Inhibition of Cerebral Ischemia/Reperfusion Injury by MSCs-Derived Small Extracellular Vesicles in Rodent Models: A Systematic Review and Meta-Analysis

Small extracellular vesicles (sEVs) secreted by mesenchymal stem cells (MSCs) have shown great therapeutic potential in cerebral ischemia-reperfusion injury (CIRI). In this study, we firstly performed a systematic review to evaluate the efficacy of MSCs-derived sEV for experimental cerebral ischemia/reperfusion injury. 24 studies were identified by searching 8 databases from January 2012 to August 2022. The methodological quality was assessed by using the SYRCLE ‘s risk of bias tool for animal studies. All the data were analyzed using RevMan 5.3 software. As a result, the score of study quality ranged from 3 to 9 in a total of ten points. Meta-analyses showed that MSCs-derived sEVs could effectively alleviate neurological impairment scores, reduced the volume of cerebral infarction and brain water content, and attenuated neuronal apoptosis. Additionally, the possible mechanisms of MSCs-derived sEVs for attenuating neuronal apoptosis were inhibiting microglia-mediated neuroinflammation. Thus, MSCs-derived sEVs might be regarded as a novel insight for cerebral ischemic stroke. However, further mechanistic studies, therapeutic safety, and clinical trials are required. Systematic review registration. PROSPERO CRD42022312227.


Introduction
Stroke has the characteristics of high economic burden, high incidence, high recurrence rate, high mortality rate, and high disability rate, among which the incidence of ischemic stroke (IS) is the highest [1]. Treatment of acute ischemic stroke (AIS) is based on the timely restoration of blood flow to the ischemic brain tissue by intravenous thrombolysis (IVT) and/or mechanical thrombectomy (MT) [2]. However, recanalization may aggravate the neurological deficit after cerebral ischemia, that is, cerebral ischemia-reperfusion injury (CIRI).
CIRI refers to the phenomenon that ischemic injury of the brain leads to the injury of brain cells, which is further aggravated after the recovery of blood reperfusion [3]. The specific mechanism may be related to oxygen free radicals through lipid peroxidation, protein degeneration, mitochondrial apoptosis, and activation of death receptors during reperfusion [3]. This kind of injury can further lead to aggra-vation of brain injury and neurological dysfunction, and even nerve cell death [4]. At present, CIRI has attracted increasing attention. However, neuroprotective drugs used to treat the neuronal injury caused by CIRI have limited therapeutic effects. It is of great significance to develop a more effective new approach for the treatment of CIRI.
Extracellular vesicles (EVs) are membrane vesicles that are released into the surrounding extracellular environment and can be divided into the subgroups of microvesicles and exosomes [15,16]. Exosomes are vesicles released by a cell that is between 30 and 100 nm in diameter, which is composed of a multiprotein complex, containing receptors, enzymes, transcription factors, extracellular matrix (ECM) proteins, nucleic acids (mtDNA, ssDNA, dsDNA, mRNA, and miRNA), and also lipids [17][18][19]. Previous studies have reported that exosomes showed similar or equivalent therapeutic function to MSCs to reduce injury caused by ischemia/reperfusion in a variety of tissues and organs, including the spinal cord [20], kidney [21], liver [22], heart [23], lung [24], brain , and intestine [50]. Treatment with exosomes overcomes the limitations associated with cell-based therapies and offers several advantages such as easy entry into the ischemic brain after their administration owning to their lipophilicity, less or no immunogenicity and tumorigenicity, and less incidence of occlusion in the microvasculature [27]. Due to difficulties existing in the isolation of a pure population of exosomes through the method used in present studies, we will use the term "small extracellular vesicles" (sEVs) to refer to EVs less than 200 nm in diameter, according to the updated guidelines of the International Society for Extracellular Vesicles of 2018 (MISEV2018) [51].
In this study, 24 published literatures  were systematically reviewed and meta-analyzed to evaluate the safety and efficacy of exocrine derived from mesenchymal stem cells in the treatment of CIRI. Thus, it can provide a reference basis for the clinical use of exocrine derived from mesenchymal stem cells in the future treatment of CIRI, promote the study of more and larger-scale exocrine derived from mesenchymal stem cells in the treatment of CIRI and put it into clinical applications in a timely manner.

Search Strategies. Relevant papers published between
January 2012 and August 2022 were screened in PubMed, Web of Science, Embase, Cochrane Library, CNKI (China National Knowledge Infrastructure), VIP Database for Chinese Technical Periodicals, Wanfang Database, and Chinese Biomedical Literature Database. The following keywords were used for literature retrieval: "exosomes", "extracellular vesicles", "EVs", "Reperfusion Injury", "MSCs", and "Mesenchymal Stem Cells." All searches used combinations of keywords and free words, while appropriate adjustments were made according to the corresponding database. The relevant systematic reviews and references cited in the searched articles were also filtered to avoid leaving out any potentially usable studies. Besides, other related articles were also available by examining the reference list by hand. There is no restriction on publication language or publication status. Take PubMed as an example, the specific retrieval strategies are shown in the Supplementary 2 File.

Inclusion and Exclusion
Criteria. Studies meeting the following criteria at the same time were included in this paper: (1) animal model: rats or mice of any age or gender exposed to cerebral ischemia-reperfusion injury; (2) intervention: sEVs derived from mesenchymal stem cells without any restriction on the source of cells, the administration dose, and the site of transplantation; (3) comparison: saline, phosphate buffer saline (PBS), or no treatment; (4) outcome measure: cerebral infarct volume, apoptosis rate, neurological impairment scores, brain water content, Caspase-3, tumor necrosis factor α (TNF-α), interleukin 1β (IL-1β), and interleukin 6 (IL-6). Articles meeting any of the following criteria were excluded: (1) duplicate literature; (2) reviews, conference abstracts, editorials, letters to editors, case reports, and other meta-analyses; (3) studies in humans, or in vitro studies; (4) other sEVs than derived from mesenchymal stem cells; (5) incorrect or incomplete literature data that could not be included in the statistical analysis; (6) No relevant outcomes reported.

Quality
Assessment. The quality assessment of the studies included in the present research was independently performed by two researchers using SYRCLE 's risk of bias tool for animal studies [52] recommending ten items of evaluation, evaluation results with "Y", "N" and "U" represent, respectively, low risk of bias, bias risk, and uncertain risk of bias. The disagreements between the 2 investigators were settled by means of discussion until an agreement was reached with the third investigator.

Data Extraction.
Two independent authors extracted the following details from the included studies and made a data extraction sheet: (1) the name of the first author; (2) year of publication; (3) country; (4) animal species, sex, and weight; (5) kind of anesthetic; (6) the source of MSCs; (7) MSCs isolation method; (8) MSCs characterization method; (9) MSCs positive marker; (10) EVs isolation method; (11) EVs characterization method; (12) the diameter of EVs; (13) EVs positive marker; (14) model of cerebral I/R; (15) the information of treatment group, including therapeutic drug dosage, method of administration, duration of treatment, and the same information of control group; (16) time point of extracting brain tissue; (17) mean value and standard deviation of outcomes. Because some records' published data were only in graphical format, we made efforts to contact authors for further information. When the response was not received, the numerical values were measured from the graphs by GetData Graph Digitizer 2.26 software.
2.5. Statistical Analysis. The pooled analyses were carried out with RevMan 5.3 software. I 2 statistics is calculated and reported to assess the degree of heterogeneity. A fixedeffects model (I 2 <50%) or a random-effects model (I 2 >50%) was used depending on the value of I 2 . Funnel plots were used to visually estimate publication bias. We calculated the standard mean difference (SMD) with 95% confidence intervals (CIs). Neural Plasticity

Characteristics of
①: Was the allocation sequence adequately generated and applied; ②: were the groups similar at baseline or were they adjusted for confounders in the analysis; ③: was the allocation adequately concealed; ④: were the animals randomly housed during the experiment; ⑤: were the caregivers and/or investigators blinded from knowledge which intervention each animal received during the experiment; ⑥: were animals selected at random for outcome assessment; ⑦: was the outcome assessor blinded; ⑧: were incomplete outcome data adequately addressed; ⑨: are reports of the study free of selective outcome reporting; ⑩: was the study apparently free of other problems that could result in high risk of bias. Y: yes; N: no; U: uncertain.  [35,37,39,42,45,47,49] showed significant effects of MSCs-derived sEVs for reducing the expression of proinflammatory factor IL-1β compared with control group (n = 39, SMD: −2.57, 95% CI: −3.27 to −1.86, P < 0:00001; heterogeneity: X 2 = 10:41, df = 6 (P = 0:11), I 2 = 42)% (Figure 8).
3.6. Publication Bias Analysis. The publication bias was analyzed by funnel plot, and the volume of cerebral infarction was selected to draw the funnel plot. The funnel plot of

Discussion
Ischemic stroke remains a leading cause of mortality and disability worldwide, placing a huge economic burden on society. During ischemic cerebrovascular events, the most crucial goal for treatment is to restore blood flow to the    Figure 5: The forest plot: effects of MSCs-derived sEVs for decreasing the brain water content compared with control group.

Study or subgroup Mean SD Total
Total (95% CI) Heterogeneity: 2 = 9.00, df = 5 (P = 0.11); I 2 = 44% Test for overall effect: Z = 9.17 (P < 0.00001)   Neural Plasticity ischemic penumbra. However, the restoration of blood flow will cause reperfusion injury, which eventually leads to neuronal death in ischemic penumbra via apoptosis and necrosis. Apoptosis is one of the major mechanisms of cell death during cerebral ischemia and reperfusion injury, which is the main cause of neuronal death in the central nervous sys-tem during cerebral ischemia [26]. Future focus could be directed towards inhibiting neuronal apoptosis to recover neuronal structure and function of rats after CIRI [31]. Additionally, inflammation also serves an important role during cerebral ischemia-reperfusion injury [28]. Despite efforts to reduce cerebral ischemia/reperfusion injury, an    10 Neural Plasticity ideal therapeutic approach for clinical neuroprotection against ischemia/reperfusion injury is still lacking.

SD Total Mean
Previous studies suggested that cell-based therapy using MSCs may not only be an effective reparative treatment but also a brain-protective therapy that improves neurological recovery [53][54][55]. Recently, exosomes derived from MSCs have been found to carry various kinds of mediators, miRNAs, and proteins, which can mediate the function of MSCs [56][57][58]. There is growing evidence that MSCsderived exosomes can play important roles in repairing brain-injured tissues [26]. However, we were unable to locate any systematic reviews or reviewers regarding the attenuation of CIRI by exosomes derived from MSCs in PUBMED or Web of Science. There are 24 control trials  published from 2016 to 2022 providing new evidence. Thus, an updated meta-analysis is essential. This meta-analysis is based on 24 controlled preclinical trials to demonstrate that MSCs-derived sEVs could significantly inhibit CIRI, in terms of cerebral infarct volume, apoptosis rate, neurological impairment scores, brain water content, and neuroinflammation.
Mesenchymal stem cells (MSCs) have been widely used in the experimental or clinical treatment of various ischemic diseases, but the therapeutic efficacy of MSCs on CIRI requires more research. The ethical issue is the main factor hindering advancement in clinical research. Human umbilical cord MSCs (hUMSCs), autologous adipose-derived MSCs, and autologous bone marrow-derived MCSs (BMSCs) are associated with minimal ethical controversy compared to other stem cells. Among the 24 studies included in this meta-analysis, the sources of MSCs were bone marrow MSCs in 12 studies [29, 30, 34-36, 38, 40, 42, 43, 45, 47, 49], human umbilical cord MSCs in 7 studies [27,31,32,37,44,46,48], and adipose MSCs in 5 studies [26,28,33,39,41]. Relative to human BMSCs, hUMSCs are more readily obtained, exhibit superior viability, are compatible with therapeutic methods featuring higher levels of patient acceptability and compliance, and are not susceptible to immune-mediated graft rejection [59,60]. Stroke occurs frequently between the ages of 45 and 65, and there is an autologous bone marrow aging problem. In clinical research, it can minimize pain during bone marrow extraction and enhance volunteer compliance [37].
The sEVs derived from MSCs could mitigate nerve injury after cerebral I/R confirmed by some studies. Cheng et al. demonstrated that MSCs-derived exosomes attenuate ischemia-reperfusion brain injury and inhibit microglia apoptosis might via exosomal miR-26a-5p mediated suppression of CDK6 [36]. Furthermore, Li et al. drew a conclusion that exosomal miR-26b-5p could mitigate nerve injury after cerebral I/R by targeting CH25H and inactivating the TLR pathway [32]. Hou et al. also found that negative regulation of PTEN and activation of Akt mediated the effects of miR-29b-3p on the amelioration of brain injury caused by hypoxic ischemia [29]. Subsequently, they found out that miR-29b-3p delivered in exosomes from BMSCs accelerated angiogenesis of BMECs and hindered neuronal apoptosis after ischemic stroke via targeting PTEN and activating the Akt signaling pathway [29].
In the current analysis, the quality of included studies was considered as moderate, which ranged from three to nine out of a ten. The main drop points are that no study reported the allocation scheme concealment, whether the participants and the investigator adopted the blind method and whether the blind method was applied to the result evaluation. Secondly, only 2 studies [28,46] (8.33%) reported the random allocation method. Therefore, future research should pay more attention to the application of the blind method in experimental design, and at the same time, the 11 Neural Plasticity specific experimental implementation details should be reported comprehensively, so as to improve the repeatability and reliability of animal experimental results.
Various pharmacological agents have been shown to reduce CIRI in animal models. However, lack of neuroprotectant has been routinely utilized for clinical CIRI so far. One of the major results of this meta-analysis was that MSCs-derived sEVs significantly alleviated neurological impairment scores, reduced the volume of cerebral infarction and brain water content, and attenuated neuronal apoptosis in mice or rats MCAO model, and heterogeneity was not evident, indicating that sEVs showed consistent therapeutic potential in inhibiting CIRI and alleviating neuron damage. Furthermore, the main pathways of apoptosis include extracellular signal-triggered caspase activation and intracellular apoptotic enzyme release from mitochondria, which activate caspase [61]. As we can see, caspase plays an important role in apoptosis and is involved in the common pathway of various apoptotic signals. Among them, caspase-3 is the most important terminal cleavage enzyme in the process of cell apoptosis [62]. Our results provide evidence that MSCs-derived sEVs reduced the level of caspase-3 with no obvious heterogeneity observed, which confirmed that sEVs can suppress neuron apoptosis via caspase-3 pathway.
Acute ischemic stroke has been demonstrated to induce the inflammatory response accompanied by a significant increase in the expression levels of inflammatory and proinflammatory cytokines markers [63]. Microglia release proinflammatory cytokines such as IL-1β, IL-6, and TNF-α in the acute phase of ischemic stroke, impeding postinjury neural regeneration and producing poorer long-term neurological outcomes [64,65]. Studies have proven that decreasing microglia-mediated neuroinflammation is beneficial during stroke recovery [65,66]. The other of the major results of this meta-analysis was that MSCs-derived sEVs inhibited the expression of proinflammatory factors (TNF-α, IL-1β, IL-6) and attenuated microglia-mediated neuroinflammation after ischemic stroke and heterogeneity were not observed obviously. So, MSCs-derived sEVs can reduce CIRI by anti-inflammation.
Although results of this meta-analysis were supported by powerful proof, some limitations were worth noting. First, due to the relatively short number of trials, we were unable to conduct an in-depth metaregression analysis and subgroup analysis. Second, the neurological impairment scores included in this study varied over time, and the long-term follow-up effect could not be further analyzed, so that the long-term effect was not supported by corresponding evidence. Third, the parameters we chose are insufficient to demonstrate the full range of exosomes functions. Fourth, we acknowledge that the study did not retrieve unpublished literature and was limited to Chinese and English research, and the funnel chart suggests that there may be a certain publication bias, which could exaggerate the positive results. Fifth, most of the included studies did not report allocation concealment or blind method, which has a certain risk of bias. Sixth, because some data cannot be obtained directly in research, we measured the numerical values from the graphs, which led to possible deviations between estimated and actual statistical data. Finally, 24 studies included in this meta-analysis all used healthy adult rats or mice that fail to account for preexisting stroke risk factors such as hypertension, obesity, diabetes, sex, and aging. Generally, various factors exert important effects on the outcome in this stroke model, necessitating further research.

Conclusions
MSCs-derived sEVs could effectively attenuate CIRI in vivo and inhibit microglia-mediated neuroinflammation, which might be regarded as a novel insight for cerebral ischemic stroke. The preclinical results are encouraging for preparing and using feasibility studies in humans. However, more indepth research is needed in the future to validate the therapeutic safety, in order to draw a more reliable and persuasive conclusion.

Data Availability
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

Conflicts of Interest
The authors confirm that this article's content has no conflict of interest.