The activity of primate ventrolateral thalamic neurones during motor adaptation.

Three monkeys were trained to perform stereotyped wrist movements to track a target (phase 1). Changing the gain between the wrist movement and visual display required the monkey to adapt its wrist movement. This adaptation consisted of progressive reduction of movement amplitude over a number of trials (phase 2) until a stereotyped movement accommodating the new gain was learned (phase 3). The experiment's aim was to investigate whether cerebellar thalamic neuronal discharge (ND) changed during motor adaptation and whether this change was related to scaling of kinematic parameters or movement error. Extracellular single-cell recordings were made from "wrist-related" neurones in the cerebellar thalamus (59) and the nucleus ventro-posterior lateralis caudalis (VPLc) (37) of each monkey while they performed the movement paradigm. Neurones were selected for further analysis (37/59 cerebellar thalamic and 23/37 VPLc) if phase-1 movements were stereotyped and motor adaptation occurred in phase 2 (according to statistical definitions). When the gain initially changed, there were positional errors in the form of overshoot. Adaptation to the new gain was achieved by a variety of strategies, including modification of the amplitude of kinematic parameters and positional error in addition to reduction of time to peak velocity and movement time. During stereotyped movements, most cerebellar thalamic neurones fired before movement onset and before VPLc neurones. During adaptation, this order of onset of firing was reversed, and cerebellar thalamic neurones discharged after VPLc neurones and close to the onset of movement. During motor adaptation, the mean rate of phasic ND rose in a large proportion of cerebellar thalamic and VPLc neurones, and the proportion of cerebellar thalamic neurones that encoded a signal about positional error and movement amplitude also increased. In addition, there is set-related activity in the discharge of a majority of cerebellar thalamic and VPLc neurones. This does not appear to be specifically related to motor adaptation, but is related to the movement amplitude. We have discussed the role of the cerebello-thalamo-cortical pathway in error detection in the light of the similarities between discharge patterns of cerebellar thalamic and VPLc neurones. We speculate that, when learned movements are performed, the discharge of cerebellar thalamic neurones occurs before movement, perhaps representing an efference copy of the intended movement. During adaptation, this signal is gated out, and later-arriving peripheral afferent input dominates cerebellar thalamic discharge.

Study of projections from the entopeduncular nucleus to the thalamus of the rat.

The entopeduncular nucleus (EP) is a major outflow nucleus of the basal ganglia and innervates the lateral habenula, parafascicular, pedunculopontine, ventrolateral (VL), ventromedial (VM), and mediodorsal thalamic nuclei. This study investigated the morphology of single axons of entopeduncular neurons projecting to the motor thalamus by placing small injections of dextran biotin into the EP and reconstructing drawings of single axons from serial sections. There were two populations of entopeduncular-thalamic projection axons: those that projected only to the motor thalamus (VL and VM) and those that projected to both the motor thalamus and other nuclei (e.g., the habenula). The neurochemistry of EP neurons projecting to the thalamus was investigated by injecting the retrograde tracer FluoroGold into the VL and VM thalamic nuclei to retrogradely fill entopeduncular projection neurons. These were subsequently immunohistochemically labeled for choline acetyl transferase, gamma-aminobutyric acid (GABA), and glutamate. Consistent with previous studies, significant proportions of these neurons were GABA immunoreactive. In addition, approximately half of the entopeduncular-thalamic projecting neurons were found to be cholinergic. This excitatory input is most likely derived from axons that branch as they pass through the motor thalamus to the lateral habenula.

Long-term potentiation across rat cerebello-thalamic synapses in vitro.

We previously demonstrated that rat cerebello-thalamic synapses undergo an ultrastructural change, consistent with expression of long-term potentiation, in association with motor adaptation. The aim of these experiments was to determine if long-term potentiation can be expressed at these synapses physiologically. Excitatory post-synaptic potentials evoked by electrical stimulation of cut cerebello-thalamic afferents were recorded intracellularly from thalamic neurons in 17-24 day old rat brain slice preparations. The experimental protocol involved long periods of low frequency single-shock control stimuli interrupted by brief high frequency trains of conditioning stimuli. Conditioning at 100 Hz evoked long-term potentiation in 5/6 cells and long-term depression in 1/6 cells. Conditioning at 200 Hz was unsuccessful in evoking long-term potentiation but did evoke long-term depression in 1/3 cells and no change in 2/3 cells. We conclude that long-term potentiation can be evoked across rat cerebello-thalamic synapses in vitro.

Ultrastructural change at rat cerebellothalamic synapses associated with volitional motor adaptation.

Our ability to develop or modify motor skills is thought to involve persistent changes in the efficacy of synaptic transmission (synaptic plasticity) in the cerebellum. Previous work from our laboratory and others, examining synapses between neurons in the deep cerebellar nuclei and neurons in the thalamus revealed ultrastructural characteristics that have been implicated in the expression of synaptic plasticity at other locations in the brain. The present study sought evidence of ultrastructural plasticity at cerebellothalamic synapses associated with volitional motor adaptation. Adult rats were subject to 21 days of training, throughout which a novel load (overcome by predominantly shoulder adduction) was applied to the left forelimb while they fed (the right forelimb acted as an internal control). The behavioral paradigm was observed to produce a profound unilateral motor adaptation that was complete by day 15. Three days before the end of training, intracortical microstimulation was performed to identify the regions of primary motor cortex responsible for execution of shoulder adduction movements on the experimental (right) and control (left) sides of the brain. A retrograde neuronal tracer was injected into these cortical regions and the animals were returned to the training cage. Following training, small blocks of thalamic tissue containing retrogradely labeled cells were removed from the brains for ultrastructural analyses of presumed cerebellothalamic synapses (see Materials and Methods section). The only ultrastructural change observed to occur in association with the volitional motor adaptation was an increase in the proportion of dendritic shaft active zone with docked synaptic vesicles.

Arborisation and termination of single motor thalamocortical axons in the rat.

The aim of this study was to examine the arborisations and terminations of individual thalamocortical axons in the motor system of the rat. Small, extracellular injections of an anterograde tracer (dextran-biotin) were made into the ventrolateral (VL) or ventral posterolateral (VPL) thalamic nuclei to label thalamocortical projections. Eleven motor axons and one somatosensory axon were reconstructed through serial sections just rostral from the injection site to their terminations in sensorimotor cortex. The smallest arbor arising from a single motor axon extended approximately 0.9 mm rostrocaudal and 0.9 mm mediolateral, the largest extended 3.9 mm rostrocaudal and 1.0 mm mediolateral. In some cases, two distinct plexuses of terminals were formed by an axon. In addition, motor axons formed terminals in cortical layer V only or in layers I, III, and V. By contrast (and in keeping with previous reports), the somatosensory axon formed a single plexus of terminals in layer IV of the cortex that extended approximately 0.3 mm rostrocaudal and 0.4 mm mediolateral. It is concluded that individual motor thalamocortical neurones are in a position to influence much more widespread cortical regions than somatosensory thalamocortical neurones.

Neuronal activity in the monkey ventrolateral thalamus following perturbations of voluntary wrist movements.

Extracellular single-cell recordings were made from the cerebellar thalamus (89 neurones) and the VPLc (53 neurones) of three conscious monkeys. The animals were trained to perform wrist movement paradigms including: (a) visually triggered skilled, voluntary movements; (b) 100-ms duration torque pulse perturbations applied during a hold period (termed Pa perturbations); (c) 100-ms perturbations that commenced 100 ms after the visual trigger but during preparation before a skilled, voluntary movement (termed Pb perturbations); and (d) 100-ms perturbations during the skilled, voluntary movement (termed Pm perturbations). These Pb and Pm perturbations were used to identify central and peripheral influences on patterns of neuronal discharge in the ventrolateral thalamus. There was no systematic difference between the responses to Pb and Pm perturbations of neurones in the cerebellar thalamus and those in VPLc. The responses of VPLc and cerebellar thalamic neurones to Pa perturbations were considered to represent transduction of peripheral afferent input, and these responses were compared with the responses to the other types of perturbations. Up to 40% of neurones in cerebellar thalamus and VPLc responded to Pb and Pm perturbations in a similar pattern to that which followed Pa perturbations, and therefore most likely represented faithful transduction of peripheral input. However, the response of over half the neurones in VPLc and cerebellar thalamus to Pb or Pm perturbations differed from Pa perturbations in a manner suggesting that central influences had gated the peripheral input. The short-latency response in cerebellar thalamus which was modified by central influences is appropriately timed to contribute to the "intended" response to perturbations of motor cortical neurones.

A comparison of methods used to detect changes in neuronal discharge patterns.

The discharge pattern of two thalamic neurones was recorded from a conscious monkey performing voluntary movements about the wrist joint. The neuronal discharge was displayed as a raster centred on movement of the wrist. The discharge patterns of both neurones was very strongly correlated with movement. Three experienced researchers were asked to examine the data and to classify every part of each trial as background discharge, 'on' (increased firing rate) or 'off' (decreased or zero firing rate) and to mark the times that neuronal discharge changed state. A 'standard output' was made from these classifications. A back-propagation artificial Neural Network (the Network) was used to model the standard output and cumulative sums (CUSUMs) and maximum likelihood was then performed on the data and compared with the Network. There was a high correlation between the output of each observer (r > 0.61) and the standard output and between the Network and the standard output (r > 0.99). However the correlation between standard output and CUSUMs (r = 0.06) and standard output and maximum likelihood (r = 0.36) was much lower. The Network could be trained with as few as 12 trials, indicating a high degree of constancy in the methods employed by the observers. The Network was also highly efficient at detecting changes in state of neuronal activity (r > 0.99). In summary, when used on single trial data, visual inspection is a reliable method for detecting timing of change neuronal discharge and is superior to CUSUM and maximum likelihood. As well it is capable of detecting neuronal discharge state: that is whether firing rate is increased, normal or decreased. Neural Networks promise to be a useful method of confirming the consistency of visual inspection as a means of detecting changes in neuronal discharge pattern.

A comparison of the ultrastructure of synapses in the cerebello-rubral and cerebello-thalamic pathways in the rat.

The aim of this study was to compare the ultrastructure of anterogradely labelled cerebellar terminals in the red nucleus (RN), ventrolateral (VL), parafascicular (PF) and central medial (CM) thalamic nuclei, as well as in the zona incerta (ZI). No differences were found in the morphology of synapses in any of the nuclei. Terminals in RN and VL were larger than those in PF, CM and ZI and synapsed proximally. In contrast, synapses in PF, CM and ZI formed mainly on distal dendrites. These findings indicate that cerebellar output neurones (a) form morphologically similar synapses (Gray's type I) on neurones in functionally different nuclei, and (b) form larger, more proximal synapses in RN and VL than in PF, CM and ZI.

The role of the cerebello-thalamo-cortical pathway in skilled movement.

Studies of lesions of the primate cerebellum leave little doubt that the cerebellum is necessary for the execution of smooth and accurate movements. How the cerebellum fulfills this role at a neuronal level remains unknown. It is likely that the cerebellum exerts the same effect on a number of different efferent targets. In order to influence voluntary movement, a major output from the cerebellum projects to the motor cortex via the cerebello-thalamo-cortical (CTC) pathway. By examining neuronal activity in the cerebellar thalamus, and comparing this with activity recorded from its connections with the deep cerebellar nuclei and motor cortex, conclusions can be made regarding cerebellar function. Current data does not support a role for the CTC pathway in the initiation of movement or the control of trans-cortical reflexes. Also, the evidence does not support the hypothesis that the cerebellum prevents terminal movement oscillations by predictively sending a message to the antagonist muscle to brake the movement. The available literature supports the Eccles theory that during normal movement, the CTC pathway receives a form of efference copy from the motor cortex and compares this message with that derived from peripheral afferents about the actual progress of the movement. However, there is not a significant degree of kinematic information passing through this pathway in the course of a voluntary movement. Therefore the actual site of comparison or error-detection in this system awaits further elucidation.

Projections from the lateral and interposed cerebellar nuclei to the thalamus of the rat: a light and electron microscopic study using single and double anterograde labelling.

The lateral and interposed cerebellar nuclei may have different functions in the control of movement. Efferent fibres from both nuclei project predominantly to areas of the thalamus, which in turn project to the motor cortex. In this study, single and double anterograde-tracing techniques have been used to examine and compare the pathways from the lateral and interposed nuclei to the thalamus in the rat by using both light and electron microscopy to look for evidence of organisational or structural features that may underlie the proposed functional differences between these nuclei. Terminals from the lateral nucleus were found to be located most medially in the thalamus, predominantly in the ventral lateral nucleus and the rostral pole of the posterior nuclear group. Terminals from the posterior interposed nucleus were located slightly rostral and lateral to those from the lateral nucleus, mainly around the border between the ventral lateral nucleus and the ventral posterior medial nucleus. Terminals from the anterior interposed nucleus were located slightly rostral and lateral to those from the posterior interposed nucleus, predominantly in the rostral pole of the ventral posterior lateral nucleus. Terminals from the lateral and interposed nuclei were also found in double anterograde-tracing experiments to be nonoverlapping in the regions between these main areas of termination. The structure of terminals from the lateral and interposed nuclei, however, as well as their synaptic relationship with thalamic neurones, were found to be similar. The terminals are large and form synapses with proximal dendrites of thalamic neurones. They contained round vesicles and formed multiple synaptic contacts with dendritic shafts, as well as dendritic spines. The findings indicate that information from the lateral and interposed nuclei is processed in separate regions of the thalamus but that the mode of synaptic transfer to thalamic neurones is likely to be similar for the two projections.

Sensory characteristics of monkey thalamic and motor cortex neurones.

1. Extracellular single-cell recordings were made from the cerebellar thalamus, the ventro-posterior lateralis par caudalis (VPLc) and motor cortex of three conscious monkeys. Recordings were made from the thalamus as well as the cortex in two monkeys. In all, recordings were made from the thalamus in four hemispheres and from the motor cortex in four hemispheres. The animals were trained to permit a detailed examination when relaxed. Unexpected perturbations were applied to the wrist. Seventy-seven wrist-related neurones were recorded in the cerebellar thalamus, forty-two neurones from the VPLc and eighty-four neurones in motor cortex. 2. Cerebellar nuclear stimulation was used to physiologically identify thalamic neurones receiving input from the cerebellum. The location of all neurones was verified histologically. 3. The majority of cerebellar thalamic neurones had deep sensory receptive fields related to a single muscle, a group of synergists or a single joint. There was a distinct topographical organization. These fields were similar to sensory fields in motor cortical neurones, but had higher thresholds. 4. VPLc neurones had discrete deep or cutaneous sensory fields, or a combination of these fields, which suggests convergence. VPLc neurones had fields with lower thresholds than cerebellar thalamic neurones. The somatotopically located forelimb area in the VPLc was posterior to and continuous with the forelimb area in the cerebellar thalamus. 5. VPLc neurones responded with a shorter latency to wrist perturbations than did cerebellar thalamic neurones. VPLc neurones with deep sensory fields changed firing significantly earlier than those with cutaneous fields. The VPLc is likely to be the major source of sensory input to the motor cortex, and based on the results of this study we suggest that the VPLc is the thalamic nucleus best placed to transmit short-latency afferent input from the forelimb. 6. The timing of the neuronal discharge of cerebellar thalamic and VPLc cells, which resulted from perturbations of the wrist, was best linked to the duration of movement rather than its amplitude. The cells began firing as soon as the velocity changed sign and continued firing until the sign of the velocity changed again. In subsequent corrective movements neuronal discharge in the VPLc appeared to also encode movement acceleration.

The activity of monkey thalamic and motor cortical neurones in a skilled, ballistic movement.

1. Three monkeys were trained to perform a reaction-time task of the wrist and single-cell recordings were made from the motor cortex (eighty-four cells), ventro-posterior lateralis par caudalis (VPLc) (forty-two cells) and cerebellar thalamus (seventy-seven cells). 2. The majority (43/77, 56%) of cerebellar thalamic neurones fired phasically during movement, whereas in the motor cortex most neurones (53/84, 64%) had a phasic-tonic discharge pattern. Most neurones in both locations discharged in relation to the direction of movement (reciprocal pattern). 3. The cerebellar thalamus is unlike the motor cortex in that it does not usually encode a signal for force or joint position in its discharge. 4. Twenty-two per cent (17/77) of cerebellar thalamic neurones had a period of reduced discharge rate before the phasic burst of activity, and represent a pattern of discharge not seen in motor cortex or VPLc neurones. 5. The onset of phasic activity in the cerebellar thalamus was significantly later (average 94 ms) than in the motor cortex but occurred just before electromyogram (EMG) activity. The phasic activity in the cerebellar thalamus usually ended before the phasic component of motor cortex discharge was completed. 6. Phasic activity in VPLc neurones commenced after the onset of EMG discharge and on average 26 ms after the commencement of movement. Most neurones with deep sensory receptive fields fired with a reciprocal pattern, while neurones with cutaneous fields usually fixed bidirectionally in relation to the task. Almost one-third of neurones signalled force and a similar number had discharge levels that encoded characteristics of the joint position. 7. The duration of discharge of VPLc neurones during the voluntary movement was marginally less than the duration of the movement velocity peak and the VPLc may therefore be signalling the duration of the velocity. Phasic activity in cerebellar thalamic neurones fired for a duration similar to the VPLc neurones, but commenced before the movement. Therefore, if the cerebellar thalamus is carrying information about the duration of the velocity, it does so before the movement starts. 8. The phasic burst of activity in cells of the cerebellar thalamus is timed so that it can contribute to the later component of the phasic burst of motor cortical discharge. Thus we speculate that in skilled, ballistic movements, the cerebellum may provide a response which travels via the cerebellar thalamus and helps to determine the magnitude and duration of the phasic part of cortical discharge.(ABSTRACT TRUNCATED AT 400 WORDS)

A frequency analysis of neuronal activity in monkey thalamus, motor cortex and electromyograms in wrist oscillations.

1. Extracellular recordings were made in three monkeys while recording from neurones in the motor cortex (eighty-four cells), ventro-posterior lateralis pars caudalis (VPLc, forty-two cells) and cerebellar thalamus (seventy-seven cells). 2. This experiment was designed to produce active and reflex movements of varying velocities in order to study the relationship between amplitude of velocity and magnitude of neuronal discharge of thalamic neurones. The active movements were voluntary rapid alternating movements (RAMs) of the wrist and the reflex movements were produced by forcibly oscillating the wrist joint between frequencies of 1 and 7 Hz (forced oscillations). 3. This study was also designed to examine cerebellar influences on a reflex path, namely the transcortical reflex loop. Forced oscillations were predicted to provide circumstances where active damping was required to prevent excessive oscillations in the reflex path. Rapid alternating movements of the wrist were predicted to provide circumstances where oscillations at the natural frequency in that reflex path would support and propagate the movements. 4. Forced oscillations from 1 to 7 Hz produced movements of different velocities. VPLc and cerebellar thalamic neurones discharged in relation to the duration of movement in a particular direction, but their discharge levels were unrelated to the magnitude of the velocity. Motor cortex neurones fired in a pattern which was related to the timing but not the magnitude of the acceleration. 5. In forced oscillations of the wrist the resonant frequency was between 3 and 7 Hz. They may be controlled in part by a transcortical reflex. The cerebellar thalamic neurones did not fire before motor cortex neurones. Therefore, it is unlikely that the cerebello-thalamo-cortical pathway is necessary to damp these potentially unstable oscillations by an effect on antagonist-related cortical neurones. 6. Rapid alternating movements (RAMs) of monkeys' wrists were performed in a stereotyped fashion over a narrow range of frequencies with the greatest displacement in joint angle and peak velocity at the natural frequency of 3-5 Hz. 7. During the performance of RAMs, neuronal discharge modulated sinusoidally in the VPLc, cerebellar thalamus and motor cortex. There was no relationship between velocity and neuronal discharge of the cerebellar thalamic and motor cortical neurones but there did appear to be a relationship between velocity and VPLc neuronal discharge. 8. The onset of electromyogram (EMG) discharge changed earlier than neuronal discharge in the motor cortex and thalamus during the performance of RAMs.(ABSTRACT TRUNCATED AT 400 WORDS)