Abstract Introduction Advanced glycation end products (AGEs) play a critical role in the development of vascular diseases in diabetes. Although stem cell therapies often involve exposure to AGEs, the impact of this environment on extracellular vesicles (EVs) and endothelial cell metabolism remains unclear. Methods Human umbilical cord mesenchymal stem cells (MSCs) were treated with either 0 ng/ml or 100 ng/ml AGEs in a serum-free medium for 48 hours, after which MSC-EVs were isolated. The EVs were characterized by morphology, particle size, and protein markers of MSC-EVs, and microRNA (miRNA) sequencing was performed to identify differentially expressed miRNAs. MSC-EVs were cocultured with human umbilical vein endothelial cells (HUVECs) to assess effects on cell viability, metabolic activity, oxidative stress, and antioxidant capacity. Tube formation and glucose transporter protein analyses were conducted to evaluate the angiogenic ability and glucose metabolism capacity. Results MSC-EVs ranged from 30 to 150 nm, which is consistent with exosomal properties. AGEs treatment reduced MSC viability but had minimal effect on EV morphology and protein markers. miRNA sequencing showed downregulation of hsa-miR-223-3p and hsa-miR-126-3p_R-1, with upregulation of hsa-miR-574-5p, implicating changes in glycolytic and oxidative phosphorylation pathways. MSC-EVs treated with AGEs decreased HUVEC viability (P < 0.05), pH (P < 0.05), adenosine triphosphate (ATP) metabolism (P < 0.05), glucose metabolism (P < 0.05), while enhancing glycolysis processes, including glycolytic activity, capacity, and reserve (P < 0.05). This likely resulted from impaired mitochondrial function, including reduced ATP production, maximal respiration, basal respiration, and spare respiratory capacity (P < 0.05), or increased reactive oxygen species (ROS) (P < 0.05) and glucose-6-phosphate dehydrogenase (G6PD) activity (P < 0.05). In addition, AGEs reduced glucose transporter types 1, 3, and 4 (GLUT1, GLUT3, GLUT4), and synthesis of cytochrome c oxidase 2 expression (P < 0.05), along with angiogenic capacity (P < 0.05) in HUVECs. Conclusion Exposure to AGEs diminishes the therapeutic potential of MSC-derived EVs by disrupting energy metabolism and promoting metabolic reprogramming in endothelial cells. These findings suggest that adjusting the dosage or frequency of MSC-EVs may enhance their efficacy for treating diabetes-related vascular conditions. Further research is warranted to evaluate AGEs' broader impact on various cell types and metabolic pathways for improved exosome-based therapies. Keywords: advanced glycation end products, angiogenesis, energy metabolism, extracellular vesicles, mesenchymal stem cells, oxidative stress Graphical abstract [45]graphic file with name ga1.jpg [46]Open in a new tab __________________________________________________________________ AGEs play a significant role in the development of complications associated with diabetes and cardiovascular diseases. These compounds are produced through nonenzymatic reactions involving sugars and proteins, lipids, or nucleic acids, and they accumulate in the body, initiating various pathophysiological responses that lead to tissue damage and functional impairment.[47]1, [48]2, [49]3 For example, upon binding to their receptor (RAGE), the cytoplasmic domain of RAGE engages with diaphanous-related formin-1 and formin homology 1 domain, activating Rho GTPases Rac-1 and Cdc42.[50]^4 This activation then leads to the initiation of the c-jun n-terminal kinase and p38 mitogen-activated protein kinase pathways,[51]^5 which in turn stimulate the proinflammatory transcription factors nuclear factor kappa-light-chain-enhancer of activated B cells[52]^6^,[53]^7 and activator protein-1.[54]^8 Nuclear factor kappa-light-chain-enhancer of activated B cells subsequently move into the nucleus, driving the expression of various proinflammatory genes, such as interleukin (IL)-6 and tumor necrosis factor-alpha, along with adhesion molecules such as vascular cell adhesion molecule-1.[55]^9 Furthermore, the AGE/RAGE interaction, through a protein kinase C–dependent mechanism, recruits adaptor proteins myeloid differentiation primary response 88 and toll-IL-1 receptor adaptor protein to form a signaling complex that transmits signals to downstream targets.[56]^10 Through this toll-IL-1 receptor adaptor protein/myeloid differentiation primary response 88 complex, RAGE ligands activate protein kinase B, which sustains nuclear factor kappa-light-chain-enhancer of activated B cell activation, thereby intensifying the inflammatory response.[57]^11 The binding of AGE to RAGE on the cell surface increases the expression of nicotinamide adenine dinucleotide phosphate (NADPH) oxidase subunits, which activates NADPH oxidase[58]^12 and fosters ROS generation.[59]^13 These ROS induce oxidation and nitrosylation of proteins, resulting in cellular damage and disruption of enzymatic functions.[60]^14 The initial ROS produced further stimulate mitochondrial ROS production, causing a cascade of oxidative reactions within the cell.[61]^2 The accumulation of ROS leads to oxidative damage to lipids, proteins, and DNA, resulting in changes to the structure and function of endothelial cells.[62]^15 Pathological alterations inflict harm upon vascular endothelial cells, promote the proliferation and migration of cells, cause intimal thickening, and accelerate the development of atherosclerosis.[63]^16 In addition, AGEs can cross-link extracellular matrix proteins such as collagen and elastin, resulting in matrix stiffening and vascular rigidity. It compromises vascular elasticity and function.[64]^17 In diabetic retinopathy, AGEs accumulate in retinal microvessels, damaging retinal vessels and cells, and causing retinopathy.[65]^18 Recent studies highlighted a bidirectional feedback mechanism between cardiac structural remodeling and mitochondrial bioenergetic function,[66]^19 and improving mitochondrial function can prevent diabetic nephropathy.[67]^20 Thus, mitochondrial metabolic function is crucial for the prevention and treatment of diabetes and cardiovascular diseases. MSCs and their secreted vesicles have demonstrated remarkable potential in vascular repair and functional regulation.[68]^21^,[69]^22 MSCs are multipotent stem cells capable of self-renewal and differentiation into various cell types. Meanwhile, their vesicles act as crucial mediators of intercellular communication, carrying numerous bioactive molecules including proteins, RNA, and miRNAs.[70]^23^,[71]^24 Studies have shown that EVs-MSC are rich in angiogenic factors such as vascular endothelial growth factor, fibroblast growth factor-2, and angiopoietin-1.Then they substantially boost endothelial cell proliferation and angiogenesis.[72]^25 In animal models, MSC-EVs have exhibited significant therapeutic effects. For example, they promote angiogenesis and improve mitochondrial function in models of diabetic foot ulcers and diabetic nephropathy.[73]^26^,[74]^27 Certain miRNAs, such as miR-126 and miR-210, are enriched in MSC vesicles,[75]^28 and these miRNAs can modulate inflammatory and oxidative stress responses, affecting mitochondrial function, reducing endothelial cell damage, and maintaining vascular function.[76]^29^,[77]^30 Therefore, the utilization of MSC vesicles in addressing diabetic complications and vascular diseases holds a significant promise. Nevertheless, when treating these conditions, MSCs are often exposed to an environment rich in AGEs.[78]^31 The impact of this environment remains largely unknown on the morphology, quantity, and miRNA production of EVs-MSC, as well as on mitochondrial function and energy metabolism in endothelial cells. This study, through high-throughput miRNA sequencing analysis, investigates the effects of AGEs on MSC vesicles and their influence on the metabolic activity and function of vascular endothelial cells. The insights gained will assist in assessing the metabolic alterations prompted by stem cell vesicles therapy in diabetes and cardiovascular diseases, providing theoretical guidance for the practical application of MSC EVs. Methods Cell Culture Human umbilical cord MSCs were obtained from XIAMEN Anti-hela Biological Technology Trade Co. Ltd (Xiamen, China). To assess the effects of AGEs on cell viability, MSCs were treated with 0 ng/ml, 50 ng/ml, 100 ng/ml, 200 ng/ml, and 400 ng/ml of AGEs (bs-1158P, Bioss) for 48 hours. AGEs were naturally purified proteins with a purity > 98%. The lyophilized protein was dissolved in 10mM phosphate-buffered saline (PBS) (pH 7.4) and diluted according to the desired concentration; avoid repeated freeze/thaw cycles were avoided. We inoculated 1×10^7 MSCs into a 3D FloTrixTM miniSPIN bioreactor (CytoNiche, miniSPIN S1-01) containing 200 ml culture medium, at 37 °C in a 5% CO₂ incubator. After 48 hours, they were replaced with serum-free human umbilical cord MSCs (IMMOCELL, IML-103-1) supplemented with final concentrations of 0 ng/ml and 100 ng/ml AGEs for 48 hours. Removed the bioreactor and allowed the microcarriers with cells to settle. Collected the supernatant to extract EVs. This study was divided into the following 3 groups: * • Mock: HUVECs were cultured in complete endothelial cell medium (IMC-309, IMMOCELL) for 48 hours. * • EVs-MSC: HUVECs were cocultured with 5 μg/96-well, 20 μg/24-well, 80 μg/6well EVs derived from MSCs treated with 0 ng/ml AGEs in complete medium for 48 hours. * • EVs-MSC/AGEs: HUVECs were cocultured with 5 μg/96-well, 20 μg/24-well, 80 μg/6well EVs derived from MSCs treated with 100 ng/ml AGEs in complete medium for 48 hours. EVs Extraction and Characterization EVs were extracted by ultracentrifugation (Hitachi, CP100MX). First, the samples were transferred into a new centrifuge tube and centrifuge at 2000 g for 30 minutes at 4 °C; then carefully transferred the supernatant to a new tube and centrifuged again at 10,000 g for 45 minutes at 4 °C to remove larger vesicles. The supernatant was filtered through a 0.45 μm filter (Millipore, R6BA09493), collecting and transferring the filtrate to a new tube, then was centrifuged at 100,000g for 70 minutes at 4 °C. Discarded the supernatant, resuspended the pellet in 10 ml of 1× PBS (Sangon Biotech, E607008), and repeated ultracentrifugation at 100,000 g for another 70 minutes at 4 °C. The supernatant was removed, resuspended in 100 μl PBS, 20 μl for transmission electron microscopy, 10 μl for nanoparticle tracking analysis, and the remaining exosomes were stored at −80 °C. Exosome characterization was performed using nanoparticle tracking analysis to determine particle size and concentration, transmission electron microscopy to observe exosome morphology and structure, and Western blotting to detect exosome marker proteins (such as differentiation [CD]9, CD63, and CD81) to confirm purity and characteristics. The concentration of EVs was determined using the BCA protein quantification assay kit (KGB2101, Keygen). 3-(4,5-Dimethylthiazol-2-yl)-2,5-Diphenyltetrazolium Bromide Proliferation Assay and pH Measurement After treating the cells according to the specified groups, 20 μl of 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide solution (40201ES72, Yeasen) was added to each well and incubated for 4 hours. The supernatant was then discarded, and 150 μl of dimethyl sulfoxide was added to dissolve the formazan crystals by shaking for 10 minutes. The optical density was measured at 490 nm (MOLECULAR), and the relative cell viability was calculated based on the results. For pH measurement, the supernatant was collected after treatment, and the pH was measured using a Leici pH meter (PHS-3C), with the values recorded. ROS and 5,5',6,6'-Tetrachloro-1,1',3,3'-Tetraethylbenzimidazolylcarbocyanine Iodide (JC-1) Flow Cytometry Detection After treating the cells according to the groups, ROS detection was performed using the ROS assay kit (50101ES01, Yeason). Cells were collected by centrifugation at 300 g for 5 minutes at 4 °C and resuspended in serum-free medium at a density of 1.0×10^6 to 2.0×10^7 cells/ml. The cells were incubated with the DCFH-DA probe at a final concentration of 10 μM in the dark for 20 minutes, mixing every 3 to 5 minutes. After washing the cells 1 to 2 times with serum-free medium, they were resuspended in 500 μl of serum-free medium and analyzed by flow cytometry. For JC-1 mitochondrial membrane potential (mΔψm) detection, cells were collected after treatment, and 100,000 to 600,000 cells were resuspended in 0.5 ml of cell culture medium. JC-1 staining working solution (0.5 ml) was added, and the cells were incubated at 37 °C for 20 minutes. Cells were then centrifuged at 600 g for 3 to 4 minutes at 4 °C, the supernatant was discarded, and cells were washed twice with 1 × JC-1 staining buffer, each time centrifuging at 600 g for 3 to 4 minutes at 4 °C. Finally, the cells were resuspended in an appropriate amount of 1× JC-1 staining buffer and analyzed by flow cytometry to calculate the relative MFI (Green/Red). Metabolic Function Detection ATP Detection After group treatments, the culture medium was removed, and 200 μl of lysis buffer (S0026, Beyotime) was added to each well to lyse the cells. The lysates were centrifuged at 12,000 g for 5 minutes at 4 °C, and the supernatant was used for ATP detection. The ATP detection reagent was dissolved on ice, and the ATP standard solution was diluted to concentrations of 0.01, 0.03, 0.1, 0.3, 1, 3, and 10 μM. After preparing the ATP detection working solution, 100 μl was added to each well and left to sit for 3 to 5 minutes to consume the background ATP. Then, 20 μl of the sample or standard was added, mixed well, and the relative light units or counts per minute was measured using a chemiluminescence detector. The ATP concentration in the samples was calculated based on the standard curve. G6PD Detection After group treatments, the cells were collected in centrifuge tubes, the supernatant was discarded, and 1 ml of extraction buffer (BC0260, Solarbio) was added. The cells were disrupted by ultrasonication (20% power, sonicate for 3 seconds, rest for 10 seconds, repeated 30 times). The lysates were centrifuged at 8000g for 10 minutes at 4 °C, and the supernatant was collected and kept on ice for analysis. The ultra violet spectrophotometer was preheated for more than 30 minutes, and the wavelength was set to 340 nm, with distilled water as the blank. Reagent 1 was preheated in a 37 °C water bath for 30 minutes. A working solution (950 μl) was mixed with 50 μl of the sample or distilled water, and the absorbance change at 340 nm was measured over 5 minutes. The absorbance at 0 seconds was recorded as A1 and at 300 seconds as A2. The change in absorbance (ΔA) was calculated as ΔA_sample = A2_sample − A1_sample; and ΔA_blank = A2_blank − A1_blank. The G6PDH activity units were calculated using the following equation: [MATH: G6PDH(U/104cel l)=[(ΔAsampleΔAblank)÷(ε×d)×109×Vreaction]÷(Vsample÷Vextraction×500)÷T=1.286×(ΔAsampleΔAblank) :MATH] Glucose Detection The glucose test strip (Sannuo) was removed and the electrode end was inserted into the glucometer (GA-3 model, Sannuo); 200 μl of the treated supernatant was placed in a glass dish; the suction port was brough to the top end of the test strip into contact with the supernatant. The sample was automatically drawn into the reaction chamber of the test strip; and the measurement value recorded. Lactic Acid Detection After treatment, cells were lysed using PBS as a homogenization medium and centrifuged to collect the supernatant for measurement. The enzyme working solution and color reagent were prepared according to the lactate (LD) assay kit (A019-2-1, Nanjing Jiancheng). For each tube, the system included 1 ml of enzyme working solution, 0.2 ml of color reagent, and 0.02 ml of distilled water per 3 mmol/l standard per sample. The system was mixed well and incubated at 37 °C for exactly 10 minutes in a water bath; 2 mL of stop solution was added, mixed well, and the absorbance (A) of each sample at a wavelength of 530 nm was measured. Then 200 μl of the reaction mixture was transferred to a 96-well plate for absorbance measurement. The calculation formula was as follows: [MATH: LD(mmol/gpro t)=Asa< mi>mpleAbla nkAstandardAbla nk×Cstandard÷Cpr :MATH] Seahorse Cell Mitochondrial Stress Test Seahorse XFe/XF analyzer was preheated for at least 5 hours. Cells were seeded into the Seahorse XF cell culture microplate and incubated in a 37 °C, CO2-free incubator. A sensor probe plate was hydrated with Seahorse XF calibrant overnight. The next day, the assay medium was prepared by supplementing Seahorse XF Dulbecco's Modified Eagle medium with additives (1 mmol/l sodium pyruvate, 2 mmol/l glutamine, and 10 mmol/l glucose). The assay medium warmed to 37 °C in a water bath. Oligomycin (blue cap), carbonyl cyanide-p-trifluoromethoxyphenylhydrazone (yellow cap), and rotenone/antimycin A (red cap) compounds were retrieved from the Seahorse XF Cell Mito Stress Test Kit, the compounds in the assay medium dissolved, 2 to 3 ml of working solution prepared, and added to the injection ports on the probe plate. According to the manual, the recommended starting volume of assay medium for a 96-well cell plate was 180 μl per well. The Seahorse XF cell culture microplate was removed from the incubator, confluency confirmed, the culture medium replaced with prewarmed assay medium, and the cell culture microplate incubated in a 37 °C CO2-free incubator for 45 minutes. Finally, the calibration plate and the probe plate with compounds were placed into the instrument's tray for calibration and analysis. Cellular Glycolytic Stress Test The Agilent Seahorse XFe/XF analyzer was preheated for at least 5 hours. Cells were seeded at a predetermined optimal density into the Seahorse XF cell culture microplate using the appropriate cell growth medium (containing the corresponding EVs). A sensor probe plate was hydrated with Seahorse XF calibrant overnight in a CO2-free incubator. The next day, the assay medium was prepared with 2 mmol/l glutamine as the starting condition, the pH adjusted to 7.4 using 0.1 N NaOH, and warmed to 37 °C for use. Oligomycin (light blue cap), glucose (blue cap), and 2-deoxyglucose (green cap) were retrieved from the Seahorse XF Glycolysis Stress Test Kit, dissolved in the assay medium, and added to the appropriate injection ports on the hydrated probe plate. They were then added as indicated in [79]Table 1. The Seahorse XF cell culture microplate was removed from the incubator, cell confluency confirmed, the culture medium was replaced with prewarmed assay medium and incubate the cell culture microplate was incubated in a 37 °C, CO2-free incubator for 45 minutes. Finally, the calibration plate and the probe plate with the compounds were placed into the instrument's tray for calibration and analysis. Table 1. Cellular glycolytic stress test volume instruction Agilent Seahorse XFe/XF96 Fixed volume __________________________________________________________________ Fixed concentration __________________________________________________________________ Initial well volume: 175 μl assay medium Initial well volume: 180 μl assay medium Injection Port A Glucose Final Well Concentration (mmol/l) Stock Solution Volume (μl) Assay Medium Volume (μl) 8X (Injection Port Volume (μl) Added to Injection Port Volume (μl) Final Well Concentration (mmol/l) Stock Solution Volume (μl) Assay Medium Volume (μl) 10X (Injection Port Volume [μl]) Added to Injection Port Volume (μl) 10 3000 750 80 25 10 3000 0 100 20 Injection Port B Oligomycin Final Well Concentration (mmol/l) Stock Solution Volume (μl) Assay Medium Volume (μl) 9X (Injection Port Volume [μl]) Added to Injection Port Volume (μl) Final Well Concentration (mmol/l) Stock Solution Volume (μl) Assay Medium Volume (μl) 10X (Injection Port Volume [μl]) Added to Injection Port Volume (μl) 1.0 270 2730 9 25 1.0 300 2700 10 22 Injection Port C 2-DG Final Well Concentration (mmol/l) Stock Solution Volume (μl) Assay Medium Volume (μl) 10X (Injection Port Volume [μl]) Added to Injection Port Volume (μl) Final Well Concentration (mmol/l) Stock Solution Volume (μl) Assay Medium Volume (μl) 10X (Injection Port Volume [μl]) Added to Injection Port Volume (μl) 50 3000 0 500 25 50 3000 0 500 25 [80]Open in a new tab 2-DG, 2-deoxyglucose. RT-Quantitative Polymerase Chain Reaction Total RNA was extracted from samples using the Trizol method, and its quality was assessed. RNA was then reverse-transcribed into cDNA using the HiScript II Q RT SuperMix (R222-01, Vazyme Biotech, Nanjing). The quantitative polymerase chain reaction system was prepared, including cDNA template, specific primers, and fluorescent dye (Q431-02, Vazyme Biotech, Nanjing); and amplification was performed on a quantitative polymerase chain reaction instrument. The relative expression levels of target genes were calculated using the ΔΔCt method by analyzing the Ct values. The reference gene and target gene primer information were shown in [81]Table 2. Table 2. Primer sequence The reference gene Upstream primer sequences (5 ′ −3 ′) Downstream primer sequences (5 ′ −3 ′) 18S CGACGACCCATTCGAACGTCT CTCTCCGGAATCGAA CCCTGA GLUT3 TGCCTTTGGCACTCTCAACCAG GCCATAGCTCTTCAGACCCAAG GLUT1 TTGCAGGCTTCTCCAACTGGAC CAGAACCAGGAGCACAGTGAAG GLUT4 CCATCCTGATGACTGTGGCTCT GCCACGATGAACCAAGGAATGG SCO2 GACCACTCCATTGCCATCTACC CTCAAGACAGGACACTGCGGAA [82]Open in a new tab GLUT, glucose transporter type; SCO2, synthesis of cytochrome c oxidase 2. Western Blot (WB) Proteins were extracted from cells using lysis buffer and centrifuged at 12,000 g for 30 minutes. The supernatant was collected, and protein concentration was determined using a BCA assay kit. Proteins were separated by SDS-PAGE electrophoresis and then transferred from the gel to a PVDF membrane. The membrane was blocked with 5% BSA at room temperature for 1 hour, incubated with specific primary antibodies overnight at 4 °C, and then with HRP-conjugated secondary antibodies. Finally, the protein bands were visualized using chemiluminescent reagents (e.g., ECL), and recorded and analyzed using an imaging system. The dilution ratios for the primary antibodies were: GLUT1 (21829-1-AP, Proteintech) 1:500, GLUT3 (ab314193, Abcam) 1:500, synthesis of cytochrome c oxidase 2 ([83]Ab169042, Abcam) 1:500, and GLUT4 (ab313775, Abcam) 1:500; Actin (SA00001-1, Proteintech) 1:1000. The secondary antibodies HRP-Anti-Rat IgG (SA00001-2, Proteintech) and HRP-Anti-Mouse IgG (SA00001-1, Proteintech) were diluted 1:10000. The original western blot is shown in [84]Supplementary Figure S1. Tube Formation Assay The day before the experiment, Matrigel was thawed overnight at 4 °C, and pipette tips, 96-well plates, and centrifuge tubes were precooled. Matrigel was diluted with Dulbecco's Modified Eagle medium at a 1:1 ratio, and 50 μl of Matrigel was added to each well of the precooled 96-well plate. The plate was incubated at 37 °C for 45 minutes to 1 hour to solidify the Matrigel. Cells were digested at 70% confluency, washed with PBS, digested with trypsin, resuspended in exosome-containing medium, and counted. The cell suspension was adjusted to 50 μl per well (30,000 cells/well), with triplicates for each group, and incubated at 37 °C or 5% CO2 for 3 hours. Finally, tube formation images were captured using 40× and 100× magnifications. miRNA Sequencing and Differential Analysis Total RNA of high purity was extracted from 2 types of EVs. The small RNA library was constructed using an end repair, adaptor ligation, reverse transcription, and polymerase chain reaction amplification kit. The constructed library was sequenced on a high-throughput sequencing platform (Illumina HiSeq). The sequencing data underwent quality control, adaptor removal, and low-quality read filtering. Clean reads were aligned to the reference genome or known small RNA databases using alignment tools. The expression levels of each miRNA were calculated based on the alignment results, and differential expression analysis tools (e.g., DESeq2 or EdgeR) were used to compare miRNA expression levels between different groups, identifying significantly differentially expressed miRNAs. Functional annotation and pathway analysis were performed on these differentially expressed miRNAs to predict their target genes and explore their biological functions. Statistical Analysis The statistical analysis was performed using IBM SPSS Statistics, version 26.0, available at [85]https://www.ibm.com/products/spss-statistics. The data visualization was carried out using GraphPad Prism, version 9.0, available at [86]https://www.Gr.with. Results are presented as mean ± SEM, with a minimum of 3 independent replicates per group. For comparisons between 2 groups, a 2-tailed t test was employed. For multiple group comparisons, 1-way analysis of variance was conducted, followed by Tukey’s post hoc test for pairwise comparisons. A P-value < 0.05 was considered statistically significant; and results were denoted as nonsignificant P > 0.05, ∗P < 0.05, ∗∗P < 0.01, ∗∗∗P < 0.001 and ∗∗∗P < 0.0001, where applicable. Results Characterization of AGE-treated EVs-MSC and the Cell Viability of MSC As the concentration of AGEs increased (0, 50, 100, 200, 400 ng/ml), the cell viability of MSCs significantly decreased. High concentrations of AGEs (200 and 400 ng/ml) significantly reduced cell viability (P < 0.01); whereas low concentrations of AGEs (50 and 100 ng/ml) had no significant effect on cell viability ([87]Figure 1a). We selected 100 ng/mL AGEs for subsequent experiments. The size distribution of EVs from both groups was primarily between 40 and 100 nm, with the particle size of EVs-MSC/AGEs being 47.87 nm, slightly smaller than that of EVs-MSC, which was 56.78 nm ([88]Figure 1b). The particle sizes of MSC-EVs range from 30 to 150 nm, classifying them as exosome. Both EVs-MSC and EVs-MSC/AGEs expressed exosome marker proteins (CD9, CD81, CD63), regardless of AGEs treatment, suitable for comparative experiments ([89]Figure 1c). The morphology of EVs was similar in both groups, with EVs-MSC/AGEs slightly oval and EVs-MSC more spherical ([90]Figure 1d). In summary, AGEs treatment reduced MSC cell viability but minimally affected EV particle size, distribution, morphology, and characteristic proteins. Figure 1. [91]Figure 1 [92]Open in a new tab Characterization of AGE-treated EVs-MSC and the cell viability of MSC. (a) Cell viability of MSCs treated with different concentrations of AGEs (0, 50, 100, 200, 400 ng/ml) for 48 hours. ns: P > 0.05, ∗∗: P < 0.01. Compared to 0 ng/ml, 200 and 400 ng/ml significantly reduced cell viability. (b) Nanoparticle tracking analysis showed the size distribution of extracellular vesicles. (c) Western blot analysis of exosome markers (CD9, CD81, CD63) in EVs-MSC and EVs-MSC/AGEs, with actin as the loading control. (d) Transmission electron microscopy (TEM) images showing the morphology and size of EVs-MSC and EVs-MSC/AGEs at different magnifications (1 μm, 100 nm, 200 nm, 500 nm). AGEs, advanced glycation end products; EVs, extracellular vesicles; MSC, mesenchymal stem cells; ns, nonsignificant. Effects of AGE-Treated EVs-MSC on HUVEC Proliferation and Metabolic State This study examined the impact of EVs-MSC and EVs-MSC/AGEs on HUVEC cell viability and metabolism. Compared to the Mock group, EVs-MSC significantly increased HUVEC cell viability (P < 0.01), ATP content (P < 0.05), and glucose consumption (P < 0.05), with no significant effects on pH, lactate production, or mΔψm in HUVECs (P > 0.05). However, EVs-MSC/AGEs treatment led to reduced cell viability (P < 0.001) ([93]Figure 2a), lower pH (P < 0.01) ([94]Figure 2b), decreased ATP content (P < 0.001) ([95]Figure 2c) and glucose consumption (P < 0.001) ([96]Figure 2d), increased lactate production (P < 0.05) ([97]Figure 2e), and higher mΔψm in HUVECs (P < 0.001) ([98]Figure 2f). These results suggested that AGE-treated EVs-MSCs impaired cell viability, acidified the cellular environment, reduced energy metabolism, and may promote glycolysis, potentially due to mitochondrial dysfunction or increased cellular stress. The AGEs environment diminished the positive effects of EVs-MSC on endothelial cells and altered energy metabolism pathways. Figure 2. [99]Figure 2 [100]Open in a new tab Effects of AGE-treated EVs-MSC on HUVEC proliferation and metabolic state. (a) Cell viability of HUVECs. Compared to the Mock group, EVs-MSC increased cell viability, EVs-MSC/AGE reduced this increase in viability. (b) pH levels in the culture medium of HUVECs. Compared to the Mock group, EVs-MSC had no difference in pH. Compared to EVs-MSC, EVs-MSC/AGE had a significant reduction in pH. (c) ATP content in HUVECs. Compared to the Mock group, EVs-MSC increased ATP content, EVs-MSC/AGE reduced this increase in ATP content. (d) Glucose consumption of HUVECs. Compared to the Mock group, EVs-MSC increased glucose consumption, EVs-MSC/AGE reduced this increase in glucose consumption. (e) Lactic acid levels in the culture medium of HUVECs. Compared to the Mock group, EVs-MSC had no difference in Lactic acid levels. Compared to EVs-MSC, EVs-MSC/AGE had a significant increase in Lactic acid levels. (f) Mitochondrial membrane potential (mΔψm) in HUVECs. Compared to the Mock group, EVs-MSC had no difference in mΔψm. Compared to EVs-MSC, EVs-MSC/AGE had a significant increase in mΔψm. (g) Flow cytometry analysis of JC-1 fluorescence in HUVECs treated with mock, EVs-MSC, or EVs-MSC/AGEs. ns: P > 0.05, ∗P < 0.05, ∗∗P < 0.01, ∗∗∗P < 0.001. AGEs, advanced glycation end products; ATP, adenosine triphosphate; EVs, extracellular vesicles; HUVEC, human umbilical vein endothelial cells; JC-1, 5,5',6,6'-Tetrachloro-1,1',3,3'-tetraethylbenzimidazolylcarbocyanine iodide; MSC, mesenchymal stem cells. Effects of AGE-Treated EVs-MSC on Glycolytic Metabolic Activity in HUVECs The glucose concentration and lactate accumulation levels suggested that AGE-treated EVs-MSC might promote the glycolysis process in HUVECs. We used the Seahorse XF Glycolysis Stress Test to observe the effects of EVs-MSC/AGEs on the glycolytic function of HUVECs in real-time ([101]Figure 3a). The real-time extracellular acidification rate curve showed that the glycolytic activity of EVs-MSC/AGEs was markedly elevated post–glucose addition, compared to HUVEC control group (P < 0.0001), whereas there was no significant difference compared to the EVs-MSC group (P = 0.5509) ([102]Figure 3b). After adding oligomycin (an ATP synthase inhibitor), the glycolytic capacity of EVs-MSC/AGEs remained significantly higher than the HUVEC control group (P < 0.0001), with a nonsignificant difference compared to the EVs-MSC control group (P = 0.2028) ([103]Figure 3c). Following the addition of 2-deoxyglucose, a glycolysis inhibitor, the glycolytic reserve of EVs-MSC/AGEs was significantly higher than that of the HUVEC control group (P = 0.0005), with no significant difference compared to the EVs-MSC control group (P = 0.2135) ([104]Figure 3d). These findings indicated that AGE-treated MSC-EVs exhibited enhanced glycolytic activity, glycolytic capacity, and glycolytic reserve, mainly affecting cellular energy metabolism through the glycolysis pathway. Figure 3. [105]Figure 3 [106]Open in a new tab Effects of AGE-Treated EVs-MSC on glycolytic metabolic activity in HUVECs. (a) Extracellular acidification rate (ECAR) profile of HUVECs. Measurements were taken following sequential injections of glucose, oligomycin, and 2-deoxyglucose (2-DG). (b) Glycolysis rate of HUVECs. Glycolysis rate was calculated from ECAR data following glucose injection. Compared to the HUVEC group, EVs-MSC had no difference in glycolysis rate (P = 0.5509). Compared to the HUVEC group, EVs-MSC/AGE had a significant increase in glycolysis rate (P < 0.0001). (c) Glycolytic capacity of HUVECs. Glycolytic capacity was calculated from ECAR data following oligomycin injection. Compared to the HUVEC group, EVs-MSC had no difference in glycolytic capacity (P = 0.2028). Compared to the HUVEC group, EVs-MSC/AGE had a significant increase in glycolytic capacity (P < 0.0001). (d) Glycolytic reserve of HUVECs. Glycolytic reserve was calculated from ECAR data by subtracting the glycolysis rate from the glycolytic capacity. Compared to the HUVEC group, EVs-MSC had no difference in glycolytic reserve (P = 0.2135). Compared to the HUVEC group, EVs-MSC/AGE had a significant increase in glycolytic reserve (P = 0.0005). 2-DG, 2-deoxyglucose; AGEs, advanced glycation end products; EVs, extracellular vesicles; HUVEC, human umbilical vein endothelial cells; MSC, mesenchymal stem cells. Effects of AGE-Treated EVs-MSC on Mitochondrial Function of HUVECs The JC-1 flow cytometry assay indicated an increase in mΔψm. To further investigate the impact of EVs-MSC/AGEs on the mitochondrial function of HUVECs, we performed Seahorse XF Mitochondrial Stress Test. HUVECs treated with EVs-MSC exhibited higher oxygen consumption rates, especially reaching peak respiratory capacity following carbonyl cyanide-p-trifluoromethoxyphenylhydrazone injection, indicating enhanced mitochondrial function. In contrast, HUVECs treated with EVs-MSC/AGEs showed lower oxygen consumption rates throughout process, suggesting impaired mitochondrial function ([107]Figure 4a). The proton leak (P < 0.0001), ATP production (P = 0.0002), maximal respiratory capacity (P < 0.0001), basal respiration (P < 0.0001), and spare respiratory capacity (P = 0.0379) were significantly higher in HUVECs treated with EVs-MSC than in the control group, whereas these parameters were significantly diminished in HUVECs treated with EVs-MSC/AGEs (P < 0.05) ([108]Figure 4b–f). Seahorse XF analysis results demonstrated that EVs-MSC significantly enhanced the mitochondrial function of HUVECs, including ATP production, maximal respiratory capacity, basal respiration, and spare respiratory capacity. Conversely, EVs from an AGEs environment notably impaired these metabolic functions. AGEs not only altered the metabolic regulatory capacity of EVs but also decreased the mitochondrial function of endothelial cells. Figure 4. [109]Figure 4 [110]Open in a new tab Effects of AGE-treated EVs-MSC on mitochondrial function of HUVECs. (a) Oxygen consumption rate (OCR) profile of HUVECs. Measurements were taken following sequential injections of oligomycin, FCCP, and rotenone/antimycin A (Rot/AA). (b) Proton leak of HUVECs. Compared to the HUVEC group, EVs-MSC had significant increase in proton leak (P < 0.0001). Compared to the EVs-MSC, EVs-MSC/AGE had a significant reduction in proton leak (P < 0.0001). (c) ATP production in HUVECs. Compared to the HUVEC group, EVs-MSC had significant increase in ATP production (P = 0.0002). Compared to the EVs-MSC, EVs-MSC/AGE had a significant reduction in ATP production (P < 0.0047). (d) Maximal respiration of HUVECs. Compared to the HUVEC group, EVs-MSC had significant increase in maximal respiration (P < 0.0001). Compared to the EVs-MSC, EVs-MSC/AGE had a significant reduction in maximal respiration (P < 0.0001). (e) Basal respiration of HUVECs. Compared to the HUVEC group, EVs-MSC had significant increase in basal respiration (P < 0.0001). Compared to the EVs-MSC, EVs-MSC/AGE had a significant reduction in basal respiration (P < 0.0001). (f) Spare respiratory capacity of HUVECs Compared to the HUVEC group, EVs-MSC had significant increase in spare respiratory capacity (P = 0.0379). Compared to the EVs-MSC, EVs-MSC/AGE had a significant reduction in spare respiratory capacity (P = 0.0141). AGEs, advanced glycation end products; ATP, adenosine triphosphate; EVs, extracellular vesicles; FCCP, carbonyl cyanide-p-trifluoromethoxyphenylhydrazone; HUVEC, human umbilical vein endothelial cells. Effects of AGE-Treated EVs-MSC on Oxidative Stress and Antioxidant Capacity in HUVECs An increase in mΔψm may also be attributed to excessive cellular stress. We examined how EVs-MSC/AGEs impact oxidative stress and antioxidant capacity in HUVECs by measuring ROS levels and G6PD activity. The mock group showed baseline levels of ROS and G6PD activity. EVs-MSC/AGEs significantly increased ROS levels (P < 0.0001) and G6PD activity (P < 0.01) in HUVECs, whereas EVs-MSC had no significant impact on ROS levels and G6PD activity ([111]Figure 5a–d). G6PD is a key enzyme in the pentose phosphate pathway that generates NADPH. NADPH helps maintain glutathione in its reduced state, protecting cells from oxidative stress. Our results indicated that EVs-MSC/AGEs induced intracellular oxidative stress, leading to an upregulation of G6PD activity as a cellular response to elevated ROS levels, thereby enhancing antioxidant capacity. This suggested that AGE-treated EVs-MSC can induce metabolic reprograming to maintain redox balance in HUVECs. Figure 5. [112]Figure 5 [113]Open in a new tab Effects of AGE-treated EVs-MSC on oxidative stress and antioxidant capacity in HUVECs. (a) Flow cytometry histograms showing ROS levels in HUVECs treated with mock, EVs-MSC, and EVs-MSC/AGEs. ROS levels were measured using the DCFDA assay, where DCF fluorescence indicates ROS production. (b) Overlay of flow cytometry. (c) Quantification of relative mean fluorescence intensity (MFI) of DCF in HUVECs. (d) Glucose consumption of HUVECs. ns: P > 0.05, ∗∗P < 0.01, ∗∗∗P < 0.0001. AGEs, advanced glycation end products; DCFDA, 2',7' -dichlorofluorescin diacetate; EVs, extracellular vesicles; HUVEC, human umbilical vein endothelial cells; MSC, mesenchymal stem cells; ns, nonsignificant. Effects of AGE-Treated EVs-MSC on Angiogenic Capacity of HUVECs Angiogenic capacity is crucial for addressing diabetic complications, wound healing, and ischemic conditions. At 40× and 100× magnification, HUVECs created sparse tubular structures with limited connections, resulting in a nondense network. EVs-MSC treatment significantly enhanced the tubulogenic capacity of HUVECs, creating more, continuous, and denser tubular structures, indicating potent angiogenesis. Conversely, EVs-MSC/AGEs treatment reduced this capacity, leading to fewer and incomplete tubular structures with a sparser network than in control and EVs-MSC alone ([114]Figure 6). These results suggested that the AGEs environment negatively impacted exosome function, reducing their angiogenic capacity. Figure 6. [115]Figure 6 [116]Open in a new tab Effects of AGE-treated EVs-MSC on angiogenic capacity of HUVECs. AGEs, advanced glycation end products; EVs, extracellular vesicles; HUVEC, human umbilical vein endothelial cells; MSC, mesenchymal stem cells. Effects of AGE-Treated EVs-MSC on the Expression of Glucose Transporters and Metabolic Enzymes in HUVECs RT-quantitative polymerase chain reaction and Western blot analyses revealed that EVs-MSC significantly increased the expression of glucose transporters GLUT1, GLUT3, GLUT4 ([117]Figure 7a–c), and metabolic enzyme synthesis of cytochrome c oxidase 2 ([118]Figure 7d) at both mRNA and protein levels ([119]Figure 7e and f). Conversely, EVs-MSC/AGEs significantly reduced the expression of these genes, though their levels remained higher than those in the mock group ([120]Figure 7). These results indicated that EVs-MSC contributed to enhancing glucose uptake and energy metabolism in cells, supporting their normal function and health. However, EVs-MSC/AGEs significantly decreased the expression of these key proteins, weakening the beneficial effects of EVs-MSC. This reduction may be due to impaired cellular energy metabolism and mitochondrial function, or decreased insulin sensitivity and glucose metabolism disorders. Figure 7. [121]Figure 7 [122]Open in a new tab Effects of AGE-treated EVs-MSC on the expression of glucose transporters and metabolic enzymes in HUVECs. (a–d) mRNA levels of (a) GLUT1, (b) GLUT3, (c) GLUT4, and (d) SCO2 in HUVECs. (e) Western blot images showing protein levels of GLUT1, GLUT3, GLUT4, and SCO2 in HUVECs. (f) Quantification of protein levels of GLUT1, GLUT3, GLUT4, and SCO2 in HUVECs, normalized to actin. ∗P < 0.05, ∗∗P < 0.01, ∗∗∗P < 0.001, ∗∗∗∗P < 0.0001. AGEs, advanced glycation end products; EVs, extracellular vesicles; GLUT; glucose transporter type; HUVEC, human umbilical vein endothelial cells; MSC, mesenchymal stem cells; SCO2, synthesis of cytochrome c oxidase 2. Differential miRNA Expression in AGE-Treated EVs-MSC and Potential Functional Prediction To elucidate the mechanisms of AGEs environment affecting EVs-MSC, we performed miRNA sequencing on EVs from both groups. Annotation using Rfam ([123]https://rfam.org/) indicated that both EV types mainly contained approximately 60% ribosomal RNA, 2% small nucleolar RNA, with small nuclear RNA in EVs-MSC/AGEs being twice that in EVs-MSC, and the proportion of transfer RNA being nearly 1.5 times higher ([124]Figure 8a). Other non-miRNA entities were excluded from the analysis. Figure 8. [125]Figure 8 [126]Open in a new tab Differential microRNA expression and functional enrichment analysis of miRNAs in EVs-MSC and EVs-MSC/AGEs. (a) Pie charts of distribution in different types of RNAs (rRNA, snRNA, snoRNA, and tRNA) in EVs-MSC and EVs-MSC/AGEs. (b) Venn diagram of differentially expressed miRNAs between EVs-MSC and EVs-MSC/AGEs. (c) Volcano plot of differentially expressed miRNAs between EVs-MSC and EVs-MSC/AGEs, highlighting significantly upregulated (red) and downregulated (blue) miRNAs. (d) Correlation scatter plot comparing the expression levels of miRNAs in EVs-MSC and EVs-MSC/AGEs. (e) Bar graph of the number of significantly upregulated and downregulated miRNAs. (f) Heatmap of differentially expressed miRNAs between EVs-MSC and EVs-MSC/AGEs. (g) Gene ontology (GO) enrichment analysis of target genes of differentially expressed miRNAs, categorized into biological processes, cellular component, and molecular function. (h) Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis of target genes of differentially expressed miRNAs, highlighting the most significantly enriched pathways. AGEs, advanced glycation end products; EVs, extracellular vesicles; miRNA; microRNA; MSC, mesenchymal stem cells; snRNA, small nuclear RNA; snoRNA, small nucleolar RNA; tRNA, transfer RNA. Following differential analysis of miRNAs, we conducted clustering analysis and visualization ([127]Figure 8f). Among the top 10 differentially expressed miRNAs, hsa-miR-223-3p, hsa-miR-126-3p_R-1, pal-mir-10001-p5_1ss1AT, PC-3p-6123_268, mmu-miR-2137_L-2R-1_1ss16AG, mmu-miR-2137_L-2_1ss16AG, and hsa-miR-26a-5p were significantly downregulated; whereas hsa-miR-574-5p, hsa-miR-199a-3p_R-1, and hsa-miR-125b-5p were significantly upregulated ([128]Figure 8c and d). The top 10 differential miRNA are shown in [129]Table 3. Annotation revealed 473 common genes between DSS-Exo and Exo ([130]Figure 8b), with the number of differential genes at different P-values shown in [131]Figure 8e. Kyoto Encyclopedia of Genes and Genomes pathway enrichment analysis revealed significant enrichment in metabolic and oxidative phosphorylation pathways ([132]Figure 8h), and in protein binding-related biological functions ([133]Figure 8g). Table 3. The top 10 differential miRNA Index 1 2 3 4 5 6 7 8 9 10 miR_name hsa-miR-223-3p hsa-miR-126-3p_R-1 pal-mir-10001-p5_1ss1AT PC-3p-6123_268 mmu-miR-2137_L-2R-1_1ss16AG mmu-miR-2137_L-2_1ss16AG hsa-miR-574-5p hsa-miR-26a-5p hsa-miR-199a-3p_R-1 hsa-miR-125b-5p miR_seq TGTCAGTTTGTCAAATACCCCA TCGTACCGTGAGTAATAATGC TGGGCGGGCGGGGCCGGGGG AGTTCTTGTGAGCTTTCTCG CGGCGGGAGCCCCGGGGA CGGCGGGAGCCCCGGGGAG TGAGTGTGTGTGTGTGAGTGTGT TTCAAGTAATCCAGGATAGGCT ACAGTAGTCTGCACATTGGTT TCCCTGAGACCCTAACTTGTGA up/down down down down down down down up down up up fold_change (EVs_AGES (mean)/EVs (mean) 0.32 0.38 0.16 0.07 0.25 0.24 4.12 0.39 1.92 3.38 Log 2 (fold_change) −1.66 −1.39 −2.63 −3.79 −1.99 −2.04 2.04 −1.34 0.94 1.76 P value (t_test) 3.05E-06 3.34E-06 7.75E-06 8.53E-06 9.82E-06 1.61E-05 2.08E-05 2.22E-05 2.25E-05 2.65E-05 EVs (mean) 15,299 12,028 6299 224 12,338 38,966 1255 59,142 13,027 2716 EVs_AGES (mean) 4829 4602 1020 16 3109 9445 5172 23,349 24,959 9176 EVs_AGES/EVs_AGES_1 (norm) 5050 4391 1061 11 2985 8866 5334 24,804 24,987 9425 EVs_AGES/EVs_AGES_2 (norm) 4944 4556 859 16 2779 8420 4963 23,769 24,303 8988 EVs_AGES/EVs_AGES_3 (norm) 4494 4860 1139 22 3561 11,049 5218 21,474 25,586 9115 EVs/EVs_1 (norm) 15,354 12,221 6528 221 12,148 40,445 1109 61,319 13,444 2583 EVs/EVs_2 (norm) 15,614 12,009 6199 233 12,062 37,446 1281 58,642 12,459 2789 EVs/EVs_3 (norm) 14,928 11,853 6169 218 12,803 39,006 1375 57,464 13,178 2775 EVs_AGES/EVs_AGES_1 (raw) 5426 4718 1140 12 3208 9527 5732 26,653 26,849 10,128 EVs_AGES/EVs_AGES_2 (raw) 7158 6596 1243 23 4024 12,192 7185 34,414 35,187 13,014 EVs_AGES/EVs_AGES_3 (raw) 5775 6245 1464 29 4577 14,199 6705 27,596 32,880 11,714 EVs/EVs_1 (raw) 13,526 10,766 5751 195 10,702 35,629 977 54,018 11,843 2276 EVs/EVs_2 (raw) 9083 6986 3606 136 7017 21,784 745 34,114 7248 1623 EVs/EVs_3 (raw) 9252 7346 3823 135 7935 24,174 852 35,614 8167 1720 Expression level high high high middle high high high high high high [134]Open in a new tab AGEs, advanced glycation end products; EVs, extracellular vesicles; miRNA; microRNA. We further analyzed the metabolic pathways,[135]^32 focusing on glycolysis, the tricarboxylic acid (TCA) cycle, amino acid metabolism, lipid metabolism, and nucleotide metabolism. Key steps and metabolites of glycolysis and the TCA cycle, like glucose, pyruvate, and acetyl-CoA, exhibited significant changes ([136]Supplementary Figure S2). Shown in [137]Supplementary Figure S2, genes marked in green are enriched by differential miRNAs, playing key roles in oxidative phosphorylation. Complex I (NADH dehydrogenase) include genes ND1 to ND6, which initiate NADH oxidation and electron transfer, creating a proton gradient. Complex II (succinate dehydrogenase) with genes SDHA to SDHD l links the TCA cycle to the electron transport chain through succinate oxidation. Complex III (cytochrome bc1 complex) genes Core 1, Core 2, cytochrome b, and iron-sulfur protein transfer electrons to cytochrome C and pumping protons. Complex IV (cytochrome c oxidase) genes COX I to IV transfer electrons to oxygen, forming water; and in addition, pump protons. Complex V (ATP synthase) subunits use the proton gradient for ATP production. These genes are critical in the processes of electron transport and ATP synthesis, illustrating the crucial role of oxidative phosphorylation in cellular energy metabolism. In summary, the AGEs environment influenced the expression of miRNAs in MSC-derived EVs, regulating the expression of these genes and playing important roles in cellular energy metabolism and related diseases. Discussion This study has shown that MSC-derived EVs obtained from an AGE environment reduced the positive impact of MSC-derived exosomes on endothelial cells by altering energy metabolic pathways and inducing metabolic reprograming to maintain redox balance in HUVEC cells. Therefore, this study highlighted the potentially overlooked metabolic and functional deficiencies of these MSCs in clinical applications, which were like metabolic disorders caused by AGEs in diabetic complications. AGEs are a significant etiological factor to contribute to diabetes and vascular diseases. MSCs and MSC-EVs represent promising therapeutic approaches for treating diabetic complications, including diabetic foot and chronic kidney disease.[138]33, [139]34, [140]35, [141]36 However, many patients with chronic hyperglycemia accumulate substantial amounts of endogenous and exogenous AGEs in their blood and body tissues, which affects them during MSC and MSC-EVs treatment.[142]^37^,[143]^38 AGEs impact not only MSCs but also their EVs. Increasing evidence suggests that MSCs promote tissue repair through paracrine immunomodulatory and regenerative factors, with the majority of these factors being transported by EVs released by MSCs.[144]^28 Previously, we compared the effects of various inflammatory microenvironments, including vascular cell adhesion molecule-1, tumor necrosis factor-alpha, and IL-6, on the miRNA differences in MSC-derived EVs. We found that in tumor necrosis factor-alpha and IL-6 environments, miRNAs related to angiogenesis were more downregulated; and the angiogenic ability of MSCs-exo decreased in the IL-6 environment,[145]^39 indicating that the microenvironment significantly affects MSCs. In an inflammatory environment, MSCs induce glycolytic reprograming of macrophages,[146]40, [147]41, [148]42 and MSCs-exo independently exert antiinflammatory effects, which has been confirmed in inflammatory arthritis.[149]^43 MSC-EVs exhibit functions similar to those of their parent cells, making it essential to understand how disease conditions (such as AGEs) influence MSC-secreted EVs, because this may affect their therapeutic effectiveness. We hypothesize that an AGEs environment may also impair the function and metabolic activities of MSC-EVs. This study found that EVs-MSC/AGEs significantly impaired HUVEC viability, ATP production, and glucose consumption while increasing lactate production and mitochondrial membrane potential. These findings indicated a shift toward glycolysis and suggested mitochondrial dysfunction. The upregulation of glycolysis may serve as a compensatory mechanism for impaired oxidative phosphorylation. Seahorse XF analysis showed that HUVECs treated with EVs-MSC/AGEs exhibited lower oxygen consumption rates across various parameters, indicating reduced mitochondrial function. Elevated ROS levels and G6PD activity in these cells suggested that EVs-MSC/AGEs induced oxidative stress, leading to increased antioxidant responses. This underscored a potential link between AGEs-induced mitochondrial dysfunction and oxidative stress. Tube formation assays showed that EVs-MSC/AGEs significantly inhibited this effect, whereas EVs-MSC enhanced angiogenesis in HUVECs. The decreased tube formation suggested that AGEs modified the proangiogenic properties of MSC-derived exosomes, which could influence tissue repair and regeneration. EVs-MSC increased the expression of glucose transporters (GLUT1, GLUT3, and GLUT4) and the metabolic enzyme synthesis of cytochrome c oxidase 2, enhancing glucose uptake and energy metabolism in HUVECs. In contrast, EVs-MSC/AGEs downregulated these proteins, impairing glucose metabolism. Overall, the AGEs microenvironment diminished the beneficial effects of EVs-MSC on endothelial cells in terms of energy and glucose metabolism. We conducted high-throughput miRNAs sequencing on both types of exosomes and found significant differences in miRNA expression between EVs-MSC and EVs-MSC/AGEs. These miRNAs were related to glycolysis, TCA metabolism, and oxidative phosphorylation pathways, suggesting that AGE-treated exosomes may influence cells through miRNA-driven gene regulation. Enrichment analysis further supported our hypothesis, aligning with the cellular experiments. Some reports indicated a vicious cycle between AGEs and oxidative stress: AGEs increased oxidative stress, and excessive oxidative stress, in turn, accelerates the production of AGEs, further exacerbating diabetic cardiovascular complications. AGEs activated signaling cascades that increased ROS production, increasing oxidative stress by upregulating inducible nitric oxide synthase and NADPH oxidase, which depleted antioxidants like glutathione and lower cellular antioxidant capacity.[150]44, [151]45, [152]46, [153]47, [154]48, [155]49, [156]50 Previous studies showed that in patients with diabetes, the AGE-RAGE axis promoted endothelial dysfunction and the development of chronic kidney disease through multiple mechanisms. When AGEs bound to RAGE, ROS were produced in endothelial cells, which activated the p38 and extracellular signal-regulated kinase 1/2 signaling pathways, inhibited the expression of endothelial nitric oxide synthase, and reduced nitric oxide production.[157]^51 The generation of ROS also induced the secretion of atherogenic cytokines, such as monocyte chemoattractant protein-1 and matrix metallopeptidase 9.[158]^52^,[159]^53 Furthermore, AGEs suppressed of endothelial nitric oxide synthase production, increased microvascular permeability, and promoted the migration of macrophages and T cells to the vessel wall, thereby accelerating inflammation and leading to the formation and instability of atherosclerotic plaques. In diabetic nephropathy, the AGE/RAGE axis activated the Rho and β-catenin pathways, leading to the degradation of vascular endothelial cadherin at endothelial cell junctions, which exacerbated interendothelial gaps, increased endothelial permeability, and resulted in microalbuminuria.[160]^54^,[161]^55 In addition, AGEs increased the concentration of asymmetric dimethylarginine by activating protein arginine methyltransferase, further inhibiting nitric oxide production and of endothelial nitric oxide synthase activity, thereby causing vascular endothelial dysfunction.[162]^56^,[163]^57 AGEs also promoted mesenchymal transition and fibrosis by downregulating sirtuin 1 and activating Smad, which exacerbated renal and vascular damage in diabetic patients.[164]58, [165]59, [166]60 These studies all indicated that AGEs had a negative impact on endothelial cell function, including inhibited proliferation, increased inflammatory responses, and oxidative stress, primarily involving NADPH oxidase and mitochondria in oxidative stress. However, there may have been crosstalk between NADPH oxidase, mitochondria, and glycolysis.[167]^47 Therefore, our findings further illustrated the real-time impact of AGEs on mitochondrial energy metabolism and redox state in stem cell–treated vascular endothelial cells. There was a complex relationship between glycolysis and oxidative stress. Under high glucose conditions, glycolysis accelerated, generating pyruvate that entered mitochondria and produced ATP through the tricarboxylic acid cycle, but also increased mitochondrial load and ROS production. To alleviate the mitochondrial burden, cells may have upregulated glycolysis through metabolic reprograming, thereby reducing ROS production. In addition, glycolytic byproducts, such as lactate, suppressed NADPH oxidase activity and reduced oxidative stress. In this process, oxidative stress decreased the expression of glucose transporters GLUT1 and GLUT4, limiting glucose uptake and indirectly slowing glycolysis, which helped reduce mitochondrial load and ROS generation. Glycolysis and oxidative stress influenced each other: under high glucose or oxidative stress conditions, cells accelerated glycolysis to reduce mitochondrial ROS production, whereas oxidative stress also affected glycolysis by modifying glycolytic enzymes or altering the activity of glucose transporters. The results of this study were consistent with existing literature emphasizing the adverse effects of AGEs on cellular function and metabolism. The novelty of this research lies in the detailed analysis of how AGE-treated exosomes mediate these effects in endothelial cells. Other studies have shown that miRNAs could transfer to mitochondria, regulating mitochondrial function and metabolic activity. For example, miR-149 increases mitochondrial function and biogenesis by activating sirtuin 1 and peroxisome proliferator-activated receptor gamma coactivator 1. Other miRNAs, such as miR-326,[168]^61 miR-25,[169]^62 miR-338,[170]^63 and miR-181c,[171]^64 directly or indirectly targeted key enzymes in mitochondrial metabolic pathways, regulating oxidative phosphorylation and the TCA cycle. MSC-EVs protected cardiac stem cells from oxidative damage and promote cardiomyocyte regeneration by transferring miRNAs such as miR-214[172]^65 and miR-21.[173]^66 We proposed that the AGEs environment affected a complex network of miRNA regulation because a single miRNA could not fully explain its impact on endothelial cells. However, these differentially expressed miRNAs point to specific metabolic signaling pathways, helping us understand how AGEs influence cell communication and function through exosomes. Although some studies suggested that MSC-derived exosomes have potential therapeutic effects, our findings underscored the need to consider the impact of AGEs and other environmental factors on exosome function. These results highlighted the potential necessity of increasing stem cell therapy doses or targeting exosome pathways to mitigate the adverse effects of AGEs in therapeutic strategies. The importance of developing exotic-based therapies that can withstand or counteract these negative influences is particularly crucial in managing diabetes-related complications. This study has certain limitations, because it was conducted in vitro, and the results may not fully replicate in vivo conditions. Future research should validate these findings in animal models and clinical trials. This study focused on specific aspects of HUVECs cell function and metabolic pathways. Comprehensive omics analyses, including proteomics and metabolomics, on various cell types are needed in the future to gain a broader understanding of the impact of AGEs. Conclusion In conclusion, the AGEs environment diminishes the positive effects of MSC-derived exosomes on endothelial cells by disrupting energy metabolic pathways and triggering metabolic reprograming to maintain the redox balance in HUVEC cells. This suggested that adjustments in the dosage or frequency of stem cell exosome therapy may be necessary when treating diabetes and related vascular lesions in clinical applications. in addition, further research should explore the effects of AGEs on other cell types and metabolic processes to optimize the application of stem cell exosomes in various metabolic diseases. Disclosure All the authors declared no competing interests. Acknowledgments