Acacetin targets STING to alleviate the destabilization of the medial meniscus induced osteoarthritis in mice

Osteoarthritis (OA), a widespread joint disorder affecting approximately 7% of the global population, is primarily characterized by the gradual loss of articular cartilage. This degeneration results from local in�ammation, matrix depletion, and direct cartilage damage. A critical element in this cascade is the activation of the Stimulator of the Interferon Genes (STING) pathway. Emerging evidence underscores its potential as a therapeutic target, with natural products showing promise as inhibitors. This study centers on Acacetin, a basic unit of polyketides known for its anti-inammatory attributes. Prior research has highlighted its potential interaction with STING based on the structure. Therefore, this study aimed to assess Acacetin's effectiveness as a STING inhibitor and its protective role against OA. In vitro experiments demonstrated that Acacetin pretreatment not only mitigated interleukin-1 β (IL-1 β )-induced cytotoxicity but also decreased the in�ammatory response and degeneration in IL-1 β -stimulated chondrocytes. In vivo studies revealed that Acacetin administration signi�cantly reduced articular cartilage destruction, abnormal bone remodeling, and osteophyte formation in a model of OA induced by destabilization of the medial meniscus (DMM). Mechanistically, Acacetin was found to directly interact with STING, and inhibit IL-1 β -induced activation of STING and subsequent phosphorylation of the TBK1/NF κ B pathway in chondrocytes. In conclusion, our �ndings con�rm that Acacetin is an effective inhibitor of STING, offering protection to chondrocytes against IL-1 β -induced damage and attenuating the progression of OA in mice.

License:   This work is licensed under a Creative Commons Attribution 4.0 International License.

Read Full License
Additional Declarations: No competing interests reported.

Introduction
Osteoarthritis (OA), affecting approximately 7% of the global population, is predominantly observed in individuals aged 55 to 64.It is characterized by a gradual loss of articular cartilage, accompanied by synovial in ammation and subchondral bone sclerosis [1,2].The pathophysiology of OA is complex due to its chronic and progressive nature, involving a multifaceted interplay of mechanical and biological factors.Initially, biomechanical injuries to various structures trigger the release of mediators, activating pro-in ammatory signaling pathways [3][4][5].This underscores the potential of in ammation prevention as a key strategy in slowing or arresting OA progression [6][7][8].
The Stimulator of Interferon Genes (STING) plays an important role in the pathogenesis of OA by sensing damaged cytosolic DNA and activating the innate immune response [9][10][11].It is notably upregulated in OA chondrocytes and cartilage in both mouse and human models [12,13].By triggering the NF-κB signaling cascade, STING contributes to imbalances in extracellular matrix metabolism, apoptosis, and senescence in chondrocytes.Genetic disruption of STING in mice mitigated OA symptoms [12,13].Inhibition of STING has been shown to alleviate OA manifestations, including cartilage destruction and subchondral bone sclerosis, as well as hindpaw mechanical allodynia, which is associated with OA pain [14].This evidence highlights its potential as a therapeutic target.The recent surge in developing STING inhibitors, particularly from natural products, further emphasizes its signi cance [15].
Acacetin, a avone in the natural gum, has gained attention for its anti-in ammatory properties.It generally exists in polymers that combine with glycose to form glycosides or carbohydrates, and has important value for medical pharmacy and animal husbandry.It effectively reduces pro-in ammatory cytokines like TNF-α and IL-1β, offering protection against chemically induced in ammation [16,17].Notably, Gao et al. suggested a potential interaction between Acacetin and the STING protein, revealing its role in inhibiting STING-induced expression in macrophages and reducing interferon levels [18].Prior research indicates that Acacetin can protect nucleus pulposus cells, a type of chondrocyte, by reducing in ammatory mediators and preventing ECM degradation.It has also demonstrated protective effects against disc degeneration and osteoarthritis in animal models [19].
Therefore, we propose that Acacetin, which may work as a STING inhibitor, could be effective in treating osteoporosis.In this study, we con rmed the direct interaction between Acacetin and STING in chondrocytes and proved that the Acacetin exerts protective effects of in mice with experimentallyinduced OA caused by destabilization of the medial meniscus.Collectively, we reported plant derived natural product Acacetin as a novel STING inhibitor and the potential of Acacetin as a promising therapeutic agent in OA treatment.
7-week-old wild-type male C57BL/6 mice (collected from Changzhou Cavens Experimental Animals Co., Ltd.Suzhou, China) were raised in the speci c pathogen-free (SPF) animal facility at the Institute of Translational Medicine of Shanghai University.The mice were maintained indoors at an ambient temperature of 22-25℃ with a humidity of 50%.The lighting conditions were maintained on a 24-hour light-dark cycle.The animal study received permission from the Ethics Committee of Shanghai University (approval number ECSHU2021-168) and was conducted following the ARRIVE guidelines.

Isolation and culture of mouse peritoneal macrophages
Two 6-week-old C57BL/6J mice (male or female) were anesthetized with pentobarbital and sacri ced and immersed in 75% ethanol for 10 min.Subsequently, PBS was injected into the abdominal cavity using a 2.5 mL syringe after cutting the abdominal skin of the mice with sterile scissors.The abdomen of the mice was gently massaged, and subsequently the peritoneal washing liquid were collected and centrifuged.With Red blood cell pyrolysis liquid handling and centrifugal again, cells were maintained with RIPM 1640 medium adding 10% FBS and 1% streptomycin/penicillin [20,21].

Isolation and culture of mouse primary chondrocytes
Four 1-day-old neonatal mice (C57BL/6J, both males and females) were euthanized with pentobarbital anesthesia.The mice were dissected and immersed in 75% ethanol for 10 minutes.The processed cartilage was then minced using sterile scissors and transferred to a solution of 1% type II collagenase (0.5 mg/mL) dissolved in DMEM/F-12 (L310KJ, Gibco, USA).The digestion was carried out in a cell culture incubator for 12-16 hours.Subsequently, the digestion mixture was collected, and the cells were centrifuged after being ltered via a 70 µm cell strainer.After washing with 1×PBS and centrifugation, the cells were cultured in DMEM/F-12 (L310KJ, Gibco, USA) supplemented with 10% FBS and 1% streptomycin/penicillin antibiotics.Only the rst and second generations of chondrocytes were employed in the tests for this investigation [14].

Cell viability assay
The primary chondrocytes were cultivated in a 96-well plate at a density of 2×10 4 cells per well.The cells were then treated according to the experimental protocol.Cell viability of the chondrocytes was analyzed using the CCK-8 assay kit (CK04-11, Dojindo Molecular Technologies).
Subsequently, 1 µg of RNA underwent reverse transcription into cDNA using the PrimeScriptTM RT Kit (RR037, Takara).Gene ampli cation followed using the 2 -ΔΔCt method.Normalization involved comparing target gene expression to GAPDH gene expression.Refer to Table 1 for primer sequences.

Western Blot
RIPA lysis buffer (P0013B, Beyotime Institute of Biotechnology, Shanghai, China) was used to extract cellular proteins.The protein levels were quanti ed using a Bicinchoninic acid assay kit (Pierce #23227).
After performing SDS-PAGE denaturing gel electrophoresis, after transferring the proteins to a PVDF membrane (R1BB06918, Millipore), it was subjected to incubation with 5% skim milk at room temperature for 1 hour to inhibit non-speci c binding.Following this, the membrane was subjected to an overnight incubation in a refrigerated environment with the primary antibody.Following that, the secondary antibody was incubated with it at room temperature for a duration of 2 hours.After washing three times with TBST, the membrane was subjected to protein immunoblot detection using an enhanced chemiluminescence (ECL) protein blotting kit (PZ3020-1, Beyotime Institute of Biotechnology) following the manufacturer's operating guidelines.

Destabilization of the Medial Meniscus model
The medial meniscus (DMM) was surgically destabilized to create a mouse model of osteoarthritis [22].The mice were anesthetized using 2.5% iso urane gas.Once the mice were fully anesthetized, they were positioned in a supine position.The knee joint was prepared by using a depilatory agent to remove hair and then sterilized with alcohol.A medial incision was made on the knee joint's patellar ligament, resulting in the exposure of the skin and fascia layers.The extensor muscles and patellar ligament were laterally displaced, and the fat pad was bluntly dissected.Subsequently, the medial meniscus and ligament (MMTL) were transected.The extensor muscles and patellar ligament were then repositioned, and the incision was sutured.
The mice were randomly separated into ve randomized groups after the procedure: sham operation group, DMM group and treatment group, which were treated with intraperitoneal injections of either saline, Acacetin (1 mg/kg and 5 mg/kg), or Astin C (0.5 mg/kg) for 12 weeks.Simulated operations were performed without transecting the medial meniscus ligament in the sham group.They received an equivalent amount of saline injection.All groups of mice were allowed free movement, and the administration of drugs was performed via intraperitoneal injections every three days.

Micro Computed Tomography Assay (Micro-CT)
The intact knee joint of the mice was scanned using a micro-CT scanner (Micro-CT) with the following settings: rotation step of 0.20, resolution of 8 µm, voltage of 55 kV, and current of 60 µA.The threedimensional reconstruction and quantitative analysis of those bone microstructure parameters were accomplished by Software CT Analyser V1.1 (Bruker, Germany), including trabecular separation (Tb.Sp), trabecular number (Tb.N.), bone volume/total volume (BV/TV) and trabecular thickness (Tb.Th.).

Histology and immunohistochemistry
According to the instructions, each joint specimen was stained with Hematoxylin-Eosin (HE) staining, Toluidine Blue O (TB) staining, and Saffron-O and Fast Green staining.These stained specimens were then evaluated by a panel of experienced and blinded histopathologists to assess the cellular structure and morphology of the cartilage and subchondral bone.The assessment of the medial femoral condyle and the medial tibial plateau was performed utilizing the International Osteoarthritis Research Society (OARSI) [23] scoring system, while a grading system for synovitis was used to determine the severity of the condition.Additionally, immunostaining with antibodies targeting MMP3 and Col2A was performed on the tissue sections.

Molecular docking
Molecular docking was performed by the program Autodock 4.2.6.The crystal structure of human STING (PDB ID: 6DNK) is downloaded from the Protein Data Bank Database.The 3D structure of small molecules is downloaded from the PubChem Compound database, which is initialized by adding Gasteiger charges, merging non-polar hydrogen bonds, and setting up rotatable buttons, and then rewritten into PDBQT format.The docking box is set to the range containing the STING active pocket.100 runs of the genetic algorithm are performed using default parameters.The lower the docking binding energy, the higher the a nity of the ligand to the receptor.The docking is effective when the root mean square deviation of the docking binding energy is less than 2. The ligand and the receptor can be combined in the natural state when the docking binding energy is less than 0 kcal/mol.The docking is considered effective when the docking binding energy is less than − 1.2 kcal/mol.

Statistical analysis
The statistical analysis was applied with the GraphPad Prism tool (version 9.0, GraphPad Software, San Diego, CA, USA).Mean ± SD represented the data.Group comparisons involved either two-way ANOVA followed by Bonferroni t-test or one-way ANOVA followed by Tukey's post hoc analysis.Statistical signi cance was denoted by P < 0.05.

Acacetin is a promising STING inhibitor candidate
Natural products are important sources for candidate drugs and drug lead structures.We have constructed a candidate compound library comprising 89 natural products (Table 2).Among these, 38 compounds were sourced from a commercially available compound library (HY-LD-000001447) obtained from MedChemexpress (MCE, USA), while the remaining 51 compounds were isolated from the medicinal mushroom Ganoderma lucidum and purchased from Chengdu Biopurify Phytochemicals Ltd.All compounds were pre-dissolved in dimethyl sulfoxide (DMSO) at a concentration of 10 mM (100 µL).
Based on the inhibitory effects on IFN-β production under DMXAA stimulation, we aimed to assess the anti-in ammatory effects of these compounds and identify potential STING inhibitors.The drug screening results indicated that Acacetin exhibited the most potent anti-in ammatory effect among the 89 candidate compounds (Fig. 1b).Acacetin, a small molecule derived from the leguminous plant Albizia julibrissin, demonstrated an inhibition e ciency of 92.83 ± 5.66% against IFN-β (Fig. 1c and Fig. S1).

Acacetin binds to the STING and inhibits IL-1β-induced activation of the STING pathway in Chondrocytes
The interaction of human STING with Acacetin (PDB ID code: 6DNK) was studied using in silico docking.
The base of the crevice formed by the dimerized STING-CTD contained the c-di-GMP or 2'3'-cGAMP molecule.In silico simulations indicated that Acacetin is potentially bound to the identical pocket where cyclic dinucleotides (CDNs) bind.(Fig. 2a, left).The primary interaction forces between Acacetin and STING involve hydrophobic interactions and hydrogen bonding.Key binding sites encompass ARG232, ARG238, GLU260, and THR 263.Furthermore, a π-π stacking interaction occurs between the benzene ring of STING's TYR167 and the oxygen-containing heterocycle of Acacetin (Fig. 2a, right).It was con rmed by biotin pull-down assay that biotin-Acacetin bound to STING (Fig. 2b).Chondrocytes were pretreated with AstinC and Acacetin (1 and 10 µM) for 3 hours, followed by 1-hour IL-1β stimulation.IL-1β stimulation in chondrocytes enhanced phosphorylation of STING, Tbk1 and p65 in chondrocytes, which was reversed by pre-treatment with Acacetin or Astin C (Fig. 2c-d), at least in part.IL-1β stimulation in chondrocytes enhanced MMP3 protein levels and decreased aggrecan and Col2A protein levels in chondrocytes, which was partially reversed by pre-treatment with Acacetin (Fig. 2e-f).Pre-treatment with Astin C reduced MMP3 protein levels and enhanced Col2A protein levels, but did not alter aggrecan protein expression.

Acacetin ameliorated the development of osteoarthritis induced by DMM in Mice
Seven-week-old male C57BL/6 mice underwent the DMM surgical procedure and were treated with Acacetin (1 and 5 mg/kg) and Astin C (0.5 mg/kg) for 12 weeks.Histopathological analysis conducted 12 weeks post-surgery revealed signi cant cartilage degeneration, characterized by surface structure damage, reduced cartilage area, and notably decreased Osteoarthritis Research Society International (OARSI) scores.This degeneration was alleviated by Acacetin or Astin C treatment (Fig. 3a-c).
Additionally, the effect of Acacetin on reducing the TUNEL-positive cells is comparable to Astin C in mice exposed to DMM surgery (Fig. 4a).DMM surgery induced abnormal cartilage matrix metabolism in mice, which was evidenced by reduced Col2A staining areas and increased MMP3-positive cell counts.This effect was partially reversed by Acacetin or Astin C (Fig. 4b-c).The phosphorylation of STING and its downstream targets TBK1 and p65 was elevated in DMM model and signi cantly inhibited by high dose acacetin and Astin C (Fig. 4d-e).

Acacetin ameliorated osteophyte formation and synovium calci cation induced by DMM in Mice
The progression of osteoarthritis (OA) was also characterized by abnormal bone remodeling and the formation of osteophytes in the underlying bone.Micro-CT scans highlighted an increased subchondral bone volume at the medial tibial plateau, characterized by denser and fewer trabeculae, increased trabecular thickness, and reduced separation in mice subjected to DMM.These changes were partially reversed with Acacetin or Astin C treatment (Fig. 5a-c).
These ndings led us to investigate whether Acacetin could in uence osteoclast activity under in ammatory conditions.Thus, we used bone marrow-derived macrophages (BMMs) to assess the impact on osteoclast differentiation in vitro.Both Acacetin and Astin C treatments signi cantly reduced the number of TRAP-positive osteoclasts compared to the untreated control group, with the Astin Ctreated group showing a further reduction (Fig. 5d, e).The expression of genes associated with osteoclast markers also exhibited similar trends in line with these cellular effects.(Fig. 5f).The presence of Acacetin and Astin C treatment led to signi cant suppression of gene expression involved in osteoclast differentiation and precursor cell fusion, including c-fos and Nfatc1.Likewise, the gene expression levels associated with bone resorption, such as Trap and Ctsk, exhibited a dose-dependent reduction.These results indicated that the STING inhibitors possess the capability to inhibit osteoclast differentiation.
Further analysis revealed that DMM surgery induced osteophyte formation and synovium calci cation in mice.Treatment with both Acacetin and Astin C reduced osteophyte numbers, calci ed cartilage and synovium volumes in OA mice (Fig. 6a-d).

Discussion
This study demonstrates that Acacetin, a avone derived from plants, functions as an effective STING (Stimulator of Interferon Genes) inhibitor.Acacetin protects chondrocytes from IL-1β-induced damage and attenuates the progression of osteoarthritis (OA) in mice.The primary characteristic of OA is the degradation of articular cartilage, with the activation of the STING pathway identi ed as a key factor contributing to this degradation.Our investigation reveals that Acacetin treatment signi cantly ameliorates joint damage in destabilization of the medial meniscus (DMM) mice, mitigating in ammatory responses in chondrocytes by inhibiting STING activation and the TBK1/NF-κB pathway.
STING has emerged as a pivotal research target for immune therapy in OA, with numerous biopharmaceutical companies, including Merck Sharp & Dohme (MSD), Bayer, and Novartis, actively developing innovative therapies targeting the STING pathway.While several STING agonists have entered clinical trials, the development of inhibitors is primarily in the preclinical research stage.Notable STING inhibitors include H-151 and its analogs [15], Astin C derived from Centella asiatica [25], and Gelsevirine isolated from Gelsemium elegans Benth [26,27].H-151 reduces TBK1 phosphorylation and inhibits STING palmitoylation [15], Astin C and Gelsevirine competitively bind to STING with cyclic dinucleotides (CDNs) to exert inhibitory effects.Additionally [25,26], Astin C directly modulates the STING signalosome and binds to the C-terminal domain of STING.Acacetin was previously reported to potentially interact with STING, but the speci c binding mode, site, and effects have not been con rmed [18].Our research reveals that acacetin is an e cient STING inhibitor, with an inhibition e ciency of 92.83 ± 5.66% against DMXAAstimulated IFN-β.Through in silico docking simulation, we predicted the binding site and mode of acacetin with STING: speci cally, the oxygen-containing heterocycle of acacetin can form a π-π stacking interaction with the benzene ring Tyr167 site of STING through hydrophobic interactions and hydrogen bonding, thereby preventing the binding of STING to 2'3'-cGAMP.We further demonstrated the direct binding of acacetin to STING through biotin immunoprecipitation.Meanwhile.
The onset of OA is associated with in ammation, aging, and DNA damage [28].The cGAS-STING pathway, as a crucial sensor for cytoplasmic DNA, triggers immune responses and autophagy in response to damaged DNA.Hyperactive STING signaling promotes the expression of MMP3 and ADAMTS5 through the NF-κB pathway, inhibiting the production of essential structural proteins Aggrecan and Col2A in the cartilage matrix [13].Additionally, STING activation induces cellular senescence and apoptosis, contributing to the development of osteoarthritis [12,14].Elevated levels of damaged DNA in in ammation-induced mouse chondrocytes have been observed, validating this nding in human osteoarthritis tissues [12].Our evidence suggests that Acacetin has the potential to inhibit STING dimerization in chondrocytes, resulting in the downregulation of downstream factors TBK1 and p65 activation.At 10 µM, Acacetin effectively reduces the expression levels of in ammatory factors such as IL-6 and TNF-α in macrophages, aiding in the inhibition of in ammatory cell in ltration and alleviation of OA in ammation, ultimately reducing joint tissue damage.Astin C has been con rmed to speci cally inhibit the cGAS-STING signaling pathway, attenuating immune responses [25].The current study provides evidence that Acacetin has the potential to inhibit STING dimerization in chondrocytes, leading to the downregulation of downstream factors TBK1 and p65 activation.Compared to Astin C, Acacetin partially reverses the elevated levels of MMP3 protein and decreased levels of aggrecan and Col2A proteins in chondrocytes stimulated by IL-1β, whereas Astin C does not alter aggrecan protein expression levels.This suggests that these two STING inhibitors exhibit different impacts on the downstream genes.
Acacetin also affects bone remodeling in the subchondral bone by inhibiting osteoclast activation, ultimately improving arthritis-induced functional impairment [29].
Acacetin has been widely reported bene cial in various in ammation-related diseases.These broad pharmacological effects may be associated with its role as a STING inhibitor.The cGAS-STING signaling pathway has emerged as a crucial mediator of in ammation in the context of infection, cellular stress, and tissue damage [11].Activation of the STING signaling pathway induces in ammatory responses, exacerbating in ammatory damage in various disease processes, including sepsis [30], non-alcoholic fatty liver disease [31], rheumatoid arthritis [32], and neurodegenerative diseases [33].Acacetin has been shown to reduce the gene expression of IL-6 and TNF-α in the liver of non-alcoholic fatty liver mice, to reduce hepatic in ammation [34].Acacetin signi cantly alleviates the in ammatory response in acute lung injury caused by sepsis, primarily manifested by its reduction of in ammatory cytokine concentrations and in ltrating in ammatory cells in BALF [35].By inhibiting NF-κB activation and downregulating the expression of in ammatory iNOS and COX-2 genes in macrophages, acacetin can suppress in ammation-related tumor occurrence [36].Acacetin also exhibits inhibitory effects on neuroin ammation by suppressing the activation of microglial cells [37,38], which may be associated with STING mediated immune responses in microglia [39].Therefore, as a promising STING inhibitor, acacetin holds great potential for application in many STING-related in ammatory diseases beyond OA.
This study has certain limitations.We employed in silico simulations to predict the binding site of Acacetin with STING and con rmed their direct interaction.The validation of our hypothesis could be further strengthened by utilizing cell lines and animal models with STING TYR167 mutations.Additionally, Acacetin has been recognized for its diverse pharmacological functions, and the current study focused on the STING-related mechanism.It cannot be excluded that Acacetin may have other targets, which are also involved in its anti-in ammatory protective effects against OA.Furthermore, it is noteworthy that the decision to administer Acacetin via intraperitoneal injection was in uenced by its reported low oral bioavailability.Considering the potential future clinical application of the compound, efforts should be directed toward enhancing its water solubility to facilitate oral administration.Figure 6

Figures
Figures

Table 1
Minxuan Han, Yongsheng Yu and Li Su conduct background and mechanism research, participate in experimental design, analyze data statistically, and draft manuscripts.Cell culture is handled by Dingjun Xu, Linjie Zhang, and Chenyu Song, who also participate in all in vitro and in vivo experiments.The tissue samples after the in vivo experiment are histologically examined by Dinglei Zhang, Dingjun Xu, and Linjie Zhang.Statistical analysis and sample collection were performed by Dinglei Zhang, Chunlei Xing, Huihui Bian, and Juan Lv.In addition to designing and organizing the research, Yongsheng Yu, Minxuan Han , Minyu Zhu, and Li Su contributed to writing the manuscript.It is the responsibility of Minxuan Han, Minyu Zhu, Li Su, and Yongsheng Yu to write and revise the manuscript drafts.The nal manuscript was read and approved by all authors.Jiangsu Province Capability Improvement Project through Science, Technology and Education-Jiangsu Provincial Research Hospital Cultivation Unit (YJXYYJSDW4) and Jiangsu Provincial Medical Innovation Center (CXZX202227).Institutional Review Board Statement: According to ARRIVE guidelines, the animal study was approved by the Ethics Committee of Shanghai University (approval number ECSHU2021-168).
Funding: This work was supported in part by grant from the National Natural Science Foundation of China (82273937 and 81981340417 to Li Su), China Oral Health Foundation (A2021-025), Angelalign Scienti c Research Fund (EARD20221214053), Natural Science Foundation of Zhejiang Province (LY21H070003),