Stem cell-based therapies hold great promise in treating PD by addressing the potential for replacing or regenerating the depleted mDA neurons. After the discovery of adult neural stem cells (NSCs) in the human brain (32), numerous studies have explored whether neurogenesis occurs in the adult brain (33) and if restoration of mDA neurons using resident NSCs is possible in the brains of PD patients (30, 34, 35). Adult NSCs have been isolated from rodent models and human brain tissues (36), confirming their limited but verified ability to differentiate into mDA neurons (30). However, due to fate commitment issue primarily located in the anatomical subventricular zone, the efficiency of differentiation into mDA neurons from these cells was highly restricted, and the effects of animal model transplantation were modest (37). On the other hand, transplantation of fetal brain tissue into PD patients showed promising results (38, 39), but faced limitations such as tissue acquisition challenges and ethical issues. The concept of differentiating and transplanting mDA neuronal cells from human ESCs (hESCs) has emerged (40-42), leading to the investigation of two primary approaches: the transplantation of mDA neuron progenitors derived from ESCs or iPSCs, and the transplantation of NSCs (43).

ESCs and iPSCs have shown significant potential in PD therapy due to their ability to differentiate into mDA neurons suitable for transplantation (31, 44). Multiple studies have successfully differentiated ESCs and iPSCs into mDA neurons, providing a renewable source of cells for transplantation. Kriks et al. (45) generated mDA neurons from hESCs and human iPSCs (hiPSCs), demonstrating functional properties similar to native mDA neurons. Transplanting these cells into a rat PD model resulted in functional recovery and substantial improvements in motor behaviors. Kim et al. (46) developed a scalable and reproducible protocol for differentiation of hESCs and hiPSCs into mDA neurons through biphasic WNT activation. They effectively transplanted hiPSC-derived mDA neuron precursors into a primate and rat model of PD, resulting in significant enhancements in motor function and behavioral recovery (46). Schweitzer et al. (47) reported that a single patient undergoing hiPSC-derived mDA precursor transplantation exhibited stabilized or improved clinical measures of PD symptoms 18 to 24 months post-implantation. Recently, they advanced their research by co-transplanting dopamine cells with autologous regulatory T cells, effectively increasing the engraftment rate of the transplanted cells. This development signifies a shift towards not just exploring the therapeutic possibilities of cell transplantation but also seeking more efficient strategies for cell engraftment therapy (48). Several research groups have initiated clinical trials employing human PSC-derived dopamine cells, highlighting their potential for cell replacement (47, 49-52). Moreover, Piao et al. (50) and Kirkeby et al. (53) have commenced clinical trials by producing hESC-derived precursor cells of mDA neurons that meet quality, safety, and efficacy criteria for transplantation into patients, respectively. In this process, they have overcome the limitations of personalized therapy based on hiPSCs by incorporating the assistance of immunosuppressive agents (54), aiming to achieve the concept of transplantation using cryopreserved, ready-to-use cells based on hESCs.

In addition to ESC and iPSC-based transplantation, NSC-based transplantation has also shown promise for PD therapy. NSCs are multipotent cells capable of differentiating into various neural cell types, including mDA neurons. Transplantation of NSCs directly into the affected brain regions of PD patients has demonstrated positive outcomes in both preclinical and clinical studies. Roy et al. (55) successfully engrafted and differentiated NSCs into mDA neurons in a rat PD model, leading to behavioral improvements. The transplanted cells expressed mDA markers and exhibited functional integration within the host brain. In a primate model of PD, Redmond et al. (56) transplanted NSCs derived from the ventral mesencephalon, observing improved motor function, indicating the therapeutic potential of NSC-based approaches. Bachoud-Lévi et al. (57) conducted a clinical trial where NSCs were transplanted into the striatum of PD patients, demonstrating long-term survival, functional integration of the transplanted cells, along with improvements in motor symptoms and quality of life.

These studies underscore the potential of ESC, iPSC, and NSC-based therapies for PD by providing a renewable source of mDA neurons or neural progenitors capable of differentiation and functional integration. Further research and clinical trials are essential to optimize transplantation protocols, assess long-term effects, and establish the safety and efficacy of these stem cell-based therapies for PD.

While stem cell-based therapies offer great potential for the treatment of PD, several limitations need to be addressed at the clinical level before their extensive application. These limitations encompass concerns regarding efficacy, efficiency, potential side effects, immunological considerations, limited availability, scalability, and ethical and regulatory considerations.

Efficacy: In stem cell-based therapies for PD, efficacy is of paramount importance. Considering the diversity in genetic profiles, environmental factors, and the stages of disease progression that have resulted in variations in the effects of nearly all medications, it remains necessary to firmly establish efficacy by applying attempts to treat PD patients with dopaminergic cell transplants to a larger patient population. These efforts are still in their challenging and exploratory early stages, and, as a result, they have so far been limited to single patients or small cohorts, making it difficult to precisely evaluate the efficacy and identify influencing factors. Additionally, long-term tracking of efficacy is a vital aspect to consider. Although early clinical trials have provided promising results in terms of ameliorating motor symptoms, translating these findings into meaningful and sustained clinical benefits remains a formidable challenge. The heterogeneity in patient responses to treatment and the enduring effects of therapy necessitate further investigation. Turning to the analysis of therapeutic outcomes from the transplantation of fetal midbrain tissue in PD patients, postmortem studies revealed long-term survival of transplanted dopaminergic and serotonin neurons for over a decade (58-60). However, the presence of Lewy bodies in the transplanted neurons raises concerns about the longsuffering and stable effectiveness of stem cell-derived neuron transplants (59, 60). Conversely, another study reported no Lewy bodies observed in postmortem tissue even 14 years after cell transplantation. This divergence underscores the need for validation and optimization of stem cell-derived dopamine cell transplantation from a long-term perspective (58). Factors influencing therapeutic efficacy, including viability, functional integration, and the establishment of appropriate neural circuits within the transplanted cells, are substantial and necessitate optimization to maximize clinical advantages (61, 62).

Efficiency: Efficiency represents another significant limitation in the clinical application of stem cell-based therapies for PD. Generating a sufficient number of functional mDA neurons from stem cells in a cost-effective and scalable manner is a complex task. Crucial steps include streamlining and standardizing the manufacturing process, ensuring reproducibility and consistency of produced cells, and optimizing cell transplantation protocols. Advances in cell manufacturing technologies, including automation and quality control measures, are being explored to enhance the efficiency of stem cell-based therapies (63, 64).

Side effects: Safety concerns and potential side effects are of vital significance in the context of stem cell-based therapies. While these therapies aim to restore the function of mDA neurons, there is an inherent risk of adverse effects. One notable concern pertains to the development of graft-induced dyskinesias, characterized by abnormal involuntary movements resulting from the excessive release of dopamine by transplanted cells (65). Strategies to mitigate this risk encompass refining transplantation protocols, optimizing cell dosages, and implementing immunosuppressive regimens when necessary. Additionally, long-term monitoring and evaluation of patients undergoing stem cell-based therapies are essential to assess potential side effects, including tumorigenicity and immune responses (31, 66). From this perspective, the development of therapeutic cells has also involved “fail-safe approach” efforts in methods for controlling their fate. These endeavors include recognizing markers associated with cancerous transformations in cells that fail to complete their final differentiation or markers signifying a continuing proliferative state in cells remaining in a pro-mitotic state (67, 68). In such cases, cell engineering is employed to activate artificial cell suicide programs, facilitating self-destruction. Alternatively, more assertive techniques, such as gamma-ray irradiation, are being explored to induce damage specific to undifferentiated cells that still retain proliferative potential post-transplantation (69). These methods are currently under investigation for their potential utility as safety measures.

Immunological considerations: Another challenge in stem cell-based therapies is the potential for immune rejection of transplanted cells. ESCs, iPSCs, and NSCs may elicit immune responses in recipients, which can compromise the survival and functionality of transplanted cells. Strategies to address immunological compatibility include the use of immunosuppressive medications or the application of immune-matched or patient-specific stem cells. For example, successful transplantation of autologous hiPSC-derived mDA neurons has been reported, circumventing the need for immunosuppression (70). Further research, such as the development of hypo-immune hiPSCs (universal hiPSCs) (71), is necessary to optimize immunosuppression regimens and develop immunomodulatory approaches to enhance the long-term survival of transplanted cells.

Limited availability and scalability: The ability to apply stem cell-based therapies for PD on a broader scale is constrained by the limited availability and scalability of suitable cells (72). The challenge of producing high-quality and standardized cell populations for transplantation remains prominent. There is a need for scalable protocols to enable the large-scale production of mDA neurons or NSCs to meet the demands of potential clinical applications. Furthermore, establishing robust manufacturing processes, implementing quality control measures, and ensuring compliance with regulatory standards are essential steps for translating stem cell-based therapies into clinical practice.

Ethical and regulatory considerations: Stem cell-based therapies bring forth ethical and regulatory considerations, particularly when utilizing hESCs or fetal-derived cells (73). The use of hESCs raises ethical concerns related to the destruction of embryos (74). However, the development of hiPSCs provides an alternative source of patient-specific cells without ethical controversies. It is of paramount importance to strictly adhere to ethical guidelines and regulatory frameworks to ensure patient safety and the responsible advancement of stem cell-based therapies (75, 76).

Stem cell-based disease models have brought about a revolution in PD research. These models provide a platform to study the underlying mechanisms of the disease and facilitate drug development, as illustrated in Fig. 2.

Disease modeling studies utilizing hESCs have offered insights into disease mechanisms and potential therapeutic targets for PD. For example, Cooper et al. (83) successfully differentiated hESCs into mDA neurons, identifying dysregulation of the WNT signaling pathway as a potential contributing factor in PD pathology. Byers et al. (84) used hESC-derived mDA neurons to demonstrate the involvement of the unfolded protein response and endoplasmic reticulum (ER) stress in PD pathogenesis. These studies showcase the utility of hESC-based models in investigating disease mechanisms and identifying novel therapeutic targets. In recent years, there has been a notable upswing in the utilization of hiPSCs for the purpose of modeling PD. A significant emphasis within the realm of hiPSC models lies in the exploration of α-syn (SNCA) mutations and gene multiplications associated with PD (4, 84-86). Especially, hiPSC-derived mDA neurons from individuals afflicted with SNCA mutations revealed elevated levels of α-syn and the resulting deleterious consequences, such as oxidative stress and ER stress (84, 86-88). Furthermore, α-syn oligomers have been identified as disruptors of axonal transport, mitochondrial function, and synaptic connectivity in hiPSC-derived DA neurons (89, 90), as detailed in Table 1 (84, 87-117).

LRRK2 mutations, considered another genetic risk factor for PD, have undergone extensive examination via the use of hiPSC models (107, 108, 118, 119). These models have brought to light a spectrum of pathological characteristics linked to LRRK2 mutations, encompassing mitochondrial dysfunction, oxidative stress, and compromised autophagy (83, 109, 110, 120). Importantly, hiPSC-derived DA neurons have served as a valuable tool for screening potential therapeutic compounds in the context of LRRK2-related PD (121). Exploration of mutations in the GBA gene, known to heighten the risk of PD, has been undertaken employing hiPSC models (98, 99, 101). These models have furnished valuable insights into the role of GBA mutations in α-syn accumulation, lysosomal dysfunction, and mitochondrial impairment (100, 101). Furthermore, hiPSC-based investigations have uncovered promising therapeutic strategies for GBA-associated PD (122). Additionally, hiPSC models have been employed to validate other PD-related genes, including PINK1 (123), PARK2/PARKIN (104), and VPS35 (124). These investigations have unveiled critical facets of mitochondrial dysfunction, impaired mitophagy, and disrupted endo-lysosomal trafficking in the pathogenesis of PD (104, 117, 125). Beyond the study of monogenic forms of the disease, hiPSC models have significantly contributed to our comprehension of sporadic PD (126-128). They have illuminated aberrations in oxidative stress, epigenetic modifications, and proteostasis pathways, offering valuable insights into the intricate nature of sporadic PD. Collectively, these models hold tremendous promise for advancing our understanding of PD pathogenesis, the identification of novel therapeutic candidates, and the rigorous evaluation of their effectiveness, constituting a pivotal component of PD research.

Stem cell technology presents several advantages in the context of PD drug development. One of the key benefits is the ability to generate disease-relevant cell types affected in PD. For instance, hiPSCs derived from PD patients can be differentiated into mDA neurons, enabling the investigation of disease-specific mechanisms. Utilizing this approach, numerous groups have revealed PD-associated phenotypes in mDA neurons and identified therapeutic compounds capable of alleviating the phenotypes, as detailed in Table 2 (83, 95, 99, 121, 122, 125, 127-139).

Additionally, stem cell-based models facilitate high-throughput screening assays and validation processes for identifying potential therapeutic compounds. For example, Tabata et al. (121) conducted a high-throughput screen on hiPSC-derived mDA neurons and discovered compounds that suppressed rotenone-induced apoptosis while promoting neuronal survival, suggesting their potential as therapeutic agents. Similarly, Yamaguchi et al. (137) utilized a semi-automated phenotypic analysis system to screen a library of compounds using hiPSC-derived mDA neurons from familial PD patients. They successfully identified four compounds that improved mitochondrial clearance and enhanced the functionality of mDA neurons.

However, despite its advantages, stem cell technology encounters certain limitations that require addressing for its effective application in drug discovery. One such limitation pertains to the complexity of modeling a multifactorial disease like PD, which entails intricate interactions between various cell types and environmental factors (140). To address this challenge, recent studies have concentrated on the development of multicellular systems, such as three-dimensional (3D) organoid models created from hiPSCs. For example, Kwak et al. (141) engineered a 3D midbrain organoid model, enabling exploration into the effects of midbrain-related cell types on the survival of mDA neurons and providing valuable insights into disease mechanisms. Another obstacle lies in maturing stem cell-derived mDA neurons to faithfully replicate the PD phenotypes observed in neurons resembling those in PD patients. To tackle this issue, diverse strategies have been explored, including extended culture periods (142), co-cultivation with supporting cells (143, 144), and genetic manipulation (145). Furthermore, the variability observed among hiPSC-derived neuronal cultures and differentiation protocols complicates the generation of consistent and reproducible results, posing challenges for the development of high-throughput screening assays. Efforts have been made to standardize protocols and optimize differentiation conditions to reduce variability and enhance the reliability of stem cell-based models. For instance, Nolbrant et al. (146) devised a defined and scalable differentiation protocol for generating mDA neurons from hiPSCs, which demonstrated improved efficiency and consistency across different hiPSC lines.

The enactment of the U.S. Food and Drug Administration (FDA) Modernization Act 2.0 in December 2022 has underscored the imperative need for non-animal testing alternatives in drug development (147, 148). This regulatory shift promotes the utilization of human-centric models, such as hiPSC technology and organoids, for non-clinical tests in the field of drug development and safety evaluations. Consequently, there exists an urgency to devise advanced hiPSC-based models capable of accurately recapitulating human disease pathology and offering enhanced preclinical assessment of drug candidates, aligning with the FDA’s strategic objectives. While PD is a multifaceted disorder with various contributory factors, including genetic mutations, environmental influences, oxidative stress, and mitochondrial dysfunction, a common hallmark of PD is the accumulation of α-syn proteins within mDA neurons. This molecular event remains a fundamental feature of PD pathogenesis, leading to the formation of α-syn aggregates known as Lewy bodies. These aggregates are believed to selectively exert toxicity in mDA neurons, resulting in neuronal dysfunction and eventual death-key characteristics of PD (1, 2, 5). Given the central role of α-syn aggregation in PD, there is a pressing necessity for a novel hiPSC-based model to expedite the development of interventions that can impede or reverse the progression of this debilitating disorder. However, substantial challenges persist, primarily due to the somatic reprogramming involved in generating hiPSCs, which resets the cells to an embryonic state and makes it challenging to model late-onset diseases like PD characterized by α-syn aggregation (149, 150). Consequently, additional strategies are required to accurately replicate the aberrant protein aggregation intrinsic to the disease progression (151, 152). Furthermore, it is crucial to recognize that most non-cell-based compound screenings suffer from limited reproducibility in human neurons, thereby underscoring the urgent need to establish reliable, consistent, and cost-effective human neuronal models for high-throughput drug discovery platforms. In conclusion, the establishment of a stem cell-based model focusing on α-syn aggregates in PD is of utmost importance. Such a model promises not only to enrich our comprehension of the disease but also to expedite the discovery of potential treatments, marking a significant advancement in the battle against PD.

To address these challenges, we have recently developed the OASIS model (153, 154). This model presents several advantages for studying PD and sheds light on its potential integration with other technologies and future directions (Fig. 3). One key advantages of the OASIS model is its unique ability to induce α-syn aggregation using optogenetic proteins within a short timeframe. This distinctive feature allows precise control and manipulation of the formation of pathological α-syn aggregates, closely resembling the Lewy bodies observed in PD. By investigating the impact of α-syn aggregation on cellular processes and neurodegeneration, the OASIS model offers valuable insights into the underlying mechanisms of PD pathogenesis. Another advantage of the OASIS model is its provision of a simplified, cost-effective platform for drug screening and the identifying potential therapeutic candidates. Through controlled induction of α-syn aggregation, compounds can be efficiently screened to reduce aggregate formation, yieling highly consistent results. Moreover, the OASIS model employs hiPSC-derived mDA neurons and midbrain organoids from PD patients, enabling the generation of patient-specific multiple cell types. This allows for a more accurate assessment of the effectiveness of potential hit compounds in reducing cellular toxicity and restoring normal function. As such, the OASIS model serves as a non-clinical testing platform for PD, holding promise for discovering new therapeutic targets and developing disease-modifying treatments.

Nevertheless, it is crucial to consider the limitations of the OASIS model. One constraint is its primary focus on the aggregation of α-syn in SNCA triplication PD iPSCs. Another concern is its exclusive emphasis on α-syn aggregation, despite PD involving a complex interplay of diverse cellular processes originating from other genetic mutations or sporadic PD patients’ hiPSCs. These aspects are linked to the limitations of the OASIS model, which, despite its ability to mimic α-syn aggregation, is confined to representing the molecular phenotype at the endpoint of cellular pathology. The model faces limitations in depicting the cellular pathological mechanisms initiating PD, including the recently investigated propagation of misfolded α-syn in the gut-brain axis and functional impairment due to misfolded α-syn within cells (155, 156). Falling short, the model fails to capture the broader molecular pathological mechanisms occurring before α-syn aggregation, such as issues stemming from a low functional protein dose of α-syn (157) and dominant-negative mutants (158) caused by α-syn mutations (159, 160). This underscores the necessity for model improvements to achieve a more precise understanding of molecular and cellular pathology related to the α-syn function on various biological events (161-164), as well as a more comprehensive understanding that encompasses the spread of pathological molecular features in PD.

Utilizing the OASIS model based on patient-specific hiPSCs facilitates the exploration of pathogenic mechanisms reflecting the individual genetic profiles, encompassing various genetic mutations implicated in PD onset. This approach allows for the identification of personalized therapeutic interventions and assessment of their potential efficacy, making it possible to evaluate individual responses to treatment and optimize therapeutic strategies. Conversely, introducing diverse SCNA mutations into normal cells with the same genetic background and utilizing OASIS advances knowledge regarding the penetrance and diversity of PD onset and symptoms for each genetic mutation type, contributing to a precise understanding of PD pathogenesis. Moreover, the OASIS model can be enhanced by integrating with other advanced technologies, such as microfluidic systems (165), organoids (166, 167), or organ-on-a-chip platforms (168), enabling the mimicry of the physiological environment of PD pathology-relevant tissues. This allows for the exploration of cell-cell interactions, neuronal circuitry, and synaptic activity within the context of α-syn aggregation. Additionally, incorporation of advanced imaging techniques and multi-omics approaches can provide a more detailed characterization of cellular and molecular changes associated with PD.

In conclusion, the OASIS model offers advantages such as precise control over α-syn aggregation, utilization of patient-specific hiPSCs, and potential for drug screening. However, it is crucial to carefully address the limitations of this model and consider future integration with other technologies to enhance its validity and comprehensiveness. The direction of the OASIS model lies in the potential development of personalized medicine and targeted therapies for PD. Depending on the extent of improvements, it can also contribute to the development of various pathological models and the elucidation of fundamental mechanism of PD onset and progression.