Autoradiographic labelling of metabotropic glutamate type 2/3 receptors in the hemi-parkinsonian rat brain

L-3,4-dihydroxyphenylalanine (L-DOPA) is the treatment of choice for Parkinson ’ s disease (PD) motor symptoms, but its chronic use is hindered by complications such as dyskinesia. Pre-clinical studies discovered that activation of metabotropic glutamate type 2 and 3 (mGlu 2/3 ) receptors alleviates L-DOPA-induced dyskinesia. To gain mechanistic insight into the anti-dyskinetic activity of mGlu 2/3 activation, we performed autoradiographic binding with [ 3 H]-LY-341,495 in brain sections from L-DOPA-treated 6-hydroxydopamine (6-OHDA)-lesioned rats that developed mild or severe dyskinesia, as well as L-DOPA-untreated


Introduction
The motor symptoms of Parkinson's disease (PD) manifest due to the progressive degeneration of dopaminergic neurons of the substantia nigra (SN) pars compacta (SNc) and the subsequent alterations in neurotransmission within the cortexbasal gangliathalamuscortex circuit (DeLong and Wichmann, 2007;Hornykiewicz and Kish, 1987).Dopamine replacement therapy with L-3,4-dihydroxyphenylalanine (L-DOPA) is the standard symptomatic treatment for PD (Fox et al., 2018).However, chronic L-DOPA therapy can lead to debilitating complications, with nearly 95 % of PD patients developing dyskinesia after 15 years of treatment (Ahlskog and Muenter, 2001;Hely et al., 2005).Currently, amantadine, a non-selective N-methyl-D-aspartate (NMDA) receptor antagonist, remains the sole anti-dyskinetic treatment approved by the FDA (Perez-Lloret and Rascol, 2018), but its use is limited by emergence of tolerance to the anti-dyskinetic effects (Thomas et al., 2004) and a tendency to cause hallucinations (Hauser et al., 2017).Although the exact mechanisms underlying dyskinesia remain unknown, excessive glutamatergic activity within the striatum is largely regarded as a tenet of L-DOPA-induced dyskinesia (Cenci, 2014;Huot et al., 2013).As such, the glutamatergic system is a key target in the pursuit of discovering novel therapies for PD patients afflicted with dyskinesia, and a dampening of glutamatergic transmission is thought to underlie, at least in part, the anti-dyskinetic benefits of amantadine (Rascol et al., 2021).
Although mGlu 2 PAMs and mGlu 2/3 orthosteric agonists have shown promising results, the anti-dyskinetic activity of these molecules has been investigated only at the behavioural level, and mechanistic studies are lacking.Hence, a better understanding of brain areas in which mGlu 2/3 receptor levels are altered may provide insight into the possible mechanisms that underlie the anti-dyskinetic effects of mGlu 2/3 activation.Autoradiographic binding studies using the mGlu 2/3 orthosteric antagonist [ 3 H]-LY-341,495 have reported mGlu 2/3 receptor binding in the caudate nucleus, putamen, and globus pallidus (GP), both pars interna (GPi) and pars externa (GPe) in PD patients (Samadi et al., 2009) and MPTP-lesioned macaques with and without dyskinesias (Samadi et al., 2008).However, these pioneering studies did not explore mGlu 2/3 receptor binding levels along the other structures of the motor cortexbasal gangliathalamusmotor cortex loop (McGregor and Nelson, 2019), i.e. the subthalamic nucleus (STN), motor cortex and thalamus (ventral anterior and ventral lateral [VA/VL] nuclei), and the full picture therefore remains incomplete.Here, we sought to quantify mGlu 2/3 receptors along the motor cortexbasal gangliathalamusmotor cortex loop by performing autoradiographic binding with [ 3 H]-LY-341,495 in the 6-OHDA-lesioned rat.In addition, we tested for correlations between the severity of L-DOPA-induced dyskinesia of 6-OHDA-lesioned animals and [ 3 H]-LY-341,495 specific binding levels in the examined brain areas.

Animals
Adult female Sprague-Dawley rats (250-275 g, Charles River, Canada) (N = 32) were housed in groups of three under conditions of controlled temperature (21 ± 1 • C), humidity (55 %) and light (12-hour light/dark cycle, lights on at 07:00).Rats were selected from a local rodent brain bank created to study the molecular underpinnings of parkinsonism and dyskinesia (Frouni et al., 2024).Additional details about this brain bank and the methodology employed to create it can be found in (Kwan, 2023).Rats were allowed unrestricted access to food and water.Upon arrival, rats were left undisturbed for one week to acclimatise to housing conditions.All animal care and experimental procedures were approved by the Montreal Neurological Institute-Hospital (The Neuro) and McGill University Animal Care Committees, adhering to the regulations established by the Canadian Council on Animal Care.

Assessment of hemi-parkinsonism
At the end of the 3-week recovery, the cylinder test to evaluate the hemi-parkinsonian severity (Schallert et al., 2000).Rats were put in a transparent cylinder (14 cm diameter × 28 cm height), and were filmed for 10 minutes (Frouni et al., 2019b;Kwan et al., 2020).The recordings were visualised and scored by a blinded rater who quantified the number of contacts each forepaw made with the wall of the cylinder (use of forepaw contralateral or ipsilateral to the lesion, and concurrent bilateral forepaw use).Rats that preferentially used the forepaw ipsilateral to the lesion in ≥ 70 % of their rearing, a threshold that indicates ≥ 88 % striatal dopamine depletion (Schallert et al., 2000), were selected for the experiments detailed below.

Experimental design
The experimental flow and grouping of animals are presented in Fig. 1.Rats were randomly assigned to 4 different groups.Within any given treatment group, drug administration order was randomised daily.Group A (sham-lesioned, daily treatment with vehicle, N = 8) animals were injected with 6-OHDA vehicle and were later injected with L-DOPA/benserazide vehicle (0.9 % saline with 0.1 % ascorbate) s.c.once daily for 14 days.The rats that were unilaterally injected with 6-OHDA were assigned to groups B, C, and D. Group B (6-OHDA-lesioned, L-DOPA-untreated, N = 9) rats received a daily s.c.injection of L-DOPA/ benserazide vehicle for 14 days.The remaining 6-OHDA-lesioned rats were treated with L-DOPA/benserazide (10/15 mg/kg, s.c.) once daily for 14 days.The randomisation followed simple randomisation strategy and was performed using the GraphPad online random number generator (https://www.graphpad.com/quickcalcs/randomize1/,GraphPad Software Inc, Boston, MA, USA).
After this 14-day treatment period, the severity of abnormal involuntary movements (AIMs), based on the axial, limbs and orolingual (ALO) subtypes of dyskinesia (described in Section 2.3.2) was quantified.Rats with a cumulative ALO AIMs of score < 50 were assigned to Group C (6-OHDA-lesioned, L-DOPA treatment, low dyskinesia, N = 7), while rats that received a cumulative ALO AIMs score ≥ 50 were allocated to Group D (6-OHDA-lesioned, L-DOPA treatment, severe dyskinesia, N = 8).

Dyskinesia rating
The severity of ALO AIMs was assessed by an observer blinded to the lesion status and treatment group.Both AIMs duration and AIMs amplitude for each of the ALO body segments were evaluated, according to a scale previously described (Cenci and Lundblad, 2007).AIMs duration and amplitude scores for each body segment were measured using a 0-4 scale during 2 minutes, every 20 minutes, over a 180-minute experimental session (Cenci and Lundblad, 2007).For each of duration and amplitude, individual ALO body segment AIMs scores at any observation point were summed to produce the ALO AIMs score of said time point.The cumulative ALO AIMs score for an animal was obtained by the addition of the duration and amplitude components over the totality of the 3-hour testing session.

Tissue preparation
Forty-five minutes after receiving their regular treatment based upon their group assignment, rats were anaesthetised with 4 % isoflurane in 100 % oxygen (1 L/min) and perfused with 0.9 % NaCl transcardially.Brains were extracted and flash frozen in isopentane at − 56 • C, after which they were stored at − 80 • C until use.Brains were sectioned in optimal cutting temperature compound into 12-µm thick coronal sections using a cryostat (Leica CM3050 S, Leica Biosystems, Concord, ON, Canada).Sections were thaw-mounted on SuperFrost Plus slides (Thermo Fisher Scientific, Saint-Laurent, QC, Canada) and, after drying at room temperature, slides were stored at − 80 • C until use.

Immunohistochemistry
Immunohistochemical staining was performed on striatal brain sections from all sham-lesioned and all 6-OHDA-lesioned rats.Brain sections were thaw-mounted on slides and dried overnight at room temperature.They were post-fixed by immersion in acetone (-20 • C) for 10 minutes, air dried for 20 minutes (Matsumoto, 1985), quenched in 0.5 % H 2 O 2 for 10 minutes, then rinsed in Tris buffered saline (TBS: 240 mM NaCl in 100 mM Tris-HCl, pH 7.40) 3 times for 5 minutes each.Sections were then incubated in a TBS solution containing 0.3 % Triton X-100, 10 % normal goat serum (NGS) and 5 % bovine serum albumin (BSA) for 1 hour, following which they were rinsed with TBS 3 × 5 minutes as before.Afterwards, sections were incubated overnight at 4 • C with mouse monoclonal antibody raised against tyrosine hydroxylase (TH) (catalogue number: MAB318; dilution: 1:1000; Rats injected with 6-OHDA were evaluated for the severity of their hemi-parkinsonsim, after which they were administered L-DOPA and the severity of their abnormal involuntary movements was quantified (A).Rats were divided into four different groups; sham-lesioned, vehicle-treated rats were assigned to Group A; 6-OHDAlesioned, vehicle-treated rats were assigned to Group B; 6-OHDA-lesioned, L-DOPA-treated rats with ALO AIMs score < 50 were assigned to Group C; and 6-OHDA-lesioned, L-DOPA-treated rats with ALO AIMs scores ≥ 50 were assigned to Group D (B).6-OHDA, 6-hydroxydopamine; AIMs, abnormal involuntary movements; ALO, axial, limbs and orolingual; L-DOPA, L-3,4-dihydroxyphenylalanine.
Brain sections were imaged with a Zeiss Axio Observer Z1 microscope (The Neuro Microscope Core Facility) using the Zeiss ZEN Pro Microscopy software (Carl Zeiss AG, Germany, version 3.0).THimmunoreactivity was measured in the dorsolateral striatum of both hemispheres through optical densitometry using the ImageJ software (NIH, version 1.53a).The relative TH-immunoreactivity of the lesioned hemisphere was calculated as a percentage of the non-lesioned hemisphere.This measurement was performed on four adjacent brain sections of each animal, after which an average relative TH optical density was determined for each animal.

[ 3 H]-LY-341,495 autoradiographic binding
The brain regions of interest were selected based upon their involvement in the motor loop of the cortexbasal gangliathalamuscortex circuit.The brain areas were identified according to a rat brain atlas (Paxinos and Watson, 2007).Autoradiographic binding to mGlu 2/3 receptors with [ 3 H]-LY-341,495 was adapted from previously published protocols (Frank et al., 2011;Wright et al., 2001).Slides encompassing the brain regions under investigation were removed from their storage at − 80 • C and thawed overnight at room temperature.On the day of the experiment, sections were first incubated in a 50 mM Tris buffer (pH 7.6) containing 10 mM KH 2 PO 4 , 100 mM KCl twice for 10 minutes, then dried for 60 minutes at room temperature.Slides selected to define total binding were then incubated in the same Tris buffer solution as above, which contained 5.0 nM of [ 3 H]-LY-341,495 (METIS Laboratories, Valley Stream, NY, USA; specific activity: 54 Ci/mmol) for 60 minutes at room temperature.For their part, slides employed to define non-specific binding were incubated in the same Tris buffer solution as above, in the presence of both 5.0 nM of [ 3 H]-LY-341,495 and 1 mM of L-glutamic acid (MilliporeSigma) for 60 minutes at room temperature.Sections were then washed twice for 1 minute each at 4 • C in the Tris buffer solution previously described, and rinsed for 30 seconds with ddH 2 O at 4 • C. Lastly, they were air dried at room temperature for 2 hours.Once dried, sections were apposed to [ 3 H]-sensitive Biomax MR films (Milli-poreSigma) along with [ 3 H]-microscale standards (ART0123B and ART0123C, American Radiolabeled Chemicals, St. Louis, MO, USA, 5 mm × 7 mm) for 3 weeks at room temperature.
Following the 3-week incubation period, films were developed and autoradiograms were analysed through optical densitometry measurement with ImageJ software.The [ 3 H]-microscale standards were used to calculate a reference curve of radioactivity cf grey values of autoradiograms (Zilles et al., 2002) which enabled to quantify signal intensity as nanocuries per milligram of tissue (nCi/mg).For each rat, for each brain area of interest, 4 consecutive sections were allocated for total binding, and another 4 were allocated for non-specific binding.Background intensity was subtracted from each signal reading.Signal intensity obtained from total binding and non-specific binding was calculated the same way.Specific binding was determined by subtracting non-specific binding value from the total binding.Signal intensity (nCi/mg) was divided by the specific activity of the radioligand (Ci/mmol) for conversion into femtomole of receptor per milligram of tissue (fmol/mg).

Statistical analysis
Relative striatal TH optical density values are presented as a percentage of the hemisphere contralateral to 6-OHDA injection.After ensuring that combining the TH data of all 6-OHDA-lesioned animals would not violate the Levene's test for equality of variances, TH optical density values were pooled together and compared to sham-lesioned animals with the Student's t test.ALO AIMs scores of 6-OHDA-lesioned animals treated chronically with L-DOPA are displayed as the median ± semi-interquartile range.ALO AIMs scores between different groups were analysed by Mann-Whitney U test.[ 3 H]-LY-341,495 specific binding levels are shown as the mean value ± standard deviation (SD).Multiple unpaired t-tests with Welch's correction were performed to compare each animal group with every other group (group A: shamlesioned; group B: 6-OHDA-lesioned, vehicle-treated; group C: 6-OHDA-lesioned, L-DOPA-treated, mildly dyskinetic; group D: 6-OHDAlesioned, L-DOPA-treated, severely dyskinetic).Correlations between ALO AIMs scores and [ 3 H]-LY-341,495 specific binding levels were investigated using the Pearson correlation coefficient.Statistical significance was set to P < 0.05.Statistical analyses were computed with GraphPad Prism 9.2.0 (GraphPad Software Inc).

Assessment of ALO AIMs
6-OHDA-lesioned rats assigned to the severely dyskinetic group had ALO AIMs scores > 3 times higher than the scores of rats that were classified as mildly dyskinetic (median ± semi-interquartile range ALO AIMs scores: mildly dyskinetic: 32.0 ± 7.0; severely dyskinetic: 102.3 ± 24.6; U = 2, P < 0.05, Mann-Whitney U test).Of note, whereas some rats classified into the mildly dyskinetic group exhibited mild orolingual AIMs, none of them harboured axial and limbs subtype AIMs and the cumulative AIMs scores of these animals were significantly lower than the corresponding scores in 6-OHDA-lesioned rats that were part of the severely dyskinetic group.
[ 3 H]-LY-341,495 binding levels in the remaining brain areas of the ipsilateral hemisphere of both 6-OHDA-lesioned, L-DOPA-treated mildly dyskinetic and severely dyskinetic rats were similar to these obtained in the other groups.Lastly, in comparison to the sham-lesioned group, vehicle-treated 6-OHDA-lesioned rats did not exhibit significant changes in [ 3 H]-LY-341,495 binding in any brain area of the hemisphere ipsilateral to the lesion.

Discussion
In this study, we assessed the distribution of mGlu 2/3 receptors in 6-OHDA-lesioned rats administered with L-DOPA.Notably, we found that mGlu 2/3 receptor binding is elevated in the ipsilateral primary motor cortex of 6-OHDA-lesioned animals displaying severe L-DOPA-induced dyskinesia when compared to animals that were chronically L-DOPAtreated but developed only mild dyskinesia.We also demonstrated that there is a regionally selective down-regulation of [ 3 H]-LY-341,495 binding in 6-OHDA-lesioned L-DOPA-treated mildly dyskinetic rats.This down-regulation manifested bilaterally in the EPN, GP and primary motor cortex.In the contralateral GP, a smaller degree of downregulation was also exhibited by rats with severe L-DOPA-induced dyskinesia.In contrast, the striatum, SN, STN and VA/VL thalamus did not exhibit significant alterations in [ 3 H]-LY-341,495 binding, although a positive correlation between ALO AIMs scores and [ 3 H]-LY-341,495  binding was observed in the ipsilateral striatum.ALO AIMs scores were also positively correlated with [ 3 H]-LY-341,495 specific binding in the ipsilateral EPN, as well as in both hemispheres of the primary motor cortex.These results indicate that L-DOPA administration may be reducing mGlu 2/3 receptor levels in specific brain regions of 6-OHDAlesioned rats, suggesting alterations in mGlu 2/3 receptor expression may be part of an endogenous compensatory mechanism against dyskinesia.
The ligand LY-341,495 was demonstrated to be a potent orthosteric antagonist of group II mGlu receptors (Kingston et al., 1998;Ornstein et al., 1998), with similar affinity for mGlu 2 and mGlu 3 receptors (Wright et al., 2001).Due to the current lack of orthosteric antagonists capable of differentiating between the two receptor types and because of the lack of commercially-available more selective mGlu 2 PAMs or mGlu 2 NAMs, characterising the individual roles of mGlu 2 and mGlu 3 in conferring anti-dyskinetic activity remains a difficult task, and we cannot ascertain whether the results we discovered here represent differences in mGlu 2 receptors, mGlu 3 receptors or combination thereof.Similarly, we cannot exclude that, in brain areas where no changes were detected, a variation in mGlu 2 or mGlu 3 receptor levels might have gone undetected, as it might have been offset by an opposite variation of the other receptor density.Lastly, mGlu 2 and mGlu 3 receptors may form hetero-dimers in the rodent brain (Lee et al., 2020); our results do not allow us to make any speculations regarding the state of these dimers in the 6-OHDA-lesioned rat.For these reasons, the following paragraphs refer to variations of mGlu 2/3 receptors without attempting to ascribe any change to a given receptor.
E. Kim et al.
higher levels of radioligand binding observed in cortical areas, with lower levels in the SN and VA/VL thalamus, are consistent with the previous study (Wright et al., 2001).This distribution of specific mGlu 2/3 binding seen in our results is also in broad agreement with previous autoradiographic binding studies in the rat brain that utilised a different radioligand, the mGlu 2/3 orthosteric agonist [ 3 H]-LY-354,740 (Richards et al., 2005;Schaffhauser et al., 1998).The weaker binding signal that we observed may reflect our use of binding buffer containing potassium chloride, as chloride ions have been shown to enhance the binding of the cold ligand, thereby minimising the impact of non-specific binding on total binding (Mena et al., 1982), as opposed to potassium bromide used elsewhere (Wright et al., 2001).Sex differences in the distribution and expression of mGlu receptors in rodents have been noted, although mGlu 2/3 receptors seem to be quite similarly expressed in rats from either sex (Fabian et al., 2023), which may underlie, at least partly, the discrepancy in binding signal intensity between our female adult Sprague-Dawley rats vs the male adult Sprague-Dawley rats as used by Wright et al. (Wright et al., 2001), though differences pertaining to mGlu 2/3 expression specifically have not yet been thoroughly explored.Indeed, although behavioural studies from our lab thus far do not suggest a difference between male and female animals in response to the effect of mGlu 2/3 receptor activation on dyskinesia (Frouni et al., 2019a;Frouni et al., 2021;Kang et al., 2023b;Kwan et al., 2021;Sid-Otmane et al., 2020), a caveat of the current study is that we did not include male rats, and we acknowledge that our results therefore cannot be generalised to male animals.Furthermore, as our study focusses on investigating the development of L-DOPA-induced dyskinesia and the expression of mGlu 2/3 receptors in the 6-OHDA-lesioned parkinsonian rat, our study lacks a shamlesioned, L-DOPA-treated group to examine the effects L-DOPA administration may have on mGlu 2/3 receptors in the non-parkinsonian animal.We acknowledge this limitation and recognise that we cannot conclusively infer if the changes in mGlu 2/3 receptor expression are a direct result of interaction with L-DOPA, or from the pathophysiology underlining the development of dyskinesia and ALO AIMs.We have tried to speculate on how mGlu 2/3 receptor levels may be altered in the 6-OHDA-lesioned rat upon chronic L-DOPA exposure, especially in regard to the results we have obtained in our study, but these limitations must be considered while reading our interpretations below.
In our current study, significant reductions in [ 3 H]-LY-341,495 binding were seen in specific brain regions in mildly dyskinetic rats, with the most noticeable down-regulation occurring in the primary motor cortex.In this brain region, [ 3 H]-LY-341,495 binding in mildly dyskinetic animals was reduced to nearly half of that of sham-lesioned and vehicle-treated 6-OHDA-lesioned animals.Given the results of our study, it appears that chronic administration of L-DOPA may be downregulating mGlu 2/3 receptors in the primary motor cortex of 6-OHDAlesioned rats.A reduction in mGlu 2/3 receptors on its own does not seem to be sufficient to elicit L-DOPA-induced dyskinesia, as mildly dyskinetic animals remained without axial and limbs dyskinesia despite the relatively lower levels of mGlu 2/3 expression.Given that overexcessive glutamatergic activity within the motor loop of the cortex - basal gangliathalamuscortex circuit is regarded as a tenet of L-DOPA-induced dyskinesia (Frouni and Huot, 2022;McGregor and Nelson, 2019), it is possible that this initial reduction of mGlu 2/3 receptors caused by L-DOPA administration may be permissive for the development of dyskinesia in the 6-OHDA-lesioned rat.However, contrary to our expectations, [ 3 H]-LY-341,495 binding in the primary motor cortex was higher in severely dyskinetic animals when compared to mildly dyskinetic animals, with a significant elevation observed in the ipsilateral hemisphere.Furthermore, increasing ALO AIMs severity was significantly correlated with higher [ 3 H]-LY-341,495 binding in the primary motor cortex, with a stronger positive correlation seen in the ipsilateral hemisphere.These results suggest the alternate possibility that in the 6-OHDA-lesioned rat, upon development of L-DOPA-induced dyskinesia, there may be an endogenous up-regulation of mGlu 2/3 receptors in the primary motor cortex, in an attempt to compensate for the overactive glutamatergic activity present in the dyskinetic state.This endogenous compensatory mechanism is likely insufficient to alleviate dyskinesia on its own, and may be enhanced by the administration of mGlu 2 PAMs or mGlu 2/3 orthosteric agonists, with a resulting anti-dyskinetic effect (Frouni et al., 2021;Hamadjida et al., 2020;Sid-Otmane et al., 2020).In rats that develop mild dyskinesia, this compensatory mechanism may not be evoked, which may explain the lower levels of [ 3 H]-LY-341,495 binding seen in this group of rats.More generally, our results support previous studies implicating the primary motor cortex in L-DOPA-induced dyskinesia, along with the characteristic cortical overexcitation observed in dyskinesia (DeLong and Wichmann, 2007;McGregor and Nelson, 2019), metabolic neuroimaging studies have also revealed excessive activity of the motor cortex in PD patients with dyskinesia (Rascol et al., 1998).Our results also indicate that the primary motor cortex may be a key brain region involved in L-DOPA-induced dyskinesia, suggesting that activation of mGlu 2/3 receptors in the primary motor cortex may be contributing to the anti-dyskinetic efficacy of mGlu 2 PAMs and mGlu 2/3 orthosteric agonists.
In addition to the primary motor cortex, [ 3 H]-LY-341,495 binding was also diminished in the EPN and GP of mildly dyskinetic rats.According to the classic model of the basal ganglia (McGregor and Nelson, 2019), diminished mGlu 2/3 receptor levels in the EPN and GP would result in opposite effects on the thalamus and cortical activation.In the ipsilateral hemisphere, while the GP did not show show a significant correlation between ALO AIMs severity and [ 3 H]-LY-341,495 binding, a positive correlation was observed in the EPN.It is therefore possible, although speculative, that mGlu 2/3 receptors are part of a compensatory strategy that normalises thalamic and cortical activation so that no dyskinesia occurs.Ultimately however, this compensation mechanism might be overcome, and dyskinesia becomes manifest.Further studies are required to more clearly elucidate the role that mGlu 2/3 receptors in these two brain regions may play in the mediation or expression of L-DOPA-induced dyskinesia in PD.
Prior to the current study, only two autoradiographic binding studies had investigated mGlu 2/3 receptor levels in relation to dyskinesia in PD; in one, a reduction in mGlu 2/3 binding was also observed in the GPi and GPe of L-DOPA-treated MPTP-lesioned macaques in which dyskinesia development was prevented through pharmacological strategies (Samadi et al., 2008), but in the other, there were no differences in mGlu 2/3 binding in the GPi and GPe when human PD patients with dyskinesia were compared with PD patients without dyskinesia (Samadi et al., 2009).Factors that may account for the discrepancy between the human study and our rat study include medication taken by PD patients (several individuals were on dopamine agonists, amantadine, anticholinergics, in addition to L-DOPA, whereas the rats in our study were only administered L-DOPA), as well as the long post-mortem delays in the human study (14 hours), compared to brain collection immediately after euthanasia in the current rat study.
Autoradiographic receptor binding presents a valuable quantitative method of investigating the localisation of receptors.However, autoradiographic binding studies are unable to reveal dynamic information about a receptor, such as changes in receptor functional state, downstream signalling, subcellular vs membranal distribution of a receptor, etc., which may represent additional ways whereby mGlu 2/3 receptors may adapt to chronic administration of L-DOPA and an eventual dyskinetic phenotype.The discovery of positron emission tomography (PET) ligands selective for mGlu 2/3 receptors is an active, exciting field of research (Kang et al., 2023a;Perkins et al., 2023;Yamasaki et al., 2020;Yuan et al., 2022a;Yuan et al., 2022b;Zhang et al., 2021), and the identification of a ligand suitable for imaging mGlu 2/3 receptors might enable to shed new light, in vivo, on mGlu 2/3 receptor expression in dyskinesia.
In summary, our study suggests that, in the 6-OHDA-lesioned rat, chronic administration of L-DOPA alters mGlu 2/3 receptor expression in the EPN, GP and primary motor cortex, and that mGlu 2/3 receptor regulation may be part of an endogenous, however limited, compensatory response to attenuate the dyskinetic phenotype.These findings shed light on a possible mechanism by which pharmacological modulation of mGlu 2/3 receptors may be alleviating dyskinesia in PD and provide insight into brain regions worth further investigation in the study of L-DOPA-induced dyskinesia.

Fig. 1 .
Fig.1.Schematic representation of the experimental design and division of animals.Timeline of in vivo experiments performed in 6-OHDA-lesioned rats.Rats injected with 6-OHDA were evaluated for the severity of their hemi-parkinsonsim, after which they were administered L-DOPA and the severity of their abnormal involuntary movements was quantified (A).Rats were divided into four different groups; sham-lesioned, vehicle-treated rats were assigned to Group A; 6-OHDAlesioned, vehicle-treated rats were assigned to Group B; 6-OHDA-lesioned, L-DOPA-treated rats with ALO AIMs score < 50 were assigned to Group C; and 6-OHDA-lesioned, L-DOPA-treated rats with ALO AIMs scores ≥ 50 were assigned to Group D (B).6-OHDA, 6-hydroxydopamine; AIMs, abnormal involuntary movements; ALO, axial, limbs and orolingual; L-DOPA, L-3,4-dihydroxyphenylalanine.

Fig. 3 .
Fig. 3. Autoradiograms illustrating [ 3 H]-LY-341,495 binding across the rat brain regions of interest.All autoradiograms are from sham-lesioned animals.The top row represents the total binding, while the bottom row depicts the non-specific binding.Brain regions of interest are delineated on the side ipsilateral to the lesion.Brain regions of interest include the striatum (A, F), STN (B, G), VA/VL nuclei of the thalamus (C, H), SN (D, I), primary motor cortex (E red, J), GP (E yellow, J), and EPN (E purple, J).Scale bar: 400 μm.

Table 1
[ 3 H]-LY-341,495 binding levels in the rat brain hemisphere ipsilateral to lesion.

Table 2
[ 3 H]-LY-341,495 binding levels in the rat brain hemisphere contralateral to lesion.