Dopamine replacement with l-DOPA is the most effective therapy in Parkinson disease. However, with chronic treatment, half of the patients develop an abnormal motor response including dyskinesias. The specific molecular mechanisms underlying dyskinesias are not fully understood. In this study, we used a well-characterized animal model to first establish the molecular differences between rats that did and did not develop dyskinesias. We then investigated the molecular substrates implicated in the anti-dyskinetic effect of buspirone, a 5HT1A partial agonist. Striatal protein expression profile of dyskinetic animals revealed increased levels of the dopamine receptor (DR)D3, ΔFosB and phospho (p)CREB, as well as an over-activation of the DRD1 signalling pathway, reflected by elevated ratios of phosphorylated DARPP32 and ERK2. Buspirone reduced the abnormal involuntary motor response in dyskinetic rats in a dose-dependent fashion. Buspirone (4 mg/kg) dramatically reduced the presence and severity of dyskinesias (by 83%) and normalized DARPP32 and ERK2 phosphorylation ratios, while the increases in DRD3, ΔFosB and pCREB observed in dyskinetic rats were not modified. Pharmacological experiments combining buspirone with 5HT1A and DRD3 antagonists confirmed that normalization of both pDARPP32 and pERK2 is required, but not sufficient, for blocking dyskinesias. The correlation between pDARPP32 ratio and dyskinesias was significant but not strong, pointing to the involvement of convergent factors and signalling pathways. Our results suggest that in dyskinetic rats DRD3 striatal over-expression could be instrumental in the activation of DRD1-downstream signalling and demonstrate that the anti-dyskinetic effect of buspirone in this model is correlated with DRD1 pathway normalization.