Molecular adaptations of striatal spiny projection neurons during levodopa-induced dyskinesia

Abstract

Levodopa treatment is the major pharmacotherapy for Parkinson's disease. However, almost all patients receiving levodopa eventually develop debilitating involuntary movements (dyskinesia). Although it is known that striatal spiny projection neurons (SPNs) are involved in the genesis of this movement disorder, the molecular basis of dyskinesia is not understood. In this study, we identify distinct cell-type-specific gene-expression changes that occur in subclasses of SPNs upon induction of a parkinsonian lesion followed by chronic levodopa treatment. We identify several hundred genes, the expression of which is correlated with levodopa dose, many of which are under the control of activator protein-1 and ERK signaling. Despite homeostatic adaptations involving several signaling modulators, activator protein-1-dependent gene expression remains highly dysregulated in direct pathway SPNs upon chronic levodopa treatment. We also discuss which molecular pathways are most likely to dampen abnormal dopaminoceptive signaling in spiny projection neurons, hence providing potential targets for antidyskinetic treatments in Parkinson's disease.

Fig. 1.

Genome-wide heat-map of statistically significant expression changes over experimental contrasts. Contrasts were defined by cell type (Drd1a dSPNs, Drd2 iSPNs) and dose regimen (dopamine depletion only, chronic low-dose levodopa, or chronic high-dose levodopa). A single representative probe-set is shown for each gene. Color indicates direction of and magnitude of log₂ fold-change between mean expression of each probe-set in each treatment group, compared with the appropriate matched control group (red, increased; blue, decreased; white, no significant change). Genes were filtered for statistically significant (Benjamini–Hochberg adjusted P value ≤ 0.10) changes and sorted by hierarchical clustering using a cosine distance metric.

Fig. 2.

Genome-wide analysis of gene expression changes induced by dopamine depletion and levodopa treatment. (A) Venn diagrams showing the total numbers of genes changing across treatments in Drd1a (dSPN) cells (Left) and Drd2 (iSPN) cells (Right) for statistically significant changes (Benjamini–Hochberg adjusted P value < 0.10) of 1.5-fold or greater. (B) Venn diagrams comparing the numbers of genes up-regulated and down-regulated by each treatment between Drd1a (dSPN) and Drd2a (iSPN) cells for statistically significant changes (Benjamini–Hochberg adjusted P value < 0.10) of 1.5-fold or greater.

Fig. 3.

Representative examples of dose-dependent, AIM-correlated probe-sets with different patterns of responses to dopamine depletion and levodopa treatments in Drd1a dSPNs. (Upper) Scatterplots of total AIM score vs. log₂ gene expression. Each point represents the gene expression measurement and AIM score from a single mouse. Colors indicate treatment groups (see key, Upper Left). (Lower) Box-plots summarizing gene expression across treatment groups. (A) Gpr39 (1432260_at): Expression decreases with dopamine depletion, increases significantly with chronic levodopa treatment, and expression depends on levodopa dose. (B) Fosl1 (1417488_at): Expression is unchanged by dopamine depletion, increases significantly with chronic levodopa, and expression depends on levodopa dose. (C) Ier3 (1419647_a_at): Expression is unchanged by dopamine depletion, increases dramatically with chronic levodopa treatment, and depends on levodopa dose. (D) Itch (1415769_at): Expression is unchanged by dopamine depletion, decreases significantly with levodopa treatment, and depends on levodopa dose.