Ubiquitin-specific peptidase 25 exacerbated osteoarthritis progression through facilitating TXNIP ubiquitination and NLRP3 inflammasome activation

Rat chondrocytes were collected following previous steps [21]. Isolated chondrocytes were cultured in DMEM/F12 containing 10% fetal bovine serum (FBS, Gibco), 100 U/ml penicillin and streptomycin (Sigma-Aldrich) at 37 ℃ with 5% CO₂. For IL-1β induction, chondrocytes were subjected to IL-1β (10 ng/ml, R&D System) for 24 h.

USP25 overexpression plasmid (pcDNA3.1/USP25) was constructed by inserting USP25 cDNA fragment into pcDNA3.1 vector and empty vector was used as control. USP25 shRNA (shUSP25) and TXNIP shRNA (shTXNIP) and their control (shNC) were synthesized by Shanghai Genepharma. The overexpression plasmids and shRNAs were transfected into rat chondrocytes with Lipofectamine 3000 (Invitrogen).

The concentration of IL-18, MMP-3, and MMP-13 in the supernatants of chondrocytes was detected using ELISA Kit (R&D System) as per the supplier’s instructions.

RNA was extracted with TRIzol Reagent (Invitrogen). RevertAid First Strand cDNA Synthesis kit was used for cDNA synthesis. Quantification of mRNA levels was conducted using SYBR Green polymerase chain reaction (PCR) Mix (Thermo Fisher) on ABI Prism 7300 SDS System (Applied Biosystem). GAPDH was used as internal control. The primer sequences used were USP25-forward, 5′-CCCTACCATCACAGTCCTTACC-3′; USP25-reverse, 5′-CTGGAGGTATCCGAGACTGAGT-3′; TXNIP-forward, 5′-CAGCAGTGCAAACAGACTTCGG-3′; TXNIP-reverse, 5′-CAGCAGTGCAAACAGACTTCGG-3′; GAPDH-forward, 5′-GTCTCCTCTGACTTCAACAGCG-3′; and GAPDH-reverse, 5′-ACCACCCTGTTGCTGTAGCCAA-3′.

Western blot was performed following steps previously described [22]. Briefly, proteins were isolated from cartilage tissues of experimental rats or chondrocytes with RIPA lysis buffer and quantified using BCA protein kit (Both from Beyotime). 10 µg protein/lane was separated by 10% SDS-PAGE and transferred onto PVDF membranes. Subsequently, the membranes were incubated with primary antibodies against USP25, NLRP3, GSDMD-N, active caspase-1, TXNIP, and GAPDH at 4 °C overnight, followed by an incubation with horseradish peroxidase-conjugated secondary antibody for 1 h. Membranes were visualized with chemiluminescence reagent (Bio-Rad) with a ChemiDoc imaging machine (Bio-Rad).

IL-1β-challenged chondrocytes were transfected with shNC or shUSP25. 48 h after transfection, the original medium was replaced with the fresh medium containing Cycloheximide (10 μg/ml). Then, the expression of TXNIP was determined using the western bolt at 0, 3, 6, 9 h after Cycloheximide treatment.

Cell Counting Kit 8 (CCK-8, Dojindo) was used to assess the proliferation of chondrocytes. Cells were seeded into 96-well plates and CCK-8 reagent (10 μL) was added at indicated time points (0, 24, 48, 72 h). Cell viability was determined by measuring the optical density at 450 nm (OD 450 nm) with a microplate reader (Bio-Rad).

The pyroptosis of chondrocytes was assessed using caspase-1 Detection Kit (ImmunoChemistry) following manufacturer’s instructions. Briefly, cells were incubated with FAM-YVAD-FMK in the dark and then resuspended in PI staining buffer. Pyroptosis was defined as double positive with FAM-YVAD-FMK and PI staining and determined by flow cytometer (BD Biosciences).

ROS production was detected using DCFH-DA provided in the reactive oxygen species assay kit (Beyotime Biotechnology) as per supper’s manual. Briefly, rat chondrocytes were incubated with DCFH-DA for 20 min at 37 °C in the dark. Relative fluorescent intensity was analyzed using flow cytometry with excitation wavelength at 488 nm and emission wavelength at 525 nm.

Co-IP and ubiquitination assay was performed as previously described [23]. Cell lysates were extracted with RIPA buffer and incubated with specific antibodies at 4 °C overnight. Subsequently, the mixture of protein and antibodies was incubated with protein A/G beads (Santa Cruz) at 4 °C for 4 h. Bound protein was detected by western blot. For ubiquitination assay, the cells were treated with Mg132 and then lysed by SDS-free RIPA buffer and immunoprecipitated with indicated antibodies followed by A/G plus agarose. The supernatant was analyzed by western blot using ubiquitin antibody.

Male Sprague–Dawley rats (200 ± 20 g) were obtained from the Animal Center of Chinese Academy of Sciences (Shanghai, China). The OA model was established by injection of monosodium iodoacetate (MIA) into the knee joint cavities of experimental rats as previously described [24]. The rats were randomly divided into three groups: Control, MIA, and MIA + AZ1. For AZ1 treatment, AZ1 was dissolved in DMSO (100 mM, 42.2 mg ml⁻¹) to make a stock. AZ1 was diluted in 100–200 μl of PBS and administered to the rats through intragastric gavage (20 mg/kg). After 6 weeks of indicated treatments, the knee joint tissues were collected for further experiments.

To analyze the histological changes, knee joint tissues were embedded in paraffin and then cut into Sections (5 μm) using a rotary microtome. Then, the sections were stained with hematoxylin–eosin (H&E) and Safranin O to be observed under a light microscope. Cartilage destruction was examined by Safranin O staining and the score was assessed using the Osteoarthritis Research Society International (OARSI) grading system.

Firstly, the effect of IL-1β on the injuries of isolated rat chondrocytes was investigated. The chondrocytes were treated with IL-1β for different time points (0, 24, 48, 72 h), and the ROS production, cell proliferation, and USP25 expression were detected. As shown in Fig. 1A–D, the ROS level and USP25 expression were gradually increased and cell proliferation was gradually decreased in a time-dependent manner. These results suggested that USP25 might be involved in IL-1β-induced ROS production and cell viability suppression.

To evaluate the biological role of USP25, USP25 was successfully silenced in IL-1β-treated OA chondrocytes as indicated by western blot (Fig. 2A). Flow cytometry indicated that USP25 depletion reversed the increase of pyroptosis and ROS production caused by IL-1β stimulation (Fig. 2B and C). ELISA results showed that IL-1β-induced upregulation of pro-inflammatory cytokine (IL-18) and MMPs (MMP-3 and MMP-13) were suppressed by USP25 depletion (Fig. 2D). The protein levels of NLRP3, GSDMD-N, and active caspase-1 were markedly elevated in IL-1β-treated cells, which were decreased by silencing USP25 (Fig. 2E). To sum up, USP25 depletion alleviated IL-1β-induced pyroptosis, ROS production, ECM degradation, and activation of NLRP3 inflammasome.

To investigate the role of ROS in USP25-mediated pyroptosis and NLRP3 inflammasome activation, apocynin (ROS inhibitor) was used in subsequent studies. Firstly, western blot confirmed the successful overexpression of USP25 and the participation of apocynin partially reversed the effect of USP25 overexpression (Fig. 3A). Flow cytometry analysis indicated that apocynin decreased the pyroptosis and ROS production of IL-1β-treated chondrocytes induced by USP25 abundance (Fig. 3B and C). Besides, the expression of IL-18, MMP-3, MMP-13, NLRP3, GSDMD-N, and active caspase-1 increased by USP25 overexpression was decreased by apocynin (Fig. 3D and E). Taken together, USP25 increased ROS production to exacerbate IL-1β-induced injuries in OA chondrocytes.

DUBs modulate biological activities by affecting the degradation or function of substrate proteins [25]. Previous studies showed that TXNIP was associated with ROS production and the activation of NLRP3 inflammasome in OA [26, 27]. Besides, reinforced ubiquitination has been proved to cause decreased TXNIP expression [28, 29]. Therefore, we speculated that the deubiquitinase USP25 might interact with TXNIP and promoted its expression through inhibiting ubiquitylation. A Co-IP assay first confirmed that USP25 bind with TXNIP in IL-1β-treated chondrocytes (Fig. 4A). To verify whether USP25 has a regulatory role in TXNIP, USP25 was silenced in chondrocytes, which substantially reduced the mRNA and protein expression of USP25 but had no significant effect on the level of TXNIP (Fig. 4B and C). Interestingly, the protein expression but not the mRNA expression of TXNIP was decreased by shUSP25 in IL-1β-treated chondrocytes (Fig. 4D and E), suggesting that USP25 deficiency might facilitate the intracellular degradation of TXNIP in 1β-treated chondrocytes. Expectedly, the rate of TXNIP protein degradation was significantly increased in USP25-silenced chondrocytes when cycloheximide was added to prevent new protein synthesis (Fig. 4F and G). Moreover, ubiquitination assay results showed that USP25 knockdown significantly increased the ubiquitination of TXNIP (Fig. 4H). These results suggested that USP25 binds with TXNIP in IL-1β-treated chondrocytes and the knockdown of USP25 decreased TXNIP expression by reinforcing its ubiquitination.

The biological role of the interaction between USP25 and TXNIP was further investigated by overexpressing USP25 while silencing TXNIP in IL-1β-induced chondrocytes (Fig. 5A and B). Flow cytometry results revealed that inhibition of TXNIP reversed the promotive effect of USP25 overexpression on pyroptosis and ROS production (Fig. 5C and D). In addition, TXNIP deficiency abrogated the increase in levels of IL-18, MMP-3, MMP-13, NLRP3, GSDMD-N, and active caspase-1 resulted from USP25 abundance in the synovial tissues of knee joint (Fig. 5E-H). To summarize, USP25 interacted with TXNIP to affect IL-1β-induced injuries in OA chondrocytes.

Afterwards, we investigated the effect of USP25 on OA rat model by using USP25 inhibitor (AZ1). H&E and Safranin O staining indicated that MIA-induced damage in articular tissues was effectively ameliorated by AZ1 administration (Fig. 6A and B). Next, IL-18, MMP-3, and MMP-13 concentration in the synovial fluid was measured by ELISA. The results showed that AZ1 administration caused a significant decrease in the upregulated concentration of IL-18, MMP-3, and MMP-13 in MIA-challenged rats (Fig. 6C-E). Additionally, western blot detected that MIA-induced increased levels of TXNIP, NLRP3, GSDMD-N, active caspase-1 were attenuated by the inhibition of USP25 with AZ1 in cartilage tissues of experimental rats (Fig. 6F). To conclude, USP25 inhibition suppressed OA progression in vivo.