MnO2-armored human embryonic stem cell exosomes delay intervertebral disc degeneration via synergistic antioxidant defense and regenerative signal delivery Authors: Honglei Li, Hanlin Mao, Runze Mao, Qingsheng Ma, Qi Tan, Luoyuan Liu, Hui Ma, Zhi Ruan Published: 2026 ABSTRACT AbstractThe oxidative and inflammatory microenvironment impairs the regeneration and homeostasis of nucleus pulposus cells. This microenvironment is characterized by high levels of reactive oxygen species (ROS), mitochondrial dysfunction, and NLRP3 inflammasome-induced pyroptosis. These changes accelerate cellular senescence and disrupt the metabolic homeostasis of the extracellular matrix (ECM). Exosomes derived from human embryonic stem cells (hESCs) hold great potential for regeneration and inflammation control but lack inherent antioxidant protection. In the highly oxidative microenvironment of degenerated intervertebral discs, these exosomes degrade rapidly, which significantly limits their therapeutic efficacy. To address this issue, we developed a manganese dioxide nanoplate-armored exosome system (hESC-Exos@MnO2).The system exerts its therapeutic effects through a synergistic, multilevel mechanism. First, the MnO2 nanoarmor shields the exosomes; second, MnO2 is degraded in the acidic, H2O2-rich degenerative microenvironment to scavenge reactive oxygen species (ROS); finally, the released Mn2+ may exert a synergistic effect with exosomes on ECM remodeling and microenvironment repair. In the restored microenvironment, exosomes efficiently deliver regenerative signaling molecules to regulate key signaling pathways such as p53 and can partially alleviate oxidative stress-induced cell-cycle arrest, inhibit cell death and senescence, and promote ECM synthetic metabolism. Our study revealed that hESC-Exos@MnO2 attenuates oxidative injury, partially preserved mitochondrial membrane potential, and ameliorates ECM homeostasis. In a rat model of intervertebral disc degeneration, this treatment effectively maintenance intervertebral disc height. This work introduces a new “catalytic scavenging–bioremediation" strategy, providing a modular nanoplatform with high applicability for intervertebral disc regeneration and potential utility in treating other degenerative diseases.Keywords: Intervertebral disc degeneration, hESC-derived exosomes, ROS scavenging, Pyroptosis, Cell senescence, Extracellular matrix regeneration Introduction Intervertebral disc degeneration (IVDD) is the primary cause of chronic low back pain, which is the most disabling condition worldwide and imposes a heavy burden on global health and economies [1,2]. The progression of IVDD is driven by multiple core pathological events, including excessive oxidative stress, inflammatory responses, nucleus pulposus cell (NPC) senescence, and extracellular matrix (ECM) metabolic imbalance [3,4]. Extremely high concentrations of reactive oxygen species (ROS) disrupt mitochondrial homeostasis and activate the NLRP3 inflammasome, thereby inducing the pyroptosis of NPCs. In the final stage of this pathological process, ECM degradation and accelerated cellular senescence lead to intervertebral disc collapse [[5], [6], [7]]. For these reasons, a successful therapeutic strategy should simultaneously achieve several goals, such as scavenging reactive oxygen species (ROS), inhibiting cell death, delaying cell aging and restoring the balance of the ECM. Multitarget strategies can either delay or even reverse IVDD. Although researchers have tried to use antioxidants and stem cell therapy, these approaches still cannot achieve good results because of the harsh internal environment of the intervertebral disc. Exosomes, as a cell-free replacement therapy for traditional cell therapy, have emerged in recent years, and their low immunogenicity, high yield and strong paracrine signal transduction have attracted attention [8]. Exosomes can regulate inflammatory reactions, immune responses, cell survival and other biological activities by delivering functional substances; thus, they are recognized as a new generation of tools in the field of regenerative medicine [[9], [10], [11]]. Among all exosomes, human embryonic stem cell-derived exosomes (hESC-Exos) are the most prominent. Compared with adult stem cell-derived exosomes, hESC-Exos contain more growth factors, their genetic characteristics are more stable, and their regeneration ability is stronger [12]. These exosomes serve as delivery vehicles for repair signals, transporting a diverse array of miRNAs, proteins, and lipids that govern cell growth, division, and self-protection. They can activate aging cells and push them to re-enter the cell cycle and restore their functions [13,14]. There is evidence that the bioactive components in hESC-Exos not only have antiaging effects but also help nucleus pulposus cells regenerate and reduce local inflammation. In addition, they promote the production of ECM [[15], [16], [17]]. Despite these benefits, clinical application is still very difficult. In the harsh environment of degenerated intervertebral discs, exosomes decompose too quickly and cannot remain in place for a long time. This lack of persistence hinders their effectiveness and weakens the success of treatment [18]. To overcome these restrictions, it is necessary to identify a microenvironment-responsive carrier that can not only protect exosomes but also respond to the diseased microenvironment. In recent years, manganese dioxide MnO2 nanosheets have attracted attention because of their ability to scavenge ROS. Compared with traditional antioxidant materials, MnO2 nanosheets have different advantages as nanocatalysts. When used as nanoenzymes, MnO2 nanosheets can decompose H2O2 into O2 and H2O to directly remove excess reactive oxygen species [19,20]. The oxygen produced during this process reduces the local hypoxia of degenerated intervertebral discs, which limits the further production of ROS to some extent [21,22]. MnO2 can balance acidic conditions and improve the environment, which is harmful to cell survival [23,24]. In acidic and ROS-rich pathological areas, MnO2 is continuously degraded into biocompatible manganese ions. This degradation makes the material safe and biodegradable. The released manganese may have a role in promoting extracellular matrix repair and chondrocyte differentiation [[25], [26], [27]]. The structure of MnO2 is sensitive to pathological triggers. Decomposition, protection and release of loaded therapeutic molecules under inflammatory or oxidative stress. MnO2 nanosheets have good biocompatibility, regulate cell metabolism and support cell function recovery, which make them ideal carriers for the complex environment of degenerative intervertebral discs [28,29]. With these advantages, MnO2 nanosheets can protect hESC-Exos from oxidative damage and improve the microenvironment of intervertebral discs. MnO2 degradation occurs especially in acidic and highly reactive oxygen regions, which can ensure the accurate and sustained release of exosomes. The oxygen produced during this process also increases the delivery efficiency of hESC-Exos proteins, thus leading to better therapeutic effects [30]. Based on the above considerations, we designed a manganese dioxide nanoparticle exosome system (hESC-Exos@MnO2) to remove excessive ROS from degenerated intervertebral discs, inhibit apoptosis and restore the ECM steady state. MnO2 nanosheets have two functions: first, they can be used as a protective layer to increase the stability and retention of exosomes at the lesion site; second, ROS are actively removed by their nanoenzyme activity to alleviate oxidative stress. In the degenerative microenvironment, MnO2 is activated by pathological signals and degraded, thus protecting hESC-Exos and promoting regeneration and anti-inflammatory activity (Fig. 1). The synergistic effect of MnO2 and hESC-Exos can disrupt the vicious cycle of oxidative damage, inflammation and ECM degradation and provide a combined treatment scheme for intervertebral disc degeneration. Schematic overview of the preparation and proposed therapeutic mechanism of hESC-Exos@MnO2. Schematic overview of hESC-Exos@MnO2, in which the MnO2 armor scavenges ROS and safeguards the exosomes, thereby coordinating multi-pathway modulation to suppress nucleus pulposus cell pyroptosis, enhance cellular regeneration, and maintain extracellular matrix homeostasis, ultimately driving intervertebral disc regeneration. In summary, the hESC-Exos@MnO2 treatment approach combines the environmental response ability of MnO2 nanozymes and the strong regeneration ability of hESC-Exos. This method creates a new nanoscale treatment platform that combines active protection and cooperative repair. We tested this approach in vitro and in vivo. MnO2 nanomaterials can remove reactive oxygen species while maintaining the structural integrity of exosomes so that functional miRNAs, proteins and other bioactive components carried by hESC-Exos can be delivered. This system can mitigate many pathological processes associated with intervertebral disc degeneration and can simultaneously act on oxidative stress, pyroptosis, cell aging and ECM metabolic disorders. If we intervene in these key processes, we can treat intervertebral disc degeneration, which means that the platform has great potential for future clinical translation. Results Design, construction, and characterization of hESC-Exos@MnO2 nanocomposites An environment of a degenerated intervertebral disc is usually characterized by high ROS levels, acidity and a lack of nutrition. Such harsh conditions make it very difficult for traditional biological treatments to function [31]. To solve this problem, we created a hESC-Exos@MnO2 nanocomposite according to an “active defense and collaborative repair” strategy. In this system, MnO2 nanosheets serve as a “nanoarmor”. This MnO2 nanoarmor eliminates ROS via catalysis, thus easing oxidative stress, and when it is destroyed, the Mn2+ may exert a promoting effect on the repair of the ECM [[32], [33], [34]]. Moreover, the bioactive molecules in hESC-Exos can promote the regeneration of cells and the repair of tissues. We isolated hESC-Exos from the culture medium of human embryonic stem cells. Their integrity was verified by transmission electron microscopy (TEM) and Western blotting (Fig. 2A and F). Western blot analysis confirmed the expression of the exosomal markers CD9, CD63, and TSG101, whereas the negative marker Calnexin was undetectable. These results indicate that the isolated hESC-Exos were of high purity (Fig. 2F). Nanoparticle tracking analysis (NTA) provided more details about the size and concentration (Fig. 2E). Next, the MnO2 nanosheets were synthesized. TEM revealed an ultrathin sheet structure, and a lattice structure was also visible, demonstrating that the synthesis was successful (Fig. 2B; S1A). Next, we coated the exosomes with the nanosheets using a mild physical coincubation approach that depends on including coordination, hydrophobic and van der Waals forces interactions [35]. TEM revealed a distinct thin layer on the exosomal surface, indicating the successful coating of MnO2 and the formation of a typical core–shell structure (Fig. 2C). Elemental mapping further revealed a uniform distribution of Mn and O along the exosomal contour (Fig. 2D). Zeta potential analysis demonstrated that the surface charge of hESC-Exos@MnO2 shifted from an exosome-characteristic profile to a MnO2-dominated profile, clearly differing from that of the simple mixture and further confirming the successful coating of MnO2 onto the exosomal surface (Fig. S2D). Characterization and functional assessment of MnO2 nanosheets, hESC-Exos, and hESC-Exos@MnO2. (A–C) Representative transmission electron microscopy (TEM) images of (A) hESC-Exos, (B) MnO2 nanosheets, and (C) the hESC-Exos@MnO2 nanocomplex. (D) Elemental mapping analysis of hESC-Exos@MnO2. (E) Nanoparticle tracking analysis (NTA) of hESC-Exos and hESC-Exos@MnO2. (F) Western blot analysis confirming the presence of exosomal markers, including CD9, CD63, and TSG101, and the absence of negative markers Calnexin. (G) Time-dependent degradation profiles of hESC-Exos, hESC-Exos + MnO2, and hESC-Exos@MnO2 under 100 μmol/L TBHP treatment, as well as hESC-Exos in PBS. (H, I) In vivo imaging and corresponding quantitative analysis one week after treatment with hESC-Exos, hESC-Exos + MnO2, and hESC-Exos@MnO2. (H) Representative fluorescence images showing the localization of PKH26-labeled hESC-Exos, hESC-Exos + MnO2, and hESC-Exos@MnO2 within the caudal intervertebral discs in vivo. (I) Quantification of fluorescence intensity reflecting the retention of hESC-Exos in different groups. (J) Representative confocal images showing the cellular uptake of PKH26-labeled hESC-Exos, hESC-Exos + MnO2, and hESC-Exos@MnO2 (red) by NP cells under different conditions. F-actin and nuclei were stained with phalloidin (green) and DAPI (blue), respectively. (scale bar: 20 μm). Data are presented as mean ± SD of at least three independent experiments. Statistical significance was determined by one-way ANOVA: P < 0.05(*), P < 0.01(**), P < 0.001(***), P < 0.0001(****), ns = not significant. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.) To mimic the pathological microenvironment of intervertebral disc degeneration, acidic conditions (pH 6.5) combined with oxidative stress stimulation by TBHP were applied in this study [[36], [37], [38]]. Previous studies have shown that lactate accumulation in degenerative intervertebral discs induces local acidification, accompanied by excessive reactive oxygen species (ROS) production [39,40]. Compared with the release from neutral PBS, Inductively Coupled Plasma Optical Emission Spectroscopy (ICP–OES) analysis revealed that Mn2+ release from hESC-Exos@MnO2 was markedly accelerated under acidic and oxidative stress conditions, whereas only a slow-release profile was observed in neutral PBS (Fig. S2E). These findings indicate favorable microenvironment-responsive degradability. Such responsiveness allows the system to maintain structural stability and protect exosomes under physiological conditions while enabling functional release in pathological microenvironments. Combined with the observed structural preservation and environmental responsiveness, these results further suggest that the formation of this system is mainly mediated by noncovalent interactions, including coordination, hydrophobic, van der Waals interactions. The MnO2 loading amount was quantified by inductively coupled plasma (ICP) analysis. The MnO2 concentration in the hESC-Exos@MnO2 was 34.78 μg/mL, corresponding to a coating efficiency of approximately 69.56% (Fig. S2F). NTA revealed that the hydrodynamic diameter of hESC-Exos@MnO2 was slightly increased, while excellent monodispersity was maintained, conditions that are favorable for tissue penetration and cellular uptake (Fig. 2E). Collectively, these complementary structural and quantitative results demonstrate that hESC-Exos@MnO2 represents a well-defined MnO2-armored nanostructure rather than a simple physical mixture. TEM and NTA revealed that hESC-Exos clearly degraded after stimulation with 100 μmol/L tert-butyl hydroperoxide (TBHP) (Fig. S1B and C). Protein quantification analysis of hESC-Exos, hESC-Exos + MnO2, and hESC-Exos@MnO2 demonstrated that naked hESC-Exos underwent rapid degradation, with only 21% of the protein remaining after 24 h. In contrast, the simple mixture group of hESC-Exos and MnO2 exhibited a relatively slower degradation rate, although the protein retention rate at 24 h was still only 39%. Notably, hESC-Exos@MnO2 showed markedly enhanced stability, maintaining more than 80% protein retention over the same period (Fig. 2G). These results indicate that MnO2 armor provides dual protection through both a physical barrier and chemical antioxidative activity by scavenging local ROS, thereby prolonging exosomal stability and enhancing the potential role of these compounds in regenerative repair. To evaluate the in vivo protective effect and delivery efficiency of the MnO2 armor, further in vitro and in vivo experiments were performed. After caudal intervertebral disc injection of PKH26-labeled hESC-Exos@MnO2, both the retention time and fluorescence intensity at the lesion site were significantly greater than those in the bare exosome group and the simple exosome/MnO2 nanosheet mixture group (Fig. 2H and I; S1D and E). In addition, dissolved oxygen detection revealed that although TBHP caused certain interference in the assay, the dissolved oxygen level in the MnO2+TBHP group was markedly increased after the effects of PBS and TBHP itself were excluded (Fig. S2A and B). Furthermore, after nucleus pulposus cells were treated with the classical hypoxia inducer CoCl2 at 200 μmol/L for 24 h, Western blot analysis confirmed that MnO2 promoted oxygen generation in the TBHP-associated oxidative microenvironment and partially alleviated hypoxia in nucleus pulposus cells (Fig. S2C). An in vitro uptake assay further demonstrated that under TBHP stimulation, MnO2 armor not only effectively protected the exosomes but also improved the oxidative microenvironment and promoted oxygen generation, thereby significantly enhancing the uptake of hESC-Exos by nucleus pulposus cells (NPCs) (Fig. 2J; S1F). Thus, the MnO2 “armor” strategy proposed in this study can directly protect exosomes from oxidative degradation by scavenging ROS and may indirectly promote their uptake by target cells through the remodeling of the local microenvironment, such as H2O2 depletion and hypoxia alleviation. Therefore, the protective role of MnO2 nanosheets is not limited to reducing oxidative damage to hESC-Exos but may further increase their delivery efficiency to target cells by ameliorating the pathological microenvironment. In summary, these findings demonstrate that MnO2 nanoarmor effectively protects hESC-Exos, scavenges ROS, and modulates the local microenvironment, thereby markedly alleviating exosomal degradation under oxidative stress and providing important experimental evidence for the development of subsequent therapeutic strategies. Excellent biocompatibility and efficient antioxidant performance of hESC-Exos@MnO2 We evaluated the safety of hESC-Exos@MnO2 from three aspects in a systematic manner to ensure its compatibility with the body. The cytotoxicity test results were as follows. The CCK-8 assay revealed that the number of nucleus pulposus cells (NPCs) was not inhibited by hESC-Exos@MnO2 in the concentration range of 0–100 μg/mL. The nanocomposites had a proliferative effect at 50 μg/mL; however, the effect was slightly lower at 100 μg/mL (Fig. S3A). This dose-dependent increase in biocompatibility also implies that growth factors from the exosomes lead to an increase in cell proliferation and that MnO2 at the right concentration does not lead to cell death. Based on these results, we used a working concentration of 50 μg/mL in the following experiments. The results of the hemolysis test revealed that the use of hESC-Exos@MnO2 in blood is safe. PBS, MnO2, hESC-Exos, and hESC-Exos@MnO2 at 50 μg/mL resulted in no obvious red blood cell lysis; in contrast, deionized water induced marked hemolysis (Fig. S3C). Quantitative analysis also revealed that, in addition to the positive control group, all the remaining groups exhibited a below-acceptable hemolysis rate (Fig. S3B). These results are important for potential IV or tissue diffusion applications. The systemic toxicity tests revealed no pathological changes in the major organs (heart, liver, spleen, lung, or kidney) after the acute toxicity experiments with MnO2, hESC-Exos, or hESC-Exos@MnO2. Histology sections did not show any indicators of inflammation, and there was no cell degeneration or death (Fig. S3D). Serum biochemical analysis confirmed these findings. Liver markers (ALT, AST, ALP), kidney markers (Cr, UA), and nutritional indicators (TP) remained normal and did not significantly change compared with those in the control group (Fig. S4). These results also demonstrated that hESC-Exos@MnO2 is showed acceptable short-term biocompatibility at the tested dose. In IVDD, more ROS are present inside a person's body. General ROS include hydroxyl radicals (·OH), superoxide anions (O2·-), and hydrogen peroxide (H2O2). These ROS overproduce, flooding the cell and damaging proteins, lipids and nucleic acids. This process contributes to the development of several types of acute and chronic illnesses [41]. The ROS-scavenging capacity of hESC-Exos@MnO2 is very important because it protects disc tissue and maintains the shape and function of the carried hESC-Exos. We comprehensively tested the antioxidant performance of hESC-Exos@MnO2 at 50 μg/mL at both the molecular and cellular levels and investigated its multiple protective mechanisms at the molecular level. hESC-Exos@MnO2 could effectively eliminate DPPH·(2,2-diphenyl-1-picrylhydrazyl) and ABTS+·(2,2′-azino-bis(3-ethylbenzothiazoline-6-sulfonic acid) cation) radicals and ·OH radicals. DPPH is a stable free radical that appears purple after it accepts electrons. It shows a characteristic peak change at 517 nm [42]. The hESC-Exos@MnO2 solution changed from purple to yellow, and its absorbance also decreased (Fig. 3A and D). The DPPH· scavenging rate determined by quantitative measurement was 80.27% (Fig. 3G). The color of the ABTS+·:hESC-Exos@MnO2 solution changed from blue to colorless, and the 743 nm absorption peak clearly decreased, indicating an antioxidant capacity of 0.739 mM (Fig. 3B, E and H). Hydroxyl radicals (·OH) are the most reactive ROS and can damage biological molecules via electron transfer, addition, and hydrogen abstraction reactions [43]. The ·OH concentration was measured by a Fenton-like reaction assay. H2O2+FeSO4 generates ·OH, which is coincubated with PBS, hESC-Exos, MnO2, or hESC-Exos@MnO2, followed by MB, ·OH causes MB bleaching. The blank control exhibited obvious bleaching, and a decrease in the absorption peak at 665 nm was detected; however, this phenomenon was not detected in the other groups (Fig. 3C and F). The percentage of •OH scavenging in the hESC-Exos@MnO2 group reached 83.64% (Fig. 3H). ExcessO2·- can destroy redox balance; damage mitochondrial function; damage DNA, protein, and lipids; and activate inflammatory pathways such as those involving NLRP3, ultimately inducing apoptosis and ferroptosis [44]. PBS, hESC-Exos, MnO2, or hESC-Exos@MnO2 was used for testing, and the efficiency of the nanocomposite was 80.67%, as shown in (Fig. 3J). These results indicate that hESC-Exos@MnO2 is good at clearing different ROS, such as DPPH·, ABTS+·, ·OH, and O2•-, effectively lowering the level of extracellular ROS and maintaining the integrity and bioactivity of hESC-Exos, improving cell survival. In vitro antioxidant performance of MnO2, hESC-Exos, and hESC-Exos@MnO2. (A–C) UV–vis absorption spectra demonstrating the radical scavenging capacity against (A) DPPH·, (B) ABTS+·, and (C) ·OH by MnO2, hESC-Exos, and hESC-Exos@MnO2 at 50 μg/mL (D–F) Visual color changes of (D) DPPH·, (E) ABTS+·, and (F) ·OH solutions following treatment with the respective agents. (G–J) Quantitative analysis of the scavenging rates for (G) DPPH·, (H) ABTS+·, (I) ·OH, and (J) O2·-. (K, L) Evaluation of nanozyme-mimetic activities: (K) Catalase (CAT)-like and (L) superoxide dismutase (SOD)-like activities. Data are presented as mean ± SD of at least three independent experiments. Statistical significance was determined by one-way ANOVA: P < 0.05(*), P < 0.01(**), P < 0.001(***), P < 0.0001(****), ns = not significant. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.) In addition to having a strong ability to eliminate different ROS, hESC-Exos@MnO2 also has clear nanozyme catalytic activity, with both CAT- and SOD-like functions that can effectively breakdown H2O2 and O2·- at the same time (Fig. 3K and L). High CAT activity removes H2O2 and can reduce oxidative stress-related cell damage. Moreover, it supports cell survival and growth by maintaining normal cell metabolism and signaling [45]. SOD also removes a large amount of O2·-. This action limits the damage to cells caused by free radicals and helps maintain a balance in terms of oxidation and antioxidant activity inside cells [46]. Together, this enzyme-like activity solves the problems associated with antioxidants that just remove ROS once and allows for continuous and dynamic control of ROS in the diseased microenvironment. In summary, hESC-Exos@MnO2 can remove large amounts of ROS. The system itself also possesses CAT- and SOD-like nanozyme activities. These activities act in concert to catalyze ROS breakdown and allow long-term control. Moreover, loaded and released hESC-Exos can also deliver bioactive molecules. These molecules enhance antioxidant defense through control and repair. By combining these functions, a protection network with many layers is formed. This network is markedly better at addressing oxidative stress altogether. This strategy can provide a powerful “shield” for the delivery of bioactive contents of hESC-Exos@MnO2. It can also greatly improve the antiaging effect on nucleus pulposus cells. At the cellular level, hESC-Exos@MnO2 has a strong redox-balancing capacity. To test the effect on the removal of ROS, the cells were stained with 2,7-dichlorofluorescein diacetate (DCFH-DA), and FCM was performed [47]. The results revealed that TBHP upregulated intracellular ROS. However, hESC-Exos@MnO2 reduced the amount of ROS produced by TBHP (Fig. 4A and C). Furthermore, ECM analysis indicated that TBHP clearly decreased MitoTracker Red fluorescence intensity, indicating a loss of mitochondrial membrane potential. The fluorescence intensity largely recovered upon treatment, indicating that the treatment improved the structure and functionality of the mitochondrial network (Fig. 4B and D). To further understand the function of mitochondria, we used a JC-1 fluorescence probe to study the dynamic change in the mitochondrial membrane potential (ΔΨm) [48]. The confocal images clearly revealed that TBHP caused a decrease in the mitochondrial membrane potential and increased the amount of JC-1 monomers. However, MnO2, hESC-Exos and hESC-Exos@MnO2 increased the number of mitochondria and increased JC-1 aggregation (Fig. 4E and F). These results demonstrate that hESC-Exos@MnO2 can not only eliminate ROS but also significantly improve ROS-induced decrease in the mitochondrial membrane potential. This effect can help maintain the activity of the electron transport chain and steady cell energy metabolism. hESC-Exos@MnO2 modulates ROS levels and mitochondrial function in NPCs under oxidative stress. (A, C) Flow cytometry (FCM) analysis of DCFH-DA staining (A) and the corresponding quantification (C) of relative fluorescence intensity in NPCs under different treatments. (B, D) FCM analysis of MitoTracker Red CMXRos staining (B) and the quantification (D) of relative fluorescence intensity. (E, F) Representative JC-1 staining images (E) (scale bar: 100 μm) and quantitative analysis (F) of the aggregate/monomer fluorescence intensity ratio in differently treated NPCs. (G, H) Representative DCFH-DA staining images (G) (scale bar: 50 μm) and quantification (H) of relative fluorescence intensity. (I, J) FCM analysis of JC-1 staining (I) and quantification (J) of the aggregate/monomer ratio. Data are presented as mean ± SD of at least three independent experiments. Statistical significance was determined by one-way ANOVA: P < 0.05(*), P < 0.01(**), P < 0.001(***), P < 0.0001(****), ns = not significant. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.) To test whether hESC-Exos@MnO2 affects mitochondrial superoxide (O2·-), we used MitoSOX Red for testing [49]. The results demonstrated that TBHP induced in a rapid increase in mitochondrial O2·- levels. In contrast, compared with the control, hESC-Exos@MnO2 decreased O2·- levels close to baseline. This effect was greater than that of hESC-Exos or MnO2 alone (Fig. S5). These results indicate that hESC-Exos@MnO2 synergistically targeted mitochondrial oxidative stress. The effect fits its SOD-like activity. Confocal microscopy also revealed that hESC-Exos @MnO2 efficiently decreased intracellular ROS levels (Fig. 4G and H). The flow cytometry results for JC-1 also verified these findings. The data revealed that hESC-Exos@MnO2 decreased the amount of JC-1 monomers and increased the amount of JC-1 aggregates (Fig. 4I and J). These results demonstrate that hESC-Exos@MnO2 is a good ROS scavenger that greatly decreases oxidative stress in cells and mitochondria, regulates cell redox, and maintains the form and function of the mitochondria. This system provides layering of protection, the effects of which include ROS clearance, membrane potential repair, and mitochondrial network stabilization. Therefore, hESC-Exos@MnO2 has a clear and well-coordinated effect on cell antioxidant defense and mitochondrial maintenance. Transcriptomics-based mechanism analysis: hESC-Exos@MnO2 reverses the degenerative phenotype of NPCs through coordinated multipathway regulation To obtain a complete understanding of the effects of the global regulation of hESC-Exos@MnO2 on IVD degeneration, we performed RNA-seq to obtain the transcriptomes of NPCs subjected to different treatments [50]. The experimental groups included the control, TBHP injury and hESC-Exos@MnO2 treatment groups. NPCs in the injury and treatment groups were first exposed to 100 μmol/L TBHP for 24 h to establish an injury model. After this step, the control group was cultured in basic medium, the injury group received PBS, and the treatment group received medium supplemented with 50 μg/mL hESC-Exos@MnO2. All the groups were cultured for another 48 h. Principal component analysis (PCA) revealed that the control, TBHP injury, and hESC-Exos@MnO2 treatment groups formed three distinct clusters in the transcriptome space (Fig. 5A). This result clearly revealed that the treatment altered the gene expression disorder caused by TBHP. Volcano plot analysis showed that compared with the control group, the TBHP injury group had 2387 genes whose expression levels were significantly downregulated and 2178 genes whose expression levels were upregulated (Fig. S6A). These results indicate that TBHP induced global gene expression changes. When the TBHP injury group was compared with the hESC-Exos@MnO2 treatment group, 2268 genes were significantly downregulated, and 3148 genes were significantly upregulated in the injury group. Many of these genes were connected to pyroptosis, cell aging, and ECM pathways (Fig. 5B). These findings indicate that the treatment strongly altered the changes in gene expression resulting from injury. Venn diagram analysis revealed that the different groups shared important regulated genes, but each group also had its own unique genes. These findings indicate that hESC-Exos@MnO2 acts at multiple levels and along multiple paths (Fig. 5C). The heatmap results revealed that many genes were abnormally expressed in response to TBHP injury. After hESC-Exos@MnO2 treatment, the expression patterns of some genes were restored or altered (Fig. 5D; S6B). Together, these data show that hESC-Exos@MnO2 broadly regulates gene expression in injured cells and has strong potential to control cell function and state. Next, we performed pathway enrichment analysis to identify important targets. Gene Ontology (GO) analysis revealed that the DEGs were enriched in oxidative stress response-, DNA damage repair-, and inflammation-related processes (Fig. 5E). To define which pathways were involved, we performed Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis [51]. The results revealed that hESC-Exos@MnO2 mainly regulated pathways related to pyroptosis, cell aging, and ECM–receptor interactions (Fig. 5F). These findings provide direct molecular evidence of the wide and synchronized effects of hESC-Exos@MnO2. Transcriptomic analysis of NPCs after TBHP treatment and hESC-Exos@MnO2 intervention. (A) Principal component analysis (PCA) plot showing distinct clustering of the Control, TBHP, and hESC-Exos@MnO2 treatment groups. (B) Volcano plot of differentially expressed genes (DEGs) between the TBHP and hESC-Exos@MnO2 groups. (C) Venn diagram illustrating the overlap of DEGs among the Control vs. TBHP, TBHP vs. hESC-Exos@MnO2, and Control vs. hESC-Exos@MnO2 comparisons. (D) Heatmap of DEGs between the TBHP and hESC-Exos@MnO2 groups. (E) GO enrichment analysis of DEGs between TBHP and hESC-Exos@MnO2 groups, highlighting biological processes related to oxidative stress, DNA damage, and inflammation. (F) KEGG pathway enrichment analysis showing the impact of hESC-Exos@MnO2 treatment on pyroptosis, senescence, and ECM-related pathways compared to the TBHP group. (G–I) GSEA enrichment plots for reactive oxygen species metabolic process (G), DNA damage response (H), and inflammatory response (I). (J–L) GSEA enrichment plots for the NOD-like receptor signaling pathway (J), p53 signaling pathway (K), and ECM-receptor interaction pathway (L). At least three independent experiments were performed. Gene set enrichment analysis (GSEA) identified the core regulatory network driven by hESC-Exos@MnO2 treatment. Our analysis revealed that TBHP treatment reversed the abnormal upregulation of the expression of genes related to ROS metabolism, the DNA damage response, and inflammation (Fig. 5G–I). Compared with the treatment group, the TBHP injury group had upregulated expression of genes related to the Toll-like receptor and NOD-like receptor signaling pathways (Fig. 5J; S6C). The process of innate immune activation and activation related to inflammasomes is very important for injuries. The Toll-like receptor pathway provides the priming signal for the inflammasome. The NOD-like receptor then sends out the downstream execution order. This signaling cascade activates caspase-1 and gasdermin d and ultimately causes pyroptosis [52]. The TBHP injury group also exhibited very strong p53 signaling activation and continuous inhibition of the cell cycle pathway (Fig. 5K; S6D). p53-regulated cell cycle arrest controls downstream target gene expression. This arrest holds NLPs at the G1/S checkpoint, decreasing cell growth and resulting in SASP. These changes ultimately promote the aging of NPCs [53]. In addition, the ECM–receptor interaction pathway was inhibited in the injury group, but treatment with hESC-Exos@MnO2 effectively restored the activity of the ECM–receptor interaction pathway (Fig. 5L). This restoration implies that treatment can help rebuild extracellular matrix (ECM) homeostasis and enhance cell adhesion and signaling. Such changes would likely increase the amount of the matrix and decrease how much of the matrix is lost [54]. In summary, we report that hESC-Exos@MnO2 operates through ROS clearance and blocking downstream inflammation and DNA damage signals. These actions halt the progression of cell pyroptosis and senescence and promote the metabolic production of the ECM, and comprehensive transcriptome analyses revealed that hESC-Exos@MnO2 acts through multiple pathways together. These structures do not just remove the cells breaking down because of too much oxygen at a tiny scale; they also repair stable cells and wake up growing cells. Verification of the core mechanism: inhibiting pyroptosis, delaying aging, and remodeling the ECM Based on the transcriptome data, we carried out systematic verification at the molecular, protein and cellular levels. hESC-Exos@MnO2 inhibits NLRP3 inflammasome-mediated pyroptosis by regulating pattern recognition receptor pathways IVDD involves an inflammatory oxidative stress destruction loop [55]. Abnormal ROS levels are endogenous danger signals that trigger the expression of PRRs, namely, the NLRP3 inflammasome, to cause a very inflammatory type of cell death called pyroptosis [56]. RT‒qPCR revealed that TBHP strongly increased the expression of pyroptosis-related factors such as NLRP3, GSDMD, caspase-1, IL-1β, and IL-18. Both MnO2 and hESC-Exos alone were able to provide a small amount of relief; however, hESC-Exos@MnO2 was able to suppress the expression of these genes much more effectively (Fig. 6A–E). Western blotting was performed to verify the protein levels (Fig. 6F and G). More specifically, hESC-Exos@MnO2 decreased the expression of upstream proteins such as NLRP3 and pro-IL-β. In addition, the key downstream executors GSDMD, caspase-1, IL-1β, and IL-18 were reduced. Most notably, the treatment inhibited the active forms of these proteins, such as cleaved caspase-1 and the GSDMD-N fragment. The mRNA and protein expression findings confirmed that hESC-Exos@MnO2 suppressed NLRP3/Caspase-1-associated pyroptotic responses, and functional tests provided even more direct evidence. Live/dead cell staining revealed that TBHP stimulation led to the death of many PI-positive cells. However, the death rate of the hESC-Exos@MnO2 group was much lower (Fig. 6H and I). These effects were verified by testing the caspase-1 enzyme in pyroptosis. Compared with TBHP treatment, hESC-Exos@MnO2 treatment effectively inhibited caspase-1 activity (Fig. 6J). These findings are consistent with our protein data and confirm that hESC-Exos@MnO2 prevent caspase-1 activity. The system halts this activation to stop GSDMD cleavage and the maturation of inflammatory cytokines, such as IL-1β and IL-18. Finally, we performed immunofluorescence staining and showed that hESC-Exos@MnO2 successfully reduced intracellular caspase-1 and NLRP3 expression under TBHP-induced stress, as shown in (Fig. 6K–M). In total, some anti-pyroptosis effects were detected in all the groups, but compared with MnO2 alone, hESC-Exos performed better, while hESC-Exos@MnO2 had the most inhibitory effect. hESC-Exos@MnO2 suppresses NPC pyroptosis via the pattern recognition receptor pathway. (A–E) RT-qPCR analysis of mRNA levels for NLRP3, GSDMD,caspase-1, IL-1β, and IL-18 in NPCs under different treatments. (F, G) Representative western blot images (F) and densitometric quantification (G) of pyroptosis-related proteins. (H, I) Representative live/dead staining images (H) (scale bar: 200 μm) and quantification (I) of PI-positive cells. (J) Caspase-1 enzymatic activity in NPCs after different treatments. (K–M) Representative immunofluorescence images for Caspase-1 and NLRP3 (red) (K) (scale bar: 50 μm). F-actin and nuclei were stained with phalloidin (green) and DAPI (blue), respectively. Quantification of relative fluorescence intensity for Caspase-1 (L) and NLRP3 (M) is shown. Data are presented as mean ± SD of at least three independent experiments. Statistical significance was determined by one-way ANOVA: P < 0.05(*), P < 0.01(**), P < 0.001(***), P < 0.0001(****), ns = not significant. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.) To determine whether the antipyroptotic effect of hESC-Exos@MnO2 depends on the NLRP3/caspase-1 pathway, the NLRP3 inhibitor MCC950 and the caspase-1 inhibitor VX-765 were used. TBHP treatment markedly increased IL-1β and IL-18 secretion, as well as LDH release, indicating the activation of pyroptosis. hESC-Exos@MnO2 significantly reduced these pyroptosis-associated readouts, with an inhibitory efficacy comparable to that of MCC950 alone. Notably, compared with single treatments, combined treatment with hESC-Exos@MnO2 and either VX-765 or MCC950 further decreased IL-1β and IL-18 secretion and LDH release (Fig. S7A–C), suggesting that the antipyroptotic activity of hESC-Exos@MnO2 involves not only the NLRP3/caspase-1 axis but also its antioxidative and mitochondrial ROS-scavenging capacities. Consistently, TBHP markedly upregulated the mRNA expression of caspase-1 and GSDMD, whereas hESC-Exos@MnO2, either alone or in combination with the inhibitors, significantly suppressed their expression (Fig. S7D and E). Mechanistically, hESC-Exos@MnO2 exerts its antipyroptotic effects through multilayered regulation. On the one hand, MnO2 nanosheets scavenge ROS through enzyme-mimetic activity, thereby attenuating upstream signals required for NLRP3 inflammasome activation [57]. On the other hand, hESC-Exos may regulate related signaling pathways by delivering bioactive molecules and suppressing inflammation-associated gene expression at the transcriptional level [15]. Thus, hESC-Exos@MnO2 integrates antioxidative and anti-inflammatory functions, enabling multitarget intervention in the oxidative–inflammatory cascade, which may explain its superior therapeutic efficacy. Collectively, these findings suggest that hESC-Exos@MnO2 suppresses NLRP3 inflammasome-mediated pyroptosis by modulating pattern-recognition receptor–related inflammatory signaling through multiple regulatory layers, thereby restraining the inflammatory cascade and preserving nucleus pulposus cell function. hESC-Exos@MnO2 promotes NPC regeneration by regulating the p53 signaling pathway and cell cycle network Senescence of intervertebral disc nucleus pulposus cells (NPCs) is a key factor in reduced tissue anabolic function, increased catabolism, and impaired regenerative capacity. This process is controlled mainly by the classical p53–p21 signaling axis [5,58]. At the molecular level, DNA damage caused by oxidative stress activates p53 signaling and triggers cell cycle arrest and senescence [59]. Our results revealed that hESC-Exos@MnO2 treatment significantly suppressed the TBHP-induced overexpression of p53 and its downstream effector p21 at the mRNA level (Fig. 7A and B). Moreover, the secretion of key proinflammatory factors of the senescence-associated secretory phenotype (SASP), including IL-6 and TNF-α, was markedly reduced (Fig. 7C and D). Western blot analysis further confirmed that hESC-Exos@MnO2 decreased the protein levels of senescence markers and SASP-related proteins (Fig. 7E and F). These results indicate that hESC-Exos@MnO2 can block overactivation of the p53 pathway at both the transcriptional and translational levels. It also improves cell cycle progression and promotes cell homeostasis and regenerative potential at the molecular level. At the cellular level, SA-β-gal staining revealed that NPCs treated with hESC-Exos@MnO2 under oxidative stress had a lower percentage of positive cells (Fig. 7G and H). Immunofluorescence confirmed that hESC-Exos@MnO2 reduced the expression of P21, P53, and IL-6 in TBHP-stimulated cells. In addition, the formation of γ-H2AX, a marker of DNA double-strand breaks, was clearly decreased (Fig. 7I–M). To evaluate the effect of hESC-Exos@MnO2 on the proliferative capacity of NPCs under oxidative stress, EdU staining and cell cycle analysis were performed. TBHP treatment markedly reduced the proportion of EdU-positive cells, indicating impaired DNA synthesis and suppressed cell proliferation. Although MnO2, hESC-Exos, and the simple mixture of hESC-Exos and MnO2 partially restored the EdU-positive population, hESC-Exos@MnO2 produced a more pronounced proproliferative effect (Fig. S8A and B). hESC-Exos@MnO2 promotes NPC regeneration by regulating the p53 pathway and cell cycle network. (A–D) RT-qPCR analysis of mRNA levels for TP53, CDKN1A (p21), TNF-α, and IL-6. (E, F) Representative western blot images (E) and quantification (F) of senescence and SASP-related proteins. (G, H) Representative SA-β-gal staining images (scale bar: 200 μm) (G) and quantification (H) of positive cells. (I–M) Representative immunofluorescence images for IL-6, γH2AX, p21, and p53 (red) (I) (scale bar: 10 μm). Quantification of relative fluorescence intensity is shown for IL-6 (J), γ-H2AX (K), p21 (L), and p53 (M). F-actin and nuclei were stained with phalloidin (green) and DAPI (blue). Data are presented as mean ± SD of at least three independent experiments. Statistical significance was determined by one-way ANOVA: P < 0.05(*), P < 0.01(**), P < 0.001(***), P < 0.0001(****), ns = not significant. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.) Flow cytometric analysis further revealed that TBHP induced prominent G0/G1-phase arrest, accompanied by a reduction in the proportion of S-phase cells. In contrast, MnO2, hESC-Exos, and the simple hESC-Exos + MnO2 mixture partially alleviated this cell cycle disturbance, whereas hESC-Exos@MnO2 significantly reduced the G0/G1-phase population and increased the S-phase fraction, indicating partially alleviated oxidative stress-induced cell-cycle arrest (Fig. S8C and D). Together, these results demonstrate that hESC-Exos@MnO2 markedly promotes NPC proliferation and mitigates TBHP-induced cell cycle arrest, highlighting its potential to restore cellular regenerative capacity in an oxidative stress microenvironment. The above results show that because they are strong antioxidants, hESC-Exos@MnO2 has good protective effects on the structure and function of hESC-Exos. Moreover, it greatly reduces DNA damage from oxidative stress, stops continuous abnormal p53 signaling from continuing, and promotes NPC growth and renewal, along with hESC-Exos@MnO2, not only helping cells overcome senescence-related cell cycle arrest but also decreasing the amount of proinflammatory factors produced by the SASP. A pro-cell proliferation and anti-aging microenvironment is created [60]. In short, hESC-Exos@MnO2 provides cell protection and tissue regeneration and maintain the activity and homeostasis of existing functional cells through antiaging mechanisms. However, it also enhances cell proliferation and renewal to improve the self-repair capacity of the tissue. These results demonstrate that this composite system can regulate key signaling pathways associated with senescence, promoting the proliferation and renewal of NPCs. These findings provide cellular-level support for regenerative repair in intervertebral disc degeneration. hESC-Exos@MnO2 maintains metabolic homeostasis in NPCs by regulating ECM–receptor interactions The balance between ECM synthesis and degradation is essential for maintaining intervertebral disc structure and function. Disruption of this balance is a key factor in IVDD development [61,62]. At the molecular level, qRT‒PCR revealed that in NPCs treated with hESC-Exos@MnO2, the levels of SOX9, type II collagen (COL2A1), and aggrecan (ACAN), which are closely related to matrix synthesis, significantly increased. Moreover, the expression of matrix-degrading enzymes, including MMP13 and ADAMTS5, was clearly suppressed (Fig. 8A–E). Western blot analysis further confirmed that hESC-Exos@MnO2 effectively reduced the protein levels of matrix-degrading enzymes while increasing the expression of matrix synthesis-related proteins (Fig. 8F and G). At the cellular level, Alcian blue and safranin O staining revealed that TBHP strongly inhibited the synthetic function of NPCs [63]. MnO2 or hESC-Exos alone provided only moderate protection, suggesting that Mn2+ released from both MnO2 and hESC-Exos can support ECM repair. In contrast, hESC-Exos@MnO2 markedly improved the proteoglycan synthesis phenotype of NPCs. These findings suggest that the armor exosome system more effectively restored the synthetic capacity of NPCs and exhibited a synergistic effect in promoting ECM repair (Fig. 8H–J). These combined results indicate that hESC-Exos@MnO2 can shift ECM metabolism toward a more balanced anabolic profile. These changes in protein expression were verified by immunofluorescence. In the cells treated with hESC-Exos@MnO2, abundant COL2A1 and ACAN protein expression was detected surrounding the NPCs. In comparison, the fluorescence signal for MMP13 was much weaker in this group (Fig. 8K–O). hESC-Exos@MnO2 maintains ECM metabolic homeostasis in NPCs by regulating ECM-receptor interaction. (A–E) RT-qPCR analysis of mRNA levels for catabolic (MMP13, ADAMTS5) and anabolic (COL2A1, SOX9, ACAN) ECM-related genes. (F, G) Representative western blot images (F) and quantification (G) of ECM catabolic and anabolic proteins. (H–J) Representative Alcian Blue and Safranin O staining images (H) (scale bar: 200 μm) and quantitative analysis of staining intensity (I, J). (K–O) Representative immunofluorescence images for MMP1, MMP13, ACAN, and COL2A1 (green) (K) (scale bar: 20 μm). Quantification of relative fluorescence intensity is shown for MMP1 (L), MMP13 (M), ACAN (N), and COL2A1 (O). F-actin and nuclei were stained with phalloidin (red) and DAPI (blue). Data are presented as mean ± SD of at least three independent experiments. Statistical significance was determined by one-way ANOVA: P < 0.05(*), P < 0.01(**), P < 0.001(***), P < 0.0001(****), ns = not significant. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.) In our study, we demonstrated that hESC-Exos@MnO2 could generally regulate the expression of genes related to ECM metabolism. hESC-Exos@MnO2promotes a metabolic state change from mainly breakdown to mainly synthesis, which maintains a physiological balance. The regulation of ECM metabolism by hESC-Exos@MnO2 is not solely due to regenerative signals from hESC-Exos directly activating synthesis pathways; it also depends on its antioxidant and anti-inflammatory effects, which improve the cellular microenvironment. These effects help cells restore their own anabolic metabolic function. This homeostasis-oriented regulation is closer to a natural repair process, as it helps rebuild the dynamic balance between ECM synthesis and degradation. Moreover, this mechanism reduces the risk of pathological changes, such as abnormal fibrosis caused by excessive stimulation of synthesis [64]. Therefore, by regulating ECM–receptor interactions and related signaling networks, hESC-Exos@MnO2 provide an important material basis and regulatory support for maintenance of the intervertebral disc height. hESC-Exos@MnO2 shows significant therapeutic effects in a rat model of intervertebral disc degeneration Based on clear evidence from in vitro studies, this study further evaluated the in vivo effects of MnO2, hESC-Exos, and hESC-Exos@MnO2 in a rat tail IVDD model. We induced IVDD by injecting TBHP into rat intervertebral discs using a needle [65,66]. After 14 days, the TBHP, TBHP + MnO2, TBHP + hESC-Exos, and TBHP + hESC-Exos@MnO2 groups received weekly injections of PBS, MnO2, hESC-Exos, and hESC-Exos@MnO2, respectively [30]. After 2 weeks of treatment, we assessed the degree of disc degeneration (Fig. 9A). hESC-Exos@MnO2 attenuates IVDD progression and preserves disc structure in rats. (A) Schematic of the experimental timeline for rat caudal IVDD modeling and treatment. (B, C) Representative T2-weighted MRI images (B) and Pfirrmann degenerative grades (C) of rat caudal IVDs after treatment. (D) Representative X-ray images of rat caudal IVDs. (E) Representative 3D reconstructed micro-CT images of rat caudal IVDs. (F) Quantification of the disc height index (DHI%) based on X-ray images. (G) Representative H&E, Safranin O/Fast Green (SO/FG), and Masson's trichrome staining of rat caudal IVD tissues (scale bars: 500 μm or 50 μm). (H) Histological scores based on staining results. Data are presented as mean ± SD of at least three independent experiments. Statistical significance was determined by one-way ANOVA: P < 0.05(*), P < 0.01(**), P < 0.001(***), P < 0.0001(****), ns= not significant. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.) Imaging analysis confirmed that hESC-Exos@MnO2 significantly restored the overall structure of the IVD [67]. Analysis of MR T2-weighted images revealed that although MnO2 and hESC-Exos offered only minor relief from TBHP-induced degeneration, the hESC-Exos@MnO2 treatment group showed significant improvement. The signal intensity of the caudal discs in the hESC-Exos@MnO2 treatment group rats was significantly higher, and the Pfirrmann grades remained close to normal levels (Fig. 9B and C). These results show that both the hydration status and the internal architecture were effectively restored by hESC-Exos@MnO2. X-ray and micro-CT 3D reconstruction also revealed that the disc height index (DHI%) was well maintained in the hESC-Exos@MnO2 group. Compared with the control groups included in the same in vivo therapeutic assessments, the hESC-Exos@MnO2 group exhibited attenuated disc degeneration (Fig. 9D–F). This composite system exhibited a synergistic repair effect on TBHP-induced disc degeneration in the present study. Treatment with hESC-Exos@MnO2 promoted ECM homeostasis restoration of the disc while concurrently attenuating degenerative progression at the macroscopic level. Histological analysis revealed the microscopic changes and repair status of NPCs and ECM in the discs after treatment. H&E, SO/FG, and Masson staining showed that the hESC-Exos@MnO2 group maintained the structure of the NP. The annulus fibrosus (AF) was more regularly arranged, and tissue continuity was preserved. A clear boundary appeared between the NP and AF regions, and the cell density in the NP increased markedly. The histological scores indicated a significant improvement in disc degeneration (Fig. 9G and H). Compared with that in the hESC-Exos group, the AF lamellar structure was better preserved. These findings suggest that compared with the direct delivery of hESC-Exos, the delivery of hESC-Exos via hESC-Exos@MnO2 has a stronger regenerative effect on disc tissue. These morphological changes indicate that the treatment not only promotes cell survival and proliferation but also effectively promotes normal ECM remodeling. Molecular expression analysis further linked the in vivo efficacy to the in vitro mechanisms. After treatment, the expression levels of the proinflammatory cytokine IL-1β and the matrix-degrading enzyme MMP13 were significantly reduced in rat disc tissues, whereas the expression of the matrix synthesis marker COL2A1 was markedly increased (Fig. 10A–D). Fluorescence imaging further confirmed that the treatment not only led to better-maintained disc height but also resulted in reduced expression of MMP3 and increased expression of aggrecan in both NPCs and AFCs (Fig. 10E–G). hESC-Exos@MnO2 improves disc tissue homeostasis by modulating inflammation and ECM metabolism in vivo. (A–D) Representative immunohistochemical images (A) (scale bar: 50 μm) and quantitative analysis of protein expression levels for IL-1β (B), MMP13 (C), and COL2A1 (D) in rat caudal IVD tissues. (E–G) Representative immunofluorescence images of rat caudal IVD tissues (E) (scale bars: 500 μm or 50 μm) and quantitative analysis of relative fluorescence intensity for MMP3 (F) and Aggrecan (G). Data are presented as mean ± SD of at least three independent experiments. Statistical significance was determined by one-way ANOVA: P < 0.05(*), P < 0.01(**), P < 0.001(***), P < 0.0001(****), ns = not significant. This research has proven the feasibility of the integrated therapeutic strategy of hESC-Exos@MnO2. Compared with traditional multidrug combinations or complex delivery systems, only one preparation is needed to achieve multiple repair goals, such as clearing the pathological microenvironment, promoting cell regeneration and rebuilding matrix homeostasis. In summary, multilevel evidence has shown that hESC-Exos@MnO2 can systematically maintenance the height, tissue morphology and molecular expression steady state of intervertebral discs by synergistically regulating oxidative stress, inflammatory reactions and matrix metabolism. This study provides evidence from mechanism analysis to efficacy verification, highlighting the promising potential of this composite system for disc regeneration and repair. Discussion The enzyme-mimetic activity of MnO2 nanozymes confers unique biological advantages and has been shown to improve therapeutic outcomes in multiple diseases through the efficient scavenging of reactive oxygen species (ROS) [25,68,69]. Moreover, hESC-Exos have been demonstrated to mediate regenerative signaling, promote cellular repair, and delay cellular senescence [13,70]. IVDD is a prevalent yet therapeutically challenging degenerative disorder characterized by chronic inflammation, excessive ROS accumulation, and progressive ECM destruction, which together limit the efficacy of conventional treatments [31,71]. Although stem cell- and exosome-based biotherapies have shown reparative potential, their vulnerability to inactivation and rapid loss within the hostile degenerative microenvironment remains a major barrier to clinical translation [72]. To address this challenge, we developed a microenvironment-responsive nanobiocomposite, hESC-Exos@MnO2, based on an “active defense–synergistic repair” design principle, aiming to integrate delivery protection, microenvironmental modulation, and regenerative repair within a single platform. The central feature of this system lies in the armored protective layer formed by MnO2 nanosheets. In the ROS-enriched pathological microenvironment of IVDD, MnO2 continuously scavenges ROS through its nanozyme activity [28,73], thereby directly alleviating oxidative stress while protecting the loaded hESC-Exos from oxidative degradation. As a result, hESC-Exos@MnO2 treatment markedly improves exosome retention, preservation, and biodistribution at the lesion site, enabling a transition from effective delivery to functional tissue repair. Moreover, unlike conventional exosome delivery strategies, such as hydrogels, microspheres, or membrane coatings, which rely mainly on passive encapsulation and sustained release, the MnO2-armored system developed in this study combines microenvironment-responsive behavior with active regulatory capacity. ICP‒OES analysis revealed that the release of Mn2+ from hESC-Exos@MnO2 was accelerated under acidic and oxidative conditions but remained relatively stable in neutral PBS, suggesting that responsive degradation occurred within the pathological microenvironment of degenerative discs and enabled both exosome protection and on-demand release. In parallel, MnO2 not only acts as a physical barrier against ROS-mediated exosomal damage but also continuously scavenges ROS and ameliorates the inflammatory microenvironment through its nanozyme activity. Thus, MnO2 functions beyond acting as a passive carrier and cooperates with hESC-Exos to generate synergistic therapeutic effects. Consistently, TEM, elemental mapping, ICP, and zeta potential analyses confirmed the stable loading of MnO2 onto the exosomal surface and the formation of a responsive “nanoarmor” architecture rather than a simple physical mixture. Compared with conventional delivery platforms that rely solely on encapsulation or sustained release, this system integrates delivery, responsiveness, and therapy into a unified platform, which may partly account for its enhanced reparative efficacy in intervertebral disc regeneration. Therefore, the MnO2 armor strategy may offer a promising alternative to traditional passive exosome delivery systems for disc regeneration. More importantly, hESC-Exos@MnO2 exerts therapeutic effects through a multilayered and multipathway cooperative network. First, its potent antioxidative capacity markedly reduces both intracellular and mitochondrial ROS levels, thereby attenuating direct ROS-mediated cellular injury and blocking ROS-driven upstream activation of the NLRP3 inflammasome. This suppresses NLRP3 inflammasome-mediated pyroptosis, as evidenced by the downregulation of key pyroptotic molecules, including NLRP3, GSDMD, and caspase-1, as well as the downstream proinflammatory cytokines IL-1β and IL-18, ultimately interrupting the vicious cycle between inflammatory amplification and pyroptotic cell death [74,75]. ROS clearance also reduces DNA damage, thereby helping delay p53-associated cellular senescence [[76], [77], [78]]. In parallel, hESC-Exos were protected and successfully delivered by MnO2 armor, further mediating regenerative signaling through their bioactive cargos and regulating key pathways such as pattern recognition receptor signaling, the p53 pathway, and cell cycle progression. These effects not only synergistically suppress nucleus pulposus cell pyroptosis but also promote NPC proliferation and regeneration [79,80]. Furthermore, this system upregulates the expression of ECM anabolism-related genes, including SOX9, COL2A1, and ACAN, while suppressing the expression of matrix-degrading enzymes such as MMP13 and ADAMTS5, thereby reprogramming ECM metabolism from a catabolic state to an anabolic state and restoring the matrix homeostasis required for disc structure and function [81,82]. In a rat caudal disc degeneration model, hESC-Exos@MnO2 exhibited a robust ability to promote disc regeneration and repair. Multimodal imaging, including MRI, together with histological analyses, consistently demonstrated that this treatment restored disc height, improved tissue hydration, preserved relatively normal tissue architecture, and recapitulated anti-inflammatory, anti-senescent, and pro-anabolic molecular profiles. Taken together, these findings provide robust evidence that hESC-Exos@MnO2 exerts therapeutic benefits spanning from molecular modulation to macroscopic tissue restoration. Furthermore, the markedly enhanced therapeutic efficacy of this composite system underscores the synergistic interplay between MnO2-mediated protection and hESC-Exos-driven regenerative signaling. Despite these encouraging mechanistic and in vivo findings, several issues warrant further investigation. First, the key bioactive components within hESC-Exos, such as specific miRNAs or proteins, should be further defined. Second, the material fabrication process needs to be optimized to establish standardized, reproducible, and scalable production protocols. Regarding delivery strategies, integration with local sustained-release carriers could be explored to prolong intradiscal retention, enhance targeting efficiency, and reduce dosing frequency. Although the hESC-Exos + MnO2 simple-mixture group was included to demonstrate the superiority of the armored formulation over simple co-administration, additional control groups—such as inactive MnO2-coated hESC-Exos or non-exosomal particles@MnO2—would be valuable for more clearly elucidating the synergistic specificity of this nanosystem. These controls will be incorporated into our future studies to better distinguish the specific contributions of MnO2 nanozyme activity from the regenerative signals derived from hESC-Exos. The long-term storage stability of hESC-Exos@MnO2 at 4 °C or −20 °C also remains to be investigated. Future investigations should systematically assess changes in particle size, zeta potential, exosomal protein integrity, MnO2 coating stability, ROS-scavenging activity, and cellular uptake over a storage period of 1–3 months. In addition, the potential benefits of cryoprotectants or lyophilization for improving stability warrant further exploration. In addition, the in vitro ROS-scavenging and anti-inflammatory activities of hESC-Exos@MnO2 may help re-establish a more favorable intervertebral disc microenvironment, thereby contributing to attenuating degenerative progression and preserving disc structural integrity in vivo. Nevertheless, the exact quantitative relationship between ROS-scavenging efficiency and long-term functional restoration of the intervertebral disc remains to be further investigated. Finally, systematic evaluation in larger animal models (e.g., sheep, dogs, or pigs) is necessary to assess long-term efficacy, biosafety, potential immune responses, metabolic fate, and tissue compatibility, thereby providing stronger support for future clinical trials [83]. In summary, hESC-Exos@MnO2 represents a nanobiocomposite system that integrates delivery protection, microenvironmental regulation, and multidimensional repair, offering a translationally promising strategy for IVDD treatment. Its integrated design concept of “microenvironmental modulation–bioactive signal delivery” may also provide new insights for treating other degenerative diseases arising within complex pathological microenvironments, such as osteoarthritis and cartilage defects. From a translational medicine perspective, this multipathway cooperative regulatory mode is of particular clinical relevance. IVDD progression involves intertwined pathological processes, including oxidative stress, inflammation, cellular senescence, and metabolic imbalance of the ECM, making single-target interventions insufficient to achieve durable therapeutic benefits [58,84,85]. Owing to its pleiotropic properties, hESC-Exos@MnO2 can simultaneously act on multiple key nodes within this pathological network and may therefore produce more sustained and comprehensive therapeutic effects. The synergistic mode revealed in this study further highlights the unique advantage of exosome-based therapy in complex diseases. Compared with conventional single-gene or single-protein therapies, exosomes can exert more physiological and balanced regulation of multiple signaling pathways through their intrinsic repertoire of bioactive cargos, a property that more closely resembles endogenous repair processes and is supported by our experimental findings [86]. By combining this multitarget regulatory capacity of exosomes with the active protective function of nanomaterials, this study establishes a highly promising therapeutic paradigm. Conclusion In summary, this study developed hESC-Exos@MnO2 as an integrated nanotherapeutic system that combines delivery, responsiveness, and therapy. By leveraging the MnO2 nanoarmor to scavenge ROS, alleviate oxidative stress, and preserve exosomal bioactivity within the degenerative microenvironment while concurrently enabling hESC-Exos-mediated delivery of regenerative bioactive signals, this system achieves coordinated intervention across multiple key pathological processes of IVDD, including oxidative stress, inflammatory pyroptosis, cellular senescence, and metabolic imbalance of the ECM. Both in vitro and in vivo results demonstrate that hESC-Exos@MnO2 effectively preserves nucleus pulposus cell function and molecular homeostasis, promotes ECM anabolic remodeling, and improves disc structural integrity. This study not only provides a new integrated treatment strategy with great potential for IVDD, but also provides a new specific direction for regenerative medicine research of other degenerative diseases through the design concept of regulating the microenvironment of intervertebral disc and delivering regeneration-related signals. Materials and methods Ethics approval and consent to participate All experimental protocols involving patient samples were approved by the Ethics Committee of the First Affiliated Hospital of Shihezi University (Approval No. KJ2024-558-02). All animal experimental procedures were approved by the Bioethics Committee of Shihezi University (Approval No. A2024-378). The entire experimental design was conducted in accordance with the Declaration of Helsinki. Isolation and characterization of hESC-Exos Human embryonic stem cell line H9 was provided by the National Key Laboratory of Stem Cell at Sun Yat-sen University. hESCs were cultured in RP01020 medium (Nuwacell Biotechnologies) and harvested at ∼80% confluence. Cell culture supernatants were sequentially centrifuged at 300 × g for 10 min, 2000 × g for 15 min, and 10,000 × g for 30 min to remove cells and debris. Exosomes were then isolated using the automated EXODUS system (H-450, Huixin). Cells were resuspended in phosphate-buffered saline (PBS) (Wuhan Procell Biotechnology Co., Ltd). Exosomes were characterized by transmission electron microscopy (TEM), nanoparticle tracking analysis (NTA), and Western blot for exosomal markers CD9, CD63, TSG101, and negative controls Calnexin. Preparation and characterization of MnO2 nanosheets MnO2 nanosheets were prepared by a controlled redox reaction in an aqueous system [69]. First, 20 mL of TMA·OH (0.6 M) and 30% H2O2 were mixed. This mixture was then quickly added to 10 mL of MnCl2·4H2O solution (0.3M) within 15s. The solution was stirred strongly for 10 min. It was then stirred at 600 rpm overnight to allow full formation of MnO2 nanosheets. The product was collected by centrifugation. It was washed several times with deionized water. It was then freeze-dried to obtain a black MnO2 solid. To obtain uniform MnO2 nanosheets, 100 mg of the dried MnO2 solid was added to 10 mL of deionized water. The mixture was sonicated for 10h at 4 °C using a cell disruptor. The suspension was then centrifuged at 8000 rpm for 10 min. The purified MnO2 nanosheets were collected. The nanosheets were re-dispersed in deionized water. They were sonicated for 20 min in an ice bath to prepare the MnO2 dispersion for later use. Finally, the morphology of MnO2 nanosheets was examined by transmission electron microscope. Preparation and characterization of hESC-Exos@MnO2 The extracted hESC-Exos were mixed with the prepared MnO2 nanosheet dispersion at a mass ratio of 1:1. The mixture was gently stirred and incubated at room temperature for 24 h, allowing the MnO2 nanosheets to uniformly coat the surface of hESC-Exos through including coordination, hydrophobic and van der Waals forces interactions. After incubation, unbound hESC-Exos and MnO2 nanosheets were removed by centrifugation. The obtained hESC-Exos@MnO2 complexes were collected and re-dispersed in PBS to form a stable colloidal suspension. Subsequently, the concentration of hESC-Exos@MnO2 was quantified based on the protein content of the encapsulated hESC-Exos to ensure consistent hESC-Exos concentration for subsequent experiments. The morphology and physicochemical properties of hESC-Exos@MnO2 were characterized using transmission electron microscopy (TEM), elemental mapping analysis, and nanoparticle tracking analysis (NTA). The manganese content was determined by inductively coupled plasma optical emission spectrometry (ICP-OES) and converted into MnO2 content according to the molecular weight. The coating efficiency was calculated as the ratio of the measured MnO2 content to the theoretical input amount. hESC-exos clearance assay A 100 μmol/L TBHP solution (MS5603, Maokangbio) was separately incubated with hESC-Exos, hESC-Exos + MnO2, and hESC-Exos@MnO2 at 37 °C. For the blank control group, hESC-Exos were supplemented with an equal volume of PBS only. Incubation solutions were collected at 0, 2, 4, 6, 8, 10, and 12 h. Subsequently, RIPA lysis buffer (P0013B, Beyotime) containing 1% PMSF protease inhibitor (AR0162, Boster) was added to each sample, followed by lysis on ice for 30 min. The lysates were then centrifuged at 12,000 rpm for 15 min. Protein concentrations in the supernatants were quantitatively determined using a BCA Protein Assay Kit (AR1189, Boster). For BCA analysis, MnO2-containing blank controls without hESC-Exos were included, and the corresponding absorbance background was subtracted to minimize potential colorimetric interference from MnO2. All experiments were performed in triplicate at a minimum. Dissolved oxygen (DO) measurement The dissolved oxygen (DO) levels under different treatment conditions were measured using a Dissolved Oxygen Assay Kit (eBox, Guangdong Huankai Biotechnology Co., Ltd.) to evaluate the catalytic oxygen-generating capacity of MnO2 under oxidative stress conditions. Briefly, different treatment systems, including PBS, MnO2, TBHP, and MnO2 + TBHP, were separately added into centrifuge tubes and incubated at 37 °C for a designated period before measurement. Considering that TBHP itself may interfere with DO detection, PBS and TBHP control groups were included for background correction, and the influences of PBS and TBHP on the detection signal were excluded during data analysis. Subsequently, the assay was performed according to the manufacturer's instructions. Briefly, an appropriate volume of each sample was transferred into a BOD bottle, followed by sequential addition of the dissolved oxygen detection reagents. After thorough mixing and incubation, the dissolved oxygen content was determined by titration. The final DO concentration (mg/L) was calculated using the formula provided by the manufacturer. The results demonstrated that, after correction for the background effects of PBS and TBHP, the MnO2 + TBHP group exhibited a markedly increased dissolved oxygen level, indicating that MnO2 could catalytically decompose peroxides and promote oxygen generation under oxidative stress conditions. All experiments were performed in triplicate at a minimum. Cell culture and treatment To isolate human nucleus pulposus cells (NPCs), NP tissues were obtained from patients undergoing discectomy surgery (n = 10). The study subjects were recruited from patients diagnosed with lumbar disc herniation by magnetic resonance imaging (MRI) and requiring surgical treatment. Patients with Pfirrmann grades II–V intervertebral disc degeneration were included. The inclusion criteria were patients aged between 20 and 60 years with lumbar disc herniation, while patients younger than 20 years or older than 60 years were excluded. All specimens were collected with informed consent from the patients. The NP tissues were minced into small pieces and digested with 0.2% type II collagenase (Gibco) for 6 h to prepare NPC suspensions. The cell suspensions were subsequently centrifuged at 1200 rpm for 5 min and washed twice with PBS. The resulting cell pellets were re-suspended in a complete F12 culture medium (DMEM/F12, BaobioTM([300130], Hangzhou, China)) supplemented with 10% fetal bovine serum (FBS, Gibco) and 1% penicillin–streptomycin (Sigma-Aldrich). Cells were cultured at 37 °C in a humidified incubator containing 5% CO2. The culture medium was first replaced one week after initial seeding and subsequently changed every 3–4 days until the cells reached more than 95% confluence at the bottom of the culture flask. Cells were passaged at a ratio of 1:3 or 1:4. Second-passage NPCs were considered to be in a stable physiological state, and morphologically healthy NPCs were selected for subsequent experiments. All biological replicate experiments were performed using cells derived from different donors. Exosome Labeling and uptake assay Purified exosomes were labeled with the BeyoExo™ Exosome Labeling and Tracking Kit (PKH26) (C3637S, Beyotime) according to the manufacturer's instructions. The labeled exosomes were re-suspended in culture medium. The exosomes were incubated with NPCs at 37 °C for 24 h. The cells were then stained with FITC phalloidin (BL1188A, Biosharp) for 30 min to label the cytoskeleton. The nuclei were stained with Hoechst 33342 (C1022, Beyotime) for 10 min. Images were collected using a fluorescence microscope (Olympus). The mean red fluorescence intensity was measured with ImageJ (NIH) to assess exosome uptake. All experiments were performed in triplicate at a minimum. CCK-8 assay Cell proliferation was detected using CCK-8. Nucleus pulposus cells were planted in 96-well plates with 5000 cells per well and then grown with different amounts of hESC-Exos@MnO2 for 24, 48, or 72 h. At every time point we added new medium with 10% CCK-8 reagent (HIYCEZMBI0, Wuhan, China) and replaced old one. Incubated for 60 min at 37 degree C the plates. Then transferred 100 μL of the supernatant from each well to a new plate and measured absorbance at 450 nm using microplate reader. The cell viability was calculated using the following formula:Cell viability(%) = (OD samples-OD blank)/(OD control-OD blank)×100% Hemolysis assay This experiment was performed using erythrocytes isolated from Sprague–Dawley (SD) rats. Briefly, rats were properly restrained under anesthesia, and the hair over the precordial region was removed. The site of obvious cardiac pulsation was identified and marked by palpation. After disinfection, a 25-gauge needle connected to a syringe was inserted into the precordial region, allowing blood to be automatically aspirated into the syringe. Approximately 1 mL of blood was subsequently transferred into a heparinized collection tube and gently mixed to prevent coagulation. After erythrocyte isolation, the cells were suspended in PBS buffer and centrifuged at 3000 rpm for 3 min. The erythrocyte pellets were washed three times with normal saline and then re-suspended. Subsequently, hESC-Exos, MnO2, and hESC-Exos@MnO2 were added to the erythrocyte suspensions to achieve a final concentration of 50 μg/mL. Deionized water was used as the positive control, while the negative control group received no treatment. Considering that the intrinsic color of nanomaterials might interfere with absorbance measurements, corresponding concentrations of hESC-Exos, MnO2, and hESC-Exos@MnO2 solutions without erythrocytes were prepared as background controls. All samples were incubated at 37 °C for 1 h, followed by centrifugation at 200 × g for 15 min. Subsequently, 100 μL of the supernatant was collected, and the absorbance was measured at 540 nm. The hemolysis rate was calculated according to the following equation: Antioxidant activity assays In this study, the antioxidant capacities of hESC-Exos, MnO2, and hESC-Exos@MnO2 were evaluated at a concentration of 50 μg/mL. Total antioxidant capacity and free radical scavenging activities were assessed using DPPH and ABTS assays. Briefly, for the DPPH assay, samples were incubated with 0.15 mmol/L DPPH solution for 30 min, and the absorbance was recorded at 517 nm. For the ABTS assay, the ABTS+. radical was generated by reacting 7 mmol/L ABTS with 2.45 mmol/L potassium persulfate. The radical solution was diluted to an absorbance of 0.700 ± 0.02 at 734 nm before use. Samples were then incubated with the diluted ABTS + solution, and the absorbance was measured at 734 nm. An equal volume of distilled water was used as a blank control for both assays. Hydroxyl radical scavenging activity was assessed using a Fenton-like reaction inhibition assay, H2O2 (5 mmol/L) and FeSO4 (1.8 mmol/L) were incubated at room temperature for 10 min to generate ·OH. Then it was incubated with hESC-Exos, MnO2, or hESC-Exos@MnO2 for 15 min. Then it added methylene blue (1.8 mmol/L) and incubated for another 15 min at room temperature. Absorbance was measured at 665 nm. Additionally, antioxidant activity was determined according to the manufacturers’ instructions using commercial kits for total antioxidant capacity (S0121, Beyotime), DPPH scavenging (BC4750, Solarbio), superoxide anion scavenging (HYF5129F, Huayunbio), and hydroxyl radical scavenging (BC1325, Solarbio). The activities of catalase (CAT) and superoxide dismutase (SOD) were also measured using commercial kits (S0051 and S0101M, respectively; Beyotime). Western blot analysis Cells were pretreated with 100 μmol/L TBHP for 24 h, followed by co-culture with MnO2, hESC-Exos, or hESC-Exos@MnO2 (50 μg/mL) for 2 days, after which they were collected. Lysis protein of cells and exosomes by RIPA buffer (P0013B, Beyotime) with 1% PMSF protease inhibitor (AR0162, Boster) on ice for 30 min. Then it was centrifuged at 12, 000 rpm for 15 min, and we determined the protein concentration of the supernatant with a BCA assay kit (AR1189, Boster). We added 5× SDS-PAGE loading buffer (Boster) and denatured the 30 - 40 μg protein samples according to the manufacturer's instructions. Samples were separated by SDS-PAGE at a constant voltage of 120 V. The running time was adjusted according to protein size, at approximately 1 kDa per minute. After electrophoresis, proteins were transferred to 0.45 μm PVDF membrane (Millipore) for 60min at 300 mA. We blocked membranes for 1hr at RT with 5% BSA. After that the membranes were incubated overnight at 4 °C with the primary antibodies. Washed the membranes three times with Tris buffered saline with 0.1% Tween 20 (TBST), incubated with HRP-conjugated goat anti-rabbit or goat anti-mouse IgG secondary antibodies (SA00001-2/1, Proteintech) for 1.5 h at room temperature. Finally, we applied enhanced chemiluminescence (ECL) reagents (KF001, Affinity) and captured the protein bands using a ChemiDoc imaging system (12003154Hercules, Bio-Rad). All experiments were performed in triplicate. Specific details for all antibodies and dilution ratios are provided in Table S1 of the Supplementary Materials. Quantitative real-time PCR (RT-qPCR) NPCs were seeded in 6-well plates. The cells were first treated with 100 μmol/L TBHP for 24 h. The cells were then co-cultured for 2 days with MnO2, hESC-Exos, or hESC-Exos@MnO2 at a concentration of 50 μg/mL. There were five groups included in the study: Control, TBHP, MnO2, hESC-Exos, hESC-Exos@MnO2. After incubation, total RNA was isolated from cells using a cell/tissue RNA kit (R0027, Beyotime) based on the manufacturer's protocol. Use a spectrophotometer (Applied Biosystems) to measure RNA concentration. Add RNA into mixture HiScript III RT SuperMix (R323-01, Vazyme). cDNA synthesis on reverse transcription system (Applied Biosystems). cDNA was combined with gene-specific primers, SYBR Green dye (Q713-02, Vazyme) and DEPC-treated water (BL510A, Biosharp). The study measured gene expression through a real-time PCR system (Bio - Rad Laboratories). First denature at 95 °C for 30 s. Then amplification was run 40 cycles.Each cycle included annealing at 55 °C for 5 s and extension at 72 °C for 10 s. GAPDH was employed as the internal control, and Cq values were normalized to this reference. Relative gene expression was then quantified using the 2^−ΔΔCT method. Primer sequences are provided in Supplementary Table S2. Flow cytometry NPCs were seeded into T25 culture flasks and cultured for 3 days. Subsequently, oxidative stress was induced by treating the cells with 100 μmol/L TBHP for 24 h. The cells were then co-incubated with MnO2, hESC-Exos, or hESC-Exos@MnO2 at a final concentration of 50 μg/mL for an additional 2 days. After treatment, the cells were digested with trypsin (C0201, Beyotime) and collected by centrifugation. According to the manufacturers’ instructions, the cells were stained using JC-1 (M8650, Solarbio), MitoTracker Red CMXRos (M9940, Solarbio), ROS Detection Kit (E-BC-K138-F, Elabscience), and Cell Cycle Assay Kit (E-CK-A351, Elabscience), respectively. Following staining, the cells were washed three times with phosphate-buffered saline (PBS) and analyzed using a flow cytometer (FACSCalibur, BD Biosciences). All experiments were performed in triplicate at a minimum. Immunofluorescence analysis For immunofluorescence (IF) staining, NPCs were seeded onto sterile coverslips at a density of 1 × 106 cells. The cells were then pretreated with 100 μmol/L TBHP for 24 h, co-cultured with MnO2, hESC-Exos, or hESC-Exos@MnO2 (50 μg/mL) for 2 days, and finally fixed with 4% paraformaldehyde for 30 min. For immunostaining, cells were first permeabilized with 0.5% Triton X-100 for 20 min at room temperature. After rinsing three times with PBS, the samples were blocked using immunostaining blocking buffer (P0102, Beyotime) for 30 min at room temperature. Primary antibodies were diluted in antibody dilution buffer and incubated with the samples at 4 °C for 12–16 h. The following day the plates were washed 3 times on a horizontal shaker with PBST (PBS with 0.1% Tween - 20). We added species - matching secondary antibodies and incubated for 90 min at room temperature. And then we did 3 washes. To visualize cytoskeletons we used FITC-Phalloidin or TRITC-Phalloidin (BL1188A/BL1189A, Biosharp) and incubated slides in a dark and humidified chamber for 30 min. We stained the nuclei with DAPI (C1002, Beyotime) for 12 min in darkness finally. All the Figs were obtained with an Olympus confocal laser scanning microscope. To evaluate mitochondrial condition and oxidative stress, we applied JC -1 (M8650, Solarbio), ROS fluorescence kits (E − BC - K138 - F, Elabscience), or MitoSOX Red (M4830, Solarbio) to NPCs as directed by the producers. We counterstained all these samples with Hoechst 33342 (C1022, Beyotime) and stained them for 10 min in the dark. Washed 3 times with 0. 1% PBST for 5 min each time, took images by confocal microscopy. And also observe the occurrence of pyroptosis and its results through using a Calcein/PI cell viability and cytotoxicity kit (C1371S, Beyotimes). We processed and analyzed all fluorescence images by using the ImageJ software platform (NIH). All experiments were performed in triplicate at a minimum. Senescence-associated β-galactosidase and proliferation-related EdU staining The study analyzed treated cells using a cell senescence staining kit (C0602, Beyotime) and a cell proliferation assay kit (C0078S, Beyotime), following the manufacturer's instructions. The study performed SA-β-gal staining to assess cell senescence. The study performed EdU staining to assess cell proliferation. For SA-β-gal staining, the study washed treated cells three times with PBS. The study fixed the cells with 4% paraformaldehyde for 15 min. The study washed the cells three times again. The study added the working solution to the samples. The samples were incubated overnight at 37 °C in a constant-temperature incubator. The study captured and analyzed images using an Olympus light microscope. For EdU staining, the study added EdU working solution to the cell culture system. The cells were incubated at 37 °C for 12 h. The study fixed the cells with 4% paraformaldehyde. The study washed the cells and performed membrane permeabilization. The study added the Click reaction solution and incubated the cells for 30 min at room temperature in the dark. The study stained cell nuclei with Hoechst 33342 for 10 min. The study observed and captured images using an Olympus microscope. The study collected fluorescence images using a fluorescence microscope (Olympus). The study analyzed all data using ImageJ software.All experiments were performed in triplicate at a minimum. Caspase-1 activity assay The study measured caspase-1 activity using a commercial kit (C1101, Beyotime). The study seeded NPCs in 6-well plates and divided them into five groups: Control, TBHP (100 μmol/L), TBHP + MnO2 (50 μg/mL), TBHP + hESC-Exos (50 μg/mL), and TBHP + hESC-Exos@MnO2 (50 μg/mL). The cells were co-cultured for 2 days. The study then collected the cells. The study lysed the cells on ice according to the kit protocol. The study centrifuged the lysates at 18,000 g and 4 °C to collect the supernatant. The study measured protein concentration using a small aliquot. The study prepared pNA standards from 0 to 200 μM to generate a standard curve at 405 nm. During detection, the study mixed samples or lysates with assay buffer and Ac-YVAD-pNA substrate to a final volume of 100 μL. The samples were incubated at 37 °C for 90 min. The study measured absorbance at 405 nm. The study subtracted the blank value and calculated pNA production using the standard curve. The study calculated caspase-1 activity based on the definition that 1 U produces 1 nmol pNA per hour at 37 °C. The study normalized caspase-1 activity to protein content.All experiments were performed in triplicate at a minimum. Enzyme-linked immunosorbent assay (ELISA) The secretion levels of IL-1β and IL-18 in the culture supernatants of different treatment groups were determined using enzyme-linked immunosorbent assay (ELISA). NPCs were seeded into 6-well plates and subjected to the indicated treatments after cell attachment. For pharmacological inhibition experiments, the NLRP3 inhibitor MCC950 (10 μmol/L) and the Caspase-1 inhibitor VX-765 (20 μmol/L) were added 2 h prior to TBHP-induced pyroptosis stimulation. Subsequently, a pyroptosis model was established using TBHP treatment, followed by administration of hESC-Exos@MnO2 or other corresponding treatments according to the experimental grouping. After treatment, culture supernatants from each group were collected and centrifuged at 4 °C and 3000 rpm for 10 min to remove cell debris. The obtained supernatants were then subjected to subsequent analysis. The concentrations of IL-1β and IL-18 were measured using ELISA kits (E-EL-H0149 and E-EL-H0253, Elabscience) according to the manufacturer's instructions. Briefly, standards and samples were added to 96-well plates pre-coated with capture antibodies and incubated at 37 °C. After washing, biotinylated detection antibodies and HRP-conjugated secondary reagents were sequentially added. Subsequently, TMB substrate solution was added for color development in the dark, and the reaction was terminated using stop solution. The absorbance (OD value) was measured at 450 nm using a microplate reader, and the concentrations of IL-1β and IL-18 were calculated based on the corresponding standard curves. All experiments were performed in triplicate at a minimum. Lactate dehydrogenase (LDH) release assay The release level of lactate dehydrogenase (LDH) in the cell culture supernatants was determined using an LDH Assay Kit (E-BC-K771-M, Elabscience) to evaluate cell membrane damage and the extent of pyroptosis. Briefly, NPCs were seeded into 96-well plates and subjected to the indicated treatments after cell attachment. Following treatment, the culture supernatants from each group were collected. According to the manufacturer's instructions, the supernatants and LDH reaction working solution were added to 96-well plates and incubated at 37 °C in the dark for the specified duration.Subsequently, the absorbance (OD value) was measured at 450 nm using a microplate reader, with 600 nm used as the reference wavelength for background correction. The LDH release level was calculated as the ratio of the experimental group to the maximum enzyme release control group and expressed as a percentage. All experiments were performed in triplicate at a minimum. Alcian blue and safranin O staining The study treated cells in the same way as described above. The study fixed the cells with 4% paraformaldehyde for 30 min. The study then pretreated the cells with 3% acetic acid at room temperature to create an acidic environment. The study next added Alcian blue solution (C1101, Beyotime) and safranin O solution (C0621S, Beyotime). The study incubated Alcian blue with NPCs for 1 h. The study incubated safranin O with NPCs for 10 min. The study washed the cells three times with 3% acetic acid. The study then rinsed the cells three times with distilled water for 10 s each time. In this study, Olympus microscope was used to observe the staining results, and photos were taken. All the stained photos were analyzed by ImageJ software, and each experiment was repeated three times. RNA sequencing The study split NPCs into three groups: Control, TBHP, and TBHP + hESC-Exos@MnO2. After culture, use the TRIzol reagent (15596018CN, Invitrogen) to obtain the total RNA of the study. We evaluate RNA quality and integrity in the study. Studied the RNA. The study went ahead and made RNA sequencing libraries according to a typical method. Biotree (Shanghai, China) performed sequencing using the NovaSeq 6000 platform (Illumina). The study mapped raw reads to the human reference genome using default settings. The study analyzed differentially expressed genes using DESeq2 (version 1.38.3). The study defined significance as P < 0.05 and a fold change of at least 2. The study then performed functional enrichment analysis and KEGG pathway analysis. The study defined significant enrichment as P < 0.05. The RNA-seq datasets generated in this study are available in the Genome Sequence Archive (GSA) at the National Genomics Data Center (NGDC), under accession number HRA018797. Animal experiments Thirty 6-week-old male Sprague–Dawley rats (220 ± 5 g) (Hubei Biont Biological Technology Co., Ltd) were randomly assigned into five groups, with six animals in each group. The rats were anesthetized with 2% sodium pentobarbital at a dose of 40 mg/kg. An intervertebral disc degeneration (IVDD) model was established by injecting 10 μL of TBHP solution (100 mmol/L) into the Co7/8 intervertebral disc using a 33-gauge needle, whereas the control group received no treatment. Following surgery, the animals were maintained under standard conditions, and postoperative complications, including urinary retention and infection, were carefully monitored during the first postoperative week. Fourteen days after model establishment, 10 μL (50 μg/mL) of MnO2, hESC-Exos, or hESC-Exos@MnO2 was injected into the Co7/8 intervertebral discs of the MnO2, hESC-Exos, and hESC-Exos@MnO2 groups, respectively, while the IVDD group received an equal volume of PBS. Injections were administered once weekly. To minimize potential mechanical injury caused by repeated intradiscal injections, all injections were performed using a 33G ultrafine needle. Previous studies have demonstrated that repeated intradiscal administration using the same injection volume (10 μL) and 33G needle in rat IVDD models did not significantly affect disc height index (DHI), MRI signals, or histological morphology in PBS-treated control discs, indicating that the injection procedure itself did not introduce substantial mechanical confounding effects [30]. Furthermore, long-term observations in rat intervertebral disc models have shown that low-volume intradiscal injections did not induce significant structural degeneration, supporting the safety and feasibility of intradiscal injection strategies in IVDD research [87]. Therefore, in the present study, all IVDD-related experimental groups underwent identical injection procedures, including the same needle gauge, injection volume, and injection frequency, thereby minimizing injection-related bias in DHI, MRI, and histological analyses. After 14 days of treatment, the rats were deeply anesthetized with 3% sodium pentobarbital and euthanized, followed by radiological and histological analyses. Throughout the experimental period, all animals were housed in a specific pathogen-free (SPF) animal facility with free access to food and water under a standard light/dark cycle. Acute toxicity test The rats with established IVDD was treated with PBS, MnO2, hESC-Exos, or hESC-Exos@MnO2. Following a 2-week treatment period, the rats were anesthetized with 3% sodium pentobarbital, and serum samples were collected for subsequent biochemical analyses to evaluate liver and kidney function. The animals were then euthanized, and major organs, including the heart, liver, spleen, lungs, and kidneys, were harvested for histopathological staining to assess the biocompatibility of the different treatments. In vivo animal imaging According to the research design, hESC-Exos and hESC-Exos@MnO2 were labeled as PKH26 in this study. In the experiment, rats were anesthetized and made to lie on their back. In order to perform imaging operation, the tail of animals was straightened. FX PRO imaging system produced by Germany Brooke Company is used in the experiment, which uses 550 nm excitation wavelength and 600 nm emission wavelength. In-vivo imaging was carried out at different time points to measure the fluorescence intensity of each intervertebral disc. With the help of ImageJ software, all fluorescence signals were analyzed, and each experiment was repeated three times. Imaging evaluation Intervertebral disc degeneration in rat tails was evaluated using X-ray, micro-computed tomography (micro-CT), and magnetic resonance imaging (MRI) (Wuhan Center for Magnetic Resonance, Innovation Academy for Precision Measurement Science and Technology, Chinese Academy of Sciences). Two weeks after treatment, the rats were anesthetized with 3% sodium pentobarbital and euthanized, and tail tissues were harvested for further analysis. Radiographic assessment of rat tails was conducted on a DRX Ascend X-ray system (Carestream, Canada) operated at 50 kV and 160 mA, with an exposure time of 0.06 s and a source-to-detector distance of 100 cm. Micro-CT imaging was carried out using a SkyScan 1176 scanner (Bruker, Germany) at 60 kV and 100 mA, with an isotropic voxel size of 18μm. CT-Vox software (Bruker, Germany) was subsequently employed for three-dimensional reconstruction of coccygeal intervertebral discs. For MRI, T2-weighted images were acquired on a 4.7 T animal-dedicated scanner using a fast spin-echo sequence (TR/TE = 3000/70 ms), with 0.5-mm contiguous slices (0 mm gap). The disc height index (DHI) was calculated from X-ray images, and Pfirrmann grades were determined based on MRI images. Histological analysis After imaging analysis, the study washed rat tail intervertebral disc (IVD) samples twice with PBS. The study fixed the samples in 4% paraformaldehyde for two days. The study then decalcified the samples, except for the heart, liver, spleen, and kidney, using 10% EDTA. The study dehydrated the samples and embedded them in paraffin. The study cut the samples into 5 μm sections. The study then performed H&E staining, Masson staining, and Safranin O–fast green staining (Biossci biotechnology, Wuhan, China). These stains were used to evaluate the degree of IVD degeneration. Immunohistochemical (IHC) staining The study dewaxed paraffin sections of rat tail intervertebral discs with xylene. The study rehydrated the sections through graded ethanol solutions. The study then performed antigen retrieval using citrate buffer. The study blocked non-specific binding sites with normal goat serum. The study incubated the sections with primary antibodies at 4 °C for 12–16 h. The study then incubated the sections with secondary antibodies and developed the signal using DAB. Finally, the study observed the sections under a microscope and performed histological analysis. Statistical analysis All experiments clearly defined the meaning of “n” by distinguishing between two types of replicates: biological replicates and technical replicates. Biological replicates refer to independent experiments performed using NPCs derived from different donors or independent animals, whereas technical replicates refer to repeated measurements performed within the same experiment. All biological experiments were performed with at least three independent biological replicates, while non-biological experiments were conducted with at least three independent technical replicates.In addition, the inclusion and exclusion criteria for animal experiments have been clearly specified. No animals were excluded during the study. All experimental data are presented as the mean ± standard deviation (mean ± SD). Statistical analyses were performed using Origin software (version 2024) or GraphPad Prism software (version 10.0). Comparisons between two groups were analyzed using Student's t-test. For comparisons among multiple groups, one-way analysis of variance (one-way ANOVA) was performed, followed by Tukey's multiple comparisons test for pairwise group comparisons. Statistical significance was defined as follows: P > 0.05 indicated no significant difference (ns), and P < 0.05 indicated a significant difference. Statistical significance was denoted by asterisks as follows: P < 0.05(*), P < 0.01(**), P < 0.001(***), P < 0.0001(****). CRediT authorship contribution statement Honglei Li: Data curation, Formal analysis, Investigation, Writing – original draft, Writing – review & editing. Hanlin Mao: Supervision, Visualization. Runze Mao: Methodology, Project administration. Qingsheng Ma: Conceptualization, Data curation. Qi Tan: Resources, Software. Luoyuan Liu: Software. Hui Ma: Resources, Supervision. Zhi Ruan: Conceptualization, Funding acquisition, Investigation, Supervision. Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.