Human samples

All OA cartilage specimens were collected from patients who underwent surgery to replace the knee joint at Tongji Hospital. The control cartilage was obtained from patients who underwent amputation surgery. The informed consent forms for cartilage specimen collection were signed by the patients and the collection of and experiments using human cartilage were approved by the Ethics Committee of Tongji Hospital (TJ-IRB20210905). The basic information of the amputees and the OA patients is shown in Additional file 1: Tables S1-2.

Materials and reagents

The following reagents were obtained commercially: tumor necrosis factor-α (TNF-a) was obtained from R&D systems (Minneapolis, MN, USA; #554,589). IL-1β was obtained from R&D Systems (Minneapolis, MN, USA; 401-ML-010). Cyclosporin A (CsA) was obtained from MedChemExpress (New Jersey, USA; HY-B0579). Fetal bovine serum was obtained from NEW-ZEALAND (New York, USA). iNOS Rabbit mAb (13,120; Western blot: 1:1000), COX2 Rabbit mAb (12,882; Western blot: 1:1000) and Phospho-NF-κB p65 (Ser536) Rabbit mAb (3033; Western blot: 1:1000; IF:1:800) were purchased from Cell Signaling Technology, Inc. (Beverly, USA). dsDNA Mouse mAb (ab27156; Western blot: 1:1000; IF: 1:600) antibodies were obtained from Abcam (Cambridge, UK). COL2A1 Rabbit Polyclonal antibody (28459-1-AP; Western blot: 1:1000; IF: 1:200; IHC: 1:800), β-ACTIN Recombinant Rabbit antibody (81115-1-RR; Western blot: 1:10000), Phospho-RIPK1 (Ser161) Mouse Monoclonal antibody (66854-1-Ig; Western blot: 1:5000; IF: 1:400; IHC: 1:500), IRF1 Rabbit Polyclonal antibody (11335-1-AP; Western blot: 1:500; IF: 1:250), GAPDH Mouse mAb (60004-1-Ig; Western blot: 1:10000) and MMP13 Rabbit Polyclonal antibody (18165-1-AP; Western blot: 1:1000; IF: 1:200; IHC: 1:100) were acquired from Proteintech Group (Wuhan, China). The ZBP1 Rabbit Polyclonal Antibody was obtained from Thermo Fisher (Massachusetts, USA, PA5-20455; Western blot: 1:1000; IF: 1:200; IHC: 1:500). MMP3 Rabbit monoclonal antibody (BM4074; Western blot: 1:1000), FITC-conjugated goat anti-mouse and anti-rabbit secondary antibodies, type II collagenase, Dulbecco’s modified Eagle’s culture medium F12 (DMEM/F12) and trypsin were purchased from Boster (Wuhan, China). IP lysis buffer (p0013) and MitoTracker were obtained from Beyotime (Shanghai, China).

Chondrocyte isolation and culture

Primary chondrocytes were isolated from the knee joints of 5-day-old house mice. First, the cartilage was dissected into pieces in sterilized phosphate-buffered saline (PBS), and the cartilage slices were subsequently digested with 0.25% trypsin at 37 °C for 30 min. Then,, the cartilage was digested with 0.2% type II collagenase in DMEM for 6 h. Afterward, the cell suspension was transferred to a centrifuge tube and centrifuged at 1,200-1,800 rpm at room temperature for 5 min. Primary chondrocytes were resuspended and cultured in DMEM/F12 containing 10% fetal bovine serum. Next, the chondrocytes were passaged in a T25 flask for culture, and the medium was changed every 2 days. Only the first or second passages of chondrocytes were used in the experiments.

Small interfering RNAs (siRNAs), plasmids and transfection

The negative control siRNA and siRNAs targeting ZBP1 and IRF1 were synthesized by Tsingke Biotechnology. The transfection procedure was conducted as previously described [23]. After the chondrocytes had grown to 60–70% confluence on six-well plates, they were transfected with 50 nmol of the negative control small interfering RNA (siNC), ZBP1 siRNA (siZBP1 #1: CCCTCAATCAAGTCCTTTA; siZBP1 #2: GCCTGCAACATGGAGCATA; siZBP1 #3: CCTGTATTCCATGAGAAAT), or IRF1 siRNA (siIRF1 #1: CACTGATCTGTATAACCTA; siIRF1 #2: GATGGACATTATACCAGAT; siIRF1 #3: GCAGATGGACATTATACCA) using Lipofectamine 3000 in accordance with the manufacturer’s protocol. Then, the chondrocytes were treated with IL-1β or TNF-α before total RNA and protein extraction. Western blot and qRT‒PCR were performed to assess the knockdown efficiency compared to that in the siNC group. The negative control plasmid and flag-tagged ZBP1 plasmid were synthesized by Aoke Biosciences.

The RHIM domain of ZBP1 was mutated and loaded in an adenovirus

Recombinant lentivirus vectors which expressing ZBP1 or ZBP1 which mutant the RHIM domain were constructed by DesignGene (Wuhan, China). Briefly, DNA fragments encoding ZBP1 or ZBP1 which mutant the RHIM domain were subcloned into the lentivirus vectors (NM_021394.2). The empty lentivirus vector was used as a negative control. Subsequently, chondrocytes were cultured in the 10 cm dish, and transfected with lentivirus when the cell density reached to about 80%. The transfect time duration was 24 h. Then, the chondrocytes were cultured useing the fresh medium for another 24 h. Finally, the chondrocytes were treated with TNF-α for 15 min. Chondrocytes were infected with lentivirus at 30 multiplicities of infection (MOI).

In vivo downregulation of ZBP1 using an adeno-associated virus

Adeno-associated virus 9 containing ZBP1 (AAV9-shZBP1) (WZ Biosciences, Inc.) was administered to eight-week-old C57BL/6J mice by intra-articular injection. The doses (10 µl, 1 × 1012 vg/ml) used for the joint injection of AAV9 were chosen according to a previous report [24]. Three weeks after the injection of adeno-associated virus into the joint cavity, destabilization of the medial meniscus (DMM) surgery was performed to establish the animal model. At the eighth week after surgery, the knee joints of the mice were collected for subsequent experiments. Other groups that did not require an injection of AAV9-shZBP1 were treated with a negative control (AAV9-GFP) for the same time.

Animal experiments

Eight-week-old male C57BL/6 mice were used for the animal experiments. The in vivo study was approved by the Institutional Animal Care and Use Committee of Tongji Hospital (IACUC No.: TJH202301008). DMM surgery was conducted according to previously described instructions [25] to induce posttraumatic OA when the C57BL/6 mice were anesthetized. Mice in the control group underwent a sham operation via an incision in the medial capsule of the knee joint. Thirty-two male C57BL/6 mice were indiscriminately separated into four groups: Sham + AAV9-GFP, Sham + AAV9-shZBP1, DMM + AAV9-GFP, and DMM + AAV9-shZBP1. For knee joint capsule administration of CsA, we indiscriminately separated thirty mice into three groups: DMM + PBS, DMM + CsA (2 µg/kg), and DMM + CsA (20 µg/kg). Then, 10 µl of CsA or PBS was injected into the knee cavity once a week for seven weeks. The mice were sacrificed eight weeks after DMM or Sham surgery.

Western blot analysis

The treated chondrocytes were lysed with RIPA buffer (Biocolors Biotechnology Co., Shanghai, China), and the samples were sonicated 3 times. Then, the samples were centrifuged at 12,000 rpm for 30 min at 4 °C. The supernatant was quantified using the bicinchoninic acid method. Next, the samples were heated with loading buffer at 95 °C for 10 min. The 25 µg protein samples were separated by SDS-PAGE and transferred to polyvinyl difluoride membranes. Afterward, the PVDF membranes were blocked with blocking solution (5% skim milk) for 1 h at room temperature and then incubated with primary antibodies against MMP13, MMP3, GAPDH, COL2A1, IRF1, COX2, AGGRECAN, iNOS, SOX9, ZBP1, β-actin, P65, P-P65, TAK1, P-TAK1, RIPK1, and P-RIPK1 at 4 °C. After an overnight incubation, the membranes were washed with Tris-buffered saline with Tween (TBST) four times for 7 min. The next step was to incubate the membranes with the secondary antibody, after which the membranes were washed as described above. Finally, the membranes were subjected to chemiluminescence. We selected GAPDH or β-actin as the reference. A Bio-Rad scanner was used to detect the signal strength on the membranes.

Quantitative real-time PCR

A total RNA extraction kit (Omega Biotek, R6834-01) was used to extract the RNA for qPCR. Complementary DNA was synthesized using Hifair® III 1st Strand cDNA Synthesis SuperMix (Yeasen, 11141ES60). Then, the cDNA was amplified using SYBR Green Master Mix (Yeasen, 11203ES03). The amplification method was described previously. Gapdh served as a reference. All mRNA expression levels were standardized to Gapdh. The primer sequences utilized in the study were Zbp1 (F) 5’-AAGAGTCCCCTGCGATTATTTG-3’ and (R) 5’-TCTGGATGGCGTTTGAATTGG-3’; Irf1 (F) 5’-ATGCCAATCACTCGAATGCG-3’, (R) 5’-TTGTATCGGCCTG TGTGAATG-3’; Inos (F) 5’-GTTCTCAGCCCAACAATAC AAGA-3’, (R) 5’-GTGGACGGGTCGATGTCAC-3’; and Gapdh (F) 5’-AGGTTGTCTCCTGCGACTT CA-3’, (R) 5’-GGGTG GTCCAGGGTTTCTTA-3’. The tests were repeated 3 times.

Detection of mtDNA in cytosolic extracts

mtDNA release was assayed as described previously [25, 26]. The cell fraction was collected as described previously [26, 27]. Briefly, chondrocytes were washed with PBS, 10% of which was used to extract DNA from the entire cell. The other chondrocytes were resuspended in precooled mitochondrial extraction buffer 1 (70 mM sucrose, 2 mg/ml bovine serum albumin, 20 mM HEPES-KOH, 220 mM mannitol, pH 7.5, and 1 mM EDTA) supplemented with a protease inhibitor cocktail and a phosphatase inhibitor. Next, the homogenized cells were centrifuged for 15 minutes (1000 × g, 4°C). The supernatant was centrifuged again at 10,000 × g for 10 minutes at 4°C to pellet the mitochondria from the supernatant of the cytosolic fraction. One part was extracted using a DNA extraction kit (Tiangen Biotech, China), and the extracts were used as standardization controls for total DNA. The other sample was centrifuged to obtain pure cytosolic fractions. DNA was subsequently extracted from the fractions via QIAQuick Nucleotide Removal Columns (QIAGEN, Germany). Then, the total DNA and mitochondrial DNA were amplified using qPCR. The Ct value for the mtDNA number in the whole-cell extracts was regarded as a standardization control for comparing the variation in mtDNA levels in the cytosol. The presence of TERT DNA indicated that nuclear lysis did not occur. The sequences of the primers used in this experiment are D-loop (F) 5’-AATCTACCATCCTCCGTGAAACC-3’ and (R) 5’-TCAGTTTAGCTACCCCCAAGTTTAA-3’; Loop2 (F) 5’-CCCTT CCCCATTTGGTCT-3’, (R) 5’-TGGTTTCACGGAGGATGG-3’; Loop3 (F) 5’-TCCTCCGTGAAACCAACAA − 3’, (R) 5’-AGCGAGAAGAGGGGCATT-3’; and TERT (F) 5’-CTAGCTCATGTGTCAAGACCCTCTT-3’, (R) 5’-GCCAGCACGTTTC TCTCGTT-3’.

Coimmunoprecipitation (co-IP) assay

The co-IP assay was conducted as previously reported [28]. Briefly, chondrocytes were cultured and treated with TNF-α for 15 min. Afterward, the cells were lysed in 1 mL of IP lysis buffer supplemented with protease inhibitors. The samples were centrifuged at 4 °C for 15 min after being lysed on ice for 10 min. Protein A + G magnetic beads were added to the cell lysates, which were then shaken for 1 h to eliminate nonspecific binding. Subsequently, the mixture was shaken overnight at 4 °C after being mixed with the primary antibody. Next, the samples were centrifuged, the supernatant was removed, and the immune complex was washed 3 times with lysis buffer. Finally, lysis buffer and 5× loading buffer were added to the beads. After the mixture was boiled for 10 min at 97 °C, the supernatant was removed for western blot analysis.

Immunofluorescence (IF) staining

First, chondrocytes were seeded in confocal dishes, and the expression of ZBP1 or IRF1 was knocked down according to previous methods. Next, the cells were treated with or without TNF-α for a suitable time period. Then, the chondrocytes were fixed with 4% paraformaldehyde for 10 min, and 0.5% Triton X-100 was used to permeabilize the cell membrane. Afterward, the chondrocytes were incubated with blocking goat serum for 1 h, after which they were incubated with primary antibodies against MMP13, COL2A1, ZBP1, P-P65, P-RIPK1, and double-stranded DNA (dsDNA), as well as the corresponding secondary antibodies. After the incubations, the nuclei were stained with DAPI for 3 min. Next, the cells were washed with PBST between each step. A laser scanning confocal microscope (FV3000, Olympus Corporation, Japan) was used to capture the images.

Histological and immunohistochemical (IHC) analyses

Mouse knee joints were fixed with paraformaldehyde for 1 day. The samples were decalcified for 21 days with 10% EDTA (pH 7.4) and cut into 5-µm thick sagittal sections after they were embedded in paraffin. Next, the samples were stained with safranin O/fast green and hematoxylin–eosin (HE). The OARSI histopathology scoring system was applied in a blinded manner to evaluate cartilage degeneration. After the samples were deparaffinized and rehydrated, BSA containing 0.1% Triton X-100 was applied to block the samples for 1 h. Next, the samples were incubated with anti-COL2A1, anti-P-P65, anti-P-RIPK1, anti-MMP13 or anti-ZBP1 antibodies, incubated with the secondary antibody, and stained with DAB. Finally, hematoxylin was used to counterstain the samples. Images were obtained using a general microscope (BX53, Olympus Corporation, Japan).

Microcomputed tomography (micro-CT) analysis

First, the joints were evaluated with a Viva CT 80 scanner (Scanco Medical AG, Switzerland), and the samples were scanned at a thickness of 10.5 μm. Images of reconstructed 3-dimensional (3D) regions were obtained using built-in software. The dimensions of the tibial plateau osteophytes were detected and quantified in cross-sectional images. The analysis of subchondral bone in the medial condyle of the tibia began at the distal 10th layer of the tibial plateau superior border and ended at the 40th layer. Morphological parameters, such as the bone volume/tissue volume (BV/TV), the number of trabeculae (Tb.N), the trabecular space (Tb.Sp) and the thickness of the trabecula (Tb.Th), were analyzed.

Statistical analysis

The unpaired two-tailed Student’s t test was used to compare two different groups, and one-way ANOVA followed by Tukey’s post hoc test and two-way ANOVA followed by Sidak’s multiple comparisons test for multiple comparisons. A. The data are presented as the means ± standard deviations (SDs). At least three biological replicates were analyzed in each experiment. The statistical analysis was performed with GraphPad Prism 9.0. P < 0.05 was regarded as significant, NS represents not significant, * indicates P < 0.05, and ** indicates P < 0.01.