Mesenchymal stem cells as a multimodal treatment for nervous system diseases.

Neurological disorders are a massive challenge for modern medicine. Apart from the fact that this group of diseases is the second leading cause of death worldwide, the majority of patients have no access to any possible effective and standardized treatment after being diagnosed, leaving them and their families helpless. This is the reason why such great emphasis is being placed on the development of new, more effective methods to treat neurological patients. Regenerative medicine opens new therapeutic approaches in neurology, including the use of cell-based therapies. In this review, we focus on summarizing one of the cell sources that can be applied as a multimodal treatment tool to overcome the complex issue of neurodegeneration-mesenchymal stem cells (MSCs). Apart from the highly proven safety of this approach, beneficial effects connected to this type of treatment have been observed. This review presents modes of action of MSCs, explained on the basis of data from vast in vitro and preclinical studies, and we summarize the effects of using these cells in clinical trial settings. Finally, we stress what improvements have already been made to clarify the exact mechanism of MSCs action, and we discuss potential ways to improve the introduction of MSC-based therapies in clinics. In summary, we propose that more insightful and methodical optimization, by combining careful preparation and administration, can enable use of multimodal MSCs as an effective, tailored cell therapy suited to specific neurological disorders.

This concise review summarizes the results of preclinical and clinical trials in neurological diseases of different etiologies. This review focuses on possible mechanisms of action of mesenchymal stem cells (MSCs) but also discusses approaches to augment their effects. A summary of the properties of MSCs reveals their broad therapeutic potential, which can orchestrate regenerative processes after neural injuries.

INTRODUCTION

Injury of the nervous system leads to a cascade of events that eventually ends with neuronal loss and acute or chronic dysfunction. Such processes can be caused by neurodegeneration (eg, amyotrophic lateral sclerosis [ALS], Parkinson's disease [PD], and Alzheimer's disease [AD]), autoimmunological reactions (multiple sclerosis [MS]), ischemia (stroke), mechanical injury (traumatic brain injury [TBI], spinal cord injury [SCI]), and other factors (drug‐resistant epilepsy [DRE], cerebral palsy [CP]). , , These diseases impose serious economic and financial burdens on patients, their families, and society as a whole.

Over several years, myths related to the lack of neurogenesis de novo and lymphatic drainage in the nervous system as well as regarding the immune‐privileged state of the nervous system have been debunked. , Thanks to recent research, the current view of the central nervous system includes a system that is relatively able to recover. As many diseases and injuries of the nervous system are still untreatable or not efficiently curable by standard medical and pharmaceutical practices, alternatives featuring regenerative medicine might overcome existing barriers. Transplantation of cells and tissues into the nervous system, which was first performed in the 1980s, aims to promote regeneration through direct replacement of lost cells. Obtained fetal tissues and implants derived from various sources of neural progenitor cells (NPCs) and neurons still evoke debate around ethical and safety issues. Apart from the source of new neurons, NPCs have also been shown to have immunomodulatory characteristics. Nevertheless, discontinuation of the need for treatment with immunosuppressive drugs that comes with allogeneic treatment diminishes or even removes the positive effects of therapies. Recently, autologous transplantation strategies featuring iPSC technology have appeared. However, clinical translation of this approach is far from realized because the tumorigenic and long‐term immunogenic potentials of these cells have not been tested.

Strategies for treating diseases and injuries of nervous system appoint a less direct but still beneficial source of cells for transplant to cure such yet incurable diseases—mesenchymal stem cells (MSCs). They can be easily obtained from tissues, expanded ex vivo, and transplanted in an autologous or allogeneic manner. Due to their immunomodulative properties, MSCs can resolve inflammation triggered by injury or degeneration. Via their secretome, they can support the survival of neurons and affect the regeneration of tissue loss by influencing local neurogenic niches.

In this review, we introduce unique MSC characteristics valuable for the repair of the nervous system in various diseases based on in vitro and preclinical studies. Taking into consideration the clinical application of MSCs, this review is focused only on the properties of human MSCs from the three most common sources: bone marrow (BM), Wharton's jelly (WJ), and adipose tissue (AT). Clinical studies will be reviewed, focusing on their safety and efficacy. We will also explore discrepancies between clinical studies and suggest potential ways to enhance the effectiveness of MSC therapies.

This narrative review was prepared based on publications found in the PubMed database using the following keywords: MSCs, nervous system, neurodegeneration, and neurological diseases (or each disease specifically, eg. PD, AD, epilepsy, and SCI). For clinical trials, the name of each neurological disease and the term “mesenchymal stem cells” were used as key words, adding “clinical trial” as a filter in the PubMed database. Additionally, clinical trials were filtered from the ClinicalTrials.gov database.

MSCs AND THEIR PROPERTIES

MSCs, which possess self‐renewal potential and multipotent properties, can be found in neonatal and adult tissues. These adherent, fibroblast‐like cells were first isolated from BM in 1970 by Friedenstein et al. Over the years, these cells have been called by many names, such as mesenchymal stromal stem cells, multipotent adult progenitor cells, medicinal signaling cells, and mesenchymal progenitor cells (MPCs). Currently, MSCs is the most common terminology, but is sometimes used interchangeably with mesenchymal stromal stem cells to underline their origin from the nonhematopoietic compartment of BM. In addition, MPCs are occasionally presented as a distinct population.

Apart from BM, MSCs have also been identified in AT, umbilical cord blood, the umbilical cord lining, subendothelial layers, the perivascular zone, WJ, dental pulp, synovial fluid and the synovial membrane, amniotic fluid, fetal liver and even urine or endometrium. Recently, pericytes with MSC‐like characteristics were also found in the brain.

Independent of the tissue source, the isolated cells need to express common characteristics to be defined as MSCs. As no single marker has been specified for these cells yet, analysis of a set of surface antigens needs to be performed. According to the International Society for Cellular Therapy gold standard, MSCs need to be positive for CD73, CD90 and CD105 (all >95%) and negative for CD34, CD45, CD11b/integrin alpha M or CD14, CD79 alpha or CD19, and HLA class II (all <2%). The status of HLA class II can change upon cell stimulation but the expression of costimulatory molecules, such as CD40, CD80, CD86, CD134, and CD142, cannot be changed. , Moreover, the multipotent character of MSCs needs to be proven by their differentiation into adipocytes, chondrocytes, and osteocytes when cultured in vitro. Some studies suggest that MSCs are also capable of transdifferentiating in vitro to cells outside mesenchymal lineages, such as neural and glial cells, cardiomyocytes, skeletal myocytes, hepatocytes, and endothelial cells; however, these studies have been questioned by recent findings. ,

As a distinct entity from the multipotency understood as differentiation potential per se, the term functional multipotency has been coined. This characteristic refers to the ability of different types of stem cells to exert pleiotropic influence on injured tissue to support the maintenance of homeostasis, which remains crucial during development but also during tissue repair after injury. Interestingly, studies have shown that sustaining the stemness of MSCs by incorporating specially adjusted scaffolds can highly augment the therapeutic potential of MSCs in treating spinal cord injuries. Such results should encourage abandonment of the uncertain approach of using transdifferentiated cells.

The most key and characteristic feature of MSCs is their broad secretome, which can influence tissue regeneration. It has been shown that MSCs can produce many immunomodulatory, proangiogenic, tissue remodeling, antiapoptotic, growth, and trophic factors that can support survival of host cells, reconstruction of injured tissue and activation and differentiation of local progenitors. However, depending on the source of MSCs, they can differ in their properties. , ,

Differences between sources

Although MSCs from various sources share common characteristics, some differences can be found between them. These variations in MSC populations may reflect particular regional properties of the niches from which they originate. MSC features are also susceptible to variations between cell culture conditions and isolation protocols. It has been shown that MSCs obtained from the same patient can vary in their properties between isolations. Additionally, discrepancies between different subpopulations of BM‐derived MSCs (BM‐MSCs) have been shown. In the case of Wharton's jelly‐derived MSCs (WJ‐MSCs), depending on the chosen isolation method (isolated enzymatically by collagenase, trypsin, or hyaluronidase, or by extraction directly from explants), cells can slightly differ in features such as the expression of pluripotency markers and cell proliferation rates. However, in some reports, it was noted that autologous MSCs differ from those obtained from healthy donors and that such differences can influence the final outcomes of therapies.

Amable et al. published three studies in which the properties of BM‐MSCs, adipose tissue‐derived MSCs (AT‐MSCs) and WJ‐MSCs were analyzed in different culture conditions (medium supplemented with fetal bovine serum or platelet‐rich plasma [PRP]) or during differentiation. , , They confirmed that the higher proliferative potential of WJ‐MSCs compared to cells from other sources was independent of cell culture conditions. , AT‐MSCs have a moderate proliferation rate, and BM‐MSCs have the lowest proliferation rate. Authors have shown the influence of the cell culture on MSCs secretome. AT‐MSCs and BM‐MSCs cultured in fetal bovine serum (FBS) produced high amounts of extracellular matrix (ECM) components, which was not observed for WJ‐MSCs. Only AT‐MSCs were able to produce collagen (I, II, and III). However, supplementation of medium with PRP compensated for these differences, although AT‐MSCs were still the only producers of collagen II and IV. Independent of cell culture conditions, BM‐MSCs maintained their highly proangiogenic features. AT‐MSCs cultured in PRP had lower secretion than those cultured in FBS while demonstrating the most pronounced proangiogenic characteristics. Amable et al. also showed that in nonstimulated conditions, WJ‐MSCs produce higher amounts of chemokines able to attract a wide range of inflammatory cells (chemokine ligand 5, monocyte chemotactic protein 1 [MCP‐1], and interferon gamma‐induced protein 10) and interleukin (IL)‐6 than other types of MSCs. The high expression of IL‐6 can be crucial in treating liver fibrosis. In PRP‐supplemented media, IL‐6 secretion by BM‐MSCs was significantly increased. In other studies, functional tests with phytohemagglutinin‐activated T lymphocytes or peripheral blood mononuclear cells have shown that BM‐MSCs are the most immunosuppressive cells. This feature has been preserved independently of cell culture conditions. , , Interestingly, Amable et al. also evaluated how the levels of secreted chemokines, cytokines, ECM proteins, proangiogenic factors, and growth factors changed in differentiated cells. Those experiments have shown that some characteristics are preserved, for example, high expression of vascular endothelial growth factor (VEGF) by BM‐MSCs and very low expression of ECM proteins by WJ‐MSCs, whereas some of the characteristics dramatically changed, for example, expression of interleukins or collagen II by differentiated AT‐MSCs. These results show that the differentiation of cells can influence some properties of MSCs. These findings also underlie the importance of cell culture conditions on the final properties of cells that can be later administered to patients.

MSCs IN NEURODEGENERATION AND BRAIN INJURY—PRECLINICAL TRIALS

The previously mentioned features of MSCs make them a perfect tool in cellular therapies for pathological processes in the nervous system driven by excessive inflammation and neurodegeneration. MSCs secrete a variety of factors, including neurotrophic factors. MSCs from different sources can also differentiate into neuronal lineages by forming primary neurospheres; however, only WJ‐MSCs and BM‐MSCs could form secondary neurospheres. Moreover, differentiated WJ‐MSCs secreted more neurotrophic factors than BM‐MSCs and AT‐MSCs.

Routes of MSC transplantation

Preclinical studies have established ways of MSC implementation via intravenous, intra‐arterial, intrathecal, intranasal, intraperitoneal, intraspinal, intracerebroventricular, intracerebral, or direct administration to particular structures. The route of administration is important because it determines the number of successfully grafted cells in the injured site, which can be correlated with therapeutic outcome. , Additionally, taking into consideration that neurological disorders may not be localized, indirect administration, for example, intrathecal injection, may be of great importance.

To understand the pros and cons of each route of administration, an invaluable tool is cell tracking. Various methods have been developed for intravital imaging. Studies have shown that although the most feasible method of MSC transplantation is through intravenous injection, in such conditions, most of the cells become trapped in the lungs. Nevertheless, such entrapped MSCs can release microvesicles and immunomodulative factors and affect the overall state of the patient by modulating peripheral immune cells. Moreover, cell tracking can visualize MSC migratory potential.

Active migration of MSCs toward injury

Once administered indirectly, MSCs need to actively migrate to the injury region. Active migration of MSCs is possible due to the expression of receptors and cell adhesion molecules. Pivotal roles are played by receptors, integrins, selectins and proteolytic enzymes. One of the pathways crucial for MSC migration is the METR/HIF‐1/CXCR4 pathway. It was shown that preconditioning MSCs with stroke patients' sera enhanced the METR/HIF‐1/CXCR4 pathway and increased the migratory potential of MSCs, which translated into improved recovery in a transient middle cerebral artery occlusion (tMCAO) stroke model in rats. Important chemoattractants that can enhance MSCs to regions of injury in the brain are MCP‐1 and stromal cell‐derived factor‐1 (SDF‐1). In a study by Lee et al., it was shown in the MCAO rat model of stroke that MCP‐1 and SDF‐1 have both region‐ and time‐dependent differential expression, which directs intravenously injected MSCs to migrate either to the cortex 1 day after injury or to the striatum in later days. That MSC migration dependent on CXCR4 receptor expression was also shown in elegant in vitro studies with microfluidics systems.

MSCs act through immunomodulation

MSCs are mostly recognized as immunomodulatory cells that can balance inflammation in the tissue environment by upregulating anti‐inflammatory signaling and decreasing pro‐inflammatory signaling and thus regulate immunological cells such as lymphocytes, macrophages or microglia and astrocytes. , MSCs can influence inflammation by secreting soluble factors or direct cell‐cell contact. MSCs constitutively or upon stimulation secrete indoleamine 2,3‐dioxygenase, transforming growth factor beta, hepatocyte growth factor, IL‐6 and IL‐10, prostaglandin E2, heme oxygenase 1 and soluble HLA‐G5.

Inflammation is an integral part of pathological processes that emerge in the nervous system. Depending on the mechanism of injury or neurodegeneration, the innate or adaptive system plays a more important role. Due to their migratory potential, MSCs can migrate to the site of injury and, due to their immunomodulatory properties, decrease inflammation. Excessive inflammation in the brain is the most devastating force causing degeneration of the central nervous system. , Acute injuries due to ischemia (hypoxia‐ischemia encephalopathy [HIE] and stroke), mechanical‐driven trauma (such as TBI) or progressive neurodegeneration lead to moderate activation of microglia followed by activation of astrocytes, the main sources of inflammatory cytokines. If this state persists, damage to the blood‐brain barrier (BBB) occurs, and intensified inflammation appears due to the migration of peripheral immune cells (lymphocytes and monocytes). Such processes suppress neurogenesis and endogenous repair. Thus, MSCs could be a perfect tool in brain injuries, as they can modulate the inflammatory state. It has been shown that MSCs can attenuate microgliosis and astrogliosis in rats with induced HIE, SCI, or epilepsy. , , MSCs can also suppress the proliferation and differentiation of B lymphocytes. Moreover, transplanted MSCs can switch activated M1‐phenotype microglia to the regenerative M2 phenotype. , , Additionally, in an AD model of APP/PS1 double transgenic mice, transplantation of MSCs led to reduced β‐amyloid (Aβ) peptide deposition by microglia but without secretion of proinflammatory factors. , , MSCs have also been shown to improve BBB integrity in a rat TBI model. This effect was mainly mediated by the activity of metalloproteinase inhibitor 3 (TIMP‐3) released by MSCs.

MSCs support regeneration through their neurotrophic activity

Decreasing the inflammation status may contribute to the onset of direct repair of neuronal circuitry and activation of endogenous neurogenesis by MSCs. It has been shown that timing of extinguished inflammation and activation of local progenitors overlap in MSC‐treated mice with induced HIE. MSCs and MSC conditioned media (MSC CM) alone can activate local progenitors in healthy adult rat brains and their differentiation into immature neurons in the subventricular zone (SVZ). However, injection of MSC CM into the brain alone cannot protect the brain against inflammation and has a short‐lasting effect. Moreover, repeated transplantations of MSCs in a D‐galactose‐induced mouse model of cognitive decline have shown functional improvement measured in cognitive tests by enhancing synaptic plasticity and endogenous neurogenesis. This effect was shown to be mediated by activation of the mitogen‐activated protein kinase‐ERK‐CREB signaling pathway in the aged hippocampus. Transplantation of MSCs also rescued long‐term potentiation impairment in aging mice through impacts on electrophysiology.

The neuroregeneration ability of MSCs is also based on the secretion of a wide range of paracrine substances by host cells and MSCs. Several growth factors are secreted by MSCs: brain‐derived neurotrophic factor (BDNF), nerve growth factor, insulin‐like growth factor 1, glial cell‐line‐derived neurotrophic factor (GDNF) and VEGF. , , The neuroprotective function of transplanted MSCs is based on a reduction in neuronal sensitivity to glutamate receptor ligands and altered gene expression, suggesting a link between the therapeutic effects of MSCs and the activation of cell plasticity in damaged nervous structures. Experimental models proved that the MSC secretome promotes axonal growth and neuroprotection and minimizes cavity formation in SCI. , Neurotrophic and neurotropic effects of MSCs were also clearly presented in some elegant ex vivo studies employing adult rat dorsal root ganglia organotypic cultures. , Interestingly, this type of culture can be used to decipher other effects of MSCs on injured tissue, for example, immunomodulation. Lu et al. investigated the nature of chronic scars and their ability to block axon growth. Chronically injured spinal cord axons can regenerate through the gliotic scar in the presence of local growth‐stimulating factors. MSCs may provide a source of growth factors to enhance axonal elongation across spinal cord lesions and minimize cavity formation in SCI. , Interestingly, growth factor secretion, neurogenesis, and survival of stem cells are improved when MSCs are harvested under hypoxic conditions.

MSC modifications and neuronal priming

Despite their innate, multipotent character, MSCs can be even more perfectly tailored to the exact type of injury. This advancement can be achieved by appropriate MSC preparation for the specific environment before transplantation. Up to date, various ways of priming of MSCs with cytokines, hypoxia, pharmacological agents, biomaterials, and other molecules have been established (comprehensively reviewed in Reference 89).

In some neurological disorders, such as PD, ALS, and stroke, gene therapies are proposed as a method of treatment. One of the challenges of such approaches is the delivery of genes of interest, especially in nonfocal neurodegenerations. MSCs can thus serve as carriers for genes whose expression is needed in specific neurological disorders: GDNF (PD and ALS), , VEGF (PD), GDNF and VEGF (ALS), BDNF (Huntington's disease [HD], SCI, and stroke), , , conserved dopamine neurotrophic factor, and PlGF (stroke). It has been shown that modified cells have a more pronounced therapeutic effect than unmodified MSCs. This result stems from the fact that despite the desired alterations, modified MSCs maintain the rest of their characteristics, such as a capacity for immunomodulation. However, more studies are needed to determine whether transient delivery of such growth factors is sufficient or repeated transplantation is needed depending on the treated disease. In contrast, MSCs can also be used to decrease undesirable genes to promote repair. One study has shown that modification of MSCs with lentiviral RNAi downregulating adenosine kinase, the major adenosine‐removing enzyme, may be beneficial for treating epilepsy. Indeed, transplantation of such modified MSCs resulted in a decrease in seizures, and this effect was strictly connected to elevated levels of adenosine. Moreover, modified MSCs transplanted directly into the injury site demonstrated a better ability to promote neuron survival and decrease damage than unmodified MSCs.

Additionally, taking into consideration the impact of miRNAs in the regeneration of tissue, MSCs can serve as carriers of different miRNAs to the nervous system. , However, in some cases, a decrease in miRNAs can be beneficial. For example, in the case of SCI, suppression of miR‐383 enhances the therapeutic potential of MSCs in SCI.

There have also been studies with an established cell line, SB623 hBM‐MSCs (SanBio Inc.), which overexpresses the NOTCH 1 intracellular domain. These cells transplanted in preclinical trials have ameliorated damage in models of PD or after TBI, , probably due to enhanced neuropoietic and proangiogenic activity. , Moreover, in a TBI model, transplanted SB623 cells formed “biobridges” between the neurogenic niche and the site of injury. This effect has been correlated with metalloproteinase‐9 (MMP‐9) activity. Functional formation of these biobridges can play an important role not only in TBI but also in ischemic injuries in the brain. These results can partially explain the regenerative potential of SB623 cells in clinical trials with stroke patients. Currently, a clinical trial is underway in which SB623 cells are being applied to treat patients with TBI (STEMTRA, NCT02416492).

MSCs have been shown to differentiate into many types of cells from the nervous system, including dopamine neurons, acetylcholine‐secreting motor neuron‐like cells, cholinergic‐like neurons, GABAergic neurons, and oligodendrocytes. However, the functionality of these cells, manifested by, for example, the ability of the cells to be engrafted into recipient models and to form connections with existing circuits, remains controversial. Notably, the conditions required to force MSCs to differentiate into neuronal cells necessitate the presence of factors, such as 5‐aza‐deoxycytidine, that are not naturally found in living organisms. On the other hand, priming the MSCs to a neural phenotype may pronounce the therapeutic effect of MSCs alone. The differentiation potential of MSCs can be exploited during their preparation before transplantation. It was shown in a PD model that neuro‐primed MSCs have an enhanced restorative effect. , Such an effect was also observed in MSCs that were primed by overexpression of Lmx1α and neurturin, which are important factors for differentiation and survival of dopaminergic neurons. Another method of genetic priming features overexpression of neurogenin 1 (Ngn‐1) in MSCs. These cells have pronounced therapeutic effects in a mouse model of ALS and brain ischemia in comparison to unmodified MSCs. , Although MSCs overexpressing Ngn‐1 exhibited neuron‐like characteristics in these studies, all transplanted cells vanished within 8 weeks. Additionally, the authors have not yet evaluated the electrophysiological recordings of MSCs after transplantation, which will support their claims. Other preclinical studies have also demonstrated that MSCs are more likely to differentiate into neuronal‐like cells when in the presence of other neuronal cells.

MSCs cotransplantation with other cells

MSCs can also be immunosuppressive in xenotransplantation models. Intrastriatal cotransplantation of syngeneic MSCs with porcine neuroblasts into 6‐OHDA unilaterally lesioned rats resulted in successful neural stem cells grafting in four out of six rats. Beyond changes at the cellular level, motor recovery was also observed due to transplantation. This result suggests that MSCs can be used in xenotransplantation instead of immunosuppressants, for example, cyclosporin A, which cause side effects and are not able to protect long‐lasting grafts. On the other hand, in rat models of PD, it has been shown that human MSCs can evoke inflammation after transplantation. This may be due to differences between the characteristics of syngeneic and xenogeneic transplanted MSCs. These data suggest that the immunomodulatory effect can be inefficient in xenotransplanted MSCs or that other mechanisms were triggered in those studies.

The aforementioned interactions of MSCs with diseased nervous tissue, including modulation of inflammation and neurogenesis by MSCs, seem to be the key to modifying the environment to pursue regeneration (Figure 1). The multimodal activity of transplanted MSCs in models of neurodegenerative diseases has led not only to a pathophysiological view of the course of disease but also, in some cases, to a therapeutic effect, such as better cognitive outcomes in models of AD, better motor activity in models of PD and ALS or a decrease in the amount and severity of recurrent seizures in epilepsy. Understanding which conditions are crucial to boost efficiency by optimizing the route, time and number of administrations of MSC transplantations in preclinical models will improve our knowledge and enhance translation into clinical trials.

FIGURE 1

Therapeutic mechanisms of mesenchymal stem cells (MSCs) in the nervous system. Source: Servier Medical Art, modified. BBB, blood‐brain barrier

CLINICAL TRIALS

Selected clinical trials using MSCs to treat neurodegeneration and brain injury are presented in Table 1. In the numerous presented studies, expanded ex vivo autologous MSCs were transplanted. However, in a clinical trial for CP, neural‐primed autologous MSCs were used, and allogeneic administration of the previously mentioned SB623 cells in stroke patients has been described.

TABLE 1

Selected clinical trials in neurodegeneration and brain injury involving mesenchymal stem cells (MSCs)

Disorder	Trial number	Phase	Type of trial	Cells applied	Amount of cells	Time from onset	Delivery method	Results	Reference
TBI	—	—	Nonrandomized, uncontrolled	Auto BM‐MSCs	1° 1 × 107‐1 × 109 2° 1 × 108‐1 × 1010	Undisclosed	1° at injury site (intracerebral) 2° intravenous (after 4‐12 days)	Safety (CTCAE v3.0); improved scores of BI	123
TBI	—	—	Nonrandomized, uncontrolled	Auto BM‐MSCs	1 × 106	Undisclosed	Intrathecal	Safety; improvements in motor functions (27/73), improvements in consciousness (PVS; 11/24); better results for younger patients and for therapies earlier after injury	145
Stroke	NCT01287936	I/IIa	Nonrandomized, uncontrolled	Allo SB623 (modified BM‐MSCs)	2.5‐10 × 106	7–36 mo	At injury site (intracerebral)	Treatment‐emergent adverse events without sequelae (18/18); improvements in the mean scale scores of ESS, NIHSS, F‐M total score, and F‐M motor function total score	110
Stroke	—	I/II	Randomized, controlled	Auto BM‐MSCs	1‐2° 1 × 108	32‐61 days	1‐2° intravenous (1°‐4 to 5; 2°‐7 to 9 wk after symptoms onset)	Safety; improved scores of BI	133
Stroke	—	—	Randomized, controlled, observer‐blinded	Auto BM‐MSCs	1‐2° 1‐2 × 107	~5 wk	Intravenous (2° 2 wk later)	Safety; increased number of patients with mRS scores 0‐3	134
Stroke	—	I	Nonrandomized, uncontrolled	Auto BM‐MSCs	0.6‐1.6 × 108	36‐133 days	Intravenous	Safety; improvement in NIHSS scores	135
Stroke	—	—	Nonrandomized, controlled	Auto BM‐MSCs	5–6 × 107	7‐12 mo	Intravenous	Safety	127
CP	ChiCTR‐TRC‐12002056	—	Nonrandomized, controlled, observer‐blinded	Auto neural‐primed BM‐MSCs	1‐2° 1‐2 × 107	1–32 yr	Intrathecal (2° 3 wk later)	Safety; motor recovery (better in group with GMFCS levels IV and V)	136
PD	—	—	Nonrandomized, uncontrolled	Auto BM‐MSCs	n/a	14.76 ± 7.56	Intracerebral, sublateral ventricular zone	Safety; modest clinical improvement (UPDRS in “on” and “off” periods)	137
SCI	NCT00695149	I/II	Nonrandomized, uncontrolled	Auto BM‐MSCs	3–5 × 107	<3 wk	Intrathecal	Safety (CESCT); motor improvement (4/5), stabilized condition (1/5)	139
SCI	—		Nonrandomized, uncontrolled	Auto BM‐MSCs	1 × 106/kg body weight	<6 mo; >6 mo	Intrathecal	Safety; improved scores of BI (5/30)	138
SCI	NCT01274975	I	Nonrandomized, uncontrolled	Auto AT‐MSCs	4 × 108	>12 mo	Intravenous	Some adverse effects, slightly improved scores of BI (1/8)	125
SCI	—	I/II	Nonrandomized, controlled	Auto BM‐MSCs	0.7‐1.2 × 106	14‐43 days	Intrathecal	Safety; marked improvement in study group in ASIA scale from A to C (5/11) in comparison to control group (3/20)	142
SCI	—	—	Nonrandomized, controlled	Auto BM‐MSCs	5‐10 × 106/kg body weight	3.6 ± 2.5 yr	1‐6° intrathecal (monthly for 6 mo)	Several side effects; improvement in ASIA scale A patients (12/40)	124
SCI	—	—	Nonrandomized, uncontrolled	Auto BM‐MSCs	1° 1‐4 × 106/kg body weight 2–3° 1‐2 × 106/kg body weight	3‐132 mo	1° at injury site 2–3° intrathecal (up to 21 days after 1°)	Safety; sensory (2/13) and motor functions (1/13) recovery	140
SCI	—	—	Nonrandomized, uncontrolled	Auto BM‐MSCs	1° 8 × 106 2° 4 × 107 3° 5 × 107	Undisclosed	1° at injury site 2° intradural space (up to 21 days after 1°) 3° intrathecal (4‐8 wk after 1 and 2°)	Safety; significant improvement in motor and sensory functional recovery (6/10)	126
SCI	—	—	Case report	Auto 1° BMNCs 2° BMNCs 3° BM‐MSCs	1° 3.2 × 109 2° 5 × 108 3‐7° 1.3‐3.65 × 107	3 mo	1° intravenous 2° intrathecal 3‐7° intrathecal (every 3‐4 mo)	Safety; improvement in ASIA scale A to C/D; sensation level decreased from Th1 to L3‐4; restored trunk control and bladder filling and control; restoration of the spinal cord continuity; improved muscle strength	128
MS	NCT01228266	II	Randomized, double‐blind, placebo‐controlled, crossover	Auto BM‐MSCs	1‐2 × 106/kg body weight	2–10 yr	Intravenous	Safety; trend to lower cumulative number of gadolinium‐enhancing lesions on MRI; nonsignificant decrease of Th1 and modest decrease in Th17 lymphocytes population in peripheral blood	129
MS	NCT00395200	IIa	Nonrandomized, uncontrolled	Auto BM‐MSCs	1.1‐2 × 106/kg body weight	5‐26 yr	Intravenous	Safety; improved visual acuity, contrast sensitivity and visual evoked response latency, increase in optic nerve area	130
MS	—	IIa	Nonrandomized, uncontrolled	Auto BM‐MSCs	Mean: 29.5 × 106	8.6 ± 2.4 yr	Intrathecal	Safety; slight adverse postinjection events due to lumbar puncture; stability (16/22) and improvement (3/22) in the course of the disease measured by EDSS score; halting of T2 lesions development on MRI	131
DRE	NCT02497443	I	Randomized, controlled	Auto 1° BM‐MSCs 2° neural‐primed BM‐MSCs	1° 4–10 × 107 2° 2.7‐8 × 106	15‐32 yr	1° intravenous 2° intrathecal	Safety; significant endpoint improvement vs AEDs only group in monthly seizure frequency, NHS seizure severity score and anxiety score	141
DRE	—	—	Nonrandomized, uncontrolled	Auto 1° BMNCs 2° BMNCs 3‐6° BM‐MSCs	1° 0.38‐1.72 × 109 2° 0.5 × 109 3‐6° 18.5‐40 × 106	8 mo‐6 yr	1° intravenous 2° intrathecal 3‐6° intrathecal	Safety; neurological and cognitive improvement in all patients; reduction in the number of epileptic seizures; absence of status epilepticus episodes; improved motor functions	132

Abbreviations: AEDs, antiepileptic drugs; allo, allogenic; ASIA, American Spinal Injury Association; AT‐MSCs, adipose tissue mesenchymal stem cells; Auto, autologous; BI, Barthel index; BM‐MSCs, bone marrow mesenchymal stem cells; BMNCs, bone marrow nucleated cells; CESCT, Committee on Effectiveness and Safety of Clinical Treatment; CP, cerebral palsy; CTCAE, Common Terminology Criteria for Adverse Events; DRE, drug‐resistant epilepsy; EDSS, expanded disability status scale; ESS, European Stroke Scale; F‐M, Fugl‐Meyer; GMFCS, Gross Motor Function Classification System; L, lumbar region; MRI, magnetic resonance imaging; mRS, modified Rankin scale; MS, multiple sclerosis; NHS, National Health Service; NIHSS, National Institutes of Health Stroke Scale; PD, Parkinson's disease; PVS, persistent vegetative state; SCI, spinal cord injury; TBI, traumatic brain injury; Th, thoracic region; UPDRS, Unified Parkinson's disease rating scale.

The majority of clinical trials have shown the safety of a variety of MSC applications. Severe side effects were noted when cells were transplanted during stereotactic surgery in TBI patients. Nevertheless, most adverse events were correlated with the conducted procedure, and none of them were followed with sequelae. Side effects (spasticity, neuropathic pain, and encephalomyelitis) were also observed in a trial performed by Kishk et al., although the American Spinal Injury Association (ASIA) scales of the patients were rated down from type A to type B. In a study by Ra et al., the authors reported adverse effects following transplantation in all patients enrolled in the study, although their electrophysiological (somatosensory evoked potential [SSEP] and motor evoked potential [MEP]) recordings were not significantly different than before implantation. In most of the studies, the safety and efficiency of therapies were assessed additionally by magnetic resonance imaging (MRI). , , , , , , , , , , , , , , , , No significant anatomical, structural, or parenchymal abnormalities or tumor formation were observed.

In most of the studies, bacterial tests were performed. , , , , , , , , , , , In the majority of studies, karyotype analysis was also carried out. , , , , , , , , Additionally, the differentiation potential of MSCs was assessed in a few clinical studies. , , , , , , Only in a minority of studies parallel testing in animals was conducted. , In some clinical trials, safety evaluation was performed, proving the lack of tumorigenicity of MSCs after administration to immunodeficient mice via subcutaneous or intraspinal transplantations. ,

Transplantation of MSCs resulted in neurologic improvement in the majority of clinical trials. Superior performance in activities of daily living (in TBI, stroke, and SCI) and motor recovery (in TBI, stroke, CP, and SCI) was noted. In clinical studies in groups of patients with SCI, significant improvements were observed in sensory level and motor function, as well as in general outcome with the ASIA scale. , , Tian et al. demonstrated that in TBI patients in a vegetative state, consciousness improved after MSC transplantation. In patients with MS, a tendency toward a decline in active inflammation processes and stabilization of disease progression was observed. , An enormous improvement has been noted in patients diagnosed with DRE, in whom transplantation of autologous MSCs resulted in amelioration of epileptic incidences. This suggests a profound role of MSCs in the repair of epileptic brains. , Modest outcomes have been noted in clinical trials with PD patients, in whom only slight improvements in the Unified Parkinson's disease rating scale scale were obtained. However, most patients claimed subjective marginal improvement of symptoms.

In several studies, slight changes on neuroimaging were noted. Steinberg et al. reported signal changes on T2 fluid attenuation inversion recovery (FLAIR) MRI (13/18) and in the number of contrast‐enhancing areas (15/18) in stroke patients 1 or 2 weeks after transplantation. Moreover, they found a significant correlation between these changes and clinical outcomes at the 12‐month follow‐up.

Slight changes on MRI were observed in a study performed by Bang et al. Researchers noticed less prominent atrophic changes following stroke in the MSC‐treated group than in the control group.

Honmou et al. found a reduction in infarct size on FLAIR MRI in 7 out of 12 stroke patients after treatment. However, because of the lack of a control group in this study, the authors cannot exclude spontaneous recovery as a cause for changes in infarct size. Nevertheless, they also noted a significant correlation between neuroimaging results and mean changes in the National Institutes of Health Stroke Scale score.

In a study by Lee et al., researchers noted a correlation between clinical improvements of stroke patients and SVZ damage defined by diffusion MRI. They noticed a relationship between less SVZ damage and clinical improvement in the MSC‐treated group of patients but did not see this relationship in the rest of the groups (MSC‐treated patients with more SVZ damage and corresponding controls). Moreover, they determined SDF‐1α levels in patient plasma at the time of first transplantation. Their analysis revealed a significant correlation between SDF‐1α plasma levels and patient outcome, defined as scores on the Barthel index and modified Rankin scale.

Currently, there are two ongoing clinical trials concerning the application of autologous MSCs in stroke, one of which is a randomized, controlled, and observer‐blinded trial. The detailed methodology of these studies has already been published. ,

In clinical trials concerning patients with SCI, Park and colleagues reported for the first time improvement on electrophysiological results, assessed by SSEP and MEP recordings, and MRI examination. This improvement may be related to several factors. Direct administration of MSCs into the site of injury is a more effective method for SCI recovery than intrathecal injection. On the other hand, three patients from the group who showed motor recovery had an incomplete SCI with residual neurological function.

In a study by Jarocha et al. concerning one patient with total SC interruption at the Th2‐3 level, muscle strength at the left lower extremity improved from plegia to deep paresis (1° on the Lovett scale). Moreover, the ability to move her lower extremities against gravity supported by the movements in her quadriceps was restored. Neurophysiologic examination including electromyography, electroneurography SEP, and MEP recordings objectively confirmed the improvement. Moreover, MRI demonstrated restoration of spinal cord continuity.

In a study concerning the application of MSCs in MS, Bonab et al. showed that one intrathecal injection of autologous MSCs in patients with secondary progressive MS resulted in stabilization of MRI findings in approximately 70% of participants. As the authors underlined, these results are very appealing in comparison to those gained from standard pharmacological treatments with interferon beta‐1a and mitoxantrone. MRI studies were also performed to separate the “honey‐moon effect” from the real effect of the therapy. Interestingly, the results obtained by Bonab et al. showed that 1 year after transplantation of cells, new lesions started to appear on MRI but without clinical manifestations. This observation may be caused by the diminishing effect of therapy with time and stresses the necessity of repeated transplantations.

In the study by Connick et al., researchers proposed a new approach to measure therapeutic outcome after single intravenous administration of autologous MSCs by assessing functioning of the anterior visual pathway. They showed that visual evoked response latency and an increase in optic nerve area appeared as a result of the improvement in visual acuity from the applied therapy. Despite the important conclusions from the above clinical trials, the lack of a control group of patients is a major drawback of these studies.

In fact, in a randomized placebo‐controlled phase II trial performed by Llufriu et al., therapeutic effects were seen in only groups treated with MSCs. Compared with the placebo group, the MSC‐treated group had changes in the mean cumulative number of gadolinium‐enhancing lesions on MRI and a tendency for lower numbers of Th1 lymphocytes in blood; however, none of the differences was statistically significant. These observations underline the importance of controlled studies with larger groups of enrolled patients, especially for testing MSC therapies in diseases with unstable clinical performance, as is the case for MS.

Only some studies have implemented repeated transplantation of MSCs. , , , , , In other studies, deterioration of the clinical status of patients was noted, which may have been due to the single MSC administration.

As shown above, MSCs ameliorate functional deficits in several central nervous system diseases in both experimental animal models and in the clinic. Therapeutic mechanisms may include neuroprotective effects, immunomodulation, tissue remodeling, and activation of local progenitors. Therefore, MSCs prepare the environment for axonal ingrowth, stimulate angiogenesis, and result in functional recovery.

DISCUSSION AND FUTURE REMARKS

Growing knowledge about MSC regenerative potential raises great hopes for applying them in the clinic. However, there is still space for improvement, and more preclinical studies have to be performed to evaluate cell culture conditions, the potential for neuronal priming, and the timing and route of administration to obtain the best improvements for patients.

First, there is a lack of consistent data showing the optimal source of MSCs for transplantation. As shown in basic studies, MSCs can differ between sources in their regenerative potential, as shown by, for example, the different levels of secreted trophic factors or the propensity of cells to differentiate toward different lineages. However, there are many discrepancies between published data on the properties of BM‐, AT‐, and WJ‐MSCs. Therefore, a comprehensive study is needed to obtain reliable results. Such inquiry could also improve cell preparation methods for clinical trials and define optimal cell culture conditions. Another issue is the reported differences between MSCs from healthy donors and MSCs from diseased patients that could influence the final outcomes of therapies. More detailed studies are also needed to obtain a consensus on this phenomenon.

Second, similar numbers of transplanted cells have led to improved clinical outcomes in neurological disorders. However, there is a lack of comprehensive studies discussing these issues by evaluating different doses in one clinical study.

Another issue is a lack of consensus for the best MSC transplantation method. Intravascularly transplanted MSCs may become trapped in the lungs and liver, from which they can release microvesicles and immunomodulative factors. In some preclinical studies, such entrapment resulted in serious side effects, such as increased risk for iatrogenic atelectasis and lethal pulmonary thromboembolism; however, such observations were made when MSCs were transplanted into healthy animals. In those animals, active migration of MSCs cannot be enhanced, so the cells inertly accumulate in the lungs. In many studies (in stroke and MS), intravascular administration of MSCs despite their presence near the injury area has a significant influence on outcome after treatment. , , This suggests that MSCs can influence inflammation in the brain through the peripheral circulatory system. However, the residual presence of MSCs can also be caused by their migration and poor survival in the nervous system. In contrast, studies have shown the presence of MSCs in the nervous system for up to 8 weeks, which could be long enough to exert therapeutic effects.

Invasive ways of transplantation of MSCs can provoke many side effects and potentially enhance inflammation in the nervous system, which can (a) cause additional changes in the nervous system and/or (b) diminish therapeutic potential. Therefore, intrathecal implementation of MSCs can transplant them directly to cerebrospinal fluid without any significant adverse effects. On the other hand, SCI clinical trials have shown that the best results occur when MSCs are transplanted directly into damaged tissue. However, our clinical trials have shown that intrathecal implementation can also be very successful. , Therefore, the combination of intrathecal and intravenous administration can be considered an effective way to modulate the environment inside the nervous system as well as the adaptive immune response. Importantly, both of these methods of transplantation can be used for repeated injections.

As shown in clinical and preclinical studies, the mainstays of MSC regenerative potential are the capabilities for immunomodulation and secretion of many trophic factors. Therefore, multiple injections can provide continuous stimulation for repair. , , , , , Several clinical trials have shown that repeated transplantations of cells are beneficial for patient outcome, which necessitates the standardization of a reasonable cell injection strategy. In our hands, such a method seems to be intrathecal injection, which is also popular in other studies. , Although some side effects may appear, there is no danger to patient health or life. Moreover, this route of administration has undoubted advantages in being transplanted directly to the nervous system, allowing the cells to easily migrate to regions of degeneration. Nonetheless, because some studies reported complications after this type of administration, such an approach still remains to be evaluated. Less invasive methods of MSC transplantation might be intranasal injections; however, to date, this method of injection has only been evaluated in clinical studies for drug administration.

As mentioned above, the trophic character of MSCs underlies clinical improvement in patients. It was also shown in preclinical studies that MSCs are present up to weeks after transplantation, and this time is sufficient for promotion of neurogenesis in adult mice with HD. The issue, which fortunately is being evaluated, is whether MSCs in fact can engraft and differentiate into neurons and replace the lost ones. As studied in MSCs overexpressing Ngn‐1, modified MSCs have enhanced electrophysiological characteristics, and some of them show neuronal‐like characteristics after transplantation. Nevertheless, it cannot be claimed that MSCs transdifferentiate into neurons. Additionally, the cells disappear after 8 weeks. To date, no study has performed electrophysiological tests on transplanted MSCs with neuronal‐like characteristics. Though studies have claimed that modification of MSCs or priming can enhance neuronal‐like features, calling those cells neurons, without a proper battery of tests, is incorrect.

Another emerging issue is related to the necessity of side‐to‐side clinical and preclinical studies. Such an approach has been implemented in only a minority of clinical studies. , This type of scientific approach is of necessary to not only identify potential improvements but also ensure solid scientific methods are being employed to explain observed phenomena.

It is also worth mentioning that the discrepancy between preclinical and clinical trials can be caused by the selection of patients in clinical trials. Unfortunately, because there are still not well‐standardized procedures concerning MSC transplantation, patients enrolled in clinical trials are usually incurable by every other common method or their disorder has lasted for a long time. The abovementioned reasons can diminish the clinical outcomes in patients in clinical trials in comparison to those seen in preclinical trials. This also addresses the problem of the best therapeutic window. It was shown in vast of preclinical studies that MSC transplantation shortly after injury is most promising for obtaining the greatest therapeutic effect. , This is especially crucial for neurological disorders such as stroke, TBI, SE or SCI, in which there is no time to obtain and expand autologous MSCs from patients. Currently, this approach is not possible in clinical trials. In the future, cell banks can store ready‐made products to be transplanted in an allogeneic manner.

It has been shown in preclinical studies that one of the ways to enhance the positive impact of MSCs on the repair of the nervous system is to modify them with trophic factors. For safety reasons, allogeneic transplantations will be the first choice. In other cases, in vitro priming of autologous MSCs can also be efficient, which has been shown in CP clinical trials. Detailed in vitro studies combined with in vivo testing are needed to evaluate standardized protocols of such priming/modification strategies. The effect of modification must be profound to risk transplantation of genetically modified cells in clinical studies. However, as shown in studies using the stable SB623 cell line, the usage of such cells can be safe and may increase the therapeutic potential of MSCs. , ,

CONCLUSIONS

Published results of MSC transplantation in various neurological diseases, especially in clinical trials, show that MSCs have great potential to improve patient symptoms and quality of life, whereas traditional medicine offers no efficient treatment. However, the design of experiments within clinical studies and preclinical studies leaves vast space for reasonable criticism of the regenerative potential of MSCs. Therefore, more consistent evaluation of MSCs is needed. Taking into consideration the multimodal characteristics of MSCs, studies evaluating their role in repair should also consider such characteristics. More advanced and parallel methods should be involved, such as proteome and secretome studies, in vitro and in vivo functional studies and strictly controlled preclinical studies (Figure 2). As it is difficult to implement a placebo control in clinical trials, especially in pediatric diseases, there is a great need to perform detailed studies in preclinical trials involving advanced methodology and focusing on a deep understanding of the mechanism of the therapeutic potential of MSCs. Therefore, multidisciplinary studies involving clinicians and scientists from various specialties can pave a more reliable way for introducing MSCs into the clinic. We hope that our review uniformly summarizes both the promise and potential routes of advancement of MSCs that can provide information for consideration by clinicians planning future clinical trials.

FIGURE 2

Schematic overview of the most essential aspects of mesenchymal stem cell (MSC)‐based cell therapies in treatment of nervous system diseases. Created with BioRender.com

CONFLICT OF INTEREST

The authors declared no potential conflicts of interest.

AUTHOR CONTRIBUTIONS

B.B., M.S., O.M., and M.M.: manuscript writing, final approval of the manuscript.