Neural stem cells of the subventricular zone: A potential stem cell pool for brain repair in Parkinson’s disease

Summary

Parkinson’s disease is a neurodegenerative disease caused by the degeneration of dopaminergic neurons in the substantia nigra. There are no curative treatments, and therefore, there is an urgent need for new approaches. One potential strategy being investigated is stem cell-based approaches to replace lost neurons, by, for example, harnessing endogenous neural stem cells (NSCs). These cells are found in the subventricular zone (SVZ) aligning the lateral ventricles and remain in a dormant state in the aged and diseased mammalian brain. However, with the appropriate stimuli, NSCs can shift into an activated state, proliferate, and differentiate. In this review, we discuss how PD pathology affects the behavior of NSCs and current pharmacological strategies to boost regeneration in PD. NSCs of the SVZ could be a stem cell source for brain repair, and future studies should shed light on whether these stem cells have the potential to produce functional neuronal cells.

Graphical abstract

Verkerke and colleagues review the effect of Parkinson’s disease (PD) on adult neurogenesis and briefly highlight some of the molecular mechanisms that regulate neurogenesis in the subventricular zone in the adult mammalian brain. Finally, they discuss potential new therapeutic strategies that could boost neurogenesis in PD. These insights will help to advance the development of therapies for brain repair.

Introduction

Parkinson’s disease (PD) is the second most common neurodegenerative disease and the fastest growing neurological disorder (Dorsey and Bloem, 2018). In 2023, 8.5 million people worldwide had PD, and this number is expected to increase to 12.9 million people by 2040 (Dorsey and Bloem, 2018; WHO, 2023). The majority of PD cases are sporadic, while only 10%–15% are familial PD cases (Hop et al., 2024). Several genes have been identified that can harbor mutations, which cause monogenic PD or increase the risk of developing PD. Other important factors that can increase the risk of developing PD include aging and environmental factors, such as pesticides (Tsalenchuk et al., 2023; Deliz et al., 2024). PD is a progressive disease, clinically characterized by motor symptoms, such as tremors, bradykinesia, and rigidity, and non-motor symptoms, such as cognitive impairment, sleep disorders, and depression (Figure 1) (Jankovic, 2008; Aarsland et al., 2021). One of the main pathological hallmarks of PD is the loss of A9 dopaminergic neurons located in the substantia nigra pars compacta, which results in dopamine depletion in the striatal projections (Panicker et al., 2021). Another pathological hallmark is the presence of cytoplasmic Lewy bodies containing aggregates of alpha-synuclein (aSyn) (Spillantini et al., 1997; Panicker et al., 2021). Physiologically, aSyn is expressed throughout the brain where it is involved in synaptic transmission (Panicker et al., 2021). Pathological accumulation of aSyn was initially thought to start in the dorsal motor nucleus of the glossopharyngeal and vagal nerves in the brainstem and in the anterior olfactory nucleus, from which it would spread to the rest of the brain (Braak et al., 2003; Jellinger, 2003). However, the extent of aSyn burden does not always correlate with clinical symptoms, which has led to alternative hypotheses on the origin and spreading of pathological aSyn accumulation and the involvement of co-pathologies (Parkkinen et al., 2005; Horsager and Borghammer, 2024). Non-motor symptoms are associated not only with deficits in the dopaminergic system but also with deficits in other neurotransmitter systems, such as the cholinergic and adrenergic system (Schapira et al., 2017; Parmar et al., 2020).

Figure 1.

Clinical and neuropathological features of Parkinson’s disease

Parkinson’s disease is characterized by motor and non-motor symptoms. The disease is caused by a progressive loss of dopaminergic neurons in the substantia nigra, which leads to dopamine deficiency in the nigrostriatal pathway. Other neuropathological hallmarks are the intracellular Lewy bodies containing alpha-synuclein and a neuroinflammatory response mediated by microglia and astrocytes. Created with BioRender.com.

Despite the identification of several risk factors and pathological hallmarks, no curative treatment is yet available. Current treatment of PD mainly focusses on the restoration of dopamine tone through the administration of levodopa, a precursor of dopamine. Another treatment option is deep brain stimulation of specific brain regions to modulate the output of the basal ganglia to restore motor function. Additional treatment of non-motor symptoms, for example with antidepressants, can improve the patient’s quality of life (Parmar et al., 2020; Poewe et al., 2017). However, current treatment options cannot stop or reverse disease progression nor fully restore brain function. Therefore, there is an urgent need for new therapeutic strategies that provide genuine repair. One promising option is to replace lost neuronal cells through stem cell-based therapies (Parmar et al., 2020).

Stem cell-based therapies for PD

Due to the characteristic loss of a specific neuronal population in a localized brain region, PD has been regarded for quite some time as a prime target disease for cell replacement therapy. This began with the surgical stereotactic injection of human fetal stem cells (FSCs) in the striatum of two patients in 1989 (Lindvall et al., 1989), which resulted in modest clinical improvement in these patients. Importantly, this was the first proof-of-concept study showing the feasibility of transplantation therapies. In the past decades, more preclinical and clinical studies have been performed with various sources of stem cells, such as embryonic stem cells (ESCs) and induced pluripotent stem cells (iPSCs) (Wang et al., 2018; Kirkeby et al., 2023; Song et al., 2020; Schweitzer et al., 2020; Takahashi, 2020). Advances in the use of autologous iPSC-derived cells for transplantations circumvent the ethical concerns regarding FSCs and ESCs and eliminate the need for immunosuppressants. This further enabled the development of new dopaminergic cell replacement therapy, which has been reviewed recently (Parmar et al., 2020).

Another potential stem cell-based approach that is being investigated for different neurodegenerative diseases and traumatic brain injury is the recruitment of endogenous stem cells to repair the brain. Neural stem cells (NSCs) can be found in neurogenic niches in the adult mammalian brain and remain in both the aged and diseased human brain (Donega et al., 2019; Van Den Berge et al., 2011; Leonard et al., 2009). Therefore, these stem cells could be a potential source for brain repair. However, NSCs remain mostly in a quiescent state in the adult brain, that is, they do not proliferate or give rise to new neurons. Thus, exploring therapeutic strategies that can re-activate endogenous NSCs and stimulate them to produce new functional neurons to replace lost neuronal cells could prove beneficial for the treatment of PD. Indeed, endogenous NSCs could be a stem cell source to repair different brain regions and different neuronal and glial cell populations and would be invaluable to promote both motor and cognitive recovery in PD where also the adrenergic and cholinergic systems are affected.

In this review, we will briefly discuss some molecular mechanisms that regulate neurogenesis in the adult mammalian brain. Subsequently, the effect of PD pathology on the behavior of NSCs will be addressed. Finally, we will discuss current pharmacological strategies to stimulate neurogenesis to boost regeneration in PD, focusing on the generation of dopaminergic neurons. These insights will help to advance the development of new therapeutic approaches by paving new avenues to boost the capacity of the adult human brain to repair.

Neural stem cells in the adult SVZ: Balancing quiescence and activation

There are two neurogenic niches in the adult mammalian brain: the subgranular zone in the dentate gyrus of the hippocampus and the subventricular zone (SVZ) aligning the lateral ventricles (Morrens et al., 2012). The SVZ is composed of NSCs, progenitor cells committed to different cell lineages (i.e., neural progenitor cells [NPCs] and oligodendrocyte progenitor cells), neuroblasts, niche astrocytes, microglia, and oligodendrocytes (Figure 2) (Doetsch et al., 1997, 1999; Morrens et al., 2012; Quinones-Hinojosa et al., 2006).

Figure 2.

Schematic overview of the human SVZ

(A) Sagittal view of the human brain with the lateral ventricle in green, the striatum in purple , and the substantia nigra in orange.

(B) Coronal section of the human brain at the level of the basal ganglia with a detailed overview of the subventricular zone. A layer of ependymal cells aligns the lateral ventricle, which is followed by a hypocellular gap devoid of cell nuclei. The subventricular zone consists of neural stem cells (NSCs), neural progenitor cells (NPCs), oligodendrocyte progenitor cells (OPCs), neuroblasts, astrocytes, microglia, oligodendrocytes, and blood vessels (BVs). Created with BioRender.com.

In rodents, NSCs from the adult SVZ were shown to arise from a population of slow-cycling/quiescent embryonic progenitors, which emerge between E13.5 and E15.5 in the ganglionic eminence (Fuentealba et al., 2015; Furutachi et al., 2015). This is likely also the case in the human SVZ, where quiescent embryonic progenitors could accumulate and give rise to the NSC population in the adult SVZ. How the switch from an activated to a quiescent state is induced is not yet clear. Cyclin-dependent kinase inhibitor 1C (P57) expression was shown to play an important role in regulating NSC quiescence (Furutachi et al., 2013). A recent study showed that the expression of Notch and its effector Hey1 is upregulated following cell-cycle arrest in slow-cycling embryonic progenitors and is necessary for their long-term maintenance (Harada et al., 2021). Another study demonstrated that vascular cell adhesion molecule 1 (Vcam1) expression by slow-cycling NSCs of the embryonic brain plays an important role in regulating quiescence and that its loss reduces the postnatal NSC pool (Hu et al., 2017). Interestingly, this quiescent state is reversible, and in the adult rodent SVZ, quiescent NSCs re-activate and differentiate into neurons that integrate into the olfactory bulb network under homeostatic conditions (Furutachi et al., 2015; Calzolari et al., 2015). In contrast, in the human brain, quiescent NSCs give rise to a few calretinin interneurons that migrate to the striatum (Ernst et al., 2014). A more detailed review of NSC identity and adult neurogenesis can be found in the study by Obernier et al. (Obernier and Alvarez-Buylla, 2019).

Quiescent NSCs can be distinguished from activated NSCs by their lack of proliferation markers and lack of epidermal growth factor receptor (Egfr) and Nestin expression (Figure 3). While activated NSCs are neurogenic in vivo and produce neurospheres in vitro, quiescent NSCs are slowly cycling and remain largely dormant in vivo, where, in rodents, they give rise to olfactory bulb neurons although with slower kinetics (Codega et al., 2014). Recent studies show that quiescent and activated NSCs also differ in their metabolic state (recently reviewed by Scandella et al. (2023)). Single-cell RNA sequencing highlighted the heterogeneity of NSCs revealing a gradient from quiescence to activation (Llorens-Bobadilla et al., 2015).

Figure 3.

Differences between active and quiescent neural stem cells

With aging, neural stem cells (NSCs) transition from an active to a quiescent state. Active NSCs are characterized by proliferation, activation of the Egfr and Wnt pathways, expression of Egfr and Nestin, and a highly active lysosomal pathway. In contrast, quiescent NSCs do not proliferate, and the Bmp-4 and Notch signaling pathways are activated. Quiescent NSCs lack the expression of Egfr and Nestin but do express the cell cycle inhibitor P57. Additionally, aging reduces the activity of the lysosomal pathway in quiescent NSCs. Egfr, epidermal growth factor receptor; Bmp-4, bone morphogenetic protein 4; Fgf-2, fibroblast growth factor 2. Created with BioRender.com.

Understanding the mechanisms that regulate NSC quiescence and re-activation is crucial for the development of strategies to boost neurogenesis in the aged brain. In tissues other than the brain, quiescent stem cells re-activate following injury. For instance, in the bone marrow, hematopoietic stem cells switch between quiescence and activation under physiological conditions and following injury (Mendelson and Frenette, 2014). Muscle stem cells enter a so-called pre-proliferative, reversible “alert state” after injury, enhancing their capacity to participate in muscle regeneration (Rodgers et al., 2014). It would be interesting to determine if distinct subpopulations of NSCs exist that are only active following injury (Pietras et al., 2015). Indeed, recent studies suggest that quiescent NSCs are more responsive to injury signals, possibly because of their higher expression of membrane receptors, which may make them more sensitive to changes in the environment (Codega et al., 2014; Llorens-Bobadilla et al., 2015; Morizur et al., 2018; Donega and Raineteau, 2017).

Several intrinsic and extrinsic mechanisms have been identified that control NSC quiescence and activation (Figure 3). NSC quiescence increases with aging, which correlates with an increase in inflammatory signaling and inhibition of the Wnt pathway in the SVZ (Kalamakis et al., 2019). Indeed, the Wnt pathway inhibitor secreted frizzled-related protein 5 (Sfrp5) was increased in NSCs of the aged SVZ (Chavali et al., 2018). Interestingly, the authors showed that quiescent NSCs are amenable to manipulation to re-activate and proliferate through activation of the Wnt pathway after sequestration of Sfrp5. A downstream target of the non-canonical Wnt signaling, Rho-GTPase Cdc42, maintains NSCs in a quiescent state (Chavali et al., 2018). Both Notch2 and Notch3 were shown to repress cell cycle progression genes, thereby promoting stem cell quiescence (Engler et al., 2018; Kawai et al., 2017). Bone morphogenetic protein 4 (Bmp-4) signaling induces a quiescent NSC state, while the combination of Bmp-4 and fibroblast growth factor 2 (Fgf-2) signaling promotes a so-called primed-quiescent state when NSCs are preparing to activate. These primed-quiescent NSCs express high levels of leucine-rich repeats and immunoglobulin-like domains 1, which when inhibited results in increased NSC proliferation, thereby regulating NSC exit from quiescence (Marques-Torrejon et al., 2021). Expression of active Egfr was shown to promote NSC proliferation through phosphatidylinositol 3-kinase/Akt, Mek/Erk, and mechanistic target of rapamycin (mTor) signaling (Cochard et al., 2021). Another mechanism that regulates NSC activation is through increased expression of miR-17-92 that promotes NSC proliferation and neurogenesis instead of oligodendrogenesis (Favaloro et al., 2022). One cell-intrinsic mechanism is through lysosome activation, which is necessary for NSCs to exit quiescence. However, with aging, the activation of lysosomes decreases, impairing the ability of NSCs to re-enter the cell cycle (Figure 3) (Leeman et al., 2018). Indeed, enhanced lysosomal degradation was shown to maintain NSCs in a quiescent state (Kobayashi et al., 2019). For a more thorough review on the intrinsic and extrinsic mechanisms regulating NSC activation and quiescence, please see the study by Urban et al. (2019).

Neurogenesis in PD

To be able to boost neurogenesis in PD, the first question that needs to be addressed is whether this NSC pool remains in the diseased PD brain. Importantly, studies showed that not only are NSCs still present in the SVZ of patients with PD but are also not depleted and can be easily detected in the SVZ. Indeed, no significant differences were found in the number of NSCs between PD and age-matched non-demented controls (Van Den Berge et al., 2011; Donega et al., 2019; Leonard et al., 2009). One also needs to keep in mind when working with postmortem human brain tissue that we are looking at a snapshot of the response of NSCs to PD. We are investigating the end stage of the disease and therefore cannot extrapolate our findings to early disease stages. We know that NSCs are still present in the SVZ of patients with PD, but whether this pool would be large enough to replenish lost dopaminergic cells and improve motor function needs to be investigated. NSCs have the potential to self-renew, and therefore, the possibility that NSCs could be stimulated to self-renew to increase the stem cell pool, if properly stimulated, cannot be excluded. More research is needed to fully understand the biology of NSCs of both the aged and diseased human SVZ.

Pathological hallmarks and multiple genetic factors of PD were shown to affect the proliferation and differentiation capacity of NSCs (Figure 4) (Planas-Ballvé and Vilas, 2021). For the successful development of a therapeutic approach that stimulates neurogenesis, the effect of pathology and gene mutations on NSC behavior needs to be determined. In the following section, we will discuss the current knowledge on how NSC behavior is affected by dopamine depletion and gene mutations in rodent models for PD. As every model, rodent models for PD have strengths and limitations. These models were shown to recapitulate some key hallmarks of PD, such as degeneration of dopaminergic neurons in the substantia nigra and motor impairment. Pesticides and neurotoxins known to induce PD in humans also induce Parkinson-like phenotype in rodents. These models also have limitations, such as the lack of Lewy body formation.

Figure 4.

Pathological hallmarks of Parkinson’s disease that affect adult neurogenesis

(A) NSCs express both D2 and D3 dopamine receptors. Dopamine stimulates NSC proliferation, which is partially mediated via Egf. Degeneration of dopaminergic neurons in the substantia nigra leads to dopamine deficiency in the nigrostriatal pathway. Rodent models of dopaminergic neuron loss show conflicting results on the proliferative capacity of NSCs. This apparent contradiction could be explained by differences in the severity of neuronal loss, BrdU paradigm used, how long after induction of neuronal loss the proliferative capacity of NSCs was assessed, and the region of the SVZ that was analyzed. It has yet to be determined whether these pathological hallmarks also affect adult neurogenesis in patients with PD.

(B) Three genes that carry mutations that can increase the risk of developing PD are LRRK2, PARK2, and VPS35. LRRK2 interacts with several components of the Wnt/β-catenin pathway, and mutations in LRRK2 have been shown to decrease the proliferative capacity and differentiation capacity of NSCs. VPS35 regulates Wnt signaling and decreases LRRK2 kinase activity and might therefore affect proliferative and differentiation capacity of NSCs. PARK2 is an E3 ubiquitin ligase, and two of its targets are p21, a cell cycle inhibitor, and β-catenin, a component of the Wnt pathway. It is likely that mutations in PARK2 affect the proliferative capacity of NSCs. However, this has yet to be determined.

(C) Accumulation of monomeric alpha-synuclein into fibrils is a key hallmark of PD. In vivo, overexpression of wild-type alpha-synuclein leads to the reduction in differentiation capacity of NSCs in the rodent SVZ, while the pathological A53T also affects the proliferative capacity of NSCs. In vitro, it has been shown that this effect is mediated through the Notch pathway. LRRK2, leucine-rich repeat kinase 2; PARK2, Parkin RBR E3 ubiquitin protein ligase 2; Egf, epidermal growth factor. Created with BioRender.com.

Dopamine depletion

The characteristic degeneration of dopaminergic neurons in the substantia nigra results in dopamine deficiency in the nigrostriatal pathway. The SVZ receives dopaminergic input from the midbrain, and NSCs in the SVZ also express dopamine receptors D2 and D3, and therefore, dopamine depletion could affect NSC behavior (Winner et al., 2009; Coronas et al., 2004; Hoglinger et al., 2004; Freundlieb et al., 2006). However, contradictory findings have been reported on the effect of dopamine deficiency on NSC proliferation in the SVZ of preclinical PD models (Van Den Berge et al., 2013). In the preclinical 6-hydroxydopamine (6-OHDA) PD model, the neurotoxin 6-OHDA is injected in the medial forebrain bundle or the striatum where it is taken up by dopaminergic neurons and noradrenergic neurons. The toxin causes progressive neurodegeneration and thereby depletion of dopamine in the nigrostriatal pathway in the injected side, although there is no accumulation of aSyn (Dovonou et al., 2023). Most studies that inject 6-OHDA in the medial forebrain bundle find a reduction in the number of proliferating cells, either BrdU or Pcna⁺, in the SVZ of the lesioned side compared to the contralateral side (Chiu et al., 2014; Singh et al., 2018; Winner et al., 2008a; O'Keeffe et al., 2009). However, Liu and colleagues show an increase in BrdU⁺ cells in the SVZ at 2 weeks after 6-OHDA injection in the medial forebrain bundle, but this effect disappears 4 weeks after injection (Liu et al., 2006). In contrast, Aponso and colleagues injected 6-OHDA in the striatum and found an increase in the number of BrdU⁺ cells in the SVZ even before substantial loss of tyrosine hydroxylase positive (Th+) neurons in the substantia nigra was observed (Aponso et al., 2008). In another preclinical PD model, the neurotoxin 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) is administered systemically, and once it has crossed the blood-brain barrier, it is converted by glial cells into the toxic metabolite MPP⁺. This is subsequently taken up by dopaminergic neurons leading to dopaminergic neuron loss (Dovonou et al., 2023; Zhao et al., 2003). Höglinger and colleagues show an acute decrease in Pcna⁺ cells in the SVZ 2 days after MPTP administration, while Peng and colleagues show an increase in BrdU⁺ cells in the SVZ at 5 and 14 days after MPTP administration (Figure 4A) (Hoglinger et al., 2004; Peng and Andersen, 2011; Peng et al., 2008).

These apparently contradictory findings could be explained by several differences in study design. First, the injection site of 6-OHDA seems to be important and might correlate to the severity of dopaminergic neuron loss as the injection into the medial forebrain bundle results in rapid neurodegeneration, while injection in the striatum leads to more progressive neurodegeneration (Dovonou et al., 2023). Second, the timing of BrdU injection is also an important aspect, since NSC proliferation is affected differently as dopaminergic neuron loss progresses. In this regard, it is interesting that in human postmortem brain tissue, no differences have been found in the proliferative capacity of NSCs in the SVZ of patients with PD compared to controls (Van Den Berge et al., 2011). Thus, the changes in proliferative capacity might reflect an adaptive response that occurs early on during disease onset and disappears as the disease progresses. Finally, how the number of proliferating cells is determined may result in different findings regarding the proliferative capacity of NSCs. BrdU is incorporated in the DNA of actively dividing cells and depends on the administration regime and timing of BrdU administration, while staining for Pcna will identify cells that are in the cell cycle at that time, regardless of cell cycle phase. Finally, the SVZ shows spatial heterogeneity in both lineage commitment and proliferative capacity (Donega and Raineteau, 2017; Azim et al., 2016). It can be subdivided into three walls, the dorsal SVZ that gives rise to both glutamatergic and GABAergic progenitors and oligodendrocytes, while the lateral and medial SVZ give rise to GABAergic progenitors and oligodendrocytes (Brill et al., 2009; Delgado et al., 2021; Winpenny et al., 2011; Fiorelli et al., 2015; Young et al., 2007; Merkle et al., 2007). Indeed, Th+ cells have been shown to be mainly produced by the dorsal SVZ wall (Fernandez et al., 2011). Most studies, however, do not report the region of the SVZ that was assessed. As discussed, analysis of different SVZ regions could lead to contradictory findings.

Additionally, other pathological processes, besides dopamine deficiency, are mimicked in the preclinical PD models, making it difficult to conclude whether dopamine deficiency alone affects NSC proliferation. Nevertheless, D2 receptor stimulation via intraperitoneal injection of the dopamine agonist quinpirole induced proliferation in the SVZ of adult wild-type mice demonstrated by an increase in the number of BrdU⁺ cells (Yang et al., 2008). Moreover, in vitro treatment of adult mouse NSCs with dopamine also increased the number of BrdU⁺ cells, and this effect was mediated via epidermal growth factor (Egf) (O'Keeffe et al., 2009). Dopamine depletion is one of the main pathological hallmarks of PD. Several genes have been identified that cause PD or increase the risk of developing PD in specific subgroups of patients. These PD-related genes and their potential effect on neurogenesis will be discussed further.

PARK8/LRRK2 gene

Mutations in the leucine-rich repeat kinase 2 (LRRK2) gene are the cause of 5% of the familial PD cases and are also present in 1% of the idiopathic PD cases (Planas-Ballvé and Vilas, 2021). In the mammalian brain, LRRK2 is highly expressed in the SVZ (Melrose et al., 2007). It is an enzyme involved in autophagy and endocytosis and has been implicated in several signal transduction pathways, among which is the canonical Wnt/β-catenin pathway (Habig et al., 2008; Berwick et al., 2017). Wild-type LRRK2 inhibits the Wnt/β-catenin pathway via interaction with low-density lipoprotein (LDL) receptor-related protein 6 and glycogen synthase kinase-3 beta (GSK-3β). Several PD-related mutations in LRRK2 increase its kinase activity. Moreover, a newly identified risk gene, RAB32, was shown to interact with LRRK2 and increase its kinase activity (Hop et al., 2024; Gustavsson et al., 2024). This increased kinase activity could alter the activity of the Wnt/β-catenin pathway (West et al., 2005). As the Wnt/β-catenin pathway is important for NSC activation and differentiation to dopaminergic neurons, the altered activation of the pathway could potentially influence adult neurogenesis (Marchetti et al., 2020). The most prevalent LRRK2 mutation, G2019S, has been shown to increase Lrrk22 kinase activity and decrease proliferation in the SVZ in young adult transgenic mice (4 months old) (Winner et al., 2011; Berwick et al., 2017). A similar effect has been found in vitro, where the G2019S and R1441G mutations decreased the clonal expansion and differentiation capacity of NSCs (Figure 4B) (Liu et al., 2012; Bahnassawy et al., 2013). However, Wetzel and colleagues showed an increase in active β-catenin and thus increased Wnt signaling in adult (7 months old) G2019S transgenic mice as well as in Lrrk2 knockout mice. At protein level, they showed that components of the Wnt signaling pathway changed in a region-dependent manner in the G2019S mice. For example, the ratio of active β-catenin to total β-catenin increased in the cortex but decreased in the striatum and remained unchanged in the hippocampus. Thus, the Wnt signaling pathway might be affected differently in neurogenic niches, such as the SVZ and hippocampus, when compared to the cortex (Wetzel et al., 2024).

Lrrk22 not only affects neurogenesis via the Wnt pathway but also through the regulation of several microRNAs (miRNAs). Mutated Lrrk22 inhibits the activity of different miRNAs (Gehrke et al., 2010; Gonzalez-Cano et al., 2018; Bahnassawy et al., 2013). Among these miRNAs is Let-7a, which is involved in NSC differentiation. Mutated Lrrk2 inhibits neuronal differentiation via Let-7a, which is mediated through its kinase activity (Bahnassawy et al., 2013; Gehrke et al., 2010). An in vitro study showed that pharmacological inhibition of Lrrk2 kinase activity leads to increased proliferation of adult murine NSC-containing neurospheres (Salado et al., 2017). Moreover, Lrrk2 inhibitors are being currently assessed in clinical trials for their safety and tolerability and their effect on slowing down PD symptom progression (Jennings et al., 2023). Overall, a growing number of studies suggest a role for Lrrk2 in adult neurogenesis and its potential as a therapeutic target to boost adult neurogenesis.

PARK17/VPS35 gene

The PARK17/VPS35 gene encodes for vacuolar protein sorting 35 (VPS35) and was identified as a cause of autosomal-dominant familial PD (Vilarino-Guell et al., 2011; Zimprich et al., 2011). VPS35 is a subunit of the retromer, a protein complex mediating endosome-Golgi trafficking, which regulates Wnt signaling among other signaling pathways (Belenkaya et al., 2008; Coudreuse et al., 2006). The PD-related mutation D620N in VPS35 causes a redistribution and enlargement of retromer-containing endosomes (Follett et al., 2014). Mice with the D620N mutation exhibit late-onset (at 16 months) dopaminergic neuron loss in the substantia nigra and reduced protein expression of Wnt1 and β-catenin (Chiu et al., 2020). The stem cell pool in the SVZ of D620N mice has not yet been investigated, but in another neurogenic niche, the hippocampus, it was shown that these mice exhibit a reduced pool of Sox2⁺ cells and proliferating cells (either Ki67⁺ or BrdU⁺) at 3 months (Jiang et al., 2021). The authors also showed a decrease in Dcx⁺ BrdU⁺ cells and Neun⁺ BrdU⁺ cells in the dentate gyrus of the hippocampus suggesting impaired proliferation, differentiation, and/or cell survival. The D620N mutation has also been linked to increased Lrrk2 kinase activity, indicating a converging pathophysiology with PD-associated mutations in LRRK2 (Bu et al., 2023).

PARK2 gene

Another PD-related gene is the Parkin RBR E3 ubiquitin protein ligase (PARK2), which is mutated in 15% of the familial PD cases and in 4% of the idiopathic PD cases (Planas-Ballvé and Vilas, 2021). PARK2 encodes for Parkin, which is an E3 ubiquitin ligase involved in the proteasome- and autophagy-mediated degradation pathway, particularly of mitochondria. PD-related mutation in PARK2 results in the loss of ubiquitin ligase activity (Shimura et al., 2000). Physiologically, Parkin interacts with β-catenin targeting it for degradation, thereby inhibiting Wnt signaling (Rawal et al., 2009). Another target of Parkin is the cyclin-dependent kinase inhibitor 1A (p21), which is involved in the maintenance of the NSC pool (Figure 4B) (Park et al., 2017; Marques-Torrejon et al., 2013). Thus, the targets of Parkin suggest that it could play a role in neurogenesis, and mutation-induced loss of its ligase activity might affect neurogenesis. However, current studies on the effect of mutated Parkin on neurogenesis show contradictory findings and remain inconclusive (Le Grand et al., 2015). It has yet to be determined how specific mutations in the Park2 gene affect adult neurogenesis.

PARK1/SNCA gene

Wild-type aSyn

The SNCA gene encodes for aSyn, and mutations as well as multiplications of this gene are associated with familial PD (Planas-Ballvé and Vilas, 2021). Viral overexpression of wild-type human aSyn in fetal mouse NSCs in vitro led NSCs to exit the cell cycle and to a reduction in the formation of secondary neurospheres (Tani et al., 2010). However, overexpression with the same viral vector in vivo did not affect the number of proliferating NSCs as there were no changes in the number of BrdU⁺ cells in the SVZ (Tani et al., 2010). In line with the in vivo data of Tani et al., overexpression of wild-type human aSyn under the platelet-derived growth factor (Pdgf) promoter did not change the number of Pcna⁺ cells in the adult (3 months old (Winner et al., 2004)) or aged (15 months old (Winner et al., 2008b)) mouse SVZ. Several studies report a decrease in newborn neurons (BrdU⁺/NeuN⁺ or BrdU⁺/Th⁺) in the olfactory bulb suggesting decreased neurogenesis in the olfactory bulb of mice overexpressing wild-type human aSyn (Tani et al., 2010; Winner et al., 2004; May et al., 2012). Thus, overexpression of wild-type human aSyn does not affect the number of stem cells in the SVZ but instead decreases the number of newborn neurons in the olfactory bulb.

Mutated aSyn

Mouse models bearing mutant forms of human aSyn are valuable preclinical models to study the effect of pathological aSyn on adult neurogenesis. A preclinical model that recapitulates the accumulation of aSyn is the A53T aSyn mouse model, in which the A53T mutant form of human aSyn is expressed under the mouse prion protein promoter. In this model, the mutant aSyn is expressed six times higher than the endogenous mouse aSyn and aggregates in typical inclusion-like structures mimicking the progressive nature of PD. The number of proliferating NSCs, either Pcna⁺ or BrdU⁺, was reduced in the SVZ of adult (8 months old) and aged (15 months old) A53T aSyn mice, while the total number of Sox2⁺ NSCs remained comparable to wild-type mice. This suggests that aSyn accumulation does not lead to NSC depletion (Fuchigami et al., 2023; Winner et al., 2008b). Similar to wild-type aSyn, the overexpression of mutant A53T aSyn led to reduced neurogenesis in the olfactory bulb although this was more pronounced in the A53T mice than in wild-type aSyn (Winner et al., 2008b). Another transgenic mouse model overexpressing the mutant A30P aSyn also reported a reduction in olfactory bulb neurogenesis (Zhang et al., 2019; Neuner et al., 2014; Marxreiter et al., 2009; Martin-Lopez et al., 2023). However, the findings on NSC proliferation in the A30P mice are contradictory as some studies report a reduction in proliferation (Zhang et al., 2019), while other studies show no difference in the number of proliferating NSCs (Marxreiter et al., 2009; Neuner et al., 2014).

Overexpression of aSyn, either wild-type or A53T, in adult rat NPCs in vitro decreased the activity of the Notch pathway by downregulating Notch1 and its downstream target Hes5, which is a signaling pathway involved in stem cell quiescence and activation (Desplats et al., 2012; Crews et al., 2008). The downregulation of Notch1 was mediated by p53, a tumor suppressor gene that can inhibit cell cycle progression and induce apoptosis (Meletis et al., 2006; Desplats et al., 2012).

Altogether, most studies reported a decrease in neurogenesis in the olfactory bulb and either an increase or decrease in NSC proliferation in the SVZ (Figure 4C). These differences might arise from the use of different aSyn variants, generic or cell type-specific overexpression, differences in the age of the animals, assessing proliferation at different time points after PD induction, or differences in the onset and severity of dopaminergic neuron loss.

Other genetic factors

Lastly, there are some studies that suggest that other genes with known PD-related mutations, such as glucosylceramidase beta 1, PTEN-induced kinase 1, and Parkinsonism-associated deglycase, can affect proliferation or differentiation of NSCs and regulate the Wnt pathway (Agnihotri et al., 2017; Awad et al., 2017; Sun et al., 2018). However, more research is needed on specific PD-related mutations to clarify the potential effect of these mutations on neurogenesis. An in-depth overview of the effect of PD-associated mutations on neurogenesis is provided by Le Grand and colleagues (Le Grand et al., 2015).

In conclusion, PD-related mutations as well as pathological hallmarks of PD affect the proliferative capacity of NSCs in the SVZ. Importantly, a growing number of studies show that the stem cell pool in the SVZ is not depleted in PD. This opens up opportunities for boosting endogenous neurogenesis as a potential therapeutic approach, which will be discussed in the next section.

Boosting neurogenesis in PD

It has been shown in rodent models that quiescent NSCs can re-activate if properly stimulated. For example, an increase in BrdU⁺ and Th⁺ cells was observed in the substantia nigra of the MPTP mouse model compared to wild-type mice (Zhao et al., 2003). Therefore, it is important to look further into different approaches that could be used to boost the regenerative capacity of NSCs from the SVZ, including the use of dopamine agonists, small molecules, growth factors, and lipids. In the following section, we discuss the efficacy of different approaches being studied to re-activate NSCs and stimulate neurogenesis in the SVZ of rodent PD models (Figure 5; Table S1).

Figure 5.

Strategies to boost neurogenesis as a therapeutic approach for Parkinson’s disease

The capacity of NSCs to proliferate and self-renew can be stimulated through different strategies (left). Growth factors such as Cdnf, Lgf, and Fgf together with Egf enhance proliferation of NSCs in the SVZ. P21 is a negative regulator of the cell cycle, and intranasal administration of GD3 inhibits p21 thereby stimulating proliferation. As NSCs express several dopamine receptors, increasing the dopamine concentration in the brain with medication, such as quinpirole, pramipexol, levodopa, or selegiline was shown to stimulate proliferation of NSCs. For therapeutic relevance, NSCs need to differentiate into dopaminergic neurons (right). The Wnt pathway is important for neuronal development, and stimulating Wnt signaling via inhibition of Gsk-3β can stimulate NSC self-renewal as well as differentiation. Ganglioside GM1 improves NSC differentiation through Nurr1 and Gdnf. Besides Gdnf, Bdnf, another growth factor, also stimulated differentiation. Cdnf, cerebral dopamine neurotrophic factor; Lgf, liver growth factor; Fgf, fibroblast growth factor ; Egf, epidermal growth factor; p21, cyclin-dependent kinase inhibitor 1A; Gsk-3β, glycogen synthase kinase-3 beta; Nurr1, nuclear receptor-related 1; Gdnf, glial cell-derived neurotrophic factor; Bdnf, brain-derived neurotrophic factor. Created with BioRender.com.

Dopamine agonists promote NSC self-renewal

Current treatment for PD is focused on restoring dopamine levels in the nigrostriatal pathway by administering levodopa or dopamine agonists. Besides alleviating the motor symptoms, these therapies may also increase neurogenesis. The reduction in NSC proliferation in the lesioned side of 6-OHDA mice can be rescued by administration of several dopamine agonists, such as pramipexol (D₂/D₃ receptor) and quinpirole (D₂ receptor) (Winner et al., 2009; Yang et al., 2008). Additionally, increasing dopamine levels by the systemic administration of levodopa or inhibiting the degradation of dopamine by the systemic administration of the monoamine oxidase inhibitor selegiline also increased the number of Pcna⁺ cells in the lesioned side to similar levels as the contralateral side (Chiu et al., 2014). However, another study showed that dopamine agonists or dopamine precursors did not increase the number of proliferating cells in the SVZ of a 6-OHDA rat model (Marin et al., 2014). This apparent contradictory result could be explained by differences in the administration method, that is, pramipexol was injected either subcutaneously (Marin et al., 2014) or orally (Winner et al., 2009). Moreover, Marin et al. administered BrdU before starting treatment while Winner et al. administered BrdU during treatment, which might also contribute to discrepancies in the results as they could be labeling different populations of proliferating NSCs (Table S1).

Future studies should determine whether proliferating NSCs in the SVZ can differentiate into new dopaminergic neurons in the striatum. It was shown that after withdrawal of the dopamine agonist pramipexol, the number of proliferating cells in the SVZ decreased to the level of untreated 6-OHDA rats, while the number of BrdU⁺ and Th⁺ cells increased in the olfactory bulb but not in the substantia nigra or striatum (Winner et al., 2009). The discrepancy with the study of Zhao and colleagues, who found an increase in BrdU⁺ and Th⁺ cells in the substantia nigra of MPTP mice even in the absence of any pro-neurogenic treatment, might be attributed to a difference in timing and duration of BrdU administration, severity of neuronal loss, or species-dependent differences.

Activation of the canonical Wnt pathway promotes neurogenesis

As discussed before, the Wnt/β-catenin pathway plays an important role in regulating adult neurogenesis, and thus, members of this pathway could be potential therapeutic targets to boost neurogenesis (Marchetti et al., 2020). Increased levels of activated Gsk-3β have been shown in PD rodent models as well as in patients with PD suggesting a reduced activity of the Wnt/β-catenin pathway (Wills et al., 2010; Singh et al., 2018; L’Episcopo et al., 2011). Several studies from the Marchetti lab have shown that stimulation of the Wnt/β-catenin pathway increased neurogenesis in the SVZ. For example, intraperitoneal injection of the Gsk-3β inhibitor AR-AO14418 increased the number of dopaminergic neurons in the substantia nigra and the expression of the dopamine transporter Dat in the striatum in the MPTP mouse model (L’Episcopo et al., 2011). This correlated with an increase in Fzd1 and β-catenin expression, while Gsk-3β decreased. Similar to AR-AO14418, intraperitoneal injection of another Gsk-3β inhibitor, SB216763, has been shown to increase the number of newly generated neurons in the rostral migratory stream and striatum of the 6-OHDA mouse model (Figure 5) (Singh et al., 2018) (Table S1).

Growth factors stimulate NSC self-renewal or differentiation

Growth factors, among which are neurotrophic factors, are important for NSC self-renewal and differentiation in the developing brain. Glial cells, such as niche astrocytes, in the adult SVZ can provide these growth factors to the NSCs, and NSCs are also in contact with the growth factor-rich cerebral spinal fluid. It has been investigated in several rodent PD models whether the administration of growth factors can stimulate NSC self-renewal and/or differentiation. In a genetic mouse model for progressive dopaminergic neurodegeneration (Girk2 mutation), increasing the concentration of brain-derived neurotrophic factor (Bdnf) via daily intraperitoneal administration of compound BNN-20 from P14 to P60 weeks led to an increased number of BrdU⁺ and Th⁺ cells in the substantia nigra, while the number of Pcna⁺ cells in the SVZ remained the same. Lineage tracing with an intraventricular DiI injection at P45 confirmed that the newborn dopaminergic neurons at P60 originated from the SVZ (Table S1). This suggests that Bdnf stimulates both proliferation and differentiation of NSCs (Mourtzi et al., 2021).

Interestingly, several growth factors were shown to increase the pool of dividing cells in the SVZ of rodent PD models. Intracerebroventricular infusion of Egf, Fgf-2, liver growth factor (Lgf), or cerebral dopamine neurotrophic factor increased the number of dividing cells in the SVZ of 6-OHDA rats (Winner et al., 2008a; Nasrolahi et al., 2019; Gonzalo-Gobernado et al., 2009). Some studies reported that these BrdU⁺ NSCs were also Dcx⁺ and migrated toward the adjacent striatum (Nasrolahi et al., 2019; Winner et al., 2008a). Future studies should determine the fate of Dcx⁺ cells and whether these cells give rise to mature functional neurons in the striatum and alleviate the PD-like motor symptoms.

Gangliosides stimulate NSC self-renewal

Dysregulation of lipid metabolism in the brain is a major risk factor for multiple neurodegenerative diseases, including PD (Fanning et al., 2020). One class of lipids in the brain are gangliosides, which are sialic acid-containing glycolipids predominantly found in the outer cell membrane (Fuchigami et al., 2023). Deficiency of ganglioside GM1 has been correlated with the development of PD in mice and humans. Heterozygous knockout of GM1 in mice resulted in loss of Th⁺ neurons and accumulation of aSyn in the substantia nigra and motor impairments (Wu et al., 2012). Additionally, the concentration of several gangliosides, among which is GM1, is significantly reduced in the substantia nigra and occipital cortex of patients with PD compared to controls (Wu et al., 2012; Hadaczek et al., 2015; Ledeen et al., 2022).

The correlation between gangliosides and PD is possibly mediated by the interaction of gangliosides with aSyn, where GM1 inhibits the accumulation of aSyn and stimulates neuronal survival by increasing glial cell-derived neurotrophic factor signaling (Hadaczek et al., 2015; Martinez et al., 2007; Wu et al., 2012). Moreover, ganglioside GD3 is required for the maintenance of the quiescent NSC pool in the adult mouse SVZ (Wang et al., 2014; Wang and Yu, 2013; Fuchigami et al., 2024). The Yu lab has investigated the therapeutic potential of gangliosides in PD extensively. The intranasal administration of GD3 to transgenic A53T aSyn adult mice increased the percentage of BrdU⁺ and Sox2⁺ cells in the SVZ (Fuchigami et al., 2023). This effect could be mediated via the suppression of the cell cycle inhibitor p21. However, GD3 did not restore the number of Th⁺ neurons in the olfactory bulb. Interestingly, intranasal administration of GM1 did not affect the number of BrdU⁺ and Sox2⁺ cells but instead increased the number of Th⁺ cells in the olfactory bulb (Fuchigami et al., 2023). The effect of GM1 on dopaminergic neurogenesis could be mediated by Nurr1, a transcription factor involved in dopaminergic neuron development and maintenance through the regulation of Th expression, among other genes (Jankovic et al., 2005). The intranasal administration of GM1 increased nuclear expression of Nurr1 and its binding to the Th promoter region (Figure 5) (Itokazu et al., 2021) (Table S1). Thus, combinatorial and/or sequential administration of different gangliosides might prove beneficial to boost NSC self-renewal and their differentiation into dopaminergic neurons.

Future directions

There is evidence supporting the regenerative potential of both endogenous NSCs and pluripotent stem cell-derived neurons. Both are valid stem cell sources to study, as cell transplantation is a viable approach to more localized damages, for example the substantia nigra in PD. However, for diseases such as Alzheimer’s disease with more diffuse neuronal loss or even in PD where cortical acetylcholinergic neurons are lost, neuronal grafting would not be sufficient. Therefore, we should invest in both approaches, to ensure the development of a more personalized approach to regenerate the human brain following both localized and diffuse brain damage.

For endogenous NSCs to be an effective stem cell source to promote brain regeneration, we need to determine the permissiveness of quiescent NSCs to re-activate, their competence to produce different neuronal cell types including dopaminergic neurons, and their capacity to fully integrate into the local neuronal network. Indeed, future work needs to assess whether NSCs from the SVZ can produce functional dopaminergic neurons of the correct subtype (e.g., dopaminergic neurons of the substantia nigra or dopaminergic neurons of the olfactory bulb) that integrate into the local network and survive in the long term.

Our knowledge of NSCs in the adult and aged human SVZ is still limited due to scarcity of human brain tissue and technical limitations that are inherent to ex vivo studies. Nevertheless, recent technological advances have enabled the study of human NSCs with unprecedented resolution. Large community efforts to map NSCs of the newborn, infant, adult, and aged human brains are starting to reveal their molecular fingerprints and neurogenic potential (Puvogel et al., 2024; Donega et al., 2019, 2022; Baig et al., 2024; Bhaduri et al., 2021; Eze et al., 2021; Velmeshev et al., 2023). Moreover, the rise in single-cell atlases of the human brain provides an important map of the different neuronal subtypes, revealing an intricate network of specialized cells (Siletti et al., 2023). A better understanding of neuronal diversity is important for the development of successful regenerative therapies.

Studies show that several subtypes of dopaminergic neurons exist (La Manno et al., 2016; Kamath et al., 2022; Siletti et al., 2023). Olfactory bulb dopaminergic neurons are functionally distinct from substantia nigra dopaminergic neurons. Thus, understanding the molecular factors and temporal regulation of dopaminergic neuron specification is crucial for the development of therapeutic strategies to stimulate endogenous NSCs to produce functional dopaminergic neurons. In a study from the Studer lab, dopaminergic neurons generated from embryonic PSCs were transplanted into the rodent striatum (Kriks et al., 2011). They showed that dopaminergic neurons generated through an intermediate floor-plate progenitor state specified more efficiently into dopaminergic neurons than previous protocols that differentiated dopaminergic neurons directly from embryonic PSCs. Activation of the Wnt pathway, which induced co-expression of LMX1A in FOXA2-positive cells, improved the specification of dopaminergic neurons. Interestingly, the authors showed that activation of the Wnt pathway stimulated LMX1A expression, which is an important transcription factor for the specification of midbrain dopaminergic neurons during development (La Manno et al., 2016). This highlights the importance of understanding how different subtypes of dopaminergic neurons are specified for successful repair of the dopaminergic system.

To fully recover brain function, the newly generated neuronal cells need to migrate to the brain region where neuronal loss has taken place and functionally integrate into the local neuronal network. A previous study from the Studer lab showed that grafting human dopaminergic neurons derived from human ESCs into the striatum of PD mouse and rat models resulted in functional recovery 5 months after transplantation (Kriks et al., 2011). The maturation pace of a human neuron is determined epigenetically by the expression of a specific set of epigenetic factors (Ciceri et al., 2024). This so-called epigenetic barrier is inherent to the cell and sets the maturation pace with species specificity. Different studies using rodent models for PD showed that engrafted human dopaminergic neurons derived from human ESCs formed correct synaptic contacts with surrounding striatal medium spiny neurons and neurons from the medial prefrontal cortex. These studies showed that synaptic integration started 6 weeks after engraftment into the rodent striatum or substantia nigra (Grealish et al., 2015). An interesting example highlighting the regenerative potential of NSCs is shown in an adult mouse model of targeted cortical ablation where NSCs from the SVZ were recruited and developed into mature neurons that formed cortico-thalamic connections (Magavi et al., 2000). These exciting results suggest that following targeted injury, chemotactic factors and axon guidance molecules are re-expressed in the adult brain. It also suggests that NSCs from the SVZ are still competent to produce different types of cortical neurons and are able to migrate significant distances to repair the brain. Although in a different injury context, this study underlies the potential of NSCs to differentiate, migrate, and form the appropriate connections.

Remarkably, this was also observed in the human brain following successful engraftment of fetal ventral midbrain tissue to the substantia nigra or striatum (Li et al., 2008, 2016; Hallett et al., 2014; Mendez et al., 2005). Follow-up 3 to 24 years after engraftment confirmed successful re-innervation and long-term survival of dopaminergic neurons. However, the appearance of Lewy bodies in some of the grafted dopaminergic neurons was reported, which correlated with a decline in clinical improvement (Li et al., 2008, 2016). This underlines the importance of a combination of therapeutic approaches to inhibit pathology and stimulate repair at the same time. Nevertheless, these studies provide strong evidence that the human brain maintains an unexpected level of plasticity, even in aged and diseased conditions. The mechanisms that regulate this remarkable plasticity are not yet understood, but it suggests that the aged and diseased human brain remain permissive to regeneration, challenging the long-standing view that brain plasticity is lost with aging. This provides an important framework for the stimulation of endogenous NSCs to repair the brain. The appearance of PD pathology in newly generated dopaminergic neurons underlines that replacing lost dopaminergic neurons with healthy functional neurons and restoring dopaminergic innervation alone is not sufficient to provide long-term clinical improvement.

Thus, for successful treatment of PD, a multifaceted approach is needed that combines both stimulation of brain regeneration and disease-modifying strategies to prevent new dopaminergic neurons from developing disease pathology and stopping disease progression. Exploring the biology of NSCs under homeostatic conditions and in the diseased mammalian brain has uncovered greater capacity for plasticity than previously thought. Together with the growing number of studies in rodents suggesting that NSCs of the SVZ remain amenable to activation, proliferation, and differentiation, these NSCs are a valid stem cell source that warrants further research to chart the scope of their reparative capacity.

Acknowledgments

The figures were created with BioRender.com.

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article. The work was supported by a grant from the Stichting ParkinsonFonds.

Declaration of interests

The authors declare no competing interests.