Articular cartilage tissues were collected from 30 OA patients (male, 21; female, 9; age range 52–65 years) undergoing total knee replacement surgery and 30 non-OA patients (male, 19; female, 11; age range 54–67 years) with femoral neck fracture undergoing surgery at the Second People's Hospital of Huai'an (Huai'an, China) between August 2015 and August 2016. The non-OA patients had no known history of OA or rheumatoid arthritis. All patients provided written informed consent and the present study was approved by the ethics committee of the Second People's Hospital of Huai'an.

Human chondrocyte cell line CHON-001 was obtained from American Type Culture Collection (cat no. CRL-2846; Manassas, VA, USA) and incubated in Dulbecco's modified Eagle's medium/nutrient mixture F12 (DMEM/F12; Gibco; Thermo Fisher Scientific, Inc., Waltham, MA, USA) with 10% fetal bovine serum (FBS; Gibco; Thermo Fisher Scientific, Inc.), 1% penicillin-streptomycin solution and 25 µg/ml amphotericin B (Fungizone; Gibco; Thermo Fisher Scientific, Inc.) in an incubator at 37°C with 5% CO₂.

For cell transfection, 5×10⁴ CHON-001 cells/well were seeded into a 24-well plate. miRNA mimic and inhibitor were obtained from Shanghai GenePharma Co., Ltd., (Shanghai, China). The next day, cells were transfected with 50 nM miR-31 mimic (forward: 5′AGGCAAGAUGCUGGCAUAGCU3′, reverse: 5′CUAUGCCAGCAUCUUGCCUUU3′), 50 nM miR-31 inhibitor (5′AGCUAUGCCAGCAUCUUGCCU3′), miR-control (forward: 5′UUCUCCGAACGUGUCACGUTT3′, reverse: 5′ACGUGACACGUUCGGAGAATT3′), 2 µl control plasmid (cat. no. sc-108083; Santa Cruz Biotechnology, Inc., Santa Cruz, CA, USA), 2 µl CXCL12-plasmid (cat. no. sc-422854-ACT; Santa Cruz Biotechnology, Inc.), miR-31 mimic + control plasmid or miR-31 mimic+CXCL12-plasmid using Lipofectamine^® 3000 (Thermo Fisher Scientific, Inc.). After 48 h, the cells were harvested for subsequent experimentation.

Total RNA was extracted by TRIzol^® (Invitrogen; Thermo Fisher Scientific, Inc.). RNA was reverse transcribed into cDNA using a Reverse Transcription system (Promega Corporation, Madison, WI, USA). The qPCR reaction mixture was as follows: SYBR Premix Ex Taq™ II (Bio-Rad Laboratories, Inc., Hercules, CA, USA), 2 µl cDNA, 5 µl 2X master mix, 0.5 µl forward/reverse primer and 2 µl nanopure water. qPCR was performed on an ABI 7900HT machine (Applied Biosystems; Thermo Fisher Scientific, Inc.). Amplification conditions were as follows: 95°C for 5 min, followed by 40 cycles of 95°C for 15 sec and 60°C for 1 min. The primer sequences were obtained from Jinsite Science and Technology (Nanjing) Co., Ltd. (Nanjing, China) and listed in Table I. Data were calculated by 2^−ΔΔCq method (10).

CHON-001 cell lysates were washed with PBS and homogenized in radioimmunoprecipitation assay buffer (Roche Diagnostics, Basel, Switzerland). Proteins (30 µg per lane) were separated by 10% SDS-PAGE and transferred to polyvinylidene fluoride membranes, which were blocked with 5% milk in Tris buffered saline with 0.1% Tween-20 for 2 h at room temperature. Next, they were incubated with primary antibodies against β-actin (cat no. 4970; 1:1,000; Cell Signaling Technology, Inc., Danvers, MA, USA), CXCL12 (cat no. 3530; 1:1,000; Cell Signaling Technology, Inc.), type I collagen (cat no. ab34710; 1:1,000; Abcam) and aggrecan (cat no. ab36861; 1:1,000; Abcam) at 4°C overnight. Membranes were subsequently incubated with horseradish peroxidase-conjugated anti-rabbit IgG secondary antibody (cat no. 7074; 1:5,000; Cell Signaling Technology, Inc.) for 1 h at room temperature. Bands were visualized by SuperSignal^® West Pico Chemiluminescent Substrate (Thermo Fisher Scientific, Inc.). Band density was normalized to β-actin.

Cell viability was assessed by MTT assays at 24, 48 and 72 h after cell transfection. Cells (5×10³ cells/well) were seeded in 96-well plates and incubated for 24 h at 37°C. Next, MTT (20 µl; 5 mg/ml) was added to each well and incubated for another 4 h. Wells were then washed in PBS, dried and 150 µl dimethyl sulfoxide was added. Data were measured by a microtiter plate reader at 490 nm (Thermo Fisher Scientific, Inc.).

The migratory ability of CHON-001 cells was detected using 24-well Transwell chambers. Cells (2×10⁴/well) were plated into the upper chambers and incubated in 200 µl serum-free DMEM (Gibco; Thermo Fisher Scientific, Inc.). The bottom chamber contained DMEM (0.5 ml) with 20% FBS (Gibco; Thermo Fisher Scientific, Inc.). Following a 48 h incubation, cells on the upper chambers were removed, and the migrated cells on the lower chambers were fixed with 10% formalin for 30 min at room temperature and stained with hematoxylin and eosin (H&E) for 15 min at 37°C. Cell migratory ability was assessed by counting the number of H&E-stained cells of five fields of view under a light microscope (Olympus Corporation, Tokyo, Japan).

Prediction of miRNA target sites was performed using microRNA.org (www.microrna.org).

A luciferase reporter (Promega Corporation) for the wild-type (WT) CXCL12 3′untranslated region (UTR) and mutant (MUT) CXCL12 3′UTR was constructed by site-directed mutagenesis with Phusion™ High-Fidelity DNA Polymerase (Thermo Fisher Scientific, Inc.). CHON-001 cells (5×10⁴) were seeded into each well of a 24-well plate. The next day, cells were co-transfected with 50 nM CXCL12-3′UTR-WT or CXCL12-3′UTR-MUT and miR-31 mimics or control using Lipofectamine^® 3000 (Thermo Fisher Scientific, Inc.). After 72 h, cells were harvested to determine luciferase activity with the Dual-Luciferase Reporter Assay system (Beyotime Institute of Biotechnology, Haimen, China) and a luminometer. Firefly luciferase activity was normalized to Renilla.

Data were analyzed by Student's t-test or one-way analysis of variance followed by Tukey's test, using SPSS version 21 (IBM Corp., Armonk, NY, USA). Data were expressed as the mean ± standard deviation. P<0.05 was considered to indicate a statistically significant difference.

The expression of miR-31 in the articular cartilage tissues of patients with and without OA was determined by RT-qPCR. It was found that that miR-31 expression was significantly decreased in OA patients, compared with the non-OA patients (Fig. 1).

To investigate the role of miR-31 in OA progression, bioinformatics software (www.microrna.org) was used to predict the potential targets of miR-31. A total of 6,829 targets were identified, including CXCL12 (Fig. 2A). As it has been previously reported that CXCL12 has critical function in the development of OA (11,12), this gene was selected for further analysis.

Next, the correlation between miR-31 and CXCL12 was verified by dual-luciferase reporter assays, which showed that the luciferase activity of the CXCL12-WT group significantly declined following miR-31 mimic transfection compared with the miR-control group, with no evident alteration in the CXCL12-MUT groups (Fig. 2B).

In addition, the expression of CXCL12 in the articular cartilage tissues of patients was determined by RT-qPCR and western blotting. It was demonstrated that CXCL12 protein (Fig. 2C) and mRNA (Fig. 2D) expression was significantly increased in OA patients compared with non-OA patients.

To further investigated whether miR-31 regulated CXCL12 expression in CHON-001 cells, they were transfected with miR-31 mimics, inhibitor or control mimics. Transfection efficiency was determined 48 h after transfection by RT-qPCR (Fig. 3A). It was revealed that CXCL12 expression was markedly decreased by miR-31 mimic transfection, and increased by miR-31 inhibitor transfection, compared with the control group (Fig. 3B and C).

Compared with the control group, CHON-001 cells transfected with miR-31 mimics exhibited a significant increase in cell viability, while miR-31 inhibitor transection results in a significant decrease in cell viability (Fig. 4A). In addition, it was found that the number of migrated cells in the miR-31 mimic group was increased compared with the control group, whereas miR-31 inhibitor transfection significantly prevented CHON-001 cell migration (Fig. 4B and C).

During the progression of OA, the collagen matrix undergoes a transitional degradation (13,14). Therefore, the expression of type I collagen and aggrecan was detected in the present study. The findings indicated that the protein and mRNA expression of type I collagen and aggrecan was increased by miR-31 mimic transfection, and miR-31 inhibitor had the opposite effect in CHON-001 cells (Fig. 5).

To examine whether miR-31 affected CHON-001 cells by directly targeting CXCL12, control or CXCL12-plasmid was initially transfected into CHON-001 cells to assess transfection efficiency by western blotting (Fig. 6A) and RT-qPCR (Fig. 6B). Following this, miR-31 mimics + control plasmid or miR-31 mimics+CXCL12-plasmid was transfected into CHON-001 cells. The protein and mRNA expression of CXCL12, type I collagen and aggrecan was detected by western blot and RT-PCR, 48 h after transfection. It was found that the protein (Fig. 6C) and mRNA expression (Fig. 6D and E) of type I collagen and aggrecan was notably enhanced by miR-31 mimics, whereas CXCL12 overexpression eliminated these effects.

Furthermore, compared with the cells transfected with miR-31 mimics + control plasmid, cell viability (Fig. 7A) and migration (Fig. 7B and C) significantly decreased in the miR-31 mimic+CXCL12-plasmid transfected cells, compared with the mimics + control plasmid transfected cells (Fig. 7). Taken together, these data indicated that miR-31 promoted CHON-001 cell proliferation and migration, as well as type I collagen and aggrecan expression by directly targeting CXCL12.