Abstract Achieving effective drug delivery and therapeutic efficacy poses significant challenges in intervertebral disc degeneration (IDD). Here, we developed a dual-pathological cascade delivery system utilizing therapeutic mesenchymal stem cell-derived apoptotic vesicles (ApoVs). These vesicles are engineered with MMP13-responsive cell-penetrating peptides (MR-ApoVs) for targeted modulation of senescence. A reactive oxygen species (ROS)-responsive hydrogel incorporating CD44 aptamers (Apt-Gel) was developed to provide high-affinity retention and spatiotemporal controlled release of MR-ApoVs. In this system, MR-ApoV release is first triggered by hydrogel degradation in response to elevated ROS levels. Subsequently, the MMP13-responsive peptides on MR-ApoVs are activated to enhance their internalization into senescent nucleus pulposus (NP) cells, thereby achieving a sequential response to pathological signals within the degenerative disc microenvironment. In a rat model of IDD, MR-ApoV@Apt-Gel effectively attenuated NP cell senescence, restored extracellular matrix homeostasis, preserved disc hydration, and maintained intervertebral disc height. This dual-pathological cascade-responsive strategy represents a promising therapeutic approach for IDD treatment. Keywords: Apoptotic vesicles, Aptamer, Hydrogel, Cellular senescence, Intervertebral disc degeneration Graphical abstract [47]Image 1 [48]Open in a new tab 1. Introduction Intervertebral disc degeneration (IDD) is a leading contributor to lower back pain and spinal dysfunction [[49]1]. With the rapidly aging global population, the healthcare burden of IDD continues to escalate [[50]2]. However, current treatments remain largely palliative, focusing on symptom control rather than addressing the underlying pathology. Oxidative stress plays a central role in the pathogenesis of IDD by disrupting redox homeostasis and initiating a cascade of degenerative processes [[51]3]. Elevated levels of reactive oxygen species induce DNA damage, mitochondrial dysfunction, and endoplasmic reticulum stress, ultimately leading to cellular senescence and tissue degeneration [[52][4], [53][5], [54][6]]. Recent studies have highlighted that inhibiting or reversing cellular senescence, as well as selectively eliminating senescent cells, represents a promising therapeutic approach for IDD [[55][7], [56][8], [57][9]]. Senescent cells adopt a senescence-associated secretory phenotype (SASP), characterized by the secretion of pro-inflammatory cytokines and matrix-degrading enzymes [[58]10,[59]11]. Among these factors, matrix metalloproteinase-13 (MMP13) contributes to the degradation and degeneration of the disc matrix [[60]12]. Importantly, MMP13 expression is directly upregulated by ROS via activation of redox-sensitive signaling pathways such as NF-κB and MAPK [[61]13,[62]14], and is also a key component of the SASP profile. These insights underscore the pathological synergy between ROS and MMP13 in driving IDD progression and highlight the therapeutic potential of simultaneously targeting both factors. In this context, mesenchymal stem cell-derived apoptotic vesicles are emerging as promising therapeutic candidates in regenerative medicine. These nanoscale vesicles retain functional biomolecules (e.g., miRNAs, proteins) from their parent cells, while avoiding the safety and regulatory concerns associated with live cell therapies [[63]15,[64]16]. Notably, ApoVs leverage evolutionarily conserved apoptosis-regeneration signaling pathways to promote tissue repair and mitigate age-related degeneration [[65]17,[66]18]. Recent advances in orthopedic research have demonstrated the therapeutic potential of ApoVs derived from various cell types. For example, macrophage- and lymphocyte-derived ApoVs have demonstrated efficacy in alleviating arthritis through immunomodulation [[67]19,[68]20], while MSC-derived ApoVs have been reported to facilitate osteochondral repair, ameliorate osteoporosis, and enhance bone regeneration [[69][21], [70][22], [71][23]]. Despite these findings, it remains unclear whether MSC-ApoVs can modulate nucleus pulposus (NP) cell senescence and disrupt the SASP-driven feedback loop in IDD. The unique avascular and high-pressure environment of the intervertebral disc presents significant barriers to therapeutic delivery [[72]24]. Conventional extracellular vesicle injection is limited by poor penetration, rapid degradation, and a lack of responsiveness to the pathological microenvironment [[73]25]. To overcome these challenges, we engineered a cascade-responsive delivery system guided by IDD's dual pathological cues: elevated reactive oxygen species (ROS) levels [[74]3] and MMP13 overexpression [[75]12]. Specifically, we functionalized ApoVs with MMP13-responsive cell-penetrating peptides, which facilitating enhanced penetration into senescent NP cells via an enzyme-activated mechanism. For spatiotemporal release control, MR-ApoVs were encapsulated within a ROS-sensitive polyvinyl alcohol (PVA) hydrogel modified with CD44-specific DNA aptamers. These aptamers enable high-affinity vesicle retention and modulate release kinetics through a reaction–diffusion process [[76]26,[77]27]. The resulting aptamer-conjugated hydrogel (Apt-Gel) enables ROS-triggered vesicle release in response to the oxidative microenvironment of the degenerative disc. This cascade delivery system integrates pathological targeting with on-demand release, representing a smart therapeutic platform for the precision treatment of IDD ([78]Scheme 1). Scheme 1. [79]Scheme 1 [80]Open in a new tab Schematic illustration of the dual-pathological cascade-responsive delivery system for intervertebral disc degeneration therapy. (a) Therapeutic apoptotic vesicles (ApoVs) derived from mesenchymal stem cells were functionalized with MMP13-responsive cell-penetrating peptides (MR-ApoVs) to enable enzyme-triggered targeting of senescent NP cells. (b) MR-ApoVs were embedded in a ROS-sensitive PVA hydrogel modified with CD44-specific DNA aptamers (Apt-Gel), which ensured high-affinity retention and on-demand release under oxidative stress. This dual-pathological cascade system responded sequentially to elevated ROS and MMP13 levels in the degenerative disc microenvironment. (c) In vivo, MR-ApoV@Apt-Gel treatment reduced NP cell senescence, promoted ECM restoration, and preserved disc structure, demon-strating the potential of this cascade-responsive system for effective IDD therapy. 2. Results 2.1. Engineering of MSC-derived ApoVs for targeted delivery to senescent NP cells [81]Fig. 1a presents a schematic overview of the isolation of ApoVs from MSCs through differential centrifugation. Apoptosis was induced in MSCs by treatment with 0.5 μM staurosporine (STS), resulting in hallmark morphological changes including cell shrinkage and membrane blebbing, as observed under optical microscopy ([82]Fig. 1b). To confirm apoptosis, Annexin-V staining was performed, capitalizing on the externalization of phosphatidylserine (PtdSer), a key apoptotic marker [[83]28]. Both immunofluorescence imaging ([84]Fig. 1c) and flow cytometry ([85]Fig. 1d) demonstrated a high proportion of apoptotic cells. Fig. 1. [86]Fig. 1 [87]Open in a new tab Preparation, characterization, and modification of MSC-derived apoptotic vesicles. (a) Schematic overview of ApoV isolation from MSCs via differential centrifugation. (b) Optical microscopy images showing morphology of apoptotic MSCs. (c,d) Confirmation of apoptosis using Annexin-V staining via (c) immunofluorescence imaging and (d) flow cytometric analysis. Scale bar: 50um. (e) Characteristic staining of Annexin-V and complement protein C1q of ApoVs. Scale bar: 20um. (f) Transmission electron microscopy (TEM) image of ApoVs. Scale bar: 200um. (g, h) Dynamic light scattering (DLS) and Zeta potential analysis of ApoVs. (i) Western blot analysis of caspase-3, Hsp70, CD90, CD73 and CD44 in MSCs, Apoptotic MSCs and ApoVs. (j) Confirmation of MMP13-mediated cleavage of peptides (n = 3). (k–m) Characterization of MMP13-responsive ApoVs (MR-ApoVs) showing no significant changes in (k) morphology (TEM), Scale bar: 200um. (l) size distribution (DLS), or (m) surface potential after modification. (n) Flow cytometry analysis of vesicle uptake under different treatment conditions. (o) Mean fluorescence intensity (MFI) is quantified (n = 3). (p) Percentage of fluorescence-positive NP cells (n = 3). All data are presented as mean ± SD. Statistical analysis was determined by one-way ANOVA test followed by Tukey's multiple comparison analysis (j, o, p). ∗p < 0.05; ∗∗p < 0.01, ∗∗∗p < 0.001, ns indicates no significance. Following induction, ApoVs were collected through stepwise differential centrifugation and resuspended in PBS. Immunofluorescence staining confirmed the expression of Annexin-V and complement protein C1q on the vesicles ([88]Fig. 1e). Transmission electron microscopy (TEM) revealed that ApoVs possessed a spherical to oval morphology ([89]Fig. 1f), while dynamic light scattering (DLS) analysis indicated an average hydrodynamic diameter of 218.9 nm ([90]Fig. 1g). Zeta potential measurements showed a moderately negative surface charge, with an average potential of −11.9 mV ([91]Fig. 1h). The stability of the apoptotic vesicles under physiologically conditions was also confirmed ([92]Fig. S1). To further verify the apoptotic characteristics and MSC origin of ApoVs, we conducted Western blot analysis. The results showed elevated caspase-3 levels in both apoptotic MSCs (Apo-MSCs) and their derived ApoVs ([93]Fig. 1i), validating their apoptotic origin. We also observed increased expression of HSP70, a stress-induced protein commonly upregulated during apoptosis, which provides supplementary evidence of apoptotic processing in both Apo-MSCs and ApoVs. To confirm the mesenchymal origin of the vesicles, we assessed the expression of canonical MSC surface markers. Western blot analysis demonstrated the presence of CD90 and CD73 in both parental MSCs and ApoVs ([94]Fig. 1i). Additionally, ApoVs retained expression of CD44 a transmembrane glycoprotein involved in cell adhesion and homing via interactions with the ECM [[95]29]. To enhance the targeting specificity of ApoVs, vesicles were functionalized with MMP13-responsive cell-penetrating peptides. An α-helical sequence mimicking apolipoprotein facilitated strong lipid-binding to ApoVs. Given the upregulation of MMP13 as a component of the SASP, MR-ApoVs were expected to exhibit selective affinity for senescent cells. The cleavage of peptides by MMP13 was confirmed ([96]Fig. 1j). Characterization of MR-ApoVs by TEM and DLS showed that surface modification did not significantly alter vesicle morphology ([97]Fig. 1k), size distribution ([98]Fig. 1l), or zeta potential ([99]Fig. 1m). To evaluate targeting efficiency, fluorescently labeled ApoVs and MR-ApoVs were incubated with senescent NP cells. Confocal imaging revealed stronger fluorescence in the MR-ApoV group, indicating enhanced targeting ([100]Fig. S2). Flow cytometry further confirmed significantly higher fluorescence intensity with MR-ApoVs compared to unmodified ApoVs ([101]Fig. 1n–p). However, MMP13 knockdown markedly reduced MR-ApoV uptake ([102]Fig. 1n–p, S3), suggesting an MMP13-dependent targeting mechanism. 2.2. Synthesis and characterization of CD44 aptamer-functionalized PVA hydrogel (Apt-Gel) The unique anatomical and biochemical environment of the intervertebral disc presents considerable challenges for the effective delivery of ApoVs, including rapid clearance, enzymatic degradation, and limited tissue retention [[103]24]. To overcome these barriers, we engineered a PVA-based hydrogel with affinity for ApoVs by exploiting aptamer-ligand interactions. This strategy enables localized, sustained release of ApoVs within the IVD, enhancing therapeutic retention and bioactivity. CD44, a transmembrane glycoprotein highly expressed on MSC, facilitates cell adhesion and homing through interaction with extracellular matrix components [[104]29]. Notably, MSC-derived ApoVs retain surface CD44 expression, presenting a targetable feature for enhancing hydrogel-vesicle interactions. To harness this, we exploited a CD44-specific aptamer [[105]30] and grafted it onto PVA via an acetal reaction using 5′-aldehyde-modified aptamers, yielding the CD44 Apt-PVA conjugate ([106]Fig. 2a). To verify successful conjugation, we performed a series of characterization experiments, using aptamer and PVA before or after reaction. Fluorescence visualization using DNA-specific dye GelGreen revealed aptamer retention in ultrafiltration tubes due to covalent binding with PVA (right tube), confirming successful conjugation ([107]Fig. 2b). The aptamer grafting efficiency was determined by comparing fluorescence intensity before and after reaction, yielding a high conjugation rate of 91.3 %. Ultraviolet–visible (UV–vis) spectroscopy further confirmed the presence of the aptamer by its characteristic absorption peak ([108]Fig. 2c), while agarose gel electrophoresis showed reduced migration of Apt-PVA compared to free aptamers, reflecting increased molecular weight due to conjugation ([109]Fig. 2d). Fig. 2. [110]Fig. 2 [111]Open in a new tab Development and Characterization of ApoV-Affinitive PVA Hydrogel (Apt-Gel). (a) Schematic illustration of the synthetic route for CD44 Apt-PVA and Apt-Gel, along with the ROS-triggered hydrolysis mechanism of Apt-Gel. (b) Ultrafiltration results of Aptamer and PVA with (right) and without (left) reaction, visualized using Gelgreen staining. (c) UV–visible absorption spectra of PVA and Apt-PVA. (d) Electrophoretic analysis of Aptamer, PVA, and Apt-PVA on an agarose gel, demonstrating reduced migration of Apt-PVA compared to free aptamers. (e) ^1H NMR spectrum of TSPBA, the crosslinker. (f) Gelation process observed upon mixing Apt-PVA and TSPBA solutions, leading to the formation of Apt-Gel. (g) FTIR spectra of Apt-PVA and Apt-Gel. (h) Representative SEM images of Apt-Gel, showing a well-defined porous microstructure. Scale bar: 10 μm. (i) Rheological measurements showing the storage modulus (G′) surpassing the loss modulus (G″) upon gelation. (j) Self-healing properties of Apt-Gel after shear-induced disruption. (k) 3D confocal microscopy images of Live/dead staining of NP cells encapsulated in Apt-Gel, with corresponding z-axis maximum projection images. Left scale bar: 400 μm; right scale bar: 250 μm. (l) Cytotoxicity assessment of Apt-Gel using the CCK-8 assay (n = 3). All data are presented as mean ± SD. Statistical analysis was determined by two-tailed Student's t-test (l). ∗p < 0.05; ∗∗p < 0.01, ∗∗∗p < 0.001, ns indicates no significance. To induce hydrogel formation, we synthesized a phenylboronic acid (PBA)-based crosslinker, N1-(4-boronobenzyl)-N3-(4-boronophenyl)-N1,N1,N3,N3-tetramethylpropane- 1,3-diaminium (TSPBA). Diol groups on PVA reacted with PBA groups in TSPBA to form phenylboronic ester linkages, producing a ROS-responsive hydrogel network. This network remains stable under physiological conditions but undergoes degradation in response to elevated ROS levels, thereby triggering ApoVs release. Additionally, the PBA groups act as ROS scavengers, providing intrinsic antioxidative effects. The chemical structure of TSPBA was validated by ^1H NMR spectroscopy ([112]Fig. 2e). Upon mixing Apt-PVA with TSPBA, rapid gelation was observed to form Apt-Gel ([113]Fig. 2f). FTIR spectroscopy showed new peaks at 1310 cm^−1 and 1430 cm^−1, consistent with the formation of phenylboronic ester bonds ([114]Fig. 2g). Scanning electron microscopy (SEM) analysis revealed a well-defined porous microarchitecture in the freeze-dried Apt-Gel, facilitating distribution and diffusion of ApoVs ([115]Fig. 2h). Rheological measurements demonstrated successful hydrogelation, with the storage modulus (G′) exceeding the loss modulus (G″) upon mixing ([116]Fig. 2i). Moreover, Apt-Gel exhibited excellent self-healing capability, with G′ recovering rapidly after mechanical disruption, suggesting its resilience to the biomechanical stresses present in the IVD environment ([117]Fig. 2j). Biocompatibility of Apt-Gel was evaluated using live/dead staining of encapsulated NP cells, which maintained high viability ([118]Fig. 2k). Additionally, CCK-8 assays confirmed that Apt-Gel extracts did not impair cell proliferation, supporting its cytocompatibility and safety for in vivo applications ([119]Fig. 2l). Together, these findings demonstrate that Apt-Gel as a promising injectable platform for the targeted and responsive delivery of ApoVs within the intervertebral disc, potentially enhancing regenerative outcomes in disc degeneration. 2.3. Apt-Gel enables ApoVs affinity and ROS-triggered release We systematically investigated the ROS-responsive behavior and controlled release characteristics of the Apt-Gel under varying oxidative conditions. Hydrogel degradation was assessed at 3, 7, and 14 days ([120]Fig. 3a). In both low (0.25 mM) and high (1 mM) H[2]O[2] environments, Apt-Gel exhibited accelerated degradation compared to the PBS control. Quantitative analysis of hydrogel weight loss showed that Apt-Gel retained 86 % of its original weight in PBS at 14 days, indicating high stability under physiological conditions ([121]Fig. 3c). In contrast, weight loss increased to 53 % in 0.25 mM H[2]O[2] and further to 75 % in 1 mM H[2]O[2]. To further characterize the hydrogel's ROS-responsiveness, degradation profiles were fitted to a first-order kinetic model. The resulting daily degradation rate curves and apparent degradation rate constants (K[obs]) demonstrated a dose-dependent pattern in response to increasing H[2]O[2] concentrations ([122]Fig. S4), confirming the controllable and responsive degradation behavior of Apt-Gel under oxidative stress. Additionally, SEM was employed to evaluate microstructural changes following ROS exposure ([123]Fig. 3b). Hydrogels exposed to 1 mM H[2]O[2] exhibited pronounced disruption of their porous network structure, consistent with the observed macroscopic degradation. These results collectively highlight the ROS-sensitive and controllable degradation profile of Apt-Gel, supporting its suitability for spatiotemporal drug delivery in pathological oxidative microenvironments. Fig. 3. [124]Fig. 3 [125]Open in a new tab Functional evaluation of Apt-Gel. (a) Photographs displaying the degradation process of Apt-Gel under different conditions (PBS, 0.25 mM H[2]O[2] and 1 mM H[2]O[2]). (b) SEM images of Apt-Gel after one day of exposure to 1 mM H[2]O[2]. Scale bar: 50um. (c) Quantitative assessment of hydrogel weight loss over 14 days under different conditions (n = 3). (d) ROS scavenging capacity of Apt-Gel assessed in a 1 mM H[2]O[2] solution (n = 3). (e) Intracellular ROS assessment of NP cells cultured with or without Apt-Gel by DCFH-DA staining. Scale bar: 100um. (f) Quantification of intracellular ROS by fluorescence intensity (n = 3). (g) Confocal microscopy images showing co-localization of FAM-labeled aptamers (green) and Dil-stained MR-ApoV (red) in Apt-Gel. Scale bar: 3um. (h) Co-localization analysis of aptamer and MR-ApoV signals. (i) MR-ApoV release kinetics from Gel and Apt-Gel in PBS or H[2]O[2] solution (n = 3). (j) Confocal imaging of residual MR-ApoV within Gel and Apt-Gel after 12 h in PBS or ROS conditions (n = 3). Scale bar: 10um. All data are presented as mean ± SD. Statistical analysis was determined by one-way ANOVA test followed by Tukey's multiple comparison analysis (f, j). ∗p < 0.05; ∗∗p < 0.01, ∗∗∗p < 0.001, ns indicates no significance. (For interpretation of the references to color in this figure legend, the