Neutrophil membrane-camouflaged SiRNA nanoplatform targeting IL-33 attenuates osteoarthritis through autophagy-dependent senescence regulation

Abstract

Osteoarthritis (OA) imposes a substantial health and economic burden globally. Currently, there is a lack of disease-modifying osteoarthritis drugs (DMOADs). This study aimed to elucidate the relationship between chondrocyte senescence and OA progression, as well as to develop an effective siRNA nanodelivery platform for OA treatment. We engineered neutrophil membrane-coated, siIL33-loaded nanoparticles (NM-NP-siIL33) for OA management. The therapeutic efficacy of NM-NP-siIL33 was evaluated through both in vitro and in vivo experiments. Our findings revealed that IL-33 expression was significantly upregulated in damaged articular cartilage in both young and aged mice following anterior cruciate ligament transection (ACLT) surgery. In vitro experiments demonstrated that IL-33 promotes chondrocyte senescence by inhibiting cellular autophagy via activation of the p38 mitogen-activated protein kinase (MAPK) pathway. Additional in vivo studies showed that NM-NP-siIL33 effectively delivered siIL33 to target cells within OA tissues, thereby mitigating the degradation of articular cartilage. Our results suggest that IL-33 plays a critical role in OA progression by accelerating chondrocyte senescence. Furthermore, NM-NP-siIL33 represents a promising therapeutic strategy for managing OA.

Graphical abstract

Supplementary Information

The online version contains supplementary material available at 10.1186/s12951-025-03686-3.

Introduction

Osteoarthritis (OA), the most prevalent age-related joint disorder, characterized by cartilage deterioration, subchondral bone remodeling, and synovial inflammation [1]. Global epidemiological projections indicate that the prevalence of OA will affect approximately 67 million individuals worldwide by 2030, posing an escalating socioeconomic burden paralleling population aging trends [2]. Despite decades of therapeutic development, no disease-modifying osteoarthritis drugs (DMOADs) have demonstrated clinical efficacy in halting early-stage OA progression. Current end-stage management remains predominantly reliant on artificial joint replacement, underscoring the urgent need for targeted molecular therapies [3].

Emerging evidence implicates cellular senescence as a central mechanism driving age-related pathologies, including OA, through metabolic dysregulation and chronic inflammation [4–7]. Chondrocytes, the sole cell type found in cartilage, play a crucial role in maintaining the homeostasis and integrity of joint articular tissue by synthesizing the extracellular matrix (ECM) [8]. Multifactorial stressors-including age-related mitochondrial dysfunction, traumatic injury, and aberrant mechanical loading – converge to induce irreversible senescence marked by cell-cycle arrest and senescence-associated secretory phenotype (SASP) activation [9]. Notably, senescent chondrocytes amplify OA pathogenesis through sustained release of catabolic mediators (IL-1β, IL-6, TNF-α) and matrix metalloproteinases (MMPs), establishing senolysis as a promising therapeutic frontier [10]. Herein, we developed a nanoparticle platform for delivering siRNA to OA lesions. These nanoparticles effectively mitigate articular cartilage damage in murine models following ACLT. Mechanistically, these nanoparticles prevent chondrocyte senescence by restoring impaired autophagy, thereby attenuating OA progression.

Autophagy plays a crucial role in maintaining cellular homeostasis and facilitating adaptation to stress. Dysregulation of autophagic processes can disrupt tissue homeostasis and adversely affect individual health. Increasing evidence suggests that autophagy activity decreases with organismal aging, while cellular senescence phenotypes can be inhibited by activating autophagy [11]. For instance, Tai et al. report that metformin alleviates vascular smooth muscle cell (VSMC) senescence by enhancing autophagic flux at the autophagosome-lysosome fusion level [12]. However, the precise mechanistic interplay between autophagy dysregulation and chondrocyte senescence in OA remains undefined. Recent studies indicate that IL-33 is involved in both autophagy and cellular senescence [13, 14]. Furthermore, bioinformatics analysis reveals that IL-33 is highly expressed in OA tissues. Therefore, elucidating the role of IL-33 in OA with respect to chondrocyte autophagy and senescence could aid in the development of novel pharmacological therapeutic strategies for OA.

Biomimetic nanoparticles (BMNPs) represent an innovative drug delivery system characterized by their adjustable composition and size, alongside superior biocompatibility. Recent studies have demonstrated that nanoparticles functionalized with neutrophil membranes substantially improve the delivery efficiency of therapeutic agents [15]. In this study, we assessed the expression of IL-33 in a murine osteoarthritis (OA) model. Gain-of-function experiments were conducted to investigate the impact of IL-33 on autophagy and senescence in chondrocytes. Furthermore, we developed si-IL33-loaded neutrophil membrane-coated nanoparticles (NM-NP-siIL33) to enhance the delivery efficiency of siIL33 to chondrocytes. Our results confirmed that NM-NP-siIL33 effectively prevented OA progression in our models, and we elucidated the molecular mechanisms underlying this effect. These findings contribute to a deeper understanding of the mechanisms involved in OA pathogenesis and provide a novel pharmacological therapeutic strategy for OA treatment.

Materials and methods

Bioinformatics analysis

Candidate genes were screened using the Gene Expression Omnibus (GEO) database (www.ncbi.nlm.nih.gov/geo/). The datasets GSE249509 and GSE246425 were selected for analysis. Differentially expressed genes (DEGs) were identified based on the following criteria: a fold change greater than 2 and a p-value less than 0.05. Data analysis and graphical representation were conducted utilizing the OmicStudio platform (https://www.omicstudio.cn/).

Cell culture and treatment

Chondrocytes were isolated from C57BL/6 mice following a previously described protocol with modifications [16]. Primary chondrocytes were cultured in DMEM-F12 supplemented with 15% fetal bovine serum (FBS) and 1% penicillin-streptomycin. Senescence was induced by culturing chondrocytes for four hours in a medium containing hydrogen peroxide (H₂O₂, 200 µM). To assess the pro-senescent effects of IL-33, primary chondrocytes were stimulated with recombinant IL-33 (10 µg/mL). Additionally, Astegolimab (10 ng/mL) and Adezmapimod (20 ng/mL) were administered in combination with IL-33 according to the specified experimental groupings. For cellular internalization testing, NM-NP-siIL33 (100 µg/mL) were incubated with chondrocytes for six hours. After rinsing excess nanoparticles (NPs) with phosphate-buffered saline (PBS), the internalization of NPs by chondrocytes was observed using confocal microscopy (Leica Biosystems, Germany).

Animals

All animal experiments were approved by the Animal Ethics Committee of Shanghai General Hospital, affiliated with the Shanghai Jiao Tong University School of Medicine. An anterior cruciate ligament transection (ACLT)-induced mouse OA model was established in accordance with previously described methodologies [17]. Briefly, mice were anesthetized with 1% isoflurane (RWD Life Science, China). A longitudinal incision was made on the right knee skin, and the anterior cruciate ligament was severed under a surgical microscope using an ophthalmic scalpel. For IL-33 overexpression, an adeno-associated virus (AAV)-mediated delivery system (VectorBuilder, China) was used. Equal volumes of AAV vectors (5 µL) carrying scrambled control or IL-33 fragments were injected into the joint cavity. In the NM-NP-siIL33 therapeutic effectiveness study, mice were randomly divided into four groups: OA (ACLT mice injected with PBS), NM-NPs (ACLT mice injected with empty neutrophil membrane-coated NPs), siIL33 (ACLT mice injected with naked siIL33), and NM-NP-siIL33 (ACLT mice injected with siIL33-loaded, neutrophil membrane-coated NPs). Knee injections were administered on days − 7, 3, and 14 post-ACLT surgery. Whole knee joint tissues from the right limb were harvested at week 8 post-euthanasia for further analysis.

Micro-CT analysis

The knee joints of mice were initially fixed with 4% PFA. Imaging and analysis were subsequently conducted utilizing the high-resolution SkyScan1076 micro-computed tomography system (Bruker, Belgium). Specific regions of interest (ROIs) within the subchondral bone trabecular area were identified for data collection. Subsequently, two-dimensional (2D) and three-dimensional (3D) image reconstructions were performed using 3D MED image processing software. A quantitative analysis of the trabecular bone microstructure was executed, focusing on parameters, including bone volume relative to total volume (BV/TV) and trabecular thickness (Tb.Th).

CCK8 assay

CCK8 assays were conducted to assess the cytotoxicity of NM-NP-siIL33 on chondrocytes according to the manufacturer’s instructions (Solarbio, China). Briefly, chondrocytes (1 × 10⁴ cells) were seeded in a 96-well plate and various concentrations of NM-NP-siIL33 were treated with these cells for 48 h. Following this, each well was subjected to incubation with the CCK8 reagent for a duration of 2 h. Subsequently, the optical density (OD) value was measured at a wavelength of 450 nm using a microplate reader (BioTec Instruments, Inc., USA).

Histological analyses

Mouse specimens were fixed in 4% paraformaldehyde (PFA), embedded in paraffin, and sectioned for further analysis. Cartilage and synovium sections were stained with Safranin O-Fast Green (Solarbio, China) and H&E (Solarbio, China), respectively. The OARSI cartilage scoring system, along with the synovitis scores, was employed in accordance with the methodologies outlined in prior research [18, 19].

Immunohistochemistry (IHC) and Immunofluorescence (IF) analyses

The sections designated for IHC staining were incubated in a 3% hydrogen peroxide solution for 15 min. Subsequently, all sections underwent blocking with a 1% goat serum solution containing 0.15% Triton X-100 for 1 h, after which they were incubated with primary antibodies. For IHC staining, the tissue sections were treated with horseradish peroxidase (HRP)-conjugated secondary antibodies, followed by staining with 3,3′-diaminobenzidine (DAB). In contrast, for IF staining, the sections were incubated with secondary antibodies conjugated to fluorescent dyes. IHC slides were examined using a Pannoramic MIDI (3DHISTECH, Hungary), while IF slides were captured using confocal microscopy.

Synthesis and characterization of NM-NP-siIL33

Mouse bone marrow cells were harvested and stimulated with LPS (100 ng/mL) for 2 h. Following stimulation, the cells were centrifuged at 500 × g for 10 min to isolate neutrophils. The isolated cells were then resuspended in a precooled separation buffer containing 225 mM mannitol, 75 mM sucrose, 30 mM Tris-HCl, 0.5 mM EDTA, and 1% (v/v) protease inhibitor, maintained on ice, and sonicated at 100 W for 5 min. Subsequently, the homogenate was centrifuged at 10,000 × g for 10 min at 4 °C. The resulting supernatant was collected, lyophilized, and stored at − 80 °C.

PLGA-PEG-MAL (MCE, USA) was dissolved in acetone and combined with deionized water containing 20 µM siIL33. The solution was dried in a vacuum oven to evaporate acetone, forming the NP core. Fluorescently labeled NPs were synthesized using sulfo-cyanine5 (Cy5)-labeled siIL33 or siNC. Neutrophil membranes were isolated from HL60-derived neutrophils using a previously established method with minor modifications. Neutrophil membranes were incorporated with siIL33-loaded NPs at a weight ratio of 3:1 (w/w) and processed through 20 cycles of extrusion using nucleopore membranes with initial pore sizes of 400 nm, followed by 200 nm, employing an Avanti mini extruder (Avanti Polar Lipids, USA). The size, surface zeta potential, and morphology of the resultant nanoparticles were characterized using dynamic light scattering (DLS, Malvern, UK) and transmission electron microscopy (TEM, JEOL Ltd., Japan).

In vivo localization analysis

Cy5 labeled siIL33 and NM-NP-siIL33 were injected through the joint cavity. The hind legs were harvested and analyzed using an imaging visualization and infrared spectroscopy (IVIS) system (PerkinElmer, Inc., USA), in conjunction with immunofluorescence staining.

RT-qPCR

Total RNA was isolated from cellular samples or articular cartilage utilizing TRIzol reagent. RT-qPCR was performed employing the BeyoFast™ SYBR Green One-Step qRT-PCR Kit (Beyotime, China) on a Real-Time PCR Detection System (Bio-Rad, USA), following the manufacturer’s guidelines. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) served as the internal control. The specific primers utilized in this investigation are outlined in Table S1.

Reactive oxygen species (ROS) detection

The Reactive Oxygen Species Assay Kit (Solarbio, China) was employed to assess the levels of ROS. In summary, cell or tissues samples was incubated with 200 µL of a 10 mM solution of 2ʹ,7ʹ-dichlorodihydrofluorescein diacetate (DCFH-DA), which was then incubated at 37 °C for a duration of 30 min. The fluorescence intensity of DCFH-DA was measured using a microscope to quantify the intracellular levels of reactive oxygen species.

Western blot

Protein isolation, quantification, and electrophoresis were conducted in accordance with previously established protocols. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was utilized as a loading control. The analysis of protein blots was carried out using Image Pro Plus version 6.0.

Senescence-associated β-galactosidase (SA-β-Gal) assay

SA-β-gal staining was conducted utilizing an SA-β-Gal staining kit (Solarbio, China), following the methodologies outlined in prior research [16]. Briefly, chondrocytes (1.0 × 10⁵) were inoculated onto six-well culture plates and incubated for 48 h. Following treatment with recombinant IL-33 (rIL-33), cytochemical staining for senescence-associated β-galactosidase (SA-β-Gal) was conducted at pH 6, and the number of positively stained cells was quantified.

Statistical analysis

All quantitative data are expressed as mean ± standard deviation (SD). Data analysis and image generation were performed utilizing GraphPad Prism 9.0 (GraphPad Software, USA). For quantitative comparisons between two groups, an unpaired Student’s t-test was employed. One-way ANOVA or two-way ANOVA, followed by a post-hoc Bonferroni test, was utilized for comparisons among multiple groups. A p-value of less than 0.05 was deemed statistically significant.

Results

IL-33 is significantly upregulated and closely associated with cellular senescence in a mouse osteoarthritis (OA) model

The expression of IL-33 in senescent chondrocytes was initially investigated. Bioinformatics analysis based on the GEO database (www.ncbi.nlm.nih.gov/geo/) identified 47 genes that were upregulated in both human (GSE249509) and murine (GSE246425) senescent chondrocytes (Fig. 1A). Subsequently, the differentially expressed genes (DEGs) in chondrocytes were validated. Chondrocytes isolated from the articular cartilage of mice were treated with hydrogen peroxide to induce cellular senescence. Results demonstrated that the number of SA-β-gal-positive chondrocytes in the H2O2 group was significantly higher than in the PBS group (Fig. 1B and C). Further RT-qPCR analysis revealed that among the top five candidate genes screened from the database, IL-33 exhibited the most substantial increase in relative expression in senescent chondrocytes (Fig. 1D). IL-33 expression was also evaluated in a mouse OA model. The results indicated evident cartilage destruction in both young and aged mice at 8 weeks post-anterior cruciate ligament transection (ACLT) surgery, along with a higher Osteoarthritis Research Society International (OARSI) score in aged mice (Fig. 1E and F). Immunohistochemical (IHC) staining further confirmed an increased number of IL-33-positive cells in the cartilage of aged mice (Fig. 1G and H). Comparable findings were observed in reactive oxygen species (ROS) testing (Supplementary Fig. 1 A) and senescence-associated secretory phenotype (SASP) analysis (Supplementary Figs. 1 B-E).

Fig. 1.

IL-33 is significantly upregulated in senescent chondrocytes both in vitro and in vivo. A Bioinformatics analysis to screen candidate genes. B SA-β-gal staining for the identification of senescent chondrocytes. Scale bars = 50 μM. C Quantification of SA-β-gal-positive chondrocytes. (n=4), ***p<0.001. D Quantitative analysis of candidate gene expression utilizing RT-qPCR. (n=4), ***p<0.001.E The degradation of articular cartilage was assessed using SO/FG staining. Scale bars = 100 μM. F OARSI scoring was performed according to staining results. (n=8), **p<0.01. G IL-33-positive chondrocytes detected by IHC. H Quantitative IL-33-positive chondrocytes. (n=4). ***p<0.001 vs. 3M group, ^&&& p<0.001 vs. 3M + ACLT

Overexpression of IL-33 promotes chondrocyte senescence and exacerbates cartilage destruction in a mouse OA model

The timeline for AAV-IL-33 injection and ACLT surgery is illustrated in Fig. 2A. RT-qPCR results showed a significant increase in IL-33 gene expression in mouse articular cartilage two weeks after AAV-IL-33 injection (Supplementary Fig. 2 A). Safranin-O/Fast Green (SO/FG) staining revealed more severe articular cartilage destruction in the AAV-IL-33 injection group 8 weeks post-ACLT surgery (Fig. 2B and C). Synovitis was assessed by hematoxylin and eosin (H&E) staining and scored using a semi-quantitative scoring system. Results indicated that AAV-IL-33 treatment significantly increased synovial tissue proliferation and inflammatory cell infiltration (Fig. 2D and E). Immunofluorescence (IF) staining was used to examine senescence markers in chondrocytes. Results showed a significant increase in γ-H2A.X+ (Fig. 2F and G) and p21+ (Fig. 2H and I) chondrocytes in the AAV-IL-33 injection group compared to the OA model group.

Fig. 2.

IL-33 overexpression promotes OA progression. A Timeline of different treatments. B The degradation of articular cartilage was assessed using SO/FG staining. Scale bars = 100 μM. C OARSI scoring was performed according to staining results. (n=8), **p<0.01. D Synovial inflammation was detected by H&E staining. Scale bars = 100 μM. E Synovitis scoring was performed according to staining results. (n=8),**p<0.01. F DNA-damaged chondrocytes (γ-H2A.X⁺) was detected by IF staining. Scale bars = 100 μM. G Quantification of γ-H2A.X-positive chondrocytes. (n=4), ***p<0.001. H Senescent chondrocytes (p21⁺) were detected by IF staining. Scale bars = 100 μM. G Quantification of p21-positive chondrocytes. (n=4), ***p<0.001

Recombinant IL-33 (rIL-33) protein promotes chondrocyte senescence

In vitro gain-of-function experiments were conducted to evaluate the effects of rIL-33 on chondrocyte senescence. The optimal concentration of rIL-33 (50 ng/mL) required to induce chondrocyte senescence was determined using SA-β-Gal staining (Fig. 3A and B). IF staining revealed that rIL-33 treatment exacerbated DNA damage in chondrocytes, as evidenced by an increase in γ-H2A.X-positive speckles within the nucleus (Fig. 3C and D). Western blotting results indicated that rIL-33 significantly elevated the expression of senescence-associated proteins, including p53, p21, and p16 (Fig. 3E). Ki67 detection suggested that rIL-33 treatment inhibited chondrcyte proliferation (Fig. 3F and G). RT-qPCR analysis of SASP components, including IL-1β, IL-6, CCL2, ICAM-1, TGF-β, IFN-γ, EGF, and bFGF, demonstrated that rIL-33 significantly promoted chondrocyte senescence. These effects were attenuated by the application of the IL-33 inhibitor, Astegolimab (Fig. 3H).

Fig. 3.

IL-33 promotes chondrocyte senescence. A SA-β-gal staining for the identification of senescent chondrocytes. Scale bars = 50 μM. B Quantification of SA-β-gal-positive chondrocytes. (n=4), **p<0.01, ***p<0.001. C Chondrocyte intranuclear damaged DNA labelled using IF (γ-H2A.X) staining. Scale bars = 10 μM.D Quantification of γ-H2A.X-positive speckles in each nucleus. (n=4), ***p<0.001.E Senescence-related proteins detected by WB. F Proliferating (Ki67⁺) chondrocytes detected by IF staining. Scale bars = 50 μM. G Quantification of Ki67-positive chondrocytes. (n=4), **p<0.01, ***p<0.001 vs. Ctrl, ^&&& p<0.001 vs. IL-33. H The mRNA expression of SASP was detected by RT-qPCR. **p<0.01, ***p<0.001 vs. Ctrl,^&&p<0.01,^&&&p<0.001 vs. IL-33

Recombinant IL-33 (rIL-33) inhibits cellular autophagy through the activation of p38 signaling

The impact of rIL-33 treatment on autophagic function in chondrocytes was assessed using IF staining. Findings indicated that rIL-33 treatment reduced the expression of the autophagy-related protein LC3B in chondrocytes (Fig. 4A and B). Database screening revealed that IL-33 exerts its effects by binding to the IL1RL1 receptor, thereby modulating the downstream p38 signaling pathway (Fig. 4C). Western blot analysis further corroborated that IL-33 inhibits autophagy and promotes cellular senescence via the activation of the p38 pathway (Fig. 4D and E). These effects were attenuated by the application of the p38 inhibitor, Adezmapimod.

Fig. 4.

IL-33 inhibits autophagy in chondrocytes through the activation of p38 MAPK. AAutophagy-related protein LC3B detected by IF staining. Scale bars = 50 μM. BQuantification of LC3B relative fluorescence intensity. (n=4), ***p<0.001. C Schematic of IL-33 and IL1RL1 binding prediction. D The expression of p38, p-p38, LC3B, p53 was detected by WB. E Quantification of grey scale values of protein bands. *p<0.05, **p<0.01, ***p<0.001 vs. Ctrl,^&&& p<0.001 vs. IL-33.

Synthesis and characterization of neutrophil membrane-coated siIL33 loaded NPs

NM-NP-siIL-33 was synthesized through a systematic process, as illustrated in Fig. 5A. The morphology of NM-NP-siIL-33 was examined using transmission electron microscopy (TEM), as depicted in Fig. 5B. Both NM-NP-siIL-33 and unencapsulated siRNA were incubated with ribonuclease (RNase). Results indicated that unencapsulated siRNA underwent rapid degradation, whereas siRNA extracted from nanoparticles retained its structural integrity when exposed to RNase for up to 120 min (Fig. 5C). The average particle sizes of NP-siIL33 and NM-NP-siIL33 were 121.75 ± 9.07 nm and 141.25 ± 15.41 nm, respectively (Fig. 5D). The corresponding average zeta potentials were − 15.3 ± 1.71 mV and − 17.55 ± 1.58 mV, respectively (Fig. 5E). The in vitro release profile demonstrated that approximately 65% of siIL-33 was released from NM-NP-siIL-33 after 96 h in PBS (Fig. 5 F). CCK-8 assays indicated that NM-NP-siIL-33 exhibited low cytotoxicity in chondrocytes (Fig. 5G). Protein blotting results confirmed the successful coating of nanoparticles with neutrophil membranes (Fig. 5H). Cellular uptake of NM-NP-siIL-33 was assessed in chondrocytes, and findings revealed that the neutrophil membrane coating significantly enhanced the delivery efficiency of siIL-33-loaded nanoparticles (Fig. 5I and G).

Fig. 5.

Synthesis and Characterization of NM-NP-siIL33. ASchematic diagram of NM-NP-siIL33 synthesis. B NM-NP-siIL33 morphology observed by TEM. Scale bars = 50 nM. C siIL33 degradation was assessed by gel electrophoresis. D and E Particle size and zeta potential were assessed by DLS. F siIL33 release profile. G Cell viability was measured using the CCK8 assay. H Neutrophil membrane proteins was detected by WB. I The internalization of CY5-labeled siIL33 by chondrocytes was detected using IF techniques. Scale bars = 100 μM. J Quantification of CY5 relative fluorescence intensity. n=4. ***p<0.001

Neutrophil membrane coating enhances the cartilage delivery efficiency of NP in vivo

The cartilage delivery efficiency of siIL-33 was evaluated using a mouse OA model (Fig. 6A). For in vivo localization analysis, siIL-33 was labeled with Cy5. Bioluminescence analysis suggested that NM-NP-siIL-33 exhibited higher relative radiant efficiency in knees compared to naked siIL-33 (Fig. 6B and C). Additionally, immunofluorescence (IF) staining revealed a higher accumulation of Cy5-labeled siIL-33 in cartilage tissues, as evidenced by increased fluorescence intensity (Fig. 6D and E).

Fig. 6.

NM-NP-siIL33 in vivo residency study.ATimeline of in vivo imaging.B Detection of in vivo location of cy5-labeled siIL33 (red) by IVIS. C Quantification of the fluorescence intensity of Cy5. n=6. ***p<0.001.D Detection of the location of Cy5-labeled siIL33 in the cartilage by IF staining. Scale bars = 50 μM. E Quantification of CY5 relative fluorescence intensity. n=4. ***p<0.001

NM-NP-siIL33 treatment preserves articular cartilage integrity in a mouse OA model

The therapeutic efficacy of NM-NP-siIL33 was assessed in a mouse model of OA induced by ACLT (Fig. 7A). RT-qPCR results showed a significant decrease in IL-33 gene expression in mouse articular cartilage after NM-NP-siIL33 injection (Supplementary Fig. 2 A). Gait analysis revealed that the distance traveled by the hind paw in the treatment group was significantly greater than in the OA group, suggesting that NM-NP-siIL33 may delay OA progression (Fig.7B). Cartilage degradation was evaluated eight weeks post-ACLT surgery using micro-computed tomography (micro-CT), SO/FG staining, and H&E staining. Micro-CT analysis indicated that NM-NP-siIL33 treatment increased the number of trabecular bones in the coronal plane and enhanced subchondral bone volume in the tibia, as observed in the sagittal plane. Trabecular bone parameters, including bone volume fraction (BV/TV) and trabecular thickness (Tb.Th), exhibited significant improvements following NM-NP-siIL33 treatment (Fig. 7C and E). SO/FG staining and OARSI score analysis demonstrated enhanced roughness of the articular surface cartilage and mitigation of cartilage degeneration following NM-NP-siIL33 administration (Fig. 7F and G). H&E staining indicated that NM-NP-siIL33 diminished synovial inflammation in the knee joints of mice (Fig. 7H and I).

Fig. 7.

NM-NP-siIL33 treatment inhibit OA progression. A Timeline of in vivo administration. B Recordings of walking distances on the hind legs of mice. C The degradation of articular cartilage was assessed using micro-CT.DandE Quantification of BV/TV (%) and Tb.Th (mm). n=6. ***p<0.001. vs. OA.^&p<0.01,^&& p<0.001 vs. siIL33. F The degradation of articular cartilage was assessed using SO/FG staining. Scale bars = 100 μM. G OARSI scoring was performed according to staining results. (n=10), ***p<0.001. vs. OA.^& p<0.01. vs. siIL33. H Synovial inflammation was detected by H&E staining. Scale bars = 100 μM. I Synovitis scoring was performed according to staining results. (n=8), ***p<0.001. vs. OA.^& p<0.01. vs. siIL33

NM-NP-siIL33 inhibits chondrocyte senescence by activating autophagy in a murine model of OA

Following NM-NP-siIL-33 treatment, IL-33 expression in mouse knee tissues was assessed using Western blot analysis. Results indicated a significant reduction in IL-33 expression in the NM-NP-siIL-33-treated group compared to the OA group. Additionally, NM-NP-siIL-33 treatment activated p38-mediated autophagy while concurrently decreasing the expression levels of the senescence-associated protein P53 (Fig. 8A and F). RT-qPCR assays confirmed that NM-NP-siIL-33 treatment reduced the expression of SASP molecules (Fig. 8G and M).

Fig. 8.

NM-NP-siIL33 treatment activated p38 in vivo. A The expression of IL33, p38, p-p38, LC3B, p53 in OA tissues was detected by WB. B-F Quantification of grey scale values of protein bands. n=4. ***p<0.001. vs. OA.^&&& p<0.001. vs. siIL33. G-M The mRNA expression of SASP was detected by RT-qPCR. ***p<0.001 vs. OA,^&& p<0.01, ^&&& p<0.001 vs. siIL33

Discussion

Osteoarthritis (OA) imposes a significant health and economic burden globally. Despite extensive research efforts, no clinically effective disease-modifying osteoarthritis drugs (DMOADs) are currently available to halt the progression of early-stage OA [2]. In this study, we developed si-IL33-loaded nanoparticles encapsulated in neutrophil membranes and evaluated their therapeutic efficacy for the treatment of OA in a murine model. The key findings of our study are as follows: (1) IL-33 was significantly increased in senescent chondrocytes and OA tissues; (2) IL-33 promotes chondrocyte senescence by suppressing chondrocyte autophagy via the p38 MAPK signaling pathway; (3) The encapsulation of NPs within neutrophil membranes improves the efficiency of their internalization by chondrocytes; (4) NM-NP-siIL33 effectively inhibits the progression of OA in a murine model. These findings provide robust evidence that NM-NP-siIL33 may represent a promising therapeutic strategy for managing OA.

The onset of OA is significantly correlated with aging. However, the direct relationship between aging and the pathogenesis of OA remains incompletely understood. Emerging evidence suggests that chondrocyte senescence is one of the primary pathological mechanisms underlying OA [20, 21]. T Senescent chondrocytes exhibit characteristic features such as proliferation arrest, DNA damage, and elevated expression of senescence-associated proteins and reactive oxygen species (ROS) [22]. In this study, we observed aggregates of senescent chondrocytes in the knee joints of both young (3-month-old) and old (18-month-old) mice following anterior cruciate ligament transection (ACLT) surgery. Further analysis of senescence-associated proteins, damaged DNA, and ROS levels in joint tissues revealed that the accumulation of senescent chondrocytes was significantly more pronounced in older mice. In addition, senescent cells influence the homeostasis and functionality of adjacent cells through the secretion of the senescence-associated secretory phenotype(SASP) [10]. Our findings confirm that the expression of the SASP is significantly elevated in the joint tissues of older mice undergoing ACLT surgery. Furthermore, histological examinations reveal a greater degree of synovial inflammation in older mice compared to their younger counterparts. These results indicate that senescent chondrocytes play a critical role in articular cartilage degradation and the initiation of synovial inflammation.

IL-33 is a cytokine belonging to the IL-1 family and is expressed in various cell types, including endothelial cells, cardiomyocytes, fibroblasts and chondrocytes. Recent studies have demonstrated that IL-33 is involved in multiple pathological conditions, including lung fibrosis, cancer, and cardiovascular disease [14, 23, 24]. This involvement occurs through the IL33/ST2 signaling pathway, which is activated upon IL-33 binding to the transmembrane receptor ST2L. For instance, a recent study reported that IL-33 promotes chronic obstructive pulmonary disease (COPD) via an ST2-independent RAGE/EGFR signaling pathway [25]. However, the precise molecular mechanisms by which IL33/ST2 influences chondrocyte senescence and potentially facilitates the progression of OA remain to be fully elucidated. Previous studies have established that autophagy dysfunction contributes to cellular senescence promotion [26]. Our findings align with these observations, demonstrating that IL-33 inhibits autophagy, thereby promoting chondrocyte senescence through the activation of the p38 MAPK signaling pathway. Furthermore, the application of the autophagy agonist Adezmapimod partially mitigates the senescence-promoting effects of IL-33 on chondrocytes. These results indicate that the mechanism through which IL-33 facilitates chondrocyte senescence involves the regulation of p38 MAPK-mediated autophagy dysfunction.

Various gene therapy agents hold potential for addressing OA at the cellular level. However, their clinical application is limited due to challenges in maintaining an effective therapeutic concentration at the lesion site through systemic administration. Articular cartilage, being a highly differentiated tissue with limited vascularity, poses additional challenges. Surface-modified nanoparticle delivery systems offer significant potential in overcoming these limitations [27]. For instance, Li et al. synthesized rapamycin-loaded, ROS-responsive polymer-modified NPs. These NPs effectively mitigate the progression of OA by promoting intracellular autophagy, thereby successfully repolarizing M1 macrophages into the M2 phenotype [28]. In this study, we employed neutrophil membrane encapsulation of NPs loaded with siIL33. In vitro experiments demonstrated that the modification of NPs with neutrophil membranes significantly enhanced their binding efficiency to chondrocytes. Furthermore, in vivo imaging revealed that neutrophil membrane encapsulation prolonged the residence time of NPs within the joint. Subsequent in vivo experiments confirmed that the encapsulation of neutrophil membranes significantly enhanced the therapeutic efficacy of siIL33-loaded NPs in the treatment of OA.

Herein, we developed a NPs platform which facilitates the delivery of siRNA to OAs. These NPs mitigate articular cartilage damage in murine models following ACLT. Mechanistically, these NPs play a crucial role in preventing the chondrocytes senescence by restoring their impaired autophagy, thereby mitigating the progression of osteoarthritis OA.

Conclusion

This study identifies IL-33 as a key driver of chondrocyte senescence and osteoarthritis (OA) progression through p38 MAPK-mediated autophagy suppression. We engineered a new neutrophil membrane-coated siRNA nanoparticles (NM-NP-siIL33) to overcome articular delivery challenges and attenuate OA progression in ACLT mice. These findings contribute to a deeper understanding of the mechanisms involved in OA pathogenesis and provide a novel pharmacological therapeutic strategy for OA treatment.

Author contributions

Jian Chen and Zeze Fu wrote the main manuscript text. Jiahao Chen prepared Figs. 1, 2, 3 and 4 and DengShuo Sun prepared figures 5-8. Siqi Zhang and Jian Chen did all the experiments. All authors reviewed the manuscript.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Competing interests

The authors declare no competing interests.