MEG3 inhibits proliferation and promote apoptosis in osteoarthritis chondrocytes by miR-361-5p/wnt/ β-catenin axis

Background: This study aimed to investigate the role of long noncoding RNA (lncRNA) maternally expressed 3 (MEG3) and related molecular mechanisms in osteoarthritis (OA). Methods: Patients with OA and patients undergoing thigh amputation were involved in OA group and control group, respectively. Cartilage tissues of all patients were isolated and cultured. Based on different transfection, MEG3 cells were grouped into Blank, pcDNA3.1-NC, pcDNA3.1-MEG3, si-NC, si-MEG3, pcDNA3.1-NC + mimics NC, pcDNA3.1-MEG3 + mimics NC, pcDNA3.1-NC + miR-361-5p mimics and pcDNA3.1-MEG3 + miR-361-5p mimics group. The cells transfected with pcDNA3.1-NC and pcDNA3.1-MEG3, and then cultured with XAV939 was named as pcDNA3.1-NC +XAV939 group and pcDNA3.1-MEG3 + XAV939 group respectively. The RT-qPCR was used to detect the expression of MEG3 and miR-361-5p . Moreover, Western blot, luciferase reporter assay, RIP, CCK-8 and ow cytometry analysis were performed to reveal the morphology, proliferation and apoptosis in cartilage cells. Finally, the histological analysis and immunostaining were performed on OA rat model. Results: The expression of lncRNA MEG3 and miR-361-5p in OA was signicantly decreased and increased respectively than that in normal. Meanwhile, MEG3 was competitive binding with miR-361-5p in OA chondrocytes. Moreover, the Western blot and CCK-8 assay showed that MEG3 might inhibit cell proliferation and promote cell apoptosis via Wnt/β-catenin pathway. Finally, rat model analysis showed that MEG3 contributed to the cartilage matrix degradation. Conclusion: MEG3 and miR-361-5p might down-regulated and up-regulated respectively in the chondrocytes of OA patients. Furthermore, MEG3


Background
Osteoarthritis (OA) is a disease increasing with age and even induces serious pain and disability [1]. It lead to various pathological changes including articular cartilage degradation, synovial in ammation and subchondral osteoblast activation [2]. Since various genetic, biological, and biomechanical components are proved to be associated with OA [3], the molecular mechanisms underlying the progression of OA is not completely understood.
The lncRNAs play important regulatory roles in oncogenic pathways [4][5][6][7][8]. The biological function of lncRNAs is proved to be associated with knee OA progression [9]. As a member of lncRNA, maternally expressed 3 (MEG3) is proved to participated in the development of tumors [10]. Recently, MEG3 is found to be a potential target for OA, which was con rmed to participate in various cancer including lung cancer [11], breast cancer [12] and esophageal cancer [13]. Su et al. showed that MEG3 was suppressed and inversely related to vascular endothelial growth factor A levels in OA. [14]. Actually, the biological function of MEG3 is realized by targeting certain pathways such as wnt/β-catenin signaling pathway [15]. During this process, the wnt/β-catenin signaling related factors such as Matrix metallopeptidase 1 (MMP-1) and MMP-13 are highly overexpressed in OA [16]. Cui et al. showed that MEG3 was closed associated with wnt/β-catenin signaling pathway, and further play an important role in glioblastoma [17]. Xia et al.
indicated that the suppression of MEG3 take part in the biological function of lung cancer cells via promoting the expression of wnt/β-catenin [16]. Meanwhile, pervious study shows that miR-361-5p hindered the tumor cells mobility via this pathway [18]. Interestingly, MEG3 is proved to be downregulated and inversely associated with VEGF levels in OA [19]. Jin et al. indicated that the downregulation of MEG3 leaded to OA progression via certain miRNA-target gene axis [20]. Although the effect of MEG3 in cancer progression via certain genes or pathways are mentioned in previous studies, the potential mechanism of MEG3 during OA progression is still unclear.
In the current analysis, the mechanism of lncRNA MEG3 and related pathways in OA was explored based on cartilage cells obtained from patients with OA (OA group) and arti cial joint replacement (normal group). The RT-qPCR, Western blot, luciferase reporter assay, RNA immunoprecipitation (RIP), cholecystokinin (CCK-8) and ow cytometry analysis were performed to reveal the morphology, proliferation and apoptosis in cartilage cells. Finally, the histological analysis and immunostaining were performed on OA rat model. The study might provide a new target and theoretical basis for OA treatment.

Patients and grouping
The cartilage tissue of OA patients came from the knee joints of 30 patients who had undergone total knee arthroplasty. Meanwhile, the healthy cartilage tissue came from 20 patients who had not undergone OA or RA (rheumatoid arthritis). All patients voluntarily signed the informed notice. The current study obtained the approval of Ethics Committee of the hospital (ethic vote 198/203).

Cell culture
The femoral articular cartilage of the femur was obtained and cut. Then, it was digested by 0.2% type II collagenase, oscillation resolving for 40min at 37℃, washed by D-Hanks, and diluted in DMEM/F12 (Gibco, USA) combined with 10% FBS, 100 u/ml penicillin (Gibco, USA) as well as 100 mg/ml streptomycin (Gibco, USA). Then cells were culture in a saturated humidity incubator (37°C, 5% CO 2 ). The solution was changed every two days until the chondrocytes grew into sheets and covered more than 85% of the wall of the bottle [21]. The cells of 2 or 3 generations were enrolled for the further investigation.
Real-time uorescent quantitative PCR Total RNA (500 ng) from sample of each group was obtained as well as quanti ed using TRIzol reagent, and cDNA template was synthesized by reagent kit (invitrogen, San Diego, USA). GAPDH was used as reference ( Table 1). The associated conditions were: 95°C for 3 min, 39 cycles at 95 °C for 10 s, 55 °C.
Fluorescence signals were collected at the end points of each cycle extension, followed by the ampli cation curve investigation. Relative expression of candidate genes were calculated by 2 -∆∆CT method [22].

Luciferase reporter assay
The regulatory relation between MEG3 and miR-361-5p was revealed by using StarBase. A wild type (MEG3-WT) or mutant (MEG3-MUT) fragment of the MEG3 3'UTR containing miR-361-5p was synthesized. Then, the wild type or mutant were cloned into the pmirGLO reporter vector (Promega, USA). After inoculation of human chondrocytes in 24-well plates (5 × 105/well) for 24 hours, miR-361-5p mimic or miR-361-5p NC, MEG3-WT or MEG3-MUT were co-transfected into human chondrocytes by Lipofectamine 3000 (Thermo Fisher Scienti c), followed by Kit (Promega, E1910, USA) detection 48 hours after transfection. RIP assay RIP was determined by using a Magna RIPTM RNA kit (Millipore, USA). Brie y, the cultured chondrocytes were further suspended in RIP lysis buffer (Solarbio). Then, cell extract was incubated overnight with RIP buffer containing human anti-Ago2 antibody beads (Millipore) (Input and normal IgG served as controls). Finally, RNA samples was extract using TRIzol reagent, followed by relative enrichment of MEG3 and miR-361-5p detection.

Flow cytometry assay
Simply, the cells were treated by trypsin combining with 200 μL Annexin V-FITC, incubated for 10 min in dark, and then washed with 200 μL PBS and 10 μL PI. Cell cycle was then detected by using ow cytometry (Beckman Coulter), followed by the data analysis.

OA rat model construction
Male SD rats (200-250 g) were obtained from the Experimental Animal Center of Taishan Medical College. All rats were anesthetized with intramuscular injection of sodium pentobarbital (0.05 mg/g, Chuangdong Co., Chongqing, China). Then, SD rats was traversed by the medial collateral ligament and destabilized by the medial meniscus (DMM). One week after the operation, si-NC, si-MEG3 (1×109 PFU, 20 μl) was injected into the knee joint of the recipient rat (n = 6 for each group, 20 μL per joint) twice a week for 4 weeks. Eight weeks after the operation, the rats were sacri ced with cervical dislocation method (external force dislocated the cervical vertebra of rat and disconnects the spinal cord from the cerebrospinal cord), and then the knee joints were harvested. All experiment was performed in Experimental Center of Taishan Medical College. This study was approved by the Laboratory Animal Ethics Committee of Taishan Medical College (No. 2019146), and all experiments were in accordance with the guide for the care and use of laboratory animal.

Histological and immunostaining investigation
Cartilage samples was placed in paraformaldehyde (4%), embedded in para n and cut into sample slice (5μm/slice). The cartilage destruction was evaluated by using the Safranin 'O' staining. Histological scores were performed according to the International Osteoarthritis Research Association (OARSI) grading system, which ranging from 0 (normal) to 6 (>80% represented the cartilage loss). Scores were determined from multiple serial sections of the knee joint of each mouse.

Statistical Analysis
The SPSS 18.0 (Chicago, IL, USA) was selected as the software for data analysis. Meanwhile, all the data in current investigation were represented as the mean ± standard deviation (SD). Signi cant differences between two groups were assessed using Student's t-test, while least signi cant difference between means (LSD)-t multiple comparison test was processed for more than two groups. P < 0.05 was considered as statistically signi cant [24].

Results
MEG3 regulated OA cell proliferation, apoptosis, and cartilage matrix degradation Compared with Normal group, MEG3 in chondrocytes of OA group was signi cantly decreased (P < 0.001) ( Figure 1A). After transfection with pcDNA3.1-MEG3 or MEG3 siRNA into OA chondrocytes, the expression of MEG3 in pcDNA3.1-MEG3 group was dramatically promoted compared with that in pcDNA3.1-NC group (P < 0.05). Moreover, compared with the si-NC group, the expression of MEG3 in the si-MEG3 group was signi cantly suppressed ( Figure 1B) (P < 0.05). Moreover, the cell proliferation ability of the pcDNA3.1-MEG3 group was dramatically decreased than that of the pcDNA3.1-NC group (P < 0.05) ( Figure 1C), while the cell proliferation ability of the si-MEG3 was dramatically promoted compared to the si-NC group ( Figure 1D). Furthermore, compared with the pcDNA3.1-NC group, PCNA and the marker ki67 in the pcDNA3.1-MEG3 group was dramatically suppressed, and the expression of the PCNA and the marker ki67 in the pcDNA3.1-MEG3 group was dramatically promoted ( Figure 1E). In addition, the apoptosis rate of pcDNA3.1-MEG3 group was signi cantly promoted compared with pcDNA3.1-NC group, and the apoptosis rate of si-MEG3 group was dramatically suppressed than that of si-NC group ( Figure  1F). Compared with pcDNA3.1-NC group, Bcl-2 was up-regulated and Bax was suppressed in pcDNA3.1-MEG3 group, respectively. Meanwhile, compared with si-NC group, Bcl-2 was suppressed, while Bax was promoted in si-MEG3 group, respectively ( Figure 1G). The rat OA model analysis showed that compared with pcDNA3.1-NC group, MMP1, MMP13, ADAMTS-5 was up-regulated, while Collagen II and Aggrecan expression were down-regulated in cDNA3.1-MEG3 group ( Figure 1H). Compared with si-NC group, MMP1, MMP13, ADAMTS-5 was down-regulated, while Collagen II and Aggrecan was promoted in si-MEG3 group.

MEG3 contributed to the cartilage matrix degradation in rats
Compared with the OA group, the expression of MEG3 in and control group and OA + si-MEG3 group was both decreased ( Figure 6A). Injection of si-MEG3 effectively reduced the cartilage damage of the operation, protected the cartilage from degradation, and reduced the loss of proteoglycan and joint soft cell ( Figure 6B). Moreover, the interference with MEG3 inhibited cartilage bone marrow matrix degradation ( Figure 6C).

Discussion
Although OA affecting about 237 million people worldwide [25], the detail molecule mechanism of OA is still not fully investigated until now. In the current study, the RT-qPCR analysis revealed that the lncRNA MEG3 and miR-361-5p in OA was signi cantly decreased and increased respectively than that in normal. Meanwhile, StarBase prediction showed that the MEG3 was competitive binding miR-361-5p in OA chondrocytes. Moreover, the Western blot and CCK-8 assay showed that MEG3 might inhibit cell proliferation and promote cell apoptosis via Wnt/β-catenin pathway.
As a maternally expressed imprinted gene, MEG3 can be found in various normal tissues except for tumor cells [11,26]. A previous study shows that the downregulation of MEG3 is associated with the expression of tumor genes in epithelial cells [27]. Ying et al. showed that the suppression of MEG3 was related with the progression of bladder cancer [28]. A previous study shows that the low expression induced by gene knockdown can inhibit apoptosis of chondrocyte in OA rat model [30]. Actually, the low expression of MEG3 has also been reveled in OA. MEG3 has been proved to participate in the development of OA via miR-16/AMAD7 interaction [20]. A previous study indicates that MEG3 is down-regulated and inversely associated with vascular endothelial growth factor A levels in OA. [14]. In the current study, the RT-qPCR analysis showed that the expression of MEG3 in OA group was signi cantly decreased than that in normal group. Thus, we speculated that MEG3 might down-regulated in the chondrocytes of patients with OA.
MEG3 has been proved to play an vital role in the process of tumor in previous studies [31]. Xu et al. revealed that knockdown of MEG3 induced proliferation of rat chondrocytes and inhibited apoptosis [20].
Sun et al. revealed that down-regulated MEG3 promoted cell proliferation in gastric cancer [29]. A previous study shows that MEG3 take part in the process of lung cancer cells function, and suppress apoptosis [32]. Actually, the biological function of MEG3 is realized by involving certain signaling pathway. The previous mechanics, in ammatory mediators and aging collide studies indicate that wnt/βcatenin signaling is critically involved in these processes [34]. In fact, the down-regulated MEG3 can activate in wnt/β-catenin signaling pathway, and then enhance cisplatin resistance in lung and glioma cancer [16]. Liu et al. indicated that MEG3 participated in the progression of carcinoma by via the this pathway [35]. Gao et al. showed that the MEG3 low-expression contributed to retinoblastoma progression via wnt/β-catenin pathway [36]. In adult β-catenin conditional activation rat, OA-like phenotype were shown in articular chondrocytes [37]. In this study, the Western blot, CCK-8 and ow cytometry assay showed that MEG3 not only promoted the proliferation, but also inhibited apoptosis in OA chondrocytes cells. Meanwhile, StarBase prediction showed that MEG3 competitive binding miR-361-5p in OA chondrocytes. Moreover, the RT-qPCR analysis for expression of PCNA and ki67 showed that the effect of MEG3 on proliferation, apoptosis and cartilage matrix degradation of OA chondrocytes was inhibited by miR-361-5p. Importantly, to reveal whether a target gene participates in certain pathway, the detection of core protein (β-catenin) and related factors (e.g. MMP-1, MMP-133) in wnt/β-catenin signaling pathway is needed [38]. A previous study indicates that the low expression of MMP13 is an effective method for decrease articular cartilage loss in OA models [39]. In the current study, MMP-1, MMP-13 and β-catenin detection proved that EMG3 involved in the regulation of Wnt/β-catenin during OA progression. Thus, based on these results in current study, we speculated that MEG3 might inhibit cell proliferation and promote cell apoptosis via miR-361-5p/Wnt/β-catenin axis in OA chondrocytes.

Conclusions
In conclusion, MEG3 and miR-361-5p might down-regulated and up-regulated respectively in the chondrocytes of OA patients. Furthermore, MEG3 might inhibit cell proliferation and promote cell apoptosis via miR-361-5p/Wnt/β-catenin axis in OA chondrocytes. Availability of data and material: All data generated or analyzed during this study are included in this published article and its supplementary information les.

Abbreviations
Competing interests: The authors declare that they have no competing interests.
Funding: Not applicable.
Authors' contributions: AYW, NXH and YFZ designed and analyzed the experiment, and was a major contributor in writing the manuscript. YZC, CHS, YL and YS performed the experiment. All authors read and approved the nal manuscript. Table   Table 1 The amplification primer used for current RT-qPCR analysis

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download. ARRIVEchecklist.docx