Long noncoding RNA LINC00314 facilitates osteogenic differentiation of adipose-derived stem cells through the hsa-miR-129-5p/GRM5 axis via the Wnt signaling pathway

Many studies have shown that long noncoding RNAs (lncRNAs) are closely related to the stimulation of osteogenic differentiation of adipose-derived stem cells (ADSCs) and the prevention of osteoporosis. Current research aimed to investigate the novel lncRNA and explored the function and molecular mechanism of the LINC00314/miR-129-5p/GRM5 axis in regulating osteogenic differentiation of ADSCs. LncRNA and miRNA sequencing was performed in normal and osteogenic differentiation-induced ADSCs (osteogenic group). Abnormally expressed lncRNAs and miRNAs were obtained by the R software and the relative expression of LINC00314, miR-129-5p, and GRM5 during osteogenic induction was measured by RT-PCR. ADSCs were then transfected with pcDNA3.1-sh-LINC00314 and agomiR-129-5p. Alizarin red staining (ARS) and alkaline phosphatase (ALP) staining were performed to identify the mechanism of the LINC00314/miR-129-5p/GRM5 axis in regulating osteogenic differentiation of ADSCs. LINC00314 was significantly upregulated in the group of osteogenic-induced ADSCs. LINC00314 and GRM5 mimics increased the early and late markers of osteogenic differentiation, which manifest in not only the markedly increased ALP activity but also higher calcium deposition, while miR-129-5p mimic had the opposite effects. LINC00314 directly targeted miR-129-5p through luciferase reporter assay, and miR-129-5p suppressed GRM5 expression. Moreover, the LINC00314/miR-129-5p/GRM5 regulatory axis activated the Wnt/β-catenin signaling pathway. LINC00314 confers contributory function in the osteogenic differentiation of ADSCs and thus the LINC00314/miR-129-5p/GRM5 axis may be a novel mechanism for osteogenic-related disease.


Background
Osteoporosis is a serious degenerative disease characterized by the reduction of bone mass and destruction of the bone microstructure [1,2]. Osteoporosis can result in fractures and may increase economic cost and societal burden [3]. Previously, bone marrow mesenchymal stem cells (BMSCs) were administered as seed cells to improve BMD and used for tissue engineering [4]. However, the proliferation ability of BMSCs was limited, especially in older patients. Adipose-derived stem cells (ADSCs) could be a potential alternative for BMSCs due to their advantages of being easy to obtain and widely sourced [5,6].
Long noncoding RNAs (lncRNAs) belongs to a family of small RNAs, which are often 200 or more nucleotides long and play crucial roles in bone disease [7]. Zhang et al. [8] reported that lncRNA NKILA indirectly regulates RXFP1/AKT signaling pathway to regulate osteogenesis. Another study also found lncRNA HOTAIRM1 directly targeting with JNK/AP-1 signaling pathway to control the osteogenic differentiation [9]. Wu et al. [10] analyzed the differentially expressed long noncoding RNAs in the normal and induced groups and found that HIF1A-AS2 has a positive role in promoting ADSCs osteogenic differentiation through the PI3K/Akt signaling pathway. LncRNA LINC00314 is located on chromosome 21 with open reading frame 94. Currently, no study has explored the role and mechanism of LINC00314 in the regulation of ADSC osteogenic differentiation.
There is now overwhelming evidence that alterations in microRNA (miRNA) expression levels are linked to osteogenic differentiation [11]. Several miRNAs have been identified to have potential roles in promoting or inhibiting the osteogenic differentiation of ADSCs [12]. Li et al. [12] found that miR-154-5p could directly bind to the 3′-UTR of Wnt11 and thus inhibit the osteogenic differentiation. Liu et al. [11] revealed that miR-145-5p suppresses osteogenic differentiation of ADSCs by targeting semaphorin 3A.
MiR-129-5p participates in regulating the proliferation, migration, and invasion of retinoblastoma cells and affects spinal cord injury [13,14]. In a study on the regulation of osteogenic differentiation, Valenti et al. [15] found that physical exercise modulates miR-129-5p expression in progenitor cells and thus promotes osteogenesis. Further study also revealed that miR-129-5p could regulate osteoblast differentiation of BMSCs.
Metabotropic glutamate receptor 5, named GRM5, has been shown to activate phospholipase C and participate in multiple biological processes [16]. Many studies have shown the Wnt/β-catenin signaling pathway participant into the osteogenic differentiation of ADSCs. However, whether LINC00314 could activate the Wnt/β-catenin signaling pathway and stimulate osteogenic differentiation of ADSCs remains unclear.
In this study, LINC00314 was found to be abnormally expressed during the osteogenic differentiation of ADSCs through gene chip and further confirmed by qRT-PCR. Further bioinformatics analysis revealed LINC00314 directly regulating with miR-129-5p and its targeting gene GRM5.

Tissue specimens and ADSC cultures
Human adipose tissue was collected from patients who prepared for total knee arthroplasty. Each patient signed an informed consent form, and this research was approved by the ethics committee of the Second Hospital of Hebei Medical University. Collected adipose tissue was washed with PBS three times and cut into sections. Then, these sections were digested with 0.1% type I collagenase (Gibco, Grand Island, NY, USA) at 37°C for 1 h. Next, we added 5 mL DMEM containing 10% fetal bovine serum (FBS, Thermo Fisher Scientific, Inc., USA) to inhibit the type I collagenase activity. The cell suspension was centrifuged at 300×g to collect the cells. Cells were cultured in highglucose Dulbecco's modified Eagle medium (DMEM, Gibco, Grand Island, NY, USA) with 20% FBS and 1% penicillin at 5% CO 2 and 37°C. ADSCs at passage 3 were used for subsequent studies. A previous study revealed that less than 5% of the ADSCs showed senescence when expanded to generation 10 [17].
Trilineage differentiation was performed to show the differentiation potency of ADSCs. In brief, ADSCs were seeded into 6-well plates (5.0 × 10 5 cells/well) with normal medium until cell confluence reached approximately 70%. The nonadherent cells were removed by replacing the medium, and the attached cells were cultured until confluence. The cells were then grown for 21 days in the adipogenic, osteogenic, and chondrogenic medium (Cyagen, Guangzhou, China). Alizarin red S staining (ARS, Solarbio, Beijing, China) was performed to assess osteogenic differentiation. Adipogenic differentiation was visualized using Oil Red O (Sigma-Aldrich, St. Louis, MO, USA) staining. Alcian Blue (Sigma-Aldrich, St. Louis, MO, USA) staining was used to assess chondrogenic differentiation.

Microarray
Microarray analyses of lncRNA and microRNA expression were performed as described previously [18]. Briefly, total RNA from the normal and induced groups was extracted by TRIzol as described previously. cDNA was synthesized, labeled with fluorescent dye, and hybridized with a lncRNA Human Gene Expression Microarray v4.0 (4 × 180 K; Cloud-Seq Biotech, Shanghai, China) platform (LC Sciences, Houston, TX, USA) for lncRNA and microRNA, respectively. Differentially expressed lncRNAs and mRNAs were obtained by the limma package. Moreover, heatmaps and volcano plots were generated by using Bioconductor (http://www.bioiconductor.org) in R software (Free Software Foundation Inc., Boston, MA, USA).

Bioinformatic analysis
First, the limma package was utilized to identify differentially expressed lncRNAs and miRNAs in the induced and non-induced groups using the screening criteria |logFC (foldchange)| ≥ 1 and adjusted P < 0.05. Cluster enrichment of gene ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway was performed using the Database for Annotation, Visualization and Integrated Discovery (DAVID) database (https://david.ncifcrf.gov/).
The GO terms including three categories: biological process (BP), cellular component (CC), and molecular function (MF). Target genes network were constructed by the Search Tool for the Retrieval of Interacting Genes (STRING) database.

qRT-PCR
Trizol reagent (Takara Bio Inc., Otsu, Shiga, Japan) was applied for total RNA extraction from cells. Afterwards, mRNA levels were determined using the SYBR Green qPCR Mix Kit on the ABI 7500 Fast Real-Time PCR System (Thermo Fisher Scientific Inc., Waltham, MA, USA). All primers were purchased from Invitrogen (Shanghai, China). Glyceraldehyde-3phosphate dehydrogenase (GAPDH) and U6 were utilized for internal control as appropriate. Sequences of primers are presented in Table 1. The relative quantitative expression of interest genes was expressed as fold change (2− ΔΔCt method). ΔΔCt = [Ct (target gene) − Ct (GAPDH/U6)] experimental group − [Ct (target gene) − Ct (GAPDH/U6)] control group. Each sample was examined in triplicate.

Western blot analysis
The expression level of osteogenic differentiationrelated genes was determined by Western blot analysis. The equivalent cracking solution was added to phenylmethyl sulfonylfluoride until the concentration was 1 mM (No. ST506, Beyotime Biotechnology Co.,

ALP activity and staining
ADSCs were washed with precooled PBS three times and lysed in precooled 1% Triton X-100 on ice for 30 min. Cell lysate was subjected to ALP activity determination, and the value at 405 nm was normalized to that of total protein concentration. ADSCs were washed with PBS three times and fixed with 4% paraformaldehyde for 10 min and then added NBT-BCIP solution (BiYunTian Biotech Company, Shanghai, China) for incubation for another 15 min. Images were captured under a microscope (Olympus DP73 Microscope, Olympus, Tokyo, Japan).
In the microRNA array, a total of 105 microRNAs were identified as differentially expressed. Among these differentially expressed microRNAs, 50 were downregulated and 55 were upregulated (Fig. 2c, d). The top 100 differentially expressed miRNAs between OIM and control groups were listed in Supplementary Table S2. In particular, miR-129-5p was found significantly downregulated (logFC = − 1.93, P < 0.05).

Bioinformatic analysis
Analysis in DAVID was performed using the data profile of differentially expressed mRNAs identified by R software. GO terms that included BP, CC, and MF are listed in Fig. 3a, b, and c, respectively. The most significantly enriched biological processes, cellular component, and molecular function terms were transcription, DNA-templated, Z disc and translation repressor activity, and nucleic acid binding, respectively. In the KEGG pathway analysis, the Wnt/β-catenin signaling pathway was considered to be significantly enriched during osteogenic differentiation of ADSCs (Fig. 3d).
We then constructed a LINC00314-miRNA-target gene network using Cytoscape to visualize their interrelationships based on our miRNA-seq and target gene data (Fig. 4a). A Venn diagram was generated to assess the overlap between miRNAs targeting LINC00314, the miRNA profile, and miRNAs affecting Wnt/β-catenin signaling, and miR-129-5p was identified (Fig. 4b). A PPI network was constructed to reveal the target gene interaction (Fig. 4c).
The network was then analyzed using the MCODE plugin, and four clustering modules were identified according to the chosen screening conditions. Clustering module 1 scored 5.2 with 5 nodes and 9 edges (Fig. 4d), clustering module 2 scored 4.8 with 11 nodes and 24 edges (Fig. 4e), clustering module 3 scored 4.2 with 4 nodes and 6 edges (Fig. 4f), and clustering module 4 scored 3.5 with 9 nodes and 15 edges (Fig. 4g).

Effects of LINC00314 on osteogenic differentiation of ADSCs
Compared with the normal group, the induced group had significantly increased LINC00314 expression at 1, 2, and 3 weeks (Fig. 5a, P < 0.01). The expression of LINC00314 was significantly increased or decreased after transfection of LINC00314 or sh-LINC00314, respectively, and thus the successful plasmid construction was verified (Fig. 5b). Moreover, osteogenic markers (Osterix, RUNX2, and osteocalcin) were significantly upregulated or downregulated after transfection of LINC00314 or sh-LINC00314, respectively (Fig. 5c-f). A large number of calcified nodules were identified by ARS and ALP activity was increased by LINC00314 overexpression, whereas these effects were suppressed by LINC00314 silencing (P < 0.05, P < 0.01, Fig. 5g). These data indicated that overexpression LINC00314 promoted the osteogenic differentiation of ADSCs.

MiR-129-5p regulated ASC osteogenic differentiation through GRM5
PCR analysis showed that as the induction time increased, the relative GRM5 expression gradually increased (P < 0.05) in the induced group compared with the normal group (Fig. 7a). There was a clear significant negative correlation between the miR-129-5p and the GRM5 expression (R 2 = 0.871, P = 0.001, Supplementary Fig. 2).
Compared with the mimic group, the GRM5 mimic+-XAV939 (10 μM) group was associated with a reduction in p-β-catenin and Wnt5a expression. These findings suggested that GRM5 regulated the Wnt/β-catenin signaling pathway and thus promoted osteogenic differentiation of ADSCs.

Discussion
We firstly identified LINC00314 plays a critical role in the regulation of osteogenic differentiation of ADSCs. Gain and loss of LINC00314/miR-129-5p/GRM5 function studies revealed that LINC00314 could positively regulate the early and late osteogenesis of ADSCs. Furthermore, the LINC00314/miR-129-5p/GRM5 axis exerts its osteogenic effects on ADSCs through the Wnt/βcatenin signaling pathway.
Recently, ADSCs have gained considerable attention in tissue-engineered bone due to the fat tissue is a rich source of MSCs with minimal injury and less painful [20]. Previous studies have demonstrated that ADSCs can trigger differentiation into adipocytes, chondrocytes, and osteoblasts under specific optimal medium condition [21][22][23]. Therefore, the promotion of osteogenic differentiation of ADSCs played an important role.
In this study, microarray was firstly conducted to reveal the differentially expressed lncRNAs and identified LINC00314 as the most statistically significant lncRNA.
We predicted the LINC00314 binding site by using the bioinformatics databases (miRcode). The overlap between the predicted microRNAs and microRNA array results found that miR-129-5p was linked with both LINC00314 and GRM5. Previous studies showed that miR-129-5p function as oncogenes or tumor suppressors during the occurrence and development of multiple cancers, including retinoblastoma [13], non-small cell lung cancer [25], and hepatocellular carcinoma [26]. Its functional role in osteogenic differentiation of ADSCs is poorly demonstrated.
We found that overexpressing miR-129-5p could significantly downregulate ALP activity and calcium deposits. Thus, we concluded that LINC00314 is a sponge of miR-129-5p. Han et al. [27] found that miR-129-5p directly targeting ROCK1 and suppresses osteosarcoma progression, which is consistent with our research findings. Luciferase assays showed that miR-129-5p directly bind to the 3′UTR of GRM5. Finally, GRM5 could regulate the Wnt/β-catenin signaling pathway to induce the osteogenic differentiation of ADSCs. Wnt/β-catenin is a very complex signaling pathway and is involved in ADSC differentiation [28]. Xu et al. [29] found miR-889 directly targeting WNT7A and subsequently suppressing bone formation.
The Wnt/β-catenin signaling pathway is closely related to osteoblast differentiation of ADSCs. Anabolic agents that stimulate the Wnt/β-catenin signaling pathway can be used to treat osteoblast-related diseases such as osteoporosis. Recently, antisclerostin and anti-dickkopf antibodies were used to conduct the preliminary clinical trial and show great potential for the treatment of osteoporosis [30][31][32]. Western blotting and loss/gain-of-function assays of the Wnt pathway confirmed that upregulated expression of GRM5 induced the activity of the Wnt signaling pathway, which promoted the osteogenic differentiation of ADSCs.

Conclusions
In conclusion, we propose that the LINC00314/miR-129-5p/GRM5 axis positively regulates the early and late osteogenic differentiation ability by modulating the Wnt/β-catenin signaling pathway in ADSCs (Fig. 9). LINC00314 was upregulated during osteogenic differentiation of ADSCs, which enhanced its miR-129-5p sponge function, thereby promoting GRM5 expression by activating the Wnt/β-catenin signaling pathway. Our findings could provide a new target for controlling osteogenesis in ADSCs, which are crucial to bone tissue engineering and treatment for bone diseases.