Striatal mRNA expression patterns underlying peak dose L-DOPA-induced dyskinesia in the 6-OHDA hemiparkinsonian rat

L-DOPA is the primary pharmacological treatment for relief of the motor symptoms of Parkinson's disease (PD). With prolonged treatment (⩾5 years) the majority of patients will develop abnormal involuntary movements as a result of L-DOPA treatment, known as L-DOPA-induced dyskinesia. Understanding the underlying mechanisms of dyskinesia is a crucial step toward developing treatments for this debilitating side effect. We used the 6-hydroxydopamine (6-OHDA) rat model of PD treated with a three-week dosing regimen of L-DOPA plus the dopa decarboxylase inhibitor benserazide (4 mg/kg and 7.5 mg/kgs.c., respectively) to induce dyskinesia in 50% of individuals. We then used RNA-seq to investigate the differences in mRNA expression in the striatum of dyskinetic animals, non-dyskinetic animals, and untreated parkinsonian controls at the peak of dyskinesia expression, 60 min after L-DOPA administration. Overall, 255 genes were differentially expressed; with significant differences in mRNA expression observed between all three groups. In dyskinetic animals 129 genes were more highly expressed and 14 less highly expressed when compared with non-dyskinetic and untreated parkinsonian controls. In L-DOPA treated animals 42 genes were more highly expressed and 95 less highly expressed when compared with untreated parkinsonian controls. Gene set cluster analysis revealed an increase in expression of genes associated with the cytoskeleton and phosphoproteins in dyskinetic animals compared with non-dyskinetic animals, which is consistent with recent studies documenting an increase in synapses in dyskinetic animals. These genes may be potential targets for drugs to ameliorate L-DOPA-induced dyskinesia or as an adjunct treatment to prevent their occurrence.

Phase relationships support a role for coordinated activity in the indirect pathway in organizing slow oscillations in basal ganglia output after loss of dopamine

The goal of the present study was to determine the phase relationships of the slow oscillatory activity that emerges in basal ganglia nuclei in anesthetized rats after dopamine cell lesion in order to gain insight into the passage of this oscillatory activity through the basal ganglia network. Spike train recordings from striatum, subthalamic nucleus (STN), globus pallidus (GP), and substantia nigra pars reticulata (SNpr) were paired with simultaneous local field potential (LFP) recordings from SNpr or motor cortex ipsilateral to a unilateral lesion of substantia nigra dopamine neurons in urethane-anesthetized rats. Dopamine cell lesion induced a striking increase in incidence of slow oscillations (0.3-2.5 Hz) in firing rate in all nuclei. Phase relationships assessed through paired recordings using SNpr LFP as a temporal reference showed that slow oscillatory activity in GP spike trains is predominantly antiphase with oscillations in striatum, and slow oscillatory activity in STN spike trains is in-phase with oscillatory activity in cortex but predominantly antiphase with GP oscillatory activity. Taken together, these results imply that after dopamine cell lesion in urethane-anesthetized rats, increased oscillatory activity in GP spike trains is shaped more by increased phasic inhibitory input from the striatum than by phasic excitatory input from STN. In addition, results show that oscillatory activity in SNpr spike trains is typically antiphase with GP oscillatory activity and in-phase with STN oscillatory activity. While these observations do not rule out additional mechanisms contributing to the emergence of slow oscillations in the basal ganglia after dopamine cell lesion in the anesthetized preparation, they are compatible with 1) increased oscillatory activity in the GP facilitated by an effect of dopamine loss on striatal 'filtering' of slow components of oscillatory cortical input, 2) increased oscillatory activity in STN spike trains supported by convergent antiphase inhibitory and excitatory oscillatory input from GP and cortex, respectively, and 3) increased oscillatory activity in SNpr spike trains organized by convergent antiphase inhibitory and excitatory oscillatory input from GP and STN, respectively.