats were purchased from Charles River (France) and housed so as to minimize animal discomfort [see 131. Animals were anesthetized with halothane in a mixture of l/3 O,, 2/3 N,O, given atropine, paralyzed with an i.v. infusion of gallamine triethiodide (Flaxedil) and artificially ventilated. The animals were maintained under deep anesthesia during the surgical procedure (2-2.5% halothane). The percentage of halothane was then reduced to OS-0.6% and maintains at this level during the
recording period which began when anesthesia reached a stable level, typically 1 h after reduction of the halothane concentration. The level of anesthesia was checked throughout the experiment [see 131. Body temperature was maintained at 37m-37S”C. and the heart rate was monitored continuously. A craniotomy was performed over the sagittal suture in the region overlying thalamus, and the dura mater was removed. The recording electrodes were glass micropipettes filled with a mixture of 5% NaCl and Pontamine Sky Blue. Usually only 2 tracks per animal. one on each side, were made. Standard extracellular single unit recording techniques were employed. The firing rate of each unit under study was displayed continuously on a chart recorder. Receptive fields were roughly determined by stimulating the paws and sometimes the tail, ears and lips. The unit’s responses to various stimuli were characterized by brushing, rubbing, pinching, stroking, tapping, moving joints, and in some cases noxious thermal stimulation. The number of sites from which responses could be evoked (paws, tail, ears, lips), the type of response (slowly adapting, rapidly adapting, degree of after-discharge), the spontaneous activity (low: < 2 Hz, medium: 2-10 Hz, high: > 10 Hz, variable, bursting), the magnitude of the response relative to the spontaneous activity (weak: < 1.S times spontaneous; medium: 1.5-3.0 x spontaneous, high: > 3 x spontaneous), and the variability of the response or after-discharge were noted. All recording sites were reconstructed from histological sections containing electrode tracks and dye marks as described previously [ 131. Only a few sites were in the medial part of mediodorsal (MD) nucleus and no attempt was made to distinguish this region termed intermediodorsal by Paxinos and Watson [33] from MD. At the levels sampled there was no clear demarcation between the anteromedial (AM) and the interanteromedial thalamic nucleus, and neurons in these regions have not been categorized separately. Results Results were obtained normal and 23 arthritic
from 773 neurons in 23 rats. Of these, 262 re-
95
sponded to stimulation of the periphery; 246 of these neurons were excited and the remaining 16 inhibited. With a single exception none of the neurons in normal rats responded to low threshold innocuous tactile stimulation of the skin. In the arthritic rats many neurons responded to joint
movement or pressure applied to the inflamed joints. However, such stimuli in the arthritic rat have been shown to activate nociceptors [16] and to evoke nociceptive responses [6,19,30,38], and we have considered these responses to be nociceptive responses.
A 7.2
oN0 RESPONSE
*PINCH
-INHIBITION
Fig. 1. Reconstruction of all the recording sites in medial thalamus of normal animals. The symbols indicate whether the neuron at each site was excited (closed circle), inhibited (-) or unaffected (open circle) by somatosensory stimulation. The sites have been plotted on slightly modified and simplified outlines taken from the atlas of Paxinos and Watson [33]. All the same nomenclature has been used except that nucleus submedius has been designated SM instead of gelatinosus (G) as used in the atlas. Abbreviations: AD, anterodorsal; AM, anteromedial; AV, anteroventral; CL, centrolateral; CM, centromedial; LD, laterodorsal; LHb, lateral habenular; MD, mediodorsal; MDL, lateral mediodorsal; MHb, medial habenular; mt, mammillothalamic tract; PC, paracentral; PO, posterior group; PV, paraventricular; PVP, posterior paraventricular; PT, paratenial; Re, reuniens; Rh, rhomboid; RT, reticular thalamic; SM, submedius; SPF, subparafascicular; VL, ventrolateral; VM, ventromedial; 21, zona incerta; 3V, third ventricle.
96
The distributions of the locations of responsive and unresponsive neurons are shown in Figs. 1 and 2. The figures reveal that only a small number of regions in medial thalamus failed to contain neurons responding to noxious stimulation. Of those where a reasonable number of neurons were sampled, only anteroventral (AV) failed to contain responsive neurons. The number of neurons sampled in anterodorsal (AD), laterodorsal (LD), lateral posterior (LP), lateral and medial habenula (LHb, MHb), paraventricular (PV), posterior nucleus (PO), subparafascicular (SPF) and ventroposterior medial (VPM) were too small to allow any reasonable conclusions to be made regarding incidence of responding neurons. The incidence of responsive neurons in those sites where more than 20 neurons were sampled, anteromedial (AM), centrolateral (CL), centromedial (CM), mediodorsal (MD), lateral mediodorsal (MDL), paracentrai (PC), reuniens (Re), rhomboid (Rh). submedius (SM), ventrolateral (VL) and ventromedial (VM). A 5.7
,
A 6.7
oN0
RESPONSE
A 7.2
*JOINT
PRESS/Mvt
-INHIBITION
Fig. 2. Reconstruction of all the recording sites in medial thalamus of arthritic animals. Symbols and abbreviations as in Fig. 1.
TABLE
I
INCIDENCE
OF RESPONSIVE
Numbers in parentheses: ilow as in Fig. 1. Nucleus
I______5%resp norm
AM
36(14)
AV
0 (6)
<‘I. CM
PC Rk RH
26 (23) 49 (35) 39 (72) 24(17) 49 (35) 14 (14) 3x (8)
SM VI. VM
49 (66) 32 (19) 61 (23)
MD MDL.
NEURONS
total number
of neurons.
Abbrevia-
% resp arthritic ___._..___~~_. 17 (18) If (14) 11 (9) 2x (47) 49 (R3) 29 (17) 36 (251 0 (13) 33 (12) 41 (81) 23 (22) 36 (36) --_-._---I.
varied from a low of 7% for Re to a high of 46% for VM. However, these differences in incidence of nociceptive neurons in the various nuclei were not statistically significant except between SM (44%; total n = 147) and CL (22%; n = 32). The incidence of inhibitory responses was small for all regions and no statistically significant differences between regions were noted. In all regions sampled except for VM and CM the incidence of responsive neurons was not significantly different in normal compared to arthritic animals. Table I shows the data obtained from both groups of animals for those structures where reasonable numbers of units were recorded. In the case of VM and CM the incidence of responsive neurons was significantly higher in the normal series compared to the arthritic. A total of 190 neurons of which 71 responded to somatosenso~ stimuli were recorded in the region designated as MD and MDL in the atlas of Paxinos and Watson [33). The boundary between these two regions is not always clear in all sections and some of our sites may have been us-classified. Most of the tracks passed through the larger main portion. Only one neuron was tested with noxious heat and it responded. Fig. 3 shows an example from a normal rat of the responses of a neuron in MD to noxious mechanical stimulation of the ipsilateral and contra~atera~ hind paw, ear and tail and noxi-
p.i
t Ear
I
48’p.c.
t I Pinch tail
Fig. 3. An example of the responses of a neuron located in MD of a normal rat. The diagram at the top tight side of figure shows the reconstruction of the recording site. The 5 segments of recordings show the responses of this neuron to noxious pressure applied to the ipsilaterai and contralateral hind paws (pi. and p.c. respectively), the ear and the tail. The response to immersion of the contralateral hind paw in water at 48“C is also shown. In each case the activity of the neuron is represented as frequency of discharge histograms (2 set bin width). Abbreviations as in Fig. 1.
ous thermal stimulation by immersion of the contraiateral paw into water at 48 o C. Fig. 4 shows an example of a neuron in MD in an arthritic rat that responded to pressure applied to the limbs and to flexion of the contralateral leg. A total of 174 neurons of which 63 responded to noxious stimuli were recorded in the region comprising CL, PC and CM. The boundaries between PC and CM and PC and CL are indistinct and thus the breakdown into these groups (Table I) is only approximate. Fig. 5 shows responses of a neuron located on the border between CM and PC in a normal rat. The neuron responded with Iittle or no after-discharge to pinch stimuli applied to the ipsilateral and contralateral hind paws and tail (other regions not tested for this neuron). This neuron also responded to noxious (48” C) thermal stimulation. Two of 3 neurons located in CM and all 6 neurons in PC of normal rats that responded
to noxious mechanical stimulation and that were tested for activation by noxious thermal stimulation were excited by the thermal stimuli. A total of 59 recording sites were found to be within the region denoted by VM [33]. Fig. 6 shows an example of the responses of a neuron in Vm in a normal animal. This neuron responded to pinch stimuh applied to the tail and ipsilateral and contralateral limbs. The responses to ipsilateral hind paw pinch and tail pinch were delayed; delayed responses were only rarely encountered in medial thalamus (3%). Fig. 7 shows examples of the responses of two neurons in arthritic rats to mechanical and thermal stimulation. Note the long after-discharges of the neuron in part B to pressure, flexion and thermal stimulation. The neuron in part A, in contrast, had very little after-discharge even to thermal stimulation. Note also that this neuron did not respond to pressure applied to
t1
tt
FLEX. p.c.
PRESS. p.i.
QL!. a i
QLf. a.c.
Fig. 4. An additional example of a neuron in MD. In tkis case the recordings were obtained from an arthritic rat. Tke focation of the recording site is shown in the reconstruction. The continuous rate-meter record (2 set bin width) shows the responses to pressure applied to the inflamed ipsilateral posterior (p.i.) and anterior (ai.) paws and contralateral anterior paw (a.c.) as well as to flexion of the contralateral posterior anklejoint. Abbreviations as in Fig. 1.
I
lair
t
t
Pinch p.i
t
t
Pinch p.c
t
c
pinch Tail
t t 48 pi.
Fig. 5. Tkis figure consists of 4 segments from a rate-meter recording that show the responses uf a ttenron locsted on the border between CM and PC of a normal rat (see location of recording site on reconstruction). The first 3 responses are to pinch stimuli applied to the ipsilateral and contralateral kind paws (p.i. and p.c. respectively) and to the tail. The last segment shows the response to immersing the ipsilateral kind paw into water at 48OC. Abbreviations as in Fig. 1.
t ai.
1
lOti.?
30s
&_&IL
Pinch L
tp.c.’
t p.i
I
‘tail ’
Fig. 6. Examples of responses to noxious mechanical stimuli of a neuron located in medial VM in a normal rat. The location of the recording site is shown on the diagram at the right. The rate-meter segments show the responses of the neuron to noxious stimulation of anterior contralateral and ipsilateral paw (a.c. and a.i. respectively). The third segment on the top row shows lack of clear response to pinching the contralateral ear. On the bottom row are shown the responses to pinching the contralateral and ipsilateral posterior paws (p.c. and p.i. respectively) and the tail. Note the delayed response to stimuli delivered to the ipsilateral hind paw and tail. Abbreviations
the inflamed forelimbs. The response to thermal stimulation at 50 o C was much higher than that to 48 o C stimulation. Three of 4 neurons located in VM of arthritic rats and 1 of 3 in normals that responded to noxious mechanical stimulation and that were tested for activation by noxious thermal stimulation were excited by the thermal stimuli. Most (85%) of the neurons that responded could be activated by stimuli applied bilaterally and frequently to all body parts tested. However, it is important to note that 33% of the neurons responding to noxious stimuli that were tested for inputs from at least 3 different sites could not be activated from at least one of the sites tested, since this indicates that the observed effects are probably due to specific sensory inputs rather than generalized changes in excitability elicited by any noxious stimulus. There were no significant dif-
as in Fig. 1.
ferences in incidence of neurons with bilateral receptive fields or neurons that could not be activated from all sites on the body depending on their locations in thalamus. Most (86%) of the neurons responded for the entire duration of the 15 set stimulus. Many (51%) displayed after-discharges ranging from 15 set (32%) to over 1 min. The responses and degree of after-discharge of many of the neurons varied from stimulus to stimulus and/or from site to site (e.g., see Figs. 5 and 6). Less than 10% of the neurons showed weak responses whereas the responses of 37% of the neurons were classified as high. No significant differences in types of responses or their magnitude relative to the locations of the neurons in thalamus were noted. Most of the neurons were spontaneously active. 47% had low rates of firing and 13% high. The
A
t I
t
FLEX
I
PRESS.
t
I
p.0.
48-p.i.
50’p.i
Fig. 7. This figure illustrates the responses of 2 VM neurons in 2 arthritic rats. The neuron in A, whose location is indicated on the reconstruction, responded with long after-discharges to extension and flexion of the ipsilateral and contralateral ankles of the hind paw (p.i. and p.c. respectively) as well as to immersion of the ipsilateral hind paw (p.i.) into water at 4X and 50°C. Note that this neuron did not respond to pressure applied to the inflamed ankle joint of the contralateral and ipsilateral anterior paws (a.c.). In B (next page) are illustrated the responses of the other neuron to extension and flexion of the ankle joint of the ipsilateral and contralateral hind paws (p.i. and p.c.) and to immersion of the ipsilateral hind paw in water at 50 o C. This neuron also responded to pressure delivered to the anterior contralateral paw. Abbreviations as in Fig. 1.
firing rate of 50% of the neurons was variable. 25% of the neurons tended to fire in bursts. There were no significant differences in the incidences of neurons in the various regions of medial thalamus with respect to spontaneous rate (low, medium and high), bursting type of activity or variable rates of firing.
Discussion The results of the present study clearly indicate that neurons excited by noxious stimuli exist over a large part of the medial thalamus and are not
limited to the intralaminar nuclei and SM. The existence of such neurons in CM, PC and CL is not surprising since the intralaminar nuclei have been implicated in nociception and previous studies have reported the existence of nociceptive neurons in these regions in a variety of species including the rat [1,3,7,12,14,17,34,37,39]. The finding of nociceptive neurons in regions surrounding the intralaminar nuclei is perhaps more surprising. The existence of nociceptive neurons in SM is consistent with recent findings of a dense spinothalamic and trigeminothalamic input to this region [9,25,35]. Perhaps the most surprising finding was the clear existence of nociceptive neurons
101
I
10Hz
t
I 5O'p.i.
t I PRESSa.c. Fig. JB.
in MD. The responses of neurons spanning large regions of medial thalamus including VM and VL are consistent with recent findings of spinothalamic inputs to many of these regions [9,10,25,26, 351, although obviously less direct polysynaptic pathways may be involved in providing the nociceptive afferent information to these regions. Most earlier studies that examined the effects of nociceptive stimuli on responses of neurons in medial thalamus were conducted on cats or monkeys. Most of the studies concentrated on the more caudal parts of medial thalamus and reported the existence of neurons responsive to noxious inputs in centrolateral and parafascicular regions [1,12,34]. More recently there have been a number of studies in rats reporting nociceptive neurons in medial thalamus [3,8,27,37]. Again most
of these have concentrated on the caudal parafascicular region or on centrolateral nucleus. Thus, the present study constitutes the first extensive exploration of the more rostra1 and medial parts of the rat medial thalamus. All the responsive neurons in this study had large, usually bilateral, receptive fields. These characteristics are in many ways similar to those of the nociceptive neurons in the centre median-parafascicular region and centrolateral nucleus reported in the previous studies [1,3,8,12, 27,34,37]. However, the incidence of neurons responding to nonnoxious stimuli was practically zero in this study whereas in some of the other studies there was a significant proportion of neurons activated by taps or low threshold mechanical stimuli [7,12,37]. The characteristics of the
102
neurons recorded in this study are consistent with the possible involvement cf these nuclei in mediating the affective-motivational aspects of pain. Some marked or reconstructed sites occurred on or adjacent to a cytoarchitectonic border. In other cases the cytoarchitectural borders were difficult to delineate. Thus there may be errors in the placement of some of the reconstructed recording within the indicated boundaries of the nuclei in Figs. 1 and 2. However. in all of the major regions studied at least some of the marked recording sites were found to be well within the boundaries of the nucleus. Thus there is no doubt that there are at least some nociceptive neurons within the boundaries of CL/PC/CM, MD/MDL. SM, AM. VM and VL. It should be pointed out that all the VL neurons recorded in the present study (see Fig. 1. AS.7) were located near the medial boundary of the nucleus and no more lateral than many of the neurons that were located in VM. The neurons recorded in AM generally were located close to the rostra1 border of CM. T‘hc anatomical studies have shown that SM is quite distinct from the other surrounding regions in terms of its anatomical connections in that it receives a dense input from the spinal cord and trigeminal nucleus caudalis [9,25,35] and has reciprocal connections with a circumscribed region of frontal cortex [9,11.23]. These anatomical differences suggest functional differences and it was thus surprising that no marked differences were noted in the response characteristics or receptive fields of the SM neurons in comparison with those of surrounding regions. This study provides the first report of a substantial number of nociceptive neurons in MD. Casey [7] reported the existence of a nociceptive neuron in monkey MD. In the cat it has been reported that tooth pulp stimulation excites some neurons in MD [24.32]. MD has been reported to receive a sparse input from the SIT in monkey and cat [25,26]. It has also been shown to receive inputs from the reticular formation, possibly as part of the spino-reticula-thalamic tract [36] as well as from the hypothalamus [31]. The MD is well known to have extensive projections to frontal cortex and receive afferent inputs from limbic structures [lg], connections consistent with a pos-
sible involvement of this region in motivational and affective components of pain. Although some previous studies of nociceptive responsive neurons in thalamus have noted an increase in incidence of nociceptive neurons in the arthritic rat [15,20.22.30], in the present study no significant differences in incidence were noted between the two groups of animals. Presumably this is due to more medial and rostra1 recording sites in the present study and suggests a difference between the functional properties of the neurons sampled. Nonetheless, it is obvious that afferent inputs affected by the arthritic condition were reaching these parts of thalamus since in the arthritic animals all of the responsive neurons in these regions responded to low threshold stimuli applied to the inflamed tissue and/or to joint movements, stimuli which in the normal animals were ineffective in eliciting responses. The responses elicited in the arthritic animals by mechanical stimulation and movements of the inflamed joints were assumed to result from activation of nociceptive afferents. Such stimuli have been shown to result in nociceptive behavior and to activate small diameter fibers [6,16.19,38]. Moreover, such stimuli are ineffective in exciting neurons in ventrobasal thalamus and medial thalamus in the normal animals [15,20,37]. The receptive fields of most of the neurons were large and in many cases included the whole body. Due to the fact that nociceptive stimuli had to be used, it was not feasible to examine the whole body and thus we are only assuming that some of the neurons could be excited from the whole body since all sites (4 to a maximum of 7) tested were effective. It is, however, noteworthy that in many cases although the receptive fields are large they did not include the whole body since stimulations at some sites were ineffective. This observation indicates that the responses observed, at least in these neurons, were not due to a general non-specific arousal type response but rather were related in some manner to nociception. This conclusion is also consistent with the fact that the stimuli used in these experiments did not cause changes in the electroencephalogram. The large size of the receptive fields of the neurons in our sample indicates that these neurons
103
are unlikely to be involved in those aspects of nociception that are related to localization of the stimulus, although they do not rule out a role in intensity discrimination. This conclusion is consistent with the proposed role of the medial structures in mediating affective and motivational rather than sensory-discriminative aspects of pain [17,29,39].
9
10
11
Acknowledgements 12
The authors wish to thank Michele Gautron for her excellent help in preparing the animals, data collection and preparing figures and to acknowledge the excellent help of Jacqueline Carroue in preparing the histological sections and of Mary Teofilo in preparing some of the figures. This study was supported by the French INSERM, a joint Canadian MRC-INSERM travel grant to J.O.D. and a U.S. NIH Grant DE05404 awarded to J.O.D.
13
14
15
16
References 17 1 Albe-Fessard, D. and Kruger, L., Duality of unit discharges from cat centrum medianum in response to natural and electrical stimulation, J. Neurophysiol., 25 (1962) 3-20. 2 Applebaum, A.E., Leonard, R.B., Kenshalo, Jr., D.R., Martin, R.F. and Willis, W.D., Nuclei in which functionally identified spinothalamic tract neurons terminate, J. Comp. Neural., 188 (1979) 575-586. 3 Benabid, A.L., Hemiksen, S.J., McGinty, J.F. and Bloom, F.E., Thalamic nucleus ventro-postero-lateralis inhibits nucleus parafascicularis response to noxious stimuli through a non-opioid pathway, Brain Res., 280 (1983) 217-231. 4 Boivie, J., The termination of the spinothalamic tract in the cat. An experimental study with silver impregnation methods, Exp. Brain Res., 12 (1971) 331-353. 5 Bowsher, D., Termination of the central pain pathway in man. The conscious appreciation of pain, Brain, 80 (1957) 606-622. 6 Calvino, B., Crepon-Bernard, M.-O. and Le Bars, D., Parallel clinical and behavioural studies of adjuvant-induced arthritis in the rat: possible relationship with ‘chronic pain,’ Behav. Brain Res., 24 (1987) 11-29. 7 Casey, K.L., Unit analysis of nociceptive mechanisms in the thalamus of the awake squirrel monkey, J. Neurophysiol., 29 (1966) 727-750. 8 Condes-Lara, M. and Zapata, LO., Suppression of noxious
18 19
20
21
22
23
24
thermal evoked responses in thalamic central lateral nucleus by cortical spreading depression, Pain, 35 (1988) 199-204. Craig, Jr., A.D. and Burton, H.. Spinal and medullary lamina I projection to nucleus submedius in medial thalamus: a possible pain center, J. Neurophysiol., 45 (1981) 443-465. Craig, Jr., A.D. and Burton, H., The distribution and topographical organization in the thalamus of anterogradely transported horseradish peroxidase after spinal injections in cat and raccoon, Exp. Brain Res.. 58 (1985) 227-254. Craig, Jr., A.D., Wiegand, S.J. and Price, J.L.. The thalamo-cortical projection of the nucleus submedius in the cat. J. Comp. Neurol., 206 (1982) 28-48. Dong, W.K.. Ryu, H. and Wagman. LH., Nociceptive responses of neurons in medial thalamus and their relationship to spinothalamic pathways, J. Neurophysiol., 41 (1978) 1592-1613. Dostrovsky, J.O. and Guilbaud, G.. Noxious stimuli excite neurons in nucleus submedius of the normal and arthritic rat, Brain Res., 460 (1988) 269-280. Duncan, G.H., Bushnell, M.C. and Feine. J.S., Single-unit responses in monkey medial thalamus during nociceptive and nonnociceptive discrimination tasks, Sot. Neurosci. Abst., 14 (1988) 581. Gautron, M. and Guilbaud, G., Somatic responses of ventrobasal thalamic neurones in polyarthritic rats. Brain Res., 237 (1982) 459-471. Guilbaud, G.. Iggo, A. and Tegner, R., Sensory receptors in ankle joint capsules of normal and arthritic rats, Exp. Brain Res., 58 (1985) 29-40. Guilbaud, G., Peschanski, M. and Besson, J.-M., Experimental data related to nociception and pain at the supraspinal level. In: P.D. Wall and R. Melzack (Eds.), Textbook of Pain, Edn. 2, Churchill Livingstone. Edinburgh. 1989, pp. 141-153. Jones, E.G., The Thalamus, Plenum Press, New York, 1985. Kayser, V. and Guilbaud, G., The analgesic effects of morphine, but not those of the enkephalinase inhibitor thiorphan are enhanced in arthritic rats, Brain Res., 267 (1983) 131-138. Kayser, V. and Guilbaud, G.. Further evidence for changes in the responsiveness of somatosensory neurons in arthritic rats: a study of the posterior intralaminar region of the thalamus, Brain Res., 323 (1984) 144147. Kerr, F.W.L., The ventral spinothalamic tract and other ascending systems of the ventral funiculus of the spinal cord, J. Comp. Neurol., 159 (1975) 335-356. Lamour, Y., Guilbaud, G. and Wilier, J.C., Altered properties and laminar distribution of neuronal responses to peripheral stimulation in the SmI cortex of the arthritic rat, Brain Res., 273 (1983) 183-187. Ma. W., Peschanski, M. and Ohara, P.T., Fine structure of the dorsal part of the nucleus submedius of the rat thalamus: an anatomical study with reference to possible pain pathways, Neuroscience, 26 (1988) 147-159. Mancia, M.. Mariotti, M.. Caraceni. A., Formenti, A.,
104
25 26
27
28
29
30
31
Imeri, L. and Palestini. M., Center median-parafascicular thalamic complex and mediodorsal nucleus unitary responses to noxious stimuli and their conditioning by limbic and mesencephalic stimulations. In: M. Tiengo et al. (Eds.), Advances in Pain Research and Therapy, Vol. 10. Raven Press, New York, 1987, pp. 17-29. Mantyh. P.W., The terminations of the spinothalamic tract in the cat, Neurosci. Lett., 38 (1983) 119-124. Mantyh, P.W.. The spinothalamic tract in the primate: a re-examination using wheatgerm agglutinin ConJugated to horseradish peroxidase, Neuroscience, 9 (1983) 847-862. McClung, R.E. and Dafny. N., The parafascicular nucleus of thalamus exhibits convergence input from the dorsal raphe and the spinal tract of the trigeminal nerve, Brain Res., 197 (1980) 525-531. Mehler, W.R., Feferman, M.E. and Nauta, W.J.H. Ascending axon degeneration following anterolateral cordotomy. An experimental study in the monkey. Brain, 83 (1960) 7188751. Melzack, R. and Casey, K.L., Sensory, motivational, and central control determinants of pain: a new conceptual model. In: D. Kenshalo (Ed.), The Skin Senses, Thomas, Springfield, IL, 1968, pp. 423-443. Menttrey. D. and Besson, J.-M., Electrophysiological characteristics of dorsal horn cells in rats with cutaneous inflammation resulting from chronic arthritis. Pain. 13 (1982) 343-364. Gno. K. and Niimi, K., Direct projections of the hypo-
32
33 34
35 36
37
38
39
thalamic nuclei to the thalamic mediodorsal nucleus in the cat. Neurosci. Lett., 57 (1985) 283-287. Palestini, M., Mariotti, M., Velasco, J.M., Formenti, A. and Mancia, M., Medialis dorsalis thalamic unitary response to tooth pulp stimulation and its conditioning by brainstem and limbic activation, Neurosci. Lett., 78 (1987) 161-165. Paxinos, G. and Watson, C., The Rat Brain in Stereotaxic Coordinates, Academic Press, Sydney, 1982. Pert. E.R. and Whitlock, D.G.. Somatic stimuli excitmg spinothalamic projections to thalamic neurons in cat and monkey, Exp. Neural.. 3 (1961) 256-296. Peschanski. M., Trigeminal afferents to the diencephalon in the rat, Neuroscience, 12 (1984) 4655487. Peschanski. M. and Besson, J.-M., A spino-reticula-thalamic pathway in the rat: an anatomical study with reference to pain transmission, Neuroscience, 12 (1984) 165-178. Peschanski, M., Guilbaud, G. and Gautron, M., Posterior intralaminar region in rat: neuronal responses to noxious and nonnoxious cutaneous stimuli, Exp. Neural.. 72 (1981) 2266238. Pircio, A.W., Fedele, C.T. and Bierwagen, M.E., A new method for the evaluation of analgesic activity using adJuvant-induced arthritis in the rat, Eur. J. Pharmacol.. 31 (1975) 2077215. Willis, Jr.. W.D.. The pam system. The neural basis of nociceptive transmission in the mammalian nervous system. In: P.L. Gildenberg (Ed.), Pain and Headache, Karger. Basel. 1985. pp. l-346.