Cortical Serotonergic and Catecholaminergic Denervation in MPTP-Treated Parkinsonian Monkeys

Abstract

Decreased cortical serotonergic and catecholaminergic innervation of the frontal cortex has been reported at early stages of Parkinson’s disease (PD). However, the limited availability of animal models that exhibit these pathological features has hampered our understanding of the functional significance of these changes during the course of the disease. In the present study, we assessed longitudinal changes in cortical serotonin and catecholamine innervation in motor-symptomatic and asymptomatic monkeys chronically treated with low doses of 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP). Densitometry and unbiased stereological techniques were used to quantify changes in serotonin and tyrosine hydroxylase (TH) immunoreactivity in frontal cortices of 3 control monkeys and 3 groups of MPTP-treated monkeys (motor-asymptomatic [N = 2], mild parkinsonian [N = 3], and moderate parkinsonian [N = 3]). Our findings revealed a significant decrease (P < 0.001) in serotonin innervation of motor (Areas 4 and 6), dorsolateral prefrontal (Areas 9 and 46), and limbic (Areas 24 and 25) cortical areas in motor-asymptomatic MPTP-treated monkeys. Both groups of symptomatic MPTP-treated animals displayed further serotonin denervation in these cortical regions (P < 0.0001). A significant loss of serotonin-positive dorsal raphe neurons was found in the moderate parkinsonian group. On the other hand, the intensity of cortical TH immunostaining was not significantly affected in motor asymptomatic MPTP-treated monkeys, but underwent a significant reduction in the moderate symptomatic group (P < 0.05). Our results indicate that chronic intoxication with MPTP induces early pathology in the corticopetal serotonergic system, which may contribute to early non-motor symptoms in PD.

dopamine, noradrenaline, prefrontal cortex, primary motor cortex, serotonin

Introduction

Parkinson’s disease (PD) is a progressive neurodegenerative disorder clinically characterized by motor disturbances such as resting tremor, slowness of movement (bradykinesia), rigidity, and postural instability. These motor symptoms are mainly related to the progressive degeneration of the nigrostriatal dopaminergic pathway. It is well recognized that PD symptoms extend beyond motor deficits and include cognitive, psychiatric, and autonomic dysfunctions which, on their own, can be a major source of decreased quality of life for PD patients (Arie et al. 2017; Corallo et al. 2017; Huang et al. 2019). Cognitive impairments in these patients, particularly at the early stages of the disease, primarily affect attention, planning, and executive functions, whereas visuospatial and other cognitive functions are mostly unimpaired (Owen et al. 1997; Marinus et al. 2003; Mollion et al. 2003; O'Callaghan and Lewis 2017; Weintraub et al. 2018; Baiano et al. 2020; Fang et al. 2020). Attention and executive function deficits in PD patients are characterized by impaired performance in attentional set-shifting and sustained-attention tasks, increased distractibility and impaired cognitive flexibility (Downes et al. 1989; Sharpe 1990; Lange et al. 1992; Owen et al. 1992; Sharpe 1992). Patients with more advanced disease may exhibit a broader range of cognitive deficits which may be related to dysfunction of medial temporal lobe structures (Owen et al. 1997; Halliday et al. 2014; O'Callaghan and Lewis 2017). These symptoms display variable responses to dopaminergic medication (Chaudhuri et al. 2006; Burn 2010; Prediger et al. 2012; O'Callaghan and Lewis 2017).

Although the pathobiological substrate(s) of these changes remain to be elucidated, there is evidence from postmortem and imaging studies of PD patient brains that monoaminergic cortical denervation may contribute to the cognitive and psychiatric signs commonly associated with PD (Scatton et al. 1983; Kim et al. 2003; Azmitia and Nixon 2008; Giza et al. 2012; Pavese et al. 2012; O'Callaghan and Lewis 2017; Wen et al. 2017; Weintraub et al. 2018; Fang et al. 2020). Recent optogenetic and chemogenetic studies suggested that relatively subtle changes of dorsal raphe (DR) 5-HT neurons may have a significant impact on executive functions related to attention and cognitive flexibility (Clarke et al. 2004; Ohmura et al. 2014; Fonseca et al. 2015; Matias et al. 2017; Lottem et al. 2018; Miyazaki et al. 2018), 2 cognitive domains that are particularly affected in PD patients (Sawada et al. 2012; Estanga et al. 2014; Rustamov et al. 2014; Martinez-Horta and Kulisevsky 2019). Further animal studies have demonstrated that dopamine and norepinephrine in the frontal cortex play an essential role in modulating cognitive processes, especially working memory (Murphy et al. 1996; Watanabe et al. 1997; Bian et al. 2020) and extradimensional shifting, a behavior in which animals must reorient their attentional reserves to novel stimuli to obtain food reward, respectively (McGaughy et al. 2008; Newman et al. 2008).

Imaging studies have led to variable results on the state of the serotonergic and catecholaminergic cortical innervation in early PD patients. (Laihinen et al. 1995; Brooks and Piccini 2006; Guttman et al. 2007; Albin et al. 2008; Politis et al. 2010; Strecker et al. 2011; Politis 2014; Brockmann et al. 2017; Johar et al. 2017; Pagano et al. 2017). The stage and condition of PD patients and the sensitivity of positron emission tomography (PET) radio ligands used in these studies may account for this variability. However, postmortem studies showed a significant loss of cortical dopamine and serotonin in advanced PD patients (Guttman et al. 2007; Buddhala et al. 2015).

Work achieved in our laboratory and others demonstrated that long-term chronic administration of 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) in nonhuman primates leads to PD-like neuronal degeneration that extends beyond the dopaminergic nigrostriatal system (Schneider 1990; Taylor et al. 1990; Pifl et al. 1991b; Roeltgen and Schneider 1994; Pessiglione et al. 2004; Fornai et al. 2005; Fox and Brotchie 2010; Karachi et al. 2010; Masilamoni et al. 2011b; Villalba et al. 2014; Masilamoni et al. 2016; Kanazawa et al. 2017; Masilamoni and Smith 2018). Furthermore, behavioral studies have shown that chronically MPTP-treated monkeys display early cognitive impairments prior to the development of motor symptoms and significant degeneration of the nigrostriatal dopaminergic system (Decamp and Schneider 2004; Decamp et al. 2004; Schneider 2006). In the present study, we used this animal model to assess the extent of cortical monoaminergic denervation in motor-asymptomatic and motor-symptomatic (mild and moderate parkinsonian) MPTP-treated monkeys.

Materials and Methods

Animals

Ten adult female and 1 male (total 11) rhesus monkeys (Macaca mulatta, 4.5–8.5 kg) from the Yerkes National Primate Research Center colony were used in this study (Table 1). All procedures were approved by Emory’s Animal Care and Use Committee in accordance with guidelines from the National Institutes of Health. The animals were housed in a temperature-controlled room and exposed to a 12-h light/dark cycle. They were fed twice daily with monkey chow supplemented with fruits or vegetables. The animals had free access to water.

Table 1

Subject demographics and clinical data

Monkey	Age, years	Gender	MPTP dosage (mg/kg)	Cumulative MPTP (mg)	Days between last MPTP dose and sacrifice	MDS	Anti-parkinsonian drug (mg/kg)	Clinical status
A	12	F	NA	NA	NA	0/27	NA	Naïve
B	11	F	NA	NA	NA	0/27	NA	Naïve
C	5	F	NA	NA	NA	0/27	NA	Naïve
D	17	F	0.05–0.2	1.6	21	0/27	NA	Asymp
E	7	M	0.05–0.2	1.4	21	0/27	NA	Asymp
F	11	F	0.3	3.0	14	7/27	NA	Mild symp
G	10	F	0.2–0.4	3.8	31	7/27	NA	Mild symp
H	10	F	0.2–0.4	3.6	140	8/27	NA	Mild symp
I	8	F	0.2–0.5	5.4	39	17/27	NA	Moderate symp
J	8	F	0.2–0.5	5.4	39	20/27	NA	Moderate symp
K	11	F	0.2–0.7	2.7	770	17/27	LDOPA: 200 Mirapex: 50	Moderate symp

Monkeys A to C: Control/naive (No MPTP); monkeys D and E: intravenous MPTP injections once every two weeks; monkeys F–K: intramuscular injections of MPTP once a week. MDS ranges from 0 to 27: 0–4 = no impairment, 5–10 = mild impairment; 11–20 = moderate impairment; and 21–27 = severe impairment. Data are mean of 3 or more behavioral assessments to determine stability of the model. F = female; M = male.

MPTP Treatment

Following the collection of baseline measures, monkeys D and E received intravenous MPTP injections once every 2 weeks (0.05–0.2 mg/kg, Sigma-Aldrich) until they displayed cognitive impairment (Tompkins et al. 2011). On the other hand, monkeys F–K received intramuscular MPTP injections once a week (0.2–0.8 mg/kg, Sigma-Aldrich) until they displayed mild or moderate stable parkinsonian motor symptoms (Table 1; Masilamoni et al. 2010; Masilamoni, Bogenpohl et al. 2011; Lin et al. 2015). The monkeys were divided into 4 groups. Group 1 comprised 3 untreated control monkeys (control group). Group 2 consisted of 2 motor-asymptomatic MPTP-treated monkeys without any significant nigrostriatal dopaminergic degeneration (Table 1). Group 3 comprised 3 monkeys that were treated with low doses of MPTP until they displayed mild parkinsonian motor signs (mild symptomatic group), including 1 monkey that was sacrificed 140 days (monkey H) after the last MPTP injection (Table 1). Group 4 consisted of 3 monkeys that progressively developed moderate parkinsonian motor signs (symptomatic group). One monkey in this group was sacrificed 770 days (monkey K) after the last MPTP injection and received 10 doses of 20 mg/kg L-dihydroxyphenylalanine (DOPA) and 50 mg Mirapex (Table 1).

To assess the state of the serotonergic and catecholaminergic cortical afferent systems in these animals, the following measurements were collected from various motor and non-motor cortical areas: 1) Densitometry analyses of serotonin- and TH-immunoreactive neuropil, 2) Stereological counts of 5HT-immunoreactive varicosities, and 3) Stereological counts of 5-HT and TH-immunoreactive cell bodies in the DR and ventral midbrain.

Behavioral Observations

Changes in the severity of parkinsonian motor features were documented using a parkinsonian motor disability score (MDS) routinely used in our laboratory and others to assess the state of parkinsonism in MPTP-treated monkeys (Masilamoni et al. 2010; Masilamoni et al. 2011b; Potts et al. 2015; Masilamoni and Smith 2018). In brief, the animals were brought in an observation cage, of which one of the side walls was made of Plexiglas for easy visibility of the monkey. After a 15-min habituation period, the animals were videotaped for an additional 15 min per session. Their behavior was later monitored and evaluated from videos using a modified Parkinson’s disease rating scale. The scale used in this study evaluated 9 criteria: gross motor activity, balance, posture, arm bradykinesia, arm hypokinesia, leg bradykinesia, leg hypokinesia, arm tremor, and leg tremor. Each criterion received a score between 0 and 3 (0 = normal, 1 = mild, 2 = moderate, and 3 = severe), for a maximum MDS of 27 points. The total number of points was used as the clinical score to compare the severity of parkinsonian motor symptoms across animals in the different experimental groups. The behavioral scores for each animal (average from 3 observations) used in this study are provided in Table 1.

Termination of the Experiments

At the end of the experiments, the monkeys were deeply anesthetized with an overdose of pentobarbital (100 mg/kg, intravenous), and perfused transcardially with cold oxygenated Ringer’s solution, followed by 2 l of fixative containing 4% paraformaldehyde and 0.1% glutaraldehyde in phosphate buffer (0.1 M, pH 7.4). After perfusion, the brains were removed from the skull and cut into 10-mm-thick blocks in the frontal plane. The blocks were further cut into 50-μm-thick sections with a vibratome and used for postmortem immunostaining and cell counting procedures.

Immunostaining

In order to determine the extent of serotonin and catecholaminergic cortical denervation in MPTP-treated monkeys, serial sections at the level of the dorsolateral prefrontal cortex (DLP; Brodmann’s Areas 9 and 46), limbic cortices (Li; Brodmann’s Areas 24 and 25), and sensory motor cortical samples (SM; Brodmann’s Areas 4 and 6) from control and MPTP-treated monkeys were immunostained with specific antibodies against 5HT or TH (Table 2). To avoid inter-individual variability in the intensity of immunostaining due to slight changes in experimental conditions, brain sections used for densitometry measurements of TH or 5HT terminal immunostaining in the various cortices were incubated at the same time using the same antibody solutions and reagents. To relate changes in cortical innervation to the extent of neuronal loss in the potential sources of inputs to these regions, sections at the level of the ventral midbrain were immunostained for TH, whereas sections at the level of the DR were immunostained for 5HT. At the midbrain level, additional sections were immunostained for calbindin-D28K to differentiate calbindin-positive cells in the dorsal tier of the substantia nigra pars compacta (SNCd) and ventral tegmental area (VTA) from the calbindin-negative ventral tier of the substantia nigra pars compacta (SNCv) neurons (Gerfen et al. 1987; Damier et al. 1999; Masilamoni et al. 2010; Masilamoni et al. 2011b; Lin et al. 2015; see Table 2 for details on sources, specificity tests, RRID and dilutions of antibodies). Additional sections including the striatum were immunostained for TH and 5HT using the same procedure.

Table 2

Primary antibodies used in this study

Antibody	Immunogen	Manufacturer data	Dilution
Tyrosine hydroxylase	Tyrosine hydroxylase purified from PC12 cells	Millipore (MAB 318), Mouse monoclonal	1:1000
Calbindin-D-28K	Bovine kidney calbindin-D	Sigma (C-9848) Mouse monoclonal	1: 4000
Serotonin	Serotonin conjugated to BSA	Sigma (MAB 352), Rat monoclonal	1:500

The immunostaining protocols used in this study were similar to those described in our previous studies (Masilamoni et al. 2010; Masilamoni et al. 2011a; Hadipour-Niktarash et al. 2012; Bogenpohl et al. 2013; Galvan et al. 2014; Mathai et al. 2015; Devergnas et al. 2016; Lottem et al. 2018). In brief, sections were treated at room temperature (RT) with 1% sodium borohydride for 20 min followed by a preincubation for 1 h in a solution containing 1% normal horse serum (NHS) or normal goat serum (NGS), 0.3% Triton-X-100, and 1% bovine serum albumin (BSA) in PBS. Sections were then incubated for 24 h at RT in a solution containing the subsequent primary antibodies in 1% NHS or NGS, 0.3% Triton-X-100, and 1% BSA in PBS. On the following day, sections were thoroughly rinsed in PBS and then incubated in a PBS solution containing either (secondary) biotinylated goat anti-rat IgGs or horse anti-mouse IgGs (1:200; Vector) combined with 1% NHS or NGS, 0.3% Triton-X-100, and 1% BSA for 90 min at RT. Sections were exposed to an avidin–biotin–peroxidase complex (ABC; 1:100; Vector) for 90 min followed by rinses in PBS and Tris buffer (50 mM; pH 7.6). Sections were then incubated within a solution containing 0.025% 3,3′-diamino-benzidine tetrahydrochloride (DAB; Sigma), 10-mM imidazole, and 0.005% hydrogen peroxide in Tris buffer for 10 min at RT, rinsed with PBS, placed onto gelatin-coated slides, and cover slipped with Permount. The slides were digitized with an Aperio ScanScope CS system (Aperio Technologies).

Digital Image Analysis

Using Image Scope viewer software (Aperio), the digital images of the stained tissue slides were examined, and 10X-magnification images covering areas of the dorsolateral prefrontal cortex, limbic cortices and sensorimotor cortices were obtained. Adjacent Nissl-stained tissue sections were used to delineate cortical lamina in the 5HT- and TH-stained tissue. Separate optical density (OD) measurements of immunostaining and background were obtained from Layer I, Layers II–III, and Layers IV–VI in all cortical regions analyzed for 5HT (see Supplementary Fig. 1), whereas TH OD measurements were integrated across all cortical lamina (see Supplementary Fig. 2). Four to 8 images were captured from adjacent anteroposterior tissue sections per area analyzed in each animal, depending on the size of the region of interest (Paxinos et al. 1999). The images were then imported into ImageJ (v1.41, National Institutes of Health) for additional processing. For OD measurements, the images were inverted to a dark field such that dark immunoreactive elements on a light shaded background were converted to bright immunoreactive elements on a dark background. Comparable areas of analysis were highlighted within the cortical regions of interest, and the integrated OD within the selected area was measured. To control for differences in background staining, 3 OD measurements within the highlighted area without immunoreactive elements were determined and averaged. The background OD value was then subtracted from the initial OD value within each cortical region of interest.

For striatal TH and 5-HT OD measurements, images were captured at 0.4X magnification and imported into ImageJ for additional processing. The images were converted into 8-bit grayscale format and calibrated using a step tablet, gray scale values were converted to OD units using the Rodbard function, and the mean OD for each area of interest was recorded. To control for differences in background staining, the OD measurement in the internal capsule was subtracted from that obtained in striatal measurements. Mean values were calculated, using one out of every 12 sections. With this measuring scheme, 5–7 sections were used per region of interest in each animal.

Stereological Analyses

Estimation of the Total Number of TH-Positive Neurons in Ventral Midbrain and 5HT-Immunoreactive Neurons in DR

The unbiased stereological estimation of the total number of dopamine neurons in the ventral SNc, dorsal SNc and VTA, or serotonergic neurons in the DR was achieved using the optical fractionator principle (StereoInvestigator, MicroBrightField, Inc.), a stereological approach that combines the optical dissector with a fractionator sampling scheme. This sampling technique is not affected by tissue volume changes and does not require reference volume determinations. The random systematic sampling of counting areas was done using the Leica DMR microscope. TH- and 5HT-positive cells were counted using a 100X oil-immersion objective in one out of every twelfth section through the rostrocaudal extent of the ventral midbrain and DR nucleus. To perform unbiased stereology, counting frames (65 × 65 μm) were randomly placed that is based on the sampling grid size (250 × 250 μm), by the stereology software within the chosen region of interest. The software also controlled the position of the x–y stage of the microscope, so that the entire brain region could be scanned by successively meandering between counting frames.

To count midbrain TH-positive neurons, we first manually delineated the borders of the ventral SNc, dorsal SNc, and VTA based on the presence or absence of calbindin-positive neurons (Masilamoni et al. 2010; Masilamoni, Bogenpohl, et al. 2011). Then, the borders of the different ventral midbrain regions were manually delineated on TH-immunolabeled slides adjacent to those immunostained for calbindin. On average, 12 sections/animal were analyzed and ~300 cells were counted in controls. CE values were ≤0.045 for control and MPTP-treated monkeys, which meet the criteria of acceptable CE values as previously established elsewhere (Gundersen and Osterby 1981).

The serotonergic cell group of DR was delineated by the expression of 5HT- immunoreactive neurons confined within the following anatomical landmarks: The ventral tip of the cerebral aqueduct as the dorsal limit, the ventral border of the medial longitudinal fasciculus as the ventral limit, and the midline as the medial limit. 5HT-positive cell counts were made from 6 serial sections collected at regular intervals through the full rostro-caudal extent of DR (4.13 to −0.15 mm from the interaural line; Paxinos et al. 1999). Adjacent sections were processed for TH immunoreactivity to count TH-positive cells in DR nucleus. Findings from these additional experiments will allow us to determine whether serotonergic and dopaminergic DR neurons are differentially affected by chronic MPTP administration. A minimum of 200 cells immunostained for 5HT and 75 cells labeled for TH were counted in each series of sections, resulting in a coefficient of error (Gundersen, m = 1) that was ≤0.08. d MPTP-treated animals.

Estimation of 5HT-Positive Varicosities in Frontal Cortical Regions

As a complement to the densitometry measurement data of 5HT immunostaining, we assessed the cortical 5HT innervation at a finer level of resolution through quantification of 5HT-positive varicosities using stereo investigator. To address this issue across various cortical regions and Layers (1, 2/3, and 4/5), sections were prepared as discussed above (see Supplementary Fig. 2). The 5HT-positive varicosities were defined as individual round or oval-shaped bulbous structures spaced irregularly along labeled axons that varied in size from 0.5 to 3.0 μm in diameter (Figs 3 and 4). Stereological analysis was made from 1 of 48 serial sections through the anteroposterior extent of the various cortical areas. The number of sections analyzed to estimate the total number of labeled varicosities in the various cortical regions was as follows: Area 4:4; Area 6:4; Area 24:7; Area 25:3; Area 9:4; and Area 46:7. This design resulted in a coefficient of error of 0.027–0.067 (Gundersen, m = 1; Gundersen and Osterby 1981). The density of labeled varicosities was calculated by dividing the total number of varicosities counted in each region of interest (ROI) by the estimated volume of the ROI. We used the Cavalieri’s principle to estimate the volume of the cortical areas and layers examined (Gundersen and Osterby 1981; Schmitz and Hof 2005).

Statistical Analysis

Data were statistically analyzed using Graphpad Prism software (version 8.2). One-factor analysis of variance (ANOVAs) for repeated measures followed by the Tukey post hoc test was used to compare TH- and 5HT-positive neuronal loss, density measurements, and behavioral tests between control and 3 different MPTP treatments. Significance was taken at P < 0.05^*, P < 0.001^**, and P < 0.0001^***. All results are expressed as mean ± standard deviation (SD).

Data Availability

All data presented in this manuscript will be made available upon reasonable request.

Results

Motor Impairment and Nigrostriatal Dopamine Loss in the MPTP-Treated Monkeys

Eight of the 11 monkeys used in this study received chronic injections of low doses of MPTP. Subject demographics and preclinical data for each of these animals are shown in Table 1. Based on the appearance and severity of motor symptoms induced by the MPTP treatment, the animals were divided into 3 groups: 1) Asymptomatic (N = 2; MDS 0/27), 2) Mild symptomatic (N = 3; MDS 5–10/27), and 3) Moderate symptomatic (N = 3; MDS 11–20/27). Three more monkeys were used as control (no MPTP treatment; Fig. 2C).

Figure 1 compares the level of TH immunostaining in the substantia nigra and striatum between the control and the 3 groups of MPTP-treated monkeys (Fig. 1A–H). As depicted, both the mild and moderate symptomatic animals displayed a variable loss of TH-immunoreactive innervation of the striatum (see Supplementary Fig. 3F) accompanied with a significant loss of TH-positive neurons in the ventral tier of the SNc (SNCv) compared with controls (Figs 1A–H and 2A; P < 0.001). In contrast, asymptomatic monkeys displayed patchy reduction of TH immunostaining mainly confined to the postcommissural putamen, and no significant loss of TH-immunoreactive neurons in the SNCv (Figs 1B, F and 2A). In the SNCd and VTA, a significant loss of TH-immunoreactive neurons was only seen in the symptomatic animals (P < 0.05; Fig. 2A).

Figure 1

TH- and 5HT immunoreactivity in a control and the 3 groups of MPTP-treated monkeys (asymptomatic, mild symptomatic, and moderate symptomatic). (A–D) TH-immunostained cell bodies in the SNC and VTA. (E–H) TH-immunostained nigrostriatal terminals in the postcommissural putamen. (I–L) 5HT-immunoreactive neurons in the dorsal (DR) and median (MnR) raphe. Scale bars in A, E, I are valid for micrographs displayed in each row.

Figure 2

(A, B) Quantitative assessment of changes in the number of TH+ (A) and 5HT+ (B) neurons in the 3 groups of MPTP-treated monkeys compared with controls. A significant decrease of TH+ neurons was found in the ventral tier of SNC and VTA in the mild symptomatic and moderate symptomatic animals, whereas the difference did not reach significance in the asymptomatic group in any of the ventral midbrain regions. There was a significant decrease in the total number of 5HT-positive neurons in the DR of the moderate symptomatic monkeys. Although both the asymptomatic and mild symptomatic monkeys also displayed a loss of 5HT+ neurons, the significance of these differences could not be assessed because of the low number of animals available in each group. (C) depicts the average clinical rating score of the 4 groups of monkeys used in this study. Only the mild and moderate symptomatic monkeys displayed significant parkinsonian motor signs. In each graph, data are represented as mean ± standard error of the mean (SEM) and each symbol indicates the value for individual monkeys.

Changes in the Number of 5HT-Positive Neurons in the Raphe Nucleus of MPTP-Treated Monkeys

To assess potential degeneration of serotonergic neurons in MPTP-treated monkeys, unbiased stereological count of 5HT-positive cells in the DR was performed. As shown in Figures 1 and 2, MPTP treatment resulted in a significant reduction of the total number of 5HT-positive neurons in the raphe nucleus of the moderate symptomatic monkeys (P < 0.05). Due to mechanical tissue damage at the level of the DR in some monkeys, animal’s tissue availability for cell counts in the DR was limited. Thus, because of the low number of animals, we were not able to perform statistical analysis of the extent of neuronal loss in the asymptomatic and mild symptomatic group. However, the 3 monkeys examined in these groups displayed a loss of 5HT-positive neurons in the DR that ranged from 9% to 27% (Fig. 2B).

To determine if neuronal loss in DR was specific for 5-HT-positive cells, we counted TH-positive neurons in the same region (Stratford and Wirtshafter 1990), and found no significant difference between control and MPTP-treated monkeys (see Supplementary Fig. 4A–E), suggesting a preferential MPTP-induced neurotoxic effect towards serotonergic DR neurons.

Reduced Serotonergic Innervation of the Dorsolateral Prefrontal, Limbic, and Motor Cortices in Symptomatic and Asymptomatic MPTP-Treated Monkeys

As described in previous studies, the morphology of 5HT-positive axon- and terminal-like structures in the dorsolateral prefrontal (Areas 9 and 46), limbic (Areas 24 and 25), and motor (Areas 4 and 6) cortices was heterogeneous (Beaudet and Descarries 1976; Smiley and Goldman-Rakic 1996; Way et al. 2007; Raghanti et al. 2008) comprising non-varicose large diameter axon-like profiles or fine axonal processes with large or small varicosities (Fig. 3A–E). These labeled elements were present in all cortical areas examined, but their regional and laminar distributions differed. In control monkeys, densitometry measurements revealed significant 5HT innervation of all cortical areas, with slightly larger values in limbic (Fig. 4C and D) than motor (Areas 4 and 6) and dorsolateral prefrontal (Areas 9 and 46) regions (Fig. 4A, B, E, F). Although not significant, Layer 1 harbored a stronger intensity of 5HT immunoreactivity than deeper layers in all cortical regions (Fig. 4).

Figure 3

Morphology of typical fine and beaded serotonergic axons in control monkey. In most areas of neocortex, beaded axons predominate in superficial Layers 1 and 2 (A, B, D, F), and fine axons predominate in deep layers (C). Scale bars in A, D are valid for micrographs displayed in each row.

Figure 4

Average (±SEM) OD measurements of 5HT immunostaining in motor (Areas 4 and 6) and prefrontal (Areas 24, 25, 9, and 46) cortices in control animals and 3 groups of MPTP-treated monkeys (motor asymptomatic, mild symptomatic, moderate symptomatic). In each graph, the X axis indicates the cortical layers from where the measurements were taken. The different symbols in each bar show data collected from individual monkeys. Note the significant loss of 5HT immunostaining across all layers in Areas 24 and 25 of both asymptomatic and symptomatic monkeys. (Stats data, P values)

Densitometry measurements revealed a significant reduction of 5HT immunoreactivity in all cortical regions of the asymptomatic and symptomatic MPTP-treated monkeys compared with control (P < 0.05; Figs 4 and 5). When analyzed at a layer-specific level, the MPTP treatment resulted in a homogenous significant reduction of 5HT-positive nerve fibers in Layers 1, 2/3, and 4/5 of the asymptomatic monkeys. These observations were confirmed and extended by our stereological quantitative assessment of the number of 5HT-positive varicosities (P < 0.05–0.0001; Fig. 6). Both approaches revealed that the mild and symptomatic MPTP-treated monkeys underwent the most severe 5HT denervation in comparison to control (P < 0.0001) and asymptomatic monkeys (P < 0.05; Figs 4 and 6). The degree of terminal loss in the cortical regions appears to be more pronounced than the magnitude of DR 5HT-positive neuronal loss, suggesting that cortical 5HT-positive nerve terminals are the primary target of the degenerative process and that neuronal death in MPTP-treated monkeys may result from a “dying back” process. It is noteworthy that one of MPTP-treated moderate parkinsonian monkeys that was sacrificed 770 days after the last MPTP injection and received some L-DOPA treatment (Monkey K in Table 1) exhibited a larger density of 5HT-immunoreactive varicosities that the 2 other animals in this group in prefrontal, but not in motor, cortical regions (Compare Fig. 6C–F with A–B).

Figure 5

5HT-immunoreactive fibers and varicosities in frontal cortical regions (Areas 6, 46 and 25) of control (A, F, K) and MPTP-treated motor asymptomatic (B, G, L), mild symptomatic (C, H, M), and moderate symptomatic (D, I, N) monkeys. The last column (S770-E, J, O) depicts the cortical labeling in one of the symptomatic monkeys that was sacrificed 770 days after the last MPTP injection. In this monkey, the density of 5HT+ axonal profiles was larger than in the 2 other symptomatic monkeys that were sacrificed 39 days after the last MPTP injection (Table 1). The scale bar in A is valid for all micrographs.

Figure 6

Average (±SEM) density of 5HT+ varicosities in motor (Areas 4 and 6) and prefrontal (Areas 24, 25, 9, and 46) cortices in control animals and 3 groups of MPTP-treated monkeys (motor asymptomatic, mild symptomatic, and moderate symptomatic). In each graph, the X axis indicates the cortical layers from where the measurements were taken. The different symbols in each bar show data collected from individual monkeys. Note the significant loss of 5HT immunostaining across all layers in Areas 24 and 25 of both asymptomatic and symptomatic monkeys. (Stats data, P values). Note: Difference between densitometry measurements and varicosities counts in area 25-symptomatic monkeys

To determine if the pattern of cortical and striatal serotonergic denervation follows the same trajectory, 5HT densitometry analysis was performed on various striatal regions in these monkeys. In contrast to the cerebral cortex, the striatum of MPTP-treated asymptomatic monkeys did not exhibit any significant loss of 5HT innervation (see Supplementary Fig. 3B and E). On the contrary, the intensity of 5-HT immunoreactivity in the 2 animals of this group was either the same or slightly increased compared with controls (see Supplementary Fig. 3B and E). A significant decrease in OD of 5HT immunoreactivity was only found in the caudate nucleus and the motor territory of the putamen in the mild and moderate parkinsonian monkeys (see Supplementary Fig. 3A–E).

Reduced Catecholaminergic Cortical Innervation in Symptomatic, but not Asymptomatic, Parkinsonian Monkeys

To determine if MPTP-treated monkeys displayed any change in cortical catecholaminergic innervation, tissue from the dorsolateral prefrontal (Areas 9 and 46), limbic (Areas 24 and 25), and motor (Areas 4 and 6) cortices was immunostained for TH (marker of dopaminergic and noradrenergic axon terminals) and subjected to OD measurements using the Image J software. Overall, the pattern of distribution of TH-labeled fibers and varicosities in the different cortical regions of control monkeys was consistent with that reported in previous studies (Akil and Lewis 1993; Raghanti et al. 2008; Martin and Spuhler 2013). In brief, TH-immunoreactive processes were distributed throughout the entire dorsoventral extent of each cortical area with a preponderance of labeling in Layers 1–3 compared with deep cortical layers (Fig. 7 and see Supplementary Fig. 2). Overall, the intensity of TH labeling in control monkeys was comparable between 5 cortical regions examined (Areas 4, 6, 9, 24, and 25; Fig 8). The only exception was Area 46 that displayed lower OD measurements than other regions (Fig. 8). In the 3 groups of MPTP-treated monkeys, only the “moderate symptomatic” animals displayed a significant loss of TH innervation in Areas 4 and 9 and 24 (Fig. 8). In contrast, none of the cortical regions exhibited a significant reduction of TH immunolabeling in either the asymptomatic or mild symptomatic monkeys, except for Area 9 in mild symptomatic animals (Figs 7 and 8). Despite this lack of evidence for catecholaminergic denervation, it is noteworthy that TH-immunoreactive varicosities in all cortical regions of the asymptomatic and mild symptomatic monkeys displayed abnormal enlargement, a pathological feature commonly associated with early signs of degeneration (Fig. 9).

Figure 7

TH-immunoreactive fibers and varicosities in frontal cortical regions (Areas 4, 9, and 25) of control (A, E, I) and MPTP-treated motor asymptomatic (B, F, J), mild symptomatic (C, G, K) and moderate symptomatic (D, H, L) monkeys. The scale bar in A is valid for all micrographs. Not the significant decrease in immunostaining in the moderate symptomatic monkeys compared with controls and the 2 other groups of MPTP-treated monkeys.

Figure 8

Average (±SEM) OD measurements of TH immunostaining in motor (Areas 4 and 6) and prefrontal (Areas 24, 25, 9, and 46) cortices in control animals and 3 groups of MPTP-treated monkeys (motor asymptomatic, mild symptomatic, moderate symptomatic). The different symbols in each bar show data collected from individual monkeys. Note the decrease in TH labeling reached statistical significance only in Areas 4, 24, and 9 of the moderate symptomatic monkeys. (Stats data, P values).

Figure 9

Representative low and high magnification of TH-immunoreactive fibers and varicosities in frontal cortical regions (Areas 46 and 9) of MPTP-treated motor asymptomatic monkey. The scale bar in A, B, C are valid for micrographs displayed in each row. Note the presence of enlarged and swollen axonal varicosities in the cortical regions (C, F).

Discussion

Our findings demonstrate a significant decrease in serotonergic innervation of motor, pre-motor, prefrontal, and limbic cortical regions in adult rhesus monkeys chronically treated with low doses of MPTP. Although these changes were particularly profound in animals with mild or moderate parkinsonian motor signs, they were also significant in motor asymptomatic monkeys. In contrast, striatal serotonergic innervation was not affected in motor asymptomatic animals. These major changes in cortical serotonergic innervation across the 3 MPTP-treated animal groups were accompanied with a slight decrease of 5HT-positive neurons in the DR, which was most pronounced in moderate symptomatic parkinsonian monkeys. Changes in catecholaminergic cortical innervation were less prominent such that only areas 4, 9, and 24 of moderate symptomatic monkeys displayed significant decreases in TH immunoreactivity. However, pathological enlargement of TH-positive varicosities indicative of degeneration was found in all cortical regions of asymptomatic and mild symptomatic monkeys. Overall, our data suggest that cortical serotonergic denervation might be an early pathology that could contribute to the emergence of non-motor cognitive and psychiatric deficits in the parkinsonian state.

Early and Progressive Cortical Serotonergic Denervation in Chronically MPTP-Treated Monkeys

Our findings show a significant reduction in the intensity of 5HT immunoreactivity and in the number of 5HT-positive varicosities across motor, limbic, and dorsolateral prefrontal cortical regions in the 3 groups of MPTP-treated monkeys. Literature reports about cortical serotonergic pathology in MPTP-treated monkeys have been variable (Pifl et al. 1990; Mihatsch et al. 1991; Gaspar et al. 1993; Perez-Otano et al. 1994; Mounayar et al. 2007; Boulet et al. 2008; Zeng et al. 2010; Beaudoin-Gobert et al. 2015; Engeln et al. 2015; Kanazawa et al. 2017), ranging from a significant loss of serotonin or serotonin transporter binding in various cortical regions of symptomatic and asymptomatic monkeys (Perez-Otano et al. 1991; Pifl et al. 1991b; Kanazawa et al. 2017) to no change in cortical serotonergic innervation (Beaudoin-Gobert et al. 2015; Engeln et al. 2015; Ballanger et al. 2016). Variations in MPTP dosage schedules, survival times after intoxication and methods of serotonin innervation measurements (biochemistry, immunostaining, and PET imaging) may account, at least in part, for these differences. In the present study, monkeys were treated with a chronic low-dose MPTP administration regimen spread over many months. The choice of this MPTP intoxication approach was based on a significant amount of literature showing that chronic administration of MPTP results in neuronal loss that extends beyond the nigrostriatal dopaminergic system to include brainstem noradrenergic and serotonergic neurons (Schneider and Kovelowski 1990; Schingnitz, et al. 1991; Herrero et al. 1993; Hornykiewicz 1998; Masilamoni, Weinkle, et al. 2011; Porras et al. 2012; Halliday et al. 2014; Beaudoin-Gobert et al. 2015; Masilamoni and Smith 2018). In regard to the human literature, PET imaging studies and postmortem data have reported profound and widespread neocortical decrease of 5HT transporter ligand binding and serotonin levels in advanced PD patients (Ogawa et al. 1992; Kish 2003; Guttman et al. 2007; Albin et al. 2008; Azmitia and Nixon 2008; Politis et al. 2010; Buddhala et al. 2015). However, data from early PD patients are scarce and controversial; imaging studies based on small cohorts of patients reported either a reduction (Guttman et al. 2007; Albin et al. 2008; Politis 2014) or no alteration (Strecker et al. 2011) in cortical serotonin transporter (SERT) binding in non-depressed PD patients, whereas another small study suggested an elevated SERT binding in the prefrontal and dorsolateral prefrontal cortices of depressed PD patients (Boileau et al. 2008). More recently, a study demonstrated no change in cortical SERT binding during the pre-motor phase of parkinsonism in LRRK2 mutation carriers (Wile et al. 2017). Based on these observations, it is difficult to make a firm conclusion about the status of the cortical serotonergic innervation in early PD patients (Pagano et al. 2017). Larger cohorts of depressed and non-depresssed PD patients must be studied. The comparison between our postmortem immunohistochemical data from MPTP-treated monkeys and the human in vivo SERT binding imaging results must also be done with caution because various factors, other than changes in the number of serotonin terminals, can influence the SERT binding potential (Politis et al. 2010; Porras et al. 2012; Politis 2014).

Regional Pattern of Cortical Serotonergic Terminals Loss in Parkinsonian Monkeys

The decreases in serotonin innervation of frontal and prefrontal cortices reported in our study were based on both densitometry measurements of 5HT immunoreactivity and unbiased stereological counts of labeled varicosities across cortical layers. Both approaches revealed a homogeneous reduction in 5HT-positive profiles throughout the full dorsoventral extent of the cortical regions examined. Because cortical layers are organized into distinct cytoarchitecture, connectivity and function (Sawaguchi et al. 1989, 1990; Goldman-Rakic 1995; Kritzer and Goldman-Rakic 1995; Arnsten et al. 2012), a layer-specific alteration in serotonergic innervation would have been indicative of dysregulation of the neuromodulatory influences of serotonin on specific cortical microcircuits. However, the homogeneous decrease in serotonin innervation described in our study suggests a more global disruption of cortical functions. To our knowledge, our data provide the first layer-specific quantitative analysis of changes in the density of 5HT-positive profiles between control and parkinsonian monkeys. Evidence for layer-specific 5HT cortical denervation has been reported in the prefrontal cortex (PFC) of aged A53T α-synuclein-expressing mice model of PD (Wihan et al. 2019). In contrast, a marked decrease of 5HT-positive axons in all cortical layers has been shown in MPTP-treated mice 16 weeks after the MPTP administration, whereas only superficial layers were affected after 3 weeks post-MPTP survival (Nayyar et al. 2009). It is noteworthy that these findings and those presented in our study do not provide unequivocal evidence for a loss of cortical serotonin terminals in MPTP-treated animals. The reported changes in 5HT immunoreactivity could result from downregulation of serotonin expression in individual terminals. Another limitation of our study is the low number of animals/group, which reduces the statistical power of some of our comparative analyses. Despite these shortcomings, our results, combined with previous biochemical data showing decrease in cortical serotonin levels in motor asymptomatic MPTP-treated monkeys (Pifl et al. 1991b), suggest that early dysfunction of the corticopetal serotonergic system could contribute to the development of pre-motor cognitive impairments in PD (Schneider and Kovelowski 1990; Tompkins et al. 2011).

There are 2 main types of serotonergic axons in the cerebral cortex, those that are thin with small varicosities, which mainly originate from the DR and large highly varicose axons that originate predominantly from the median raphe (Kosofsky and Molliver 1987). Although we did not attempt at dividing these 2 types of varicosities in our study, the fact that both the densitometry and stereological quantitative methods indicated significant losses of 5HT innervation across all cortical layers suggest that both types of serotonin axons were affected in MPTP-treated monkeys. In MPTP-treated mice, large 5HT-containing beaded axons appeared to be preferentially affected over small fibers (Nayyar et al. 2009). Whether this differential pattern of deafferentation indicates a genuine species difference between primates and rodents, or merely relies on different doses and regimen of MPTP intoxication used in either species, remains to be established.

Loss of Raphe Serotonergic Neurons in the Parkinsonian State

Despite significant evidence for changes in serotonin innervation of frontal cortices in animal models of parkinsonism and PD patients, less is known about the extent of neuronal loss in the raphe nuclei (D'Amato et al. 1987; Halliday et al. 1990; Paulus and Jellinger 1991; Doder et al. 2003; Halliday et al. 2014; Jellinger 2017; Pagano et al. 2017). Our findings demonstrate a significant loss of 5HT-positive neurons in the DR of moderate parkinsonian monkeys, a result consistent with that of a recent study using a chronic MPTP treatment monkey model of PD and tryptophan hydroxylase 2 as a marker of raphe serotonin neurons (Beaudoin-Gobert et al. 2015). Although the cellular mechanisms of DR serotonergic neuronal loss remain unclear, given that MPP+ can gain access to 5HT neurons via the serotonin transporter (Kanazawa et al. 2017), the retrograde “dying back” hypothesis similar to what has been suggested for degeneration of the nigrostriatal dopaminergic projection may be considered (Herkenham et al. 1991; Kanazawa et al. 2017). However, because our findings do not provide direct evidence for 5HT terminal degeneration (vs. downregulation of 5HT expression), future studies are needed to confirm this hypothesis. Albeit less pronounced, a decrease in 5HT-immunoreactive neurons was also found in the DR of the 2 other groups of MPTP-treated monkeys used in our study, but the statistical significance of these observations could not be assessed due to the low number of animals available for these experiments. Nonetheless, these data suggest that the decrease in cortical serotonergic innervation seen in MPTP-treated monkeys is partly accounted for by DR neuronal loss. In contrast to serotonin neurons, no significant loss of TH-positive cells was found in the DR of MPTP-treated monkeys, highlighting the specific neurotoxic effects of MPTP towards the serotonergic cell group.

In humans, the few semi-quantitative observations available reported either a loss (D'Amato et al. 1987; Paulus and Jellinger 1991; Watanabe et al. 1997; Doder et al. 2003; Brooks and Piccini 2006; Jellinger 2017; Pagano et al. 2017) or no apparent change (Halliday et al. 1990) in the number of DR neurons in PD patients. Halliday et al. (1990) suggested that the median raphe was more affected than DR. Because none of the neuropathological data were gathered through rigorous unbiased cell count methods, a direct comparison between these data sets and ours must be made with caution. Similarly, caution must be exercised in comparing our findings with those obtained through the PET imaging studies because of the various factors that can affect the binding properties of serotonin transporter or serotonin receptor ligands during the course of the disease in PD patients (Doder et al. 2003; Brooks and Piccini 2006; Pagano et al. 2017). Finally, alpha-synuclein pathology may contribute to raphe neuronal loss in PD patients, but not in chronically MPTP-treated parkinsonian monkeys (Halliday et al. 1990; Jellinger 2012; Masilamoni and Smith 2018).

Cortical Catecholaminergic Denervation in Chronically MPTP-Treated Monkeys

Our findings demonstrate a significant decrease in the intensity of TH immunostaining in the cortical areas 4, 9, and 24 of chronically MPTP-treated parkinsonian monkeys. In contrast to the widespread early serotonergic depletion of cortical innervation prior to the development of parkinsonian motor signs, the reduced TH labeling was restricted to fewer cortical regions, and was found only in parkinsonian animals. These observations are in part consistent with our cell count data, which showed a significant loss of VTA TH-positive neurons, the main source of the meso-cortical dopaminergic system, in moderate symptomatic monkeys, but not in asymptomatic animals. However, even if motor-asymptomatic monkeys did not display a significant decrease in cortical TH immunoreactivity, or a reduction in the number of midbrain TH-positive neurons, the evidence of pathologically enlarged TH-positive varicosities in motor and prefrontal cortices suggest early signs of cortical catecholaminergic dysregulation. Similar morphological changes in TH-positive varicosities have been reported in the striatum of PD patients (Huot et al. 2007; Zeng et al. 2010). Although the exact mechanisms underlying the differential sensitivity of serotonin versus TH-positive cortical terminals in MPTP-treated asymptomatic monkeys are unclear, it is noteworthy that the bulk of cortical dopamine innervation originates from VTA neurons which, in contrast to SNC neurons, express a much lower level of dopamine transporter (DAT), thereby limiting their sensitivity to MPTP toxicity. Furthermore, the fact that these neurons express calbindin D28K may also account for their relative sparing in response to MPTP. In regard to the early loss of serotonin terminals after MPTP intoxication in monkeys, our findings are in line with those of (Pifl et al. 1991b) who also reported a more profound reduction of 5HT than DA levels in various cortices of asymptomatic MPTP-treated monkeys. It is also important to note that a significant decrease in the density of cortical 5HT-positive terminals, without degeneration of DR neurons, has been reported in animal models and humans intoxicated with 3,4-methylenedioxymetamphetamine (Molliver et al. 1990; Beaudoin-Gobert et al. 2015) indicating the sensitivity of cortical serotonin terminals to neurotoxins.

Our previous findings showed that monkeys rendered parkinsonian under the same chronic low-dose regimen of MPTP as used in the present study exhibit significant loss of noradrenergic neurons in the locus coeruleus (LC; Masilamoni, Bogenpohl, et al. 2011; Masilamoni et al. 2016; Masilamoni and Smith 2018). Thus, because TH is expressed in all catecholaminergic neurons, the reduced intensity of cortical TH immunoreactivity reported in the present study could also be due to the death of LC noradrenergic neurons. However, various data suggest that TH is predominantly expressed in dopaminergic, over noradrenergic, terminals in the primate cerebral cortex (Lewis et al. 1987; Berger et al. 1988) (Schmidt and Bhatnagar 1979). In human postmortem material, only 10–50% of dopamine-beta-hydroxylase (DβH)-positive terminals, a marker of noradrenergic neurons, express TH immunoreactivity (Gaspar et al. 1989). Based on these observations, it is likely that changes in TH immunostaining intensity reported in our study are mainly accounted for by degeneration of the meso-cortical dopaminergic system. Knowing that the noradrenergic system undergoes early degeneration in human PD, our findings must be interpreted with caution because they may not reflect the full extent of cortical catecholaminergic denervation associated with early and late stages of PD.

Previous studies of cortical catecholaminergic innervation in MPTP-treated parkinsonian monkeys led to variable results. On one hand, some authors reported over 70% loss of TH immunostaining in the sensorimotor and associative cortices of vervet monkeys (Jan et al. 2003), whereas biochemical data indicated either no significant change (Engeln et al. 2015) or moderate to profound loss of dopamine and noradrenaline (Elsworth et al. 1990; Schneider and Kovelowski 1990; Pifl et al. 1991a) in motor and prefrontal cortices of symptomatic and asymptomatic macaque monkeys. Despite some limitations in reconciling these variable data due to differences in MPTP administration regimen, state of parkinsonism, monkey species, and catecholamine measurement approaches, most studies concur that motor and prefrontal cortices undergo noradrenaline and dopamine denervation in parkinsonian monkeys. These observations are consistent with PET imaging data showing decreased binding for dopaminergic and noradrenergic markers in M1 and prefrontal cortex of PD patients compared with age-matched control subjects (Brooks and Piccini 2006; Moriguchi et al. 2017; Sommerauer et al. 2018; Andersen et al. 2020). Similarly, postmortem immunohistochemical and biochemical studies demonstrated a significant decrease in cortical noradrenaline content and a modest cortical dopaminergic denervation in PD patients (Scatton et al. 1983; Gaspar et al. 1991; Buddhala et al. 2015). Given the importance of prefrontal cortical dopamine and noradrenaline in regulating cognition, mood and other complex limbic-related behaviors these results suggest that dysregulation of either transmitter system may contribute to a wide range of non-motor deficits (executive dysfunction, depression, anxiety, sleep disorders, psychosis, and other neuropsychiatric symptoms) commonly seen in PD patients (Rodriguez-Oroz et al. 2009; Brichta et al. 2013; O'Callaghan and Lewis 2017; Ryan et al. 2019). Although the extent of cortical dopaminergic denervation described in our study and in PD patients is not as profound as in the striatum, it is important to consider that even subtle deviations from cortical dopamine levels may lead to cognitive impairments (Leblois et al. 2006; Guthrie et al. 2013), which is consistent with the inverse U-shape regulation of cortical functions by dopamine (Goldman-Rakic 1996; Cools et al. 2001).

Concluding Remarks

Our findings demonstrate that chronically MPTP-treated monkeys exhibit widespread changes in serotonergic and catecholaminergic innervation of motor and prefrontal cortices reminiscent of those seen in PD patients. In concert with other studies, these data suggest that this animal model may be useful to study the potential consequences of cortical monoaminergic denervation associated with motor and non-motor symptoms of PD (Masilamoni and Smith 2018). Most importantly, evidence that monkeys treated with chronic low-doses of MPTP exhibit changes in attention, cognitive flexibility and executive memory prior to the development of motor symptoms (Schneider and Kovelowski 1990; Schneider and Roeltgen 1993; Tompkins et al. 2011; Vezoli et al. 2011; Barth et al. 2020) highlight the potential use of this model towards a deeper understanding of the underlying substrates of early cognitive impairments in PD.

Funding

NIH (grant P50-NS098685; Udall Center grant); the NIH/ORIP Yerkes National Primate Center (base grant P51-OD011132).

Notes

The authors thank Jean-Francois Pare and Susan Jenkins for their excellent technical assistance. Conflict of interest: The authors declare that they have no conflict of interest.