Address correspondence and reprint requests to M. Angela Cenci, Basal Ganglia Pathophysiology Unit, Department of Experimental Medical Science, Lund University, BMC F11, Lund, Sweden. E-mail: email@example.com
Metabotropic glutamate receptor type 5 (mGluR5) modulates dopamine and glutamate neurotransmission at central synapses. In this study, we addressed the role of mGluR5 in l-DOPA-induced dyskinesia, a movement disorder that is due to abnormal activation of both dopamine and glutamate receptors in the basal ganglia. A selective and potent mGluR5 antagonist, 3-[(2-methyl-1,3-thiazol-4-yl)ethynyl] pyridine, was tested for its ability to modulate molecular, behavioural and neurochemical correlates of dyskinesia in 6-hydroxydopamine-lesioned rats treated with l-DOPA. The compound significantly attenuated the induction of abnormal involuntary movements (AIMs) by chronic l-DOPA treatment at doses that did not interfere with the rat physiological motor activities. These effects were paralleled by an attenuation of molecular changes that are strongly associated with the dyskinesiogenic action of l-DOPA (i.e. up-regulation of prodynorphin mRNA in striatal neurons). Using in vivo microdialysis, we found a temporal correlation between the expression of l-DOPA-induced AIMs and an increased GABA outflow within the substantia nigra pars reticulata. When co-administered with l-DOPA, 3-[(2-methyl-1,3-thiazol-4-yl)ethynyl] pyridine greatly attenuated both the increase in nigral GABA levels and the expression of AIMs. These data demonstrate that mGluR5 antagonism produces strong anti-dyskinetic effects in an animal model of Parkinson’s disease through central inhibition of the molecular and neurochemical underpinnings of l-DOPA-induced dyskinesia.
Dopamine (DA) replacement with l-DOPA remains the most effective pharmacological treatment for Parkinson’s disease (PD). However, after a few years from treatment onset, the majority of PD patients exhibit fluctuations in motor response and dyskinesia [abnormal involuntary movements (AIMs)] (Fahn 2000; Obeso et al. 2000, 2004). These complications result from the combined effects of nigrostriatal DA degeneration and l-DOPA treatment (Chase 1998; Rascol et al. 1998). DA-denervation lowers the threshold for the dyskinetic effect of l-DOPA, whereas plastic changes induced by the treatment itself cause an increasing incidence and severity of dyskinesia over time (Jenner 2000; Cenci and Lundblad 2006). These changes are long-lasting (Andersson et al. 2003) and prime the brain for a dyskinetic motor response to l-DOPA even if the drug treatment is temporarily discontinued. An overactive glutamate transmission in the basal ganglia plays a key role in the maladaptive plasticity at the basis of l-DOPA-induced dyskinesia (LID; for review see (Chase et al. 2000; Bezard et al. 2001; Cenci and Lundblad 2006). Accordingly, ionotropic glutamate receptor antagonists (iGluR) have shown anti-dyskinetic efficacy in some clinical and experimental studies (for review see Brown et al. 2002). However, these drugs may produce significant side effects (Meldrum 1998), which limit their utility in many patients. Metabotropic glutamate receptors (mGluRs) are a family of G protein-coupled receptors classified in three major groups (group I–III) on the basis of their pharmacological properties (for review see Pin and Duvoisin 1995; Nakanishi et al. 1998). Preliminary reports indicate that pharmacological antagonism of mGluR5 can inhibit the expression of dyskinesia in both rodent (Dekundy et al. 2006) and non-human primate models of PD (Hill et al. 2001). Type 5 mGluRs are abundantly expressed in striatal projection neurons (Testa et al. 1994; Kerner et al. 1997), where they modulate excitatory synaptic transmission (Kearney and Albin 1995; Pisani et al. 1997; Ferre et al. 1999; Pintor et al. 2000) and nuclear signalling processes (Mao and Wang 2002, 2003;Voulalas et al. 2005).
In the present study, we set out to determine whether pharmacological blockade of mGluR5 can attenuate dyskinesia and its associated plastic changes in 6-hydroxydopamine (6-OHDA) -lesioned rats treated with l-DOPA. Central blockade of mGluR5 was achieved using a potent and highly selective compound, i.e. 3-[(2-methyl-1,3-thiazol-4-yl)ethynyl]pyridine (MTEP) (Busse et al. 2004). Different doses of MTEP were first screened for their ability to prevent striatal prodynorphin (PDyn) and preproenkephalin (PPE) gene induction by acute l-DOPA treatment (Experiment 1). Thereafter, a selected, effective dose of MTEP was co-administered with l-DOPA in a chronic course of treatment, during which the development of dyskinesia was monitored using a rat AIMs scale (Experiment 2). To obtain a neurochemical correlate of the anti-dyskinetic effect of MTEP, we monitored GABA release in the substantia nigra pars reticulata (SNr) and globus pallidus (GP) in vivo, simultaneously with the rating of AIMs (Experiment 3).
Our results indicate that MTEP can block both the acute expression of abnormal movements and the priming for dyskinesia induced by l-DOPA through central normalization of molecular and neurochemical effects of the treatment.
Materials and methods
The study was performed in young adult Sprague–Dawley rats (Harlan, Stockholm, Sweden; Harlan Italy; S. Giorgio al Natisone, Italy) weighing ∼200 g at the beginning of the experiments. The animals were housed under a 12-h light/dark cycle. The treatment of the animals and their conditions had been approved by the Malmö-Lund Ethical Committee on Animal Research (Experiments 1–2) and by the Ethical Committee of the University of Ferrara and Italian Ministry of Health (licence no. 71-2004-B; (Experiment 3).
Experiment 1. Acute MTEP and l-DOPA co-administration in l-DOPA-naïve rats
Thirty-four rats with unilateral 6-OHDA lesions were allocated to five groups to receive treatment with either vehicle or l-DOPA (10 mg/kg plus 15 mg/kg benserazide, i.p.) alone or combined with three different doses of MTEP (0.625, 1.25 and 6.25 mg/kg, i.p.). The rats were killed 3 h post-injection, and their brain tissue was prepared for in situ hybridization histochemistry.
Experiment 2. Chronic MTEP and l-DOPA co-administration in l-DOPA-naïve rats
Forty rats with unilateral 6-OHDA lesions were allotted to four treatment groups, which received (i) MTEP (5 mg/kg/day, i.p.), (ii) l-DOPA (6 mg/kg/day plus 15 mg/kg/day benserazide, i.p.), (iii) MTEP plus l-DOPA and (iv) vehicle (consisting of benserazide, 15 mg/kg/day, dissolved in the MTEP vehicle). These treatments were given daily for 21 days. During this period, rats were assessed on the following behavioural tests: AIM ratings (9 times), rotarod (5 times) and cylinder test (4 times). At the end of the chronic treatment, all the animals were given a 24-h drug washout period and were then exposed to a challenge dose of l-DOPA (6 mg/kg plus 15 mg/kg benserazide, i.p.), followed by a 3-h AIM rating session. After an additional 2 days, rats were killed and the brains were processed for in situ hybridization histochemistry.
Experiment 3. Acute MTEP and l-DOPA co-administration in rats rendered dyskinetic by a previous course of l-DOPA treatment
Eighteen rats with unilateral 6-OHDA lesions were treated with l-DOPA (6 mg/kg/day plus 15 mg/kg/day benserazide, i.p.) for 21 days and were then sorted in a dyskinetic group (comprising rats with maximally severe AIMs, n = 11) and a non-dyskinetic group (n = 7). Control groups included drug-naïve, intact and lesioned animals. Two days after discontinuation of l-DOPA treatment, microdialysis probes were implanted in the SNr and GP ipsilaterally to the 6-OHDA lesion. One day after probe implantation, GABA release was monitored in the SNr and GP in parallel with the scoring of AIMs after a challenge injection of MTEP and l-DOPA.
l-DOPA methyl ester and Benserazide-HCl (Sigma-Aldrich, AB, Sweden) were dissolved in physiological saline and administered within 1 h at the volume of 1.0 mL/kg body weight. l-DOPA was given 20 min before the start of each AIM-rating session. In experiment 1, l-DOPA was administered at the dose of 10 mg/kg, which can acutely induce a significant up-regulation of PDyn mRNA in the DA-denervated striatum (Andersson et al. 1999, 2001). For chronic drug treatments (experiments 2–3), l-DOPA was administered at the dose of 6 mg/kg/day, which produces a gradual development of dyskinesia when given for 3 weeks (Lundblad et al. 2002). Benserazide was co-administered with l-DOPA at a fixed dose of 15 mg/kg per injection. This dose of benserazide is within the range required to ensure a sufficient duration (>2 h) of the biochemical and motor effects of single l-DOPA doses in rodents (Da Prada et al. 1987; Grange et al. 2001; and own unpublished data). MTEP (synthetized by Merz Pharmaceuticals GmbH, Frankfurt, Germany) was suspended in a solution of 5% Tween 80/water and administered in a final volume of 2.0 mL/kg body weight by i.p. injection. The compound was administered simultaneously with l-DOPA (thereby 20 min before the start of AIM-rating sessions and 45–60 min before the start of rotatod and cylinder test). In experiment 1, we tested three MTEP doses (0.625, 1.25 and 6.25 mg/kg) that are estimated to produce low (∼20–30%), high (∼60–70%) and complete (100%) mGluR5 receptor occupancy in vivo (Anderson et al. 2003; Busse et al. 2004). The dose of MTEP selected for Experiment 2 (5 mg/kg) was chosen based on preliminary behavioural studies (Dekundy et al. 2006), and it is known to produce >90%in vivo receptor occupancy (Busse et al. 2004). All drug-treatments were started 5–6 weeks post-lesion (and 1 week after the last pre-training session on the rotarod test).
Lesion surgery and behavioural screening
Unilateral 6-OHDA lesions of the right ascending DA fibre bundle were performed according to our standard procedure (Cenci et al. 1998; Andersson et al. 1999; Lundblad et al. 2002, 2003; Carta et al. 2006). Briefly, 6-OHDA-HCl was dissolved in 0.02% ascorbate-saline at a concentration of 3 µg free-base 6-OHDA per microlitre and was injected into the right ascending DA fibre bundle at the following co-ordinates (in mm; Paxinos and Watson, 1997) relative to bregma and the dural surface: (i) A = −4.4, L = –1.2 and V = –7.8, tooth bar at −2.4 (7.5 µg deposit); (ii) A = −4.0, L = –0.75 and V = –8.0, tooth bar at +3.4 (6 µg deposit). Animals with good lesions were selected using both a test of forelimb akinesia (cylinder test) and an automated recording of amphetamine-induced rotation (2.5 mg/kg d-amphetamine i.p.; 90 min testing), which were carried out 2–3 weeks post-lesion. All the animals selected for the study showed severe motor impairments, consisting of (i) rates of amphetamine-induced rotation >5 full turns per minute in the direction ipsilateral to the lesion and (ii) <30% contralateral paw usage in the cylinder test (which was <<3 standard deviations below the mean of normal control animals, defining the 99.73% confidence limit). Our previous studies have shown that these behavioural scores correspond to >90% depletion of DA fibre terminals (Winkler et al. 2002; Lundblad et al. 2004) and >95% tissue DA depletion (Carta et al. 2006) on the side of the striatum ipsilateral to the lesion. Cylinder-test scores and amphetamine-induced rotations were then used as balancing criteria for the allocation of animals to different experimental groups in all experiments. Immunohistochemical verification of the lesion extent was carried out in all the animals included in experiments 1 and 2 using an antibody against tyrosine hydroxylase as described in Andersson et al. (1999). All the animals showed a virtually complete disappearance of tyrosine hydroxylase-positive cell bodies and fibres in the substantia nigra pars compacta on the side ipsilateral to the 6-OHDA lesion.
All behavioural testing was carried out between 9.00 a.m. and 5.00 p.m. by an experimentally blinded investigator. The cylinder test (Schallert and Tillerson 2000; modified by Lundblad et al. 2002), evaluates spontaneous forelimb use for weight shifting during vertical exploration. Briefly, each animal was placed in a glass cylinder (21 cm diameter and 34 cm height) and videotaped for 3 min. The number of wall contacts performed independently with the left and the right forepaw were counted and noted down up to a total number of 20 wall contacts per rat per session. Only wall contacts where the animal supported one’s body weight on the paw with extended digits were counted. The data are presented as contralateral (left) paw use (i.e. percentage of left-paw wall contacts over the total number of wall contacts).
Four weeks after the amphetamine-induced rotation test, and 1–2 weeks before the initiation of MTEP-l-DOPA treatment, the animals used in the Experiments 1 and 2 were pre-trained on the rotarod apparatus (Rotamex 4/8; Columbus Instruments, OH, USA) according to the protocol described by Lundblad et al. (2003). By the last pre-training sessions, all animals had reached a stable rotarod performance. In the Experiments 1 and 2, the rotarod test was performed 45–60 min after the administration of l-DOPA and/or MTEP. These testing sessions consisted of 3 × 90 s trials (plus a shorter and easier session in-between trials and after the last trial; see Lundblad et al. 2003). The same acceleration speed and end speeds were used in this part of the study as in the pre-training sessions. The rotarod performance was measured as the total time spent on the rotarod (longer time indicating better performance). The performance after treatment is expressed in this study as a percentage of the control performance (i.e. the stable baseline values achieved in each animal after the pre-training session).
Quantification of the AIMs induced by l-DOPA was carried out as extensively described in previous papers (Lee et al. 2000; Lundblad et al. 2002; Winkler et al. 2002). Briefly, rats were observed individually for 1 min every 20 min during the 3 h that followed an injection of l-DOPA. Dyskinetic movements were classified based on their topographic distribution into four subtypes: (i) axial AIMs, i.e. twisted posture or choreiform twisting of the neck and upper body toward the side contralateral to the lesion; (ii) forelimb AIMs, i.e. jerky or dystonic movements of the contralateral forelimb and/or purposeless grabbing movement of the contralateral paw; (iii) orolingual AIMs, i.e. orofacial muscle twitching, empty masticatory movements and contralateral tongue protrusion; (iv) locomotive AIMs, i.e. increased locomotion with contralateral side bias. Each AIM subtype was rated on a severity scale from 0 to 4 (1 = occasional; 2 = frequent; 3 = continuous but interrupted by external stimuli; 4 = continuous, severe and not interrupted by external stimuli) on each monitoring period. In addition to this scale, which is based on the frequency and persistence of the dyskinetic movements, axial and limb AIMs were also given an amplitude score from 0 to 4 on each monitoring period as described in Winkler et al. (2002). As locomotive AIMs differ greatly from the other three AIM subtypes in terms of pharmacological features and anatomical substrate (Andersson et al. 1999; Lundblad et al. 2002; Winkler et al. 2002), they are presented and analysed as a separate item. By contrast, axial, forelimb and orolingual AIMs are presented together as a global AIM score per session (sum of the products of amplitude and frequency scores from all monitoring periods; (Carta et al. 2006).
In vivo microdialysis
Dual probe microdialysis (Morari et al. 1996; Marti et al. 2002, 2005) was used to monitor GABA release simultaneously in the SNr and GP of awake, freely moving rats. Briefly, microdialysis probes of concentric design were stereotaxically implanted under isoflurane anaesthesia (1.5% isoflurane/air mixture) into the lesioned SNr and GP (1 mm and 2 mm dialysing membrane, respectively), according to the following co-ordinates from bregma and the dural surface: GP, A = −1.3, L = +3.3 and V = −7.5; SNr, A = −5.5, L = −2.2 and V = −8. Forty-eight hours after surgery, probes were perfused with a modified Ringer solution (CaCl2 1.2 mmol/L, KCl 2.7 mmol/L, NaCl 148 mmol/L and MgCl2 0.85 mmol/L) at 3 µL/min flow rate. After 6 h rinsing, samples were collected every 15 min for a total of 3 h. At least three baseline samples were collected before administering l-DOPA and/or MTEP. At the end of experiment, animals were killed and the correct placement of the probes was verified histologically. Endogenous GABA levels were measured by HPLC-coupled to fluorimetric detection according to Marti et al. (2003). Briefly, 45 µL samples were pipetted into glass microvials and placed in a thermostated (4°C) Triathlon autosampler (Spark Holland, Emmen, The Netherlands). Forty microlitres of o-pthaldialdheide/boric acid solution were added to each sample, and 60 µL of the solution was injected onto an analytical column (3 mm inner diameter, 10 cm length; Chrompack, Middelburg, The Netherlands). The column was eluted at a flow rate of 0.48 mL/min with a mobile phase containing 0.1 mol/L sodium acetate, 10% ethanol and 2.5% tetrahydrofurane (pH 6.5). GABA retention time is about 17 min, and the sensitivity of the method is 150 fmol/sample.
Three hours after drug administration in the Experiment 1 and 2 days after the final l-DOPA challenge in the Experiment 2, the animals were deeply anaesthetized with sodium pentobarbitone (240 mg/kg, i.p.; Apoteksbolaget AB, Sweden) and killed by decapitation. Brains were rapidly extracted, frozen on powdered dry-ice and stored at −80°C. Coronal sections through the striatum were cut on a cryostat at 16-µm thickness and thaw-mounted onto adhesive glass slides (Superfrost End; Electron Microscopy Sciences, Hatfield, PA, USA). The slides were air dried and stored at −20°C. In the Experiment 3, animals were killed by decapitation immediately after completing the microdialysis session and brains were sectioned for histological verification of probe placement.
In situ hybridization histochemistry
In situ hybridization histochemistry was performed using synthetic oligomers complementary to PPE and PDyn mRNA, which have been extensively characterized in previous papers (Cenci et al. 1998; Andersson et al. 1999; Winkler et al. 2002; Lundblad et al. 2003). Oligonucleotides (0.2 µmol/L) were labelled at the 3′ end with 4 µmol/L [γ-35S] dATP (>37 TBq/mmol; Amersham Pharmacia Biotech, UK) using 15 U of terminal deoxynucleotidyltransferase (Amersham Pharmacia Biotech) for 2 h at 37°C. The labelled probes were purified by spin-column chromatography (Chroma Spin Columns; Clontech Laboratories; Palo Alto, CA, USA) to specific activities of >109 cpm/µg. Sections were air dried for 10 min and then incubated with the hybridization mixture, which comprised 50% formamide (deionized), 4x SSC (1x SSC = 0.15 mol/L NaCl and 0.015 mol/L sodium citrate), 1x Denhardt’s solution (0.02% Ficoll, 0.02% polyvinyl pyrolidone and 10 mg/mL of RNAse free bovine serum albumine), 1% sarcosyl, 10% dextran sulphate, 500 µg/mL sheared and denatured salmon sperm DNA, 25 µL/mL tRNA, 200 mmol/L dithiothreitol, 0.2 mg/mL heparin and 107 cpm/mL of 35S-labelled oligonucleotide probe. Hybridization cocktail of about 40 µL were added to each section (i.e. 240 µL/slide). Slides were coverslipped with parafilm and incubated for 18 h at 42°C in a humid chamber. After hybridization, the parafilm coverslips were floated off in 1x SSC at 55°C and the sections were given 4 × 15 min washes in 1x SSC at 55°C, plus a final wash beginning at 55°C and cooling down to 22°C. The slides were then rinsed twice in distilled water, dehydrated in 70% and 95% ethanol (2 min each) and exposed to Fuji imaging plates (Fujifilm AB, Sweden) for 1 h (PPE) or 16 h (PDyn). To obtain publication-quality pictures, the sections were also exposed to autoradiographic film (β-max, Amersham Pharmacia Biotech) for 3–12 days at −20°C. The films were developed in Kodak D-19 and fixed in Snabbfix (Scanfors, Sweden).
Image analysis was performed by an experimentally blinded investigator.
Phosphorimager plates. For a quantification of PPE or PDyn mRNA levels, imaging plates were scanned in a phosphorimager (BAS-5000; Fujifilm) and analysed using the software, Tina Adobe 2.10 (Fujifilm). The photostimulated luminescence emitted by the hybridized sections was calibrated against radiolabelled standards (14C-microscales; Amersham Pharmacia Biotech), which had been coexposed with the sections. Measurements of PPE and PDyn mRNA levels were carried out on five coronal sections per animal spanning the mid part of the caudate–putamen (CPu), i.e. between levels +1.60 to −0.48 mm relative to bregma (Paxinos and Watson 1997). This portion of the CPu is highly innervated by the somatosensory and primary motor cortex in the rat (McGeorge and Faull 1989; Kincaid and Wilson 1996) and exhibits pronounced molecular changes following chronic dyskinesiogenic treatment with l-DOPA (Andersson et al. 1999). The hybridization signal measured on both the DA-denervated and the intact side of the CPu was expressed as photostimulated luminescence/mm2 after subtraction of the tissue background (as determined on the corpus callosum in each section).
AIM scores collected on a single rating session (priming test in Experiment 2) were analysed non-parametrically by Kruskal–Wallis test and post hoc Mann–Whitney test. Like other behavioural data that had been collected on sequential testing sessions (rotarod and cylinder test), the AIMs scores recorded during a course of chronic drug treatment (Experiment 2) were analysed using repeated measures anova. This test provides valuable information on the interaction between a time factor (repeated testing) and a group category (treatment), which is not readily available in non-parametric tests. The group differences in AIM scores disclosed by repeated measures anova were verified non-parametrically (using Kruskal–Wallis test and post hoc Mann–Whitney test at single testing session). In situ hybridization data were examined using a two-factor anova with side (contralateral vs. ipsilateral to the lesion) and group as independent variables. Post hoc comparisons were carried out where appropriate using the Tukey ‘honestly significant difference method’. In the microdialysis experiment, statistical analysis was performed by a two-factor anova on absolute data or area-under-the-curve values followed by the Newman–Keuls test for multiple comparisons. The alpha level of statistical significance was set at p < 0.05. In the following, behavioural and in situ-hybridization data are expressed as group means ± SEM. GABA release is expressed as percentage of basal values (calculated as mean of the three samples before drug treatment) ± SEM.
Rats with 6-OHDA lesions were challenged with an acute injection of l-DOPA alone or in combination with different doses of MTEP. The levels of PDyn and PPE mRNA in the striatum were examined 3 h after drug administration, an interval sufficient to detect changes in the expression of opioid precursor genes after acute treatment with l-DOPA (Andersson et al. 2001).
The hybridization signal for PDyn mRNA differed significantly among the treatment groups only on the side ipsilateral to the lesion (Fig. 1a; treatment effect: F(1,60) = 1.17, p = 0.23; side effect: F(4,60) = 2.67, p = 0.04; side × treatment interaction: F(4,60) = 3.88, p < 0.001). In rats injected with vehicle, striatal levels of PDyn mRNA were reduced by ∼40% ipsilaterally to the lesion (#p < 0.05 vs. contralateral CPu), which is consistent with previous reports (Cenci et al. 1998; Andersson et al. 1999; Johansson et al. 2001; Fredduzzi et al. 2002). Treatment with l-DOPA alone markedly elevated the expression of PDyn mRNA on the lesioned side (Fig. 1a; increase by ∼40% vs. contralateral CPu, #p < 0.05; increase by ∼110% vs. the DA-depleted CPu in the vehicle group, *p < 0.05). The mGluR5 antagonist dose-dependently blocked this effect. Indeed in rats co-treated with l-DOPA and the two higher doses of MTEP (1.25 and 6.25 mg/kg), the expression of PDyn mRNA in the DA-denervated striatum did not differ significantly from that measured in vehicle-injected controls but it differed (at the highest dose of MTEP) from the values in l-DOPA-only animals (°p < 0.05 vs. the DA-depleted CPu in the l-DOPA-only group).
PPE mRNA levels were significantly higher in the DA-denervated striatum than in the contralateral intact side in all groups (Fig. 2a: treatment effect: F(4,66) = 0.75, p = 0.56; side effect: F(1,66) = 211.66, p < 0.001; side x treatment interaction: F(4,66) = 1.16, p = 0.34). Acute injection of l-DOPA alone or together with MTEP (at any of the doses tested) did not modify the up-regulation of PPE mRNA-induced by DA denervation (Fig. 2a). Acute injections of MTEP alone did not modify the expression of either PDyn or PPE mRNA compared with the levels measured in vehicle-treated 6-OHDA rats (data not shown).
mGluR5 antagonism inhibits the development of dyskinesia during chronic l-DOPA treatment (Experiment 2)
We next examined whether MTEP could block the development of AIMs, which occurred gradually during a 21-day course of l-DOPA treatment (Experiment 2). For this experiment, we used a dose of MTEP (5 mg/kg/day) that is known to produce >90%in vivo mGluR5 occupancy (Busse et al. 2004) and had been characterized in previous behavioural studies (Dekundy et al. 2006).
The development of dyskinesia during chronic drug treatment differed greatly among the groups (Fig. 3a; treatment effect: F(3,288) = 10.53, p < 0.001; time effect: F(8,288) = 5.30, p < 0.001; treatment x time interaction: F(24,288) = 2.65, p < 0.001). Indeed, chronic treatment with l-DOPA alone produced increasingly severe AIMs affecting trunk, limb and orolingual muscles. From the second to the last testing session, rats co-treated with l-DOPA and MTEP showed significantly lower AIM scores (*p < 0.05 vs. l-DOPA group), although they also differed significantly from vehicle-injected controls and from animals treated with MTEP only, which showed no sign of AIMs at all. Locomotive AIM scores showed a similar pattern of changes (Fig. 3b; treatment effect: F(3,288) = 9.04, p < 0.001; time effect: F(8,288) = 3.76, p < 0.001; time x treatment interaction: F(24,288) = 2.00, p = 0.004), being significantly lower in animals treated with l-DOPA plus MTEP compared with l-DOPA alone in most of the testing sessions. By the end of the chronic treatment period, the global AIM scores for axial, limb and orolingual dyskinesia were reduced by more than 60% in the animals treated with MTEP and l-DOPA compared with l-DOPA alone and the locomotive AIM scores were reduced to a similar extent.
In order to exclude that the anti-dyskinetic action of MTEP was associated with adverse motor effects, we used tests of general motor co-ordination (rotarod test) and spontaneous forelimb use (cylinder test). In the rotarod test, we detected significant overall differences among groups and testing sessions (Fig. 3c; treatment effect: F(3,112) = 13.36, p < 0.001; time effect: F(12,112) = 2.05, p = 0.025; time and treatment interaction: F(4,112) = 1.17, p > 0.33). l-DOPA alone produced a stable 40–60% increase in rotarod performance compared with control values throughout the experiment (*p < 0.05 vs. vehicle group in all testing sessions). Co-administration of l-DOPA and MTEP produced a gradual motor improvement, as rotarod performance in the combined treatment group was not different from control values on the first testing session, but showed a steady increase thereafter. In the last three testing sessions, the time spent on the rod was increased by >70% in the rats co-treated with l-DOPA and MTEP compared with vehicle-injected animals (*p < 0.05 vs. control in the testing session 2, 3, 4 and 5), comparably with the improvement seen after l-DOPA only. When given alone, MTEP did not have any significant effect on rotarod performance compared with vehicle (Fig. 3c; filled diamonds).
In the cylinder test, the proportion of wall contacts performed by the paw contralateral to the lesion (left) amounted to ∼10–15% in all groups when the test was performed off drugs (baseline in Fig. 3d). During the course of chronic drug treatment, the performance on this test showed significant differences among groups and testing sessions (Fig. 3d; treatment effect: F(3,108) = 2.95, p = 0.045; time effect: F(3,108) = 8.00, p < 0.001; time and treatment interaction: F(9,108) = 2.88, p = 0.004). l-DOPA produced an improvement in limb use asymmetry, raising the percentage of wall contacts performed by the left paw to 35–40% of total (*p < 0.01 for l-DOPA vs. vehicle in all testing session). The groups co-treated with l-DOPA and MTEP showed a gradual improvement in limb use asymmetry, reaching ∼25% contralateral paw usage in the last two testing sessions (*p < 0.05 vs. vehicle; non-significant difference from l-DOPA-only). Even in this test, MTEP alone did not produce any significant improvement in motor performance (∼18% contralateral forelimb use; non-significant difference from vehicle).
mGluR5 antagonism interferes with the dyskinesia priming process
The results presented so far show that MTEP significantly reduced the development of AIMs at a dose that did not interfere with the anti-akinetic action of l-DOPA. However, this may reflect a physical suppression of dyskinesia as opposed to its actual prevention. In order to verify that MTEP had indeed prevented the neural priming for dyskinesia, rats from all the treatment groups were given 1 day of drug washout and were then challenged with one injection of l-DOPA (6 mg/kg + benserazide 15 mg/kg), followed by a dyskinesia-rating session. As shown in Fig. 4, the median global AIM score for axial, limb and orolingual dyskinesia was over 13-fold larger in animals from the chronic l-DOPA-treatment group compared with the rats that had previously received vehicle. The global AIM score was reduced by 87% in rats previously co-treated with l-DOPA and MTEP (#p < 0.05 vs. l-DOPA-only group), and it was not significantly elevated above the values measured in l-DOPA-naïve animals (p = 0.47 vs. vehicle group; Fig. 4). Rats previously treated with MTEP alone did not differ from vehicle-injected controls, showing low levels of AIMs (#p < 0.05 vs. l-DOPA group).
A well-established molecular correlate of the dyskinesia priming process is the up-regulation of PDyn gene expression in the striatum (Cenci et al. 1998; Andersson et al. 1999; Westin et al. 2001; Lundblad et al. 2003), which persists for up to 16 days after discontinuation of chronic l-DOPA treatment (Andersson et al. 2003). We therefore measured striatal levels of PDyn mRNA in all animals 2 days after the challenge injection of l-DOPA (a time sufficient for the effect of a single l-DOPA injection to subside). The expression of this transcript differed significantly among the groups ipsilaterally to the lesion (Figs 5a–d and i; two-factor anova, treatment effect: F(3,70) = 10.40, p < 0.001; side effect: F(1,70) = 10.20, p = 0.002; side-treatment interaction: F(3,70) = 9.20, p < 0.001). On the side of the CPu ipsilateral to the lesion, PDyn mRNA levels were significantly reduced in animals from the vehicle-control group (decrease by ∼34% vs. the intact side; #p < 0.05), which is consistent with the transcript down-regulation that is expected after DA denervation (cf. vehicle group in Fig. 1). In the animals that had previously received chronic treatment with l-DOPA, the levels of PDyn mRNA on the DA-denervated side were ∼70% larger than on the intact side (#p < 0.05) and were increased by 147% above the values measured in vehicle-treated animals (*p < 0.05 vs. lesion side in the vehicle group; Fig. 5i). An up-regulation of PDyn mRNA by ∼30% on the lesion over the intact side was measured in the rats previously co-treated with l-DOPA and MTEP (#p < 0.05 vs. the contralateral CPu; *p < 0.05 vs. DA-depleted CPu in the control group, Fig. 5i), but this increase was significantly less pronounced than that observed in the l-DOPA-only group (°p < 0.05 vs. DA-depleted CPu in the l-DOPA group). Animals previously treated with MTEP alone did not show any significant side difference in the expression of PDyn mRNA, suggesting reversal of the lesion-induced down-regulation of PDyn mRNA by mGluR5 antagonist treatment (Figs 5d and i).
We also measured the striatal expression of PPE mRNA, which is potentially affected by the pharmacological modulation of group I mGluRs (Wardas et al. 2003) and has been implicated in the dyskinesia priming process by some investigators (for review see Calon et al. 2000). PPE mRNA levels were significantly up-regulated on the DA-denervated side of the CPu in all experimental groups (Figs 5e–h and j; side effect: F(1,70) = 309.62, p < 0.001; treatment effect: F(3,70) = 1.72, p > 0.17; side and treatment interaction: F(3,70) = 2.97, p = 0.037 on a two-factor anova). Chronic treatment with l-DOPA alone exacerbated this up-regulation, causing an additional 20% increase in PPE mRNA levels in the DA-denervated CPu over the values measured in vehicle-injected controls (*p < 0.05 vs. lesion side of vehicle group; cf. Figs 5e and f). Co-administration of MTEP partially blocked this additional up-regulation (Fig. 5g), as indicated by the fact that animals in the l-DOPA + MTEP group were not significantly different from vehicle-injected controls, though they did not differ significantly from the l-DOPA group either (Fig. 5j). In rats that had been treated with MTEP only (Fig. 5h), striatal levels of PPE mRNA did not differ from those measured in any other group and showed a significant increase ipsilaterally to the lesion compared with the intact side, indicating that chronic treatment with the mGluR5 antagonist had not reversed the up-regulation of PPE mRNA due to the lesion.
mGluR5 antagonism inhibits the expression of dyskinesia and the concomitant increase in nigral GABA output (Experiment 3)
Preliminary reports have indicated that pharmacological antagonism of mGluR5 can acutely reduce the expression of already established dyskinesias (Hill et al. 2001; Dekundy et al. 2006). Such an effect is unlikely to depend on a normalization of the long-term changes in striatal gene expression that are induced by chronic treatment with l-DOPA and must therefore rely on some alternative mechanism.
Based on current pathophysiological models (for review see Brotchie et al. 2005), we hypothesized that the expression of dyskinesia is dependent on changes in GABA release in the target of the striatal efferent projections. We therefore measured GABA levels in the SNr and GP in freely moving dyskinetic rats simultaneously with AIMs rating (Fig. 6).
Baseline levels of GABA in the SNr were not significantly different between dyskinetic (14.8 ± 4.2 nmol/L; n = 11) and non-dyskinetic rats (11.4 ± 2.1 nmol/L; n = 7). Following an i.p. injection of l-DOPA (6 mg/kg + benserazide 15 mg/kg, i.p.), extracellular levels of GABA in the SNr showed a gradual increase in dyskinetic animals, rising by 70% above baseline at 45–75 min post-injection and gradually declining thereafter (Fig. 6a, empty triangles; main effect of l-DOPA: F(1,23) = 5.19, p = 0.032). Interestingly, this time course closely paralleled the expression of AIMs (Fig. 6e). This neurochemical change was specifically associated with dyskinesia because it did not occur in l-DOPA-treated rats that did not express AIMs (Fig. 6b). Co-administration of MTEP (5 mg/kg, i.p.) with l-DOPA completely abolished the surge in GABA levels associated with dyskinesia (Fig. 6a, filled triangles; main effect of MTEP: F(1,23) = 1.73, p > 0.20; l-DOPA–MTEP interaction: F(1,23) = 14.21, p < 0.001). This effect was paralleled by a marked attenuation (∼60%) of the rat AIM score (Fig. 6ep < 0.01 vs. l-DOPA alone). Differently from the SNr, no change in GABA release was measured in the GP during the expression of AIMs (Figs 6c–d). In this structure, GABA levels did not differ between dyskinetic and non-dyskinetic rats either at baseline (13.0 ± 1.3 nmol/L and 14.5 ± 1.8 nmol/L, respectively, n = 7) or after the administration of l-DOPA, MTEP or their combination.
This study focuses on mGluR5, a phospholipase C-coupled receptor that is abundantly expressed in striatal neurons (Standaert et al. 1994), where it positively modulates DA-dependent and NMDA-dependent signalling and synaptic plasticity (for review see Gubellini et al. 2004). A previous study demonstrated that mGluR5 gene expression is up-regulated in the striatum in dyskinetic animals (Konradi et al. 2004), but the possible contribution of mGluR5 to the pathogenesis of LID has thus far remained unknown. In this study, we provide the first demonstration that pharmacological antagonism of mGluR5 inhibits not only the acute expression, but also the long-term development of LID in an experimental model of PD. These behavioural effects are paralleled by a normalization of molecular and neurochemical changes that are closely associated with the dyskinetic action of l-DOPA, i.e. the striatal overexpression of PDyn mRNA and the excessive overflow of GABA in the SNr. As discussed below, striatal neurons of the ‘direct pathway’ are likely to provide the common locus of the behavioural, molecular and neurochemical effects of mGluR5 antagonism in this study.
Type 5 mGluRs are most abundantly expressed in striatal neurons than in any other structure within the basal ganglia (Testa et al. 1994). The principal neurons within the striatum give rise to two efferent pathways, one of which projects directly to the output stations of the basal ganglia (i.e. SNr and entopeduncular nucleus in the rat), while the other one (termed ‘indirect pathway’) projects to the GPe and modulates the basal ganglia output structures via polysynaptic circuits (Albin et al. 1989). A body of independent studies indicates that both the acute expression and the gradual development of dyskinesia are due to an excessive responsiveness of ‘direct pathway’ neurons to l-DOPA. In rats with 6-OHDA lesions, the maladaptive behavioural effects of l-DOPA are closely associated with an up-regulation of transcription factors and plasticity genes in dynorphinergic (‘direct pathway’) but not in enkephalinergic striatal neurons (Andersson et al. 1999; Carta et al. 2005; St-Hilaire et al. 2005;Sgambato-Faure et al. 2005; St-Hilaire et al. 2005). In the same rat model, studies of 2-deoxyglusose uptake have suggested increased axon terminal activity of striatal efferent pathways to the entopeduncular nucleus and SNr, but not to the GPe (Trugman and Wooten 1986). Finally, antagonism of D1 DA receptors, which are predominantly expressed in direct pathway neurons (Gerfen 1992) can completely block both the development of dyskinesia and the associated abnormal molecular plasticity in 6-OHDA-lesioned rats (St-Hilaire et al. 2005; Westin et al. 2007). The blockade of l-DOPA-induced PDyn gene up-regulation by MTEP in this study indicates that the compound normalized the response to l-DOPA in striatal neurons of the ‘direct pathway’. Although the anatomical site of action of MTEP was not addressed, several independent studies performed in cell cultures and slice preparations have shown that pharmacological antagonists of mGluR5 can directly modulate gene expression and synaptic responses in striatal neurons by a local effect (Mao and Wang 2001, 2002, 2003; Voulalas et al. 2005).
Up-regulation of PDyn mRNA by l-DOPA provides a molecular marker of the dyskinesia priming process in both rodent and non-human primate models of PD (Cenci et al. 1998; Tel et al. 2002; Henry et al. 2003). We have previously shown that an acute injection of l-DOPA is able to enhance PDyn mRNA expression in the DA-denervated CPu, although PDyn mRNA levels increase further with repeated l-DOPA administration (Andersson et al. 2001). Our data show that mGluR5 antagonism can block the up-regulation of PDyn mRNA induced by both acute and chronic l-DOPA treatment. Along with this effect, the compound can attenuate both the gradual development of AIMs and the persistent priming for dyskinesia that are induced by chronic l-DOPA treatment. Thus, our data indicate that mGluR5 is involved in the maladaptive molecular changes that induce and maintain a dyskinesia-prone state. Although group I mGluRs have been reported to modulate striatal gene expression in different model systems (Mao and Wang 2001, 2002, 2003), this is the first report showing that l-DOPA-induced gene expression can be inhibited by mGluR5 blockade. This data is particularly important in light of the fact that NMDA antagonists are reportedly unable to inhibit D1 receptor dependent- and/or l-DOPA-induced gene expression in DA-denervated striatal neurons (Keefe and Gerfen 1996; Adams et al. 2000; Ganguly and Keefe 2000). Both D1-like and mGlu5 receptors co-operate to enhance striatal extracellular signal-regulated kinases 1 and 2 phosphorylation through a common pathway (Voulalas et al. 2005). Blockade of l-DOPA-induced gene PDyn up-regulation by MTEP may therefore reflect an action of the compound on D1-dependent nuclear signalling processes in ‘direct pathway’ neurons.
In addition to blocking the priming for dyskinesia, MTEP acutely reduced the expression of AIMs when dyskinesia had already been established by a prior course of l-DOPA treatment. This effect was associated with an inhibition of l-DOPA-induced GABA release in the SNr. This is the first report demonstrating that the physical manifestation of AIMs is closely linked with an enhanced GABA overflow in the basal ganglia output stations. The maladaptive value of such a neurochemical change is suggested by a comparison between dyskinetic and non-dyskinetic rats, where the latter group did not exhibit any significant modification in nigral GABA release after the administration of l-DOPA. Excessive GABA release may mediate a reduction in neural activity associated with a change in firing patterns in the basal ganglia output nuclei, and these changes have a prime pathophysiological importance in LID (for review see Bezard et al. 2001; Cenci 2007). The SNr is the largest output nucleus of the basal ganglia in the rat and contains GABAergic neurons along with a prominent GABAergic innervation from striatal and pallidal afferents (Windels et al. 2005). Drug treatments that enhance DA release in the striatum can cause excess GABA overflow in the SNr through D1 receptor-dependent activation of the ‘direct pathway’ (Mark et al. 2004). Accordingly, we have obtained preliminary evidence that the surge in nigral GABA release induced by l-DOPA can be blocked by intrastriatal infusion of a D1 receptor antagonist (own unpublished data). The striatonigral origin of the change in nigral GABA release observed in dyskinetic rats is further supported by the lack of any parallel change in the GPe, which is a target of striatopallidal (‘indirect pathway’) projections.
In the present study, we demonstrate that mGluR5 receptor blockade provides specific protection against dyskinesia in a rat model of PD, acting both on the sensitization (priming process) and on the acute expression of AIMs as induced by l-DOPA treatment. We also provide the first demonstration that the physical expression of AIMs goes hand in hand with an abnormal increase in GABA outflow within the SNr. We propose that the anti-dyskinetic effects of mGluR5 antagonism are mediated through normalization of molecular responses in striatal neurons of the direct pathway, which show an abnormal sensitivity to DA (D1) receptor stimulation following DA-denervation, and become further hyperresponsive during the course of chronic l-DOPA treatment (Aubert et al. 2005; Konradi et al. 2004). Although MTEP may also block extrastriatal mGlu5 receptors, an action on extrastriatal sites [e.g. the STN and SNr (Testa et al. 1995)] would not be expected to reduce dyskinesia, but on the contrary, may cause its further aggravation.
Previous studies have provided variable results on anti-parkinsonian-like activity of mGluR5 receptor blockade in rodents (Spooren et al. 2000; Ossowska et al. 2001; Breysse et al. 2003; Dekundy et al. 2006). Accordingly, we did not observe significant anti-akinetic effects of MTEP in rats with 6-OHDA lesions when the compound was administered alone. However, the compound did not interfere with the therapeutic-like actions of l-DOPA in this rat model, while inhibiting both the acute expression and the long-term development of dyskinetic movements. The present data thus suggest that pharmacological antagonism of mGluR5 may be a useful adjunct to l-DOPA therapy in order to prevent disabling motor complications in PD.
The study was supported by grants from the MJ Fox Foundation for Parkinson’s Research, the Swedish National Research Council, The Greta and Johan Kock Foundation, The Elsa and Thorsten Segerfalk Foundation, The King Gustav V and Queen Victoria Foundations and The Swedish Parkinson Foundation (to M.A.C) and by a FIRB and PRIN 2005 grant from the Italian Ministry of University (to MM). We thank Inga-Lill Bertilsson for technical help and Jonas Björk (Competence Centre for Clinical Research, Lund University Hospital) for his advice on statistics.