Abstract
Treatment of Parkinson’s disease (PD) is based on the use of dopaminergic drugs, such as L-Dopa and dopamine receptor agonists. These substances counteract motor symptoms, but their administration is accompanied by motor and non-motor complications. Among these latter conditions a neurobehavioral disorder similar to drug abuse, known as dopamine dysregulation syndrome (DDS), is attracting increasing interest because of its profound negative impact on the patients’ quality of life. Here we replicate DDS in a PD mouse model based on a bilateral injection of 6-hydroxydopamine (6-OHDA) into the dorsal striatum. Administration of L-Dopa induced locomotor sensitization and conditioned place preference in 6-OHDA lesion, but not in control mice, indicative of the acquisition of addictive-like properties following nigrostriatal dopamine depletion. These behavioral effects were accompanied by abnormal dopamine D1 receptor (D1R) signaling in the medium spiny neurons of the dorsal striatum, leading to hyperactivation of multiple signaling cascades and increased expression of ΔFosB, a stable transcription factor involved in addictive behavior. Systemic administration of the D1R antagonist, SCH23390, abolished these effects and the development of place preference, thereby counteracting the psychostimulant-like effect of L-Dopa. The rewarding properties of L-Dopa were also prevented by chemogenetic inactivation of D1R-expressing neurons in the dorsal striatum. Our results indicate the association between abnormal D1R-mediated transmission and DDS in PD and identify potential approaches for the treatment of this disorder.
Introduction
Parkinson’s disease (PD) is classically defined by the progressive loss of dopamine neurons located in the substantia nigra pars compacta (SNc) and projecting to the dorsal striatum, and by the emergence of rigidity, tremor, and bradykinesia [1, 2]. These motor symptoms are commonly treated with dopamine replacement therapy (DRT). However, the prolonged use of L-Dopa and dopamine receptor agonists leads to the development of motor (i.e., dyskinesia) and non-motor complications [3]. These latter conditions include a neurobehavioral disorder similar to drug abuse, referred to as dopamine dysregulation syndrome (DDS). DDS is more frequently observed in patients treated with short-acting drugs, such as L-Dopa or apomorphine, and is characterized by a pathological overconsumption of dopaminergic medications, far beyond that necessary to correct motor disabilities. DDS patients feel under-medicated, ignore advised dose schedules and self-medicate to a state where they only feel “on” when notably dyskinetic [4,5,6,7]. This psychiatric condition represents a major problem for the patients and their relatives and seriously limits the use of drugs commonly employed to manage PD. Whereas the study of the motor complications associated with administration of L-Dopa has been the subject of intense research [8], less is known about the mechanisms implicated in the non-motor side effects caused by this drug.
In PD, administration of L-Dopa regulates the activity of two major populations of striatal projection neurons, referred to as medium spiny neurons (MSN), expressing dopamine D1 or D2 receptors (D1R and D2R). A large body of evidence indicates that the depletion of dopamine occurring in PD leads to enhanced sensitivity of D1R expressed in MSN [8,9,10]. In experimental PD, this phenomenon leads to the hyperactivation of multiple signaling pathways, including the cAMP, extracellular signal-regulated kinase (ERK) and mammalian target of rapamycin (mTOR) cascades, ultimately resulting in long-term modifications of gene expression and protein synthesis. In the striata of rodent and non-human primate models of PD, the transcription factor ΔFosB, a stable splice variant of the immediate early gene FosB, is up-regulated following chronic L-Dopa treatment. These signaling abnormalities have been linked to the emergence of motor complications, such as levodopa-induced dyskinesia [8, 9, 11,12,13]. Importantly, enhanced expression of ΔFosB has been implicated in the long-term effects produced by substances of abuse and natural rewards [14, 15] but its involvement in the acquisition of motor programs at the basis of compulsive behavior in DDS remains to be assessed.
L-Dopa gained rewarding properties in a rat model of PD with a partial loss of dopamine neurons in the SNc induced by viral-mediated expression of α-synuclein [16]. In line with this finding, apomorphine showed a higher propensity to develop psychostimulant-like properties in rats with a 6-hydroxydopamine (6-OHDA) lesion of nigrostriatal dopamine neurons [17]. These studies indicate the possibility of using rodent preclinical models to identify pathological changes responsible for DDS.
In this study, we reproduced features of DDS in a mouse model of PD based on bilateral injection of the neurotoxin 6-OHDA in the dorso-lateral striatum, which leads to a partial degeneration of midbrain dopamine neurons [18]. The rewarding effect of L-Dopa was examined in the conditioned place preference (CPP) and locomotor sensitization paradigms, in parallel to biochemical analyses of different components of the D1R signaling cascade. Pharmacological and chemogenetic inhibition of D1R was tested for its ability to counteract DDS-like behavior and associated biochemical changes produced by L-Dopa.
Material and methods
Animals
Studies were performed in adult (2–4 months old) mice of both sexes. We used C57BL/6J mice (20–25 g; Charles River, Sulzfeld, Germany) and mice expressing enhanced green fluorescent protein (EGFP) under the control of the promoter for D2R (D2-EGFP mice) [19]. Chemogenetic experiments were performed in heterozygous Drd1a-Cre (D1-Cre) transgenic mice (EY262 line, GENSAT project). All mice were acclimatized for one week before surgery. Mice were group-housed under a 12:12 h light-dark cycle with access to food and water ad libitum. Experiments were performed in accordance with the guidelines of Research Ethics Committee of Karolinska Institutet, Swedish Animal Welfare Agency, and European Communities Council Directive 86/609/EEC.
Drugs
6-Hydroxydopamine hydrochloride (6-OHDA, Sigma-Aldrich, Stockholm, Sweden) was dissolved in 0.02% ascorbic acid and 0.9% sterile saline at a free base concentration of 4 μg/μl and injected in the dorso-lateral striatum. 3,4-dihydroxy-l-phenylalanine (L-Dopa, 10 mg/kg in sterile saline; Sigma-Aldrich, Stockholm, Sweden) was administered intraperitoneally (i.p.) together with the DOPA decarboxylase inhibitor benserazide hydrochloride (7.5 mg/kg in saline; Sigma-Aldrich, Stockholm, Sweden) at a volume of 10 ml/kg bodyweight. Mice were treated with L-Dopa (+Benserazide) alone or in combination with selective inhibitors: SCH23390 (SCH, 0.25 mg/kg in 0.9% sterile Saline; Tocris‐Biotechne Ltd., Abingdon, UK) was administered i.p. 10 min before L-Dopa at 10 ml/kg bodyweight. Rapamycin (Rapa, 5 mg/kg in 5% DMSO, 5% Tween20, 15% PEG in dH2O; LC Laboratories, Woburn, USA) and PD0325901 (PD03, 5 mg/kg in 10% DMSO, 5% Tween20, and 15% PEG in dH20; Selleckchem, Planegg, Germany) were injected i.p. at 2 ml/kg and 45 min before L-Dopa injection. For chemogenetic experiments, Clozapine N-oxide (CNO, 1 mg/kg in 5% DMSO in 0.9% sterile saline; Tocris‐Biotechne Ltd., Abingdon, UK) was administered at 10 ml/kg bodyweight 20 min before L-Dopa. Control mice were injected with vehicle, accordingly.
Stereotaxic surgery
All surgical procedures were performed in 2.5–3-month-old mice. Animals received a bilateral partial lesion with the neurotoxin 6-OHDA using a well-established protocol [18]. Briefly, mice were anesthetized with isoflurane and positioned in a stereotaxic frame (Stoelting, Wood Dale, Il, USA) equipped with a heating pad to maintain normothermia. Animals were injected subcutaneously with an analgesic (0.1 mg/kg Temgesic; Apoteket, Stockholm, Sweden) and anesthetic cream (EMLA, 2.5% lidocaine, 2.5% prilocaine; Apoteket, Stockholm, Sweden) was applied locally. Mice received ophthalmic ointment to prevent corneal dryness. 6-OHDA hydrochloride was freshly dissolved and 1.25 µl was injected into each dorsal striatum at a rate of 0.2 µl/min according to the following coordinates (mm from bregma): anteroposterior +0.6; medio-lateral ±2.2 and dorsoventral −3.2. The injector was left in place for an additional 5 min, allowing the solution to diffuse. Control (sham) mice received bilateral injections of 1.25 μl vehicle. Mice were allowed to recover for 3 weeks before experiments. We followed an enhanced pre- and post-operative care protocol to minimize mortality to <5% [18]. For chemogenetic experiments D1-Cre (+/-) or wildtype (-/-) mice were injected with 6-OHDA bilaterally (CPP experiment), or unilaterally (rotation test), following the same protocol as described above. After 3 weeks recovery, mice received bilateral (CPP experiment) or unilateral (rotation test) infusions of a Cre-inducible-adeno-associated virus carrying the gene for the inhibitory Designer Receptors Exclusively Activated by Designer Drugs (Gi-DREADD) and were allowed to recover for three more weeks before behavioral experiments. The Gi-DREADD (pAAV5-hSyn-DIO-hM4D(Gi)-mCherry; titer ≥7 × 1012 vg/mL, Addgene plasmid #44362) was injected (0.5 µl per hemisphere) at a rate of 0.1 µl/min according to the following coordinates (mm from bregma): anteroposterior +0.6; medio-lateral ±2.2 and dorsoventral −3.2.
Conditioned place preference
CPP is a form of Pavlovian conditioning used to assess the rewarding and motivational properties of addictive drugs [20]. The effect of L-Dopa was tested in a CPP paradigm performed over 8 consecutive days. The apparatus consisted of a white acrylic two-compartment chamber (20 × 20 × 50 cm each compartment). The two compartments were separated by detachable doors and differed in tactile cues on the floor (grooved vs. smooth). CPP procedures consisted of 3 phases: pre-conditioning, conditioning, and post-conditioning. During pre-conditioning, mice were given free access to both compartments for 15 min. Spontaneous preference was evaluated by measuring the average time each animal spent in either of the two compartments and animals spending >85% of time in one compartment were excluded from analysis due to a strong preference bias. During the conditioning phase, mice were injected with L-Dopa alone or in combination with SCH23390, rapamycin, PD0325901 or CNO (administered 10, 45, 45 or 20 min before L-Dopa, respectively) and immediately placed for 40 min in the “least-preferred” compartment (biased research design of CPP [21]). In the next session, mice received a vehicle injection and were placed in the opposite compartment. The procedure was repeated for a total of 12 times (6 days, 2 sessions per day). Accordingly, mice received 6 injections of drug paired to one specific compartment and the same number of vehicle injections paired to the opposite compartment. Control groups were injected with vehicle in both compartments during the entire conditioning phase. In the post-conditioning phase (24 h after the last conditioning session), mice were given free access to both compartments for 15 min in a drug-free state and the time spent in each of the two compartments was measured. In a separate experiment, CPP was examined 7 days after the last conditioning session. To test the effect of mTORC1 and ERK inhibitors on CPP memory recall [22], control mice and mice conditioned with L-Dopa alone received vehicle, rapamycin or PD0325901 45 min before the post-conditioning test. Reward-seeking behavior was assessed by comparing the time spent in the drug-paired compartment during pre-conditioning and post-conditioning stages. Preference score was calculated as post- minus pre-conditioning time spent in the drug-paired compartment. Behavior was video-recorded and analyzed by a video tracking software (Ethovision XT16, Noldus, The Netherlands).
Locomotor activity
Locomotor activity was initially assessed during the CPP test by measuring the distance (m) moved during the 40 min daily conditioning trials. These results were further characterized by testing the mice in locomotor activity boxes (45 × 45 × 35 cm). In this experiment, locomotor activity was assessed in 15 min trials during 4 different phases: habituation, treatment, extinction and reinstatement. During habituation, mice were allowed to explore the arena in a drug-free state. In the treatment phase, locomotor activity was measured daily for 6 days following saline or L-Dopa injections. Extinction was measured 24 h and 7 days after the last treatment. Finally, mice received a priming injection of saline or L-Dopa and reinstatement of locomotor activity was assessed. Mice were video-recorded, and motor behavior was analyzed by a video tracking software (Ethovision XT16, Noldus, The Netherlands).
Rotation behavior
Unilateral 6-OHDA lesion D1-Cre (+/-) or wildtype (-/-) mice expressing Gi-DREADD in the lesion hemisphere were placed in individual glass cylinders (12 cm diameter) and motor activity was video-recorded. Following a 20 min habituation, mice were injected with CNO and, 20 min later, with L-Dopa. The number of ipsilateral and contralateral rotations was manually counted for 90 min after L-Dopa injection by an observer blind to experimental groups.
Western blot
Mice were sacrificed by decapitation, heads were immediately immersed in liquid nitrogen for 5 s, and the brains were rapidly removed. Tissue punches of dorsal and ventral striata from both hemispheres were dissected, sonicated in 1% SDS, and boiled for 10 min in sample buffer. The resulting homogenates were stored at −20 °C until further use. Aliquots of 2 × 5 µl of each homogenate were used for protein quantification with a BCA assay kit (Pierce, Rockford, IL, USA). Equal amounts of protein (25 µg/sample) were separated by SDS-PAGE and transferred overnight to nitrocellulose membranes (0.45 micron, Thermo Scientific, Germany). After the transfer, membranes were placed on a shaker, washed for 5 min in phosphate-buffered saline (PBS), and then blocked in Odyssey blocking buffer (Li-Cor Bioscience, Lincoln, NE, USA) for 1 h at room temperature. Next, membranes were incubated in primary antibodies: anti-tyrosine hydroxylase (TH; 1:2000, #AB152, Merck); anti-p44/42 MAPK ERK1/2 and anti-phospho-p44/42 MAPK ERK1/2 (1:2000, #9107 and #4377, Cell Signaling); anti-S6 ribosomal protein, anti-phospho-S6 ribosomal protein Ser 235/236 (1:1000, #2317 and #2215, Cell Signaling); anti-beta-actin (1:30000, #A5316, Merck); anti-FosB (1:1000, #sc-48, Santa Cruz Biotechnology) for two hours at room temperature and afterwards washed with PBS/0.1% Tween. Detection was based on fluorescent secondary antibody binding (IR Dye 800CW and 680RD; Li-Cor Bioscience, Lincoln, NE, USA), 1.5 h at room temperature. Quantification was performed with a Li-Cor Odyssey infrared fluorescent detection system, using the software Image Studio to quantify signal intensities. Data were calculated as % of control and phosphoproteins were normalized for the amount of the corresponding total protein detected in the sample or normalized for the amount of the corresponding beta-actin protein.
Immunohistochemistry
Mice were anesthetized with pentobarbital (1:1 in 0.9% sterile saline) and perfused with 4% paraformaldehyde (PFA, Sigma-Aldrich, Darmstadt, Germany) in PBS (pH 7.4). Brains were dissected, post-fixed overnight in 4% PFA under continuous agitation at 4 °C, and cut into coronal sections (40 µm) using a vibratome (Leica VT 1000S, France). Tissue sections were stored at −20 °C in a solution containing 30% ethylene glycol, 30% glycerol and 0.1 M PBS until they were processed for immunofluorescence. Staining was performed on selected free-floating sections of the striatum and midbrain. Sections were permeabilized with PBS/0.1% Triton X-100 solution and blocked in PBS/0.3% Triton X-100 and 10% normal goat serum for 1.5 h. Sections were then incubated at 4 °C overnight with primary antibodies: anti-TH (1:1000, #AB152, Merck); anti-phospho-p44/42 MAPK ERK1/2 (1:1000, #4377, Cell Signaling); anti-phospho-S6 ribosomal protein Ser 235/236 (1:1000, #2215, Cell Signaling); anti-FosB (1:200, #sc-48, Santa Cruz Biotechnology); anti-GFP (1:1000, GFP-1020, AvesLabs); anti-RFP (1:1000; #600-401-379, Rockland) and subsequently with secondary antibody (1:500, Alexa Fluor 488, 594, 647; Jackson and ImmunoResearch) for one hour at room temperature. Finally, sections were mounted on poly-L-lysine prepared glass slides (Sigma diagnostic, USA) and covered with DABCO media (1,4-Diazabicyclo [2.2.2] octane powder in glycerol solution).
Imaging and cellular counting
Representative images of the full striatum and midbrain were acquired at 10x (stitched tiling) and images for cell counting were acquired at 20x by confocal microscopy (LSM880, Zeiss, Germany) with the Zen software (Zeiss, Germany). Cell counts were performed in 3 sections per animal in the dorso-lateral striatum in proximity to the injection site or in 6 midbrain sections per animal, containing SNc and ventral tegmental area (VTA). Images were processed with ImageJ software (Scion Corporation, USA) and cell counting was performed by a researcher blind to treatment and groups.
Statistical analyses
Sample size was calculated using G Power 3.1 [23] and was based on previously published studies. Data were analyzed using GraphPad Prism (Version 9, GraphPad Software, La Jolla, CA, USA) and tested for normality with the D’Agostino & Pearson test. Mice were randomly assigned to experimental groups and analysis was performed by a researcher blind to experimental groups. Data are presented as mean ± SEM unless stated otherwise. Statistical significance is presented as: *p < 0.05, **p < 0.01, ***p < 0.001. Figure legends contain p-values and number of animals and/or replicates. For 2-group comparisons, unpaired one-tailed Student’s t tests were used. For 3 or more groups, a one-way ANOVA with Dunnett’s post-hoc test (vs. control group) was used. For experiments with multiple groups and longitudinal data, two-way ANOVA, followed by Bonferroni post-hoc test (vs. control for each time point) was performed.
Results
Partial dopamine depletion in the bilateral striatal 6-OHDA lesion model
Dopaminergic degeneration after bilateral injection of 6-OHDA in the dorsal striatum was examined 3 weeks post-surgery by western blot and immunohistochemistry using an antibody against TH, the rate-limiting enzyme of catecholamine biosynthesis. Western blot analysis of the whole (total) striatum showed a 54.3% reduction of TH levels in 6-OHDA lesion mice compared to sham lesion mice (Fig. 1A left). 6-OHDA caused a larger reduction of TH levels in the dorsal striatum (77.3%, Fig. 1A right and B upper panel) compared to the ventral striatum (nucleus accumbens) (13.7%, Fig. 1A right and B lower panel). Immunohistochemical analysis of the SNc revealed a 59.8% reduction in the number of dopaminergic cell bodies (Fig. 1C left and D), which mainly project to the dorsal striatum. In the VTA, which preferentially projects to the ventral striatum, dopaminergic cell bodies were less affected (Fig. 1C right and D; 15.9% reduction). In line with previous work [18], we confirmed that the infusion volume and concentration of 6-OHDA (1.25 µl, 4 μg/μl) results in a restricted lesion, limited to the dorsal striatum and its dopaminergic afferents from the SNc….
