In vivo evidence for a differential contribution of striatal and nigral D1 and D2 receptors to l-DOPA induced dyskinesia and the accompanying surge of nigral amino acid levels

Abstract

Evidence for an involvement of striatal D1 receptors in levodopa-induced dyskinesia has been presented whereas the contribution of striatal D2 receptors remains controversial. In addition, whether D1 and D2 receptors located in the substantia nigra reticulata shape the response to levodopa remains unknown. We therefore used dual probe microdialysis to unravel the impact of striatal and nigral D1 or D2 receptor blockade on abnormal involuntary movements (AIMs) and striatal output pathways in unilaterally 6-hydroxydopamine lesioned dyskinetic rats. Regional perfusion of D1/D5 (SCH23390) and D2/D3 (raclopride) receptor antagonists was combined with systemic administration of levodopa. Levodopa-induced AIMs coincided with a prolonged surge of GABA and glutamate levels in the substantia nigra reticulata. Intrastriatal SCH23390 attenuated the levodopa-induced AIM scores (~ 50%) and prevented the accompanying neurochemical response whereas raclopride was ineffective. When perfused in the substantia nigra, both antagonists attenuated AIM expression (~ 21–40%). However, only intranigral SCH23390 attenuated levodopa-induced nigral GABA efflux, whereas raclopride reduced basal GABA levels without affecting the response to levodopa. In addition, intranigral raclopride elevated amino acid release in the striatum and revealed a (mild) facilitatory effect of levodopa on striatal glutamate. We conclude that both striatal and nigral D1 receptors play an important role in dyskinesia possibly via modulation of the striato-nigral direct pathway. In addition, the stimulation of nigral D2 receptors contributes to dyskinesia while modulating glutamate and GABA efflux both locally and in the striatum.

Introduction

l-DOPA still represents the most effective treatment for Parkinson's disease (PD), although long-term therapy with l-DOPA is burdened by side-effects such as motor fluctuations and abnormal involuntary movements (dyskinesia; Fabbrini et al., 2007, Nutt, 1990). l-DOPA-induced dyskinesia (LID) results from maladaptive pre- and postsynaptic changes in dopamine (DA) transmission (Cenci, 2007). DA activates 5 receptor subtypes which are classified in the D1-like (D1 and D5) and D2-like (D2, D3 and D4) classes (henceforth D1 and D2), based on structural and pharmacological analogies (Seeman and Van Tol, 1994, Sibley and Monsma, 1992). A wealth of studies has demonstrated a major role for D1 receptors in LID. In particular, unregulated DA release from DA and non-DA neurons causes up-regulation and abnormal trafficking of D1 receptors in striatal neurons (Aubert et al., 2005, Berthet et al., 2009, Konradi et al., 2004), along with abnormal downstream signaling responses (reviewed in Cenci and Konradi, 2010). Altered D1 receptor trafficking leads to a relative enrichment of D1, but not D2 receptors at the plasma membrane in dyskinetic rats (Berthet et al., 2009). These changes are likely to alter the functions of D1-expressing striato-nigral GABAergic neurons (the so-called direct pathway), which monosynaptically inhibit nigro-thalamic output neurons causing thalamic disinhibition and movement initiation (Deniau and Chevalier, 1985). Supporting a pivotal role for D1 receptors in LID, D1 receptor agonists have strong dyskinesiogenic properties, whereas D1 receptor antagonists prevent LID in both nonhuman primate (Grondin et al., 1999) and rat (Lindgren et al., 2009, Monville et al., 2005, Taylor et al., 2005, Westin et al., 2007) models of PD. Consistently, D1 receptor knockout mice are poorly susceptible to LID, while D2 receptor knockout mice do not differ from wild-type controls in this regard (Darmopil et al., 2009). These findings do not however exclude a role of D2 receptors in LID. In fact, D2 receptor agonists precipitate dyskinesia in l-DOPA primed animals, whereas D2 receptor antagonists can attenuate LID (Grondin et al., 1999, Lindgren et al., 2009, Monville et al., 2005, Taylor et al., 2005). Furthermore, striatal overexpression of RGS-9, a GTPase accelerating protein that terminates signaling at D2 receptors, improves LID in macaques and rodent models of PD (Gold et al., 2007). Although most studies thus far have focused on the striatum, it will be important to also consider changes occurring in other areas were D1 and D2 receptors are highly expressed, such as the substantia nigra. Indeed, contralateral turning induced by l-DOPA in unilateral 6-OHDA lesioned rats correlates with the dynamics of DA release in the substantia nigra reticulata (SNr), and can be blocked by local infusion of the D1/D5 receptor antagonist SCH23390 in this brain area (Robertson and Robertson, 1989). Moreover, l-DOPA-treated dyskinetic rats show abnormally large elevations in extracellular DA levels not only in the striatum but also in the SNr (Lindgren et al., 2010). Finally, abnormal oscillatory activity in the theta/alpha band (Meissner et al., 2006) and pronounced microvascular plasticity (Westin et al., 2006) have been detected in the SNr of dyskinetic rats. Further strengthening the contribution of the SNr to LID, we found a temporal correlation between the expression of abnormal involuntary movements (AIMs) and a large elevation of extracellular GABA levels within the SNr of dyskinetic rats (Mela et al., 2007), suggesting that GABA release from striato-nigral neurons is involved in generating LID. This previous study did not however clarify whether the surge in extracellular GABA depended on a stimulation of D1 or D2 receptors by l-DOPA-derived DA, nor did it address the anatomical location from which the effect was generated. In addition, it did not examine whether changes of nigral glutamate (GLU) levels were associated with LID. Indeed, the role of GLU in dyskinesia has long been established (Calabresi et al., 2000, Chase and Oh, 2000) and elevated in vivo GLU levels have been documented in the striatum and SNr of dyskinetic rats under basal conditions but not following l-DOPA (Robelet et al., 2004). Conversely, other studies have reported increased (Dupre et al., 2011) or reduced (Morgese et al., 2009) striatal GLU levels in response to l-DOPA in dyskinetic rats.

The present study was undertaken to investigate the role of striatal and nigral D1 and D2 receptors in LID and the associated changes in GABA and GLU release within the basal ganglia network. This was accomplished using a reverse microdialysis approach in awake rats, whereby D1/D5 and D2/D3 antagonists (SCH23390 and raclopride, respectively) were infused in the dorsolateral striatum (DLS) or SNr while l-DOPA-induced AIMs were monitored.