In the past year, we have explored the specifics of dysfunctional alterations in basal ganglia output in animal models of Parkinsons disease. In particular, we have focused on beta range (12-35 Hz) activity which is enhanced in the subthalamic nucleus of parkinsonian patients and has been hypothesized to be antikinetic in nature. Previous studies used the 1 Hz rhythm dominant in the cortex of the anesthetized rat to develop hypotheses about how loss of dopamine leads to entrainment of basal ganglia output to cortical rhythms. These studies suggested that loss of dopamine affects the ability of the striatum to properly filter input from the cortex, and promotes entrainment of the basal ganglia to cortical rhythms. To further examine this hypothesis, we developed an awake behaving rat model of Parkinsons disease which allows us to investigate changes in basal ganglia activity after loss of dopamine in conjunction with different behavioral states. The results show that two qualitatively different patterns are present in measures of power and coherence in motor cortex and substantia nigra pars reticulata (SNpr) by day 7 after dopamine cell lesion. During inattentive rest, modest but significant increases motor cortex-SNpr LFP coherence are evident over a broad 8 to 40 Hz range coinciding with significant increases in SNpr LFP power in the 8 25 Hz range but no significant increases in motor cortex LFP power. This observation supports the view that loss of dopamine can lead to increased entrainment of basal ganglia output to dominant cortical rhythms in the absence of notable increases in cortical power in the motor cortex. These results are consistent with the hypothesis that loss of dopamine facilitates striatal transmission of cortical input, and with data from recordings in urethane anesthetized rats, where the 1 Hz rhythm that is dominant in the cortex during anesthesia becomes notably more prominent throughout the basal ganglia after dopamine cell lesion, although it is not enhanced in the cortex itself. On the other hand, during epochs when rats are engaged in ongoing motor activity, more dramatic increases in LFP power emerge in the motor cortex as well as the SNpr, focused around peak frequencies in the 30-35 Hz range and accompanied by a notable increase in motor cortex-SNpr coherence. This increase in 30-35 Hz activity coincides with the loss of a modest but significant peak in synchronized activity in the 40-45 Hz range evident in the control/sham hemispheres during treadmill walking. These observations raise questions about how loss of dopamine leads to increases in oscillatory activity in a relatively focused high beta/low gamma band in both motor cortex and SNpr, in the context of ongoing motor activity. To gain insight into this phenomenon, additional experiments investigated the effect of L-DOPA treatment on the strength of this oscillation, and the lag times between coherent oscillations in motor cortex and SNpr in the context of spiking activity in these two areas. L-DOPA administration reversed the changes associated with dopamine cell lesion, confirming the critical role of tonic loss of dopamine receptor simulation and, in particular, D2 receptor stimulation, in mediating the increases in LFP and coherence in motor cortex and SNpr associated with loss of dopamine. To gain insight into the spiking activity that supports these prominent LFP oscillations, we next examined the directionality and timing of transmission of the 30 - 35 Hz LFP oscillations between cortex and SNpr. Anatomical considerations of connections between the motor cortex and basal ganglia have, historically, led to a focus on the flow of information from cortex to the striatum, and from the striatum to the SNpr via the direct and indirect pathways, with the latter involving synapses in the external globus pallidus and the subthalamic nucleus on the way to the SNpr. Another route, the hyperdirect pathway, involves projections from the cortex to the STN, and from the STN to the SNpr. These pathways would be predicted to produce a lag time between cortical output and SNpr activation sufficient for transmission across 2 - 4 synapses. Studies involving electrical stimulation of the cortex in conjunction with recordings in the SNpr have reported lag times in the range of 10 15 ms. It is generally believed, at least as an initial assumption in thinking about spike-LFP relationships, that the troughs of the LFP oscillations in a given nucleus reflect peak depolarization of the neurons surrounding the recording electrode, and thus also reflect the time when these neurons are most likely to be spiking. Therefore, as LFP or EEG activity is a readily available marker of synchronized spiking, determination of the lag time between peaks of LFPs recorded in two different areas is typically the starting point for considering timing relationships between synchronized spiking in the two areas. Lag times between coherent LFP oscillations in motor cortex and SNpr during walking epochs were calculated as the shift at maximum cross correlation between the two LFP signals filtered at 25 40 Hz. The results show a strong effect of loss of dopamine on lag times between LFP oscillations in the 25 40 Hz range. However, lag times between motor cortex and Snpr were very short and showed inconsistent directionality of information flow between the two sites during walking epochs in the dopamine-depleted hemisphere. This data was not consistent with predictions of information flow from motor cortex to SNpr output and anatomical considerations of cortico-basal ganglia connectivity. To further clarify the significance of the lag times observed in the LFP cross correlation analysis, spike triggered LFP waveform averages (STWAs) were used to assess the degree of phase-locking of individual SNpr and motor cortex pyramidal neurons and interneurons with LFP activity in the 12-18 Hz and 25 40 Hz ranges during rest and treadmill walking in the hemiparkinsonian rats. A second approach involved comparing the relative timing of spiking activity of interneurons and pyramidal in the cortex with timing of spiking in SNpr for those spike trains showing significant correlation with the dominant LFP oscillation in the 25 40 Hz range during walking. Data showed that during treadmill walking epochs, in the presence of exaggerated high beta/low gamma oscillations, SNpr neurons fired relatively consistently during the downslope of the LFP oscillation, clustering around 90 degrees, with respect to the peak (0 degrees) of the SNpr (and motor cortex) LFP oscillation. In contrast, neurons in layer 5/6 of the motor cortex fired later in the oscillatory cycle, with peak activity of both pyramidal neurons and interneurons occurring on the rise of the LFP oscillation, around 270 degrees. Data makes several interesting points. First, the use of the troughs of the LFPs as a measure of timing of neuronal spiking is misleading with respect to this data set. Although motor cortex and SNpr LPFs had close to zero phase lag, the time of the spiking of the SNpr neurons and the cortical pyramidal neurons were off-set by about half the period of the main oscillation. This could support a scenario wherein the dominant frequency of this oscillations is determined by the time required to allow passage of a pulse of activity through the motor cortex basal ganglia- thalamocortical loop, although other scenarios should be considered, including generation of the 30 35 Hz rhythm within the cortical network and/or within subsets of basal ganglia-thalamic networks. Future studies will further explore the underlying mechanisms and functional consequences of the abnormal rhythms emerging in the basal ganglia in the rodent model of Parkinson's disease.
Showing the most recent 10 out of 13 publications
