Increased D1 Dopamine Receptor Signaling in Levodopa-Induced Dyskinesia Incarnation Aubert, PhD,1 Céline Guigoni, PhD,2 Kerstin Håkansson, PhD,3 Qin Li, BsC,4 Sandra Dovero,2 Nicole Barthe, MD, PhD,5 Bernard H. Bioulac, MD, PhD,2 Christian E. Gross, PhD,2 Gilberto Fisone, PhD,3 Bertrand Bloch, MD, PhD,1 and Erwan Bezard, PhD2 Involuntary movements, or dyskinesia, represent a debilitating complication of levodopa therapy for Parkinson’s disease. Although changes affecting D1 and D2 dopamine receptors have been studied in association with this condition, no causal relationship has yet been established. Taking advantage of a monkey brain bank constituted to study levodopainduced dyskinesia, we report changes affecting D1 and D2 dopamine receptors within the striatum of normal, parkinsonian, nondyskinetic levodopa-treated parkinsonian, and dyskinetic levodopa-treated parkinsonian animals. Whereas D1 receptor expression itself is not related to dyskinesia, D1 sensitivity per D1 receptor measured by D1 agonist-induced [35S]GTP␥S binding is linearly related to dyskinesia. Moreover, the striata of dyskinetic animals show higher levels of cyclin-dependent kinase 5 (Cdk5) and of the dopamine- and cAMP-regulated phosphoprotein of 32kDa (DARPP-32). Our data suggest that levodopa-induced dyskinesia results from increased dopamine D1 receptor–mediated transmission at the level of the direct pathway. Ann Neurol 2005;57:17–26 Long-term L-3,4-dihydroxyphenylalanine (L-dopa) treatment of Parkinson’s disease (PD)1–3 induces adverse fluctuations in motor response and involuntary movements, known as L-dopa–induced dyskinesia (LID) (for a review, see Bezard and colleagues4). Denervation-induced supersensitivity of dopamine (DA) receptors (D1-like and D2-like) has been widely suggested as the most plausible mechanism of LID. Indeed, striatal D2 receptor–binding sites are increased in postmortem tissue of untreated parkinsonian patients and in animal models.5,6 Although supersensitivity of D2 receptors is expected when parkinsonism is first apparent, the first L-dopa dose administered does not generally induce dyskinesia, but dyskinesia develops gradually over time.7 Accordingly, the D2/D3 receptor agonists exert an antiparkinsonian effect with a reduced propensity to elicit dyskinesia when administered de novo in PD patients.8 There is some evidence that D1 messenger RNA (mRNA) levels are increased after dopaminergic treatment of the DA-depleted striatum in animal models of LID9; that downstream signal transduction cascades is abnormal in LID,10,11 including increased phosphorylation of cAMP-regulated phosphoprotein of 32kDa (DARPP-32)12; and that an altered subcellular localization of D1 receptors is involved in LID.13 Moreover, a DA D1 receptor agonist with proven antiparkinsonian action14 induced LID similar to that induced by L-dopa in PD patients,15 further suggesting that D1 supersensitivity plays a key role in LID occurrence. Together, these observations call for a reassessment of the changes affecting D1 and D2 DA receptors in LID. In this study, taking advantage of a nonhuman primate (NHP) brain bank constituted to study the pathophysiology of LID,16 we determined changes affecting D1 and D2 DA receptors within the striatum of four experimental groups: normal, parkinsonian, parkinsonian chronically treated with L-dopa without exhibiting dyskinesia, and parkinsonian chronically treated with L-dopa that shows overt dyskinesia. We show that LIDs are linked to a modification of both D1 receptor expression and sensitivity of the D1signaling cascade, reinforcing the hypothesis of the piv- From the 1Centre National de la Recherche Scientifique Unite Mixte de Recherche 5541 and 2Basal Gang Centre National de la Recherche Scientifique UMR 5543, Bordeaux Cedex, France; 3 Karolinska Institutet, Department of Neuroscience, Stockholm, Sweden; 4Lab Animal Research Center, China Agricultural University, Beijing, China; and 5Institut National de la Sante et de la Recherche Médicale U443, Université Victor Segalen-Bordeaux 2, Bordeaux Cedex, France. Received Jun 3, 2004, and in revised form Aug 19. Accepted for publication Aug 24, 2004. Published online Oct 27, 2004, in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/ana.20296 Address correspondence to Dr Bezard, Laboratoire de Physiologie et Physiopathologie de la Signalisation Cellulaire, Centre National de la Recherche Scientifique Unite Mixte de Recherche 5543, Université Victor Segalen, 146 rue Léo Saignat, 33076 Bordeaux Cedex, France. E-mail: firstname.lastname@example.org © 2004 American Neurological Association Published by Wiley-Liss, Inc., through Wiley Subscription Services 17 otal role played by the so-called direct pathway in LID genesis. Materials and Methods Animals We used 17 female cynomolgus monkeys (Macaca fascicularis; Shared Animal Health, Beijing, China) for this study (Table). Animals were housed in individual primate cages under controlled conditions of humidity (50% ⫾ 5%), temperature (24°C ⫾ 1°C) and light (12-hour light/12-hour dark cycles, time lights on 8:00 AM), food and water were available ad libitum, and animal care was supervised by veterinarians skilled in the health care and maintenance of NHPs. Experiments were carried out in accordance with European Communities Council Directive of 24 November 1986 (86/609/EEC) for care of laboratory animals. Experimental Protocol Four monkeys were kept normal (control group), and 13 were intoxicated with 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) hydrochloride. Once a bilateral parkinsonian syndrome had stabilized (ie, unchanged disability score over several weeks), four monkeys were kept without any dopaminergic supplementation (MPTP group), and the nine were treated chronically with twice daily administration of L-dopa (Modopar; L-dopa/carbidopa, ratio 4:1) for 6 to 8 months at a tailored dose designed to produce a full reversal of parkinsonian condition (see Table). Five monkeys developed severe and reproducible dyskinesia (MPTP-intoxicated, dyskinetic monkey group), whereas four did not (MPTPintoxicated, nondyskinetic monkey group). Animals were killed by sodium pentobarbital overdose (150mg/kg of body weight, intravenously). Brains were removed quickly after death. Each brain was bisected along the midline, the most rostral part of the striatum was removed for western blotting experiments, and the two hemispheres were immediately frozen by immersion in isopentane (⫺45°C) and then stored at ⫺80°C. Tissue was sectioned coronally at 20m in a cryostat at ⫺17°C, thaw-mounted onto gelatine-subbed slides, dried on a slide warmer, and stored at ⫺80°C. Behavioral Assessment Parkinsonian condition (and reversal) was assessed on a parkinsonian monkey rating scale using videotape recordings of monkeys as previously described.16,17 A score of 0 corresponds to a normal animal, and a score above 6 corresponds to a parkinsonian animal. The severity of dyskinesia was rated using the Dyskinesia Disability Scale: 0, dyskinesia absent; 1, mild, fleeting, and rare dyskinetic postures and movements; 2, moderate, more prominent abnormal movements, but not interfering significantly with normal behavior; 3, marked, frequent and, at times, continuous dyskinesia intruding on the normal repertoire of activity; or, 4, severe, virtually continuous dyskinetic activity, disabling to the animal and replacing normal behavior. Assessment of Lesion DA transporter binding using [125I](E)-N-(3-iodoprop-2enyl)-2␤-carboxymethyl-3␤-(4⬘-methylphenyl)-nortropane (PE2I; Chelatec, Nantes, France) was measured as previously described.18 Processing of mesencephalic sections for tyrosine hydroxylase (TH) immunohistochemistry, counterstaining with cresyl violet (Nissl staining), and cell counts (Visioscan version 4.12; Biocom, Les Ulis, France) were performed as Table. Characteristics of the Monkey Subgroups L-Dopa 0 min Group Control C1 C2 C3 C4 M1 M2 M3 M4 L1 L2 L3 L4 D1 D2 D3 D4 D5 MPTP MPTP/L-dopa nondyskinetic MPTP/L-dopa dyskinetic 90 min Age (yr) Weight (kg) mg P D P D 3 3 3 3 3 3 3 3 3 3 3 3 3 3 4 3 3 3.0 2.7 3.0 3.1 3.4 2.9 3.0 3.1 2.7 2.5 2.7 3.3 3.2 3.3 4.5 2.8 3.0 — — — — — — — — 54 59 56 50 47 60 66 53 50 0.0 0.0 0.0 0.0 9.0 9.2 8.2 8.8 9.4 10.6 9.2 8.4 10.0 8.8 7.5 7.2 7.5 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.2 0.0 0.0 0.0 2.2 0.7 0.2 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.2 0.0 0.0 0.4 1.9 3.7 2.8 3.7 3.3 L-Dopa dose (mg) is given for each individual as well as their parkinsonian (P) and dyskinetic (D) scores before administration (0 min) and 90 min after administration. L-Dopa dose was tailored to produce a full reversal of parkinsonian motor abnormalities as shown by the drastic decrease in parkinsonian score after 90 min. Despite comparable levels of lesion and duration of treatment, only the “dyskinetic” animals displayed severe LID 90 min after L-dopa administration. 18 Annals of Neurology Vol 57 No 1 January 2005 previously described.18 The boundaries of the substantia nigra pars compacta (SNc) were chosen on three consecutive sections corresponding to a representative median plane of the SNc by examining the size and shape of the different tyrosine hydroxylase–immunoreactive (TH-IR) neuronal groups, cellular relationships to axonal projections, and nearby fiber bundles. The number of both TH–IR and Nissl-stained neurons per SNc representative plane was counted three times by one examiner blind about the experimental condition. Split cell counting error was corrected by using the formula of Abercrombie.19 Mean cell number per plane and standard error of the mean were then calculated for each group of monkeys. Receptor Binding Both the D1 and D2 receptors were labeled using ligands specific for D1-like sites ([3H]SCH 23390, 75Ci/mmol; New England Nuclear, Paris, France) or D2-like sites ([3H]YM09151-2, 85Ci/mmol, New England Nuclear) as previously described18: tissue sections were incubated for 1 hour at room temperature in a buffer solution (50mM Tris-HCl, 120mM NaCl, 5mM KCl, 2mM CaCl2, 1mM MgCl2, pH 7.4) containing either 2nM [3H]SCH 23390 or 0.3nM [3H]YM-09151-2. Nonspecific binding was defined in the presence of 10M of (⫹)butaclamol for both subtypes of DA receptor. Sections were exposed to a ␤-imager (Biospace, Paris, France) to assess directly the radioactivity bound to regions of interest.20,21 GTP␥S Binding Labeling of monkey brain sections with [35S]GTP␥S (Amersham, Uppsala, Sweden) was carried out essentially as described by Sim et al. (1995) with minor modifications. The slides were incubated for 10 minutes at 25°C in assay buffer (50mM Tris-HCl, 3mM MgCl2, 0.2mM EGTA, 100mM NaCl, pH 7.7). Slides were then incubated with 2mM GDP in assay buffer for 15 minutes at 25°C. Agonist-stimulated activity was determined by incubation in [35S]GTP␥S (0.01nM) with 2mM GDP and D1 agonist (SKF38393) in assay buffer for 2 hours at 25°C. In each experiment, basal activity was assessed with GDP in absence of agonist, and nonspecific binding was assessed in presence of 10M unlabelled GTP␥S. Slides were exposed to ␤-imager (Biospace) to assess directly the radioactivity bound to regions of interest.20,21 Image Analysis Densitometric analysis of autoradiographs (in situ hybridization) and direct measurement of radioactivity of ␤-imager images (binding) was performed using, respectively, an image analysis system (Visioscan version 4.12; Biocom) and ␤-vision (version 4.2; Biospace), at three rostrocaudal levels in accordance with the functional organization of the striatum as previously described18,25: a rostral level including the caudate, putamen, and nucleus accumbens (A21.0); a midlevel including the caudate, putamen, and globus pallidus pars externalis (A17.2); and a caudal level including the body of the caudate, the putamen, and both parts of the globus pallidus (ie, pars externalis and pars internalis) (A14.6). Where appropriate, both caudate and putamen were divided into dorsolateral, dorsomedial, ventrolateral, and ventromedial quadrants for analysis. Three sections per animal per striatal level were analyzed by an examiner blind about the experimental condition. For autoradiographs, optical densities were averaged for each region in each animal, converted to amount of radioactivity bound by comparison to the standards, and expressed in femtomoles per milligram of tissue equivalent (mean ⫾ standard error of the mean). Since ␤-imager images allow direct measurements of the radioactivity, data are expressed in counts per minute per square micrometer. Western Blotting Pieces of monkey caudate–putamen were sonicated in 1ml of 1% sodium dodecyl sulfate and boiled for 10 minutes. Aliquots (5l) of the homogenate were used for protein determination using the BCA (bicinchoninic acid) assay kit (Pierce, Oud Beijerland, The Netherlands). Equal amounts of protein (30g) from each sample were loaded onto 10% polyacrylamide gels, separated by sodium dodecyl sulfate polyacrylamide gel electrophoresis, and transferred to polyvinylidene difluoride membranes (Amersham). The membranes were immunoblotted using affinity-purified polyclonal antibodies against cyclin-dependent kinase 5 (Cdk5; Santa Cruz Biotechnology, Santa Cruz, CA) and DARPP-32 (Cell Signaling Technology, Beverly, MA). Antibody binding was indicated by incubation with goat anti–rabbit horseradish peroxidase–linked immunoglobulin G (diluted 1:10,000; Pierce Europe, Oud Beijerland, The Netherlands) and the enhanced chemiluminescence (ECL) immunoblotting detection system. Autoradiograms were quantified with NIH Image software (version 1.62). In Situ Hybridization Histochemistry The in situ hybridization procedure was performed as previously described22 with probes designed to recognize the human D1R23 or the human D2R24. 35S-labeled antisense and sense complementary RNA probes were prepared by in vitro transcription from 100ng of linearized plasmid using [35S]UTP (⬎1,000Ci/mmol; New England Nuclear), and SP6, T3, or T7 polymerases. After alkaline hydrolysis to obtain 0.25kb complementary RNA fragments, the probes were purified on G50-Sephadex and precipitated in sodium acetate (0.1 vol)–absolute ethanol (2.5 vol). Sections were then hybridized as described by Aubert et al.22 Slides were then exposed to Biomax film (Kodak) with autoradiographic microscale standard (Amersham). Statistical Analysis For multiple comparisons of cell counting, DAT binding and GTP␥S binding, one-way analysis of variance (ANOVA) was used. For multiple comparisons of DA receptor binding and DA receptor mRNA levels, a two-way ANOVA, with group and striatal subregions as factors, was used. ANOVAs were followed when allowed by post hoc t tests corrected for multiple comparisons by the method of Bonferroni. For multiple comparisons of behavioral assessments, the Kruskal– Wallis nonparametric test was used to estimate overall significance followed by post hoc t tests corrected for multiple comparisons by the method of Dunn. These analyses were completed using the Stata program (Intercooled Stata 6.0; Aubert et al: Increased D1 Signaling in Dyskinesia 19 Stata Corporation, College Station, TX). A probability level of 5% ( p ⬍ 0.05) was considered statistically significant. Results Changes in Motor Behavior Five monkeys developed severe and reproducible dyskinesia (MPTP-lesioned, dyskinetic monkey group, score ⫽ 3.08 ⫾ 0.34, of a maximum of 4) (see Table), whereas four did not (MPTP-lesioned, nondyskinetic monkey group, score ⱕ 0.4) (see Table). Parkinsonism was comparable between the different MPTP-lesioned groups ( p ⬎ 0.5), and motor abnormalities were fully reversed by the L-dopa treatment (see Table). Dyskinetic animals presented choreic–athetoid (characterized by constant writhing and jerking motions), dystonic, and sometimes ballistic movements (large-amplitude flinging, flailing movements). At the peak of dose (80 – 150 minutes after injection), dystonic rolling and writhing on cage floor were common. Dyskinesia developed by MPTP animals was similar to the LID observed in PD patients. Extent of Lesion Is Homogeneous among the MPTPLesioned Groups Since the extent of nigrostriatal degeneration is an obvious variable that may play a role in the susceptibility to develop LID, we first assessed whether all MPTPlesioned monkeys had similar loss in the number of TH-IR neurons in the SNc and in striatal DA nerve endings by measuring DA transporter (DAT) binding. MPTP treatment induced a strong loss both in the number of TH-IR cells and in the total number of neurons, that is, Nissl-stained cells (Fig 1A). In addition, MPTP induced striatal dopaminergic denervation, as shown by a decrease in DAT binding, both in the dorsolateral caudate nucleus and in the dorsolateral putamen (see Figs 1B, C). There was no significant difference in the decrease in the number of TH-IR neurons and DAT binding between MPTP-lesioned group, nondyskinetic L-dopa–treated MPTP-lesioned group, and dyskinetic L-dopa–treated MPTP-lesioned group. Thus, MPTP intoxication generated a similar lesion in all MPTP-lesioned groups (see Fig 1C). L-Dopa Does Not Normalize D2 Levels Previous studies have shown that DA denervation causes an increase in striatal D2 DA receptor binding sites in the postmortem tissue of untreated patients with PD and animal models.5,6,18 Accordingly, we found that MPTP alone produced a significant increase in the levels of D2 mRNA in comparison with control animals (Figs 2A, B). A similar effect was produced on D2 DA receptor binding (see Figs 2C, D). When compared with the control group, the increase in D2 binding was particularly striking in the dorsal part of the caudate nucleus and in the whole putamen (see Figs 2C, D). L-Dopa treatment, in addition to MPTP intoxication, differentially affected D2 mRNA in nondyskinetic and dyskinetic animals. In the nondyskinetic NHP, L-dopa induced a normalization of D2 mRNA levels, which were not significantly different from those determined in the control group (see Figs 2A, B). In the dyskinetic animals, D2 mRNA levels remained signifi- Fig 1. Effect of 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) intoxication and L-dopa treatment on (A) the estimated number of tyrosine hydroxylase-immunoreactive (TH-IR) neurons and Nissl-stained cells in the substantia nigra pars compacta (SNc), and (B) on the striatal DA transporter (DAT) binding. MPTP treatment induced a strong loss in the estimated number of TH-IR neurons (F(3,15) ⫽ 293.7; p ⬍ 0.0001) and Nissl-stained cells (F(3,15) ⫽ 180.9; p ⬍ 0.0001) and specific [125I]PE2I binding (caudate: F(3,15) ⫽ 488.47; p ⬍ 0.0001; putamen: F(3,15) ⫽ 1911.82; p ⬍ 0.0001). Results represent the mean ⫾ standard error of the mean (* ⫽ statistically significant difference compared with control animals, p ⬍ 0.05). (C) Example of DAT binding autoradiographs showing the striatal denervation at the caudal level. 20 Annals of Neurology Vol 57 No 1 January 2005 Fig 2. Effect of 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) intoxication and L-dopa treatment on (A, B) striatal D2 messenger RNA (mRNA) expression (group effect: F(3,95) ⫽ 55.38; p ⬍ 0.0001; interaction between group and striatal subregion: F(21,95) ⫽ 1.08) (DL ⫽ dorsolateral; DM ⫽ dorsomedian; VL ⫽ ventrolateral; VM ⫽ ventromedian) and (C, D) striatal D2 binding levels (group effect: F(3,135) ⫽ 104.9; p ⬍ 0.0001; interaction between group and striatal subregion: F(21,135) ⫽ 0.61). Results are expressed as the percentage of control animals. For example, in the dorsolateral (DL) putamen of control animals, mRNA level was 30.73 ⫾ 1.23fmol/mg of tissue equivalent, and binding level was of 2.07 ⫾ 0.17cpm/m2 (* ⫽ statistically significant difference compared with control animals, p ⬍ 0.05). (B) Examples of D2 mRNA expression autoradiographs and (D) of D2 binding ␤-imager images at the rostral level of striatum. cantly higher than those of control monkeys, particularly in the ventromedian caudate nucleus and the putamen (see Figs 2A, B). Despite these differences in mRNA, L-dopa therapy had no effect on D2 receptor levels. Indeed, the D2 binding of the nondyskinetic and dyskinetic groups remained comparable to that observed after MPTP treatment (see Figs 2C, D). Our results are in agreement with previous studies that showed a normalization of the D2 mRNA receptor after L-dopa treatment in patients with PD6,26,27 and in MPTP-lesioned monkeys.28 They confirm the lack of direct correlation between D2 regulation and the occurrence of LID and rule out the hypothesis of a predominant role for the indirect pathway in LID occurrence. L-Dopa Increases D1 Levels D1 receptor expression does not follow the same pattern. In MPTP-lesioned animals and in MPTPlesioned animals treated with L-dopa, D1 mRNA expression was similar to that of control group in all striatal quadrants (Fig 3) but not in the ventrolateral quadrant of the putamen, where MPTP alone had a significant effect on the D1 mRNA levels (see Fig 3). In this region, MPTP decreased the D1 mRNA expression (see Fig 3). Even if, in other subregions of the striatum, the MPTP had not a significant effect, a decrease in the D1 mRNA levels was found in MPTPlesioned group compared with control group (⫺34% in mean). L-Dopa treatment, in nondyskinetic NHP, did not induce modifications of D1 mRNA expression. The level was similar to D1 mRNA level observed in the MPTP-lesioned group. In dyskinetic NHP, the D1 mRNA level displayed a tendency to increase, and the difference with control animals became less pronounced (see Figs 3A, B). Similar to the D1 mRNA level, D1 binding was not affected by MPTP treatment alone. However, L-dopa administration increased D1 binding in the striata of all MPTP-lesioned animals (eg, nondyskinetic and dyskinetic monkeys) (see Figs 3C, D). Overall, these results showed increased levels of D1 protein in the striatum after L-dopa treatment. In addition, in MPTP-lesioned animals, the expression of D1 mRNA is decreased, though not significantly (⫺34%), in comparison with control animals and was normalized by further L-dopa treatment. This decrease is more marked in the ventrolateral putamen. How- Aubert et al: Increased D1 Signaling in Dyskinesia 21 Fig 3. Effect of 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) intoxication and L-dopa treatment on (A, B) striatal D1 messenger RNA (mRNA) expression (group effect: F(3,119) ⫽ 39.46; p ⬍ 0.0001; interaction between group and striatal subregion: F(21,119) ⫽ 0.51), and (C, D) D1 binding levels (group effect: F(3,135) ⫽ 34.79; p ⬍ 0.0001; interaction between group and striatal subregion: F(21,135) ⫽ 1.12). The results are expressed as the percentage of control animals. For example, in the putamen dorsolateral of control animals, messenger RNA (mRNA) level was of 78.28 ⫾ 10.59 fmol/mg of tissue equivalent, and binding level was 5.74 ⫾ 0.62 cpm/m2 (* ⫽ statistically significant difference compared with control animals; p ⬍ 0.05). (B) Examples of D1 mRNA expression autoradiographs and (D) of D1 binding ␤-imager images at the rostral level of striatum. ever, the D1 binding is not affected by the DA denervation.6,29 Whereas the expression of D1 mRNA is not affected by the L-dopa therapy, this treatment provokes a strong increase in the D1 binding in both nondyskinetic and dyskinetic NHP. These results show dissociation between D1 mRNA and protein, indicating that modification of the levels in D1 mRNA is not necessarily correlated with a comparable change in D1 protein levels, and they suggest a modification of D1 mRNA transcription regulation after chronic L-dopa treatment. Changes in Overall D1 Agonist–Stimulated GTP␥S Binding Although no significant differences in receptor binding were found between nondyskinetic and dyskinetic NHPs, it could still be possible that dyskinesia was due to more subtle changes at the level of dopamine D1 receptor transmission. We have previously hypothesized that D1 receptors are subjected to a differential trafficking13 by showing that L-dopa induces a cytoplasmic localization of D1 receptors in striatal neurons of 6-hydroxydopamine–treated rats and parkinsonian patients, although none of the rats or PD cases were 22 Annals of Neurology Vol 57 No 1 January 2005 dyskinetic at the time of their death.13 We also wondered whether those D1 receptors have the same sensitivity to pharmacological stimulation. To assess the difference of D1 sensitivity between dyskinetic and nondyskinetic groups, we studied D1 agonist–stimulated GTP␥S binding. The [35S]GTP␥S autoradiography detects functionally active receptors by indicating their ability to interact with G proteins. MPTP and L-dopa treatment had a significant effect on the [35S]GTP␥S binding in the striatum, independently of the D1 agonist concentration (Figs 4A, B). In basal condition, that is, without D1 agonist stimulation, [35S]GTP␥S binding was similar in all four groups studied. MPTP treatment alone increased [35S]GTP␥S binding in the striatum, regardless of the D1 agonist concentration (see Fig 4A). After L-dopa treatment, the [35S]GTP␥S binding was different in nondyskinetic and dyskinetic animals. In the nondyskinetic group, L-dopa reduced the MPTP-induced increase in [35S]GTP␥S binding, which returned to control levels (see Figs 4A, B). In the dyskinetic group, 35 L-dopa induced a strong increase of the [ S]GTP␥S 35 binding. This [ S]GTP␥S binding was higher than Fig 4. Effect of 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) intoxication and L-dopa treatment on the striatal D1 agonist–stimulated GTP␥S binding. (A) Raw values (0.1M, F(3,14) ⫽ 26.45, p ⬍ 0.0001; 0.3M, F(3,14) ⫽ 64.08, p ⬍ 0.0001; 1M, F(3,14) ⫽ 22.49, p ⬍ 0.0001; 3M, F(3,14) ⫽ 58.09, p ⬍ 0.0001; 10M, F(3,14) ⫽ 27.64, p ⬍ 0.0001), and (C) normalized values (raw values/D1 binding) (0.1M, F(3,14) ⫽ 30.26, p ⬍ 0.0001; 0.3M, F(3,14) ⫽ 61.92, p ⬍ 0.0001; 1M, F(3,14) ⫽ 21.70, p ⬍ 0.0001; 3M, F(3,14) ⫽ 51.06, p ⬍ 0.0001; 10M, F(3,14) ⫽ 15.80, p ⫽ 0.0003) of D1stimulated GTP␥S binding expressed in cpm/m2 (* ⫽ statistically significant difference compared with control animals, p ⬍ 0.05, # ⫽ significant difference compared with MPTP-alone animals, p ⬍ 0.05; § ⫽ statistically significant difference compared with nondyskinetic animals, p ⬍ 0.05). (B) Examples of D1-stimulated (1.0M) GTP␥S binding images obtained with the ␤-imager at the rostral level of striatum. that in the control group, in the MPTP group, and in the nondyskinetic group (see Figs 4A, B). Normalized D1 Agonist–Stimulated GTP␥S Binding Is Linearly Related to L-Dopa–Induced Dyskinesia Severity However, D1 binding levels are increased in dyskinetic NHP (see Fig 3). The increased sensitivity thus might well be the direct consequence of the increased number of proteins or of an increased availability of receptor at the membrane surface,13 and not necessarily of sensitivity. Consequently, we normalized the GTP␥S binding according to the D1 binding. Once normalized, the results were different, even if MPTP and L-dopa treatment still had significant effects (see Fig 4C). MPTP lesions induced an increase of the [35S]GTP␥S binding only for the low D1 agonist concentrations (see Fig 4C), and for the other concentration the [35S]GTP␥S binding was similar to the control binding. In the nondyskinetic NHP, the [35S]GTP␥S binding was decreased compared with the control group (see Fig 4C) and to the MPTP-lesioned group (see Fig 4C). In the dyskinetic group, L-dopa treatment increased the [35S]GTP␥S binding above the levels reported in the control group, showing that the sensitivity per receptor is increased in the dyskinetic situation (see Fig 4C). Interestingly, these [35S]GTP␥S binding values were similar to the values obtained in the MPTP-alone group. These results suggest that the L-dopa treatment does not act in the same way on the D1 receptor sensitivity. In nondyskinetic animals, L-dopa would induce a decrease of D1 sensitivity, whereas in dyskinetic animals, D1 sensitivity is strongly increased. Moreover, the D1 agonist–stimulated GTP␥S binding levels in putamen correlated with occurrence and severity of LID (r2 ⫽ 0.96, p ⬍ 0.05, n ⫽ 8). These results show that LIDs were accompanied by an increased responsiveness of the D1-mediated signaling, a result to compare with the previously demonstrated correlation between D3 receptor binding and LID in the same experimental conditions.16 Thus, these results suggest a correlation between D1 supersensitivity and LID. Increased Cyclin-Dependent Protein Kinase 5 and DARPP-32 Levels Are Associated with D1 Receptor Supersensitivity and L-Dopa–Induced Dyskinesia Recent evidence indicates that Cdk5 is involved in long-term synaptic changes. Moreover, this kinase participates in the regulation of DARPP-32, a DA- and cAMP-regulated phosphoprotein of 32kDa that plays a critical role in DA D1 receptor–mediated transmission by modulating the state of phosphorylation and activity of a variety of downstream physiological effectors.30 Unfortunately, quantitative assessments of phosphorylated DARPP-32 and effector proteins are not possible in our animals. Indeed, accurate measurement of phosphorylation state is prevented by the time required to dissect the brains, which is known to be crucial. Microwave killing is not available for the primate, and this technique is mandatory for studying phosphorylated proteins. We therefore determined the total levels of Cdk5, p35, a Cdk5 activator, and DARPP-32 in the striata of NHP by western immunoblotting. We found that the striata of animals affected by LID contained significantly higher levels of Cdk5 and DARPP-32 compared with the striata of control, MPTP-lesioned, and MPTP-lesioned nondyskinetic NHP (Fig 5). No Aubert et al: Increased D1 Signaling in Dyskinesia 23 Fig 5. Effect of 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) intoxication and L-dopa treatment on the levels of Cdk5 and DARPP-32. Lower panels show autoradiograms of Western blots for all animals tested (four in the control group, four in the MPTP-lesioned group, four in the nondyskinetic group, and three in the dyskinetic group) obtained using polyclonal antibodies against Cdk5 (A) or DARPP-32 (B). Upper panels show the amount of Cdk5 and DARPP-32 as median ⫾ standard deviation and are expressed as percentage of control group. * ⫽ p ⬍ 0.01 (One-way analysis of variance followed by Newman–Keuls test; Cdk5: F(3,14) ⫽ 8.09; DARPP-32: F(3,14) ⫽ 7.73). difference was observed between the levels of p35 in the different experimental groups (data not shown). Discussion In this study, we have utilized the MPTP-lesioned monkey model of PD to demonstrate that the severity of LID is linearly correlated with normalized D1 agonist–stimulated GTP␥S binding levels. Furthermore, we provide evidence indicating that the levels of Cdk5 and DARPP-32, two pivotal players in DA signaling, are increased in the striata of dyskinetic animals. Together, these data suggest that LID is caused by increased G protein coupling efficiency at the level of DA D1 receptors, most likely occurring in the striatonigral neurons of the direct pathway. Interestingly, from the point of view of D1 receptor levels, we cannot distinguish the dyskinetic from the nondyskinetic monkeys. MPTP intoxication induces a significant increase of D1 agonist–stimulated [35S]GTP␥S binding without affecting the number of D1 receptors. This finding suggests that striatal DA depletion induces a sensitization of D1 receptors. In nondyskinetic animals, the D1 agonist–induced [35S]GTP␥S binding is decreased, becoming lower than that in control animals. In the nondyskinetic situation, the enhancement of the D1 protein expression, induced by L-dopa, seems to offset a decrease in the GTP binding, suggesting that the D1 receptors are either desensitized or subjected to a differential trafficking.13 Although the phenomenon of receptor desensitization is common, it cannot simply be explained by the mechanism of homologous desensitization through arrestin– receptor interaction31 since there are many potential regulators of receptor–G protein coupling, that is, by definition, all scaffolding proteins present in a synapse such as the membrane-associated guanylate kinase su- 24 Annals of Neurology Vol 57 No 1 January 2005 perfamily of synapse-associated proteins.32 It nevertheless remains that internalization and desensitization would represent two aspects of a common mechanism that either fails or is overactive in dyskinetic monkeys, representing an attempt of the system to compensate for DA receptor overstimulation. Theoretically, if desensitization is impaired or delayed, DA receptors would appear functionally supersensitive. This hypothesis is in agreement with our results since despite the absence of difference of D1 protein expression between dyskinetic and nondyskinetic NHP, the D1 agonist– induced [35S]GTP␥S binding is strongly increased, showing a D1 supersensitivity. In fact, it is linearly correlated with LID severity. Thus, in dyskinetic animals, internalization of the D1 receptor is probably associated with the reported increase in the activity of signal transduction pathway of the still functional receptors present at the membrane surface.10,11,33 Such a mechanism, though compensatory in nature, fails since NHPs are dyskinetic. We show that the signal transduction pathway is increased in dyskinetic animals. Previous work performed in the 6-OHDA–lesioned rat model of PD showed that, in dyskinetic animals, DARPP-32 is hyperphosphorylated at Thr34,12 but its expression levels are not modified at all.12 These results show, however, an increased expression of DARPP-32 in dyskinetic animals, a finding that is at odds with rodent data.12 Such a discrepancy could be attributed to a species difference to a certain extent, but more likely to a difference in duration of dopamimetic treatment. Indeed, rats were treated with L-dopa for 16 to 22 days,12 whereas our monkeys were treated for several months, a time frame that allows further dysregulation of signaling cascade. We have not been able to address the phosphorylation issue for technical reasons (see above), but we could at least assume that the D1 receptor pulsatile stimulation by L-dopa induces an increased phosphorylation of DARPP-32 at Thr34 through protein kinase A (PKA) activation.30 Cdk5 phosphorylates DARPP-32 at Thr75,34 thereby converting this phosphoprotein into an inhibitor of PKA.34 We hypothesize that the increase in Cdk5 expression found in dyskinetic NHP may represent a homeostatic response to hyperactivation of the D1–PKA pathway. Interestingly, the role of Cdk5 has been studied in another model of hyperdopaminergia, that is, in animals sensitized after chronic exposure to cocaine.35 They found a reduction in PKA-dependent phosphorylation of DARPP-32 (at Thr34) in striatal tissue dissected from rats chronically treated with cocaine. This effect was proposed to depend on increased Cdk5 expression, phosphorylation of DARPP-32 at Thr75, and inhibition of PKA.35 We hypothesize that a similar mechanism is present in dyskinetic animals, where Cdk5 is also overexpressed. If Cdk5 regulates activation of the D1–PKA pathway in the striatal neuron, it would negatively regulate DA release from DA terminals.36 Although few terminals remain in the MPTP-denervated striatum, Cdk5 might attempt controlling newly formed DA from exogenous L-dopa. However, these two negative feedbacks at both presynaptic and postsynaptic levels are not efficient enough to correct for D1 hyperactivation. The hypothesis that Cdk5 overexpression represents a compensatory mechanism remains to be experimentally addressed. This study clearly shows that, whereas D2 receptor levels are not significantly impaired by L-dopa treatment, the D1 receptor expression, sensitivity, and integrity of signaling cascade is modified by the chronic pharmacological stimulation. Interestingly, we have recently shown in the very same animals that the DA D3 receptor binding level is also linearly correlated with the severity of LID.16 D3 receptor mRNA is expressed in the striatal medium spiny neurons of the direct pathway, that is, those that express the D1 receptor.37 Considering that (1) the D1 receptor expression is increased after L-dopa treatment, (2) that the sensitivity of the D1 signaling cascade is enhanced in LID, (3) that D1 and D3 receptors are likely coexpressed in the direct pathway neurons, and (4) that D2 receptor levels expressed by medium spiny neurons of the indirect pathway are neither normalized nor increased after L-dopa treatment, these results support the hypothesis of a predominant role for the direct pathway in LID manifestation. Moreover, our results lead us to hypothesize that Cdk5-increased expression in LID is compensatory upon DARPP-32 activity. Unraveling of this machinery may have tremendous consequences on the development on therapeutic tools for the management of LID. This work was supported by grants from the Michael J. Fox Foundation for Parkinson Research (C.G., B.B., E.B.), the Fédération pour la Recherche sur le Cerveau (C.E.G., E.B.) and the Fondation pour la Recherche Médicale (E.B., C.E.G.). The University Victor Segalen, the Centre National de la Recherche Scientifique, the IFR of Neuroscience (Institut National de la Sante et de la Recherche Médicale No. 8; Centre National de la Recherche Scientifique No. 13) provided the infrastructural support. We thank L. Cardoit and C. Imbert for technical assistance. The authors declare that they have no competing financial interests. References 1. Birkmayer W, Hornykiewicz O. The L-dihydroxyphenylalanine (L-dopa) effect in Parkinson’s syndrome in man: on the pathogenesis and treatment of Parkinson akinesis. Arch Psychiat Nervenkr 1962;203:560 –574. 2. Birkmayer W, Hornykiewicz O. The L-3,4-dioxyphenylalanine (dopa)-effect in Parkinson-akinesia. Wien klin Wschr 1961;73: 787–788. 3. Lees AJ. Levodopa substitution: the gold standard. Clin Neuropharmacol 1994;17:S1–S6 4. Bezard E, Brotchie JM, Gross CE. Pathophysiology of levodopa-induced dyskinesia: potential for new therapies. Nat Rev Neurosci 2001;2:577–588. 5. Creese I, Burt DR, Snyder SH. Dopamine receptor binding enhancement accompanies lesion-induced behavioral supersensitivity. Science 1977;197:596 –598. 6. Lee T, Seeman P, Rajput A, et al. Receptor basis for dopaminergic supersensitivity in Parkinson’s disease. Nature 1978;273: 150 –151. 7. Yahr MD, Duvoisin RC, Hoehn MM, et al. L-Dopa (L-3,4dihydroxyphenylanine): its clinical effects in parkinsonism. Trans Am Neurol Assoc 1968;93:56 – 63. 8. Rascol O, Brooks DJ, Korczyn AD, et al. A five-year study of the incidence of dyskinesia in patients with early Parkinson’s disease who were treated with ropinirole or levodopa. N Engl J Med 2000;342:1484 –1491. 9. Gerfen CR, Engber TM, Mahan LC, et al. D1 and D2 dopamine receptor-regulated gene expression of striatonigral and striatopallidal neurons. Science 1990;250:1429 –1432. 10. Gerfen CR. Molecular effects of dopamine on striatalprojection pathways. Trends Neurosci 1995;23:S64 –S70. 11. Berke JD, Paletzki RF, Aronson GJ, et al. A complex program of striatal gene expression induced by dopaminergic stimulation. J Neurosci 1998;18:5301–5310. 12. Picconi B, Centonze D, Hakansson K, et al. Loss of bidirectional striatal synaptic plasticity in L-dopa-induced dyskinesia. Nat Neurosci 2003;6:501–506. 13. Muriel MP, Bernard V, Levey AI, et al. Levodopa induces a cytoplasmic localization of D1 dopamine receptors in striatal neurons in Parkinson’s disease. Ann Neurol 1999;46: 103–111. 14. Rascol O, Blin O, Thalamas C, et al. ABT-431, a D1 receptor agonist prodrug, has efficacy in Parkinson’s disease. Ann Neurol 1999;45:736 –741. 15. Rascol O, Nutt JG, Blin O, et al. Induction by dopamine D-1 receptor agonist ABT-431 of dyskinesia similar to levodopa in patients with Parkinson disease. Arch Neurol 2001;58: 249 –254. 16. Bezard E, Ferry S, Mach U, et al. Attenuation of levodopainduced dyskinesia by normalizing dopamine D3 receptor function. Nat Med 2003;9:762–767. Aubert et al: Increased D1 Signaling in Dyskinesia 25 17. Boraud T, Bezard E, Bioulac B, Gross C. Dopamine agonistinduced dyskinesias are correlated to both firing pattern and frequency alteration of pallidal neurons in the MPTP-treated monkey. Brain 2001;124:546 –557. 18. Bezard E, Dovero S, Prunier C, et al. Relationship between the appearance of symptoms and the level of nigrostriatal degeneration in a progressive MPTP-lesioned macaque model of Parkinson’s disease. J Neurosci 2001;21:6853– 6861. 19. Abercrombie E. Estimation of nuclear population from microtome sections. Anat Rec 1946;94:239 –247. 20. Charpak G, Dominik W, Zaganidis N. Optical imaging of the spatial distribution of ␤-particles emerging from surfaces. Proc Natl Acad Sci U S A 1989;86:1741–1745. 21. Theler JM, Lakhdar-Ghazal N, Pevet P, et al. Mapping of [3H]vasopressin binding sites in the brain of jerboa (Jaculus orientalis) by an high resolution ␤-radio imager. J Neurosci Methods 1993;49:231–240. 22. Aubert I, Ghorayeb I, Normand E, Bloch B. Phenotypical characterization of the neurons expressing the D1 and D2 dopamine receptors in the monkey striatum. J Comp Neurol 2000; 418:22–32. 23. Dearry A, Gingrich JA, Falardeau P, et al. Molecular cloning and expression of the gene for a human D1 dopamine receptor. Nature 1990;347:72–76. 24. Grandy DK, Marchionni MA, Makam H, et al. Cloning of the cDNA and gene for a human D2 dopamine receptor. Proc Natl Acad Sci U S A 1989;86:9762–9766. 25. Gross CE, Ravenscroft P, Dovero S, et al. Pattern of levodopainduced striatal changes is different in normal and MPTPlesioned mice. J Neurochem 2003;84:1246 –1255. 26. Brooks DJ, Ibanez V, Sawle GV, et al. Striatal D2 receptor status in patients with Parkinson’s disease, striatonigral degeneration, and progressive supranuclear palsy, measured with 11C-raclopride and positron emission tomography. Ann Neurol 1992;31:184 –192. 26 Annals of Neurology Vol 57 No 1 January 2005 27. Shinotoh H, Hirayama K, Tateno Y. Dopamine D1 and D2 receptors in Parkinson’s disease and striatonigral degeneration determined by PET. Adv Neurol 1993;60:488 – 493. 28. Graham WC, Sambrook MA, Crossman AR. Differential effect of chronic dopaminergic treatment on dopamine D1 and D2 receptors in the monkey brain in MPTP-induced parkinsonism. Brain Res 1993;602:290 –303. 29. Shinotoh H, Inoue O, Hirayama K, et al. Dopamine D1 receptors in Parkinson’s disease and striatonigral degeneration: a positron emission tomography study. J Neurol Neurosurg Psychiatry 1993;56:467– 472. 30. Greengard P, Allen PB, Nairn AC. Beyond the dopamine receptor: the DARPP-32/protein phosphatase-1 cascade. Neuron 1999;23:435– 447. 31. Luttrell LM, Lefkowitz RJ. The role of ␤-arrestins in the termination and transduction of G-protein-coupled receptor signals. J Cell Sci 2002;115:455– 465. 32. Craven SE, Bredt DS. PDZ proteins organize synaptic signaling pathways. Cell 1998;93:495– 498. 33. Walaas SI, Aswaad SW, Greengard P. A dopamine- and cyclic AMP-regulated phosphoprotein enriched in dopamineinnervated brain regions. Nature 1984;301:69 –71. 34. Bibb JA, Snyder GL, Nishi A, et al. Phosphorylation of DARPP-32 by Cdk5 modulates dopamine signalling in neurons. Nature 1999;402:669 – 671. 35. Bibb JA, Chen J, Taylor JR, et al. Effects of chronic exposure to cocaine are regulated by the neuronal protein Cdk5. Nature 2001;410:376 –380. 36. Chergui K, Svenningsson P, Greengard P. Cyclin-dependent kinase 5 regulates dopaminergic and glutamatergic transmission in the striatum. Proc Natl Acad Sci USA 2004;101:2191–2196. 37. Bordet R, Ridray S, Schwartz JC, Sokoloff P. Involvement of the direct striatonigral pathway in levodopa-induced sensitization in 6-OHDA-lesioned rats. Eur J Neurosci 2000;12: 2117–2123.