Global dysrhythmia of cerebro-basal ganglia-cerebellar networks underlies motor tics following striatal disinhibition

Abstract

Motor tics, a cardinal symptom of Tourette syndrome (TS), are hypothesized to arise from abnormalities within cerebro-basal ganglia circuits. Yet noninvasive neuroimaging of TS has previously identified robust activation in the cerebellum. To date, electrophysiological properties of cerebellar activation and its role in basal ganglia-mediated tic expression remain unknown. We performed multisite, multielectrode recordings of single-unit activity and local field potentials from the cerebellum, basal ganglia, and primary motor cortex using a pharmacologic monkey model of motor tics/TS. Following microinjections of bicuculline into the sensorimotor putamen, periodic tics occurred predominantly in the orofacial region, and a sizable number of cerebellar neurons showed phasic changes in activity associated with tic episodes. Specifically, 64% of the recorded cerebellar cortex neurons exhibited increases in activity, and 85% of the dentate nucleus neurons displayed excitatory, inhibitory, or multiphasic responses. Critically, abnormal discharges of cerebellar cortex neurons and excitatory-type dentate neurons mostly preceded behavioral tic onset, indicating their central origins. Latencies of pathological activity in the cerebellum and primary motor cortex substantially overlapped, suggesting that aberrant signals may be traveling along divergent pathways to these structures from the basal ganglia. Furthermore, the occurrence of tic movement was most closely associated with local field potential spikes in the cerebellum and primary motor cortex, implying that these structures may function as a gate to release overt tic movements. These findings indicate that tic-generating networks in basal ganglia mediated tic disorders extend beyond classical cerebro-basal ganglia circuits, leading to global network dysrhythmia including cerebellar circuits.

Figure 1.

Anatomical reconstruction of injection and recording sites. A, Image of coronal section of the left hemisphere from Monkey R, showing gliosis formed by recording electrode trajectories, and microinjection sites that were targeted to dorsolateral putamen AC-2. Pu, Putamen; OT, optic tract. B, Electrolytic marking lesions made before animals were killed, targeted to the right CbllCx and dentate nucleus, at regions that were highly active before tic initiation. Den, Dentate nucleus. C, Outline drawings reconstructed from the coronal sections of the left hemisphere of Monkey R. The projected injection sites are derived from electrophysiological mapping and postmortem reconstructions from each animal overlaid on the same sections and are summarized in Table 1. Circles, Monkey R; stars, Monkey B.

Figure 2.

Tic-related firing of cerebellar neurons. A, Examples of raw data from EMG (top trace) and activity of a single CbllCx neuron (bottom trace) after administration of bicuculline. EMG is recorded from the orofacial region. The large voltage transients are tic events. B, Quantification of the increase in tic-related rate firing expressed as a perievent raster and histogram. Neuronal activity is aligned to tic onset. C, Histogram representing the distribution of response latencies in CbllCx relative to tic onset. D, Examples of a raw neuronal recording of an excitatory dentate nucleus cell associated with tic events. E, Perievent raster and histogram of the cell shown in D. F, Histogram representing the distribution of response latencies of excitatory dentate nucleus neurons. G, Raw neuronal recording in conjunction with EMG activity of a typical inhibitory dentate cell. H, Perievent raster and histogram of the cell shown in G. I, Histogram representing the distribution of response latencies of inhibitory dentate nucleus neurons. Note the late onset of activity relative to tic activity.

Figure 3.

Simultaneous recording of cerebro-basal ganglia activity during tics. A, Examples of raw data from EMG (top trace) and simultaneously recorded single cell activity in putamen, GPe, GPi, and M1 after administration of bicuculline. EMG was recorded from the orofacial region. The large voltage transients are tic events. Str, Striatum. B, Perievent raster and histogram of the putamen cell shown in A. Neuronal activity is aligned to tic onset. C, Histogram representing the distribution of response latencies in putamen relative to tic onset. D, Perievent raster and histogram of the multiphasic GPe cell shown in A. E, Histogram representing the distribution of response latencies of GPe neurons. F, Perievent raster and histogram of the inhibitory GPi cell shown in A. G, Histogram representing the distribution of response latencies of all recorded GPi neurons. Note the early onset of activity. H, Perievent raster and histogram of the M1 cell shown in A. I, Histogram showing the distribution of response latencies for all of the recorded M1 neurons.

Figure 4.

Comparison of neuronal response latency between each recorded node. The cumulative plots of neuronal onset latency relative to EMG onset are made using those neurons that discharged earlier than tic-related EMG onset (M1, 67; CbllCx, 21; dentate, 16; putamen, 57; Gpe, 45; GPi, 25).

Figure 5.

Response latencies of basal ganglia and cerebellar neurons measured relative to pair-recorded M1 neurons. A, Distribution of response latencies of putamen neurons. Negative values indicate that putamen neurons fired before M1 neurons. Note that nearly all activity in the putamen occurred before activity in M1. B, Response latencies of GPe neurons. C, Response latencies of GPi neurons. D, Response latencies of CbllCx neurons. E, Response latencies of dentate nucleus neurons.

Figure 6.

Local field potential recordings during tic states. A, Examples of raw data from EMG (bottom trace) and simultaneously recorded LFPs in putamen (dorsal and ventral regions), GPe, GPi, M1, and CbllCx after administration of bicuculline. The EMG is recorded from the orofacial region. Gray shading shows periods of LFP spike activity in the basal ganglia in the intertic interval; note their absence in M1 and CbllCx. B, Average LFP spike amplitude in the dorsal putamen obtained during tic periods (red trace) and intertic interval (black trace). The inset plot shows statistical comparison of LFP spike shape; note the lack of significant difference between tic-associated LFPs and those in the intertic interval. C, Average LFP spike amplitude in the ventral putamen. Same conventions as in B. D, Average LFP spike amplitude in the GPe. E, Average LFP spike amplitude in the GPi. F, Average LFP spikes in M1. Note the presence of statistical difference in waveform structure between tic state and intertic interval (dashed line in the inset shows the significance level, p = 0.001). G, Average LFP spikes in the CbllCx. Note the presence of statistical difference in waveform structure between tic state and intertic interval.

Figure 7.

Behavioral analysis between intertic and tic periods. A–D, Examples of raw collapsed EMG traces from an injection day when arm and orofacial tics were present. The data are aligned with the onset of tic-related LFP spikes (red trace) and intertic LFP spikes (black trace) in the striatum. Note that there is no detectable EMG voltage deflection in the intertic interval. E, Probability of a spontaneous movement occurring just before and after intertic LFP spikes. Note the lack of consistent probability increases associated with intertic LFP spikes.

Figure 8.

Firing rate analysis between the intertic and tic period. A, Comparison of striatal single-cell activity associated with LFP spikes in the intertic (black trace and raster) and tic (red trace and raster) interval. Data are aligned with striatal LFP spike onset. Note the preservation of the response during the intertic interval. B, An example of spiking activity from the GPe between the two states. Note the partial preservation of the response during the intertic interval, with the early inhibitory peak maintained and the later excitatory peak reduced. C, The preservation of response in the GPi between the two intervals. D, Differential M1 activity between the intertic and tic intervals. E, Differential CbllCx activity between the two intervals. F, Differential dentate nucleus activity between the two intervals.