Exercise your graft – An important lesson for cell replacement therapy for Parkinson’s disease

Abstract

Parkinson’s disease (PD) is a complex multisystem, chronic and so far, incurable disease affecting millions of people worldwide. With the continuing need for better therapeutic options for PD, there is a global renewed interest in cell replacement therapy due to progress in using pluripotent stem cells as an unlimited source of dopaminergic (DA) neurons for cell transplantation. Despite the significant progress made, obstacles remain that interfere with the restoration of functional circuits by DA grafts. The functional connectivity between DA grafts and host cells may be enhanced by adjunctive therapies, such as physical activity. Exercise modalities, such as use of treadmill, enhance neuroplasticity and improve motor and cognitive functions in PD patients. The patients are able to re-learn movement and adjust their posture, which, in turn, results in short term-reduced rigidity and improved stride length and cadence. By stabilizing selected active inputs and eliminating inactive ones, activity-dependent mechanisms fine-tune new neural circuits for optimal connection and physiological function. This communication will review the mechanisms and synergies between cell replacement therapy and physical and cognitive training to enhance induced pluripotent stem cell-mediated functional reinnervation of the striatum in PD.

Parkinson’s disease (PD) is a progressive neurodegenerative disease caused by death of dopamine (DA) neurons in the substantia nigra pars compacta (SNc). PD manifests with tremor, bradykinesia, rigidity and with significant non-motor dysfunctions, such as cognitive deficits and sleep disturbances among others (Braak et al., 1995; Braak and Del Tredici, 2017; Chaudhuri and Schapira, 2009; Espay et al., 2017; Jellinger, 2012, 2017; Lang and Lozano, 1998). In the US, over a million people suffer from PD and 90,000 new cases are diagnosed each year with an estimated cost of $52 billion per year (Willis et al., 2022; Yang et al., 2020). Since the incidence of PD increases with aging, with an additional trend of increased incidence over the age of 60 (Savica et al., 2016), the expected burden will continue to escalate surpassing Alzheimer ‘s disease (Dorsey and Bloem, 2018; GBD 2015 Neurological Disorders Collaborator Group, 2017). Although there are a number of therapeutic options available, current symptomatic therapies, such as levodopa pharmacotherapy and deep brain stimulation (DBS) (Castrioto et al., 2011; Cyron, 2016; Mestre et al., 2014) do not work uniformly well in all patients, and eventually are significantly curtailed by waning efficacy and debilitating side-effects with disease progression. Furthermore, symptomatic treatments generally do not address cognitive and other non-motor dysfunctions. Although antibody therapies are promising interventions for PD, none have met their primary endpoints in clinical trials (Devos et al., 2022; Lang et al., 2022; Pagano et al., 2022) and it remains to be demonstrated whether eliminating aggregated proteins, such as α-synuclein would be efficacious in late PD.

Since PD stems from the degeneration of one main cell type, the dopaminergic (DA) neurons, in a defined brain location, PD patients are considered optimal candidates for cell replacement therapy, based on the rational of grafting DA neurons back into the brain to restore the DA innervation of the striatum and the functions associated with it (Fig. 1). For more than three decades, experimental therapies have aimed at restoring lost dopaminergic (DA) terminals in the parkinsonian striatum through grafting replacement neurons (Björklund et al., 1980; Perlow et al., 1979). Grafting primary fetal ventral mesencephalon (VM) DA neurons has been clinically promising with efficacy reported in PD patients (Barker et al., 2013; Freed et al., 2011; Lindvall et al., 1990; Mendez et al., 2008; Olanow et al., 1996; Olanow et al., 2009). This therapeutic approach, however, encountered many challenges including practical, ethical and safety issues of using fetal tissues. In addition, some patients that showed therapeutic benefits developed adverse dyskinesias (Greene et al., 2021). In other clinical trials, sham-operated patients appeared to improve to an extent indistinguishable from transplanted patients (Freeman et al., 1995). With the recent progress in human pluripotent stem cell technologies, grafting trials in PD are beginning to re-emerge world-wide with a new focus on human embryonic stem cells (hESCs), human induced pluripotent stem cells (hiPSCs), or human parthenogenic stem cells (Gonzalez et al., 2016) as an unlimited source of DA neurons (Barbuti et al., 2021) (e.g.: Clinical Trial Identifiers NCT03119636, NCT03815071, NCT02452723, NCT03309514) (Table 1). Indeed, hESC- and hiPSC-derived DA grafts have thus far provided some striatal reinnervation and functional benefit (Adler et al., 2019; Daadi et al., 2024; Hallett et al., 2015; Kikuchi et al., 2017; Kriks et al., 2011; Tao et al., 2021). Despite this renewed interest, obstacles remain that interfere with restoration of functional circuits (Kordower et al., 2017; Li et al., 2016). The essential criteria for judging a successful neural cell therapeutic product for PD would seem to be: (1) a graft which robustly elaborates neurites that functionally reinnervate the host striatum with DA input; (2) efficient delivery procedure that ensures the cells safely occupy a precisely targeted area with no adverse effects; and (3) functional recovery associated with the transplantation.

Fig. 1.

Induced Pluripotent Cell-based Therapy for Parkinson’s Disease. Schematic representation of the cell source, derivation process and therapeutic strategy of our drug product. The iPSCs are derived by reprograming neural stem cells (NSCs) isolated from the developing human brain. The iPSC clones are isolated, characterized and processed for the isolation and perpetuation of self-renewable, multipotent NSCs giving rise to neurons, astrocytes and oligodendrocytes. The NSCs are scaled up in 3D culture systems, characterized and banked as an intermediate stage of the production process. The last step of the manufacturing process involves the differentiation of NSCs into midbrain DA neurons using the DIF method (Daadi, 2019). The therapeutic strategy involves the transplantation of the drug product into the DA denervated putamen of PD patients and implementation of a personalized adjunctive physical and cognitive training rehabilitation to promote homeostatic and functional graft-derived DA synaptic reinnervation of the striatum. Created with BioRender.com.

Table 1.

Current clinical trials for pluripotent stem cell replacement therapy for Parkinson’s Disease.

Sponsor	BlueRock Therapeutics (US)	Region Skane (UK, Sweden)	Aspen Neuroscience (US)	UCSD / Sumitomo / Kyoto University (US, Japan)
US Government Trial ID (ClinicalTrials.gov)	NCT04802733 Phase 1	NCT05635409 Phase 1	NCT06344026 Phase 1	NCT06482268 Phase 1
Drug Product	MSK-DA01 (Embryonic stem cell-derived dopaminergic neurons)	STEM-PD (Embryonic Stem Cell Derived Dopaminergic Progenitors)	ANPD001 (iPS Cell-derived Dopaminergic Progenitors)	CT1-DAP001 (iPS Cell-derived Dopaminergic Progenitors)
Cell Source	Allogeneic	Allogeneic	Autologous	Allogeneic
Rehabilitation Program	No	No	No	No
Start Date	May 3rd, 2021	November 30th, 2022	January 23rd, 2024	June 1st, 2024
Primary Outcome Measures	Safety (SAE §, abnormal tissue overgrowth)	Safety (SAE, abnormal tissue overgrowth)	Safety (SAE)	Safety (SAE, graft overgrowth)

^§“Serious Adverse Events”

Most transplant studies in the nervous system passively observe whether grafting, as the sole intervention, meets these criteria. Investigators tend not to address the fact that PD patients receive an entire suite of therapeutic interventions, including physical and cognitive rehabilitation. Indeed, physical therapy alone can be beneficial in many neurological conditions including PD. The design of these studies also does not address that neural development is critically reliant on activity-dependent cues. Certain aspects of the stem cell-based engraftment process recapitulate the functional connectivity of the nigrostriatal and cortico-striatal systems during development. A recent study (Daadi et al., 2024) combining physical and cognitive training with an iPSC-based cell replacement therapy in a nonhuman primate (NHP) model of PD brings back adjunctive behavioral training that Steve Dunnett, a pioneer in neural transplantation, advocated decades ago with the expression: “train your graft”.

Compelling evidence suggest that physical activity, such as treadmill exercise, transiently enhances neuroplasticity and improves motor and cognitive functions in animal models and PD patients (Abbruzzese et al., 2016; Hotting and Roder, 2013; Rhyu et al., 2010; Stern et al., 2019; Tomlinson et al., 2013), although, admittedly, it does not stop disease progression (Alberts et al., 2011; Alberts et al., 2016; Fisher et al., 2013; Fisher et al., 2004; Fisher et al., 2008; Herman et al., 2007; Pelosin et al., 2009; Petzinger et al., 2007; Schenkman et al., 2012; Stuckenschneider et al., 2016; Tillerson et al., 2003). Indeed, many rehabilitation facilities train their PD patients in ballroom dance, tango, yoga, boxing, cycling, treadmill, and other physical activities. Activity-dependent beneficial effects appear to be mediated by a variety of mechanisms, including an increase in mitochondrial function, reduced inflammation, enhanced angiogenesis, neurotrophic factor secretion, and synaptogenesis (Cotman and Berchtold, 2002; Fredriksson et al., 2011; Gerecke et al., 2012; Lau et al., 2011; Petzinger et al., 2013; Real et al., 2013; Tajiri et al., 2010; Toy et al., 2014; Tuon et al., 2012; Wu et al., 2011).

The critical role of training and use-dependent neuroplasticity for functional restoration and circuit refinement has been studied during development in association with recovery from injuries and after cell transplantation (Brasted et al., 1999; Dobrossy and Dunnett, 2001; Mayer et al., 1992; Sofroniew, 2018). During development and motor learning newly formed circuits become functional through activity-dependent use, which activates the interplay between neural activity and physiological synapse formation (Crair et al., 1998; Inglis et al., 2000; Martin et al., 2004). The absence of such activity-dependent use prevents the acquisition of function. Specifically, when muscle activity is prevented by paralyzing selected forelimb muscles with intramuscular injections of botulinum toxin, the development of cortico-spinal innervation is defective in adult animals (Martin et al., 2004). There are reductions in the topographic distribution of axon terminals, in terminal and pre-terminal branching, and in varicosity density (Martin et al., 2004). By stabilizing selected inputs and eliminating inactive ones, activity-dependent mechanisms fine-tune connections for optimal, physiological function (Inglis et al., 2000). Interestingly, in animal models of spinal cord injury (SCI), combining activity-dependent spinal stimulation with physical training promoted functional recovery (McPherson et al., 2015), while combining forelimb skilled grasping rehabilitation tasks with neural progenitor cell grafts boosted functional recovery (Lu et al., 2022). In rat models of SCI (Tashiro et al., 2016) and PD (Torikoshi et al., 2020), the combination of embryonic fetal brain neural cell transplantation with treadmill exercise enhanced graft survival and maturation. In a mouse model of skeletal muscle injury, adjunctive exercise enhanced the engraftment of stem cells, muscle reinnervation, and functional recovery (Ambrosio et al., 2010; Quarta et al., 2017). In animal models of Huntington disease, those that received grafts of embryonic fetal striatal tissue into the striatum and underwent adjunctive physical activity training were able to learn new tasks, whereas lesioned grafted animals that did not receive the physical training were unable to learn the same task (Brasted et al., 1999; Brasted et al., 2000; Mayer et al., 1992).

In the above-referenced paper (Daadi et al., 2024), the authors report on the positive impact that treadmill exercise combined with a cognitive task had on the engraftment of hiPSC-derived DA neurons in the NHP model of PD (Choudhury and Daadi, 2018). Six months post-transplant, the NHP group that received adjunctive physical and cognitive training returned to a normal baseline in their daily activity (measured by actigraphy), and significantly improved their performance in sensorimotor and cognitive tasks; histological examination of their brains showed enhanced synapse formation between grafted and host cells.

PD patients suffer from complex sensorimotor and cognitive deficits including loss of automaticity, and impaired procedural learning, cognitive flexibility, and response inhibition (Cools et al., 2001; Downes et al., 1989; Knowlton et al., 1996; Owen et al., 1992; Shohamy et al., 2004). With running as an exercise, PD patients are transiently able to re-learn movement and adjust their posture, which, in turn, results in short term-reduced rigidity and improved stride length, cadence, postural stability, gait rhythmicity, and joint excursion (Frenkel-Toledo et al., 2005; Herman et al., 2007; Herman et al., 2009). The repeated training of motor tasks results in the gradual learning of “association”, a characteristic of habit learning, which is mediated, in large part, by the striatum (Knowlton et al., 1996). One could imagine that a new source of DA innervation (pluripotent stem cell derived) of the striatum might only enhance these gains and render them more functionally long-lasting.

We envision that the design and implementation of PCT in clinical trials would be patient-tailored depending on age, stage of PD and physical abilities of the patient and include training modalities engaging sensorimotor and cognitive functions in accordance with the World Health Organization (WHO) recommendations for achieving a moderate-to-vigorous intensity to gain motor and cognitive benefits (World Health Organization, 2020). Typical forms of exercise engaging physical and cognitive performances that benefit patients include treadmill, exoskeleton robot-aided for patients with disability or advanced PD, bimanual tasks, touch screen tasks, video games, action observation training & motor imagery, dual-task gait/balance training, self-reported exercise, swimming, yoga, tai-chi, dancing, jogging, cycling and walking.

The preclinical and clinical findings discussed above, now extended to clinically relevant NHP models (Daadi et al., 2024), suggest that even the best cell-based intervention for PD with demonstrable graft-derived DA synaptic reinnervation of the striatum would perform even better, conferring more impactful and long lasting functional recovery, if it were combined with an adjunctive physical and cognitive rehabilitation program.

Data availability

No data was used for the research described in the article.