at SciVerse ScienceDirect

Neuropharmacology 73 (2013) 216e231Contents lists availableNeuropharmacology

journal homepage: www.elsevier .com/locate/neuropharmChronic treatment with MPEP, an mGlu5 receptor antagonist,normalizes basal ganglia glutamate neurotransmission inL-DOPA-treated parkinsonian monkeys

Nicolas Morin a,b, Marc Morissette a, Laurent Grgoire a, Baltazar Gomez-Mancilla c,Fabrizio Gasparini c, Thrse Di Paolo a,b,*aNeuroscience Research Unit, Laval University Medical Center (CHUQ), Quebec, QC, Canadab Faculty of Pharmacy, Laval University, Quebec, QC, CanadacNovartis Institute for BioMedical Research, Basel, Switzerlanda r t i c l e i n f o

Article history:Received 29 October 2012Received in revised form17 May 2013Accepted 18 May 2013

Keywords:mGlu5 receptormGlu2/3 receptorNMDA receptorAMPA receptorL-DOPAMPEPAbbreviations: PD, Parkinsons disease; DA,dihydroxyphenylalanine; LID, L-DOPA-induced dyskglutamate; MPEP, 2-methyl-6-(phenylethynyl)pyridin1,2,3,6-tetrahydropyridine; Bmax, density; Kd,hydroxydopamine; MTEP, 3-[(2-methyl-1,3-thiazoglobus pallidus; GPi, internal globus pallidus; GPe, extN-methyl-D-aspartate; DM, dorsomedial; VM, ventrodorsolateral; AMPA, a-amino-3-hydroxy-5-methyl-4-* Corresponding author. Neuroscience Research U

Center (CHUQ), 2705 Laurier Blvd., Quebec, Qc, Canad2296; fax: 1 418 654 2761.

E-mail addresses: therese.dipaolo@crchul.upha.ulaval.ca (T. Di Paolo).

0028-3908/$ e see front matter 2013 Elsevier Ltd.http://dx.doi.org/10.1016/j.neuropharm.2013.05.028a b s t r a c t

Metabotropic glutamate 5 (mGlu5) receptor antagonists reduce L-3,4-dihydroxyphenylalanine (L-DOPA)-induced dyskinesias (LID) in Parkinsons disease (PD). The aim of this study was to investigate the long-term effect of the prototypal mGlu5 receptor antagonist 2-methyl-6-(phenylethynyl)pyridine (MPEP) onglutamate receptors known to be involved in the development of LID in the de novo chronic treatment ofmonkeys lesioned with 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP). MPTP monkeys weretreated for one month with L-DOPA and developed dyskinesias while those treated with L-DOPA andMPEP (10 mg/kg) developed significantly less. Normal control and saline-treated MPTP monkeys werealso included. All MPTP monkeys were extensively and similarly denervated. The basal ganglia [3H]ABP688 specific binding (mGlu5 receptors) was elevated in L-DOPA-treated MPTP monkeys compared tocontrols but not in those treated with L-DOPA and MPEP; dyskinesia scores of these monkeys correlatedpositively with their [3H]ABP688 specific binding. Striatal density (Bmax) of [3H]ABP688 specific bindingincreased in L-DOPA-treated MPTP monkeys compared to other groups and affinity (Kd) remained un-changed. Striatal mGlu5 receptor mRNA remained unchanged following treatments. Elevated basalganglia specific binding of [3H]Ro 25-6981 (NMDA NR1/NR2B receptors), [3H]Ro 48-8587 (AMPA re-ceptors) but not [3H]CGP-39653 (NMDA NR1/NR2A receptors) was observed only in L-DOPA-treatedMPTP monkeys; dyskinesias scores correlated with binding. By contrast, basal ganglia [3H]LY341495specific binding (mGlu2/3 receptors) decreased in L-DOPA-treated MPTP monkeys compared to controls,saline and L-DOPA MPEP treated MPTP monkeys; dyskinesias scores correlated negatively with thisbinding. Hence, chronic MPEP treatment reduces the development of LID and is associated with anormalization of glutamate neurotransmission.

2013 Elsevier Ltd. All rights reserved.dopamine; L-DOPA, L-3,4-inesias; mGlu, metabotropice; MPTP, 1-methyl-4-phenyl-

affinity; 6-OHDA, 6-l-4-yl)ethynyl]pyridine; GP,ernal globus pallidus; NMDA,medial; VL, ventrolateral; DL,isoxazolepropionic acid.nit, Laval University Medicala G1V 4G2. Tel.: 1 418 654

laval.ca, Therese.DiPaolo@

All rights reserved.1. Introduction

Parkinsons disease (PD) is a progressive neurodegenerativedisorder characterized by tremor, rigidity, and bradykinesia and islikely to increase due to the aging of populations (de Lau andBreteler, 2006; Siderowf and Stern, 2003). Instability in posturalreflexes is primarily attributed to loss of dopamine (DA) neurons inthe substantia nigra compacta (Wichmann and DeLong, 2003).Although L-3,4-dihydroxyphenylalanine (L-DOPA) remains the goldstandard for symptomatic treatment of PD (Mercuri and Bernardi,2005), various complications including motor fluctuations andabnormal involuntary movements, such as L-DOPA-induced dys-kinesias (LID), limit the quality of life in PD patients and can be very

Delta:1_given nameDelta:1_surnameDelta:1_given nameDelta:1_surnameDelta:1_given nameDelta:1_surnamemailto:therese.dipaolo@crchul.ulaval.camailto:Therese.DiPaolo@pha.ulaval.camailto:Therese.DiPaolo@pha.ulaval.cahttp://crossmark.dyndns.org/dialog/?doi=10.1016/j.neuropharm.2013.05.028&domain=pdfwww.sciencedirect.com/science/journal/00283908http://www.elsevier.com/locate/neuropharmhttp://dx.doi.org/10.1016/j.neuropharm.2013.05.028http://dx.doi.org/10.1016/j.neuropharm.2013.05.028http://dx.doi.org/10.1016/j.neuropharm.2013.05.028

N. Morin et al. / Neuropharmacology 73 (2013) 216e231 217difficult to manage (Fabbrini et al., 2007; Gottwald and Aminoff,2011; Meissner et al., 2011). Glutamatergic transmission isincreased in the basal ganglia in PD (Klockgether and Turski, 1993)and is also believed to be involved in LID (Calon et al., 2003; Chaseand Oh, 2000). Amantadine, a non-competitive antagonist at N-methyl-D-aspartate (NMDA) receptors, is currently the only drugused in the clinic shown to havemodest benefit in LID (Crosby et al.,2003; Stocchi et al., 2008; Verhagen Metman et al., 1998b).Amantadine is also reported to reduce LID in 6-hydroxydopamine(6-OHDA)-lesioned rats (Dekundy et al., 2007; Lundblad et al.,2002) and in non-human primate models of PD and LID (Blanchetet al., 1998). However, NMDA antagonists have significant adverseeffects in many patients such as cognitive impairment, whichsignificantly limits their use (Stocchi et al., 2008).

In addition to ionotropic receptors, metabotropic glutamate(mGlu) receptors are of particular interest because of their abun-dance in the basal ganglia and because of their higher affinity forglutamate than ionotropic glutamate receptors (Conn et al., 2005;Marino et al., 2002). The mGlu receptors constitute a family of G-protein coupled receptors comprising 8 subtypes that are classifiedinto 3 groups based on the signal transduction pathway, homologyof the amino acid sequence and receptor pharmacology. The ma-jority of mGlu receptors of Group I (w90%), including mGlu5 re-ceptor, appear to be located post-synaptically on the peri-synapticannulus of dendritic spines, which lead to enhanced neuronalexcitation (Lujan et al., 1997). By contrast, pre-synaptically localizedGroup II mGlu receptors, including mGlu2/3 receptor, are thoughtto represent the classical inhibitory autoreceptor mechanism sup-pressing excess glutamate release from pre-synaptic terminals(Schoepp, 2001).

mGlu5 receptor specific binding is increased in the basal gangliaof parkinsonian monkeys with LID and in parkinsonian patientswith motor complications (Ouattara et al., 2010a, 2011; Samadiet al., 2008b). mGlu5 receptor antagonists 2-methyl-6-(phenyl-ethynyl)pyridine (MPEP) and 3-[(2-methyl-1,3-thiazol-4-yl)ethy-nyl]pyridine (MTEP) inhibit LID in the 6-OHDA-lesioned rat model(Dekundy et al., 2006; Gravius et al., 2008; Levandis et al., 2008;Mela et al., 2007). The mGlu5 receptor antagonists MPEP, MTEP,fenobam and AFQ056 (mavoglurant) reduce LID in 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP)-lesioned monkeys(Gregoire et al., 2011; Johnston et al., 2010; Morin et al., 2010;Rylander et al., 2010) or prevent the development of L-DOPAinduced motor complications (Morin et al., 2013a). Moreover,mavoglurant and recently ADX-48621 (dipraglurant) were shownto reduce LID in parkinsonian patients (Addex Therapeutics, 2012;Berg et al., 2011).

The mechanisms underlying the development of LID are stillunknown, but evidence suggests that LID is the result of mal-adaptive plasticity at striatal synapses (Calabresi et al., 2010; Cenciand Lundblad, 2006; Iravani et al., 2012; Jenner, 2008). Altereddopaminergic and nondopaminergic neurotransmission in thebasal ganglia are observed in LID (Blandini and Armentero, 2012).Glutamate neurotransmission plays a crucial role in themodulationof corticostriatal inputs and striatal output to downstream nuclei ofthe basal ganglia circuit (Blandini and Armentero, 2012). Increasedglutamate transmission involves pre-synaptic changes, such asincreased striatal concentration of extracellular glutamate (Dupreet al., 2011) or changes in mGlu2/3 receptors (Samadi et al.,2008a, 2009). Post-synaptic changes in a-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA), NMDA and mGlu5 re-ceptors also seem to play a major role in the development of LID(Duty, 2012). mGlu5 receptors are highly expressed in the striatumand other basal ganglia nuclei including subthalamic nucleus (STN),substantia nigra and globus pallidus (GP) (Ferraguti and Shigemoto,2006). Numerous interactions betweenmGlu5 receptor and NMDA,D1R DA, D2R DA, A2A adenosine receptors suggest that these re-ceptors may function together as closely associated signalingpartners in the appearance of LID (Samadi et al., 2007). mGlu5 re-ceptor activation of NMDA receptors (Pisani et al., 2001) as well asNMDA and mGlu5 receptor co-localization are described (Perroyet al., 2008). The blockade of mGlu5 receptors leads to anti-dyskinetic actions, which is associated with normalization of firingof striatal signals (Duty, 2012).

mGlu5 receptor antagonists are very promising drugs for themanagement of various brain disorders and one of its potentialapplications is LID therapy. Therefore, it is essential to know thelong-term behavioral effects of this class of compounds andensuing possible biochemical adaptations of the brain; it is criticalto learn as much as possible from MPTP primate models, whichclosely mimic human PD conditions (Morin et al., 2013b). Werecently reported that development of LID over a month of treat-ment was lower by overall 72%with addition ofMPEP to the L-DOPAtreatment in de novo MPTP monkeys (Morin et al., 2013a). mGlu5receptor specific binding increased in the putamen of these MPTPmonkeys treated with L-DOPA and was prevented with the L-DOPA MPEP treatment (Morin et al., 2013a). The mechanismunderlying the antidyskinetic activity of mGlu5 receptor antagonistis hypothesized to be related to the blockade of mGlu5 receptorsand its interactions with other receptors that overall restore anormal glutamate neurotransmission. The aim of the present studywas thus to evaluate if a chronic treatment with L-DOPA MPEPcan normalize more generally glutamate neurotransmission with afocus on mGlu5, mGlu2/3, NMDA and AMPA receptors in the basalganglia of MPTP monkeys.2. Materials and methods

2.1. Animals and treatments

This experiment was carried out using 18 drug naive female ovariectomizedmonkeys (Macaca fascicularis) (aged 4.7e7.7 y; weight: 2.7e4.2 kg; CovanceResearch Products, China) in agreement with the National Institute of Health Guidefor the Care and Use of Laboratory Animals. The motor behavior of these monkeyswas previously reported (Morin et al., 2013a). Animals were maintained in atemperature-controlled environment and artificial daylight (lights on 06:00e18:00 h). The Laval University committee for the protection of animals approved thisstudy. The animals were divided into four experimental groups. One non-parkinsonian group (n 4) remained untreated and served as healthy controls(control group), and three groups were rendered parkinsonian with the neurotoxinMPTP (SigmaeAldrich Canada Ltd., Oakville, ON, Canada) administered continuouslyusing subcutaneous Alzet minipumps (0.5 mg/24 h in saline solution) until sus-tained parkinsonian features were achieved. Treatment was started after the bilat-eral parkinsonian syndrome had stabilized (after an average period of 6.6 months).Five of these MPTP monkeys remained untreated and received saline (saline MPTPgroup), while four others received a chronic daily treatment with L-DOPA/benser-azide alone (100/25 mg; Prolopa) (Hoffmann-La Roche limited, Mississauga, ON,Canada) and the five remaining animals were treated with L-DOPA/benserazide(same dosage) and MPEP (10 mg/kg, administered 15 min prior to L-DOPA, providedby Novartis Pharma AG, Basel, Switzerland). MPEP and L-DOPA/benserazide watersolutions were prepared freshly on every experimental day and administered bynasogastric gavage. The same dose of L-DOPA/benserazide per day was administeredto each monkey in the L-DOPA and L-DOPA MPEP groups. Since there was nodifference in the weight of monkeys between the L-DOPA and the L-DOPA MPEP-treated groups (Morin et al., 2013a) they thus received the same amount of L-DOPA/benserazide per kg and per day.

The dose of MPEP administered to the animals was based on our previous resultsin an acute treatment study in dyskinetic MPTP monkeys where MPEP 10 mg/kgreduced already developed LID (Morin et al., 2010).

The animals were first evaluated following vehicle (water) administration alone.Behavioral measures after L-DOPA/benserazide alone or L-DOPA/benserazide andMPEP were performed every other day for 30 days. The animals were observedthrough a one-way screen and were scored immediately every 15 min for anti-parkinsonian (disability scale) and dyskinetic responses (LID scale) for the wholeduration of the L-DOPA response (Morin et al., 2013a). An independent evaluatorblind to treatment assignment rated the behavioral measures.

Moreover, we showed in monkeys of the present study that there was elevatedMPEP blood concentrations at time of the motor effect of L-DOPA as described in the

N. Morin et al. / Neuropharmacology 73 (2013) 216e231218behavioral evaluation of these monkeys recently reported (Morin et al., 2013a).L-DOPA administered to MPTP monkeys induced a decrease (improvement) ofparkinsonian scores after the first administration and this effect was similar with theaddition of MPEP to L-DOPA (Morin et al., 2013a). Moreover, the antiparkinsonianresponse in MPTP-lesioned monkeys treated with L-DOPA MPEP was maintainedduring the one-month treatment as compared to the L-DOPA-treated group. Themean dyskinesia score increased over a month in the L-DOPA-treated group,whereas it was significantly lower by 81%, 65%, 68% and 74% in L-DOPA MPEPcompare to the L-DOPA-treated monkeys, respectively for each consecutive week ofthe one-month treatment (Morin et al., 2013a). The duration of the L-DOPA anti-parkinsonian effect significantly decreased in the L-DOPA-treated group on weeks 3and 4 (a reduction of 49 and 56 min, respectively) compared to the first weeksuggesting wearing-off, while the decrease was not significant in the presence ofMPEP (Morin et al., 2013a).

2.2. Tissue preparation

Animals were killed 24 h after the end of the chronic treatments by an over doseof sodium pentobarbital (30 mg/kg intracardiac; Bimeda-MTC Animal Health Inc.,Cambridge, Ontario, Canada). The brains were then immersed in isopentane(40 C) and stored at80 C until use. Brains were cut into 12-mm coronal sectionson a cryostat (18 C). Samples of the posterior striatum, containing the putamen,the caudate nucleus, the external GP (GPe) and the internal GP (GPi), from levelsA15eA18 as defined in the atlas of Szabo and Cowan (Szabo and Cowan, 1984) wereinvestigated. Slices were thaw-mounted onto SuperFrost Plus slides (Brain ResearchLaboratories, Newton, MA) and stored at 80 C until use for assays.

2.3. Biogenic amine assays

The caudate nucleus and putamen were dissected from coronal brain sectionsand homogenized in 250 ml of 0.1 M HClO4 at 4 C and centrifuged at 10 000 g for20 min to precipitate proteins. The supernatants were kept at 80 C in smallpolyethylene tubes until time of assay, while the pellets were solubilized in 100 ml of0.1 M NaOH for determination of protein content. The concentration of DA and itsmetabolites 3,4-dihydroxyphenylacetic acid (DOPAC), 3-methoxytyramine (3-MT)and homovanillic acid (HVA) in the caudate nucleus and putamenwere measured byhigh-performance liquid chromatography with electrochemical detection, accord-ing to our previously published procedures (Goulet et al., 1999; Morissette et al.,2006).

2.4. DA transporter (DAT) autoradiography

The DAT was labeled with 3b-[4-125I-iodophenyl]tropane-2b-carboxylic acidisopropyl ester ([125I]RTI-121; 2200 Ci/mmol, Mandel, Boston, MA) binding ac-cording to our previously published procedure (Calon et al., 2002a; Morissette et al.,2006). Specific binding to the DAT was measured using 25 pM [125I]RTI-121. Non-specific binding was determined in the presence of 100 nM mazindol (SandozPharmaceuticals, Dorval, Quebec, Canada) added to the incubation buffer. Allexperimental groups were simultaneously analyzed in the same assay. A standardgrayscale strip was used to generate a calibration curve for optical densities. Forquantification of autoradiograms with [125I]RTI-121, optical gray densities weretransformed into nCi/mg of tissue equivalent using a standard curve generated withstandard [125I]-strips (Amersham, USA).

2.5. [3H]ABP688 autoradiography

mGlu5 receptor specific binding was evaluated with [3H]ABP688 binding(81.6 Ci/mmol, Novartis, Switzerland), a high-affinity and selective mGlu5 receptorantagonist (Hintermann et al., 2007), according to our published conditions(Ouattara et al., 2010a, 2011) using 5 nM [3H]ABP688 and 10 mM MPEP to estimatenon-specific binding. Slide-mounted tissue sections were exposed to BioMax MRfilms (Kodak, USA; along with tritium standards [3H]-microscales, Amersham, USA)for 20 days at room temperature.

2.6. [3H]ABP688 binding assay to mGlu5 receptors in tissue homogenates

The caudate nucleus and putamen were dissected from coronal brain sectionsand homogenized individually with a glasseTeflon homogenizer in 100 vol. (w/v) ofKrebseRinger Hepes buffer (KRH buffer) containing 20 mM Hepes (2-[4-(2-hydroxyethyl)-1-piperazinyl] ethanesulfonic acid), 118 mM NaCl, 4.8 mM KCl,2.5 mM CaCl2, 1.2 mM MgSO4, and 10 mM NaOH, pH 7.4 (Hintermann et al., 2007)and centrifuged at 16 000 g for 15 min at 4 C. Supernatants were discarded andpellets resuspended and centrifuged under the same condition. Supernatants werediscarded and the final pellets were resuspended in 100 vol. of the same KRH in-cubation buffer as described above. To estimate mGlu5 receptors densities (Bmax)and affinities (Kd), [3H]ABP688 (six concentrations, 0.5, 1.0, 2.0, 3.0, 4.0 and 5.0 nM;81.6 Ci/mmol, Novartis, Switzerland) saturation binding isotherms were performedon homogenates supplemented with 0.05 mg/ml bovine serum albumin (BSA). Inthese assays, 40 ml of membrane (20 mg of protein) was incubated in a final volume of1 ml for 60 min at room temperature. Incubation was stopped by rapid filtration(Cell Harvester Me48N, Brandel Co., MD, USA) with three rapid 3 ml washes withKRH buffer at 4 C through Whatman GF/C fiberglass filter (Brandel Co., MD, USA).Filters were placed in scintillation counting vials with 10 ml scintillation cocktail(Ultima Gold High Flash-point LSC cocktail, PerkinElmer Inc., MA, USA). Non-specificbinding was estimated using 10 mM MPEP$HCl. Radioligand binding was quantifiedin a LKB beta-counter with 45e55% efficiency. Protein determinationwas performedby themethod of micro bicinchoninic acid (BCA) protein assay (Thermo Scientific, IL,USA).

2.7. mGlu5 receptors in situ hybridization

A complementary DNA (cDNA) oligonucleotide probe was synthesized corre-sponding to bases 196e240 of M. fascicularis cDNA (accession no. of GenBankdatabase DQ417747) according to our previously published procedure (Ouattaraet al., 2010a). The oligonucleotide was labeled with 35S-ATP (1250 Ci/mmol;Perkin-Elmer, Woodbridge, Ontario, Canada) using a 30-terminal deoxynucleotidyltransferase enzyme kit (GE Heathcare, Baie dUrf, Quebec, Canada). The in situhybridization conditions were as we previously described (Ouattara et al., 2010a).Brain slices were exposed to Kodak BioMax MR films along with standards [14C]-microscale (Amersham, USA) for 15 days at 22 C.

2.8. [3H]LY341495 autoradiography

mGlu2/3 receptor specific binding was evaluated with [3H]LY341495 (40.0 Ci/mmol) (American Radiolabelled Chemicals, MO, MO, USA) binding according to ourpreviously published procedure (Samadi et al., 2008a). Brain sections were incu-bated with 5 nM [3H]LY341495. Non-specific binding was determined in the pres-ence of 1 mM L-glutamic acid (RBI Natick, MD, USA) in the buffer solution. The brainsections were exposed to [3H]-sensitive films along with [3H]-microscale standardsfor 14 days at room temperature.

2.9. [3H]CGP-39653 and [3H]Ro 25-6981 autoradiographies

Autoradiography of NMDA receptors composed of NR1/NR2A and NR1/NR2Bsubunits was performed using [3H]CGP-39653 (50 Ci/mmol, PerkinElmer Inc., MA,USA) and [3H]Ro 25-6981 (25.7 Ci/mmol; Hoffmann-La Roche, Basel, Switzerland)respectively according to our previously published procedures (Calon et al., 2002b).Briefly, sections were incubated with 20 nM of [3H]CGP-39653 or 5 nM [3H]Ro 25-6981 to estimate total binding to the NR1/NR2A and NR1/NR2B subunits respec-tively. [3H]CGP-39653 autoradiography was performed in the absence of Mg2 inorder to bind only to NR2A-containing NMDA receptors (Kendrick et al., 1996; Laurieand Seeburg, 1994). Non-specific binding was estimated by adding respectively500 mMNMDA (NR1/NR2A subunit) or 10 mMRo 04-5595 hydrochloride (Hoffmann-La Roche, Basel, Switzerland) (NR1/NR2B subunit) to the incubation buffers. Thebrain sections were exposed to [3H]-sensitive films along with tritium standards([3H]-microscales) during 10 weeks or 4 weeks at room temperature for [3H]CGP-39653 and [3H]Ro 25-6981 respectively.

2.10. [3H]Ro 48-8587 autoradiography

Autoradiography of AMPA receptor specific binding using [3H]Ro 48-8587 wasperformed according to experimental conditions previously published (Ouattaraet al., 2010b). The brain sections were incubated with 5 nM [3H]Ro 48-8587 (52 Ci/mmol from Novartis, Switzerland). Non-specific binding was determined in thepresence of 50 mM of the AMPA receptor antagonist (1,4-dihydro-6-(1H-imidazol-1-yl)-7-nitro-2,3-quinoxalinedione hydrochloride, YM90K) (Tocris Biosciences, USA).Autoradiogramswere generated by apposing the labeled tissues to BioMaxMR Filmswith tritium standards ([3H]-microscales) for 1 month at room temperature.

2.11. Image, data and statistical analysis

The intensity of all autoradiograms was quantified with a Power Macintosh G4connected to a video camera (XC-77; Sony) and a constant-illumination light tableusing computerized densitometry (NIH image v.1.63). Optical gray densities weretransformed into nCi/mg of tissue using a standard radioactivity curve. Results werefinally converted into fmol/mg of tissue using the radioligand specific activity. Foreach brain regions investigated, we used 4 brain slices for each animal for eachautoradiographic analysis.

For analysis, the caudate nucleus and putamenwere divided in four sub-regions(medialelateral and dorsaleventral) while the GP was divided in its external (GPe)and internal (GPi) sub-regions. The NMDA, AMPA, mGlu5 and mGlu2/3 receptorswere quantified in the striatum and the GP and their sub-regions because of theirabundance in these brain regions and their reported change in PD and LID (Morinet al., 2013a; Ouattara et al., 2009, 2010b; Samadi et al., 2008a, 2008b). Statisticalcomparison of data was performed by two-way analysis of variance with brain sub-regions and treatments as factors, followed by post-hoc pairwise comparisons withFishers probability of least significant difference test, except for the striatal cate-cholamine concentrations where a one-way analysis of variance was performed.

N. Morin et al. / Neuropharmacology 73 (2013) 216e231 219A simple regression model was used to determine coefficients of correlation and thesignificance of the degree of linear relationship between the mean dyskinesia scoresof these monkeys (n 18) and specific binding. The mean dyskinesia scores used inthe correlations was calculated by averaging all 15min scores obtained during all theduration of the antiparkinsonian effect of the L-DOPA treatment of the last week ofbehavioral observations (cumulative score of total period). A value of p 0.05 wasconsidered significant.Fig. 1. Effect of MPTP on striatal DA markers (A) Representative autoradiograms of coronal bof control and MPTP-lesioned monkeys treated with saline, L-DOPA and L-DOPA MPEP. (B)monkeys (n 4), MPTP-lesioned monkeys treated with saline (n 5), L-DOPA (n 4) and Lcontrol values (fmol/mg of tissue) were for the caudate nucleus, DM 0.76, VM 0.80, DL of MPTP lesion and treatments on catecholamine concentrations in the caudate nucleus ahundred percent control values were for caudate nucleus, DA 112.1, DOPAC 4.1, 3-MTMT 11.8, and HVA 132.2 ng/mg protein. ***p < 0.0001 vs. control monkeys. DM, dorso3. Results

3.1. DA denervation

Representative autoradiograms of [125I]RTI-121 binding in con-trols and MPTP-lesioned monkeys are shown in Fig. 1A. MPTPrain sections showing [125I]RTI-121 binding to the DA transporter (DAT) in the striatumEffect of MPTP lesion and treatments on striatal [125I]RTI-121 specific binding in control-DOPA MPEP (n 5). Data are expressed as the mean SEM. One hundred percent0.89, VL 0.75; for the putamen DM 0.69, VM 0.75, DL 0.80, VL 0.85. (C) Effectnd the putamen of the same monkeys. Data are expressed as the mean SEM. One 12.9, and HVA 104.0 ng/mg protein; for putamen, DA 94.3, DOPAC 3.9, 3-

medial; VM, ventromedial; VL, ventrolateral; DL, dorsolateral.

N. Morin et al. / Neuropharmacology 73 (2013) 216e231220

N. Morin et al. / Neuropharmacology 73 (2013) 216e231 221lesion induced a decrease of [125I]RTI-121 specific binding to theDAT similarly in the lateral, medial, dorsal and ventral parts of thecaudate nucleus (F3,14 141.9, p < 0.0001; Fig. 1B) and putamen(F3,14 459.9, p < 0.0001). The three groups of MPTP monkeys hada similar loss of [125I]RTI-121 specific binding (94%, 94%and 97% in the caudate nucleus and 95%, 93% and 97% in theputamen in each group respectively) (Fig. 1B).

An extensive decrease of DA (98%, 99% and 99% in thecaudate nucleus: F3,14 292.3, p < 0.0001 and 99%, 98%and 99% in the putamen: F3,14 590.5, p < 0.0001) and its me-tabolites DOPAC (93%, 89% and 84% in the caudate nucleus:F3,14 211.3, p< 0.0001 and92%,87% and88% in the putamen:F3,14 321.7, p < 0.0001), 3-MT (96%, 90% and 92% in thecaudate nucleus: F3,14 165.0, p < 0.0001 and 90%, 77%and 79% in the putamen: F3,14 43.0, p < 0.0001) and HVA(89%, 83% and 75% in the caudate nucleus: F3,14 69.1,p < 0.0001 and 77%, 68% and 79% in the putamen: F3,14 31.0,p < 0.0001) were measured in MPTP-lesioned monkeys, comparedto the controls (Fig. 1C). Thus, the three groups of MPTP monkeyshad a similar loss of DA and its metabolites.

3.2. mGlu5 receptor

Representative autoradiograms of basal ganglia [3H]ABP688binding in control and the groups of MPTP-lesioned monkeys aswell as non-specific binding are shown in Fig. 2A. Following L-DOPAtreatment, [3H]ABP688 specific binding increased significantly andsimilarly in sub-regions of the putamen (F3,14 7.18, p < 0.01;Fig. 2B) and GP (F3,14 5.21, p < 0.05; Fig. 2C) of L-DOPA-treatedMPTP monkeys compared to controls as well as saline and L-DOPA MPEP treated MPTP groups (41%, 39% and 28% for theputamen and 66%, 104% and 45% for the GP vs. each grouprespectively). In the caudate nucleus (F3,14 3.69, p< 0.05; Fig. 2B),[3H]ABP688 specific binding increased significantly following L-DOPA treatment in L-DOPA-treated MPTP monkeys compared tocontrols and saline-treated MPTP groups only (24% and 21% vs.each group respectively). By contrast, [3H]ABP688 specific bindingdid not significantly change in the L-DOPA MPEP MPTP groupcompared to controls and saline-treated MPTP monkeys in thecaudate nucleus, putamen and GP. There was a significant positivecorrelation between the mean dyskinesia scores of these monkeys(n 18) and [3H]ABP688 specific binding in the caudate nucleus(r 0.615, p 0.006; Fig. 2D), putamen (r 0.739, p 0.0005;Fig. 2D), GPe (r 0.598, p 0.008; Fig. 2D) and GPi (r 0.641,p 0.004; Fig. 2D).

Examples of [3H]ABP688 saturation binding curves to caudatenucleus and putamen membranes of a control monkey are shownin Fig. 3A. MPTP lesion and treatments did not change the affinity(Kd) in the striatum of these monkeys (F3,26 0.93, p 0.44;Fig. 3B). The density (Bmax) of the saturation binding curves of [3H]ABP688 specific binding in the striatum increased significantly in L-DOPA-treated MPTP monkeys (F3,26 3.60, p 0.028; Fig. 3B)compared to controls, as well as saline and L-DOPA MPEP treatedMPTP groups (17%, 12% and 11% vs. each group respectively)and it did not change in the L-DOPA MPEP group compared tocontrols and saline-treated MPTP monkeys.

Fig. 4A shows examples of in situ hybridization for mGlu5 re-ceptor in these monkeys. The MPTP lesion and treatments leftFig. 2. Autoradiography of brain mGlu5 receptors showing the effect of MPTP lesion, L-DOPcoronal brain sections of control and MPTP-lesioned monkeys treated with saline, L-DOPA anof control monkeys (n 4), and MPTP-lesioned monkeys treated with saline (n 5), L-DOP(GPe) and internal (GPi) globus pallidus of these monkeys. (D) Correlation between mean dynucleus, putamen, GPe and GPi; each point represents an individual monkey. **p < 0.01 antreated MPTP monkeys; p < 0.05 and p < 0.01 vs. MPTP L-DOPA-treated MPTP monkunchanged mRNA levels of mGlu5 receptor both in the caudatenucleus (F3,14 0.19, p 0.91; Fig. 4B) and putamen (F3,14 0.24,p 0.87; Fig. 4B) of controls and all MPTP-lesioned monkeystreated with saline, L-DOPA and L-DOPA MPEP. No effect of lesionand treatments were measured for mGlu5 mRNA levels in the GPeand GPi in these experimental groups (data not shown).

3.3. mGlu2/3 receptor

Representative autoradiograms of basal ganglia [3H]LY341495binding in controls and different groups of MPTP-lesionedmonkeysas well as non-specific binding are shown in Fig. 5A. Following L-DOPA-alone treatment, [3H]LY341495 specific binding decreasedsignificantly in all sub-regions of the caudate nucleus (F3,14 8.37,p < 0.01; Fig. 5B) and putamen (F3,14 8.12, p < 0.01; Fig. 5B)compared to controls as well as saline and L-DOPA MPEP treatedMPTP groups (35%, 48% and 37% for the caudate nucleusand 25%, 19% and 17% for the putamen in each group respec-tively). In both the GPi and GPe, there was a decrease (F3,14 4.10,p < 0.05; Fig. 5C) for the L-DOPA-treated group compared to salineand L-DOPA MPEP-treated MPTP groups (19%, 31% and 33%for theGP in eachgroup respectively). [3H]LY341495 specificbindingdid not significantly change in the L-DOPA MPEP MPTP groupcompared to controls and saline-treated MPTP monkeys in thecaudate nucleus, putamen, GPe and GPi. There was a significantnegative correlation between the mean dyskinesia scores of thesemonkeys (n 18) and [3H]LY341495 specific binding in the caudatenucleus (r 0.739, p 0.0005; Fig. 5D), putamen (r 0.500,p 0.03; Fig. 5D), while no correlation was observed in the GPe(r0.111, p 0.66; Fig. 5D) andGPi (r0.414, p 0.087; Fig. 5D).

3.4. NMDA/NR2B and NMDA/NR2A receptors

Examples of basal ganglia [3H]Ro 25-6981 autoradiography areshown in Fig. 6A. Following L-DOPA treatment, [3H]Ro 25-6981specific binding increased significantly in all sub-regions of thecaudate nucleus (F3,14 7.87, p < 0.01; Fig. 6B), putamen(F3,14 6.84, p< 0.01; Fig. 6B) and GP (F3,14 4.78, p< 0.05; Fig. 6C)of L-DOPA-treated MPTP monkeys compared to controls as well assaline and L-DOPA MPEP treated MPTP groups (33%, 55%and 68% for the caudate nucleus, 45%, 57% and 91% for theputamen and 150%, 129% and 739% for the GP in each grouprespectively). By contrast, [3H]Ro 25-6981 specific binding did notsignificantly change in the L-DOPA MPEP MPTP group comparedto controls and saline-treated MPTP monkeys in the caudate nu-cleus, putamen, GPe and GPi. There was a significant positive cor-relation between the mean dyskinesia scores of these monkeys(n 18) and [3H]Ro 25-6981 specific binding in the caudate nucleus(r 0.569, p 0.01; Fig. 6D), putamen (r 0.757, p 0.0003;Fig. 6D), GPe (r 0.724, p 0.0007; Fig. 6D) and GPi (r 0.667,p 0.003; Fig. 6D).

Representative autoradiograms of [3H]CGP-39653 binding inthe basal ganglia of controls and different groups of MPTP-lesionedmonkeys as well as non-specific binding are shown in Fig. 7A. TheMPTP lesion and treatments left unchanged [3H]CGP-39653 bind-ing of NMDA/NR2A both in the caudate nucleus (F3,14 0.51,p 0.68; Fig. 7B) and putamen (F3,14 0.19, p 0.90; Fig. 7B) ofcontrols and all MPTP-lesioned monkeys treated with saline, L-A alone and with MPEP (A) Representative autoradiograms of [3H]ABP688 binding ind L-DOPA MPEP. (B) [3H]ABP688 specific binding in the caudate nucleus and putamenA (n 4) and L-DOPA MPEP (n 5). (C) [3H]ABP688 specific binding in the externalskinesia scores of these monkeys and their [3H]ABP688 specific binding in the caudated ***p < 0.001 vs. control monkeys; ##p < 0.01 and ###p < 0.001 vs. MPTP saline-eys. DM, dorsomedial; VM, ventromedial; VL, ventrolateral; DL, dorsolateral.

Fig. 3. Saturation for binding to striatal mGlu5 receptors showing the effect of MPTP lesion, L-DOPA alone and with MPEP (A) Example of [3H]ABP688 saturation binding curves tocaudate nucleus and putamen membranes of a control monkey. Each point represents the mean SEM of triplicate determinations. (B) Effect of MPTP lesion and treatments on theaffinity (Kd) and density (Bmax) of the saturation binding curves of [3H]ABP688 specific binding to mGlu5 receptors in the striatum of control monkeys (n 4), and MPTP-lesionedmonkeys treated with saline (n 5), L-DOPA (n 4) and L-DOPA MPEP (n 5). **p < 0.01 vs. control monkeys; #p < 0.05 vs. MPTP saline-treated MPTP monkeys; p < 0.05 vs.MPTP L-DOPA-treated MPTP monkeys.

N. Morin et al. / Neuropharmacology 73 (2013) 216e231222DOPA and L-DOPA MPEP. The GPe and GPi showed weak bindingof [3H]CGP-39653 and no effect of lesion and treatments weremeasured in the experimental groups studied (data not shown).

3.5. AMPA receptor

Examples of basal ganglia [3H]Ro 48-8587 autoradiography areshown in Fig. 8A. Following L-DOPA treatment, [3H]Ro 48-8587specific binding increased significantly in all sub-regions of thecaudate nucleus (F3,14 6.80, p < 0.01; Fig. 8B) and putamen(F3,14 12.44, p < 0.001; Fig. 8B) of L-DOPA-treated MPTP monkeyscompared to control as well as saline and L-DOPA MPEP treatedMPTP groups (23%, 16% and 27% for the caudate nucleusand 35%, 33% and 52% for the putamen vs. each group respec-tively). [3H]Ro 48-8587 specific binding did not change in the L-DOPA MPEP MPTP group compared to controls and saline-treatedMPTPmonkeys in the caudate nucleus and putamen, as observed for[3H]ABP688 and [3H]Ro 25-6981 specific binding. There was a sig-nificant positive correlation between the mean dyskinesia scores ofthese monkeys (n 18) and [3H]Ro 48-8587 specific binding in thecaudate nucleus (r 0.675, p 0.0021; Fig. 8D) and putamen(r 0.773, p 0.0002; Fig. 8D). The GPe and GPi showed weakbinding of [3H]Ro 48-8587 and no effect of lesion and treatmentswere measured in the experimental groups studied (data notshown).

4. Discussion

4.1. Glutamate neurotransmission in PD and LID

Glutamate is the brain most abundant excitatory neurotrans-mitter mediating as much as 70% of synaptic transmission in thecentral nervous system and its overactivity is well documented inPD and LID (Bezard et al., 2001; Blandini and Armentero, 2012;Klockgether and Turski, 1993). The present study showed for thefirst time that a chronic treatment with MPEP in MPTP monkeystreated with L-DOPA normalized changes produced by L-DOPA onseveral ionotropic and metabotropic glutamate receptors (Table 1).All MPTP-lesioned monkeys investigated were extensively andsimilarly denervated in the caudate nucleus and the putamen.Hence, the glutamate receptor changes in the MPTP-lesionedmonkeys can be related to the drug treatments.

4.2. Effects of a chronic treatment with MPEP on mGlu5 receptors

An increase of mGlu5 receptor specific binding was observed inall sub-regions of the putamen, the caudate nucleus and the GP of L-DOPA treatedMPTPmonkeys. This is in agreement and expands ourprevious report in the putamen of these monkeys (Morin et al.,2013a). It also agrees with our previous results with other mon-keys showing an elevation of striatal mGlu5 receptor in L-DOPAtreated MPTP monkeys that have developed LID (Ouattara et al.,2010a, 2011; Samadi et al., 2008b). Moreover, the present resultsagree with human data where we observed elevated striatal mGlu5receptors in parkinsonian patients with motor complications(Ouattara et al., 2010a, 2011; Samadi et al., 2008b). By contrast,mGlu5 receptor specific binding did not increase in the striatumand GP of monkeys treated with L-DOPA MPEP compared tocontrol monkeys and untreated MPTP-lesioned monkeys in thesame brain regions. These results suggest that MPEP prevented theLID-related increase in mGlu5 receptor levels associated withdevelopment of LID.

Moreover, a positive correlation between the mean dyskinesiascores of these monkeys and mGlu5 receptor specific binding inthese four regions was observed, suggesting that higher levels ofmGlu5 receptor are associated with LID.

We showed for the first time in MPTP monkeys that thedensity (Bmax) of [3H]ABP688 specific binding increased signifi-cantly in L-DOPA-treated monkeys compared to the other groupsand the affinity (Kd) remain unchanged. [3H]ABP688 bindingcapacity was modified by the chronic treatment with L-DOPA-alone and not with the L-DOPA MPEP treatment. These resultsextend our results obtained by [3H]ABP688 autoradiography,where an increase in specific binding was observed. Theincreased Bmax and the unchanged Kd observed in the saturationbinding curves support that the L-DOPA treatment increasedthe density in the striatum and this was prevented in theL-DOPA MPEP-treated group. In accordance with the presentfindings in MPTP monkeys, in 6-OHDA-lesioned rats, striatal [3H]MPEP binding capacity did not change by a chronic treatment oftwo weeks with MPEP (3 mg/kg), Bmax and Kd remained un-changed, demonstrating that neither the lesion, the chronictreatment with MPEP or their combination modified the affinityof the MPEP binding sites (Domenici et al., 2005). Hence, fromthe relevant literature and the present study, the effect of MPEPtreatment on mGlu5 receptors density is suggested, rather than adirect regulation of mGlu5 receptors, to be related to thenormalization of excessive glutamate neurotransmission thatwas exacerbated by L-DOPA.

Striatal and pallidal mGlu5 receptor specific binding wereincreased in L-DOPA treated MPTP monkeys compared to the threeother experimental groups, and were correlated with dyskinesiascores suggesting that these receptor increases have functionalsignificance. Using [3H]ABP688 autoradiography in the striatum

Fig. 4. In situ hybridization of mGlu5 receptor mRNA showing no effect of MPTP lesion, L-DOPA alone and with MPEP (A) Representative autoradiograms of in situ hybridization ofmGlu5 receptors in coronal brain sections showing control and MPTP-lesioned monkeys of the experimental groups investigated. (B) Histograms of mean values of in situ hy-bridization in the caudate nucleus and putamen from control monkeys (n 4), and MPTP-lesioned monkeys treated with saline (n 5), L-DOPA (n 4) and L-DOPA MPEP (n 5).Values are presented in relative units of optical densities as the mean SEM. DM, dorsomedial; VM, ventromedial; VL, ventrolateral; DL, dorsolateral.

N. Morin et al. / Neuropharmacology 73 (2013) 216e231 223(Fig. 2), a mean increase of 21e41% in the MPTP L-DOPAdyskinetic group was measured as compared to the other exper-imental groups. Using saturation of [3H]ABP688 binding to striataltissue homogenates, the Bmax (Fig. 3) was increased by 11e17%only in the L-DOPA group as compared to the other groups, whilethe Kd remained unchanged. In a similar de novo protocol, the L-DOPA treatment inducing LID, increased by 25e39% striatal [3H]ABP688-specific binding measured by autoradiography ascompared to control and saline-treated MPTP monkeys (Ouattaraet al., 2010a). Moreover using [3H]MPEP autoradiography, LIDwere associated with an increase of 41% of mGlu5 receptor specificbinding in the posterior putamen compared to controls (Samadiet al., 2008b). Prevention of LID, with CI-1041, an NMDA recep-tor antagonist, or low doses of the DA D2 receptor agonist,cabergoline, was associated with a decrease of 37% of striatalmGlu5 receptor specific binding compared to dyskinetic animals(Samadi et al., 2008b). PD patients without motor complicationshad w25% lower [3H]ABP688 specific binding compared to con-trols whereas PD patients with motor complications had w25%higher binding compared to controls and patients without motorcomplications (Ouattara et al., 2011). In MPTP monkeys, [3H]ABP688 specific binding was also elevated by 40% in the striatumof dyskinetic L-DOPA-treated MPTP monkeys but not in MPTPmonkeys without dyskinesias compared to controls (Ouattaraet al., 2011). Thus, previous studies in MPTP monkeys and hu-man PD patients show similar 25e40% increases in striatal mGlu5receptor density when dyskinetic (functional change) subjectswere compared to non-dyskinetic and control subjects supportingthe behavioral relevance of the mGlu5 receptor changes. Otherexamples of small mGlu5 receptor changes from 4 to 21% are re-ported to have functional effects in different brain areas and pa-thologies such as smoking (Akkus et al., 2013), depression(Deschwanden et al., 2011) and sleep deprivation (Hefti et al.,2013). These data suggest that in different brain areas, a small5e10% change in mGlu5 receptor Bmax, with a Kd unchanged, canlead to functional changes.

The present study showed that MPTP lesion, L-DOPA andL-DOPA MPEP treatments left unchanged mRNA levels of mGlu5receptor in the striatum. We previously reported in others groupsof monkeys that striatal mGlu5 receptor mRNA levels remainedunchanged following a MPTP lesion (Ouattara et al., 2010a). Theseresults are also in accordance with previous observations in the ratstriatum, where mRNA levels of mGlu5 receptor remained un-changed following a 6-OHDA lesion (Rodriguez-Puertas et al.,1999). Moreover, as observed in the present study, chronicL-DOPA treatment for one month was reported to leave unchangedstriatal mGlu5 receptor mRNA levels in MPTP monkeys killed 24 hafter their last L-DOPA dose. For MPTP monkeys treated for onemonth with L-DOPA and killed 4 h after their last L-DOPA treatmentstriatal mGlu5 receptor mRNA levels were elevated (Ouattara et al.,2010a). The mGlu5 receptor specific binding changes were not inparallel with mRNA changes of this receptor suggesting a post-transcriptional effect that modifies mGlu5 receptor specific bind-ing. Amismatch between themGlu5 receptor protein and its mRNAlevels was also observed using the reserpinized rat model, where18 h after a reserpine injection, reducing striatal DA concentrations,

Fig. 5. Autoradiography of brain mGlu2/3 receptors showing the effect of MPTP lesion, L-DOPA alone and with MPEP (A) Representative autoradiograms of [3H]LY341495 binding incoronal brain sections showing control and MPTP-lesioned monkeys treated with saline, L-DOPA and L-DOPA MPEP. (B) Histograms of mean values of [3H]LY341495 specificbinding in the caudate nucleus and putamen of control monkeys (n 4), and MPTP-lesioned monkeys treated with saline (n 5), L-DOPA (n 4) and L-DOPA MPEP (n 5).(C) [3H]LY341495 specific binding in the GPe and GPi of these monkeys. (D) Correlation between mean dyskinesia scores of these monkeys and their [3H]LY341495 specific bindingin the caudate nucleus, putamen, GPe and GPi; each point represents an individual monkey. **p < 0.01 and ***p < 0.001 vs. control monkeys; ##p < 0.01 and ###p < 0.001 vs.MPTP saline-treated MPTP monkeys; p < 0.01 vs. MPTP L-DOPA-treated MPTP monkeys. DM, dorsomedial; VM, ventromedial; VL, ventrolateral; DL, dorsolateral.

N. Morin et al. / Neuropharmacology 73 (2013) 216e231224

Fig. 6. Autoradiography of brain NMDA receptors composed of NR1/NR2B subunits showing the effect of MPTP lesion, L-DOPA alone and with MPEP (A) Representative autora-diograms of [3H]Ro 25-6981 binding in coronal brain sections showing control and MPTP-lesioned monkeys treated with saline, L-DOPA and L-DOPA MPEP. (B) Histograms ofmean values of [3H]Ro 25-6981 specific binding in the caudate nucleus and putamen of control monkeys (n 4), and MPTP-lesioned monkeys treated with saline (n 5), L-DOPA(n 4) and L-DOPA MPEP (n 5). (C) [3H]Ro 25-6981 specific binding in the GPe and GPi of the same monkeys. (D) Correlation between mean dyskinesia scores of these monkeysand their [3H]Ro 25-6981 specific binding in the caudate nucleus, putamen, GPe and GPi; each point represents an individual monkey. *p < 0.05 and **p < 0.01 vs. control monkeys;#p < 0.05, ##p < 0.01 and ###p < 0.001 vs. MPTP saline-treated MPTP monkeys; p < 0.01 and p < 0.001 vs. MPTP L-DOPA-treated MPTP monkeys. DM, dorsomedial; VM,ventromedial; VL, ventrolateral; DL, dorsolateral.

N. Morin et al. / Neuropharmacology 73 (2013) 216e231 225

Fig. 7. Autoradiography of brain NMDA receptors composed of NR1/NR2A subunits showing no effect of MPTP lesion, L-DOPA alone and with MPEP (A) Representative autora-diograms of [3H]CGP-39653 binding in coronal brain sections showing control and MPTP-lesioned monkeys treated with saline, L-DOPA and L-DOPA MPEP. (B) Histograms of meanvalues of [3H]CGP-39653 specific binding in the caudate nucleus and putamen of control monkeys (n 4), and MPTP-lesioned monkeys treated with saline (n 5), L-DOPA (n 4)and L-DOPA MPEP (n 5). DM, dorsomedial; VM, ventromedial; VL, ventrolateral; DL, dorsolateral.

N. Morin et al. / Neuropharmacology 73 (2013) 216e231226mGlu5 receptor mRNA levels were decreased in the striatumwhereas [3H]MPEP specific binding to mGlu5 receptor wasincreased (Ismayilova et al., 2006). These results suggest that theincreased receptors in the membrane available for binding dependon the balance between synthesis/insertion and internalization/degradation rates of the receptors and may take more time thanchanges of mRNA levels. mGlu5 receptor gene transcription seemsto return to normal 24 h post L-DOPA treatment and the increasedmGlu5 receptor protein in the membrane to last longer. This wasnot affected by the MPEP treatment. Indeed, our results suggest arapid response of mGlu5 receptor to the presence of DA (between 4and 24 h) during the one-month treatment with daily adminis-trations of L-DOPA and the fluctuating levels of mGlu5 receptorsmay promote the development of LID. Alternatively, MPEP mayblock this excessive and fluctuating glutamate neurotransmission.

4.3. Effects of a chronic treatment with MPEP on mGlu2/3 receptors

In the present study, by contrast to mGlu5 receptors, mGlu2/3specific binding decreased in the striatum and the GP of L-DOPA-treated monkeys compared to control, saline or L-DOPA MPEP-treated MPTP monkeys and mean dyskinesia scores of thesemonkeys correlated negatively with striatal mGlu2/3 specificbinding. These results suggest that the decreased pre-synapticmGlu2/3 receptor specific binding was less efficacious to controlthe excessive striatal glutamate release and this was corrected bythe chronic MPEP treatment. In 6-OHDA rats, a lesion-inducedincreased expression of striatal mGlu2/3 receptor proteins wasobserved (Picconi et al., 2002), no change of striatal mGlu2/3 re-ceptors (Testa et al., 1998), or a decrease of striatal and pallidalmGlu3 receptor mRNA expression (Messenger et al., 2002). Previ-ous results from our laboratory showed that mGlu2/3 receptorspecific binding was reduced in the caudate nucleus and GPi ofpatients without wearing-off compared to controls and in patientswho experienced wearing-off while there was no difference amongPD patients with or without dyskinesias (Samadi et al., 2009). InMPTP monkeys, no change of mGlu2/3 receptor specific bindingwas observed in the basal ganglia after nigrostriatal denervation byMPTP and L-DOPA treatment as compared to controls, while areduction was observed in non-dyskinetic monkeys receivingcabergoline with L-DOPA as compared with L-DOPA-treated mon-keys with LID (Samadi et al., 2008a). These findings are somewhatat variance with the present results and could be because monkeysin the present study were treated after a longer time after theMPTPlesion. Moreover cabergoline, a long acting agonist, could affect theadaptation of brain receptors to glutamate overactivity. Changes inmGlu2/3 receptors can reflect crosstalk between the pre-synapticmGlu2/3 and D2 DA receptors. Indeed, studies revealed a close

Fig. 8. Autoradiography of brain AMPA receptors showing the effect of MPTP lesion, L-DOPA alone and with MPEP (A) Representative autoradiograms of [3H]Ro 48-8587 binding incoronal brain sections showing control and MPTP-lesioned monkeys treated with saline, L-DOPA and L-DOPA MPEP. (B) Histograms of mean values of [3H]Ro 48-8587 specificbinding in the caudate nucleus and putamen of control monkeys (n 4), and MPTP-lesioned monkeys treated with saline (n 5), L-DOPA (n 4) and L-DOPA MPEP (n 5).(C) Correlation between mean dyskinesia scores of these monkeys and their [3H]Ro 48-8587 specific binding in the caudate nucleus and putamen; each point represents an in-dividual monkey. **p < 0.01 and ***p < 0.001 vs. control monkeys; ##p < 0.01 and ###p < 0.001 vs. MPTP saline-treated MPTP monkeys; p < 0.001 and p < 0.0001 vs.MPTP L-DOPA-treated MPTP monkeys. DM, dorsomedial; VM, ventromedial; VL, ventrolateral; DL, dorsolateral.

N. Morin et al. / Neuropharmacology 73 (2013) 216e231 227

Table 1Summary of glutamate receptor results in the basal ganglia of MPTP-lesionedmonkeys treated with L-DOPA and MPEP.

Receptor Treatment Correlation

MPTP MPTP L-DOPA

MPTP L-DOPA MPEP

LID vs.receptor

mGlu5 0,Y 0,Y PositivemGlu2/3 0,[ 0,[ NegativeNMDA NR1/NR2B 0,Y 0,Y PositiveNMDA NR1/NR2A 0 0 0 NoAMPA 0,Y 0,Y Positive

0, , : No effect, decreased or increased receptor levels vs. respective controlmonkeys; [, Y: higher or lower receptor levels vs. respective MPTP L-DOPAmonkeys; LID: Levodopa-induced dyskinesias (mean dyskinesia scores).

N. Morin et al. / Neuropharmacology 73 (2013) 216e231228interaction between mGlu2/3 receptors and DA systems both pre-synaptically and post-synaptically (David and Abraini, 2001;Morishima et al., 2005). However, the precise role of mGlu2/3 re-ceptors in the basal ganglia function are not well known andfurther studies are required to fully understand their functionalimplications.

4.4. Effects of a chronic treatment with MPEP on NMDA and AMPAreceptors

In the present study, we observed an increase of NMDA NR1/NR2B receptors specific binding in the striatum and GP as well asfor AMPA receptors specific binding in the striatum of L-DOPA-treated monkeys compared to control and saline-treated MPTPmonkeys while the NMDA NR1/NR2A receptor specific bindingremained unchanged by these treatments in all the brain regionsinvestigated. These results are in agreement with our previousfindings in post-mortem brains of parkinsonian patients displayingmotor complications; NMDA receptors composed of the NR2Bsubunits are increased in the putamen whereas those containingthe NR2A subunits remain unchanged (Calon et al., 2003). We alsoshowed that [3H]AMPA specific binding was increased in the pu-tamen of parkinsonian patients with LID compared to patientswithout LID (Calon et al., 2003). Moreover, an increased uptake of11C-CNS 5161, a non-competitive NMDA receptor antagonist whichbinds with high affinity to its MK801 site (Biegon et al., 2007), wasobserved by positron emission tomography in the striatum of onstate PD patients expressing LID (Ahmed et al., 2011).

In accordance with the human results and the present results inmonkeys, we previously observed that NMDANR1/NR2B and AMPAreceptors were enhanced in L-DOPA-treated MPTP monkeys dis-playing LID whereas those comprised of NR2A subunits weregenerally unchanged (Ouattara et al., 2009, 2010b). Moreover,western blotting studies have shown that L-DOPA increased striatalNMDA/NR2B receptor levels compared to lesioned MPTP monkeys(Hallett et al., 2005) or 6-OHDA lesioned rats (Dunah et al., 2000).Similarly, following chronic L-DOPA treatment, striatal NMDA re-ceptors were substantially increased in 6-OHDA rats with motorcomplications compared to controls (Xu et al., 2009).

Interestingly, NMDA NR1/NR2B and AMPA receptor specificbinding in the striatum and GP of L-DOPA MPEP-treated MPTPmonkeys were not elevated. This has not been reported previouslyand suggests that MPEP prevented the LID-related increase inNMDA and AMPA receptor levels associated with development ofLID. Previously, we observed that CI-1041 administered withL-DOPA prevented LID and the augmentation of NMDA and AMPAreceptors (Ouattara et al., 2009, 2010b).

The role of NMDA receptors containing the NR2B subunits in LIDis also supported by a clinical study with parkinsonian patientsshowing that CP-101,606, a NR2B selective NMDA glutamateantagonist, reduced the maximum severity of LID in PD patients(Nutt et al., 2008). Hence, this evidence strongly suggests thatNMDA receptors containing NR2B subunits are involved in thepathophysiology of LID and might contribute to their development.

Although less receptor specific, dextrorphan, dextromethor-phan and amantadine, known to block NMDA receptors, also showantidyskinetic activity in humans (Blanchet et al., 1996, 1997;Luginger et al., 2000; Metman et al., 1999; Rajput et al., 1998;Ruzicka et al., 2000; Snow et al., 2000; Verhagen Metman et al.,1998a). The role of AMPA receptors in LID is also supported by re-ports of improved LID with blockade of AMPA receptors in theMPTP monkey model (Bibbiani et al., 2005; Konitsiotis et al., 2000).NMDA and AMPA receptor antagonists were also shown to block L-DOPA-induced motor complications in 6-OHDA rats (Bibbiani et al.,2005; Engber et al., 1994; Marin et al., 2000,1996; Papa et al., 1995).

The present results in MPTP monkeys treated with L-DOPA andMPEP showed positive correlations between mean dyskinesiascores of the monkeys and both their respective NMDA and AMPAspecific binding in the same brain regions. These findings support aclose interaction between mGlu5, NMDA and AMPA receptors andthat MPEP blocking one of these receptors affected the others oralternatively that MPEP might have normalized glutamate neuro-transmission and consequently these glutamate receptors.

4.5. Effects of a chronic treatment with MPEP in the GP and basalganglia

The striatal medium spiny neurons project to the output nucleiof the basal ganglia, substantia nigra pars reticulata and GPi viadirect and indirect pathways, mono-synaptically or poly-synaptically through relays in the GPe and the STN (Levesque andParent, 2005; Nadjar et al., 2006). GPe seems to play a major rolein the control of the basal ganglia circuitry and should not beconsidered as a simple relay structure in the striatopallidal pathway(Heimer et al., 2006). In GPe and GPi, mGlu5 receptor is distributedpost-synaptically in the main body of GABAergic striatal synapsesand they regulate GABAergic and glutamatergic transmission(Rouse et al., 2000). For example, the release of GABA is activatedwhen the mGlu5 receptor is stimulated at the GABAergic synapses(de Novellis et al., 2003). Thus, the reduction of LID in MPTPmonkeys treated with L-DOPA MPEP could result from disinhi-bition of GPi via the STNeGPeeGPi circuit. The addition of MPEP toL-DOPA was associated with the prevention of an upregulation ofmGlu5 receptor specific binding induced by the L-DOPA treatmentin the pallidal neurons. The decrease of cortico-subthalamic andSTNeGPe activity with MPEP may restore the disruption of GPeeGPi balance and the abnormal discharge patterns in the GPiinduced by the L-DOPA treatment.

4.6. Biochemical effects of mGlu5 receptor antagonists

The mGlu5 receptor subtype is highly expressed in striatalmedium spiny neurons (Conn et al., 2005; Paquet and Smith, 2003;Testa et al., 1994) and plays a key role in modulating the responsesmediated by NMDA receptors and L-type calcium channels(Gubellini et al., 2004). In addition, an antagonistic interactionbetween the D2R DA receptor and mGlu5 receptors is reported(Fuxe et al., 2008).

The response of the basal ganglia of primates to a chronictreatment with a mGlu5 receptor antagonist is not yet documentedbut in rodent models of PD, striatal molecular changes relevant toLID are reported to be reversed by MPEP or MTEP, including deltaFosB protein (Jimenez et al., 2009), prodynorphin mRNA (Melaet al., 2007), glutamic acid decarboxylase (GAD65 and GAD67)mRNA (Yamamoto and Soghomonian, 2009) and phosphorylated

N. Morin et al. / Neuropharmacology 73 (2013) 216e231 229extracellular signal-regulated kinases 1 and 2 (ERK1/2) proteinlevels (Rylander et al., 2009). Hence, mGlu5 receptor antagonistsreverse hyperactive GABAergic transmission in the basal ganglia inrodent models of PD and its downstream molecular changesassociated with LID. In the present study, an mGlu5 receptorantagonist reversed hyperactive glutamate transmission in thebasal ganglia of a primate model of PD thus showing the wide-spread normalization activity of mGlu5 receptor antagonists in thebasal ganglia in PD animal models.4.7. Conclusion

The present study showed that the mGlu5 receptor antagonistMPEP normalized glutamate neurotransmission through regulationof the expression and distribution of glutamate receptors in thebasal ganglia and reduced the development of LID in L-DOPA-treated MPTP-lesioned monkeys. Taken together, results in rodentmodels and results presented herein with MPTP-lesioned monkeysargue in favor of the contribution of mGlu5 receptors in the path-ophysiology of LID and clearly indicate that blockade of this re-ceptor might constitute a novel therapeutic treatment for LID in PD.To this date, no therapy has yet been approved for LID treatmentand mGlu5 receptor antagonist treatment might bring the firsttherapy for attenuating LID.Acknowledgments

This workwas supported by a grant from the Canadian Institutesof Health Research to TDP. N.M. holds a professional health carestudentship from the Fonds de la recherche en sant du Qubec(FRSQ).References

Addex Therapeutics, 2012 Mar 21. Geneva, Switzerland (Media Release) (online).Available from URL: http://www.addextherapeutics.com/investors/press-releases/news-details/article/addex-reports-positive-top-line-phase-iia-data-for-dipraglurant-in-parkinsons-disease-levodopa-indu/ (accessed 11.02.13.).

Ahmed, I., Bose, S.K., Pavese, N., Ramlackhansingh, A., Turkheimer, F., Hotton, G.,Hammers, A., Brooks, D.J., 2011. Glutamate NMDA receptor dysregulation inParkinsons disease with dyskinesias. Brain 134, 979e986.

Akkus, F., Ametamey, S.M., Treyer, V., Burger, C., Johayem, A., Umbricht, D., GomezMancilla, B., Sovago, J., Buck, A., Hasler, G., 2013. Marked global reduction inmGluR5 receptor binding in smokers and ex-smokers determined by [11C]ABP688 positron emission tomography. Proc. Natl. Acad. Sci. U. S. A. 110,737e742.

Berg, D., Godau, J., Trenkwalder, C., Eggert, K., Csoti, I., Storch, A., Huber, H., Morelli-Canelo, M., Stamelou, M., Ries, V., Wolz, M., Schneider, C., Di Paolo, T.,Gasparini, F., Hariry, S., Vandemeulebroecke, M., Abi-Saab, W., Cooke, K.,Johns, D., Gomez-Mancilla, B., 2011. AFQ056 treatment of levodopa-induceddyskinesias: results of 2 randomized controlled trials. Mov. Disord. 26, 1243e1250.

Bezard, E., Brotchie, J.M., Gross, C.E., 2001. Pathophysiology of levodopa-induceddyskinesia: potential for new therapies. Nat. Rev. Neurosci. 2, 577e588.

Bibbiani, F., Oh, J.D., Kielaite, A., Collins, M.A., Smith, C., Chase, T.N., 2005. Combinedblockade of AMPA and NMDA glutamate receptors reduces levodopa-inducedmotor complications in animal models of PD. Exp. Neurol. 196, 422e429.

Biegon, A., Gibbs, A., Alvarado, M., Ono, M., Taylor, S., 2007. In vitro and in vivocharacterization of [3H]CNS-5161ea use-dependent ligand for the N-methyl-D-aspartate receptor in rat brain. Synapse 61, 577e586.

Blanchet, P.J., Konitsiotis, S., Chase, T.N., 1998. Amantadine reduces levodopa-induced dyskinesias in parkinsonian monkeys. Mov. Disord. 13, 798e802.

Blanchet, P.J., Metman, L.V., Mouradian, M.M., Chase, T.N., 1996. Acute pharmaco-logic blockade of dyskinesias in Parkinsons disease. Mov. Disord. 11, 580e581.

Blanchet, P.J., Papa, S.M., Metman, L.V., Mouradian, M.M., Chase, T.N., 1997. Modu-lation of levodopa-induced motor response complications by NMDA antago-nists in Parkinsons disease. Neurosci. Biobehav Rev. 21, 447e453.

Blandini, F., Armentero, M.T., 2012. New pharmacological avenues for the treatmentof L-DOPA-induced dyskinesias in Parkinsons disease: targeting glutamate andadenosine receptors. Expert Opin. Investig. Drugs 21, 153e168.

Calabresi, P., Di Filippo, M., Ghiglieri, V., Tambasco, N., Picconi, B., 2010. Levodopa-induced dyskinesias in patients with Parkinsons disease: filling the bench-to-bedside gap. Lancet Neurol. 9, 1106e1117.Calon, F., Birdi, S., Rajput, A.H., Hornykiewicz, O., Bedard, P.J., Di Paolo, T., 2002a.Increase of preproenkephalin mRNA levels in the putamen of Parkinson diseasepatients with levodopa-induced dyskinesias. J. Neuropathol. Exp. Neurol. 61,186e196.

Calon, F., Morissette, M., Ghribi, O., Goulet, M., Grondin, R., Blanchet, P.J., Bedard, P.J., DiPaolo, T., 2002b. Alteration of glutamate receptors in the striatum of dyskinetic 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine-treated monkeys following dopa-mineagonist treatment. Prog.Neuropsychopharmacol. Biol. Psychiatry 26,127e138.

Calon, F., Rajput, A.H., Hornykiewicz, O., Bedard, P.J., Di Paolo, T., 2003. Levodopa-induced motor complications are associated with alterations of glutamate re-ceptors in Parkinsons disease. Neurobiol. Dis. 14, 404e416.

Cenci, M.A., Lundblad, M., 2006. Post- versus presynaptic plasticity in L-DOPA-induced dyskinesia. J. Neurochem. 99, 381e392.

Chase, T.N., Oh, J.D., 2000. Striatal mechanisms and pathogenesis of parkinsonian signsand motor complications. Ann. Neurol. 47, S122eS129. Discussion S129e130.

Conn, P.J., Battaglia, G., Marino, M.J., Nicoletti, F., 2005. Metabotropic glutamatereceptors in the basal ganglia motor circuit. Nat. Rev. Neurosci. 6, 787e798.

Crosby, N.J., Deane, K.H., Clarke, C.E., 2003. Amantadine for dyskinesia in Parkin-sons disease. Cochrane Database Syst. Rev., CD003467.

David, H.N., Abraini, J.H., 2001. Differential modulation of the D1-like- and D2-likedopamine receptor-induced locomotor responses by group II metabotropicglutamate receptors in the rat nucleus accumbens. Neuropharmacology 41,454e463.

de Lau, L.M., Breteler, M.M., 2006. Epidemiology of Parkinsons disease. LancetNeurol. 5, 525e535.

de Novellis, V., Marabese, I., Palazzo, E., Rossi, F., Berrino, L., Rodella, L., Bianchi, R.,Maione, S., 2003. Group I metabotropic glutamate receptors modulate gluta-mate and gamma-aminobutyric acid release in the periaqueductal grey of rats.Eur. J. Pharmacol. 462, 73e81.

Dekundy, A., Lundblad, M., Danysz, W., Cenci, M.A., 2007. Modulation of L-DOPA-induced abnormal involuntary movements by clinically tested compounds:further validation of the rat dyskinesia model. Behav. Brain Res. 179, 76e89.

Dekundy, A., Pietraszek, M., Schaefer, D., Cenci, M.A., Danysz, W., 2006. Effects ofgroup I metabotropic glutamate receptors blockade in experimental models ofParkinsons disease. Brain Res. Bull. 69, 318e326.

Deschwanden, A., Karolewicz, B., Feyissa, A.M., Treyer, V., Ametamey, S.M.,Johayem, A., Burger, C., Auberson, Y.P., Sovago, J., Stockmeier, C.A., Buck, A.,Hasler, G., 2011. Reduced metabotropic glutamate receptor 5 density in majordepression determined by [(11)C]ABP688 PET and postmortem study. Am. J.Psychiatry 168, 727e734.

Domenici, M.R., Potenza, R.L., Martire, A., Coccurello, R., Pezzola, A., Reggio, R.,Tebano, M.T., Popoli, P., 2005. Chronic treatment with the mGlu5R antagonistMPEP reduces the functional effects of the mGlu5R agonist CHPG in the stria-tum of 6-hydroxydopamine-lesioned rats: possible relevance to the effects ofmGlu5R blockade in Parkinsons disease. J. Neurosci. Res. 80, 646e654.

Dunah, A.W., Wang, Y., Yasuda, R.P., Kameyama, K., Huganir, R.L., Wolfe, B.B.,Standaert, D.G., 2000. Alterations in subunit expression, composition, andphosphorylation of striatal N-methyl-D-aspartate glutamate receptors in a rat 6-hydroxydopamine model of Parkinsons disease. Mol. Pharmacol. 57, 342e352.

Dupre, K.B., Ostock, C.Y., Eskow Jaunarajs, K.L., Button, T., Savage, L.M., Wolf, W.,Bishop, C., 2011. Local modulation of striatal glutamate efflux by serotonin 1Areceptor stimulation in dyskinetic, hemiparkinsonian rats. Exp. Neurol. 229,288e299.

Duty, S., 2012. Targeting glutamate receptors to tackle the pathogenesis, clinicalsymptoms and levodopa-induced dyskinesia associated with Parkinsons dis-ease. CNS Drugs 26, 1017e1032.

Engber, T.M., Papa, S.M., Boldry, R.C., Chase, T.N., 1994. NMDA receptor blockadereverses motor response alterations induced by levodopa. Neuroreport 5,2586e2588.

Fabbrini, G., Brotchie, J.M., Grandas, F., Nomoto, M., Goetz, C.G., 2007. Levodopa-induced dyskinesias. Mov. Disord. 22, 1379e1389. Quiz 1523.

Ferraguti, F., Shigemoto, R., 2006. Metabotropic glutamate receptors. Cell Tissue Res.326, 483e504.

Fuxe, K., Marcellino, D., Rivera, A., Diaz-Cabiale, Z., Filip, M., Gago, B., Roberts, D.C.,Langel, U., Genedani, S., Ferraro, L., de la Calle, A., Narvaez, J., Tanganelli, S.,Woods, A., Agnati, L.F., 2008. Receptor-receptor interactions within receptormosaics. Impact on neuropsychopharmacology. Brain Res. Rev. 58, 415e452.

Gottwald, M.D., Aminoff, M.J., 2011. Therapies for dopaminergic-induced dyskine-sias in Parkinson disease. Ann. Neurol. 69, 919e927.

Goulet, M., Morissette, M., Grondin, R., Falardeau, P., Bedard, P.J., Rostene, W., DiPaolo, T., 1999. Neurotensin receptors and dopamine transporters: effects ofMPTP lesioning and chronic dopaminergic treatments in monkeys. Synapse 32,153e164.

Gravius, A., Dekundy, A., Nagel, J., More, L., Pietraszek, M., Danysz, W., 2008.Investigation on tolerance development to subchronic blockade of mGluR5 inmodels of learning, anxiety, and levodopa-induced dyskinesia in rats. J. NeuralTransm. 115, 1609e1619.

Gregoire, L., Morin, N., Ouattara, B., Gasparini, F., Bilbe, G., Johns, D., Vranesic, I.,Sahasranaman, S., Gomez-Mancilla, B., Di Paolo, T., 2011. The acute anti-parkinsonian and antidyskinetic effect of AFQ056, a novel metabotropicglutamate receptor type 5 antagonist, in L-DOPA-treated parkinsonian mon-keys. Parkinsonism Relat. Disord. 17, 270e276.

Gubellini, P., Pisani, A., Centonze, D., Bernardi, G., Calabresi, P., 2004. Metabotropicglutamate receptors and striatal synaptic plasticity: implications for neuro-logical diseases. Prog. Neurobiol. 74, 271e300.

http://www.addextherapeutics.com/investors/press-releases/news-details/article/addex-reports-positive-top-line-phase-iia-data-for-dipraglurant-in-parkinsons-disease-levodopa-indu/http://www.addextherapeutics.com/investors/press-releases/news-details/article/addex-reports-positive-top-line-phase-iia-data-for-dipraglurant-in-parkinsons-disease-levodopa-indu/http://www.addextherapeutics.com/investors/press-releases/news-details/article/addex-reports-positive-top-line-phase-iia-data-for-dipraglurant-in-parkinsons-disease-levodopa-indu/http://refhub.elsevier.com/S0028-3908(13)00242-6/sref2http://refhub.elsevier.com/S0028-3908(13)00242-6/sref2http://refhub.elsevier.com/S0028-3908(13)00242-6/sref2http://refhub.elsevier.com/S0028-3908(13)00242-6/sref2http://refhub.elsevier.com/S0028-3908(13)00242-6/sref3http://refhub.elsevier.com/S0028-3908(13)00242-6/sref3http://refhub.elsevier.com/S0028-3908(13)00242-6/sref3http://refhub.elsevier.com/S0028-3908(13)00242-6/sref3http://refhub.elsevier.com/S0028-3908(13)00242-6/sref3http://refhub.elsevier.com/S0028-3908(13)00242-6/sref3http://refhub.elsevier.com/S0028-3908(13)00242-6/sref3http://refhub.elsevier.com/S0028-3908(13)00242-6/sref4http://refhub.elsevier.com/S0028-3908(13)00242-6/sref4http://refhub.elsevier.com/S0028-3908(13)00242-6/sref4http://refhub.elsevier.com/S0028-3908(13)00242-6/sref4http://refhub.elsevier.com/S0028-3908(13)00242-6/sref4http://refhub.elsevier.com/S0028-3908(13)00242-6/sref4http://refhub.elsevier.com/S0028-3908(13)00242-6/sref5http://refhub.elsevier.com/S0028-3908(13)00242-6/sref5http://refhub.elsevier.com/S0028-3908(13)00242-6/sref5http://refhub.elsevier.com/S0028-3908(13)00242-6/sref6http://refhub.elsevier.com/S0028-3908(13)00242-6/sref6http://refhub.elsevier.com/S0028-3908(13)00242-6/sref6http://refhub.elsevier.com/S0028-3908(13)00242-6/sref6http://refhub.elsevier.com/S0028-3908(13)00242-6/sref7http://refhub.elsevier.com/S0028-3908(13)00242-6/sref7http://refhub.elsevier.com/S0028-3908(13)00242-6/sref7http://refhub.elsevier.com/S0028-3908(13)00242-6/sref7http://refhub.elsevier.com/S0028-3908(13)00242-6/sref7http://refhub.elsevier.com/S0028-3908(13)00242-6/sref7http://refhub.elsevier.com/S0028-3908(13)00242-6/sref8http://refhub.elsevier.com/S0028-3908(13)00242-6/sref8http://refhub.elsevier.com/S0028-3908(13)00242-6/sref8http://refhub.elsevier.com/S0028-3908(13)00242-6/sref9http://refhub.elsevier.com/S0028-3908(13)00242-6/sref9http://refhub.elsevier.com/S0028-3908(13)00242-6/sref9http://refhub.elsevier.com/S0028-3908(13)00242-6/sref10http://refhub.elsevier.com/S0028-3908(13)00242-6/sref10http://refhub.elsevier.com/S0028-3908(13)00242-6/sref10http://refhub.elsevier.com/S0028-3908(13)00242-6/sref10http://refhub.elsevier.com/S0028-3908(13)00242-6/sref11http://refhub.elsevier.com/S0028-3908(13)00242-6/sref11http://refhub.elsevier.com/S0028-3908(13)00242-6/sref11http://refhub.elsevier.com/S0028-3908(13)00242-6/sref11http://refhub.elsevier.com/S0028-3908(13)00242-6/sref12http://refhub.elsevier.com/S0028-3908(13)00242-6/sref12http://refhub.elsevier.com/S0028-3908(13)00242-6/sref12http://refhub.elsevier.com/S0028-3908(13)00242-6/sref12http://refhub.elsevier.com/S0028-3908(13)00242-6/sref13http://refhub.elsevier.com/S0028-3908(13)00242-6/sref13http://refhub.elsevier.com/S0028-3908(13)00242-6/sref13http://refhub.elsevier.com/S0028-3908(13)00242-6/sref13http://refhub.elsevier.com/S0028-3908(13)00242-6/sref13http://refhub.elsevier.com/S0028-3908(13)00242-6/sref14http://refhub.elsevier.com/S0028-3908(13)00242-6/sref14http://refhub.elsevier.com/S0028-3908(13)00242-6/sref14http://refhub.elsevier.com/S0028-3908(13)00242-6/sref14http://refhub.elsevier.com/S0028-3908(13)00242-6/sref14http://refhub.elsevier.com/S0028-3908(13)00242-6/sref15http://refhub.elsevier.com/S0028-3908(13)00242-6/sref15http://refhub.elsevier.com/S0028-3908(13)00242-6/sref15http://refhub.elsevier.com/S0028-3908(13)00242-6/sref15http://refhub.elsevier.com/S0028-3908(13)00242-6/sref16http://refhub.elsevier.com/S0028-3908(13)00242-6/sref16http://refhub.elsevier.com/S0028-3908(13)00242-6/sref16http://refhub.elsevier.com/S0028-3908(13)00242-6/sref17http://refhub.elsevier.com/S0028-3908(13)00242-6/sref17http://refhub.elsevier.com/S0028-3908(13)00242-6/sref17http://refhub.elsevier.com/S0028-3908(13)00242-6/sref17http://refhub.elsevier.com/S0028-3908(13)00242-6/sref18http://refhub.elsevier.com/S0028-3908(13)00242-6/sref18http://refhub.elsevier.com/S0028-3908(13)00242-6/sref18http://refhub.elsevier.com/S0028-3908(13)00242-6/sref19http://refhub.elsevier.com/S0028-3908(13)00242-6/sref19http://refhub.elsevier.com/S0028-3908(13)00242-6/sref20http://refhub.elsevier.com/S0028-3908(13)00242-6/sref20http://refhub.elsevier.com/S0028-3908(13)00242-6/sref20http://refhub.elsevier.com/S0028-3908(13)00242-6/sref20http://refhub.elsevier.com/S0028-3908(13)00242-6/sref20http://refhub.elsevier.com/S0028-3908(13)00242-6/sref21http://refhub.elsevier.com/S0028-3908(13)00242-6/sref21http://refhub.elsevier.com/S0028-3908(13)00242-6/sref21http://refhub.elsevier.com/S0028-3908(13)00242-6/sref22http://refhub.elsevier.com/S0028-3908(13)00242-6/sref22http://refhub.elsevier.com/S0028-3908(13)00242-6/sref22http://refhub.elsevier.com/S0028-3908(13)00242-6/sref22http://refhub.elsevier.com/S0028-3908(13)00242-6/sref22http://refhub.elsevier.com/S0028-3908(13)00242-6/sref23http://refhub.elsevier.com/S0028-3908(13)00242-6/sref23http://refhub.elsevier.com/S0028-3908(13)00242-6/sref23http://refhub.elsevier.com/S0028-3908(13)00242-6/sref23http://refhub.elsevier.com/S0028-3908(13)00242-6/sref24http://refhub.elsevier.com/S0028-3908(13)00242-6/sref24http://refhub.elsevier.com/S0028-3908(13)00242-6/sref24http://refhub.elsevier.com/S0028-3908(13)00242-6/sref24http://refhub.elsevier.com/S0028-3908(13)00242-6/sref25http://refhub.elsevier.com/S0028-3908(13)00242-6/sref25http://refhub.elsevier.com/S0028-3908(13)00242-6/sref25http://refhub.elsevier.com/S0028-3908(13)00242-6/sref25http://refhub.elsevier.com/S0028-3908(13)00242-6/sref25http://refhub.elsevier.com/S0028-3908(13)00242-6/sref25http://refhub.elsevier.com/S0028-3908(13)00242-6/sref26http://refhub.elsevier.com/S0028-3908(13)00242-6/sref26http://refhub.elsevier.com/S0028-3908(13)00242-6/sref26http://refhub.elsevier.com/S0028-3908(13)00242-6/sref26http://refhub.elsevier.com/S0028-3908(13)00242-6/sref26http://refhub.elsevier.com/S0028-3908(13)00242-6/sref26http://refhub.elsevier.com/S0028-3908(13)00242-6/sref27http://refhub.elsevier.com/S0028-3908(13)00242-6/sref27http://refhub.elsevier.com/S0028-3908(13)00242-6/sref27http://refhub.elsevier.com/S0028-3908(13)00242-6/sref27http://refhub.elsevier.com/S0028-3908(13)00242-6/sref27http://refhub.elsevier.com/S0028-3908(13)00242-6/sref28http://refhub.elsevier.com/S0028-3908(13)00242-6/sref28http://refhub.elsevier.com/S0028-3908(13)00242-6/sref28http://refhub.elsevier.com/S0028-3908(13)00242-6/sref28http://refhub.elsevier.com/S0028-3908(13)00242-6/sref28http://refhub.elsevier.com/S0028-3908(13)00242-6/sref29http://refhub.elsevier.com/S0028-3908(13)00242-6/sref29http://refhub.elsevier.com/S0028-3908(13)00242-6/sref29http://refhub.elsevier.com/S0028-3908(13)00242-6/sref29http://refhub.elsevier.com/S0028-3908(13)00242-6/sref30http://refhub.elsevier.com/S0028-3908(13)00242-6/sref30http://refhub.elsevier.com/S0028-3908(13)00242-6/sref30http://refhub.elsevier.com/S0028-3908(13)00242-6/sref30http://refhub.elsevier.com/S0028-3908(13)00242-6/sref31http://refhub.elsevier.com/S0028-3908(13)00242-6/sref31http://refhub.elsevier.com/S0028-3908(13)00242-6/sref31http://refhub.elsevier.com/S0028-3908(13)00242-6/sref32http://refhub.elsevier.com/S0028-3908(13)00242-6/sref32http://refhub.elsevier.com/S0028-3908(13)00242-6/sref32http://refhub.elsevier.com/S0028-3908(13)00242-6/sref33http://refhub.elsevier.com/S0028-3908(13)00242-6/sref33http://refhub.elsevier.com/S0028-3908(13)00242-6/sref33http://refhub.elsevier.com/S0028-3908(13)00242-6/sref33http://refhub.elsevier.com/S0028-3908(13)00242-6/sref33http://refhub.elsevier.com/S0028-3908(13)00242-6/sref34http://refhub.elsevier.com/S0028-3908(13)00242-6/sref34http://refhub.elsevier.com/S0028-3908(13)00242-6/sref34http://refhub.elsevier.com/S0028-3908(13)00242-6/sref35http://refhub.elsevier.com/S0028-3908(13)00242-6/sref35http://refhub.elsevier.com/S0028-3908(13)00242-6/sref35http://refhub.elsevier.com/S0028-3908(13)00242-6/sref35http://refhub.elsevier.com/S0028-3908(13)00242-6/sref35http://refhub.elsevier.com/S0028-3908(13)00242-6/sref36http://refhub.elsevier.com/S0028-3908(13)00242-6/sref36http://refhub.elsevier.com/S0028-3908(13)00242-6/sref36http://refhub.elsevier.com/S0028-3908(13)00242-6/sref36http://refhub.elsevier.com/S0028-3908(13)00242-6/sref36http://refhub.elsevier.com/S0028-3908(13)00242-6/sref37http://refhub.elsevier.com/S0028-3908(13)00242-6/sref37http://refhub.elsevier.com/S0028-3908(13)00242-6/sref37http://refhub.elsevier.com/S0028-3908(13)00242-6/sref37http://refhub.elsevier.com/S0028-3908(13)00242-6/sref37http://refhub.elsevier.com/S0028-3908(13)00242-6/sref37http://refhub.elsevier.com/S0028-3908(13)00242-6/sref38http://refhub.elsevier.com/S0028-3908(13)00242-6/sref38http://refhub.elsevier.com/S0028-3908(13)00242-6/sref38http://refhub.elsevier.com/S0028-3908(13)00242-6/sref38

N. Morin et al. / Neuropharmacology 73 (2013) 216e231230Hallett, P.J., Dunah, A.W., Ravenscroft, P., Zhou, S., Bezard, E., Crossman, A.R.,Brotchie, J.M., Standaert, D.G., 2005. Alterations of striatal NMDA receptorsubunits associated with the development of dyskinesia in the MPTP-lesionedprimate model of Parkinsons disease. Neuropharmacology 48, 503e516.

Hefti, K., Holst, S.C., Sovago, J., Bachmann, V., Buck, A., Ametamey, S.M.,Scheidegger, M., Berthold, T., Gomez-Mancilla, B., Seifritz, E., Landolt, H.P., 2013.Increased metabotropic glutamate receptor subtype 5 availability in humanbrain after one night without sleep. Biol. Psychiatry 73, 161e168.

Heimer, G., Rivlin-Etzion, M., Bar-Gad, I., Goldberg, J.A., Haber, S.N., Bergman, H.,2006. Dopamine replacement therapy does not restore the full spectrum ofnormal pallidal activity in the 1-methyl-4-phenyl-1,2,3,6-tetra-hydropyridineprimate model of Parkinsonism. J. Neurosci. 26, 8101e8114.

Hintermann, S., Vranesic, I., Allgeier, H., Brulisauer, A., Hoyer, D., Lemaire, M.,Moenius, T., Urwyler, S., Whitebread, S., Gasparini, F., Auberson, Y.P., 2007.ABP688, a novel selective and high affinity ligand for the labeling of mGlu5receptors: identification, in vitro pharmacology, pharmacokinetic and bio-distribution studies. Bioorg. Med. Chem. 15, 903e914.

Iravani, M.M., McCreary, A.C., Jenner, P., 2012. Striatal plasticity in Parkinsons dis-ease and L-DOPA induced dyskinesia. Parkinsonism Relat. Disord. 18 (Suppl. 1),S123eS125.

Ismayilova, N., Verkhratsky, A., Dascombe, M.J., 2006. Changes in mGlu5receptor expression in the basal ganglia of reserpinised rats. Eur. J. Pharmacol.545, 134e141.

Jenner, P., 2008. Molecular mechanisms of L-DOPA-induced dyskinesia. Nat. Rev.Neurosci. 9, 665e677.

Jimenez, A., Bonastre, M., Aguilar, E., Marin, C., 2009. Effect of the metabotropicglutamate antagonist MPEP on striatal expression of the Homer family proteinsin levodopa-treated hemiparkinsonian rats. Psychopharmacology (Berl.) 206,233e242.

Johnston, T.H., Fox, S.H., McIldowie, M.J., Piggott, M.J., Brotchie, J.M., 2010. Reductionof L-DOPA-induced dyskinesia by the selective metabotropic glutamate receptor5 antagonist 3-[(2-methyl-1,3-thiazol-4-yl)ethynyl]pyridine in the 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine-lesioned macaque model of Parkinsonsdisease. J. Pharmacol. Exp. Ther. 333, 865e873.

Kendrick, S.J., Lynch, D.R., Pritchett, D.B., 1996. Characterization of glutamatebinding sites in receptors assembled from transfected NMDA receptor subunits.J. Neurochem. 67, 608e616.

Klockgether, T., Turski, L., 1993. Toward an understanding of the role of glutamate inexperimental parkinsonism: agonist-sensitive sites in the basal ganglia. Ann.Neurol. 34, 585e593.

Konitsiotis, S., Blanchet, P.J., Verhagen, L., Lamers, E., Chase, T.N., 2000. AMPA re-ceptor blockade improves levodopa-induced dyskinesia in MPTP monkeys.Neurology 54, 1589e1595.

Laurie, D.J., Seeburg, P.H., 1994. Ligand affinities at recombinant N-methyl-D-aspartate receptors depend on subunit composition. Eur. J. Pharmacol. 268,335e345.

Levandis, G., Bazzini, E., Armentero, M.T., Nappi, G., Blandini, F., 2008. Systemicadministration of an mGluR5 antagonist, but not unilateral subthalamic lesion,counteracts L-DOPA-induced dyskinesias in a rodent model of Parkinsons dis-ease. Neurobiol. Dis. 29, 161e168.

Levesque, M., Parent, A., 2005. The striatofugal fiber system in primates: a reeval-uation of its organization based on single-axon tracing studies. Proc. Natl. Acad.Sci. U. S. A. 102, 11888e11893.

Luginger, E., Wenning, G.K., Bosch, S., Poewe, W., 2000. Beneficial effects ofamantadine on L-DOPA-induced dyskinesias in Parkinsons disease. Mov. Dis-ord. 15, 873e878.

Lujan, R., Roberts, J.D., Shigemoto, R., Ohishi, H., Somogyi, P., 1997. Differentialplasma membrane distribution of metabotropic glutamate receptors mGluR1alpha, mGluR2 and mGluR5, relative to neurotransmitter release sites. J. Chem.Neuroanat. 13, 219e241.

Lundblad, M., Andersson, M., Winkler, C., Kirik, D., Wierup, N., Cenci, M.A., 2002.Pharmacological validation of behavioural measures of akinesia and dyskinesiain a rat model of Parkinsons disease. Eur. J. Neurosci. 15, 120e132.

Marin, C., Jimenez, A., Bonastre, M., Chase, T.N., Tolosa, E., 2000. Non-NMDAreceptor-mediated mechanisms are involved in levodopa-induced motorresponse alterations in Parkinsonian rats. Synapse 36, 267e274.

Marin, C., Papa, S., Engber, T.M., Bonastre, M., Tolosa, E., Chase, T.N., 1996. MK-801prevents levodopa-induced motor response alterations in parkinsonian rats.Brain Res. 736, 202e205.

Marino, M.J., Awad, H., Poisik, O., Wittmann, M., Conn, P.J., 2002. Localization andphysiological roles of metabotropic glutamate receptors in the direct and in-direct pathways of the basal ganglia. Amino Acids 23, 185e191.

Meissner, W.G., Frasier, M., Gasser, T., Goetz, C.G., Lozano, A., Piccini, P.,Obeso, J.A., Rascol, O., Schapira, A., Voon, V., Weiner, D.M., Tison, F., Bezard, E.,2011. Priorities in Parkinsons disease research. Nat. Rev. Drug Discov. 10,377e393.

Mela, F., Marti, M., Dekundy, A., Danysz, W., Morari, M., Cenci, M.A., 2007. Antag-onism of metabotropic glutamate receptor type 5 attenuates L-DOPA-induceddyskinesia and its molecular and neurochemical correlates in a rat model ofParkinsons disease. J. Neurochem. 101, 483e497.

Mercuri, N.B., Bernardi, G., 2005. The magic of L-DOPA: why is it the gold standardParkinsons disease therapy? Trends Pharmacol. Sci. 26, 341e344.

Messenger, M.J., Dawson, L.G., Duty, S., 2002. Changes in metabotropic glutamatereceptor 1-8 gene expression in the rodent basal ganglia motor loop followinglesion of the nigrostriatal tract. Neuropharmacology 43, 261e271.Metman, L.V., Del Dotto, P., LePoole, K., Konitsiotis, S., Fang, J., Chase, T.N., 1999.Amantadine for levodopa-induced dyskinesias: a 1-year follow-up study. Arch.Neurol. 56, 1383e1386.

Morin, N., Gregoire, L., Gomez-Mancilla, B., Gasparini, F., Di Paolo, T., 2010. Effect ofthe metabotropic glutamate receptor type 5 antagonists MPEP and MTEP inparkinsonian monkeys. Neuropharmacology 58, 981e986.

Morin, N., Gregoire, L., Morissette, M., Desrayaud, S., Gomez-Mancilla, B.,Gasparini, F., Di Paolo, T., 2013a. MPEP, an mGlu5 receptor antagonist, reducesthe development of L-DOPA-induced motor complications in de novo parkin-sonian monkeys: biochemical correlates. Neuropharmacology 66, 355e364.

Morin, N., Jourdain, V.A., Di Paolo, T., 2013b. Modeling dyskinesia in animal models ofParkinson disease. Exp. Neurol.. http://dx.doi.org/10.1016/j.expneurol.2013.1001.1024. (Epub ahead of print).

Morishima, Y., Miyakawa, T., Furuyashiki, T., Tanaka, Y., Mizuma, H., Nakanishi, S.,2005. Enhanced cocaine responsiveness and impaired motor coordination inmetabotropic glutamate receptor subtype 2 knockout mice. Proc. Natl. Acad. Sci.U. S. A. 102, 4170e4175.

Morissette, M., Dridi, M., Calon, F., Hadj Tahar, A., Meltzer, L.T., Bedard, P.J., DiPaolo, T., 2006. Prevention of levodopa-induced dyskinesias by a selectiveNR1A/2B N-methyl-D-aspartate receptor antagonist in parkinsonian monkeys:implication of preproenkephalin. Mov Disord. 21, 9e17.

Nadjar, A., Brotchie, J.M., Guigoni, C., Li, Q., Zhou, S.B., Wang, G.J., Ravenscroft, P.,Georges, F., Crossman, A.R., Bezard, E., 2006. Phenotype of striatofugal mediumspiny neurons in parkinsonian and dyskinetic nonhuman primates: a call for areappraisal of the functional organization of the basal ganglia. J. Neurosci. 26,8653e8661.

Nutt, J.G., Gunzler, S.A., Kirchhoff, T., Hogarth, P., Weaver, J.L., Krams, M.,Jamerson, B., Menniti, F.S., Landen, J.W., 2008. Effects of a NR2B selective NMDAglutamate antagonist, CP-101,606, on dyskinesia and Parkinsonism. Mov Disord.23, 1860e1866.

Ouattara, B., Belkhir, S., Morissette, M., Dridi, M., Samadi, P., Gregoire, L.,Meltzer, L.T., Di Paolo, T., 2009. Implication of NMDA receptors in the anti-dyskinetic activity of cabergoline, CI-1041, and Ro 61-8048 in MPTP monkeyswith levodopa-induced dyskinesias. J. Mol. Neurosci. 38, 128e142.

Ouattara, B., Gasparini, F., Morissette, M., Gregoire, L., Samadi, P., Gomez-Mancilla, B., Di Paolo, T., 2010a. Effect of L-DOPA on metabotropic glutamatereceptor 5 in the brain of parkinsonian monkeys. J. Neurochem. 113, 715e724.

Ouattara, B., Gregoire, L., Morissette, M., Gasparini, F., Vranesic, I., Bilbe, G.,Johns, D.R., Rajput, A., Hornykiewicz, O., Rajput, A.H., Gomez-Mancilla, B., DiPaolo, T., 2011. Metabotropic glutamate receptor type 5 in levodopa-inducedmotor complications. Neurobiol. Aging 32, 1286e1295.

Ouattara, B., Hoyer, D., Gregoire, L., Morissette, M., Gasparini, F., Gomez-Mancilla, B.,Di Paolo, T., 2010b. Changes of AMPA receptors in MPTP monkeys withlevodopa-induced dyskinesias. Neuroscience 167, 1160e1167.

Papa, S.M., Boldry, R.C., Engber, T.M., Kask, A.M., Chase, T.N., 1995. Reversal oflevodopa-induced motor fluctuations in experimental parkinsonism by NMDAreceptor blockade. Brain Res. 701, 13e18.

Paquet, M., Smith, Y., 2003. Group I metabotropic glutamate receptors in themonkey striatum: subsynaptic association with glutamatergic and dopami-nergic afferents. J. Neurosci. 23, 7659e7669.

Perroy, J., Raynaud, F., Homburger, V., Rousset, M.C., Telley, L., Bockaert, J., Fagni, L.,2008. Direct interaction enables cross-talk between ionotropic and group Imetabotropic glutamate receptors. J. Biol. Chem. 283, 6799e6805.

Picconi, B., Pisani, A., Centonze, D., Battaglia, G., Storto, M., Nicoletti, F., Bernardi, G.,Calabresi, P., 2002. Striatal metabotropic glutamate receptor function followingexperimentalparkinsonismandchronic levodopa treatment. Brain125,2635e2645.

Pisani, A., Gubellini, P., Bonsi, P., Conquet, F., Picconi, B., Centonze, D., Bernardi, G.,Calabresi, P., 2001. Metabotropic glutamate receptor 5 mediates the potentia-tion of N-methyl-D-aspartate responses in medium spiny striatal neurons.Neuroscience 106, 579e587.

Rajput, A.H., Rajput, A., Lang, A.E., Kumar, R., Uitti, R.J., Galvez-Jimenez, N., 1998.New use for an old drug: amantadine benefits levodopa-induced dyskinesia.Mov. Disord. 13, 851.

Rodriguez-Puertas, R., Herrera-Marschitz, M., Koistinaho, J., Hokfelt, T., 1999.Dopamine D1 receptor modulation of glutamate receptor messenger RNA levelsin the neocortex and neostriatum of unilaterally 6-hydroxydopamine-lesionedrats. Neuroscience 89, 781e797.

Rouse, S.T., Marino, M.J., Bradley, S.R., Awad, H., Wittmann, M., Conn, P.J., 2000.Distribution and roles of metabotropic glutamate receptors in the basal gangliamotor circuit: implications for treatment of Parkinsons disease and relateddisorders. Pharmacol. Ther. 88, 427e435.

Ruzicka, E., Streitova, H., Jech, R., Kanovsky, P., Roth, J., Rektorova, I., Mecir, P.,Hortova, H., Bares, M., Hejdukova, B., Rektor, I., 2000. Amantadine infusion intreatment of motor fluctuations and dyskinesias in Parkinsons disease. J. NeuralTransm. 107, 1297e1306.

Rylander, D., Iderberg, H., Li, Q., Dekundy, A., Zhang, J., Li, H., Baishen, R., Danysz, W.,Bezard, E., Cenci, M.A., 2010. A mGluR5 antagonist under clinical developmentimproves L-DOPA-induced dyskinesia in parkinsonian rats and monkeys. Neu-robiol. Dis. 39, 352e361.

Rylander, D., Recchia, A., Mela, F., Dekundy, A., Danysz, W., Cenci, M.A., 2009.Pharmacological modulation of glutamate transmission in a rat model of L-DOPA-induced dyskinesia: effects on motor behavior and striatal nuclearsignaling. J. Pharmacol. Exp. Ther. 330, 227e235.

Samadi, P., Gregoire, L., Morissette, M., Calon, F., Hadj Tahar, A., Belanger, N.,Dridi, M., Bedard, P.J., Di Paolo, T., 2008a. Basal ganglia group II metabotropic

http://refhub.elsevier.com/S0028-3908(13)00242-6/sref39http://refhub.elsevier.com/S0028-3908(13)00242-6/sref39http://refhub.elsevier.com/S0028-3908(13)00242-6/sref39http://refhub.elsevier.com/S0028-3908(13)00242-6/sref39http://refhub.elsevier.com/S0028-3908(13)00242-6/sref39http://refhub.elsevier.com/S0028-3908(13)00242-6/sref40http://refhub.elsevier.com/S0028-3908(13)0024