A total of 40 glioma tissues and matched adjacent tissues were used in the present study. The diagnosis of glioma was performed by surgeons and pathologists. The exclusion criteria were as follows: Patients with recurrent glioma; patients who received immunotherapy, radiotherapy or chemotherapy prior to surgery; and other severe organ or autoimmune diseases. The protocol of the present study was approved by The Human Ethics Committee of The First Affiliated Hospital of China Medical University. All the experiments were performed in accordance with the Declaration of Helsinki and subsequent updates. Written informed consent was obtained from each patient.

A total of 500 ng of RNA was subjected to Custom RT² Profiler PCR Array (cat. no. 330171; Qiagen China Co., Ltd.). According to the list of gene names, symbols, UniGene and Genbank IDs, primers were synthesized by the manufacturer (GenePharma Co., Ltd.). Three biological replicates (experimental replicate) were used per group, and each was measured in duplicate (technical replicates). Comparisons for significance were performed using a Student's t-test.

The normal human astrocyte cell line CC-2565 was obtained from Lonza Group, Ltd. U87 MG cells, a glioblastoma cell line of unknown origin, were obtained from the American Type Culture Collection (cat. no. HTB-14). The human glioma cell line U251 (cat. no. KCB 200965YJ) was purchased from Kunming Institute of Zoology, Chinese Academy of Sciences. Normal human astrocytes (NHA) were cultured in Astrocyte Growth Medium (Lonza Group Ltd.) supplemented with 0.03% FBS (Thermo Fisher Scientific, Inc.). U87 and U251 cells were cultured in DMEM (Thermo Fisher Scientific, Inc.) supplemented with 10% FBS. Cells were maintained at 37°C in an incubator with a humidified atmosphere with 5% CO₂ and 95% air. All the reagents were purchased from Gibco (Thermo Fisher Scientific, Inc.). miR-489 mimics, miR-489 inhibitor (miR-489 inh) and negative control RNA were purchased from GenePharma Co., Ltd. The oligonucleotide sequences were as follows: miR-489 mimic, 5′-GUGACAUCACAUAUACGGCAGC-3′; miR-489 inh, 5′-GCTGCCGUAUAUGUGAUGUCAC-3′; negative control (control), 5′-UUGUCCGAACGUGUCACGUTT-3′. Cells were transfected with Lipofectamine^® 3000 (Invitrogen; Thermo Fisher Scientific, Inc.), according to the manufacturer's protocol. pDONR223 PAK5 (Addgene plasmid no. 60532) was a gift from Dr John Brognard (Professor of Signalling Networks in Cancer Group, Cancer Research UK, Paterson Institute for Cancer Research, University of Manchester) (14). pDONR223 (Addgene plasmid no. 12536017) was obtained from Thermo Fisher Scientific, Inc.

Wild-type (wt) or mutant (mut) 3′-untranslated region (3′-UTR) of the PAK5 promoter was amplified and cloned into a pGL474-vector, and the 3′-UTR of PAK5 and corresponding miRNA vectors were co-transfected into U87 and U251 cells, respectively. The primer sequences were as follows: Wt-PAK5 forward, 5′-CTTTGATGTCATGTAGCCATTG-3′ and reverse, 5′-CAAAGTTCCTAAAGAATCATTGGT-3′; and mut-PAK5 forward, 5′-CTTTGATGTCCTAATGCCATTG-3′ and reverse, 5′-CAAAGTTCCTAATTAGCATTGGT-3′. At 24 h after co-transfection, luciferase activity was measured using a Dual-Luciferase^® Reporter Assay System (Promega Corporation).

Total RNA was extracted using TRIzol^® (Ambion; Thermo Fisher Scientific, Inc.) according to the manufacturer's protocol. Reverse transcription of RNA to cDNA was performed using a reverse transcription system (Promega Corporation). PCR amplification of miR-489 and PAK5 mRNA was performed using an Mx3000P real-time PCR system (Applied Biosystems; Thermo Fisher Scientific, Inc.) according to the manufacturer's protocol. U6 and GAPDH were used as the reference genes to normalize the expression of miR-489 and PAK5 mRNA, respectively. The primer sequences for the detection of mRNA expression were as follows: PAK5: Forward, 5′-GTCTCCTCTGACTTCAACAGC-3′ and reverse, 5′-ACCACCCTGTTGCTGTAGCCAA-3′; miR-489: Forward, 5′-GTGACATCACATATACGG-3′ and reverse, 5′-GAACATGTCTGCGTATCTC-3′; U6: Forward, 5′-CTCGCTTCGGCAGCACA-3′ and reverse, 5′-AACGCTTCACGAATTTGCGT-3′; c-Myc: Forward, 5′-CCTGGTGCTCCATGAGGAGAC-3′ and reverse, 5′-CAGACTCTGACCTTTTGCCAGG-3′; cyclin D1: Forward, 5′-ATGTTCGTGGCCTCTAAGATG-3′ and reverse, 5′-CAGGTTCCACTTGAGCTTGTTC-3′; Bcl-2: Forward, 5′-ATCGCCCTGTGGATGACTGAG-3′ and reverse, 5′-CCAGGAGAAATCAAACAGAGG-3′; Bax: Forward, 5′-TCAGGATGCGTCCACCAAGAA-3′ and reverse, 5′-TGTGTCCACGGCGGCAATCATC-3′; MMP2: Forward, 5′-AGCGAGTGGATGCCGCCTTTA-3′ and reverse, 5′-CATTCCAGGCATCTGCGATGAG-3′; GAPDH: Forward, 5′-GTCTCCTCTGACTTCAACAGC-3′ and reverse, 5′-ACCACCCTGTTGCTGTAGCCAA-3′.

Total proteins from 1×10⁶ cells were extracted using lysis buffer (Thermo Fisher Scientific, Inc.) and protein concentration was determined using a bicinchoninic acid assay (Pierce; Thermo Fisher Scientific, Inc.). A total of 40 µg protein was resolved on a 10% gel using SDS-PAGE (Thermo Fisher Scientific, Inc.). Subsequently, proteins were transferred to a PVDF membrane (EMD Millipore). The membranes were blocked in 5% skimmed milk in TBS-Tween for 1 h at room temperature, and probed with the following primary antibodies: Rabbit polyclonal antibodies PAK5 (cat. no. ab110069; Abcam), p-S338-RAF1 (cat. no. ab51042; Abcam), RAF1 (cat. no. ab137435; Abcam) and matrix metalloproteinase 2 (MMP2; cat. no. ab37150; Abcam); rabbit monoclonal antibodies c-Myc (cat. no. ab32072; Abcam), cyclin D1 (cat. no. ab134175; Abcam), Bcl-2 (cat. no. ab32124; Abcam) and Bax (cat. no. ab32503; Abcam) at 4°C overnight, after which time the membranes were incubated with horseradish peroxidase-conjugated secondary antibodies (cat. nos. ab222772 and ab222759; Abcam). GAPDH was used as the loading control (mouse monoclonal anti-GAPDH, KC-5G4, obtained from Zhejiang Kangchen Biotech Co., Ltd.). The dilution of primary antibodies was 1:1,000 and the dilution of secondary antibodies was 1:5,000. Signals were visualized using enhanced chemiluminescence (Pierce; Thermo Fisher Scientific, Inc.).

Glioma cell lines were transfected with negative control (NC), miR-489 mimic or miR-489 inhibitor using Lipofectamine^® 3000. After 24 h, 1×10⁴ cells were plated in medium containing 10% FBS in 96-well plates. Cell proliferation was measured using a modified MTT assay. Briefly, MTT solution in PBS was added to each well. After a 4 h incubation at 37°C, the medium was replaced with DMSO. After a 15 min incubation at 37°C, the optical density at 490 nm was measured using a microplate reader (Bio-Rad Laboratories, Inc.).

Glioma cells transfected with miR-489 mimic or inh were plated at a density of 2×10³ cells/well into 96-well plates at 37°C in an incubator with 5% CO₂. Cell proliferation was assessed every 24 h using a CCK-8 assay (Sigma-Aldrich; Merck KGaA) according to the manufacturer's protocol. For each sample at each time-point, six wells were analyzed and the experiment was repeated three times.

Migration and invasion assays were performed using modified Boyden chambers with Corning^® Transwell^® polycarbonate membrane cell culture inserts (Sigma-Aldrich; Merck KGaA). The invasion assay was performed using BioCoat Matrigel invasion chambers (BD Biosciences). A total of 1×10⁵ cells in 100 µl serum-free DMEM supplemented with 0.1% BSA were placed in the upper part of each chamber, and the lower compartments were filled with 600 µl DMEM supplemented with 10% FBS. The cells were incubated for 18 h at 37°C, and the cells that had invaded or migrated through the membrane were stained with 0.5% crystal violet solution and counted.

Glioma cells were transfected with NC, miR-489 mimic, or miR-489 inh for 24 h. After incubation, the cells were fixed with 4% polyoxymethylene, washed twice with PBS, treated with 10 µg/ml Hoechst 33258 (Molecular Probes; Thermo Fisher Scientific, Inc.) for 5 min at room temperature, and subsequently washed with PBS three times, for 10 min per wash. Cells were observed using a fluorescence microscope (Leica Microsystems GmbH).

The results are presented as the mean ± standard deviation of at least three experimental repeats. Comparisons among different groups were performed using a two-way ANOVA. Dunnett's multiple comparisons test was performed to compare the mean of each group with the mean of the control group. Paired data were compared using a Student's t-test. P<0.05 was considered to indicate a statistically significant difference.

Recently, numerous miRNAs have been demonstrated to be aberrantly expressed in glioma tissues compared with adjacent tissues. In the present study, it was demonstrated that the difference in miR-489 expression between glioma and adjacent tissues was the largest amongst the various miRNAs measured (Table I). Using the miRDB database (9,15), miR-489 was identified as a candidate miRNA targeting PAK5. The relative RNA expression levels of miR-489 and PAK5 were determined in 40 samples from patients with glioma (Fig. 1). The results demonstrated that the expression of miR-489 was decreased in cancer tissues (0.583±0.080) compared with that in adjacent tissues (0.774±0.110), whereas the expression of PAK5 increased in cancer tissues (0.732±0.072) compared with that in adjacent tissues (0.572±0.070) (Fig. 1A and C). To determine whether there was an association between miR-489 expression and clinical outcome in patients with glioma, the 40 cases were divided into two groups according to miR-489 expression, the high miR-489 (n=14) and low miR-489 (n=26) groups. As shown in Fig. 1B, patients with higher miR-489 expression levels had longer survival. In addition, as shown in Fig. 1D, patients with lower PAK5 expression levels (n=20) had longer survival.

To further determine the association between miR-489 and PAK5 in glioma, a Spearman's rank correlation analysis was performed (Fig. 1E). miR-489 and PAK5 expression were found to be negatively correlated. These results suggest that miR-489 expression is negatively correlated with PAK5 expression and may be used to predict the clinical outcome in patients with glioma.

miRNA target prediction data were analyzed using the miRDB database (http://mirdb.org/index.html), and it was predicted that miR-489 targeted the 3′-UTR of PAK5 (Fig. 2A). A dual luciferase reporter assay was performed in U87 and U251 cells to demonstrate the putative binding sites. Glioma cells were transfected with miR-489 mimic or miR-489 inh, combined with wt-PAK5 promoter or mut-PAK5 promoter.

As shown in Fig. 2B, the relative activity of the wt-PAK5 promoter was reduced in cells transfected with miR-489 mimic in U87 cells. Ectopic expression of miR-489 inh restored the relative activity of wt-PAK5 compared with the control. The relative activity of the mut-PAK5 promoter was not altered in cells transfected with either miR-489 mimic or miR-489 inh. Similar results were observed in the U251 cells (Fig. 2C).

To further investigate expression of miR-489 and PAK5 in glioma cells, RT-qPCR assays were performed. As shown in Fig. 2D, miR-489 expression was lower in U87 and U251 cells compared with NHA, whereas PAK5 expression was higher in U87 and U251 cells compared with NHA (Fig. 2E). Additionally, western blotting was performed to examine changes in PAK5 protein expression following transfection with miR-489 mimic or miR-489 inh. PAK5 expression increased following transfection of miR-489 inh, and decreased significantly following transfection with miR-489 mimic in both U87 and U251 cells (Fig. 2F). Similar changes were observed in the relative RNA expression levels (Fig. 2G and H). These results suggest that miR-489 targets the 3′-UTR of PAK5 and decreases the expression of PAK5 in glioma cells.

Recent studies have demonstrated that miR-489 is a tumor suppressor targeting multiple oncogenes in a number of different types of cancer. miR-489 mimic or miR-489 inh was transfected into U87 and U251 cells to determine the effects of miR-489 on glioma cells in vitro. MTT assays were performed to determine the effects of miR-489 on cell viability. The results demonstrated that miR-489 mimic decreased cell viability, whereas miR-489 inh increased the viability of both U87 and U251 cells (Fig. 3A and B). The differences in cell viability after 36 and 48 h were significant. A CCK-8 assay was used to evaluate the effect of miR-489 on cell proliferation. As shown in Fig. 3C, proliferation was increased in cells transfected with miR-489 mimic, whereas it was decreased in cells transfected with miR-489 inh. Similar results were observed in both U251 and U87 cells (Fig. 3D).

As PAK5 phosphorylates RAF1, but not B-raf, at Ser338, thereby activating the RAF1/ERK/MAPK signaling pathway (16), the expression of phospho-Ser338 RAF1 (p-RAF1), RAF1, and the effectors c-Myc and cyclin D1 was determined in U87 and U251 cells using western blotting. As shown in Fig. 3E, following Ser338 phosphorylation, c-Myc and cyclin D1 decreased in U87 cells, whereas PAK5 expression was decreased following transfection with miR-489 mimic, although total RAF1 remained unchanged. By contrast, p-RAF1, c-Myc and cyclin D1 expression decreased following transfection with miR-489 inh. Similar results were observed in U251 cells (Fig. 3F). To confirm the changes in the protein expression levels, RT-qPCR assays were performed to evaluate changes in the relative RNA levels. As shown in Fig. 3G, transfection of miR-489 mimic decreased PAK5, c-Myc and cyclin D1 RNA expression levels significantly, but did not affect RAF1 levels in U87 cells. Transfection with miR-489 inh significantly increased PAK5, c-Myc and cyclin D1 levels (Fig. 3H). Similar results were observed in the U251 cells (data not shown). These results suggest that miR-489 decreases cell viability by inactivating PAK5/RAF1-mediated pathways.

As previously described, the pro-oncogene RAF1 and apoptosis-associated proteins Bax/Bcl-2 were involved in the regulation of apoptosis in a variety of different types of cancer (17,18). U87 and U251 cells were transfected with control, miR-489 mimic or miR-489 inh. Hoechst 33258 staining was performed to determine whether miR-489 increased the apoptosis of glioma cells. Increased condensation of chromatin was observed in glioma cells following transfection with the miR-489 mimic (Fig. 4A and C), whereas the opposite was observed in cells transfected with miR-489 inh (Fig. 4B and D). Furthermore, apoptosis was determined by flow cytometry 24 h after transfection, and the results demonstrated that apoptosis in cells transfected with miR-489 mimic was significantly increased, whereas apoptosis in cells transfected with miR-489 inh was significantly decreased compared with control.

Western blotting was performed to assess changes in expression of the pro-apoptotic protein Bax, and pro-survival protein Bcl-2. As shown in Fig. 4E and H, Bax expression decreased significantly, whereas Bcl-2 expression increased significantly following transfection with miR-489 mimic in both U87 and U251 cells. Transfection of miR-489 inh significantly upregulated Bax expression and downregulated Bcl-2 expression. Furthermore, RT-qPCR analysis demonstrated that transfection of miR-489 mimic increased Bax expression and decreased Bcl-2 expression (Fig. 4F), whereas the opposite results were observed in cells transfected with miR-489 inh in U87 cells (Fig. 4G). The effects of miR-489 on the expression of Bax and Bcl-2 in U251 cells were consistent with those in U87 cells (Fig. 4I and J). These results suggest that miR-489 induced apoptosis in glioma cells by inhibiting the PAK5/RAF1/Bax/Bcl-2 axis.

To investigate the effects of miR-489 on migration and invasion, Transwell assays with or without Matrigel^® were performed. As shown in Fig. 5A and B, miR-489 mimic significantly decreased migration and invasion, whereas miR-489 inh significantly increased migration and invasion in both U87 and U251 cells. The cell counts of invading or migrating cells are presented in Table II. To further verify the results, western blotting and RT-qPCR analyses were performed in U87 and U251 cells. As shown in Fig. 5C-E, MMP2 expression increased significantly following transfection with miR-489 mimic, whereas the opposite was observed in cells transfected with miR-489 inh. Consistently with the results of RNA expression, the protein levels of MMP2 were decreased in cells transfected with miR-489 mimic, whereas they were significantly increased in cells transfected with miR-489 inh. Similar results were observed in U251 cells (Fig. 5F-H).

These results suggest that miR-489 decreased cell migration and invasion by attenuating the PAK5-MMP2 signaling pathway.

U251 cells were transfected with control, miR-489 mimic or a combination of miR-489 mimic and PAK5 for 24 h. MTT assays, flow cytometry and Transwell invasion assays were performed to assess cell viability, apoptosis and invasiveness, respectively. The results demonstrated that overexpression of PAK5 reversed the effects of miR-489 on cell growth (Fig. 6A), apoptosis (Fig. 6B) and invasion (Fig. 6C). Furthermore, western blot analysis demonstrated that the overexpression of PAK5 attenuated the miR-489-induced dephosphorylation of RAF1 (Fig. 6D). Additionally, RT-qPCR analysis confirmed that overexpression of miR-489 decreased the expression of c-Myc, cyclin D1, Bcl-2 and MMP2, whereas miR-489 enhanced Bax expression, which was accompanied by a reduction in PAK5 expression (Fig. 6E). Overexpression of PAK5 enhanced the expression of c-Myc, cyclin D1, Bcl-2 and MMP2, and reduced the expression of Bax. These results suggest that miR-489 targeted PAK5 and downregulated PAK5/RAF1-mediated signaling.