Are oligodendrocytes bystanders or drivers of Parkinson’s disease pathology?

Oligodendrocytes in Parkinson’s disease pathology: What do we know?

Magnetic resonance imaging (MRI) studies have shown that PD patients with dementia have significant structural changes in the majority of white matter tracts of the brain [12,13], often interpreted as demyelination. Although this type of measurement cannot directly assess the myelin content in patients, it may indicate cellular oligodendroglial changes or even a decline in the number of oligodendrocytes. Examining the peripheral nervous system, one study observed a significant loss of unmyelinated axons in the epicardial nerve of PD patients [26]. Even though this study does not strictly involve oligodendrocytes, as they only reside in the central nervous system, it highlights the neuroprotective function of myelin in the peripheral nervous system. In addition to the main pathological hallmarks, brain-wide oxidative stress is also an important feature of PD, and oligodendrocytes and oligodendrocyte progenitors cells (OPCs) are known to be highly vulnerable to oxidative damage [27]. This is due to a combination of high metabolic rates with potentially dangerous by-products, high intracellular iron, and low concentrations of antioxidants [28], further indicating that oligodendrocyte death may occur and contribute to the exacerbation of neuronal pathology.

Perhaps surprising for the field was the discovery of α-synuclein inclusions in oligodendrocytes in samples from PD patient brains [29,30]. This protein is thought to be expressed primarily in neurons, where it is known to regulate synaptic functions, and was therefore not expected to be present in oligodendrocytes. The source of α-synuclein remains under debate, as single nucleus RNA-sequencing approaches revealed endogenous expression of SNCA, the gene that encodes for α-synuclein, in oligodendrocytes in the human brain [31]. Oligodendrocytes from a rodent immortalized cell line are also capable of taking up monomeric and oligomeric forms of this protein in vitro [32]. Although this phenomenon is not very well defined in the context of PD, oligodendroglial α-synuclein inclusions are the main hallmark of multiple system atrophy, a rare neurodegenerative disease which also presents with parkinsonism [33]. Given that these diseases share some clinical and histopathological features, the question naturally arises as to how oligodendrocytes may be affected in PD and how they, in turn, respond to the pathological progression of the disease.

Imaging- and histology-based techniques for assessing oligodendrocytes in the context of PD are sparse, and interpretations of how these observations relate to the pathological developments are difficult. Due to technological advances in next-generation sequencing, studying oligodendroglial transcriptomic signatures in human patients has become easier. Indeed, recent single-cell and single-nuclei transcriptomic studies on human samples of the whole midbrain, the substantia nigra and other associated structures [11,15,34–39], have identified that oligodendrocytes are, after neurons, likely the second most affected cell type in PD (a detailed summary of study results is given in Table 1). Although these datasets largely vary in size, ranging from 17,000 to 316,000 analyzed nuclei, common findings were that oligodendrocytes always comprise the largest cell population in these brain regions and that the transcriptional landscape of oligodendrocytes was always altered. Depending on the detail and focus of the analysis, oligodendrocyte numbers are significantly depleted [39], or shifted in their proportional composition of the subpopulations, with some being depleted, but some also enriched [38]. The identified transcriptional alterations that were identified in PD oligodendrocytes were largely variable across studies, including an increase in stress response [34], inflammation [36], or response to unfolded protein [39] alongside an apparent loss of their myelinating capacity [36,39]. The differences in these findings affecting functional changes could likely be derived from different samples sizes, with largely variable gene numbers and different bioinformatic methods being applied. Most of these studies have combined these data with genome-wide association studies (GWAS) [9,11,35] or other means to include genetic risk-loci [15] and found that the genetic risk in PD is associated not only with dopaminergic neurons, as expected, but also with oligodendrocytes, and that oligodendrocytes can also be associated with clinical outcome prediction [37]. Furthermore, simultaneous use of single-nucleus RNA- and ATAC-sequencing of the human substantia nigra has also revealed that differential changes present in oligodendrocyte transcriptomes can be particularly associated to pathology progression [34].

Particularly surprising in this context is the apparent down regulation of genes involved in myelination, possibly indicating that this particular cell type simply becomes dysfunctional, or that oligodendrocytes need to take on other, as yet unknown, functions [34,36]. This finding is puzzling because dopaminergic neurons in the substantia nigra, the primary cell type affected in PD, are considered to be sparsely myelinated [14], at least according to common knowledge (discussed in the next section). In addition, a stress responding and inflammatory transcriptomic profile has been brought to attention in oligodendrocytes from samples of the striatum and the entire midbrain [34,35]. These results rather suggest a complex functional influence of oligodendrocytes in the pathogenesis of PD that—if oligodendrocytes become affected—goes beyond simple myelin loss and potential cell demise, although clear conclusions are difficult to draw, as human single-cell transcriptomic data come from a single time point of pathology.

In summary, the limited studies on oligodendrocytes in PD provide first evidence of non-cell-autonomous changes associated with reduced functionality of oligodendrocytes that may have widespread effects on the susceptibility of dopaminergic neurons, and therefore raise the hypothesis that these changes may potentially play a role in contributing to disease pathogenesis. While valuable, these studies are largely descriptive and indicate a change in the transcriptional profile of oligodendrocytes. Thus, further studies are needed to shed light on the functional impact of these changes on oligodendrocyte dynamics, and potentially reveal underlying molecular mechanisms contributing to pathology.

How severely are oligodendrocytes functionally altered in Parkinson’s disease?

Given the discoveries of transcriptional changes in oligodendrocytes in PD highlighted above, several important questions for the field need to be raised. First, do oligodendrocytes play an active role in the pathogenesis of the disease or are they merely secondary victims of the pathological changes in affected neurons? Second, do these cells initiate and promote disease progression in early stages, or do they only respond to global damage from neighboring cells? Literature indeed suggests the possibility of an early active pathogenic role of oligodendrocytes. For example, the transcriptomic studies from human samples reported a shift in their gene expression profile as early as Braak stages 1 and 2 [11]. Nevertheless, these data only represent a single time point in the course of the disease. It is therefore impossible to say what happens in earlier stages of the disease and whether these transcriptomic changes are preceded by other physiological changes. We believe that appropriate human in vitro models could aid address these pathologies and, in long term, could lead to the development of new research strategies and potential therapies in the future.

The ability to combine single-cell transcriptomic and GWAS has enabled researchers to better understand which cell types carry disease-associated single-nucleotide polymorphisms (SNPs) and has contributed tremendously to bringing oligodendrocytes into focus [11]. However, how these SNPs influence the function of oligodendrocytes and in turn disease progression remains to be elucidated. As demonstrated in one study [34], the multi-omics approach that combines single-cell RNA-sequencing with single-cell ATAC-sequenzing (to assess genome-wide chromatin accessibility) on the very same cell, offers the possibility—albeit limited—of understanding which genes are affected by the SNP in a particular cell type. Further development of these combined technologies will greatly improve our understanding of the cell-specific pathological mechanisms in PD. Identifying and understanding these mechanisms requires an enormous collaborative scientific effort, and longitudinal studies using animal models and human patient material would be of great value to the field. This may also answer the question of whether oligodendroglial changes progress linearly or whether the observed pathological alterations also depend on different disease stages. This could open a whole new set of biomarkers that would aid patient care and prognosis.

Does demyelination in the substantia nigra play a role in Parkinson’s disease?

Myelination in patients with PD has been largely assessed by MRI, which, as previously mentioned [12,13], is only an indirect measurement of myelin content; therefore, the question of whether demyelination is a common feature in PD remains largely speculative (Fig 2). Therefore, a thorough examination of the myelin content in patients at different stages of the disease will help us understand whether myelin changes precede and therefore could be causative of neuronal loss, or whether sublethal damaged neurons cause oligodendroglial changes. The general assumption in this field of research is that the degenerating dopaminergic neurons in the substantia nigra are only sparsely myelinated [14], which is probably one reason why the study of oligodendrocytes and myelin has not been the focus of research to date. However, it seems that the low degree of myelination of human dopaminergic neurons has fallen into a state of common knowledge, with actually little experimental evidence to support this statement.

Fig 2. Function and dysfunction of oligodendrocyte lineage cells in healthy and PD brains.

(A) In the healthy brain, the primary role of oligodendrocytes is the myelination of axons, though dopaminergic neurons are generally thought to have minimal myelination [14]. Moreover, both OPCs and non-myelinating oligodendrocytes are recognized for their crucial roles in signaling and providing metabolic support [10], although the exact function of the non-myelinating oligodendrocytes is even less clear. (B) In the PD brain, the particular consequences of damaged or non-functional oligodendrocytes is not well understood. Although there are reports of demyelination in patient brains [12,13], it is still unclear to what extent this affects dopaminergic neuron health or whether demyelination precedes neuronal damage. α-synuclein positive inclusions have been found in oligodendrocytes in PD patient brains [29,30]; however, it remains unclear whether this is due to cell-autonomous build-up of α-synuclein or protein cell-to-cell transfer from neurons to oligodendrocytes. It is hypothesized that diseased oligodendrocytes and OPCs may lose their direct signaling and metabolic support functions (indicated by dashed red lines) to neurons due to pathological changes in their environment, such as the accumulation of ROS and iron (Fe) accumulation that could exacerbate neuronal damage. Created with BioRender.com. OPC, oligodendrocyte progenitor cell; PD, Parkinson’s disease; ROS, reactive oxygen species.

https://doi.org/10.1371/journal.pbio.3002977.g002

On the contrary, newer studies, showing, for example, high proportions of oligodendrocytes [15,34,35,38] (summarized in Table 2) and high levels of myelination in the substantia nigra [40], suggest a higher degree of myelination of dopaminergic neurons than previously assumed. Additionally, improved technologies have shown that many neurons in regions of gray matter are only intermittently myelinated [41], demonstrating that even small fragments of myelin along axons are critical for complex neuronal functions [42]. Given this updated information, the myelin content of dopaminergic neurons should be reassessed using state-of-the-art technologies. Furthermore, it needs to be considered that PD is characterized by more widespread pathology in other brain regions and involves non-dopaminergic neurons as well. Hence, demyelination and malfunction of other neuronal populations may also contribute to motor and non-motor symptoms associated with disease progression. Finally, the high number of oligodendrocytes in areas with little apparent myelination suggests that myelination is likely not the sole purpose of these cells, and that their contribution to normal brain physiology may be far more pleiotropic than previously thought.

Do oligodendrocyte functions deteriorate due to oxidative stress?

There is strong evidence for an imbalance of brain iron in PD associated with pathology, and the presence of elevated oxidative stress levels is an established pathological feature [45]. Oligodendroglial iron has not yet received much scientific attention in the context of PD, despite oligodendrocytes having the highest iron content in the central nervous system [46], essential for their metabolism. Given the complicated and intimate nature of communication between neurons and oligodendrocytes (Fig 2) and their antioxidant function related to iron metabolism [47], we raise the question whether oligodendrocytes exacerbate or mitigate neuronal iron overload and its associated neurotoxicity. Oligodendrocytes have been shown to be particularly vulnerable to oxidative damage [48], but there is still little known about the sources of reactive oxygen and nitrogen species and the effects on specific functions of oligodendrocytes, particularly with regard to direct coupling of neurons. We speculate that these toxic agents in the environment lead to impaired oligodendrocyte metabolism, rendering them incapable of providing their physiological protection to neurons, or, alternatively, oligodendrocytes themselves could be the source of these toxic components. Therefore, it should be considered that antioxidant support from oligodendrocytes on neurons [47] may still be present, but disease conditions overwhelm the process as the disease progresses.

What is known about the spread of α-synuclein pathology between oligodendrocytes and neurons?

Evidence of oligodendroglial α-synuclein aggregates in PD has recently emerged [29,30] (Fig 2), which was initially thought to be due to oligodendroglial uptake of monomeric and oligomeric forms of α-synuclein [32]. However, more sensitive approaches have revealed that oligodendrocytes do express α-synuclein mRNA and protein [49]; yet the functional and disease-associated implications of this finding remains to be studied. To this end, we propose that advanced disease modeling platforms may be beneficial, such as co-culture systems of neurons with oligodendrocytes, to help assess whether oligodendrocytes are capable of taking up neuronal α-synuclein, thereby reducing the pathological burden, and how this may affect progression of pathology. Or, as another possibility, whether an abnormal function or level of α-synuclein in oligodendrocytes may account as a triggering determinant of pathology affecting α-synuclein levels and accumulation in neighboring neurons. In-depth investigation of this topic is critical to better understand possible risk factors and treatment strategies in PD, as well as other synucleinopathies.

What role might oligodendrocyte progenitor cells play in the pathogenesis of Parkinson’s disease?

Since OPCs were considered merely precursors of oligodendrocytes for a long time after their discovery, many non-canonical functions of OPCs have been discovered, which have been extensively discussed previously [50]. Although not much is known about the contribution of OPCs to PD pathology yet, many of their functions could be of potential importance in disease pathogenesis (Fig 2). Several publications have shown that OPCs are important for neuronal function and their deficiency induces depression-like behavior [51] or obesity [52] in mice. Due to their strong synaptic coupling with dopaminergic neurons [53] (Fig 2), malfunctioning OPCs could disrupt cellular crosstalk, which could, in the long term, lead to functional impairment of this vulnerable cell type. Moreover, OPCs are capable of synapse pruning [54], which may be of pathological significance given the severe synapse loss in PD [55]. Lastly, OPCs have also been shown to have immunomodulatory functions via direct cytokine signaling [56] or through the acquisition of properties of antigen-presenting cells in multiple sclerosis [57,58], a chronic inflammatory condition resulting in oligodendrocyte death and demyelination. With regard to PD, an immunomodulatory function of OPCs could potentially exacerbate neuroinflammation.

Oligodendrocytes in Parkinson’s disease pathology: What are the major challenges?

In recent years, studies using human brain tissue and next-generation transcriptomic approaches have proven indispensable for describing previously unknown pathological changes. However, the tissue is a limited resource and possible artefacts (due to non-standardized handling protocols of the postmortem tissue) is a source of great variability in the data. Furthermore, human brain tissue only provides a snapshot of the disease, and cellular and molecular changes that occurred long before disease onset remain speculative, making additional approaches essential.

Based on the Alzforum database (www.alzforum.org), there are currently 33 genetic and several other drug-induced rodent models of PD available. Although they span over a variety of risk genes, pathological features, and several molecular mechanisms, they all aim at reproducing well-known key pathological hallmarks of the disease and may therefore have limited relevance to oligodendrocyte biology. Moreover, mouse oligodendroglia differ significantly from human oligodendroglia at a fine transcriptomic level [59], which could potentially result in unique species-specific responses to pathological stimuli, further complicating the use of these animal models to study oligodendrocytes. This was indeed shown to be the case in a comparative transcriptomic analysis of human and murine oligodendrocytes in Alzheimer’s disease research [60]. In fact, this study confirmed that the oligodendrocyte activation signature observed in human Alzheimer’s disease is largely distinct from those observed in mice. Possibilities to better understand the pathological mechanisms are, for example, the generation of novel animal modes based on identified molecular pathways that are critical for oligodendrocytes in PD, as well as the “humanization” of animal models through transplantation of human oligodendrocytes.

Human iPSC models have made significant progress, and their use, particularly for studying dopaminergic neurons, has increased significantly in the PD field [8,16]. However, obtaining iPSC-derived oligodendrocytes remains a challenge (see section “Why has the role of oligodendrocytes in PD pathogenesis remained so elusive?”). Although stem cell-derived models do not fully recapitulate the human brain in general, and a neurodegenerative disease due to several limitations in particular, they are highly valuable for studying the underlying molecular mechanisms of cells in isolation. Having robust oligodendrocyte models that can also be kept for an extended amount of time and expanding these to more complex systems, such as co-cultures and oligodendrocyte-harboring and myelinating 3D organoid models, will allow the examination of cell–cell communication and network building. These models should therefore—in parallel to advancing suitable in vivo models—be of highest priority.

Therefore, due to the limitations of all model systems, a concerted scientific approach combining the use of human imaging data and postmortem material, together with human in vitro iPSC platforms and rodent models, may provide some further insight to open scientific questions on this topic.