microRNA-613 exerts anti-angiogenic effect on nasopharyngeal carcinoma cells through inactivating the AKT signaling pathway by down-regulating FN1

Background: Nasopharyngeal carcinoma (NPC) is a disease highly sensitive to radiotherapy with the unclear etiology. However, the specific effects of microRNA-613 (miR-613) on NPC still remain elusive. Therefore, the present study probes into the underlying mechanism of miR-613 in NPC via AKT signaling pathway by regulating Fibronectin 1 (FN1). Methods: First, microarray analysis was used to screen differentially expressed genes (DEGs) and regulatory miRs associated with NPC. Next, miR-613 and FN1 expression in NPC cells was determined, followed by verification of target relationship between miR-613 and FN1. With NPC cells exposed to miR-613 mimic, si-FN1 and LY294002 (inhibitor of AKT signaling pathway), the regulatory effects of miR-613 on proliferation, apoptosis, invasion, migration and angiogenesis of NPC cells were detected with ratio of B-cell lymphoma 2/Bcl-2-associated X protein (Bcl-2/Bax), Cleaved-caspase3, matrix metallopeptidase 2 (MMP-2), MMP-9, vascular endothelial growth factor (VEGF), results of tube formation assay, which was employed to assess angiogenesis of HONE1 and CNE1 cells after transfection, suggested that the angiogenesis rate did not differ in the NC-mimic group and the si-NC group compared with the blank group ( P > 0.05). However, the LY294002 group exhibited a significant decline in angiogenesis rate in comparison with the blank group ( P < 0.05). NPC cells transfected with miR-613 mimics showed obviously reduced angiogenesis rate versus cells transfected with NC-mimic ( P < 0.05). Compared with the si-NC group, the angiogenesis rate of the si-FN1 group was significantly decreased ( P < 0.05). Relative to the miR-613 mimic group, the miR-613 mimic + oe-FN1 group exhibited enhanced angiogenesis rate P < 0.05, a conclusion could be that overexpression of miR-613, silencing of FN1 or LY294002 treatment exerted inhibitory effects on angiogenesis of NPC cells by inactivating the AKT signaling pathway.

radiation [6]. In spite of these treatments, over 1/3 patients still suffer from local recurrence mainly due to distant metastasis [7]. Therefore, it is in urgent need to develop new biomarkers for NPC treatment.
MicroRNAs (miRs) have been extensively reported to be closely involved in the developmental processes of tumors by regulating the expression of genes, including cell proliferation, migration, apoptosis, etc. [8,9]. miR-613 expression is remarkably low in human hepatocellular carcinoma (HCC) cells, and overexpression of miR-613 suppresses proliferation of HCC cells [10]. Meantime, a prior study showed that restored expression of miR-613 suppressed NPC cell invasion induced by Snail2 [11]. Moreover, miR-613 was revealed to target Fibronectin 1 (FN1) in a previous study [12]. FN1, a member belonging to the class of extracellular matrix glycoprotein, executes its functions in cellular adhesion, migration, tissue remodeling, and infection resistances [13]. High FN1 expression is responsible for metastasis, carcinogenesis, and invasion in various human carcinomas including renal cancer and colorectal cancer [14,15]. It was previously revealed that overexpressed miR-9-3p could lead to inhibited metastases and proliferation of NPC cells by reducing FN1 [13]. Furthermore, Wu et al. revealed that miR-613 modulated the AKT signaling pathway in ischemia reperfusion (I/R) [16]. A previous study has found that AKT signaling pathway regulates several cellular courses such as cell proliferation, apoptosis, and tumorigenesis, which eventually leads to NPC progression [17]. It has been revealed that the inactivation of the AKT signaling pathway could contribute to the inhibition of metastasis of NPC cells [18]. Based on these findings, it is suggested that miR-613, FN1 as well as the AKT signaling pathway may work in the development of NPC. Therefore, the present study furnishes a new potential effect of miR-613 on modulating invasion, migration, and angiogenesis of NPC cells via the AKT signaling pathway by regulating FN1.

Microarray analysis
The Gene Expression Omnibus (GEO) database (https://www.ncbi.nlm.nih.gov/geo/) was employed to retrieve the microarray data related to NPC, and the 'limma' package in R language was used for differential analysis of the NPC-related microarray data. The differentially expressed genes (DEGs) were screened with |logFC| > 2 and P value < 0.05 used as the screening threshold, and the 'pheatmap' package was applied to construct the heatmap for DEGs. The STRING database (https://string-db.org/) was applied for gene interaction analysis, with the analysis results exported. Then, the exported analysis results were imported into the cytoscape software, and then the core degree values of 22 genes in interaction network were calculated using the statistical tool of the cytoscape software. Based on the degree values, a map of gene interaction network was constructed, with the degree values of genes labeled using different colors, the deeper color indicated the higher degree value of gene and the higher core level of gene in the interaction network. The DIANA database (http://diana.imis.athena-innovation. gr/DianaTools/index.php?r=microT CDS/index), miRDB database (http://mirdb.org/miRDB/index.html), mirDIP database (http://ophid.utoronto.ca/mirDIP/index.jsp#r), miRSearch database (https://www.exiqon.com/miRSearch), starBase database (http://starbase.sysu.edu.cn/) and Target Scan database (http://www.targetscan.org/vert 71/) were used to retrieve the miRs that regulated FN1, with the intersection of the predicted results obtained.

Cell culture and transfection
A total of four NPC cell lines 5-8F, CNE2, CNE1, and HONE-1 and one immortalized human nasopharyngeal epithelial cell line NP69 (American Type Culture Collection [ATCC), Manassas, VA, U.S.A.) were incubated in an incubator containing RPMI-1640 complete medium consisting of 10% fetal bovine serum (FBS), 100 μg/ml streptomycin and 100 U/ml penicillin at 37 • C with 5% CO 2 and 95% saturated humidity with the medium replaced 3-4 times per week depending on the cell growth. Cells were sub-cultured when the cell confluence reached about 80%. Reverse transcription quantitative polymerase chain reaction (RT-qPCR) was carried out to measure the level of miR-613 in each cell line in order to screen out two cell lines with the lowest miR-613 level for following cell experimentations.
was then added to the culture wells and cultured in an incubator with 5% CO 2 at 37 • C. After 4 h, with medium changed to complete medium, cells continued to be cultured for 48 h and were collected for subsequent experiments.

Dual-luciferase reporter gene assay
In order to confirm whether miR-613 acted as transcription factor to target FN1, the 3 untranslated region (

5-Ethynyl-2 -deoxyuridine assay
The 5-ethynyl-2 -deoxyuridine (EdU) solution (cell medium:EdU solution = 1000:1) was added into the cell culture plate, which was then incubated at room temperature for 2 h and washed with PBS. Then the cells were fixed with 40 g/L paraformaldehyde for 30 min, followed by incubation in glycine solution for 8 min and PBS washing. After being rinsed with PBS containing 0.5% Triton X-100 and incubated with Apollo ® staining solution at room temperature avoiding exposure to light for 30 min, the cells were washed twice with methanol and PBS, respectively. Then, the cells were incubated with Hoechst 3334 reaction solution at room temperature for 20 min avoiding exposure to light. Images were captured under a fluorescence microscope. When photographed with green light at the excitation wavelength of 488 nm, the green-stained cells were proliferating cells. When photographed with purple light at the excitation wavelength of 350 nm, the blue-stained cells were total cells. With three visual fields selected, the EdU-stained cells (proliferating cells) and Hoechst 33342-stained cells (total cells) were counted. Cell proliferation rate = number of proliferating cells/number of total cells × 100%. The parallel experiment was repeated three times.

Flow cytometry
The apoptosis of CNE1 and HONE-1 cells after 24-h culture was detected by Annexin V-fluorescein isothiocyanate (FITC)/propidium iodide (PI) double staining kits (556547, Shanghai Shuojia Biotechnology Co., Ltd., Shanghai, China). The 10 × Binding Buffer was diluted to 1 × Binding Buffer with deionized water. Cells were collected after centrifugation at 2000 rpm for 5 min at room temperature. The cells were then resuspended by precooled 1 × PBS, centrifuged at 200 rpm for 5-10 min, washed, and suspended with 300 μl 1 × Binding Buffer. After mixing with 5 μl Annexin V-FITC, the cells were incubated for 15 min at room temperature avoiding exposure to light. Finally, the cells were ice-bathed with the addition of 5 μl PI avoiding exposure to light for 5 min. The flow cytometer (Cube 6, Partec, Munster, Germany) was applied for detection of FITC at the excitation wavelength of 480 and 530 nm and PI at the excitation wavelength more than 575 nm.

Transwell assay
After transfection for 48 h and starvation in serum-free medium for 24 h, the cells were detached, washed twice with PBS, and resuspended with serum-free Opti-MEMI medium (Invitrogen, Carlsbad, California, U.S.A.) supplemented with 10 g/L BSA, with the cell density adjusted to 3 × 10 4 cells/ml. A 24-well plate and an 8 μm Transwell chamber (Corning Inc., Corning, NY, U.S.A.) were adopted with three chambers in each group and 100 μl cell suspension in each chamber. The basolateral chamber was incubated with the addition of 600 μl 10% RPMI-1640 medium with 5% CO 2 at 37 • C. For cell migration detection, after 48 h, cells were fixed with 4% paraformaldehyde for 30 min and the chamber was placed in 0.2% Triton X-100 solution (Sigma-Aldrich, St. Louis, MO, U.S.A.) for 15 min, and stained in 0.05% gentian violet solution for 5 min. As to cell invasion detection, 50 μl Matrigel (Sigma-Aldrich, St. Louis, MO, U.S.A.) was spread in the chamber prior to the experiment, and 48 h later, fixation and staining were performed using the above-mentioned methods. The number of stained cells was counted under an inverted microscope. Five visual fields were randomly selected for cell counting, and the number of cells was presented using mean. The parallel experiment was repeated three times.

Tube formation assay
Following transfection, the cancer cells were co-cultured with human umbilical vein endothelial cell (HUVEC) (ATCC, Manassas, VA, U.S.A.). Each group was set with three duplicated wells. A total of 60 μl Matrigel was added into each well of a 96-well plate which was gently shaken to enable the Matrigel spread evenly. After the Matrigel gel was coagulated, the HUVEC (ATCC, Manassas, VA, U.S.A.) were resuspended with the supernatant of the transfected HONE-1 cells, with cell concentration adjusted to 2 × 10 5 ml −1 . Afterwards, the above-mentioned cell suspension (150 μl) was cultured in the 96-well plate coated with Matrigel at 37 • C with 5% CO 2 for 24 h. Then, the tube formation of HUVEC was observed under a phase contrast microscope (×100). Five visual fields were randomly selected from each group to assess the angiogenesis of HUVEC co-cultured with cancer cells in each group, which was expressed as the number of tube formation.

Xenograft in nude mice
Forty-two 5 weeks old male BALB/c nude mice weighing 18-20 g purchased from Shanghai Laboratory Animal Center of Chinese Academy of Sciences (Shanghai, China) were housed in specific pathogen-free (SPF) environment with constant temperature of 25 + − 2 • C, constant humidity of 45-50% and 12 h of dark/light cycle. The nude mice were fed with high-pressure sterilization standard laboratory diet with free access to sterile water. All nude mice were randomly classified into seven groups with six mice in each group: blank, NC-mimic, miR-613 mimic, si-NC, si-FN1, miR-613 inhibitor + si-FN1 and LY294002 groups. The transfected HONE-1 cells of seven groups in exponential growth in the incubator were treated with 0.25% trypsin. The collected cells were mixed with serum-free medium to prepare single cell suspension with a concentration of 1 × 10 7 cells/ml. Subsequently, 100 μl of cell suspension was inoculated subcutaneously into the dorsal root of the right hind limb in nude mice, and each nude mouse was inoculated with approximately 1 × 10 6 tumor cells. Tumors were measured with vernier calipers every 3 days. The nude mice were killed under anesthesia after 6 weeks, and the skin was cut, followed by dissection of the intact transplanted tumors. The tumors were then weighed by electronic balance and the mass of terminal tumor was recorded.

Immunohistochemistry
The microvessel density (MVD) in tumor tissues was measured in accordance with the Weidner criteria. Sections were observed under a low-power microscope. Five areas with the most stained microvessel were selected to search location rich in vessels, which were located in or near the tumor tissue area. Tumor tissues were collected for immunohistochemical staining, with the primary antibody against CD34 (ab81289, 1/2000, Abcam, Cambridge, MA, U.S.A.) used, and the vascular endothelial cells stained brown were considered as positive. MVD was counted under a high-power microscope according to the criteria for identifying any stained endothelial cells or cells as independent vessels. Five different visual fields were selected on each slide, and the stained vessels were simultaneously counted under a multi-eye fluoroscopy microscope to obtain the mean value.

Statistical analysis
The data in the present research were analyzed by SPSS 21 (IBM Corp. Armonk, NY, U.S.A.). The measurement data were presented in the form of the mean + − standard deviation. First, the normality and the homogeneity of variance were tested. If the data were in accordance to the normal distribution and homogeneity of variance, intra-group comparison was conducted using paired t-test, inter-group comparison using unpaired t-test and data among multiple groups were compared by one-way analysis of variance (ANOVA) or repeated measurement ANOVA, and pairwise comparison by the post-hoc test. The data with skewed distribution or unequal variances were analyzed using the rank sum test.

miR-613 influences the development of NPC via the AKT signaling pathway by regulating FN1
Four NPC-related microarray data GSE13597, GSE12452, GSE53819, and GSE64634 were retrieved from the GEO database. Differential analysis concerning the NPC samples and normal control samples in these four microarray data was conducted, obtaining 602, 766, 478, and 1033 DEGs, respectively. The heat maps of top 50 DEGs in four microarray data are shown in Figure 1A-D. Venn analysis of the four microarray data was conducted to further obtain DEGs in NPC ( Figure 1E), revealing that 22 genes existed in the intersection of the four microarray data, which were all differentially expressed in the four NPC-related microarray data. The interaction among the 22 genes was further analyzed and the gene interaction network was constructed ( Figure 1F), showing that FN1 gene was in the core of the 22 genes. Through the Kyoto Encyclopedia of Genes and Genomes (KEGG) database, FN1 gene was found to be located in the upstream of the AKT signaling pathway (map05205). Besides, the AKT signaling pathway has been reported to act an essential role in NPC [19,20], suggesting that FN1 might regulate NPC through the AKT signaling pathway. In order to further understand the upstream regulatory mechanism of FN1, the upstream regulatory miR of FN1 was predicted in databases such as DIANA, and the intersection of the predicted results was obtained ( Figure 1G). Finally, six potential miRs regulating FN1 were obtained from the intersection, among which miR-613 was rarely studied in cancers, especially in NPC. Nevertheless, the other five miRs have already been reported in cancers including NPC [21][22][23]. Taken together, miR-613 participated in the development of NPC via the AKT signaling pathway by mediating the expression of FN1 gene. Then, miR-613 and FN1 expression in CNE1, CNE2, HONE1, 5-8F, and NP69 cell lines were measured by RT-qPCR and Western blot analysis ( Figure 1H,I). In comparison with the NP69 cell line, The measurement data were expressed as the mean + − standard deviation; data between two groups were compared by the paired t-test; data among multiple groups were compared by one-way ANOVA; the experiment was repeated three times.
the NPC cell lines exhibited significantly decreased miR-613 expression and obviously increased FN1 expression, of which HONE1 and CNE1 exhibited the relatively lower miR-613 expression but the relatively higher FN1 expression. Therefore, HONE1 and CNE1 cells were selected for follow-up experiment.

FN1 is the target gene of miR-613
The Target Scan website showed that a binding relationship between miR-613 and FN1 (Figure 2A). On this basis, dual luciferase reporter gene assay was conducted to verify the target relationship between miR-613 and FN1, and the results illustrated that compared with the NC group, the luciferase activity in FN1-Wt was obviously decreased (P<0.05) but that in FN1-Mut was not affected (P>0.05) with the overexpression of miR-613 ( Figure 2B).
At the same time, protein expression of FN1 in HONE1 and CNE1 cell lines transfected with miR-613 mimic or si-FNI was detected by Western blot analysis. The results in CNE1 cell line revealed that compared with the blank group, the protein expression of FN1 did not differ in the NC-mimic group and the si-NC group (P>0.05). The protein expression of FN1 was greatly lower in the miR-613 mimic group than that in the NC-mimic group (P<0.05). Compared with the si-NC group, the si-FN1 group showed marked declines in protein expression of FN1 (P<0.05, Figure 2C,D). The results in HONE1 cell line were consistent with the results of CNE1 cell line ( Figure 2E,F). These results suggested that miR-613 could target and down-regulate FN1.

Overexpressed miR-613 or silenced FN1 represses cell proliferation, invasion, and migration, and promotes apoptosis in NPC by inactivating the AKT signaling pathway
Results of EdU assay for the detection of cell proliferation of HONE1 and CNE1 cells after transfection showed that compared with the blank group, there was no obvious difference in cell proliferation in the NC-mimic group and si-NC group (P>0.05), but cell proliferation was significantly reduced in the LY294002 group (P<0.05). Cell proliferation of the miR-613 mimic group was evidently inhibited versus that in the NC-mimic group (P<0.05). Relative to the si-NC group, the si-FN1 group exhibited dramatically suppressed cell proliferation (P<0.05). In comparison with miR-613 mimic-treated cells, proliferation was promoted in cells transfected with both miR-613 mimic and oe-FN1 (P<0.05, Figures 3A and 4A). Taken together, overexpression of miR-613, silencing of FN1 or LY294002 treatment in HONE1 cell. *P<0.05 compared with the NC-mimic group; # P<0.05 compared with the si-NC group; the measurement data were expressed as the mean + − standard deviation; data between two groups were compared by the unpaired t-test; data among multiple groups were compared by one-way ANOVA; the experiment was repeated three times; Wt, wild type; Mut, mutant type.

had inhibitory effects on proliferation of NPC cells.
Results of flow cytometry, which was performed to assess cell apoptosis of HONE1 and CNE1 cells after transfection, showed that the apoptosis rate did not differ greatly in the NC-mimic group and the si-NC group compared with blank group (P>0.05). However, in contrast to the blank group, the LY294002 group presented obviously increased apoptosis rate of NPC cells (P<0.05). Compared with the NC-mimic group, the apoptosis rate of NPC cells in the miR-613 mimic group was significantly elevated (P<0.05). Compared with the si-NC group, the apoptosis rate of NPC cells in the si-FN1 group was greatly increased (P<0.05). In comparison with the NPC cells treated with miR-613 mimic, the NPC cells treated with both miR-613 mimic and oe-FN1 displayed down-regulated apoptosis rate (P<0.05, Figures 3B and 4B). These findings demonstrated that miR-613 overexpression, FN1 silencing, or LY294002 treatment could promote apoptosis of NPC cells.
Subsequently, Western blot analysis was applied to examine the protein expression of proliferation-related factors and apoptosis-related factors in HONE1 and CNE1 cells. It revealed that in comparison with the blank group, the NC-mimic and si-NC groups showed no significant difference in the expression of Bcl-2, Bax, and Cleaved-caspase3 (P>0.05), while the LY294002 group exhibited significantly lower ratio of Bcl-2/Bax but obviously higher expression of Cleaved-caspase3 (P<0.05). In contrast to the NC-mimic group, significantly decreased ratio of Bcl-2/Bax and obviously increased expression of Cleaved-caspase3 were found in the miR-613 mimic group (P<0.05). Compared with the si-NC group, the si-FN1 group depicted lower ratio of Bcl-2/Bax but higher expression of Cleaved-caspase3 (P<0.05). Besides, obviously increased ratio of Bcl-2/Bax and decreased expression of Cleaved-caspase3 were found after co-treatment with miR-613 mimic and oe-FN1 compared with the miR-613 mimic treatment alone (P<0.05, Figures 3C and 4C). From these findings, a conclusion could be drawn that miR-613 overexpression, FN1 silencing, or LY294002 treatment decreased the ratio of Bcl-2/Bax and increased expression of Cleaved-caspase3.
Transwell assay was performed for detection of cell migration and invasion of HONE1 and CNE1 cells after transfection. The results suggested that versus the blank group, there was no evident difference concerning the migration  .05 compared with the si-NC group; @ P<0.05 compared with the miR-613 mimic group; the measurement data were expressed as the mean + − standard deviation; data among multiple groups were compared by one-way ANOVA; the experiment was repeated three times. and invasion of cells in the NC-mimic group and the si-NC group (P>0.05) while cell migration and invasion were significantly attenuated in the LY294002 group (P<0.05). Compared with the NC-mimic group, the miR-613 mimic group exhibited inhibited cell migration and invasion (P<0.05). Cell migration and invasion in the si-FN1 group were markedly attenuated in contrast to that in the si-NC group (P<0.05). Compared with the NPC cells treated with miR-613 mimic, the migration and invasion in NPC cells transfected with miR-613 mimic and oe-FN1 were dramatically promoted (P<0.05, Figures 3D and 4D). Therefore, miR-613 elevation, FN1 down-regulation or LY294002 treatment inhibited cell migration and invasion in NPC.
Here, the levels of factors related to migration (Figures 3E and 4E) and invasion in HONE1 and CNE1 cells after transfection was measured by Western blot analysis. It revealed that the levels of MMP-2 and MMP-9 in the NC-mimic and si-NC groups were not remarkably different from those in the blank group (P>0.05), but the expression of MMP-2 and MMP-9 in the LY294002 group was evidently down-regulated versus the blank group (P<0.05). Compared with the NC-mimic group, the expression of MMP-2 and MMP-9 in the miR-613 mimic group was greatly decreased (P<0.05). The si-FN1 group presented evident declines in the expression of MMP-2 and MMP-9, relative to the si-NC group (P<0.05). In comparison with the miR-613 mimic group, the expression of MMP-2 and MMP-9 in the miR-613 mimic + oe-FN1 group was greatly up-regulated (P<0.05, Figures 3F and 4F). From these results, the conclusion could be made that overexpressed miR-613, silenced FN1 or the LY294002 treatment could lead to suppressed migration and invasion of NPC cells.

Overexpression of miR-613 or silencing of FN1 inhibits angiogenesis of NPC cells
The results of tube formation assay, which was employed to assess angiogenesis of HONE1 and CNE1 cells after transfection, suggested that the angiogenesis rate did not differ in the NC-mimic group and the si-NC group compared with the blank group (P>0.05). However, the LY294002 group exhibited a significant decline in angiogenesis rate in comparison with the blank group (P<0.05). NPC cells transfected with miR-613 mimics showed obviously reduced angiogenesis rate versus cells transfected with NC-mimic (P<0.05). Compared with the si-NC group, the angiogenesis rate of the si-FN1 group was significantly decreased (P<0.05). Relative to the miR-613 mimic group, the miR-613 mimic + oe-FN1 group exhibited obviously enhanced angiogenesis rate (P<0.05, Figure 5A,B). Thus, a conclusion could be made that overexpression of miR-613, silencing of FN1 or LY294002 treatment exerted inhibitory effects on angiogenesis of NPC cells by inactivating the AKT signaling pathway.
Expression of angiogenesis-related factors VEGF and CD31 in HONE1 and CNE1 cells after transfection was measured by Western blot analysis. The results showed that the expression of VEGF and CD31 did not differ in the NC-mimic group and the si-NC group in comparison with the blank group (P>0.05), but the levels of VEGF and CD31 were significantly reduced in the LY294002 group (P<0.05). The expression of VEGF and CD31 was greatly down-regulated in the miR-613 mimic group (P<0.05) compared with the NC-mimic group, reduced in the si-FN1 group versus the si-NC group (P<0.05), and elevated in miR-613 mimic + oe-FN1 group relative to the miR-613 mimic group (P<0.05, Figure 5C,D). Taken together, the levels of angiogenic factors were decreased due to the overexpression of miR-613, silencing of FN1 or LY294002 treatment.

miR-613 inactivate the AKT signaling pathway through inhibiting FN1 in NPC
In order to determine the effects of miR-613 on the AKT signaling pathway, HONE1 and CNE1 cells were transfected with miR-613 overexpression or FN1 silencing to conduct Western blot analysis for the determination of the protein expression of factors related to the AKT signaling pathway protein and phosphorylated proteins. Compared with the blank group, there was no significant difference in the expression of AKT and mTOR as well as the phosphorylation extents of AKT and mTOR in the NC-mimic group and the si-NC group (P>0.05). In the LY294002 group, AKT and mTOR expression and their phosphorylation extents were remarkably inhibited versus the blank group (P<0.05). Compared with the NC-mimic group, the expression and phosphorylation extents of AKT and mTOR were significantly down-regulated in the miR-613 mimic group (P<0.05). The si-FN1 group exhibited evident declines in expression and phosphorylation extents of AKT and mTOR, relative to the si-NC group (P<0.05). Compared with the miR-613 mimic group, the expression and phosphorylation extents of AKT and mTOR in the miR-613 mimic + oe-FN1 group were significantly up-regulated (P<0.05, Figure 6A-D). Hence, overexpression of miR-613 down-regulated FN1, thus inactivating the AKT signaling pathway. .05 compared with the NC mimic group; # P<0.05 compared with the si-NC group; @ P<0.05 compared with the miR-613 mimic group; the measurement data were expressed as the mean + − standard deviation; data among multiple groups were compared by one-way ANOVA; the experiment was repeated three times.

Overexpression of miR-613 or silencing of FN1 inhibits tumorigenesis and angiogenesis
Tumorigenic ability and neovascular MVD in HONE1 cell after transfection were detected by xenograft in nude mice and immunohistochemistry, respectively. The results suggested that in comparison with the blank group, tumor growth rate, tumor weight, and MVD did not differ in the NC-mimic group and the si-NC group (all P>0.05) but the tumor growth rate, tumor weight and MVD were significantly reduced in the LY294002 group (P<0.05). Mice in the miR-613 mimic group exhibited lower tumor growth rate, tumor weight, and MVD, relative to mice in the NC-mimic (P<0.05). In comparison with the si-NC group, the tumor growth rate, tumor weight, and MVD were significantly decreased in the si-FN1 group (P<0.05). In contrast to the miR-613 mimic group, the tumor growth rate, tumor weight, and MVD were significantly elevated in the miR-613 mimic + oe-FN1 group (P<0.05, Figure  7A-E). The results demonstrated that overexpression of miR-613, silencing of FN1, or LY294002 treatment could inhibit the tumorigenesis and angiogenesis.

Discussion
NPC is a squamous cell carcinoma that derives from the nasopharynx epithelium, which presents remarkable geographical and racial differences [3,5]. Up to now, several EBV-encoded and human miRs have been evidenced to be dysregulated in NPC, thus exerting functions in the tumorigenesis, metastasis and invasion of NPC cells [1]. Nevertheless, the potential mechanism of miR-613 in NPC is still unclear, and the present study probes into a new potential role of miR-613, FN1, and AKT signaling pathway in modulating invasion, migration, and angiogenesis of NPC cells. As a consequence, the present study demonstrates that up-regulation of miR-613 suppresses NPC cell invasion, metastasis, and angiogenesis through inactivation of the AKT signaling pathway by inhibiting FN1.
Initially, our results demonstrated that down-regulation of miR-613 and overexpression of FN1 was found in NPC tissues. Similarly, Wang et al. found that miR-613 expression was remarkably reduced in HCC cells [10]. Results of the study conducted by Guan and his team indicated the down-regulation of miR-613 in serum samples and cancerous tissue of patients with esophageal squamous cell carcinoma [12]. Meanwhile, a previous study emphasized that up-regulation of miR-613 suppressed NPC cell invasion induced by Snail2 [11]. Previous studies found that FN1 expression is significantly high in NPC tissues and FN1 plays a crucial role in cell metastasis and differentiation [13,24].  Next, the results of dual luciferase reporter gene assay demonstrated that miR-613 could target FN1; miR-613 suppresses invasion, metastasis, and angiogenesis of NPC cells by targeting FN1. Overexpressed miR-613 has been revealed to inhibit the bladder cancer cell invasion, proliferation, and metastasis via regulation of the expression of Sphk1 [25], which was partly consistent with our results. Interestingly, a very recent study proved that up-regulation of miR-613 suppressed cell invasion, metastasis, and proliferation through directly targeting and inhibiting VEGFA [26]. Up-regulated miR-15a and miR-16 inhibited angiogenesis multiple myeloma by targeting VEGF, which was proved by a recent study [27]. Interestingly, overexpressed miR-92a in angiogenic endothelial cells was reported to exert an anti-angiogenic function in cancer [28]. Moreover, it is revealed that FN1 is a target gene of miR-613 [12]. FN1, a member of ECM glycoprotein family, plays a key role in cellular adhesion, migration polarity, and tissue remodeling; FN1 is also conducive to microvascular integrity maintenance and infection resistances [13]. Furthermore, it has been recently verified that FN1 was closely correlated to cell metastasis, differentiation, and adhesion in various cancers, and the down-regulation of FN1 causes suppression of the invasion and metastasis in cancer cells [24]. Consistently, FN1 down-regulation could repress cell invasion, metastasis, and proliferation in esophageal cancer cells [29]. Additionally, it was suggested that attenuated FN1 could effectively repress the metastasis of gastric cancer cells, and that miR-200c could lead to suppression of the invasion, metastasis, and proliferation of gastric cancer cells via the down-regulation of FN1 [30].
Besides, our research also revealed that overexpression of miR-613 reduces tumor invasion, metastasis, and angiogenesis in NPC through inactivating the AKT signaling pathway by inhibiting FN1. AKT signaling pathway acts an essential function in modulating several cellular processes, such as cell survival, proliferation, apoptosis, and nutrient metabolism, which function in both normal and cancer cells [31]. A previous study suggested that AKT signaling pathway played a key role in carcinogenesis of NPC, and that down-regulation of AKT signaling pathway suppressed the development of NPC cells [32]. In addition, suppressing the activation of AKT signaling pathway was demonstrated to inhibit invasion and metastasis in NPC cells [33]. miR-613 was demonstrated to influence I/R-induced cardiomyocyte apoptosis through regulating the AKT signaling pathway, reported by a very recent study by Wu et al. [16]. The results of the current study provide evidence that FN1 regulates NPC via the AKT signaling pathway.

Conclusions
In line with the previous studies, we have confirmed down-regulation of miR-613 and overexpression of FN1 in NPC tissues. We also demonstrate that FN1 is the target gene of miR-613, and up-regulation of miR-613 suppresses invasion, migration, and angiogenesis in NPC cells through inactivation of AKT signaling pathway via down-regulation of FN1 ( Figure 8). Therefore, we speculate that miR-613 may be potential new direction in the treatment of NPC. Nevertheless, further study is required to probe into the mechanism through which miR-613 regulates the AKT signaling pathway in the cancer.