Animal experiments

Animal experiments were conducted with the approval of the Zhejiang University Experimental Animal Welfare and Ethics Committee under Institutional Animal Care and Use Committee guidelines (ZJU20230014 for mouse model, ZJU20230341 for pig model). Previous hemarthrosis mice model, which mainly focused on local joint damage, was established by whole-body deletion of the coagulation factor together with joint puncture.39 The hemophilic model would cause systemic changes. To investigate the hemarthrosis-caused liver responses in normal organism, we referenced IA injection model previously used for large animal.40 Though single bleeding would damage the joint function in human,41 mice that only received three or four doses of IA blood injection exhibited significant decrease of sport ability (Supplementary Fig. 16). The serological markers of liver dysfunction were increased after three or four doses of blood injection and the group treated with four doses of blood showed significant increase compared with control (Supplementary Fig. 3A). Therefore, to imitate clinical phenomenon (worse serological liver indexes), the knee hemarthrosis model was established on 10-week-old mice by 4-time blood injection (8 μl per leg) into the bilateral knee articular cavity, once a week. The blood injected was autologous. After collecting the blood through cutting a small area at the terminal of tail, the blood was collected and immediately injected without removing or adding any factors. PBS was used in the Control group. Blood containing 25 μg/ml sc79 (S7863, Selleck, USA) was used in Blood+sc group. The C57BL/6 mice were sacrificed 1 week after final injection, and the joints, the liver, the spleen, and the peripheral blood were collected for subsequent studies. The hemarthrosis model was also established on 6-month-old Yorkshire pigs by 4-week blood injection (2 ml per leg) into the two knee articular cavities, twice a week. PBS was used in the Control group and blood containing 25 μg/ml sc79 was used in Blood+sc group. The Yorkshire pigs were sacrificed 1 week after final injection, and the peripheral blood were collected for subsequent studies. Mouse DMM model was referenced by previous study.42

For systemic COL II antibody (ab34712, Abcam, UK) treatment to block the elevated serological COL II, the antibody were intraperitoneal injection at a dose of 3 mg/kg mice. According to the change of serological COL II level, for the first and second blood injection, the antibody were injected once a week, 1 h after blood treatment. For the third and fourth blood injection, the antibody were injected twice a week, 1 h and 3 d after blood injection. IgG (I5006, Sigma, Germany) was used as a control. For IL-17 antibody (clone 17F3, BioXCell, USA) treatment, the antibody was intraperitoneal injection with 300 μg/mice an hour after very blood treatment at the 1st and 2nd IA blood injection, every 3 d after the 3rd IA blood injection. For the intravenous injection of conditioned media, the frequency is same as antibodies injection, 40 μL/injection. The fibroblasts used in this study were L929 cells from ATCC. The conditioned media were centrifugated to exlude cells before injection.

Clinical data

The data of clinical serum biochemical test of the study were collected with the approval of Second Affiliated Hospital of Zhejiang University School of Medicine (20230735). Baseline characteristics is listed as Supplementary Table 1. During the process of data collection of PTJD group, patients with trauma-induced muti-organ damage were excluded, while those only diagnosed with joint damage were collected. Patients with drug, surgery and diseases history that may influence liver functions were also excluded.

Chondrocyte isolation and cell experiments

Mice chondrocytes were isolated from the knee cartilage of postnatal mice (day 0–4) following a previously published method.43 The collagenaseII-digested cells were cultured with DMEM/F12 medium (Gibco, USA) with 10% fetal bovine serum and 1% penicillin/streptomycin (P/S, Gibco, USA) at 37 °C with 5% CO2. For the detection of ROS level in chondrocytes, the DCFH-DA (S0033S, Beyotime, China) was used for flow cytometry detection according to the manufacturer’s instructions. Hepatocytes used in this study were AML12 cells from ATCC. The cells were cultured and treated with serum (10%, v/v) from the Control or Blood group for 24 h before detection, referenced by the previous study.25 For the activation of splenocytes and PBMCs in vitro, 50 μg/ml (terminal concentration) mouse sCOL II (MC22, Elastin Products Company, USA) was directly added into the media of the cells, which were derived from the hemarthrosis model. After 24-h co-culture, the cells and supernatant were detected or used for the treatment of hepatocytes. 15 μg/ml (terminal concentration) of IL-17 antibody or IgG control was used for in vitro study. For in vitro IGF-2 treatment, 10 ng/ml IGF-2 (292-G2, R&D Systems, USA) was used.

For in vitro blood treatment, the transwell system was used to deal with the cells during blood (5% or 10%, v/v) or blood (5%) containing 5 μg/ml sc79 treatment. Ly294002 (ly) (20 μM, S1105, Selleck, USA) and VitE (100 μM, S4686, Selleck, USA) and pyrintegrin (2 or 10 μM, E0462, Selleck, USA) were also used to treat chondrocytes together with blood (5%).The whole blood was added into the upper well and the chondrocytes were cultured in the lower well. The blood was removed after three days of treatment and the chondrocytes were cultured for another 3 days before collection. Cell viability was assayed by CCK-8 kit (CK04, DOJINDO, Japan) according to the manufacturer’s instruction. For treatment of cartilage explant, 5% (v/v) blood, blood+sc (5 μg/ml) or PBS was directly added into culture medium and cultured for 3 days before detection.

Micro-CT analysis

The fixed joint samples were scanned using micro-CT (U-CT-XUHR, MILabs, Netherlands) at 4 μm resolution. The three-dimensional (3D) structures of the calvarium were reconstructed through MILabs-Rec, and analyzed by IMALYTICS Preclinical 2.1. software of the micro-CT.

Von Frey test

The von Frey test was performed mimicking a previous study.44 The package of the experiment was bought from Stoelting, USA. Briefly, allodynia was evaluated by applying von Frey hairs in ascending order of bending force (force range: 0.02, 0.04, 0.07, 0.16, 0.40, 0.60, 1.0, 1.4 or 2.0 g). The von Frey hair was applied perpendicular to the plantar surface of the hind paw (avoiding the toe pads) for 2–3 s. If no response, the next higher strength of hair was used. If a withdrawal response occurred, the paw was re-tested, starting with the next descending von Frey hair until no response occurred. Four more measurements were made after the first difference was observed. The 50% PWT was determined by using the following formula:

$$50 \% }=(^}+\kappa \delta })/10,000,$$

where Xf is the exact value (in log units) of the final test of von Frey hair, κ is the tabular value for the pattern of the last six positive/negative responses, and δ is the mean difference (in log units) between stimuli. The threshold force required to elicit paw withdrawal (median 50% withdrawal) was determined twice on each hind paw (and averaged) on each testing day, with sequential measurements separated by at least 10 min.

Histology, immunohistochemistry (IHC), and immunofluorescence (IF) assays

Tissue specimens were fixed in 4% paraformaldehyde, washed with running water, dehydrated in a graded ethanol series, vitrified with dimethylbenzene, and embedded in paraffin. Paraffin sections (7 μm) were deparaffinized in xylene, hydrated with gradient ethanol, and stained with standard hematoxylin and eosin (HE) and safranin O/Fast green staining (SO) staining procedures. Cartilage histological scores were calculated from the results of the staining according to the OARSI atlas.45

For IHC or IF analysis, sections were incubated at 4 °C overnight with primary antibodies (Supplementary Table 3). For IHC, the sections were then incubated with horseradish peroxidase-linked secondary antibodies (Supplementary Table 3) for 1 h, and the staining was visualized with diaminobenzidine (DAB) solution (DA1010, Solarbio, China). For IF, sections or cells were subsequently incubated with fluorescein-conjugated secondary antibodies for 1 h and observed under a confocal fluorescence microscope (Olympus, Japan).

Treadmill running

To assess the exercise performance of the mice, they were subjected to treadmill activity at an initial speed (1 m/min) and a gradually increasing speed (2 m/min2). An electric foot shock was used to stimulate mice to run. The number of electric foot shocks was calculated, and the test was terminated when the maximum shock count reached 350.

RNA-seq

The cartilage samples from the newborn mice were collected in the dishes and dealt with blood. The liver samples were collected from the joint-bleeding mice model. RNAseq and library construction was performed by the Beijing Genomics Institution (BGI, China). Briefly, total RNA was extracted from the tissues using Trizol (Invitrogen, USA) according to manual instructions. About 60 mg of tissues were ground into powder by liquid nitrogen in a 2-mL tube for RNA extraction. Then, 25–100 μL of DEPC-treated water was added to dissolve the RNA. Total RNA was qualified and quantified using a Nano Drop and Agilent 2100 bioanalyzer (Thermo Fisher Scientific, USA). Next, purified mRNA was fragmented with fragment buffer at appropriate temperature. First-strand cDNA was generated using random hexamer-primed reverse transcription, followed by a second-strand cDNA synthesis. The cDNA fragments obtained from the previous step were amplified by PCR, and products were purified by Ampure XP Beads. The final library was amplified with phi29 to make DNA nanoball (DNB, China) which had more than 300 copies of one molecular, DNBs were loaded into the patterned nanoarray and single end (liver tissue) or double end (cartilage tissue) 50 bases reads were generated on BGIseq500 platform (BGI-Shenzhen, China).

Quantitative real-time PCR

Quantitative real-time PCR (qPCR) was conducted to evaluate the mRNA levels of genes. At the endpoint of each experiment, the cells or tissues were washed 2 times in PBS and then added with Trizol reagent (9109, TaKaRa, Japan) for RNA extraction. Reverse transcription was conducted by a commercial kit (R323, Vazyme, China) according to the manufacturer’s instructions. The primers used for the qPCR analysis in this study are listed in Supplementary Table 2. qPCR reactions were performed using the TB Green PCR kit (Q711, Vazyme, China) according to the manufacturer’s instructions.

Western blot

RIPA lysis buffer containing protease and phosphatase inhibitors (78442, ThermoFisher Scientific Inc, USA) were used for cell lysis, BCA assay (P0012S, Beyotime, China) was used for quantification of the protein concentration. Overall, 25 μg total proteins were loaded. The gels were run for an hour and half, and the separated proteins were subsequently transferred to PVDF membranes. The membranes were blocked for an hour and then incubated with primary antibodies (Supplementary Table 3) at 4 °C overnight. After they were incubated with secondary antibodies (Supplementary Table 3) at room temperature for 1 h, the blot was imaged (Clinx Science Instruments, China). For detection of p-STAT3 protein, after treatment of sCOL II to mixed splenocytes, CD4+ T cells were sorted and detected by western blot.

Flow cytometry analysis

For flow cytometry analysis of blood and spleen sample, followed by lysis of red blood cells, more than 1 × 105 cells were suspended in 100 μl PBS and incubated with fluorescence-conjugated primary antibody (Supplementary Table 3) on ice for 1 h. The cells were then washed and suspended in PBS and was detected using a CytoFLEX S flow cytometer (Beckman Coulter, USA). FlowJo software (BD Biosciences, USA) was used to analyze the data.

Blood analysis

The cytokines in mouse serum were detected using ELISA assay kits (for IL-6, ml063159, Mlbio, China) (for IL-17, DY421, R&D system, USA) according to the manufacturer’s instructions. Mice (ml063001, Mlbio, China) and human (ml057473, Mlbio, China) sCOL II was detected using ELISA assay kit. The serum biochemical markers were detected by an automatic biomedical analyzer (BS-240vet, Mindray, China).

Single-cell mass cytometry

CyTOF (PLT Tech., China) staining was performed following a previously published method.46 Cells were washed with 1× PBS once and then stained with 100 μL of 250 nM cisplatin (Fluidigm, USA) for 5 min on ice to exclude dead cells, and then incubated in Fc receptor blocking solution before stained with surface antibodies cocktail for 30 min on ice. Cells were washed with FACS buffer (1xPBS + 0.5%BSA) for twice and fixed in 200 μL of intercalation solution (Maxpar Fix and Perm Buffer containing 250 nM 191/193Ir, Fluidigm, USA) overnight. After fixation, cells were washed once with FACS buffer and then perm buffer (eBioscience, USA), and stained with antibodies cocktail (Supplementary Table 4) for 30 min on ice. Cells were washed and resuspended with deionized water, added into 20% EQ beads (Fluidigm, USA), and acquired on a mass cytometer (Helios, Fluidigm, USA).

Cell clusters were identified based on marker expression distribution according to standard definitions of cell type: CD4+ T cells (B220− Gr1− CD3e+ TCRb+ CD4+ CD8−), CD8+ T cells (B220− Gr1− CD3e+ TCRb+ CD4− CD8+), gdT (B220− Gr1− CD3e+ TCRb− TCRgd+), NK (NK1.1+ CD3e− B220− CD19−), B cells (CD3e- B220+ CD19+), monocyte (CD11b+ Ly6C+ NK1.1− CD3- CD19− B220− FcerIa− siglecF− Ly6G−), DC (CD3e− B220+ CD19- CD317+), Neutrophil (CD11b+ Ly6G+), eosinophil (CD11b+ SiglecF+), basophils (CD11b+ Fcer1a+ SiglecF− Ly6G−).

Single-cell real-time quantitative PCR and analysis

Single-Cell Real-Time Quantitative PCR (Single-cell qPCR) was performed according to standard protocols provided by Fluidigm (USA). Briefly, cartilage samples from the joint-bleeding mice model were digested and suspended. Single cells were captured by micropipette under a 40x microscope. Amplified complementary DNA libraries were generated using the C1 Single-Cell Auto Prep Module 1 and Module 2 kits (Fluidigm, USA) and a mixture of outer primers specific to 96 genes (Supplementary Table 5). Genes were selected on the basis of established markers for cell adhesion, inflammation, and chondrocyte lineage based on the scientific literature. Single-cell QPCR analysis were performed with inner (nested) primers on the BioMark platform (Fluidigm, USA) using 96 Dynamic Array IFC chips (Fluidigm, USA), according to the manufacturer’s instructions. Single-cell qPCR data with Ct values were analyzed by the SINGuLAR Analysis Toolset 3.6.2 software (Fluidigm, USA).

CD4+ T-cell isolation and transfer to Rag1-deficient mice

Spleens from C57BL/6 mice were used to isolate CD4+ T cells by isolating CD45+CD3+CD4+CD19-CD11b-NK1.1- cells using a Beckman moflo Astrios EQ flow sortor. Cells were washed and resuspended in saline for intravenous injection. In all, 2.5 × 106 sorted cells were injected to Rag1-deficient mice (GemPharmatech, China) 3 days prior to blood injection.

Data analysis

All data were shown as the mean ± SE. Statistical analysis was performed and charts were constructed with GraphPad Prism version 8.0. Statistical comparisons were calculated using a two-tailed Student’s t test (2 groups) or one-way ANOVA with Tukey’s correction (multiple groups). A P value of <0.05 was considered statistically significant. In transcriptomic data, P value of <0.05 and fold change ≥2 were considered as DEGs.