Neural Stem Cell Therapy and Rehabilitation in the Central Nervous System: Emerging Partnerships

Stem Cells for Neural Regeneration

A fundamental goal of CNS regenerative medicine is to circumvent the inherent inability of CNS tissue to regenerate.^19,20 Tools used for neural regeneration have been reviewed elsewhere,²¹ and here we expand on stem cells as one promising choice. The very biology of a stem cell makes it an attractive option for neural regeneration (Fig. 1). The term “stem cell” refers to a cell that can: (1) repopulate a given tissue through division into additional stem cells (self-renewal) and (2) progress or differentiate from an immature cell to a postmitotic, mature cell such as a neuron or astrocyte (multipotency). To further explain self-renewal, stem cells asymmetrically divide, generating one identical daughter stem cell and one daughter progenitor cell. The former daughter stem cell replenishes the original stem cell reservoir, thus “self-renewing” the population. The latter progenitor cell is more “committed” and thus closer to its final differentiative state. To further explain multipotency, daughter progenitors divide at least once and ultimately become the mature cell types of a given tissue (ie, neurons and glia in the CNS).

• Download figure
• Open in new tab
• Download powerpoint

Figure 1.

General lineage of neural stem cells. Neural stem cells undergo asymmetric division periodically for normal maintenance or in response to injury or other stimuli. One daughter cell (black) is an identical stem cell that renews or repopulates stem cell numbers. The other daughter cell (white) is a more committed progenitor cell that also may divide, one or more times, to create an even greater number of presumptive progeny cells. In the neural system, progeny cells will become postmitotic and mature into the mature central nervous system cell types: neurons, astrocytes, and oligodendrocytes. Reprinted with permission from Elsevier Sciences Ltd from: Weiss S, Reynolds BA, Vescovi AL, et al. Is there a neural stem cell in the mammalian forebrain? Trends Neurosci. 1996;19:387–393.

Stem cells in the CNS are robustly present during embryonic development. They also persist in the postnatal and adult CNS, albeit at much lower numbers²² and only in specific locations.^23–26 The 2 most well-characterized adult stem cell regions of the CNS are the subgranular zone of the dentate gyrus in the hippocampus and the subventricular zones of the lateral ventricles.^22,27–29 The hippocampal lineage produces neuronal progenitors that divide once and integrate into the dentate gyrus, likely contributing to hippocampal homeostasis.³⁰ The subventricular zone lineage divides several times to replenish interneurons in the olfactory bulb to support the chemical sense of smell in lower mammals.³¹ In higher mammals, some evidence suggests the subventricular lineage may respond to injury^32,33 by homing to tumor or stroke foci, although the significance of this phenomenon is unclear.

Self-renewal, multipotency, and a presence in the adult brain are central features of the therapeutic application of stem cells for CNS repair. First, through a sustained ability to divide (self-renewal)—stem cells can generate numerous new cells that replace lost cells or provide nutritive support (ie, secretion of molecules that promote healing) to residual tissue. Second, because they are multipotent, stem cells generate multiple cell types for any given tissue. Accordingly, stem cells from the CNS (or those that can be programmed to make CNS cell types) possess the added benefit of producing not just any cell, but neural cells in particular (ie, neurons, astrocytes, and oligodendrocytes). These qualities are particularly important when cell loss occurs, a characteristic of many CNS pathologies.* Third, the persistence of stem cells into adulthood raises possibilities relevant to rehabilitation. How do these pools contribute to recovery following injury, if at all? Can we use clinically relevant physical therapy protocols to augment neurogenesis in the injured or diseased CNS and promote functionally meaningful, endogenous cell replacement and self-repair? These are intriguing, yet elusive, issues that are currently being addressed by experimental strategies involving transplantation of stem cells from various sources in preclinical models of neurological disease and trauma.

Limitations Associated With Stem Cell Therapies in the CNS

Each year, an estimated 50 million people sustain CNS damage, including stroke, traumatic brain injury, SCI, demyelinating diseases, and neurodegenerative disorders.³⁴ Although CNS pathologies are variable in mechanism and presentation, the uniform and resultant regenerative failure is the principal rationale for pursuing regenerative medicine–based repair—including stem cell transplantation. The primary regenerative goals are the genesis of new neurons and supporting glial cells, as well as long-distance regrowth of injured nerve fibers. The injured CNS is characterized by a complex cascade of biochemical events that contribute to secondary tissue loss beyond the initial insult^21,35 (summarized in Fig. 2). In this hostile host environment, donor or experimentally introduced cells are susceptible to conditions that can negatively affect cell engraftment (as noted below). Thus, donor cell populations shown to be capable of long-term survival in all neural cell types in pretransplant culture systems^27,28,36 demonstrate significant reductions in proliferation and differentiation following transplantation into the injured CNS environment. This fact dramatically limits transplantation efficiency and underlies the theory that additional cotherapies such as rehabilitation may be required to optimize cell therapy outcomes.

• Download figure
• Open in new tab
• Download powerpoint

Figure 2.

Overview of damage following central nervous system injury.

The selection process for therapeutic stem cells is a topic fueled by both ethical and logistical considerations. Stem cell sources from early developmental time points (eg, embryonic stem cells) are attractive given their high rate of proliferation. However, cell division must be harnessed in order to avoid tumor formation—a major risk factor in using embryonic stem cells. Conversely, cells obtained from a later developmental time point (eg, fetal or postnatal stem cells) are less likely to form tumors. Yet, this option could pose difficulties in generating ample transplant material due to the age-related decline in stem cell robustness with increasing age of the donor source. Feasibility also remains an important consideration for obtaining stem cell populations. Accessing embryonic and fetal material is both controversial and heavily regulated in the United States. On the other hand, harvesting of autologous stem cells from the postnatal or adult brain germinal zones is difficult to envision at this time. Another selection factor is using a donor cell that makes the mature cell type required. “Cell fate” is the identity of a mature cell (ie, neuron, astrocyte, oligodendrocyte) obtained following lineage progression from a stem cell to a final progeny cell. The importance of the potential fates of engrafted cells is dependent on the therapeutic goals. The general consensus is that, for robust tissue-specific cell replacement, postnatal donor material must come from the tissue of origin.^†,37 Reports to the contrary for neural cell replacement using stem cells from other tissues are controversial, but should not be ignored.^38–40 Additionally, donor cells from other tissue types (eg, mesenchymal stem cells for neural injury) offer the potential for non-cell replacement-related support, such as nutrition and counteraction of immune responses.^41,42

CNS Stem Cell Transplantation Studies

Because of the nature and properties inherent to stem cells, and in light of the limitations presented by a lesion to an essentially nonregenerative CNS, stem cell transplantation is first theorized to replace neural cells lost due to injury through cell replacement.⁴³ The notion that stem cell grafts could additionally provide trophic or nutritive support⁴⁴ to healing tissue is potentially just as powerful. In the case of SCI, transplantation of many cell types (eg, fetal material, embryonic and adult neural stem cells, marrow stromal stem cells, olfactory ensheathing cells, Schwann cells) from various donor sources has been tested in different experimental SCI models.^43,45–47 Collectively, these reports have demonstrated safety with rare exception,⁴⁸ and donor cell survival is seen weeks to months following transplantation. However, optimal cell delivery protocols (eg, cell types versus times of delivery versus means of delivery)—for human applications especially—remain to be defined.⁴⁹ Furthermore, a “transplant effect” is often assigned to behavioral improvements in experimental animals, but the underlying mechanism of a graft-mediated effect is often insufficiently demonstrated. Overall, experimental recovery is quite variable, and there is little information that correlates recovery with metrics of functional engraftment. For example, data demonstrating host-graft synaptic connectivity are sparse.

Translation of clinical stem cell transplantation to the clinic is an ongoing effort that is characterized by both promise and limitation. Clinical fetal cell transplantation studies initiated more than 3 decades ago provided an early road map for translating the basic science of neural tissue replacement to the clinical arena, with safety and feasibility being the endpoints of primary interest.^50–53 Translational studies initially focused on treating people with Parkinson disease with dopamine-producing embryonic CNS donor tissue, which has since been recognized to be a rich source of neural stem cells.^50,51 Preclinical results obtained from animal models of Parkinson disease and SCI led to human trials, which demonstrated that human neural progenitors could be safely grafted into human recipients.^50–53 Subsequently, clinical trials have been initiated in the United States and overseas using stem cells to treat people with stroke, amyotrophic lateral sclerosis, and SCI^43,54–56 (Table). An in-depth review of that work is beyond the scope of this article, but here we focus on relevant general comments. In addition to being an active and exciting area of translational research around the world, current and future clinical trials can broaden our knowledge through thoughtful consideration of experimental variability in areas such as nonuniform recruitment base, nonstandardized study designs, variability in procedural control, cell source definition, and peer-reviewed documentation. Limitations to preclinical central nervous system transplant literature include: heterogeneity of starting cell material, differences in cell isolation/preparation between laboratories, timing of transplant postinjury, mode of cell delivery, and lack of positive finding in a chronic injury model. Considerable caution must be exercised when drawing conclusions about the safety of different cell sources, efficacy of engraftment, or reliability of clinical benefits.⁴³

These overall results would suggest that, although an incredibly promising venture, CNS transplantation is still an evolving approach that will benefit from more extensive study in preclinical animal models. Current and future studies should seek to ensure that the optimal model is in use so that rehabilitation has the best opportunity to elicit biologically relevant improvements.

Building Evidence for Rehabilitation to Augment CNS Stem Cell Therapies

Stem cell transplantation offers the potential benefit of introducing an additional and sustained source of neuroplasticity. Neuroplasticity in this context refers to changes in the structure or function of a neural system in response to a stimulus. Therefore, following injury, stem cells would contribute structurally or functionally to what is essentially a novel neural substrate, within which lies an opportunity for the application of rehabilitation approaches to maximize the efficacy of current transplantation approaches. Activation of the nervous system through injury or sensory or motor stimuli provides a rationale to pursue combined therapies.

Stem cells in the CNS are responsive to injury and, therefore, could contribute to the recovery process. For example, following SCI in rodents, ependymal spinal cord stem cells rapidly proliferate, primarily producing glial cells.^57–59 In animal models of stroke, endogenous, subventricular, zone-derived progenitors proliferate and migrate toward the lesion.^33,60,61 Even more striking, these cells can differentiate into mature neurons⁶² and appear to integrate with host tissue.⁶³ In these studies, the time frame of donor cell integration occurs over a period of several months, an interval that, perhaps only coincidentally, corresponds to a typical poststroke recovery. In support of a possible endogenous neural stem cell contribution to recovery, chemical reduction of the host stem cell population results in a worse functional outcome following stroke compared with naive animals that generate a normal stem cell response to injury.⁶⁴ These data suggest that CNS stem cells are mobilized following injury and could contribute to recovery.⁶⁵ What clinical outcomes affirm, however, is that this endogenous stem cell response to injury is inadequate to mount a complete recovery of function following injury without further intervention.

Stem cells also respond to externally applied sensory and motor stimuli, which underlies the potential for rehabilitation to augment innate stem cell-mediated recovery. This finding was established in studies on the effects of sensory input and motor activity on the birth of new neurons (neurogenesis) in the hippocampus (ie, through proliferation or survival of stem cells and their progeny cells).^66,67 Animals housed in an enriched environment (ie, increased sensory input and social interaction) exhibited increased survival of newborn hippocampal neurons over time. In addition, voluntary exercise on a running wheel resulted in a temporally significant increase in the birth of new hippocampal neurons and improved their overall survival.⁶⁷ In aged animals, an average of 3.9 km per day (SD=0.1) on a running wheel results in a 50% reversal of the natural decline in hippocampal neurogenesis.⁶⁸ Thus, appropriately timed and dosed stimuli may influence endogenous stem cell responses. Furthermore, the same stimuli could complement transplantation by enhancing donor cell survival, fate, and integration, thereby leading to more consistent and biologically significant functional outcomes.

The mechanisms underlying stem cell responses to general exercise and environmental enrichment are slowly emerging.⁶⁹ Teng and colleagues⁶⁵ have proposed that activity stimulates endogenous or transplanted stem cells to secrete neurotrophins (molecules that promote neuronal health and plasticity). This proposal is supported by correlative data on the effects of activity at the anatomical and physiological levels. At the anatomical level, exercise can affect dendritic morphology and increase vascularity.⁷⁰ At the physiological level, exercise influences expression of synaptic and receptor proteins.⁷⁰ Such plasticity can directly influence neural stem cell pools.^67,68 On the other hand, activity or physiological modalities could counteract negative factors within the injury environment.

Physical rehabilitation approaches that focus on activation of the neuromuscular system could have an impact at the cellular level, with the aim of augmenting a behavioral change and improving outcomes.⁷¹ In this light, rehabilitation is a promising partner to stem cell technologies. A recent review emphasized the convergent themes of CNS regenerative medicine and neurorehabilitation approaches.²¹ The authors discussed that functional gains seen using these approaches in isolation, to date, are modest. They posited that the introduction of complementary approaches (eg, motor stimulation combined with small molecule therapy) will be needed to achieve improved outcomes. Applied to the current topic, patterned input through modalities and activity could address current limitations to cell replacement therapies by conditioning transplanted cells and the injured environment for survival and function over long periods.

Experimental evidence is needed to determine whether rehabilitation, with the appropriate timing and dosage of administration, can augment stem cell therapies aimed at CNS repair. Such evidence is emerging across the neuromuscular spectrum. For example, in the musculoskeletal system, exercise augments stem cell transplantation following muscular injury in a mouse model.¹³ In the CNS, locomotor training alone has elicited gains in hind-limb functional recovery in a number of neurological populations.^72–74 Yet, recovery is enhanced when locomotor training is paired with regenerative modalities, such as electrical stimulation⁵ or pharmacological agents.⁷⁵ Another recent report demonstrated that improvements in rodent motor function following cervical SCI were absent following gene therapy unless combined with rehabilitation.⁷⁶

A final example of particular interest is intermittent hypoxia, a systemic and physical therapy–administered physiologic stimulus. Relevant to stem cell therapies, recent in vitro findings outline that CNS neural stem cells harvested following intermittent hypoxia are more proliferative and yield a greater percentage of neurons than those harvested from control animals.^77,78 This is a compelling finding in that individuals with chronic incomplete SCI demonstrate greater plantar flexion force and ambulatory speed and endurance following acute mild intermittent hypoxia protocols.^12,79 Similar to the aforementioned findings in conjunction with locomotor training, the combination of ambulatory training with mild intermittent hypoxia results in an additive benefit to gait endurance.⁷⁹ Together, these findings suggest that systemic intermittent hypoxia is sufficient to elicit changes in neural stem cell biology that support study of their combined use to offset some limitations to CNS stem cell therapies (specifically, reduced proliferation or differentiation of transplanted cells with preconditioning and conditioning of a toxic injury environment).

Thus, rehabilitation modalities, in theory, gain ground as a component of multipronged approaches⁷⁵ to potentially optimize stem cell transplantation. This notion is in keeping with the concept of rehabilitating the patient on the cellular level^13,80 and, if carefully optimized for timing and dose of administration, may prove to be effective beyond the capabilities of the injured system alone.

Professional Opportunities to Forward Regenerative Rehabilitation

The potential for physical therapy to augment advanced technologies is supported by initiatives already set forth to address progressive therapeutics. In 2006, the American Physical Therapy Association (APTA) House of Delegates' Vision 2020⁹ mandated innovative practice models as part of the profession's mission. Subsequently, emphasis has been placed on the effective translation of advanced technologies into clinical practice to meet this mission. In 2010, the Physical Therapy and Society Summit (PASS) produced recommendations for professional growth to meet scientific advances.^81,82 These recommendations called for effective collaborations in testing new technologies, progression of technology-based intervention to health care delivery, establishment of mechanisms to promote clinical competencies with new technologies, and the promotion of multidisciplinary efforts in science and technology developments. Finally, during the 2011 National Institute of Child Health and Human Development Scientific Vision Workshop on Plasticity, basic and clinical scientists identified emerging developments regarding plasticity-related research. As outlined in the white paper report,⁸³ relevant questions raised included:

1. Do different modes of plasticity converge to promote healing?

2. How do therapeutic interventions tap into plasticity, and are multiple interventions needed to optimize access to multiple forms of biological plasticity?

3. What new tools and approaches are needed to study plasticity?

4. How do we evaluate the effects of combination therapies on plasticity?

5. Do we have the most appropriate preclinical models?

Therefore, there is a significant drive for physical therapists to engage in advanced technologies research, such as those technologies encompassed by regenerative medicine.

In the spirit of acting on these initiatives, several recent sessions have focused on the growth of regenerative rehabilitation. Regenerative rehabilitation has been the focus of an educational session (2011) and the Eugene Michels Researchers Forum (2012) at the APTA Combined Sections Meetings. Building on this educational platform, Ambrosio and colleagues have initiated a series of symposia on regenerative rehabilitation that overview current research efforts and progress toward combined therapies, with other similar symposia now occurring internationally. Regenerative rehabilitation was the subject of the 2012 APTA Section on Research retreat and has been highlighted in print⁸⁴ and via podcast in PT in Motion.⁸⁵ Finally, with the inception of the Alliance for Regenerative Rehabilitation Research and Training (AR³T), major research institutions are joining together to fund research and to educate and disseminate current research to the broadest audience possible.⁸⁶

Clearly, more physical therapists are becoming involved in regenerative medicine research. For example, a survey of the top 27 programs listed in the US News and World Report revealed that 43% actively participate in some form of regenerative medicine–oriented research (unpublished observation). This observation supports the growing interface between physical therapy and regenerative medicine research teams. Building partnerships between physical therapy researchers and clinicians will become a critical next step, where each can play a significant role in the design and application of emerging regenerative rehabilitation protocols. Early collaboration between the 2 disciplines is the root of a fusion model proposed by Ambrosio and Russell,¹⁵ which contends that rehabilitation and regenerative medicine interactions should be assessed concurrently during research development, rather than in a separated and linear fashion. The rationale is to assess rehabilitation approaches in the context of developing advanced technologies and as a component of preclinical trials to ensure that dosing parameters and clinical outcomes are optimized.

What does this conceptualization mean for the experienced clinician and for interested clinical doctor of physical therapy (DPT) students? Experienced clinical insight will be critical to the successful and wide implementation of regenerative rehabilitation. Advanced knowledge of patient responses to regenerative approaches will be important to the accurate assessment of treatment efficacy. The development of appropriate clinical outcomes that capture nuanced aspects of recovery will also be important.⁸⁷ A preliminary step in establishing prerequisite clinician involvement could be the creation of specialized clinical trial sites that will interface novel rehabilitation approaches with specific patient groups receiving novel regenerative approaches. As an example, the mission of the Christopher and Dana Reeve Foundation NeuroRecovery Network⁸⁸ is to fast-track the translation of scientific evidence for activity-based therapies into clinical practice in concert with ongoing program evaluation. Essential to the NeuroRecovery Network is the use of: (1) standardized therapy protocols for specific patient populations, (2) standardized outcome measures, and (3) standardized training of staff across the NeuroRecovery Network. Parameters and dosage of locomotor training (current activity-based therapy), patient benefit (eg, physical performance, health, quality of life), population benefit, and rate of recovery are among the many aspects of clinical data collected from the patients. Thus, program evaluation provides data with which to track the progress of developing technologies, problem-solve within the clinical team (eg, physician, therapist, administrator, manager, scientist), and ultimately inform and guide clinical practice.

Although perhaps premature at this stage, the practice of physical therapy, specifically neurorehabilitation, will change as advances continue in brain and spinal cord repair. This change will require new material within academic programs and advanced continuing education to prepare physical therapists for safe and effective implementation of evidence-based rehabilitation protocols in patients having undergone treatments targeting neuroregeneration. Finally, clinicians will need to acknowledge the current state of practice, as patients are exposed to considerable unfiltered information regarding treatment options, benefits, and controversies. A well-informed clinician must have resources to best help their patients to sift through the “hype versus the hope.” This concern is important for all therapists dedicated to best informing their patients, whether or not engaged in regenerative rehabilitation research.