Novel Therapeutic Opportunities for Neurodegenerative Diseases with Mesenchymal Stem Cells: The Focus on Modulating the Blood-Brain Barrier

1. Introduction

The evidence indicates that longevity is currently increasing worldwide and the occurrence of cumulative disabilities is the price to pay for living longer [1]. It is widely recognized that aging is a natural, progressive, and inevitable process that occurs in all organisms, although the functional and morphological changes affecting the body tissues and organs during its progression are highly variable [2]. Specifically, aging-associated changes in the central nervous system (CNS) are of crucial relevance, and the prevention and treatment of neurodegenerative diseases represent one of the greatest challenges for modern societies. However, therapeutic options to treat CNS-related diseases are very limited, mainly due to the fine-tuned status of the brain, the complexity of neurological diseases, and the lack of knowledge of their etiology and pathophysiology.

An additional challenge in this scenario comes from the singular presence in the CNS of the blood-brain barrier (BBB), a complex, dynamic, and structured network of cells and proteins responsible for protecting the brain and regulating the transport of substances and cells from the peripheral circulation to the CNS. Because keeping the integrity of BBB is critical for maintaining a constant environment in the CNS in healthy conditions, investigating how this structure changes during aging and specifically in pathological conditions, has lately received special attention worldwide. Although some researchers have demonstrated minor and variable BBB leakage in healthy aging without immune infiltration and neuropathological signs [3], evidence indicates that BBB integrity is compromised in most neurological diseases, including those associated with aging [4,5,6,7,8,9,10,11]. Therefore, BBB disruption emerges as a key innovative element to design new therapeutic approaches to treat neurodegeneration during aging and neurological disorders.

Recently, stem cells isolated from adult mesenchymal tissues (MSCs) have emerged as attractive candidates for the treatment of aging-associated neurological diseases [12,13]. While many studies have explored the involvement of the neuroprotective, neurotrophic, and immunomodulatory capacities of MSCs in their therapeutic actions in neuroinflammatory and neurodegenerative disorders, the potential and additional role played by MSCs in improving the sealing and modulation of the BBB has been scarcely addressed.

This review aims to examine the structure and functions of the BBB and how its impairment, along with changes in transporters, extracellular matrix, and cell-junctional components, influences the onset and progression of several neurodegenerative disorders. We will specially focus on four neurodegenerative disorders, namely Alzheimer´s disease (AD), Parkinson´s disease (PD), multiple sclerosis (MS), and stroke. These disorders exhibit different clinical and pathological signs, and although their etiologies remain mostly unknown, they all course with neuroinflammation and neurodegeneration, are significantly associated with aging, and display profound alterations in cerebral vasculature and microvessels components. Importantly, they all positively respond to MSC-based therapies. Our review will also delve into the molecular mechanisms involved in the MSCs-treatment of neurological pathologies, with special attention to their ability to restore the integrity of the BBB. Finally, we will discuss how to further improve MSCs therapies by specifically targeting BBB.

2. Structure and Role of the Blood-Brain Barrier

The BBB is a complex and dynamic microvascular structure composed of several types of cells, which maintains the homeostasis of the CNS by regulating the supply of molecules and filtering potentially harmful compounds from the bloodstream to brain tissues and back [14]. Its main scaffold is structured in a neurovascular unit (NVU) that is shaped by a tight layer of brain microvascular endothelial cells (BMVECs) surrounded by astrocyte end-feet and pericytes, all of them embedded in an extracellular matrix network and the basement membrane [15] (Figure 1).

The BMVECs are highly specialized endothelial cells that show unique structural and biological properties compared to peripheral endothelial cells [16], including low pinocytic activity, absence of fenestration, low levels of leukocyte adhesion molecules, high expression of intercellular junctions, mainly tight-junctions (TJs) and adherent-junctions (AJs) [17,18], and increased number of mitochondria to supply the energy that is required for the active transendothelial transport of molecules [19]. Additionally, BMVECs regulate BBB permeability by mainly controlling the intercellular and intracellular transport of cells and molecules via cellular junctions and specific membrane carriers, channels, and transporters [20]. Moreover, these cells present a very remarkable apicobasal polarity, based on a differential membrane composition (lipids, glycoproteins, receptors, and transporters) between the luminal and the abluminal sides [21]. For instance, enzymes like γ-glutamyl-transpeptidase [22] or alkaline phosphatases [23] are located in the luminal face of the endothelium, while Na⁺-K⁺ ATPase [24] and the Na⁺-dependent neutral amino acid transporter [25] are present at the basal membrane.

BMVECs are wrapped by pericytes which contribute to the regulation of endothelial cell proliferation, survival, migration, differentiation, vascular branching, and blood flow control [26]. Smooth muscle cells are also found around large vessels (arteries, arterioles, venules, and veins), providing strength and elasticity, and playing a significant role in basal tone maintenance, blood pressure and blood flow distribution [27,28]. Pericytes are localized along capillaries and imbedded within the BM. They regulate blood flow, modulate immune and phagocytic responses after brain injury, and promote angiogenesis in the adult CNS [29,30].

Astrocytes are the major glial cell that enfolds the endothelium of the BBB [31]. They connect to BMVECs through their end-feet, contributing to BBB consistency and determining its properties [30]. Astrocytic end-feet contain a set of proteins that interact with the vascular tube, such as dystroglycan-dystrophin complex or aquaporin 4 (AQP-4), among others. The latter is critical for regulating water homeostasis in the CNS, while the dystroglycan-dystrophin complex links the astrocytic skeleton to the BM [32]. Since astrocytes serve as the cellular linkage between the neuronal circuitry and the vascular system in the CNS, they release signals that regulate the blood flow in response to neuronal activity. For instance, they control the contraction and dilation of SMCs and pericytes [33]. Additionally, they contribute to the formation of endothelial cell TJs through vascular endothelial growth factor (VEGF)-mediated signals [34] and regulate tissue inhibitor metalloproteinases (TIMPs), which maintain the balance between deposition and degradation of the extracellular matrix components [35].

The basement membrane (BM) is a highly organized sheet composed prominently of extracellular matrix proteins (collagen IV, laminins, nidogen, and perlecan). It plays an important role in providing structural support, cell anchoring, and signaling transduction [36]. Two types of BM of the BBB have been characterized: the inner vascular BM, secreted by BMVECs and pericytes, which contains laminins α1 and α2; and the outer parenchymal BM, secreted by astrocytes, which contains laminins α4 and α5 [37,38]. Nevertheless, the BM is largely understudied, in comparison to the cellular components of BBB, probably due to its intrinsic complexity and the lack of research tools [39].

The exchange of molecules and cells across the BBB requires specific transporters, channels, and receptors. Two major mechanisms, named paracellular and transcellular transport have been identified. Table 1 describes some of the main transporters. Paracellular transport is the main pathway used for the exchange of small hydrophilic substances. On the other hand, the transcellular transport involves the movement of molecules through the cell membrane of BMVECs. It occurs via several mechanisms depending on the nature of the molecule, such as passive diffusion, facilitated diffusion, active transport, and receptor-mediated transport. Importantly, gases and small lipophilic molecules do not require transporters to cross the BBB and freely diffuse across the endothelium [40].

Both, paracellular and transcellular processes are precisely regulated by junctional structures, mainly TJs, with some contribution from gap junctions and AJs [41]. TJs are intricate structures located along the membranes of adjacent BMVECs, intermingled distributed with AJs, which provide stability and consistency to the BBB [42,43] (Figure 1). Among TJ proteins, claudin-1, -3, -5, -12, and occludin control the transportation of solutes and ions [44]. These proteins are associated with the cytoskeleton, primarily based on actin and vinculin, through scaffolding proteins, such as ZO-1, -2, and -3. Moreover, dystrophin acts as a scaffold protein that mobilizes actin and vinculin proteins [45]. AJs create inter-endothelial contact connections maintained by proteins such as VE-cadherin and platelet endothelial cell adhesion molecule-1 [46,47] (Figure 1), that contribute to the continuous crosstalk with TJs for paracellular transportation. Similarly to TJs, AJs are attached to the cytoskeleton, contributing to the regulation of cellular transportation of lymphocytes, monocytes, or neutrophils [48,49,50]. Pericyte-endothelial junctions also contain cadherins, with N-cadherins forming homophilic interactions between pericytes and BMVECs, thereby maintaining vascular integrity [51]. Furthermore, BMVECs interact with the BM stablishing AJs via α- and β-integrin receptors, which are transmembrane glycoproteins involved in the extracellular matrix connection to the endothelial cytoskeleton [44]. Conversely, gap junctions serve as intercellular channels facilitating cytoplasmic connections between neighboring cells, enabling selective communication of molecules primarily dependent on molecular size, driven by passive diffusion [52]. In the brain, endothelial cells express the gap junctions connexin 37 (Cx37), Cx40, and Cx43 [53], while astrocytes express Cx30 and Cx43 [54]. In addition to their channel function, certain connexins also play a regulatory role in the expression of other junctional molecules, such as Cx43 interacting with N-cadherin [55].

3. MSCs as a Therapeutic Option in CNS Diseases

Despite notable advancements in the management of symptoms of neurodegenerative diseases such as AD, PD, MS, and stroke, with treatments that enhance quality of life and increase lifespan, the available drugs only slow the progression of neuronal death. Given the multifactorial and complex nature of these diseases, the primary causal agent remains unclear, and it is imperative to develop multi-target therapies that address the different causes/consequences of these disorders such as neuroinflammation, neuronal cell death and dysfunction, and BBB disruption.

MSCs are emerging as one of the most promising cell therapies against different immune-mediated diseases due to their unique properties. MSCs are multipotent cells able to differentiate into mesodermal lineages (fibroblast, osteocyte, adipocyte, and chondrocyte) and in some cases, into endodermal or ectodermal (neuronal) fates [69]. The scarce expression of the major histocompatibility complex and other co-stimulatory molecules makes MSCs immune-privileged cells. This immune status allows MSCs to be used in an allogenic manner without requiring additional immunosuppression [70].

The International Society for Cellular Therapy has defined MSCs based on their expression of CD90, CD73, CD105, and CD44 while lacking the expression of CD45 and CD31 [71]. These markers help distinguishing MSCs from other cell types and are used to identify and isolate these cells for research and therapeutic purposes.

In adults, several tissues act as MSCs reservoirs [72]. The first type of MSCs to be described were bone marrow-derived mesenchymal stem cells (BM-MSCs) [73], making BM the primary source for MSCs isolation. However, the process of obtaining BM-MSCs involves a highly invasive and painful procedure that requires anesthesia, posing a risk of infection [74]. Alternatively, adipose tissue-derived MSCs (ASCs) can be isolated from biological material generated during liposuction or lipectomy after medical interventions. The natural abundance of MSCs in adipose tissue, which is approximately 500 times higher than in BM, accompanied by easier isolation, has led to an increased utilization of ASCs [75]. Additionally, a recent study has demonstrated that ASCs exhibit lower immunogenicity and transcriptomic heterogeneity compared to BM-MSCs [76]. Apart from adult tissues, MSCs can also be derived from birth-associated tissues, such as the umbilical cord Wharton's jelly (WJ). WJ-MSCs have emerged as an ideal source of MSCs for therapy due to several advantages: they can be harvested painlessly in abundance without causing donor site morbidity, are easy to isolate and culture, possess a high proliferative rate and retain their stemness properties in vitro [77].

In the context of neurodegeneration, there is growing interest and promise in therapies based on MSCs. As of July 2023, 249 clinical studies were found on www.clinicalstrials.gov [78] with the terms “Nervous System Diseases” and “Mesenchymal Stem Cell”, reflecting the potential of these cells in addressing neurodegenerative disorders. A summary of trials focused on AD, PD, MS and stroke identifying the tissue for MSC isolation, donor, route of administration, and pathological target are shown in Table 2.

The main mechanisms exerted for MSCs that contribute to their potential efficacy include:

• Neuroprotective effect: MSCs have demonstrated to have an important neuroprotective effect, as they secrete neurotrophic growth factors such as glial cell-derived neurotrophic factor, VEGF, brain-derived neurotrophic factor and nerve growth factor (NGF) [79], as well as anti-apoptotic factors like Bcl-2 [80]. These factors enable MSCs to promote nervous regeneration, inhibit neuronal apoptosis, and induce endogenous neurogenesis. For example, Oh et al. [81] demonstrated that intravenous injection of MSCs increased hippocampal neurogenesis and differentiation of neural progenitor cells into mature neurons in Aβ-treated mice (AD model) by augmenting the Wnt signaling pathway. Additionally, MSCs may inhibit stroke-associated apoptosis through the Bcl-2 pathway in neurons and astrocytes from rats [82]. Besides, MSCs can transfer healthy mitochondria to damaged cells, protecting neural stem cells from neurotoxic agents. MSCs may transfer this organelle in various ways, including gap junctions, cell fusion, microvesicles, and through tunnelling nanotube formation [83]. Mitochondria play a crucial role in maintaining metabolic homeostasis, and defects such as membrane leakage, electrolyte imbalances, activation of pro-apoptotic pathways, and mitophagy have been implicated in the pathogenesis of various CNS disorders [84]. It has been demonstrated that the ability of MSCs to transfer healthy mitochondria to damaged cells protects neural stem cells from neurotoxic agents [85], and has garnered significant attention in the field of cellular therapy for CNS disorders.
• Immunomodulatory role: MSCs can interact with the immune system and participating in both innate and adaptive immunity due to their significant immunoregulatory functions. This indicates that, depending on the environment in which MSCs are introduced, they can modulate the response. Thus, in an inflammatory environment, MSCs exhibit anti-inflammatory behavior. By expressing different molecules such as transforming growth factor β, indoleamine 2,3-dioxygenase, prostaglandin E2, nitric oxide, and interleukin-10 (IL-10), they can interact with immune cells either through direct cell-to-cell contact or via paracrine activity [86,87,88,89,90]. MSCs can also modulate the macrophage/microglia polarization upregulating the ratio of anti- versus pro-inflammatory responses [91], suppress Th1 and Th17 responses, enhance the maturation of DCs from monocytes, and enhance the Th2 response and the generation of Forkhead Box P3 positive Treg population. Moreover, some studies reported that the secretion of IL-6 by MSCs can inhibit astrocyte apoptosis, increase the neuroprotective population of astrocytes, and reduce neuron damage post-injury [92].
• Regulation of protein clearance: treatment with MSCs has been shown to induce the secretion of neprilysin in vitro and in vivo, improving the endogenous machinery for the degradation of Aβ-plaques and enhancing the clearance of these aggregates [93]. This is particularly relevant as abnormal protein aggregation is one of the major hallmarks of neurodegenerative diseases like PD and AD [94].

4. MSCs as Promising Modulators of the BBB in Neurodegenerative Diseases

As described before, while it remains unclear whether BBB disruption is a cause or a consequence of neuroinflammation, it is undoubtedly a crucial component of CNS pathologies. Unfortunately, the BBB is often viewed as a challenge that hinders the delivery of drugs to the CNS and reduces the efficacy of conventional treatment approaches for neurodegeneration, Therefore, the potential of pharmacological interventions targeting the BBB could represent a promising therapeutic strategy for the neuroinflammatory-mediated neurodegenerative diseases [12,13]. Besides the neuroprotective and immunomodulatory roles of MSCs in neurodegeneration, recent reports have pointed out a beneficial effect of MSCs on modulating the disrupted BBB.

In general, the delivery of MSCs to the CNS is highly diverse, although systemic administration, particularly intravenous infusion, is the preferred method (Table 2). When MSCs are infused intravenously, they transiently accumulate in the lungs for 1-3 hours, followed by a gradual movement to other tissues such as the liver, spleen, kidney, and bone marrow [95]. Interestingly, MSCs have shown the ability to reach brain vessels and adhere to them after 6 hours post-injection, according to Rüster et al. [96], thanks to specific interactions with endothelial cells through adhesion molecules such as P-selectin and VCAM-1/VLA-4. While some reports indicate that after a middle cerebral artery occlusion (MCAO) model of stroke, injected MSCs accumulate in the vessels of the infarcted region [97,98], other studies describe no MSCs were detected in the cerebral parenchyma after an intra-arterial injection in an Alzheimer’s disease mouse model [99]. Therefore, although MSCs are capable of rolling and executing a coordinated extravasation through activated endothelia in other tissues, allowing them to access sites of damage [96], it remains unclear whether this cellular therapy can cross the BBB and exert its function within the brain tissue.

Nevertheless, despite their uncertain ability to cross the BBB and penetrate to the brain, MSCs can directly contact with endothelial cells in the damaged area. This interaction likely enables MSCs to exert their paracrine functions to other cells of the NVU and the BBB from this location (Figure 1). In fact, the scientific community is currently exploring two interesting derivatives of MSCs: genetically modified MSCs and the use of MSC-derived extracellular vesicles (MSC-EVs). Genetically modified MSCs provide the opportunity to enhance the therapeutic effect of MSCs by improving their inherent functions or enabling them to synthesize drugs or active compounds. On the other hand, direct use of the secretome in the form of MSC-EVs improves the penetration through the BBB. In the following sections, we will provide examples of these strategies and the cellular and molecular beneficial effects exerted by MSCs in preclinical models of neurodegenerative diseases characterized by a severe disruption of the BBB (Figure 2, Figure 3 and Figure 4).

4.1. Alzheimer´s disease

Alzheimer´s disease (AD) is a progressive neurodegenerative disease characterized by a cerebrovascular and neuronal dysfunction, resulting in a gradual decrease in cognitive functions [100]. In 2019, it was estimated that 50 million people suffered from AD [101]. The principal pathological hallmarks are extracellular amyloid-β (Aβ) deposition and neuronal accretion of phosphorylated tau-forming neurofibrillary tangles [102]. Aβ is a proteolytic by-product derived from the amyloid precursor protein produced by several cleavages via β- and γ-secretases. There are two main types of AD: early onset AD, which is a rare form affecting <1% of AD cases in subjects <65 years-old and is caused by genetic mutations, and late onset AD, which is the most frequent form, and primarily affects to patients >65 years-old. While genetic factors may potentially contribute to its development, specific mutations that directly cause an increase of proteolytic cleavage in patients have not been observed [103].

4.1.1. Dysfunctional BBB in AD

BBB dysregulation plays a significant role in the pathogenesis of AD, affecting various components of the NVU (Figure 2).

For instance, the presence of Aβ disrupts the organization of TJs and AJs (i.e, occludin, claudin-5 and ZO-1) in BMVECs, leading to compromised barrier activity [104,105]. Additionally, the reduction of GLUT-1 in cerebral microvessels of AD patients contributes to vessel degeneration and further exacerbates the disease [106,107]. Conversely, patients with mild cognitive impairment, which is the precursor of AD, display increased BBB permeability that correlates with high levels of soluble platelet-derived growth factor receptor β in the cerebrospinal fluid, which is indicative of pericyte damage [108]. In fact, the decreased number of pericytes in AD patients may worsen the accumulation of Aβ both in brain parenchyma and blood vessels.

On the other hand, astrocytes in AD patients showed reduced expression of AQP4 in perivascular end-feet and increased levels of astrocytic activation markers [109]. In fact, the accumulation of Aβ in brain leads to pericyte degeneration and loss, dysregulated BM and astrocytic end-feet depolarization with loss of AQP4, which will decrease the Aβ clearance, fueling a pathogenic feedback loop. Thickening of BM [110] and increased collagen levels in these structures [111,112] are also common features in AD patients. MMP2 and MMP9 are significantly activated in the NVU, contributing to BM remodeling in AD [113,114].

Furthermore, transporters of the BBB, such as RAGE, LRP1, or P-gp, are key elements in the regulation of the Aβ clearance. In fact, analysis of microvessels and BMVECs in postmortem AD brains, showed high expression of RAGE, which mediates Aβ entry into the brain [115], and reduced expression of LRP1 and P-gp, involved in the clearance of cerebral Aβ [116,117,118]. The immune system is also compromised in AD and cells like monocytes, lymphocytes or neutrophils can cross the BBB in response to Aβ accumulation and the augmentation of vascular adhesion molecules, contributing to the pathogenesis of AD [119,120,121].

4.1.2. Therapeutic Opportunities for MSCs Targeting BBB in AD

Currently there are no effective treatments to cure or slow AD progression. However, emerging evidence suggest that MSCs therapy could be a promising approach. In general, MSCs transplantation has been found to decrease Aβ deposits and plaques, and tau-related cell death in vivo. The paracrine effects of MSCs stimulate neurogenesis, synaptogenesis, and neuronal differentiation, demonstrating neuroprotective functions. Beside this, their immunoregulatory properties, which modulate microglia/astrocytes activity state, can deactivate neuroinflammatory responses via several transcription factor signaling pathways [122].

The BBB represents a major challenge in treating AD, and different studies have focused on the action of MSCs on cerebral vasculature (Figure 2). For instance, Garcia et al. [123] demonstrated the ability of intracerebrally transplanted MSCs in a 2xTg-AD mouse model to promote neovascularization in the hippocampus. Specifically, they found that MSCs genetically modified to express VEGF, enhanced their therapeutic efficacy in promoting neovascularization. Focusing on transporters involved in AD, Son et al. [124] modified MSCs to express the secreted isoform of RAGE (sRAGE), which inhibits the interaction between RAGE and its ligands, thus preventing the adverse effects of this signaling pathway. To note, when activated by Aβ oligomers, RAGE can lead to cell stress, generation of ROS, and RAGE-mediated inflammation and neurodegeneration. Transplantation of sRAGE-MSCs into 5xFAD transgenic mice reduced the deposition of Aβ, cell death, and inflammation.

In a rat model of cerebral small vessel disease, a pathology characterized by Aβ deposition equivalent to AD, the intravenous infusion of MSCs restored the polarity/distribution of AQP4 to the end-feet of astrocytes, relieving cerebral edema and promoting the clearance of Aβ [125]. In another study, Tachibana et al. [126] implanted mouse MSC-derived pericytes into the brains of APP/PS1 mice and observed a reduction in Aβ levels in the hippocampus, an effect that was mediated by LRP1. Interestingly, a recent study has shown that, in a model of microfluidic BBB-like microvasculature, BM-MSCs emulate more efficiently the function of perivascular pericytes than induced pluripotent stem cell-derived pericytes, leading to greater restoration of TJs and the abluminal BMs [127]. In fact, there are several similarities between MSCs and pericytes. Pericytes express a similar pattern of immunological markers (CD44, CD90, CD73, CD105, and CD45), are self-renewable, and have the capacity to differentiate into nervous cells, mainly glial cells, in vivo [128]. Therefore, MSCs could potentially supply the loss of pericytes in AD. In summary, these results highlight the need for further research in this field, as understanding the role of MSCs in modulating the BBB in the context of AD is essential to develop effective therapies.

4.2. Parkinson´s disease

Parkinson´s disease (PD) is the second most common neurodegenerative disease after AD. This progressive disorder is characterized by the loss of dopamine neurons in the substantia nigra pars compacta (SNpc) and the accumulation of filamentous and oligomeric inclusion bodies (Lewy bodies) composed of misfolded α-synuclein proteins. These structures provoke cellular processes disruption and lead to neuronal degeneration mainly leading to motor dysfunction. Its etiology is still unknown and current treatments are mainly focused on symptoms [129].

4.2.1. Dysfunctional BBB in PD

Although decades ago, NVU was not recognized as an important element of the pathogenesis of PD, it is now well stablished that BBB disruption is associated with PD (Figure 3).

Nevertheless, further research is needed to completely understand the molecular mechanism of BBB disruption in this context [130]. Brain tissue from PD patients showed perivascular deposits of fibrinogen or fibrin, IgG and haemosiderin in specific regions, indicating BBB disruption [131,132,133]. In fact, degeneration of BMVECs and TJs, as well as disorganization of the components of the BM, have been reported in PD brain tissues [132]. In addition, patients with idiopathic PD show genetic mutations that affect NVU components. For example, mutations in leucine-rich repeat kinase 2 in BMVECs lead to increased monocyte attachment in PD patients [134]. Moreover, several PD patients have mutations in the MDR1 gene, which encodes P-gp in BMVECs, resulting in reduced pump function [135].

Angiogenesis is also affected in PD. Although PD patients show augmented vascular density in SNpc, in the proximity of neuronal damage, these new microvessels display impaired maturation processes and altered diameters [136]. While new microvascular architecture may allow the supply of nutrients and cellular debris, it also raises the risk of leakage and infiltration of toxins, drugs, and immune cells, potentially exacerbating the pathology [137,138]. Besides, proinflammatory cytokines secreted by activated immune and glial cells (such as TNF-α, IL-1β, and IL-6) [139] can decrease the expression of ZO-1 and occludin, leading to a disruptive state of TJs and subsequently contributing to BBB breakdown [140].

4.2.2. Therapeutic Opportunities for MSCs Targeting BBB in PD

Over the last decade, preclinical studies investigating the potential of MSCs in the context of PD have been performed (Figure 3). For instance, Chao et al. conducted an extensive study on the effect of MSCs using the preclinical model of PD induced by 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) [141]. They found that MSCs, intraperitoneally infused 24 hours after the last MPTP injection, migrated to the SNpc and efficiently rescued dopaminergic TH+ neurons. Moreover, the treatment with MSCs restored BBB integrity, as indicated by the retrieval of TJ-protein expression (claudin-1, claudin-5 and occludin).

Other studies have recently demonstrated that the treatment with MSC-derived exosomes regulates genes associated with the angiogenesis of human BMVECs in vitro, resulting in increased expression of Angpt1 and Flk1, as well as the secretion of the intercellular adhesion molecule 1 (ICAM1) protein. Injecting these exosomes into an MPTP-induced PD model resulted in their homing to the injured brain and a significant recovery from the disease. Additionally, there was an increase in the expression of ICAM1 and CD31 markers in the striatum and SNpc [142]. Conversely, in a LPS injection model into the SNpc, the treatment with MSCs increased the expression of P-gp in endothelial cells and restored BBB integrity [143]. The study suggested that MSCs decreased the proinflammatory activation of microglia and modulated the VEGF-A signalling through astrocytes, leading to an increase the astrocytic end-feet density. This process stabilized the expression of TJ-proteins. These results suggest the relevance of modulating the BBB in PD for developing effective therapies against this debilitating disease.

4.3. Multiple sclerosis

Multiple sclerosis (MS) is an inflammatory neurodegenerative autoimmune disease that affects the CNS. In MS, the immune system generates a complex response against myelin sheaths that wraps nerve axons, eventually leading to inflammation, demyelization, axonal degeneration and ultimately neuronal loss [144]. Immune cells cross the damaged BBB and release proinflammatory cytokines such as TNF-α, IFN-γ or IL-17 which directly attacks myelinating oligodendrocytes [145] or provoke a pro-inflammatory polarization of microglia and astrocytes that finally cause oligodendrocytes loss [146].

There are different types of MS depending on the evolution of the disease: relapsing-remitting MS (RRMS) and primary/secondary progressive MS (PPMS/SPMS). RRMS is characterized by alternating periods of symptoms enhancement (relapses) and partial or complete recovery of the neurologic function (remissions). Conversely, progressive MS (15% of MS patients) is marked by a gradual worsening of symptoms without periods of relapses or remissions. On the contrary, SPMS follows the initial relapsing-remitting course [144].

4.3.1. Dysfunctional BBB in MS

Although the etiology of the disease is not completely understood, a vast body of evidence suggests the importance of BBB disruption in the pathology of MS (Figure 4).

Inflammation affects to several components of the NVU and hampers the physiologic function of numerous transport mechanisms. BBB permeability may be an important early step that is correlated to the initiation of a CNS-specific immune response [147]. Pro-inflammatory mediators such as IL-1β [148], IL-6 [149], TNF-α [150], and chemokine (C-C motif) ligand 2 (CCL2) [151] reduces the expression of TJ and AJ proteins in different in vitro BBB models. Specifically, the expression of toll-like receptors, which play a significant role modulating MS, is significantly increased in BMVECs in response to ROS and TNF-α [152,153]. The expression of other transporters like GLUT-1 [154], LAT1 [155] or P-gp [156], is also affected by the exposure of BMVECs to inflammatory mediators. Moreover, pro-inflammatory molecules can cause pericyte detachment from BMVECs and undergo transformation into phagocytic or fibroblastic-like cells [147]. Regarding the autoimmune component of MS, activation of BMVECs with Th1 cytokines (IL-2, TNF-α, IFN-γ) modulates the BBB phenotype and stimulates the expression of endothelial cell adhesion molecules, such as ICAM-1 and VCAM-1 [157].

4.3.2. Therapeutic Opportunities for MSCs Targeting BBB in MS

The disruption of the BBB in MS allows immune cells to infiltrate into the CNS, contributing to the pathogenesis of MS, as previously mentioned. In vitro studies using a BBB model exposed to TNF-α have shown that treatment with embryonic MSCs modulates barrier permeability, increases the expression of TJ proteins, and decreases the expression of pro-inflammatory chemokines like CCL2 and CXCL12 [158] (Figure 4). In vivo studies have also demonstrated that co-administration of MSCs expressing IFN-β along with minocycline, tames the disruption of the Blood-Spinal Cord Barrier (the functional equivalent of the BBB in the spinal cord) [159]. Recent studies in the MS preclinical model (experimental autoimmune encephalomyelitis) have further shown that MSC transplantation reduced BBB disruption, as evidenced by reduced IgG leakage. Additionally, MSC transplantation led to the adequate expression of TJ-proteins occluding and ZO-1 in BMVECs and the restoration of AQP4 levels in astrocytes [160].

4.4. Stroke

Stroke is a complex and heterogeneous affection that is influenced by genetic predisposition, life habits, and chronic diseases. It is the second leading cause of death and third leading cause of disability worldwide. To note, 60% of the survived patients are affected by cognitive impairment, dementia, or depression. Stroke is classified into two types: hemorrhagic and ischemic, with the latter being the most prevalent (87% of the total). Ischemic stroke is characterized by a sudden cessation of oxygen and blood supply due to thrombus blocking blood flow in the brain vasculature [161].This initiates a rapid and complex cascade of pathophysiology events at genomic, molecular, and cellular levels that may evolve over hours to days and weeks after the onset, including energy failure, acidosis, loss of cell homeostasis, excitotoxicity, oxidative stress, activation of glial cells, inflammation, and disruption of the BBB with infiltration of leukocytes.[162,163]

4.4.1. Dysfunctional BBB in Brain Ischemia.

BBB plays a significant role in the pathophysiology of ischemic stroke, and its dysfunction varies depending on the severity, and duration of the ischemia. Predominantly, during human stroke, BBB presents a continuous opening pattern with biphasic peaks distributed along four stages [164] (Figure 5).

Hyperacute stage is the first phase that evolves within the first 6h after the ischemic onset. Along this phase, the first BBB opening is documented. Due to oxygen and glucose deprivation, Na⁺ and Ca⁺ accumulates inside the cells of the NVU, including astrocytes or BMVECs, leading to cytotoxic edema [165], glutamate excitotoxicity [166], oxidative damage associated with ROS generation, and mitochondrial dysfunction [164]. Moreover, MMPs (mainly MMP2) directly degrade TJ proteins and BM components, contributing to BBB leakage [167].

The next stage corresponds the acute phase, which occurs after the first 6h of the onset for a period of 72-96h. The second permeability peak is observed at this stage. Since this point, immune components start to participate more significantly to the stroke pathophysiology. Neutrophils are the predominant peripheral immune cells in the acute post-stroke period [168]. They contribute to the BBB disruption by producing excessive ROS [169], proteases (MMP9, proteinase-3, elastase) [170] and neutrophil gelatinase-associated lipocalin [171], generating neutrophil extracellular traps [172] and secreting inflammatory cytokines (IL-1β, IL-6, IL-8, TNF-α) and chemokines (CCL2, CCL3, and CCL5) [173]. Moreover, P-selectin glycoprotein ligand-1 and macrophage-1 antigen in neutrophils, and their respective receptors, P-selectin and ICAM-1, in BMVECs, are up-regulated shortly after ischemic stroke induced by IL-1β. Hence, paracellular BBB permeability is increased [174,175]. Natural killer (NK) cells are also present during this stage producing IFN-ϒ and ROS [176]. Monocytes and cerebral immune cells, microglia and astrocytes display a pro-inflammatory phenotype which exacerbate BBB breakdown through the secretion of proinflammatory factors (IL-1α, IL-1β, IL-6, TNF-α, IFN-γ, CCL2), MMP9, VEGF, ROS and activation of inducible nitric oxide synthase [162]. Finally, upregulation of MMP-9 is crucial in this phase (24-48h after onset) and it plays a dual role in the proteolytic degradation of the BM components of the BBB. Its capability to digest TJ proteins, enhances the disruption of the BBB [177]. Concomitantly, vascular remodeling is induced by VEGF, leading to the mobilization of progenitor BMVECs, which implies an immature and damaged BBB during this neovascularization process [178].

The subacute phase starts one week after the stoke onset, a timepoint from which BBB begins its recovery process. Monocytes and glial cells shift to anti-inflammatory phenotypes expressing cytokines (IL-10, IL-4, TNF-β) and neurotrophic factors that help prevent inflammation [179]. Moreover, during this phase, angiogenesis plays a vital role in restoring the blood flow and oxygen supply in ischemic tissues. Angiogenesis promotes the proliferation and migration of BMVECs and pericytes, enhances tube formation, branching, and anastomosis, all of which are modulated by the inflammatory microenvironment. VEGF, MMP9 and angiopoietin 2 (Ang-2), as well as its receptor Tie-2, are deeply involved in this process [164]. While these agents may temporarily contribute to BBB leakage, a higher degree of angiogenesis has been linked to increased survival in patients [180] and greater stability of the BBB [181].

Finally, the chronic post stroke phase starts approximately six weeks after the ischemic event, where the BBB is still disrupted but to a lesser extent compared to the previous phases. During this stage, NVU components are restored in order to seal the BBB, with an overexpression and new distribution of TJ proteins. Some of the factors that stabilize the BBB include Ang-1, which maintains BMVECs in a quiescent state and contributes to junction generation, and sphingosine-1 phosphate and activated protein C, which help balance junction proteins and cytoskeleton [182]. Immune cells also shift to anti-inflammatory phenotypes and develop mechanisms to diminish BBB breakdown. For example, anti-inflammatory astrocytes release pentraxin-3, which inhibits VEGF in the ischemic cerebral tissue and specifically supports BBB integrity [183]. At last, neuronal progenitor cells migrate to the ischemic tissue to favor neurogenesis and neuroplasticity and restore BBB components [184,185].

4.4.2. Therapeutic Opportunities for MSCs Targeting BBB in Brain Ischemia.

Several studies have demonstrated the therapeutic effect of MSCs in preclinical models of stroke (Figure 5). Thus, understanding their effect on the BBB has become a topic of great interest in research [186].

For example, MSCs have shown the ability to protect neurons against oxidative damage. A study conducted by Huang et al. [187] showed that MSCs express high levels of antioxidant enzymes from the peroxiredoxin (PRDX) family. In this report, MSCs were able to rescue BBB integrity in an in vitro model of oxidative damage with the bEnd.3 cell line, by reducing TJ degradation and excessive ROS generation. This effect was found to be partially mediated by the secretion of PRDX4. Interestingly, silencing PRDX4 in MSCs attenuated their protective effect on BBB integrity, both in vitro and in the in vivo MCAO ischemic model. Conversely, Cheng et al. [188] investigated the effect of the treatment of MSCs administered 15 minutes after MCAO and found a reduction in IgG leakage in the brain parenchyma, along with the reversal of the TJ-protein gap formation (ZO-1, occludin and claudin-5). This treatment also suppressed MMP-9 upregulation, reduced neuroinflammation, and decreased neutrophil infiltration with downregulation of ICAM-1 expression. In a related study, Liu et al. [189] demonstrated that co-culture of MSCs with HUVECs in an in vitro ischemic-reperfusion model, rescued injured endothelial cells via the generation of TNT-like structures, which was dependent on F-actin polymerization. They further investigated this effect in vivo in a model of MCAO in rats [190], in which the injection of MSCs 24 h after the ischemic induction led to a significant reduction in the infarct volume and higher microvessel densities in the peri-infarct areas. This was attributed to rescuing brain microvasculature through mitochondrial transfer with TNT formation in vivo. Moreover, Zacharek et al. [191] demonstrated that the administration of MSCs 24 h after MCAO in rats reduced BBB leakage and promoted angiogenesis and vascular stabilization in vivo and in vitro. This effect was achieved by increasing endogenous Ang-1/Tie2 and TJ proteins, and promoting the cross-talk between BMVECs and astrocytes.

Finally, different studies have demonstrated that the administration of MSC-derived EVs can lead to a reduction in the infarct volume, improve neurological recovery, and enhanced angiogenesis. These effects are particularly significant when MSCs are exposed to hypoxic conditions prior to EV isolation [192]. These results suggest that it is not necessary to use the whole MSC for the treatment of ischemic stroke [193,194], evidencing MSC-derived EVs as a promising alternative for stroke therapy. Overall, these studies suggest that MSC treatment has a significant impact on BBB integrity and a great potential to modulate microvasculature and reduce the pathologic processes associated to stroke.

5. Challenges and Future Directions

While preclinical studies have shown promising results regarding the therapeutic potential of MSCs in neurodegenerative and neuroinflammatory diseases, the clinical efficacy of using MSCs is still uncertain (Table 2). Recently, Kvistad et al. [195] conducted a meta-analysis to evaluate the safety and efficacy of MSC therapy in various neuroinflammatory diseases, including MS, ischemic stroke, and traumatic spinal cord injury. The study found that the treatment was generally safe and well-tolerated by patients. However, the efficacy results were inconclusive, with no significant improvements, likely due to a considerable placebo effect. This lack of efficacy may be attributed to several factors, including the retention of a significant proportion of the transplanted cells in the lungs rather than reaching the target brain tissue, as well as the absence of a robust mechanism for efficient arrival to the brain. Of note, the therapeutic role for MSCs seems to be not only based on their homing ability and/or to their capability to differentiate into diverse NVU cell types, but mainly dependent on their paracrine secretion of trophic factors, immunomodulatory molecules, microvesicles, microRNAs and mitochondrial transfer. Additionally, the specific source of MSCs, the dosage of transplanted cells, the timing and route of delivery, and the culture and isolation protocols [196] can also impact the therapeutic outcome. Interestingly, the clinical trials using ASCs against neurodegenerative disorders have been significantly increased in the last years (Table 2). These cells are easily available in adults, abundant, and readily to isolate and expand. Additionally, ASCs allow allogenic treatments characterized by many advantages from a therapeutical perspective as they can be used for generating a standardized allogenic cost-effective donor bank and avoid the autologous associated issues like donor-receiver diseases.

Besides, age, gender, genetic traits, and medical history of the donor are important factors that challenge the quality of the MSCs [197]. Therefore, the development of live imaging techniques to track and assess the bioavailability and biodistribution of these cells, the standardization of isolation and expansion protocols for MSCs, the incorporation of clear biomarkers that can accurately reflect the therapeutic effects of MSCs in patients, and the implementation of well-designed randomized controlled studies among others, will minimize variability and improve the reproducibility and reliability of MSCs-based therapies.

To address the challenges and improve the clinical outcomes of MSC therapy in neuroinflammatory pathologies, it is imperative gaining a clearer understanding of the underlying biology and mechanisms of action of MSCs as well as identifying new cellular and/or molecular targets involved in these diseases. This knowledge will serve as a foundation for optimizing the treatment approach and developing targeted strategies that can enhance therapeutic efficacy. On the other hand, understanding the role of the BBB in the pathophysiology of neurodegenerative and neuroinflammatory diseases is crucial for the development of new multi-target-based therapies. The ability of MSCs to restore the integrity and functionality of the disrupted BBB during aging-associated CNS diseases, together with their capacity to modulate the immune response and neurodegeneration makes them an attractive therapeutic option. By focusing on the BBB in the development of MSC-based therapies, we may be able to improve the clinical outcomes of patients suffering from these debilitating conditions.