Introduction
Glioblastoma multiforme (GBM) is the most malignant form of human astrocytomas (1). Despite recent advances in our understanding of the molecular mechanisms of the disease and improvement in therapeutic strategies, the median survival of GBM patients still remains dismal, with less than one year after diagnosis (2). GBM patients succumb to death due to rapid, aggressive and infiltrative growth of the tumor, rendering the tumor unresectable. Moreover, GBM is recalcitrant to radiotherapy and chemotherapy (3,4) and patients eventually succumb to GBM despite the best available therapeutic regimen due to the emergence of chemoresistance to anticancer drugs.
MicroRNAs (miRNAs) are evolutionarily conserved, small non-coding endogenous RNA molecules that function as posttranscriptional gene regulators by pairing with the 3′ untranslated regions (3′-UTR) of specific target messenger RNAs (mRNAs) to suppress translation or induce mRNA degradation (5,6). Emerging evidence has revealed that many miRNAs function as oncogenes or anti-oncogenes and may play key roles in GBM initiation by regulating a myriad of target genes, especially PTEN, TP53 and EGFR, which are key regulators of GBM progression (7–10). miR-27a is located at chromosome 19 and has been shown to be expressed in a variety of cancer types including breast cancer, gastric adenocarcinoma and cervical cancer (11). miR-27a was shown to be an oncomir in prostate cancer by targeting prohibin (12). It was upregulated in breast cancer tissues (13) and has recently been proposed as a novel marker of breast cancer progression and worse prognosis (14). miR-27a was also found to be upregulated in glioma tissues (15) and may target multiple cellular signaling pathways important for gliomagenesis including the MAPK signaling pathway, the apoptotic signaling pathway and the transforming growth factor-β (TGF-β) signaling pathway (16). miR-27a was shown to drive glioma cell growth by targeting the tumor suppressor gene MXI1 (17) and targeting miR-27a by anti-miRNAs suppressed the growth and invasiveness of GBM U87MG cells (18). These findings implicate miR-27a in gliomagenesis and progression.
The B-cell translocation gene 2 (BTG2), a p53-inducible anti-proliferation gene and a tumor suppressor gene (19), regulates the proliferation of neural progenitor cells during neurogenesis and promotes neurogenic asymmetric divisions (20). In a mouse model of platelet-derived growth factor B (PDGF-B) overexpression-induced glioma, Calzolar et al found that BTG2 downregulation was implicated in the process of malignant transformation and progression of low to high grade gliomas (21). Appolloni et al has further demonstrated that reduced BTG2 expression in human glioma tissues correlates with a significantly worse prognosis (22).
Given the important role of miR-27a in oncogenesis and progression of multiple cancer types, particularly glioma, we were interested in identifying downstream target genes of miR-27a that mediated the actions of miR-27a in GBM cells. In the present study, we investigated the effects of miR-27a on the proliferation and invasiveness of GBM cells in vitro and in a mouse xenograft model and further studied the relation between miR-27a expression and its target gene BTG2, which was identified by computation prediction algorithms.
Materials and methods
Animals
Eighteen female BALB/c-nu mice, 7–8 weeks of age, were obtained from the Institute of Zoology, Chinese Academy of Sciences, Shanghai, China and kept under specific pathogen-free conditions in accordance with procedures and guidelines set by the Institutional Animal Care and Use Committee (IACUC) of the Second Military Medical University, Shanghai, China. The study protocol was approved by the local institutional review board.
Plasmid construction
We amplified a DNA fragment containing the pri-miR-27a gene by PCR using the primer pair: 5′-ACG GGA TCC ACA TTG CCA GGG ATT TCC A-3′ (sense) and 5′-CGC GAA TTC AGC TCA GTA GGC ACG GGA G-3′ (antisense). The fragments were then inserted between BamHI and EcoRI sites of pcDNA3 (Invitrogen, Carlsbad, CA, USA), yielding pcDNA3/pri-miR-27a. For construction of the enhanced green fluorescence protein (EGFP) expression vector carrying the 3′-UTR segment of the BTG2, pcDNA3/EGFP-BTG2 3′-UTR, we amplified the 3′-UTR segment of the BTG2 gene carrying the predicted miR-27a binding site by PCR using the primer set: 5′-GAT CCA GAT GTG AGA GGG AGC AAG CAA GGT TAG CAA CTG TGA ACA GAG AGG TCG GGA TTT GCC CTG-3′ (sense) and 5′-AAT TCA GGG CAA ATC CCG ACC TCT CTG TTC ACA GTT GCT AAC CTT GCT TGC TCC CTC TCA CAT CTG-3′ (antisense). The PCR products were then cloned into the pcDNA3/EGFP vector between BamHI and EcoRI sites. To construct the siR-BTG2 vector, we performed an annealing reaction using two single strands 5′-CAC UAC AAA CAC CAC UGG UdTdT-3′ and 5′-ACC AGT GGT GTT TGT AGT GdTdT-3′. To construct a BTG2 overexpression plasmid pcDNA3/BTG2 the full length human BTG2 gene (without the 3′-UTR) was PCR-amplified using a cDNA clone (GenBank™, AF361937) as a template with the primer set: 5′-ACG GGA TCC ATG AGC CAC GGG AAG GGA A-3′ (sense) and CGC GAA TTC TAG CTG GAG ACT GCC ATC-3′ (antisense), followed by insertion into the pcDNA3 vector between the EcoRI and XhoI sites. All insertions were verified by DNA sequencing.
Cells and transfection
The human GBM cell lines U87MG and U251MG (Cell Bank, Type Culture Collection, Chinese Academy of Sciences, Shanghai, China) were cultured in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% fetal bovine serum (FBS) and 100 U/ml penicillin/streptomycin at 37°C in a humidified atmosphere with 5% CO2.
To generate GBM cells expressing miR-27a or BTG2, we transfectedthecellswithpcDNA3/pri-miR-27a,pcDNA3/BTG2, or the control vector pcDNA3/EGFP using Lipofectamine 2000 (Invitrogen) according to the manufacturer’s instructions. GBM cells with downregulated expression of BTG2 were also transfected with Lipofectamine 2000. Stable cell lines with downregulated expression of miR-27a were established by transfecting GBM cells with 2′-O-methyl-modified antisense oligonucleotide (ASO-miR-27a, Saierbio, Tianjin, China) followed by selection for 20–30 days in complete medium supplemented with 500 μg/ml G418 (Invitrogen). Control cells were transfected with ASO-NC using the same procedure.
RT-PCR
Total cellular RNAs were isolated with the mirVana™ miRNA Isolation kit (Ambion, Austin, TX, USA) according to the manufacturer’s protocol. For miRNA quantification, 1 μg RNA was used as the template and cDNA was generated with the stem-loop reverse transcript primer and M-MLV reverse transcriptase (Promega, Madison, WI, USA), followed by quantitative polymerase chain reaction (PCR) with SYBR Premix EX Taq (Takara, Otsu, Shiga, Japan) on an iQ5 real-time PCR detection system (Bio-Rad). To quantify the target gene BTG2, 2 μg RNA was reverse-transcribed to cDNA with oligo(dT) primers and M-MLV reverse transcriptase (Promega). The sequences of the primers used for the RT-PCR are provided in Table I. Raw data analysis was performed with iQ5 real-time PCR software. The relative fold-change in the transcripts was calculated using the 2−ΔΔCt method with U6 as a normalizing internal control for miRNAs and β-actin for BTG2. Each experiment was carried out in triplicate at least three times independently.
MTT and clonogenic assays
Logarithmically growing cells were plated into 96-well plates at 2×103 cells/well. Cellular proliferation was evaluated using the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay as instructed by the manufacturer (Sigma-Aldrich, St. Louis, MO, USA). Absorbance was read at 570 nm using an uQuant Universal Microplate Spectrophotometer (Bio-Tek Instruments, Winooski, VT, USA). For clonogenic assays, GBM cells were seeded in 12-well plates at 3×102 cells per well. The number of colonies was counted on the 9th day after seeding and a colony was counted only if it contained >50 cells. Each experiment was carried out in triplicate and at least three times independently.
Cell migration assays
Transwell cell migration assay was done to assess the migration of transfected GBM cells. Briefly, the GBM cells were seeded in Transwell inserts with 8.0-μm pore Transwell filters (BD Biosciences, NJ, USA) in 0.2 ml DMEM without FBS. The lower chamber was filled with 600 μl DMEM containing 20% FBS acting as chemoattractants. Twenty-eight hours later, the migrant cells that had attached to the lower surface were fixed with 20% methanol and stained for 20 min with crystal violet. The membranes were then carved and embedded under cover slips with the cells on the top. Cells in three different fields of view at ×100 magnification were counted and expressed as the average number of cells per field of view. Each experiment was carried out in triplicate at least three times independently.
Tumor xenografts
For GBM xenograft experiments, 5×106 stable U87MG cells with downregulated miR-27a were inoculated subcutaneously at the axillary fossae of female athymic nude mice. Tumor size was monitored by measuring the length and width with calipers and volumes were calculated with the formula: (L×W2) × 0.5 mm3, where L is the length and W is the width of each tumor.
Cell cycle analysis
Transfected GBM cells were plated in 6-well plates in duplicate for 24 h in complete medium. One group of cells was deprived of serum for 24 h before harvesting, while another group of cells was returned to complete medium for another 24 h before harvesting. Cells were gathered by centrifugation, fixed in 95% (v/v) ethanol and stored at −20°C overnight. After washing with phosphate-buffered saline (PBS), the cells were resuspended in propidium iodide staining buffer (PBS, 0.1% Triton X-100, 60 μg/ml propidium iodide, 0.1 mg/ml DNase-free RNase and 0.1% trisodiumcitrate) for 30 min on ice. The DNA content was analyzed with FACSCalibur flow cytometer and CellQuest software (BD Biosciences). The proliferation index (PI) was calculated using the following equation:
PI
=
(
S
+
G
2
)
/
G
1
where S, G2 and G1 represent the percentage of cells in the S, G2 and G1 phase, respectively.
Computational algorithm analysis
We obtained predicted miRNA target sites of miR-27a on human mRNA transcripts from PicTar (http://picta.mdc-berlin.de/), TargetScan (www.targetscan.org/) (Whitehead Institute for Biomedical Research, Cambridge, MA, USA) and miRBase Targets databases (23).
Fluorescent reporter assays
For determination of direct interaction between miR-27a and BTG2 mRNA, 1.5×104 U87MG cells per well were seeded into 48-well plates and after 24-h incubation, cells were transfected with pri-miR-27a or ASO-miR-27a together with the pcDNA3/EGFP-BTG2 3′-UTR reporter vector, the RFP expression vector pDsRed2-N1 (Clontech) was used as an internal control. Forty-eight hours after transfection, cellular lysates were prepared using RIPA lysis buffer (150 mM NaCl, 50 mM Tris-HCl pH 8.0, 1% Triton X-100 and 0.1% SDS) and EGFP and RFP intensity was then measured with an F-4500 fluorescence spectrophotometer (Hitachi, Tokyo, Japan). All experiments were performed at least three times independently.
Western blot assays
Cellular lysates were prepared using RIPA lysis buffer as described above and subjected to 15% SDS denaturing polyacrylamide gel electrophoresis. The immunoblotting procedure was performed as previously described (24) and anti-BTG2 or anti-GAPDH antibodies (Abcam) were used for the procedure. The protein bands were visualized by enhanced chemiluminescence and analyzed using the LabWorks image acquisition and analysis software (UVP, LLC).
Statistical analysis
The data are reported as mean ± SD of at least three independent experiments with the two-sided Student’s t-test. A P-value of<0.05 was considered significant.
Results
miR-27a promotes the clonogenic growth and migration of U87MG cells
We were interested in the effect of miR-27a, which has been shown to be significantly overexpressed in GBM tissues (25), on the proliferation and migration of GBM cells. Our quantitative RT-PCR assays revealed that transfection with pri-miR-27a caused a significant 3.97±0.35-fold increase in miR-27a levels compared with controls (P<0.001) in GBM U87MG and U251MG cells while transfection with ASO-miR-27a caused a marked 78±6% reduction in miR-27a levels in the two cell lines (P<0.01) (Fig. 1A). The MTT assays showed that miR-27a overexpression was associated with an ~1.47±0.02-fold increase in the viability of U87MG and U251MG cells over the controls (P<0.05 or 0.01) (Fig. 1B). A corresponding significant increase in the number of colonies of U87MG and U251MG cells was also observed as revealed by clonogenic assays (P<0.05 or 0.01) (Fig. 1C). By contrast, miR-27a downregulation noticeably suppressed the viability and clonogenicity of U87MG and U251MG cells (P<0.001 or 0.01) (Fig. 1B and C). Transwell assays further revealed that miR-27a overexpression significantly increased the number of migrated U87MG and U251MG cells (P<0.05 or 0.001) while miR-27a downregulation by ASO-miR-27a caused a 50% reduction in the number of migrated cells (P<0.05 or 0.001) (Fig. 1D). We also found that miR-27a downregulation markedly reduced the growth of mouse xenografts bearing U87MG cells (P<0.05 vs. controls) (Fig. 2). The above findings together indicated that miR-27a overexpression potentiates the growth and migration of GBM cells both in vitro and in vivo.
miR-27a facilitates G1/S phase transition in GBM cells
We determined whether increased proliferation of GMB cells by miR-27a overexpression was due to changes in cell cycle distribution. Our flow cytometric analysis showed that miR-27a overexpression by pri-miR-27a caused an apparent increase in the percentage of U87MG (pri-miR-27a, 27.94±0.72% vs. controls, 21.65±0.39% P<0.05) and U251MG cells (pri-miR-27a, 33.97±1.63% vs. controls, 26.64±2.12% P<0.05) in the S phase (Fig. 3A and B). By contrast, downregulation by ASO-miR-27a noticeably decreased the percentage of U87MG (ASO-miR-27a, 15.87±0.95 vs. controls, 24.53±0.24% P<0.05) and U251MG cells (ASO-miR-27a, 17.58±1.39 vs. controls, 28.56±3.51% P<0.05) in the S phase (Fig. 3A and B). Consistently, U87MG and U251MG cells with miR-27a over-expression showed a significantly higher proliferation index than that of controls (P<0.05 while GBM cells with downregulated miR-27a had a noticeably lower proliferation index compared with controls (P<0.01 or 0.05) (Fig. 3C). These results suggest that miR-27a may promote the clonogenic growth of GBM cells by increasing the percentage of GBM cells in the S phase.
miR-27anegativelyregulatestheexpressionofBTG2bydirectly targeting its 3′-UTR
We used TargetScan, PicTar and miRBase targets to identify candidate target genes of miR-27a and found that the BTG2 gene, whose encoded protein is involved in cell cycle control (26), contained a putative binding site for miR-27a in its 3′-UTR (Fig. 4A). To confirm whether BTG2 was a bona fide target of miR-27a, we co-transfected U87MG cells with pcDNA3/EGFP-BTG2 3′-UTR and pri-miR-27a or ASO-miR-27a. We found that pri-miR-27a caused a 19.7±2.4% reduction in EGFP intensity compared to controls (P<0.05) while ASO-miR-27a caused a 37.2±0.7% increase in EGFP intensity (P<0.001 vs. controls) (Fig. 4B). Our quantitative RT-PCR assays showed that miR-27a overexpression caused a marked 39.3±1.1% decrease while miR-27a downregulation by ASO-miR-27a caused a noticeable 108.6±9.6% increase in the mRNA transcript levels of BTG2 (P<0.001 vs. controls in both) (Fig. 4C). Immunoblotting assays further revealed that miR-27a overexpression led to a marked 79.3±2.1% decrease (P<0.001 vs. controls) while miR-27a downregulation caused a significant 22.7±4.4% increase (P<0.05 vs. controls) in endogenous BTG2 levels (Fig. 4D). These data reveal that miR-27a negatively regulates endogenous BTG2 expression by targeting its 3′-UTR.
BTG2 modulates the growth and migration of GBM cells
We examined whether BTG2 exerted any effect on the clonogenic growth and migration of GBM cells. Our clonogenic assays demonstrated that suppression of BTG2 expression by siRNA caused a 1.6-fold increase in the number of colonies from U87MG cells (P<0.001 vs. controls) while BTG2 overexpression caused a 60% reduction in the number of colonies (P<0.01 vs. controls) (Fig. 5A). Moreover, BTG2 knockdown was associated with 1.4-fold increase in the number of migrated U87MG cells (P<0.01 vs. controls) (Fig. 5B). By contrast, BTG2 overexpression was associated with a 55% decrease in the number of migrated U87MG cells (P<0.01 vs. controls). We further assessed the effect of BTG2 on cell cycle distributions by flow cytometry. We found that BTG2 knockdown markedly increased the proportion of cells in the S phase (P<0.05 vs. controls) while BTG2 overexpression led to a significant decline in the proportion of U87MG cells in the S phase (P<0.05 vs. controls) (Fig. 5C). Consistently, BTG2 downregulation significantly increased while BTG2 overexpression markedly reduced the proliferation index (P<0.05 in both) (Fig. 5D). These data suggest that BTG2, similar to miR-27a, modulated the clonogenic growth and migration of human GBM cells.
BTG2 functions downstream of miR-27a in modulating the proliferation and migration of human GBM cells
To determine whether the effects of miR-27a expression on cell growth and migration of GBM cells were mediated by BTG2, we transfected U87MG cells with ASO-miR-27a or siR-BTG2 and appropriate controls. Consistently, miR-27a downregulation by ASO-miR-27a markedly suppressed the clonogenic growth of U87MG cells (P<0.01 vs. controls); this decrease, however, was abrogated by siRNA knockdown of BTG2 (Fig. 6A). Transwell assays further showed that knockdown of BTG2 reversed the suppression of U87MG cell migration by ASO-miR-27a from 50–90% of controls (Fig. 6B). These results indicate that BTG2 functions downstream of miR-27a in modulating the proliferation and migration of human GBM cells.
Discussion
miR-27a and BTG2 have been shown to be implicated in gliomagenesis and glioma progression. However, hitherto, a link between miR-27a and BTG2 in glioma has not been suspected. The present study demonstrated that miR-27a modulates the growth and invasiveness of GBM cells and directly targets BTG2.
We showed that miR-27a overexpression significantly increased the clonogenic growth of GBM cells in vitro. Furthermore, miR-27a overexpression was associated with markedly increased invasiveness of GBM cells in vitro. These findings are consistent with those by Xu et al (17) who found that miR-27a was upregulated in glioma tissues and its overexpression drove the proliferation of glioma cells. miR-27a has been reported to regulate cell growth and division in a dose-dependent manner (11). We found that increased proliferation of GBM cells from miR-27a overexpression was associated with a marked increase in the proportion of GBM cells in the S phase and a corresponding increase in the proliferation index. We further showed that miR-27a downregulation caused a significant reduction in the number of migrated GBM cells as revealed by Transwell assays. This is in agreement with the findings by Feng et al who showed inhibition of miR-27a substantially impaired the invasiveness of U87MG cells in vitro (18).
BTG2 downregulation was involved in the progression of low to high grade gliomas (21) in a mouse glioma model and reduced BTG2 expression was found to correlate with a worse prognosis (22) of glioma patients, suggesting that BTG2 may play an important role in gliomagenesis and progression. Our computational prediction algorithms revealed BTG2 as a candidate target gene of miR-27a, which was confirmed by our findings from EGFP reporter assays showing an inverse relation between miR-27a expression and endogenous BTG2 expression. Though molecular targets of miR-27a such as Sprouty have been reported (27), no previous study has shown that miR-27a targets BTG2. We showed that BTG2 expression impacts on the proliferation and invasiveness of GBM cells and BTG2 functions downstream of miR-27a in modulating the proliferation and migration of GBM cells.
It has been reported that BTG2 is a direct transcriptional target gene of p53 (28–31). BTG2 was found to trigger cell cycle arrest by inhibiting G1-S transition via decreasing cyclin D1 expression (19,32,33). Apart from glioma (22), downregulation of BTG2 expression has also been observed in several tumor types, including prostate cancer, clear renal cell carcinomas and breast cancer (34–36). We speculate that miR-27a functions as an oncomir through negative regulation of BTG2, which then promotes G1→S transition.
Collectively, our results indicate that miR-27a and BTG2 promote the migration and the growth of GBM in vitro and in vivo and miR-27a directly targets BTG2 in GBM cells. Our findings suggest that miR-27a and BTG2 may represent novel biomarkers for GBM patients and could potentially be manipulated therapeutically.