Tissue specimens from 76 GBM and corresponding normal brain tissues were obtained from the Guangzhou Women and Children's Medical Center and Sun Yat-Sen Memorial Hospital between January 2013 and October 2015. Each sample was frozen in liquid nitrogen within 2 h of extraction. The medical history of each patient was recorded as shown in Table I. GBM diagnosis was made according to the revised WHO classification system (22). Patients who had received treatment, including chemotherapy and radiation therapy, before surgery, were excluded. All experiments were performed in accordance with the guidelines approved by the Ethics Committee of Sun YatSen Memorial Hospital, and informed consent was obtained from each patient.

The human glial HEB and glioblastoma cell lines U87MG, SHG-44, U251, and A172 were maintained in Dulbecco's modified Eagle's medium (DMEM) with low glucose, supplemented with 10% fetal bovine serum (Gibco, MD, USA). Cells were incubated at 37°C with 5% CO₂. Control RNA mimics and miR-205 mimics were obtained from Ribobio (Guangzhou, China), and transfected into cells at a working concentration of 50 nM using Lipofectamine 2000 reagent (Invitrogen, Carlsbad, CA, USA) according to the manufacturer's instructions.

The full length ZEB1 gene open reading frame (ORF) was amplified via PCR and cloned into pcDNA-3.1 (Invitrogen) to generate a pcDNA-3.1+ZEB1 construct (hereinafter, ZEB1). The empty pcDNA-3.1 served as the control (hereinafter, vector). SHG-44 and A172 cells were transfected with the miR-205 mimic (50 nM), in 6-well plates, followed by co-transfection with 2.0 µg of either pcDNA-3.1+ZEB1 or control vector for 48 h.

Total RNA was extracted using TRIzol reagent (Invitrogen) according to the manufacturer's protocol, and reverse-transcription was conducted using PrimeScript™ RT-PCR kit (Takara, Otsu, Shiga, Japan). qRT-PCR was performed to quantify expression levels using the SYBR Green PCR Master mix (Applied Biosystems, Foster City, CA, USA) according to the manufacturer's instructions. qRT-PCR was performed at 94°C for 2 min followed by 40 cycles at 94°C for 10 sec, 60°C for 1 min, and 30 sec at 72°C on an ABI PRISM 7500 Real-Time PCR system (Applied Biosystems). Expression of mature miR-205 was assayed using the Bulge-Loop™ miRNA qRT-PCR Primer Set and the miRNA qRT-PCR Control Primer Set (RiboBio, Guangdong, China). GAPDH and U6 were used as internal controls. All sequences used are shown in Table II. Each reaction was performed in triplicate. Relative quantification of gene expression levels was expressed as fold change, normalized against internal controls, using the ΔΔCq method. Each sample was detected in triplicate.

Total protein was extracted from cells using 1% RIPA lysis buffer (Beyotime, Jiangsu, China), and the BCA method was used for protein quantitation. Equal concentrations of protein were separated by SDS-PAGE, transferred onto PVDF membranes, and probed with ZEB1, mTOR, phospho-mTOR (Abcam, MA, USA), Akt, phospho-Akt, E-cadherin, N-cadherin, vimentin (Cell Signaling Technology, Beverly, MA, USA) and GAPDH (Abcam) antibodies overnight at 4°C. The membranes were incubated with horseradish peroxidase (HRP)-conjugated secondary antibodies (Abcam) for 2 h. Immunocomplexes were visualized using the ECL detection reagent (Beyotime) and the intensities of the signals were quantified using ImageJ software 6.0.

Approximately 10⁶ cells were seeded into each well of 6-well plates. A cell scratch spatula was used to scratch the cell after transfection. Plates were washed using warmed phosphate-buffered saline three times to remove cellular debris, and the cells were incubated at 37°C for 48 h, examined and photographed under a microscope (Olympus Corp., Tokyo, Japan) 0 and 48 h after wounding. Each cell condition was assayed in triplicate.

Transwell chambers (8-µm pore size; Millipore, Billerica, MA, USA) coated with Matrigel (BD Biosciences, San Jose, CA, USA) were used for the invasion assays. Following trans-fection, cells were seeded onto the upper wells, and 500 µl DMEM containing 10% fetal bovine serum (FBS) was used as a chemoattractant in the lower chambers. The cells were allowed to migrate for 24 h, the Matrigel and the cells in the upper surface were removed, and the cells that had migrated to the lower surface were fixed in 4% paraformaldehyde and stained with 0.1% crystal violet (Sigma, St. Louis, MO, USA). The number of invaded cells were counted under a light microscope (Axiovert 200 inverted microscope; Zeiss, Germany) in five fields, and presented as the average number of cells per field of view.

Potential targets of miR-205 were predicted using miRbase, and finally the putative complementary sequence of miR-205 was identified in the 3′UTR of ZEB-1 mRNA. Luciferase reporter vectors were constructed using the full 3′UTR of ZEB-1 and inserted into the pEZX-MT01 vector (GeneCopia, Labomics S.A, Nivelles, Belgium). Empty vectors were used as a negative control. Renilla luciferase, encoded by the same vector, served as an internal control. Each luciferase reporter construct, including the Luc+miR205, Luc+ZEB1 3′UTR, and negative control vectors, was co-transfected into SHG-44 cells and A172 cells using Lipofectamine 2000 (Invitrogen). After 24 h of incubation, the cells were re-seeded in 96-well plates and then Firefly and Renilla luciferase activities were determined using Luc-Pair™ miR Luciferase assay kits (GeneCopia) according to the manufacturer's instructions. All transfection experiments were performed in triplicate and repeated three times.

To analyze immuno-fluorescence, SHG-44 and A172 cells were transfected with miRNA-NC or miRNA-205 mimics for 48 h and cultured on circular coverslips (BD Biosciences) in 6-well plates, then washed and fixed with 4% paraformaldehyde for 15 min. The cells were then permeabilized with 0.1% Triton X for 5 min and blocked for 30 min with 3% bovine serum albumin (BSA)-PBS at room temperature. The cells were incubated overnight at 4°C with anti-ZEB1 primary antibodies, then incubated with a fluorescent secondary antibody for 1 h at room temperature. To visualize the nuclei, the wells were stained with DAPI (Sigma) for 5 min. Fluorescent images (at ×200 magnification) were acquired using a fluorescence inverted microscope (Olympus) and images were processed with Image-ProPlus 6.0 (Media Cybernetics, USA).

Data were expressed as the mean ± standard deviation (SD). Student's t-test was used to compare differences between two groups. One-way analysis of variance (ANOVA) was used to compare the differences between more than two groups, and significance was determined using the LSD t-test. Pearson correlation was used to analyze the relationship between miR-205 and ZEB1 expression. A P-value of <0.05 was considered statistically significant. All statistical analyses were performed using SPSS 20.0 (SPSS, Inc., Chicago, IL, USA), and GraphPad Prism 6.0 (GraphPad Software, Inc., CA, USA) was used to generate graphs.

Previous studies have shown miR-205 expression is reduced in GBM tissues and cell lines (14,15). Therefore, we investigated miR-205 expression, via qRT-PCR, from surgically excised specimens. Our results showed that miR-205 was significantly downregulated in GBM tissues, as compared to the corresponding non-neoplastic tissues (Fig. 1A). Furthermore, we found miR-205 expression was also substantially decreased in several glioma cell lines, as compared to human glial HEB cells (Fig. 1B).

To determine the downstream targets of miR-205 and its role in GBM, we queried candidate target genes from the human microRNA database, and found ZEB1 was a putative target of miR-205. Therefore, we decided to investigate ZEB1 expression in GBM tissues and glioma cell lines. As shown in Fig. 2A, the levels of ZEB1 in GMB tissues were lower than those in the non-tumor tissues. A subsequent correlation analysis found ZEB1 was negatively correlated with miR-205 expression (Fig. 2B). Our qRT-PCR results showed that miR-205 expression in SHG-44 and A172 cells increased more than 130-fold in the miR-205 mimics group compared to miR-NC group (Fig. 3A and B). Fig. 3C shows the putative position of the miR-205 target site in the 3′UTR of ZEB1 mRNA. To confirm these findings, we performed a luciferase assay, and found that in SHG-44 and A172 cells overexpression of miR-205 suppressed luciferase activity, while transfection of a scrambled sequence had no effect (Fig. 3D and E). These results suggest that miR-205 regulates ZEB1 expression by targeting its 3′UTR.

In order to further elucidate the molecular mechanisms of miR-205 in GBM metastasis, we examined the downstream signaling pathways activated by ZEB1. We found ZEB1 protein expression in SHG-44 and A172 cells transfected with the miR-205 mimics were significantly inhibited, as detected by western blotting (Fig. 4A–D). Along with the decrease in ZEB1 levels, overexpression of miR-205 also inhibited p-Akt and p-mTOR expression levels, but had no effect on total Akt and mTOR expression levels (Fig. 4A–D). These results indicate that miR-205 regulates the Akt/mTOR signaling pathway by targeting ZEB1. We also performed immunofluorescent staining on SHG-44 and A172 cells transfected with miR-205 mimics or miR-NC mimics. As shown in Fig. 4E, the ZEB1 protein is expressed in the cytoplasm of the SHG-44 and A172 cells, and Fig. 4F shows its level is reduced in the miR-205 mimics group as compared with the miR-NC group, which corresponds with our western blotting results.

We also tested the effects of miR-205 on the migration and invasion properties of glioma cells. Our wound healing assay showed that the cell migration was suppressed when miR-205 was overexpressed in both SHG-44 and A172 cells (Fig. 5A and B). Furthermore, our Transwell invasion assay indicated that ectopic expression of miR-205 significantly inhibited the invasive response of both SHG-44 and A172 cells (Fig. 5C and D). To determine the regulatory effect of miR-205 on EMT, we performed qRT-PCR and western blot analyses. The E-cadherin mRNA expression level was markedly increased in the miR-205 mimics group, while N-cadherin and vimentin were decreased in both SHG-44 and A172 cells (Fig. 6A and B). Consistent with the qRT-PCR results, E-cadherin, N-cadherin, and vimentin protein expression showed similar changes (Fig. 6C and D).

We performed a rescue assay to investigate if ZEB1 is involved in the miR-205-mediated regulation of glioma cells (23–25). Following transfection, the mRNA and protein levels of ZEB1 increased in glioma cells transfected with pcDNA-3.1+ZEB1 (Fig. 7). We also found an increase in the migratory ability of glioma cells transfected with the ZEB1 construct following treatment with the miR-205 mimics (Fig. 8A and B). In addition, upregulation of ZEB1 partially reversed the effects of miR-205 on cell invasion ability (Fig. 8C and D). Overexpression of ZEB1 inhibited E-cadherin expression and induced the expression of N-cadherin and vimentin as detected by qRT-PCR (Fig. 9A and B) and western blotting (Fig. 9C and D). These data indicate that miR-205 inhibits cell migration and invasion, and reverses EMT by downregulating ZEB1 in GBM cells.