Stroke is one of the three major diseases causing the highest lethality and disability rate globally [1]. Current management for acute ischemic stroke consists of intravenous thrombolysis and endovascular recanalization [2]. However, given the short therapeutic window, many patients failed to receive the necessary treatment and developed lifelong disabilities [3]. Current treatments have rarely been effective in neurogenesis and functional recovery during the chronic phase, resulting in an adverse impact on patients’ lives and socioeconomic conditions [4].

Therefore, neuroregenerative approaches are being developed to facilitate the repair of damaged neural networks and reduce the risk of disability from ischemic stroke [5]. Advances in regenerative medicine indicated the possibility of tissue repair and functional improvement. Stem cell therapy is a promising therapeutic strategy for ischemic stroke, owing to the stem cells’ capacity of self-renewal, homing, and multi-lineage differentiation [6]. The stem cells used for this purpose include mesenchymal stem cells (MSCs), neural stem cells (NSCs), embryonic stem cells (ESCs) and induced pluripotent stem cells (iPSCs) [7].

In this review, we focus on MSCs. MSCs are defined as cells that can self-renew and develop the characteristics of mesenchymal tissues [8]. MSCs can be obtained from the bone marrow, muscle, dental pulp, adipose tissue, or umbilical cord [9,10,11,12,13]. The ability of orientable differentiation makes MSCs an ideal cell source for nerve regeneration [14]. Meanwhile, sufficient evidence has shown that MSCs promote recovery through angiogenesis, secretion of neurotrophic factors, inhibition of apoptosis and modulation of the immune system [15,16,17,18,19,20]. MSCs administration can help reconstruct function area [21], promote synaptogenesis and stimulate nerve regeneration [22]. In clinical research, the potential tumorigenic risks limit the utility of ESCs and iPSCs [23,24,25]. The unique immunomodulatory effect of MSCs can modulate inflammatory response caused by ischemic stroke, thereby reducing brain tissue damage. These factors make MSCs the focus of stem cell studies. Here, we review the research progress on MSC therapy, especially the therapeutic mechanisms in functional recovery. We also discuss the experimental evidence and clinical trials on the use of MSCs in ischemic stroke patients and prospect the future direction of MSC research.

The Mesenchymal and Tissue Stem Cell Committee of the International Society for Cellular Therapy (ISCT) defined the minimum criteria for MSCs as follows:

  1. (i)

    Isolated cells showing adherence to plastic in culture.

  2. (ii)

    Cells expressing mesenchymal or endothelial surface markers (CD73, CD90, and CD105) and negative for hematopoietic markers (CD11b, CD14, CD19, CD34, CD45, CD79α and HLA-DR).

  3. (iii)

    Ability to differentiate into osteoblasts, adipocytes, and chondroblasts in vitro [26,27,28].

Recent studies identify several new markers, such as SSEA1/4, CD44, CD146, and CD271. And CD271 is considered one of the most specific MSC markers [29,30,31,32]. These surface markers are related to the stemness within MSCs and contribute to the identification of MSCs in vivo. The identification by markers can improve the purity of MSCs than isolation based on traditional plastic adherence. MSCs differentiate into neural cells by expressing neuronal markers (NeuN and MAP-2) and migrate to brain lesions [33,34,35,36,37]. Meanwhile, MSCs can also play a regenerative role by secreting a variety of paracrine factors, such as vascular endothelial growth factor (VEGF), hepatocyte growth factor (HGF), brain-derived neurotrophic factor (BDNF), glial-derived neurotrophic factor (GDNF), and fibroblast growth factor (FGF) [38,39,40,41,42]. In preclinical studies, MSC transplantation has been found to be a safe and recommended recovery strategy for treatment.

Principles of mesenchymal stem cell (MSC) therapy

Generally, MSC transplantation is carried out by intracranial and intravascular methods. The main intracranial methods are stereotactic injection and intraventricular injection [43]. Stereotactic injection helps transport the MSCs to the infarction area directly. This method requires high precision to reach the target area [44]. Intraventricular injection distributes MSCs to a wider range of cerebral regions. The therapeutic efficacy in such cases depends on the number of cells [45].

Intravascular methods include intravenous and intra-arterial injection. The level of inflammatory cytokines increases after infarction attracting MSCs to the center of the ischemic area [46]. MSCs by intravenous injection are difficult to pass through pulmonary vessels due to the large volume, and few cells are able to reach the target area [47]. Intra-arterial injection through the internal carotid artery can deliver MSCs in a short time. However, intra-arterial injection may cause occlusion of the end arteries in the brain [48, 49].

Intracranial transplantation may cause mechanical damage, but it is better at transporting the cells to the target sites than other methods. Intravascular transplantation requires more cells and is better for large infarctions than intracranial method. In clinical applications, intravascular delivery is easier to perform than the intracranial method.

Intranasal injection, a new transplantation method, needs further research [50]. Cells can be transported from the nasal mucosa to the injured area by blood circulation [51]. Intranasal injection, as a noninvasive method, could be a prospective method of cell transplantation [52, 53].

Recent studies have shown that extracellular vesicles (EVs) derived from MSCs have therapeutic effects comparable to those of direct cell transplantation. EVs can reduce neuroinflammation, enhance angiogenesis, and increase neurogenesis. Compared to MSCs transplantation, EVs show unique advantages in stroke treatment, such as no first-pass effect, ability to pass the blood–brain barrier (BBB), and ability to reduce the risk of cell-related infarction. EVs therapy is a new therapeutic approach for neuroprotection in acute ischemic stroke [54,55,56].

Although there are clinical trials of MSCs in ischemic stroke treatment, there is still no consensus on the optimal dose in cell therapy. Currently, the recommended dose for clinical trials is 1–2 × 106/kg of weight [57, 58]. Furthermore, it is still questionable whether the MSCs dose should be personalized according to the infarct size.

The different phases of pathological ischemic process offer different targets for MSC therapy. In the early phase, MSC delivery may reduce the inflammatory response, regulate the dynamic environment against toxicity, and decrease the injury in the peri-infarct area. At 2–3 weeks after ischemia, late cell transplantation can modulate the reparative processes in favor of angiogenesis and neurogenesis [59]. Clinically, MSCs administration has been found to be safe in stroke populations in early-phase trials [60, 61]. However, the effectiveness of MSC therapy in subacute and chronic ischemic stroke has yet to be validated. Therefore, the optimal timing of administration needs further evaluation.

Therapeutic mechanisms of MSC therapy

MSCs can differentiate into neurons and glial cells to repair structural damage [62, 63]. However, MSCs lack the voltage-gated ion channels expressed in functional nerve cells for generating action potentials [64]. Thus, direct cellular replacement may not be the primary method for achieving a therapeutic effect. Currently, many studies show that the paracrine actions of MSCs exert neurotrophic effects, improving the functional benefit directly or indirectly. Paracrine signaling may be the main condition for the recovery process [65,66,67].

Reparative mechanisms of MSCs mainly include cell migration, angiogenesis, immunomodulation, neuroprotection and neural circuit reconstruction.

Cell migration

Most of the current MSC trials use intravascular delivery methods. Therefore, MSCs need to cross the BBB to migrate towards the target regions. The BBB is composed of endothelial cells (ECs), basal layer, pericytes, and astrocytes. Tight junctions between ECs and membrane transport proteins make the BBB a selective barrier. Influenced by ischemic stroke, the junctions between ECs disrupt, leading to paracellular permeability elevation. MSCs engage with ECs by multistage homing cascade, including selectin-mediated rolling, integrin-associated adhesion, and chemokine-directed migration. MSCs release CXCL-11 and bind with CXCR-3 on ECs. This can activate ERK1/2 signaling and open the tight junctions [68]. Moreover, activating PI3K/Akt and inhibiting Rho/ROCK signaling lead to the disassembling of tight junctions and opening of paracellular pathways for MSCs transmigration [69]. The interactions between vascular EC adhesion molecule (VCAM)-1 and very late antigen (VLA)-4 also regulate the passage across the barrier [70]. However, some studies have found that the transmigration action of MSCs is independent of VCAM-1 [71]. The mechanism and molecular pathway in the passage of MSCs across the BBB need further clarification.

MSCs migration is usually achieved by responding to different chemotactic signals. In the infarct zone, microglia and astrocytes secrete chemokines such as SDF-1 in the surrounding environment. The expression of CXCR-4, as the physiological receptor for SDF-1, increases on the surface of MSCs [72]. The interaction between SDF-1 and CXCR-4 mediates cell migration [73]. Recent studies show that CXCR-7 may exert a synergistic effect with CXCR-4 in promoting MSC migration [74]. Valproate can promote MSCs migration by inducing CXCR-4 overexpression, and lithium can upregulate MMP-9 to enhance chemotaxis [75]. MSCs are mainly transported to the ischemic penumbra and subventricular zone in response to chemotactic signals, including those of MCP-1 and MIP-1a [76].

In addition, c-Met signaling induces MSC migration to the damaged areas [77]. Recent studies have shown that overexpression of neurogenin-1 can increase the homing ability of MSCs and enhance the engraftment efficiency in the ischemic area [78].

Angiogenesis

MSC transplantation can improve revascularization in the ischemic zone, resulting in recovery effects. MSC infusion can enhance microvascular regeneration [79]. Reformation of neurovascular units can be beneficial to neuronal regeneration and functional recovery in ischemic regions [80,81,82,83,84]. The trophic factors secreted by MSCs such as VEGF, induce the formation of immature vessels [81, 85,86,87]. Other growth factors secreted by MSCs, including BDNF, IGF-1, GDNF, bFGF, Ang-1, and Ang-2, probably contribute to enhanced angiogenesis in the ischemic core and border zone [88,89,90,91,92,93,94]. However, VEGF may induce the increased vascular permeability in the BBB causing cerebral edema [95]. In contrast, the anti-edemic effect of Ang-1may counteract vessel leakage [96]. Meanwhile, Ang-1 can specifically combine with Tie-2 on vascular endothelial cells, phosphorylate Tie-2, and promote the maturation and stability of new blood vessels [89]. Ang-1 plays an important role in neovascularization.

In preclinical models, MSCs were found to differentiate into endothelial cells and activate endothelial progenitor cells that can enhance the proangiogenic effect. Notch signaling pathway [97, 98] and mitochondrial nanotube transportation [99] are considered the key mechanisms of MSC-induced angiogenesis. Angiogenesis increases the blood flow in the brain tissue, which is beneficial for endogenous neurogenesis. MicroRNAs (miRNAs) are also important for angiogenesis. Recent trials showed that miRNA-210 was associated with angiogenesis promotion [100].

Immunomodulatory effects

In comparison with other types of stem cells, MSCs have immunomodulatory effects that mediate immune responses. Inflammatory cell proliferation reduced after coculture with MSCs in vitro [101]. Leukocytes gather in the infarct zone after stroke. The strong inflammatory response leads to secondary nerve cells apoptosis [102]. MSC transplantation can modulate the immune response by inhibiting cytotoxic T cells and promoting regulatory T cells [103,104,105]. MSCs suppress maturation and secretion of B-cell antibodies. MSCs reduce the cytotoxicity of immune cells and antibody secretion [106, 107].

In addition to immune cells modulation, MSCs promote immunosuppression by regulating the expression of cytokines. TGF-β secreted by MSCs can block the upregulation of MCP-1 and the infiltration of CD68 + cells [108]. Meanwhile, MSCs attenuate astrocyte reactivity by increasing the expression of IL-10 and decreasing the expression of TNF-α [109, 110]. Inhibition of TNF-α has been shown to limit monocyte maturation, resulting in the lack of antigen-presenting functions in dendritic cells [111]. Although several growth factors and cytokines are involved in MSC-mediated immunomodulatory including IL-6 [112], IL-23/IL-17 [113], MMP2, TGF-β1, HGF, NGF, pGe2, TLR-4, and RAGE [114, 115], the underlying mechanisms have yet to be validated.

Neuroprotection

MSCs can also exert neuroprotective effects, inhibit apoptosis and promote endogenous repair. MSCs increase the expression of neurotrophic factors, achieving neuroprotective effect directly and/or indirectly, such as VEGF, GDNF, BDNF, NGF, IGF-1, HGF, EGF, and bFGF [58, 92, 116, 117]. BDNF interacts with tyrosine kinase receptors promoting neuronal survival [118]. IL-10, as an anti-inflammatory cytokine, induced immune tolerance in preclinical models [119]. MSCs increase the expression of IL-10 [120], which can inhibit microglial activation, reduce inflammatory cytokine expression (IL-1β, IL-6 and TNF-α), and subsequently decrease neuronal degeneration after stroke [112].

In the acute phase of ischemic stroke, the microglia can help reduce brain damage. However, excessive microglia activation or sustained immune response can lead to apoptosis. CXC3CL1 secreted by MSCs has been found to induce the production of the neuroprotective phenotype of the microglia and suppress neurotoxic microglia activity [121]. MSCs can effectively suppress activated microglia and inhibit apoptosis. Apoptotic response to the astrocytes reduced and bFGF expression increased in rat models after MSC transplantation [122, 123]. A recent study showed that MSCs can activate miRNA-29b-3p mediated Akt-dependent anti-apoptotic cascade and inhibit apoptosis [124].

Neural circuit reconstruction

Several mechanisms are involved in MSC-mediated neural circuit reconstruction of the infarct zone and boundary area, including neurogenesis inducement [86], axonal sprout [125, 126], and synaptogenesis enhancement [127].

Endogenous neurogenesis and axonal plasticity may be the basic mechanisms by which MSCs improve the neurological function after ischemic stroke [80, 114]. Axonal sprouting increases the link between different cerebral areas, thus reconstructing neural connections. t-PA activation [125] and PAI-1 downregulation [128] in the boundary area can promote axonal formation and establishment of synaptic connections. In addition, MSCs crosslink peripheral cells, astrocytes, and endothelial cells to repair the BBB [129] and establish a microenvironment promoting neurogenesis and neural circuit recovery.

MSCs can stimulate the production of TGF-β activating the multiplication of endogenous neural stem cells located in the hippocampus and subventricular zone [130, 131]. Neuroblast migration induced by MSCs [114] enhances the survival of cortical cells in the peri-infarct zone and helps repair the neural network [126, 132]. The expression of IL-6 increases in the ischemic penumbra via the nuclear factor kappa-B signaling pathway [133]. This can activate resident stem cells and promote endogenous repair. Meanwhile, angiogenesis induced by MSCs promotes endogenous neurogenesis; miRNA-184 promotes neurogenesis after ischemic stroke [100].

Moreover, the paracrine actions of MSCs can stimulate neurogenesis. MSCs promote the expression of synaptic vesicle protein and BDNF which induce the differentiation of astrocytes and synaptogenesis directly or indirectly [134, 135]. However, the molecular mechanism of MSC-induced neural plasticity requires further research.

Preclinical trials

MSCs have the multidirectional differentiation potential, e.g., osteoblasts, chondrocytes, adipocytes and neurons [62]. The beneficial effects of MSCs include neural lineage trans-differentiation, neurogenesis, angiogenesis induction and synapse formation. Therefore, MSCs are suitable for treating ischemic stroke and preclinical studies.

Researchers found that both allogeneic and heterogenetic MSCs produced a significant recovery in middle cerebral artery occlusion (MCAO) models. In an early preclinical trial, researchers tested the therapeutic efficacy of MSCs in the MACO rat model. Approximately 21% of bone marrow-derived MSCs (BMSCs) were distributed in the MCA territory after intracarotid arterial injection. The treated rats exhibited functional improvement as compared with controls [88]. Then, the same team carried out research on the treatment of rats with ischemic stroke rats using intravenous injection of human MSCs. The researchers observed significant functional recovery in the treated rats. The expression of BDNF and NGF increased in the ischemic tissue, and the level of apoptosis reduced in the penumbral area. The proliferation of endogenous neural stem cells and the formation of new cells occurred in the subventricular area [122].

Although MSCs have the ability to cross the BBB, a limited number of cells can migrate to the ischemic boundary sites. Several studies have shown that despite a small number of MSCs migrating to the injury site, the trophic factors and cytokines secreted by MSCs have a therapeutic effect [136].

Paracrine secretion

After MSCs injection, cells migrate to the infarct zone and differentiate into neuronal, glial, and endothelial cells to enhance neuroplasticity. However, the paracrine action of MSCs can also induce the regenerative process by increasing the level of growth factors or receptors, such as VEGF, NGF, EPOR, TROY, RAGE, and neuropilin. Meanwhile, cytokines or chemokines (IL-13, MMP2, and MIP) are modified after MSC administration [137].

In the ischemic hemisphere, β1-integrin could promote angiogenesis and increase blood supply to the local cortex. In addition, SDF-1, GDNF, and BDNF expression increased significantly after MSC therapy [138]. IGF-1 plays an important role in neurological recovery. The expressions of IGF-1 and IGF-1 receptor (IGF-1R) in MSC-treated rats significantly increased in the ischemic brain tissue. IGF-1 is associated with neurogenesis due to MSC transplantation [139].

Genetic modification with exogeneous cytokines can enhance the roles of MSCs. In a preclinical trial, MSCs transfected with the Ang-1 gene led to significantly more functional recovery than uninfected MSCs in MCAO rat models [89]. To enhance the recovery effect, the research team transfected MSCs with the Ang-1 and VEGF genes. The Ang-VEGF-MSCs showed the greatest structural–functional recovery in all groups [140]. Exogenous gene transinfection has been suggested to enhance the therapeutic effect of MSCs. Combined gene transfection in MSC therapy represents a new strategy.

Angiogenesis and neurogenesis

The multidirectional differentiation potential of MSCs causes their trans-differentiation into endothelial cells and neural cells. Angiogenesis and neurogenesis constitute the mechanisms of structural repair. In rats with ischemic stroke, MSC transplantation may produce functional recovery by inducing angiogenesis [80]. The expression of endogenous growth factors increased after MSCs transplantation and induced the formation of small vessels in the infarct boundary, including VEGF, EGF, and bFGF [90, 141]. The infarct volumes of MCAO rats reduced after MSC therapy. MSCs induced the proliferation of subventricular zone cells, which may promote endogenous neurogenesis [92]. It significantly reduced the mortality of the rats and facilitated behavioral and neurological recovery. MSCs promote the reconstruction of the neurovascular units and recovery of brain function [142].

Immunomodulation and neuroprotective effect

The immunomodulation and neuroprotective effects of MSCs are the basic mechanisms that reduce secondary brain injury after ischemic stroke. MSCs suppress the activation of microglia and delay neuronal death [143]. Besides suppressing microglial activation, MSCs induced an increase in IL-10 expression and reduction in neuronal apoptosis in the peri-infarct area of MACO rats [144]. MSCs can inhibit the production of CD4+ and CD8+ T cells and promote the production of regulatory T cells [145]. Meanwhile, MSCs can significantly inhibit expressions of Bax, caspase-3, IL-18, TLR-4, and PAI-1 [114, 146].

Heterogeneity

MSCs from different sources share morphological, regenerative, and immunomodulatory characteristics [147]. However, these cells show variations in other features such as paracrine functions and neurogenic potential. Adipose-derived MSCs (AD-MSCs) secrete more VEGF, HGF, and TGF-β [148, 149] than BMSCs. Meanwhile, the conditioned media of AD-MSCs have higher levels of MMP3 and MMP-9 and enhance angiogenesis [150].

Clinical trials

Safety is the primary concern in stem cells application in clinic. Currently, almost all studies showed no cases of acute toxicity, thromboembolism, abnormal cell growths, neurological deterioration, or death after MSC transplantation. Transient febrile reaction was the only side effect related to cell administration [151, 152]. MSCs are easy to obtain and proliferate. The immunomodulatory characteristics and potential of nerve regeneration make MSCs the ideal candidate for clinical therapy.

Safety and efficacy

In a study of autologous BMSCs, 30 patients with ischemic stroke were divided into an MSC group (n = 5) and control group (n = 25). The patients in the MSC group received an intravenous infusion of 1.0 × 108 cells. During the 12-month follow-up period, the Barthel index and modified Rankin Scale (mRS) score of the MSC group improved consistently as compared to the scores of the control group [153]. This shows the safety of MSC transplantation and improvement in patients’ neurological condition. Another clinical trial evaluated the safety and feasibility in stereotactic implantation of autologous BMSCs. This study recruited five patients with cerebral infarction. No adverse events occurred in this study, and improvements were observed in the neurological recovery [154]. However, due to the limited sample size of the two trials, it is difficult to obtain conclusive results.

In a recent randomized controlled trial (RCT), 16 of 31 patients accepted autologous MSCs therapy. During the 2-year follow-up, the MSCs group showed significant improvements in motor National Institute of Health stroke scale (NIHSS), motor Fugl-Meyer score, and task-related fMRI activity in the primary motor cortex. The result shows that MSC treatment for subacute ischemic stroke was safe and feasible. MSCs improved functional recovery via sensorimotor neuroplasticity [155].

Long-term efficacy

Some researchers explores whether MSCs can maintain their therapeutic effect over time. In a 5-year follow-up clinical trial, researchers randomly allocated 85 patients to the MSC group and control group. The MSC group received intravenous autologous MSCs. They were followed for 5 years; finally 52 patients were examined. Compared with the control group, in the MSC group, mRS score decreased and the number of patients with mRS score 0–3 increased significantly. The clinical recovery may be associated with SDF-1 serum levels and subventricular region involvement of the lateral ventricle [156].

Efficacy of different transplantation routes

One study examined the effectiveness of different administration methods used in MSC transplantation. The researchers found that intra-arterial infusion led to higher biological distribution than intravenous delivery. This study also assessed the safety and efficacy of catheter delivery. They recruited four patients with stroke (ischemic stroke three; hemorrhagic stroke (1). The patients received a single dose of 2 × 107 umbilical cord MSCs. The researchers infused the cells to the M1 segment of the MCA via catheterization. Muscle strength and mRS score improved in the two ischemic stroke patients. However, these two patients experienced ischemic stroke again at 3 and 6 months after MSCs infusion [157], probably due to the short-term effects of stem cell therapy.

Different sources of MSCs for cell therapy

Commonly, BMSCs are selected for trials, but the procurement is difficult and the cell number is limited. Compared to BMSCs, AD-MSCs can be obtained by relatively safe methods such as liposuction procedures. This makes AD-MSCs an attractive resource for clinical applications [158].

Intravenous administration of allogeneic AD-MSCs could be a safe therapy for early stages of acute stroke. A clinical trial researching the safety and efficacy of allogeneic MSC transplantation recruited 20 patients with acute ischemic stroke. The enrolled patients were equally divided into two cohorts treated with allogeneic AD-MSCs or placebo. They received a single dose intravenously within the first 2 weeks after symptom onset. During the 2-year follow-up period, the researchers recorded the mRS score, NIHSS score, infarct size, and levels of biochemical markers for efficacy analysis. The result showed that allogeneic AD-MSCs are beneficial for cerebral recovery [159].

Currently, autologous MSCs are the preferred option because of the low risk of rejection. Due to the long culture time, autologous MSCs can hardly be used in the acute phase of ischemic stroke treatment. Allogeneic MSCs could break the limitation. Because MSCs lack HLA-II antigens, allogeneic MSCs could hardly cause an immunologic response, thus eliminating the risk of rejection [160, 161].

In another clinical trial, 15 patients in three groups were treated with allogeneic MSCs. The three doses of 0.5, 1.0, and 1.5 × 106/kg were found to be safe. Then, 21 patients, as an expanded safety cohort, received allogeneic MSCs at a dose of 1.5 × 106/kg. The Barthel score and number of patients achieving excellent functional outcomes in this cohort increased significantly over the 12-months of follow-up. The result supported the fact that transplantation of allogeneic MSCs was safe and effective. Allogeneic MSCs are promising candidates for MSC therapy in acute ischemic stroke [162].

Genetic technology and preconditioning

Gene transfection-induced MSCs may have high therapeutic value. A phase I/IIa study explored the clinical outcomes of modified BMSC transplantation. SB623 cells, as allogenic modified BMSCs, are transinfected with a plasmid coding for the intracellular domain of Notch-1. The study enrolled 18 patients with chronic stroke and divided then into three groups that received stereotactic single doses of 2.5, 5.0, and 10 × 106 SB623 cells. In all, 16 were followed up for 12 months. Comparing to the baseline, significant improvements were observed in the European Stroke Scale (ESS), NIHSS, and Fugl-Meyer score. Thus, SB623 cells were found to be safe and effective for clinical use [163].

Preconditioning with stroke serum before transplantation can activate MSCs into a primed state and reinforce resistance to ischemic microenvironments. The cytoprotective effect of preconditioning enhances the migration and survival of MSCs [164]. In an autologous MSC transplantation study, 12 patients received intravenous auto serum-expanded MSCs 36–133 days after stroke. In the first week of infusion, the median daily rate of NIHSS change increased. The MRI results showed that the mean ischemic area decreased to over 20%. Although the study did not rule out the placebo effects and the effect of natural recovery on treatment outcomes, the result suggested the role of relevant preconditioning in improving the efficacy of MSCs [165].

These studies support that MSC transplantation is safe and effective in cerebral recovery after ischemic stroke. In future, determining the timing of administration and optimum infusion dose could be the predominant challenge in MSC therapy [93, 117, 166].

Conclusion

Mature neurons cannot proliferate and differentiate. Neurological impairment caused by ischemic stroke has been considered difficult to treat. Currently, the capacity of MSCs in neuronal differentiation and functional recovery in ischemic therapy has generated immense interest. The general mechanisms by which MSCs induce improvement include cell replacement, angiogenesis, paracrine actions, neuroprotective effect, immunomodulation and neural circuit reconstruction. However, the molecular pathways of neurogenesis and angiogenesis need further examination.

In recent years, AD-MSCs have become popular in stem cell therapy due to the extensive tissue sources, short culture time and strong ability of differentiation into neural-like cells. Future studies need to focus on the safety and efficacy of MSCs and monitor adverse events before widely using them in clinical practice. Furthermore, large clinical trials are still limited, the optimal timing and doses for MSC transplantation are not yet known. In summary, MSC therapy provides a new therapeutic approach and research direction in the neurological recovery of stroke patients.