Forkhead Box M1 Is Essential for Nuclear Localization of Glioma-associated Oncogene Homolog 1 in Glioblastoma Multiforme Cells by Promoting Importin-7 Expression*

Background: The transcription factors GLI1 and FOXM1 play critical roles in cancer development and progression. Results: FOXM1 bound to the importin-7 promoter to up-regulate its expression; FOXM1 deficiency inhibited importin-7 expression and nuclear localization of GLI1. Conclusion: FOXM1 is essential for nuclear localization of GLI1 by promoting importin-7 expression. Significance: FOXM1 and GLI1 form a positive feedback loop that contributes to glioblastoma multiforme development. The transcription factors glioma-associated oncogene homolog 1 (GLI1), a primary marker of Hedgehog pathway activation, and Forkhead box M1 (FOXM1) are aberrantly activated in a wide range of malignancies, including glioma. However, the mechanism of nuclear localization of GLI1 and whether FOXM1 regulates the Hedgehog signaling pathway are poorly understood. Here we found that FOXM1 promotes nuclear import of GLI1 in glioblastoma multiforme cells and thus increases the expression of its target genes. Conversely, knockdown of FOXM1 expression with FOXM1 siRNA abrogated its nuclear import and inhibited the expression of its target genes. Also, genetic deletion of FOXM1 in mouse embryonic fibroblasts abolished nuclear localization of GLI1. We observed that FOXM1 directly binds to the importin-7 (IPO7) promoter and increases its promoter activity. IPO7 interacted with GLI1, leading to enhanced nuclear import of GLI1. Depletion of IPO7 by IPO7 siRNA reduced nuclear accumulation of GLI1. In addition, FOXM1 induced nuclear import of GLI1 by promoting IPO7 expression. Moreover, the FOXM1/IPO7/GLI1 axis promoted cell proliferation, migration, and invasion in vitro. Finally, expression of FOXM1 was markedly correlated with that of GLI1 in human glioblastoma specimens. These data suggest that FOXM1 and GLI1 form a positive feedback loop that contributes to glioblastoma development. Furthermore, our study revealed a mechanism that controls nuclear import of GLI1 in glioblastoma multiforme cells.

The transcription factors glioma-associated oncogene homolog 1 (GLI1), a primary marker of Hedgehog pathway activation, and Forkhead box M1 (FOXM1) are aberrantly activated in a wide range of malignancies, including glioma. However, the mechanism of nuclear localization of GLI1 and whether FOXM1 regulates the Hedgehog signaling pathway are poorly understood. Here we found that FOXM1 promotes nuclear import of GLI1 in glioblastoma multiforme cells and thus increases the expression of its target genes. Conversely, knockdown of FOXM1 expression with FOXM1 siRNA abrogated its nuclear import and inhibited the expression of its target genes. Also, genetic deletion of FOXM1 in mouse embryonic fibroblasts abolished nuclear localization of GLI1. We observed that FOXM1 directly binds to the importin-7 (IPO7) promoter and increases its promoter activity. IPO7 interacted with GLI1, leading to enhanced nuclear import of GLI1. Depletion of IPO7 by IPO7 siRNA reduced nuclear accumulation of GLI1. In addition, FOXM1 induced nuclear import of GLI1 by promoting IPO7 expression. Moreover, the FOXM1/IPO7/GLI1 axis promoted cell proliferation, migration, and invasion in vitro. Finally, expression of FOXM1 was markedly correlated with that of GLI1 in human glioblastoma specimens. These data suggest that FOXM1 and GLI1 form a positive feedback loop that contributes to glioblastoma development. Furthermore, our study revealed a mechanism that controls nuclear import of GLI1 in glioblastoma multiforme cells.
The Hedgehog (Hh) 3 signaling pathway is a developmental signaling pathway that controls numerous developmental processes (1). Although this pathway is silenced in normal adult mature cells, it is aberrantly activated in a wide range of malignancies, including glioma (2,3). The central components of the mammalian Hh pathway include three secreted ligands (Sonic hedgehog, Indian hedgehog, and Desert hedgehog), a negative regulatory receptor (PTCH), a positive regulatory protein (SMO), and the glioma-associated oncogene transcription factors (glioma-associated oncogene homolog 1 (GLI1), GLI2, and GLI3) (4). In particular, the zinc finger transcription factor GLI1 is the primary marker of Hh pathway activation. Without Hh ligands, PTCH inhibits the activity of SMO to keep it from transmitting the Hh signal. Upon ligand binding, this inhibition of SMO by PTCH is relieved, and the Hh signal is activated. Activated SMO orchestrates a signaling cascade that eventually results in activation and release of the glioma-associated oncogene transcription factors from a protein complex (2)(3)(4). Activated glioma-associated oncogenes translocate into the nucleus to regulate the expression of various context-specific genes. However, the mechanism of the nuclear localization of GLI1 is poorly understood.
Forkhead box M1 (FOXM1) is a member of the Forkhead family of transcription factors (5). Many studies have demonstrated that FOXM1 is involved in different aspects of tumorigenesis, including angiogenesis, invasion, and metastasis (6,7). FOXM1 is often overexpressed in various human malignancies. For example, high expression of FOXM1 in glioblastoma multiforme (GBM) correlates with tumorigenicity of the GBM cells (8). Authors have reported FOXM1 to be a downstream target of GLI1 in the Hh pathway that contributes to tumorigenesis (9). However, how FOXM1 regulates the Hh signaling pathway remains unknown.
Given that both FOXM1 and GLI1 are activated in many cancers, including glioma, we investigated the potential relationship between these two key transcriptional factors in GBM cases. We found that FOXM1 is essential for nuclear import of GLI1 by promoting importin-7 (IPO7) expression. We also demonstrated that IPO7 binds to GLI1 and thus leads to enhanced nuclear import of GLI1. These data suggest that FOXM1 and GLI1 form a positive feedback loop that contributes to GBM development. Additionally, this study revealed a mechanism that controls nuclear import of GLI1 in GBM cells.
Subcellular Fractionation-Cells were harvested, and nuclear protein fractions were separated using a CelLytic nuclear extraction kit (Sigma-Aldrich) according to the manufacturer's recommended procedures.
IP and IB Analysis-For co-IP, cells were grown in 10-cm dishes and transfected with the appropriate plasmids. Cell lysates were incubated with 2 g of antibody on a rotator overnight at 4°C. Protein-antibody-protein A/G-agarose complexes were prepared by adding 50 l of protein A/G-agarose beads (Amersham Biosciences) for 1 h at 4°C. After extensive washing with radioimmunoprecipitation assay lysis buffer, immunoprecipitated complexes were resuspended in a reducing sample buffer and boiled for 10 min. After centrifugation to pellet the agarose beads, supernatants were subjected to SDS-PAGE and IB analysis as described previously (11).
Immunofluorescent Analysis-Cells were rinsed twice with PBS, fixed with 4% buffered paraformaldehyde, and permeabilized with 0.5% Triton X-100 for 15 min. Cells were then incubated with a primary antibody overnight at 4°C and then with a secondary antibody conjugated to FITC (Molecular Probes, Eugene, OR) for 1 h at room temperature. Cells were then examined using a deconvolutional microscope (Carl Zeiss, Oberkochen, Germany) as described previously (11).
Promoter Reporters and Dual-Luciferase Assay-Cells were transfected with the IPO7 human promoter reporter plasmid together with pRL-TK and analyzed as described previously (11). Luciferase activity was measured using a Dual-Luciferase assay system (Promega, Madison, WI).
Cell Proliferation and Colony Formation Assays-For a cell proliferation assay, after transfection, U87 cells (5 ϫ 10 3 ) were plated in 96-well plates. Cell growth was assessed using a standard 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay. For a colony formation assay, after transfection, cells were trypsinized and reseeded in 6-well plates at a density of 5 ϫ 10 3 cells per well. For crystal violet staining, plates were washed once each with PBS and cold methanol and then incubated in a crystal violet solution for 5 min. The plates were subsequently washed twice with doubledistilled water, air-dried, and scanned using an Epson (Long Beach, CA) scanner.
Wound Healing Assay-Following transfection, U87 cells were plated overnight to achieve a subconfluent cell layer in 6-well plates. The cell layer was scratched with a micropipette tip, and cultures were washed twice with a serum-free medium to remove floating cells. Wound healing was visualized by comparing photographs of the cell layer taken at 0 and 24 h and processed using the Photoshop CC software program (Adobe Systems, San Jose, CA). Differences in cell migration distances were determined using a Student's t test for comparing mean values.
Transwell Invasion Assay-An invasion assay was carried out in Transwell chambers containing polycarbonate filters (8-m pore size; BD Biosciences) in which the upper surfaces of the filters in the invasion chambers were coated with a growth factor-reduced Matrigel matrix. Following transfection, U87 cells (5 ϫ 10 4 ) in a 500-l volume of serum-free medium were placed in the upper chambers and incubated at 37°C for 16 h for the invasion assay. The cells that penetrated through the Matrigelcoated filters were counted at a magnification of ϫ200 in 15 randomly selected fields, and the mean number of cells per field was recorded.
Human Tissue Specimens and Immunohistochemical Analysis-Sections of paraffin-embedded human GBM specimens were stained with antibodies against GLI1 and FOXM1. Immunohistochemical analysis of GBM tissue arrays was performed using a standard immunostaining protocol as described previously (11). The use of human brain tumor specimens was approved by the MD Anderson Institutional Review Board.
Statistical Analysis-The significance of the data on the patient specimens was determined using the Pearson correlation coefficient. The significance was determined using the Student's t test (two-tailed). p values less than 0.05 were considered significant.

FOXM1
Promotes Nuclear Import of GLI1 in a DNA Bindingdependent Manner-We analyzed the expression of nuclear FOXM1 and GLI1 protein in Hs683, SW1783, HFU251, and U87 cells. We observed markedly higher FOXM1 expression in HFU251 and U87 cells than in Hs683 and SW1783 cells (Fig.  1A). Also, the expression of nuclear GLI1 was positively correlated with the expression of nuclear FOXM1 in these cell lines. Because GLI1 is a nuclear cytoplasmic shuttling protein (12), we attempted to determine whether FOXM1 modifies the nuclear translocation of GLI1. As nuclear export of GLI1 is dependent on chromosome region maintenance homolog 1 (12), we examined GLI1 localization in the presence of LMB, a chromosome region maintenance homolog 1-dependent export inhibitor. We found that exposure to LMB increased the nuclear localization of GLI1, which was partially attenuated by knockdown of FOXM1 expression, confirming that FOXM1 is essential for nuclear import of GLI1 (Fig. 1B). Furthermore, deletion of FOXM1 in Foxm1 fl/fl mouse embryonic fibroblasts virtually abolished nuclear localization of GLI1 (Fig. 1C), further confirming that FOXM1 is required for nuclear import of GLI1.
Next, we induced overexpression of FOXM1 without a DNA-binding domain, the FOXM1 mutant R286A/H287A (which destroys the DNA binding ability of FOXM1), and WT FOXM1 in Hs683 cells. WT FOXM1 increased endogenous nuclear GLI1 levels, whereas the other two proteins did not (Fig. 1D), suggesting that FOXM1 promotes nuclear import of GLI1 in a DNA binding-dependent manner. Conversely, depletion of FOXM1 by siRNA in U87 cells decreased the level of nuclear GLI1 (Fig. 1E). To investigate the changes in mRNA expression of the GLI1 target genes in the presence of FOXM1, we performed quantitative real-time RT-PCR analysis of Hs683 cells overexpressing FOXM1. As shown in Fig. 1F, FOXM1 increased the expression of Ptch1, Ptch2, CCND2, and IL-6. In contrast, knockdown of FOXM1 expression decreased the expression of these genes in U87 cells (Fig. 1G). Taken together, these results suggested that FOXM1 promotes nuclear import of GLI1 in a DNA binding-dependent manner.
IPO7 Is a Direct Transcriptional Target of FOXM1-IPO protein plays a key role in moving other proteins into the nucleus (13). The nuclear import pathway of GLI1 has yet to be fully elucidated. To understand the molecular mechanism of FOXM1 regulation in nuclear import of GLI1, we screened Hs683 cells to determine whether FOXM1 regulated the expression of several IPOs. As shown in Fig. 2A, overexpression of FOXM1 markedly increased IPO7 expression but did not change IPO8, IPO9, or IPO13 expression in Hs683 cells. Next, we analyzed the sequence of the human IPO7 promoter using the FOXM1 consensus sequences. We identified three putative FOXM1-binding sites in the human IPO7 promoter (Fig. 2B). All of the FOXM1-binding regions of the IPO7 promoter bound to endogenous FOXM1 protein in U87 cells in our ChIP assays (Fig. 2C). Moreover, we examined whether FOXM1 transactivates the IPO7 promoter using a human IPO7 promoter luciferase reporter. Overexpressed WT FOXM1 but not FOXM1 without a DNA-binding domain or the FOXM1 mutant R286A/H287A up-regulated IPO7 promoter activity in HEK 293T and Hs683 cells (Fig. 2, D and E). In comparison, IPO7 promoter activity decreased in U87 and HFU251 cells with FOXM1 siRNA (siFOXM1) (Fig. 2, F and G). Collectively, these results indicated that IPO7 is a direct transcriptional target of FOXM1 and that FOXM1 transactivates the IPO7 promoter in a DNA binding-dependent manner.
IPO7 Binds to GLI1-To determine whether nuclear import of GLI1 by IPO7 is mediated by interaction between the two, we performed a co-IP assay. In HEK 293T cells transfected with a FLAG-GLI1 fusion construct, FLAG-GLI1 protein was coimmunoprecipitated with endogenous IPO7 (Fig. 3A). Furthermore, we examined the interaction of endogenous GLI1 and IPO7 at the physiological level. As shown in Fig. 3B, endogenous GLI1 was associated with IPO7 in U87 cells in a co-IP assay. Next, we analyzed the effect of IPO7 knockdown by siRNA on localization of endogenous GLI1 as detected via immunostaining. In contrast with control siRNA, IPO7 siRNA reduced nuclear accumulation of GLI1 (Fig. 3C). Consistent with these results, knockdown of IPO7 expression decreased GLI1 target gene expression according to quantitative real-time RT-PCR analysis (Fig. 3D). These results demonstrated that IPO7 is a key nuclear transporter for GLI1 by interacting with GLI1 protein.
FOXM1 Induces Nuclear Import of GLI1 by Promoting IPO7 Expression-We examined whether FOXM1 increases nuclear import of GLI1 via up-regulation of IPO7 expression. U87 cells exhibited reduced nuclear accumulation of GLI1 in the presence of siFOXM1, but overexpression of IPO7 rescued nuclear localization of GLI1 in these cells according to immunofluores-cent staining (Fig. 4A). Using Western blotting of U87 cells, we found that knockdown of FOXM1 expression resulted in decreased nuclear localization of GLI1, which was abrogated by IPO7 overexpression (Fig. 4B). Moreover, knockdown of FOXM1 expression decreased GLI1 target gene expression, and IPO7 overexpression rescued the inhibitory effect of FOXM1 depletion (Fig. 4C). Taken together, these results demonstrated that FOXM1 induces nuclear import of GLI1 by promoting IPO7 expression.
The FOXM1/IPO7/GLI1 Axis Promotes Cell Proliferation, Migration, and Invasion-Next, we examined whether the FOXM1/IPO7/GLI1 axis has a role in the proliferation, migration, and invasion of U87 cells. We found that U87 cells with FOXM1 depletion proliferated much more slowly than control cells did. Furthermore, the siFOXM1-transfected U87 cells were much less able to form colonies than were control cells; however, IPO7 and GLI1 overexpression rescued siFOXM1transfected U87 cell proliferation and increased the number of colonies of these cells (Fig. 5, A and B). We also performed wound healing and Transwell invasion assays to determine whether FOXM1 affects GLI1-mediated cell migration and invasion. The results demonstrated that FOXM1 down-regulation inhibited the migration of U87 cells more than control cells. Moreover, the inhibitory effect of siFOXM1 on the migration of U87 cells was rescued by IPO7 and GLI1 overexpression (Fig. 5C). In addition, knockdown of FOXM1 expression decreased the invasion of U87 cells relative to that of control cells, and expression of IPO7 or GLI1 restored the invasion of these cells (Fig. 5D). Taken together, these results demon-  IPO7 is a direct transcriptional target of FOXM1. A, increased IPO7 expression induced by FOXM1. Hs683 cells were transfected with an empty vector (CTRL) or FOXM1 expression vector. Cell lysates were collected at 48 h after transfection and subjected to IB analysis with the indicated antibodies. B, schematic of the human IPO7 promoter. The sequences of the FOXM1-binding elements are shown. C, results of a ChIP assay performed with U87 cells. Chromatin fragments of the cells were immunoprecipitated with an anti-FOXM1 antibody or control IgG and subjected to PCR using primers for three FOXM1-binding sites. One percent of the total cell lysates was subjected to PCR before IP as an input. D and E, HEK 293T (D) and Hs683 (E) cells were transfected with the human IPO7 promoter and indicated plasmids. The luciferase activity of the cells was then determined. Each error bar indicates that the variation in the mean results from three independent experiments. ***, p Ͻ 0.001. F and G, U87 (F) and HFU251 (G) cells were transfected with the human IPO7 promoter and siFOXM1 or a control siRNA (siCTRL). The luciferase activity of the cells was then determined. Each error bar indicates the variation in the mean results from three independent experiments. **, p Ͻ 0.01; ***, p Ͻ 0.001.

FIGURE 3. IPO7 is a key nuclear transporter for GLI1 by interacting with GLI1 protein.
A, expression of FLAG-GLI1 in HEK 293T cells and IP of it using an anti-FLAG antibody. GLI1-bound IPO7 protein was detected in the cells using an IB with an anti-IPO7 antibody. Whole-cell lysates were directly subjected to immunoblotting with FLAG or IPO7 antibody as an input. B, co-IP of endogenous GLI1 with IPO7 in U87 cells. IPO7 was immunoprecipitated, and the amount of IPO7 bound to GLI1 was determined using an IB with an anti-GLI1 antibody. C, attenuation of nuclear import of GLI1 by knockdown of IPO7 expression. Top: U87 cells were transfected with siRNAs targeting IPO7 (siIPO7) or control siRNA (siCTRL). After 24 h of transfection, cells were treated with 5 nM LMB for 8 h. Cells were then stained with an anti-GLI1 antibody and an FITC-conjugated anti-rabbit secondary antibody (green, for GLI1), and the nuclei were visualized via staining with DAPI (blue). Scale bars, 20 m. Bottom: results are indicated as the percentage of cells with mostly cytoplasmic fluorescence, mostly nuclear fluorescence, or both cytoplasmic and nuclear fluorescence. D, knockdown of IPO7 expression decreased GLI1 target gene expression. U87 cells were transfected with siRNAs targeting IPO7 or control siRNA. Expression of Ptch1, Ptch2, CCND2, and IL-6 mRNA was measured using quantitative real-time RT-PCR. The mean Ϯ S.D. values for triplicate samples from a representative experiment are presented. **, p Ͻ 0.01; ***, p Ͻ 0.001. strated that FOXM1 increases IPO7 expression and thus promotes nuclear import of GLI1, which results in increased cell proliferation, migration, and invasion.
FOXM1 Expression Correlates Positively with GLI1 Expression-Finally, to determine whether our findings are clinically relevant, we examined the FOXM1 and GLI1 expression levels in serial sections of 40 human primary GBM specimens via immunohistochemical analysis. The expression of FOXM1 correlated positively with that of GLI1 in the specimens (Fig. 6A). Quantification of staining demonstrated that this correlation was statistically significant in all 40 specimens (r ϭ 0.803; p Ͻ 0.001) (Fig. 6B). These results further supported a critical role for FOXM1 in induction of nuclear import of GLI1 in human GBM cells.

Discussion
Our study demonstrated strong evidence supporting a critical role for FOXM1 in regulation of GBM development. We showed that FOXM1 enhances the growth and invasion of human glioma cells via the Hh signaling pathway. Specifically, FOXM1 stimulates transcription of IPO7 by binding directly to its promoter at three sites, which in turn leads to nuclear localization of the transcription factor GLI1, which is the most important indicator of Hh pathway activation. We also found that the level of FOXM1 expression is highly correlated with that of GLI1 expression in human GBM specimens. Because GLI1 may up-regulate FOXM1 expression (9), the regulatory feedback between FOXM1 and GLI1 in our study may represent a critical mechanism of the proliferation and invasion of human brain tumors (Fig. 6C).
FOXM1 is a member of the Forkhead family of transcription factors, whose members promote tumorigenesis and tumor metastasis by activating a series of oncogenes (14 -17). FOXM1 is involved in regulation of several signaling pathways, such as Wnt/␤-catenin (18), TGF-␤/Smads (19), Akt (20), and MAPK/ERK (21). However, whether cross-talk between FOXM1 and the Hh pathway exists is unclear. In the present study, we found that FOXM1 regulated nuclear localization of GLI1 in a DNA binding-dependent manner, a crucial finding suggesting that the transcriptional activity of FOXM1 is necessary for nuclear localization of GLI1. Also, we analyzed the mechanism by which GLI1 may be transported into the nucleus. Specific knockdown of IPO7 expression using siRNA resulted in reduced nuclear accumulation of GLI1, indicating the dependence of nuclear import of GLI1 on IPO7. Previous studies demonstrated that several molecules may affect nuclear localization of GLI1 (22)(23)(24)(25). For example, nuclear entry of GLI1 was regulated by its nuclear import factor Imp␤1 (importin ␤1) and SuFu (22); WIP1 (or PPM1D) enhanced the function of GLI1 by increasing its nuclear localization (23); Rab23 reduced the FIGURE 4. FOXM1 induces nuclear import of GLI1 by promoting IPO7 expression. A, left: after 24 h of transfection with the indicated siRNA and/or plasmids, U87 cells were treated with 5 nM LMB for 8 h. Cells were then stained with an anti-GLI1 antibody and an FITC-conjugated anti-rabbit secondary antibody (green, for GLI1), and the nuclei were visualized via staining with DAPI (blue). Scale bars, 20 m. Right: results are indicated as the percentage of cells with mostly cytoplasmic fluorescence, mostly nuclear fluorescence, or both cytoplasmic and nuclear fluorescence. B, U87 cells were transfected with the indicated siRNA and/or plasmids, nuclear extracts were prepared, and the GLI1 expression was measured. Lamin B was used as a loading control for nuclear fractions. C, U87 cells were transfected with the indicated siRNA and/or plasmids. Expression of Ptch1, Ptch2, CCND2, and IL-6 mRNA was measured using quantitative real-time RT-PCR. The mean Ϯ S.D. values for triplicate samples from a representative experiment are presented. **, p Ͻ 0.01; ***, p Ͻ 0.001. NS, not significant.

FOXM1 Is Essential for Nuclear Localization of GLI1
JULY 24, 2015 • VOLUME 290 • NUMBER 30 JOURNAL OF BIOLOGICAL CHEMISTRY 18667 nuclear localization of GLI1 (24); and SHP (small heterodimer partner) inhibited GLI1 nuclear localization (25). Thus, determining whether FOXM1 also modulates these molecules' expression will be very interesting.
GLI1 is the primary nuclear effector of the Hh pathway (26), and infiltrative invasion and uncontrolled proliferation are hallmarks of GBM development (27)(28)(29). In our study, functional experiments using gain-or loss-of-function studies demonstrated that the FOXM1/IPO7/GLI1 axis promotes GBM cell proliferation, migration, and invasion. A well known characteristic of GLI1 is that it is often overexpressed in human glioma cells (30,31), and it directly regulates expression of oncoproteins, including cyclin D1/D2 (32), FOXM1 (13), and epithelialto-mesenchymal transition-related proteins (33), and thus promotes the proliferation and invasion of tumor cells. We found that FOXM1 expression correlated directly with GLI1 expression in human GBM specimens, further confirming that these two important transcriptional factors have potential clinical relevance.
We also discovered that FOXM1 and GLI1 form a novel positive regulatory feedback loop. Such loops are common mechanisms of consecutive activation of factors and signaling pathways in tumor progression. Specifically, we found that FOXM1 activates the Hh pathway by stimulating nuclear localization of GLI1 via up-regulation of IPO7 expression. Moreover, a previous study demonstrated that GLI1 activates FOXM1 expression (13). Therefore, this regulatory feedback loop made up of FOXM1 and GLI1 maintains consecutive activation of the Hh signaling pathway, which may be a critical mechanism of brain tumor progression.
In summary, our findings demonstrated cross-talk between FOXM1 and the Hh signaling pathway in human glioma cells. Because of the great importance of the FOXM1/IPO7/GLI1 axis to development of human cancers, our findings strongly FIGURE 5. The FOXM1/IPO7/GLI1 axis promotes cell proliferation, migration, and invasion. A, U87 cells were transfected with the indicated siRNA and/or plasmids, and their proliferation was analyzed using a 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay. Data are presented as the mean Ϯ S.D. values from three independent experiments. siCTRL, control siRNA. ***, p Ͻ 0.001. B, left: U87 cells were transfected with the indicated siRNA and/or plasmids, and their growth was examined using a monolayer colony formation assay. Scale bars, 10 mm. Right: results of quantitative analysis of colony numbers shown as the mean Ϯ S.D. values from three independent experiments. ***, p Ͻ 0.001. C, U87 cells were transfected with the indicated siRNA and/or plasmids, and their migration was detected using a wound healing assay. The cell motility was quantified by measuring the distance between the invading front of cells in six randomly selected microscopic fields for each condition and time point. The degree of motility is expressed as the percentage of wound closure as compared with the zero time point. The mean Ϯ S.D. values from three independent experiments are presented. **, p Ͻ 0.01. D, left: U87 cells were transfected with the indicated siRNA and/or plasmids and subjected to an in vitro invasion assay. Scale bars, 200 m. Right: quantitative analysis of invasive U87 cells shown as the mean Ϯ S.D. results from three independent experiments. ***, p Ͻ 0.001.
suggest that targeting FOXM1 is a therapeutic strategy for GBM.
Author Contributions-J. X. conceived and designed the project, performed most of the experiments, and wrote the manuscript. A. Z., C. T., Y. W., and H. T. L. helped with some experiments. W. L. and K. X. provided reagents and advice. S. H. supervised the study.