MicroRNA-566 activates EGFR signaling and its inhibition sensitizes glioblastoma cells to nimotuzumab

We previously demonstrated that the expression of five microRNAs (miR-181d, miR-518b, miR-524-5p, miR-566 and miR-1227) were correlated with the survival of glioblastoma patients [12]. Previous studies have clarified the functions of the miR-181 family, miR-518b, miR-524-5p and miR-1227. However, the function of miR-566 has not been reported. To address the function of miR-566, we first examined the expression of miR-566 in a panel of five glioma cell lines and normal astrocytes by qPCR. The results showed that miR-566 was overexpressed in all the glioma cell lines, especially in U87 and LN229 cells. The expression level of miR-566 was 5.1 fold in U87 and 4.7 fold in LN229 cells than miR-566 in control astrocytes (Figure 1A). We selected these two cell lines for further functional analysis and examined the expression of EGFR at both the mRNA and protein levels. Interestingly, the same expression profile was detected in five glioma cell lines, including normal astrocytes (Figure 1B). Therefore, we hypothesized that the EGFR pathway maybe regulated by miR-566. To confirm this, a lentivirus containing the miR-566 inhibitor segment was constructed to down-regulate the expression of miR-566. qPCR analysis indicated that the lentivirus could successfully reduce miR-566 expression by 79% and 75% in U87 and LN229 cells, respectively (Figure 1C). Finally, EGFR and Akt expression were detected both at the mRNA and protein levels, which revealed that the miR-566 inhibitor deactivated EGFR/Akt signaling (Figure 1D and E).

After demonstrating that miR-566 deactivates the EGFR/Akt pathway, we examined miR-566’s function in glioma cell proliferation, invasion, apoptosis and cell cycle distribution. We investigated the effects of miR-566 inhibition on the growth and viability of U87 and LN229 glioma cells. Lenti-AS-566 significantly suppressed the colony forming abilities of these cells (to 47% and 39%, respectively; Figure 2A) compared to lenti-NC treated cells. To examine the potential role of miR-566 in tumorigenesis, we infected U87 cells with lenti-AS-566 or lenti-NC and used these cells in an intracranial model, and tumor size was assessed weekly by a luciferase bioluminescence imaging system. Notably, the inhibition of miR-566 expression in U87 cells resulted in delayed tumor formation and a dramatic reduction in tumor size compared to the lenti-NC group (Figure 2B). This result demonstrated that miR-566 inhibits the tumorigenicity of glioma cells both in vitro and in vivo. We then investigated how miR-566 affects the invasive and apoptotic behaviors of glioma cells. U87 and LN229 cells were infected with lenti-AS-566, and 48 h after infection, a cell invasion assay was performed. Significantly increased G₁ cell cycle arrest was observed in lenti-AS-566-infected cells (Figure 2C). We also found a significant decrease in cell invasion (Figure 2D) compared to lenti-NC-treated cells. Apoptosis and cell cycle analysis based on flow cytometry were performed with infected glioma cells (U87 and LN229). A significant increase in the number of apoptotic cells in lenti-AS-566-infected cells (U87 and LN229) was also detected (Figure 2E). miR-566 inhibitor had no effect on the apoptosis of normal astrocytes (Additional file 1: Figure S1).

Having confirmed that a miR-566 inhibitor could deactivate the EGFR pathway and inhibit the proliferative and invasive behavior of glioma cells, we then demonstrated whether the functions of miR-566 were mainly through the EGFR pathway. U87 glioma cells were first infected with lenti-AS-566 and then EGF was introduced to activate EGFR signaling. Results showed that EGF partially reversed the effects of lenti-AS-566 (Additional file 1: Figure S2).

We have clarified that a miR-566 inhibitor blocks proliferation and invasion and accelerates apoptosis in glioma cells partially through the EGFR pathway. We then further confirmed the target of miR-566. VHL was shown to be involved in the tumorigenesis of glioma [9, 21], and we found that the 3’ UTR of VHL contained 2 potential complimentary binding sites for miR-566. To validate this, we performed luciferase reporter assays using 3’ UTR sequence fragments of the VHL transcript containing the two predicted binding sites for miR-566 (Figure 3A). Western blot analysis demonstrated that VHL protein expression was significantly increased in lenti-AS-566-infected glioma cells (Figure 3B). Transient cotransfection of human U87 and LN229 cells with lenti-AS-566 and a wild-type VHL 3’ UTR plasmid led to a significant increase in luciferase reporter activity compared to infection with lenti-NC (Figure 3C). ‘Rescue’-experiments were performed by introducing a VHL expression plasmid lacking the miR-566 3’ UTR region into U87 and LN229 glioma cells to confirm that the growth and invasion inhibition, cell cycle arrest and apoptosis promotion by lenti-AS-566 was mediated, at least in part, through the induction of VHL expression. The successful transfection of the VHL plasmid was monitored at the VHL protein level (Figure 3D). We then performed colony formation assays, invasion assays and flow cytometry in VHL plasmid-transfected glioma cells. As shown in Figure 3E and G, cell proliferation and invasion activity were significantly inhibited in VHL-transfected glioma cells (48 h after transfection). As shown in Figure 3 F and H, G₁ cell cycle arrest was induced and apoptosis was increased in VHL-transfected glioma cells.

TOP/FOP FLASH reporter and luciferase constructs were introduced to detect the activity of canonical Wnt signaling. TOPFLASH activity was measured in U87 and LN229 cells infected with lenti-AS-566 or transfected with VHL plasmid in the presence of LiCl or not. Our Results showed that miR-566 inhibition or VHL overexpression could reduce the activity of the β-catenin/TCF4 pathway (Figure 4A). Western blot was used to examine the expression of β-catenin both in the cytoplasm and nucleus. We found that lenti-AS-566 or VHL plasmid decreased the expression of β-catenin, especially in the nucleus (Figure 4B). Confocal images showed the same results (Figure 4C). Together, these findings suggested that miR-566 regulated the activity of β-catenin/TCF signaling through targeting VHL. The same results were found in the HIF-1α pathway (Figure 4D and E). All together, we demonstrated that miR-566 could regulate both β-catenin and HIF-1α signaling by targeting VHL.

VHL is responsible for the degradation of β-catenin and HIF-1α [22], transcription factors that regulate the expression of EGFR [23, 24]. Therefore, we examined whether HIF-1α forms a complex with β-catenin in glioma cells by performing co-IP experiments. The results showed that HIF-1α forms a complex with β-catenin (Figure 5A). Furthermore, lenti-AS-566 partially inhibited the formation of this complex. Western blot analysis was used to determine the endogenous expression levels of HIF-1α or β-catenin in the glioma cell lysates used in these experiments (Figure 5A). These results suggested that HIF-1α and β-catenin form a complex and cooperatively promote EGFR synthesis (Figure 5B).

We have shown that miR-566 is an important oncomiR in EGFR pathway regulation. We then determined if miR-566 inhibition has synergistic effects with nimotuzumab administration. U87 and LN229 cells were treated with nimotuzumab (100 μg/ml), and after 24 h, infected with lenti-AS-566 or mock control. Cell proliferation (Figure 5C), cell cycle distribution (Figure 5D), in vitro invasion (Figure 5E) and apoptosis (Figure 5 F) were evaluated four days after-lentiviral infection. Lenti-AS-566 enhanced the effects of nimotuzumab with suppression of cellular proliferation and invasion (Figure 5C and E). Flow cytometric analysis revealed that more cells were arrested in the G₁ phase in the combination group (Figure 5D). In addition, more apoptotic cells were detected after treatment with nimotuzumab combined with lenti-AS-566 (Figure 5 F). To evaluate the effects of the combined therapy of nimotuzumab and miR-566 inhibition on tumor growth in vivo, we established tumors as intracranial xenografts in nude mice. U87 cells were pretreated with lentivirus containing a luciferase reporter. Compared with nimotuzumab alone, the combination of nimotuzumab with lenti-AS-566 significantly decreased the tumor burden (Figure 6A and B). To analyze the survival times of the treatment groups, we generated Kaplan-Meier survival curves (Figure 6C), which demonstrated that combined therapy significantly prolonged survival.

The human glioma cell lines U87, LN229, SNB19, LN308 and U251 were obtained from the American Type Culture Collection (ATCC, Manassas, VA, USA). Human astrocytes (Invitrogen, Carlsbad, CA) were derived from human brain tissues. The human glioma cell lines were cultured in Dulbecco’s modified Eagle medium (DMEM) supplemented with 10% heat-inactivated fetal bovine serum (FBS, Hyclone, Waltham, MA). Astrocytes were cultured in GIBCO Astrocyte Medium supplemented with N-2, FBS and EGF. Cells were cultured in a humidified 10% CO₂ atmosphere at 37˚C. LiCl (Acros Organics, New Jersey, USA) and CoCl₂ (Sinopharm Chemical, Shanghai, China) were diluted in phosphate-buffered saline (PBS).

Lentiviruses containing a miR-566 inhibitor segment (lenti-AS-566) or negative control (lenti-NC) segment were obtained from Genepharma (Shanghai, China). The human glioma cell lines U87 and LN229 were infected with the viral suspension. pcDNA3 and pcDNA3-VHL plasmids were transfected using Lipofectamine 2000 (Invitrogen, Carlsbad, CA) following the manufacturer’s instructions. Cells were harvested 48 h after infection or transfection, and RNA and protein extractions were performed. TRIzol (Invitrogen, Carlsbad, California) was used to isolate total RNA. To detect miR-566, stem-loop reverse transcription-polymerase chain reaction (RT-PCR) was performed with a one-step RNA PCR kit (Takara, Otsu, Shiga, Japan) according to the manufacturer’s instructions. Real-time PCR was performed by SYBR green detection with a forward primer for the mature miRNA sequence and a universal adaptor reverse primer. For the analysis of EGFR, AKT1, AKT2 and AKT3 messenger RNA (mRNA) expression, complementary DNA (cDNA) synthesis was performed using random primers under standard conditions. mRNA expression was quantified using the ΔΔCt method. GAPDH served as the internal control. All miRNA expression data were normalized to a U6 small nuclear RNA from the same sample. All reactions were performed in triplicate.

Glioma cells were washed in PBS and lysed with ice-cold RIPA buffer (Pierce, Brebieres, France) containing the protease inhibitor PMSF (Sigma, St. Louis, MO). Protein quantification was performed with a NanoDrop ND-1000 Spectrophotometer (NanoDrop Technologies, Wilmington, USA). A DUALXtract Cytoplasmic and Nuclear Protein Extraction Kit (Dualsystems Biotech, Schlieren, Switzerland) was used to isolate cytoplasmic and nuclear proteins from cultured glioma cells.

The protein lysates were subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and transferred to a polyvinylidene difluoride membrane (Roche, Basel, Switzerland). Membranes were immunostained with specific antibodies according to standard protocols. Antibody-labeled protein bands on the membranes were detected with a G:BOX F3 (Syngene, Cambridge, UK).

For immunoprecipitation, cells were lysed with IP lysis buffer (Pierce, Rockford, USA). The cell lysates were then subjected to immunoprecipitation with 1–5 mg of antibodies and Protein A/G agarose beads (Pierce, Rockford, USA) overnight at 4°C with constant agitation. Control samples were incubated with agarose beads after immunoprecipitation with a control immunoglobulin. The immunoprecipitated complexes were then washed with wash buffer. The proteins were eluted, boiled and subjected to SDS-PAGE analysis.

For immunofluorescence analysis, glioma cells were seeded on poly-L-Lysine treated coverslips (BD, USA). The cells were then infected with lenti-566 or lenti-NC with or without LiCl or CoCl₂. After 48 h, the cells were fixed in cold methanol for 2 min. The cells were then washed 3 times in PBS and incubated in blocking buffer for 30 min at room temperature. Next, the cells were washed in PBS and incubated overnight at 4°C with β-catenin and HIF-1α primary antibodies (Cell Signaling Technology). The cells were again washed in PBS, followed by incubation with a fluorescent secondary antibody for 1 h at room temperature. Nuclei were stained with DAPI solution for 5 min. Confocal images of the cells were acquired on a confocal microscope (FV500) with a 40 × water immersion lens and a 1.20 numerical aperture using FluoView software (Olympus, Japan).

For the colony formation assay, 2,000 glioma cells treated with lenti-AS-566, lenti-NC, VHL plasmid or nimotuzumab were plated in complete growth media in a fresh 6-well plate and allowed to grow until visible colonies formed. Cold methanol was used to fix the cell colonies, and colonies were stained with 0.1% crystal violet for 15 min, washed, air dried, photographed and counted.

Corning transwell insert chambers (Corning, New York) and BD Matrigel Invasion Chambers (BD Biosciences, Bedford, MA) were used for the cell invasion experiment. The prepared cells were added to the chamber and incubated for 24 h at 37˚C. Cells that invaded the lower chamber through the membrane were fixed with 20% methanol and stained with 0.1% crystal violet, imaged and counted.

The cell cycle was analyzed by flow cytometry. Pretreated U87 and LN229 cells were washed with PBS, trypsinized, fixed in 70% ethanol, washed and incubated in phosphate-buffered saline containing propidium iodide and RNase A (Sigma, St. Louis, MO) for 30 min at 37˚C. The cell cycle distributions were determined using a DNA stain (4’,6-diamidino-2-phenylindole). The data are the mean ± SD of 3 independent experiments.

Forty-eight hours after transfection, cells were harvested, washed, resuspended in staining buffer and examined using an Annexin V FITC Apoptosis Detection Kit (KeyGEN Biotech, Nanjing, China). The apoptotic distribution of the cells in each sample was then determined using fluorescence-activated cell sorting. Annexin V-positive cells were regarded as apoptotic cells.

Athymic mice (4 weeks of age) were intracranially implanted with 5 × 10⁵ U87 cells (pretreated with lentivirus containing the miR-566 inhibitor segment or negative control segment) under the direction of a stereotactic instrument. Four days after cell implantation, mice were injected intraperitoneally with nimotuzumab or control PBS every other day. Bioluminescence imaging was used to detect intracranial tumor growth. Mice were anesthetized, injected with D-luciferin (Promega, Agora, Fitchburg Center, Fitchburg, Wisconsin) at 50 mg/mL intraperitoneally and imaged with the IVIS Imaging System (Caliper Life Sciences) for 10–120 s. To quantify bioluminescence, identical circular regions of interest were drawn around the entire head of each animal, and the integrated flux of photons (photons per second) in each region of interest was determined by using the Living Images software package (Caliper Life Sciences). Data were normalized to the bioluminescence at the initiation of treatment for each animal. The error bars shown in the figures indicate SDs. All protocols involving animals were performed in accordance with an approved Institutional Animal Care and Use Committee protocol.