Intrastriatal administration of botulinum neurotoxin A normalizes striatal D2 R binding and reduces striatal D1 R binding in male hemiparkinsonian rats

Wedekind et al. - page 1 of 35 Intrastriatal administration of botulinum neurotoxin A normalizes striatal D2R binding and reduces striatal D1R binding in male hemiparkinsonian rats Franziska Wedekind1, Angela Oskamp1, Markus Lang2, Alexander Hawlitschka3, Karl Zilles4,5, Andreas Wree3*, Andreas Bauer1,6* 1 Institute of Neuroscience and Medicine, INM-2, Research Center Jülich, Jülich, Germany 2 Institute of Neuroscience and Medicine, INM-5, Research Center Jülich, Jülich, Germany 3 Institute of Anatomy, Rostock University Medical Center, Rostock, Germany 4 Institute of Neuroscience and Medicine, INM-1, Research Center Jülich, Jülich, Germany 5 Department of Psychiatry, Psychotherapy and Psychosomatics, RWTH Aachen, Aachen, and JARA - Translational Brain Medicine, Aachen, Germany 6 Department of Neurology, Medical Faculty, Heinrich-Heine-University Düsseldorf, Düsseldorf, Germany * Andreas Wree and Andreas Bauer contributed equally to this article. Keywords: Parkinson disease, 6-hydroxydopamine, botulinum neurotoxin A, dopamine receptors, positron emission tomography in vitro receptor autoradiography Correspondence: Franziska Wedekind, Institute of Neuroscience and Medicine, INM-2, Research Center Jülich, Leo-Brandt-Straße, 52425 Jülich, Germany. Email: f.wedekind@fzjuelich.de Wedekind et al. - page 2 of 35 Abstract Cerebral administration of botulinum neurotoxin A (BoNT-A) has been shown to improve disease-specific motor behavior in a rat model of Parkinson disease (PD). Since the dopaminergic system of the basal ganglia fundamentally contributes to motor function, we investigated the impact of BoNT-A on striatal dopamine receptor expression using in vitro and in vivo imaging techniques (positron emission tomography and quantitative autoradiography, respectively). 17 male Wistar rats were unilaterally lesioned with 6-hydoxydopamine (6-OHDA) and assigned to two treatment groups seven weeks later: 10 rats were treated ipsilaterally with an intrastriatal injection of 1 ng BoNT-A, while the others received vehicle (n = 7). All animals were tested for asymmetric motor behavior (apomorphine-induced rotations and forelimb usage) and for striatal expression of dopamine receptors and transporters (D1R, D2R and DAT). The striatal D2R availability was also quantified longitudinally (1.5, 3, and 5 months after intervention) in 5 animals per treatment group. The 6-OHDA lesion alone induced a unilateral PD-like phenotype and a 13% increase of striatal D2R. BoNT-A treatment reduced the asymmetry in both apomorphine-induced rotational behavior and D2R expression, with the latter returning to normal values 5 months after intervention. D1R expression was significantly reduced while DAT concentrations showed no alteration. Independent of the treatment, higher interhemispheric symmetry in raclopride binding to D2R was generally associated with reduced forelimb akinesia. Our findings indicate that striatal BoNT-A treatment diminishes motor impairment and induces changes in D1 and D2 binding site density in the 6-OHDA rat model of PD. Wedekind et al. - page 3 of 35 Significance The pharmacological treatment of Parkinson disease (PD) is characterized by complex and disabling side effects. The reduction of dominant striatal acetylcholine by cerebral application of botulinum neurotoxin A (BoNT-A) proved to be promising in preclinical PD research. We demonstrate that intrastriatal BoNT-A administration normalizes D2 receptor binding in hemiparkinsonian rats, in correlation to behavioral improvements. Additionally, striatal D1 receptor binding is reduced, while presynaptic dopamine transporter expression is not affected. Our longitudinal investigation on the dopamine system points to a complex pattern of reorganization and provides evidence for BoNT-A as a potential alternative or amendment to dopaminergic pharmacotherapy in PD. Wedekind et al. - page 4 of 35 1) Introduction Parkinson disease (PD) is caused by degeneration of neuronal, particularly dopaminergic cells in the substantia nigra pars compacta (Dauer and Przedborski, 2003). The loss of dopaminergic input interferes with the overall transmitter availibility in the striatum leading to an imbalance of transmitter systems including glutamate, gamma-aminobutyric acid, and acetylcholine (ACh) (Calabresi et al., 2006). The strong branching of striatal cholinergic interneurons (ChIs), the large population of medium spiny neurons (MSNs), and the high expression of dopamine receptors underline the particular importance of the cholinergic and dopaminergic transmitter system (Lim et al., 2014). They modulate each other directly via metabotropic muscarinic receptors (on MSNs) and dopaminergic receptors (on ChIs) and indirectly via autoreceptor activation and altered glutamatergic and GABAergic transmission (Kljakic et al., 2017). In the case of dopamine depletion as observed in PD, reduced cholinergic autoreceptor function contribute to the abundance in ACh (Ding et al., 2006). Therefore, the increased activation of metabotropic acetylcholine receptors (AChRs) (accompanied by elevated GABAergic und changed glutamatergic receptor activation) alters the signal transduction of MSNs as the main striatal output, which is already less modulated by dopamine. Besides strategies to compensate dopamine deficits, attempts have been made to restore the transmitter balance – for instance, by antagonizing ACh. As an example, it was recently shown that the ablation of ChIs, the main source of striatal ACh (Oldenburg and Ding, 2011), diminished the risk of induced dyskinesia in L-DOPA-treated hemiparkinson mice (Won et al., 2014). A promising strategy to functionally antagonize striatal cholinergic neurotransmission is local application of botulinum neurotoxin A (BoNT-A), which blocks the release from presynaptic terminals via degradation of the synaptosomal–associated proteins of 25 kDa (Dong et al., 2006; Verderio et al., 2006). BoNT-A is already in use, particularly in symptomatic therapies Wedekind et al. - page 5 of 35 of drooling in PD (Egevad et al., 2014) and in disorders of cholinergic hyperactivity (e.g. dystonia, hyperhidrosis, strabismus, blepharospasm) (Lim and Seet, 2010). We have previously investigated a potential therapeutic application of BoNT-A in the 6hydroxydopamine (6-OHDA) model of PD (Wree et al., 2011). In this model, unilateral injection of 6-OHDA into the medial forebrain bundle (MFB) leads to rapid degeneration of nigrostriatal neurons (Meredith et al., 2008), asymmetry in forelimb use (Schallert et al., 2000), and drug-induced rotations (Ungerstedt and Arbuthnott, 1970). On the molecular level, the loss of nigrostriatal dopamine input causes an imbalance of dopaminergic and cholinergic neurotransmission in the caudate-putamen (CPu). The degeneration of nigrostriatal terminals is also reflected in reduced concentrations of proteins associated with dopamine production and transport (tyrosine hydroxylase and dopamine transporters [DATs] (Sun et al., 2011)), whereas long-lasting impacts of 6-OHDA on proteins related to ACh is not yet clear. For instance, while Ma et al. (2014) reported a decrease in striatal choline acetyltransferase concentration following 6-OHDA infusions, no changes could be observed by Pezzi et al. (2005). We used the 6-OHDA model to test the efficacy of a single intrastriatal administration of BoNT-A. We found significant improvements in motor impairment for up to 6 months after BoNT-A application (Wree et al., 2011). Histologically, a remarkable number of varicosities were observed at neuronal processes, while the overall quantity of striatal neurons, ChIs, and tyrosine hydroxylase levels remained unchanged (Itakura et al., 2014; Mehlan et al., 2016; Wree et al., 2011). The number of varicosities declined over time, while remaining varicosities exhibited permanently elevated volumes (Mehlan et al., 2016). There were no changes in cognitive abilities or striatal volume at the BoNT-A injection site (Antipova et al., 2013; Holzmann et al., 2012). The present study focuses on the molecular changes that underlie the observed improved motor behavior. Downregulation of pathologically increased dopamine D2 receptor (D2R) concentrations is often considered as Wedekind et al. - page 6 of 35 hint for a promising treatment strategy. We hypothesize that the positive impact of BoNT-A on motor impairments in the unilateral 6-OHDA rat model will be mirrored by more balanced dopamine receptor concentrations (especially for D2R) in the CPu, since motor actions are regulated by basal ganglia circuits in which striatal output function is strongly modulated by dopamine. Moreover, we assume that the long-lasting impact on behavior improvement will also be accompanied similarly over time by changes in dopamine receptor concentrations. We designed a longitudinal behavioral and imaging study to investigate components of the dopaminergic system. 6-OHDA rats were treated either with BoNT-A or vehicle and subsequently examined with regard to apomorphine-induced rotation behavior, forelimb use, and dopamine D1 receptor (D1R), D2R and DAT availability with longitudinal positron emission tomography (PET) and quantitative in vitro autoradiography. Wedekind et al. - page 7 of 35 2) Materials and Methods 2.1) Animals Seventeen adult male Wistar rats (Charles River WIGA, Germany, RRID: RGD_737929) with an age of around 2 months were housed under standard conditions (22 °C ± 2 °C; 12 h light/dark cycle) with free access to food and water. The animals were randomly assigned to two treatment groups (Figure 1): 1. Sham treatment (n = 7): All animals underwent unilateral lesioning, intrastriatal vehicle administration, behavioral testing and in vitro examinations (histology and autoradiography). A subset of 5 rats were additionally studied by repeated PET. 2. BoNT-A treatment (n = 10): All animals underwent unilateral lesioning, intrastriatal BoNTA administration, behavioral testing and in vitro examinations (histology and autoradiography). A subset of 5 rats were additionally studied by repeated PET. All procedures were conducted according to protocols approved by the local animal welfare committees (LALLF M-V Rostock and LANUV NRW Recklinghausen, Germany, no. LALLF M-V/TSD/7221.3-1.2-046/10 (1.1-019/08)) 2.2) Unilateral lesioning and treatment To produce unilateral degeneration of dopaminergic neurons and hemiparkinsonian symptoms, rats weighing 280 ± 12 g (n = 16; one animal’s record was lost) were lesioned by injection of 6-OHDA (Sigma-Aldrich, Taufkirchen, Germany; product number H4381) into the right MFB: after intraperitoneal injection of ketamine (50 mg/kg body weight) and xylazine (4 mg/kg body weight) for anesthesia, 4 µl of 6-OHDA (24 µg) dissolved in 0.1 M citrate buffer was delivered over 4 min via a 26-gauge 5-µl Hamilton syringe with the following coordinates referenced from bregma: anterior-posterior = -2.3 mm, lateral = -1.5 mm and ventral = -9.0 mm (Paxinos and Watson, 2007). Wedekind et al. - page 8 of 35 Seven weeks later, rats (356 ± 30 g) received either 1 ng BoNT-A (n = 10) or vehicle (n = 7) into the right CPu; after ketamine-xylazine anesthesia, rats were treated with two individual injections of 1-µl volume containing either BoNT-A (List, USA, purchased via Quadratech, UK) or 0.1% bovine serum albumin in phosphate-buffered saline (“vehicle”). Each intrastriatal injection was delivered over 4 min and according to the following coordinates (with reference to bregma): anterior = 1.3/-0.4 mm, lateral = -2.6/-3.6 mm and ventral = -5.5 mm (Paxinos and Watson, 2007). 2.3) Behavioral testing After successful unilateral 6-OHDA-induced dopamine depletion, animals tend to turn contralaterally to the lesioned side after dopaminergic stimulation (Ungerstedt and Arbuthnott, 1970). Here, 2 weeks before treatment (thus, 5 weeks after 6-OHDA lesioning) and 2 weeks after treatment, all animals were tested on rotational behavior (net contraversive turns per minute) after subcutaneous injection of 0.25 mg/kg body weight apomorphine (Teclapharm, Lüneburg, Germany). Rotations were measured over 40 min, using a selfconstructed automated rotometry device (modified according to Ungerstedt and Arbuthnott (1970)). Rotations were only analyzed in case of complete 360° turns in the left or right direction. All rats were also tested for non-drug-induced forelimb use in the same period (2 weeks before and after treatment) by applying the cylinder test (Schallert et al., 2000): In brief, animals were placed into a transparent cylinder of 20-cm diameter and 30-cm height and videotaped (Sony), with the recordings being analyzed for the number of forelimb-to-wall contacts by a person blinded for animal condition. Out of 30 subsequent contacts, the quotient of left-to-right (L/R) forelimb use was calculated (around 1 = well-balanced use, < 1 = asymmetric use). 2.4) In vivo [11C]raclopride PET Wedekind et al. - page 9 of 35 D2R availability was assessed by longitudinal [11C]raclopride PET in a subset of animals at 1.5, 3, and 5 months after treatment. Two animals (one BoNT-A and one sham-treated rat) were measured simultaneously: Each rat was anesthetized with around 1.6% isoflurane (ReboPharm, Bocholt, Germany) vaporized in 1.25 L/min oxygen gas and placed in a supine position (opposite each other, with the heads aligned, but radial shifted to the center of field of view) in the INVEON dedicated PET scanner (Siemens, Knoxville, TN, USA). Anesthesia duration was longer in the first scan session (F1.22,9.78 = 8.15; n = 10 per session; p = .014 with simple contrasts F1,8 ≥ 8.43; p ≤ .02), but not different between groups at any scan session. Attenuation correction was performed based on a 57 Co-transmission scan (20 min). Animals received an intravenous injection of [11C]raclopride (over 1 min), either at the beginning of emission scan (first rat) or 3 min later (second rat). Alternating the groups for being injected first, times of injection (Ti) did not differ between sham and BoNT-A-treated animals and took place at 11:28 a.m. ± 0:22 (10 rats; 3 time-points per rat). [11C]raclopride was prepared as described previously (Schott et al., 2008) with a radiochemical purity > 95% and a mean specific activity (SA) = 73 ± 34 GBq/µmol (n = 15) at the end of synthesis. SA did not differ between groups, but was higher at the second scan session (59.85 ± 5.40 GBq/µmol [55.68, 64.03] vs. 30.12 ± 20.43 GBq/µmol [14.37, 45.88] at the first session and 27.02 ± 6.95 GBq/µmol [21.66, 32.38] at the last session with F1.15,9.17 = 14.79; n = 10 per session; p = .003 and repeated contrasts F1,8 ≥ 16.10; p ≤ .004). More PET parameters are listed in Table 1. PET data were acquired in list mode for 70 min, normalized and scatter corrected. A dynamic histogram sequence of 6× 10 s, 3× 20 s, 8× 60 s, 4× 300 s and 3× 600 s (24 frames, starting at Ti (Alexoff et al., 2004) was applied. Images were reconstructed 2D as filtered backprojection and with a voxel size of 0.7764 x 0.7764 x 0.796 mm3 (matrix size 128 x 128 x 159). Volumes of the first 17 frames (first 10 min post injection, p. i.) and of all 24 frames were integrated for the coherent registration to the Schiffer’s rat brain template, implemented Wedekind et al. - page 10 of 35 in PMOD software 3.4 (PMOD Technologies, Zurich, Switzerland). The resultant transformation was applied to all individual frames producing reasonable volumes of interests such as 43.55 mm3 (CPu) and 196.98 mm3 (cerebellum) (Mengler et al., 2014). Regional time-activity curves (TACs) were calculated and striatal binding potentials (BPNDs), defined as ratio of specifically bound and nondisplaceable radioligand in tissue at equilibrium, were generated separately for each CPu and for all three consecutive investigations. Modeling based on the simplified reference tissue model (SRTM), with cerebellum as reference and Poisson weighting on measures (Lammertsma and Hume, 1996). For comparison, BPNDs derived from the late phase (starting at 20 min p. i.), with fixed tissue-to-plasma clearance (k2’ = 0.24/min) and referenced to the cerebellum, were determined via Logan reference modeling (Lettfuss et al., 2012). 2.5) Tissue preparation for in vitro studies All rats were sacrificed by decapitation under isoflurane anesthesia 5 months after treatment (rats which received a PET scan were decapitated immediately after the last scan). Each brain was dissected quickly (5 ± 1 min), snap frozen in 2-methylbutane (Sigma-Aldrich, Taufkirchen, Germany; product number 320404), stored at -80 °C and cut into 20 µm coronal cryostat slices of CPu and substantia nigra. The sections (thaw-mounted onto silane-coated slides) were stored at -80 °C until in vitro experiments were performed. 2.6) In vitro [3H]SCH23390, [3H]raclopride and [3H]mazindol autoradiography For D1R binding, striatal sections were pre-treated with 50 mM Tris-HCl buffer (pH 7.4) containing 5 mM KCl, 2 mM CaCl2 (Merck, Darmstadt, Germany; product number 102382), 1 mM MgCl2 (Merck, Darmstadt, Germany; product number 105832) and 120 mM NaCl (Merck, Darmstadt, Germany; product number 106404) for 20 min. Afterwards, binding sites were labeled with 1.67 nM [3H]SCH23390 (SA = 2978.5 GBq/mmol) in buffer supplemented Wedekind et al. - page 11 of 35 with 1 µM mianserin (90 min). For determination of non-specific binding, adjacent slices were incubated likewise in parallel, but in the presence of the D1R antagonist SKF 83566 (1 µM). Finally, sections were washed in ice-cold buffer (2x 10 min) and quickly dipped into ice-cold deionized water. In a similar manner, striatal slices were pre-incubated in 50 mM Tris-HCl buffer (pH 7.4) containing 0.1% ascorbic acid and 150 mM NaCl (Merck, Darmstadt, Germany; product number 106404) for 20 min, followed by incubation in buffer supplemented with 0.57 nM [3H]raclopride (SA = 2753 GBq/mmol) for 45 min to label D2R. Adjacent sections were investigated in the same way, but in the presence of the D2R antagonist (+)-butaclamol (1 µM) for determination of non-specific binding. Slides were rinsed in ice-cold buffer (6x 1 min) and quickly dipped into ice-cold deionized water. For DAT binding, striatal slices were incubated in 50 mM Tris-HCl buffer (pH 7.9) with 120 mM NaCl (Merck, Darmstadt, Germany; product number 106404) and 5 mM KCl (30 min) and with 300 mM NaCl, 5 mM KCl, 300 nM desipramine and 3.98 nM [3H]mazindol (SA = 765.9 GBq/mmol) for 60 min. Sections were washed in buffer (2x 3 min) and quickly dipped in deionized water. All steps were performed on ice and in parallel with the dopamine reuptake inhibitor nomifensine (100 µM during main incubation) to determine non-specific binding on adjacent slices. After binding procedure, slides were dried under a cold ventilator stream and exposed together with [3H] standards of known radioactivity to [3H]-sensitive films (Carestream Health Inc., Rochester, NJ, USA) (12 - 15 weeks). Following film processing and digitalization, grey values were non-linearly transformed into concentrations of radioactivity as described previously (Zilles et al., 2004). Receptor and transporter concentrations (fmol/mg protein) were calculated using the corresponding SA, ligand concentration and dissociation constant (KD; 1.67 nM, 0.55 nM and 4 nM for [3H]SCH23390, [3H]raclopride and Wedekind et al. - page 12 of 35 [3H]mazindol, respectively). Binding was calculated and displayed in color-coded linearized images using Axiovision (Zeiss, Göttingen, Germany, RRID: SCR_002677) and Matlab (Mathworks, Aachen, Germany, RRID: SCR_001622). Per brain, dorsal left and right CPu were defined as regions of interest (ROI) in four to five sections (including one adjacent section with non-specific binding), based on a standard rat atlas (Paxinos and Watson, 2004) and cell body staining. Binding was averaged for each hemisphere and calculated as specific binding, which was finally averaged for each treatment. Since [3H]SCH23390 and [3H]raclopride also bind to other receptor subtypes (e.g. to D5R and to D3R, respectively), binding experiments were optimized for D1R and D2R: they were performed with radioligand concentrations around the corresponding KDs, which were determined in saturation experiments. In addition, specific binding was calculated by subtracting unspecific binding (<13% and 43% of the total binding in receptor and transporter binding assays) from total binding. Accordingly, the data presented in this study reflect D1R and D2R concentrations rather than D1/5R and D2/3R binding sites. Nonradioactive chemicals and tritiated radioligands were purchased from Sigma-Aldrich (Taufkirchen, Germany; product numbers T2913, P9333, M2525, S110, A5960, D033, D3900, N1530), if not noted otherwise, and Perkin Elmer (Rodgau, Germany; product numbers NET-930, NET975, NET816), respectively. 2.7) Histology For optimal ROI definition, histology was performed on corresponding brain sections, which were fixated in neutral-buffered formalin-solution (30 min), washed in deionized water (2x 5 min) and stained with cresyl violet (30 min) for Nissl substance. Additionally, cell bodies or myelin were stained in alternating formalin-fixated sections following the protocols of Merker and Gallyas, respectively (Gallyas, 1979; Merker, 1983). In brief, after several pre-treatment Wedekind et al. - page 13 of 35 and washing steps, slides were transferred to developer solutions containing silver nitrate (70 min for cell body or 30 min for myelin staining, respectively). Sections were washed, fixated in a T-max fixation solution (Kodak, Germany; product number 5089198) and washed again. All sections were dehydrated in ascending solutions of propanol (70% - 100%), cleared in XEM-200 (2x 10 min; Vogel, Germany, purchased via Diatec, Bamberg, Germany; product number 1-019614), mounted in DPX (Sigma-Aldrich, Taufkirchen, Germany; product number 44581) and dried until digital processing using Axiovision (Zeiss, Göttingen, Germany, RRID: SCR_002677). Pretreatment and washing steps in silver-based cell body and myelin stainings were as follows: Pretreatment: overnight incubation in formic acid and hydrogen peroxide (cell bodies) or in pyridine and acetic anhydride (Sigma-Aldrich; Taufkirchen, Germany; product number 822278) (myelin), respectively. Additional incubation in ammoniacal silver nitrate for myelin staining only (30 min). Washing: rinse in demineralized water. Before and after incubation in the developer solutions, washing was performed with 1% acetic acid. Chemicals were obtained from Merck (Darmstadt, Germany; product numbers 104002, 105235, 1015120, 1009966, 100264, 108597, 109728, 100062) if not noted otherwise. 2.8) Statistics Values are expressed as mean ± standard deviation (SD) and statistical analyses were performed using IBM SPSS Statistics 22 (SPSS Inc., Chicago, IL, USA, RRID: SCR_002865) with p < .05 being considered significant. Where appropriate, 95% confidence intervals are displayed in squared parentheses. Gaussian distribution was checked inspecting skewness and kurtosis (raw data and zvalues) and the Shapiro-Wilk normality test. Wedekind et al. - page 14 of 35 Homogeneity of within-variances was revised with Levene test and Hartley Fmax < critical value. Data on forelimb usage was compared between groups with 1-way analysis of covariance (ANCOVA) using the pretreatment scores as covariate, with regression slopes and covariate (unpaired 2-tailed Student t test) being equal across groups. Data on rotational behaviour displayed unequal regression slopes (“treatment*covariate” F1,13 = 9.51; n = 7 – 10; p = .009), hence a general linear model (GLM) analysis was reported. Absolute data from binding experiments were compared between groups by a mixed repeated measure analysis of variance (RM-ANOVA with ‘hemisphere’ as within-factor for autoradiographic data, or ‘hemisphere’ and ‘scan’ as within-factors for PET data, respectively). Homogeneity of between-subject variance was tested and corrected according to Mauchly sphericity test. RMANOVA was followed either by contrasts (scan parameters) or by 2-tailed Student t tests with Bonferroni-correction (binding data), respectively. The p values were multiplied by 3 due to paired (within-subject comparison in each group) and unpaired (between-group comparison of contralateral hemispheres) Student t testing. Binding data, expressed as percentage of contralateral binding, was compared by unpaired 2-tailed Student t test for each scan session and target separately. Parametric analyses of total correlation (thus neglecting the separation into treatment groups) are performed, reporting Pearson r. Wedekind et al. - page 15 of 35 3) Results Effects of intrastriatal BoNT-A treatment on 6-OHDA rats were investigated longitudinally with behavioral testing and PET, and invitro with autoradiography and histological techniques. The time course of experimental procedures is depicted in Figure 1. 3.1) Unilateral 6-OHDA lesion All investigated animals (n = 17) underwent unilateral dopamine depletion by application of 4 µl 6-OHDA into the right MFB. Individual lesion effects were verified both in vivo by assessing apomorphine-induced rotations and spontaneous forelimb-usage 5 weeks after lesioning and in vitro by DAT autoradiography. It is to note that tyrosine hydroxylase quantification was not feasible in non-fixated samples and thus outcome of behavioral tests and quantification of striatal DAT were used to determine nigrostriatal degeneration (Heuer et al., 2012; Liu et al., 2014; Molinet-Dronda et al., 2015). Behavioral testing showed typical asymmetrical rotational movements (8 ± 2.72 contraversive turns/min after apomorphine stimulation; n = 17) and forelimb usage (L/R ratio of 0.36 ± 0.36; n = 17). There were no behavioral differences between animals assigned to treatment and sham groups. After completion of the in vivo investigations, all brains underwent in vitro DAT autoradiography in order to quantify the lesion effect and the impact of BoNT-A on presynaptic DAT. All lesioned brains displayed a strong ipsilateral reduction of DAT (90.01% ± 8.89%; n = 17) without any difference between BoNT-A and sham-treated animals. Furthermore, there was a massive reduction of silver-stained Nissl-positive cell bodies in the ipsilateral substantia nigra (Figure 2A-C). 3.2) BoNT-A treatment The 6-OHDA lesioned rats were assigned to two groups with either BoNT-A (n = 10) or sham (n = 7) treatment. There were no differences between these groups with regard to 6-OHDA Wedekind et al. - page 16 of 35 lesion effects and body weight. All animals underwent behavioral testing before and after treatment (Figure 1). Before treatment, 6-OHDA animals showed turns in opposite direction to the lesioned side and prevalent right forelimb usage. There were no differences between the treatment groups. After BoNT-A treatment, scores for rotational behavior were significantly different with regard to reduced asymmetry (1.32 ± 1.05 vs. 6.78 ± 3.72 contraversive turns/min, GLM with F1,14 = 21.21; n = 7 - 10; p = .0004). In contrast, BoNT-A did not affect forelimb preference, although differences in L/R usage tend to be diminished (0.62 ± 0.37 vs. 0.24 ± 0.36) (Figure 2C). 3.3) Longitudinal PET studies Pairs of BoNT-A and sham-treated 6-OHDA rats (n = 5 per group) were simultaneously investigated with [11C]raclopride PET at 1.5, 3, and 5 months after treatment. Scan parameters and striatal tracer delivery (referenced to cerebellar influx, R1; data not shown) did not differ between both groups (Figure 1 and Table 1). There was a constant trend towards higher [11C]raclopride binding to striatal D2R after 6OHDA-lesioning (17% to 25.7% in ipsilateral CPu; paired Student t tests with t4 ≥ -2.57; n = 5; p = .062 to.114) which tended to return to normal values after BoNT-A administration (8.2% to 1% in ipsilateral CPu) (Figure 3 and Table 2). Mean [11C]raclopride BPND values were not significantly different (even when they were analyzed separately for each single scan session). However, percent changes in [11C]raclopride BPNDs became significantly different between groups at 5 months after treatment (17% in sham-treated vs. -8.2% in BoNT-Atreated animals, unpaired Student t test with t8 = 2.37; n = 5; p = .045). Rats with an ipsilateral increase in interstriatal D2R binding show higher forelimb asymmetry (Pearson r = -0.84, n = 10, p = .002 for the first scan; r = -0.733, n = 10, p = .016 for the second scan and r = -0.773, n = 10, p = .009 for the third PET scan; Figure 4). 3.4) In vitro autoradiography Wedekind et al. - page 17 of 35 Striatal D1R and D2R binding sites were studied with quantitative autoradiography in all animals 5 months after treatment (Figure 5 and Table 3). There was a significant decrease in the mean ipsilateral D1R concentrations in BoNT-Atreated rats (mixed 1-way RM-ANOVA with “hemisphere*treatment” F1,15 = 11.51; n = 7 10; p = .004 and Bonferroni-corrected Student t test, p < .0001) in contrast to sham-operated animals that showed no changes 5 months after treatment. Mean striatal [3H]raclopride binding to D2R was fairly balanced between hemispheres, while it was increased in the lesioned CPu of sham-treated animals (“hemisphere*treatment” F1,15 = 11.61; n = 7 - 10; p = .004 with Bonferroni-adjusted Student t test, p = .039). All sham-treated animals (except for the one with decreased ipsilateral [11C]raclopride binding in PET and smaller reduction in DAT concentration) showed increased D2R concentrations. BoNT-A treatment increased D1R and reduced D2R percentage differences in striatal binding (unpaired Student t test with t15 = 3.17; n = 7 - 10; p = .006 and t15 = 3.49; n = 7 - 10; p = .003, respectively). For in vitro raclopride (D2R) data, there was an excellent correlation to in vivo interhemispheric changes, measured by PET 5 months after treatment (Pearson r = 0.964, n = 10, p < .0001). Wedekind et al. - page 18 of 35 4) Discussion In the present study we explored the longitudinal effects of a BoNT-A application in 6-OHDA lesioned rats on striatal markers of the dopamine system. PET and quantitative in vitro autoradiography showed that SCH23390 binding to D1R was reduced, while raclopride binding to D2R returned to normal 5 months after BoNT-A treatment. Treatment also alleviated turning asymmetry, and individuals with reduced interhemispheric differences in raclopride binding displayed improved symmetry in forelimb use, although BoNT-A was without significant effect on this spontaneous behavior. This is to our knowledge the first study demonstrating that the beneficial effects of BoNT-A on motor behavior are accompanied by a recovery of striatal D2R binding and a reduction in D1R binding. Our findings point to a potential treatment strategy in PD that is not primarily dopamine based but instead initiates a complex pattern of chemical and functional reorganization. A broad range of short and long-term side effects limits the use of anticholinergic medications in PD (Calabresi et al., 2006; Katzenschlager et al., 2003; Lim et al., 2014). BoNT-A injected into the ipsilateral striatum of hemiparkinsonian rats reduces hypercholinism directly at the site of action (Holzmann et al., 2012) and improves motor function for up to 6 months without inflammatory reactions or side effects on cognitive behavior (Antipova et al., 2013; Wree et al., 2011). To understand the molecular mechanisms underlying these actions of BoNT-A, we investigated postsynaptic dopamine receptors which are of crucial importance for basal ganglia circuitry and connectivity. Our findings indicate that BoNT-A affects striatal D1 and D2 binding sites similarly, since both were reduced 5 months after intervention. It is unlikely that this downregulation is caused by neurotoxic effects as numbers of striatal neurons are unchanged up to 12 months after treatment (Antipova et al., 2013). Moreover, it appears implausible that alleviation in 6OHDA-induced D2R binding is independent of BoNT-A, because interhemispheric Wedekind et al. - page 19 of 35 differences have been shown to persist at later stages (Blunt et al., 1992; Sun et al., 2011) and were also detected in the sham-treated animals here. Likewise, results from PET investigations on patients with PD and parkinsonian monkeys suggest that the downregulation of pathologically increased D2R concentrations is most probably caused by long-term treatment with dopamine agonists rather than by disease progression (Doudet et al., 2002a; Seeman and Niznik, 1990). In an earlier study it has also been shown that blockade of muscarinic interactions leads to reduced [11C]raclopride binding in monkeys (Tsukada et al., 2000), we therefore assume that BoNT-A treatment normalizes striatal raclopride binding by an inhibition of ACh neurotransmission. Whether this impact of BoNT-A on D2R availability is time dependent has yet to be clarified, since the variability in our PET data is comparable to that described in the literature (Lettfuss et al., 2012), but the lesion-induced differences may be too small to evaluate treatment effects (differences between sham and BoNT-A treatment did not reach significance in the microPET experiments at 1.5 and 3 months). Regarding D1R and their striatal expression during PD, data are conflicting, showing either no or bidirectional changes in untreated and treated conditions (Blunt et al., 1992; Fornaretto et al., 1993; Guigoni et al., 2005; Hurley and Jenner, 2006). Down-regulation of D1R mRNA and receptor protein by 6-OHDA intervention was small, was only found in animals with medium-sized lesions, and unaffected by intrastriatal implantation of dopamine cells (Blunt et al., 1992; Da Cunha et al., 2008). We therefore hypothesize that, in the present study, the reduction in D1R concentrations is attributable to BoNT-A instead of 6-OHDA intervention. On the other hand, no change in striatal D1R expression was observed in mice lacking the muscarinic M4 AChR (M4 mAChR) (Schmidt et al., 2011), despite the reciprocal involvement of both transmitter targets in striatonigral signaling (Mao et al., 2016) and transmitter release (Tzavara et al., 2004). Thus, whether M4 mAChRs are directly involved in the effect of BoNT-A on striatal D1R needs to be proven. Wedekind et al. - page 20 of 35 Changes in dopamine receptor concentrations as a function of transmitter concentration should be addressed by microdialysis and by multireceptor (especially muscarinic receptor) studies (the latter is currently under investigation within a cooperative study). Furthermore longer time periods should be included, as BoNT-A exhibits its enzymatic activity up to 10 months in primary rat spinal neurons (Whitemarsh et al., 2014), and BoNT-A-induced varicosities on cholinergic neurons were observed up to 12 months (Hawlitschka et al., 2013; Mehlan et al., 2016). Long-lasting effects ≥ 5 months might indicate a persistent disruption of ACh release by BoNT-A or a delayed adaption of receptor expression to changed ACh levels. Preclinical studies on rodents illustrate that unbalanced limb use and turning behavior are reduced when abating the predominance of ACh by nigrostriatal intervention (Maurice et al., 2015; Won et al., 2014). Blockade of M4 mAChR, located on D1 MSNs (Lim et al., 2014), alleviates the akinetic dysfunction in a 6-OHDA lesion model (Ztaou et al., 2016). AChinduced changes in electrophysiology of D2 MSNs (Day et al., 2008) provoke PD-like akinesia in mice (Kondabolu et al., 2016; Maurice et al., 2015). In the present study, we detected dopamine receptor changes after BoNT-A treatment, along with alleviated turning asymmetry described in the literature (Antipova et al., 2013; Wree et al., 2011). We confirmed that BoNT-A dosage was insufficient to improve forelimb deficits (Wree et al., 2011), but individual forelimb akinesia correlated to interhemispheric differences in raclopride binding during the entire time course. These findings raise the question of whether there is a causal link between ACh levels, behavioral alterations, and changes in dopamine receptor binding. Although the link between forelimb asymmetry and imbalance in striatal dopamine receptor concentration is not yet clear, former studies have shown that turning behavior of 6-OHDA lesioned rats was linked to changes in receptor expression. An increase in D2R concentration was detected in contraversive turning rats, and a decrease in D1R expression was found in ipsiversive turning rats 1 week after 6-OHDA lesion (Da Cunha et Wedekind et al. - page 21 of 35 al., 2008). Doudet et al. (2002a) showed that a D2R increase is only present in symptomatic, while D1R downregulation was detected in both, symptomatic and asymptomatic MPTPinduced, untreated parkinsonian monkeys (Doudet et al., 2002b). Blunt et al. (1992) report that in rats a 6-OHDA-induced D2R upregulation was reversed after intrastriatal implantation of fetal dopamine cells, while apomorphine-induced asymmetry in rotation was not completely normalized. Zhou et al. (1996) showed that direct inhibition of striatal D2R mRNA in 6-OHDA lesioned mice temporarily alleviates drug-induced imbalanced turning response independent of mAChR stimulation. Based on a complex underlying circuitry, we assume that blocked ACh release improves asymmetric rotational behavior in part via M4 mAChR on D1 MSNs, possibly by increasing the number of ipsiversive turns, which in turn might be mirrored in reduced D1R concentrations. It would be desirable to investigate BoNTA effects on dopamine receptors along behavioral assessment, to illuminate the exact contribution of both receptor subtypes to long-term behavioral improvements. It should be noted that although the unilateral 6-OHDA model cannot be transferred directly to PD (PD is not 1-sided, PD symptoms occur already with smaller nigrostriatal damage, beneficial effects in the model might be restricted in their efficacy in patients,…), it remains reasonable to obtain an initial estimation of the antiparkinsonian potency of new drugs because of the direct, quantifiable link between nigrostriatal neuron loss and behavior (Przedborski, 2017). In conclusion, our data support previous reports on behavioral improvements after BoNT-A treatment and indicate that BoNT-A leads to receptor-dependent effects on the nigrostriatal dopamine system in the 6-OHDA lesion model of PD. D2R levels, which were ipsilaterally increased by 13% after 6-OHDA lesion, returned to normal at 5 months after striatal BoNT-A treatment. D1Rs were significantly reduced, while DAT showed no alteration. Rotational behavior was alleviated, and individual weak asymmetric forelimb use was correlated with reduced interhemispheric differences in raclopride binding to D2R. These findings on the Wedekind et al. - page 22 of 35 dopamine receptor system point to a complex pattern of reorganization following striatal BoNT-A application and provide evidence for BoNT-A potentials in the treatment of PD. However, further experiments are required to provide direct causal evidence for receptor changes in association to behavioral alterations. Wedekind et al. - page 23 of 35 Acknowledgements The authors thank U. Haase, S. Krause, T. Kroll, S. Lehmann, M. Vögeling and F. Winzer for excellent technical assistance, S. Beer for PET scanner quality assurance, and D. Nabbi, D. Schneider, S. Laskowski, and D. Elmenhorst for helpful discussions. N. Kornadt-Beck, T. Juraschek, S. Holz and N. Hartwigsen are gratefully acknowledged for their support in animal housing. Conflict of interest The authors declare no potential conflicts of interests, including financial, personal, or other relationships, that could inappropriately influence or be perceived to influence the work presented here. Author contributions A.B., A.W., K.Z., and F.W. designed and planned the project. F.W. performed binding experiments and analyzed data; and prepared the manuscript. A.W. and A.H. performed lesioning, BoNT-A treatment, and behavioral testing. A.O. assisted with binding experiments, and M.L. performed [11C]raclopride syntheses. F.W., A.W., K.Z., and A.B. wrote the manuscript. Wedekind et al. - page 24 of 35 References Alexoff DL, Vaska P, Logan J. 2004. Imaging dopamine receptors in the rat striatum with the MicroPET R4: kinetic analysis of [11C]raclopride binding using graphical methods. Methods Enzymol 385:213-228. Antipova V, Hawlitschka A, Mix E, Schmitt O, Drager D, Benecke R, Wree A. 2013. Behavioral and structural effects of unilateral intrastriatal injections of botulinum neurotoxin a in the rat model of Parkinson's disease. J Neurosci Res 91(6):838-847. Blunt SB, Jenner P, Marsden CD. 1992. Autoradiographic study of striatal D1 and D2 dopamine receptors in 6-OHDA-lesioned rats receiving foetal ventral mesencephalic grafts and chronic treatment with L-dopa and carbidopa. Brain Res 582(2):299-311. Calabresi P, Picconi B, Parnetti L, Di FM. 2006. A convergent model for cognitive dysfunctions in Parkinson's disease: the critical dopamine-acetylcholine synaptic balance. Lancet Neurol 5(11):974-983. Da Cunha C, Wietzikoski EC, Ferro MM, Martinez GR, Vital MA, Hipolide D, Tufik S, Canteras NS. 2008. Hemiparkinsonian rats rotate toward the side with the weaker dopaminergic neurotransmission. Behav Brain Res 189(2):364-372. Dauer W, Przedborski S. 2003. Parkinson's disease: mechanisms and models. Neuron 39(6):889-909. Day M, Wokosin D, Plotkin JL, Tian X, Surmeier DJ. 2008. Differential excitability and modulation of striatal medium spiny neuron dendrites. J Neurosci 28(45):11603-11614. Wedekind et al. - page 25 of 35 Ding J, Guzman JN, Tkatch T, Chen S, Goldberg JA, Ebert PJ, Levitt P, Wilson CJ, Hamm HE, Surmeier DJ. 2006. RGS4-dependent attenuation of M4 autoreceptor function in striatal cholinergic interneurons following dopamine depletion. Nat Neurosci 9(6):832-842. Dong M, Yeh F, Tepp WH, Dean C, Johnson EA, Janz R, Chapman ER. 2006. SV2 is the protein receptor for botulinum neurotoxin A. Science 312(5773):592-596. Doudet DJ, Jivan S, Ruth TJ, Holden JE. 2002a. Density and affinity of the dopamine D2 receptors in aged symptomatic and asymptomatic MPTP-treated monkeys: PET studies with [11C]raclopride. Synapse 44(3):198-202. Doudet DJ, Jivan S, Ruth TJ, Wyatt RJ. 2002b. In vivo PET studies of the dopamine D1 receptors in rhesus monkeys with long-term MPTP-induced Parkinsonism. Synapse 44(2):111-115. Egevad G, Petkova VY, Vilholm OJ. 2014. Sialorrhea in patients with Parkinson's disease: safety and administration of botulinum neurotoxin. J Parkinsons Dis 4(3):321-326. Fornaretto MG, Caccia C, Caron MG, Fariello RG. 1993. Dopamine receptors status after unilateral nigral 6-OHDA lesion. Autoradiographic and in situ hybridization study in the rat brain. Mol Chem Neuropathol 19(1-2):147-162. Gallyas F. 1979. Silver staining of myelin by means of physical development. Neurol Res 1(2):203-209. Guigoni C, Aubert I, Li Q, Gurevich VV, Benovic JL, Ferry S, Mach U, Stark H, Leriche L, Hakansson K, Bioulac BH, Gross CE, Sokoloff P, Fisone G, Gurevich EV, Bloch B, Wedekind et al. - page 26 of 35 Bezard E. 2005. Pathogenesis of levodopa-induced dyskinesia: focus on D1 and D3 dopamine receptors. Parkinsonism Relat Disord 11 Suppl 1:S25-S29. Hawlitschka A, Antipova V, Schmitt O, Witt M, Benecke R, Mix E, Wree A. 2013. Intracerebrally applied botulinum neurotoxin in experimental neuroscience. Curr Pharm Biotechnol 14(1):124-130. Heuer A, Smith GA, Lelos MJ, Lane EL, Dunnett SB. 2012. Unilateral nigrostriatal 6hydroxydopamine lesions in mice I: motor impairments identify extent of dopamine depletion at three different lesion sites. Behav Brain Res 228(1):30-43. Holzmann C, Drager D, Mix E, Hawlitschka A, Antipova V, Benecke R, Wree A. 2012. Effects of intrastriatal botulinum neurotoxin A on the behavior of Wistar rats. Behav Brain Res 234(1):107-116. Hume SP, Gunn RN, Jones T. 1998. Pharmacological constraints associated with positron emission tomographic scanning of small laboratory animals. Eur J Nucl Med 25(2):173176. Hurley MJ, Jenner P. 2006. What has been learnt from study of dopamine receptors in Parkinson's disease? Pharmacol Ther 111(3):715-728. Itakura M, Kohda T, Kubo T, Semi Y, Nishiyama K, Azuma YT, Nakajima H, Kozaki S, Takeuchi T. 2014. Botulinum neurotoxin type A subtype 2 confers greater safety than subtype 1 in a rat Parkinson's disease model. J Vet Med Sci 76(8):1189-1193. Wedekind et al. - page 27 of 35 Katzenschlager R, Sampaio C, Costa J, Lees A. 2003. Anticholinergics for symptomatic management of Parkinson's disease. Cochrane Database Syst Rev(2):CD003735. Kljakic O, Janickova H, Prado VF, Prado MAM. 2017. Cholinergic/glutamatergic cotransmission in striatal cholinergic interneurons: new mechanisms regulating striatal computation. J Neurochem. Kondabolu K, Roberts EA, Bucklin M, McCarthy MM, Kopell N, Han X. 2016. Striatal cholinergic interneurons generate beta and gamma oscillations in the corticostriatal circuit and produce motor deficits. Proc Natl Acad Sci U S A 113(22):E3159-E3168. Lammertsma AA, Hume SP. 1996. Simplified reference tissue model for PET receptor studies. Neuroimage 4(3 Pt 1):153-158. Lettfuss NY, Fischer K, Sossi V, Pichler BJ, von Ameln-Mayerhofer A. 2012. Imaging DA release in a rat model of L-DOPA-induced dyskinesias: a longitudinal in vivo PET investigation of the antidyskinetic effect of MDMA. Neuroimage 63(1):423-433. Lim EC, Seet RC. 2010. Use of botulinum toxin in the neurology clinic. Nat Rev Neurol 6(11):624-636. Lim SA, Kang UJ, McGehee DS. 2014. Striatal cholinergic interneuron regulation and circuit effects. Front Synaptic Neurosci 6:22. Liu L, Zhang W, Gong X, Liang X, Wang X. 2014. Relation between microPET imaging and rotational behavior in a parkinsonian rat model induced by medial forebrain bundle axotomy. Behav Brain Res 265:148-154. Wedekind et al. - page 28 of 35 Ma Y, Zhan M, OuYang L, Li Y, Chen S, Wu J, Chen J, Luo C, Lei W. 2014. The effects of unilateral 6-OHDA lesion in medial forebrain bundle on the motor, cognitive dysfunctions and vulnerability of different striatal interneuron types in rats. Behav Brain Res 266:37-45. Mao LM, Wang HH, Wang JQ. 2016. Antagonism of Muscarinic Acetylcholine Receptors Alters Synaptic ERK Phosphorylation in the Rat Forebrain. Neurochem Res. Maurice N, Liberge M, Jaouen F, Ztaou S, Hanini M, Camon J, Deisseroth K, Amalric M, Kerkerian-Le GL, Beurrier C. 2015. Striatal Cholinergic Interneurons Control Motor Behavior and Basal Ganglia Function in Experimental Parkinsonism. Cell Rep 13(4):657666. Mehlan J, Brosig H, Schmitt O, Mix E, Wree A, Hawlitschka A. 2016. Intrastriatal injection of botulinum neurotoxin-A is not cytotoxic in rat brain - A histological and stereological analysis. Brain Res 1630:18-24. Mengler L, Khmelinskii A, Diedenhofen M, Po C, Staring M, Lelieveldt BP, Hoehn M. 2014. Brain maturation of the adolescent rat cortex and striatum: changes in volume and myelination. Neuroimage 84:35-44. Meredith GE, Sonsalla PK, Chesselet MF. 2008. Animal models of Parkinson's disease progression. Acta Neuropathol 115(4):385-398. Merker B. 1983. Silver staining of cell bodies by means of physical development. J Neurosci Methods 9(3):235-241. Wedekind et al. - page 29 of 35 Molinet-Dronda F, Gago B, Quiroga-Varela A, Juri C, Collantes M, Delgado M, Prieto E, Ecay M, Iglesias E, Marin C, Penuelas I, Obeso JA. 2015. Monoaminergic PET imaging and histopathological correlation in unilateral and bilateral 6-hydroxydopamine lesioned rat models of Parkinson's disease: a longitudinal in-vivo study. Neurobiol Dis 77:165-172. Oldenburg IA, Ding JB. 2011. Cholinergic modulation of synaptic integration and dendritic excitability in the striatum. Curr Opin Neurobiol 21(3):425-432. Paxinos G, Watson C. 2004. The Rat Brain in Stereotaxic Coordinates. Amsterdam; Boston: Elsevier Academic Press. Paxinos G, Watson C. 2007. The Rat Brain in Stereotaxic Coordinates. Amsterdam, Boston: Elsevier Academic Press. Pezzi S, Checa N, Alberch J. 2005. The vulnerability of striatal projection neurons and interneurons to excitotoxicity is differentially regulated by dopamine during development. Int J Dev Neurosci 23(4):343-349. Przedborski S. 2017. The two-century journey of Parkinson disease research. Nat Rev Neurosci 18(4):251-259. Schallert T, Fleming SM, Leasure JL, Tillerson JL, Bland ST. 2000. CNS plasticity and assessment of forelimb sensorimotor outcome in unilateral rat models of stroke, cortical ablation, parkinsonism and spinal cord injury. Neuropharmacology 39(5):777-787. Wedekind et al. - page 30 of 35 Schmidt LS, Thomsen M, Weikop P, Dencker D, Wess J, Woldbye DP, Wortwein G, FinkJensen A. 2011. Increased cocaine self-administration in M4 muscarinic acetylcholine receptor knockout mice. Psychopharmacology (Berl) 216(3):367-378. Schott BH, Minuzzi L, Krebs RM, Elmenhorst D, Lang M, Winz OH, Seidenbecher CI, Coenen HH, Heinze HJ, Zilles K, Duzel E, Bauer A. 2008. Mesolimbic functional magnetic resonance imaging activations during reward anticipation correlate with rewardrelated ventral striatal dopamine release. J Neurosci 28(52):14311-14319. Seeman P, Niznik HB. 1990. Dopamine receptors and transporters in Parkinson's disease and schizophrenia. FASEB J 4(10):2737-2744. Sun W, Sugiyama K, Asakawa T, Yamaguchi H, Akamine S, Ouchi Y, Magata Y, Namba H. 2011. Dynamic changes of striatal dopamine D2 receptor binding at later stages after unilateral lesions of the medial forebrain bundle in Parkinsonian rat models. Neurosci Lett 496(3):157-162. Tsukada H, Harada N, Nishiyama S, Ohba H, Kakiuchi T. 2000. Cholinergic neuronal modulation alters dopamine D2 receptor availability in vivo by regulating receptor affinity induced by facilitated synaptic dopamine turnover: positron emission tomography studies with microdialysis in the conscious monkey brain. J Neurosci 20(18):7067-7073. Tzavara ET, Bymaster FP, Davis RJ, Wade MR, Perry KW, Wess J, McKinzie DL, Felder C, Nomikos GG. 2004. M4 muscarinic receptors regulate the dynamics of cholinergic and dopaminergic neurotransmission: relevance to the pathophysiology and treatment of related CNS pathologies. FASEB J 18(12):1410-1412. Wedekind et al. - page 31 of 35 Ungerstedt U, Arbuthnott GW. 1970. Quantitative recording of rotational behavior in rats after 6-hydroxy-dopamine lesions of the nigrostriatal dopamine system. Brain Res 24(3):485-493. Verderio C, Rossetto O, Grumelli C, Frassoni C, Montecucco C, Matteoli M. 2006. Entering neurons: botulinum toxins and synaptic vesicle recycling. EMBO Rep 7(10):995-999. Whitemarsh RC, Tepp WH, Johnson EA, Pellett S. 2014. Persistence of botulinum neurotoxin a subtypes 1-5 in primary rat spinal cord cells. PLoS One 9(2):e90252. Won L, Ding Y, Singh P, Kang UJ. 2014. Striatal cholinergic cell ablation attenuates LDOPA induced dyskinesia in Parkinsonian mice. J Neurosci 34(8):3090-3094. Wree A, Mix E, Hawlitschka A, Antipova V, Witt M, Schmitt O, Benecke R. 2011. Intrastriatal botulinum toxin abolishes pathologic rotational behaviour and induces axonal varicosities in the 6-OHDA rat model of Parkinson's disease. Neurobiol Dis 41(2):291-298. Zhou LW, Zhang SP, Weiss B. 1996. Intrastriatal administration of an oligodeoxynucleotide antisense to the D2 dopamine receptor mRNA inhibits D2 dopamine receptor-mediated behavior and D2 dopamine receptors in normal mice and in mice lesioned with 6hydroxydopamine. Neurochem Int 29(6):583-595. Zilles K, Palomero-Gallagher N, Schleicher A. 2004. Transmitter receptors and functional anatomy of the cerebral cortex. J Anat 205(6):417-432. Ztaou S, Maurice N, Camon J, Guiraudie-Capraz G, Kerkerian-Le GL, Beurrier C, Liberge M, Amalric M. 2016. Involvement of Striatal Cholinergic Interneurons and M1 and M4 Wedekind et al. - page 32 of 35 Muscarinic Receptors in Motor Symptoms of Parkinson's Disease. J Neurosci 36(35):91619172. Wedekind et al. - page 33 of 35 Figure Legends Figure 1 Study design. Male Wistar rats with unilateral 6-OHDA injection into the right MFB received either 1 ng BoNT-A (n = 10) or vehicle (n = 7) into the ipsilateral CPu. Two weeks before and after treatment, animals were tested for motor performance by determining apomorphine (APO)-induced rotational behavior and forelimb use. D2R availability was longitudinally quantified with PET and [11C]raclopride at 1.5, 3, and 5 months after treatment. Finally, D1R, D2R and DAT autoradiography and histological stainings for cell bodies and myelin were performed in vitro subsequent to the last PET scan. Figure 2 Unilateral 6-OHDA lesion. Unilateral 6-OHDA lesioning was proven by asymmetry in striatal DAT concentration, nigral cell body number, and motor behavior. (A) Striatal ipsilateral (ipsi) DAT concentration is reduced compared with contralateral (contra) and comparable between groups. Representative, color-coded coronal images show the total binding (upper row) and the non-specific binding (lower row) with the scale assigning colors to DAT concentrations in picomoles per milligram protein. Error bars denote SD; filled and open symbols indicate BoNT-A and sham treatment, respectively. Iindividual changes are displayed in the insert on the right side. Significance indications: **** p < .0001 (Bonferronicorrected paired Student t test). (B) Nigral cell bodies are present mainly at the contralateral hemisphere (arrowheads), while ipsilateral cell bodies could hardly be found (silver staining, at the top). Cell bodies and myelinated bundles are identified bilaterally in striatal slices adjacent to autoradiographic slices (cresyl violet staining at the middle and silver staining at the bottom, respectively). (C) Both groups showed asymmetric turning behavior in response to apomorphine (on the left) and forelimb use (on the right) 2 weeks before (b) treatment (n = 7 - 10 per group, with error bars denoting SD; filled and open symbols indicate BoNT-A and sham treatment, respectively). There was no difference between groups, but GLM analysis Wedekind et al. - page 34 of 35 detected reduced asymmetry 2 weeks after (a) BoNT-A treatment (only turning behavior). Significance indications: *** p < .001. Figure 3 Effect of BoNT-A on D2R in vivo availability. Longitudinal [11C]raclopride PET in 6-OHDA rats showed that 5 months after treatment, BoNT-A diminishes the interstriatal differences compared with sham treatment. BoNT-A treated rats display reduced interstriatal BPND difference compared with sham-treated rats 5 months after treatment. (A) Coronal (upper row) and horizontal (lower row) parametric images of sham and BoNT-A-treated rats (n = 4 - 5 per group) show increased and normalized [11C]raclopride binding in the ipsilateral (ipsi) CPu compared with contralateral (contra), respectively. At the top, a rat brain atlas (Schiffer) is depicted to illustrate high striatal (red) and low extrastriatal (dark blue) SRTM2modeled BPNDs. (B) Individual regional TACs of [11C]raclopride concentrations (kBq/cc) and the corresponding SRTM-derived modeling curves (solid lines) show that ipsilateral concentrations in the CPu of BoNT-A-treated rats (filled symbols) appear to be normalized compared with contralateral (contra). Both BoNT-A and sham-treated rats (open symbols) were simultaneously investigated 3 months after treatment. (C) BoNT-A diminished the interhemispheric differences in the CPu 5 months after treatment. Filled and open symbols indicate BoNT-A and sham treatment, respectively (n = 5 per group). Top: Individual SRTMderived BPNDs illustrate the interhemispheric difference with stacked lines for increased and solid lines for reduced ipsilateral BPND. Bottom: The mean percent change (with error bars denoting SD) was calculated as (ipsilateral BPND–contralateral BPND)/contralateral BPND*100. Significance indications: * p = .045 (unpaired Student t test; n = 5). Figure 4 Correlations of behavioral outcome and in vivo D2R availibility. Percent changes in BPND, determined in PET 1.5 months after treatment (first scan), were modestly correlated to net contraversive rotational behavior (A) and forelimb usage (B) of 6-OHDA rats (n = 5 per Wedekind et al. - page 35 of 35 group), tested 2 weeks after treatment. Percent changes in SRTM-derived BPND were calculated with (ipsilateral BPND – contralateral BPND)/contralateral BPND *100. Figure 5 Effect of BoNT-A on striatal in vitro dopamine receptor concentrations. BoNT-A reduced striatal D1R and D2R in the CPu of 6-OHDA rats (5 months after treatment). (A) Striatal ipsilateral (ipsi) D1R concentration is reduced compared with contralateral (contra). Representative, color-coded coronal images show the total binding (upper row) and the nonspecific binding (NS, lower row) with the scale assigning colors to D1R concentrations in picomoles per milligram protein. D1R concentrations decreased ipsilaterally only after BoNTA treatment (p = .004; error bars denoting SD). Filled and open symbols indicate BoNT-A and sham treatment, respectively. Individual changes are diplayed in the insert on the right side as solid lines for a decrease and as stacked lines for an increase. Significance indications: **** p < .0001 (Bonferroni-corrected paired Student t test). (B) BoNT-A treatment normalized striatal ipsilateral (ipsi) D2R concentration compared with contralateral (contra). Representative, color-coded coronal images show the total binding (upper row) and the nonspecific binding (NS, lower row) with the scale assigning colors to D2R concentrations in picomoles per milligram protein. D2R concentrations increased ipsilaterally only in shamtreated animals (p = .004; error bars denoting SD). Filled and open symbols indicate BoNT-A and sham treatment, respectively. Individual changes are displayed in the insert on the right side as solid lines for a decrease and as stacked lines for an increase. In vitro changes were highly correlated to in vivo changes in SRTM-derived BPND determined 5 months after treatment (n = 5 per group) and prove a consistent definition of striatal ROIs in both binding approaches. Percentage changes in SRTM-derived BPND were calculated with (ipsilateral binding – contralateral binding)/contralateral binding *100. Significance indications: * p = .039 (Bonferroni-corrected paired Student t test). Wedekind et al. - page 1 of 2 Table 1: Parameters of PET scans. PET 1 PET 2 PET 3 1.5 months after treatment 3 months after treatment 5 months after treatment Sham BoNT-A Sham BoNT-A Sham BoNT-A Weight (g) 471 (44) 453 (32) 527 (56) 490 (30) 573 (70) 525 (46) Anesthesia (min) 99 (21) 99 (18) 89 (14) 84 (15) 88 (15) 83 (9) Activity (MBq) 10.8 (8) 8.9 (5.9) 22.5 (6.7) 20.7 (4.7) 11.9 (4.5) 10.1 (2) Dose (MBq/kg) 23.3 (16.3) 20.2 (14.6) 42.4 (11.1) 42.1 (7.9) 20.6 (6.6) 19.2 (3.7) Mass (nmol/kg) 0.78 (0.1) 0.71 (0.09) 0.71 (0.18) 0.72 (0.16) 0.75 (0.03) 0.73 (0.05) Calculated receptor Occupancy (%) 4.4 (0.6) 4 (0.5) 4 (1.0) 4.0 (0.9) 4.2 (0.2) 4.1 (0.3) Mixed 1-way RM-ANOVA showed no significant differences for each parameter between sham- and BoNT-A-treated animals at any of the 3 PET scans. Parameters are given as mean (SD), n = 5 per treatment group. Activity, dose and mass are adjusted to time of injection (Ti); duration of anesthesia is displayed only as time elapsed until start of emission, and receptor occupancy is calculated based on mass at Ti and ED50 = 17.1 nmol/kg with receptor occupancy = activity/(weight*specific activity*ED50) + 1 (Hume et al., 1998). Wedekind et al. - page 1 of 1 Table 2: Striatal in vivo binding potentials (BPNDs). PET 1 PET 2 PET 3 1.5 months after treatment 3 months after treatment 5 months after treatment Contralateral Ipsilateral SRTM-derived Contralateral Ipsilateral Contralateral SRTM-derived Ipsilateral SRTM-derived Sham 2.02 (0.33) 2.45 (0.43) 1.91 (0.18) 2.4 (0.47) 1.94 (0.20) 2.27 (0.44) BoNT-A 2.03 (0.19) 2.03 (0.22) 1.97 (0.19) 2 (0.4) 1.83 (0.1) 1.68 (0.29) Logan-derived Logan-derived Logan-derived Sham 2.08 (0.31) 2.58 (0.46) 2.00 (0.21) 2.60 (0.54) 2.03 (0.25) 2.44 (0.52) BoNT-A 2.12 (0.22) 2.13 (0.23) 2.1 (0.2) 2.16 (0.44) 1.80 (0.27) 1.72 (0.34) Mixed 2-way RM-ANOVA showed no significant differences between sham- and BoNT-A-treated animals at any of the 3 PET scans. BPNDs are displayed as mean (SD) with n = 5 per treatment group. SRTM and Logan reference modeling were performed on striatal (referenced to cerebellar) TACs. Modeling was performed from minute 20 onwards and with fixed k2’ = 0.24/min in case of Logan-derived BPND. Wedekind et al. - page 1 of 1 Table 3: Striatal in vitro receptor concentrations. D1R Contralateral Ipsilateral (fmol / mg protein) D2R DAT Change Contralateral Ipsilateral Change (%) (fmol / mg protein) (%) Contralateral Ipsilateral (fmol / mg protein) Change (%) Sham 3542 (204) 3418 (294) -3.5 (5.4) 600 (62) 671 (48) $ 12.5 (9.6) 2518 (192) 349 (296) $$$$ -86.6 (10.2) BoNT-A 3714 (213) 3330 (261) $$$$ -10.4 (3.5) ** 637 (28) 610 (64) -4.2 (9.8) ** 2429 (323) 189 (186) $$$$ -92.4 (7.5) Mixed 1-way RM-ANOVA with post hoc testing detected a decrease in ipsilateral D1R (p = .004) and no significant difference in ipsilateral D2R concentrations 5 months after BoNT-A treatment. Unpaired Student t test identified that these findings are different when compared with sham treatment (p = .006 and p = .003, respectively). The decrease in DAT concentration was detected in both groups without intergroup differences (p < .0001). Values are displayed as mean (SD) with n = 7 - 10 per group. The interhemispheric change is calculated for each animal separately by (ipsilateral binding – contralateral binding)/contralateral binding*100 and shown as average mean percent change per group. Absolute values: $ p < .05 and $$$$ p < .0001 versus contralateral binding (Bonferroni-corrected post hoc paired Student t test); percent change: ** p < .01 versus sham treatment (unpaired Student t test).