An inflammatory TMJOA model

The protocols for animal studies were approved by the Ethics Committee for Animal Research, School and Hospital of Stomatology, Wuhan University (protocol No. S07918060A), and in accordance with the National Research Council’s Guide for the Care and Use of Laboratory Animals. A CFA-induced mouse TMJOA model was employed in the present study, with 27 male C57BL/6 J mice (aged 8–10 weeks, weighing 20–25 g) purchased from the Experimental Animal Centre of Hubei Province. Mice were fed a standard diet and housed in pathogen-free cages with 12 h circadian rhythms and randomly divided into saline, CFA, and CFA + Resatorvid groups, with each group containing 9 mice. CFA-induced TMJOA mice were elicited by bilateral intra-articular injections of 20 μL CFA (Sigma, F5881) using an anterosuperior puncture technique as previously described [23]. The saline group received bilateral intra-articular injections of isopyknic saline. Mice in the saline, CFA, and CFA + Resatorvid groups were analyzed at 1, 2, and 4 weeks after intra-articular injection, with 3 mice per time point (6 joints, n = 6).

Administration of Resatorvid

Resatorvid (HY-11109, MCE), a specific TLR4 antagonist interfering with the interactions between TLR4 and its intracellular adaptors, was employed in the present study. The CFA + Resatorvid group was treated with intraperitoneal injections of Resatorvid at a dose of 10 mg/kg [24] 1 day before intra-articular CFA induction and maintained twice a week post intra-articular injection.

Micro-CT analysis

Collected temporomandibular joint (TMJ) tissues from each group were fixed in 4% paraformaldehyde solution for 24 h and flushed overnight. Afterwards, TMJ specimens used for bone morphological analysis were scanned by micro-CT (filter Al 0.2 mm, 50 kV, 500 μA, 12.59 μm, SkyScan1176) to record the change in subchondral bone. The scanned data were reconstructed by NRecon, with sagittal sections processed with CTAn. Furthermore, 3D images were dimensionally reconstructed by CTvox for morphological assessment.

Histological detection

After paraformaldehyde fixation and ethylene diamine tetra-acetic acid (EDTA) decalcification, all TMJ specimens were subjected to gradient dehydration and embedded in paraffin. Then, the embedded specimens were cut into continuous sagittal sections, each at 5 μm. After dewaxing in xylene and gradient hydration, H&E, Safranin-O/Fast Green, and Masson staining were performed according to the manufacturer’s protocols to examine synovial inflammation, cartilage degeneration, and bone destruction, respectively. H&E staining of synovial tissues was quantified by the number of synovial lining layers. Cartilage degradation was evaluated by condylar cartilage thickness of H&E staining and modified Mankin OA score of Safranin-O/Fast Green staining, on the basis of predecessors’ methods [25]. The degeneration of subchondral bone was measured by the percentage of unmineralized bone based on Masson staining, which averaged from anterior, middle and posterior, 3 selected areas of the subchondral bone. Moreover, TRAP staining was utilized to detect osteoclast activity within the subchondral bone, with the number of TRAP-positive cells used for quantification.

Immunohistochemistry and immunofluorescence

Immunohistochemistry and immunofluorescence of tissue slices were performed as described previously [26], with primary antibodies utilized as follows: rabbit anti-TLR4 (1:400, 19811–1-AP, Proteintech), rabbit anti-CD34 (1:300, GB111693, Servicebio), rabbit anti-F4/80 (1:100, GB113373, Servicebio), rabbit anti-IL-1β (1:200, ab9722, Abcam), rabbit anti-TNF-α (1:200, ab6671, Abcam), rabbit anti-iNOS (1:100, GB11119, Servicebio), rabbit anti-COX-2 (1:300, 12375–1-AP, Proteintech), rabbit anti-ADAMTS5 (1:200, ab41037, Abcam), rabbit anti-MMP13 (1:200, 18165–1-AP, Proteintech), rabbit anti-Caspase1 (1:200, BA2220, Boster), rabbit anti-GSDMD (1:200, ab219800, Abcam), rabbit anti-Aggrecan (1:600, 13880–1-AP, Proteintech), rabbit anti-MyD88 (1:300, 23230–1-AP, Proteintech), rabbit anti-NF-κB p65 (1:300, 10745–1-AP, Proteintech), and rabbit anti-NLRP3 (1:300, 19771–1-AP, Proteintech). The positive expression of condylar cartilage and synovial tissues was quantified by the rate of positive cells, average optical density (AOD), or integral optical density (IOD) of positive staining, respectively.

Chondrocyte and macrophage culture

Mouse ATDC5 chondrogenic cells (BFB Biotechnology Development Co., LTD) were cultured in DMEM/F-12 and induced by insulin-transferrin-selenium (ITS) as previously mentioned [2]. After chondrogenic induction, ATDC5 cells were stimulated with lipopolysaccharide (LPS) at a dose-course for TLR4 activation. Resatorvid (10 μM, HY-11109, MCE), PDTC (50 μM, HY-18738, MCE), and MCC950 (10 μM, HY-12815A, MCE) were added to ATDC5 cells before IL-1β (10 ng/ml, 200-01B, PeproTech) incubation. Moreover, ATDC5 cells were transfected with NLRP3 lentiviruses (Lv-NLRP3, MOI = 10) from Genechem Co., Ltd. for NLRP3 overexpression.

RAW264.7 macrophages (Procell Life Science & Technology Co., Ltd.) were cultured in DMEM with 10% fetal bovine serum (FBS) and incubated with LPS (100 ng/ml, L4391, Sigma) combined with IFN-γ (20 ng/ml, CM41, Novoprotein) for M1 phenotype polarization. Pretreatment with Resatorvid (10 μM), MCC950 (10 μM) and N-acetylcysteine (NAC, 10 mM, HY-B0215, MCE) was utilized to analyze the effect of TLR4, NLRP3, or ROS suppression on the inflammatory responses of M1 macrophages. Furthermore, M1 macrophages were enriched in ROS by incubation with H2O2 (100 μM, 323381, Sigma) for 24 h. Each experiment was repeated thrice independently.

ELISA analysis

One day following phenotype induction and mediator administration, RAW264.7 macrophages were replaced with fresh medium and collected 24 h after culture. After centrifugation to remove precipitation, the concentrations of IL-1β, TNF-α, and IL-6 in macrophage supernatants were detected by ELISA analysis. The experimental procedure of ELISA analysis was performed based on the manufacturer’s protocols, including a mouse IL-1β ELISA kit (RX203063M, Ruixin bio, Quanzhou, China), mouse TNF-α ELISA kit (JL10484, Jianglai bio, Shanghai, China), and mouse IL-6 ELISA kit (EMC004, NeoBioscience, Shenzhen, China). Each experiment was independently repeated thrice.

Immunocytofluorescence

RAW264.7 macrophages and ATDC5 cells used for immunocytofluorescence detection were permeabilized with 0.3% Triton X-100 after 4% paraformaldehyde fixation. Afterwards, the cells were incubated with 2.5% bovine serum albumin (BSA) for antigen blocking and treated with primary antibody against NLRP3 (1:300, 19771–1-AP, Proteintech) overnight at 4 °C. Subsequently, Dylight 488, goat anti-rabbit IgG or DyLight 594, and goat anti-rabbit IgG were added and reacted at 37 °C for 60 min, and the nuclei were stained with 4′,6-diamidino-2-phenylindole (DAPI). Positive reactivity of cells was captured under a fluorescence microscope and quantified by the value of AOD.

Pyroptosis detection and ROS measurement

To morphologically analyze chondrocyte pyroptosis, ATDC5 cells were cultured to 80% confluence and then treated with Hoechst staining combined with propidium iodide (PI) staining, according to previous reports [27]. Briefly, cells were stained with PI for 15 min at 37 °C, and nuclei were stained with Hoechst 33342. Lytic cell death was visualized and measured by PI incorporation to label dying cells. Fluorescence images were captured using a fluorescence microscope, and cellular morphology was observed through brightfield images.

For ROS measurement, RAW264.7 macrophages were labeled with cell-permeable dihydroethidium (DHE) fluorogenic probes for 30 min at 37 °C, followed by nuclear staining with DAPI. The released ROS from macrophages were detected by fluorescence microscopy under an excitation wavelength of 594 nm and quantified by the AOD of positive labeling.

Western blotting

After cells were lysed with RIPA buffer and centrifuged to separate sediment, the protein samples were loaded and separated by electrophoresis in a 10% SDS–PAGE gel. The proteins were transferred to a PVDF membrane and incubated with the following primary antibodies at 4 °C overnight after blocking with 5% skim milk: rabbit anti-TLR4 (1:1000, GB11519, Servicebio), rabbit anti-NLRP3 (1:2000, ab263899, Abcam), rabbit anti-ADAMTS5 (1:2000, ab41037, Abcam), rabbit anti-MMP13 (1:1000, GB11247-1, Servicebio), rabbit anti-Caspase1 (1:2000, BA2220, Boster), rabbit anti-GSDMD (1:1000, ab219800, Abcam), rabbit anti-COX-2 (1:1000, #12282, CST), mouse anti-iNOS (1:1000, MA5-17139, Invitrogen), rabbit anti-NF-κB (1:2000, 10745–1-AP, Proteintech), rabbit anti-IL-1β (1:6000, GB11113, Servicebio), rabbit anti-IL-6 (1:4000, GB11117, Servicebio), and mouse anti-GAPDH (1:5000, RAB0101, Frdbio). On the second day, the membranes were incubated with HRP-conjugated secondary antibodies at 37 °C for 1 h, and protein bands were visualized by a hypersensitive ECL kit (PMK003, BioPM). The expression of proteins was quantified and normalized to GAPDH using the ImageJ 1.8.0 software.

Statistical analysis

Statistical analysis within this study was conducted by Prism 8.0 (GraphPad). Quantitative data are presented as the mean ± SEM, and p < 0.05 was considered statistically significant. Data between groups were analyzed by Student’s t test or one-way ANOVA followed by Dunnett’s multiple comparisons test, and two-way ANOVA followed by Sadak’s multiple comparisons test was used for group analysis. Beforehand, a Shapiro–Wilk test for normality was conducted to determine whether the data were parametric, with F tests utilized to check the homogeneity of variance.