FAM64A Promotes Osteosarcoma Cell Growth and Metastasis and Is Mediated by miR-493

Aberrant expression of FAM64A was correlated with cell proliferation in various cancer types. We examined the expression of FAM64A and the upstream gene miR-493 in OS. The functions of miR-493 were revealed through extensive experiments. We found an increase of FAM64A gene and protein in OS tissues. Overexpression of FAM64A resulted in promoting tumor proliferation, migration, and invasion. The miR-493 targeted and negatively regulated FAM64A. Our data showed that upregulation of FAM64A in OS correlated with poor prognosis.


Introduction
Osteosarcomas are originated from primitive mesenchymal cells and defined as the most prevalent primary solid tumor of the bone. Due to the heterogeneity of osteosarcoma, the etiology of OS in most patients is still obscure. Fletcher et al. determined increased incidence of OS in cases with altered tumor suppressor genes [1]. For example, CATS (FAM64A) is confirmed to be highly expressed in leukemia, lymphoma, and a range of tumor cell lines. Moreover, it has been reported that its protein levels have intense relationship with proliferation in both tumor cells and nonmalignant cells. Silencing of FAM64A resulted in decreased proliferation and cell cycle progression of hematopoietic cells [2]. However, the research of FAM64A is inadequate; thus, the role of FAM64A in OS cells is still poorly understood. e dysregulation of intracellular signaling pathways such as Notch1, Akt, Wnt pathway, and JAK2/STAT3 was reported to be participated in the development of OS [3][4][5][6]. Xu et al. found that FAM64A served as a positive regulator of STAT3, which is linked to various cancer types [7]. Here, we tried to find a new mechanism that interacts with FAM64A in OS. In this study, we provide an insight into the expression patterns of genes in OS and control samples. We identified the genomic aberrations and the molecular mechanism associated with OS. We discovered the tumorderived miR-493 targeted to FAM64A and regulated the cell growth and metastasis of OS.

Genomic Data of OS.
From Gene Expression Omnibus (GEO; available at http://www.ncbi.nlm.nih.gov/geo/) database, we retrieved the gene expression profiling of OS. e keywords including Homo sapiens and OS were used. e informatics analysis of FAM64A levels in OS was performed in the Cancer Genome Atlas (TCGA) database.

Cell Viability Assay. OS cell viability was measured with
A CCK-8 assay (Dojindo Molecular Technologies, Japan). Specifically, cells with a density of 7,000 cells/well were firstly seeded in 96-well plates. After 6 h of culture, cells were transfected with MG-63-NC, MG-63-FAM64, U-2 OS-NC, and U-2 OS-FAM64A, respectively, and then incubated at 37°C with 5% CO 2 for 0, 24, 48, and 72 h, respectively. At each time-point, a CCK-8 reagent of 10 μl was added into each well, and the incubation was subsequently extended for an additional 2 h at 37°C. e absorbance of each well was measured with a microreader (Bio-Rad, USA) at 450 nm.

Cell Invasion Assay.
e invasiveness of OS cells was measured with a Transwell assay using Transwell chambers (8 μm, Corning) preburden Matrigel (BD, USA). In detail, cells were first collected and then resuspended in the FBSfree culture medium at a density of 2 × 10 5 cells/ml. After that, the upper chambers were then seeded with 200 μl cell suspension, while the lower chambers were filled with 600 μl DMEM containing 10% FBS. Following a further incubation for 48 h at 37°C, the noninvasive cells were removed by cotton-tipped swabs. Images of invasive cells were taken, and the number was counted under Nikon ECLIPSE TS100 (Nikon) at ×100 magnification.
2.8. Dual-Luciferase Activity Assay. PCR was used to amply the 3′UTR of FAM64A containing the potential miR-493 binding site. 293T cells were then cotransfected with NC/494 mimics and psiCHECK2-FAM64A3-UTR WT/psi-CHECK2-FAM64A3-UTR MT. e dual-luciferase reporter assay (Promega) was then used to measure the relative luciferase activity.

Western Blot
Analysis. Cellular lysates were electrophoresed in a 10% SDS-PAGE gel and then transferred to nitrocellulose membrane (GE Healthcare). Primary antibody anti-CATS 2C4 (1 : 1000) and GAPDH were purchased from Abcam. e membranes were rinsed with TBS (Tris-Buffered Saline) and Tween (Sigma-Aldrich) twice before being incubated with Goat Antimouse IgG H&L (HRP) for 1 h at room temperature in the dark. A Bio-Rad ChemiDoc ™ XRS system was then used for membrane visualization.

Wound Healing Assay.
e cells with a density of 8 × 10 4 cells/well were seeded in a 24-well plate. A vertical line was drawn among them with a sterile pipette tip after approximately 80% of the confluency was reached. e suspended single cells on the surface were then washed away with warm PBS. Fresh DMEM containing 10% FBS was added to plates, and the cells were then cultured in an incubator at 37°C with 5% CO 2 . e cells were imaged under a phase contrast light microscope, at 0, 24, and 48 h, respectively. Cells' migration ability was then measured with Image J (National Institutes of Health).

Statistical Analysis.
All data in this study were expressed as mean ± SD (standard deviation). SPSS 22.0 was used throughout this study for the statistical analysis, and one-way analysis of variance was used for comparison. Survival of mice in this study was measured with Kaplan-Meier analysis. A P < 0.05 was designated as statistically significant.

Overexpression of FAM64A in OS Tissues Predicted Poor
Prognosis. As shown in Figure 1(a), GSE12865 and GSE28425 were selected on GEO to select differentially expressed genes, with P < 0.05 and logFC absolute value >1.5.
ere were 1,504 GSE28425 differentially screened genes GSE12865 and 617 intersections, and a total of 147 genes were obtained (Figure 1(b)). e expression level of FAM64A in OS tissues and adjacent nontumor tissues were analyzed. As demonstrated in Figures 1(c) and 1(d), the gene expression level of FAM64A and its protein in tumor tissues were significantly heightened compared with the control sample (P < 0.05). To GO annotate these 147 genes, we focused on this FAM64A. GO: 0009987 cellular processes. Online software (http://www.oncolnc.org) was used to analyze the SARC data of TCGA, and it was found that the difference of KM curve logrank analysis between patients with high and low expression of this gene, suggesting patients with higher expression of FAM64A, had poor outcome (Figure 1(e)).

FAM64A Is the Target of miR-493.
rough bioinformatics analysis, we found that many miRNAs may regulate FAM64A. MiR-493 has been proved to play the role of tumor suppressor gene in osteosarcoma and inhibit the cell biological process of osteosarcoma, playing an opposite role with FAM64A [8,9]. So, we chose miR-493 to study its regulation on FAM64A expression. Our data showed that contains a potential complimentary binding site for miR-493 within FAM64A 3'-UTR ( Figure 5(a)). e possible participation of miR-493 in the FAM64A pathway is indicated in Figure 5 ese assays demonstrated that miR-493 inhibited proliferation, migration, and invasion of OS cells via regulating FAM64A.

Discussion
Recent studies have shown that FAM64A, also named as CATS and PIMREG, participates in malignant transformation [10][11][12][13]. FAM64A was first studied in hematologic carcinomas, which was known as CALM/AF10 interacting proteins. As reported, higher levels of FAM64A was associated with the tumor proliferation process. However, the role of FAM64A in solid tumor is still few. Here, we evaluated the gene expression profiling of OS in the database and found the FAM64A. Based on the data from GEO, we chose FAM64A and evaluated its expression in tumor samples. We found elevated expression of FAM64A in tumor tissues in comparison to the nontumor tissues.
Moreover, we identified the promotion of tumorigenicity by establishing overexpressing FAM64A OS cells model, which indicated underlying molecular mechanism of how miR-493 participated in upregulating migration and invasion of tumor cells. Previous evidence suggested that miR-493 inhibits the biological behavior of lung tumor [14]. Meanwhile, miR-493 also acts as a suppressor in various cancers including colon cancer, bladder cancer, and ovarian cancer via multiple intracellular signaling pathways [15][16][17].
Next, we performed a xenograft model of tumor and further investigated the prognostic role of FAM64A in mice. We found a sharp decrease of tumor volume and tumor weight in siRNA-transfected mice, indicating silencing FAM64A suppressed tumor growth. Kaplan-Meier curves for OS in vivo revealed that upregulation of FAM64A was positively correlated with worse outcomes. To confirm the interaction of FAM64A and miR-439, we performed the luciferase reporter assay. e effects of FAM64A knockdown on OS cells mimicked those induced by miR-493 mimics and were reversed by miR-493 inhibitor.      Taken together, our data demonstrated that miR-493 negatively regulates FAM64A via binding to its 3′ UTR in OS. Our study found that higher expression of FAM64A in vitro and in vivo is correlated with poor survival in OS. is is the first study to shed light on the participation of FAM64A regulated by miR-439 in the malignancy of osteosarcoma.
Data Availability e data and materials used to support the findings of this study are included within the published article.

Conflicts of Interest
e authors declare that they have no conflicts of interest.
Acknowledgments is work was supported by grant NSFC81300346 from the National Natural Science Foundation of China. e data and materials used to support the findings of this study are included within the published article.