Cuttlefish ink nanoparticle hydrogels ameliorate osteoarthritis via concurrent inhibition of macrophage pyroptosis and chondrocyte senescence

Abstract

Osteoarthritis (OA) is a debilitating joint disorder characterized by synovial inflammation and cartilage degradation; however, current therapeutic options fail to simultaneously address these dual pathological progressions. Here, we present clinical evidence identifying synovial macrophage (Mφ) pyroptosis and chondrocyte senescence as pivotal features of OA, which amplify joint damage through sustained inflammatory cascades and metabolic dysfunction. To address this, we developed cuttlefish ink-derived melanin nanoparticle (CIMNP) hydrogels designed to target the ROS/HIF-1α/glycolysis axis, thereby concurrently mitigating synovial Mφ pyroptosis and chondrocyte senescence. The CIMNP hydrogels exhibit potent ROS-scavenging capacity and shear-thinning injectability, making them suitable for intra-articular delivery. In vitro, CIMNP/Alg hydrogels suppressed ROS-induced activation of the HIF-1α/glycolysis axis in Mφs, thereby attenuating inflammasome activation, reducing pore formation, and preventing pyroptotic cell death. Meanwhile, the hydrogels alleviated the ROS/HIF-1α/glycolysis-driven impairment of anabolic marker expression and counteracted the senescence phenotype in chondrocytes. In vivo, intra-articular delivery of CIMNP/Alg hydrogels synchronously suppressed ROS-driven HIF-1α/glycolytic signaling in both synovial and cartilage tissues, leading to inhibition of Mφ pyroptosis, alleviation of chondrocyte senescence, and mitigation of synovitis and cartilage erosion. This coordinated regulation of inflammation and senescence further improved subchondral bone microarchitecture and restored joint homeostasis in OA rats. In sum, this study demonstrates a simple yet effective metabolic reprogramming strategy against OA, with promising potential for treating other degenerative diseases, thereby paves the way for its broader clinical translation.

Graphical Abstract

Cuttlefish ink nanoparticle hydrogels alleviate osteoarthritis via targeting metabolic ROS/HIF-1α/glycolysis axis to inhibit macrophage pyroptosis and chondrocyte senescence.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12951-026-04396-0.

Introduction

Osteoarthritis (OA) is a chronic and degenerative joint disease, affecting over 595 million people globally with a projected increase to 1 billion by 2050 [1, 2]. OA presents clinical manifestations such as joint pain, functional impairment, and progressive structural damage, which are closely linked to its pathological feature of whole-joint tissue degeneration-prominently involving synovial inflammation and cartilage matrix degradation [3–5]. Clinically, intra-articular therapies such as hyaluronic acid injections are commonly used in early-to-moderate osteoarthritis to relieve pain and improve joint lubrication; however, their therapeutic efficacy remains largely symptomatic and does not effectively halt disease progression [6]. For patients with advanced OA, surgical interventions including microfracture, autologous chondrocyte implantation, and the use of biomaterial- or scaffold-based implants have been increasingly explored to support cartilage repair [7]. Nevertheless, these strategies are often associated with limitations such as invasive procedures, inconsistent clinical outcomes, limited durability of repair tissue, and high costs. Collectively, these limitations highlight the urgent need for more effective and durable therapeutic approaches capable of modifying disease progression.

In synovial inflammation, activated macrophages (Mφs) exacerbate inflammatory responses by releasing proinflammatory mediators, further promoting joint structural damage [5, 8, 9]. As a pivotal pathogenesis in OA progression, synovitis implicates intricate interactions between multi-immune cells and local articular microenvironment, characterized by joint pain and cartilage structural damage [3, 10]. Extensively proliferated and infiltrated Mφs perpetuate the synovial inflammation, and the persistently activated phenotype of synovial Mφs correlate strongly with cartilage deterioration and OA severity [5, 9]. Pyroptosis is a death pathway of Mφs that triggered by pathological signals derived from pathogen-associated molecular patterns (PAMPs) or danger-associated molecular patterns (DAMPs), specifically characterized by NLRP3-based inflammasome self-formation, Cl-Caspase-1 release, GSDMD cleavage, membrane pore generation and inflamed cytokines escape, such as IL-1β and IL-18 [11, 12]. By orchestrating the release of pro-inflammatory mediators, including cytokines and chemokines, pyroptotic Mφs create a cascade that amplifies synovitis, cartilage matrix degradation, and joint damage perpetuation [9, 13–15]. Therefore, the efficient reversal of Mφ pyroptosis remains a major therapeutic challenge in OA.

Cartilage degradation primarily results from dysfunctional chondrocytes, whose impaired ECM metabolic homeostasis disrupts matrix turnover and accelerates tissue deterioration [8, 9, 16]. Emerging evidence indicated that several hallmarks exacerbate cartilage erosion, including epigenetic dysregulation, metabolic sensing impairment, mitochondrial damage, cellular senescence, and disrupted intercellular signaling [17]. As the typically phenotypic changes of stromal chondrocytes, senescence, is hallmarked by the following features, implicating cyclin dependent kinase inhibitor induced irreversible cell cycle arrest, pro-inflammatory molecules secretory patten changes formulated senescence-associated secretory phenotype (SASP), energy dysregulation, and redox disequilibrium medicated metabolic disorder [18, 19]. The SASPs of senescent chondrocytes containing diversiform pro-inflamed mediators, such as iNOS, COX2, ILs, ADAMTS, and MMPs, amplifies localized inflammation and reinforces the pathological cycle in cavum articulare [4, 20]. In addition, aging phenotypic chondrocytes play indispensable role in shifting ECM metabolic homeostasis variation from anabolism to catabolism via perpetuating SASP generation [9, 13]. These findings highlight that targeting chondrocyte senescence represents an effective strategy to restore cartilage homeostasis and promote OA repair.

Ascribe the propelling effect of synovitis and cartilage erosion to the pathogenic interplay of Mφ pyroptosis and chondrocyte senescence, unearthing upstream metabolic nodes that orchestrate pyroptosis and aging is pivotal for achieving OA reversal [16, 21]. Thus, there is an urgent need to identify upstream regulatory strategies capable of simultaneously modulating Mφ pyroptosis and chondrocyte senescence. As a conserved cytosolic pathway that splits glucose into pyruvate with lactate as an anaerobic product, glycolysis is notably heightened in pathological joint microenvironments, particularly within OA progression [10, 22]. Perpetuated synovial inflammation and cartilage degradation were occurred in glycolytic metabolism upregulated OA, through mediating abnormal activation of several pathological signal pathways [10, 15]. Glycolytic upregulation in OA orchestrates the form of pyroptosis phenotypic synovial Mφs by facilitating NLRP3 inflammasome activation through accumulated metabolic intermediates (e.g., lactate) and energy metabolism imbalance, thereby propagating synovial inflammatory response [8, 11, 23]. In addition, the shaping of senescence phenotypic chondrocytes is synchronously participated by perpetuated glycolysis via the analogously metabolic mechanism, precipitating the expression of SASP related mediator [11, 15, 16, 24]. Collectively, based on the simultaneously modulation role in Mφ pyroptosis and chondrocyte senescence, glycolysis could act an essential pointcut for formulating the coordinate therapeutic strategy against synovitis and cartilage corrosion [10, 23].

Emerging evidences have manifested the reactive oxygen species (ROS) mediated HIF-1α axis as a pivotal upstream modulator governing glycolysis in OA. Activated ROS/HIF-1α signal initiated by redox imbalance, subsequently propels the recalibration of glycolytic flux through orchestrating the functional status of rate-limiting enzymes [10, 16]. Cuttlefish ink-derived melanin nanoparticles (CIMNPs) exhibit inherent biocompatibility, satisfying degradability, and efficient capacity of multi-radicals neutralization ascribed to the abundant phenolic hydroxyl groups anchored on particle surface [19]. As a natural polysaccharide from ocean, sodium alginate (Alg) possess the features of economy, biosafety, and biocompatibility, can form a continuous three-dimensional network for hydrogel matrix by cation cross-linked method [13]. Melanin and melanin-like nanomaterials possess abundant catechol/quinone moieties, enabling broad-spectrum ROS/RNS scavenging and intrinsic redox-buffering properties that distinguish them from conventional antioxidants [19, 25, 26]. Cuttlefish ink-derived melanin nanoparticles, CIMNPs, sourced from abundant marine resources, exhibit favorable biocompatibility and biodegradability, making them promising for long-term intra-articular applications [27–30]. Accumulating evidence demonstrated that CIMNPs exhibited robust and sustained antioxidant activity, effectively mitigating oxidative stress in diverse pathological microenvironments [27–30]. Alg is a biocompatible polysaccharide widely used to construct injectable Ca²⁺-crosslinked hydrogels for intra-articular drug delivery [19, 31]. Alg-based hydrogels have shown therapeutic potential in osteoarthritis by mitigating synovial inflammation and protecting cartilage [32–34].

In our study, CIMNP/Alg gel was fabrication via incorporating centrifugation-prepared CIMNPs into Alg solution, and subsequently generating Ca²⁺ mediated crosslinking by dropwise adding calcium gluconate solution refer to prior report [35]. We then determined the joint therapeutic effect of CIMNP/Alg gels on synovial pyroptosis and cartilaginous aging by targeting ROS/HIF-1α/glycolysis axis and ultimately alleviating pathological symptom of OA.

Results

The joint of patients with OA is characteristic of synovial Mφ pyroptosis and chondrocyte senescence

Given the complex and multi-factorial etiology of OA, it is vital to explicitly investigate the specifically pathological mechanism for fabricating high efficiently therapeutic platform against OA. Therefore, healthy and OA articular tissues from clinical samples were collected to comprehensively dissect the pathological discrepancy. As shown in Fig. 1A, inflammatory cell infiltration occurred and the synovitis score obviously escalated by 10.50-fold in OA articular synovium (P < 0.0001), as determined by H&E [36]. ASC functions as an essential adaptor that nucleates the assembly of the NLRP3 inflammasome, whose activation subsequently drives caspase-1–dependent maturation and release of IL-1β and IL-18, thereby executing the canonical pyroptotic program [37]. Furthermore, the positive cell rates of ASC, NLRP3, IL-18, and IL-1β significantly increased in the articular synovium of OA patients, by approximately 3.90, 6.13, 4.68, and 7.66-fold, respectively, compared to normal group (P < 0.0001) (Fig. 1B-E). Specifically, the statistical MFI of ROS was prominently upregulated in the OA synovium, showing a 5.43-fold increase compared with healthy tissues (P < 0.0001) (Fig. 1F). Protein expression levels of NLRP3 (P < 0.01, 0.58-fold), ASC (P < 0.0001, 0.61-fold), Cl-Caspase-1 (P < 0.001, 3.24-fold), GSDMD-N (P < 0.0001, 0.67-fold), IL-1β (P < 0.0001, 3.61-fold), IL-18 (P < 0.0001, 0.51-fold), TNF-α (P < 0.0001, 0.59-fold), ADAMTS-4 (P < 0.0001, 1.37-fold), ADAMTS-5 (P < 0.0001, 4.21-fold), MMP1 (P < 0.001, 0.80-fold), MMP3 (P < 0.0001, 2.12-fold), MMP9 (P < 0.01, 0.54-fold), MMP13 (P < 0.001, 0.79-fold), and HMGB1 (P < 0.0001, 1.87-fold) were dramatically increased in OA synovium, which was closely linked with synovial inflammation (Fig. 1G-J). The cartilages from OA tissues revealed acute discontinuity, denudation of the surface, and diminution of extracellular matrix, as determined by H&E, SO&FG, and TB staining (Fig. 1K). Further analysis revealed that protein expression level of p21, p16, γ-H2AX, HMGB1, iNOS, COX-2, IL-1β, IL-6, ADAMTS-4, ADANTS-5, MMP1, MMP3, MMP9 and MMP13 was markedly attenuated by 2.44 (P < 0.001), 0.75 (P < 0.0001), 4.28 (P < 0.001), 1.26 (P < 0.0001), 0.36 (P < 0.1), 1.17 (P < 0.001), 6.75 (P < 0.0001), 0.53 (P < 0.1), 0.93 (P < 0.0001), 2.00 (P < 0.001), 0.65 (P < 0.1), 2.57 (P < 0.0001), 2.09 (P < 0.0001), 3.35 (P < 0.0001) -fold in OA articular cartilage (Fig. 1L-O). Protein expression level of Cyclin D1, CDK4, CDK6, p-Rb/Rb, and E2F1 were 0.35 (P < 0.1), 0.43 (P < 0.001), 0.42 (P < 0.001), 0.63 (P < 0.0001), 0.47 (P < 0.001) -fold decrease than healthy group, respectively, demonstrating an exacerbation of chondrocyte senescence in osteoarthritic cartilage (Fig. 1L-O). Briefly, pyroptosis of Mφs and senescence of chondrocytes may serve as the indispensable mediators to perpetuate synovitis as well as cartilage erosion, and eventually accelerate the OA progression.

Fig. 1.

Pathological disparity between OA and healthy knee joints. (A) Representative images and synovitis score of H&E staining between healthy and OA articular synovium (n = 6). (B-E): Representative images of immunohistochemical staining and positive cell rates of ASC, NLRP-3, IL-18 and IL-1β between healthy and OA articular synovium (n = 3). (F) ROS levels in healthy articular synovium and OA articular synovium were observed by fluorescence microscope (n = 6). (G-J) WB analysis of relative protein level to NLRP3, ASC, cleaved caspase-1, GSDMD-N, IL-1β, IL-18, TNF-α, ADAMTS-4, ADAMTS-5, MMP1, M-MP3, MMP9, MMP13 and HMGB1of cartilage tissues with the same treatment (n = 10). (K) Representative images of H&E, SO&FG, and TB staining between healthy and OA articular cartilage. (L-O) WB analysis of relative protein level of p21, p16, cyclin D1, CDK4, CDK6, Rb, p-Rb, E2F1, γ-H2AX, HMGB1, iNOS, COX-2, IL-1β, IL-6, ADAMTS-4, ADAMTS-5, MMP1, MMP3, MMP9 and MMP13 between healthy and OA articular cartilage (n = 10). Data are presented as mean ± SD. ^*P < 0.05, ^**P < 0.01, ^***P < 0.001, ^****P < 0.0001, vs. the indicated groups

Preparation and characterization of CIMNP hydrogels

The CIMNPs and sodium alginate were extracted from cuttlefish ink sacs and seaweed, respectively. The CIMNP/Alg hydrogels were fabricated by ionically crosslinking sodium alginate with calcium gluconate to encapsulate CIMNPs (Fig. 2A). The morphology of CIMNPs was characterized using scanning electron microscopy (SEM) and transmission electron microscopy (TEM) (Fig. 2B and C). The CIMNPs exhibited a spherical morphology, a monodisperse size distribution, and an average diameter of about 150 nm. The hydrodynamic size distribution of CIMNPs was similar to the results from SEM and TEM (Fig. 2D). Dynamic light scattering revealed that CIMNPs possessed a hydrodynamic size of ~187 nm (PDI 0.126) with a moderately negative surface charge (−27.3 mV). In addition, the CIMNPs were dispersed in water, normal saline, and culture medium to investigate their dimensional stability (Fig. 2E), which revealed that they were highly stable and manifested no signs of obvious aggregation in all three solutions over a seven-day period.

Fig. 2.

Preparation and characterization of CIMNP/Alg gels. (A) Preparation process of CIMNP/Alg gels. (B) Representative SEM image of CIMNPs (scale bar = 200 nm). (C) Representative TEM image of CIMNPs (scale bar = 200 nm). (D) Hydrodynamic size distribution of CIMNPs. (E) The stability of size of CIMNPs in water, saline, and medium during 7 days. (F) Representative SEM images of Alg gels, and CIMNP/Alg gels (scale bar = 50 μm). (G) The photographs of CIMNPs, Alg gels and the CIMNP/Alg gels. (H) DPPH radical scavenging capacity of CIMNP/Alg gels with increasing CIMNPs concentrations (0.05, 0.1, 0.2, 0.5, and 1.0 mg/mL). (I-L) The rheology behavior, (I) Storage modulus and loss modulus of Alg gels, (J) Storage modulus and loss modulus of CIMNP/Alg gels, (K) Viscosity versus shear rate, (L) Viscoelastic response of Alg gels and CIMNP/Alg gels. Data are presented as mean ± SD (n = 3). ^**P < 0.01^***P < 0.001, ^****P < 0.0001, vs. the indicated groups

Then, the Alg gel and CIMNP/Alg gel were observed by SEM (Fig. 2F). Loose porous structures at the microscopic level were observed in the Alg gel, and the enlarged image showed that the surface of the Alg gel was relatively uniform and flat, while a large number of small protrusions were exhibited on the surface of CIMNP/Alg gel, and the surface of the CIMNP/Alg gels displayed more uniform and delicate in pore distribution by the incorporation of CIMNPs, compared with Alg gels (Fig. 2F). As illustrated in Fig. 2G, the CIMNP solution with feature of black-colored transparent was well-dispersed in the Alg gel and formed homogeneous black gel. Moreover, the DPPH radical scavenging capacity of CIMNPs enhanced progressively with increasing concentration (Fig. 2H). The rheology behavior of Alg gel and CIMNP/Alg gel were performed using rotational rheometer (Fig. 2I-L). The storage modulus and loss modulus of Alg gels and CIMNP/Alg gels were determined by rheological evaluation (Fig. 2I-J). In all tested frequencies, the storage modulus of Alg gels was changed from 1130.18 to 1706.09 Pa, while the storage modulus of CIMNP/Alg gels increased from 1930.99 to 2522.81 Pa. Both Alg gels and CIMNP/Alg gels exhibited a frequency-dependent increase in storage modulus, and the storage modulus consistently exceeded the loss modulus, confirmed the hydrogels maintained a predominantly elastic, solid-like mechanical property (Fig. 2I-J). The viscosity of Alg gels was observed to decrease as the shear rate was increased (Fig. 2K), demonstrating the shear-thinning behavior that enables injectability. In addition, the curves of CIMNP/Alg gels and Alg gels were nearly overlapped, indicating that the viscosity properties of the hydrogels were negligibly influenced after CIMNP incorporation (Fig. 2K). In the low-strain region, both gels displayed linear viscoelastic behavior with stable ionic crosslinked networks, while the CIMNP/Alg gel exhibited higher moduli than the Alg gel. At higher strain, the decline of both moduli indicated network yielding and shear-thinning characteristics, confirming the injectability of the CIMNP/Alg gel (Fig. 2L). Overall, the CIMNP/Alg gels exhibit excellent resistance to oxidative stress and favourable injectability was obtained.

CIMNP/Alg gels protect Mφs against ROS mediated HIF-1α and glycolysis signal pathway to inhibit pyroptosis

The crucial effect of abnormal ROS and pyroptosis in the synovitis progression was observed from clinal healthy and OA articular synovium (Fig. 1A-J). Then, rat peritoneal Mφs were harvested by peritoneal lavage and subsequently used for in vitro evaluation of the anti-pyroptotic effects of the hydrogels. CIMNP/Alg gels demonstrated good safety toward Mφs at a concentration of 500 µg/mL for the CIMNPs (Fig. S1A). Accordingly, a series of in vitro experiments using rat peritoneal Mφs were performed to investigate the effect of CIMNP/Alg gels on the ROS/HIF-1α/glycolysis signaling pathway and pyroptosis under H₂O₂ treatment. Primarily, we detected ROS levels using a DCFH-DA probe and quantified via fluorescence imaging and flow cytometry (FCM) (Fig. 3A-B). Compared to the control group, the H₂O₂ group showed a sharp increase in MFI (3.18-fold in images and 1.42-fold in FCM; P < 0.001), confirming oxidative damage. The results in H₂O₂+Alg group were similar to the H₂O₂ group. In contrast, the CIMNP/Alg gel treatment markedly reduced intracellular ROS compared to the H₂O₂ group (P < 0.001), which was comparable to that of the NAC positive control (Fig. 4A). In addition, previous studies have demonstrated that abnormally elevated HIF-1α levels were mediated by excessive ROS accumulation under pathological conditions [38]. Western blot (WB) analysis revealed a 3.45-fold increase in HIF-1α accumulation in the H₂O₂ group compared with the control group (P < 0.001) (Fig. 3C). Notably, HIF-1α expression in the H₂O₂+CIMNP/Alg group was significantly attenuated compared to both the H₂O₂ group and the H₂O₂+Alg group, with reductions of 0.75-fold and 0.76-fold (P < 0.001), respectively. The minimal difference between the H₂O₂+CIMNP/Alg group and the H₂O₂+NAC group further suggests comparable efficacy between CIMNPs and NAC (Fig. 3C). Given that the glycolytic pathway can be modulated by HIF-1α, we also assessed the expression of key glycolysis-related proteins (HK2, PFK1, and PKM2). WB results indicated significant upregulation of HK2, PFK1, and PKM2 in the H₂O₂ group relative to the control, with increases of 5.30-fold (P < 0.0001), 2.74-fold (P < 0.001), and 3.48-fold (P < 0.001), respectively, suggesting enhanced glycolytic flux (Fig. 3C). In the H₂O₂+CIMNP/Alg group, the expression levels of HK2, PFK1, and PKM2 were markedly reduced by 0.79-fold (P < 0.0001), 0.69-fold (P < 0.001), and 0.55-fold (P < 0.001), respectively, relative to the H₂O₂ group (Fig. 3C). Both the H₂O₂+NAC and H₂O₂+CIMNP/Alg groups exhibited protein levels nearly restored to those under normal physiological conditions (Fig. 3C).

Fig. 3.

The ROS mediated HIF-1α and glycolysis signal pathway and pyroptosis adjustment role of CIMNP/Alg gels on Mφs treated with H₂O₂. (A) Representative fluorescence images and MFI of ROS in Mφs after treated with Control, H₂O₂ (300 μM), H₂O₂ (300 μM) +NAC (N-Acetyl-L-cysteine, 5 mM), H₂O₂ (300 μM) +Alg and H₂O₂ (300 μM) +CIMNP/Alg (CIMNPs, 200 μg/mL) for 24 h. (B) Flow cytometry analysis and MFI of ROS in Mφs with the same treatment. (C) WB analysis of relative protein expression level to HIF-1α, HK2, PFK1 and PKM2 in Mφs with the same treatment. (D) Representative SEM images and quantification of cell surface pore in Mφs with the same treatment. Red arrows indicated the pores on the cell surface. (E-G) WB analysis of ASC, NLRP3, Pro-Caspase-1, Cl-Caspase-1, GSDMD-N, Pro-IL-1β, IL-1β and IL-18 in Mφs with the same treatment. Data are presented as mean ± SD (n = 3). ^**P < 0.01, ^***P < 0.001, ^****P < 0.0001, vs the indicated groups. NS, no significant difference

Fig. 4.

The ROS mediated HIF-1α and glycolysis signal pathway and senescence adjustment role of CIMNP/Alg gels on chondrocyte treated with TBHP. (A) Representative fluorescence images and MFI of intracellular ROS in chondrocytes after the control, TBHP (T-butyl hydroperoxide, 50 µM), TBHP (50 µM) + NAC (N-Acetyl-L-cysteine, 5 mM), TBHP (50 µM) + Alg, TBHP (50 µM) +CIMNP/Alg (CIMNPs, 200 µg/mL) treated for 24 h. (B) WB analysis of relative protein expression level to HIF-1α, HK2, PFK1 and PKM2 with the same treatment. (C) SA-β-gal staining and analysis in positive cell rates of chondrocytes with the same treatment. (D) WB analysis of relative protein expression level to p16, CDK4, CDK6, cyclin D1, p-Rb, Rb, E2F1 and γ-H2AX with the same treatment. (E-F) WB analysis of ADAMTS-5, MMP13, Sox9 and Col2al with the same treatment. Data are presented as mean ± SD (n = 3). ^**P < 0.01, ^***P < 0.001, ^****P < 0.0001, vs. the indicated groups. NS,no significant difference

SEM imaging enabled direct visualization of pyroptotic pores, with quantitative analysis of Mφs surfaces revealing that hydrogen peroxide exposure drastically increased pore formation (Fig. 3D). In spite of that, H₂O₂+CIMNP/Alg group illustrated a 0.83-fold reduction compared to H₂O₂ group (P < 0.0001). Furthermore, the expression levels of pyroptosis relevant proteins (ASC, NLRP3, Pro-Caspase-1, Cl-Caspase-1, GSDMD-N, Pro-IL-1β, IL-1β, and IL-18) in H₂O₂+CIMNP/Alg group validated pronounced decrease after CIMNP/Alg gels treated (Fig. 3E-G). Meanwhile, the cytokines level of IL-1β and IL-18 were subsequently reduced −2.73 and 0.09-fold by CIMNP/Alg gels, compared with H₂O₂ group (Fig. S2A-B). The above results demonstrated that CIMNP/Alg gels potently inhibit pyroptosis in Mφs. Collectively, CIMNP/Alg gels suppress the intracellular ROS/HIF-1α signaling pathway and protect Mφs from pyroptosis, suggesting its therapeutic potential for OA even in hostile pathological environments.

CIMNP/Alg gels reestablish HIF-1α/glycolysis aix in chondrocytes and reduce senescence via ROS scavenging

Above clinical findings highlight cellular senescence and subsequent SASP formation as pivotal pathological drivers of OA progression, suggesting that targeting chondrocyte senescence represents an effective therapeutic approach for OA management (Fig. 1K-O). Recent reports indicated that continuous glycolytic reprogramming contributes to the establishment of chondrocyte senescence through comparable metabolic alterations, ultimately inducing the elevated secretion of SASP components [39]. In addition, several evidence identified the ROS-mediated HIF-1α axis as a key upstream regulator of glycolytic reprogramming in OA, wherein redox imbalance–induced activation of this pathway recalibrates glycolytic flux by modulating rate-limiting enzymes. Accordingly, we sought to investigate the feasibility of CIMNP/Alg gels in modulating chondrocyte anti-senescence through the ROS/HIF-1α/glycolysis axis. First, CIMNP/Alg gels exhibited none cytotoxicity in chondrocytes at concentration of CIMNPs up on to 500 µg/mL (Fig. S1B). TBHP is widely employed as a classical inducer of oxidative stress mediated cellular senescence, serving as a reliable model compound for mimicking redox imbalance driven aging in vitro. Then, we investigated the protective effect of CIMNP/Alg gels on aged chondrocytes induced by TBHP. In order to determine the optimal concentration of TBHP for inducing cellular senescence, we conducted experiments by treating cells with a series of concentrations of TBHP (0, 12.5, 25, and 50 µM) further detect relative protein expression of key markers in senescence (p16, γ-H2AX, and MMP3) and cartilage-protective factors (Sox9 and Col2a1), ultimately selected a concentration of 50 µM (Fig. S3A-F). Fluorescence imaging with DCFH-DA probe demonstrated that TBHP stimulation increased intracellular ROS by approximately 4.33 -fold compared with the control, confirming severe oxidative stress in rat articular chondrocytes. In contrast, the CIMNP/Alg gel treatment reduced ROS levels by about 0.56 -fold relative to the TBHP group, exhibiting antioxidative efficacy comparable to that of the NAC positive control. In addition, pathological dysregulation of ROS triggers a HIF-1α surge, which in turn modulates glycolytic flux. Consistent with this, our WB analysis revealed that TBHP exposure significantly increased HIF-1α levels by 2.51-fold. While the Alg group showed no significant effect, the CIMNP/Alg treatment markedly reduced HIF-1α, achieving 0.67-fold and 0.64-fold decreases relative to the TBHP and TBHP + Alg groups, respectively (Fig. 4B). Consequently, the levels of glycolytic enzymes (HK2, PFK1, PKM2) were also markedly elevated by TBHP (3.06-fold, 2.73-fold, and 2.66-fold). Both CIMNP/Alg and NAC treatments effectively reversed these changes, restoring expression levels to near baseline (Fig. 4B).

Glycolysis facilitates cellular senescence progression. SA-β-gal staining and positive cell rates of chondrocytes manifested a 0.79 -fold downregulation in TBHP group, after treated with CIMNP/Alg gels (P < 0.0001) (Fig. 4C). Lamin B1 is the classical marker of cell senescence, upregulating during aging due to nuclear envelope stress and altered chromatin organization [40]. Quantitative analysis revealed that the anomalous mRNA expression of Lamin B1 and p21 in the TBHP group was rectified upon CIMNP/Alg treatment by 5.31 -fold increase and 0.74 -fold decrease, respectively (P < 0.0001, P < 0.001) (Fig. S4A-B). The attenuation of cellular senescence was evidenced by a marked decrease in the key markers p16 (0.56-fold, P < 0.0001) and γ-H2AX (0.69-fold, P < 0.0001). This coincided with a pronounced upregulation of cell cycle promoters, with CDK4, CDK6, Cyclin D1, the p-Rb/Rb ratio, and E2F1 increasing by 0.45-fold (P < 0.01), 0.59-fold (P < 0.01), 1.53-fold (P < 0.001), 0.62-fold (P < 0.0001), and 0.60-fold (P < 0.05), respectively (Fig. 4D). Overall, CIMNP/Alg gels significantly reduced senescence markers, underscoring its strong anti-senescence potential. This effect was further supported by WB analysis, which revealed that the TBHP+CIMNP/Alg treatment consistently downregulated key catabolic markers (ADAMTS-5, MMP1, MMP3, MMP9, and MMP13) while upregulating cartilage-protective factors (Sox9 and Col2a1) (Figs. 4E-F and S4C-D). Together, these integrated analyses demonstrate that the CIMNP/Algs attenuate oxidative stress, normalizes HIF-1α signaling, and reverses senescence-associated metabolic reprogramming in chondrocytes, highlighting its multi-faceted therapeutic promise for OA under pathological conditions.

Inhibition of HIF-1α-dependent glycolysis mediates the protective effects of CIMNP/Alg gels against Mφs pyroptosis and chondrocytes senescence

We performed rescue experiments using the HIF-1α activator CoCl₂ to determine whether the protective effects of CIMNP/Alg gels could be inhibited and to clarify the role of HIF-1α in regulating downstream glycolytic metabolism and associated cellular phenotypes. As shown in Fig. 5A–B, WB analysis demonstrated that the inhibitory effects of CIMNP/Alg gels on glycolytic enzymes (HK2, PKM2) and pyroptosis-associated markers (NLRP3, IL-1β) in Mφs were partially abrogated upon HIF-1α activation by CoCl₂ (both P < 0.0001). Similarly, in chondrocytes, the suppressive effects of CIMNP/Alg gels on glycolysis (HK2, PKM2, P < 0.0001, P < 0.01) and senescence-associated markers (p16, Sox9, P < 0.05, P < 0.01) were also significantly attenuated in the presence of CoCl₂ (Fig. 5C–D). These results suggested that CIMNP/Alg gels exhibited HIF-1α-dependent anti-glycolytic, anti-pyroptotic, and anti-senescent effects. As shown in Fig. 5E–F, WB analysis demonstrated that ROS damage model combined with 2-DG treatment markedly reduced the expression of pyroptosis-related markers (NLRP3, IL-1β, P < 0.0001, P < 0.0001) in Mφs and senescence-associated markers (p16, Sox9, P < 0.0001, P < 0.01) in chondrocytes. Notably, CIMNP/Alg gel treatment produced effects comparable to those observed with 2-DG, including significant suppression of these markers.

Fig. 5.

CIMNP/Alg gels attenuated Mφs pyroptosis and chondrocyte senescence via inhibiting HIF-1α/glycolysis axis. (A–B) WB analysis of HK2, PKM2, NLRP3, and IL-1β levels in CIMNP/Alg gels combined with HIF-1α inhibitor (CoCl₂) treated Mφs under H₂O₂ condition. (C–D) WB analysis of HK2, PKM2, p16, and Sox9 levels in CIMNP/Alg gels combined with HIF-1α inhibitor (CoCl₂) treated Mφs under TBHP condition. (E) WB analysis of NLRP3 and IL-1β levels in CIMNP/Alg gels or glycolysis inhibitor (2-DG) treated Mφs under H₂O₂ condition. (F) WB analysis of p16 and Sox9 levels in CIMNP/Alg gels or glycolysis inhibitor (2-DG) treated Mφs under TBHP condition. Data are presented as mean ± SD. ^*P < 0.05, ^**P < 0.01, ^***P < 0.001, ^****P < 0.0001, vs the indicated groups. NS, no significant difference

Collectively, these results indicated that by suppressing HIF-1α-dependent glycolytic reprogramming, CIMNP/Alg gels effectively attenuated downstream inflammatory and senescence-associated phenotypes under oxidative stress conditions.

ClMNP/Alg gels alleviate subchondral bone microarchitecture and reduce osteoarthritic degeneration

Macroscopic evaluation of the joints revealed clear treatment-dependent differences in cartilage integrity (Fig. S5A). A smooth and continuous articular surface was preserved in sham group, suggesting capsulotomy alone was not sufficient to induce detectable cartilage damage. However, extensive surface erosion and large defects, particularly on the medial tibial plateau were observed in saline group, reflecting the severe mechanical abrasion resulted from joint destabilization. Although partial preservation of the articular surface, mild-to-moderate fissures and surface roughening were sustained in the DXM group, indicating that progressive degeneration of cartilage was not reversed. Moderate protection was achieved after Alg gel treated, where smoother joint surfaces with fewer defect compared with saline group, implying that lubrication provided by the hydrogel mitigated friction-induced wear. By contrast, minimal erosion and clear signs of tissue repair were observed in CIMNP/Alg gel group, not only reduced cartilage wear but also promoted structural restoration, exhibiting the most pronounced chondroprotective effect. Collectively, these macroscopic findings demonstrate that CIMNP/Alg gels afforded superior protection against OA-related cartilage degeneration compared with DXM or Alg gel alone (Fig. S5A). Except articular cartilage surface exacerbation, subchondral bone changes are considered as an indispensable clinical indication in the pathologic progress of OA. We therefore performed micro-CT to assess the microarchitectural alterations in the subchondral bone of Wistar rats treated with ClMNP/Alg gel following modified Hulth’s surgery. As shown in the micro-CT 2D images of knee joints (Fig. 6A), the saline group exhibited subchondral bone sclerosis, trabecular malformation, and osteophyte formation following modified Hulth’s surgery compared with the sham group. In contrast, the DXM and ClMNP/Alg gel groups showed an opposite trend in 2D micro-CT imaging, characterized by reduced sclerosis and fewer osteophytes (Fig. 6A). Micro-CT 3D reconstruction further revealed prominent OA-related pathological changes in the saline group, including meniscal calcification, periarticular mineralization, narrowing of the articular cavity, and a rough cartilage surface, relative to the sham group (Fig. 6B). These features were markedly alleviated in the DXM and ClMNP/Alg gel groups (Fig. 6B). Quantitative analysis of the micro-CT 2D images was performed to assess trabecular bone microstructure. Compared with the sham group, the saline group showed significant reductions in bone volume fraction (BV/TV), trabecular number (Tb.N), and trabecular thickness (Tb.Th) by 0.29-fold (P < 0.001), 0.21-fold (P < 0.001), and 0.41-fold (P < 0.0001), respectively (Fig. S5B–D). The Alg gel group exhibited only minor increases in BV/TV and Tb.Th (0.15- and 0.19-fold, respectively) compared to the saline group (Fig. S5B–D). In contrast, both DXM and ClMNP/Alg gel treatment led to significant improvements in BV/TV (0.31- and 0.33-fold increases, respectively; P < 0.001 for both) (Fig. S5B). Similarly, Tb.N and Tb.Th were significantly improved in these two groups, with increases of 0.18- and 0.21-fold in Tb.N (both P < 0.01), and 0.58- and 0.55-fold in Tb.Th (both P < 0.001), respectively (Fig. S5C–D). Furthermore, the ClMNP/Alg gel group demonstrated significantly greater improvements in BV/TV, Tb.N, and Tb.Th compared to the Alg gel group, with increases of 0.19-fold (P < 0.01), 0.16-fold (P < 0.01), and 0.30-fold (P < 0.001), respectively, underscoring the specific role of ClMNPs in ameliorating subchondral bone sclerosis (Fig. S5B–D). In summary, ClMNP/Alg gels effectively reversed severe imaging manifestations of osteoarthritis in the joint cavity induced by modified Hulth’s surgery. These results highlight its ability to restore articular surface smoothness and improve subchondral bone architecture, particularly in mitigating cartilage degradation, osteophyte formation, and trabecular bone abnormalities.

Fig. 6.

Micro-CT imaging evaluation of CIMNP/Alg gels in vivo therapeutic effects. (A) Representative micro-CT 2D images of OA joints at the sagittal plane after treatment. Red arrows denote osteophytes. (B) Representative micro-CT bone remodeling 3D images of OA joints in different treatment groups. Red arrows denote osteophytes. (C) H&E, SO&FG, TB staining features of knee cartilage sections from Sham, Saline, DXM, Alg gel and CIMNP/Alg gel groups. (D) Analysis in depth of cartilage in the joint with different treatment. (E) Relative GAG content in the joint. (F) OARSI score of the joint. (G) Representative Sox9 immunohistochemical staining and quantitative positive cell rate analysis of articular cartilage across Sham, Saline, DXM, Alg gel, and CIMNP/Alg gel groups (n = 3). (H) Representative Col2a1 immunohistochemical staining and quantitative positive area rate analysis of articular cartilage across Sham, Saline, DXM, Alg gel, and CIMNP/Alg gel groups (n = 3). (I) WB analysis of Sox9 and Col2a1 with the different treatment. Data are presented as mean ± SD (n = 6). ^*P < 0.05, ^**P < 0.01, ^***P < 0.001, ^****P < 0.0001, vs the indicated groups. NS, no significant difference

Subsequently, cartilage histological lesions were evaluated using H&E, Safranin O & Fast Green (SO&FG), and Toluidine Blue (TB) staining. H&E staining revealed severe cartilage breakdown and disorganized chondrocytes in the saline group, indicating OA progression (Fig. 6C). Alg gel treatment slightly improved cartilage surface regularity. In contrast, DXM and CIMNP/Alg gel interventions resulted in varying degrees of enhancement in cartilage structural integrity and uniformity. Notably, the CIMNP/Alg gels were most effective in mitigating cartilage erosion, restoring extracellular matrix, reorganizing chondrocyte arrangement, and achieving a smooth articular surface (Fig. 6C). In SO&FG staining, anionic polysaccharides in articular cartilage (e.g., chondroitin sulfate and keratan sulfate) bind to Safranin O, producing a red color. In TB staining, cationic groups in the dye interact with anionic polysaccharides, yielding blue staining. Compared with the sham group, the saline group showed weaker Safranin O and TB staining, reduced chondrocyte density, marked glycosaminoglycan (GAG) depletion, cartilage denudation, surface roughness, and decreased cartilage thickness (Fig. 6C). Quantitative analysis indicated that cartilage thickness in the saline group was reduced by 0.85-fold compared to the sham group (P < 0.0001) (Fig. 6D). In contrast, the CIMNP/Alg gel group showed an approximately 6.07-fold increase relative to the saline group (P < 0.0001), while the DXM and Alg gel groups exhibited intermediate improvements (Fig. 6D). The relative GAG content in the saline group was significantly increased by 8.44-fold compared to the sham group (P < 0.0001), indicating severe articular cartilage damage. Conversely, the CIMNP/Alg gel group showed a 7.36-fold higher GAG content than the saline group (P < 0.0001), outperforming both DXM and Alg gel treatments (Fig. 6E). The OARSI score was improved by 1.70-fold in the CIMNP/Alg gel group compared to the saline group (P < 0.001), representing greater amelioration than other treatments (Fig. 6F). The relative ROS level was 0.65-fold lower in the CIMNP/Alg gel group than in the saline group (P < 0.001), also surpassing the effects of DXM and Alg gel (Fig. S5E). IHC staining showed that the positive cell rate of Sox9 in the CIMNP/Alg gel group was 3.35-fold higher than that in the saline group (P < 0.001), exceeding the levels in the DXM and Alg gel groups (Fig. 6G). Similarly, the positive area rate of Col2a1 was 5.47-fold higher in the CIMNP/Alg group than in the saline group (P < 0.001), underscoring the efficacy of CIMNP (Fig. 6H). As expected, WB analysis confirmed that the relative protein levels of Sox9 and Col2a1 in the CIMNP/Alg gel group were increased by 0.25-fold (P < 0.01) and 2.20-fold (P < 0.001), respectively, compared to the saline group (Fig. 6I and Fig. S5F–G). Additionally, the body weight of rats across all treatment groups remained comparable after 28 days, supporting the biosafety of CIMNP/Alg gels (Fig. S5H). In summary, these results demonstrate that CIMNP/Alg gels effectively reverse cartilage degeneration in experimental OA, highlighting its pronounced anti-erosion and cartilage-protective capacity.

CIMNP/Alg gels protect against synovitis connected with inhibition of the ROS/HIF-1α/glycolysis axis, thereby suppressing pyroptosis

Comparative examination of synovial tissues from healthy and OA joints revealed the significant involvement of the ROS, pyroptosis, and synovitis in OA pathogenesis (Fig. 1A-H). Thus, in vivo experiments were conducted to assess whether CIMNP/Alg gels modulate the ROS/HIF-1α/glycolysis axis and pyroptosis to inhibit synovitis. First, the ROS level was significantly reversed by CIMNP/Alg gels in OA synovium (Fig. S6A). As shown by WB analysis, HIF-1α expression was upregulated by 4.75-fold in the saline group versus the sham group (P < 0.001), whereas the Alg gel group showed no significant alteration compared to the saline control. (Fig. 7A). On the contrary, the expression of HIF-1α in CIMNP/Alg group manifested prominent attenuation relative to saline group, decreasing by 0.41 -fold (P < 0.01) (Fig. 7A). The marginal disparity between CIMNP/Alg gel group and DXM group also indicated the efficacy of CIMNPs (Fig. 7A). A reduction of 0.35 -fold in HIF-1α was observed in CIMNP/Alg gel group relative to Alg gel group (P < 0.01) (Fig. 7A). We next examined key glycolysis-related proteins (HK2, PFK1, PKM2) and found that their expression trends were consistent with that of HIF-1α. Specifically, the CIMNP/Alg gel group showed significant reductions in these proteins (0.79-, 0.31-, and 0.55-fold, respectively) compared to the Alg gel control (Fig. 7A).

Fig. 7 .

Metabolic reprogramming and inhibition of pyroptosis effects in synovitis tissues from OA rats. (A) WB analysis of relative protein expression level to HIF-1α, HK2, PFK1 and PKM2 among the Sham, DXM, Saline, Alg gel and CIMNP/Alg gel groups (n = 5). (B-C) Representative IHC images and positive cell rate of NLRP3 and Cl-Caspase-1 in the synovitis tissues (n = 3). (D) WB analysis of relative protein expression level to NLRP3, ASC, Pro-Caspase-1 and Cl-Caspase-1 in the synovitis tissues (n = 5). (E-F) Representative IHC images and positive cell rate of IL-18 and IL-1β in the synovitis tissues (n = 3). (G) WB analysis of relative protein expression level to GSDMD-N, Pro-IL-1β, IL-18 and IL-1β among the Sham, DXM, Saline, Alg gel and CIMNP/Alg gel groups (n = 5). Data are presented as mean ± SD. ^*P < 0.05, ^**P < 0.01, ^***P < 0.001, ^****P < 0.0001, vs the indicated groups. NS, no significant difference

Pyroptosis, which can be activated by glycolysis in synovial tissues, was subsequently investigated. qRT-PCR analysis demonstrated that NLRP3 mRNA expression was significantly upregulated by 9.04-fold in the sham group compared to the saline group (P < 0.001), while only minor changes were observed relative to the Alg gel group (Fig. S6B). In contrast, the CIMNP/Alg gel group exhibited a 0.73-fold downregulation in NLRP3 mRNA compared to the Alg gel group (P < 0.01) (Fig. S6B). Consistently, IHC analysis of NLRP3 showed a pronounced 0.73-fold decrease in positive staining in the CIMNP/Alg gel group versus the Alg gel group (P < 0.01) (Fig. 7B). Similarly, Cl-Caspase-1 IHC staining revealed a 0.62-fold reduction in the CIMNP/Alg gel group (P < 0.01), further supporting the inhibitory effect of CIMNPs on pyroptosis (Fig. 7C). The protein level of NLRP3, ASC, Pro-Caspase-1, and Cl-Caspase-1 were elevated by 1.83-, 3.35-, 1.85-, and 0.96-fold, respectively, in the CIMNP/Alg gel group compared to the Alg gel group (Fig. 7D). The CIMNP/Alg gel group also showed marked reductions in IL-18, with mRNA levels and IHC-positive cell proportions decreased by 0.67-fold (P < 0.001) and 0.76-fold (P < 0.05), respectively, relative to the Alg gel group (Fig. 7E, S6C). IL-1β mRNA expression was downregulated by 0.48-fold (P < 0.0001), and IHC analysis indicated a 0.80-fold decrease in IL-1β-positive cells (P < 0.05) (Fig. 7F, S6D). Further supporting these findings, WB analysis revealed that the CIMNP/Alg gel group significantly suppressed the protein levels of GSDMD-N, Pro-IL-1β, IL-18, and IL-1β by 0.43-, 0.62-, 0.62-, and 0.72-fold, respectively (Fig. 7G). ELISA results also confirmed pronounced reductions in the synovial tissue concentrations of IL-18, IL-1β, TNF-α, and IL-6 following CIMNP/Alg gel treatment (Fig. S6E–H). The ECM degradation enzymes in synovium such as MMP3 and MMP9 were obviously reduced after CIMNP/Alg gel treated, compared with Saline group (Fig. S7A-C). In conclusion, CIMNP/Alg gels attenuate synovitis by coordinately suppressing the ROS/HIF-1α pathway, modulating glycolytic metabolism, and subsequently inhibiting pyroptosis, demonstrating considerable therapeutic potential.

CIMNP/Alg gels mitigate cartilage damage associated with suppression of the ROS/HIF-1α/glycolysis/senescence cascade

The adverse contribution of cellular senescence to OA progression was verified by comparing healthy and osteoarthritic articular cartilage (Fig. 1L-O). Subsequently, the effect of CIMNP/Alg gels on the ROS/HIF-1α/glycolysis pathway and chondrocyte senescence was examined in vivo cartilage tissues under different treatment conditions. HIF-1α protein expression was significantly upregulated by 3.26-fold (P < 0.001) in the saline group compared to the sham group, as determined by WB analysis, whereas the Alg gel group showed minimal change relative to the saline control (Fig. 8A). This upregulation was markedly reversed by CIMNP/Alg gel treatment, which exhibited a 0.44-fold downregulation compared to the saline group (P < 0.0001). The efficacy of CIMNPs was further highlighted by the comparable HIF-1α levels between the CIMNP/Alg and DXM groups. Furthermore, relative to the Alg gel group alone, the CIMNP/Alg gel group showed a significant 0.38-fold reduction in HIF-1α (P < 0.001). A consistent expression pattern was observed for key glycolysis-related proteins. In the CIMNP/Alg gel group, levels of HK2, PFK1, and PKM2 were significantly decreased by 0.79- (P < 0.01), 0.31- (P < 0.01), and 0.55-fold (P < 0.001), respectively, compared to the Alg gel group, according to WB analyses (Fig. 8A).

Fig. 8.

Metabolic reprogramming and inhibition of senescence capacities in cartilage tissues from OA rats. (A) WB analysis of relative protein expression level to HIF-1α, HK2, PFK1 and PKM2 among the Sham, DXM, Saline, Alg gel and CIMNP/Alg gel groups (n = 5). (B) Quantitative analysis of relative mRNA expression of p16 with the same treatment by RT-qPCR (n = 3). (C) Representative IHC images and positive cell rate of p16 in cartilage tissues (n = 3). (D) Quantitative analysis of relative mRNA expression of p21 with the same treatment by RT-qPCR (n = 3). (E) Representative IHC images and positive cell rate of p21 in cartilage tissues (n = 3). (F) WB analysis of relative protein expression level to CDK4 and CDK6 among the Sham, DXM, Saline, Alg gel and CIMNP/Alg gel groups (n = 5). (G) Quantitative analysis of relative mRNA expression of Cyclin D1 with the same treatment by RT-qPCR (n = 3). (H) Representative IHC images and positive cell rate of γ-H2AX in cartilage tissues (n = 3). (I) Quantitative analysis of relative mRNA expression of MMP13 with the same treatment by RT-qPCR (n = 3). (J) Representative IHC images and positive cell rate of MMP13 in cartilage tissues (n = 3). (K) WB analysis of relative protein expression level to MMP1, MMP3, MMP9 and MMP13 among the Sham, DXM, Saline, Alg gel and CIMNP/Alg gel groups (n = 5). Data are presented as mean ± SD. ^*P < 0.05, ^**P < 0.01, ^***P < 0.001, ^****P < 0.0001, vs the indicated groups. NS, no significant difference

Senescence markers were also significantly altered. p16 mRNA expression was profoundly elevated in the sham group (6.82-fold, P < 0.0001) over the saline group but remained unchanged in the Alg gel group (Fig. 8B). In contrast, the CIMNP/Alg gel group exhibited a 0.71-fold reduction in p16 mRNA compared to the Alg gel group (P < 0.01). This decrease was corroborated at the protein level, with IHC analysis revealing a striking 0.74-fold reduction in p16-positive cells in the CIMNP/Alg gel group (P < 0.0001, Fig. 8C). A similar regulatory effect was seen for p21. While its transcript was 2.28-fold higher in the CIMNP/Alg gel group than in the saline group (P < 0.0001, Fig. 8D), it was significantly suppressed relative to the Alg gel control. The CIMNP/Alg gel group demonstrated 0.60-fold lower p21 mRNA and a 0.69-fold reduction in p21-positive cell rate compared to the Alg gel group (both P < 0.001, Fig. 8E).

Cyclin-dependent kinases CDK4 and CDK6, which drive G1-S transition and are tightly linked to chondrocyte proliferative dysfunction in OA, were markedly declined in the saline group related to sham group (both P < 0.001; Fig. 8F). While the level of CDK4 and CDK6 were pronouncedly reversed via CIMNP/Alg gels treated relative to saline group (both P < 0.05; Fig. 8F). qRT-PCR analysis revealed a 2.06-fold increase in Cyclin D1 transcript levels in the CIMNP/Alg gel group relative to the saline group (P < 0.01; Fig. 8G). Quantitative IHC evaluation demonstrated that CIMNP/Alg gel treatment differentially down-regulated senescence marker γ-H2AX staining by 0.86-fold (P < 0.0001) compared to the saline group (Fig. 8H). In addition, the mRNA and protein expression level of MMP13 were obviously mitigated by 0.82-fold (P < 0.01) and 0.81-fold (P < 0.001) related to the saline group (Fig. 8I–J). Corroborating these findings, WB analysis confirmed the significant downregulation of the matrix-degrading enzymes MMP1, MMP3, MMP9 and MMP13 in the CIMNP/Alg gel group, with reductions of 0.41- (P < 0.001), 0.32- (P < 0.0001), 0.63- (P < 0.001) and 0.53-fold (P < 0.001), respectively, compared to the Alg gel group (Fig. 8K). Taken together, these results indicate that CIMNP/Alg gels alleviate cartilage senescence and halt osteoarthritis progression by suppressing the ROS/HIF-1α–driven metabolic reprogramming pathway.

Discussion

OA, a leading cause of disability and work loss, is a heterogeneous whole-joint syndrome characterized by pain [2]. The pathogenesis of OA involves a close interplay between synovitis and cartilage degradation, forming a self-perpetuating pathological cycle [4]. Extensive preclinical studies in rodent models have indicated the potential occurrence of Mφ pyroptosis in synovitis [41–43]. Similarly, accumulating evidence demonstrated that the hidden occurrence of chondrocytes senescence in articular cartilage [44–46]. These findings collectively suggest that Mφs pyroptosis and chondrocyte senescence may drive a pathogenic cascade leading to synovitis and cartilage degradation in OA. This underscores the urgent need for validation in human samples to substantiate the potential mechanistic relevance.

Significant inflammatory cells infiltration was observed in OA synovium compared to healthy controls, accompanied by a marked elevation in synovitis score [36]. Notable upregulation of pyroptosis markers (NLRP3, ASC, Cl-caspase-1, and GSDMD-N) and inflammatory cytokines (IL-1β, IL-18, and TNF-α) were observed in synovial tissues. In fact, the accumulation of activated Mφs is a pivotal event in establishing chronic synovitis and fueling persistent joint inflammation [13, 47]. Our clinical data revealed that numerous inflammatory cells had accumulated in the synovium and underwent pyroptosis, a pro-inflammatory cell death triggered by NLRP3 inflammasome activation [48]. The NLRP3 inflammasome is a cytosolic sensor complex that responds to microbial and danger cues. Upon recognizing PAMPs or DAMPs, NLRP3 oligomerizes via its NACHT domain, recruits ASC through PYD-PYD interactions, and subsequently engages pro-Caspase-1 via CARD domains to assemble the functional inflammasome [49, 50]. Activated NLRP3 inflammasomes are engaged to auto-activate Caspase-1, promote IL-1β/IL-18 maturation, and cleave GSDMD into pore-forming GSDMD-N, whereby cytokines are released and synovitis is intensified [49–53]. In OA, pyroptosis-driven TNF-α, IL-18, and IL-1β from the synovium are released to induce ECM-degrading proteinases upregulation, including ADAMTSs and MMPs, collectively driving progressive cartilage deterioration [52, 54]. As illustrated in Fig. 1I-L, cartilage specimens demonstrated extensive structural deterioration, characterized by surface erosion and extracellular matrix loss, which intimately correlated with senescence markers, SASP associated cytokines and enzymes. Cyclin D1-CDK4/6 normally inactivates Rb through hyperphosphorylation, releasing E2F1 to drive the G1-S transition. In contrast, senescent cells upregulate the CDK inhibitors p16 and p21, which block CDK4/6-Cyclin D1 activity and prevent Rb hyperphosphorylation, thereby enforcing permanent cell-cycle arrest [55, 56]. Thus, Rb remains hypophosphorylated, sequestering E2F1 and blocking G1-S entry, which represses DNA-replication genes and drives excessive SASP production [57, 58]. Specifically, these SASP factors (IL-6, IL-1β) inspired the generation of large contents of NO by iNOS, which in turn enhanced COX-2 synthesis [20]. Various SASP proteinases (ADAMTS, MMPs) dysregulation mediated by iNOS and COX-2 further perpetuated the destruction of type II collagen and erosion of proteoglycans in articular cartilage [59–61]. Moreover, during cellular senescence, HMGB1 as a prototypical DAMP underwent translocation from the nucleus to the cytoplasm, then secreted and activated NF-κB through Toll-like receptor signaling, which amplified cell senescence and SASP secretions [62–64]. Thus, activation of the p16/p21-Rb pathway drives chondrocyte senescence, leading to abundant SASP release (ILs, ADAMTSs, MMPs) that accelerates ECM degradation and establishes a self-reinforcing inflammatory loop that fuels OA progression [20, 65]. Overall, our investigation focused on clinical joints samples revealed, for the first time, that pyroptosis and senescence were the intensive inducements to synovitis and cartilage degradation, respectively.

In view of the indispensable roles of pyroptosis and senescence in OA progression, several studies tried to target pyroptosis or senescence in isolation for OA reversal, while those strategies lacked the capacity for synchronous phenotypic remolding to pyroptosis in synovium and senescence in cartilage [41–46]. As a conserved metabolic pathway, glycolysis is tightly linked to OA progression, evidenced by aberrant upregulation of key enzymes such as HK2, PFK1, and PKM2 in OA joint tissues [15]. Several studies further show that glycolysis acts upstream to sustain Mφ pyroptosis and chondrocyte senescence, thereby aggravating synovitis and cartilage degradation in OA [16, 66]. Elevated PKM2-dependent lactate facilitated to IL-1β, IL-18, and HMGB1 unleash by alternative activation of eukaryotic translation initiation factor 2-alpha kinase 2 (EIF2AK2)-dependent NLR family, NLRP3 inflammasome in Mφs. Pharmacological and genetic suppression of the PKM2/EIF2AK2 pathway decayed activation of NLRP3 inflammasomes and restricted the unleash of IL-1β, IL-18 and HMGB1 [67, 68]. Di Liu et al. demonstrated that PKM2-mediated glycolysis promotes NLRP3 inflammasome-dependent pyroptosis of skeletal muscle cell in dermatomyositis/polymyositis [23]. Pyruvate dehydrogenase depletion of mitochondria inhibits pyruvate oxidation and accelerates anaerobic metabolism of pyruvate, which also enhances pyroptosis [67]. Lactate as the primary metabolic product of glycolysis, mediates protein functional regulation through lactylation, which aggregates and amplifies inflammatory signaling to cause a series of acute and chronic diseases, including sepsis, lung fibrosis and sepsis-associated lung injury etc [69–71]. Jing Li et al. reported that glycolytic lactate induces H3K18la-driven NOD2 lactylation, activating RIPK2-MAPK/NF-κB signaling and thereby promoting NLRP3-dependent pyroptosis [72]. Xiaojie You et al. found that glycolytic lactate drives extracellular acidosis and activating NLRP3 inflammasomes via acid-sensing channels and NF-κB signaling [73]. Moreover, glycolytic could energy directly boosts Mφ phagocytosis and accelerates NLRP3 activation, amplifying synovial inflammation and driving OA progression [8, 11, 12, 15, 16]. Similarly, in chondrocytes, aberrant glycolytic activation disrupted cell cycle homeostasis, triggering senescence and promoting SASP secretion, which further accelerates ECM degradation [10, 15]. Prolonged glycolytic overactivity drives chondrocyte senescence, marked by γ-H2AX and elevated p21/p16, and suppresses type II collagen and aggrecan synthesis [74]. As a glycolysis-derived epigenetic modification, histone lactylation modulates gene transcription increased, inducing lactate enhancing H4K12la enrichment at SASP promoters and consequently driving SASP expression [75]. Likewise, H4K12la enhances p65/p50 promoter binding to activate NF-κB signaling and drive senescence, while its strong association with H3K27ac and H3K4me3 helps establish a senescence-specific chromatin state [75]. Overall, Glycolysis restraint bluntly suppresses pyroptosis and senescence, curbing both synovial inflammation and cartilage loss [76]. This glycolytic shift in OA is governed by the ROS/HIF-1α axis, a key mediator of metabolic adaptation to joint hypoxia. In diverse diseases, elevated ROS activates and stabilizes HIF-1α through complex III-dependent signaling [16, 38, 77]. As a central regulator of hypoxic adaptation, HIF-1α upregulates key glycolytic enzymes such as PFK1 and PKM2, driving their activation and intracellular translocation to initiate glycolysis under oxygen deprivation [16, 24, 38]. Collectively, the ROS/HIF-1α axis drives glycolysis to fuel pyroptosis-induced synovitis and senescence-mediated ECM loss, forming an integrated regulatory network linking senescence, pyroptosis, and matrix degradation, and offering a promising therapeutic target for OA [24, 78].

Melanin and melanin-like nanomaterials possess abundant catechol/quinone functional groups, conferring broad-spectrum ROS/RNS scavenging capacity and intrinsic redox-buffering behavior, which distinguishes them from many small-molecule antioxidants with limited specificity [19, 25, 26]. Compared with synthetic antioxidants, CIMNPs offer intrinsic advantages due to their natural marine origin, as cuttlefish ink is an abundant and readily accessible resource that enables facile extraction and scalable preparation without complex chemical modification [27]. CIMNPs have demonstrated excellent biocompatibility, favorable biosafety profiles, and immunological safety in multiple in vitro and in vivo systems, and are generally considered biodegradable, making them particularly suitable for intra-articular applications requiring long-term tissue compatibility [28–30]. Importantly, accumulating evidence suggests that CIMNPs function not merely as passive ROS scavengers but as bioactive redox-modulating platforms capable of regulating redox-sensitive metabolic and inflammatory pathways, which is consistent with our findings that CIMNP/Alg gel modulates the ROS/HIF-1α/glycolysis axis and downstream biological responses. Alg, a naturally derived anionic polysaccharide from brown seaweed, is widely used in biomedical applications due to its biocompatibility and mild Ca²⁺-mediated gelation, enabling injectable hydrogels for localized drug delivery and prolonged retention [19, 31]. Alg-based hydrogels have been extensively applied in intra-articular therapy and shown to alleviate synovial inflammation, protect cartilage, and modulate the inflammatory microenvironment in OA models, supporting their suitability as delivery matrices for CIMNPs [32–34]. To overcome above issues in of OA, we successfully developed a marine-derived CIMNP/Alg injectable hydrogel by encapsulating cuttlefish ink-derived CIMNPs within an Ca²⁺ crosslinked alginate network. The resulting hydrogel exhibited homogeneous nanoparticle distribution, stable porous microarchitecture, and favorable rheological properties, including shear-thinning behavior and enhanced mechanical strength, ensuring injectability and structural integrity after administration. Importantly, incorporation of CIMNPs endowed the hydrogel with robust antioxidant capacity without compromising its viscoelastic performance, providing a physicochemical foundation for sustained ROS scavenging and subsequent modulation of cellular metabolic and inflammatory responses in the OA microenvironment.

The following experimental design incorporated complementary in vitro and in vivo systems to systematically evaluate therapeutic interventions for OA. In vitro experiments revealed remarkably reduced ROS levels in Mφs following CIMNP/Alg treatment. Simultaneously, the CIMNP/Alg demonstrated significant suppression of both HIF-1α accumulation and glycolytic enzyme expression in Mφs. Moreover, the CIMNP/Alg treatment dramatically reduced pyroptotic pores formation on Mφs surfaces, while concurrently suppressing the NLRP3 inflammasome pathway and downstream cytokines. Importantly, our results indicated that CIMNP/Alg gel exerts a comprehensive inhibitory effect on Mφs pyroptosis, it not only restricts inflammasome activation, but also suppresses the execution of pyroptotic cell death, and markedly limits functional cytokine maturation and release, thereby effectively restraining the overall pyroptotic process rather than modulating a single isolated step. Significant reductions in oxidative stress and HIF-1α stabilization were observed in chondrocytes after CIMNP/Alg treatment, accompanied by normalization of glycolytic enzymes, suppression of senescence markers, restoration of cell-cycle regulators, and inhibition of cartilage-destructive proteases. Unlike transient stress-induced growth arrest or chondrocyte dedifferentiation, true cellular senescence is characterized by sustained p16/p21/Rb activation, persistent SA-β-gal positivity, and acquisition of SASP features [31]. In this study, elevated p16/p21/Rb signaling, increased SA-β-gal activity, upregulation of SASP-related factors (IL-6, IL-1β, ADAMTSs, MMPs), and their consistent association with cartilage degeneration support a senescence-associated phenotype. Importantly, CIMNP/Alg gel not only attenuated senescence markers but also restored cartilage-specific anabolic markers, indicating functional modulation of senescence rather than reversible stress responses or simple dedifferentiation. In contrast to anti-senolytic strategies that rely on direct elimination of senescent cells [79], CIMNP/Alg gel primarily functions to attenuate or partially reverse already initiated senescence-like phenotypes, rather than acting as a purely prophylactic “prevention-before-onset” intervention. Notably, treatment with the ROS scavenger NAC elicited effects comparable to those of CIMNP/Alg gel, supporting the notion that CIMNP/Alg primarily acts through ROS clearance to modulate HIF-1α stabilization, glycolytic metabolism, and associated cellular phenotypes. However, the causal relationship between glycolytic modulation and these downstream phenotypes was not fully resolved in this part of the study, prompting us to perform additional validation experiments. The rescue experiments using the HIF-1α activator CoCl₂ and the glycolysis inhibitor 2-DG highlight a critical role of HIF-1α-dependent glycolytic metabolism in linking oxidative stress to macrophage pyroptosis and chondrocyte senescence. Notably, pharmacological inhibition of glycolysis phenocopied the protective effects of CIMNP/Alg gel, suggesting that suppression of glycolytic reprogramming represents a central metabolic checkpoint underlying its anti-pyroptotic and anti-senescent activities. These findings position metabolic modulation, rather than direct inhibition of pyroptosis or senescence pathways, as a key mechanism by which CIMNP/Alg gel exerts its protective effects in the ROS-enriched microenvironment. While the present study primarily focused on molecular and ultrastructural analyses, future investigations will include complementary functional assays, such as LDH release and membrane permeability measurements (e.g., PI uptake), to provide a more comprehensive anti-pyroptosis assessment of CIMNP/Alg gel.

Collectively, the protective effects of CIMNP/Alg gels against Mφs pyroptosis and chondrocytes senescence were demonstrated through comprehensive modulation of the ROS/HIF-1α/glycolysis axis (Fig. 9). Consistent therapeutic effects were replicated in vivo by rat models, confirming the findings from in vitro experiments. Overall, coupling these physicochemical advantages with metabolic modulation of the ROS/HIF-1α/glycolysis axis, CIMNP/Alg gels emerge as a promising intra-articular therapeutic platform for OA therapy.

Fig. 9 .

Proposed mechanism of CIMNP/Alg gels. (A) Generation of CIMNP/Alg gels by Ca²⁺-mediated crosslinking. (B) i The effect of CIMNP/Alg gels to ROS/HIF-1α/glycolysis axis of Mφs and chondrocytes, ii pyroptosis in Mφs, and iii senescence in chondrocytes

Although the fixed charge density of cartilage ECM is predominantly negative and cationic nanocarriers may exhibit enhanced electrostatic-driven penetration, the disrupted matrix architecture in OA can increase cartilage permeability without relying on electrostatic interactions, and the Ca²⁺-crosslinked alginate network provides intra-articular retention and sustained local availability of CIMNPs. Therefore, the CIMNP/Alg hydrogel is expected to exert therapeutic effects through local redox buffering and suppression of the ROS/HIF-1α/glycolysis axis in joint cavity. Future work may further improve cartilage targeting by introducing cationic/cartilage-affinitive surface modifications. Moreover, the therapeutic efficacy of CIMNP/Alg gels in OA rats, CIMNPs release, cartilage penetration and chondrocytes uptake were not directly assessed and will be examined using fluorescently labeled CIMNPs in future studies. In addition, our data confirmed the therapeutic efficacy of CIMNP/Alg gels in rat OA models through coordinated suppression of Mφs pyroptosis and chondrocyte senescence, further optimization is required before these findings can be generalized to patients across diverse clinical stages of OA. Human OA exhibits pronounced heterogeneity in metabolic status, inflammatory phenotypes, and joint microenvironmental conditions, which may influence both the extent of ROS/HIF-1α activation and the responsiveness to glycolysis-oriented interventions. Therefore, additional studies across early-, mid-, and late-stage OA, as well as metabolically driven or age-associated OA subtypes, are necessary to validate the therapeutic breadth of this strategy. Future investigations using spontaneous or multi age-related OA models, large-animal systems, and long-term safety evaluations will be essential to more accurately mimic human disease progression and to strengthen the translational potential of this hydrogel platform.

Materials and methods

Materials

Cuttlefish ink was bought from Xinjie Aquatic Products Co., Ltd (Qingdao, China). Sodium alginate, calcium gluconate, hydrogen peroxide (H₂O₂) solution (30%), TBHP, Cocl₂, and 2-DG were sourced from Macklin Co., Ltd (Shanghai, China). Cell reactive oxygen species (ROS) test kit, crystal violet, 4, 6-diamidino-2-phenylindole (DAPI) staining solution, iFluor^™ 488 phalloidin, and 3,3’-diaminobenzidine (DAB) peroxidase substrate kits were got from YEASEN Biotech Co., Ltd (Shanghai, China). Prestained protein ladder were got from Thermo Fisher Scientific Inc (Massachusetts, USA). The primary antibody including NOD-like receptor thermal protein domain associated protein 3 (NLRP3) monoclonal antibody, apoptosis-associated speck-like protein containing a CARD (ASC) monoclonal antibody, Pro-Caspase-1 monoclonal antibody, Cleaved-Caspase-1 monoclonal antibody, cleaved Gasdermin D (GSDMD-N) monoclonal antibody, interleukin (IL)−1β monoclonal antibody, IL-18 monoclonal antibody, tumor necrosis factor α (TNF-α) monoclonal antibody, a disintegrin-like and metalloproteinase with thrombospondin motifs (ADAMTS)−4 monoclonal antibody, ADAMTS-5 monoclonal antibody, matrix metalloproteinase (MMP)−1 monoclonal antibody, MMP3 monoclonal antibody, MMP9 monoclonal antibody, MMP13 monoclonal antibody, high mobility group box-1 protein (HMGB1) monoclonal antibody, cyclin-dependent kinase inhibitor 1 A (p21) monoclonal antibody, cyclin-dependent kinase inhibitor 2 A (p16) monoclonal antibody, cyclin-dependent kinase (CDK) 4 monoclonal antibody, CDK6 monoclonal antibody, Retinoblastoma (Rb) monoclonal antibody, phosphorylated Rb (p-Rb) monoclonal antibody, E2 promoter-binding factor 1 (E2F1) monoclonal antibody, γ-histone 2AX (γ-H2AX) monoclonal antibody, inducible nitric oxide synthase (iNOS) monoclonal antibody, cyclooxygenase-2 (COX-2) monoclonal antibody, SRY-Box transcription factor 9 (Sox9) monoclonal antibody, collagen type II alpha 1 chain (Col2a1) monoclonal antibody, hypoxia-inducible factor 1 alpha subunit (HIF-1α) monoclonal antibody, hexokinase 2 (HK2) monoclonal antibody, phosphofructokinase 1 (PFK1) monoclonal antibody, pyruvate kinase muscle isozyme 2 (PKM2) monoclonal antibody, β-actin monoclonal antibody and β-tubulin monoclonal antibody were purchased from Affinity Biotech Co., Ltd (Jiangsu, China). Tissue ROS detect kit (DHE) was bought from Baiaolaibo Biotech Co., Ltd (Beijing, China). The sequences of the primers used in the gene expression analysis by qRT-PCR were listed in Table S1.

Pathohistological and molecular biological analysis of clinical human knee joint sample

Clinical human knee joint tissue samples used in this research were all approved by the Ethics Committee of Northern Jiangsu People’s Hospital (No. 2025ky1524). All patients had signed informed consent forms to provide knee joint samples for our study. Samples with imaging characteristics indicating a history of previous bone fractures and previous knee surgeries, including knee replacement surgeries, ligamentoplasty, and cartilage repair surgeries, were excluded. Healthy knee synovium and cartilage samples were collected from patients (aged 55–81 years, n = 10, mean age 67.4 ± 2.51 years) without OA who underwent lower limb amputation. OA knee cartilage and synovium samples were obtained from OA patients who underwent total knee arthroplasty and were aged between 57 and 76 years (n = 10, mean age 66.7 ± 2.01 years).

Pathohistological measurement including H&E staining, IHC (ASC, NLRP3, IL-1β and IL-18) and ROS detection (dihydroethidium assays) of synovium, and H&E, SO&FG and TB staining of cartilage were conducted by Servicebio Biotech Co., Ltd. (Wuhan, China).

In accordance with the guidelines by the Osteoarthritis Research Society International (OARSI), the OARSI scores of both healthy and pathological articular cartilage in humans were computed [80].

Preparation of CIMNPs and CIMNP/Alg gels

The isolation of CIMNPs and preparation of CIMNP/Alg gels were conducted according to previous reports with slight modifications [19, 81]. Firstly, cuttlefish inks were isolated from the dissected cuttlefish and dissolved in ultrapure water, the large particles were excluded by centrifugated at 5000 rpm for 10 min in 4℃. Above suspension was following by centrifugated at 14,000 rpm for 10 min in 4℃, the precipitation was washed for 3 times with ultrapure water and freeze-dried to obtain CIMNPs. For the preparation of CIMNP/Alg gels, 231.95 mg sodium alginate (Alg) was dissolved in 7.5 mL ultrapure water, the concentration of Alg was 3 wt%, and different weights of CIMNPs powder were added to the Alg solutions (CIMNP/Alg solutions). Then, 2.5 mL calcium gluconate solution (3 wt%) was slowly added into the CIMNP/Alg solutions (volume ratio: 1:3) under stirred condition to achieve CIMNP/Alg hydrogels (CIMNP/Alg gels) with various CIMNPs concentration. The Alg gel was achieved without CIMNPs added. Besides, in vitro cell related assays, solvent of the hydrogel was replaced by cell medium to limit the side effect for Mφs and chondrocytes.

Characterization of CIMNPs and CIMNP/Alg gels

The morphology of CIMNPs was observed via TEM (Tecnai 12, Philips, Holland) and SEM (Gemini SEM 300, Carl Zeiss, Germany). The hydrodynamic diameter, zeta potential and stability (dissolved in water, saline and medium for 7 days) of CIMNPs were measured by a Malvern instrument (Worcestershire, UK). The morphologic images of Alg gels and CIMNP/Alg gels were performed using SEM (Gemini SEM 300, Carl Zeiss, Germany). The general appearance of Alg gels and CIMNP/Alg gels in glass bottles were gathered by camera (EOS R6, Canon, Japan). The 1,1-diphenyl-2-picryl-hydrazyl radical (DPPH) radicals elimination performance of CIMNP/Alg gels (The concentrations of CIMNPs were 0.05, 0.1, 0.2, 0.5 and 1.0 mg/mL) were evaluated by DPPH free radical scavenging capacity assay kit according to the specification. The rheological properties of Alg gels and CIMNP/Alg gels were conducted by rotational rheometer (DHR-2, TA, USA) equipped with a pair of 40 mm parallel plates at 25℃.

Animals

Animal experiments in our research were permitted by the Institutional Animal Care and Use Committee of Yangzhou University (No. 202205010) and conducted following the guidelines of Jiangsu Laboratory Animal Welfare and Ethical of Jiangsu Administrative Committee of Laboratory Animals. Four-week-old Wistar rats (male, 180 ± 20 g) were achieved from the Comparative Medical Center of Yangzhou University (Yangzhou, China) and kept in specific pathogen-free conditions in the rat isolators.

Cells

Rat peritoneal Mφs and articular chondrocytes were obtained as previous study [82]. Rat peritoneal Mφs were collected from rat abdominopelvic cavity and resuspended in complete Dulbecco’s modified Eagle’s medium (DMEM) containing 10% FBS. Impurity cells were removed via changing the medium after culturing for 6 h for the first time, and the Mφs were gathered and counted for subsequent research. Articular chondrocytes were collected from the rat cartilage of knee joints through type II collagenase digestion. Cells were resuspended in complete DMEM/F-12 with 10% FBS, and second-passage chondrocytes were used for further studies. The cells were cultured in a humidified incubator with 5% CO₂ at 37 °C.

MTT assay of rat peritoneal Mφs and chondrocytes

Rat peritoneal Mφs (1 × 10⁴ cells per well, 100 µL) were seeded into 96-well plates for 12 h, followed by treated with CIMNPs and CIMNP/Alg gels at different concentrations of CIMNPs (0, 3.125, 6.25, 12.5, 25, 50, 100, 200, 500 µg/mL) for 48 h. Subsequently the cells were incubated with MTT solution (5 mg/mL) for 4 h and 100 uL DMSO was added, and the UV-Vis absorbance was measured at 490 nm using a microplate reader.

Anti-glycolysis and Anti-pyroptosis activity of CIMNP/Alg gels to Mφs in vitro

Rat peritoneal Mφs (1 × 10⁵ cells/well, 1 mL) were seeded in the lower chamber of a 12-well plates for 12 h. Cells were treated by 5 mM NAC, Alg and CIMNP/Alg gels (concentration of CIMNPs was 200 µg/mL) for 24 h, following by treated with 300 µM H₂O₂ for another 24 h, named H₂O₂+NAC group, H₂O₂+Alg group and H₂O₂+CIMNP/Alg group, cell with only 300 µM H₂O₂ for 24 h named H₂O₂ group, and cell with no treated named Control group. ROS level was measured by DCFH-DA probe according to instruction. The morphology of rat peritoneal Mφs was observed by SEM. The Protein levels of HIF-1α, HK2, PFK1, PKM2, ASC, NLRP3, Pro-Caspase-1, Cl-Caspase-1, GSDMD-N, Pro-IL-1β, IL-1β and IL-18 were estimated by WB. The concentration of IL-1β and IL-18 were detected by ELISA assay.

Anti-glycolysis and Anti-senescence capacity of CIMNP/Alg gels to chondrocytes in vitro

Chondrocytes (1 × 10⁵ cells/well, 1 mL) were seeded in the lower chamber of a 12-well plates for 12 h. Cells were treated by TBHP (T-butyl hydroperoxide, 0 µM), TBHP (12.5 µM), TBHP (25 µM) and TBHP (50 µM) for 24 h to detect the optimal concentration of TBHP for inducing senescence. Chondrocytes (1 × 10⁵ cells/well, 1 mL) were seeded in the 12-well plates for 12 h. Cells were treated by 5 mM NAC, Alg and CIMNP/Alg gels (concentration of CIMNPs was 200 µg/mL) for 24 h, following by treated with 50 µM TBHP for another 24 h, named TBHP + NAC group, TBHP + Alg group and TBHP+CIMNP/Alg group, cell with only 50 µM TBHP for 24 h as TBHP group, and cell with no treated as Control group. ROS level was measured by DCFH-DA probe according to instruction. The SA-β-gal of chondrocytes was observed by corresponding kits. The mRNA levels of Lamin B1 and p21 were measured by RT-qPCR. The Protein levels of HIF-1α, HK2, PFK1, PKM2, p16, CDK4, CDK6, Cyclin D1, Rb, p-Rb, E2F1, γ-H2AX, ADAMTS-5, MMP1, MMP3, MMP9, MMP13, Sox9 and Col2al were estimated by WB.

OA model establishment

The OA model was constructed by modified Hulth’s method as previous research [82]. The rats were anesthetized with Zoletil^® 50 via intramuscular injection (50 mg/kg). The right hind limb was shaved and disinfected by povidone-iodine scrubs. The rats were kept in a thermostatic operating table at 37 ℃ and draped in a sterile gauze. A 1 cm longitudinal incision was performed on the skin of the medial right-posterior knee for exposure to the capsule, followed by a second incision to open the joint capsule. The incision was extended until the patella was subluxated laterally to open the joint. Subsequently, the medial collateral ligament, anterior cruciate ligament, and the partial medial meniscus were excised. Then, the capsule and skin were sequentially seamed with polyglecaprone resorbable sutures. The sham operation only received 1 cm longitudinal incisions on the skin and joint capsule successively. After surgery, wounds were painted with erythromycin ointment for 4 weeks to prevent bacterial infection, and OA models eventually established.

Therapeutic efficacy of CIMNP/Alg gels in rat OA models

OA rats were randomly divided into seven groups (n = 6). The sham and saline groups were intravenously injected with saline as the control. The DXM group was orally administered DXM (0.1 mg/kg) daily for four weeks. Alg and CIMNP/Alg groups were intravenously injected with Alg gels (100 µL) and CIMNP/Alg gels (100 µL, the concentration of CIMNPs was 5 mg/mL) once a week for four cycles, respectively. The body weight of each group was recorded every three days. After treatment, serum and OA joints were collected for further research. Micro-computed tomography (micro-CT), histological examinations including H&E, SO&FG and TB staining were performed to evaluate therapeutic effects. The OARSI score of articular cartilage from OA rats after treatment was calculated following the guidelines of the OARSI [83]. The synovial tissues were collected and detected the metabolic reprogramming (HIF-1α, HK2, PFK2 and PKM1) and pyroptosis (ASC, NLRP3, Pro-Caspase-1, Cl-Caspase-1, GSDMD-N, Pro-IL-1β, IL-1β and IL-18) related protein by WB. Identically, the cartilage tissues were obtained and measured the metabolic reprogramming (HIF-1α, HK2, PFK2 and PKM1), senescence signal pathway (p16, CDK4, CDK6, Cyclin D1, Rb, p-Rb, E2F1), SASP (γ-H2AX, ADAMTS-5, MMP1, MMP3, MMP9 and MMP13), and ECM (Sox9 and Col2al) related protein by WB. The IHC analysis of synovium (NLRP3, Cl-Caspase-1, IL-18 and IL-1β) and cartilage (p16, p21, γ-H2AX, MMP13, Sox9 and Col2a1) were performed by Servicebio Biotech Co., Ltd. (Wuhan, China). The mRNA expression level of pyroptosis associated gene (NLRP3, IL-1β and IL-18) in synovium and senescence associated gene (p16, p21, Cyclin D1 and MMP13) in cartilage were detected by RT-qPCR. The content of IL-18, IL-1β, TNF-α and IL-6 in the synovitis tissues were measured by ELISA assay. The ROS level in synovium and cartilage was detected by tissue ROS detect kit (DHE).

Pharmacological validation of the HIF-1α/glycolysis axis underlying CIMNP/Alg gel-mediated cellular phenotypes using CoCl₂ and 2-DG

Rat peritoneal Mφs (1 × 10⁵ cells/well, 1 mL) were seeded in 12-well plates and cultured for 12 h. Cells were divided into the following groups: control, H₂O₂, H₂O₂+CIMNP/Alg, H₂O₂+CIMNP/Alg+CoCl₂, H₂O₂+2-DG, CoCl₂, and 2-DG. CIMNP/Alg gels were applied at a CIMNP concentration of 200 µg/mL for 24 h, followed by stimulation with H₂O₂ (300 µM) for another 24 h. CoCl₂ (100 µM, as the HIF-1α activator) or 2-DG (5 mM, as the glycolysis inhibitor) was added as indicated during oxidative stimulation. After treatment, the protein levels of HIF-1α, glycolytic enzymes (HK2, PKM2), and pyroptosis-related markers (NLRP3, IL-1β) were analyzed by WB.

Rat chondrocytes (1 × 10⁵ cells/well, 1 mL) were seeded in 12-well plates and cultured for 12 h. Cells were divided into the following groups: Control, TBHP, TBHP+CIMNP/Alg, TBHP+CIMNP/Alg+CoCl₂, TBHP +2-DG, CoCl₂, and 2-DG. CIMNP/Alg gels were applied at a CIMNP concentration of 200 µg/mL for 24 h, followed by stimulation with TBHP (50 µM) for another 24 h. CoCl₂ (100 µM, as the HIF-1α activator) or 2-DG (5 mM, as the glycolysis inhibitor) was added as indicated during oxidative stimulation. After treatment, the protein levels of HIF-1α, glycolytic enzymes (HK2, PKM2), and senescence-related markers (p16, Sox9) were analyzed by WB.

Statistical analysis

GraphPad Prism software 10.1.2 was used for statistical analysis and graphs preparation. For the comparison of two groups, a student’s t test was used. For more than two groups, a one-way ANOVA was used. ^*P < 0.05, ^**P < 0.01, ^***P < 0.001, ^****P < 0.0001, vs. the relevant group. NS, no significant differentiation. Data was shown as mean ± SD.

Author contributions

Zhou Xu: Investigation, Methodology, Writing – original draft. Yuan Fang: Investigation, Methodology. Yuesheng Wang: Investigation. Haixiang Miao: Methodology. Tangjie Zhang: Software. Peng Cao: Funding acquisition, Project administration. Yi Zhang: Project administration, Supervision, Writing – review & editing. Gang Chen: Conceptualization, Funding acquisition, Project administration, Supervision, Writing – review & editing.

Funding

This work was supported by the Young Taishan Scholars Program of Shandong Province (grant no. tsqn202507257), National Natural Science Foundation of China (grant no. 82230120), Shandong Provincial Natural Science Foundation (grant no. ZR2023MH352), Youth Innovation Science and Technology Program of Shandong Provincial Universities (grant no. 2024KJJ039), and Basic Research Program of Jiangsu (Grants No BK20250555).

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

Animal experiments were permitted by the Institutional Animal Care and Use Committee of Yangzhou University (No. 202205010) and conducted following the guidelines of Jiangsu Laboratory Animal Welfare and Ethical of Jiangsu Administrative Committee of Laboratory Animals. This study involves human participants and was approved by the Ethics Committee of Northern Jiangsu People’s Hospital (No. 2025ky1524).

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.