The many lives of an oligodendrocyte precursor cell

Abstract

Oligodendrocyte precursor cells (OPCs) are a population of glia that give rise to myelinating oligodendrocytes during development and in the adult brain. While OPCs have long been appreciated for their roles in oligodendrocyte production, emerging data suggest that they perform a wide range of other functions as well. For example, OPCs are the only glial cell type known to receive direct synaptic input from neurons, and they respond to neural activity through the release of factors that alter neuronal function. Moreover, OPCs directly associate with the neurovasculature to promote blood brain-barrier maintenance and integrity. Emerging data suggest that OPCs are also instrumental for refining synaptic connectivity during postnatal brain development, a process to which they contribute by phagocytosing synapses. Finally, OPCs are also involved in brain immunity, as they can adopt immune cell-like functions, such as antigen presentation, during demyelinating and neurodegenerative diseases. Altogether, these findings have identified OPCs as the major multi-taskers of the brain. In this review, we discuss the roles of OPCs that extend beyond oligodendrocyte production and their relevance for neurological function.

Introduction:

Oligodendrocyte precursor cells (OPCs) are a unique population of multifunctional glia that tile the brain, maintaining distinctive territories and retaining the ability to divide across the lifespan. At any given time, this population exhibits a diversity of states, including those associated with cell division, differentiation, migration, and cell death(Dawson 2003). OPCs are also referred to as NG2 glia based upon their expression of the cell surface protein Nerve/Glial antigen 2 (Dawson 2000). By giving rise to oligodendrocytes, OPCs participate in the production and repair of myelin, a proteinaceous, fatty substance that surrounds neuronal axons to facilitate synaptic communication between neurons (Dawson 2003, Peters 1960, Rushton 1951). In addition to their contributions to myelination, several new functions of OPCs have been recently discovered in large part thanks to technical advances in the neuroscience field, including in vivo multi-photon imaging, single-cell transcriptomics, and serial section transmission electron microscopy (ssTEM). Thus, understanding the non-canonical functions of OPCs has emerged as an exciting and technically feasible new direction in the quest to uncover the breadth of glial functions within the brain.

As the roles of OPCs in myelination have been extensively reviewed elsewhere (Emery & Wood 2024)(Hill et al 2024), in this review, we focus on the increasingly appreciated ability of OPCs to mediate neurological function through mechanisms extending beyond the generation of mature oligodendrocytes. We chronical the identification of OPCs from a historical perspective then discuss their ability to give rise to non-oligodendroglial cells such as astrocytes, their direct interactions with synapses and roles in governing synaptic connectivity, and their crosstalk with the neurovasculature. We conclude with the immunomodulatory roles of OPCs, such as phagocytosis and antigen presentation. Finally, we discuss the open questions and future directions that are most promising in their potential to improve our understanding of OPC function in the next 5 – 10 years. Our goal in this review is to highlight the remarkable ability of OPCs to juggle an impressive array of functions in the brain.

The discovery of OPCs: a historical perspective

Neurons and their synaptic specializations were first identified by Golgi(Golgi 1873) and Cajal(Cajal 1888) in the late 19^th century, preceding the discovery of glia. While the first mention of the term neuroglia is often ascribed to Virchow in 1858, it was not until more advanced tissue staining techniques were developed that the detailed structure of what he referred to as ‘putty’ or ‘glue’ was visualized (Kettenmann & Ransom 2004) (Somjen 1988). Subsequently, during the late 19^th and early 20^th centuries, anatomists and pathologists began using various heavy metals and other chemicals to visualize the brain’s microscopic structures in greater detail (Cajal 1893) (Javier DeFelipe 1991). These new histological staining methods enabled the identification of what scientists named the “third element”, later renamed neuroglia (Kettenmann & Verkhratsky 2008). With improved contrast, beautiful images emerged from these studies, and the glial cells were subdivided into astrocytes, microglia, and interfascicular oligodendroglia (Cajal 1888, Cajal 1991, Del Rio Hortega 1919).

The subsequent advent of immunological staining techniques allowed for the discovery of cell type-specific molecular markers for neuronal and glial cells. Following the initial discovery of ‘oligodendroglia’, decades elapsed before OPCs were identified as a type of bipotential glial progenitor that could differentiate into either an astrocyte or an oligodendrocyte (discussed further below)(Levine & Stallcup 1987, Peters 2004, Peters & Sethares 2004, Raff et al 1983). In particular, the detection of the cell-surface antigen NG2 identified a sub-population of cells bearing numerous processes and displaying morphological features of both neurons and glia. These NG2-positive cells, which were differentiated from astrocytes based upon their ultrastructural features, came to be defined as OPCs (Levi et al 1986, Stallcup 1981). Intriguingly, these newly identified OPCs were found to be in contact with numerous neuronal axons containing vesicles strongly resembling glutamatergic synapses (Ong & Levine 1999). These observations were prescient, as later findings established the presence of bona fide synaptic contacts between neurons and OPCs, i.e. sites at which the presynaptic inputs of neurons innervate OPCs directly via synaptic transmission.

It was not until the late 20^th and early 21^st centuries that the emergence of other advanced techniques (including genetic fate-mapping methods, transgenic and knockout animal models, array tomography, flow cytometry, single-cell RNA sequencing, and volume electron microscopy) vastly improved our knowledge of glia in general and OPCs specifically. In particular, new imaging methods, such as multi-photon microscopy in the brains of mice in which glial cells are labeled with fluorescent proteins, revealed the dynamic nature of OPCs and their numerous processes in vivo, yielding a much clearer view of OPC differentiation, migration, and division (Hughes et al 2013, Hughes et al 2018, Kachar et al 1986, Kirby et al 2006, Thornton et al 2024). In combination with fate-mapping studies, these imaging-based analyses also revealed that OPCs continue to differentiate and give rise to myelinating oligodendrocytes not just during development but in the mature brain as well, maintaining a stable population across the lifespan.

Complementary studies assessing glial morphology at nanometer resolution using ssTEM uncovered the anatomical details of OPCs, astrocytes and microglia. The overall similarities between OPCs and microglia were evident, but their ultrastructural features revealed the unique properties of each. In particular, microglia lack primary cilia because they originat from yolk sac monocytes, while neural tissue-derived OPCs retain their primary cilia until they differentiate into premyelinating oligodendrocytes(Falcon-Urrutia et al 2015). The combination of ultrastructure with dense 3D reconstructions of all cells in the volume shed light on the numerous interactions between OPCs, other glia, vasculature, neurons and synapses ((Bae et al 2023), Buchanan et al (2022). Altogether, these studies established OPCs as brain-resident glial cells of the oligodendroglial lineage that are structurally and dynamically distinct from other types of glia.

The origin and distribution of OPCs in the brain

While studies across numerous model systems have been critical for beginning to characterize the roles of OPCs in the brain, the features and functions of OPCs have been best described in mice. Between embryonic days (E)12.5 and just after birth in the mouse, OPCs are generated in three distinct regions of the developing brain in a series of waves that follow a spatiotemporal gradient from ventral to dorsal. OPCs first appear in the ventricular germinal zones (VZs) of the ventral forebrain and spinal cord (specifically the medial ganglionic eminence [MGE] and the anterior entopeduncular area) around E12.5, with a second wave of OPCs deriving from the VZs of the lateral ganglionic eminence (LGE) and the central ganglionic eminence (CGE) appearing around E16 (Kessaris et al 2005). In both cases, these early-born OPCs proliferate and migrate rapidly throughout the brain, eventually becoming evenly distributed across gray and white matter (Pringle & Richardson 1993). Finally, a third wave of OPCs emerges early postnatally from Emx1-expressing progenitors in the dorsal VZ. These later born OPCs are thought to represent a significant proportion of the OPCs that persist into adulthood, although a population of early derived OPCs is also maintained (Cai et al 2024, Kessaris et al 2005). The precise origin of OPCs in the brain is therefore variable depending on the specific region of the CNS and the stage of development analyzed (Bergles & Richardson 2015, Foerster et al 2024, Kessaris et al 2005). Thus, OPCs arise from diverse developmental origins (as extensively reviewed in Bergles and Richardson, 2015 (Bergles & Richardson 2015)), a feature that may suggest heterogeneous roles for OPC subtypes in the brain.

Potential generation of astrocytes and neurons by OPCs

One of the most apparent and earliest described myelination-independent functions of OPCs involves their ability to give rise to cells outside of the oligodendroglial lineage, most prominently astrocytes (Sawamura et al 1995). This multipotent capability was first identified in the lab of Dr. Martin Raff in the early 1980s based upon studies of OPCs isolated from rat optic nerve. Raff and colleagues found that exposing acutely isolated OPCs to fetal calf serum (FCS) in culture promotes the generation of a subset of astrocytes, called type-2 astrocytes, regardless of the status of the tightly regulated ‘intrinsic clock’ through which OPCs have been shown to differentiate into oligodendrocytes (Miller et al 1985, Raff et al 1983)(Bogler et al 1990, Chen et al 2009, Ffrench-Constant & Raff 1986, Temple & Raff 1986, Williams et al 1985). OPCs capable of giving rise to type 2 astrocytes are termed oligodendrocyte-type 2 astrocyte (O-2A) progenitor cells. These findings suggest that OPCs may differentiate into oligodendrocytes by default but that they can fate switch upon exposure to certain extrinsic cues, positioning OPCs to compensate for astrocytes when their function is perturbed.

The extrinsic cues that drive O-2A cells to differentiate into astrocytes rather than oligodendrocytes include bone morphogenetic proteins (BMPs)(Mehler et al 1997) (Mabie et al 1997) and interferon-γ IFNγ, (Bogler et al 1990, Tanner et al 2011). Exposure of OPCs to IFNγ downregulated PDGFRα expression and decreased their ability to self-renew, returning the O-2A population to a quiescent state. This IFNγ-mediated push of OPCs toward an astrocytic lineage was reversable, such that the removal of IFNγ from the media led to a return of OPCs to a baseline oligodendroglial phenotype. BMPs, a subclass of TGFβ superfamily members, also drive the differentiation of O-2A cells toward an astrocytic lineage while suppressing the induction of oligodendrocyte differentiation(Mabie et al 1997). These experiments provide evidence that extrinsic factors, including BMP and IFN family members, can influence the cell fate of an OPC in vitro.

While the majority of these studies were performed in vitro, emerging evidence suggests that OPCs may be able to give rise to non-oligodendroglial cells in vivo as well, although evidence of this is limited. Genetic fate mapping studies, such as those utilizing NG2 and PDGFRα promoters to drive Cre-dependent reporter expression in OPCs, showed that OPCs can give rise to both oligodendrocytes and a subset of neocortical, gray matter astrocytes in the intact brain(Huang et al 2014, Zhu et al 2008b). Similar result was obtained by a different group who showed that selectively labeling Sox10-positive cells (i.e. oligodendroglial lineage cells) with a virally expressed fluorophore results in the eventual labeling of not only OPCs and mature oligodendrocytes but astrocytes as well (Suzuki et al 2017). The ability of OPCs to generate astrocytes was not limited to the neocortex as OPCs can also generate astrocytes in the spinal cord(Zhu et al 2008b). Nor was the generation of astrocytes by OPCs restricted to development as it was also observed in the injured adult brain(Tatsumi et al 2005). These findings provide evidence that OPCs may give rise to astrocytes in vivo, although this issue is still debated in the field.

Interestingly, the ability of OPCs to produce astrocytes appears to vary with age, consistent with the availability of soluble cues like BMPs and IFNs changing across the lifespan. For example, OPCs transplanted into the mouse brain tended to give rise to oligodendrocytes when the transplantation took place at postnatal day (P)7 (when developmental myelination is still underway) but gave rise to GFAP-positive astrocytes when the transplantation occurred much later at 6 months of age(Sawamura et al 1995). Differences in the fate of OPCs depending upon the age at which they are transplanted into the brain is consistent with emerging evidence that OPCs become more heterogeneous with age, as discussed further below. These differences also likely reflect that OPCs are initially dedicated to generating oligodendrocytes prior to the establishment of mature patterns of myelin in the brain (which typically occurs over the first 2–3 months in the mouse), but that they can more flexibly generate astrocytes once that process is complete. Overall, these data suggest that OPCs possess an ability to differentiate into a subclass of astrocytes in vitro and potentially in vivo depending upon their exposure to specific molecular cues.

In addition to astrocytes and oligodendrocytes, OPCs may also have the capacity to give rise to neurons (Belachew et al 2003, Guo et al 2010, Kondo & Raff 2000, Liu et al 2007, Rivers et al 2008) although this topic is still hotly debated in the field. OPCs residing in neocortex mostly derive from cells born in the embryonic ganglionic eminences, regions that also produce inhibitory GABAergic neurons. Fate mapping studies have demonstrated that some neurons in the adult hypothalamus and piriform cortex derive from OPCs(Guo et al 2010, Rivers et al 2008, Robins et al 2013). Moreover, the exogenous expression of Dlx homeobox transcription factors, which are required for GABAergic neuron production, represses OPC differentiation into oligodendrocytes and pushes them instead toward an inhibitory neuron fate(Boshans et al 2021, Petryniak et al 2007). Indeed, some cortical interneurons in the mammalian forebrain were found to be derived from OPCs as well(Tsoa et al 2014). Interestingly, in addition to generating non-oligodendroglial cells during brain development, OPCs are also thought to give rise to adult-born neurons, suggesting a regenerative capacity that could make these cells appealing therapeutic candidates for addressing white matter loss in the brain(Belachew et al 2003, Dayer et al 2005, Robins et al 2013). That said, other cell fate studies have reported an absence of evidence that OPCs can give rise to neurons in vivo (Dimou et al 2008, Huang et al 2019, Huang et al 2014, Kang et al 2010, Zhu et al 2008a, Zhu et al 2011). Thus, while it appears that OPCs can be intentionally directed toward a neuronal lineage through manipulations in vitro and in vivo, whether or not OPCs represent a major progenitor pool for GABAergic neurons in the intact brain has not yet been resolved.

Regional and age-dependent diversity of OPCs

While OPCs exert a major influence on myelination during brain development, the persistence of a robust population of OPCs across the lifespan suggests that they may take on more myelination-independent functions as the organism ages. Indeed, OPCs are known to change in defined ways as a result of aging. For example, in early and mid-adulthood, the activation of the OPC cell-surface receptor PDGFRα led to the phosphorylation of downstream signaling proteins thereby promoting the transcription of genes involved in cell cycle progression and proliferation (Jiang et al 2023, Kucharova & Stallcup 2015, Medved et al 2021). However, PDGFRα levels in OPCs decrease during aging, leading to a decline in OPC proliferation and a dampening of their ability to respond to extrinsic growth factors that are crucial for their maintenance and differentiation (Calabretta et al 2018, Falcon-Urrutia et al 2015). Other features of OPCs that shift across the lifespan include their intrinsic electrophysiological properties, calcium signaling dynamics, morphological features, and transcriptomic profiles, which can also vary across different regions of the CNS. For instance, the expression and function of ion channels, such as voltage-gated sodium and potassium channels and ionotropic neurotransmitter receptors, within OPCs fluctuate significantly across the lifespan, with ion channel expression decreasing as OPCs mature (Berret et al 2017, Black & Waxman 2013). This in turn can affect the membrane properties of OPCs, altering their cellular homeostasis (Kamen et al 2024, Khawaja et al 2021). Aging is also associated with increased oxidative stress in OPCs, which can further exacerbate the decline in OPC proliferation and differentiation that occurs with age (Spaas et al 2021). Aging is also associated with alterations in the proteomic profiles of OPCs, as well as epigenetic modifications like DNA and histone methylation, which are predicted to underlie age-related shifts in OPC function by remodeling the molecular repertoire available to OPCs (Spitzer et al 2019) (Young et al 2013). Thus, OPCs change across the lifespan, indicating a level of plasticity in OPC function that is only beginning to be appreciated.

In addition to age-related changes, OPCs also exhibit different properties depending upon their location within the brain. For example, mouse brain OPCs in gray versus white matter exhibit distinct morphological characteristics, membrane properties, calcium dynamics, and transcriptional profiles (Chittajallu et al 2004, Karadottir et al 2008). Similar differences between gray and white matter OPCs were observed in humans (Osorio et al 2022, Wang et al 2020a). Transplantation experiments in mice in which white matter and gray matter OPCs were isolated and exchanged demonstrated that gray matter-derived OPCs took longer to mature into myelinating oligodendrocytes compared to white-matter OPCs, even when exposed to the white matter environment (Viganò et al 2013). Conversely, white matter–derived OPCs continued to differentiate into mature oligodendrocytes with a higher efficiency than gray matter–derived cells even when transplanted into the gray matter. These data suggest that inherent differences between OPCs in gray and white matter are retained upon transplantation into the non-originating tissue (Viganò et al 2013). More recently, long-term 3-photon imaging of the cortex and white matter areas demonstrated that OPCs in some areas exhibit enhanced oligodendrogenesis in response to demyelination, while OPCs in other areas did not(Thornton et al 2024).

Consistent with these observations, OPCs that contact brain subregions with many axons and dendritic processes exhibited higher calcium signaling and unique gene expression profiles compared to OPCs in neuronal cell body-rich areas (Spitzer et al 2019). Even within a single brain region, the mouse primary visual cortex, OPCs can display diverse ultrastructural differences such as the presence or absence of primary cilia, a harbinger of differentiation (Buchanan et al 2022). These findings suggest a level of diversity among OPCs that is consistent with different subtypes having functionally segregated roles (Spitzer et al 2019). Overall, the diversity of OPCs is emerging as a feature of these cells that is crucial for their adaptability and function in the CNS, influencing development, maintenance, and repair processes throughout life.

OPC-neurovascular interactions

Interactions between OPCs, neurons, and other types of glia represent a growing hotbed of research, including interactions between OPCs and the neurovascular unit (NVU). In particular, OPCs are thought to be key for establishing and maintaining blood-brain barrier (BBB) integrity(Seo et al 2014). The blood-brain barrier (BBB) is a specialized structure comprised of vascular endothelial cells, pericytes, and the end feet of astrocytes, and this structure represents a strong and effective physical boundary between the blood and the brain that is critical for protecting the brain from inflammatory signals originating in the periphery. Genetic ablation of NG2-positive OPCs in utero led to reduced complexity of the evolving neurovasculature, including a decrease in blood vessel density and branching (Minocha et al 2015). Postnatally, OPCs continue to induce angiogenesis particularly in the white matter via HIF and Wnt signaling, and engage TGFβ secretion and PDGF-BB signaling to increase BBB integrity by heightening the tightness of endothelial cell junctions(Kimura et al 2020, Seo et al 2014, Yuen et al 2014). Thus, OPCs are likely instrumental for the BBB’s ability to protect the brain from inflammation, and have been proposed as bona fide members of the NVU as a result (Pfeiffer 2022). Although an association between these two factors has not yet been made, the ability of OPCs to generate astrocytes, the endfeet of which help maintain BBB integrity, could serve to increase the integrity of the BBB as well.

Emerging evidence suggests that the ability of OPCs to mediate the development and maintenance of the BBB likely requires their direct association with pericytes in vivo (Girolamo et al 2019). Fluorescence and electron microscopy revealed that OPCs make tight structural contacts with pericytes surrounding the neurovasculature near the basal lamina, and in vitro studies demonstrated that signals from pericytes increase OPC numbers and maintain OPC health(Maki et al 2015). This could allow pericytes to promote CNS myelination and repair in the contexts of development and disease. Endothelial cells are also thought to secrete factors that promote OPC proliferation (Arai & Lo 2009). Reciprocally, OPCs signal to pericytes to increase their number as well, a process that may underlie the ability of OPCs to drive angiogenesis and maintain BBB integrity(Maki et al 2015). Extending these findings to an in vivo system, a study by de la Fuente et al showed that partial depletion of pericytes from the brains of mice in a demyelination disease model impaired OPC differentiation, highlighting an important role for pericytes in shaping OPC function in the brain(De La Fuente et al 2017). Altogether, these results indicate that OPCs are necessary for promoting neurovascular structure and BBB integrity, and that they likely do so through a combination of direct physical interactions and molecular signaling with pericytes.

Perhaps one reason that OPCs and pericytes interact so extensively is that OPCs utilize physical interactions with the NVU to migrate through neural tissue, particularly in the context of inflammation and injury. For example, in the Middle Cerebral Artery Occlusion (MCAo) mouse model of stroke, OPCs were recruited to regions of damage by using the vasculature as a scaffold(Bonfanti et al 2017). When OPCs were unable to effectively utilize the vasculature as a scaffold, they formed clusters around the NVU. These clusters were particularly apparent near white matter lesions in Multiple Sclerosis (MS), where OPCs become unable to properly detach from blood vessels, thereby evicting astrocytic endfeet from the BBB and disrupting its integrity (Su et al 2023). Consistent with OPC dysfunction underlying MS pathology, the transplantation of healthy OPCs in the MCAo stroke model aided in the recovery of neurological function by activating Wnt and β-catenin signaling (Wang et al 2020b). Thus, OPCs play a role in supporting the neurovasculature in health via interactions with pericytes but can compromise BBB integrity in the context of disease.

Neurons and OPCs communicate at bona fide synapses

OPCs and neurons communicate bidirectionally throughout life to support healthy neurological function. Like other types of glia, OPCs communicate with neurons through ligand-receptor signaling pathways and direct physical contact (reviewed elsewhere Akay et al (2021), (Parolisi & Boda 2018). However, OPCs are unique among all classes of glia in their ability to receive functional synaptic innervation from both inhibitory and excitatory neurons, a paradigm-shifting discovery that emerged from the labs of Dwight Bergles and Craig Jahr in the early 2000s(Bergles et al 2000, Lin & Bergles 2004). These neuron-OPC synapses were characterized at the functional level via acute slice electrophysiology and at the structural level by electron microscopy. Synaptic connections between neurons and OPCs are often localized to the tips of the fine OPC processes (Kukley et al 2010), and some of these inputs derive from nearby axons that also innervate neighboring neurons (Bergles et al 2010). Neuron-OPC synapses are similar to neuron-neuron synapses at the ultrastructural level in their inclusion of well-defined presynaptic and postsynaptic components. They also share many physiological features with neuron-neuron synapses, such as quantal neurotransmitter release, fast activation, long-term potentiation, and pre-pulse facilitation (Bergles et al 2000, De Biase et al 2010, Ge et al 2006, Osterstock et al 2018). Follow-up studies extended the observation of neuron-OPC synapses from the hippocampus and cortex (Lin SC 2005) (Chittajallu et al 2004) to numerous other regions of the brain, including brainstem(Müller et al 2009), corpus callosum (Kukley M 2007, Ziskin et al 2007), thalamus (Mount et al 2019), and cerebellum(Káradóttir et al 2005). Intriguingly, while most OPCs do not fire action potentials, stimulating a subset of white matter OPCs drove them to spike(Karadottir et al 2008). Altogether, the widespread observation of functional synaptic connections between neurons and OPCs across a swath of brain regions suggests that these nodes of communication are important for brain function.

While there has been much speculation on the biological relevance of synaptic connectivity from neurons to OPCs, a lack of tools for selectively manipulating neuron-OPC synapses without compromising myelination has challenged our understanding of the roles that these connections play in the brain. However, neural activity is known to exert a major influence over OPCs, and neuron-to-OPC synapses have been suggested as the conduit for such influences. For example, neural activity promotes OPC proliferation and differentiation, leading to enhanced myelination and improved brain function (Kukley et al 2010) (Gibson et al 2014) (Hughes et al 2018, Mitew et al 2018, Ortiz et al 2019). Conversely, enhancing neuronal activity either via chemogenetics or acute sensory stimulation also modifies OPC interactions with neural circuits without significantly impacting OPC differentiation (Auguste et al 2022) (Moura et al 2023). Thus, neuron-to-OPC synapses are poised to support the non-canonical functions of OPCs.

A pathological context in which neuron-to-OPC synapses are thought to be relevant is glioblastoma (GBM), a condition leading to tumor formation in the brain. OPCs have been implicated as a (if not the) major cell of origin for these deadly cancers (Lindberg et al 2009, Liu et al 2011, Monje et al 2011, Nagaraja et al 2017). While, in the healthy brain, OPCs receive synaptic input from neurons to shape brain function(Bergles et al 2010), GBM tumor cells also receive synaptic input from neurons via AMPA-type glutamate receptors. Experiments characterizing these neuron-to-GBM synaptic connections determined that neural activity, potentially via direct synaptic innervation, drives glioma invasion of the neural tissue (Gillespie & Monje 2018, Venkataramani et al 2019). The induction of electrically stimulated inward currents in glioma cells led to the proliferation of the tumor, worsening clinical outcomes (Pan et al 2021, Pan & Monje 2022). Recent work has shown that the interaction between neurons and cancer cells takes advantage of the adaptive neuronal plasticity of neuron-to-OPC synapses, with this plasticity resulting in increased tumor growth rates and reduced survival times (Taylor et al 2023). Thus, as thoroughly reviewed in (Moura et al 2022), synaptic connections between neurons and OPCs or OPC-derived tumor cells play both beneficial and detrimental roles in the brain.

OPC-to-neuron signaling

As described above, neurons communicate with OPCs through synapses. While OPCs have not been observed to form functional presynaptic inputs onto neurons, they can communicate with neurons by releasing neuromodulatory factors. For example, in direct response to glutamate exposure, α- and γ-secretases within OPCs cleave the cell-surface proteoglycan NG2 thereby releasing the ectodomain fragment into the extracellular matrix (Sakry et al 2014). In turn, blocking this process either genetically or pharmacologically impaired NMDA-mediated long-term potentiation at neuron-OPC synapses and disrupted the responses of OPCs to excitatory input in particular. Reciprocally, incubation of brain slices with the cleaved ectodomain of NG2 increased neuronal Fos expression, suggesting that NG2 cleavage is a mechanism whereby OPCs, upon the activation of their glutamatergic receptors, provide feedback to neurons to increase their level of activity. OPCs have also been shown to communicate with neurons through the release of the neurotransmitter GABA onto inhibitory neurons to enhance inhibition in an activity-dependent manner(Zhang et al 2021). While these findings are relatively new and have not yet been replicated by other groups, they suggest that OPC-derived molecular signals can mediate neural circuit physiology.

In addition to modifying circuit connectivity between neurons, another intriguing role for OPCs that was recently identified is in guiding interneurons to their positions in the cerebral cortex during embryonic brain development. Lepiemme et al. (Lepiemme et al 2022) showed that the first wave of ventrally derived OPCs (vOPCs) to emerge developmentally migrate to their positions within the cortex using blood vessels as a scaffold, while interneurons derived from the same region migrate directly through the parenchyma(Tsai et al 2016). Depletion of vOPCs in utero misregulated the migration of inhibitory neurons from the MGE toward developing cortex, leading to a decrease in the overall number of cortical interneurons and an abnormal distribution of interneurons across cortical layers that persisted at least until birth. OPCs mediate interneuron positioning by engaging unidirectional contact repulsion to repel the migrating interneurons away from the neurovasculature. This interaction requires signaling between OPC-derived semaphorin 6a/6b and the plexin A3 receptor on the surface of the interneuron(Lepiemme et al 2022). Although our understanding of OPC-neuron crosstalk is still largely in its infancy, these results suggest that OPCs may be powerful players in neuronal function in the brain via myelination-independent mechanisms.

OPCs restrict axonal arborization and engulf synapses

Phagocytosis is a tightly regulated process in which cells remove debris from their environment through engulfment and lysosomal degradation. Phagocytosis is critical to many forms of cellular remodeling and homeostasis throughout the body, including circuit development and plasticity in the brain. For example, phagocytosis is engaged by glia to remove excess neurons and other cells in utero, to prune transient synapses during postnatal development, and to facilitate synaptic plasticity and turnover during learning and memory (Schafer et al. 2012; Paolicelli et al. 2011; Morizawa et al. 2022; Wang, Yue, et al. 2020). In particular, microglia, a population of myeloid-derived brain-resident immune cells, are appreciated as the professional phagocytes of the brain in large part based upon their critical roles in synaptic pruning, a process that astrocytes are involved in as well (Damisah et al 2020, Hughes & Appel 2020, Lee et al 2020, Schafer et al 2012). However, the possibility that OPCs may similarly contribute to neural circuit development was until recently untested.

Relatively new work suggests that OPCs can finetune synaptic circuits via axon and synapse remodeling in zebrafish and mice. In zebrafish, OPCs in the optic tectum do not generate oligodendrocytes and myelin, but rather direct the innervation of the tissue by retinal axons. Ablation of OPCs in early developing zebrafish led to an increase in the arborization of retinal axons, whereas depletion of microglia did not cause a significant change (Xiao et al 2022). Similarly, when OPCs were ablated later in development, retinal axon remodeling and visual circuit function were both impaired. These findings suggest that OPCs restrict the territories of developing axons in brain regions devoid of myelin.

Consistent with the finding that OPCs can shape the organization and connectivity of axons in zebrafish optic tectum, two studies in mice revealed that OPCs in primary visual cortex phagocytose synaptic inputs in the intact mammalian brain. A combination of ex vivo immunofluorescence imaging, in vivo two-photon microscopy, and flow cytometry revealed the presence of thalamocortical axonal boutons within OPCs in layer 4 of primary visual cortex (Auguste et al 2022). Manipulating visual experience via dark-rearing and light re-exposure or altering the number of microglia in the brain pharmacologically, both led to changes in the ability of OPCs to engulf synapses. Specifically, re-exposure of mice to light following a week-long period of dark-rearing increased synapse engulfment by OPCs, while the short-term ablation of microglia dampened OPC-mediated synapse engulfment. At the ultrastructural level, an independent study used 3D ssTEM reconstructions (based upon the cubic millimeter data from MICrONS Explorer https://www.microns-explorer.org/) to demonstrate that OPC processes are filled with numerous phagolysosomes (PLs) that contain material derived from both inhibitory and excitatory synapses, suggesting that they assist in remodeling nascent neuronal circuits through the engulfment of pre- and postsynaptic neuronal specializations (Buchanan et al 2022). While Buchanan et al observed numerous neuron-neuron synapses in the process of ingestion by OPCs, neuron-OPC synapses were never observed to be in the process of engulfment. While this does not rule out the possibility that OPCs may engulf some of their own synapses, it provides evidence that OPCs might preferentially engulf neuron-neuron synapses in their vicinity. Although more work remains to be done to fully understand the process of synaptic phagocytosis by OPCs and the biological relevance of this mechanism, these early findings suggest critical roles for OPCs in establishing neural circuits in the brain.

OPCs adopt immune cell-like functions in health and disease

Data from studies of the healthy nervous system demonstrate that, although OPCs derive from the same neural progenitor as astrocytes and can in some cases differentiate into astrocytes, these two cell types are structurally distinct, likely reflecting their divergent physiological properties and roles in the brain (Baldwin et al 2024, Hughes et al 2013, Nimmerjahn et al 2005). On the other hand, while microglia are unique among glial cells in that they are bona fide macrophages derived from monocyte progenitors in the yolk sac, the morphologies of OPCs are much more closely aligned with those of microglia than with those of astrocytes in both the mouse and human brain (Fig. 1). For example, both OPCs and microglia have relatively small somata and numerous distinguishable ramified processes radiating outward from the soma and ending with fine tips containing filopodia that are highly dynamic. In fact, OPCs and microglia are so structurally similar that a recent report describing the connectomics of the human cortex based upon 3D ssTEM utilized a machine learning algorithm for cell identification that was unable to distinguish between them (Shapson-Coe et al 2024), a problem confronted in similar experiments of mouse cortex as well. If form follows function, these shared morphological traits suggest overlap between the roles of OPCs and microglia in the brain. For example, both microglia and OPCs engage in the phagocytosis of synapses as evidenced by both cell types containing large numbers of phagolysosomes in which synaptic material has been identified (Auguste et al 2022, Buchanan et al 2022). Hence, OPCs may protect the healthy brain from injury or disease in a similar fashion to microglia, providing an additional immune barrier to protect the brain in a manner that does not require immune cell chemotaxis from the periphery.

Figure 1. Myriad roles of OPCs in the brain.

OPCs play many roles in the brain beyond the production of oligodendrocytes. For example, OPCs have the ability to give rise to non-oligodendroglial lineage cells such as astrocytes and glioblastoma cells. OPCs also release neuromodulatory factors such as GABA, NG2, and Sema6 to mediate neuronal activity, function, and migration (box 1). Moreover, OPCs interact extensively with neuronal axons through multiple modalities, including the receipt of direct synaptic transmission and the engulfment of presynaptic inputs (box 2). OPCs also regulate blood-brain barrier integrity through direct interactions with pericytes (box 3). Finally, OPCs can perform immune cell-like functions, such as phagocytosing cellular debris and presenting antigen. Overall, this diversity of functions has led to a growing appreciation of OPCs as the multi-taskers of the brain.

Additional indicators that OPCs may serve in an innate immune capacity can be seen in the changes that OPCs undergo during neuroinflammatory disease. For example, research on the roles of OPCs in the context of demyelinating white matter disorders such as Multiple Sclerosis (MS) suggests that OPCs heighten their innate immune cell-like properties as a result of neural tissue breakdown. In particular, white matter pathology in humans and mouse models of MS is associated with the infiltration of T cells and their release of pro-inflammatory cytokines such as IFNγ in the brain. OPCs respond to IFNγ (and likely other inflammatory mediators) by expressing MHC Class I and II molecules and presenting antigen to immune cells in the microenvironment (Baxi et al 2015, Falcao et al 2018, Jakel et al 2019, Kirby et al 2019); reviewed in (Harrington et al 2020). OPCs also respond to neuroinflammation in a mouse model of MS by adopting a more amoeboid morphology, reflecting heightened activation similar to that observed in microglia (Nishiyama et al 1999). In turn, this increase in the immunocompetence of OPCs corresponds with a decrease in OPC differentiation, suggesting a trade-off between the canonical functions of OPCs in myelination and their non-canonical, immune cell-like roles that occur in the healthy brain but are heightened during neurodegeneration (Chew et al 2005, Chew et al 2019, Lin et al 2006, Tatsumi et al 2008). This finding raises the possibility that at least two subtypes of OPCs exist: a population that adopts an immunomodulatory profile in response to inflammation, and a population that maintains an ability to differentiate and thereby promote myelination even in the face of inflammatory challenge. Changes in the abundance of these populations over time as a result of chronic neuroinflammation (such as that seen in MS) via an increase in the number of OPCs that become immuno-reactive and a decrease in the number of OPCs capable of giving rise to oligodendrocytes could underlie deficits in remyelination that promote white matter loss.

While the contributions of OPCs to aging-related neurodegenerative conditions other than MS remain under active investigation, compelling work suggests that alterations in OPC morphology and function accompany pathological hallmarks in mouse models of Alzheimer’s disease (AD) and in human brain tissue from patients with AD (Chacon-De-La-Rocha et al 2020, Vanzulli et al 2020), extensively reviewed in (Zou et al 2023). For example, OPCs have been reported to exhibit a senescent phenotype in AD brains characterized by the upregulation of the markers p21 and p16, which is thought to be triggered by the aggregation of amyloid-β plaques. Senolytic treatment of AD mice relieved OPC senescence in the plaque microenvironment thereby alleviating cognitive deficits (Zhang et al 2019), suggesting that OPCs may represent viable therapeutic targets in AD. Thus, while microglia are widely thought to represent the immune cell repertoire of the brain, emerging data supports that OPCs can adopt similar functions, when necessary, either in the context of synaptic refinement during development or in response to inflammation in the processes of demyelinating and aging-related disease.

New directions and open questions

To more fully understand how OPCs contribute to neurological health and function through mechanisms that are distinct from myelination, several open questions remain to be addressed. First, the presence of synapses between (presynaptic) neurons and (postsynaptic) OPCs has been well-described, yet the functions of these specialized communication nodes are not clear. Innervation from neurons at these synapses could help an OPC decide when to divide or differentiate, could mediate the ability of an OPC to sculpt circuit connectivity via axonal remodeling and synapse engulfment, could promote some combination of both, or could underlie a currently unidentified OPC function entirely. Moreover, the diversity of functionally distinct cell states that OPCs adopt in the healthy brain, and the molecular cues that induce each state, remain to be thoroughly characterized. A challenge that has impeded such an analysis is the relative sparseness of OPCs in the brain, which has made it difficult to sequence a large enough number of OPCs in any given context to identify defined subgroups, though some insights from transcriptomic profiling of OPCs have been gained(Chamling et al 2021, Marques et al 2018). In addition, the ways in which OPCs interact with other neuronal and non-neuronal brain cells, including but not limited to mature oligodendrocytes and microglia, to coordinate neural circuit development, plasticity, and function remain to be fully described. For example, (Auguste et al 2022) found that depleting microglia from the brain during a critical period of visual circuit development significantly hindered the ability of OPCs to engulf synapses, suggesting that microglia-OPC crosstalk is important for synaptic refinement. This observation raises the possibility that microglia may ‘recruit’ a subpopulation of OPCs to shape circuit connectivity in brain regions in which synaptic remodeling is particularly active, such as during postnatal development. Reciprocally, when the brain parenchyma becomes inflamed such as during MS or AD, microglia may recruit OPCs to assist in the protection of the surrounding tissue via phagocytosis and antigen presentation. Defining how this process works could offer therapeutic insights into treating these debilitating diseases and other conditions characterized by neuroinflammation. Beyond such studies, unbiased investigations of non-canonical roles for OPCs that extend beyond both oligodendrogenesis and neural circuit development are necessary for revealing the breadth of functions that OPCs perform in the brain.

Figure 2. Ultrastructural features of neocortical OPCs.

(a) - (c) An OPC (a), a microglia (b), and an astrocyte (c) reconstructed based on serial section transmission electron microscopy (ssTEM) of mouse visual cortex. Scale bars, 20 μm. (d) Recently divided OPC with daughter cells shown in yellow and green. Individual daughter cells shown in (e) and (f). Scale bars, 15 μm. (g) Example of a neuron-to-OPC synapse. Presynaptic bouton (Pre) contacting a postsynaptic OPC branch (Post). Scale bar, 0.5 μm. (h) Organization of synapses onto the first daughter cell (pink). (i) Organization of synapses onto the second daughter cell (pink). (h) and (i) Scale bars, 15 μm.