The present study was approved by the Institutional Ethics Board of the Hospital of Stomatology, Sun Yat-sen University (Guangzhou, China). Written informed consent was obtained from all subjects.

SFs were collected, between October 2014 and April 2016, during TMJ arthrocentesis from patients with TMJ osteoarthrosis that showed no response to conservative treatment. Briefly, a no. 8 needle was punctured into the upper joint compartment. A total of 2.0 ml lidocaine was infused and then withdrawn. Diluted synovial fluid samples were collected from ~800 patients (age, 16–68 years), and SFs were obtained from 17 of these samples. These 17 patients (age, 18–61 years) had no other systemic diseases; among these patients, 3 were male and 14 were female. In addition, 8 SSSs (~0.3×0.5 cm) were obtained aseptically from patients with TMJ osteoarthrosis at the time of surgical debridement treatment for osteoarthrosis or joint disk perforation. The 8 donors (age, 25–50 years) had no other systemic diseases; among these patients, 1 was male and 7 were female.

SFs from the synovial fluid were washed three times and were then digested with 4 mg/ml type I collagenase for 2.5 h at 37°C. The specimens were dispersed by pipetting and then filtered through a 200-mesh screen. Cells were centrifuged at 300 × g at room temperature for 5 min and cultured with complete culture medium [α-minimum essential medium (α-MEM)] supplemented with 10% fetal bovine serum (FBS) and 1X GlutaMAX (all Gibco; Thermo Fisher Scientific, Inc., Waltham, MA, USA) at 37°C in 5% CO₂. The SSSCs were isolated and cultured in the same manner as the SFCs.

A total of 3 SFs and 6 SSSs samples were employed for surface antigen expression analysis. For surface marker detection, ~300,000 dissociated cells were collected. Following incubation with primary antibodies or isotype control antibodies for 30 min, the cells were centrifuged at 300 × g at room temperature for 8 min. The supernatant was discarded prior to resuspension of the cells. Flow cytometric analysis was performed using an FC 500 flow cytometer (Beckman Coulter, Miami, FL, USA), and the results were analysed using MXP Software version 2.0 (Beckman Coulter). The antibodies used are listed in Table I. Cells obtained from 3 SFs and 3 SSSs (samples 4–6) were used for subsequent experiments.

SFCs and SSSCs derived from 3 samples were mixed and seeded in 96-well plates at a density of 800 cells/well at passage 4.

Cell proliferation was evaluated using Cell Counting kit-8 (Shanghai Yeasen Biotechnology Co., Ltd., Shanghai, China). The reagent was employed with the concentration of 10% per well. Culture medium with the reagent served as a blank control. The optical density (OD) of the supernatant was measured using a microplate reader (Infinite 200; Tecan Group, Ltd., Männedorf, Switzerland) after a 2-h incubation at 37°C. Repeated measurements (n=3) were conducted at each time point. The population doubling (PD) and doubling time (DT) were evaluated using the following formulas: PD = (lnN − lnN0)/ln2 and DT = T/PD. N, OD_cells − OD_{blank control} at the end point; N0, OD_cells − OD_blank control at the initial time point; T, time interval; lnN=log(e,N).

SFCs and SSSCs (passage 4) were plated at a density of 5 cells/cm². After culturing in complete medium for 2 weeks, the cells were fixed in 4% paraformaldehyde for 10 min at room temperature and were then stained with 0.5% crystal violet for counting.

After a 4-week osteogenic induction, the cells were washed with PBS and were fixed with 4% paraformaldehyde at room temperature for 10 min. Subsequently, cells were stained with fresh 0.1% Alizarin Red S solution for 30 min at 37°C and examined under an inverted phase contrast microscope (Axiovert 40; Carl Zeiss AG, Oberkochen, Germany).

Adipogenesis was assessed by Oil Red O staining after 4 weeks of induction. The cells were washed and fixed as aforementioned. Subsequently, 0.3% Oil Red O solution was used to stain the cells for 150–180 sec. The cells were then examined under an inverted phase contrast microscope (Axiovert 40; Carl Zeiss AG).

Histological staining was employed to assess chondrogenic differentiation after a 3-week induction. Cartilage nodules formed by the cells were fixed with 4% paraformaldehyde at 4°C overnight and were then embedded in paraffin. The blocks were cut into 5 μm sections. The expression levels of collagen type II were then detected. Sections were incubated with rabbit anti-human collagen type II antibodies (Sigma-Aldrich; dilution, 1:80; cat. no. SAB4500366) in blocking buffer (10% goat serum sealant, cat. no. SL038; Solarbio, Beijing, China) for 16 h at 4°C. Biotinylated goat anti-rabbit immunoglobulin G (cat. no. SA1022; Wuhan Boster Biological Technology, Ltd., Wuhan, China) was used as a secondary antibody incubating at 37°C for 30 min and was detected using streptavidin-biotin complex reagent (cat. no. SA1022; Wuhan Boster Biological Technology, Ltd.). The staining was visualized with 3,3′-diaminobenzidine (cat. no. AR1022; Wuhan Boster Biological Technology, Ltd.) and was observed under a light microscope (Axioskop 40; Carl Zeiss AG). Sections incubated without the primary antibody served as a control.

Sulfated glycosaminoglycan (GAG) assays were performed to determine the levels of GAG. Chondrogenic nodules were digested overnight at 56°C in 50 μg/ml proteinase K solution Santa Cruz Biotechnology, Inc.) diluted in 100 mM Na₂HPO₄ (pH 8.0), followed by inactivation for 10 min at 90°C. After centrifugation, 500 μl working 1,9-dimethylmethylene blue (DMMB) solution (Santa Cruz Biotechnology, Inc.; 100 ml 1M GuHCl, 1 g sodium formate, 1 ml 98% formic acid, 25 ml 0.64‰ DMMB ethanol solution completed to 500 ml with distilled water was defined as solution A; 100 ml 1M GuHCl, 1 g sodium formate, 1 ml 98% formic acid, 25 ml 100% ethanol completed to 500 ml with distilled water was defined as solution B; working DMMB solution was made by rapidly mixing solution A with solution B) was added to 50 μl treated sample or standard sample. Subsequently, the samples were mixed vigorously for 30 min and centrifuged for 10 min at 12,000 × g at 4°C. Once the supernatant was discarded, 1 ml DMMB decomplexation solution [50 mM sodium acetate (pH 6.8) added with 10% methyl alcohol, 4M GuHCl] was added. After a further 30 min of agitation, absorbance was examined at 656 nm using a microplate reader (Infinite 200; Tecan Group, Ltd.). Double-stranded DNA, which was obtained from the chondrogenic nodules digestion fluid, as detected by Quant-iT PicoGreen dsDNA reagent (Invitrogen; Thermo Fisher Scientific, Inc.), was used as an endogenous control.

Sections were stained with Mayer's hematoxylin for 15 min and with eosin for 1 min. After gradient dehydration and clearing, the sections were mounted and observed under a light microscope (Axioskop 40; Carl Zeiss AG). Histological immunostaining of SSS sections was performed in the same manner as mentioned above. Mouse anti-human CD105 antibody (cat. no. ab11414; dilution, 1:200; Abcam, Cambridge, UK) was used as the primary antibody and biotinylated goat anti-mouae immunoglobulin G (cat. no. A1001; Wuhan Boster Biological Technology, Ltd.) was used as the secondary antibody.

Total RNA was extracted from cells using TRIzol reagent (Invitrogen: Thermo Fisher Scientific, Inc.) and cDNA was synthesized using a Transcriptor First Strand cDNA Synthesis kit (Roche Diagnostics, Basel, Switzerland) according to the manufacturers' protocols. qPCR analyses were performed using a LightCycler^® 480 Probes Master system (Roche Diagnostics) with an initial denaturation at 95°C for 10 min, followed by 40 cycles of 95°C for 15 sec and 60°C for 60 sec. A melting curve analysis was then performed. GAPDH was used as an internal control. Relative mRNA expression levels were evaluated using the following formula: (C_{qtarget gene} − C_qgapdh) _sample − (Cq_{target gene} − Cq_gapdh)_control (23).PCR primer sequences are listed in Table II.

Data were statistically analysed using SPSS 10.0 software (SPSS, Inc., Chicago, IL, USA). Experiments were repeated at least 3 times and numerical data are presented as the mean ± standard deviation. Student's t-test was used to analyze results for 2 independent groups. Comparisons between multiple groups were conducted using one-way analysis of variance (ANOVA) followed by Bonferroni's post hoc test. P<0.05 was considered to indicate a statistically significant difference.

The SFs collected from patients with TMJ osteoarthrosis were membranous and translucent, whereas SSSs were masses of tissue (Fig. 1A and B). HE staining revealed that SSSs exhibited a more complex histological structure, containing intima and subintima (Fig. 1C), whereas the SFs were formed of several layers of cells only, indicating that they were obtained from the intima (Fig. 1D).

Adherent cells were obtained from SFs and SSSs. The SFCs and SSSCs both exhibited a typical fibroblastic spindle shape (Fig. 2A and B). In addition, both cell types exhibited clone-forming potential (Fig. 2C and D). The clone-forming rate of SFCs was slightly lower than that of SSSCs (Fig. 2E; P=0.014).

Cell proliferation curves demonstrated that SFCs exhibited a growth pattern similar to that of SSSCs (Fig. 2F). The PD and DT of SFCs were 2.58±1.01 and 30.73±5.90 h, respectively, whereas the PD and DT of SSSCs were 3.18±1.38 and 24.88±5.07 h, respectively, at passage 4.

The results of a flow cytometric analysis indicated that >95% of SFCs derived from all 3 SFs expressed positive markers of MSCs, including CD90, CD44, CD73 and CD105. Negative markers of MSCs: CD11b, CD19, CD34, CD45 and human leukocyte antigen (HLA)-DR, were positive in <2% of cells (Fig. 3). In addition, 95% of SSSCs derived from all 6 SSSs expressed CD90, CD44 and CD73. However, the percentage of CD105⁺ cells in two of the SSSs was much lower (32.36 and 74.08%), whereas that in the other four samples was >95%. Negative markers of MSCs were expressed in >2% of cells for 3 SSSs (Fig. 4). The results of immunohistochemical staining demonstrated that CD105⁺ cells were located in the intima and subintima, indicating that CD105 was not a specific marker for SMSCs (Fig. 5).