Abstract Background Extracellular vesicles (EVs) secreted by mesenchymal stromal cells (MSCs) have been shown to provide significant protection against renal ischemia–reperfusion injury (IRI). Hypoxia has emerged as a promising strategy to enhance the tissue repair capabilities of MSCs. However, the specific effects of hypoxia on MSCs and MSC-EVs, as well as their therapeutic potential in renal IRI, remain unclear. In this study, we investigated the alterations occurring in MSCs and the production of MSC-EVs following hypoxia pre-treatment, and further explored the key intrinsic mechanisms underlying the therapeutic effects of hypoxic MSC-EVs in the treatment of renal IRI. Methods Human umbilical cord MSCs were cultured under normoxic and hypoxic conditions. Proliferation and related pathways were measured, and RNA sequencing was used to detect changes in the transcriptional profile. MSC-EVs from both normoxic and hypoxic conditions were isolated and characterized. In vivo, the localization and therapeutic effects of MSC-EVs were assessed in a rat renal IRI model. Histological examinations were conducted to evaluate the structure, proliferation, and apoptosis of IRI kidney tissue respectively. Renal function was assessed by measuring serum creatinine and blood urea nitrogen levels. In vitro, the therapeutic potential of MSC-EVs were measured in renal tubular epithelial cells injured by antimycin A. Protein sequencing analysis of hypoxic MSC-EVs was performed, and the depletion of Glutathione S-Transferase Omega 1 (GSTO1) in hypoxic MSC-EVs was carried out to verify its key role in alleviating renal injury. Results Hypoxia alters MSCs transcriptional profile, promotes their proliferation, and increases the production of EVs. Hypoxia-pretreated MSC-EVs demonstrated a superior ability to mitigate renal IRI, enhancing proliferation and reducing apoptosis of renal tubular epithelial cells both in vivo and in vitro. Protein profiling of the EVs revealed an accumulation of numerous anti-oxidative stress proteins, with GSTO1 being particularly prominent. Knockdown of GSTO1 significantly reduced the antioxidant and therapeutic effects on renal IRI of hypoxic MSC-EVs. Conclusions Hypoxia significantly promotes the generation of MSC-EVs and enhances their therapeutic effects on renal IRI. The antioxidant stress effect induced by GSTO1 is identified as one of the most critical underlying mechanisms. Our findings highlight that hypoxia-pretreated MSC-EVs represent a novel and promising therapeutic strategy for renal IRI. Graphical Abstract [42]graphic file with name 13287_2025_4166_Figa_HTML.jpg Supplementary Information The online version contains supplementary material available at 10.1186/s13287-025-04166-z. Keywords: Hypoxia pretreated, Mesenchymal stromal cells, Extracellular vesicles, Renal ischemia reperfusion injury, Anti-oxidative stress Background Acute kidney injury (AKI), characterized by a rapid decline in renal excretory function, poses a serious threat to patient well-being and survival [[43]1]. It is estimated that one in five adults and one in three children develop AKI during hospital stays, often as a result of sudden kidney failure or damage [[44]2]. Renal ischemia–reperfusion injury (IRI) is a leading cause of AKI, commonly occurs in patients undergoing sepsis or surgical procedures such as kidney transplantation [[45]3]. Currently, therapeutic options for AKI remain limited. While pharmacological interventions like angiotensin-converting enzyme inhibitors and angiotensin receptor blockers have been employed to modulate renal function [[46]4], they have proven ineffective in reversing the progression of kidney dysfunction. Consequently, there is an urgent need for more effective strategies to alleviate kidney tissue damage and promote regeneration are urgently needed. Mesenchymal stromal cells (MSCs) possess remarkable self-renewal capabilities and have demonstrated efficacy in treating a wide range of conditions, including tissue damage repair, inflammation suppression, and immune modulation [[47]5]. MSCs have shown significant potential in promoting kidney repair following AKI by alleviating tubulointerstitial damage [[48]6–[49]8]. Furthermore, extracellular vesicles (EVs) secreted by MSCs act as crucial cellular regulators in numerous biological processes [[50]9–[51]11]. Characterized by a lipid bilayer structure, MSC-EVs facilitate intercellular communication by delivering a diverse array of cargos, including RNAs, proteins, and lipids [[52]12, [53]13]. Notably, MSC-EVs have been reported to mitigate renal damage caused by IRI and to support cellular repair mechanisms [[54]14]. Compared to MSCs, EVs exhibit lower immunogenicity and enhanced biological safety, making them promising candidates for therapeutic applications in various diseases. Hypoxia pre-treatment significantly enhances protective capabilities of MSCs for injured tissues and organs. Hypoxia-pretreated MSCs have demonstrated the ability to alleviate spinal cord injury and promote bone repair. Additionally, hypoxic conditions influence the production and composition of MSC-EVs. However, the precise effects of hypoxic preconditioning on MSC-EV production, as well as the underlying mechanisms by which hypoxic EVs facilitate tissue repair, remain poorly understood. Therefore, in this study, we aim to investigate the impact of hypoxia on human umbilical cord MSCs and validate the therapeutic efficacy of hypoxic EVs in a rat model of IRI-induced AKI. Furthermore, we seek to elucidate the potential mechanisms underlying the therapeutic effects of hypoxic EVs through proteomic analysis. Materials and methods Cell culture MSCs were prepared and identified as described in our previous study [[55]15]. In brief, umbilical cord tissues were cut and attached individually to culture plates. The collection and subsequent use of the umbilical cord were approved by the Institutional Ethical Review Committee of Shanghai Children’s Medical Center, School of Medicine, Shanghai Jiao Tong University. Cells were cultured with a medium composed of α-MEM (12571063, Gibco) containing 5% UltraGRO™-Advanced cell culture supplement (HPCFDCGL50, Helios) and incubated in a 5% CO[2]environment at 37 °C. When the cell confluency reached 90%, MSCs were passaged at a 1:5 ratio, and the medium was replaced every 2 days. The rat renal tubular epithelial cell line NRK-52E (CL-0174, Procell Life Science & Technology Co., Ltd.) was cultured in DMEM (11965092, Gibco) supplemented with 5% fetal bovine serum (A5669701, Gibco). Cells were incubated in an incubator with 5% CO[2] at 37 °C. When cells reached 70% confluency, they were passaged at a 1:3 ratio. The medium was changed every 2 days. Hypoxia pre-treatment of MSCs MSCs were plated in 10 cm dishes and cultured until reached approximately 80%–90% confluence for downstream experiments. Hypoxia pre-treatment is conducted by different concentration of oxygen (a humidified atmosphere containing 5% CO[2] with 10%, 5% or 3% O[2] at 37 °C) for 24 h. In subsequent experiments, a 5% O[2] concentration was used for treatment. In the normoxic group, cells were incubated in an incubator with 5% CO[2] and 21% O[2] at 37 °C. CCK-8 assay The cell proliferation of MSCs and NRK-52E were determined by CCK-8 assay (K1018, APEx BIO) following the instructions. The absorbance at 450 nm wavelength was measured by a microplate reader (Thermo Fisher). Transcriptomic analysis Total RNA was extracted using Fast Pure Cell/Tissue Total RNA Isolation Kit (RC101, Vazyme) according to the manufacturer’s instructions. The primary experimental procedures for transcriptome sequencing analysis include RNA quantification and qualification, library preparation for transcriptome sequencing, clustering, and sequencing, and data analysis. HTSeq v0.6.0 was used to quantify the number of reads mapped to each gene. The FPKM of each gene was calculated based on its length and the number of mapped reads. Differential expression analysis of the two groups was performed using the DESeq2 R package (1.10.1), with an adjusted p-value < 0.05 was considered indicative of significant differential expression. GO (Gene Ontology) enrichment analysis of differentially expressed genes (DEGs) was carried out using the clusterProfiler R package. DEGs were correlated with annotations, including GO terms. Significantly enriched GO terms were identified based on corrected p-values less than 0.05. The transcriptome sequencing analysis in this studywas supported by BGIgene Co, Ltd. Isolation and purification of MSC-EVs The extraction of EVs was performed as described in the previously published literature [[56]15]. MSCs incubated under normoxic (21% O[2]) and hypoxic (5% O[2]) conditions were cultured in UltraGRO™-Advanced cell culture supplement-free media for 24 h, and the supernatants were subsequently collected for MSC-EVs extraction. In brief, the supernatants were centrifuged at 2,000 g for 20 min at 4 °C to remove cell debris, followed by a second centrifugation at 100,000 g (Beckman Coulter, Fullerton, CA) for 1 h at 4 °C. The pellet was re-suspended in cold PBS for washing and centrifuged again under the same conditions to collect the EVs. The isolated EVs were stored at   −  80 °C. Characterization of EVs The morphology of EVs was observed using transmission electron microscopy (TEM). The particle size was measured via nanoparticle tracking analysis (NTA). Surface markers including CD9, D63, and CD81 were analyzed by Western blot analysis. Protein concentration of EVs derived from 1 × 10^7 MSCs was quantified using the bicinchoninic acid (BCA) assay kit (A55860, Thermo Fisher). Western blot analysis Western blot was performed as previously described [[57]16]. The following primary antibodies were used: anti-CD9 (dilution 1:1000; ab236630, Abcam), CD63 (dilution 1:1000; ab134045, Abcam), CD81(dilution 1:1000; ab79559, Abcam), hypoxia-inducible factor-1α (HIF-1α) (dilution 1:1000; 14,179, CST), serine/threonine kinase (AKT) (dilution 1:1000; 9272, CST), phosphorylated Serine/threonine kinase (p-AKT) (dilution 1:1000; 9271, CST), phosphorylated proline-rich AKT substrate of 40 kDa (p-PRAS40) (dilution 1:1000; ab151719, Abcam), RAB22a (dilution 1:1000; ab137093, Abcam), Glutathione S-Transferase Omega 1 (GSTO1) (dilution 1:1000; 15,124–1-AP, Proteintech), Beta Actin (β-actin) (dilution 1:1000; ab8226, Abcam). Labeling and location of MSC-EVs in vivo and in vitro To label MSC-EVs, MSCs were incubated with 10 µM PKH-26 dye (HY-D1451, MCE) in serum-free culture medium at room temperature for 15 min, followed by two washes with sterile PBS to remove excess dye. The cultures were continued for MSCs 2 days and observed using fluorescence microscopy (Stellris 8, Leica). The supernatants were used to isolate PKH-26-labeled EVs in the same procedure as above. In this way, the free dye could be removed to the utmost extent. For in vivo studies, PKH-26-labeled EVs were administered into rats via the inferior vena cava. After 48 h, the kidneys of model rats underwent tissue fixation, dehydration, embedding and sectioning for subsequent observation of EVs’ location under a fluorescent microscope. Renal tissues showed green auto-fluorescence. In vitro, PKH-26-labeled EVs were added to the medium of NRK-52E cells cultured on coated coverslips for 24 h. Cells were fixed and stained for further analysis. After coverslips coated for cell culture were fixed with 4% paraformaldehyde, cytoplasm and nucleus were localized with Phalloidin (A12379, Thermo Fisher) and DAPI (62,248, Thermo Fisher), respectively. The intracellular location of EVs was then observed under a fluorescence microscope. Rats and surgical preparation The ethics Committee of Shanghai Children’s Medical Center, School of Medicine, Shanghai Jiao Tong University approved this study. Eight-week-old male Sprague–Dawley rats (180–200 g), purchased from Shanghai Jihui Experimental Animal Breeding Co., Ltd., were used in this study. All experimental rats were housed in an environment with controlled temperature and humidity and provided a normal dietAn AKI model was established as previously described [[58]17]. Rats were anesthetized with 2% isoflurane administered via the respiratory tract. In brief, rats were subjected to random allocation into four groups (n = 6), comprising those undergoing sham surgery (sham), IRI treated with PBS (IRI), IRI treated with normoxic MSC-EVs (IRI/nEVs), or IRI treated with hypoxic MSC-EVs (IRI/hEVs). A mid-abdominal incision was employed for the excision of the right kidney and temporary occlusion of the left renal pedicle for a duration of 45 min, after which the hemostatic clamp was released to restore blood flow to the kidney. Then 1 ml EVs at a protein concentration of 100 μg/ml quantified by BCA kit was administered immediately to the respective groups via the inferior vena cava. Sham-treated subjects underwent an identical surgical intervention, albeit devoid of occlusion of the renal pedicle. Rats were euthanized by excessive carbon dioxide under anesthesia conditions at 48 h after EVs injection. The kidney tissue and serum samples were collected for the following examinations. Rats were housed in the same animal facility and underwent relevant procedures performed by the same surgical personnel. All animal experiments were reported in line with the ARRIVE guidelines 2.0. Histological examination and renal function assessment Renal histological examination and renal function assessment were performed as previously described [[59]17]. Parts of the left kidney were fixed in 4% paraformaldehyde, then dehydrated in ethanol and embedded in paraffin. Kidney tissue blocks were cut into 4 µm sections and subjected to hematoxylin–eosin (H&E) staining. Subsequently, the stained sections were viewed using light microscopy. A score was given based on the grade of tubular necrosis, brush border loss, cast formation and tubular dilatation in ten randomly chosen areas. The histological scoring was determined in a blind manner based on the following criteria refer to previous literature: (0) none; (1) 0–10%; (2) 11–25%; (3) 26–45%; (4) 46–75% and (5) 76–100% [[60]17]. Renal cell apoptosis was assessed by using TUNEL staining (C1091, Beyotime). Renal tissue was analyzed using TUNEL immunohistochemistry, while coverslips coated for cell culture were analyzed using TUNEL immunofluorescence. Both renal tissues and coverslips coated for cell culture underwent fixation, permeabilization with 0.1% Triton X-100 (A110694, Sangon Biotech) for 10 min followed by wash with PBS. TUNEL staining was performed following manufacturer’s guidelines. The numbers of TUNEL-positive tubular cells in renal tissue were quantified by counting the cells in 10 randomly chosen non-overlapping fields per slide. The average fluorescence intensity in cultured cells was quantified by image J (Version 1.54 h 15, Wisconsin, USA) in 10 randomly chosen non-overlapping fields per slide. Renal cell proliferation was assessed using immunohistochemical staining for Ki67 The kidney tissue paraffin sections were permeabilized with 0.3% Triton X-100, followed by blocking with 10% donkey serum and incubation with the primary antibodies against Ki67 (dilution 1:200; ab16667, Abcam). Subsequently, the sections were incubated with secondary anti-rabbit IgG conjugated to horseradish peroxidase (HRP) (dilution 1:200; 7074S, CST). Positive staining was detected via a 3,3’-diaminobenzidine (D8001, Sigma-Aldrich) reaction. Tissue images were captured using a microscope (DMI4000 B, Leica). The blood urea nitrogen (BUN) and serum creatinine (SCr) levels were determined by the Urea Nitrogen Assay Kit (D799849, Sangon Biotech) and Creatinine Assay kit (D799853, Sangon Biotech) according to the manufacturers’ protocols. Blood samples from rats were collected 48 h after EV injection. The blood was centrifuged at 2500 rpm for 10 min, and the supernatant was collected and stored at − 80 °C for further analysis. Proteomic analysis of MSC-EVs Normoxic and hypoxic EV samples were prepared and analyzed in triplicate (n = 3). The primary experimental procedures for proteomics analysis include protein extraction, trypsin digestion, liquid chromatograph mass spectrometer (LC–MS/MS) analysis and data analysis. The resulting MS/MS data were processed using the DIA-NN search engine (Version 1.8). Tandem mass spectra were searched against the Human UniProt database (20,376 entries) concatenated with a reverse decoy databaseSubcellular localization of the differentially expressed proteins in MSC-EVs was analyzed using PSORTb software (v3.0). The Kyoto Encyclopedia of Genes and Genomes (KEGG) database was employed for KEGG pathway enrichment analysis. Reactome pathway annotation and WikiPathways pathway annotations were identified by Reactome database and WikiPathways database. The proteomic analysis in this study was supported by Jingjie PTM BioLabs. Assessment the levels of reactive oxygen species (ROS) in vivo and in vitro The fluorescent probe Dihydroethidium (DHE) (S0063, Beyotime) was used to detect the ROS levels of fresh-frozen kidney tissue. On the other hand, 10 mM 2’,7’-dichlorodihydro fluorescein diacetate (DCFH-DA, S0033S, Beyotime) was used to assess the ROS production of NRK-52E in situ visualization. Nuclei were stained with DAPI before captured the image under microscopy. Fluorescence intensity of staining was measured by Image J. Six representative visual fields of each group were counted. Assessment the levels of mitochondrial membrane potential (Δψm) NRK-52E were plated in confocal petri dishes. Cells were treated with 100 μM Antimycin (AMA) and then loaded with the potentiometric dye 500 nM TMRE (C2001S, Beyotime) at 37 °C in cell culture chamber for 20 min and Hoechst 33,342(C1027, Beyotime) for 5 min. The staining was viewed by a confocal scanning microscope after washing 3 times with PBS. Assessment of GSH and GSSG levels The concentration of GSH and GSSG of both NRK-52E and fresh rat kidney tissue were quantified by the GSH content detection kit (colorimetric method) (D799613, Sangon Biotech) and the GSSG content detection kit (colorimetric method) (D799615, Sangon Biotech). For tissue, weigh 0.1 g of fresh kidney tissue, and then add 1 mL of Reagent One. Use a homogenizer gently grind tissue samples on ice. Centrifuge samples at 8000 g and 4 °C for 10 min and save the supernatant for testing. Then, proceed with the tests as the manufacturer’s instructions. The absorbance at 412-nm wavelength was measured by a microplate reader (Aligent). Cell transfection of siRNA MSCs were seeded in a 10-cm dish at 70% confluence one day prior to transfection. Nucleotides formed transfection complexes with Lipofectamine 2000 (11,668,027, Thermo Fisher), and were added to cells and incubated for 6–8 h prior to refreshing the medium. Small interfering RNAs (siRNAs) were synthesized by Sangon Biotech Co., Ltd. based on the following sequences. siGSTO1-214: sense 5’-GCCUGAGUGGUUCUUUAAGAATT-3’, antisense 5’-UUCUUAAAGAACCACUCAGGCTT-3’. hGSTO1-411: sense 5’-CCUUGGUAGGAAGCUUUAUUATT-3’, antisense 5’-UAAUAAAGCUUCCUACCAAGGTT-3’. hGSTO1-411: sense 5’-GUUAAAUGAGUGUGUAGACCATT-3’, antisense 5’-UGGUCUACACACUUUAACTT-3’. FAM labeled siRNA was used to evaluate transfection efficiency. Cells exhibiting stable green fluorescence were considered successfully transfected. Quantitative real-time PCR (qPCR) Total RNA was initially extracted from both treated and control cell samples using a TRIzol™ Reagent (15,596,026, Thermal Fisher), following the manufacturer’s protocol. Subsequently, RNA samples were reverse transcribed into complementary DNA (cDNA) using a HiScript IV RT SuperMix for qPCR (+ gDNA wiper) kit (R423-01, Vazyme). Specific primers designed against the sequences of interest and a reference gene were utilized for qPCR amplification. The qPCR reactions were carried out in triplicate using a ChamQ Universal SYBR qPCR Master Mix (Q711-02, Vazyme) to monitor DNA synthesis in real time. Fluorescence data collected during the annealing phase were used to calculate the threshold cycle values for each target gene. Expression levels were normalized to β-actin. Here is the sequence of primers: β-actin sense 5’-CACCATTGGCAATGAGCGGTTC-3’ and antisense 5’-AGGTCTTTGCGGATGTCCACGT-3’; GSTO1 sense 5’-GAAGACGACCTTCTTTGGTGGC-3’ and antisense 5’-CTTCATGGCTGCCATCCACAGT-3’. Statistical analysis Statistical analysis was conducted using GraphPad Prism 10.0. We used the Shapiro–Wilk test to assess the normality of the data distribution. Data that were normally distributed are presented as mean ± standard deviation (SD). For comparing two groups, an unpaired t-test was used to determine statistical significance. For multiple group comparisons, one-way analysis of variance (ANOVA) was performed. P < 0.05 was deemed statistically significant. Results Hypoxia promotes proliferation and alters transcriptome of MSCs After 24 h of hypoxia pre-treatment, MSCs retained their characteristic spindle-shaped morphology with a radial distribution but exhibited a significantly increased degree of cell fusion compared to those in normoxic conditions (Fig. [61]1A). The CCK-8 assay confirmed a marked increase in cell viability in the hypoxic group, indicating that hypoxia substantially promotes cell proliferation (Fig. [62]1B). Western blot analysis further demonstrated a significant upregulation of HIF-1α expression in MSCs under hypoxic conditions. Additionally, activation of the AKT signaling pathway, known to be associated with cell proliferation, was observed, with significant increases in the levels of AKT, phosphorylated AKT (p-AKT), and the p-AKT/AKT ratio in the hypoxic group (Fig. [63]1C,D). Fig. 1. [64]Fig. 1 [65]Open in a new tab Hypoxia promotes MSCs proliferation and alters their transcriptomes. A: Morphology of MSCs under normoxic and hypoxic conditions observed via light microscope (scale bar = 200 µm). B: CCK-8 assays comparing cell viability between normoxic (n = 3) and hypoxic groups (n = 3) (*P < 0.05). C: Western blot analysis showing HIF-1α, AKT, p-AKT, and β-actin expression levels in hypoxic (n = 4) and normoxic (n = 4) conditions. D: Statistical analysis of Western blot results, showing protein expression levels of HIF-1α, AKT, and p-AKT (*P < 0.05). E: Heatmap clustering analysis of DEGs in hypoxic (n = 3) and normoxic (n = 3) conditions. F: GO enrichment analysis of the DEGs Transcriptomic analysis of hypoxia-pretreated MSCs revealed profound alterations in gene expression profiles, as illustrated by the heatmap of DEGs (Fig. [66]1E). GO analysis showed a significant enrichment of these DEGs in cellular membrane components and transporter activities, both of which are closely associated with the production and secretion of EVs (Fig. [67]1F). Alterations in the MSC characteristics and yield of EVs following hypoxia preconditioning TEM revealed that EVs derived from both normoxic and hypoxic conditions displayed similar morphologies, appearing as round or elliptical vesicles with intact structures (Fig. [68]2A). The expression levels of specific membrane protein markers associated with EVs, including CD9, CD63, and CD81 were assessed by Western blot. The results demonstrated increased expression of CD9 and CD63 in hypoxic EVs, while the alteration in CD81 expression was not statistically significant (Fig. [69]2B,C). Fig. 2. [70]Fig. 2 [71]Open in a new tab Changes in EVs production and characterization following hypoxia pre-treatment. A: TEM images showing EV morphology under normoxic and hypoxic conditions (scale bar = 200 nm). B: Western blot analysis of EV markers, including CD9, CD63, and CD81 in hypoxic (n = 3) and normoxic (n = 3) conditions. C: Relative quantification and statistical analysis of CD9, CD63, and CD81 expression levels (*P < 0.05). D: Quantification of EV protein concentration derived from 1 × 10^7 MSCs under normoxic (n = 5) and hypoxic (n = 5) conditions (*P < 0.05). E: NTA analysis showing size distribution and diameter of EVs. F: Statistical analysis of EV diameter in hypoxic (n = 5) and normoxic (n = 5) conditions. G: Statistical analysis of EV particle concentration in hypoxic (n = 5) and normoxic (n = 5) conditions (****P < 0.0001). H: Western blot analysis of HIF-1α, PRAS40, RAB22a, and β-actin levels at 21% (normoxic, n = 3), 10% (n = 3), 5% (n = 3), and 3% O[2] conditions (n = 3). I: Statistical analysis of HIF-1α protein expression levels normalized to β-actin (*P < 0.05). J: Western blot analysis of PRAS40 protein levels normalized to β-actin (*P < 0.05). K: Western blot analysis of RAB22a protein levels normalized to β-actin (*P < 0.05) Measurement of protein concentration in EVs produced by 1 × 10^7 MSCs revealed that the EVs from hypoxia pre-treated MSCs had a significantly higher concentration compared to those from the normoxic group (Fig. [72]2D). NTA demonstrated that the size of EVs ranged from 50 to 500 nm in both groups, indicating no significant difference (Fig. [73]2E,F). However, the particle concentration in the hypoxic group was significantly elevated compared to the normoxic group (Fig. [74]2G). Furthermore, we examined the expression of proteins involved in EV secretion in MSCs, specifically PRAS40 and RAB22a. PRAS40 expression exhibited a marked increase under 5% oxygen conditions, accompanied by a significantly elevated expression of HIF-1α under both 10% and 5% oxygen conditions. Although RAB22a expression did not increase under 10% oxygen conditions, but it was significantly up-regulated under 5% and 3% oxygen conditions (Fig. [75]2H–K). Hypoxic EVs exhibited superior efficacy in promoting recovery from renal IRI PKH26-labeled EVs were systemically administered to rats undergoing IRI. After 48 h, PKH26 fluorescence was clearly detectable within the kidney tissue (Fig. [76]3A). H&E staining revealed numerous necrotic areas in the proximal epithelium and abundant tubular protein casts in the IRI -affected kidneys. In contrast, treatment with both normoxic and hypoxic EVs resulted in a reduction of tubular lesions (Fig. [77]3B). Renal injury scoring, based on structural alterations, demonstrated that hypoxic EVs were significantly more effective at alleviating morphological changes associated with IRI compared to their normoxic counterparts (Fig. [78]3E). Ki67 staining, a marker of cell proliferation, showed enhanced tissue proliferation following IRI, which was further augmented by hypoxic EV treatments (Figs. [79]3C,F). TUNEL staining revealed IRI-induced apoptosis, which was markedly attenuated by the administration of both normoxic and hypoxic EVs, with hypoxic EVs exhibiting greater efficacy (Fig. [80]3D,G). Fig. 3. [81]Fig. 3 [82]Open in a new tab Hypoxia EVs attenuate renal IRI in rat A: Localization of MSC-EVs labeled with PKH-26 in IRI kidneys, with DAPI staining for nuclei; background shows spontaneous green fluorescence of the kidney (scale bar = 100 µm). B: H&E staining of rat kidneys across different groups (scale bar = 100 µm): sham surgery (Sham), renal IRI injected with PBS (IRI), with normoxic EVs (IRI/nEVs), or with hypoxic EVs (IRI/hEVs).C: Representative images of Ki67-positive cells in the tubulointerstitial area (scale bar = 100 µm). D: Representative images of apoptotic cells identified by TUNEL staining (scale bar = 100 µm). E: Statistical analysis of kidney injury scores assessed from histological slides (n = 6). F: Quantification of Ki67-positive cells per high-power field (HPF) (n = 6). G: Quantification of apoptotic cells per HPF (n = 6). H: SCr levels across different groups (µmol/L) (n = 6). I: BUN levels in different groups (µg/mL) (n = 6). (*P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001) Renal function was assessed by measuring SCr and BUN. Rats with renal IRI treated with PBS exhibited a rapid increase in SCr and BUN levels. In contrast, treatment with normoxic and hypoxic EVs significantly attenuated these increases (Fig. [83]3H,I). Notably, hypoxic EV-treated kidneys showed lower renal injury scores, higher Ki67 expression, fewer apoptotic cells, and reduced SCr levels compared to those treated with normoxic EVs. MSC-EVs promote anti-apoptosis of renal tubular epithelial cells in vitro To model hypoxic injury to renal tubular epithelial cells, rat tubular epithelial cell line NRK-52E were exposed to AMA, an inhibitor of mitochondrial electron transport chain. As shown in Fig. [84]4A, AMA induced a dose-dependent reduction in NRK-52E cell viability, with a concentration of 100 μM causing approximately a 50% decrease in cell viability. Treatment with both normoxic EVs and hypoxic EVs restored cell viability (Fig. [85]4B). To investigate the uptake of MSC-EVs by renal tubular epithelial cells in vitro, MSCs were labeled with PKH-26, enabling the secreted EVs to carry red fluorescence (Fig. [86]4C). Internalization of PKH26-labeled EVs was confirmed in NRK-52E cells. Quantitative analysis demonstrated that hypoxic EVs exhibited significantly enhanced cellular uptake compared to normoxic EVs (Fig. [87]4D,F). Fig. 4. [88]Fig. 4 [89]Open in a new tab Effects of hypoxic EVs on NRK-52E cells under AMA-induced injury. A: CCK-8 assay demonstrating the toxicity of various concentrations of AMA on NRK-52E cells (n = 5). B: Cell viability of NRK-52E exposed to 100 µM AMA for 24 h across different treatment groups (n = 5). C: MSCs labeled with PKH-26 (scale bar = 100 µm). D: Internalization of PKH-26 labeled EVs in the cytoplasm of NRK-52E (scale bar = 50 µm; blue: DAPI-stained nuclei, green: Phalloidin-stained cytoskeleton, red: PKH-26). E: Representative images of TUNEL staining in NRK-52E (scale bar = 200 µm). F: Quantification of internalized PKH-26-labeled EVs derived from normoxic (n = 3) and hypoxic MSCs (n = 3) in NRK-52E (*P < 0.05). G: Quantification of apoptotic cells HPF (n = 5) (*P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001) TUNEL staining demonstrated that AMA induced significant apoptosis in renal tubular epithelial cells. Both normoxic EVs and hypoxic EVs effectively attenuated cell apoptosis, with hypoxic EVs showing superior efficacy in protecting renal tubular epithelial cells from AMA-induced injury (Fig. [90]4E,G). Hypoxia preconditioning up-regulated antioxidant stress pathway revealed by the proteomics of EVs To elucidate the underlying mechanism by which hypoxic EVs exhibit superior efficacy for renal IRI, we conducted mass spectrometry analysis of the proteomic cargo of EVs. PCA clearly distinguished the proteomic signatures of hypoxic EVs from those of normoxic EVs, highlighting hypoxia-induced alterations in the protein profiles of MSC-EVs (Fig. [91]5A). We identified significant differences in protein expression levels between the two groups, with 264 proteins up-regulated and 288 proteins down-regulated in hypoxic EVs (P < 0.05) (Fig. [92]5B). Among these significant regulated proteins, GSTO1, an enzyme involved in the redox reaction of GSH, exhibited the most pronounced difference (Fig. [93]5C). Unsupervised hierarchical clustering analysis further validated the distinct protein patterns between normoxic and hypoxic EVs, as shown in the heatmap (Fig. [94]5D). Subcellular localization classification of the differentially expressed proteins indicated that the predominant proteins were extracellular (38.59%) and cytoplasmic (19.75%) (Fig. [95]5E). Functional enrichment analysis of up-regulated proteins demonstrated significant enrichment in GSH metabolism-related pathways, such as cysteine and methionine metabolism and glutathione metabolism, as identified by KEGG analysis (Fig. [96]5F). Additionally, there was notable enrichment of pathways associated with nuclear factor erythroid 2–related factor 2 (NRF2), a critical regulator of cellular oxidative stress resistance. This includes pathways related to nuclear events mediated by NFE2L2, the KEAP1-NFE2L2 pathway, photodynamic therapy-induced NRF2 survival signaling, and the NRF2 pathway itself (Fig. [97]5G,H). Together, these findings suggest that hypoxia preconditioning enhances the expression of antioxidant stress-related proteins in MSC-EVs. Fig. 5. [98]Fig. 5 [99]Open in a new tab Hypoxia up-regulated antioxidant proteins in MSC-EVs. A: PCA distinguishing normoxic (n = 3) and hypoxic EVs (n = 3). B: Summary of protein types in EVs, with red indicating up-regulated and blue indicating down-regulated proteins. C: Top 30 proteins showing the most significant differences in expression. D: Heatmap clustering analysis of differentially expressed proteins. E: Subcellular localization analysis of up-regulated and down-regulated proteins. F: KEGG functional enrichment analysis of up-regulated differential proteins. G: WikiPathways enrichment analysis of up-regulated differential proteins. H: Reactome pathway enrichment analysis of up-regulated differential proteins Validation of hypoxic EVs in alleviating renal IRI via antioxidant mechanisms We first evaluated the differential effects of hypoxic and normoxic EVs on oxidative stress in rat kidney tissue. The results demonstrated that IRI significantly elevated ROS levels. Notably, hypoxic EVs markedly reduced ROS levels in IRI-affected kidneys compared to normoxic EVs (Fig. [100]6A,C). In vitro assays with NRK-52E cells exposed to AMA corroborated this trend, showing that ROS production and mitochondrial membrane potential changes mirrored those observed in rat kidney tissue (Fig. [101]6B,D,E). GSH plays a pivotal role in antioxidant defense and maintaining intracellular redox balance. In NRK52E cells, we assessed the intracellular levels of GSH, GSSG, and the GSH/GSSG ratio following EV treatment. LC–MS analysis revealed that AMA significantly reduced GSH levels while increasing GSSG levels, thereby decreasing the GSH/GSSG ratio. However, hypoxic EVs significantly restored GSH and GSSG levels, surpassing the restorative effects of normoxic EVs. Ultimately, the GSH/GSSG ratio in hypoxic EV group saw a significant increase, even reaching levels comparable to the sham group (Fig. [102]6F–H). Proteomic analysis revealed significant alterations in enzymes associated with GSH metabolism, including glutathione S-transferases (GSTs), glutathione synthetase (GSS), glutathione reductase (GSR), and glutathione peroxidase 1 (GPx) (Fig. S[103]1A–F). GSTs, GSS, and GSR showed increased levels in hypoxia-preconditioned EVs, while GPx1 levels were reduced. Within the GSH metabolism pathway, GSS contributes to GSH synthesis, GSR converts GSSG back into GSH, and GSTs facilitate the conjugation of GSH with oxidative metabolic byproducts. In contrast, GPx drives the conversion of GSH to GSSG (Fig. S[104]1G). Furthermore, a significant accumulation of superoxide dismutase (SOD) was observed in hypoxic EVs (Fig. S[105]1H–I). Fig. 6. [106]Fig. 6 [107]Open in a new tab Hypoxic EVs alleviate acute renal IRI through the cargo of antioxidant proteins. A: Representative images of reactive oxygen species (ROS) fluorescence in rat kidneys (scale bars = 100 µm). B: Representative images of mitochondrial membrane potential injury in NRK-52E cells (scale bars = 50 µm). C: Quantitative analysis of ROS fluorescence intensity in rat kidneys (n = 5). D: ROS fluorescence quantified in NRK-52E following 24-h exposure to 100 µM AMA (n = 5). E: Quantitative analysis of mitochondrial membrane potential fluorescence intensity in NRK-52E cells (n = 5). F: GSH levels across different experimental groups (n = 5). G: GSSG levels across different experimental groups (n = 5). H: GSH/GSSG ratio across different experimental groups (n = 5). I: Western blot analysis of GSTO1 expression levels in MSCs and EVs following hypoxic treatment. J: Statistical analysis of GSTO1 protein expression in MSCs, normalized to β-actin, as determined by Western blot (n = 3). K: Statistical analysis of GSTO1 protein expression in EVs, normalized by total protein concentration and volume (n = 3). L: mRNA expression levels of GSTO1 post-transfection with GSTO1-targeting siRNA, assessed by qPCR (n = 3). M: Protein expression levels of GSTO1 post-transfection with GSTO1-targeting siRNA, evaluated by Western blot (n = 2). N: Representative images of ROS fluorescence in NRK-52E (scale bars = 50 µm). O: Quantitative analysis of ROS fluorescence in NRK-52E (n = 5) (*P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001) Next, the levels of TGFβ, interleukins, growth factors, and chemokines in hypoxia-preconditioned EVs were quantified. The results revealed that hypoxic EVs led to the upregulation of several soluble mediators, including TGFβ1, PDGFRA, VEGFC, and CCL2 (Fig. S[108]2A–F). Additionally, renal protein levels of IL-10 and VEGFA were measured following EV treatment. However, the changes in IL-10 and VEGFA levels in the kidneys were not statistically significant (Fig. S[109]2G–I). We further validated the protein level of GSTO1, identified as the most differentially expressed protein in hypoxic EVs. The results confirmed that GSTO1 was upregulated in both hypoxia pre-treated MSCs and derived EVs (F[110]igs. [111]6I–K). To investigate the role of GSTO1 in mitigating oxidative stress in hypoxic EVs, we knocked down GSTO1 using three different siRNAs. Except for siGSTO1-214, both siGSTO1-411 and siGSTO1-601 effectively silenced the mRNA and protein expression of GSTO1 in MSCs (Fig. [112]6L,M). Hypoxic EVs from MSCs transfected with both siGSTO1-411 and siGSTO1-601 didn’t reduce the ROS levels of renal tubular epithelial cells as effectively as those transfected with a negative control siRNA (Fig. [113]6N,O). Furthermore, GSTO1 knockdown diminished the ability of the hypoxic EVs to restore the mitochondrial membrane potential in the renal tubular epithelium (Fig. S[114]3A,C) and reduced the protective effect against AMA-induced apoptosis (Fig. S[115]3B,D). Upon treating IRI model rats with hypoxic EVs in which GSTO1 was knocked down, we observed notable changes. ROS level significantly increased compared to the control group receiving hypoxic EVs (Fig. S[116]4A,D). Additionally, there was a noticeable decrease in Ki67 expression and an increase in apoptosis (Fig. S[117]4B–C,E–F). Further analysis of kidney function revealed that hypoxic EVs with GSTO1 knockdown were less effective in mitigating elevated SCr and BUN levels resulting from IRI compared to the hypoxic EVs group (Fig. S[118]5A,B). Taken together, these findings indicate that GSTO1 in hypoxic EVs is essential for alleviating renal IRI. Discussion IRI is a leading cause of AKI, resulting in damage to renal tubules and rapid deterioration of renal function. The current study demonstrates that hypoxia not only stimulates the proliferation of MSCs but also enhances the secretion of EVs. Additionally, MSC-EVs pre-treated with hypoxia exhibit a pronounced ability to protect against renal IRI. Importantly, we have presented novel evidence that hypoxia pre-treated MSC-EVs mitigate acute renal IRI through an anti-oxidative stress mechanism, with the antioxidant protein GSTO1 playing a crucial role in this protective process. Conventional in vitro cell culture typically occurs under ambient oxygen conditions (21% O[2]), referred to as ‘normoxia’. In contrast, in vivo, MSCs are often located in niches with lower oxygen levels [[119]18]. Culturing MSCs under hypoxic conditions aims to mimic this natural microenvironment. MSCs residing in hypoxic niches, with oxygen tensions ranging from 3 to 9%, retain their capacity for self-renewal, proliferation, migration, and differentiation [[120]19, [121]20]. While previous studies have demonstrated that hypoxia preconditioning can enhance MSC proliferation and differentiation [[122]21], the precise mechanisms remained unclear. In our study, we revealed that hypoxia activates the HIF-1α/AKT signaling pathway to promote MSC proliferation. The cellular response to hypoxic stress is primarily mediated by HIF-1α [[123]22], and overexpression of HIF-1α has been shown to enhance MSC proliferation and osteogenic differentiation [[124]23]. The AKT signaling pathway plays a critical role in cell proliferation, and previous research has indicated that HIF-1α contributes to the upregulation of p-AKT protein expression [[125]24, [126]25]. These findings collectively support the involvement of the HIF-1α/AKT pathway in MSC proliferation under hypoxic conditions. MSCs contribute to tissue repair through endocrine or paracrine mechanisms, with EVs being one of their main means of communication [[127]26]. These EVs play a crucial role in tissue repair by delivering proteins and genetic materials. Importantly, EV production is influenced by the surrounding microenvironment. The impact of MSC-EVs can be amplified under certain physical and biological stimuli, such as hypoxia, lipopolysaccharide, and TNF-α. These conditions not only boost EV production but may also modify their contents, ultimately enhancing their beneficial effects [[128]27–[129]30]. The RAB family comprises key signaling molecules involved in EVs production, while PRAS40 is involved in the release of EVs following cellular stress [[130]31, [131]32]. The production of MSC-EVs significantly increased under hypoxic culture conditions, with the activation of RAB22a and PRAS40 observed in MSCs in this study. Furthermore, the plasma membrane plays a crucial role in shaping and facilitating the functions of EVs during their generation. It is actively involved in cell signaling, membrane adaptability, and uptake by recipient cells. Interestingly, our transcriptomic analysis of MSCs revealed that hypoxia activates signaling pathways related to cell membrane components and transporter activities, both of which are associated with enhanced EV production. Several studies have documentedthat hypoxic EVs are effective in mitigating organ ischemia injury, such as in the brain and limbs, by transporting microRNAs or proteins [[132]33–[133]35]. Zhang et al. [[134]36] reported that hypoxic preconditioning of MSCs can enhance the repair of injured kidneys, with increased angiogenesis and antioxidant effects playing a role. Building on this, the therapeutic effects of hypoxic EVs on renal ischemia–reperfusion injury (IRI) were explored in this study. The results indicated that hypoxia preconditioning enhances the ability of MSC-EVs to repair IRI kidneys, improving both renal morphology and function. During AKI, tubular cell apoptosis and proliferation occur. Both in vivo and in vitro experiments demonstratedthat hypoxia preconditioning of MSC-EVs significantly reduced tubular cell apoptosis and promoted their proliferation, demonstrating a stronger protective effect against injury. To further investigate the underlying mechanisms, our results revealed that hypoxic preconditioning enhances the production of MSC-derived EVs, leading to a significant increase in EV particle number and protein concentrations. This increase may partly explain the enhanced therapeutic efficacy observed with hypoxic EVs. Additionally, EVs exert their effects through biologically active internal components, including proteins, lipids, and nucleic acids. However, the primary contributor among these components remains unclear. While substantial research has focused on the alterations of miRNAs within EVs, less attention has been paid to their protein content, which can directly influence recipient cells [[135]37]. Recently, EVs have emerged as a novel mechanism for protein delivery [[136]38]. Eirin et al. [[137]39] emphasizedin their study on MSC-EV-mediated kidney injury repair that the delivery of IL-10 protein is a key factor in the protective effects of MSC-EVs. In this study, the protein levels in hypoxia preconditioned EVs were examined through proteomic analysis. Various injury repair-related factors were found to be altered, including proteins associated with oxidative stress. Oxidative stress plays a crucial role in the pathogenesis of ischemic kidney injury, and it has been established that MSC-EVs can reduce oxidative reactions and protect kidney function [[138]40, [139]41]. In this study, MSC-EVs were also found to significantly reduce oxidative stress levels in injured kidneys, with hypoxic EVs exhibiting an even greater effect, highlighting their potent antioxidant properties. We observed a significant accumulation of oxidative stress-related proteins in hypoxic EVs, particularly enzymes involved in GSH metabolism and superoxide dismutase. Among all differentially expressed proteins, GSTO1 was one of the most notably upregulated proteins, which is an important enzyme that promotes the reduction of ROS through redox reactions involving GSH. GSTO1has been shown to play a significant role in protecting the function of kidneys in end-stage renal disease [[140]42]. It has also been reported that GSTO1 is involved in regulating oxidative stress and exerting protective effects in various diseases [[141]43]. To further investigate the role of GSTO1 in hypoxic EVs, GSTO1-knockdown hypoxic EVs were obtained and used to treat ischemic renal injury. The data showed that, in the absence of GSTO1, the antioxidant effects of hypoxic EVs were significantly diminished, along with a reduction in their protective effects on tubular cells, ultimately leading to impaired kidney function protection. This finding suggests that the antioxidant protein GSTO1 plays a pivotal role in the protective effects of hypoxic EVs against renal IRI. Conclusions This study demonstrates that EVs derived from hypoxia-preconditioned MSCs provide renal protection against IRI. The antioxidant protein cargo, particularly GSTO1, plays a pivotal role in mitigating acute renal IRI. These findings highlight a promising therapeutic avenue for the clinical treatment of renal IRI. Supplementary Information [142]Additional file1.^ (6.3MB, pdf) Acknowledgements