Background
Osteoarthritis (OA) may be a response to superfluous mechanical stress or inflammation, and pro-inflammatory factors, including interleukin-1 (IL-1β), interleukin-6 (IL-6), and tumour necrosis factor-α (TNF-α), are involved in OA pathogenesis [
1,
2]. The phenotypes of cartilage injury processes induced by pro-inflammatory factors are cartilage extracellular matrix (ECM) metabolic disorder, the disruption of cartilage homeostasis, and enhanced expression of matrix degradation enzymes such as MMP13 [
3]. MMP13, a major enzyme hydrolysing type-II collagen (COL2), is a dominant protein component of the cartilage ECM [
4,
5] and a biomarker for arthritis progression and therapeutic effects [
6‐
8].
Reactive oxygen species (ROS) are products of aerobic metabolism that injure DNA, proteins, and cellular membranes [
9‐
11]. Oxidative stress plays important roles in the pathogenesis of OA and cartilage degradation, which is induced by ROS, and traumatic loading increases cartilage oxidation and causes cell death [
12]. In addition, oxidative stress-mediated regulation of the expression of redox-sensitive proteins is regarded as a key mechanism underlying age-related cellular dysfunction and disease progression [
13].
Selenoproteins (Sel) are important members of a network of antioxidant enzymatic systems and minimize damage induced by ROS. They contain selenocysteine (Sec), the 21st proteinogenic amino acid, which is named after the essential biological trace element selenium (Se) and acts as an active-site residue essential for the catalytic activity of selenoproteins [
9‐
11]. The genetic code ‘UGA’, commonly a termination codon in cells, encodes Sec into selenoproteins [
14]. Several special cis-trans elements and trans-acting factors, typically the Sec insertion sequence (SECIS) and Sec insertion sequence binding protein 2 (SECISBP2 or SBP2), regulate selenoprotein biosynthesis [
15,
16]. SECIS, which is located in the selenoprotein mRNA 3′-untranslated region (3′-UTR), binds with SBP2. The function of SBP2 is to carry Sec-tRNA
Sec into the ribosome ‘A site’ to recognize ‘UGA’ as the Sec codon during selenoprotein synthesis [
15,
16].
Intriguingly, osteo-chondroprogenitor-specific deletion of the selenocysteinyl tRNA
Sec gene results in dyschondroplasia phenotypes, particularly those showing abnormal skeletal development in mice [
17]. ‘UGA’ is recognized as a termination codon, and inactive truncated selenoproteins are produced in the presence of insufficient Sec-tRNA
Sec [
18]. Similarly, the TrxR1 short inactive fragment, a two-amino-acid truncated C-terminal motif, leads to the death of human lung carcinoma A549 cells [
19]. However, little is known about how selenoprotein biosynthesis regulates OA cartilage. In particular, the pathway from pro-inflammatory factors to selenoprotein biosynthesis mediated by SBP2 in cartilage is poorly understood.
Moreover, more than 20 miRNAs, such as the cartilage-specific
miR-140-5p, participate in chondrogenesis, cartilage homeostasis and degradation, and chondrocyte metabolism, which are closely associated with OA development [
20‐
22]. Further, miR-9, miR-34a and miR-146a are related with oxidative stress in OA chondrocytes [
23,
24]. In a previous study, we identified a repertoire of miRNAs during the development of rat femoral articular cartilage [
25] and demonstrated that
miR-337 regulates chondrogenesis through a direct target, TGFBR2 [
26]. Specifically,
miR-181a-5p, a member of the
miR-181 family, which is organized into three clusters (
miR-181a/b-1,
miR-181a/b-2, and
miR-181c/d), is positively correlated with chondrogenesis [
25]. Meanwhile, non-hypertrophic articular and hypertrophic MSC-derived chondrocytes showed differential expression of
miR-181a-5p, suggesting that its expression is altered during successive differentiation stages [
27]. Moreover,
miR-181a-5p is predicted to be a target of
hSBP2 by TargetScanHuman7.1, and it may inhibit the expression of the important ECM protein aggrecan (ACAN) in chondrocytes, simultaneously acquiring a negative feedback function for cartilage homeostasis [
28]. However, further investigation is required to understand the oxidation resistance-associated roles of
miR-181a-5p in OA.
In the present study, the glutathione peroxidase-encoding genes GPX1 and GPX4 and the selenoprotein S-encoding gene SELS were examined due to their regulation by SBP2. Hence, we investigated the detailed regulatory relationships among pro-inflammatory factors, miRNA, SBP2 and selenoproteins in the context of oxidation resistance in cartilage. Overall, this study provides the first comprehensive evidence for changes in pro-inflammatory factors in the cartilage antioxidant network during OA and describes the discovery of novel mediators of cartilage oxidative stress and OA pathophysiology. Therefore, our data suggest that miR-181a-5p may be useful for the development of novel early-stage diagnostic and therapeutic strategies for OA.
Methods
Cell culture
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.
Transient transfection of hsa-miR-181a-5p mimics or inhibitor sequences
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.
Table 1
Information of Mature miR-181a-5p
hsa-miR-181a-5p
| MIMAT0000256 | AACAUUCAACGCUGUCGGUGAGU |
mmu-miR-181a-5p
| MIMAT0000210 |
Table 2
Information of Stem-loop hsa-miR-181a
hsa-mir-181a-1 (hsa-mir-213)
| MI0000289 | 1q32.1 | UGAGUUUUGAGGUUGCUUCAGUGAACAUUCAACGCUGUCGGUGAGUUUGGAAUUAAAAUCAAAACCAUCGACCGUUGAUUGUACCCUAUGGCUAACCAUCAUCUACUCCA |
hsa-mir-181a-2
| MI0000269 | 9q33.3 | AGAAGGGCUAUCAGGCCAGCCUUCAGAGGACUCCAAGGAACAUUCAACGCUGUCGGUGAGUUUGGGAUUUGAAAAAACCACUGACCGUUGACUGUACCUUGGGGUCCUUA |
Transient transfection of siRNAs and plasmids
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.
Table 3
Information of human primers for hSBP2-CDS
hSBP2-CDS-Forward
| CAGGTCGGATCCAGACCCGGGgccaccATGGCGTCGGAGGGG |
hSBP2-CDS-Reverse
| TCTGTAGAATTCGGTCCCGGGTAAATTCAAATTCATCAT |
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.
Patients and articular cartilage collection
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).
Total RNA extraction and quantitative PCR analysis
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.
Table 4
Information of miRNA-181a-5p for Real-time PCR
hsa-miRNA-181a-5p
| MIMAT0000858 | CGCAACATTCAACGCTGTC |
hsa-let-7a
| MIMAT0000774 | CGCTGAGGTAGTAGGTTGT |
Reverse primer: GTGCAGGGTCCGAGGT |
Table 5
Information of mouse primers for Real-time PCR
Sbp2
| Forward:CTGCTCCAAAGGCCAAAG | 195 | 60 |
Reverse:GTGATTGCCCTCTGTGTCTTC |
β-Actin
| Forward:AACAGTCCGCCTAGAAGCAC | 281 | 60 |
Reverse:CGTTGACATCCGTAAAGACC |
Table 6
Information of human primers for Real-time PCR
SBP2
| Forward: CCGCAGATTCAGGGATTACT | 92 | 60 |
Reverse: CTTGGAAACGGACCAGTTCT |
ACAN
| Forward: GGCATTTCAGCGGTTCCTTCTC | 135 | 60 |
Reverse: AGCAGTTGTCTCCTCTTCTACGG |
MMP13
| Forward: AATATCTGAACTGGGTCTTCCAAAA | 102 | 60 |
Reverse: CAGACCTGGTTTCCTGAGAACAG |
COL2A1
| Forward: TGGACGATCAGGCGAAACC | 244 | 62 |
Reverse: GCTGCGGATGCTCTCAATCT |
GPx1
| Forward: AAGCTCATCACCTGGTCTCC | 124 | 60 |
Reverse: CGATGTCAATGGTCTGGAAG |
GPx4
| Forward: GCTGTGGAAGTGGATGAAGA | 105 | 60 |
Reverse: TGAGGAACTGTGGAGAGACG |
SELS
| Forward: CACCTATGGCTGGTACATCG | 130 | 60 |
Reverse: AACATCAGGTTCCACAGCAG |
GAPDH
| Forward: CACCCACTCCTCCACCTTTG | 110 | 64 |
Reverse: CCACCACCCTGTTGCTGTAG |
Protein sample preparation and western blotting
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).
Immunohistochemistry staining
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.
Statistical analysis
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).
Discussion
To explore whether
miR-181a-5p and
SBP2 are involved in OA pathogenesis, we established an IL-1β model using the chondrocyte SW1353 cell line. The results showed that IL-1β increased
hsa-miR-181a-5p and decreased
SBP2 in a time- and dose-dependent manner, while both
hsa-miR-181a-5p and
SBP2 seemed to participate in the catabolism pathway and oxidative stress in chondrocytes induced by IL-1β. This finding is in line with our expectation that pro-inflammatory cytokines induce
miR-181a-5p up-regulation in chondrocytes along with SBP2 down-regulation. Coincidentally,
miR-181a-5p up-regulates the expression of caspase-3, PARP, MMP-2, and MMP-9 while repressing chondrocyte proliferation and promoting chondrocyte apoptosis in OA [
22,
30].
Next, we used recombinant plasmids and siRNA sequences targeting
SBP2 to up- or down-regulate the expression of this gene in SW1353 cells. To investigate SBP2-mediated selenoprotein synthesis,
GPX1,
GPX4 and
SELS were selected as representative selenoproteins expressed by chondrocytes in this study not only because these proteins exhibit differential cellular localization and fulfil different functions in physiological and pathological processes in various cells but also because the affinity of their SECIS binding with ‘UGA’ recoding has been categorized as strong, moderate and weak, respectively [
31,
32]
As crucial antioxidant enzymes
, GPX1 and
GPX4 were regulated by
SBP2 up- or down-regulation, while
SELS expression levels were always stabilized; these expression patterns are attributable to the differential SECIS affinities and SBP2 binding efficiencies of these proteins. Our findings suggest that
SBP2 expression did not align with selenoprotein expression regulation, which affected total GPXs activity and oxidation resistance in chondrocytes. Oxidative damage due to the concomitant overproduction of ROS is present in ageing and OA cartilage [
33]. Predictably, oxidative stress destroys normal physiological signalling and contributes to OA [
13]. The synergy between blocked selenoprotein expression and disordered metabolism of the articular cartilage ECM induces chondrocyte apoptosis and contributes to cartilage destruction [
9,
34] In summary, selenoprotein biosynthesis leads to decreased antioxidant stress.
Additionally, we modulated miR-181a-5p expression by using mimic and inhibitor sequences in SW1353 cells. The expression of miR-181a-5p showed remarkable up-regulation, while SBP2 protein expression was significantly reduced. Unexpectedly, SBP2 expression did not change after miR-181a-5p knockdown, which implies that a very complex regulatory network and multiple modulators are involved in SBP2 expression. Furthermore, SBP2 showed a significant negative correlation with miR-181a-5p during the induced differentiation of ATDC5 cells. These results suggest that hsa-miR-181a-5p affects the chondrocyte phenotype by altering oxidation resistance.
The most effective antioxidants are members of the GPx family, but the mechanisms underlying their effects on OA chondrocytes under oxidative stress are not yet fully understood [
9]. Our results established that
miR-181a-5p regulated total GPXs activity by decreasing the expression of
SBP2 in cartilage, leading to chondrocyte apoptosis and cellular damage induced by ROS. SBP2 is required for protection against ROS-induced cellular damage and increased cell survival [
35]. For instance, gene mutations in
SBP2 decreased the expression of several selenoproteins, resulting in a complex multisystem selenoprotein deficiency disorder in humans [
36], and lipid peroxidation products mediated by free radicals increased in the blood [
37]. Further, miR-34a, miR-146a, SOD2, CAT, GPXs and NRF2 are subjected to H
2O
2 stimulus in OA chondrocytes [
24]. Meanwhile, miR-9 is a OA-related effects of oxidative stress in chondrocytes through targets SIRT1 [
23].
Finally, we discovered that miRNA-181a-5p expression was increased, and SBP2 protein and GPX1 and GPX4 mRNA expression were reduced in damaged cartilage. These results suggest that hsa-miRNA-181a-5p, GPX1, GPX4 and SBP2 all participate in the OA cartilage damage process to a certain extent. Despite the inadequate number of samples, our peripheral blood data partly support the hypothesis that miR-181a-5p is released in plasma and may facilitate early-stage diagnosis of OA because it induces ROS to damage cartilage proteins. Currently, few blood-based tests are used for the detection of early-stage OA.