Alleviation of temporomandibular joint osteoarthritis by targeting RIPK1‐mediated inflammatory signalling

Abstract Temporomandibular joint osteoarthritis (TMJOA), prevalent in adolescents and the elderly, has serious physical and psychological consequences. TMJOA is a degenerative disease of the cartilage and bone, mostly driven by inflammation, and synoviocytes are the first and most important inflammatory factor releasers. Receptor‐interacting serine/threonine‐protein kinase (RIPK1) promotes inflammatory response and cell death during an array of illnesses. This research aimed to explore the impacts of RIPK1 inhibitor therapy in TMJOA and the mechanism of RIPK1 in inducing inflammation during TMJOA. Herein, inhibition of RIPK1 suppressed the elevated levels of inflammatory factors, nuclear factor kappa B (NF‐κB), along with markers of apoptosis and necroptosis after tumour necrosis factor (TNF)‐α/cycloheximide (CHX) treatment in synoviocytes. Moreover, inflammation models were constructed in vivo through complete Freund's adjuvant (CFA) induction and disc perforation, and the findings supported that RIPK1 inhibition protected TMJ articular cartilage against progressive degradation. RIPK1 regulates NF‐κB activation via cellular inhibitor of apoptosis proteins (cIAP), apoptosis via caspase‐8, and necroptosis via RIPK3/mixed lineage kinase domain‐like (MLKL) in synoviocytes, which in turn facilitates TMJOA inflammation progression.

Inflammation-activated synoviocytes secrete chemokines to attract immunocytes and proinflammatory cytokines, a mechanism that facilitates angiogenesis and extracellular matrix (ECM) degradation. 9creasing studies indicate that synovial fibroblast-mediated inflammation can be manipulated to control arthritic degeneration. 10,11th the overproduction of pro-inflammatory signals, like interleukin (IL)-1β, and TNFα in the synovium, cartilage, or bone, there is a high level of matrix-degrading proteases, mainly matrix metalloproteinases (MMPs) and A Disintegrin and Metalloproteinase with Thrombospondin Motifs (ADAMTs). 12Therefore, new strategies for treating TMJOA should be designed that target synoviocyte-mediated inflammation in order to limit inflammatory progression at an early stage. 13PK1 regulates a series of inflammatory signalling pathways including Tumour necrosis factor receptor 1 (TNFR1), IL-1, Fas ligands, necrosis, apoptosis and others. 14As an important inflammatory signalling regulator, it promotes cell proliferation, apoptosis and necroptosis. 15As a scaffolding molecule, RIPK1 promotes inflammatory gene transcription by activating the mitogen-activated protein kinases (MAPK) and NF-κB pathways. 16RIPK1 is also an active kinase that can drive apoptosis or necrosis, based on the intracellular environment. 17TMJOA occurs when large amounts of inflammatory molecules are present in the joint cavity.Therefore, we hypothesize that these inflammatory factors could converge with RIPK1 and activate downstream NF-κB, caspase-8/MLKL, etc., regulating cellular inflammatory cytokines release, apoptosis, or necroptosis.
These manifestations of TMJOA synoviocytes may be related to the regulatory mechanism of RIPK1.Necrostatin-1 (Nec-1) is a targeted inhibitor that blocks the phosphorylation of RIPK1, thereby interfering with signalling pathways associated with the regulatory role of RIPK1. 18However, one study found that inhibition of RIPK1 also silenced NF-κB and attenuated the generation of inflammatory factors in cisplatin-induced nephrotoxic mice. 19Another study also detected apoptosis and necrosis in synoviocytes following simultaneous exposure in TNFα and CHX in vitro. 14This suggests that RIPK1 may play an important regulatory role in synoviocyte death.
In the past 2 years, several studies have confirmed that RIPK1 is a significant inflammatory signalling regulator. 20Currently, the targeted therapy of RIPK1 is highly valued internationally. 21However, little research has been done on RIPK1 in TMJOA and its contribution to the development of TMJOA is not widely recognized.Therefore, this study focuses on exploring the regulatory mechanism of RIPK1 in TMJOA and verifying the therapeutic effects of its inhibitor, in order to propose new directions for osteoarthritis treatment in clinics.

| Rat model of TMJ inflammation
Chengdu Dashuo Experimental Animal Co., Ltd.provided Sprague-Dawley (SD) rats.A total of 30 SD rats (350 g, ages 10 weeks) were grouped into TMJOA models induced using CFA (CFA, Sigma-Aldrich) or disc perforation.Each model had a sham group (n = 5), a TMJOA model group (n = 5), and a treatment group (n = 5).CFA (50 μL) was given as an injection into the upper compartment of the bilateral TMJs to induce inflammation at day 0. The same amount of saline was injected into the control group.Meanwhile, the disc was perforated to establish the other TMJOA model.In detail, the zygomatic arch was incised in an oblique manner.Then, after exposure to TMJ superior joint space, the disc was dragged out, then drilled by a ballshaped drill, till a hole (1.5 mm in diameter) was made in the centre of the TMJ disc.However, this surgery was not performed in sham rats.
The next day, the groups were injected with either 50 μL Nec-1 (1 mg/ kg) or 50 μL phosphate-buffered solution (PBS) into the bilateral TMJ.
The Nec-1/PBS was injected again on day 4. Postoperative complications were not observed.All SD rats were sacrificed at 1 week for further studies.All rats were euthanized by CO2 overdose inhalation.
Animal results were reported according to the ARRIVE guidelines.

| Histological staining
TMJ specimens were fixed in 4% paraformaldehyde and then decalcified in 10% EDTA for 3 months.After decalcification, the tissues were dehydrated, embedded in paraffin and sectioned (5 μm thick).
The orientation of slices in rat TMJ specimens was sagittal plane orientation.The tissue sections were then stained with haematoxylin and eosin (H&E) (Solarbio), sarfarin O and fast green (Solarbio).The morphology of the cartilage and synovium were scored using the Osteoarthritis Research Society International (OARSI) and arbitrary scales, respectively. 22,23
PBS was supplemented with negative controls.Diaminobenzidine was used to develop colour.After counterstaining with haematoxylin, the samples were placed on an Olympus microscope (Olympus) for observation and photography.Normal synovium samples were rinsed with PBS, then cultured (primary culture of 1 mm 3 ) in dulbecco's modified eagle medium (DMEM)/F12 containing 10% foetal bovine serum (FBS), 1% antibiotics (streptomycin sulfate, and penicillin) (37°C, 5% CO2).The synoviocytes were allowed to free from the synovium.The separated synoviocytes in passages 3 and 5 were subjected to subsequent experiments, and synoviocytes were used to construct a homogenous population.
The cartilage was sectioned into pieces sizing 3 mm 2 and cultured in DMEM/F12 plus 3 mg/mL collagenase I (37°C, 1 h).The chondrocytes were allowed to free from the cartilage.The chondrocytes were rinsed, reconstituted in DMEM/F12, and cultured.There times per week, the medium was replaced.
After 4 h of cell culture at 37°C, synoviocyte viability was checked at 450 nm by a microplate reader (Bio-Rad).
As described previously, 25 the chondrocytes were prepared from condylar fracture tissue, cultured into passage 3, then seeded into plates.Before treatment, TMJ synoviocytes-CM (N-CM, TC-CM, or TCN-CM) were diluted at 1:1 with medium.Gene expression was analysed based on samples harvested at 6, 12 or 24 h of treatment.
After washing synoviocytes with PBS, they were removed and stained with propidium iodide (PI), and cell cycle analysis was performed by employing a FACS Calibur device (BD Biosciences) and CellQuest software.

| Flow cytometry
Annexin V-FITC/PI kit was employed to assess the apoptotic of synoviocytes challenged with TNFα (10 ng/mL) and CHX (10 μg/mL) with or without Nec-1 (40 μM).After harvesting synoviocytes, they were washed with pre-cooled PBS and mixed in a binding buffer.
The mixing buffer was then supplemented with PI (5 L) and Annexin V (5 L), and stored at 4°C for 15 min.The BD FACS Calibur (BD Biosciences) was utilized to characterize the apoptotic stages of synoviocytes at early and late stages.

| Quantitative real-time polymerase chain reaction (qRT-PCR)
TMJ synoviocytes were lysed using RNAiso Plus (Takara).To obtain cDNA, total RNA was extracted and reversely transcribed by Pri-meScriptTM RT kit and gDNA Eraser (Takara).A total of 40 cycles of qRT-PCR was then run on the BIORAD real-time PCR system (CFX-Connect) and Power TB Green PCR Master Mix (Takara).GAPDH is the standard.The 2 −ΔΔct technique was employed to equalize the level of the test gene to the level of GAPDH.Table S1 presents the primer sequences.

| Statistical analysis
Experimental data were analysed and plotted using Graphpad Prism 8 software (Graphpad Prism).Data are presented as mean ± S.D. Multiple comparisons were made using one-way anova and independent samples t-tests were used to compare two groups.

| Inhibiting RIPK1 shows therapeutic effects on the synovium and cartilage in rat TMJOA models
To assess the curative effects of inhibiting RIPK1 in vivo, SD rat models of CFA-induced and disc-perforation-induced TMJOA were established (Figure 1A).After treatment with Nec-1, the degree of swelling and hyperplasia in the perichondral tissue of TMJ was less pronounced than that in the CFA/P group (Figure 1B,C).H&E staining revealed that the CFA/P group had more lymphocyte infiltration in the synovium than the other two groups.However, after Nec-1 treatment, fewer inflammatory cells were recruited (Figure 1D,F).In addition, the CFA/P group presented more severe synovial inflammation than the sham group (Figure 1E,G).
The histopathology of cartilage was assessed by the modified OARSI grading system.The Nec-1 treatment group had a lower OARSI grade than the CFA/P group (Figure 1E,G).The CFA/P group demonstrated heavier cartilage erosion and proteoglycan loss.However, the CFA/P + Nec-1 group showed smoother cartilage and milder proteoglycan loss than other groups (Figure 1D,F).
Nec-1 brought a significant cartilage-protective effect in the CFA-induced inflammation model as the expression of MMP1, MMP3 and MMP-9 dropped (Figure 2A,B).Immunohistochemistry also revealed higher protein expression of MMP1, MMP3, and MMP-9 in the P group than in the SHAM group, which were all reduced in the P + Nec-1 group (Figure 2C,D).

| Nec-1 exerts no effect on the cell cycle and proliferation but maintains the activity of TMJ synoviocytes co-stimulated by TNFα and CHX
Nec-1 (0, 20, 40 and 60 μM) did not decrease the viability or proliferation of TMJ synoviocytes at 6, 12 or 24 h (Figure 3A,B).So, the Nec-1 concentration was set at 40 μM to maintain cell viability in the following in vitro experiments. 26-stimulation with TNF and CHX reduced cell viability at a combination of concentrations (Figure 3C).When 10 ng/mL TNF and 10 μg/mL CHX were co-stimulated, the strongest effect on lowering cell viability was reported in these groups.
As shown in Figure 3B, 40 μM Nec-1 did not inhibit the proliferative viability of TMJ synoviocytes, and cell cycle analysis confirmed this finding (Figure 3D). Figure 3E includes a percentage analysis of cell cycle phases, which demonstrates that Nec-1 did not affect synoviocyte proliferation.
The results showed that MMP1, MMP3, MMP9 and ADAMTs5 appeared to be highly expressed in chondrocytes treated with TC-CM.In contrast, TCN-CM significantly reversed this high expression (Figure 4B).

| Inhibiting RIPK1 suppresses TNFαinduced inflammatory gene expression in TMJ synoviocytes and RIPK1 regulates NF-κB activation via cIAP
Pre-treatment of TMJ synoviocytes with Nec-1 for 1 h before costimulation with TNFα and CHX (Figure 4A).Synoviocytes secreted higher levels of IL-1β and IL-6 after TNF and CHX stimulation.However, pretreatment with Nec-1 notably lowered levels of the two factors (Figure 5A).
The effect of RIPK1 on the MAPK and NF-κB was investigated.Immunofluorescence was used to explore whether NF-κB p65 was activated.TNFα and CHX induced the transfer of NF-κB p65 from the cytoplasm into the cell nucleus.Interestingly, pretreatment with Nec-1 impeded this translocation (Figure 5C).The above results suggest that RIPK1 can activate synovial inflammatory pathways by driving NF-kB into the nucleus via cIAP (Figure 5D).

| RIPK1 works with caspase-8 to regulate the apoptosis and the inflammatory progression in TMJOA
We investigated the regulating mechanism of RIPK1 in TMJOA.Immunohistochemical analysis of synovial tissues from two rat TMJOA models revealed that the expression of caspase-8, an apoptotic marker, was considerably higher in both CFA/P groups compared to the SHAM group but inverted by Nec-1 (Figure 6A).
Compared to the TNFα and CHX group, Annexin V-FITC/PI staining showed a lower level of apoptosis in synoviocytes in the Nec-1 treated group (Figure 6B, C).Correspondingly, Caspase-8 expression was significantly augmented in TNFα-induced synoviocytes, then reversed by Nec-1 (Figure 6D).These observations implicate that RIPK1 in inducing apoptosis by interacting with caspase-8 and thereby influences the evolution of TNF-induced inflammation in TMJOA (Figure 6E).

| RIPK1 influences TMJOA inflammatory development by activating synoviocytes necrosis signalling pathway through interactions with RIPK3, and MLKL
The level of RIPK1/RIPK3/MLKL expression was upregulated in the CFA/P group compared to the control group, whereas it was downregulated in the Nec-1-treated group (Figure 7A,B).
The phosphorylation of PIPK1 and RIPK3, two typical markers of necroptosis, was assessed in vitro.Western blot demonstrated that TNFα and CHX resulted in protein elevation of p-RIPK1 and RIPK3, which was reduced by Nec-1 (Figure 7C).TNFα and CHX co-stimulation markedly elevated the mRNA levels of RIPK1, RIPK3 and MLKL (Figure 7D).As expected, the inflammation triggered by the RIPK1-RIPK3 is mediated by necrosis effector protein MLKL.The above results show that RIPK1 affects TMJOA inflammatory progression by cooperating with RIPK3 and MLKL to activate the synoviocyte necrosis signalling pathway (Figure 7E).

| DISCUSS ION
In the current research, we found that inhibition of RIPK1 could counteract TMJ cartilage inflammation by inhibiting apoptosis, necrosis and NF-κB activity in TMJ synoviocytes.TMJOA features pain, dental malocclusion, and limited mouth opening, always ending up with synovial inflammation, cartilage degeneration and subchondral bone reconstruction. 27,28RIPK1 has been implicated in the pathophysiology of a number of degenerative illnesses, such as amyotrophic lateral sclerosis, as well as neurodegenerative and inflammatory diseases. 21RIPK1 has a unique hydrophobic pocket between the N-and C-terminal ends of the kinase that regulates kinase activation through allosteric modulation. 21Nec-1 binds to this hydrophobic pocket and stabilizes RIPK1 in an inactive state.
Nec-1 can specifically inhibit the phosphorylation of RIPK1.It has been reported the effect of Nec-1 on pathological cartilage thinning is caused by mechanical force. 29However, the regulatory mechanism of RIPK1 in TMJOA has not been explored.Our findings uncovered that inhibition of RIPK1 suppresses synovitis to alleviate TMJOA.
TNFα stimulates cells to produce inflammatory factors or induces cell death. 30TNFα binds to TNFR1 and recruits RIPK1 to form complex I. Simultaneously, cIAPs are also recruited to complex I. Reciprocally, cIAPs ubiquitylate the components inside the complex I. 31 Complex I activates the NF-κB and MAPK pathways, resulting in the translocation of NF-κB from the cytoplasm to the nucleus, which can lead to increased expression of inflammatory genes. 32wever, cIAPs fail to ubiquitinate RIPK1, and the complex IIa builds up to activate caspase-3, thus initiating the process of apoptosis procedure. 33In addition to activating apoptotic mechanisms, RIPK1 induces the expression of inflammatory cytokines and chemokines in apoptotic cells.The combined action of apoptosis and inflammatory cytokines/chemokines may be particularly effective in recruiting and activating inflammatory cells and triggering inflammatory progression.Once caspase-8 is repressed, RIPK1 and RIPK3 associate to form complex IIb, which enhances the phosphorylation of RIPK3.
Once activated, RIPK3 then induces the phosphorylation of MLKL (p-MLKL).As p-MLKL migrates to destroy the cell wall, necroptosis starts. 34Necroptosis, always emerging with the spillover of endogenous ligands for damage-associated molecular patterns (DAMPs), can plunge innate immune cells into inflammatory responses.
In the condition of inflammation, synoviocytes are driven to secrete more proinflammatory cytokines.Meanwhile, the release of aggrecan and collagen II is suppressed, followed by the overproduction of MMP1, MMP3, MMP9 and ADAMTs5, all as key regulators of cartilage destruction. 35,36RIPK1 activity can be triggered by costimulation of TNFα and CHX.CHX also mediates the sensitization of chondrocytes to TNF-induced apoptosis. 37More necrosis and apoptosis markers are produced in the synovium of osteoarthritis mice than in normal mice. 38,39In the present in vitro experiments, we tested the viability of synoviocytes at three-time points, finding that continuous exposure to TNF and CHX at a certain concentration can realize appropriate induction of necroptosis.

2. 4 |
Sample collection and cell cultureTMJ synovium and cartilage were collected from 15 individuals (seven males and eight females, average ages 28.3 yrs, range 21-35 yrs) receiving high condylectomy in the Stomatological Hospital of Chongqing Medical University for condylar fracture.These samples were considered normal when no synovial inflammation, synovial hyperplasia and cartilage inflammation were histologically confirmed by at least two pathologists.