Cerebellar oligodendrocytic α-synuclein pathology and dentate nucleus neuronal hypertrophy in Parkinson’s disease

Abstract

Idiopathic Parkinson’s disease (iPD) research has historically prioritized basal ganglia dysfunction and neuronal Lewy body pathology, often overlooking cerebellar and non-neuronal cells contributions. This study investigates cerebellar neuropathology in iPD, specifically focusing on neuronal and non-neuronal cells of the dentate nucleus (DN) and its adjacent white matter (DN-WM). We systematically assessed autopsy-confirmed iPD brains vs. age-matched controls using immunohistochemistry and stereology-based morphometry methods. The study evaluated α-synuclein pathology, other common age-related co-occurring pathologies (β-amyloid, Tau, neuroinflammation), myelin integrity (Luxol-Fast-Blue), and neuronal volumetric changes in the DN of the cerebellum. In iPD brains, we identified intracytoplasmic α-synuclein inclusions within oligodendrocytes across the peri-DN-WM. These inclusions were distinct from classic neuronal Lewy bodies (dense core with clearer halo), which were notably absent in the DN neurons. Concomitantly, we observed myelin rarefaction in the DN and the around WM area. Moreover, stereology-based morphometric analysis revealed a significant hypertrophy of DN neuronal soma (p < 0.05), nuclei (p < 0.0001), and nucleoli (p < 0.0001) in iPD vs. C subjects, suggesting a heightened metabolic state or compensatory stress response. These findings are novel and suggest that oligodendroglial α-synucleinopathy may be a contributing factor to cerebellar abnormalities and degeneration in iPD. Due to the recent identification of direct pathways between the DN and basal ganglia in primates and the DN primary dopaminergic modulatory function, these outcomes are of a special pathophysiological relevance. In fact, we hypothesize that this oligo-synucleinopathy WM deterioration compromises the glial-neuronal metabolic coupling and physiological transmission along the cerebello-basal ganglia-cortical loops. Moreover, these alterations could contribute to explain the broader spectrum motor and non-motor symptomatology observed in specific iPD subtypes.

Introduction

The dentate nucleus (DN) - the largest of the deep cerebellar nuclei - has historically been viewed through the narrow lens of a motor control and coordination regulator only^1,2. However, this traditional perspective has become obsolete^3–6. Indeed, a wealth of anatomical, physiological, and neuroimaging data now positions the cerebellum (CRB), and the DN in particular, as a highly interconnected hub engaged in both motor and non-motor functions^7–9. Moreover, in terms of histoarchitecture and connectivity, it is now well established that the DN receives afferent input from multiple brain regions and sends efferent output via the cerebellar peduncles, projecting primarily to the thalamus, which, crucially, relays signals broadly to cortical and subcortical structures, including the basal ganglia (BG)^10,11.

A similar paradigm shift has occurred in the understanding of idiopathic Parkinson’s disease (iPD). In fact, while iPD has been classically defined by nigrostriatal dopaminergic loss and extrapyramidal motor symptoms¹², it is now recognized as a heterogeneous disorder involving widespread neurocircuitry alterations^13,14. This expanded view accounts for the diverse non-motor manifestations of the disease, including neuropsychiatric, cognitive, and autonomic dysfunctions^15–17. Moreover, advances in understanding the neuroanatomical circuits underlying iPD have further revealed that the disease pathology extends well beyond the nigrostriatal system¹⁸. Both human and animal model studies have demonstrated that multiple interconnected brain regions, including the CRB, contribute to the complex pathophysiology of iPD¹⁹. Nonetheless, the CRB, and its deep nuclei, received relatively little attention in the neuropathological context of iPD, despite its potential relevance to motor and cognitive aspects of the disease²⁰. The CRB’s underappreciated role is especially surprising considering that α-synuclein (α-syn) pathology - the hallmark of iPD mainly manifested as Lewy bodies (LBs) and Lewy neurites (LNs) - has been previously reported, although sporadically, in the CRB.^21,22. These sporadic observations after the recent studies showing direct reciprocal disynaptic connections between the CRB and BG, in particular between the DN and BG in non-human primates^23,24, and the direct cerebellar modulation of nigral dopaminergic activity, in rodents at least²⁵, urge a global re-evaluation of cerebellar involvement, and the DN involvement in particular, in the primary pathophysiological processes of iPD, including α-synucleinopathy affecting both neuronal and glial cells.

In this study, we aimed to address this gap by systematically assessing α-syn-positive lesions in the DN and its adjacent white matter (DN-WM) in autopsy-confirmed iPD brains. Using immunohistochemistry (IHC) and stereology-based morphometry methods, we evaluated not only the possible presence of glial, especially oligodendroglial, α-syn pathology, but also the neuronal integrity (i.e., morphometric changes and non-α-syn pathological inclusions) to better clarify the possible relevant DN’s contribution to the complex pathophysiological and clinical spectrum of iPD.

Materials and methods

A total of six DN from left cerebellar hemispheres and five left hemi-SN were sampled from autopsy-confirmed idiopathic PD (iPD) brains. A number of five DN from the left cerebellar hemispheres and five hemi-SN samples were collected from age-matched control subjects (C) - subjects without history of neurodegenerative disorders and autopsy-confirmed negative for Alzheimer’s Disease, PD, and other major neurodegenerative pathologies. The sample sizes for hemi-SN and hemi-DN were limited by tissue availability within the repositories. Additional brain regions (frontal cortex, hippocampus, putamen, and pons) were also assessed for the presence of α-syn pathology by performing IHC and further exclusion of other possible types of synucleinopathies such as the one typically detected in Multiple System Atrophy (MSA) brains. All samples were obtained from the University of Maryland Brain and Tissue Bank (UMBTB) and The Mount Sinai/JJ Peters VA Medical Center NIH Brain and Tissue Repository (NBTR) through the NIH NeuroBioBank consortium (https://neurobiobank.nih.gov).

All specimens were obtained after next-of-kin or legal representatives signed a written informed consent for their use in research.

Neuropathology procedures

At autopsy, a half-brain from each iPD and C subject was fixed in 10% buffered formalin for at least 2 weeks, then grossly dissected and microscopically examined on a series of coronal sections as per standard neuropathologic assessment at each corresponding brain bank (UMBTB and NBTR). All DN and SN samples were shipped to the Neuropathology Research labs, Biomedical Research Center of New Jersey (BRInj; https://brinj.org) for further assessment. All tissue-blocks were processed through an automated tissue-processor (Tissue-Tek V.I.P. 1000 Vacuum Infiltration Processor, Ames Division, Miles Laboratories, Inc. IN, USA) as per standard histology protocols. The tissue blocks were then embedded in paraffin, oriented with their rostral cross-sectional area facing the bottom of the paraffin mold and serially cut along the rostro-caudal anatomical direction using a semi-automatic microtome (Leica RM2255, Leica Biosystems, Nussloch, Germany). Tissue microsectioning was performed as previously described²⁶. The total tissue thickness received for each formalin-fixed DN and SN block ranged between 4 and 5 mm. Whereas we could not obtain the entire DN or SN region for each subject, we estimated that at least an amount equivalent to 25% of the entire DN and SN extension were indeed received for each iPD and C subject. Consequently, we were able to measure cellular (cell body) and subcellular (nucleus, nucleolus) mean volume for a high number of DN neurons in both iPD and C brains similarly to some previous similar investigations^27,28. However, as related to the stereology principles²⁹ and tissue sample size, we could not perform a reliable counting of DN or SN neurons in iPD or C subjects.

Neurohistological procedures

A single 10 μm-thick tissue section was randomly chosen and stained with hematoxylin and eosin (HE) and microscopically inspected at lower (2.5x objective) and higher (20x objective) magnification to detect possible vascular pathologies, malformations, tissue rarefactions, metastases or any other obvious histological anomaly. Furthermore, representative sections were stained with Luxol Fast Blue-PAS (LFB-PAS) to assess for potential demyelination, particularly in the peri-dentate white matter (peri-DN-WM)³⁰.

Intraneuronal Lesions (H&E)

The search for possible intraneuronal lesions in the DN of iPD and C brains was firstly based on HE stains and focused on eosinophilic (Negri bodies, Bunina bodies, Hirano bodies, and Marinesco bodies) and basophilic (Nissl bodies and Buscaino bodies) lesions. Possible granulovacuolar degeneration phenomena were considered too. The systematic HE-based assessment of DN in iPD vs. C brains included the peri-DNWM that is, all the WM region adjacent to or inside the DN (DN-gray matter [DN-GM]), the most lateral tooth-like shaped group of neurons nestled in the CRB.

Immunohistochemistry Assessment

In addition to a standard gross and microscopic neuropathological assessment for all DN and SN samples, each iPD-DN and C-DN was assessed by immunohistochemistry (IHC) protocols for the detection of: α-syn positive inclusions, hyperphosphorylated-tau neurofibrillary tangles (pTau-NFTs), pTau-neurites (pTau-Ns), extracellular deposits of insoluble 1–42 β-amyloid (βA-diffuse, βA-neuritic, βA-cored plaques). For the assessment of neuroinflammatory histo-markers, IHC protocols were performed using an anti-glial fibrillary acidic protein (GFAP) and anti-ionized calcium binding adaptor molecule 1 (IBA1) antibody for the specific evaluation of astroglia response and microglia reactivity, respectively. Briefly, a set of 5 consecutive sections were deparaffinized, hydrated and treated with endogenous peroxidase blocking with 3% hydrogen peroxide in water, rinsed in buffer (TRIS with 0.05% Triton-X), microwaved with antigen retrieval (sodium citrate pH 6.0) for 10 min, and then cooled down. Protein block was applied for 30 min using 1.5% horse serum. The following specific antibodies were used: mouse anti-β-Amyloid, 17–24 (4G8; Covance, SIG-39220; dilution 1:500); mouse anti-PHF-tau (Thermo Scientific, MN1020; dilution 1:500); mouse anti-α-synuclein (Abcam, Ab27766; dilution 1:500); mouse anti-GFAP (GA5; PA0026, Leica Biosystems, Wetzlar, Germany; dilution 1:500), rabbit polyclonal anti-ionized calcium-binding adapter molecule 1 (IBA-1, Wako 016-20001; FUJIFILM Wako Pure Chemical Corporation, Osaka, Japan; dilution 1:500). The sections were then incubated overnight at 4 °C (sections for 4G8 were pre-treated with 90% formic acid, 5 min), rinsed with buffer, and incubated with biotinylated horse anti-mouse, anti-rabbit secondary antibodies (30 min), Vector Kits (Vector Labs, Inc., CA), then with DAB substrate system (Sigma, D3939) for 5 min. Slides were counterstained with hematoxylin and dehydrated to xylene and coverslipped.

Immunofluorescence staining

To identify α-syn inclusions within oligodendrocytes,  additional sections of the DN and SN regions  -contiguous to those used for IHC -  were used for double immunofluorescence (IF) staining. Briefly, deparaffinized and rehydrated sections were subjected to antigen retrieval by boiling in citrate buffer (pH 6.0), followed by blocking with 5% normal goat serum for 1 h at room temperature (RT). Sections were then incubated overnight at 4 °C with a mixture of primary antibodies: anti-α-synuclein (phosphor-S129) antibody [EP1536Y] (Abcam, ab51253; rabbit monoclonal, 1:300) and OLIG2 monoclonal antibody 1G11 (Thermo Fisher Scientific; mouse monoclonal; 1:100). After three washes in 1xPBS at RT, sections were incubated with Alexa Fluor 488 and Alexa Fluor 546 conjugated secondary antibodies (Thermo Fisher Scientific, 1:500) for 1 h at RT in the dark. Finally, sections were rinsed in PBS and coverslipped with VECTASHIELD Antifade Mounting Medium with DAPI (Vector Laboratories, H-1200-10).

Morphometric methodologies and quantitative procedures

For stereology-based morphometric purposes, sets of separated series of 10 μm-thick sections across the entire available length of each DN from both iPD and C subjects were consecutively chosen with a rate of one out of twenty-four and a starting section randomly chosen among the first five consecutive serially-cut sections. All sections were stained with 1.0% solution of Cresyl violet (CV)³⁰, randomly re-coded by an investigator blinded for both clinical and pathologic diagnoses (HP). All volumetric measurements were performed by employing stereological tools such as Optical Fractionator and Nucleator probe³¹. The volumetric measures were performed by a second investigator (DI) blind for both clinical and pathologic diagnoses and re-coding system. Each CV-stained section was stereoscopically inspected (2.0x magnification) for the entire sectional area to precisely localize DN anatomical borders. The DN anatomical borders are typically well-defined as sharp contrast between the worm-shaped DN vs. WM background. For neuronal volumetric measurements, the following histologic criteria were applied: (a) 10 μm-thick sections; (b) a well-defined nucleolus for each randomly-sampled DN neuron; (c) nucleolus as only point of spatial reference for all volumetric measurements for the same randomly-sampled DN neuron. All stereological measurements were performed using Stereo-Investigator system, Version 10.0 (MBF Bioscience, Williston, VT), equipped with digital camera (AxioCam MRm, and MRc, Zeiss, Germany) and multiple objectives-head (2.5x-100x oil-immersion). Sampling grid area was 500 × 500 μm, counting frame of 40 × 40 μm, disector height of 10 μm with guard zones of 2 ± µm. Cell body, nuclear, and nucleolar volumes of each DN neuron were measured placing six rays automatically centered on the nucleolus, randomly intersecting the cell membrane. DN neurons were measured if their nucleolus intersected or touched the inclusion (green) line and were excluded if their nucleolus intersected or touched the exclusion (red) line²⁷. All measurements were performed using a 63x oil-immersion, NA 1.30, neofluor ∞/0.17 objective. After all measurements were completed, section codes were broken and all single stereologically-measured volumes for each DN neuron from all iPD and C subjects were clustered to obtain the mean cell body, nuclear and nucleolar volume for DN-neurons in iPD and C group.

Statistics

To detect differences between cell body, nuclear, and nucleolar mean volumes of DN neurons in iPD vs. C subjects, preliminary normality and lognormality analyses were performed to establish if data distribution was Gaussian or Logarithmic in nature. All tests used to verify the probability of Gaussian distribution of our entire stereological-based measurements dataset (D’Agostino & Pearson, Anderson-Darling, Shapiro-Wilk, and Kolmogorov-Smirnov test) showed a higher probability for a lognormal vs. Gaussian (normal) distribution. Based on these preliminary analyses, all data were analyzed as nonparametric unpaired data applying both Mann-Whitney (comparing ranks) and Kolmogorov-Smirnov test (comparing cumulative distributions). The level of statistical significance for our analyses was set when the calculated p-value was equal or less than 0.05 (p < 0.05). To avoid pseudoreplication, the mean cell body, nuclear, and nucleolar volumes were calculated for each individual subject. Statistical comparisons were then performed between the iPD group (n = 6) and Control group (n = 5) using these subject-level means.

All data and plots were elaborated using GraphPad Prism version 10.0 for Windows, GraphPad Software (www.graphpad.com).

As for HE, LFB and IHC stains, the assessment was primarily qualitative with a semi-quantitative approach for the rough estimation of the Lewy inclusions, LBs, Lewy Neurites, NFTs, Tau neurites and β-Amyloid, plaques across the targeted regions (DN and SN).

Results

Main demographic, clinical, and pathologic data are summarized in Table 1. The mean age at death between iPD (79.3 ± 3.2 years; 76–85 years range) and C (73.8 ± 11.5 years; 65–94 years range) groups did not differ. Likewise, the mean brain weight (BW; 1218 ± 142.4 g; 1102–1422 g range in iPD, and 1338 ± 221.8 g; 1002–1600 g range in C subjects) as well as the mean post-mortem interval (PMI; 17.8 ± 7.2 h; 8–28 h range in iPD, and 17.8 ± 8.0 h; 8–28 h range in C subjects) between the two groups did not differ. All available clinical and neuropathological information for both iPD and C subjects was reviewed and is summarized in Tables 1 and 2. Moreover, as based on the available clinical and neuropathological information, all iPD cases were retrospectively re-assessed for a clinical diagnosis of iPD, dementia with Lewy bodies (DLB), Parkinson’s disease with dementia (PDD), and Multiple System Atrophy (MSA) following the currently clinical and pathological consensus-based accepted criteria. Specifically, all examined subjects were confirmed to be clinically-diagnosed with iPD based on MDS-PD clinical criteria^32,33. None of the examined cases could have had a clinical diagnosis of DLB, PDD or MSA based on the DLB³⁴, PDD³⁵, and MDS-MSA³⁶ clinical criteria. Furthermore, while the neuropathological assessment of multiple brain regions of the iPD subjects suggested that a diagnosis of DLB could have been possible, none of the iPD subject could have been clinically diagnosed as PDD or MSA since none of them were described as having cognitive deficits in 1 year from the initial diagnosis of iPD or as having symptoms related to the dysfunction of multiple brain systems like in MSA cases. Moreover, all iPD subjects were reported as having a beneficial response to the dopaminergic therapies and so reinforcing the initial diagnosis of iPD.

Table 1.

Main demographic, clinical, and pathological data iPD and age-matched control brain subjects.

Subject#	Clinical diagnosis	Pathological diagnosis	Age at death	Sex	Race	Cause of death	PMI (h)	BW (g)	Age at PD diagnosis	History of dementia (last visit)	Last UPDRS part III	H&Y	Responsive to Levo-Dopa	Movement disorders features	Other relevant medical conditions
#1	iPD	iPD	78	M	W	Aspiration pneumonia	23	1422	NA	No (1 year before death)	36 (1 year before death)	3 (1 year before death)	Yes	Focal dystonia RLE (right low extremity); kinetic left-hand tremor;	orthostatic hypotension; congestive heart failure
#2	iPD	iPD	81	M	W	Complications of the disorder (PD)	28	1313	68	Yes (probable PDD; )	NA	NA	Yes	Bradykinesia Rigidity Freezing RBD	Orthostatic hypotension Falls Depression Prostate cancer
#3	iPD	iPD	79	F	W	Complications of the disorder (PD)	8	1125	59	Yes (probable PDD)	NA	NA	Yes		Post-herpetic trigeminal neuralgia; hypothyroidism; hyperlipidemia;
#4	iPD	iPD	76	M	W	Pneumonia	12	N/A	66	Yes (probable PDD)	NA	NA	Yes	RLS	Coronary artery bypass; Orthostatic hypotension; Falls; Abdominal aortic aneurysm; Peripheral artery disease; Type II diabetes; Melanoma
#5	iPD	iPD	77	F	W	Dysphagia	18	1102	70	Memory deficit (possible PDD)	NA	NA	Yes	Resting tremor left hand Softened voice Reduced facial expression Bradykinesia Micrography	Anxiety
#6	iPD	iPD	85	F	W	Complications of the disorder (PD)	18	1126	63	Mild cognitive impairment (< 1 year from death)	NA	NA	Yes	Rigidity Tremor left hand Gait disturbance	DBS Thyroid dysfunction Osteoporosis Carpal tunnel Diplopia Cervical radiculitis Lumbosacral radiculopathy Cerebral ischemia
#7	No	None	68	M	W	Fall	23	1600	Not applicable	No	Not applicable	Not applicable	Not applicable	Not applicable	None
#8	No	None	65	M	W	Natural	28	1400	Not applicable	No	Not applicable	Not applicable	Not applicable	Not applicable	None
#9	No	None	94	F	W	Cardiopulmonary Arrest	8	1002	Not applicable	No	Not applicable	Not applicable	Not applicable	Not applicable	None
#10	No	None	72	M	W	Sudden death	12	1270	Not applicable	No	Not applicable	Not applicable	Not applicable	Not applicable	None
#11	No	None	70	M	W	Sudden death	18	1420	Not applicable	No	Not applicable	Not applicable	Not applicable	Not applicable	None

The table presents individual demographic and clinical profiles for six subjects with a pathological diagnosis of iPD (Subjects #1–6) and five age-matched controls (Subjects #7–11). Data includes age at death, post-mortem interval (PMI), and brain weight (BW), alongside disease-specific metrics such as age at diagnosis, Levo-Dopa responsiveness, and Hoehn and Yahr (H&Y) staging. Clinical histories note the presence of dementia (PDD), movement disorder features (e.g., bradykinesia, rigidity, tremor), and other relevant comorbidities. iPD, idiopathic Parkinson’s Disease; PMI, Post-Mortem Interval; BW, Brain Weight; UPDRS, Unified Parkinson’s Disease Rating Scale; H&Y, Hoehn and Yahr scale; PDD, Parkinson’s Disease Dementia; DBS, Deep Brain Stimulation; RLS, Restless Leg Syndrome; RLE, Right Lower Extremity.

Table 2.

Substantia Nigra (SN) and Dentate Nucleus (DN) Lewy pathology in iPD and age-matched control brains (HE assessment).

	Lewy pathology (LB, LN) in SN and DN		Neuronal loss		Neuromelanin pigment incontinence	HE-detectable intraneuronal inclusions (LB, LN, other)		Vascular pathology
								Microhemorrhages
Subject#	SN	DN	SN	DN	SN	SN	DN	SN	DN
#1 (iPD)	Yes	Yes	Mild	No	Moderate	Yes	Yes	No	No
#2 (iPD)	Yes	Yes	Severe	No	Moderate	Yes	Yes	No	No
#3 (iPD)	Yes	Yes	Severe	No	Mild	Yes	Yes	No	No
#4 (iPD)	Yes	Yes	Moderate	No	Mild	Yes	Yes	No	No
#5 (iPD)	Yes	Yes	Mild	No	Mild	Yes	Yes	No	No
#6 (iPD)	Yes	Yes	Moderate	No	Mild	Yes	Yes	No	No
#7 (C)	No	No	Neg	Neg	Neg	Neg	Neg	No	No
#8 (C)	No	No	Neg	Neg	Neg	Neg	Neg	No	No
#9 (C)	No	No	Neg	Neg	Neg	Neg	Neg	No	No
#10 (C)	No	No	Neg	Neg	Neg	Neg	Neg	No	No
#11 (C)	No	No	Neg	Neg	Neg	Neg	Neg	No	No

This table details the neuropathological findings obtained via Hematoxylin and Eosin (HE) assessment for the study cohort. Observations for iPD subjects (#1–6) and controls (#7–11) include the presence of Lewy pathology (Lewy bodies and Lewy neurites), the severity of neuronal loss, and the presence of neuromelanin pigment incontinence1.The table also documents the presence of HE-detectable intraneuronal inclusions and vascular pathology, specifically microhemorrhages. SN, Substantia Nigra; DN, Dentate Nucleus; LB, Lewy Body; LN, Lewy Neurite; HE, Hematoxylin and Eosin; Neg, Negative.

Peri-dental and intra-dental neuronal and non-neuronal α-syn-positive inclusions

Intracellular α-syn-positive inclusions were detected mainly within oligodendrocytes across the white matter (WM) around the DN (peri-dental WM region) (Fig. 1). These peri-DNWM α-syn-positive inclusions appeared as rounded intracytoplasmic inclusions occupying most of the cell body visible area (Figs. 1, 2 and 3), which appear to be one characteristic feature of these types of lesions in comparison to the classic Lewy bodies (LBs), for example those observed in the SN that typically show a central core with a halo (see Fig. 4, upper vs. lower panels). α-syn positive neurites (Lewy neurites [LN]s) were also found but they appeared to be much rarer compared to α-syn intracytoplasmic inclusions and were localized either within or around the DN region (Fig. 3). Importantly, no typical LBs were detected in the DN of these iPD brains. Moreover, also the standard HE stains confirmed the presence of eosinophilic inclusions within oligodendrocytes in peri-DN WM region compared to C, although the HE sensitivity appeared to be much lower in comparison to IHC methods. The qualitative assessment of the Luxol Fast Blue-PAS (LFB-PAS) stain suggested a degree of myelin pallor around but mainly within the DN of iPD vs. C subjects (Fig. 5). While these observations provide a useful baseline, future studies should employ quantitative measures to confirm the extent of myelin rarefaction.

Fig. 1.

The figure shows α-synuclein (α-syn) positive intracytoplasmic inclusions within oligodendrocytes of the peri-dentate nucleus white matter (peri-DN-WM) obtained using immunohistochemistry (IHC) methods. Panels (A) and (B) shows oligodendroglial α-syn positive inclusions detected at lower and higher magnification (20x and 40x objectives, respectively) in the peri-DN-WM) area. iPD subjects display α-syn positive inclusions (red arrows) different from the typical intraneuronal Lewy bodies detected in the Substantia Nigra (SN) of the iPD brains.

Fig. 2.

Different size of α-syn positive inclusions in iPD brains. This figure shows smaller α-syn inclusions (red arrows) in comparison to those detected in Fig. 1 in the same peri-DN-WM area (Fig. 1 and Fig.2  show lesions from different iPD subjects). These smaller lesions were detected using identical IHC methods across all sections and subjects. The different sizes of α-syn inclusions across iPD cases could represent an oligodendrocytic accumulations of α-syn inclusions at different stages of the natural progression of the disease. DN-WM: dentate nucleus white matter; α-syn: α-synuclein.

Fig. 3.

α-syn inclusions and neurites in the peri-DN-WM of iPD. The figure shows the co-presence of positive α-syn inclusions and neurites in the peri-DN-WM area. The α-syn inclusions were detected by using IHC methods. DN-WM: dentate nucleus white matter; α-syn: α-synuclein. Images were obtained using bright light microscope using a 40x objective. DN-WM: dentate nucleus white matter; α-syn: α-synuclein; iPD: idiopathic Parkinson’s disease.

Fig. 4.

α-syn positive oligodendrocytic inclusions in the peri-DN-WM of iPD. The figure show (upper three panels) the presence of an α-syn positive inclusion (red) in an oligodendrocyte (green), and a merged image of those two images (yellow-orange) in the peri-DB-WM area. The lower two panels show the presence of typical LB (dense core with a peripheral clearer halo) in SN-dopaminergic neurons of the same iPD subject. The images were obtained using fluorescent light microscope using a 63x oil immersion objective. DN-WM: dentate nucleus white matter; α-syn: α-synuclein; SN: substantia nigra; LB: Lewy body; iPD: idiopathic Parkinson’s disease.

Fig. 5.

Qualitative assessment of myelin integrity in DN area of iPD vs. C subjects. Representative photomicrographs of tissue sections stained with Luxol Fast Blue-PAS (LFB-PAS) to visualize myelin sheaths (blue). (A, B) Sections from iPD subjects show more pallor in comparison to (C, D) Control subjects, especially in the intra-DN area (WM).

To notice, while the presence of oligodendrocytic inclusions was a consistent finding across all six iPD cases and absent in all controls, there was some inter-case variability in the density of the pathology, which is now detailed in Tables 3 and 4. In general, the pathology was predominantly found in the peri-dental WM, with only rare neurites found within the dentate gray matter itself and an absence of classic Lewy bodies in dentate neurons.

Table 3.

Dentate Nucleus (DN),  Substantia Nigra (SN), and other cerebral regions pathologies in iPD and ge-matched control brains (IHC assessment).

	α-syn Lewy Pathology (LB; LN [n/section])		Lewy Pathology (LB, LN) in other brain regions	pTau (NFTs, Tau-N)		Beta-Amyloid (plaques)		GFAP (astrocytic reaction/astrogliosis)		IBA-1 (microglia reactivity)
	SN	DN		SN	DN	SN	DN	SN	DN	SN	DN
1 (iPD)	LB +++ LN +++	LB++ (74/section): spherical lesions of different size mainly in peridental-WM oligodendrocytes LN+ (rare)	CA2-hippocampus (rare LB and LN) Cingulate gyrus (rare LB and LN) Locus coeruleus (mild LB and LN)	NFTs + (rare) TN + (rare)	Neg	Neg	Neg	No	rare reactive astrocytes)	Yes (rare clusters of microglial cells activation)	Yes (rare clusters of isolated microglial cells activations; elongation)
2 (iPD)	LB + LN −	LB++ (60/section): spherical lesions of different size mainly in peridental-WM oligodendrocytes LN+ (rare)	CA2-hippocampus (rare LB and LN) Frontal Cortex (mild LB and LN) Locus coeruleus (mild LB and LN)	Neg	Neg	Neg	Neg	No	rare reactive astrocytes)	Yes (rare clusters of microglial cells activation)	Yes (rare clusters of isolated microglial cells activations; elongation)
3 (iPD)	LB++ LN++	LB++ (76/section): spherical lesions of different size mainly in peridental-WM oligodendrocytes LN++ (moderate)	CA2-hippocampus (rare LB and LN) Frontal Cortex (mild LB and LN) Locus coeruleus (mild LB and LN)	NFTs + (rare) Tau-microglia?	Neg	Neg	Neg	No	rare reactive astrocytes)	Yes (rare clusters of microglial cells activation)	Yes (rare clusters of isolated microglial cells activations; elongation)
4 (iPD)	LB ++ LN +	LB++ (63/section): spherical lesions of different size mainly in peridental-WM oligodendrocytes LN+ (rare)	No	Neg	Neg	Neg	Neg	No (rare reactive astrocytes)	No	Yes (clusters of isolated microglial cells activations; elongation)	Yes (clusters of isolated microglial cells activations; elongation)
5 (iPD)	LB +++ LN +++	LB ++ (23/section): spherical lesions of different size mainly in peridental-WM oligodendrocytes LN-	CA2-hippocampus (rare LB and LN) Locus coeruleus (mild LB and LN)	NFTs + (rare) TN + (rare)	Neg	Neg	Neg	Yes (cluster of astroglial cells activated perilesional vascular; rarely activated across SN	No	Yes (clusters of isolated microglial cells activations; elongation); also perilesional vascular	No
6 (iPD)	N/A	LB ++ (38/section): spherical lesions of different size mainly in peridental-WM oligodendrocytes LN+ (rare)	CA2-hippocampus (rare LB and LN) Frontal Cortex (mild LB and LN) Locus coeruleus (mild LB and LN)	N/A	Neg	N/A	Neg	N/A	No	N/A	Yes (clusters of isolated microglial cells activations; elongation)
7 (C)	Neg	Neg	No	Neg	Neg	Neg	Neg	No	No	No	No
8 (C)	Neg	Neg	No	Neg	Neg	Neg	Neg	No	No	No	No
9 (C)	Neg	Neg	No	Neg	Neg	Neg	Neg	No	No	No	No
10 (C)	Neg	Neg	No	Neg	Neg	Neg	Neg	No	No	No	No
11 (C)	Neg	Neg	No	Neg	Neg	Neg	Neg	No	No	No	No

Table 4.

Lewy pathology and co-occurring pathologies in the Substantia Nigra and Dentate Nucleus of iPD and age-matched Control brains (HE and IHC assessment).

	α-syn Lewy pathology (LB; LN; HE)		pTau (NFTs, Tau-N; IHC)		β-Amyloid (plaques; IHC)		GFAP (astrocytic reaction/astrogliosis; IHC)		IBA-1 (microglia reactivity; IHC)
	SN	DN	SN	DN	SN	DN	SN	DN	SN	DN
1 (iPD)	Pos	Pos	Rare NFTs	Neg in GM; rare Tau lesions in WM	Neg	Neg	Neg	Scattered hypertrophic astrocytes	Normal	Normal
2 (iPD)	Pos	Pos	Neg	Neg	Neg	Neg	Neg	Neg	Normal	Normal
3 (iPD)	Pos	Pos	Rare NFTs	Neg in GM; rare Tau lesions in WM	Neg	Neg	Scattered hypertrophic astrocytes	Neg	Scattered reactive microglia cells	Normal
4 (iPD)	Pos	Pos	Rare NFTs	Neg	Neg	Neg	Neg	Neg	Isolated clusters of microglia activation	Isolated clusters of microglia activation
5 (iPD)	Pos	Pos	Rare NFTs and neurites	Neg	Neg	Neg	Scattered hypertrophic astrocytes	Neg	Normal	Normal
6 (iPD)	Pos	Pos	N/A	Neg	N/A	Neg	N/A	Neg	N/A	Isolated clusters of microglia activation
7 (C)	Neg	Neg	Neg	Neg	Neg	Neg	No	No	No	No
8 (C)	Neg	Neg	Neg	Neg	Neg	Neg	No	No	No	No
9 (C)	Neg	Neg	Neg	Neg	Neg	Neg	No	No	No	No
10 (C)	Neg	Neg	Neg	Neg	Neg	Neg	No	No	No	No
11 (C)	Neg	Neg	Neg	Neg	Neg	Neg	No	No	No	No

Other brain regions and α-syn pathology

Importantly, although IHC for α-syn revealed rare LBs and LNs in neurons of the other brain regions assessed (frontal cortex, hippocampus, putamen, and pons), we could not detect α-syn-positive inclusions within the WM or oligodendroglial cells in these regions similarly to the ones observed in the DN region of the iPD brains examined.

Stereology-based volumetric outcomes of dentate nucleus neurons

Nonparametric statistical tests results showed a difference in the mean cell body, nuclear and nucleolar volume of DN-iPD vs. C group. The Mann-Whitney test showed a p < 0.05 with a sum of ranks of 650,091 (iPD-DN group) and 671,034 (C group), and the Kolmogorov-Smirnov test showed a p < 0.05 with a Kolmogorov D (distance) of 0.007121 for the cell body mean volume. The Mann-Whitney test showed a p < 0.0001 with a sum of ranks of 748,291 (iPD-DN group) and 602,255 (C group), and the Kolmogorov-Smirnov test showed a p < 0.0001 with a Kolmogorov D (distance) of 0.2333 for the nuclear mean volume. The Mann-Whitney test showed a p < 0.00001 and the Kolmogorov-Smirnov test showed a p < 0.0001 with a Kolmogorov D (distance) of 0.1392 for the nucleolar mean volume. Fig.6 shows the mean cell body, nuclear, and nucleolar volume of the DN neurons in the DN-iPD and C group.

Fig. 6.

Neuronal Hypertrophy of DN in iPD. The figure shows the mean cell body, nuclear, and nucleolar volume of neurons in the DN of iPD vs. C subjects. DN: dentate nucleus; iPD: idiopathic Parkinson’s disease; age-matched control subjects.

Discussion

This study reveals oligodendrocytic α-synuclein pathology within the cerebellar dentate nucleus (DN) of idiopathic Parkinson’s disease (iPD) patients associated with neuronal hypertrophy of DN neurons. While previous studies have shown α-synuclein pathology in oligodendroglial cells as well as in the cerebellum in brain tissue coming from humans and animal models^21,22,37,38, no studies so far have shown the presence of oligodendrocytic α-synuclein pathology in the DN of autopsy-confirmed iPD human brains. These novel findings support the expansion of the neuropathological landscape of iPD beyond the traditionally emphasized nigrostriatal dopaminergic system by highlighting the previously underappreciated role of synucleinopathy in the oligodendrocytes of the cerebellum and its biggest deep nucleus, - dentate nucleus (DN) - in the pathogenesis of iPD. For decades, PD research has focused, almost exclusively, on the degeneration of dopaminergic neurons in the substantia nigra and the formation of neuronal Lewy bodies. However, emerging evidence, including our present findings, strongly advocates for a re-evaluation of this basal ganglia-neuron-centric view. As mentioned, while the presence of Lewy pathology in the cerebellum as well as in the cerebellar oligodendrocytes have been sporadically described before, not specific IHC methods using validated antibodies for the detection of α-synuclein positive lesions (see Methods, Figs. 1, 2, 3 and 4) and stereology-based measurements for a specific cerebellar deep nucleus, such as the DN, have been described before.

While future larger studies including more iPD autopsy-confirmed cases are necessary to completely confirm the exclusive cell type localization (oligodendrocytes) of α-synuclein positive inclusions need to be performed, as well as the analysis of iPD brains at different stages of the natural progression of the disease, the potential implications of these critical observations are not irrelevant, either in terms of possible clinico-pathological correlations and therapeutic opportunities. In fact, the oligodendrocytes have been traditionally known for their myelinating function and neuronal support^39–41, are now increasingly recognized as active participants in neurodegenerative processes⁴². The presence of α-syn inclusions within oligodendrocytes of the dentate nucleus of iPD brains suggests that this type of glial cells might not be merely passive bystanders but may be involved in the disease process, either in terms of initiation or progression, in extremely specific brain regions and related circuits, which have been recently demonstrated to regulate the dopamine function^25,43. Likely, these oligodendrocytic lesions could be directly associated with specific “cerebellar clinical manifestations” in iPD subgroups such as those characterized by prevalent resting tremor⁴⁴ or cerebellar non-motor deficits⁴⁵. These finding do align with research in Multiple System Atrophy (MSA) brains, where oligodendroglial cytoplasmic inclusions are a pathological hallmark, but also highlighting the capacity of α-synuclein to aggregate and exert pathology within oligodendrocytes in other conditions such as iPD^46,47. The presence of α-synuclein pathology in oligodendrocytes, and of peri-DN oligodendrocytes, raises critical questions about its impact on the oligodendrocytic function and white matter integrity in the context of iPD pathogenesis and clinical phenomenology. For example, α-syn inclusions within oligodendrocytes could disrupt their ability to form and maintain myelin, leading to demyelination or dysmyelination. Indeed, although in qualitative terms only, the DN of the iPD subjects assessed in this study appeared to show pallor in comparison to the age-matched C subjects (see Fig. 5). As a reminder, myelin is crucial for efficient axonal conduction and provides vital metabolic support to neurons⁴⁸. Dysfunction of oligodendrocytes, even without overt demyelination, can compromise neuronal health. For instance, impaired oligodendrocyte support could exacerbate energy deficits or oxidative stress in vulnerable neuronal populations, including those in the dentate nucleus or projecting through its white matter tracts⁴⁹. Furthermore, accumulation of misfolded α-syn within oligodendrocytes could trigger inflammatory responses, or impair cellular clearance mechanisms, creating a toxic microenvironment that could contribute to the progression of neurodegeneration mechanisms⁵⁰.

Clinically, the involvement of the cerebellar dentate nucleus and its associated white matter could have significant implications as for the diversity if symptoms, either motor or non-motor, manifested of iPD patients. While motor symptoms like tremor, rigidity, and bradykinesia are classically attributed to basal ganglia dysfunction, the cerebellum’s crucial role in motor coordination, balance, and motor learning and beyond is well-established^51,52. White matter abnormalities, often observed as white matter hyperintensities on MRI in PD patients, are increasingly linked to specific motor and non-motor symptoms. For example, as for motor symptoms, cerebellar white matter abnormalities have been associated with gait and balance impairments, which are often prominent and debilitating features of advanced PD, and frequently less responsive to dopaminergic therapies⁵³. These new findings concerning α-syn positive inclusions in oligodendrocytes of the DN could provide a direct neuropathological substrate for some clinical manifestations such as resting tremor^54,55 or cognitive changes⁵⁶, suggesting that cerebellar white matter dysfunction, driven by oligodendrocyte α-syn pathology, may contribute to motor and non-motor symptoms independent of, or in conjunction with, basal ganglia pathology.

Speculatively, the observed neuronal hypertrophy in the dentate nucleus of iPD brains could represent a compensatory mechanism for impaired connectivity, or conversely, a maladaptive response contributing to network dysfunction. As for non-motor symptoms, iPD patients are characterized by a wide array of manifestations, including neuropsychological deficits, mood disturbances, and autonomic dysfunction. As mentioned, the cerebellum is now recognized to play a significant role in non-motor functions such as cognitive and emotional processing through its mutual connections with cortical and subcortical areas⁵⁷. The white matter integrity is critical for these complex non-motor functions. Therefore, oligodendrocyte pathology and consequent white matter changes in the cerebellum could underlie or exacerbate specific non-motor symptoms in iPD, particularly those related to executive function, attention, and affective regulation as well as the correct temporal execution of all those functions⁵⁸. More specifically, DN abnormalities have been associated with dyskinesia in PD patients⁵⁹ as well as with onset of psychosis in other conditions such as schizophrenia⁶⁰.

Globally, understanding the role of oligodendrocytes and white matter α-synuclein pathology in iPD could open new avenues for the more precise characterization of various motor and non-motor manifestations along the natural history of iPD and offer the possibility to identify new or alternative therapeutic intervention. For example, targeting oligodendrocytes to prevent α-synuclein aggregation⁶¹, support myelin integrity, or promote remyelination could represent novel disease-modifying strategies⁶². Moreover, the identification of specific oligodendrocytic pathology within the dentate nucleus could lead to the development of novel imaging biomarkers (e.g., advanced MRI techniques sensitive to myelin or white matter integrity) or fluid biomarkers that reflect oligodendrocyte dysfunction, aiding in earlier diagnosis, disease monitoring, and patient stratification for clinical trials⁶³. The apparently unique α-synucleinopathy signature in iPD oligodendrocytes, distinct from neuronal Lewy pathology, might also offer insights into the heterogeneity of iPD and related synucleinopathies like MSA guiding so more accurate diagnostic and therapeutic approaches⁶⁴.

While these findings are novel, this study has some limitations. The relatively small sample size, particularly for the other brain regions beyond the DN and SN, limits the generalizability of our findings and the ability to perform more granular statistical analyses based on clinical phenotypes. Also, we could not completely exclude that other cerebellar deep nuclei (emboliform, globose and fastigial nuclei) might have α-syn positive oligodendroglial or other glial cell types inclusions as well. While all the cytological features and the double-stain immunofluorescence approach strongly suggest the specific accumulation of these α-synuclein inclusions in the oligodendrocytes, we cannot completely exclude that other glial cells types such as astroglial and microglial cells can result positive for the accumulation of α-syn inclusions as well, especially when considering the different pathological stages of the disease. This study primarily relied on immunohistochemistry for detecting α-synuclein pathology and the other common age-related co-pathologies; while this is a standard and robust method, it provides a static snapshot and does not offer insights into the dynamic processes of protein aggregation or cellular dysfunction, which could be offered by analyzing in vivo models.

A specific methodological consideration pertains to the use of 10µ-thick sections. Indeed, in a stereology-based measurements context, it is important to distinguish between the requirements for total cell counting vs. mean volume estimation. While unbiased counting using the Optical Disector principle typically requires thicker sections   (25–50 μm) to establish guard zones, our study focused exclusively on mean neuronal volume. For this purpose, the use of 10 μm sections is consistent with the Cavalieri principle, as thinner sections minimize projection artifacts (Holmes effect) and provide clearer boundary definitions for the Nucleator probe. Therefore, while we refrained from estimating total neuronal numbers to avoid the biases inherent in counting profiles on thin sections, the reported hypertrophy (mean volume) remains a robust metric derived from 2D profiles.

In general, while we identified oligodendrocytic α-synuclein inclusions in the peri-Dentate Nucleus region, the precise mechanisms leading to their accumulation in these specific types of cells, and the direct functional consequences on myelin formation and maintenance in that specific cerebellar area, will require further investigations using more detailed molecular and functional assays, which were beyond the scope of this study. The absence of a reliable counting method for DN and SN neurons due to tissue sample size also prevents conclusions about neuronal loss in these regions. Moreover, while the clinical diagnoses were re-assessed, the retrospective nature of the clinical data collection means that some detailed phenotypic information, particularly regarding non-motor symptoms and their severity, might be incomplete. Future research should aim to elucidate the precise mechanisms by which α-synuclein accumulates in oligodendrocytes, its specific toxic effects on these cells, and the resulting impact on myelin and axonal integrity within the cerebellar circuitry. Longitudinal studies combining advanced imaging techniques with detailed neuropathological examinations will be crucial to confirm the progressive nature of oligodendrocytic pathology and its correlation with clinical symptoms and disease progression in PD. Moreover, the development of in vitro and in vivo models specifically recapitulating oligodendrocytic α-synucleinopathy in the cerebellum will be essential for testing targeted therapeutic interventions. Lastly, it will require additional larger clinico-pathological correlation studies to conform that these α-syn inclusions are typical or exclusive of idiopathic PD brains in comparison to other Lewy body diseases (dementia with Lewy bodies [DLB]), MSA, AD, or normal aging relate processes in predisposed subjects.

Author contributions

D.I. conceived and designed the study. D.I. and H.P.  performed the neuropathological analyses and stereological measurements. H.P.  conducted immunohistochemistry and image processing. D.I., H.P.  interpreted the data. D.I., H.P., J.P.B., and R.K. drafted the manuscript with input from all authors. All authors reviewed and approved the final version of the manuscript.

Funding

MovemEnt Disorders Across Lifespan (MEDAL) program, Biomedical Research Institute of New Jersey (BRInj), Biomedical Sciences Research Fund 2023-2024.

Data availability

The data analyzed during the current study are not publicly available due to patient privacy and confidentiality restrictions but are available from the corresponding author on reasonable request. For data requests, please contact the corresponding author, Dr. Diego Iacono, at diego.iacono@atlantichealth.org.

Declarations

Competing interests

The authors declare no competing interests.