2.1. Human Cartilage Tissue and Chondrocyte Isolation

OA articular samples (n = 10) were obtained from patients with osteoarthritis who underwent total knee arthroplasty and the normal articular samples (n = 4) were collected from patients with trauma who underwent amputation. The harvest of chondrocytes was performed as previously described [25]. In brief, cartilage tissues were obtained from tibial plateau and cut into 1 mm³ slices, then the slices were digested in trypsin-EDTA (Gibco, USA) followed by 0.2% collagenase II (Biofroxx) treatment for 10 hr. Mixture was filtrated through 70 μm nylon mesh (Fisherbrand, USA) and centrifuged. Afterward, chondrocytes were cultured in Dulbecco’s Modified Eagle’s Medium (DMEM) containing 10% fetal bovine serum (Gibco, USA) and 1% penicillin–streptomycin (Sigma–Aldrich, USA). Chondrocytes were detached and split in new culture plates when reaching 70%–80% confluence. This study was approved by the ethics committee of Yixing People’s Hospital (Approve no. IRB-2022-RESEARCH-157), and all patients provided informed consent.

2.2. RT-qPCR

For the detection of miRNA expression level in cartilage tissues, samples were dissected and ground in liquid nitrogen. Total miRNA was isolated using miRNeasy Mini kit (Qiagen, Germany) following the instructions and the RNA samples were reverse-transcribed into complementary DNA (cDNA) with MiScript Reverse Transcription kit (Qiagen, Germany). qPCR was performed using the TB Green™ Premix Ex Taq™ II kit (Takara, Japan) on ABI 7900 fast real-time PCR system (Applied Biosystems, USA). The expression levels of miRNA were normalized to U6, and all primers were purchased from RiboBio (China).

2.3. Western Blot Assay

The proteins of chondrocytes, which were treated differently, were lysed by RIPA buffer containing protease and phosphatase inhibitor cocktail (Beyotime Biotechnology, China). The concentration was measured by BCA kits (Beyotime Biotechnology, China). After being heated with loading buffer, the proteins were separated by SDA-PAGE before transferring to polyvinylidene fluoride (PVDF) membranes (EMD Millipore Corp, USA). Membranes were blocked by 5% BSA followed by primary antibody incubation. After incubation, membranes were washed and treated with secondary antibody for 2 hr at room temperature. The expressions of proteins were visualized by ECL reagent and recorded by Tanon 5200. The antibodies used in this study were anti-collagen II (Proteintech, 28459-1-AP), aggrecan (Proteintech, 13880-1-AP), GAPDH (Proteintech, 60004-1-Ig), SOX9 (Abcam, ab185966), MMP-13 (Abcam, ab51072), cleaved caspase-3 (Abcam, ab32042), BCL-2 (Cell Signaling Technology, 15071), CX43 (Cell Signaling Technology, 3512), GPX4 (Affinity Biosciences, DF6701), and SLC7A11 (Affinity Biosciences, DF12509).

2.4. Cell Transfection

Chondrocytes were transfected with 50 nM Agomir-1, Agomir-1 negative control (Agomir NC), Antagomir-1 and Antagomir-1 negative control (Antagomir NC), all purchased from RiboBio, using Lipofectamine 3000 reagent (Invitrogen, USA). CX43 plasmids were constructed by GenePharma (China). The package of virus and the test of titer were performed as described before [26]. 1 × 10⁸ lentivirus-transducing units were used to infect chondrocytes in the presence of polybrene (GenePharma, China).

2.5. Cell Proliferation Assays

To detect the proliferation ability of chondrocytes after different treatments, CCK-8 and EdU assays were conducted. Five thousand chondrocytes treated differently were seeded into 96-well plates and incubated at 37°C. At Day 0, 1, 2, and 3, CCK-8 solution (10 μL/well, Dojindo, Japan) were added and cells were further cultured for 4 hr. The absorbance at 450 nm was detected by a microplate reader (Multiskan FC, Thermo Fisher Scientific, USA).

To further assess the proliferation, 1 × 10⁴ chondrocytes treated differently were seeded into 96-well plates. After attachment, chondrocytes were treated with 50 μMz EdU solution (RiboBio, China) and incubated for another 2 hr. Then, cells were fixed in 4% PFA and stained by Apollo solution according to the manufacturer’s instructions. The nucleus was stained with 4′, 6-diamidino-2-phenylindole (DAPI). The images were recorded using fluorescence microscopy, and the rate of positive chondrocytes was measured by ImageJ software.

2.6. Cell Migration Assays

Scratch wound healing assay and transwell assay were conducted to measure the migration ability of chondrocytes. In scratch wound healing assay, 2 × 10⁵/well chondrocytes treated differently were seeded into 6-well plates and cultured in complete medium. After 100% confluence, a scratch was made by a 200 μL pipette tip and carefully rinse the cells with PBS thrice. The scratch images were captured by a light microscope (Nikon, Japan) at 0 and 12 hr following scratching. Healing rates of the scratch were analyzed by ImageJ. In transwell assay, 1 × 10⁵ chondrocytes in a volume of 200 μL serum-free medium were seeded in the upper chamber (Corning, USA, pore size: 8 μm), and 600 μL of complete medium was introduced into the lower chamber. After 24 hr incubation, cells in the upper chamber were wiped gently with a cotton swab, and the lower surface was fixed by PFA and stained with 0.5% crystal violet for 1 min. The migrated chondrocytes were observed by microscope and counted using ImageJ.

2.7. Cell Apoptosis Assay

To assess the apoptosis of chondrocytes, cells were induced by IL-1β and the apoptosis rate was detected using Annexin V-FITC/PI Apoptosis Detection Kit (Vazyme, China) according to manufacturer’s instructions. In brief, after transfection, 4 × 10⁵-treated chondrocytes of each group were seeded into 6-well plates. After 70% confluence, the medium was replaced and 10 ng/ml IL-1β were added to induce cell apoptosis. Chondrocytes were harvested after 24 hr treatment and resuspended in binding buffer. Then, chondrocytes were stained with 5 μL Annexin V-FITC and 5 μL propidium iodide for 10 min. The apoptosis rate was analyzed using a flow cytometer (CytoFLEX, USA).

2.8. Luciferase Reporter Assay

To verify whether CX43 mRNA is the target of miR-1, the luciferase report vectors including pmir-report-CX43 wild-type (WT) and pmir-report-CX43 mutant were transfected into chondrocytes with Agomir-1 or the negative control. Twenty-four hours later, chondrocytes were lysed, and the firefly and renilla luciferase activity were detected by Dual-Luciferase® Assay Kit (Promega, USA).

2.9. Intracellular ROS and Lipid ROS Detection

To detect the intracellular ROS and Lipid ROS, fluorescent probes DCFH-DA (Beyotime Biotechnology, China) and C11-BODIPY (MCE, USA) were used following the manufacturer’s instructions. In brief, 2 × 10⁵ cells/well chondrocytes were seeded into 6-well plates. After incubation for 24 hr, cells from different groups were washed with PBS and treated with 10 μM DCFH-DA or 5 μM C11-BODIPY for 30 min at 37°C in the dark. Chondrocytes were observed and the images were captured using a fluorescence microscope (Leica, China). In the C11-BODIPY experiment, the red fluorescence associated with the unoxidized dye and the green fluorescence associated with the oxidized dye were respectively excited at 561 and 488 nm.

2.10. Detection of Lipid Peroxidation and Fe²⁺ Content

To detect the concentration of MDA (a lipid peroxidation product) in cell lysates, an MDA assay kit (Abcam, USA) was used according to the manufacturer’s instructions. An iron assay kit (Abcam, USA) was used to determine the concentration of intracellular ferrous iron level (Fe²⁺) in cell lysates as described before [27]. In brief, samples were washed with cold PBS and homogenized in iron assay buffer. Five microliter iron reducer was added to each standard well and 5 μL of assay buffer was added to each sample, and then the wells were incubated at 37°C for 30 min. After incubation, 100 µL iron probe was added to each well and incubated for 1 hr. The content was measured at 593 nm using a colorimetric microplate reader.

2.11. Experimental OA Model

To construct experimental OA model, anterior cruciate ligament transection (ACLT) surgery was conducted as described before [28]. All procedures were approved by the ethical review committee of Soochow University (Approve no. 202206A0038). In brief, male C57BL/6 mice (8 weeks) were anesthetized with pentobarbital (35 mg/kg) and the anterior cruciate ligament of the right joint was transected with a microiris scissor. After operation, mice were treated with buprenorphine (0.05 mg/kg) and gentamicin (5 mg/kg) to prevent pain and infection.

Mice in this study were divided into four groups: Sham group, ACLT group, ACLT + Agomir NC group, and ACLT + Agomir-1 group. In third and fourth group, 5 nmol Agomir NC or Agomir-1 was injected into articular cavity for 3 weeks (twice a week) after surgery. In first and second group, PBS was injected as control. After total of 8-weeks induction, all mice were sacrificed and the joints were collected. Collected joints were fixed in 4% PFA, decalcified in 10% ethylenediaminetetraacetic acid (EDTA) solution, and embedded in paraffin for further experiments.

2.12. Histological and Immunofluorescence Staining

The specimens of OA patients and mice were cut into 5 μm, and sections were dewaxed. For histological staining, sections were stained with hematoxylin and eosin (HE), Safranin O/Fast green and Toluidine Blue. The OA progression was evaluated using Osteoarthritis Research Society International (OARSI) score.

For immunofluorescence staining, sections were treated with sodium citrate antigen retrieval solution (Solarbio, China) for antigen retrieval. Sections were then permeabilized with 0.1% Triton X-100 and blocked with 5% BSA. To detect the OA progression, collagen II, aggrecan, and MMP13 primary antibodies were incubated with sections overnight at 4°C. After incubation, sections were washed and treated with fluorescent secondary antibodies. The immunofluorescence was observed by a fluorescence microscopy and the intensity of different groups were analyzed by ImageJ.

2.13. Statistical Analysis

GraphPad Prism 8.4.3 was used for statistical analysis. The student’s t-test was used to compare two groups, whereas one-way analysis of variance (ANOVA) for several groups. All results in this study were presented as the mean ± SD. p < 0.05 was considered statistically significant.

3.1. miR-1 Was Downregulated in Osteoarthritis

Histological staining revealed cartilage erosion and loss of cartilage matrix in OA samples (Figure 1(a)). To verify the expression levels of potential key miRNAs (miR-1, miR-21, miR-155, and miR-17) in OA mentioned by Chang et al. [13] qPCR was conducted. The results indicated that the expressions of miR-1 and miR-17-5p were downregulated, and the expressions of miR-21 and miR-155 were upregulated in OA cartilage (Figure 1(b)). Furthermore, we stimulated chondrocytes with IL-1β to obtain in vitro model of inflammation and examined the expression levels of corresponding miRNAs. The results of qPCR showed the same changes of miRNAs as observed in tissue (Figure 1(c)). These results confirmed that several miRNAs predicted in OA cartilage were altered. The role of miR-1 in regulating OA progression was still unclear, however, thus miR-1 was chosen for further study.

3.2. miR-1 Played an Essential Role in Regulating Chondrocyte Functions

To investigate the role of miR-1 in regulating chondrocyte functions, Agomir-1 and Antagomir-1 were used to transfect chondrocytes. As shown in Figure 2(a)–2(b), EdU assay indicated that the upregulation of miR-1 could promote the proliferation of chondrocytes, while the downregulation of miR-1 inhibited the proliferation. Same result was also observed in CCK-8 assay (Figure 2(c)). Furthermore, the migration ability of chondrocytes was tested by scratch wound healing assay and transwell assay. Results indicated that miR-1 could significantly promote the migration of chondrocytes (Figure 2(d)–2(g)). The existence of inflammation could induce chondrocytes apoptosis, thus we also assessed the protection of miR-1 in IL-1β-induced chondrocyte apoptosis by flow cytometry. The results showed that the apoptosis rate of chondrocytes was significantly reduced in Agomir-1 group, and increased in Antagomir-1 group (Figure 2(h)–2(i)), indicating miR-1 could protect the chondrocytes from inflammation-induced apoptosis. The Western blot results showed that miR-1 could promote the expression of aggrecan, Col2, SOX9, and BCL-2, while reducing the levels of MMP-13 and cleaved-caspase 3 (Figure 2(j)). These results revealed that miR-1 played an essential in regulating chondrocyte functions.

3.3. miR-1 Regulated Chondrocyte Functions by Targeting CX43

As we observed above, miR-1 was essential in regulating chondrocyte functions, however, the mechanism was unclear. Four online databases (Target Scan, miRDB, PicTar, and miRTarBase) were used to investigate the target gene of miR-1, and Connexin 43 (CX43) was predicted to be the target (Figure 3(a)). Next, we construct the WT CX43 3′-UTR containing the target sequences, or mutant sequence into luciferase reporter system to verify the prediction (Figure 3(b)). As shown in Figure 3(c), the relative luciferase activity was significantly decreased when cells were co-transfected with miR-1 and CX43 WT, while the suppression was abolished by mutating the target sites in the 3′UTR of CX43. Furthermore, the results of Western blot demonstrated that the expression of CX43 protein was inhibited by miR-1 overexpression, while the miR-1 inhibitor enhanced CX43 expression (Figure 3(d)).

To explore whether miR-1 regulated chondrocyte functions by targeting CX43, chondrocytes were transfected with Agomir-1 and CX43 plasmids (CX43^OE) or its negative control (CX43^NC). EdU and CCK-8 assays demonstrated that the overexpression of CX43 significantly attenuated the proliferation ability of chondrocytes (Figure 4(a)–4(c)). As shown in Figure 4(d)–4(g), the percentage of wound healing area and migrated cells were decreased in CX43^OE group, indicating the overexpression of CX43 inhibited the migration ability of chondrocytes. Additionally, the apoptosis rate of chondrocytes transfected with CX43 was increased after IL-1β stimulation (Figure 4(h)–4(i)). Moreover, Western blot assay revealed that the genes related to chondrogenesis such as aggrecan, Col2, Sox9, and antiapoptotic gene BCL-2 were decreased on CX43^OE group, while MMP-13 and cleaved caspase-3 were increased (Figure 4(j)). In summary, above results demonstrated that miR-1 promoted chondrocyte functions by targeting CX43.

3.4. miR-1 Inhibited Ferroptosis in Chondrocytes by Targeting CX43, Attenuating Matrix Degradation and Inflammation

Recently, several studies have demonstrated that ferroptosis plays an important role in the progression of OA [29–31]. Previous study found that the downregulation of CX43 can alleviate cisplatin-induced acute kidney injury by inhibiting ferroptosis [32]. However, whether miR-1 could regulate chondrocyte ferroptosis by targeting CX43 was unclear. First, we detected the expression level of CX43 in OA and normal cartilage samples. Western blot analysis showed that CX43 was elevated in OA samples, accompanied by decreased chondrogenic genes and increased catabolic genes (Figure 5(a)). Next, we transfected chondrocytes with Agomir-1 and CX43 to observe the expressions of ferroptosis-related genes induced by IL-1β. As shown in Figure 5(b), CX43 was elevated in chondrocytes accompanied by decreased GPX4 and SLC7A11 after IL-1β stimulation compared to untreated group. Agomir-1 treatment could attenuate the upregulation of CX43, however, and rescued the expression of proteins related to antiferroptosis. Interestingly, the ferroptosis of chondrocytes overexpressing miR-1 was aggravated again after CX43 transfection. To further confirm the role of miR-1 and CX43 in chondrocyte apoptosis, intracellular ROS and lipid ROS were detected. As shown in Figure 5(c)–5(d), IL-1β treatment could increase the level of intracellular ROS and lipid ROS, while Agomir-1 attenuated the accumulation. However, the overexpression of CX43 blocked the protection of miR-1 in preventing chondrocytes from ferroptosis. In addition, compared with the control group, IL-1β stimulated chondrocytes to produce more Fe²⁺ and MDA, and miR-1 reduced the accumulation of Fe²⁺ and MDA in chondrocytes, while CX43 partially abolished the protection of miR-1 (Figures 5(e)–5(f)). In summary, these results demonstrate that miR-1 could modulate chondrocyte ferroptosis by targeting CX43, attenuating the inflammation response and maintaining the homeostasis of chondrocytes.

3.5. miR-1 Inhibited OA Progression in Experimental OA Model

To investigate the protective role of miR-1 in OA progression, ACLT surgery was performed to induce OA in vivo, and Agomir-1 was injected. As shown in Figure 6(a), safranin O/fast green and toluidine blue staining revealed that the cartilage in ACLT and ACLT + Agomir NC group appeared collapse, erosion, and ECM loss, while Agomir-1 injection attenuated the damage. Also, the OARSI score of Agomir-1 group was lower compared with ACLT and ACLT + Agomir NC group (Figure 6(c)). In addition, immunofluorescence staining showed that the expressions of Col2 and aggrecan were significantly reduced after surgery, while Agomir-1 injection could partially reverse the degradation of matrix (Figures 6(b) and 6(d)). In contrast, miR-1 simultaneously decreased the expression of MMP13. These findings confirmed that miR-1 could inhibit the degradation of OA cartilage and attenuate OA progression. Furthermore, the results of immunohistochemical staining showed that the expression of GPX4 in the Agomir-1 group was significantly higher than that in the ACLT and ACLT + Agomir NC groups, indicating that Agomir-1 could alleviate the ferroptosis of chondrocytes (Figures 6(e) and 6(f)).