Ischemic stroke represents the most prevalent form of stroke, constituting 62.4% of all new stroke cases globally in 2019, as reported by the 2019 GBD study (Martin et al., 2024). The etiology of ischemic stroke is predominantly attributed to a decrease or cessation of cerebral blood flow, often precipitated by the narrowing or occlusion of blood vessels resulting from thrombosis or atherosclerosis. The TOAST classification system further categorizes ischemic stroke into various sub-types, including large-artery atherosclerosis, cardiac embolism, small-vessel occlusion, acute stroke of other determined etiology, and stroke of undetermined etiology (Adams et al., 1993).

In the context of ischemic stroke, the onset of hypoxia precipitates a disruption in cellular metabolism, culminating in cellular death and subsequent brain tissue damage. The pathogenesis of ischemic stroke is associated with a variety of factors, including vascular endothelial dysfunction, inflammatory response, and changes in blood rheology. Within 24 h of the onset of ischemia, brain tissues exhibit a substantial inflammatory response, accompanied by a significant escalation in pro-inflammatory cytokine levels, such as IL-1β and TNF-α, which further exacerbate brain injury (Ramiro et al., 2018). Ischemic stroke also triggers oxidative stress, which leads to the production of free radicals that damage cell membranes and DNA, thereby exacerbating cell death (Allen and Bayraktutan, 2009). Furthermore, research has identified a correlation between blood biomarkers, such as D-dimer and C-reactive protein, in patients with ischemic strokes and the underlying mechanisms of stroke. These biomarkers have been shown to be indicative of the pathophysiological status of different ischemic strokes (Katan and Elkind, 2018).

The classification of hemorrhagic stroke principally encompasses two distinct categories: intracerebral hemorrhage (ICH) and subarachnoid hemorrhage (SAH). According to the 2019 GBD study, ICH accounts for 27.9%, and SAH accounts for 9.7% (Martin et al., 2024). ICH is defined as the accumulation of blood in the brain tissue, with underlying causes including high blood pressure, ruptured cerebral aneurysms, and blood disorders. SAH, characterized by the entry of blood into the subarachnoid space of the meninges, is typically caused by a ruptured aneurysm. Research has demonstrated that patients afflicted with hemorrhagic stroke frequently exhibit a suboptimal prognosis and an elevated risk of recurrence. Consequently, the timely identification and management of risk factors are of paramount importance (Banerjee et al., 2020; Williams et al., 2023).

With regard to the underlying mechanisms, the occurrence of ICH is typically accompanied by the rupture of blood vessels and the leakage of blood, resulting in direct and secondary damage to brain tissue. Following the hemorrhage, the formation of a hematoma results in an increase in local pressure. The hemoglobin breakdown products within the hematoma can trigger secondary injury, leading to ischemia and cerebral edema in the surrounding brain tissue (Chen et al., 2021). Concurrent studies have demonstrated a close correlation between the activation of inflammatory responses and neuronal death, with the subsequent release of inflammatory mediators exacerbating brain injury (Dong et al., 2024).

Animal models play a critical role in ischemic stroke research because of their ability to mimic the physiological and pathological features of human stroke (Figure 1). Transient or permanent intraluminal lesion occlusion, middle cerebral artery occlusion (MCAO), and thromboembolic models are the most commonly used models to simulate human ischemic stroke. These models can effectively reproduce the major physiological changes and pathological processes of ischemic stroke. MCAO models are the main models used to study ischemic stroke, and studies have shown that they can be used not only to evaluate the efficacy of new therapies, but also to study the mechanisms of neuroprotection after stroke (Li and Zhang, 2021). The reproducibility and controllability of the MCAO model make it an important tool for drug screening and mechanistic studies. The reproducibility and controllability of the MCAO model make it an important tool for drug screening and mechanistic studies. In recent years, researchers have optimized the model, such as using microcatheter technology for local embolization, to improve the accuracy and reproducibility of the model (Komatsu et al., 2021). In addition to the commonly used rodents, larger animal models such as pigs and non-human primates have been used in ischemic stroke research, which are considered more clinically relevant due to the similarity of their brain structure to humans and are better able to mimic the pathophysiology of stroke in humans (Taha et al., 2022). Using these animal models, researchers can explore new therapeutic strategies for ischemic stroke and evaluate the safety and efficacy of drugs, thereby advancing translational clinical research.

Hemorrhagic stroke research also relies on animal models, particularly to study the pathological mechanisms and therapeutic strategies of hemorrhagic stroke. Common animal models include spontaneous cerebral hemorrhage models, collagenase-induced hemorrhage models, and blood injection models that mimic the pathology of human hemorrhagic stroke (Chen et al., 2021; Li and Zhang, 2021). Studies have shown that the use of these models can be effective in evaluating the potential of novel drugs to reduce post-hemorrhagic brain injury and improve functional outcomes. In addition, alterations in the microenvironment after hemorrhagic stroke have been found to be closely related to the inflammatory response, providing new ideas for the development of novel therapies (Harjunpää et al., 2024). Through in-depth study of animal models, scientists can better understand the mechanism of hemorrhagic stroke and provide a theoretical basis for clinical treatment.

The development of in vitro models has provided another effective tool for preclinical stroke research, and the combination of cellular models and in vitro experiments is playing an increasingly important role in stroke research. In recent years, advances in cell culture technology, particularly the advent of three-dimensional cell culture systems, have allowed researchers to better mimic the in vivo microenvironment and thus improve the prediction of drug responses. Using human induced pluripotent stem cell (iPSC)-derived organoid models, researchers can recapitulate the pathological features of neurodegenerative diseases in vitro to assess the efficacy and safety of new drugs. These models allow researchers to study cellular responses to ischemia and hemorrhage in a controlled environment, providing insight into the mechanisms of stroke (Pang et al., 2024). In vitro culture models of human neurons or glial cells can be used to simulate physiological changes in cells after ischemia or hemorrhage and to assess the effects of drugs on cell survival and function. In addition, the combination of using co-culture models of multiple cell types can better simulate the microenvironment in the brain and help to study cell-cell interactions and their role in stroke (Song et al., 2021). This combination of cellular modeling and in vitro experiments not only improves the reproducibility and reliability of studies, but also provides an important experimental basis for the development of new therapies and advances in stroke research (Van Breedam and Ponsaerts, 2022). Despite the advantages of in vitro models, they are still not a complete replacement for animal models because the in vitro environment cannot fully mimic the complex cellular interactions and signaling that occur under physiological conditions (Peng et al., 2021).

After ischemic stroke occurs, activation of the complement system can lead to a series of inflammatory responses that further exacerbate brain tissue damage. Excessive activation of complement leads to disruption of the blood-brain barrier, further exacerbating neuronal damage and death. The levels of complement components such as C3 and C5 are significantly elevated after ischemia, which enhances the local inflammatory response by promoting the infiltration and activation of inflammatory cells (Ma et al., 2019). C3a and C5a, as the key effectors of complement activation, are able to induce the activation of microglia and promote the release of inflammatory factors, which aggravate the brain tissue injury and attract more immune cells to reach the injury site (Ahmad et al., 2019; Couch et al., 2023).

After the onset of hemorrhagic stroke, the activation of complement can exacerbate the inflammatory response and cerebrovascular permeability, which can lead to the exacerbation of secondary brain damage. Elevated complement component C4 is strongly correlated with the severity of neurological deficits, suggesting that complement may serve as a biomarker for prognostic assessment in hemorrhagic stroke (Wu et al., 2023). Complement activation has also been associated with the formation of perihematomal edema, which is considered to be one of the important factors in the poor prognosis after hemorrhagic stroke (Hatchell et al., 2023).

The complement system plays an important role in nerve injury by regulating apoptosis, affecting blood-brain barrier integrity, and promoting axonal remodeling. Complement activation leads to an increased inflammatory response in neuronal cells, releasing a variety of proinflammatory factors that bind to specific receptors, promoting microglia activation, increasing neuroinflammatory responses, and ultimately leading to neuronal apoptosis (Kakarla et al., 2021). Elevated levels of complement cleavage products, such as C3a and C5a, are directly proportional to the degree of disruption of the blood-brain barrier, and inhibition of complement can significantly reduce the extent of blood-brain barrier damage and inflammatory response, and improved neurological function (Holste et al., 2021). Complement components C1q and C3 play a facilitating role in axonal remodeling and influence neural repair and regeneration (Kanmogne and Klein, 2021). Thus, regulation of the complement system may provide new strategies for the treatment of nerve injury.

There is a complex interaction between the complement system and inflammatory factors. Complement component C1 inhibitors and other anti-inflammatory factors can inhibit excessive inflammatory responses after ischemia and promote tissue repair and regeneration (Ma et al., 2019). Activation of complement components C3 and C5 can stimulate macrophages and other immune cells to release pro-inflammatory cytokines, which in turn promote further complement activation. While the production of C3a and C5a promotes the infiltration of inflammatory cells, they can also play an anti-inflammatory role by modulating the polarization state of macrophages and promoting the production of M2-type macrophages (Alawieh et al., 2020). This interaction is not only significant in the acute phase, but may also play an important role in the recovery process after stroke, affecting the ability of nerve repair and regeneration.

In the context of stroke therapy, stem cells can be administrated through three routes: intravenous (IV), intra-arterial (IA), and intranasal (IN). IN route, which offers numerous advantages over the other two pathways, was firstly proposed in 2013 in research focusing on ischemic stroke (Wei et al., 2013; Table 3).

Intranasal delivery enables stem cells to traverse the olfactory epithelium and gain access to the space in close proximity to the ostium of the turbinate. Subsequently, the cells enter the subarachnoid space through an extension of the olfactory filaments as they traverse the sieve plate. The specific histological features of these nasal structures facilitate the direct entry of cells into the brain (Galeano et al., 2018; Li et al., 2022). The IN delivery is non-invasive, has relatively simple administration procedures, and will not cause additional harm to patients. Consequently, patients can tolerate high-frequency administration through IN route (Chau et al., 2018). However, the unique and complex structure of the nasal cavity, along with its blood circulation, active enzymes, and defense mechanisms, creates significant challenges for delivering drugs to the brain through IN route (Miyake and Bleier, 2015). As a result, further preclinical studies are essential to provide sufficient safety and efficacy data. This research is crucial to confirm that stem cells can effectively reach the targeted areas of the brain when administered intranasally, paving the way for future clinical trials.

In addition to their use as a therapeutic modality, human pluripotent stem cells (hPSCs) can be derived to mimic the characteristics of the human BBB for application in brain and organoid models. In vitro BBB modeling can reproduce both physiological and pathological states of BBB. Consequently, BBB modeling represents a valuable instrument for elucidating disease etiology and a foundation for evaluating prospective pharmaceutical agents. As a result, it has been utilized to anticipate the permeability of CNS drugs (Aday et al., 2016; Yan et al., 2021).

Human pluripotent stem cells can be differentiated into BMECs, pericytes and astrocytes, which can be jointly applied with in vitro BBB models. Induced pluripotent stem cells (iPSCs)-derived BMECs exhibit a mixture of endothelial and epithelial characteristics at the transcriptional level and lack the expressions of key endothelial adhesion molecules, such as vascular cell adhesion molecule-1 and intercellular adhesion molecule-2, which mediate the interaction between immune cells and BBB, as reported before (Nishihara et al., 2020; Lu et al., 2021). Consequently, researchers have devised the extended endothelial cell culture method (EECM) to iPSCs into BMECs with comparable morphology, barrier properties, and expression of endothelial adhesion molecules to those of primary human BMECs. BMEC-like cells reproduce numerous functional and molecular characteristics of BMECs in vivo, thus becoming a crucial instrument for elucidating the pathophysiological mechanisms behind dysfunction of BBB and identifying novel therapeutic targets for BBB injury (Lippmann et al., 2020; Matsuo et al., 2023).

The iPSC-derived BBB model is a straightforward, two-dimensional structural representation of neurovascular unit (NVU) cells that does not form intricate three-dimensional structures in specific brain regions. In contrast, the microfluidic BBB-on-chip model reproduces more complex structure and function of BBB in vivo by integrating multiple iPSC-derived cells so as to simulate the multicellular structure of BBB in vitro. The model can precisely control various factors, including three-dimensional vessel-like structures, cell-cell interactions, cell-ECM interactions, substrate stiffness, and mechanical shear (Yan et al., 2021). In comparison to conventional models, it offers advantages in terms of cell permeability imaging and real-time monitoring of trans-epithelial electrical resistance (TEER) values (Srinivasan et al., 2015). Therefore, the combination of iPSC technology with microfluidics can serve as an instrument with great promise for mechanobiological studies and drug screening.

The interaction of stem cells with the immune system of the host is of equal importance in the process of recovery from a stroke. Stem cells have the capacity to modulate the function of local immune cells and alter the composition of the immune microenvironment, thereby promoting neuroprotection and regeneration (Verma et al., 2022). MSCs are also capable of influencing the immune microenvironment by modulating the function of macrophages, promoting tissue repair and regeneration (Manneken and Currie, 2023). This remodeling of the immune microenvironment has been shown to play a pivotal role in mitigating inflammation while promoting the formation of new blood vessels and the regeneration of nerve cells, thereby leading to significant improvements in functional recovery after stroke (Dekoninck and Blanpain, 2019). These observations underscore the potential of stem cell therapy in providing novel therapeutic approaches and strategies for BBB repair following stroke.

Stem cells have been shown to modulate the inflammatory response by secreting a variety of cytokines and exosome, which are the focus of immune mechanisms that promote the repair of BBB.

Ischemic foci in the aftermath of an acute ischemic stroke may release a considerable number of inflammatory cytokines, the accumulation of which represents a primary trigger for the destruction of BBB. Additionally, these factors can facilitate the recruitment of peripheral neutrophils, macrophages, and lymphocytes, which contribute to leukocyte infiltration and further promote inflammation, thereby leading to nerve damage and formation of a pro-inflammatory vicious cycle.

Numerous studies have identified mechanisms through which stem cells inhibit inflammatory responses when repairing damaged BBB tissue. The administration of human amniotic membrane-derived mesenchymal stem cells has been demonstrated to dose-dependently ameliorate neurological deficits caused by cerebral infarction. This is achieved by inhibiting the conversion of inflammatory cytokines from anti-inflammatory (M2-type) to pro-inflammatory (M1-type) in combination with by reducing the disruption of BBB and apoptosis in peri-infarcted areas (Yoshida et al., 2021). The inflammatory response occurring at BBB increases the absorption of VEGF by the endothelial cells in the brain. This process not only promotes the formation of new blood vessels, known as angiogenesis, but also raises the permeability of BBB itself (Croll et al., 2004). The transplantation of MSCs through gap junction-mediated cell-cell interactions has been demonstrated to inhibit VEGF uptake by brain endothelial cells in vitro, which in turn suppressed the inflammatory response in the brain (Kikuchi-Taura et al., 2021).

The nuclear transcription factor NF-κB acts as a regulatory factor with multiple transcriptional roles in a variety of cells in brain tissue. NF-κB is activated during cerebral ischemia reperfusion injury, and the resulting NF-κB activation exacerbates cerebral ischemic injury through multiple different mechanisms, including promotion of inflammation, induction of apoptosis, and mediation of free radical damage (Huang et al., 2010). In an animal model of spontaneous cerebral hemorrhage, bone marrow mesenchymal stem cells (BMSC)-secreted TSG-6 has been demonstrated to ameliorate injury of BBB by regulating activated astrocytes, which is achieved by suppressing NF-κB signaling pathway (Tang et al., 2021). An investigator in 2022 initially identified that MSC treatment through IA route after stroke exerted neuroprotective effects by modulating SIRT-1-mediated NF-κB pathways to attenuate inflammatory vesicle signaling and apoptosis (Sarmah et al., 2022). Aquaporin 4 (AQP4), the primary water channel protein in the mammalian brain, is found in astrocytes throughout the central nervous system and can control neurological water homeostasis. Besides, it shows an elevated expression in cerebral ischemia (Manley et al., 2000; Verkman et al., 2017). Furthermore, the investigator observed that administration of MSCs through IA route could modulate AQP4 by regulating expression of Protein kinase C-δ (PKC δ), thereby attenuating post-stroke edema (Datta et al., 2022). PKC δ exerts a significant regulatory effect on disruption of BBB following neuroinflammation, which is exacerbated by the downregulation of adhesion molecule expression and endothelial cell injury.

With regard to modulation of immune system and inflammatory response, the transplantation of MSCs following transient MCAO in ICR mice demonstrated a reduction in IgG leakage in brain parenchyma and a reversal of gap formation of ZO-1, occludin, and claudin-5 in tight junctions (Cheng et al., 2018). Upregulated IL-6 contributes to exacerbation of neuroinflammation in CNS-associated disorders. MSCs have been demonstrated to effectively suppress IL-6 secreted and released from immune cells. This is attributed to secretion of anti-inflammatory factors, including IL-10 and TGF-β, by activated microglia, astrocytes, and immune cells that have infiltrated the CNS. In addition, MSCs promote the production of regulatory T cells (Tregs), which exert an additional suppressive effect on secretion of IL-6 and contribute to reduction in neuroinflammation by playing an additional inhibitory role (Negi and Griffin, 2020; Kerkis et al., 2024). Neural stem cells (NSCs) may also mitigate ischemic injury by inhibiting a plethora of pro-inflammatory mediators. Above studies illustrate the pivotal role of disparate sources of stem cells in restoring the integrity and function of BBB following cerebral ischemia by modulating the extracellular microenvironment and attenuating neuroinflammation (Hamblin and Lee, 2021).

Increasing evidence indicates that oxidative stress-mediated damage to BBB is crucial in pathogenesis of various neurological disorders (Song et al., 2020). In this case, the antioxidant effects of MSCs following transplantation may become a promising strategy in repairing damage of BBB. In an in vitro cellular model of MCAO, CCR2-overexpressing MSCs demonstrated the capacity to restore BBB integrity by inhibiting degradation of tight junctions and excess ROS production. It can be attributed to partial mediation by redox-regulated peroxiredoxin-4 (PRDX4), a member of the PRDX antioxidant family, in restoring BBB integrity. Additionally, overexpression of CCR2 resulted in enhanced homing ability of MSCs (Huang et al., 2018). Moreover, Nrf2 emerges as a significant transcription factor governing antioxidant defense mechanisms, and the Nrf2 pathway is activated in astrocytes to safeguard themselves and their neighboring neurons from oxidative damage. After stimulation by oxidative stress, Nrf2 is released from Keap1 and transported to the nucleus. The RNAs and expressions of Nrf2 and heme oxygenase-1 (HO-1) were upregulated in astrocytes co-cultured with BMSCs. Additionally, MSC transplantation was observed to enhance the secretion of other antioxidant enzymes, such as HO-1, which were significantly ameliorated by the Cx43/Nrf2 signaling pathway, enhanced the antioxidant effects of astrocytes, and prevented them from apoptosis (Chen X. et al., 2020).

An additional potential mechanism through which MSCs facilitate repair of BBB is modulating MMP activity (Klein and Bischoff, 2011). MMPs have been observed to degrade tight junctions in early stages following cerebral ischemia or cerebral hemorrhage. In particular, the upregulation of MMP9 has been demonstrated to promote neuroinflammation, which in turn leads to the destruction of BBB (Yang and Rosenberg, 2011; Kurzepa et al., 2014). The administration of transplanted MSCs following MCAO inhibited MMP9 upregulation, attenuated neuroinflammation, and weakened neutrophil infiltration through downregulating intercellular adhesion molecule-1 (ICAM-1) (Cheng et al., 2018). Furthermore, controlling MMP9 upregulation has been demonstrated to promote downregulation of AQP4 and PKC δ, which in turn attenuates vasogenic cerebral edema and thereby protects the nerves (Datta et al., 2022). In addition, MMP9 has been reported to degrade ZO-1. Some studies of hNSC transplantation in mice with cerebral ischemia found that downregulating MMP9 could improve ZO-1 degradation, which is conductive to maintaining the integrity of BBB (Hamblin and Lee, 2021). Notably, a study has proved an association between elevated MMP8 concentrations in serum and an increased risk of small vessel stroke (Jia et al., 2021). In a model of acute cerebral infarction, transplantation of BMSC resulted in a decrease in MMP8 level in cerebrospinal fluid, leading to a reduction in the size of ischemic lesions and an improvement in neurological outcome (Bai et al., 2024).

After ischemic stroke, the Wnt signaling pathway is believed to play a crucial role in regulating neovascularization and repairing the damaged BBB (Tran et al., 2016). Research indicates that activating the Wnt/β-catenin signaling pathway encourages endothelial cell proliferation, migration, and lumen formation, which enhances neovascularization and improves blood flow to the brain (Fang et al., 2023; Wang J. et al., 2024). In the early stages following ischemic, the Wnt signaling pathway facilitates neovascularization by promoting the release of pro-angiogenic factors, such as VEGF, from endothelial cells while simultaneously inhibiting anti-angiogenic factors like endogenin. Additionally, this pathway can activate the AKT signaling pathway by reducing the expression of negative regulators, such as phosphatase and tensin homolog, which is vital for maintaining BBB integrity (Cheng et al., 2023). In summary, the neovascularization triggered by the activation of the Wnt signaling pathway after a stroke not only aids in repairing the damaged BBB but also enhances the oxygen and nutrient supply to brain tissues, thereby fostering neurological recovery. Furthermore, transplanted MSCs secrete Wnt signaling-related ligands such as Wnt3a and Wnt7a, which activate the Wnt signaling pathway in endothelial cells. These Wnt signaling molecules promote the proliferation, differentiation, and migration of vascular endothelial cells, leading to vascular neovascularization and the repair of BBB (Lippmann et al., 2012).

In vivo and in vitro experiments in a rat model of MCAO demonstrated that treatment with MSCs promoted angiogenesis and vascular stabilization, while reducing BBB leakage. This was achieved through the facilitation of communication between BMECs and astrocytes via the VEGF/flk1 and angiopoietin 1 (Ang1)/Tie2 pathways (Zacharek et al., 2007; Yamaguchi et al., 2022). Furthermore, MSCs safeguard neural stem cells from neurotoxic agents by transferring functional mitochondria into damaged cells (Babenko et al., 2018; Paliwal et al., 2018).

Following a stroke, some exosomes derived from brain cells have been observed to cause detriment on the nervous system and they can traverse BBB and enter the peripheral blood and cerebrospinal fluid. Furthermore, those in bodily fluids may serve as biomarkers in diagnosing stroke and its prognosis, offering a non-invasive means of assessment (Li et al., 2018; Otero-Ortega et al., 2019). Numerous recent studies have indicated that microRNAs in exosomes may represent potential biomarkers for diagnosis of ischemic stroke (Jiang et al., 2022). What’s more, prior research has indicated that plasma-derived exosomes miRNAs, including miR-21-5p and miRNA-30a-5p, can be undertaken as biomarkers assisting in diagnosing ischemic stroke and differentiating its stage. This offers clinicians a novel reference for early determination of ischemic stroke and potential insights into the utility of biomarkers for thrombolytic therapy (Wang et al., 2018). Cerebrovascular disease represents a significant cause of stroke in Asian children and young adults. Expressions of several miRNAs, including miR-574-5p, miR-3679-5p, miR-6165, and miR-6760-5p, changed in exosomes targeting cerebrospinal fluid sources from moyamoya disease patients (Wang et al., 2020). In an earlier study, dysregulated miRNAs were also found to be closely associated with subarachnoid hemorrhage (Zheng et al., 2012). In conclusion, further attention is required to research focusing on ascertaining the use of exosomes as biomarkers to enhance their immediate diagnostic value for stroke.

Exosome therapy offers enhanced safety and convenience compared to traditional cell therapy, and can mitigate the risks associated with cell transplantation (Larson et al., 2024). However, exosomes derived from disparate sources exhibit disparate mechanisms and effects when they are applied for stroke treatment. Exosomes from BMSCs can protect nerves by modulating immune cell responses. Besides, exosomes contribute to polarization of microglia and macrophages toward M2-type while suppressing astrocyte activation. Additionally, exosomes can speed up myelination, thereby promoting neurovascular regeneration and protecting the nerves (Chai et al., 2024). Furthermore, adipose-derived mesenchymal stem cells (ADMSCs)-derived exosomes exhibit anti-inflammatory role and inhibitory effect on oxidative stress production in the presence of ischemia-reperfusion injury (Chen et al., 2016). After the brain-derived neurotrophic factor (BDNF) was loaded into NSC-derived exosomes, rats middle cerebral artery occlusion possessed BDNF-hNSC-Exos, which significantly enhances cell survival and facilitates endogenous NSCs differentiating into neurons, while concurrently controlling microglia activation (Zhang et al., 2023). Given the impact of cell type specificity on the biological properties of exosomes, it is imperative to identify the optimal source of exosomes for stroke treatment.

Currently, development of biomedical nanomaterials technology has transferred people’s attention to exploring various nanotechnology approaches for delivery of drugs to treat neurological complications. Actually, exosomes emerge as one such approach. Nanomedicines based on exosomes can encapsulate neuroprotective agents and anti-inflammatory compounds, which can penetrate BBB so as to deliver bioactive drugs to target cells with precision. Exosomes possess a unique cellular tropism and remarkable specificity in delivering drugs. In certain instances, exosomes achieve receptor-mediated tissue targeting through the modification of specific ligands on their surface (Zhang et al., 2022; Singh et al., 2024).

The clinical translation of exosome therapy is still confronted with numerous challenges, including the standardization of exosome preparation, dosage control, optimization of routes of administration, and the assessment of long-term safety. It is increasingly essential to investigate the specific mechanisms of exosomes in stroke treatment and develop more efficacious therapeutic strategies based on exosomes.