MiR-135b-5p and MiR-499a-3p Promote Cell Proliferation and Migration in Atherosclerosis by Directly Targeting MEF2C

Proliferation and migration of endothelial cells (ECs) and vascular smooth muscle cells (VSMCs) are critical processes involved in atherosclerosis. Recent studies have revealed that microRNAs (miRNAs) can be detected in circulating blood with a stable form and the expression profiles differ in many cellular processes associated with coronary artery disease (CAD). However, little is known about their role, especially serum-derived miRNAs, in ECs and VSMCs phenotype modulation during atherosclerosis. We compared the miRNA expressions in serum samples from 13 atherosclerotic CAD patients and 5 healthy control subjects and identified 36 differentially expressed miRNAs. The expression of selected miRNAs (miR-135b-5p and miR-499a-3p) was further validated in 137 serum samples. Interestingly, miR-135b-5p and miR-499a-3p directly regulated a common target gene: myocyte enhancer factor 2C (MEF2C) which plays an important role in modulating cell phenotype of cardiovascular systems. Furthermore, our results indicated that the 2 elevated miRNAs could jointly promote ECs and VSMCs proliferation and migration by repressing MEF2C expression. Together, our findings demonstrated a serum-based miRNA expression profile for atherosclerotic CAD patients, potentially revealing a previously undocumented mechanism for cell proliferation and migration mediated by miR-135b-5p and miR-499a-3p, and might provide novel insights into the role of circulating miRNAs in atherosclerosis pathogenesis.


MiRNA Profiles in Atherosclerotic CAD Patients Versus That of Healthy Volunteers.
To identify aberrantly circulating miRNAs during atherosclerosis, a serum-based miRNA expression was performed in 13 atherosclerotic CAD patients and 5 healthy donors using the miRCURY LNA Array (version 11.0) system (GEO Submissions Number: GSE 53675). The levels of circulating miRNAs differed significantly between patients and healthy controls (Fig. 1). Of 1,700 miRNAs detected on the microarray, 36 were found to be differentially expressed in the patients compared with that of the controls (P < 0.05); 26 miRNAs were increased and 10 were decreased in the patient samples compared with those in the control samples. Quantification revealed that multiple miRNAs were significantly elevated in the patients compared with that in the healthy controls. According to their known and postulated functions, 13 miR-NAs were selected and further tested. The miRNA names, fold changes, and P values in the microarray cohort are presented in Table 1. In addition, the relative expressions of the 13 miRNAs in atherosclerotic patients compared with that in the healthy controls were shown in Supplementary Fig. S1 and the relative expressions of the other 26 miRNAs were presented in Supplementary Fig. S2.

Validation of MiRNAs Expressions.
To confirm the findings of the miRNA profile, the expression of these 13 dysregulated miRNAs was measured using qRT-PCR. The relative expression of each miRNA in atherosclerotic patients compared with that in the healthy controls was shown in Fig. 2A. The results demonstrated a 9.7-fold increase in miR-499a-3p (P < 0.0001) and a 2.7-fold increase in miR-135b-5p (P < 0.0001) in atherosclerotic CAD patients.
MiR-135b-5p and miR-499a-3p are of particular interest among all the validated miRNAs. MiR-499a-3p was the focus of this study because it had the highest expression level in atherosclerotic patients compared with that in healthy controls, both in the miRNA profile and quantitative PCR (Table 1 and Fig. 2A). Although miR-135b-5p is known to be associated with cancer 18,19 , recent work suggests a critical role for this miRNA regulated by the interleukin (IL)-1 receptor in mediating inflammatory responses 20 . Most importantly, computational analysis from several widely used mammalian target prediction programs, including miRBase, TargetScan, miRanda, and PicTar [21][22][23] , indicated that both miR-135b-5p and miR-499a-3p regulate several common target genes, which potentially contributes to impaired cell function in atherosclerosis.
Scientific RepoRts | 5:12276 | DOi: 10.1038/srep12276 To further validate the elevation of miR-135b-5p and miR-499a-3p in the atherosclerotic process, a series of independently validation cohorts were measured. One hundred and twenty-two serum samples were collected from 77 atherosclerotic CAD patients and 45 healthy control subjects. The clinical features of these subjects (Supplementary Table S2) did not significantly differ from that of the previous set in the miRNA profiles (Supplementary Table S1). As shown in Fig. 2B, the 2 miRNAs were found to be significantly upregulated in atherosclerotic patients compared with those in the healthy controls. Our results showed a 3.9-fold increase in miR-135b-5p and a 4.7-fold increase in miR-499a-3p expression in atherosclerotic CAD patients compared with those in healthy control subjects (both P < 0.001). To confirm the cardiac atherogenic ontogeny of the 2 miRNAs, miR-135b-5p and miR-499a-3p were also assessed in 15 non-atherogenic cardiovascular patients (3 with cardiogenic shock, 5 with heart failure, and 7 with ventricular arrhythmia). As shown in Fig. 2B, the results showed decreasing trends of miR-135b-5p and miR-499a-3p levels in non-atherogenic cardiovascular patients compared to those in healthy control Figure 1. Heat map of microRNA (miRNA) microarray expression data from serum samples of atherosclerotic patients (n = 13) and healthy control subjects (n = 5) . The expression of miRNA is hierarchically clustered on the y axis, and atherosclerotic serum samples or healthy control serum samples are hierarchically clustered on the x axis. The relative miRNA expression is depicted according to the color scale shown on the right. Red indicates upregulation; blue, downregulation. Numbers (1,9,10,12,14,15,17,22,26,31,33,36,37) indicate atherosclerotic patients' samples; numbers (C-4, 5, 6, 7, 8) indicate healthy control samples.
subjects. In summary, these data demonstrated that circulating levels of miR-135b-5p and miR-499a-3p were significantly increased in CAD patients with atherosclerosis.
Additionally, we also analyzed the association of the two miRNAs with specific factors for atherosclerosis. Both miR-135b-5p and miR-499a-3p were significantly associated with age, stoke, beta-blockers, and aspirin (Supplementary Table S3). MiR-135b-5p was highly influenced by ACE inhibitors and statin treatment was significantly associated with increased levels of circulating miR-499a-3p (Supplementary  Table S3); however, when only atherosclerotic patients were included in the analysis, hypertension was inversely correlated with the levels of circulating miR-499a-3p and none of the other factors remained a significant factor (Supplementary Table S4).
MEF2C Is a Direct Target of Both MiR-135b-5p and MiR-499a-3p. We next sought to identify direct target of miR-135b-5p and miR-499a-3p involved in atherosclerosis. We used the most commonly used bioinformatics prediction tools (TargetScan, PicTar, miRanda and miR database [21][22][23] to identify the target. We limited our search to those targets that are commonly predicted at least by two programs, and also to those that showed species conservation among vertebrates. The potential targets of both miR-135b-5p and miR-499a-3p included MEF2C, FOXN3 (Forkhead Box N3) and ATP8A1 (ATPase, aminophospholipid transporter [APLT], class I, type 8A). On MEF2C 3′ -UTR (4,363-bp length), target positions 2590-2596 was predicted for miR-135b-5p, and target positions 3659-3665 was predicted for miR-499a-3p (Fig. 3A). Next, a fragment of the 3′ -UTR of MEF2C mRNA containing the wild type or mutated putative miR-135b-5p or miR-499a-3p corresponding binding sequence (Fig. 3A) was cloned into a luciferase reporter in a dual reporter construct. The luciferase reporter assay in HEK293 cells showed that the luciferase activity was decreased by 50% when cotransfected with miR-135b-5p and its corresponding pmirGLO-MEF2C-wt (WT), and 30% when cotransfected with miR-499a-3p and its corresponding pmirGLO-MEF2C-wt (WT), whereas no significant reduction in luciferase activity was observed when the cells were cotransfected with miR-NC (Fig. 3B, WT). The construct pmirGLO-MEF2C-mut (MUT) was then used to repeat the luciferase assay experiments in HEK293 cells, which showed that mutating the seed region for miR-135b-5p or miR-499a-3p in the pmirGLO-MEF2C-wt plasmid completely abrogated its regulatory activity (Fig. 3B, MUT). The results demonstrated that miR-135b-5p or miR-499a-3p might suppress MEF2C expression at corresponding binding sites at its 3′ -UTR.
Furthermore, on each FOXN3 and ATP8A1 3′ -UTR, there are also target positions predicted for the two miRNAs (Fig. 3C,E). Then, we cloned the 3′ -UTR fragments of FOXN3 and ATP8A1 containing the putative miR-135b-5p or miR-499a-3p corresponding binding sequence into a luciferase reporter construct, respectively. Indeed, overexpression of miR-135b-5p or miR-499a-3p significantly inhibited the luciferase activity of WT 3′ -UTR of FOXN3 (Fig. 3D), whereas there was no significant reduction in luciferase activity when cotransfected with miR-135b-5p (miR-499a-3p) and its corresponding pmirGLO-ATP8A1-wt (Fig. 4F). It will be interesting to study the regulations by which miR-135b-5p and miR-499a-3p target FOXN3 gene in further studies; however, in the present study, we focused only on the investigation of the regulation of MEF2C for its important role in cell phenotypic modulation associated with cardiovascular systems. Moreover, an immunoblotting assay was conducted to verify the luciferase activity result. When mimics of miR-135b-5p or miR-499a-3p were transfected into HEK293 cells, MEF2C protein levels (but not mRNA) decreased compared to that in the negative control (Fig. 3G,H). Correspondingly, miR-135b-5p inhibitor or miR-499a-3p inhibitor could significantly increase MEF2C protein levels (Fig. 3G,H). Taken together, these results suggested that MEF2C is the valid target of miR-135-5p and miR-499a-3p.

MiR-135b-5p and MiR-499a-3p Promoted Proliferation and Migration in HUVECs By Targeting
MEF2C. Accumulating studies have demonstrated that circulating miRNAs can be actively taken up by cells, thus serving as signaling molecules mediating intercellular communication [24][25][26] . A survey of cell lines revealed that endogenous miR-135b-5p is highly enriched in muscle cells and T cells 27,28 and endogenous miR-499a-3p is enriched in cardiac and skeletal muscle cells 4 , but MEF2C is mainly expressed in endothelial cells in vivo 15 and in vitro 29 . In light of MEF2C roles in cell phenotypic modulation, we
To further determine whether the two miRNAs have an additive effect on endothelial dysfunctions, an in vitro wound closure (scratch assay) was performed to monitor the rate of cell migration into the denuded area of a confluent monolayer. As shown in Fig. 4G,H, miR-135b-5p or miR-499a-3p accelerated the rates of cell migration compared with miR-NC transfected cells. The migration was enhanced in (miR-135b and miR-499a-3p) transfection cells compared to the single miRNA-transfected cells and the migration difference was a result of the impaired expression of MEF2C (Fig. 4G,H). Moreover, ECs migration regulated by the miRNAs was also tested across Transwell filters. As shown in Fig. 4I,J, the migration rate of HUVECs treated with miR-135b-5p or miR-499a-3p was higher than that of miR-NC treated cells. Correspondingly, the migration rate of ECs treated with the two miRNAS was higher than cells treated with single miRNAs, and the increase in cell migration caused by the two miRNAs was offset in (miR-135b-5p + miR-499a-3p + pcDNA3.1-MEF2C) cells. Together, these data indicated that both miR-135b-5p and miR-499a-3p could jointly promote ECs proliferation and migration by directly repressing MEF2C.

MiR-135b-5p and MiR-499a-3p Targeted MEF2C Expression and Enhanced Proliferation and
Migration of VSMCs. In the pathogenesis of atherosclerosis, ECs dysfunction has been implicated as an early step and the following VSMCs proliferation and migration is particularly important associated with atherosclerotic progression 4,30 . As shown in Fig. 5A,B, proliferation rate of VSMCs transfected with miR-135b-5p mimic or miR-499a-3p mimic was significantly increased by approximately 1.9-fold and 2.7-fold, respectively. Particularly, the proliferation rate was enhanced in both miR-135b and miR-499a-3p transfection cells. The in vitro wound closure and transwell assays showed that transfection of miR-135b/-499a mimics significantly increased the migratory ability of VSMCs when compared with control cells (Fig. 5C-F). Furthermore, high-expression of MEF2C restored the proliferation and migration enhancement induced by the two miRNAs. Thus, miR-135b-5p/499a-3p upregulation could stimulate VSMCs proliferation and migration by directly repressing MEF2C. In summary, HUVECs and VSMCs underwent profound phenotypic modulation, adopting more proliferative and migratory phenotype after transfection with miR-135b-5p and miR-499a-3p, and the responses of HUVECs and VSMCs were regulated by the miRNAs-mediated MEF2C suppression.

MEF2C Expression in Human Atherosclerotic Plaques.
Increasing evidences suggest that circulating miRNAs protect against degradation by MVs, apoptotic bodies, or lipids and could be taken up by recipient cells, thus serving as signaling molecules mediating intercellular communication 24,31,32 . Of note, it has been reported that serum miRNAs are more commonly found in MVs, whereas plasma miRNAs are primarily associated with argonaute 2 (Ago2) or high-density lipoprotein (HDL)-associated proteins 32 . Supporting the carrier roles of MVs for the two miRNAs, circulating miR-135b-5p and -499a-3p was shown to be secreted by tumor cells, packaged into MVs and then directly delivered to ECs 16 . We also obtained MVs from serum samples of atherosclerotic CAD patients by sequential centrifugation 33 and found miR-135b-5p and miR-499a-3p were mainly stored in MVs (Supplementary Fig. S3).
Although the stimuli that trigger miRNA secretion are unclear, accumulating evidences showed that human atherosclerotic plaque contains large amounts of MVs released following cell activation or apoptosis [34][35][36] . Moreover, MVs isolated from human atherosclerotic lesions originate from multiple cells, including macrophages, lymphocytes, erythrocytes, and smooth muscle cells 34,35 . Moreover, it has been demonstrated that plaque MVs dock on nonactivated ECs and contribute to the progression and development of atherosclerotic lesions 36,37 . Given that the two miRNAs exist mainly in MVs accumulated in plaque, we hypothesized that miR-135b-5p and miR-499a-3p might be packaged into MVs from atherosclerotic plaques, secreted to circulating systems and taken by ECs and/or VSMCs of vascular Scientific RepoRts | 5:12276 | DOi: 10.1038/srep12276

Figure 5. MiR-135b-5p and miR-499a-3p promote smooth muscle cell proliferation and migration (A)
Photoimages of EdU incorporation assay of VSMCs that were treated with miRNAs or rescue plasmid. Results are representative data from three independent experiments. (B) Proliferation rates of VSMCs that were treated with miRNAs or rescue plasmid. Results are presented as means ± SD of three independent experiments. *P < 0.05; ** P < 0.01. (C) Photoimages of scratch assay of VSMCs that were treated with miRNAs or rescue plasmid. Results are representative data from three independent experiments. (D) Migration rates of VSMCs that were treated with miRNAs or rescue plasmid. Results are presented as means ± SD of three independent experiments. *P < 0.05; ***P < 0.001. (E) Photoimages of transwell assay of VSMCs that were treated with miRNAs or rescue plasmid. Results are representative data from three independent experiments. (F) Migration rates of VSMCs that were treated with miRNAs or rescue plasmid. Results are presented as means ± SD of three independent experiments. *P < 0.05; **P < 0.01. wall with MEF2C repression. To address this issue, we first detected the in situ hybridization of the miRNAs in human atherosclerotic plaques, but the poor hybridization efficiency precluded the analysis of the two miRNAs expression levels. Because we were not able to detect protein expression in lumen, Scientific RepoRts | 5:12276 | DOi: 10.1038/srep12276 immunohistochemical staining was performed to compare MEF2C protein levels in plaque with that in vascular wall (Fig. 6). As shown in Fig. 6A,B, MEF2C protein levels were significantly downregulated in human atherosclerotic plaques compared with the corresponding vascular wall. Taken together, the results indicated that miR-135b-5p and miR-499a-3p might be associated with MVs in plaques, and delivered into ECs and VSMCs to repress MEF2C expression.

Discussion
In the present study, a distinct serum-based miRNA expression pattern in atherosclerotic CAD patients was demonstrated. In particular, 2 miRNAs, miR-135b-5p and miR-499a-3p, were found to be highly expressed in atherosclerotic patients. Given that circulating miRNAs are a potential key feature of atherosclerosis-associated dysfunction of gene regulatory networks 26,33,38 , our study focused on the potential cellular properties regulated by miR-135b-5p and miR-499a-3p. The MEF2C-encoded mRNA contains a 3′ -UTR element that is partially complementary to miR-135b-5p and miR-499a-3p, indicating that the miRNAs might directly target the corresponding sites. Interestingly, the MEF2C-encoded mRNA contains a 3′ -UTR element that is also partially complementary to miR-21 and miR-98, which were downregulated in the miRNA expression profile. It has been demonstrated that the reduction of miR-98 might increase MEF2C protein levels ( Supplementary Fig. S4). It would be interesting to study the regulations by which miR-98 target MEF2C in further studies; however, in the present study, we focused on investigation of the elevated miRNAs for their potential roles in the release of MVs induced by cell activation.
In atherosclerosis, ECs activation triggers platelet aggregation, leukocyte adhesion and VSMCs proliferation 1 . The VSMCs proliferation within atherosclerotic lesions contributes significantly to luminal narrowing and to plaque instability 1,39 . Therefore, proliferation and migration of ECs and VSMCs have been implicated to be critical in the progression of atherosclerotic lesions 1,36 . Our exploration to assess the biological functional roles of the two miRNAs revealed interesting and encouraging results. We showed that both miR-135b-5p and miR-499a-3p directly regulated MEF2C and had additive enhancement effects on proliferation and migration of ECs and VSMCs. MEF2C, as the third member of the MEF2 transcription factor family (MEF 2A, -B, -C, and -D), plays a important role in modulating cell phenotype [13][14][15] . It has been reported that MEF2C is involved in proliferation of VSMCs 40 and migration of human retinal endothelial cells 13 . Here, repression of MEF2C caused by elevation of miR-135b-5p and miR-499a-3p leaded to a considerable increase in proliferation and migration. The presence of another extracellular miRNA, miR-150, supports these findings. Secreted miR-150, which is increased in the plasma of atherosclerotic patients, is taken up by endothelial cells and regulates cellular migration by repressing the c-Myb gene 33 . Thus, abnormally circulating miRNA expression is likely to contribute to the recipient endothelial cell dysfunction involved in the pathogenesis of atherosclerosis. However, circulating miRNAs that are not endogenously expressed in the recipient cell might be especially appropriate for demonstrating that transferred miRNA is essential for directly modulation in the target cell 8 . Therefore, ECs might be a better selection than VSMCs, because ECs expresses minimal endogenous miR-135-5p and miR-499a-3p, and endothelial dysfunction is particularly a driving force in the initiation and development of atherosclerosis 1 .
Next, there are several other significant factors that influence ECs and VSMCs migration in atherosclerosis, such as HDLs and statin treatment. Notably, the atheroprotective effect of HDL on endothelial cell migration has been reported 41 . We then confirmed the dose-dependent effect of mevastatin on migration 42 and demonstrated levels of MEF2C increased with both lower and higher concentrations of mevastatin ( Supplementary Fig. S5). Thus, further investigation of the influencing factors of miR-135-5p and miR-499a-3p and their target genes is needed to delineate the full range of physiological functions of the 2 miRNAs.
Furthermore, increasing evidences suggest that circulating miRNAs are taken up by recipient cells as signaling molecules of gene regulatory networks [24][25][26] . Although the stimuli that trigger miRNA secretion are unclear, it has been shown that human atherosclerotic plaque contains large amounts of MVs released following cell activation or apoptosis [34][35][36] . Several findings further confirm that MVs dock on nonactivated endothelial cells and contribute to the progression and development of atherosclerotic lesions 36,37 . Together with our data, there is a plausible explanation for the causes and consequences of circulating miR-135b-5p and miR-499a-3p in atherosclerotic CAD patients (Fig. 7). MiR-135b-5p and miR-499a-3p might be associated with MVs in atherosclerotic lesion with repression of MEF2C. Some of the MV-associated miRNAs could be secreted into circulating systems. Finally, the MVs might dock on ECs and /or VSMCs and the two miRNAs could be taken by the cells as molecular signals to facilitate atherosclerotic development by enhancing proliferation and migration. However, MVs isolated from human atherosclerotic lesions originate from multiple cells, including macrophages, lymphocytes, erythrocytes, and smooth muscle cells 34,35 . Thus, the cell origins of secreted miR-135b-5p and miR-499a-3p remain unknown because no general mechanism for MVs formation has been demonstrated.
In conclusion, these results demonstrated a serum-based miRNA profile for atherosclerotic CAD patients and indicate circulating miRNAs could regulate target gene expression and recipient cell function. The elevated miR-135b-5p and miR-499a-3p directly targeted MEF2C gene together, and had additive effect on increased proliferation and migration of ECs and VSMCs. These findings may reveal important insights into the role of circulating miRNAs in the pathogenesis of atherosclerosis, and further suggest that miR-135b-5p and miR-499a-3p serve as a novel therapeutic target for atherosclerosis. However, a large number of specific miRNAs might be involved in atherosclerotic disease. Thus, investigation of additional circulating miRNAs and their functional roles in atherosclerosis are necessary to offer insights into the pathophysiology and potential role in targeted therapy for coronary artery disease.

Methods
Study Population. The complete details of the entire study design and procedures involved were in accordance with the Declaration of Helsinki. Informed written consent was obtained from all CAD and control subjects and the study protocol was approved by the Ethics Review Committee for Human Studies of the Shandong University School of Medicine. The methods used were carried out in accordance with approved guidelines and regulations.
Ninety atherosclerotic patients with symptoms of angina pectoris were included in the study after angiographic documentation of CAD. Fifty healthy volunteers without any evidence of CAD, diabetes mellitus, or inflammatory disorders were enrolled from the Shizhong District in Eastern Jinan and served as the control group. Fifteen non-atherogenic cardiovascular patients (3 with cardiogenic shock, 5 with heart failure, and 7 with ventricular arrhythmia) served as another control group in the independent validation of miR-135b-5p and miR-499a-3p expressions. Atherosclerotic plaques were collected from patients undergoing vascular surgery or at autopsy. Serum Collection, RNA Isolation, and MiRNA Profile. Five-milliliter peripheral blood samples were separated into serum fractions by centrifugation at 1000 g for 10 min. Serum RNA was harvested with the TRI Reagent BD (Sigma-Aldrich Co., LLC, St. Louis, MO, USA) and the RNeasy mini kit (Qiagen, Inc., Valencia, CA, USA) according to the manufacturers' instructions. The miRNA microarray profiling was conducted by KangChen Bio-tech, Inc. (Shanghai, China). The miRNA expression profiling from the serum samples was performed using the miRCURY LNA Array (version. 11.0, Exiqon Inc., Woburn, MA, USA) system (Exiqon, Inc., Woburn, MA, USA). RNA samples were labeled using the miRCURY Hy3/Hy5 Power labeling kit (Exiqon Inc., Woburn, MA, USA) and hybridized on the miR-CURY LNA Array station. Scanning was performed with the Axon GenePix 4000B microarray scanner (Molecular Devices, LLC, Sunnyvale, CA, USA). GenePix Pro version 6.0 was used to read the raw data of the images. The intensity of the signal was calculated after background subtraction, and replicated spots in the same image were averaged to obtain the median intensity. The median normalization method was used to obtain normalized data (foreground minus background divided by median). The median represents the 50 th of miRNA intensity, which is >50 in all samples after background correction. The significance of the results was determined using fold change and t tests. The threshold value for significance used to define upregulation or downregulation of miRNAs was a fold change >2 and p-value <0.05. The miRNAs selected for investigation in this study were further filtered on the basis of expression levels and previously published data.
HEK293 cells were cotransfected with 20 ng luciferase reporter plasmid and the indicated miRNAs (final concentration, 25 nmol/L) using Lipofectamine 2000 (Invitrogen, Carlsbad, CA, USA) according to the manufacturer's instructions.
Antibodies against the following proteins were purchased: anti-MEF2C (SC13266, polyclonal antibody produced in goat), anti-β -actin (SC-8432, monoclonal antibody produced in mouse), anti-α -tubulin (SC-53646, monoclonal antibody produced in mouse) (Santa Cruz Biotechnology, Santa Cruz, CA) and anti-MEF2C for IHC-P (ab191428, monoclonal antibody produced in rabbit) (Abcam, Cambridge, MA). Immunoblotting and immunofluorescence staining were performed as described previously 43 . For immunohistochemisty examination of the tissue samples, hematoxylin and eosin (H&E) staining was performed according to standard procedures. Sections were sequentially blocked with 3% H 2 O 2 for 10 min, 10% normal serum matching the host of the secondary antibody for 60 min at room temperature. Primary antibody against human MEF2C (ab191428 from Abcam, Cambridge, MA) was used for immunostaining. Secondary antibody (goat anti-rabbit) was used at 1:100 dilutions. Sections were developed by DAB substrate and counterstained with hematoxylin for miscroscopic visualization.

EdU Incorporation and Migration Assays. EdU (Cell-Light TM EdU Cell Proliferation Detection
Kit, Guangzhou RiboBio, China) was added at 100 μ M and the cells were cultured for an additional 2 h. After the removal of EdU-containing medium, the cells were fixed with 4% paraformaldehyde at room temperature for 30 min, washed with glycine (2 mg/ml) for 5 min in a shaker, treated with 0.2% Trion X-100 for 10 min, washed with PBS twice. Click reaction buffer (Tris-HCl, pH 8.5, 100 mM; CuSO 4 , 1 mM; Apollo 550 fluorescent azide, 100 μ M; ascorbic acid, 100 mM) was then added. After 20 min, the cells were washed with 0.5% Triton X-100 for three times, stained with 4′ ,6-diamidino-2-phenylindole (DAPI) for 10 min at room temperature, washed with 0.5% Triton X-100 for five times, and finally, immersed in 150 μ L PBS and examined under a fluorescence microscope.
The cell migration ability was tested in Transwell Permeable Supports (Corning Incorporated, Lowell, MA, USA). After infection with the indicated miRNAs, the cells were suspended in serum-free Dulbecco's modified eagle's medium (DMEM) at a concentration of 10 5 /mL, and 200 μ L suspensions were placed in the upper chamber of the Transwell. The chamber was then transferred to a well containing 500 μ L media containing 10% fetal bovine serum. Cells were incubated for 5 h at 37 °C. Cells in the top well were removed by wiping the top of the membrane with cotton swabs. The membranes were then stained and the remaining cells were counted. The contents of 4 high-powered fields were counted for each membrane.
Statistical Analysis. All continuous variables are expressed as means ± standard deviation (SD), unless stated otherwise. Categorical variables were compared using the Chi squared (x 2 ) test, and continuous variables were compared using the Student's t-test. All tests were performed 2-sided and a significance level of P < 0.05 was considered to indicate statistical significance (*P < 0.05, **P < 0.01, ***P < 0.001 and # P < 0.0001). GraphPad Prism 5.0 (GraphPad software Inc., San Diego, CA, USA) was used for all statistical analyses. All photo images of Western Blotting, Edu incorporation, migration assays and immunohistochemical staining are representative of at least three independent experiments.