Contrasting gene expression patterns induced by levodopa and pramipexole treatments in the rat model of Parkinson's disease.

Whether the treatment of Parkinson's disease has to be initiated with levodopa or a D2 agonist like pramipexole remains debatable. Levodopa is more potent against symptoms than D2 agonists, but D2 agonists are less prone to induce motor complications and may have neuroprotective effects. Although regulation of plastic changes in striatal circuits may be the key to their different therapeutic potential, the gene expression patterns induced by de novo treatments with levodopa or D2 agonists are currently unknown. By studying the whole striatal transcriptome in a rodent model of early stage Parkinson's disease, we have identified the gene expression patterns underlying therapeutically comparable chronic treatments with levodopa or pramipexole. Despite the overall relatively small size of mRNA expression changes at the level of individual transcripts, our data show a robust and complete segregation of the transcript expression patterns induced by both treatments. Moreover, transcripts related to oxidative metabolism and mitochondrial function were enriched in levodopa-treated compared to vehicle-treated and pramipexole-treated animals, whereas transcripts related to olfactory transduction pathways were enriched in both treatment groups compared to vehicle-treated animals. Thus, our data reveal the plasticity of genetic striatal networks possibly contributing to the therapeutic effects of the most common initial treatments for Parkinson's disease, suggesting a role for oxidative stress in the long term complications induced by levodopa and identifying previously overlooked signaling cascades as potentially new therapeutic targets.

Cabergoline and pramipexole fail to modify already established dyskinesias in an animal model of parkinsonism.

Levodopa-induced dyskinesias are one of the major limiting side effects encountered in the treatment of Parkinson's disease. Dopamine agonists of the D2 family are less prone to induce these abnormal involuntary movements (AIMs), and in some instances it has been proposed that they could counteract them once already established. As differences in the plasma half-life of a given DA agonist could be related with a greater or lesser propensity to induce or to counteract AIMs, we compared the effects of two D2 agonists (cabergoline and pramipexole) with different half-lives, and levodopa, at doses producing similar improvement in purposeful forelimb use, in rats with severe nigrostriatal lesion, previously sensitized to levodopa. The same therapeutic regime was subsequently used in pharmacologically naïve rats. We found that: (i) prior induction of AIMs by levodopa administration primes rats for the occurrence of AIMs during mono-therapy with pramipexole (but not with cabergoline); (ii) an intervening period of D2 agonist mono-therapy does not modify the severity of AIMs induced by subsequent mono-therapy with levodopa; iii. de novo treatment with D2 agonists is associated with a lower risk of AIMs (regardless of the severity of the lesion) and does not modify AIMs during subsequent mono-therapy with levodopa. An unexpected finding was that prior levodopa therapy sensitized rats to the therapeutic effects of D2 agonists given in mono-therapy. In summary, the use of the rat with nigrostriatal lesion to model relevant therapeutic conditions does not support that D2 agonists prevent the development of AIMs during subsequent levodopa mono-therapy or can revert the dysfunction underlying it.

Distinct changes in evoked and resting globus pallidus activity in early and late Parkinson's disease experimental models.

The main clinical manifestations of Parkinson's disease are caused by alterations of basal ganglia activity that are tied in with the progressive loss of mesencephalic dopaminergic neurons. Recent theoretical and modeling studies have suggested that changes in resting neuronal activity occurred later in the course of the disease than those evoked by phasic cortical input. However, there is no empirical support for this proposal. Here we report a marked increase in the responsiveness of globus pallidus neurons to electrical motor cortex stimulation, in the absence of noticeable changes in resting activity, in anesthetized rats that had consistently shown a deficit in forelimb use during behavioral testing before the experiments, and had approximately 45% dopamine neurons spared in the substantia nigra. Pallidal neurons were also over-responsive to motor cortex stimulation and lost spatial selectivity for cortical inputs in rats with extensive nigrostriatal damage. After partial lesions, over-responsiveness was mainly due to an increased proportion of neurons showing excitatory responses, while extensive lesions led to an increased likelihood of inhibitory responding neurons. Changes in resting neuronal activity, comprising pauses disrupting tonic discharge, occurred across different global brain states, including an activated condition which shares similarities with natural patterns of cortical activity seen in awake states and rapid eye-movement sleep, but only after massive nigrostriatal degeneration. These results suggest that a loss of functional segregation and an abnormal temporal encoding of phasic cortical inputs by globus pallidus neurons may contribute to inducing early motor impairment in Parkinson's disease.

Mapping the effects of three dopamine agonists with different dyskinetogenic potential and receptor selectivity using pharmacological functional magnetic resonance imaging.

The mechanisms underlying dopamine agonist-induced dyskinesia in Parkinson's disease remain poorly understood. Similar to patients, rats with severe nigrostriatal degeneration induced by 6-hydroxydopamine are more likely to show dyskinesia during chronic treatment with unselective dopamine receptor agonists than with D2 agonists, suggesting that D1 receptor stimulation alone or in conjunction with D2 receptor stimulation increases the chances of experiencing dyskinesia. As a first step towards disclosing drug-induced brain activation in dyskinesia, we examined the effects of dopamine agonists on behavior and blood oxygenation level-dependent (BOLD) signal in the striatum and motor cortex of rats with unilateral nigrostriatal lesions. Rats were rendered dyskinetic before pharmacologic functional magnetic resonance imaging by means of a repeated treatment regime with dopamine agonists. The unselective agonist apomorphine and the selective D1/D5 agonist SKF-81297 induced strong forelimb dyskinesia (FD) and axial dystonia and increased BOLD signal in the denervated striatum. Besides, SKF-81297 produced a significant but smaller BOLD increase in the intact striatum and a symmetric bilateral increase in the motor cortex. The D2 family agonist quinpirole, which induced mild dyskinesia on chronic treatment, did not produce BOLD changes in the striatum or motor cortex. Further evidence to support an association between BOLD changes and dyskinesia comes from a direct correlation between scores of FD and magnitude of drug-induced BOLD increases in the denervated striatum and motor cortex. Our results suggest that striatal and cortical activation induced by stimulation of D1/D5 receptors has a primary role in the induction of peak dose dyskinesia in parkinsonism.