0306-4522/90 $3.00 + 0.00 Pergamon Press plc 0 1990 IBRO
Neuroscience Vol. 34, No. 2, pp. 379-390, 1990 Printed in Great Britain
EFFECTS OF MOTOR CORTEX AND SINGLE MUSCLE STIMULATION ON NEURONS OF THE LATERAL VESTIBULAR NUCLEUS IN THE RAT F. Istituto
LICATA,
G. LI VOLSI,
di Fisiologia
umana,
G.
MAUGERI
Viale Andrea
and
Doria
F. SANTANCELO 6, 95125 Catania,
Italy
Abstract-The neuronal responses to stimulation of motor cortical sites and of forelimb single muscles were studied in the lateral vestibular nucleus of anaesthetized rats. Of the 228 neurons tested for response to stimulation of contralateral motor cortex, 63% responded to cortical sites controlling extensor muscles and 30% to those controlling flexors. The corresponding figures for responders to ipsilateral stimulation were 34 and 21%. Vestibulospinal units responded to cortical sites controlling extensor and flexor muscles whereas the remaining lateral vestibular nucleus neurons, very reactive to cortical sites controlling extensor muscles, responded little to contralateral and not at all to ipsilateral cortical sites controlling flexor muscles. The effects evoked by contralateral cortical sites controlling extensors varied, those induced by cortical sites controlling flexors were inhibitory in 77% of cases. The responses to ipsilateral motor cortex stimulation differed not so much by cortical sites controlling extensor or flexor muscles as by whether the neuron was in the dorsal or ventral zone of the lateral vestibular nucleus: mixed in the former, all inhibitory in the latter. Of the lateral vestibular nucleus units tested for response to stimulation of ipsilateral or contralateral forelimb distal muscles, only 11% responded. All the vestibulospinal units responsive to muscle stimulation lay in the dorsal zone of the nucleus. The remainder, dorsal or ventral, were not responsive to contralateral muscles. Single lateral vestibular nucleus cells influenced both by ipsilateral muscle and by contralateral motor cortex made up 24% of the pool, vestibulospinal and non-vestibulospinal. They fell into three groups: responsive to one or both structures but responding more strongly to combined stimulation; responsive to each of the two structures but showing a response to combined stimulation not significantly different from that evoked by the cortex alone; responsive only to combined stimulation. The lateral vestibular nucleus units included in these three groups accounted for 29% of those tested for response to extensor muscles and cortical sites controlling extensors and 15% of those tested for response to flexor muscles and cortical sites controlling flexors. Twenty-five per cent of the vestibulospinal neurons responded both to contralateral muscles and to ipsilateral motor cortex stimulation but none of the non-vestibulospinal neurons responded to both. All the responders to both were in the dorsal zone of the lateral vestibular nucleus and responded to extensor stimuli, always in the same way. These results indicate that motor cortex output exerts a major influence on lateral vestibular nucleus discharges, while the muscle afferents have a modulatory influence on the lateral vestibular nucleus responses to cortex. The units more responsible for integration of cortical and peripheral information are the vestibulospinal neurons of the dorsal zone.
LVN neuronal activity to the same extent and, if so, how, has not been tested. The LVN also receives input from spinal inputs originating in neck5 and limb3*4~33~38~39 regions. The major contribution from limbs seems to come from skin and joint afferents, those of muscle and, specifically, primary afferents, contributing little.39 The aim of the present research was: (a) to study the characteristics of motor cortical control over LVN neurons; (b) to compare it with the direct muscle influence on single LVN neurons and to find out whether any integration of cortex and muscle information takes place in the nucleus. It might be useful to know whether the muscle information delivered to LVN is segregated from the motor cortical input or, converging on the same cells, helps to modify the LVN responses to cortical volleys. As some anatomical and physiological differences have been attributed to the dorsal and ventral zones
The lateral vestibular nucleus (LVN) may participate in motor control through several descending pathways, 21 the most direct of them being the vestibulospinal tract.29.30 Besides its input from labyrinthine afferents, it may receive information from the motor cortex and muscles as well. Although activation of motor cortex (area 4) in decerebellated cats was ineffective on LVN discharges,15 microstimulation of cortical motor foci in intact animals, contralateral to LVN and controlling extensor muscles, evoked responses, some of short latency, in more than 20% of LVN neurons.22 Whether the ipsilateral motor cortical foci and these controlling flexor muscles influence Abbreviations: CFD, cumulative frequency distribution; ECx, site in motor cortex controlling extensor muscles; EM, extensor muscle; EMG, electromyographic; FCx, site in motor cortex controlling flexor muscles; FM, flexor muscle; LVN, lateral vestibular nucleus; PSTH, poststimulus time histogram; VS, vestibulospinal. 379
F. LICATAet al.
380
*fthe nuc~eus,'9.26.2?.30.38 the percentages and the types of responses in each nuclear division were examined separately. This study was performed on rats and provides a comparison with preliminary studies previously performed on the cat.22
EXPERIMENTAL PROCEDURES
Animal preparation
The experiments were carried out on 32 male Wistar rats anaesthetized by injection of Ketamine hydrochloride (Ketaset: 100 mg/kg ip.). The skull was exposed and small holes drilled in the surface overlying the motor cortex (bilateral), the LVN, the lateral vestibulospinal tract at bulbar level (see later) and the ipsilateral cervical cord (in five experiments). To prevent the exposed surfaces from drying, a 3% agar in saline solution was placed on the cerebellar surface and the motor cortex. The wounds were treated by local anaesthetic (Carboicaicana, Pierre]). Blood pressure and temperature were continuously monitored and supplementary anaesthetic was provided when necessary. Recordings
and stimulations
Glass micropipettes (resistance: 558 Ma) filled with I M KCI containing 4% Pontamine Sky Blue, were lowered with a microdrive as far as the LVN (stereotaxic coordinates: P: 1.3-2.3; L: 1.8-2.6; H: +3.5 to -t-2, cf. Paxinos and Watsot?) and used for recording the unitary activity of LVN neurons extra~llularly. Two ni~kel~hromium unipolar stimulating electrodes were placed under direct vision in the forelimb areai in motor cortex and lowered to the fifth layer (depth: 100&1200 pm). In the first five experiments a tungsten electrode (resistance: OS-I MQ) was inserted at cervical level (C3) to stimulate the lateral vestibulospinal (VS) tract. In subsequent experiments we preferred a less traumatic surgical approach. With this aim and taking into account the fact that VS pathways have not been fully investigated in the rat,16 we tried to identify the same tract at bulbar level. A tungsten electrode, inserted at P: 4.54.8, was used to identify a site satisfying the following conditions: (a) a location tentatively compatible with the known data on the position of the VS tract in cat (cf. Veerhart14), monkey (cf. Emmers and Akert14) and man (cf. DeArmond ei al.iZ); (b) responsiveness with an antidromic volley to stimulation of the VS tract at cervical level; (c) the capability of evoking, when stimulated, antidromic activation of all (and only) those LVN units which displayed antidromic firing upon VS stimulation at cervical (C3) level. A site having such characteristics was identified at the follo~ng stereotaxic coordinates: P: 4.54.8, L: I-1.3, H: + I to + 1.3 (cf. Paxinos and Watso#) and tested on 31 LVN neurons, 19 of them antidromically fired by both stimulation sites (spinal and bulbar) and I2 by neither of them. The sites of bulbar and cervical s&imulation are shown in the camera lucida drawings of Fig. I The same figure shows that during a penetration at P: 2.3, three of the units histologically located in LVN were antidromically fired by stimulation of each of the two points whereas on the remaining two, located in the border of LVN, stimulation of both points evoked no antidromic spikes. In all the following experiments this bulbar site was stimulated to identify vestibulospinal LVN units. A collision test with the spontaneous discharge was routinely used. In the preliminary phase of each experiment, 50 ms trains of pulses (frequency: 300 Hz, duration: 250 pis, intensity
Neuroscience Vol. 34, No. 2, pp. 379-390, 1990 Printed in Great Britain
EFFECTS OF MOTOR CORTEX AND SINGLE MUSCLE STIMULATION ON NEURONS OF THE LATERAL VESTIBULAR NUCLEUS IN THE RAT F. Istituto
LICATA,
G. LI VOLSI,
di Fisiologia
umana,
G.
MAUGERI
Viale Andrea
and
Doria
F. SANTANCELO 6, 95125 Catania,
Italy
Abstract-The neuronal responses to stimulation of motor cortical sites and of forelimb single muscles were studied in the lateral vestibular nucleus of anaesthetized rats. Of the 228 neurons tested for response to stimulation of contralateral motor cortex, 63% responded to cortical sites controlling extensor muscles and 30% to those controlling flexors. The corresponding figures for responders to ipsilateral stimulation were 34 and 21%. Vestibulospinal units responded to cortical sites controlling extensor and flexor muscles whereas the remaining lateral vestibular nucleus neurons, very reactive to cortical sites controlling extensor muscles, responded little to contralateral and not at all to ipsilateral cortical sites controlling flexor muscles. The effects evoked by contralateral cortical sites controlling extensors varied, those induced by cortical sites controlling flexors were inhibitory in 77% of cases. The responses to ipsilateral motor cortex stimulation differed not so much by cortical sites controlling extensor or flexor muscles as by whether the neuron was in the dorsal or ventral zone of the lateral vestibular nucleus: mixed in the former, all inhibitory in the latter. Of the lateral vestibular nucleus units tested for response to stimulation of ipsilateral or contralateral forelimb distal muscles, only 11% responded. All the vestibulospinal units responsive to muscle stimulation lay in the dorsal zone of the nucleus. The remainder, dorsal or ventral, were not responsive to contralateral muscles. Single lateral vestibular nucleus cells influenced both by ipsilateral muscle and by contralateral motor cortex made up 24% of the pool, vestibulospinal and non-vestibulospinal. They fell into three groups: responsive to one or both structures but responding more strongly to combined stimulation; responsive to each of the two structures but showing a response to combined stimulation not significantly different from that evoked by the cortex alone; responsive only to combined stimulation. The lateral vestibular nucleus units included in these three groups accounted for 29% of those tested for response to extensor muscles and cortical sites controlling extensors and 15% of those tested for response to flexor muscles and cortical sites controlling flexors. Twenty-five per cent of the vestibulospinal neurons responded both to contralateral muscles and to ipsilateral motor cortex stimulation but none of the non-vestibulospinal neurons responded to both. All the responders to both were in the dorsal zone of the lateral vestibular nucleus and responded to extensor stimuli, always in the same way. These results indicate that motor cortex output exerts a major influence on lateral vestibular nucleus discharges, while the muscle afferents have a modulatory influence on the lateral vestibular nucleus responses to cortex. The units more responsible for integration of cortical and peripheral information are the vestibulospinal neurons of the dorsal zone.
LVN neuronal activity to the same extent and, if so, how, has not been tested. The LVN also receives input from spinal inputs originating in neck5 and limb3*4~33~38~39 regions. The major contribution from limbs seems to come from skin and joint afferents, those of muscle and, specifically, primary afferents, contributing little.39 The aim of the present research was: (a) to study the characteristics of motor cortical control over LVN neurons; (b) to compare it with the direct muscle influence on single LVN neurons and to find out whether any integration of cortex and muscle information takes place in the nucleus. It might be useful to know whether the muscle information delivered to LVN is segregated from the motor cortical input or, converging on the same cells, helps to modify the LVN responses to cortical volleys. As some anatomical and physiological differences have been attributed to the dorsal and ventral zones
The lateral vestibular nucleus (LVN) may participate in motor control through several descending pathways, 21 the most direct of them being the vestibulospinal tract.29.30 Besides its input from labyrinthine afferents, it may receive information from the motor cortex and muscles as well. Although activation of motor cortex (area 4) in decerebellated cats was ineffective on LVN discharges,15 microstimulation of cortical motor foci in intact animals, contralateral to LVN and controlling extensor muscles, evoked responses, some of short latency, in more than 20% of LVN neurons.22 Whether the ipsilateral motor cortical foci and these controlling flexor muscles influence Abbreviations: CFD, cumulative frequency distribution; ECx, site in motor cortex controlling extensor muscles; EM, extensor muscle; EMG, electromyographic; FCx, site in motor cortex controlling flexor muscles; FM, flexor muscle; LVN, lateral vestibular nucleus; PSTH, poststimulus time histogram; VS, vestibulospinal. 379
F. LICATAet al.
380
*fthe nuc~eus,'9.26.2?.30.38 the percentages and the types of responses in each nuclear division were examined separately. This study was performed on rats and provides a comparison with preliminary studies previously performed on the cat.22
EXPERIMENTAL PROCEDURES
Animal preparation
The experiments were carried out on 32 male Wistar rats anaesthetized by injection of Ketamine hydrochloride (Ketaset: 100 mg/kg ip.). The skull was exposed and small holes drilled in the surface overlying the motor cortex (bilateral), the LVN, the lateral vestibulospinal tract at bulbar level (see later) and the ipsilateral cervical cord (in five experiments). To prevent the exposed surfaces from drying, a 3% agar in saline solution was placed on the cerebellar surface and the motor cortex. The wounds were treated by local anaesthetic (Carboicaicana, Pierre]). Blood pressure and temperature were continuously monitored and supplementary anaesthetic was provided when necessary. Recordings
and stimulations
Glass micropipettes (resistance: 558 Ma) filled with I M KCI containing 4% Pontamine Sky Blue, were lowered with a microdrive as far as the LVN (stereotaxic coordinates: P: 1.3-2.3; L: 1.8-2.6; H: +3.5 to -t-2, cf. Paxinos and Watsot?) and used for recording the unitary activity of LVN neurons extra~llularly. Two ni~kel~hromium unipolar stimulating electrodes were placed under direct vision in the forelimb areai in motor cortex and lowered to the fifth layer (depth: 100&1200 pm). In the first five experiments a tungsten electrode (resistance: OS-I MQ) was inserted at cervical level (C3) to stimulate the lateral vestibulospinal (VS) tract. In subsequent experiments we preferred a less traumatic surgical approach. With this aim and taking into account the fact that VS pathways have not been fully investigated in the rat,16 we tried to identify the same tract at bulbar level. A tungsten electrode, inserted at P: 4.54.8, was used to identify a site satisfying the following conditions: (a) a location tentatively compatible with the known data on the position of the VS tract in cat (cf. Veerhart14), monkey (cf. Emmers and Akert14) and man (cf. DeArmond ei al.iZ); (b) responsiveness with an antidromic volley to stimulation of the VS tract at cervical level; (c) the capability of evoking, when stimulated, antidromic activation of all (and only) those LVN units which displayed antidromic firing upon VS stimulation at cervical (C3) level. A site having such characteristics was identified at the follo~ng stereotaxic coordinates: P: 4.54.8, L: I-1.3, H: + I to + 1.3 (cf. Paxinos and Watso#) and tested on 31 LVN neurons, 19 of them antidromically fired by both stimulation sites (spinal and bulbar) and I2 by neither of them. The sites of bulbar and cervical s&imulation are shown in the camera lucida drawings of Fig. I The same figure shows that during a penetration at P: 2.3, three of the units histologically located in LVN were antidromically fired by stimulation of each of the two points whereas on the remaining two, located in the border of LVN, stimulation of both points evoked no antidromic spikes. In all the following experiments this bulbar site was stimulated to identify vestibulospinal LVN units. A collision test with the spontaneous discharge was routinely used. In the preliminary phase of each experiment, 50 ms trains of pulses (frequency: 300 Hz, duration: 250 pis, intensity
