Down-regulating lncRNA LINC00485  inhibits the development of osteosarcoma by regulating miR-361-5p/Twist1 axis


 Background: Mounting researches have established that long-chain non-coding RNA (lncRNA) and microRNA (miRNA) occupy an essential position in osteosarcoma development. The present study attempted to explicate the functional role and mechanism of LINC00485 and miR-361-5p in regulating the osteosarcoma progression.Methods: RT-PCR was employed to measure LINC00485 expression in osteosarcoma tissues and adjacent healthy tissues. Moreover, the correlation between LINC00485 expression and the clinicopathological indexes was analyzed. Further, the LINC00485 low-expressed cell model was constructed, and the CCK8 assay was employed to measure cell proliferation, followed by the flow cytometry in apoptosis testing. Besides, Transwell assay was utilized to detect cell migration and invasion ability, and Western blot examined the alternations of epithelial mesenchymal transition (EMT) marker molecules (E-cadherin and Vimentin). Furthermore, bioinformatics analysis was taken advantage of predicting the downstream molecular targets of LINC00485. RT-PCR and Western blot were carried out respectively to estimate the expression variations of miR-361-5p and Twist1. Dual luciferase activity assay and RNA co-precipitation (RIP) assay were conducted to verify the targeting relationship between LINC00485 and miR-361-5p, miR-361-5p and Twist1.Results: In comparison with normal adjacent tissues, the LINC00485 expression in cancer tissues was profoundly up-regulated. LINC00485 low expression greatly attenuated proliferation, migration, invasion, and EMT in osteosarcoma cells, while enhancing apoptosis. Bioinformatics analysis claimed that miR-361-5p was a sponge miRNA for LINC00485, targeting the Twist1 expression. Further gain- and loss-experiments declared that miR-361-5p limited the osteosarcoma development by inhibiting osteosarcoma cell proliferation, metastasis and promoting apoptosis.Conclusions: LINC00485 acted as an oncogene in the osteosarcoma tumorigenesis by modulating the miR-361-5p/Twist1 axis.


Introduction
Osteosarcoma, the most prevailing cancer that starts in the bones, mainly affects children and young adults under the age of 25 [1]. High-grade osteosarcomas are the most common form and have the greatest tendency to metastasize, most often the lungs. Even with clinical improvements, patients with metastatic or recurrent tumors do not respond well to conventional therapies [2]. Moreover, although chemotherapy, combined with surgical tumor excision, partly ameliorates the clinical treatment of osteosarcoma, the poor response to chemotherapy remains a major hurdle in current antitumor research [3]. In this context, it is imperative to further make clear the molecular mechanism behind osteosarcoma, so as to identify more comprehensive and targeted therapies for osteosarcoma treatment.
Non-coding RNAs (ncRNAs) are transcribed from the mammalian genome but unable to encode proteins.
A majority of the human genome constitutes non-coding elements. These non-coding RNAs may affect gene expression and tumor progression, signi cantly contributing to cellular processes regulation [4].
Long non-coding RNA (lncRNA) is a kind of non-coding RNA that can be transcribed and processed like mRNAs, but does not encode proteins [5]. LncRNAs are underscored in a good number of biological processes, like cell cycle progression, apoptosis, development, stem cell pluripotency and canceration, etc. [7][8][9][10]. Similarly, lncRNAs also regulate the osteosarcoma progression from different dimensions. For example, FOXC2-AS1 level is elevated in osteosarcoma cell lines and tissues, promoting cell resistance to doxorubicin in vivo and in vivo, a phenomenon closely associated with the poor prognosis [11]; LncRNA SOX2-OT, functioning as a new biomarker or independent prognostic indicator for osteosarcoma, can regulate the proliferation and mobility of cells through regulating SOX2 [12]. Previous studies have substantiated LINC00485' key role in human lung adenocarcinoma carcinogenesis [13], but overlooked the molecular mechanism and biological functions of LINC00485 in osteosarcoma.
MiRNAs are non-coding RNA molecules measuring 18-22 nucleotides long, which function in posttranscriptional regulation through base-pairing with complementary sequences to promote or inhibit mRNA degradation and translation. MiRNAs are highly conserved throughout evolution [14,15]. Many papers have reported that miRNAs abnormally expressed are closely entwined with tumorigenesis and progression, and are highly involved in cell development, differentiation, and cycle regulation [16,17].
Twist family bHLH transcription factor 1 (Twist1) is a basic transcription factor encoded by the gene located at 7p21.1. During skull development, Twist1 forms both homodimers and heterodimers binding to DNA E box sequences and regulate the transcription of genes implicated in cranial suture closure [21].
Twist1 also regulates neural tube closure, limb development, together with brown fat metabolism [22 -24]. Moreover, Twist1 has been proved its essential role in carcinogenesis malignant tumors, including sarcoma and hematological malignancies, tumor stem cells, angiogenesis, invasion, metastasis and chemotherapy resistance [25][26][27]. In addition, Twist1 has been recognized as one of the main regulators of the epithelial-mesenchymal transition (EMT) process. For example, Twist1 increases vimentin expression via Cul2 circular RNA, thereby promoting EMT of hepatocellular carcinoma (HCC) [28].
However, Twist1's regulation of the EMT pathway in osteosarcoma and its in-depth mechanism still awaits further validation.
In this research, LINC00485 was reported to be up-regulated in osteosarcoma samples. LINC00485 overexpression substantially enhanced osteosarcoma cell proliferation, migration, and invasion and promoted the EMT process. Further experiments clari ed that LINC00485 was a driver as the disease progresses by regulating the miR-361-5p/Twist1 axis. Taken together, this research illustrated that the LINC00485/miR-361-5p/Twist signaling pathway was a new mechanism associated with the osteosarcoma progression, and was expected to provide new molecular strati cation options with respect to osteosarcoma therapy.

Clinical samples
The study has garnered approval from the Research Ethics Committee of the People's Hospital of Longhua, A liated Hospital of Southern Medical University and strictly abides by the Declaration of Helshnki. Between January 2014 and June 2015, 40 osteosarcoma samples and the corresponding healthy tissues in total were harvested from People's Hospital of Longhua, A liated Hospital of Southern Medical University. Before the condition was diagnosed, the patients included in the study had not received anti-tumor treatment. All eligible patients had provided an informed consent in written forms.

Real-time polymerase chain reaction (RT-PCR)
To extract the total cellular, TRIzol (Invitrogen, Carlsbad, CA, USA) was applied. Then, the total RNA extracted was treated with deoxyribonuclease I for the genomic DNA removal. The reverse transcription reaction was operated according to guidance of the reverse transcription kit (Takara, Dalian, China). The The reaction tube was put into the MX3000P Real time PCR reactor and the conditions were: 94 °C for 40 s, 55 °C for 40 s, and 72 °C for 40 s through 45 cycles, a process monitored by uorescence signal. Here, GAPDH served as the endogenous control for LINC00485 and Twist1, U6 for miR-361-5p. The 2 -ΔΔCt value represents the gene relative expression, and the ct value represents the PCR cycle number that emitted via a uorescence signal that reached the set threshold during ampli cation.
CCK-8 assay CCK-8 assay was operated in measuring the cell viability. The Saos-2 and MG-63 cells were inoculated in 96-well plates (1×10 3 cells/well), followed by a 24 h incubation, during which stable cell growth was observed. Then, the incubation continued in set periods: 12 hours, 24 hours, 48 hurs, and 72 hours. Following the supplier's instructions, CCK8 reagent (Beyotime Biotechnology, Shanghai, China) was added to each well (10 μL/well). The spectrophotometer was used to measure the OD450 value (Bio-Rad, CA, USA) after one-hour cell incubation (37 °C).

Flow cytometry
When the stably transfected cells were produced, osteosarcoma-derived cell lines (Saos-2, and MG-63) were harvested. Washed three times by PBS, 4 °C PBS was used to suspend the cells, adjusting the cell density to 5×10 5 / mL. According to the instructions of the Annexin V-FITC apoptosis detection kit (Sigma-Aldrich; St. Louis, MO, USA), the apoptosis level was evaluated. Then, the cell suspension (190 μL) prepared in step with the addition of Annexin V: FITC (5 μL). When mixed fully, the mixture was added with 5 μL of propidium iodide (PI) solution. Next, the sample was incubated shielded from light (20 minutes, room temperature), and then the ow cytometry analyzer was employed (BD Biosciences, Franklin Lakes, NJ, USA) to examine the apoptosis level.

Western Blot
Washed 3 times with cold PBS, the cells collected were then added with RIPA lysis solution (100-200 μL) (Beyotime Biotcchnology, Shanghai, China) to lyse the cells in ice water. The Bradford protein assay was performed to detect the concentration of total protein. Proteins at a concentration (equal amounts) were prepared from each group and were placed in the separation gels (10%) for the SDS-PAGE electrophoresis. Next, the proteins on the gels was transferred to PVDF membranes (Millipore, Bedford, MA, USA). Further, the membranes were blocked at 4 °C for 1 h, intervened with the following primary antibodies: E-cadherin Rabbit Polyclonal (Proteintech, Cat.No. 20874-1-AP, 1: 1000), Vimentin Rabbit Polyclonal (Proteintech, Cat.No. 10366-1-AP ), and TWIST1-speci c Rabbit Polyclonal (Proteintech, Cat.No. 25465-1-AP), overnight at 4 °C. Following that, HRP Conjugated Goat Anti-Rabbit IgG (H + L) Secondary Antibody (BOSTER, 1: 2000, BA1055) for hybridization (room temperature, 1 h), succeeded by TBST washing (3 times, 5 min each). Then, equal volumes of ECL chemiluminescent substrates A and B were mixed up and added evenly on the PVDF membranes. Five minutes were granted for color development. When the X-ray lm had been exposed for 30 seconds to 5 minutes, the exposed lm was processed in the developing solution and the xing solution in order. Next, the lm was scanned or photographed, and the gel image processing system was used to measure the gray level of the protein bands. The ratio of the gray value of the target protein bands to that of β-actin, the endogenous control was employed to correct the total amount of protein added. The corrected protein expression of the control group was taken as 100%. The percentage of protein content in the control group was calculated. On this basis, the protein expression level of each group was compared and analyzed. All experiments were repeated 3 times.

Luciferase Reporter Assay
It was carried out with the Dual-luciferase Reporter Assay System (Promega, Madison, WI, USA). Target fragments of wild-type LINC00485 (LINC00485-WT), wild-type Twist1 (Twist1-WT) and mutant LINC00485 (LINC00485-MT), mutant Twist1 (Twist1-MT) were ampli ed and integrated into pmirGLO reporter vector (Promega , Madison, WI, USA), where LINC00485-WT and Twist1-WT contain sites that bind to miR-361-5p, and LINC00485-MT and Twist1-MT do not contain sites that bind to miR-361-5p. Saos-2 cells were co-transfected with miR-361-3p mimics or negative control, respectively. In line with the manufacturer's instructions, the luciferase activity was measured forty-eight hours after transfection. All experimental procedures were done in triplicate.

RNA immunoprecipitation assay
Using a Magna RIPTM RNA kit (Millipore, USA), RIP analysis was carried out. Cultured chondrocytes were suspended in RIP lysis buffer and were further incubated overnight in RIP buffer that contained human anti-Ago2 antibody beads (Millipore) (Input and normal IgG served as controls). Afterward, RNA extraction was achieved using the TRIzol reagent and the relative enrichment of LINC00485/Twist1 and miR-361-5p was tested via the RT-PCR analysis.

Transewell assay
After the Saos-2 and MG-63 cells were detached from each other with 0.25% trypsin, the centrifugation followed. Then the cells were resuspended in a single well of a 24-well culture plate. Matrigel Chambers (8 µm poreshze; Corning, Beijing, China) were put into use in the invasion assay alone, not applied to the migration assay. Transfected cells (5×10 4 ) were inoculated into the upper chamber, and then the Matrigel Matrigel was added. Next, the lower chamber was placed with culture medium supplemented with 10% FBS and lled with 400 μL of RPMI-1640. After incubation (24 h, 37 °C), the cells failed to migrate in the upper chamber were discarded. The cells were secured for 10 min using 4% paraformaldehyde and stained with 0.5% crystal violet. Following the running water washing, the number was counted under an inverted microscope. All experimental procedures were repeated in triplicate.
Xenograft assay in nude mice The measurements were represented as the mean±SD, and student' t or one-way ANOVA was used for statistical analysis; the count data was represented as frequencies. The Chi-square test was valid for analyzing the relationship between LINC00485 and the of patients' clinicopathological indicators. The relevance between the LINC00485 expression and the osteosarcoma prognosis was determined by the KM plotter assay. P<0.05 is statistically meaningful.

Results
LINC00485 was up-regulated in osteosarcoma tissues and cell lines RT-PCR was conducted here to quantify the INC00485 level, and higher expression of LINC00485 was reported in osteosarcoma samples ( Figure 1A). Additionally, compared to the osteoblast cell line (hFOB 1.19), the LINC00485 level in osteosarcoma cells (Saos-2, MG-63, U-2OS, SOSP-9607 and MNNG/HOS) was also increased ( Figure 1B). These results demonstrated that LINC00485 may exert a carcinogenic effect in osteosarcoma.
LINC00485 expression was related to osteosarcoma patients' worse survival Analyzing the cohort of osteosarcoma patients who provided with the prognosis information, LINC00485 expression was found related to the overall survival of osteosarcoma patients. Meanwhile, patients with low LINC00485 expression had better survival than those with high LINC00485 expression (P < 0.001, Figure 1C). Besides, the correlations of LINC00485 level and pathological indexes in osteosarcoma patients were also analyzed. The results showed that osteosarcoma patients with higher level of LINC00485 had higher Enneking stage (P =0.025), larger tumor size (P =0.0267) and earlier distant metastasis (P =0.038) ( Table 1). Therefore, these results manifested that LINC00485 was an unfavorable prognostic factor for patients with osteosarcoma. LINC00485 regulated osteosarcoma cell proliferation, apoptosis, migration and invasion Studying the biological function of LINC00485 in gliomas, a knockdown LINC00485 cell model in Saos-2 and MG-63 cells (Figure 2A) was constructed. Using CCK-8 assay to measure cell proliferation, we claimed that low LINC00485 level can inhibit the cell proliferation (Figure 2 B-C). Flow cytometry results suggested that knocking down LINC00485 remarkably promoted the level of apoptosis of Saos-2 and MG-63 cells ( Figure 2D). Subsequently, the Transwell assay was operated to determine the abilities of cells in terms of migration and invasion, con rming that knocking down LINC00485 signi cantly inhibited the migration and invasion of Saos-2 and MG-63 cells (Figure 2 E, F). In addition, Western blot was also utilized to estimate the EMT of osteosarcoma cells. It was found that knocking down LINC00485 promoted the level of the epithelial cell markers E-cadherin, while inhibiting that of the interstitial cell marker Vimentin (Figure 2 G-H). Moreover, the in vivo experiments revealed that downregulation of LINC00485 prominently inhibited the growth of Saos-2 and MG-63 cells ( Figure 3A-E). Taken together, those results clari ed that LINC00485 was an oncogene of osteosarcoma progression.

LINC00485 can target miR-361-5p
To further ascertain the downstream molecular mechanism of LINC00485, lncBase v.2 (http://carolina.imis.athena-innovation.gr/diana_tools/web/index.php?r=lncbasev2%2Findex), a bioinformatics database, was put into use to retrieve potential target genes of LINC00485, showing that LINC00485 contained potential binding sites for miR-361-5p ( Figure 4A). Next, the RIP assay was employed to verify the potential endogenous interaction between LINC00485 and miR-361-5p, elucidating that after miR-361-5p mimics were transfected, the amount of LINC00485 was enriched by the anti-Ago2 antibody ( Figure 4B). In addition, luciferase reporter assay illustrated that miR-361-5p mimics greatly inhibited the luciferase activity of LINC00485-WT, but had no effect on LINC00485-MUT ( Figure 4C). Next, the miR-361-5p level in transfected sh-LINC00485 cells was detected by RT-PCR, and the results reported that in comparison with the controls, the miR-361-5p expression in the LINC00485 knockdown group was remarkably increased ( Figure 4D). Moreover, RT-PC employed to measure the miR-361-5p level in osteosarcoma tissues and normal adjacent tissues showed that the level of miR-361-5p in osteosarcoma tissues was signi cantly reduced ( Figure 4E). Pearson analysis demonstrated that LINC00485 and miR-361-5p expression was negatively correlated in osteosarcoma tissues ( Figure 4F). To sum up, the data presented here supported that LINC00485 can bind to miR-361-5p in osteosarcoma cells directly, while inhibiting its expression.
Further, we tested the expression of EMT-related proteins. In contrast to the control group, miR-361-5p over-expression notably suppressed the EMT progression of osteosarcoma ( Figure 5 G-H). These ndings substantiated that miR-361-5p inhibited osteosarcoma development.

MiR-361-5p can target Twist1
On the basis of the mirPath (http://snf-515788.vm.okeanos.grnet.gr/index.php?r=mirpath), we analyzed the KEGG pathway related to miR-361-5p. The relationship between miR-361-5p and Adherens junction pathway is reported to be the most signi cant. (Figure 6A). In exploring the downstream targets of miR-361-5p, the bioinformatics databases, micorT, miRanda, Targetscan, and PicTar, were combined in use to nd potential target genes of miR-361-5p. Building on the Wayne diagram analysis, we reported that miR-361-5p contained the potential binding sites of Twist1 (Figure 6 B-C). Moreover, RIP analysis was utilized to verify the potential endogenous interaction between miR-361-5p and Twist1, demonstrating that miR-361-5p and Twist1 were preferentially enriched in the anti-Ago2 group in contrast to the IgG control group ( Figure 6D). Besides, luciferase reporter assay elaborated that miR-361-5p dramatically inhibited Twist1-WT's luciferase activity, but exerted no effect on Twist1-MT ( Figure 6E). Next, after miR-361-5p mimics were transfected, the expression of Twist1 in osteosarcoma cells was measured via RT-PCR and Western blot. These ndings proved that the Twist1 level in the miR-361-5p mimic group was substantially lower than the control group ( Figure 6 F-G). Further, RT-PCR was taken to estimate Twist1 expression level in osteosarcoma tissues, suggesting that the Twist1 expression in osteosarcoma tissues was profoundly up-regulated compared with normal tissues ( Figure 6H). Pearson correlation analysis exhibited that Twist1 was negatively correlated with miR-361-5p in osteosarcoma tissues ( Figure 6I). Taken together, these data demonstrated that miR-361-5p can directly bind to Twist1 in osteosarcoma cells, while inhibiting its expression.
Effect of LINC00485 / miR-361-5p axis on osteosarcoma cell proliferation, apoptosis, migration, invasion and EMT process To further investigate what effects LINC00485 and miR-361-5p exerted on proliferation, migration, EMT and invasion of osteosarcoma cells, we divided MG-63 cells into three groups: sh-NC, sh-LINC00485, and sh-LINC00485+miR-361-5p inhibitor. Subsequently, the expression of LINC00485, miR-361-5p and Twist1 were measured respectively. The results clari ed that after knocking down LINC00485, the miR-361-5p level was elevated, while the LINC00485 and Twist1 expressions were down-regulated (Figure 7 A-C).

Discussion
Osteosarcoma is the most prevailing cancer that arises in bones. Through surgical resection and multidrug chemotherapy, 70% of patients with high-grade osteosarcoma and localized limb tumors have sustained long-term survival. However, treatments limited to chemotherapy associated with resistance development do not re ect well on patients who defy surgical resection, carrying with primary migration, or diseases with high recurrence [29]. Accumulating researches have shown that lncRNA is an essential cell regulator in tumor progression, including osteosarcoma, which brings new ideas for the treatment of osteosarcoma [30][31][32]. This study rst explored a new lncRNA, namely LINC00485. Our ndings suggested that LINC00485 promoted osteosarcoma carcinogenesis and progression, and regulated osteosarcoma cell proliferation, migration, and invasion by targeting the miR-361-5p/Twist1 axis.
As the latest generation of gene sequencing technology develops and bioinformatics advances, more and more non-coding RNAs (ncRNAs) are considered to be related to various human tumor diseases [33]. For example, NCK1-AS1 participates in the occurrence and development of tumors by affecting the level of related proteins. In cervical cancer, NCK1-AS1 serves as a molecular sponge for miR-134 which targets matrix metalloproteinase (MMP) -2 and MMP-9, thereby promoting the proliferation, migration, and invasion of bone cervical cancer cells [34]. Previous studies have shown that LINC00485 is up-regulated in lung adenocarcinoma (LAC) cells. Moreover, inhibiting LINC00485 attenuates the expression of various oncogenes, CHEK1, Bcl-2, VEGF and HIF-1α, in LAC cells, and also enhances the sensitivity of LAC cells to cisplatin [13]. Nevertheless, LINC00485's role in osteosarcoma has yet been fully elucidated. Here, we found a signi cant upregulation of LINC00485 in osteosarcoma samples compared to normal controls. In addition, by analyzing the biological function of LINC00485 in osteosarcoma through loss-of functions, we reported that LINC00485 knockdown effectively suppressed cell proliferation and migration in osteosarcoma, and enhanced cell apoptosis. All these results proved the carcinogenic effect that LINC00485 exerted on osteosarcoma.
In recent years, multiple lines of evidence have clari ed that various miRNAs are implicated in regulating the osteosarcoma occurrence and development, and are abnormally expressed in cancer, functioning as oncogenes to exert effects [35]. For example, Circ_0001658 propels the proliferation and metastasis of osteosarcoma cells in way of regulating the miR-382-5p / YB-1 axis [36]. Knocking down miR-9 in human osteosarcoma cell lines inhibits their metastatic potential [37]. Taking miR-361-5p as an example, it is known as a tumor-suppressor miRNA and suppresses glycolysis, proliferation, and invasion of breast cancer by targeting FGFR1 and MMP-1 [38]. Moreover, it restrains the migration and invasion of glioma by targeting SND1 [39]. Consistent with the reports mentioned above, our research clari ed that the miR-361-5p level in osteosarcoma samples was notably down-regulated in comparison with adjacent healthy tissues and cells. Further, miR-361-5p over-expression suppressed the proliferation and metastasis of osteosarcoma cells, while knocking down miR-361-5p reversed the effects. The statistics con rmed that miR-361-5p suppressed osteosarcoma progression.
Additionally, lncRNA acts as a sponge for miRNA, reducing its regulatory effects on mRNA. For example, SNHG12 is up-regulated in colorectal cancer (CRC) and promotes CRC cell proliferation and invasion as a sponge for miR-16 [40]. Notably, lncRNA-miRNA interaction was essential in the progression of osteosarcoma. For example, the highly expressed lncRNA DLEU2 in osteosarcoma cells suppresses the osteosarcoma proliferation and metastasis by sponging miR-337-3p [41]. Furthermore, lncRNA MSC-AS1 was reported to be up-regulated in osteosarcoma tissues that was associated with the poor prognosis reported by osteosarcoma patients. At the same time, lncRNA MSC-AS1 inhibition reduces the malignant phenotype of osteosarcoma cells and increases their sensitivity to cisplatin by up-regulating miR-142 [42]. Thus, exploring the interaction of lncRNA-miRNA in the progression of osteosarcoma is of great signi cance. In light of the relevance between the expression and functions of LINC00485 and miR-361-5p in osteosarcoma, we speculated that LINC00485 may be a ceRNA in osteosarcoma. Using the bioinformatics database, we identi ed that miRNAs may interact with LINC00485, namely miR-361-5p. And further studies showed that the miR-361-5p expression increased profoundly after the LINC00485 knockdown. Moreover, the luciferase activity assay con rmed that LINC00485 directly bound to miR-361-5p. Therefore, we speculated that LINC00485, after up-regulated in osteosarcoma, further inhibits the miR-361-5p expression, a tumor suppressor molecule, thereby exerting a carcinogenic effect.
EMT refers to a physiological process where epithelial cells acquire the motile and invasive properties of mesenchymal cells by losing their cell polarity and cell-cell adhesion [43]. The process where epitheliallike cells are converted into mesenchymal-like cells is characteristic of the weakened expression of epithelial marker E-cadherin and the increased expression of mesenchymal markers, Vimentin and Ncadherin [44]. As a tumor prone to spread from the initially affected bone to more sites, it is notable that osteosarcoma is also accompanied by signi cant EMT changes. Inhibiting the EMT of osteosarcoma cells can not only suppress the development of metastasis, but also decrease the resistance to osteosarcoma chemotherapy [45][46]. Twist1, belonging to the bHLH (basic helix loop helix) transcription factor family, has been con rmed by mounting studies that it can regulate the EMT of tumor cells. For example, it enhances tumor metastasis by inhibiting the E-cadherin transcription [47]. Interestingly, recent studies have described that non-coding RNA affects tumor EMT by regulating Twist1. In osteosarcoma, for instance, LncRNA AFAP1-AS1 propels tumor progression and EMT by regulating RhoC/ROCK1/p38MAPK / Twist1 Signaling Pathway [48]. MicroRNA-300, by reducing Twist1, suppresses cell viability, inhibits cell migration, and promotes osteosarcoma cell apoptosis [49]. In the present research, we explained that Twist1 was remarkably up-regulated in osteosarcoma samples. Moreover, based on bioinformatics analysis and expression regulation experiments, we determined that miR-361-5p can interfere in the 3'UTR end of Twist1 and negatively regulated its expression. This result led us to further verify whether LINC00485 can indirectly regulate Twist1 expression through miR-361-5p to exert its carcinogenic effect. These ndings exhibited that LINC00485 knockdown reduces the level of mRNA and protein expression levels of Twist1. The experimental data revealed the biological functions of LINC00485 which partly regulated the Twist1 expression through miR-361-5p.

Conclusion
In summary, this study demonstrated that LINC00485 was up-regulated in OS and promoted OS cell proliferation and invasion as a sponge for miR-361-5p. Therefore, the new LINC00485-miR-361-5p-Twist1 axis might be considered as a promising therapy in metastatic OS patients. In future research, more attention should be focused on the detailed understanding of this network at the animal level or in more cases outside the scope of this report.

Declarations
Ethics approval and consent to participate Our study was approved by the Ethics Review Board of The People's Hospital of Longhua, Shenzhen, Southern Medical University.

Consent for publication
Not applicable.

Availability of data and materials
The data sets used and analyzed during the current study are available from the corresponding author on reasonable request.