Human samples

Synovium and cartilage from 8 RA patients who underwent a total knee replacement surgery were used in the RA group. Synovium from 8 patients with an ACL injury who had no history of arthritis was used in the control group. Cartilage from the lateral tibial plateau with less cartilage damage of 8 OA patients who underwent a total knee replacement surgery was used in the control group. All human synovium and cartilage tissues were collected from Shantou Central Hospital (Shantou, China). The use of clinical sample for scientific research and written informed consent were obtained from all participants. The Ethics Committee of Shantou Central Hospital approved the study.

Animals

We purchased sixty 8-week-old male C57BL/6 J mice from Experimental Animal Centre of Southern Medical University (Guangzhou, China). To explore the effect of Rspo2 in RA mouse model, sixty 8-week-old male C57BL/6 J mice were randomly divided into five groups: An intra-articular phosphate-buffered saline (PBS) was administered as the sham group (n = 12); an intra-articular 5 μL PBS (vehicle) was administered post modeling antigen-induced arthritis (AIA) as the RA + vehicle group (n = 12); Rspo2 group an intra-articular injected with 1 ng/g recombinant mouse Rspo2 (rmRspo2, R&D systems, Minneapolis, MN, USA, #6946-RS) post AIA modeling (n = 12); Rspo2 neutralizing antibody (Rspo2-NAb) group with intra-articular injection of 10 ng/g Rspo2-NAb (R&D systems, Minneapolis, MN, USA, #773029) post AIA modeling (n = 12); DKK1 group administered with 50 ng/g recombinant mouse DKK1 (rmDKK1, ABclonal Technology Co., Ltd., Wuhan, China, #RP01350LQ) post AIA modeling (n = 12). After AIA modeling, the mice (6 mice per time point) were sacrificed either 4 or 8 weeks later. The animals in this study were provided with a consistent diet and kept in cages free of pathogens, with five mice per cage. The temperature and humidity in their living environment remained constant throughout the study. Additionally, their circadian rhythm was maintained at a 12-h cycle. Euthanasia was carried out through the administration of an overdose of ketamine/xylazine anesthesia, followed by cervical dislocation. All animal experiments were conducted with the approval of the Shantou Central Hospital Committee Animal Care and Use Committee and in accordance with their guidelines and regulations.

Antigen-induced arthritis (AIA)

The experimental model of antigen-induced arthritis (AIA) in mice was conducted according to previously established methods [20]. In brief, an injection of methylated bovine serum albumin (mBSA; Sigma-Aldrich, USA) in 50 mL phosphate-buffered saline (PBS) that was emulsified in 50 mL Freund’s adjuvant (CFA; Sigma-Aldrich, USA) was given subcutaneously to 12-week-old male C57BL/6 J mice on day 0. The mice were injected with 10μL physiological saline containing 10 μg mBSA at the right knee joint after 2 weeks. Sham mice were intra-articularly injected with PBS. The mice were administrated with vehicle, rmRspo2, Rspo2-NAb, or rmDKK1 twice per week on day 3 after modeling AIA.

Histological analysis

The right knee joints were fixed in 4% paraformaldehyde for 24 h at 4 °C, followed by decalcification with 0.5 M EDTA for 21 days and embedding in paraffin. Safranin O/Fast Green and hematoxylin–eosin (HE) staining were performed on 4-μm-thick sections of the samples. Synovium tissues were graded based on three synovial membrane features (synovial lining cell layer, stroma cell density, and inflammatory infiltrate) using HE-stained slides. A score of 0 to 3 represents no change (score: 0), slight change (score: 1), moderate change (score: 2), and strong change (score: 3) [20]. The Osteoarthritis Research Society International (OARSI) scoring system was used to grade cartilage degeneration in Safranin O/Fast Green stained sections. Using a 0–6 subjective scoring system, we evaluated each joint quadrant as follows: medial femoral condyle (MFC), medial tibial plateau (MTP), lateral femoral condyle (LFC), and lateral tibial plateau (LTP). A score of 0 to 6 represents normal cartilage (score: 0), loss of proteoglycan with an intact surface (score: 0.5), superficial fibrillation without loss of cartilage (score: 1), vertical clefts and loss of surface lamina (score: 2), vertical clefts/erosion to the calcified layer lesion over 1–25% of the quadrant width (score: 3), lesion reaches the calcified cartilage over 25–50% of the quadrant width (score: 4), lesion reaches the calcified cartilage over 50–75% of the quadrant width (score: 5), lesion reaches the calcified cartilage over > 75% of the quadrant width (score: 6). For statistics, we analyzed the average score of each section based on two blinded, independent graders.

Cell preparation

Fibroblast-like synoviocytes (FLS) were obtained from ATCC. As reported previously, primary chondrocytes were isolated from the newborn mouse tibia cartilage [20]. FLS were treated with 10 ng/ml recombinant human Rspo2 (rhRspo2, R&D systems, Minneapolis, MN, USA, #3266-RS) or 10 ng/ml recombinant human LGR5 (rhLGR5, MedChemExpress, Shanghai, China), with or without 100 ng/ml Rspo2-NAb (R&D systems, Minneapolis, MN, USA, #AF3266) or 500 ng/ml recombinant human DKK1 (rhDKK1, ABclonal Technology Co., Ltd., Wuhan, China, #RP01343) for 24 h to examine phenotype alterations. Primary chondrocytes were treated with 10 ng/ml rmRspo2, with or without 100 ng/ml Rspo2-NAb or 500 ng/ml rmDKK1 for 24 h to examine phenotype alterations.

Proliferation assay

150,000 FLS were plated per well in a 12-well plate. After 24 h treated with rhRspo2, with or without Rspo2-NAb or rhDKK1, the cells were incubated with 100 μg/mL BrdU (MedChemExpress, Shanghai, China) for 2 h. Three random sights were averaged for the quantification of BrdU-positive cells. A fluorescence microscope (Olympus) was used to obtain the images.

Migration assay

Transwell inserts (Corning #3422, Sunnyvale, CA, USA) were seeded 250 μL of FLS suspension (50,000 cells per well) in high glucose DMEM without FBS (Gibco, Gaithersburg, MD, USA). Inserts were then placed on 24-well plates filled with 750 μL high glucose DMEM with 10% FBS. After migrating to the lower side of the insert, cells were fixed with methanol and stained with 1% crystal violet for 10 min. Three representative high-power fields of each insert were counted under a microscope (Olympus).

Invasion assay

Transwell inserts (Corning #3422, Sunnyvale, CA, USA) were seeded 250 μL of FLS suspension (200,000 cells per well) in high glucose DMEM without FBS (Gibco, Gaithersburg, MD, USA). 16% Matrigel (356231; BD Biosciences, San Diego, CA, USA) high glucose DMEM without FBS was used to coat the inserts. After 5 h, inserts were then placed on a 24-well plate filled with 750 μL high glucose DMEM with 20% FBS. After invading to the lower side of the insert, cells were fixed with methanol and stained with 1% crystal violet for 10 min. Three representative high-power fields of each insert were counted under a microscope (Olympus).

Scratch assay

1,000,000 FLS were plated per well in a 12-well plate. Wounds were scratched in each well using a 200-μl sterile pipette tip after the cells had grown to 100% confluence. The cells were washed with PBS to remove cell debris and subsequently cultured in a basal high glucose DMEM with 10% FBS for 24 h. FLS migration and proliferation across the wound margins were captured using the inverted microscope (Zeiss) and measured using ImageJ software.

Immunohistochemistry and immunofluorescence

The slides were prepared according to our previous publication [20]. Immunohistochemical staining was performed on sections after blocking with 1% goat serum for an hour at 37 °C. The following primary antibodies were incubated overnight at 4 °C: Rspo2 (1:100 for IHC, 11781–1-AP, Proteintech, Wuhan, China), Ki67 (1:100 for IHC, ab15580, Abcam, Cambridge, UK), ACAN (1:100 for IHC, A11691, Abclonal, Wuhan, China), and MMP13 (1:100 for IHC, 18165–1-AP, Proteintech, Wuhan, China). The sections were incubated with species-matched HRP-conjugated secondary antibodies (1:200 for IHC, Jackson ImmunoResearch Laboratories, West Grove, PA, USA) at room temperature for an hour. Then, substrate chromogen was visualized by diaminobenzidine (DAB; ZSGB-Bio, Beijing, China), and counterstained with hematoxylin. For immunofluorescence, the sections were blocked with 5% BSA containing 0.1% Triton X-100 at room temperature for an hour and then incubated overnight at 4 °C with the following primary antibodies: LGR5 (1:100 for IF, TA503316, OriGene, Rockville, USA), β-catenin (1:100 for IF, ab32572, Abcam, Cambridge, UK), vimentin (1:100 for IF, sc-6260, Santa Cruz, Dallas, USA), MMP3 (1:100 for IHC, 17873–1-AP, Proteintech, Wuhan, China), and Col2a1 (1:100 for IF, ab34712, Abcam, Cambridge, UK). After washing with PBS containing 0.05% Tween-20, the sections were incubated with species-matched Alexa 488 or Alexa 594 fluorescent secondary antibodies (Life Technologies, Carlsbad, CA, USA) at room temperature for an hour, and then counterstained with DAPI (Thermo Fisher Scientific, Waltham, MA, USA). The images were captured by using a fluorescence microscope (Olympus).

Western blot analysis

Cells and the synovial tissue were immediately lysed with RIPA lysis buffer (FD009, Fudebio, Hangzhou, China) freshly containing phosphatase and protease inhibitors (FD1002 and FD1001, Fudebio, Hangzhou, China) for 10 min on ice. The content of total protein was separated by 10–12% sodium dodecyl sulfate–polyacrylamide gel electrophoresis and transferred to polyvinylidene difluoride (PVDF) membranes (ISEQ00010, Merck Millipore, Darmstadt, Germany). Membranes were blocked with 5% skimmed milk in TBS containing 0.1% Tween-20 (TBST) for an hour at room temperature and incubated with the primary antibodies overnight at 4 °C. After washing with TBST, PVDF membranes were incubated with species-matched HRP-conjugated secondary antibodies (1:5000, Jackson ImmunoResearch Laboratories, West Grove, PA, USA) for an hour at room temperature. Immunoreactive protein bands were visualized with FDbio-Dura ECL Kit (FD8020, Fudebio, Hangzhou, China). Antibodies used for western blotting were as follows: rabbit anti-Rspo2 (1:1000, 11781–1-AP, Proteintech, Wuhan, China), mouse anti-LGR5 (1:2000, TA503316, OriGene, Rockville, USA), rabbit anti-β-catenin (1:1000, ab32572, Abcam, Cambridge, UK), rabbit anti-Col2a1 (1:1000, ab34712, Abcam, Cambridge, UK), rabbit anti-ACAN (1:1000, 13,880–1-AP, Proteintech, Wuhan, China), rabbit anti-SOX9 (1:1000, ab185966, Abcam, Cambridge, UK), and rabbit anti-MMP13 (1:1000, 18165–1-AP, Proteintech, Wuhan, China).

Quantitative real-time PCR

Total RNA was extracted from cells using TRIzol reagent (Invitrogen, Thermo Fisher Scientific, Waltham, MA, USA) and reverse transcribed using reverse transcription reagents (Vazyme Biotech Co. Ltd, Nanjing, China). cDNA was quantitatively analyzed by using Real-Time PCR Mix (Vazyme Biotech Co. Ltd) on a light cycler (Roche, Basel, Switzerland) with the following primers: human LGR5 Forward 5′-CTC CCA GGT CTG GTG TGT TG-3′, Reverse 5′-GAG GTC TAG GTA GGA GGT GAA G-3′; human β-catenin Forward 5′-AAA GCG GCT GTT AGT CAC TGG-3′, Reverse 5′-CGA GTC ATT GCA TAC TGT CCA T-3′; human TNF-α Forward 5′-CCT CTC TCT AAT CAG CCC TCT G-3′, Reverse 5′-GAG GAC CTG GGA GTA GAT GAG-3′; human IL-1β Forward 5′-ATG ATG GCT TAT TAC AGT GGC AA-3′, Reverse 5′-GTC GGA GAT TCG TAG CTG GA-3′; human IL-6 Forward 5′-ACT CAC CTC TTC AGA ACG AAT TG-3′, Reverse 5′-CCA TCT TTG GAA GGT TCA GGT TG-3′; human β-actin Forward 5′-ACT CAC CTC TTC AGA ACG AAT TG-3′, Reverse 5′-CCA TCT TTG GAA GGT TCA GGT TG-3′; mouse LGR5 Forward 5′-GGA CCA GAT GCG ATA CCG C-3′, Reverse 5′-CAG AGG CGA TGT AGG AGA CTG-3′; mouse β-catenin Forward 5′-ATG GAG CCG GAC AGA AAA GC-3′, Reverse 5′-TGG GAG GTG TCA ACA TCT TCT T-3′; mouse β-actin Forward 5′-GTG ACG TTG ACA TCC GTA AAG A-3′, Reverse 5′-GCC GGA CTC ATC GTA CTC C-3′.

Enzyme-linked immunosorbent assay (ELISA)

Mouse Rspo2 ELISA Kits (Shanghai Enzyme-linked Biotechnology Co., Ltd., Shanghai, China, #ml037989) were used to analyze the level of Rspo2 in the serum of C57BL/6 J mice. Following the manufacturer’s instructions, the ELISA was performed.

Small interfering (siRNA) transfection

Twenty μmol/ml siRNA-LGR5 (Tsingke Biological Technology, Wuhan, China) or the corresponding negative control were transfected into FLS using 2.5μL/mL Lipofectamine 3000 (Thermo Fisher Scientific, Waltham, MA, USA) for 48 h, as per the manufacturer’s instructions. Subsequently, the cells were processed with TRIzol reagent for RNA analysis.

SiRNA1-LGR5: 5′-GGA UGA CAA UGC GUU AAC A-3′ (sense), 5′-UGU UAA CGC AUU GUC AUC C-3′ (antisense); SiRNA2-LGR5: 5′-GCU UGG UAG UUC UAC AUC U-3′ (sense), 5′-AGA UGU AGA ACU ACC AAG C-3′ (antisense); SiRNA3-LGR5: 5′-GAC ACU CUC CAA CCU UAA A-3′ (sense), 5′-UUU AAG GUU GGA GAG UGU C-3′ (antisense); SiRNA negative control: 5′-GUA UGA CAA CAG CCU CAA GTT-3′ (sense), 5′-CUU GAG GCU GUU GUC AUA CTT-3′ (antisense).

Statistical analysis

All experimental analyses were performed in triplicate and were observed by two independent observers. Student’s t-tests were used to analyze differences between two groups, whereas one-way analysis of variance (ANOVA) with Tukey’s multiple comparison test was used for analyzing differences between three groups. GraphPad Prism 6.0 (GraphPad Software Inc., La Jolla, CA, USA) was used for all statistical analyses. The results are presented as the mean values ± standard error of the mean (SEM), and P < 0.05 was considered statistically significant.