MicroRNA-744-5p Suppresses Tumorigenesis and Metastasis of Osteosarcoma Through the MAPK/ERK Signaling Pathway by Targeting TGFB1

Background: Osteosarcoma (OS) is the most common malignant bone tumor in children and adolescents, and accumulating evidence has revealed that microRNAs (miRNAs) exert a crucial part in the progression of OS. Methods: GSE65071 from the GEO database was analyzed and miR-744-5p was found to be the lowest expressed miRNA. Real-time quantitative PCR (qRT-PCR), Western blotting (WB), colony formation assay, 5-Ethynyl-2-Deoxyuridine (EdU) incorporation assay and Transwell migration and invasion assay were performed to examine the effects of miR-744-5p in vitro, Luciferase-reporter assay was performed to detect the interactions between miR-744-5p and its specic target gene. Subcutaneous tumor-forming animal models and tail vein injection lung metastatic models were conducted in animal experiments to detect the effects of miR-744-5p in vivo. Results: miR-744-5p expression was down-regulated in OS cells and tissues. Higher expression of miR-744-5p was related with better clinical prognosis and lower malignancy degree of OS, including cell proliferation, migration and invasion in vitro and vivo. Transforming growth factor-β1 (TGFB1) was negatively regulated by miR-744-5p and could reverse the effects of miR-744-5p on OS proliferation, migration and invasion. The MAPK/ERK signaling pathway was involved in the miR-744-5p/TGFB1 axis. Conclusions: In general, this study suggests that miR-744-5p is a negative regulator of TGFB1, and suppresses OS progression and metastasis via MAPK/ERK signaling pathway.


Background
Osteosarcoma (OS) is the most common malignant bone tumor in children and adolescents, originating from mesenchymal cells [1,2]. With a poor prognosis, the mortality could be over 90% before polychemotherapy was introduced to clinical practice [3]. Signi cant progress has been made in treatment of OS in the past 30 years, effective therapies like neoadjuvant chemotherapy combined with surgical resection has been introduced into clinical treament, and the prognosis and quality of life has visibly improved compared to decades before [4][5][6]. Nevertheless, the 5-year survival rate of OS patients is still less than 50% [7]. Lung metastasis is the main problem for OS therapy, and the 5-year survival rate is < 30% with metastatic OS [8]. Therefore, more potent therapeutic strategies and approaches for OS are urgently needed.
Epithelial-to-mesenchymal transition (EMT) is a complex process through which epithelial cells obtain features of mesenchymal cells and lose original polarity. This reversible phenotypic change was supposed to be a stimulative element of tumor migration and invasion [9]. Accumulating evidence have revealed that EMT is closely related to tumor occurrence and development [10,11]. Therefore, suppressing the progression of EMT may be a potentially crucial approach to OS treatment. MicroRNAs (miRNAs) are a family of endogenous small non-coding RNAs, regulating the expression of target gene by combining with the 3′-untranslated regions (3'-UTRs) [12]. Accumulating evidence suggests that miRNAs play a crucial role in the occurrence and development of various tumors by regulating multiple signaling pathways [13,14]. Moreover, partial miRNAs participate in the course of EMT and act on the regulation of tumors [15,16]. miR-744-5p has been reported to be up-regulated in several tumors and was closely related with the clinical characteristics [17][18][19][20]. However, the functions of miR-744-5p in OS has not been reported and the mechanism of action of miR-744-5p still needs further study.
Mitogen-activated protein kinases/extracellular signal-regulated kinase (MAPK/ERK) signaling pathway is reported to be related with different stages of tumor growth through diversi ed mechanisms [21,22].
The MAPK pathway plays a crucial part in partial miRNAs regulation of various tumors [23][24][25]. Suppressing the MAPK/ERK pathway is supposed to be connected with inhibiting the viability and migration, and promoting the senescence and apoptosis of tumor cells.
Transforming growth factor-β (TGF-β), the prototype of the TGF-β family, with a bifunctional of regulating cell proliferation, has been reported to have a promoting regulation effect on EMT, and su cient evidence has demonstrated that there's conspicuous increased level of TGF-ß in tumor cells [26][27][28]. Moreover, it has also been reported that MAPK/ERK was associated with TGFB1 [29,30]. Nevertheless, the speci c effects of TGFB1 in OS have not been clearly illuminated yet, and the detailed function of TGFB1/MAPK/ERK axis remains investigation.
In this study, we found that miR-744-5p was remarkably down-regulated in osteosarcoma, and it suppresses the proliferation, migration and invasion of osteosarcoma cells by negatively regulating TGFB1 and MAPK/ERK signaling pathways. These ndings may provide a new therapeutic strategy for osteosarcoma.

Tissue samples
This study was approved by the ethics committee of the Second A liated Hospital of Southern Medical University. All human osteosarcoma and para-carcinoma samples were obtained from a total of 25 patients undergoing biopsies before receiving chemotherapy and radiotherapy at the department of joint and orthopedics. Tissue samples obtained from biopsy were collected and instantly frozen in liquid nitrogen. The pathological and personal clinical information is listed in Table 1.
Obtainment and analysis of original data GES65071 from the GEO database was downloaded. The R package affy was used in background correction and normalization processing. The R package limma was used to detect the difference of the miRNA expression level between normal samples and OS samples. The lter criterion: Log [fold change (FC)] > 1 and adj. P value < 0.05. All human OS cells, including Saos-2, U-2 OS, MG-63, MNNG, 143B and normal osteoblast cell line hFOB  1.19 were obtained from the American Type Culture Collection (ATCC, Manassas, US). OS cells were cultured in Dulbecco's modi ed Eagle's medium (DMEM, Invitrogen, US) replenished with 10% fetal bovine serum (FBS) (Gibco, NY) and 1% penicillin/streptomycin (PS, Gibco, CA). The hFOB 1.19 cells were cultured in Dulbecco's modi ed Eagle's medium/Nutrient Mixture F-12 (DMEM/F12) (Life Technologies, NY) with 0.3mgl/mL of G418, 10% FBS and 1% PS. All cells were incubated with 5% CO2 at 37℃.

Establishment of transfected cells
Plasmids overexpressed miR-744-5p and TGFB1 were used in the vitro experiments. The cells were cultured in 6-well plates. After washing with DMEM, the complex liquid of transfection was added into the plates and incubated for 24 h. Then the cells were cultured with DMEM containing 10% FBS for 48 h.
G418 selective media was used to screen out the transfected cells. The lentiviral transfection was conducted in vivo experiments. The cells were cultured in 24-well plates for 24 h. The medium with 2 µg/ml polybrene was used to replace the original medium, and lentivirus transfected with miR-744-5p or TGFB1 were added into the wells. After incubating for 24 h, the cells were cultured with DMEM for another 72 h. The transfection e ciency was examined via qRT-PCR .
Quantitative real-time PCR (qRT-PCR) Total RNAs from cells and tissue samples in Trizol (Invitrogen, US) were extracted from the frozen pulverized samples according to the manufacturer's protocol. 500 ng of total RNA was reverse transcribed into cDNA. The cDNA was diluted ve times with enzyme-free water. One-step qRT-PCR was performed in a 10 µL reaction system. The purity and integrity of the total RNAs were examined through the absordance at 260nm and 280 nm with. The primers of TGFB1, U6 and GAPDH were purchased from TsingKe (Beijing, China) and the primers of miR-744-5p were designed personally based on the purchased primers. Reverse transcription (RT) was performed with SuperScriptTM Preampli cation System for First volts and then transferred to the PVDF membranes. The proteins were incubated with speci c primary antibodies at 4℃ overnight. After washing with TBST, the proteins were incubated with the secondary antibodies at indoor temperature for 2 h. Rat anti-TGFB1 (1:1000, Abcam), GAPDH (1: 10,000, Proteintech), N-cadherin (1:1000, Abcam), E-cadherin (1:1000, Abcam), Vimentin (1:1000, Abcam), p-P38 (1:1000, Abcam), t-P38 (1:1000, Abcam), antibodies were used to detect the proteins. Reacting bands were acquired with ECL reagent and the quantity analysis was performed with ImageJ normalized to GAPDH.

Colony formation assay
About 800 OS cells were seeded into the six-well plate and cultured with DMEM and 10% FBS at 37℃ for 1 week. When the colonies turned invisible they were washed with PBS and xed with 4% paraformaldehyde, then satined with 0.1% crystal violet. The images were captured with a scanner and the counts were calculated manually.

5-Ethynyl-2-Deoxyuridine (EdU) Incorporation assay
The EdU incorporation assays were performed in accordance with the manufacturer's protocol. 1 * 104 cells/well of OS cells were seeded into 96-well plate and cultured with 100 µl of 50 µM EdU medium for 2 h. Then the cells were xed with 4% paraformaldehyde and destained with 2mg/mL glycine. Next, Apollo staining was performed with 1X Apollo dyeing reaction uid. 1X Hoechst 33342 reaction mixture was used in DNA staining. At least 50 cells per well were selected randomly. The intensity was measured from ve random elds and the photos were taken with uorescent microscope (Carl Zeiss, Germany).

Transwell Migration and invasion assay
Transwell migration assay was performed to detect the cell migratory ability. Total of 4.0×104 cells were seeded in the upper chamber with 200 µL of DMEM, while the lower chamber was immersed with 600 µL of DMEM with 10% FBS. After incubating for 24 h, the lower chamber was moved away and cells were xed with 4% paraformaldehyde for 30 min. Then stained the cells with 0.1% crystal violet for 20 min, unmigratory cells on the upper chamber were wiped with swab. After removing the crystal violet, ve randomized elds were observed and photographed with a microscope. As for the Transwell invasion assay, Matrigel (BD 5mg/ml) was diluted to 1mg/ml with serum-free medium. 100 µl of resulting Matrigel was put on the upper chamber and incubated at 37℃ for 1 h. The following steps were the same with the Transwell migration assay.

Immunohistochemistry (IHC)
The slides were immersed in the miscible liquids of potassium dichromate and concentrated sulfuric acid, then ushed for 1 h. Polylysine was smeared on the surface. Tissues were conducted through para n embedding. Tissue sections were dewaxed with xylene and ethanol. Then the sections were immersed in 0.01 mol/L sodium citrate buffer for 10 min and 3% hydrogen peroxide for 30 min at indoor temperature. Next, the sections were put into phosphate-buffered saline (PBS) for 5 min, and sealed with 5% bovine serum for 0.5 h at 37℃. The tissue sections were incubated with primary antibodies overnight at 4℃. After washing three times with PBS, the sections were incubated with secondary antibodies for 0.5 h at 37℃ and then incubated with SABC for another 0.5 h. Color developing agents were added after wiping up the sections and then hematoxylin staining was performed. Finally, the sections were dehydrated with ethanol and xelene, and sealed with resinene. The main antiboies were Ki-67, E-cadherin, N-cadherin and Vimentin (Abcam, UK). Photos were captured with an orthophoto microscope.

Hematoxylin-eosin (HE) staining
The tissues were immersed in stationary liquid containing 10% methanal. After dehydrating with ethanol and xylene, the samples were embedded in the para n. The sections were dewaxed with xylene and ethanol before staining. The sections were successively immersed in Hematoxylin, hydrochioric acid, ammonium hydroxide and ushed with distilled water for 1 h. Next, the sections were dehydrated in the ethanol and dipped in eosin staining solution. Finally the sections were immersed in ethanol and xylene, and sealed with gums.

Animal experiments
Nude mice in the study were purchased from Animal Core Facility of Southern Medical University and were randomly divided into 5 groups, 5 in each group. OS cells with uorescent protein RFP were inoculated into subcutaneous tissue of the nude mice. Pulmonary metastasis models were conducted through tail vein injection. The volume and size of tumors were recorded every 3 days and the tumors were seperated and imaged on day 28th. The mice were sacri ced at the end of experiments.

Statistical analysis
All experiments were repeated at least 3 times and data were demonstrated as means ± standard deviation. Independent Students' t-test and One-way ANOVA were used to compare the difference between two groups in clinical characteristics. Paired t-test was used to evaluate the differences in the miRNA expression between TGFB1 and miR-744-5p in tissue samples. Pearson's chi-squared test was performed to detect the relation between miR-744-5p and TGFB1. Log-rank test was conducted to evaluate the prognosis and overall survival of OS patients. Statistical analyses were performed with SPSS, v. 23.0. p < 0.05 was considered statistically signi cant. Data are presented as the means ± SD. * p < 0.05, ** p < 0.01, *** p < 0.001 3. Results

miR-744-5p was down-regulated in OS cells and tissues.
In order to investigate how miRNAs expresses in OS cells and tissue samples, GSE65071 from the GEO database was analyzed with R package. The volcano plot demonstrated the differences of miRNAs expression between OS and normal tissues (Fig. 1a). According to the miRNA expression levels, the upand down-regulated miRNAs were displayed with a cluster heat map (Fig. 1b). Totally, 89 miRNAs were down-regulated in OS tissues compared to normal tissues (fold change > 1, FDR < 0.05), and the top ten down-regulated miRNAs were shown in Fig. 1c. We found that miR-744-5p was the most down-regulated miRNA among these miRNAs. Based on above results, we attempted to gure out the speci c function of miR-744-5p in OS. qRT-PCR was performed to assess miR-744-5p expression in 25 paired OS and adjacent normal tissues, and results gured that miR-744-5p was signi cantly down-regulated in OS tissues (Fig. 1d). Furthermore, low expression of miR-744-5p was found in OS cells including 143B, MNNG, U-2 OS, MG-63 and Saos-2 compared with in hFOB 1.19 cell lines (Fig. 1e).
3.2 miR-744-5p was closely related to better clinical characteristics of OS patients.
As shown in (Fig. 2a-d), expression level was signi cantly associated with OS clinical characteristics, higher expression of miR-744-5p was found in patients with earlier pathological stage, smaller tumor size, localized growth and higher long-time survival rate. Detailed clinical data was demonstrated in Table 1. 3.3 Up-regulation of miR-744-5p inhibited OS cell proliferation, invasion and migration in vitro.
MG-63 and 143B cell lines were used in vitro for further experiments in this study. qRT-PCR was performed to evaluate the transfection e ciency of miR-744-5p mimics (Fig. 3a). Colony formation and EdU assays were conducted to detect the effect of miR-744-5p on cell proliferation. Figure 3b-e demonstrated that over-expressed miR-744-5p signi cantly inhibited the proliferation of the OS cells. Transwell migration assays were performed to assess the in uence of miR-744-5p on the migratory ability of the OS cells in vitro and results showed that over-expressed miR-744-5p signi cantly decreased migration of OS cells. Similarly, the invasive ability of the OS cells were detected with Transwell invasion assays, and up-regulating miR-744-5p suppressed the invasive ability of both 143B and MG-63 cells ( Fig. 3f-i). Furthermore, WB analysis demonstrated that miR-744-5p decreased the expression of tumoregenesis-related proteins, N-cadherin, vimentin, and increased the level of E-cadherin in both MG-63 and 143B OS cells (Fig. 3j).
In order to investigate the effect of miR-744-5p in vivo, OS cells transfected with Lv-miR-NC, Lv-miR-744-5p were implanted subcutaneously in nude mice. Figure 4a showed the lung tissues obtained from the mice sacri ced in week 4. It was shown that tumors volume was smaller and the average weight was lighter in the Lv-miR-744-5p group compared with the NC group (Fig. 4b, c). IHC was performed to assess the expression of proliferation and invasion-related factors including Ki-67, E-cadherin, N-cadherin and vimentin. The results demonstrated that Ki-67, N-cadherin and vimentin were obviously down-regulated in the Lv-miR-744-5p group versus controls, indicating that miR-744-5p suppressed OS cell proliferation and invasion of OS cells in vivo. On the contrary, E-cadherin was observed with higher expression compared with Lv-miR-NC group (Fig. 4d). Less OS cells were observed in the over-expressed miR-744-5p group through HE-staining (Fig. 4e).
3.5 TGFB1 was up-regulated in OS tissues and was a target of miR-744-5p.
The main function of miRNAs was inhibiting the translation or promoting the degradation of the target genes. Thus we attempted to gure out the downstream mechanism of miR-744-5p in the occurrence and development of OS, miRDB database was checked and 111 genes in total were found to be a potential target of miR-744-5p. KEGG pathway enrichment analysis demonstrated that MAPK signaling pathway was the most relative center of the mentioned genes (Fig. 5a). Among all candidate genes in MAPK axis, we found TGFB1 was down-regulated with the overexpression of miR-744-5p in both 143B and MG-63 OS cells (Fig. 5b, c). Luciferase-reporter assay was performed to detect the speci c relationship between TGFB1 and miR-744-5p, results showed that miR-744-5p directly targeted TGFB1, and over-expressed miR-744-5p signi cantly suppressed the luciferase activity of OS cells (Fig. 5d, e). WB veri ed that miR-744-5p negatively regulated the expression of TGFB1, and MAPK-related proteins were down-regulated in the miR-744-5p mimics group (Fig. 5f). qRT-PCR was performed to investigate the expression of TGFB1 in OS cell lines and tissue samples, higher expression of TGFB1 was found in various OS cells, especially in MG-63 and 143B, and TGFB1 was signi cantly higher expressed in OS tissues compared with adjacent tissues (Fig. 5g, h). Furthermore, TGFB1 was negatively related with miR-744-5p with an r of -0.7326 in OS tissues (Fig. 5i).

TGFB1 was connected with poor clinical characteristics of OS patients.
qRT-PCR were conducted to detect the relevance between TGFB1 and clinical characteristics. Figure 6a-c demonstrated that higher expression level of TGFB1 was found in more patients with advanced, larger and metastatic tumor. Although there was no statistical difference in overall survival, patients with higher expression level of TGFB1 tend to have a poorer prognosis (Fig. 6d).

miR-744-5p suppressed OS proliferation, migration and invasion by regulating MAPK signaling pathway through TGFB1 in vitro.
In order to verify that miR-744-5p regulates proliferation, migration and invasion of OS cells through via TGFB1, a series of rescue experiments were TGFB1 was over-expressed arti cially, and qRT-PCR showed that TGFB1 was successfully transfected into the OS cells (Fig. 7a). Colony formation and EdU assays were performed, it was found that over-expressed miR-744-5p signi cantly suppressed the proliferation of OS cells, and the effects were restored with the up-regulation of TGFB1 (Fig. 7b-e). Transwell migration and invasion assays were performed, and results revealed that TGFB1 reversed the protective effects of miR-744-5p on OS migration and invasion (Fig. 7f-i). WB were performed to detect the downstream mechanism of miR-744-5p/TGFB1 axis regulating OS cells. Figure 7j, k showed that over-expressed miR-744-5p inhibited the expression of N-cadherin, vimentin, TGFB1,and t-P38, and promoted the expression of E-cadherin, indicated that EMT and MAPK pathways played essential rules in the process of miR-744-5p/TGFB1 axis regulating the OS promotion.
3.8 miR-744-5p suppressed OS growth and metastasis by regulating MAPK signaling pathway through TGFB1 in vivo.
Rescue assays were performed in animals to verify how miR-744-5p and TGFB1 worked in vivo. OS cells transfected with Lv-miR-NC, Lv-miR-744-5p and Lv-miR-744-5p with TGFB1 were respectively implanted subcutaneously in nude mice. Figure 8a-c demonstrated that over-expression of miR-744-5p suppressed the growth of tumor, and there was signi cant difference in weight and volume, however, the introduction of TGFB1 inhibited the effects of miR-744-5p, made tumors larger and heavier remarkably. Moreover, IHC showed that higher expression of TGFB1 increased the expression level of ki-67, N-cadherin and vimentin which were down-regulated by miR-744-5p (Fig. 8d). Figure 8e demonstrated that Lv-miR-744-5p inhibited the invasion of OS cells, while inversely over-expression of TGFB1 promoted the metastasis conspicuously.

Discussion
OS is the most frequent primary malignant tumor of bones which basically occurs in the adolescents [31]. However, though accumulating researches have been conducted on, poor prognosis of OS has not been soluted well. Therefore, it's necessary to develop novel potential targets on OS therapies. miRNAs have been reported to be directly related to the regulation of gene expression, and substantial evidence have revealed that abnormal miRNAs expression appears in numerous tumors [32][33][34]. Numerous studies have shown that various miRNAs have different functions in the process of OS developing [35][36][37][38][39]. Shaoxuan He [40] demonstrated that miR-217 inhibits the proliferation, migration and invasion of OS cells via directly targeting SIRT1. On the contrary, miR-652 is reported to promote tumorigenesis and metastasis through targeting RORA [41]. miR-744-5p has been reported to play a negatively regulatory role in some cancers like ovarian cancer and non-small cell lung cancer, however, few studies investigate the effects and mechanism of miR-744-5p in OS [17][18][19]. In accordance with the data from GEO database and results of qRT-PCR, we found that miR-744-5p was down-regulated in the OS cells and tissue samples, which was consistent with previous studies.
EMT participates in the migration and invasion of tumor cells, and promotes cancer progression and metastasis. During EMT process, epithelial cancer cells acquire characteristic of mesenchymal cells and lose the cell polarity and adhesion between cells. These features accelerate the migration and invasion of OS cells, promote tumor metastasis and increase drug resistance in OS therapies [42][43][44]. Furthermore, reduction of E-cadherin and induction of N-cadherin and vimentin levels are reported in the conversion from epithelial cell to mesenchymal cell [45,46]. In this study, a series of experiments were undertaken to investigate the function and molecular mechanism of miR-744-5p in OS. We found that miR-744-5p was lowly expressed in OS tissues and played a negative regulatory role in tumor developing. Results indicated that miR-744-5p downregulated the cell proliferation, migration and invasion of OS, and could be a novel target in the treatment of OS.Moreover, via Luciferase-reporter assay, we proved that there was a biding sequence between miR-744-5p and TGFB1, and TGFB1 was found to be suppressed by miR-744-5p in the study. TGFB1, a regulatory cytokine participating in multiple signaling pathways, has been reported to play dual roles in cell growth by regulating cell autophagy [47]. TGFB1 acts as a tumor suppressor in the early stage of tumor or normal tissues, and promotes tumorigenesis and metastasis in advanced tumors [26,28,48].
Recent studies have demonstrated that TGFB1 played a vital part in the progression in various tumors including ovarian cancer, colorectal cancer, cervical cancer, gastric cancer[26, 28, [49][50][51]. It was also found that TGFB1 induced EMT in the tumorigenesis and metastasis in cancer [52,53]. In this study, we found that TGFB1 was signi cantly upregulated in tumor tissues and was highly correlated to the development of OS, indicating that TGFB1 could play a stimulative role in tumorigenesis and metastasis in OS. Moreover, the relationship between clinical characteristics and TGFB1 was veri ed in the study, results demonstrated that a high expression level could result in a poor prognosis of OS patients, which was quite similar with other related studies [54,55].
Through bioinformatic analysis we have found that MAPK signal pathway was remarkably correlated with miR-744-5p. To detect the speci c mechanism of this axis in OS, we performed various experients. WB demonstrated that over-expression of miR-744-5p signi cantly suppressed the EMT-related and MAPK/ERK-related proteins in both 143B and MG-63 OS cells. Moreover, TGFB1 was then over expressed in the miR-744-5p overexpressed group, the results revealed a trend back to the results of the control group. Similar outcomes were found in the rescue assays in vivo. ERK belongs to the MAPK family, and MAPK/ERK signaling pathway is supposed to be a central section that regulates proliferation, differentiation and apoptosis in cells [56,57]. Accumulating evidence have shown that there's a close relationship between the MAPK/ERK pathway and tumor growth and metastasis, and enhanced ERK expression has been perceived in various tumors [58][59][60].
This study still remains shortcomings. We constructed the OS mice models through subcutaneous injection due to the limitation of experiment conditions, and orthotopic models were considered to be used to examine the results in the further study. Furthermore, function de ciency assay was expected to be performed to verify the effects of miR-744-5p and TGFB1. Besides, liquid biopsy has been noticed as a convenient and e cacious checkup method in recent years, and miRNAs have been detected in multiple body liquids[61-64]. miRNAs could become a potential therapeutic direction in the treatment of cancers and the expression pro le analysis of miRNAs in body liquids could be conducted to gure out whether miR-744-5p could be utilized in clinical treatment of OS patients. Except the insu ciency, this study revealed the relationship between miR-744-5p and OS for the rst time. OS patients with a high expression level of miR-744-5p was found to have better clinical characteristics and prognosis, indicating that miR-744-5p could be a latent target in prediction and assessment during treatment. Meanwhile, we detected the downstream mechanism of miR-744-5p and found that TGFB1 was a target gene of the miRNA and MAPK/ERK signaling pathway was involved in the process.

Conclusions
In conclusion, we found miR-744-5p was negatively related to the progression and metastasis of osteosarcoma via dowregulating TGFB1 through MAPK/ERK signaling pathway. We demonstrated that miR-744-5p suppressed the proliferation, migration and invasion of OS cells through MAPK/ERK signaling pathway by directly targeting TGFB1. Thus miR-744-5p/TGFB1 signal pathway could be a potential therapeutic target for OS and may provide further insight into the molecular mechanism of OS.

Declarations Ethics approval and consent to participate
The Ethics Committee of Zhujiang Hospital of Southern Medical University approved the experiments involving patients tissues in this study.
The Ethics Committee of Southern Medical University approved the animal experiments in this study.

Consent for publication
Authors involved in this paper all signed written conscent for publishing in your journal.

Availability of data and materials
The datasets generated and analysed during the current study are available in the [GEO DataSets] repository, [ https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE65071 ]

Competing interests
The authors declare that they have no competing interests.         could abolish the in uence. (j, k) Wester blotting assays showed that miR-744-5p down-regulated metastasis related, MAPK-related and TGFB1 proteins in OS cells, while TGFB1 own contrary functions.