The human astrocytoma cell lines SW1783 and U-87 were purchased from the American Tissue Culture Collection (ATCC, Rockville, MD, USA). Human full length PODXL cDNA was subcloned into pcDNA 3.1 expression vector. Human PODXL shRNA plasmid (RHS3979-98487921) and pLKO.1 empty plasmid (RHS4080) were purchased from Open Biosystems, Inc. (Huntsville, AL, USA). Anti-PODXL (3D3; 39-3800) antibody was purchased from Life Technologies (Carlsbad, CA, USA). Anti-MMP-9 (sc-13520), anti-Akt (ser473; sc-24500) and anti-P-Akt (ser473; sc-101629) antibodies were purchased from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA, USA). All the secondary antibodies were purchased from Jackson ImmunoResearch Laboratories, Inc. (West Grove, PA, USA). The DeadEnd™ Fluorometric TUNEL system was purchased from Promega (Madison, WI, USA). SuperFect^® transfection reagent was purchased from Qiagen (Valencia, CA, USA). Temozolomide, LY294002 (LY) and all the chemicals of reagent grade were purchased from Sigma (St. Louis, MO, USA).

The PODXL expression construct was transfected into SW1783 and U-87 cells using SuperFect^® transfection reagent according to the manufacturer’s instructions. Pools of stable transductants were generated via selection with G418 (800 μg/ml) using the manufacturer’s protocol. Lentiviral transduction was performed in the SW1783 and U-87 cells. Pools of stable transductants were generated via selection with puromycin (5 μg/ml).

Transwell^® cell invasion assays (Corning Life Sciences, Lowell, MA, USA) were performed. Briefly, Transwell^® cell culture chambers with 8-μm pore size (BD Biosciences, Bedford, MA, USA) for 24-well plates were coated with 50 μl Matrigel (BD Biosciences; 10 mg/ml; diluted 1:3 in RPMI-1640). The SW1783 and U-87 cells were seeded in the upper chamber at 5×10⁵ cells/well in RPMI-1640 serum-free medium. Complete medium (600 ml) was added to the lower chamber. The cells were treated with LY (50 μM) and allowed to migrate for 24 h followed by fixation and staining with crystal violet. The invasive cells were counted in 10 random fields/chamber under a microscope. Each experiment was repeated three times in triplicate.

Immunoblotting was performed with the respective antibodies. Briefly, cells were dissolved in 250 μl 2X SDS loading buffer (62.5 mM Tris-HCl, pH 6.8, 2% SDS, 25% glycerol, 0.01% bromophenol blue and 5% 2-mercaptoethanol), and incubated at 95°C for 10 min. Equal amount of proteins for each sample were separated by 10% SDS-polyacrylamide gel and blotted onto a polyvinylidene difluoride microporous membrane (Millipore, Billerica, MA, USA). Membranes were incubated for 1 h with a 1/1000 dilution of the primary antibody (1/10000 for 3D3 PODXL blotting), and then washed and revealed using secondary antibodies with horseradish peroxidase conjugate (1/5000, 1 h). Peroxidase was revealed using an ECL kit (GE Healthcare, Piscataway, NJ, USA). Proteins were quantified before being loaded onto the gel, and equal loading of extracts was verified by Ponceau coloration.

The TUNEL assay was performed using the DeadEnd™ Fluorometric TUNEL system following the instructions provided by Promega. Cells were treated with temozolamide (100 μM) in the presence or absence of LY (50 μM) for up to 8 h. Apoptotic cells exhibited a strong nuclear green fluorescence that was detected using a standard fluorescein filter. All the cells stained with DAPI exhibited a strong blue nuclear fluorescence. The slides were observed under a fluorescent microscope with relative apoptotic cells determined by counting the TUNEL-positive cells in five random fields (magnification, ×100) for each sample.

Statistical analyses were performed with SPSS for Windows 10.0. Data values were expressed as the mean ± standard deviation (SD). Comparisons of means among multiple groups were performed with one-way ANOVA followed by post hoc pairwise comparisons using the least significant difference method. The significance level of this study was set at a two-tailed P=0.05.

As shown in Fig. 1, SW1783 (grade III astrocytoma) cells had a relatively low constitutive expression of PODXL compared with U-87 (grade IV astrocytoma; gliobalstoma) cells. Thus, to investigate the functional role of PODXL in astrocytoma cells, we stably transfected SW1783 cells with PODXL expression vector to overexpress PODXL, and stably transduced U-87 cells with PODXL-shRNA to knock down PODXL. Compared with the controls, PODXL was overexpressed by >2.5-fold in SW1783 cells, and the endogenous PODXL level was knocked down ∼70% in U-87 cells. Selective PI3K inhibitor LY showed no effect on PODXL expression in either cell line.

PODXL has been reported to promote tumor cell invasion through MMPs (10). To investigate the effect of PODXL on astrocytoma cell invasion, we performed in vitro cell invasion assays and examined the MMP-9 expression level in the two cell lines. As shown in Fig. 2, PODXL overexpression in SW1783 cells increased cell invasion by ∼4-fold compared with that of the controls, and this increase was eradicated by LY. By contrast, PODXL knockdown in U-87 cells decreased cell invasion by ∼3-fold compared with the controls, and this was further decreased by LY treatment. Similar trends were observed with MMP-9 expression (Fig. 3). These results suggest that PODXL promotes astrocytoma cell invasion, potentially by upregulating MMP-9 expression in a PI3K-dependent manner.

PODXL reportedly promotes the metastatic potential of tumor cells. Since tumor cell survival is critical for metastasis (12), we next examined the effect of PODXL on astrocytoma cell survival against apoptotic stress. As shown in Fig. 4, PODXL overexpression and knockdown with or without LY treatment did not significantly alter the apoptosis rate of astrocytoma cells in normal culture conditions. Then, the cells were treated with 100 μM of temozolomide, an apoptosis-inducing chemotherapeutic agent used to treat high-grade astrocytoma. In SW1783 cells treated with temozolamide, PODXL overexpression significantly reduced cell apoptosis compared with that in the controls, and this reduction was reversed by LY (Fig. 5). In U-87 cells, PODXL knockdown significantly increased cell apoptosis in the presence of temozolamide. LY treatment further increased the apoptosis in the PODXL-knocked down cells (Fig. 5).These results suggest that PODXL promotes astrocytoma cell survival against temozolamide in a PI3K-dependent manner.

Since PODXL showed a protective effect on astrocytoma cells against temozolomide-induced apoptotic stress in a PI3K-dependent manner (Fig. 5), we investigated the effect of PODXL on the PI3K/Akt survival signaling pathway in astrocytoma cells. In SW1783 cells, PODXL overexpression significantly increased phosphorylation at serine 473 (ser473) of Akt, which is required for full activation of Akt (Fig. 6). LY treatment totally eradicated the effect of PODXL over-expression. In U-87 cells, PODXL knockdown decreased phosphorylation at serine 473 (ser473) of Akt by >3-fold compared with the control level (Fig. 6), which was further decreased by LY treatment. Taken together, these findings suggest that PODXL enhances the activation of the PI3K/Akt signaling pathway and, thereby, promotes astrocytoma cell survival against apoptotic stress.