The hsa-miR-181a-5p reduces oxidation resistance by controlling SECISBP2 in osteoarthritis

The human chondrosarcoma chondrocyte SW1353 cell line was obtained from the Chinese Academy of Sciences (Shanghai, China) and cultured in RPMI-1640 medium (HyClone, USA) with 10% foetal bovine serum (ExCell, China). The murine chondroblast ATDC5 cell line was obtained from the European Collection of Cell Cultures (ECACC) and maintained in Dulbecco’s Modified Eagle’s medium/Ham’s F12 medium (DMEM/F12, HyClone, USA) supplemented with 5% FBS (Gibco, USA). Both cell lines were maintained in a humidified incubator with 5% CO₂ at 37 °C, cultured in monolayers and grown to confluence. The medium contained 1% penicillin/streptomycin (Sigma, USA). The cells were seeded in 12-multiwell plates at 7 × 10⁴ cells/well.

For the cartilage matrix degradation model, SW1353 cells were placed in FBS-free medium for more than 10 h, and then the cells were incubated with 0 (as control), 1, 5, 10 and 20 ng/ml IL-1β (Sino Biological Inc., China) for 12 h, or 10 ng/ml IL-1β for 0 (as control), 6, 12, 24 and 48 h. For the chondrocyte differentiation model, ATDC5 cells were induced with 1× ITS supplement (1 mg/ml insulin, 0.55 mg/ml transferrin and 0.5 μg/ml selenium) added to the medium. The chondrogenic culture medium was changed every day.

SW1353 cells were seeded for 24 h, and 50 nM hsa-miR-181a-5p mimics (mimic-181a-5p) or negative control (mimic-NC) (Genepharma, China) and 200 nM hsa-miR-181a-5p inhibitor (inhibitor-181a-5p) or negative control (inhibitor-NC) (Genepharma, China) were transiently transfected into SW1353 cells by 1.5 μl/well Lipofectamine™ 2000 (Invitrogen, USA) according to the manufacturer’s instructions. Information about miR-181a-5p is provided in Tables 1 and 2.

The full-length human SBP2 CDS was cloned from SW1353 chondrocyte cDNA and inserted into a pEFGP-N1 vector (Invitrogen, USA). The primer sequences for the hSBP2-CDS clone are listed in Table 3. SW1353 cells were seeded for 24 h, and 1, 1.5, 2 and 4 μg of the pEFGP-mSBP2-N1 vector or empty vector were transiently transfected into cells by 1.5 μl/well Lipofectamine™ 2000 (Invitrogen, USA). The expression of exogenous and endogenous SBP2 was determined by western blotting with an anti-SBP2 antibody after transfection for 24 h.

Additionally, hSBP2 siRNA (si-SBP2) and control siRNA (si-NC) sequences were purchased from Genepharma Biotechnology Inc. (Genepharma, China). Cell transfection was performed according to the manufacturer’s instructions. For gene knockdown, SW1353 cells were seeded for 24 h, and 50 nM si-SBP2 (5′-GAGCCACACUACAUUGAAATT-3′) or si-NC was transiently transfected into the cells by 1.5 μl/well Lipofectamine™ 2000 (Invitrogen, USA) according to the manufacturer’s instructions. Knockdown efficiency was determined by western blotting after transfection for 48 h.

OA patients were diagnosed according to the modified Outerbridge classification by The Second Affiliated Hospital, Xi’an Jiaotong University Health Science Center. Articular cartilage samples were obtained at the time of total knee replacement (TKR) from 10 human patients with knee OA (6 women and 4 men; mean ± SEM age: 60 ± 8.3 y) from Shaanxi province, China. All patients were diagnosed with Kellgren and Lawrence grade IV OA. After washing with sterile phosphate buffered saline (PBS), portions of cartilage with a damaged articular surface and portions with a smooth articular surface were used for RNA extraction and immunohistochemistry. Smooth cartilage samples were carefully assessed for any gross signs of degeneration or injury, and only normal-appearing smooth cartilage was used as an internal control (a self control). All cartilage samples were collected without fibrillation. Peripheral blood samples were obtained from 20 OA patients (14 women and 6 men; mean ± SD age: 66.6 ± 5.7 y) and 20 normal control patients (14 women and 6 men; mean ± SD age: 65.9 ± 3.1 y).

For RNA extraction, cartilage tissues were harvested from smooth articular surfaces and damaged articular surfaces of the same patient and chopped into pieces that were smaller than 2 × 2 mm. Then, the pieces were immediately frozen in liquid nitrogen. Total RNA was isolated from cells, tissue pieces or plasma samples using TRIzol® (Invitrogen, USA). cDNA was synthesized from 2 μg of total RNA according to the manufacturer’s instructions (RevertAid™; Fermentas, Canada) in a final volume of 20 ml and stored at − 20 °C until use. Furthermore, miRNA-cDNA was obtained using the One Step PrimeScript® miRNA cDNA Synthesis Kit (Takara, Japan).

Both mRNA and miRNA expression was tested by 10 μl real-time quantitative PCR (RT-qPCR), which was performed on an iQ5 real-time PCR detection system (Bio-Rad, Hercules, CA, USA) with SYBR® Premix Ex Taq™ II (TaKaRa, Japan). Relative gene expression was normalized against GAPDH expression in SW1353 cells or β-Actin expression in ATDC5 cells. Additionally, let-7a was used as the internal reference for miR-181a-5p. The procedure for miRNA-cDNA qPCR consisted of two-step amplification: pre-denaturation at 95 °C for 10 s, followed by PCR amplification with 40 cycles of 95 °C for 5 s and 60 °C for 20 s. Information about the primers and PCR amplification is provided in Tables 4, 5 and 6.

Total protein samples from SW1353 cells or ATDC5 cells (10–20 μg) were separated by 10% SDS-PAGE and transferred to PVDF membranes (EMD Millipore, Darmstadt, Germany). After blocking with 3% non-fat milk in TBST buffer, the membranes were incubated with primary antibodies followed by secondary antibodies conjugated to horseradish peroxidase (HRP) and visualized using an ECL detection system (EMD Millipore, Darmstadt, Germany) on a chemiluminescence imaging system. The primary antibodies included anti-SBP2 (1:500, CA, USA), anti-GPX1 (1:2000, CA, USA), anti-MMP13 (1:1000, Abcam, USA) and anti-β-ACTIN (1:2000, Proteintech, China). The following secondary antibodies were purchased from Beyotime Biotech (Jiangsu, China): horseradish peroxidase-coupled anti-rabbit (1:5000) and anti-mouse (1:5000).

After measuring intrinsic peroxidase activity, articular cartilage sections were blocked with 3% hydrogen peroxide (H₂O₂) and then incubated with 1.5% BSA for 1 h. The sections were covered with anti-SBP2 antibodies (1:250, CA, USA) and incubated at 4 °C in a wet box. After 14 h, all sections were rinsed with PBS and then sequentially incubated with biotinylated secondary antibody for 1 h and DAB reagent (Boster, Wuhan, China) for 5 min at room temperature. Chromogenic reactions were terminated once claybank regions were observed under a microscope. Rabbit IgG was used as a negative control.

Data are presented as the mean ± SEM. The statistical significance of pathological data was calculated by using the Mann-Whitney U test. Means of two groups were compared using Student’s t test, and statistical significance was achieved at P < 0.05 in all tests (*: P < 0.05, **: P < 0.01 and **: P < 0.001). All analyses were performed using GraphPad Prism 6.0 (GraphPad Software, San Diego, CA, USA).

IL-1β was selected to stimulate SW1353 cells, and hsa-miR-181a-5p expression levels were determined by stem loop RT-qPCR. The expression of hsa-miR-181a-5p and MMP13 continuously and robustly increased after treatment with 10 ng/ml IL-1β for 0 (as a control), 6, 12, 24 and 48 h in SW1353 cells, while SBP2 and GPX1 expression was continuously and sharply reduced at the mRNA level (Fig. 1a). Meanwhile, SBP2, GPX1 and MMP13 expression at the protein level showed the same patterns observed at the mRNA level (Fig. 1b). The expression of hsa-miR-181a-5p increased, and the expression of SBP2 at the mRNA level reduced over time after treatment with 0 (as a control), 1, 5, 10 and 20 ng/ml IL-1β for 12 h (Fig. 1c).

To assess the role of SBP2 in chondrocytes, we constructed recombinant hSBP2-CDS clones and si-SBP2 (Fig. 2a) and transfected these constructs into SW1353 cells. Exogenous SBP2 (122 kDa, Fig. 2b) showed remarkable concentration-dependent up-regulation with pEGFP-N1-mSBP2. Overall, taking into consideration both endogenous SBP2 (95 kDa, Fig. 2b) and exogenous SBP2, 2 μg of pEGFP-N1-mSBP2 was the most suitable treatment to achieve SBP2 over-expression. SBP2 over-expression (P = 0.0003) in SW1353 cells elevated both GPX1 (P = 0.0064) and GPX4 (P = 0.0215) mRNA levels, whereas SELS (P = 0.4532) induced no evident changes (Fig. 2c). On the other hand, when SBP2 levels were specifically reduced by si-SBP2 (P = 0.0087), both GPX1 (P = 0.0097) and GPX4 (P = 0.0431) mRNA levels, but not SELS levels (P = 0.2093), were also down-regulated significantly (Fig. 2d). Meanwhile, total GPXs activity was increased (P = 0.0097) by SBP2 over-expression, and total GPXs activity was reduced (P = 0.0023) under SBP2 knockdown conditions (Fig. 2e).

To confirm the roles of miR-181a-5p in chondrocytes, a miR-181a-5p mimic (P = 0.0022) or a miR-181a-5p inhibitor (P = 0.0108) was applied to alter miR-181a-5p levels (Additional file 1: Figure S1). The expression of cartilage-specific genes such as COL2A1, ACAN and MMP13 and total GPXs activity were detected in SW1353 cells after transfection for 24 h. First, mimic-miR-181a-5p down-regulated ACAN (P = 0.0052) and up-regulated MMP13 (P = 0.0095) (Fig. 3a), while inhibitor-miR-181a-5p also significantly up-regulated MMP13 (P = 0.0319) (Fig. 3b). Furthermore, both SBP2 (P = 0.0209) and SBP2 were significantly down-regulated in SW1353 cells when miR-181a-5p was up-regulated by mimic-181a-5p (Fig. 3c and d). In contrast, neither SBP2 nor SBP2 expression changed when miR-181a-5p was down-regulated by inhibitor-181a-5p (Fig. 3c and d). Meanwhile, total GPXs activity was reduced (P = 0.0145) by miR-181a-5p over-expression, and total GPXs activity was increased (P = 0.0143) under miR-181a-5p knockdown conditions (Fig. 3e). In addition, ITS treatment was applied to cultured cells for 14 days as described previously to induce ATDC5 cells to differentiate in vitro [29], and then the expression of mmu-miR-181a-5p, Sbp2 and SBP2 was detected. With chondrocyte differentiation, the expression of mmu-miR-181a-5p showed remarkable up-regulation at D3 (P = 0.0258), D7 (P = 0.0178) and D14 (P = 0.0103), while SBP2 protein expression was significantly reduced, although the expression of Sbp2 was almost constant (Fig. 3f).

Cartilage tissues were obtained from 8 OA patients to detect the expression of miRNA-181a-5p, SBP2 and three pivotal selenoproteins. OA smooth cartilage and damaged cartilage from the same patients undergoing TKR were separated (Fig. 4a). Total RNA was extracted, and RT-qPCR was performed. According to a paired Student’s t test, miRNA-181a-5p expression levels were significantly higher (P = 0.0114) in damaged cartilage than in smooth cartilage of OA patients (Fig. 4b). Meanwhile, although SBP2 mRNA expression was unattenuated in damaged cartilage (Fig. 4c), SBP2 protein expression was reduced in damaged cartilage (Fig. 4d). Furthermore, GPX1 (P = 0.0183) and GPX4 (P = 0.0149) were down-regulated in damaged OA cartilage (Fig. 4e), while SELS showed no significant changes (Fig. 4e).

Peripheral blood was collected from 20 healthy controls and 20 OA patients. To detect the expression of miRNA-181a-5p, SBP2, GPX1, GPX4 and SELS, total RNA from peripheral blood was extracted, and RT-qPCR was performed. The expression of hsa-miRNA-181a-5p (P = 0.0329) in OA peripheral blood was significantly higher than that in normal controls (Fig. 5a), while SBP2 (P = 0.0061) and GPX1 (P = 0.0111) were both lower in OA peripheral blood than in normal controls (Fig. 5b and c). In addition, SELS (P = 0.8160) showed no statistically significant differences (Fig. 5d), and GPX4 was not detected (data not shown). These results suggested that hsa-miRNA-181a-5p is a potential diagnostic biomarker for OA.