High-mobility network hydrogel microsphere system to combat chondrocyte senescence for enhanced cartilage repair and regeneration

Abstract

Cellular senescence plays a crucial role in the progression of various diseases, and targeting senescence is a potential therapeutic strategy for osteoarthritis (OA). However, the complex biomechanical environment surrounding chondrocytes significantly affects their senescence process. Currently, few biomaterials are available that have the ability to modulate stresses and counteract chondrocyte senescence. In this study, we used cationic liposomes as the core of the crosslinked structure of the hydrogel network through imine bonding to construct a high-mobility network hydrogel microsphere system (Res@Lipo@HMs). The deformability of liposomes endowed mobility to the crosslinked structure of the hydrogel network. This system not only enhanced joint lubrication through a rolling mechanism but also distributed mechanical stress on chondrocytes by increasing the elastic deformation capacity of the microspheres. Moreover, this approach delayed chondrocyte senescence, improved chondrocyte physiological function, and slowed down OA progression by enhancing mitochondrial function and inhibiting senescence pathways. This study offers new insights into antisenescence strategies for chondrocyte therapy.

Graphical abstract

1. Introduction

Cellular senescence (SnC) is a stable state of cell cycle arrest. Senescent cells are produced early in embryonic development and accumulate progressively with age [1]. This accumulation contributes to tissue degeneration and onset of age-related conditions such as osteoporosis, diabetes mellitus, and Alzheimer's disease [2]. Osteoarthritis (OA), a prevalent degenerative joint disease, is characterized by joint inflammation, cartilage degradation, and pain [3]. In patients with OA, senescent cells are present in multiple joint tissues, including the cartilage, subchondral bone, synovium, and infrapatellar fat pads [4], and exhibit common senescence-associated characteristics [[5], [6], [7]]; thus, SnC is likely to be associated with the onset and progression of OA. Therefore, targeting SnC has emerged as a promising strategy for the treatment of OA and other age-related diseases. Furthermore, significant advances have been made in senescence-targeted therapies, indicating their potential in various diseases [8,9]. However, the complex biomechanical environment of the cartilage limits the effectiveness of these therapies in patients with OA. To achieve optimal therapeutic outcomes, OA therapies must consider these biomechanical challenges. Mechanical stress plays a crucial role in cartilage homeostasis; excessive stress drives catabolic responses [[10], [11], [12]], and abnormal mechanical stress induces the expression of senescence markers (e.g., p16INK4A and p21) and senescence-associated secretory phenotype (SASP) in chondrocytes, thereby promoting SnC and accelerating OA progression [[13], [14], [15]]. In addition, mitochondrial dysfunction is a characteristic of chondrocyte senescence. Excessive mechanical stress disrupts mitochondrial homeostasis, increasing ROS production, inhibiting autophagy, and ultimately leading to mitochondrial dysfunction and apoptosis [16]. Consequently, the development of biomaterials capable of modulating cellular stress offers a promising strategy for targeting senescence in OA treatment.

Articular chondrocytes have a complex mechanical stress environment, comprising compression, shear, and hydrostatic pressure. Excessive shear and compressive stresses are the most common source of abnormalities [[17], [18], [19], [20]]. Although the body has natural joint lubricants, aging-related pathologies lead to a decrease in lubricant and increase in friction and wear [21,22]. Hydrophilic polymer–based hydrogel microspheres (HMs) are highly biocompatible and injectable micron-sized structures that can effectively improve intra-articular lubrication through surface lubrication and self-renewal of the lubrication layer [[23], [24], [25], [26], [27]]. In recent years, significant progress has been made in research on HMs and their applicability for treating OA. For example, Lin et al. [28] developed a charge-guided hydrogel microsphere system capable of penetrating deeply into the cartilage matrix, where it targets chondrocytes for drug delivery; this method has significantly improved the efficacy of OA treatments. In another study, Wang et al. [29] used charge-guided technology to develop HMs that could activate the mitochondria of chondrocytes, thereby enhance the activity of chondrocytes and promoting the healing of OA. In general, research into how HMs can more effectively treat OA involves paying attention to the regulation of various chondrocyte pathologies. Previous studies have laid a solid foundation for the development of antiaging HMs.

Although surface lubrication is a crucial property of HMs, research has focused on enhancing microsphere hardness to reduce deformation and improve favorable mechanical properties, aiming to mimic a ball-bearing model [30,31]. However, increasing microsphere hardness can alter compressive stresses along the vertical axis, inadvertently damaging chondrocytes. Rigid microspheres transmit compressive forces directly to chondrocytes with minimal deformation, concentrating the load at the contact points and generating high localized stress. This concentration increases the risk of cell damage by exerting high pressure on the underlying chondrocytes. Conversely, soft microspheres deform under stress, expanding the area of contact with chondrocytes, distributing the same load across a larger surface, and reducing localized stress and the risk of cellular injury. In addition, the elastic deformation of soft microspheres under load allows energy to be absorbed and dispersed, reducing the transmitted load and dispersing the mechanical damage to the surrounding tissues. Consequently, biomaterials that can serve as joint space lubricants while minimizing vertical stress through enhanced elastic deformation are required to optimize chondrocyte protection and slow down cellular aging.

In this study, we used cationic liposomes as the core of the crosslinked structure of the hydrogel network through imine bonding to construct a high-mobility network hydrogel microsphere system (Res@Lipo@HMs). In addition, the deformability of liposomes endowed mobility to the crosslinked structure of the hydrogel network. This system not only enhanced joint lubrication through a rolling mechanism but also effectively dispersed mechanical stresses on chondrocytes by improving the elastic deformation properties of the microspheres (Scheme 1). In brief, stearamine-loaded, resveratrol (Res)-enriched cationic liposomes were prepared using a film hydration technique, in which the liposomes, equipped with amine groups, form chemical bonds with oxidized hyaluronic acid (HA) methacrylate (HAMA) HMs through charge interactions. This approach significantly strengthened the mechanical properties of the microspheres and mitigated the effects of mechanical stress on chondrocyte senescence by promoting microsphere elasticity. In addition, the bearing-like lubrication properties and cartilage-targeting capability of this system allowed Res to precisely target chondrocytes, maintain mitochondrial function, and delay SnC. Overall, the developed antisenescence hydrogel microsphere system shows promise in reducing cartilage damage arising from mechanical stress and is a potential treatment for OA.

Scheme 1.

Schematic illustration of A) the fabrication process of resveratrol-loaded cationic liposomes, integration with oxidized HAMA matrix, and formation of microfluidic Res@Lipo@HMs, and B) the design of the Res@Lipo@HMs system aimed at inhibiting cellular senescence and treating osteoarthritis by enhancing joint lubrication and dispersing mechanical stress.

2. Materials and methods

2.1. Reagents and chemicals

Sodium hyaluronate (MW = 74 kDa; Fruida, China), sodium periodate (110,023, 5 g, TONG GUANG FINE CHEMICAL COMPANY, Beijing, China), octadecylamine (Aladdin, China), methacrylic anhydride (Aladdin, China), and Res (Aladdin, China) were used in this study. Routine reagents were obtained from Sigma-Aldrich. All chemicals were of analytical grade and used without any further purification.

2.2. Synthesis of HAMA

HAMA was synthesized in accordance with our previously reported method [28]. In brief, 5 g of HA (MW = 74 kDa) (Bloomage Freda Biopharm Co. Ltd., China) was completely dissolved in 250 mL of phosphate-buffered saline (PBS) at 4 °C. Then, 10 mL of 5 M NaOH (1 mL/min) and 5 mL of methacrylic anhydride (94 %, 0.5 mL/min) were added slowly, and the reaction mixture was stirred at 4 °C for 4 h. The reaction mixture was then transferred to a dialysis bag (MWCO 3500, Spectrum) and dialyzed against deionized water for 3 days. The final product was lyophilized and stored at 80 °C for long-term preservation. Oxidized HAMA was prepared as previously described. In brief, 1 g of HAMA was dissolved in 100 mL of deionized water and stirred until fully dissolved. Then, 5 mL of 0.5 M sodium periodate was added dropwise, and the mixture was stirred at room temperature in the dark at 250 rpm for 2 h. Next, 1 mL of ethylene glycol was added for 1 h to quench any unreacted periodate. The solution was then dialyzed against deionized water for 3 days and freeze-dried to obtain oxidized HAMA. The percentage of methacrylate in HAMA was determined using ¹H NMR (600 MHz, Bruker, Germany).

2.3. Preparation and characterization of liposomes

Res@Lipo liposomes were synthesized via thin-film dispersion. In particular, lecithin, cholesterol, octadecylamine, and Res (at a ratio of 40:10:4:6, w/w/w/w) were dissolved in a mixture of chloroform and methanol (9:1, v/v) and evaporated at 60 °C for 30 min to form a thin lipid film. The dried lipid film was then hydrated with deionized water and sonicated for 20 min to generate dispersed multilamellar vesicles. The liposome size was subsequently reduced via extrusion through 0.45- and 0.22-μm membrane filters (Millex, Ireland), with 10 repeated passes through each filter. The morphology of the resulting liposomes was analyzed using transmission electron microscopy (TEM) (Tecnai G2 Spirit Biotwin, USA). Then, a liposome suspension (∼0.5 wt%) was prepared, and approximately 1 mL of the suspension was analyzed using dynamic light scattering (DLS) (Zetasizer Nano S, Malvern, UK) to determine particle size distribution and zeta potential.

2.4. Fabrication and characterization of Lipo@HMs

HMs were fabricated using a capillary microfluidic device (inner diameter: 0.15 mm) as described previously. In particular, an aqueous solution containing 4 wt% HAMA and 0.5 wt% liposomes, along with a 0.5 wt% photoinitiator (Lithium Phenyl (2,4,6-trimethylbenzoyl) phosphinate (LAP), Aladdin, China), was used as the dispersed phase (flow rates: 5 μL/min), and 95 wt% paraffin oil and 5 wt% Span 80 were used as the continuous phase (flow rates: 800 μL/min). At the junction point, the continuous phase sheared and dispersed the aqueous phase, forming microdroplets, which were collected using a circulating refrigeration pump (temperature control: −40 °C). The microdroplets were frozen for 30 min and subjected to UV irradiation to induce crosslinking (wavelength: 365 nm; intensity: 48w, duration: 30min). The crosslinked microspheres were washed sequentially with acetone and deionized water and finally lyophilized for subsequent experiments. The morphology and size of Lipo@HMs were observed under a bright-field microscope (LSM800, ZEISS, Germany). The morphology of the lyophilized HMs and Lipo@HMs was observed under a scanning electron microscope (Sirion 200, FEI, USA).

2.5. Rheological characterization of hydrogels

The rheological properties of the hydrogels were evaluated using a rotational rheometer (DHR-2, TA Instruments, USA). Samples with a diameter of 10 mm and a height of 5 mm were carefully placed on the rheometer's measuring plate (25 mm in diameter; plate gap set to 5 mm). Dynamic strain sweep tests were conducted under the following conditions: test temperature, 25 °C; constant oscillatory frequency, 1 Hz; and oscillatory strain amplitude, 0.1 %–1000 %.

2.6. Mechanical characterization of hydrogels

Cylindrical specimens of hydrogel samples were fabricated with a diameter of 20 mm and a height of 6 mm and subjected to mechanical testing using a universal testing machine (Instron 68TM-R5569, USA). Both the compressive strain and cyclic loading responses were evaluated. The samples were mounted onto the testing platform and subjected to a 35 % strain at a loading rate of 2 mm/min, followed by five consecutive loading–unloading cycles. Subsequently, the samples were compressed at a constant rate of 1 mm/min until rupture. A minimum of three samples were tested for each experimental condition to ensure experimental reproducibility.

2.7. Drug release test

The release kinetics of Res were determined using a UV-5100 UV–Vis spectrophotometer (Metash, China). In brief, the Res@Lipos and Res@Lipo@HMs formulations were enclosed in dialysis bags (MWCO 1000 Da) and immersed in PBS (pH 7.4) at 37 °C under continuous agitation at 80 rpm until complete release had occurred. At predefined time intervals, aliquots of the release medium were collected for UV spectroscopic analysis, and an equal volume of fresh PBS was replenished to maintain the sink conditions.

2.8. Cell culture and treatments

Primary chondrocytes (Catalog No. ZQ0935) were obtained from Shanghai Zhongqiao Xinzhou Biotechnology Co., Ltd. Cells were maintained in Dulbecco's Modified Eagle Medium supplemented with 10 % fetal bovine serum and 1 % penicillin–streptomycin under controlled conditions (37 °C, 5 % CO₂). The culture medium was refreshed every 3 days. When the cells reached approximately 80 % confluence, they were detached using trypsin–EDTA, isolated, and passaged. Experiments were initiated when the chondrocytes reached passages 3–5 and a sufficient number of cells had been obtained.

2.9. Assessment of the cytotoxicity of HMs

The cytotoxicity of HMs against cultured chondrocytes was determined using Cell Counting Kit-8 (CCK-8; Beyotime, China) in accordance with the manufacturer's protocol. In particular, the CCK-8 solution was added to the culture medium, and after 2 h of incubation, absorbance at 450 nm was measured using a FlexStation 3 microplate reader (Molecular Devices, Japan).

2.10. Establishment of a mechanical stress cell model

Chondrocytes were first seeded into 12-well plates at a density of 1 × 10⁴ cells per well. Initially, culture was performed for 8 h under standard conditions; then, the medium was changed and culture was performed for an additional 16 h. Subsequently, a 4 wt% HAMA hydrogel layer was applied on the top of the chondrocytes and cured under UV light to mimic the extracellular matrix. To simulate mechanical stress similar to that experienced by chondrocytes in joint environments, a 100-g experimental weight (Nanjing Su Measurement and Measuring Instrument Co. Ltd., China) was placed on the top of the hydrogel layer. In the control group, the weight was not applied. In separate experimental groups, HMs and Lipo@HMs were added between the weight and hydrogel. Three parallel samples were examined for each group to ensure reproducibility. Meanwhile, the cell culture device was placed on a shaking table and shaken at a frequency of 30 times per minute to simulate the dynamic environment of the knee joint.

2.11. RNA-sequencing analysis

Total RNA was isolated using TRIzol Reagent (Invitrogen, USA), and its concentration, purity, and integrity were assessed using a NanoDrop spectrophotometer (Thermo Scientific, USA). To prepare an RNA library, 3 g of total RNA was used. First, mRNA was extracted from the total RNA using magnetic beads conjugated with poly-T oligonucleotides. Then, mRNA was fragmented under high temperatures using divalent cations in Illumina's proprietary fragmentation buffer. First-strand cDNA was synthesized using random hexamer primers and SuperScript II reverse transcriptase. Subsequently, DNA polymerase I and RNase H were used to generate second-strand cDNA. The remaining overhangs were converted into blunt ends through exonuclease and polymerase activities, and then the enzymes were removed. Next, adenylation of the 3′ ends of the DNA fragments was performed, and Illumina paired-end adapters were ligated to facilitate hybridization. The AMPure XP system (Beckman Colter, USA) was used to purify the library fragments of the desired length (400–500 bp). These adapter-ligated fragments were enriched through a 15-cycle polymerase chain reaction (PCR) using Illumina's PCR Primer Cocktail. The final library products were purified and quantified using an Agilent Bioanalyzer 2100 system (Agilent, USA) and High Sensitivity DNA Analysis Kit. The prepared libraries were then sequenced and analyzed on the Illumina NovaSeq 6000 platform (Illumina, USA).

2.12. Western blotting

Western blotting was used to quantify the protein expression levels of c-Fos, CEBPB, ERK, P-ERK, P38, and P-P38. In brief, total protein was extracted from chondrocytes and quantified. Electrophoresis was conducted following standard protocols. Heat-denatured protein samples were loaded into wells and separate via SDS-PAGE. Subsequently, the proteins were transferred to polyvinylidene difluoride (PVDF) membranes and subjected to immunoblotting. Signal development was performed according to standard procedures. GAPDH was used as a loading control to ensure equal protein loading across the samples. Images were processed and background-corrected using Adobe Photoshop, and the optical density of protein bands was analyzed using the Alpha Image Processing System.

2.13. RNA extraction and real-time quantitative PCR (RT-PCR)

After 48 h of incubation, chondrocytes from the control, HMs, and Lipo@HMs groups were subjected to RT-PCR to assess mRNA expression related to SnC and mitochondrial dysfunction. In brief, total RNA was extracted from chondrocytes and reverse transcribed using the RevertAid First-Strand cDNA Synthesis Kit (Thermo Fisher Scientific, USA). Subsequently, RT-PCR was conducted using an ABI 7300 Real-Time PCR System (Applied Biosystems, USA). All primer sequences are listed in Table S2. Relative mRNA expression was determined using the comparative Ct method (ΔΔCt), with all experiments performed in triplicate to ensure reliability.

2.14. Measurement of mitochondrial respiratory chain activity

The mitochondrial respiratory chain activity was assessed using the Seahorse XF Cell Mito Stress Test Kit (Agilent Technologies, USA). The protocol consisted of the following steps: (i) After the experimental treatment, chondrocytes were rinsed twice with Seahorse XF Assay Medium, followed by the addition of 500 μL of medium, and incubated in a CO₂ incubator for 1 h. (ii) According to the kit instructions, oligomycin (1 μM), carbonyl cyanide p-trifluoromethoxyphenylhydrazone (FCCP, 1 μM), and antimycin A (1 μM) were sequentially added. (iii) The Seahorse XF analyzer (XFe24, Agilent Technologies, USA) was programmed according to the operation manual and calibrated using the provided calibration plate. (iv) The calibration plate was replaced with the cell culture plate, and the instrument was run to collect data.

2.15. Flow cytometry analysis

The apoptosis of chondrocytes induced by different experimental treatments was analyzed using the Flow Cytometry Apoptosis Detection Kit (BD, USA). In brief, chondrocytes were first digested and collected with trypsin. Then, 300 μl of binding buffer and 15 μl of Annexin V-fluorescein isothiocyanate (FITC) were added following the manufacturer's instructions, and the cells were incubated at 25 °C in the dark for 15 min. Subsequently, 5 μl of propidium iodide was added, and the apoptotic chondrocytes were analyzed via flow cytometry (FACSCalibur, USA).

2.16. Construction of an OA model

All animal experiments were approved by the Animal Ethics Committee of the Second Affiliated Hospital of Zhejiang University School of Medicine (Approval no. 2024-226). Eight-week-old male Sprague Dawley (SD) rats were randomly assigned to either a sham-operated group (n = 6) or an OA group (n = 18). Rats in the OA group were anesthetized with 3 % pentobarbital sodium (40 mg/kg), and the fur around the knee joint was shaved with scissors. A small incision was made medial to the patellar tendon to access the joint capsule. The fat pad over the cranial horn of the medial meniscus was carefully removed using microscissors, and the medial meniscotibial ligament (MMTL) was transected without damaging the articular cartilage or other soft tissues. One week after surgery, the OA rats were further randomized into three subgroups (n = 6 per group) to receive intra-articular injections of 100 μL of PBS, HMs, or Res@Lipo@HMs (concentration of drug in microspheres: 100 μM; concentration of microspheres in PBS: 10 mg/mL). The injections were repeated at 4 weeks before sacrifice.

2.17. Histological and apoptosis evaluation

After the animal experiments, the rat knee joints were completely excised and decalcified for 2 months. The specimens were embedded in paraffin and sectioned into 5-μm-thick slices. The sections were then stained with H&E and Safranin O-fast green to evaluate the morphology and matrix of chondrocytes. The samples were observed under a bright-field microscope (Nikon, Japan), and images were acquired. The severity of OA in each specimen was analyzed based on the stained images according to the currently accepted Osteoarthritis Research Society International (OARSI) scoring system, and the evaluation was performed by an observer in a double-blind manner. The quantification of GAG was based on Safranin O-fast green staining. The assessment result of GAGs in the first slice of six control group samples was used as a reference, and the data from the other groups were normalized accordingly. The relative staining intensity was statistically analyzed. Paraffin-embedded sections were stained stepwise using the TUNEL staining kit according to the manufacturer's instructions. The sections were then observed under a fluorescence microscope (Nikon, Japan), and images were acquired. Image analysis was conducted using Image-Pro Plus 6.0 software (Media Cybernetics, USA) to quantify the number of positive cells in each section, determine the total cell count, and calculate the positive rate. Paraffin sections were also stained with rabbit anti-collagen II/MMP-13 antibodies (Servicebio, China), followed by staining with a secondary antibody and DAB system. The staining results were quantified using Image-Pro Plus 6.0 software.

2.18. Statistical analysis

All data were preprocessed using Excel (Microsoft, USA) and presented as mean ± standard deviation. Statistical differences among experimental results were evaluated using either t-test or one-way analysis of variance followed by Tukey's post hoc test. All experiments were designed to include at least three parallel and repeated samples. Statistical significance was set at P < 0.05 and indicated as ∗P < 0.05, ∗∗P < 0.01, and ∗∗∗P < 0.001. Differences with P > 0.05 were considered not significant (ns).

3. Results and discussion

3.1. Characterization and mechanical properties of the Res@Lipo@HMs hydrogel system

Liposomes, comprising an aqueous core suitable for hydrophilic drugs and a phospholipid bilayer for hydrophobic drugs, are versatile drug carriers that are suitable for a range of drugs owing to their unique structure [32]. The phospholipid bilayer provides an optimal platform for loading drugs, such as Res, to achieve targeted delivery [33]. Furthermore, liposomes regulate drug release and minimize adverse reactions. In general, liposomes loaded with lipid-soluble drugs are efficiently prepared using the film dispersion method, which achieves high encapsulation efficiency [34]. Cationic liposomes loaded with Res were prepared using the thin-film hydration method (Fig. S1). TEM revealed that the particle sizes of Lipo and Res@SA-Lipo were approximately 100 nm and that both exhibited a multilayered structure (Fig. 1A and B). DLS analysis confirmed the particle sizes of Lipo (Fig. 1C) and Res@SA-Lipo (Fig. 1D) as approximately 100 nm, with a polydispersity index of 0.262 and 0.270, respectively, indicating good dispersion. These results were consistent with TEM observations. The particle size of lipid nanoparticles is crucial for their functional performance; liposomes with an average particle size of less than 200 nm are preferentially absorbed by chondrocytes [35]. As shown in Fig. 1F, unmodified liposomes were negatively charged, whereas the zeta potential of Res@SA-Lipo was 25.27 ± 4.87 mV owing to stearylamine modifications that introduce positively charged amine groups. Given that chondrocyte glycosaminoglycans are negatively charged, cationic liposomes can effectively adsorb onto chondrocytes [[36], [37], [38]].

Fig. 1.

Characterization of cationic liposomes and hydrogels. A) TEM image of unloaded liposomes (Lipo). B) TEM image of resveratrol-loaded cationic liposomes (Res@SA-Lipo). C) Particle size distribution curve and polydispersity index (PDI) value of Lipo. D) Particle size distribution curve and polydispersity index (PDI) value of Res@SA-Lipo. E) Macroscopic images of hydrogels formulated with varying concentrations of liposomes (Lipo0: 0 wt%, Lipo1.25: 0.125 wt%, Lipo2.5: 0.25 wt%, Lipo5z: 0.5 wt%). F) Zeta potential of Lipo and Res@SA-Lipo. G) Compressive stress-strain curve of hydrogels under cyclic loading. H) Compressive stress-strain curve of hydrogels under continuous loading. I) Maximum compressive load of the different hydrogels. J) Compression modulus of hydrogels. K) Maximum compressive strain of hydrogels. (n = 3 for each group), (∗P < 0.05; NS: not significant).

HA is a primary component of the extracellular matrix (ECM) in articular cartilage and is widely applied in OA treatment [39]. HAMA is synthesized by modifying the HA with methacrylate groups, facilitating UV-induced crosslinking. HA effectively anchors positively charged liposomes within the hydrogel matrix through electric dipole interactions [40]. The mechanical strength and stability of hydrogels is significantly enhanced by the formation of imine bonds between the aldehyde groups in oxidized HAMA and the amine groups in liposomes [41]. When cationic liposomes were combined with HAMA hydrogels, the transparency of HAMA decreased as the liposome concentration increased (Fig. 1E). This was also observed in rehydrated lyophilized hydrogels (Fig. S2). Furthermore, to assess the long-term stability of liposomes in the hydrogel and provide a basis for subsequent applications, we immersed the hydrogel systems that had been stored for 2 weeks and 4 weeks to release the liposomes, and then measured the particle size and zeta potential of the liposomes. The results showed that there were no significant differences in either particle size or zeta potential at each time point. This indicates that the liposomes have good stability within the hydrogel (Fig. S3).

To assess the impact of additional liposomes on the mechanical properties of the hydrogels, Lip@HMs and HMs were analyzed using a universal testing machine and rheometer.The mechanical properties of Lip@HM hydrogels with varying liposome content (0–5 mg) were evaluated using a rheometer, and the results are illustrated in Fig. 1H. Rheological analysis revealed a positive correlation between the liposome concentration and storage modulus of the hydrogels, which increased from 1630.62 to 2407.01 Pa with an increase in liposome content. A storage modulus greater than the loss modulus indicates a predominantly elastic behavior in hydrogels, and Fig. 1H shows the increase in elasticity as the liposome content increased from 0 mg to 5 mg. This improvement is likely attributed to the imine bonds formed between oxidized HAMA and stearylamine, which introduce additional crosslinking points and enhance the hydrogel's mechanical strength and stability.

During repetitive joint movements, the articular cartilage undergoes repeated compression and deformation, transferring and absorbing loads and shocks to maintain normal joint function. Consequently, cartilage tissue engineering scaffolds should exhibit excellent cyclic compressive properties. In this study, the cyclic compressive properties of hydrogels with varying liposome concentrations were assessed using a universal testing machine under the cyclic compression mode (Fig. S4). Fig. 1G presents the cyclic stress–strain curves of the hydrogels, which were similar in all samples. The compressive strength of the hydrogels progressively decreased with an increase in the number of cycles, primarily due to the irreversible breakage of the covalent bonds within the hydrogel matrix under compressive forces, resulting in a reduction of the compressive properties. In addition, the hysteresis loop area of the stress response gradually diminished as the cycle count increased, indicating that the difference between the loading and unloading stress–strain curves will decrease and eventually converge. Studies have shown that living soft tissues (such as tendons, articular cartilage, and blood vessels) exhibit hysteresis loops under cyclic loading–unloading that stabilize after multiple cycles. In this study, the residual stress percentages of lipo0@HM, lipo1.25@HM, lipo2.5@HM, and lipo5@HM after five compression cycles were 80.398 %, 88.940 %, 89.318 %, and 90.605 %, respectively; hence, lipo5@HM exhibited the best cyclic compressive performance. Furthermore, under identical compressive stress, the strain exhibited a positive correlation with the liposome concentration, indicating that the incorporation of liposomes into HAMA hydrogels enhances cyclic compressive properties and elasticity, thereby meeting the requirements for cartilage tissue engineering.

The hydrogels underwent compressive strain testing using a universal testing machine. The incorporation of various concentrations of liposomes into HAMA hydrogels resulted in no statistically significant differences in the maximum load during compressive testing, indicating that the mechanical strength of the hydrogels was not altered by liposome incorporation. However, a statistically significant increase in the maximum compressive strain was observed as the liposome content increased (Fig. 1K), suggesting that the addition of liposomes enhanced the elastic deformation range of the hydrogel while preserving the network strength. Hydrogels with liposome concentrations greater than 0.25 wt% exhibited improved elastic deformation under equivalent stress. Hydrogels with higher liposome concentrations had a lower elastic modulus (Fig. 1J), suggesting that liposome incorporation reduces material rigidity and enhances strain capacity. Compared with traditional cartilage repair hydrogel materials (such as alginate hydrogel), our hydrogel system can ensure sufficient strength while also achieving excellent deformation ability. Under the same pressure, our hydrogel system can produce greater deformation, thereby better dispersing mechanical stress.

3.2. Characterization of Res@Lipo@HMs

Using microfluidic technology (Fig. S5), we fabricated HAMA microspheres (HMs) with an appropriate size for intra-articular injection. Excessively small diameters are not ideal for intra-articular diffusion [42]. In this study, a mixture of cationic liposomes containing Res and HAMA served as the aqueous phase; the oil phase flow rate was 800–1000 μL/min and the aqueous phase flow rate was 5–10 μL/min. This configuration produced HMs with an average diameter of approximately 100 μm (Fig. S6A). The freshly formed HMs were then placed in a constant temperature bath at 40 °C, facilitating ice crystal formation and endowing the microspheres with a porous structure [43]. Under UV irradiation, the hydrogel matrix was crosslinked and cured, immobilizing the liposomes within the pores of the hydrogel structure. Res@Lipo@HM were lyophilized and subsequently stored for future use (Fig. S6B). Optical microscopy images (Fig. 2A–C) demonstrated that Res@Lipo@HMs were well dispersed, maintained an intact morphology, and had an average diameter of approximately 100 μm. Field-emission scanning electron microscopy of Res@Lipo@HMs and HMs revealed an intact porous structure that is essential (Fig. 2B–D) for drug release [44]. At the same time, we examined the drug encapsulation efficiency as 0.79 ± 0.04 % for the liposomes and 0.56 ± 0.06 % for the Res@Lipo@HMs (Fig. S7). Subsequently, to further verify the successful binding of liposomes and HMs, we performed electron microscopy and energy dispersive spectroscopy on the Res@Lipo@HMs hydrogel. This analysis confirmed that the hydrogel was rich in P, which is not present in the HMs hydrogel (Fig. S8). It further demonstrated that the liposomes and hydrogels had successfully combined.

Fig. 2.

Characterization of Res@Lipo@HMs Hydrogel Microspheres. A) Bright-field images of Res@Lipo@HMs at low magnification, illustrating the overall spherical morphology. B) SEM images of Res@Lipo@HMs: (i) showing the overall morphology, and (ii) a magnified view highlighting the internal porous network. C) Bright-field image of a single Res@Lipo@HMs microsphere. D) SEM images of HMs: (i) overview of morphology, and (ii) a magnified view showing the internal structure and porosity of the hydrogel microsphere. E) Cumulative release profile of resveratrol from Res@Lipo@HMs over a period of 21 days, demonstrating the sustained release capability. F) Cell viability assay results (measured at OD 450 nm) for various concentrations of Res@Lipo@HMs. (n = 3 for each group) (ns: not significant).

Res exhibits low bioavailability owing to its poor water solubility and short elimination half-life [45]. The encapsulation of Res in liposomes can extend its in vivo retention time and enhance its cellular uptake. Furthermore, combining liposomes with HMs stabilized drug release, potentially enabling localized drug delivery. The cumulative release profile of Res from Res@Lipo@HMs is illustrated in Fig. 2E. The release pattern approximates a linear profile, indicating effective sustained-release properties. The release rate was 44.76 % within the first 5 days, followed by a steady and gradual release. The nanoparticle release experiment continued for 25 days, with a cumulative release of 86.41 % of the total drug dose. Overall, Res@Lipo@HMs demonstrated a prolonged release effect, suggesting their suitability as a carrier for the localized delivery of Res.

To evaluate the feasibility of using Res@Lipo@HMs for OA treatment, we investigated the in vitro cytotoxicity of Res@Lipo@HMs against chondrocytes. Chondrocytes were cultured for 48 h in the presence of hydrogel microsphere leachates at various concentrations (0, 500, 1000, 1500, and 2000 μg/mL). The cell viability results, assessed using the CCK-8 assay, are shown in Fig. 2F. No significant differences in cell viability were observed between the groups, demonstrating the favorable biocompatibility of Res@Lipo@HMs.

3.3. Mechanisms of mechanical stress–induced chondrocyte senescence

Compressive stress is transmitted to chondrocytes through HMs. Owing to their high rigidity, hard microspheres exhibit minimal deformation under load, concentrating stress at the contact point, which generates high localized stress and increases the risk of chondrocyte damage (Fig. 3A). In contrast, soft microspheres undergo greater deformation, enlarging the contact area, which allows for more even load distribution and reduces localized stress (Fig. 3B). In addition, soft microspheres absorb and dissipate part of the applied energy, thereby reducing the stress transmitted to the surrounding tissues and the risk of mechanical injury.

Fig. 3.

Differential Gene Expression and Enrichment Analysis of Chondrocytes in Response to Mechanical Stress. A) Schematic representation of stress transmission to chondrocytes under mechanical loading using rigid microspheres. B) Schematic representation of stress transmission to chondrocytes under mechanical loading using soft microspheres. C) Volcano plot of the distribution of differentially expressed genes (DEGs). D) Gene Ontology (GO) enrichment analysis of DEGs. E) KEGG enrichment analysis of DEGs. F) Reactome enrichment analysis of DEGs. G) Heatmap analysis showing expression patterns of genes related to cellular senescence. H) KEGG pathway analysis of DEGs involved in cellular senescence under mechanical stress. I) Heatmap illustrating gene expression changes related to mitochondrial dysfunction under mechanical stress conditions. J) GO enrichment analysis of genes associated with mitochondrial dysfunction under mechanical stress.

To further understand how mechanical stress influences chondrocyte senescence, we performed transcriptome sequencing to explore the underlying mechanisms. Compared with normal chondrocytes, chondrocytes under mechanical stress exhibited 660 significantly differentially expressed genes (DEGs), with 396 upregulated and 264 downregulated genes (P < 0.05, log2 [fold change] > 1.5) (Fig. 3C). GO enrichment analysis revealed that these DEGs were significantly associated with external stimulus response, cell death, and apoptotic process (Fig. 3D). KEGG pathway analysis revealed that these DEGs were significantly enriched in the TNF, IL-17, NF-kappa B, and MAPK signaling pathways (Fig. 3E), which aligns with the findings of a previous study [46]. Furthermore, reactome enrichment analysis revealed that these DEGs were associated with pathways involved in the cell cycle and SnC (Fig. 3F).

After analyzing the reactome enrichment results of RNA-seq (Table S1), we identified 10 target genes (5 upregulated and 5 downregulated) significantly associated with SnC (Fig. 3G). KEGG enrichment analysis of these target genes demonstrated that the TNF, IL-17, and MAPK signaling pathways played significant roles in mechanical stress–induced chondrocyte senescence (Fig. 3H). The key genes implicated in these pathways included c-Fos and C/EBPβ. c-Fos, a critical gene in the MAPK pathway, forms the AP-1 heterodimer with c-jun. During oxidative stress–induced senescence, an increase in ROS levels promotes ERK phosphorylation, which subsequently activates AP-1 transcriptional activity, promoting SnC [47]. ERK signaling plays a pivotal role in SnC [48]. C/EBPβ is another essential transcription factor involved in SASP regulation in response to senescence, and research has shown that C/EBPβ overexpression induces senescence [49]. Overall, these findings highlight that c-Fos and C/EBPβ are key genes that regulate chondrocyte senescence.

SnC is strongly associated with mitochondrial dysfunction [50]. Mitochondrial dysfunction and oxidative stress–induced dysfunction have also been observed in OA chondrocytes [51]. However, it remains unclear if mechanical stress contributes to mitochondrial dysfunction in the progression of chondrocyte senescence. To investigate the effects of mechanical stress on mitochondrial dysfunction in chondrocytes, we identified mitochondrial function–related genes (MRGs) that were differentially expressed, obtaining a set of 16 candidate mitochondrial dysfunction–associated genes altered under mechanical stress; of these 16 genes, 9 were upregulated and 7 were downregulated (Fig. 3I). GO enrichment analysis (Fig. 3J) revealed that these DEGs were closely associated with catalytic activity, electron transport chain, and cellular respiration and were specifically associated with ATP production, mitochondrial membrane potential, and mitochondrial respiratory chain. ATP production and mitochondrial membrane potential are critical markers for assessing mitochondrial function, which directly influences SnC [52].

3.4. Restoration of chondrocyte mitochondrial function by Res@Lipo@HMs

The disruption of mitochondrial homeostasis impedes cellular energy supply, further accelerating chondrocyte senescence. Thus, the restoration of mitochondrial function in chondrocytes may provide a potential therapeutic strategy for rescuing senescent chondrocytes. To evaluate the ability of soft HMs to improve mitochondrial function by alleviating the mechanical stress on chondrocytes, we performed PCR to determine the expression of key MRGs in chondrocytes. These key genes were identified from the intersection of MRGs with DEGs (Fig. 4A). COX4-i2 plays a crucial role in regulating mitochondrial respiratory chain activity, and its deficiency impairs mitochondrial membrane potential and causes dysfunction in the electron transport chain (ETC) S [53]. AKR-1b10 regulates fatty acid synthesis, and reduced AKR1B10 levels contribute to mitochondrial membrane damage [54]. PrimPol deficiency can severely impact mitochondrial DNA replication and maintenance, affecting overall mitochondrial function and cellular health [55]. The Pif-1 protein, a 5′→3′ superfamily 1 (SF1) helicase, is essential for DNA repair, recombination, and mitochondrial genome stability [56]. AASS is involved in lysine degradation within the mitochondria, and AASS accumulation is associated with significant mitochondrial damage [57]. Overall, the dysregulation of these genes affects mitochondrial homeostasis. COX4-i2, AKR1B10, PrimPol, Pif-1, and AASS were significantly downregulated in the HM group following mechanical stress, indicating mitochondrial dysfunction (Fig. 4B–F). In contrast, when mechanical stress was applied to chondrocytes in the Lipo@HM group, the relative expression levels of COX4-i2, AKR1B10, PrimPol, Pif-1, and AASS were significantly upregulated compared with those in the HM group; however, and the expression of COX4-i2 did not differ between the HMs and control groups. COX4-i2 is recognized as a key regulator of the mitochondrial respiratory chain. These findings suggest that soft HMs are more effective than hard HMs in maintaining mitochondrial homeostasis and modulating mitochondrial respiratory chain activity.

Fig. 4.

Evaluation of Mitochondrial Function and Gene Expression in Chondrocytes Treated with Different Microspheres. A) Venn diagram illustrating the overlap between differentially expressed genes (DEGs) and mitochondrial-related genes (MRGs). B) mRNA expression levels of COX4-i2 in chondrocytes from different groups under mechanical stress. C) mRNA expression levels of AKR-1b10 in chondrocytes across the different treatment groups. D) PrimPol gene mRNA expression levels in chondrocytes after treatment with HMs and Lipo@HMs. E) mRNA expression levels of PIF-1 in chondrocytes treated with different microsphere formulations. F) Expression levels of AASS gene mRNA in chondrocytes subjected to control, HMs, and Lipo@HMs treatments. G) Oxygen consumption rate (OCR) curves depicting the mitochondrial respiration function of chondrocytes under mechanical stress conditions across each treatment group. H) Quantitative analysis of basal mitochondrial respiration. I) Quantitative analysis of the mitochondrial ATP productivity potential. J) Quantitative analysis of the maximum respiratory capacity of mitochondria. K) Quantitative analysis of the spare respiratory capacity of mitochondria. (n = 3 for each group) (∗P < 0.05, ∗∗P < 0.01, ∗∗∗P < 0.001, ns: not significant).

In this study, the mitochondrial function of chondrocytes was evaluated in vitro by analyzing the oxygen consumption rate of cells in each treatment group (Fig. 4G). Parameters such as basal respiration, ATP production, maximal respiration, and spare respiratory capacity were quantified (Fig. 4H–K). The results demonstrated that basal respiration, maximal respiration, and spare respiratory capacity in the HMs group were significantly lower than those in the control group, indicating severe impairment of the mitochondrial respiratory chain under mechanical stress. However, under the same stress conditions, basal respiration and maximal respiratory capacity in the Lipo@HMs group were significantly higher than those in the HM group, suggesting that the prepared soft HMs effectively enhance mitochondrial function in chondrocytes subjected to mechanical stress. Finally, to further demonstrate the effect of the Lipo@HMs microspheres on the mitochondrial system, we stained and quantified intracellular ROS in different experimental groups (Fig. S9A). The results showed that the cells in Lipo@HMs group produced significantly less ROS than those in HMs group (Fig. S9B). Again, it was demonstrated that the Lipo@HMs microspheres were able to efficiently activate the mitochondrial system.

3.5. Impact of Res@Lipo@HMs on chondrocyte senescence

Multiple signaling pathways are known to contribute to the senescence process in chondrocytes, with changes in gene expression being critical to mediating these pathways. In this study, reactome enrichment analysis revealed that under mechanical stress, five key genes associated with cellular senescence, namely, ETS2, RPS6K-a2, TNIK, c-Fos, and C/EBPβ, were significantly upregulated. The biological functions of these genes have been linked to the regulation of cellular senescence [[58], [59], [60], [61], [62]]. To evaluate the potential antisenescence effects of Lipo@HMs on chondrocytes subjected to mechanical stress, qRT-PCR was performed. Following the addition of rigid HMs, the mRNA expression of ETS2, RPS6K-a2, TNIK, c-Fos, and C/EBPβ was significantly elevated compared with that in the control group (Fig. 5A–E). In contrast, the expression of these genes (except for TNIK) was similar to that in normal chondrocytes after treatment with soft HMs, indicating that liposome-loaded HMs exerted a protective effect on chondrocytes under mechanical stress. Overall, Lipo@HMs exerted an antisenescence effect on chondrocytes under mechanical stress.

Fig. 5.

Evaluation of the molecular mechanisms underlying the effects of different microsphere formulations on chondrocyte senescence under mechanical stress. A) Relative expression levels of ETS2 mRNA among the control, HMs, and Lipo@HMs groups. B) Quantification of RSK gene expression across experimental groups. C) Quantification of TNIK gene expression across experimental groups. D) Analysis of c-Fos mRNA levels among the control, HMs, and Lipo@HMs groups. E) Quantification of C/EBPβ gene expression across experimental groups. F) Western blot showing protein expression of C/EBPβ across four experimental groups, including control, HMs, HMs + P38 inhibitor, and Lipo@HMs. G) Western blot analysis of c-Fos levels among the control, HMs, HMs + ERK inhibitor, and Lipo@HMs groups. H) Quantitative analysis of c-Fos by Western blotting. I) Quantitative analysis of C/EBPβ by Western blotting. J) Western blot results of ERK and P38 phosphorylation under mechanical stress. K) The phosphorylation level of ERK was assessed by the ratio of P-ERK to total ERK. L) The phosphorylation level of P38 was assessed by the ratio of P-P38 to total P38. M) Expression levels of cellular senescence markers P21 and P16INK4a in different treatment groups. N) Quantitative analysis of P21 by Western blotting. O) Quantitative analysis of P16INK4a by Western blotting. P) Flow cytometric analysis of apoptosis across different treatment groups. Q) Quantitative analysis of apoptosis in chondrocytes treated with HMs, Lipo@HMs, and Res@Lipo@HMs under mechanical stress. R) Schematic representation of the signaling pathways involved in mechanical stress–induced chondrocyte senescence (n = 3 for each group); (∗P < 0.05, ∗∗P < 0.01, ∗∗∗P < 0.001, ns: not significant).

Oxidative stress and mechanical stimulation promote phosphorylation of extracellular signal–regulated kinase (ERK), which then activates the transcriptional activity of activator protein-1 (AP-1)—a complex composed of c-Fos and c-Jun—ultimately driving SnC [63]. C/EBPβ, a key transcription factor involved in the regulation of SASP in response to senescence-inducing stimuli, acts as a downstream substrate of p38 MAPK [64]. p38 activation leads to C/EBPβ phosphorylation [65], and p38 overexpression induces SnC [49]. This is consistent with our previous findings, which highlight c-Fos and C/EBPβ as critical genes influencing chondrocyte senescence. To further elucidate the role of mechanical stress in chondrocyte senescence, i.e., in the regulation of key gene expression and their associated pathways, western blotting was performed to determine relative protein levels. Chondrocytes were divided into four groups: (1) control group without any treatment; (2) HM group comprising rigid HMs under mechanical stress; (3) HM + p38 inhibitor group comprising HMs and p38 inhibitor under mechanical stress; and (4) Lipo@HM group comprising liposome-loaded HMs under mechanical stress (Fig. 5F). The experimental setup shown in Fig. 5G was similar to that in Fig. 5F, with the exception that the third group received an ERK inhibitor instead of a p38 inhibitor and was designated as the HM + ERK inhibitor group. The results shown in Fig. 5F–I demonstrate that the expression of C/EBPβ and c-Fos was significantly higher in the HM group than in the control group but significantly lower in both the inhibitor-treated and Lipo@HM groups than in the HM group.

Fig. 5J present the expression levels of phosphorylated ERK (p-ERK) and phosphorylated p38 (p-p38). Under mechanical stress, the HM group exhibited significantly increased p-ERK and p-p38 levels. In contrast, these levels were significantly lower in the Lipo@HM group than in the HM group. These findings suggest that liposomal HMs potentially mitigate chondrocyte senescence induced by mechanical stress by inhibiting these two signaling pathways (Fig. 5K and L). Next, to further verify the effect of microspheres on chondrocyte senescence, we examined the presence of cellular senescence-related proteins (i.e., P16-INK4a and P21) (Fig. 5M). The results showed that both the P16-INK4a and P21 proteins were significantly more abundant in the HMs group than in the control group. These results indicated that chondrocyte aging in the HMs group was higher, likely due to excessive mechanical stress stimulation. Meanwhile, the abundance of P16-INK4a and P21 proteins in the Lipo@HMs group was significantly lower than that in the HMs group. Taken together, these results indicate that the aging of chondrocytes in the Lipo@HMs group was alleviated by effectively dispersing mechanical stress (Fig. 5N and O). To further verify our conclusions, we performed β-gal staining of cells from both the HMs and Lipo@HMs groups (Fig. S10A). These results showed that the staining area of the HMs group was significantly larger than that of the Lipo@HMs group (Fig. S10B). It was further demonstrated that Lipo@HMs could effectively alleviate chondrocyte senescence.

Finally, to evaluate the antisenescence effect of drug-loaded liposomal HMs through stress modulation, we evaluated the apoptosis of chondrocytes using flow cytometry (Fig. 5P). The proportion of apoptotic cells was statistically analyzed (Fig. 5Q). The rate of apoptosis in the HMs group (25.06 ± 1.13) was significantly higher than that in the control group (8.14 ± 1.14) under mechanical stress, indicating that mechanical stress promotes cellular senescence and subsequently apoptosis. Notably, the apoptosis rate in the Lipo@HMs group (20.34 ± 0.50) was significantly lower than that in the HMs group, and the rate was even further reduced in the Res@Lipo@HMs group. These findings suggest that Res@Lipo@HMs exhibited the most pronounced antiapoptotic effect on chondrocytes. Consistent with previous research [66], Res was shown to effectively reduce chondrocyte apoptosis by inhibiting p38 MAPK, highlighting its significant role in mitigating chondrocyte senescence and protecting mitochondrial function. The findings of our study also imply that the combination of Res and soft HMs synergistically reduced chondrocyte senescence (Fig. 5R).

3.6. In vivo therapeutic efficacy of Res@Lipo@HMs in OA

Destabilization of the medial meniscus (DMM) is highly clinically relevant and has been widely adopted in animal models of OA. The mechanical stability of the knee is compromised by transecting the MMTL, which subsequently leads to cartilage damage. These changes closely mimic the processes observed in OA progression in humans. In this study, rats were randomly assigned into four groups (control, OA, HMs, and Res@Lipo@HMs), and the therapeutic effects of these interventions were evaluated in an in vivo model of OA.

Firstly, we evaluated the injectability of the hydrogel microsphere system through a rheometer. We respectively compared the shear-thinning behavior of PBS solution and the hydrogel microsphere injection. The results show that the viscosity of PBS solution is 1 mPas, the same as that of pure water (Fig. S11A). And the viscosity of the hydrogel microsphere injection is 20–30 mPas and continuously decreases with the increase of shear rate (Fig. S11B). This viscosity value is lower than that of fruit juice, so the hydrogel microsphere injection has good injection performance. Subsequently, we conducted an assessment of the immunogenicity of the hydrogel microsphere injection. We evaluated the toxicity of the hydrogel microsphere by continuously monitoring the body weight of rats. The results showed that there was no significant difference in the body weight of SD rats in each group at each time point. This indicates that the hydrogel microsphere injection is of low toxicity (Fig. S12A). We used the ELISA kit to detect the content of immunoglobulin (IgG1) in the blood of rats, thereby evaluating the immunogenicity of the hydrogel microsphere injection. The results showed that there was no significant difference in the IgG1 content among different groups of rats at different time points. This indicates that the hydrogel microsphere injection does not cause an immune response in rats (Fig. S12B).

Meanwhile, we verified the retention of the hydrogel microspheres in vivo and their drug release performance through fluorescence residue experiments. We modified the fluorescent group (Cy5) onto the hydrogel microspheres, injected them into the knee joints of rats, and observed the residual fluorescence intensity after 4 weeks (Fig. S13A). The results showed that after 4 weeks, the fluorescence intensity in the rats was 23.89 % of the initial value, indicating that the hydrogel microspheres had sufficient stability in vivo (Fig. S13B). Furthermore, we encapsulated Cy5 within liposomes inside hydrogel microspheres to measure the drug release performance (Fig. S13C). The results showed that after 1–3 weeks, the drug residues were 66.11 %, 38.41 %, and 19.94 % of the initial value, respectively. By the fourth week, almost no fluorescence residue was detectable (Fig. S13D). In conclusion, the hydrogel microspheres can maintain drug release for approximately 3–4 weeks in rats. Moreover, the anti-degradation performance of the hydrogel microspheres can support their long-term sustained drug release.

Histological changes in the articular cartilage were assessed using hematoxylin and eosin (HE) staining and Safranin O-fast green staining. The articular cartilage in the control group exhibited a smooth surface, organized structural arrangement, normal cellular morphology, and strong Safranin O-fast green staining (Fig. 6A–C). In contrast, the OA group displayed severe surface abrasion, disorganized chondrocytes, and weak Safranin O-fast green staining. In the HM group, significant degenerative changes were observed compared with the control group, but the OARSI scores were better than those in the OA group (Fig. 6B). This finding suggests that intra-articular injection of HMs alone provided some alleviation of cartilage damage, possibly because of their ball-bearing lubrication effect. The Res@Lipo@HM group exhibited the best cartilage preservation, with only mild degenerative changes that were further corroborated by improved OARSI scores (Fig. 6B). These findings demonstrate that treatment with Res@Lipo@HMs was the most effective in maintaining normal articular cartilage structure and mitigating degeneration.

Fig. 6.

Animal experiments for the assessment of the treatment of OA with Res@Lipo@HMs. A) Representative images of H&E staining. B) OARSI scores of articular cartilage in each group. C) Representative images of safranin O-fast green staining (S-F staining). D) Relative GAG content in each group. E) Representative sections of terminal deoxynucleotidyl transferase-mediated deoxyuridine triphosphate nick end labeling (TUNEL) staining for apoptotic cells. F) Quantification of TUNEL-positive cells. G) Representative sections of immunohistochemical staining for collagen II. H) Quantification of collagen II-positive cells. I) Representative sections of immunohistochemical staining for MMP13. J) Quantification of MMP13-positive cells. (n = 6 for each group) (ns: not significant; ∗P < 0.05, ∗∗P < 0.01, ∗∗∗P < 0.001). (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)

We subsequently assessed the relative content of glycosaminoglycans (GAGs) using Safranin O-fast green staining to evaluate articular cartilage damage (Fig. 6C and D). The staining was lighter and the GAG content was the lowest in the OA group compared with the control group, indicating severe cartilage damage and confirming the successful establishment of the OA model. The HM group demonstrated similar pronounced degenerative changes as observed in the OA group, with no significant difference in the GAG content between the two groups. In contrast, the Res@Lipo@HM group showed more complete cartilage coverage, with stronger staining and significantly higher GAG content than the HM group. In summary, Res@Lipo@HM treatment effectively alleviated cartilage damage and degeneration in the OA model.

To verify whether drug-loaded soft HMs could effectively inhibit chondrocyte apoptosis in vivo, we employed terminal deoxynucleotidyl transferase-mediated dUTP nick end labeling (TUNEL) staining to detect apoptotic chondrocytes (Fig. 6E) and quantified the number of TUNEL-positive cells (Fig. 6F). In the OA group, TUNEL staining indicated extensive chondrocyte apoptosis arising from the lack of therapeutic intervention, indicating that the OA microenvironment leads to significant apoptosis that ultimately results in cartilage damage. The number of apoptotic cells observed in the HM group (78.93 ± 7.28) did not significantly differ from that in the OA group (67.09 ± 10.61), suggesting that HM injection alone was insufficient to inhibit chondrocyte apoptosis. However, the number of apoptotic cells was significantly lower in the Res@Lipo@HM group (39.76 ± 11.99) than in the HM and OA groups, indicating that the intra-articular injection of Res@Lipo@HMs was effective in inhibiting chondrocyte apoptosis and mitigating OA progression.

In the DMM OA model in rats, OA progression is characterized by a gradual decline in chondrocyte proliferation and extracellular matrix (ECM) synthesis. Collagen II (Col2) is the principal extracellular matrix component in the cartilage and serves as a marker for chondrocyte proliferation [67]. Matrix metalloproteinase 13 (MMP13), a critical enzyme involved in Col2 degradation, plays a vital role in the degradation of the cartilage matrix in OA [68]. To assess the effects of Res@Lipo@HMs on apoptosis levels and the balance between cellular anabolism and catabolism, protein levels of collagen II and MMP13 were determined using immunohistochemical methods. The expression levels of collagen II were significantly lower in the OA and HM groups than in the control group (Fig. 6G and H). The Res@Lipo@HM group displayed notably increased collagen II levels compared to the OA group. In addition, Fig. 6I and J demonstrate that MMP-13 protein levels were significantly elevated in both the OA and HM groups compared with the control group, indicating the DMM-induced inhibition of mitochondrial autophagy in the cartilage of rats with OA. These findings are consistent with those of previous research highlighting cartilage damage in DMM animal models. Notably, MMP-13 expression was significantly reduced in the Res@Lipo@HM group, clearly illustrating the inhibitory effects of Res@Lipo@HMs on the progression of OA in vivo.

Finally, we further verified through animal experiments that our Res@Lipo@HMs injection solution could enhance mitochondrial function and alleviate the senescence of chondrocytes. We performed immunohistochemical staining for indicators p16-INK4a and COX-4 on the tissue sections of knee joint cartilage (Fig. S14A). The experimental results show that p16-INK4a, as an indicator related to chondrocyte senescence, has a significantly increased expression level in the OA group, while the expression level in Res@Lipo@HMs group is significantly lower than that in HMs group (Fig. S14B). This indicates that an increase in mechanical stress leads to the chondrocyte senescence, while the Res@Lipo@HMs can significantly alleviate chondrocyte aging by dispersing mechanical stress. The experimental results show that COX-4, as an indicator related to mitochondrial function, has a significantly decreased expression level in the OA group, while the expression level in Res@Lipo@HMs group is significantly higher than that in HMs group (Fig. S14C). This indicates that the Res@Lipo@HMs can effectively activate mitochondrial function by alleviating mechanical stress and releasing resveratrol, laying the foundation for anti-chondrocyte senescence.

Overall, these results suggest that Res@Lipo@HMs injection can effectively target chondrocytes, whereas the enhanced mechanical properties of HMs alleviate mechanical stress on chondrocytes, thereby mitigating senescence. Thus, HMs offer a promising therapeutic strategy for OA.

4. Conclusion

In this study, we developed a high-mobility lattice HM system that incorporates cationic liposomes into a hydrogel network, integrating them into an oxidized HAMA matrix through imine bonding to achieve enhanced mechanical properties and superior elastic deformation capabilities. This system not only improves joint lubrication via a rolling mechanism but also reduces axial stress by increasing the elastic deformation of the microspheres, positively influencing the regulation of the mitochondrial respiratory chain and delaying cellular senescence. Furthermore, the controlled release of Res-loaded cationic liposomes from these HMs was able to effectively target negatively charged cartilage through electrostatic interactions, thereby inhibiting chondrocyte senescence, enhancing the physiological function of chondrocytes, and alleviating OA progression.

This study still has some limitations that require further exploration in the future. For instance, to make the application scenarios of this hydrogel microsphere system more in line with clinical practice, in vivo experiments should be conducted using large animals (such as dogs and pigs). In future research, we will not only continue to refine the content of this study but also expand the application scenarios of this hydrogel microsphere system. After all, mechanical stress overload is one of the most important pathological mechanisms of musculoskeletal diseases.

CRediT authorship contribution statement

Fangqi Xu: Writing – original draft, Methodology, Investigation. Chen Zhuang: Methodology, Investigation. Lufeng Yao: Writing – original draft, Methodology, Investigation. Yiwen Xu: Methodology, Investigation. Qihua Cao: Investigation. Zherui Fu: Methodology. Longfeng Wang: Investigation. Yuan Zhu: Investigation. Deting Xue: Project administration, Investigation. Ning Zhang: Project administration, Funding acquisition. Xiaohua Yu: Project administration, Data curation. Gangfeng Hu: Writing – review & editing, Project administration, Data curation. Feng Lin: Writing – review & editing, Project administration, Funding acquisition, Data curation, Conceptualization.

Ethics approval and consent to participate

Animal experiments were approved by the Animal Ethics Committee of the Second Affiliated Hospital of Zhejiang University School of Medicine (ethical approval number 2024-226).

Consent for publication

All authors read and agreed to submit the manuscript.

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.

Appendix A. Supplementary data

The following is the Supplementary data to this article:

Data availability

Data will be made available on request.