Pathological mechanisms and therapeutic potential of MSC-Exos in ischemic stroke management

Abstract

Cerebral ischemia–reperfusion injury (CIRI) represents a critical pathological challenge in the management of ischemic stroke, substantially limiting the clinical efficacy and success rates of neurosurgical interventions, brain tissue therapies, and cell transplantation strategies. The complex pathological mechanisms underlying CIRI, which include inflammatory responses, oxidative stress, ferroptosis, apoptosis, and blood–brain barrier disruption, significantly hinder neurological recovery. Mesenchymal stem cells-derived exosomes (MSC-Exos) have emerged as a promising acellular therapeutic strategy owing to their low immunogenicity, ability to cross the blood–brain barrier, and multi-target regulatory properties. This review compares MSC-Exos derived from multiple sources and summarises their mechanisms of action and therapeutic potential in CIRI. Despite ongoing challenges in standardisation and safety evaluation, accumulating preclinical evidence highlights their potential for future clinical translation and precision regenerative medicine.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1186/s13287-026-05120-3.

Introduction

Stroke is the second leading cause of death worldwide, with ischemic stroke accounting for approximately 87% of all cases [1]. The underlying pathophysiology of ischemic stroke involves the interruption of cerebral blood flow, resulting in localized tissue ischemia, hypoxia, and subsequent neurological damage [2]. Current therapeutic strategies primarily aim to restore cerebral perfusion as early as possible, with clinical management relying on mechanical thrombectomy and intravenous thrombolysis to re-establish blood flow [3]. However, reperfusion itself may trigger CIRI, which can further aggravate neurological dysfunction and significantly impact patient prognosis [4]. CIRI is characterized by a series of complex and interrelated pathological processes, including disturbances in energy metabolism, disruption of calcium homeostasis, oxidative stress, and various forms of cell death, ultimately leading to neuronal apoptosis or necrosis [5]. Therefore, the development of effective therapeutic strategies to mitigate CIRI is of critical importance for improving outcomes in patients with cerebral ischemia.

Mesenchymal stem cells (MSCs) are a population of adult stem cells characterized by their capacity for self-renewal and multi-lineage differentiation. Owing to their low immunogenicity and minimal ethical concerns, MSCs have been extensively investigated in the field of regenerative medicine [6, 7]. Their therapeutic effects are largely mediated through paracrine mechanisms, including the secretion of immunomodulatory and pro-angiogenic factors that facilitate neural repair and regeneration [8]. However, the in vivo homing efficiency of MSCs remains limited, and accumulating evidence suggests that their therapeutic benefits are primarily attributable to secreted extracellular vesicles, particularly exosomes [9, 10]. Notably, Lai et al. were among the first to demonstrate the secretion of exosomes by MSCs using electron microscopy and proteomic analyses [11].

Extracellular vesicles (EVs) are membrane-bound nanoparticles actively released by cells into the extracellular environment. According to the definition proposed by the International Society for Extracellular Vesicles (ISEV), EVs are “particles naturally released by cells, enclosed by a lipid bilayer, and incapable of replication, i.e., lacking a functional nucleus” [12, 13]. EVs carry a diverse array of bioactive molecules, including proteins, lipids, and various RNA species, and play a critical role in intercellular communication and the transfer of biomolecules between cells [14]. Based on their biogenesis and size characteristics, EVs can be broadly classified into several subtypes, primarily including exosomes, microvesicles, and apoptotic bodies. Among these, exosomes represent a key subtype, defined as nanovesicles originating from the endosomal pathway. Specifically, they are formed within intracellular multivesicular bodies (MVBs) and released into the extracellular space upon fusion of MVBs with the plasma membrane [15]. Exosomes typically range from 50 to 150 nm in diameter, possess a characteristic lipid bilayer structure, and are widely distributed in various biological fluids, including blood, cerebrospinal fluid, and urine [16]. As essential mediators of intercellular communication, exosomes selectively encapsulate and transport a diverse array of bioactive molecules—including proteins, lipids, and non-coding RNAs—thereby modulating key biological processes in target cells, such as inflammatory responses, cell survival, angiogenesis, and tissue remodeling [17]. The surface expression of integrin α4β1 on exosomes facilitates interaction with vascular cell adhesion molecule-1 on brain microvascular endothelial cells, enabling efficient translocation across the blood-brain barrier. This capability allows exosomes to deliver various bioactive molecules, such as anti-apoptotic microRNAs (e.g., miR-21) and neurotrophic factors (e.g., BDNF), to ischemically damaged sites. They directly inhibit neuronal apoptosis and excessive microglial activation following CIRI, thereby playing a pivotal role in the repair of ischemic brain injury [18]. Recent studies indicate that exosomes exert significant regulatory effects across multiple central nervous system disorders. Their capacity to traverse the blood-brain barrier and accumulate at lesion sites renders them particularly promising for therapeutic applications in conditions like CIRI. Compared to cell therapies, MSC-Exos are considered a safer and more controllable acellular therapeutic strategy due to their low immunogenicity, excellent biological stability, and amenability to engineered modifications, offering a novel research direction for multi-targeted intervention in CIRI [19]. Although several reviews have discussed exosome-based therapies, none have systematically compared the therapeutic mechanisms and translational readiness of MSC-Exos from distinct tissue sources.

This article provides a comprehensive elucidation of the pathophysiological mechanisms underlying CIRI, with a particular emphasis on the role of exosomes in this context. It also presents an in-depth analysis of the characteristics and therapeutic potential of MSC-Exos in the treatment of CIRI. Furthermore, the study offers a systematic comparison of the mechanisms of action and protective effects of MSC-Exos derived from four distinct sources—adipose tissue, bone marrow, umbilical cord, and pluripotent stem cells—in the treatment of CIRI, to our knowledge, relatively few studies have systematically compared the therapeutic mechanisms of MSC-Exos derived from different tissue sources in CIRI. In conclusion, we examine the advancements in clinical research concerning MSC-Exos and highlight the primary challenges encountered in their clinical translation, including safety assessment, quality control, and long-term efficacy. We also propose optimization strategies and pathways aimed at enhancing their clinical applicability, thereby offering novel insights into their potential use in the treatment of CIRI. exert a systematic and robust neuroprotective effect.

The pathophysiological mechanisms of CIRI

The pathophysiological basis of CIRI

CIRI constitutes a complex pathophysiological process characterized by multiple interacting mechanisms. During the ischemic phase, the depletion of adenosine triphosphate and the generation of lipid messengers through membrane phospholipid cleavage establish the foundation for subsequent cellular injury. Although reperfusion restores oxygen supply, persistent structural abnormalities in the membrane and oxidative stress, mediated by lipid oxidation products, further exacerbate neuronal damage. Current research suggests the involvement of various molecular mechanisms: SRC-3 exosomes inhibit ferroptosis, thereby exerting neuroprotective effects [20]; ethanol extracts of Cyperus rotundus reduce neuronal loss [21]; and the activation of TRPC3/6/7 channels exacerbates injury through pathways such as NF-κB [22].

The role of inflammatory response and oxidative stress in CIRI

The fundamental mechanisms underlying CIRI encompass intricate interactions between oxidative stress and inflammatory responses. Oxidative stress is predominantly initiated by an overproduction of reactive oxygen species (ROS) [23, 24], which directly induce lipid peroxidation in cell membranes, ultimately leading to neuronal apoptosis. Simultaneously, oxidative stress activates signaling pathways such as NF-κB [25, 26], which promotes the expression of inflammatory mediators, thereby intensifying neuroinflammation [27, 28]. The inflammatory response itself is a crucial component of CIRI, characterized by the activation of various inflammatory cytokines that trigger cascading reactions, further exacerbating brain tissue damage. For example, tumor necrosis factor-alpha (TNF-α) secreted by astrocytes exacerbates CIRI through the RIP1/RIP3/AKT pathway [29], while peroxiredoxin 1 activates TLR4-mediated inflammatory and apoptotic pathways, thereby advancing the progression of CIRI [30]. A vicious cycle is established between oxidative stress and inflammatory responses: oxidative stress induces the release of interleukin-1 beta (IL-1β) and TNF-α, which amplify inflammation and exacerbate brain injury [31, 32]; Simultaneously, inflammatory signals, such as the activation of NOX2 by NOD2, enhance the generation of ROS, thereby exacerbating oxidative stress and cellular damage [33]. In terms of therapeutic strategies, various compounds, including apigenin, quercetin, and dihydrocapsaicin, have the potential to alleviate both oxidative stress and inflammatory responses. They achieve this by inhibiting ROS production, activating antioxidant enzyme systems, blocking the NF-κB pathway, or downregulating the expression of specific inflammatory factors, thus ameliorating CIRI [25, 26]. In conclusion, a comprehensive analysis of these mechanisms, along with the development of targeted intervention strategies, will offer novel insights and potential therapeutic targets for the clinical management of CIRI [34, 35].

Apoptosis and regeneration mechanisms of neural cells

In the context of CIRI, the processes of neuronal apoptosis and regeneration are intricately regulated by multiple signaling pathways. Apoptosis encompasses various forms, including autophagy, ferroptosis, and pyroptosis, with pathways such as AMPK-mTOR and BDNF-TrkB-PI3K-AKT serving as pivotal regulators [36–39]. Simultaneously, neuroregenerative mechanisms are activated to facilitate brain injury repair. For example, remote post-ischemic treatment inhibits mitochondrial apoptosis and enhances function by activating the B2R/PI3K/AKT pathway [40]. The identification of the ferroptosis regulator FTO and lipid mediators such as 14,15-EET further elucidates novel mechanisms of cell death and identifies potential neuroprotective targets [41, 42]. Recent research highlights significant advancements in ferroptosis inhibition strategies that focus on lipid metabolism and redox homeostasis. For instance, direct inhibition of ACSL4—a critical enzyme in lipid remodeling—effectively disrupts the peroxidation cascade of polyunsaturated fatty acid phospholipids, thereby alleviating ischemia/reperfusion-induced tissue injury [43]. Furthermore, extracellular vesicles/exosomes, as novel acellular therapeutic carriers, have been demonstrated to inhibit neuronal ferroptosis signalling pathways and enhance neurological recovery by delivering antioxidant-related proteins or regulatory molecules [44]. These intertwined death and survival signalling pathways not only elucidate the pathophysiological processes of CIRI but also provide crucial theoretical foundations for developing novel therapeutic strategies.

The therapeutic mechanisms of MSC-Exos in CIRI

Regulation of inflammatory response

The investigation of exosomes in CIRI and their therapeutic mechanisms constitutes a significant area of interest in contemporary neuroscience research. Comprehensive studies have demonstrated that MSC-Exos, due to their anti-inflammatory and pro-angiogenic properties, exhibit considerable neuroprotective potential in the context of CIRI. Their mechanism of action is characterized by a multi-level, precise regulation of the inflammatory response [45]. At the molecular level, MSC-Exos modulate key inflammatory pathways through the delivery of specific non-coding RNAs. For example, miR-223-3p within exosomes directly ameliorates CIRI by inhibiting the pro-inflammatory M1 polarization of microglia [46]. Similarly, miR-330-5p reduces the release of inflammatory cytokines by targeting the SETD7 gene, thereby attenuating CIRI-associated inflammatory responses [47].

Furthermore, more complex regulatory mechanisms involving long non-coding RNAs are also implicated. For instance, the lncRNA KLF3-AS1, carried by exosomes, competitively binds miR-206 through adsorption, thereby releasing its inhibition on the deubiquitinating enzyme USP22 (Ubiquitin Specific Peptidase 22). This stabiliszes Sirt1 (Sirtuin 1) protein levels, ultimately suppressing NF-κB pathway activity and reducing the expression of inflammatory cytokines expression [48]. Yue et al. demonstrated that MSC-Exos exhibit significant therapeutic effects in other various types of ischemia-reperfusion injury. For instance, in myocardial ischemia-reperfusion injury, MSC-Exos carrying miR-182-5p target gasdermin D (GSDMD), thereby mitigating pyroptosis and inflammatory responses in cardiomyocytes [49]. In summary, MSC-Exos suppress neuroinflammation through multiple layers and targets by delivering various non-coding RNAs. This ranges from directly inhibiting pro-inflammatory pathways (e.g., miR-150-5p/TLR5/NF-κB) to constructing complex RNA regulatory networks (e.g., KLF3-AS1/miR-206/USP22/Sirt1), thereby providing a precise anti-inflammatory therapeutic strategy for CIRI. Notably, exosomes may also exhibit duality roles in inflammatory regulation. For instance, Zhang et al. demonstrated that exosomes can influence the progression of neuroinflammation by altering the polarization states of microglia. Specifically, exosomes derived from hypoxic vascular smooth muscle cells exacerbated ischemic brain injury by promoting M1 microglial polarization [50]. Similarly, macrophage-derived exosomes further aggravated brain injury following traumatic brain injury by inducing neuroinflammation and cognitive dysfunction [51]. This indicates that the inflammatory regulatory role of exosomes is source- and microenvironment-dependent, suggesting the need for precise control of their origin and functional state in therapeutic interventions.

Inhibition of neuronal apoptosis and ferroptosis

MSC-Exos have been shown to attenuate neuronal apoptosis and ferroptosis induced by CIRI through a variety of mechanisms. Zhu et al. identified that MSC-Exos can inhibit neuronal apoptosis by modulating relevant signaling pathways, including PTEN/AKT and TRAF6, through the delivery of microRNAs such as miR-486-5p and miR-345-3p [52, 53]. Li et al. reported that MSC-Exos facilitate the proliferation and lumen formation of microvascular endothelial cells in ischemic regions by activating the Wnt-β-catenin signaling pathway, which enhances local blood supply and indirectly reduces neuronal apoptosis resulting from ischemia-hypoxia [54]. Additionally, exosomes have been found to enhance cellular survival by influencing autophagy-related pathways, such as the ETV4/Sirt1 axis, thereby further alleviating brain injury [55].

Furthermore, MSC-Exos exert their protective effects by modulating ferroptosis-related signaling pathways. Gong et al. demonstrated that MSC-Exos reduce the expression of ferroptosis-associated proteins through the delivery of specific microRNAs, such as miR-1275 and miR-330-5p, thereby mitigating cellular damage [47, 56]. Li et al. have demonstrated that MSC-Exos exert neuroprotective effects in CIRI by inhibiting ferroptosis through modulation of the Nrf2/GCH1/BH4 and HSPA5/GPX4 axes [57, 58]. Collectively, MSC-Exos mitigate CIRI via multiple synergistic mechanisms. Liu et al. further discovered that MSC-Exos not only inhibit apoptosis and ferroptosis by directly regulating intracellular signaling pathways but also enhance neuroprotection by modulating inflammatory responses and oxidative stress pathways [59, 60]. Additionally, mesenchymal stem cell-derived extracellular vesicles (MSC-EVs) suppress neuronal pyroptosis by negatively regulating the stability of thioredoxin-interacting protein [61], while also delivering molecules such as miR-486-5p to regulate the PTEN signaling axis and inhibit ferroptosis, thereby alleviating CIRI through multiple pathways [52]. These findings suggest that the regulation of ferroptosis by MSC-EVs is a key mechanism underlying their neuroprotective effects.

While the inhibition of ferroptosis by MSC-Exos has been widely documented, there are notable discrepancies among studies concerning their primary effector molecules, such as USP10, SRC-3, and the circBBS2/miR-494 axis. These discrepancies highlight fundamental issues related to experimental reproducibility [20, 62]. The inconsistency primarily arises from two factors: first, the lack of standardized protocols for exosome preparation and characterization, which results in batch-to-batch variability in terms of purity, subpopulation composition, and active molecule loading; second, variations in in vitro models, including differences in OGD/R conditions and the types and concentrations of ferroptosis inducers, which may affect the dominant form of cell death. More critically, current research is significantly deficient in systematic dose-response evaluations. The units and ranges of exosome dosages are inconsistent and excessively broad, hindering the determination of minimum effective doses and safety thresholds. This greatly obstructs dose optimization for clinical translation [63]. Therefore, establishing standardised exosome preparation protocols and model systems, coupled with rigorous dose-escalation studies, represents an urgent requirement for elucidating their anti-ferroptotic mechanisms and advancing clinical applications.

Promotion of angiogenesis and neural remodeling

MSC-Exos exhibit significant potential in enhancing angiogenesis and neural remodeling. These exosomes, which carry a variety of bioactive molecules such as microRNAs (miRNAs) and proteins, are capable of modulating inflammatory responses and apoptosis subsequent to ischemia-reperfusion injury, thereby supporting neuroprotection and functional recovery [64]. Notably, miRNAs within MSC-Exos are crucial in regulating angiogenesis and neural remodeling. Bao et al. demonstrated that miR-486 within MSC-Exos substantially improves angiogenesis and neural functional recovery after ischemic brain injury by influencing the PTEN/Akt pathway [65]. Additionally, Dehghani et al. reported that MSC-Exos facilitate endogenous neural circuit remodeling, neurovascularization, and brain tissue repair, showing significant therapeutic efficacy in animal models of ischemic stroke [63]. Moreover, MSC-EVs synergistically enhance recovery from neurological deficits by modulating key signaling pathways, including AMPK and JAK2/STAT3/NF-κB. Chen et al. revealed that macrophage-derived exosomes also enhance intestinal neuronal injury via the MMP8-TGF-β pathway, suggesting potential similarities in exosomal mechanisms across different tissues [66]. In summary, MSC-Exos hold considerable promise for application in CIRI. By carrying multiple bioactive molecules, MSC-Exos can modulate diverse signalling pathways, promote angiogenesis and neural remodelling, thereby offering novel approaches for treating ischemic brain injury. Future research should further explore the specific mechanisms of MSC-Exos under different pathological conditions to advance their clinical application.

Repair of the blood-brain barrier (BBB)

MSC-Exos serve as vital mediators of intercellular communication, capable of transporting a variety of bioactive molecules, including proteins, RNA, and microRNA, which are instrumental in the repair of the BBB [67]. In the context of CIRI, the integrity of the BBB is significantly compromised, resulting in neuronal damage and dysfunction. MSC-EVs offer synergistic protection to the BBB through various mechanisms. For example, research by Li et al. has shown that MSC-Exos maintain BBB integrity following ischemic stroke by modulating the Caveolin-1-mediated autophagy degradation pathway [68]. Additionally, MSC-Exos administered intranasally have been observed to significantly upregulate tight junction proteins, such as ZO-1 and Occludin, while concurrently downregulating the expression of adhesion molecules, including ICAM-1 and VCAM-1. This results in a 40% reduction in BBB permeability [69]. This bidirectional regulation not only reduces the infiltration of inflammatory cells but also significantly mitigates ischemia-reperfusion-induced vascular endothelial injury, leading to a decrease in cerebral edema severity (evidenced by an 18% reduction in brain water content) and a lowered risk of hemorrhage [70]. This study suggests a non-invasive strategy for BBB repair through intranasal administration, thereby addressing the limitations associated with traditional intravenous delivery methods. Furthermore, by establishing bidirectional regulation of tight junction proteins and adhesion molecules as a core mechanism, it provides novel insights for targeted therapies addressing BBB injury.

Additionally, in a preterm lamb model, MSC-EVs have been shown to effectively sustain myelin basic protein expression, reduce demyelination injury, and preserve white matter integrity [71]. The study by Liu et al. further demonstrated that MSC-EVs restore lysosomal function by inhibiting the mTOR signaling pathway and promoting the nuclear translocation of transcription factor EB, thereby ameliorating autophagy-lysosomal pathway dysfunction induced by CIRI [72]. Moreover, MSC-EVs play a pivotal role in modulating oxidative stress, mitochondrial autophagy, inflammatory responses, and glial cell polarization. These combined effects collectively mitigate pathological damage associated with CIRI [73]. Additionally, exosomes derived from macrophages exhibit significant roles in various types of brain injury. For example, in the context of spinal cord injury, exosomes facilitate macrophage phenotypic switching through miRNA-miRNA networks, thereby influencing reparative processes within the immune microenvironment [74]. In cases of acute lung injury, exosomes exacerbate inflammatory responses by impairing macrophage function, further underscoring their extensive role across diverse injury contexts [74, 75]. However, the effects of exosomes on the BBB exhibit duality. Under certain circumstances, exosomes may also exacerbate cell death; for instance, TSBS1 within macrophage-derived exosomes can induce ferroptosis in endothelial cells, thereby compromising the BBB [76]. This suggests that the origin and contents of exosomes decisively influence their directional effects in CIRI (Fig. 1).

Fig. 1.

Therapeutic mechanisms of MSC-Exos in CIRI. MSCs derived from multiple sources, including AMSCs, BMSCs, hUC-MSCs, and iPSC-MSCs, release exosomes that exert multifaceted neuroprotective effects in CIRI. These MSC-Exos modulate several pathological processes in the ischemic brain: (1) inhibition of neuronal apoptosis through downregulation of pro-apoptotic genes such as Bax; (2) improvement of mitochondrial function via regulation of mitochondrial dynamics-related proteins (e.g., OPA1 and Mfn-1); (3) attenuation of neuroinflammation mediated by microRNAs such as miR-223-3p and miR-330-5p; (4) promotion of angiogenesis through the miR-486/PTEN/AKT signaling pathway; and (5) restoration of blood–brain barrier integrity by upregulating tight junction proteins, including ZO-1 and occludin. Collectively, these coordinated actions contribute to enhanced neurovascular repair and functional recovery following ischemic injury. Created by figdraw.com

Protective effects of exosomes from different sources of MSCs on CIRI

Protective effects of exosomes derived from AMSCs on CIRI

Adipose-derived mesenchymal stem cell–derived exosomes (AMSC-Exos) exhibit low immunogenicity and favorable tissue penetration, enabling them to exert multifaceted neuroprotective effects in CIRI [77]. Yao et al. reported that AMSC-Exos deliver miR-25-3p to neurons, thereby inhibiting the p53/BNIP3 signaling pathway, which alleviates ischemia-induced excessive autophagy and improves neurological function (Fig. 2) [78]. Neuroinflammation represents a key mechanism driving the progression of CIRI, and activation of microglia can exacerbate neuronal damage through the release of inflammatory mediators [79]. AMSC-Exos mitigate inflammatory injury by suppressing the NF-κB and MAPK signaling pathways, thereby reducing microglial activation and its associated cytotoxic effects [80]. Moreover, impaired angiogenesis limits post-ischemic neural repair. miR-126, a critical regulator of vascular integrity and neuroregeneration in endothelial cells [81], is downregulated in ischemia-related diseases and negatively affects angiogenesis [82]. AMSC-Exos enriched with miR-126 can downregulate the expression of pro-inflammatory cytokines such as TNF-α and IL-1β, inhibit microglial activation, and promote angiogenesis and neural regeneration, thereby further enhancing their protective effects against CIRI [83].

Fig. 2.

ROS-induced AMSC-derived EVs induce neuronal cell death via the miRNA-25-3p/p53 pathway. Upon fusion with neuronal cells, these EVs deliver miRNA-25-3p, which targets and degrades p53 mRNA in neurons. The downregulation of p53 leads to a reduction in the p53/Bnip3/autophagy signaling axis, thereby promoting neuronal cell death. This process involves the activation of caspase-3, further contributing to the apoptotic cascade. The diagram highlights the crucial role of miRNA-25-3p in mediating ROS-induced neuronal damage through AMSC-derived EVs. Created by figdraw.com

The utilization of AMSC-Exos is limited by substantial donor-dependent variability. Factors such as donor age, body mass index, metabolic status (e.g., insulin resistance), and the site of adipose tissue harvest have a direct impact on the miRNA profile and proteome of exosomes [84]. In exosomes derived from obese donors, there is an upregulation of the pro-inflammatory miR-34a and a downregulation of the neuroprotective miR-126, which may reduce their anti-inflammatory and pro-angiogenic properties [85]. Additionally, senescence induced by cell passage represents another significant challenge. AMSCs that have undergone extensive passaging exhibit an age-associated secretory phenotype, with their exosomes containing elevated levels of pro-inflammatory factors such as IL-6 and MMP-3, potentially reducing therapeutic efficacy or even worsening neuroinflammation [86]. Consequently, the establishment of donor selection criteria, the imposition of upper limits on cell passage numbers, and the development of monitoring systems for senescence markers are essential prerequisites for improving the consistency of AMSC-Exos therapy.

Protective effects of exosomes derived from BMSCs on CIRI

Among the various sources of MSC-Exos, those originating from Bone Marrow Mesenchymal stem cell–derived exosomes (BMSC-Exos) are the most extensively studied and mechanistically well-documented. These exosomes demonstrate stable and significant neuroprotective effects across the complex pathological processes associated with CIRI. Their mechanism of action involves multi-targeted synergistic effects. Firstly, BMSC-Exos effectively suppress excessive neuroinflammation by delivering specific microRNAs, such as miR-145 and miR-150-5p, which target and inhibit pro-inflammatory signaling pathways like NF-κB, thereby reducing levels of pro-inflammatory cytokines [87]. Additionally, they induce the polarization of glial cells towards the anti-inflammatory M2 phenotype, thus remodeling the immune microenvironment [88, 89]. Concurrently, BMSC-Exos modulate multiple cell death pathways. They inhibit mitochondrial apoptosis by regulating proteins such as Bcl-2/Bax [87], suppress ferroptosis through the stabilization of SLC7A11 [20], and inhibit NLRP3-mediated pyroptosis by activating AMPK-mediated autophagy [90], thereby forming synergistic protective effects (Fig. 3). Furthermore, BMSC-Exos are involved in reparative processes, including the promotion of angiogenesis and the optimization of the neurosupportive microenvironment through molecules such as miR-138-5p [91], thereby facilitating neural remodeling [92]. The combination of BMSC-Exos with rehabilitation training further enhances their therapeutic efficacy. However, the clinical application of BMSC-Exos is challenged by limitations related to sourcing. The process of bone marrow harvesting is invasive, and the age of the donor significantly affects the quality of exosomes, with those derived from elderly donors potentially containing pro-apoptotic components [93]. Additionally, prolonged in vitro expansion may result in genetic instability of the cells and diminished functional activity of their exosomes. Consequently, the establishment of standardized production and quality control systems is essential.

Fig. 3.

BMSC-Exos restore autophagic flux and attenuate pyroptosis via the AMPK/mTOR signaling pathway under OGD/R conditions. OGD/R induces cellular injury characterized by increased phosphorylation of AMPK (p-AMPK) and decreased phosphorylation of mTOR (p-mTOR), leading to impaired autophagic flux due to disrupted fusion between autophagosomes and lysosomes. This dysfunction promotes activation of the NLRP3 inflammasome, resulting in increased Cleaved-Casp1 and subsequent pyroptosis. BMSC-Exos, carrying bioactive cargos such as miRNAs and proteins, modulate the AMPK/mTOR signaling pathway by enhancing AMPK activation and inhibiting mTOR activity. This regulation restores autophagic flux, facilitates the formation of autolysosomes, and suppresses NLRP3 inflammasome activation. Consequently, downstream events including caspase-1 cleavage, GSDMD cleavage, and IL-1β secretion are reduced, ultimately attenuating pyroptosis and conferring cytoprotective effects. Created by figdraw.com

In summary, BMSC-Exos exert multi-target neuroprotective effects in CIRI through synergistic mechanisms encompassing anti-inflammation, inhibition of apoptosis and ferroptosis, promotion of angiogenesis, and facilitation of neural remodelling. This multi-targeted, cross-pathway mode of action may account for its superiority over single-molecule interventions within complex pathological environments. Future research must further elucidate the hierarchical regulatory networks of key functional molecules and validate their therapeutic stability and translational potential in models more closely approximating real-world clinical scenarios. This will advance BMSC-Exos towards becoming a viable therapeutic strategy for ischemic stroke.

Protective effects of exosomes derived from hUC-MSCs on CIRI

Human umbilical cord mesenchymal stem cell–derived exosomes (hUC-MSC-Exos) have demonstrated significant neuroprotective effects in CIRI, involving multiple mechanisms such as anti-inflammatory, antioxidative, anti-apoptotic, and pro-angiogenic activities. Studies have shown that hUC-MSC-Exos can deliver miR-330-5p, which alleviates neuroinflammation by suppressing SETD7 expression [47]. Notably, over expression of SETD7 reverses the protective effects of hUC-MSC-Exos, indicating that the miR-330-5p/SETD7 axis plays a critical regulatory role in CIRI. In addition, miR-146a-5p carried by hUC-MSC-Exos inhibits activation of the IRAK1/TRAF6–NF-κB signaling pathway, thereby reducing the release of pro-inflammatory mediators and promoting microglial polarization toward the M2 phenotype. This process improves the neuroinflammatory microenvironment, reduces infarct volume, and enhances neurological function [94]. Regarding vascular protection, hUC-MSC-Exos activate the PI3K/AKT pathway through the circDLGAP4/miR-320/KLF5 regulatory axis, leading to upregulation of VEGF expression. This effect enhances proliferation and tube-forming capacity of brain microvascular endothelial cells, promotes angiogenesis in ischemic regions, and improves endothelial function [95]. To further enhance brain-targeting efficiency, Hu et al. demonstrated that magnetically modified hUC-MSC-Exos exhibit improved BBB penetration and increased retention within brain tissue. These engineered exosomes deliver miR-1228-5p to inhibit the TRAF6-NOX1 signaling axis, thereby reducing ROS generation and mitigating mitochondrial damage, ultimately improving neuronal survival and cognitive function (Fig. 4) [96]. Moreover, hUC-MSC-Exos can regulate the NF-κB signaling pathway via miR-664a-5p to attenuate inflammatory responses and inhibit ferroptosis through activation of the GPX4 pathway mediated by the circBBS2/miR-494/SLC7A11 axis [97]. Overall, hUC-MSC-Exos exert coordinated neuroprotective effects in CIRI by modulating multiple signaling pathways involved in inflammation, oxidative stress, angiogenesis, and ferroptosis, thereby effectively alleviating ischemia–reperfusion–induced brain injury.

Fig. 4.

Spion-Exosomes enhance neuronal protection in MCAO mice through miR-1228-5p/TRAF6/Nox1 pathway. The exosomes are co-cultured with dopamine at pH 8.5, enhancing their homing ability to the ischemic brain in Middle Cerebral Artery Occlusion (MCAO) mice. After intravenous injection (I.V.), Spion-Ex exosomes home to the ischemic brain, where they deliver miR-1228-5p. This miRNA downregulates TRAF6, which in turn inhibits the NOX1 complex, reducing oxidative stress. Consequently, levels of malondialdehyde and ROS are decreased, providing neuroprotection. This study underscores the potential of Spion-Ex exosomes in modulating neuroinflammation and oxidative stress in ischemic brain injury. Created by figdraw.com

Protective effects of exosomes derived from iPSC-MSCs on CIRI

Induced pluripotent stem cell–derived mesenchymal stem cells (iPSC-MSCs) possess stable cellular characteristics and an extended lifespan [98]. Exosomes secreted by these cells (iPSC-MSC-Exos) have been shown to reduce infarct volume by enhancing neuronal tolerance to ischemic acidosis through inhibition of acid-sensing ion channel 1 (ASIC1). During cerebral ischemia, OGD shifts cellular energy metabolism from aerobic oxidation to anaerobic glycolysis. The accumulation of anaerobic glycolytic byproducts, such as lactate, leads to tissue acidosis, which subsequently activates ASIC1 [99]. Activated ASIC1 exacerbates neuronal death by mediating calcium influx and triggering necroptotic and other cell death pathways [100].

Studies have demonstrated that iPSC-MSC-Exos can deliver hsa-miR-125b-5p, which suppresses ASIC1 expression both in vivo and in vitro, thereby reducing calcium influx and associated cell death. This mechanism represents an important pathway through which iPSC-MSC-Exos alleviate CIRI [101]. In addition, iPSC-MSC-Exos attenuate neuroinflammation by inhibiting microglial activation and reducing the release of pro-inflammatory cytokines, including TNF-α, IL-1β, and IL-6. They also mitigate oxidative stress–induced damage, thereby improving BBB integrity and protecting neuronal cells [102]. Regarding vascular repair, iPSC-MSC-Exos promote angiogenesis and improve neurological recovery by modulating STAT3-related signaling pathways and inhibiting excessive autophagy [103]. This effect is associated with a reduced LC3-II/LC3-I ratio and decreased Beclin1 expression, accompanied by upregulation of P62 levels [104].

Although iPSC-MSC-Exos offer advantages in terms of stable cellular origin, concerns regarding tumorigenicity and immunogenicity remain critical barriers to their clinical translation. During the reprogramming process, residual epigenetic memory and potential insertional mutations may lead to the presence of oncogenic miRNAs (e.g., miR-372) within exosomes [105]. Furthermore, inter-batch variability during iPSC-MSCs differentiation may affect the compositional uniformity of the resulting exosomes. Therefore, future studies should establish rigorous strategies for genomic stability monitoring and tumorigenicity assessment. Overall, iPSC-MSC-Exos exert neuroprotective effects in CIRI through multiple coordinated mechanisms, including modulation of acidosis-related injury, neuroinflammation, and autophagy-associated angiogenesis (Fig. 5).

Fig. 5.

iPSC-MSC-Exos promote neuroprotection and regeneration via multiple pathways. Following endocytosis, exosomes are processed through early sorting endosomes and late sorting endosomes to MVB, from where they are secreted. Exosomal components exert several beneficial effects: (1) Antioxidant action by inhibiting ROS, leading to reduced oxidative stress. (2) Anti-inflammatory effects through NF-kB inhibition, resulting in decreased levels of pro-inflammatory cytokines (TNF-α, IL-1β, IL-6) and suppressed inflammation. (3) Inhibition of apoptosis by rendering caspase-3 inactive, reducing neuronal apoptosis. (4) Promotion of regeneration by enhancing VEGF and BDNF expression, leading to increased vascular density and neurofunctional recovery. These findings highlight the multifaceted roles of iPSC-MSC-Exos in promoting brain tissue repair and functional recovery. Created by figdraw.com

Comparative summary of exosomes from different sources

Exosomes derived from MSCs of diverse origins exhibit both commonalities and differences in their protective effects against CIRI. Their shared mechanism lies in modulating key pathological processes—including apoptosis, inflammatory responses, and oxidative stress—to mitigate CIRI (Table 1). However, the composition and functional characteristics of exosomes are significantly influenced by their donor cells, leading to notable differences in core mechanisms, efficacy levels, production feasibility, and clinical translation pathways. These distinctions provide a basis for selecting precise therapeutic strategies.

Table 1.

Comparison of Mechanisms and Effects of MSC-Exos from different sources in CIRI Treatment

Source	Core molecular mechanisms	Key targets/Pathways	Neuroprotective effects	Advantages and limitations	References
AMSC-Exos	Inhibits excessive autophagy; blocks microglial activation	miR-25-3p/p53/BNIP3;miR-126/NF-κB/MAPK	Reduces neuronal autophagic death; decreases TNF-α and IL-1β levels; improves neurological function in pMCAO model	Advantages: Easily accessible, low immunogenicity Limitations: Mechanism is relatively singular	[78, 81]
BMSC-Exos	Promotes M2 polarization; inhibits pyroptosis; activates angiogenesis	miR-145/FOXO1;miR-486-5p/PTEN/Akt༛AMPK/NLRP3	Reduces infarct size; microvessel density ↑2.3-fold; TUNEL-positive cells ↓35%; combined with exercise therapy, mNSS score ↓50%	Advantages: Well-established mechanistic research Limitations: Cell senescence affects stability	[87, 88, 90]
hUC-MSC-Exos	Multi-pathway synergy: anti-inflammatory, antioxidant, anti-ferroptosis, pro-angiogenesis	miR-330-5p/SETD7;miR-146a-5p/IRAK1/TRAF6༛circDLGAP4/miR-320/KLF5༛circBBS2/miR-494/SLC7A11	Significantly reduces infarct volume; BBB permeability ↓40%; brain water content ↓18%; improves spatial learning and memory; targeted delivery of Spion-Exos enhances efficacy	Advantages: broad multi-pathway regulatory potential and favorable engineering adaptability Limitations: High production cost	[95, 96]
iPSC-MSC-Exos	Inhibits acidosis-induced injury; regulates autophagy-angiogenesis axis	miR-125b-5p/ASIC1;STAT3/LC3-II/Beclin1	Infarct volume ↓32 ± 7%; angiogenesis ↑; ROS ↓; SOD activity ↑; long-term neurological deficits reduced	Advantages: Stable cell source; strong proliferation capacity Limitations: Long-term safety requires further validation	[101]

Mechanistic depth and specificity

Exosomes derived from mesenchymal stem cells of different sources exhibit distinct mechanistic emphases, reflecting the basis of their therapeutic specificity. AMSC-Exos primarily target autophagy regulation, where miR-25-3p suppresses excessive autophagy by targeting the p53/BNIP3 pathway. In parallel, miR-126 modulates the NF-κB/MAPK signaling pathways to inhibit microglial activation and attenuate secondary neuronal injury [78, 80]. BMSC-Exos mainly act on the inflammation–apoptosis–angiogenesis axis. They promote microglial polarization toward the M2 phenotype via the miR-145/FOXO1 pathway, enhance angiogenesis through the miR-486-5p/PTEN/Akt signaling cascade, and suppress NLRP3 inflammasome-mediated pyroptosis via AMPK-dependent autophagy [65, 89, 90]. hUC-MSC-Exos exhibit a multi-pathway synergistic regulatory profile. These exosomes integrate anti-inflammatory mechanisms mediated by the miR-330-5p/SETD7 and miR-146a-5p/IRAK1–TRAF6–NF-κB axes, along with pro-angiogenic effects driven by the circDLGAP4/miR-320/KLF5 pathway and ferroptosis inhibition via the circBBS2/miR-494/SLC7A11–GPX4 axis, thereby exerting comprehensive neuroprotective effects [62, 94, 95]. iPSC-MSC-Exos demonstrate advantages in enhancing tolerance to acidosis and regulating autophagy. Specifically, they alleviate acidosis-induced injury by suppressing ASIC1 through miR-125b-5p and promote angiogenesis by modulating endothelial autophagy via STAT3-related signaling pathways. Their stable cellular origin may support more sustained therapeutic efficacy [106]. Overall, MSC-Exos from different sources exhibit distinct yet complementary protective mechanisms, providing a theoretical basis for source-specific and potentially personalized therapeutic strategies tailored to the pathological features of CIRI. (Fig. 6)

Fig. 6.

Therapeutic effects of MSC-Exos from different sources in CIRI. AMSC-Exos exhibit anti-inflammatory effects by reducing pro-inflammatory cytokines and enhancing autophagy. BMSC-Exos protect against ferroptosis and cell death by inhibiting lipid peroxidation and ferroptosis/apoptosis pathways, mediated by Fe²⁺ regulation. hUC-MSC-Exos provide BBB protection by promoting tight junction integrity and reducing BBB permeability. iPSC-MSC-Exos stimulate angiogenesis and neurogenesis, enhancing endothelial proliferation and new blood vessel growth. Together, these findings highlight the diverse roles of MSC-Exos in promoting neuroprotection and tissue repair following CIRI. Created by figdraw.com

Therapeutic efficacy and applicable scenarios

MSC-Exos from different sources exhibit varying therapeutic efficacies, rendering them suitable for distinct clinical scenarios: hUC-MSC-Exos may demonstrate superior performance in reducing infarct volume, lowering BBB permeability, and improving cognitive function, making them appropriate for scenarios requiring comprehensive restorative effects [96]. Conversely, iPSC-MSC-Exos show potential for long-term neurological recovery and may be more suitable for chronic recovery phase treatments [104]. AMSC-Exos and BMSC-Exos demonstrate distinct value in intervening at specific pathological stages, such as acute inflammation and apoptosidemonstrated the most significant advantages in recent clinical translationss [79, 87].

Scalability, clinical readiness and translational challenges

In the context of large-scale production and clinical translation, MSC-Exos from various sources present unique opportunities and challenges. Concerning scalability, yield, and donor variability, hUC-MSCs and iPSC-MSCs demonstrate stable expansion capabilities with high batch-to-batch consistency, rendering them suitable for large-scale production [107, 108]. In contrast, AMSCs and BMSCs are significantly affected by donor age and health status, leading to considerable batch heterogeneity and limited expansion capacity, which hinder their standardized large-scale preparation [109, 110]. With regard to production processes and economic considerations, hUC-MSC-Exos currently utilize relatively advanced techniques, such as the integration of 3D culture with tangential flow filtration, although production in compliance with Good Manufacturing Practices (GMP) remains expensive [111]. Conversely, iPSC-MSC-Exos incur particularly high production costs due to the complexity of reprogramming, library establishment, and differentiation processes [112]. BMSC-Exos face additional challenges related to yield declines caused by cellular senescence, further exacerbating production uncertainties [113]. Regarding clinical readiness and evidence base, BMSC-Exos have entered early clinical exploration with substantial preclinical and preliminary clinical data [114]; as an emerging strategy, iPSC-MSC-Exos require systematic validation of long-term safety (e.g., tumorigenicity, immunogenicity) and process stability, with clinical translation still in its infancy [115].

Currently, MSC-Exos from various sources exhibit multi-level and multi-mechanistic neuroprotective potential in the treatment of CIRI. However, their clinical translation is hindered by challenges specific to each source. This paper systematically reviews preclinical findings, clinical advancements, existing gaps, and future directions for MSC-Exos derived from different sources (Table 2), with the aim of providing a foundation for the development of precision-based, phased therapeutic strategies. Overall, MSC-Exos from diverse sources represent a clearly stratified and functionally complementary therapeutic resource pool. Among these, hUC-MSC-Exos have shown promising preliminary results in recent clinical translations due to their balanced repair mechanisms, relatively mature production processes, and robust evidence base. iPSC-MSC-Exos show unique potential for long-term functional recovery owing to their unlimited sourcing and high homogeneity. While AMSC-Exos and BMSC-Exos are limited by donor variability and production consistency, they remain essential models for mechanistic studies and personalized exploration.

Table 2.

Current evidence and future directions of MSC-Exos in CIRI therapy

Source	Key preclinical effects	Clinical status	Main limitations	Future directions	References
AMSC-Exos	Anti-autophagy, pro-angiogenesis (miR-126), neuroprotection	Limited clinical evidence	Single-target focus; lack of long-term data	Multi-target mechanism studies; pilot clinical trials	[83]
BMSC-Exos	Anti-inflammatory, anti-apoptotic, pro-angiogenic, anti-ferroptotic; infarct reduction	Early-stage clinical studies with small cohorts	Incomplete ferroptosis network understanding; senescence-related instability	Mechanistic refinement; optimization of cell source and production	[20, 87, 91]
hUC-MSC-Exos	Anti-inflammatory/oxidative stress, BBB repair, enhanced delivery via engineering	Preliminary clinical exploration	Insufficient long-term biosafety data; high production cost	Systematic safety evaluation; scalable manufacturing optimization	[95, 96]
iPSC-MSC-Exos	Acidotoxicity resistance, angiogenesis, long-term functional recovery	No large-scale clinical trials	Unclear long-term safety; immature production systems	Comprehensive safety assessment; standardized production and QC	[99, 101]

Looking ahead, precision matching of exosome types to disease stages based on pathological characteristics will be essential. This involves aligning each exosome’s mechanistic strengths with appropriate production processes, establishing quality control systems grounded in functional potency to advance GMP standardisation, conducting preclinical studies in comorbidity models to enhance clinical predictive power, and designing phased clinical trials. Precise matching of exosome sources and delivery strategies to CIRI disease progression will drive therapeutic strategies towards individualisation, phased implementation, and scalable application.

Clinical practice and technological advances

Application of exosomes derived from mscs in preclinical research

With the advancement of cell-free therapeutic strategies, there has been a proliferation of preclinical studies on MSC-Exos across various disease models, consistently demonstrating their therapeutic potential, particularly in ischemic and inflammatory conditions [116]. Research has confirmed that MSC-Exos enhance tissue and functional outcomes in models of multi-organ ischemia-reperfusion injury, including those affecting the kidney, liver, lung, and brain. Their consistent efficacy across different models and organs suggests that MSC-Exos may target conserved pathological processes in these injuries, indicating strong versatility in disease treatment [116–119]. Compared to cell-based therapies, MSC-Exos offer several translational advantages, such as low immunogenicity, high safety, and the capability to cross the BBB. These properties help mitigate risks associated with cell transplantation, such as microvascular obstruction, abnormal differentiation, and tumorigenesis [120]. Furthermore, optimising culture conditions or applying engineered modifications can further enhance the therapeutic efficacy of MSC-Exos, demonstrating that they serve not only as natural therapeutic carriers but also as a malleable biotherapeutic platform, offering possibilities for precision intervention and personalised treatment [121].

While current preclinical findings are generally promising, it is essential to objectively acknowledge their limitations. The majority of studies on MSC-Exos are conducted using young, healthy animal models, which do not adequately replicate the complex comorbidities commonly observed in stroke patients—such as diabetes, hypertension, metabolic disorders, and aging—found in real clinical settings. This discrepancy in modeling may partially account for the inconsistent or limited efficacy observed during clinical translation. Previous research suggests that the chronic hyperglycemic environment associated with diabetes, along with enhanced oxidative stress and persistent low-grade inflammation, significantly exacerbates CIRI. This exacerbation occurs through excessive microglial activation and sustained release of pro-inflammatory factors, thereby impairing immune regulation and tissue repair processes [122, 123]. Additionally, metabolic abnormalities such as hypertension and high-fat diets may further amplify post-stroke inflammatory cascades, potentially limiting the neuroprotective effects of MSC-Exos [124–126]. As the most significant and irreversible risk factor for stroke, age-related immune decline and chronic low-grade inflammation substantially modify the post-ischemic brain microenvironment, potentially affecting the distribution, uptake, and biological effects of exosomes in target tissues [124, 127]. Consequently, the pronounced therapeutic effects of MSC-Exos observed in young, healthy animal models may not be directly translatable to clinical stroke populations. This highlights the critical need for systematic evaluation of their efficacy and safety in more clinically relevant models that incorporate comorbidities and aging.

Overall, existing preclinical studies consistently demonstrate promising applications for MSC-Exos across various ischemic and inflammatory disease models. Their advantages in terms of delivery flexibility, biosafety, and potential for engineered modulation position them as one of the most promising candidate strategies in the field of cell-free therapies for CIRI.

Clinical trial progress of exosomes in the treatment of CIRI

Presently, although clinical trials of MSC-Exos in the treatment of CIRI remain relatively limited, some preliminary outcomes have been achieved. Several studies have explored their safety and efficacy, laying the groundwork for further clinical application. Within the progress of clinical trials for CIRI treatment, MSC-Exos, as an emerging acellular therapeutic tool, has demonstrated its potential through early clinical exploration. In recent years, a series of clinical trials have been progressively initiated to assess the safety and preliminary efficacy of MSC-Exos in the context of ischemic stroke and associated neurological injuries. For example, a Phase I clinical trial (NCT03384433) provided preliminary evidence that the intravenous administration of BMSC-Exos is safe in patients with acute ischemic stroke, as no serious adverse events were reported following a single dose, and some patients demonstrated improved neurological function scores [114]. Another study (NCT05158101) investigated the feasibility of intranasal delivery of amniotic-derived MSC-Exos for the treatment of chronic post-stroke cognitive impairment. Preliminary data from this study’s multiple-dose regimen indicated a trend towards improvement in patients’ memory and executive function [128]. These initial clinical investigations preliminarily validate the safety of MSC-Exos in humans and suggest their potential neuro-regenerative effects. Nevertheless, large-scale, multicenter Phase II/III clinical trials are required to establish optimal treatment protocols and long-term efficacy. Moreover, MSC-Exos exhibit considerable potential across various clinical conditions. In the context of inflammatory bowel disease, MSC-Exos have demonstrated significant immunomodulatory and tissue-repair capabilities. They effectively suppress inflammatory cytokines, including TNF-α, IL-1β, and IL-6, while enhancing anti-inflammatory factors such as IL-4 and IL-10, thereby promoting the integrity of the intestinal barrier and facilitating mucosal healing [129, 130]. In the treatment of osteoarthritis (OA), MSC-Exos have shown promising therapeutic effects by regulating cartilage metabolic balance, inhibiting inflammatory mediators, and promoting cartilage repair and regeneration [131]. Vadhan et al. reported that MSC-Exos significantly improved OA histopathological scores, with potential therapeutic variations noted among different MSC-Exos sources [132]. Additionally, the application of MSC-Exos in neurological disorders, such as Alzheimer’s disease, is currently under investigation. Their capacity to cross the BBB allows for the direct delivery of therapeutic molecules to target cells, indicating potential in neural regeneration and the recovery of cognitive function [133, 134].

For instance, engineered exosomes with surface modifications have markedly enhanced the treatment of CIRI by facilitating the targeted delivery of quercetin to damaged neurons [73]. The bioengineering modifications of exosomes have significantly broadened their potential applications in CIRI. By employing surface modification techniques and pre-treatment of exosome contents, their therapeutic efficacy has been markedly enhanced. For example, surface-modified engineered exosomes have demonstrated substantial improvements in CIRI by facilitating the targeted delivery of quercetin to damaged neurons [135]. In the context of cancer therapy, MSC-Exos can be engineered to carry anticancer drugs, microRNAs, and other bioactive molecules, enabling targeted delivery to tumor cells. This approach not only enhances therapeutic outcomes but also mitigates chemotherapy-related side effects [136, 137]. The utilization of such bioengineered exosomes not only improves treatment specificity but also increases their potential for clinical translation [138]. Moreover, engineered or modified exosomes exhibit significantly improved targeting and therapeutic efficacy in stroke treatment through genetic and surface engineering of either the parent cells or the exosomes themselves, while maintaining their low immunogenicity, biocompatibility, and ability to traverse the BBB [139]. In comparison to natural exosomes, engineered exosomes offer precise loading and targeted delivery of functional nucleic acids and proteins. This capability facilitates the coordinated regulation of key pathological processes within the ischemic penumbra, such as neuronal survival, neuroinflammation, angiogenesis, and BBB repair. Consequently, engineered exosomes are considered an ideal vehicle for multi-targeted, comprehensive intervention in stroke pathological cascades [140, 141]. Notably, this approach partially bridges the gap between mechanistic research and clinical translation by combining the biological efficacy advantages of cell therapy with the controllability of nanodelivery systems [142]. Nonetheless, several challenges remain for clinical application, including large-scale production, quality consistency, payload standardization, and compliance with GMP [143]. Overall, engineered exosomes represent a significant developmental trajectory for cell-free precision therapy in stroke and are considered a promising therapeutic platform for future stroke research and translation.

Routes of administration and dose optimization for exosome therapy

Exosomes, as small extracellular vesicles secreted by stem cells, facilitate intercellular communication and tissue repair, demonstrating significant therapeutic potential in areas such as acute skin wound healing and tissue regeneration [144]. Compared to stem cell therapy, exosomes present a reduced risk of immune rejection and the unintended differentiation of stem cells into non-target cell types. Nonetheless, for successful clinical application, it is crucial to optimize administration routes and dosages while addressing safety concerns. The route of administration significantly influences the distribution, uptake efficiency, and therapeutic efficacy of exosomes in vivo [118].

Common administration routes include intravenous injection, intranasal delivery, intracerebral injection, and intraperitoneal injection. In recent years, research on exosome delivery strategies for the treatment of CIRI has evolved from merely comparing delivery routes to optimizing these routes and incorporating engineered modifications. Notably, intranasal administration has gained attention due to its non-invasive nature and direct access to the brain. Recent studies suggest that pre-treating the nasal mucosa with adhesive nanocarriers, such as chitosan nanoparticles, or permeation enhancers, like fluoxetine acid, can extend the retention time of exosomes within the nasal cavity and improve their trans-mucosal transport efficiency [145, 146]. In the context of intravenous administration, strategies to reduce pulmonary sequestration and enhance brain targeting include the pre-injection of blank exosomes to saturate pulmonary uptake, or the surface modification of exosomes with peptides such as RVG, which targets neuronal acetylcholine receptors, or c(RGDyK), which targets integrin αvβ3. These modifications significantly increase exosome accumulation in ischemic brain regions [147, 148]. For local delivery, hydrogel scaffold systems have progressed from basic physical loading to sophisticated, responsive delivery platforms [149, 150]. For instance, thermosensitive Pluronic F127 hydrogels enable in situ gelation post-injection, prolonging exosome retention at brain injury sites [151]; whereas ROS-responsive hydrogels intelligently release exosomes within the highly oxidative stress environment following ischemia-reperfusion, enabling microenvironment-triggered precision delivery [152]. Furthermore, constructing hybrid vesicles by fusing exosome membranes with liposomes has further enhanced the stability and biological activity of delivery systems [153]. Intracerebral injection allows exosomes to act directly at the injury site, though this invasive procedure carries risks such as infection [154]. Local administration enables higher exosome concentrations at target sites, offering high bioavailability with minimal systemic side effects. This approach is suitable for acute skin injuries, arthritis, and other local conditions [155]. Intraperitoneal injection allows exosome absorption into the circulatory system via peritoneal blood vessels and lymphatics. Compared to intravenous administration, intraperitoneal injection results in less pulmonary sequestration of exosomes, making it preferable in scenarios requiring avoidance of pulmonary accumulation [156]. However, absorption efficiency via intraperitoneal injection is significantly influenced by the intraperitoneal environment. Its efficacy may diminish in the presence of ascites, inflammation, or adhesions, potentially necessitating dose adjustment. Furthermore, digestive fluids or secretions within the peritoneal cavity may compromise exosome activity, requiring stability enhancement through surface modifications. Currently, intraperitoneal injection is predominantly employed in animal experimental studies [157].

Optimizing therapeutic dosages of exosomes requires a synergistic approach that integrates delivery strategies. For example, intranasal administration might necessitate significantly lower effective doses compared to intravenous routes due to the avoidance of systemic clearance. Conversely, the single-dose requirements for local hydrogel delivery systems may need reevaluation due to their sustained release capabilities. Current research trends highlight the use of pharmacokinetic/pharmacodynamic modeling to inform dose design. By monitoring the spatiotemporal distribution of exosomes in blood, cerebrospinal fluid, and brain tissue, exposure-effect relationships can be established to personalize dosing regimens. Furthermore, the “loading dose-maintenance dose” strategy is gaining traction, where an initial higher dose is administered intravenously or intranasally to quickly achieve therapeutic concentrations, followed by sustained effects through slow-release systems such as hydrogels. This approach has demonstrated superior neurological recovery and reduced side effects in animal models [158]. However, dose standardization remains a significant challenge [159]. Current recommendations advocate concurrent reporting of particle count, total protein content, and specific active molecules (e.g., miRNA), alongside establishing dose standards based on functional potency (e.g., neuroprotective activity units) to facilitate clinical translation [160]. Appropriate dose escalation may enhance therapeutic efficacy, yet excessive dosing risks adverse reactions. For specific conditions like CIRI, further research is required to determine optimal exosome dosing regimens [154]. In summary, optimising the administration route and dosage in exosome therapy is pivotal for its clinical application. Future efforts should focus on exploring the biological characteristics and mechanisms of action of exosomes, validating their therapeutic efficacy and safety through clinical trials, and developing more efficient and secure treatment protocols to advance the widespread application of exosomes in regenerative medicine.

Methodological considerations in exosome research: Western blot normalization strategy

Western blot (WB) analysis is a fundamental technique in contemporary exosome research, utilized for the validation of vesicle marker proteins, function-related proteins, and alterations in signaling pathways. It has been widely applied to elucidate the protective mechanisms of MSC-Exos in CIRI [161, 162]. However, the selection of a stable and reliable loading control, or internal reference, presents a significant challenge in the WB analysis of exosomes and EVs. Traditional cellular housekeeping proteins, such as β-actin and GAPDH, may not be appropriate for EVs, as they can be selectively under expressed [163] and their expression levels may vary under pathological conditions, thereby introducing bias [164]. Consequently, conventional cellular WB internal reference standards cannot be directly applied to EV analysis.

Currently, two primary alternative strategies are employed: using marker proteins for EV enrichment (such as CD9, CD63, TSG101, etc.) as relative internal references [165], or normalizing based on total protein loading [166]. Both approaches, however, exhibit certain limitations. The expression levels of a single marker protein can vary with cellular state or pathological conditions. On the other hand, the total protein loading method is heavily dependent on accurate sample quantification and consistent loading, necessitating strict adherence to operational protocols. Additionally, the heterogeneity of EV subtypes results in variations in protein composition across different EV populations, which must be considered in experimental design. Consequently, international guidelines, such as those from the Minimal Information for Studies of Extracellular Vesicles (MISEV), recommend against relying solely on a single reference protein. Instead, they advocate for the integration of multiple normalization strategies, such as the simultaneous assessment of marker proteins and total protein, tailored to the specific objectives of the experiment. Furthermore, they emphasize the importance of transparently reporting methodological details to enhance the reliability and comparability of results [167].

Discussion

Summary of key findings

This review systematically summarizes the protective effects and underlying mechanisms of MSC-Exos from four distinct sources in CIRI. AMSC-Exos alleviate neuronal damage by suppressing excessive autophagy and neuroinflammation; however, their key bioactive components and the multi-target regulatory networks involved remain to be fully elucidated. BMSC-Exos exert pleiotropic effects, including anti-inflammatory, anti-apoptotic, pro-angiogenic, and anti-ferroptosis activities. Nevertheless, the complex mechanisms by which they regulate ferroptosis within the ischemic microenvironment, as well as their long-term effects, require further in-depth investigation. hUC-MSC-Exos demonstrate coordinated multi-pathway regulation, exerting anti-inflammatory, antioxidative, anti-ferroptosis, and pro-angiogenic effects, while also contributing to BBB repair. However, beyond the reported miRNAs and circRNAs, the synergistic roles of other bioactive components (e.g., proteins and lipids), as well as the long-term safety of engineered exosomes, remain insufficiently characterized. iPSC-MSC-Exos enhance neuronal tolerance to acidosis and oxidative stress, while promoting angiogenesis and neuroregeneration. Despite these advantages, their storage stability, scalability for large-scale production, and long-term in vivo safety require further evaluation.

Challenges faced in clinical applications and potential solutions

Differences and standardization of exosome isolation methods

The diversity of exosome isolation methods and the lack of standardization represent major obstacles to the clinical translation of MSC-Exos in the treatment of CIRI. Currently, the principal isolation strategies include ultracentrifugation, filtration, size-exclusion chromatography (SEC), polymer-based precipitation, immunoaffinity capture, and microfluidic-based approaches, each with distinct advantages and limitations. (Table 3)

Table 3.

Exosome isolation technology

Isolation Technique	Principle	Advantages	Disadvantages	Recommended scenarios	References
Ultracentrifugation (Differential)	Separation based on density, particle size, and shape under centrifugal force.	Easy operation, large-scale preparation.	Low purity, risk of contamination, high time cost.	General use, not recommended for high-purity needs.	[168]
Ultracentrifugation (Density Gradient)	Separation based on density differences in a medium.	High purity, isolates exosome subpopulations.	Expensive, requires skilled operators, limited to small samples.	High-purity exosome isolation, small quantities.	[170]
Size Exclusion Chromatography (SEC)	Separation based on hydrodynamic radius or molecular weight.	Maintains biological activity, easy operation, scalable.	High equipment cost, low purity without enrichment.	Therapeutic applications, functional studies.	[172]
Ultrafiltration	Separation using membranes with specific molecular weight cut-offs.	Short processing time, ideal for large samples.	Risk of clogging, low separation efficiency.	Clinical trials, combination with tangential flow filtration.	[173]
Microfluidic Technology	Separation based on size, immunoaffinity, or density.	Rapid, efficient, cost-effective, automated.	Risk of non-specific binding, potential issues with purity.	High-purity isolation from micro-volumes.	[177, 178]

Ultracentrifugation remains the most widely used method and enables large-scale preparation; however, it is limited by relatively low purity and is time-consuming [168, 169]. Density gradient centrifugation provides higher purity but is associated with high equipment costs and low throughput [170]. Filtration-based methods are simple and rapid; among them, tangential flow filtration is suitable for continuous production, although parameter optimization is required to avoid vesicle damage and membrane fouling [171]. SEC preserves exosome structure and biological activity with relatively high purity and is often combined with ultrafiltration to enhance enrichment efficiency, albeit with higher equipment requirements [172, 173]. Polymer-based precipitation offers high yield and operational convenience, making it suitable for rapid processing of large sample volumes; however, it often co-precipitates protein contaminants, thereby compromising purity and downstream functional analyses [174, 175]. Immunoaffinity-based isolation achieves high specificity and purity through recognition of surface markers, making it well suited for mechanistic studies, but its low throughput and high cost limit its applicability for clinical-scale production [176]. Microfluidic technologies integrate multiple separation principles and offer advantages such as low sample consumption and a high degree of automation; nevertheless, their technological maturity and scalability remain limited [177, 178]. Notably, MSC-Exos obtained using different isolation methods exhibit significant variations in purity, yield, and biological activity, leading to poor comparability and reproducibility of therapeutic outcomes in CIRI models. Therefore, there is an urgent need to establish standardized isolation workflows, define critical operational parameters and quality control criteria, and set thresholds for impurity control and detection standards. Such efforts will provide a stable and reproducible technical foundation for the clinical translation of MSC-Exos in CIRI therapy.

Challenges in the preparation, characterization, and quality control of MSC-Exos

Although MSC-Exos have demonstrated considerable neuroprotective potential in CIRI, their clinical translation remains significantly hindered by the lack of standardized methodologies for preparation and characterization. Currently, commonly used isolation techniques include ultracentrifugation, density gradient centrifugation, SEC, tangential flow filtration (TFF), and polymer-based precipitation. However, these approaches differ markedly in terms of purity, yield, preservation of bioactivity, and scalability. For instance, ultracentrifugation, while widely applied, is prone to co-precipitation of protein aggregates [168]; SEC helps preserve vesicle structural integrity but is limited in enrichment efficiency [172]; and TFF offers potential for industrial-scale production, yet lacks standardized key process parameters [179]. In addition, the biological properties of MSC-Exos are highly dependent on donor source, cell culture conditions, and preconditioning strategies, resulting in inter-batch variability in composition and function. Although immortalized MSCs may improve production consistency, their safety requires careful evaluation, particularly regarding the potential risk of genetic material transfer [180]. Therefore, establishing stringent control over cell sources and culture conditions is essential during the production process.

At the level of characterization and quality control, a unified, potency-based evaluation system is still lacking. Conventional identification methods primarily rely on transmission electron microscopy, nanoparticle tracking analysis for size distribution, and detection of marker proteins (e.g., CD9, CD63, and TSG101) [181]. However, reliance solely on morphological features and marker expression is insufficient to reflect the true biological activity of MSC-Exos. More importantly, dosing metrics—such as particle number, total protein content, or RNA quantity—remain inconsistent across studies, substantially limiting comparability and accurate dose–response assessment. Notably, the MISEV guidelines released by the International Society for Extracellular Vesicles provide a valuable framework for reporting standards; however, further refinement is needed in areas such as optimization of isolation workflows, establishment of functional evaluation criteria, and development of GMP-compliant production systems [13].

Therefore, future studies should move beyond conventional physicochemical characterization to establish a quality control system centered on functional potency. This should incorporate validation of neuroprotective activity, standardization of biological effect–based dosing units, and rigorous assessment of batch-to-batch consistency. In parallel, the development of scalable manufacturing processes that comply with GMP standards is essential to reduce heterogeneity and enhance clinical reproducibility. Methodological standardization is not only critical for technical optimization but also represents a prerequisite for interpreting discrepancies in efficacy and mechanisms across studies. Only under a unified and reproducible production and evaluation framework can the multi-target therapeutic advantages of MSC-Exos be reliably translated into clinical practice.

Safety and immunogenicity of MSC-Exos

The safety and immunogenicity of MSC-Exos in CIRI therapy represent core clinical concerns. From a safety perspective, although MSC-Exos do not pose risks of uncontrolled cell proliferation or tumour formation compared to MSCs themselves, potential safety hazards remain. For instance, nucleic acids or proteins carried by exosomes may trigger immune responses, despite most studies indicating low immunogenicity [182].

In one study, Ha et al. conducted a toxicological assessment of hUC-MSC-Exos, demonstrating favourable outcomes in tests for skin sensitisation, photosensitisation, ocular and skin irritation, and acute oral toxicity, confirming their safety [183]. However, this does not imply absolute safety for mesenchymal stem cell exosomes from all sources under all circumstances; further validation through long-term, large-scale studies remains necessary. Regarding immunogenicity, MSC-Exos have demonstrated significant immunomodulatory effects and safety in treating various diseases [184]. MSC-Exos regulate innate and adaptive immune responses, suppress excessive inflammatory reactions, and promote tissue repair [86, 185]. Furthermore, MSC-Exos participate in antigen presentation, a process critical for immune response activation [184]. Wang et al. demonstrated via mouse models that hUC-MSC-Exos administered intravenously did not induce significant immune responses or toxicity [120]. This study confirms the low immunogenicity and high safety profile of MSC-Exos. Moreover, the immunomodulatory effects of MSC-Exos extend beyond their low immunogenicity to encompass their capacity to regulate immune cells. Qian et al. demonstrated that MSC-Exos exert immunosuppressive effects by modulating T-cell differentiation and activation, thereby inhibiting the production of inflammatory cytokines [186, 187]. For instance, MSC-Exos promote the generation of regulatory T cells while suppressing effector T-cell activation, thus maintaining immune homeostasis [188, 189]. In summary, MSC-Exos represent a promising extracellular therapeutic strategy due to their low immunogenicity and potent immunomodulatory effects. Future research should continue to explore the application potential of MSC-Exos across diverse diseases while further validating their safety and efficacy to advance their widespread clinical utilisation.

Standardization and quality control of exosome therapy

The field of exosome therapy is currently confronted with significant challenges related to standardization and quality control. At the standardization level, there is an absence of unified protocols for critical processes, including exosome isolation, purification, and characterization. The diverse separation techniques utilized by various laboratories—such as ultracentrifugation, density gradient centrifugation, and polymer-induced precipitation—result in considerable variability in the purity, yield, and biological activity of the exosomes produced. This variability complicates the comparison and replication of research findings [190]. In terms of quality control, there remains a deficiency in effective quality assessment indicators and detection methodologies. The quality of exosomes is affected by multiple factors, including donor-specific variations, cell culture conditions, and the exosome preparation process. Addressing the issues of consistency, stability, and efficacy in exosome products is an urgent concern that requires attention. In clinical trials investigating mesenchymal stem cell exosome therapy for Crohn’s disease-associated perianal fistulas, while some degree of efficacy has been observed, the absence of standardized production technology has hindered its broader dissemination and application [191]. Furthermore, the conditions under which exosomes are stored and transported critically affect their quality. Inadequate storage temperatures, durations, and other related factors can compromise the structural integrity and functionality of exosomes. Consequently, the development of a standardized system for exosome preparation, quality control, and storage and transportation is essential for the advancement of exosome therapy in clinical settings.

Exosomes in the long-term efficacy of treatment for CIRI

Preclinical studies indicate that MSC-Exos exert sustained neuroprotective effects in CIRI models, with long-term efficacy demonstrated across multiple dimensions including functional recovery, pathological markers, and molecular homeostasis, typically observed over several weeks.

In terms of neurological recovery, research indicates enduring effects. For example, a study demonstrated that a single intravenous administration of BMSC-Exos consistently enhanced motor function, as evidenced by decreased mNSS scores, and improved spatial learning and memory abilities in MCAO rats over a five-week observation period, as assessed by the Morris water maze test and other measures [73, 192]. This functional improvement is associated with the long-term support of exosome-mediated neural circuitry and synaptic plasticity. Mechanistically, exosome-carried miR-1228-5p facilitates the repair of hippocampal mitochondrial structure and synaptic vesicles within weeks post-injury by inhibiting the TRAF6/NOX1 axis, thereby sustaining long-term cognitive function [96]. Engineered modifications, such as targeting with c(RGDyK) peptides, further enhance the retention of exosomes at the lesion site, thereby extending the effects of functional recovery [193].

There is substantial evidence indicating sustained improvements in critical pathological markers. Imaging and histological analyses demonstrate that exosome therapy not only rapidly decreases infarct volume within 24 h but also maintains a reduced infarct size over several weeks, effectively mitigating cerebral edema [18, 194]. This prolonged structural protection is closely associated with sustained suppression of inflammation and regulation of cell death. For example, exosome-derived miR-146a-5p inhibits the IRAK1/TRAF6/NF-κB pathway, thereby maintaining its effect of reducing pro-inflammatory factors, such as TNF-α and IL-1β, for more than 72 h post-injury, and promoting the long-term polarization of microglia towards the anti-inflammatory M2 phenotype. Additionally, exosomes derived from BMSC-Exos consistently enhance antioxidant enzyme activity, such as superoxide dismutase (SOD), within brain tissue, while persistently inhibiting NLRP3 inflammasome-mediated pyroptosis. Collectively, these actions establish the long-term molecular foundation for mitigating secondary injury [88, 195].

Although exosomes have demonstrated significant neuroprotective potential in CIRI, their rapid clearance in vivo substantially limits target tissue accumulation and the durability of therapeutic effects [196]. Influenced by their intrinsic biodistribution profile and short plasma half-life, exosomes may be unable to maintain stable efficacy in chronic or delayed intervention settings, thereby posing a major challenge for clinical translation [197]. To address this limitation, Wang et al. encapsulated MSC-Exos within a thermosensitive hydrogel, establishing a biocompatible in situ reservoir that enables sustained local release. This strategy not only prolongs exosome retention but also enhances therapeutic efficacy through controlled delivery [196]. In addition, Song et al. reported that functionalization of exosomes using molecular modification strategies, such as aptamer-based approaches, can markedly improve tissue specificity and targeting efficiency, thereby further augmenting therapeutic outcomes [198].

Overall, MSC-Exos can exert sustained neuroprotective effects in CIRI models for several weeks. Their long-term efficacy is reflected in multiple aspects, including improved neurological function, persistent reduction in infarct volume, and sustained modulation of inflammatory and oxidative homeostasis. Notably, a single administration can initiate durable neurorestorative processes by modulating inflammatory cascades, promoting mitochondrial functional recovery, and enhancing synaptic plasticity. However, the rapid in vivo clearance of exosomes limits their stability and target tissue accumulation in the chronic phase, representing a key barrier to clinical application. To prolong bioavailability and enhance therapeutic persistence, strategies such as local sustained-release systems and surface engineering modifications have shown considerable promise. Future studies should further integrate pharmacokinetic optimization with dosing strategies to enable stable and controllable application of MSC-Exos in the treatment of CIRI.

Extracellular vesicle heterogeneity: source and functional implications

The biological activity of MSC-Exos is profoundly influenced by factors such as their cellular origin, culture conditions, passage number, and the method of isolation. Variability in MSC culture systems across different laboratories, including differences in serum source, oxygen concentration, and cell density, can substantially affect the composition of proteins, RNA, and lipids within the exosomes. This, in turn, results in batch-to-batch variations in therapeutic efficacy [15]. For example, hypoxic preconditioning can lead to the upregulation of metabolism-related mRNAs, such as ACADL, in exosomes, thereby enhancing their mitochondrial protective function [199]. Conversely, exosomes secreted by high-passage MSCs may display age-related phenotypes, potentially reducing their therapeutic efficacy [200]. Additionally, exosomes isolated through different methods, such as ultracentrifugation versus size-exclusion chromatography, exhibit significant differences in particle size distribution, surface protein expression, and RNA content. These variations directly impact their biodistribution and therapeutic effects in CIRI models [201].

To facilitate the clinical translation of MSC-Exos as a therapeutic product for CIRI, it is imperative to address three primary challenges. First, a GMP production system must be established, incorporating multi-tiered quality control measures such as cell banking, serum-free culture, scaled-up production, and SEC purification. Additionally, standards must be defined for particle size, markers, purity, and functional potency [15, 141, 202]. Second, regulatory uncertainties, including ambiguous classification as an Advanced Therapy Medicinal Product and the absence of standardized guidelines, must be resolved. This should be accompanied by comprehensive preclinical pharmacology and toxicology evaluations, informed by early clinical experiences with MSC-derived exosome therapies for stroke or central nervous system disorders, as evidenced by trials such as NCT05158101 and NCT03384433, which have preliminarily validated safety via intranasal and intravenous administration [114, 128]. Third, delivery strategies must be optimized to enhance targeting of the brain. This encompasses enhancing stability through lyophilization techniques, augmenting enrichment efficiency in ischemic regions via intranasal administration or targeted delivery methods (e.g., RVG modification), utilizing single or multiple dosing regimens, and determining the minimum effective dose and maximum tolerated dose through dose escalation studies [203–205]. Additionally, potential risks such as immune dysregulation, iPSC-derived tumorigenicity, organ accumulation, and batch heterogeneity necessitate rigorous long-term monitoring [206, 207]. Addressing these fundamental translational and clinical challenges is essential for establishing a robust scientific and regulatory framework for the clinical translation of MSC-Exos therapy in CIRI.

In conclusion, MSC-Exos demonstrate synergistic effects through various mechanisms, such as modulating inflammatory responses, inhibiting apoptosis and ferroptosis, promoting angiogenesis, and repairing the BBB, thereby enhancing neurological function and reducing tissue damage. Despite promising preclinical results, the clinical application of MSC-Exos encounters several challenges, including the standardization of preparation and purification methods, the establishment of scalable production systems, and the validation of long-term safety and efficacy. Future research should employ integrated and forward-looking strategies to address these challenges. A comprehensive standardized system encompassing cell source authentication, serum-free expansion, large-scale purification, and functional potency assessment must be developed. This system should include the identification of exosome-derived biomarkers and the use of artificial intelligence to facilitate personalized therapy. Additionally, combined treatment modalities that integrate engineered exosomes with rehabilitation training and neuroprotective drugs should be investigated. Drawing upon early trial experience, compliant multicentre clinical trials should be designed to clarify key therapeutic parameters and long-term safety, thereby advancing their development towards CIRI precision therapy products.

Conclusion and future perspectives

This review systematically summarizes the therapeutic potential of MSC-Exos as an emerging cell-free strategy for the treatment of CIRI. Current evidence indicates that MSC-Exos from different sources—such as AMSCs, BMSCs, hUC-MSCs, and iPSC-MSCs—exhibit distinct mechanistic advantages in CIRI intervention. Owing to their low immunogenicity, ability to cross the BBB, and multi-pathway neuroprotective effects, MSC-Exos have shown considerable promise in the treatment of cerebral ischemic injury. However, several critical challenges continue to hinder their clinical translation, including the lack of standardized methods for exosome isolation and purification, the establishment of scalable production systems, and comprehensive evaluation of long-term safety. To address these issues, it is essential to integrate advanced isolation technologies (e.g., microfluidic platforms) and establish standardized quality control frameworks that enable systematic assessment of particle size distribution, characteristic protein markers, and biological activity. In parallel, long-term studies on immunogenicity and safety are warranted. Future research should focus on the development of engineered MSC-Exos with enhanced brain-targeting capabilities, for example through surface ligand modification, genetic engineering, and nanotechnology-based approaches to improve delivery efficiency to ischemic brain regions. In addition, high-throughput screening strategies should be employed to evaluate and compare the neuroprotective efficacy of exosomes derived from different sources. Efforts are also needed to establish standardized manufacturing workflows and to optimize storage and transportation conditions in order to improve yield, purity, and batch-to-batch consistency. Moreover, rigorously designed multicenter clinical trials are required to systematically evaluate the long-term efficacy and safety of MSC-Exos in the treatment of CIRI. With continued technological advancements, MSC-Exos are expected to drive innovation in post-stroke neurorestorative strategies and expand the therapeutic landscape of cell-free approaches for neurological disorders.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

Xiaoming Wei, Wanting Han and Zhenwang Zhang wrote, edited and revised the manuscript, and confirmed the authenticity of the raw data. Wangxianyu Du and Tongzi Li drawed the schematic illustrations. Jiajia Deng, Qing Yang and Tie Peng participated in the collation of article tables. Dan Zhu provided direction and guidance throughout the preparation of this manuscript. All authors read and approved final version of manuscript. All authors contributed to the article and approved the submitted version.

Funding

This study was supported by the Natural Science Foundation of Hubei Province (Grant Number 2024AFB1025 to Dan Zhu) and Foundation of Hubei University of Science and Technology “Doctoral Initiation Fund” (Grant Number BK202414 to Dan Zhu).

Data availability

The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Generative AI and AI-assisted technologies in the writing process

In the course of preparing this manuscript, the authors utilized ChatGPT to enhance the readability and quality of the text. Subsequently, the authors meticulously reviewed and edited the content as necessary, assuming full responsibility for the final publication.

Competing interests

The authors declare no competing interests.