The LN-229 and U87MG human glioma cell lines were purchased from the Cell Bank of the Chinese Academy of Sciences (Shanghai, China). They were cultured in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal bovine serum (FBS), 100 U/ml penicillin and l00 µg/ml streptomycin (Invitrogen; Thermo Fisher Scientific, Inc., Waltham, MA, USA) at 37°C in a humidified atmosphere of 5% CO₂.

Full-length human PROX1 cDNA (OriGene Technologies, Inc., Rockville, MD, USA) was cloned into the expression vector pcDNA3.1(+) (Invitrogen; Thermo Fisher Scientific, Inc.). The sequence identity of the pcDNA3.1/PROX1 construct was confirmed. The NF-κB-luciferase reporter gene plasmid (pNF-κB-luc) was purchased from Stratagene; Agilent Technologies, Inc. (Santa Clara, CA, USA) and the Renilla luciferase expression plasmid (pRL-TK) was purchased from Promega Corporation (Madison, WI, USA). A plasmid expressing dominant negative inhibitor of κBα (IκBα) containing serine to alanine mutations at positions 32 and 36 (pCMV-IκBαM) was purchased from Clontech Laboratories, Inc. (Mountainview, CA, USA).

At ~80% confluence, U87MG and LN-229 cells were transfected with 0.5 µg pcDNA3.1(+) empty vector or pcDNA3.1/PROX1 using Lipofectamine^® 2000 (Invitrogen; Thermo Fisher Scientific, Inc.) according to the manufacturer's protocol. Selection was performed with G418 (Sigma-Aldrich; Merck Millipore, Darmstadt, Germany) at 800 µg/ml and resistant clones were obtained and pooled two weeks after transfection. PROX1 overexpression was confirmed by western blot analysis. For the NF-κB reporter assay, vector or PROX1 stably-transfected cells or parental cells were seeded in triplicate at 1×10⁵ cells per well into 12-well plates 24 h prior to transfection. Cells were co-transfected with 0.5 µg pNF-κB-luc together with 0.1 µg pRL-TK. Luciferase activity was measured 24 h after transfection with the Dual-Luciferase Reporter assay system (Promega Corporation) according to the manufacturer's protocol. The relative luciferase activity was determined by normalization to Renilla luciferase activity. For inhibitory studies, vector or PROX1 stably-transfected cells were co-transfected with 0.5 µg pCMV-IκBαM or pCMV plasmid. At 24 h post-transfection, cells were collected and subjected to gene expression analysis and cell proliferation and invasion assays.

Cell proliferation was determined using the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay (Sigma-Aldrich; Merck Millipore). Briefly, U87MG and LN-229 cells (3×103) were seeded into 96-well plates and cultured for three days. MTT solution at a final concentration of 5 mg/ml was added to each well and incubated at 37°C for 4 h. Dimethyl sulfoxide was added to dissolve formazan crystals. The number of viable cells was determined by measuring absorbance at a wavelength of 570 nm using a microplate reader.

For the colony formation assay, U87MG and LN-229 cells (600 per well) were plated into 6-well plates and cultured for 10 days in DMEM containing 10% FBS. Cells were fixed with methanol and stained with 0.1% crystal violet (Sigma-Aldrich; Merck Millipore). Individual colonies consisting of 50 or more cells were counted.

A total of 12 male BALB/c nude mice at 5 weeks of age were obtained from the Animal Center of Zhengzhou University (Zhengzhou, China) and housed at 24°C under a 12-h light/dark cycle with free access to water and standard laboratory food. The experimental protocols involving animals were approved by the Institutional Animal Care and Use Committee of Zhengzhou University. Vector and PROX1 stably-transfected U87MG cells (2×106) were injected subcutaneously into the right flank of mice (n=6 per group). Tumor size was measured with calipers every seven days and the tumor volume was calculated. Animals were anesthetized 35 days after injection of tumor cells with intraperitoneal injection of ketamine (60 mg/kg) and xylazine (6 mg/kg), which were purchased from Shanghai First Biochemical Pharmaceutical Co., Ltd. (Shanghai, China). Xenograft tumors were extracted and imaged. Tumor volume and weight were measured.

Transfected and parental U87MG and LN-229 cells in serum-free DMEM were plated into the upper chamber of an insert (8-µm pore size) precoated with Matrigel (BD Biosciences, Franklin Lakes, NJ, USA). The lower chamber was filled with media containing 10% FBS. Cells were allowed to invade Matrigel-coated inserts at 37°C for 24 h. Subsequently, non-invaded cells were removed with a cotton swab. The cells that had invaded through the insert were stained with 0.1% crystal violet in methanol and imaged using an inverted light microscope. Cells in five random fields per insert were counted.

For the extraction of total cellular proteins, cells were washed with ice-cold phosphate-buffered saline and lysed with lysis buffer (50 mM 4-(2-hydroxyethyl)-1-piperazineethanesulphonic acid at, pH 7.5, with 150 mM NaCl, 1 mM EDTA, 10% glycerol and 1% Triton X-100) containing a Protease Inhibitor Cocktail (Sigma-Aldrich; Merck Millipore). Lysates were centrifuged at 12,000 × g for 10 min at 4°C to remove debris, and the supernatants containing total proteins were collected for further analysis. Nuclear and cytoplasmic sub-fractions were prepared using a Nuclear/Cytosol Fractionation kit (BioVision, Inc., Milpitas, CA, USA) according to the manufacturer's protocol. Equal quantities of protein samples (50 µg per lane) were separated by 10–12% SDS-polyacrylamide gel electrophoresis and transferred onto nitrocellulose membranes. Following the blocking of non-specific binding sites with 5% non-fat milk, the membranes were incubated with primary antibodies overnight at 4°C, followed by incubation with horseradish peroxidase-conjugated goat anti-mouse IgG (catalog no. sc-2005, Santa Cruz Biotechnology, Inc., Dallas, TX, USA; 1:2,000 dilution) or goat anti-rabbit IgG (catalog no. sc-2004, Santa Cruz Biotechnology, Inc.; 1:2,000 dilution) for 1 h. The signal was detected using an Enhanced Chemiluminescence system (GE Healthcare Life Sciences, Chalfont, UK). The primary antibodies (all at 1:500 dilution) used were as follows: Rabbit polyclonal anti-PROX1 (catalog no. ab191019), rabbit monoclonal anti-cyclin D1 (catalog no. ab137875), rabbit monoclonal anti-matrix metallopeptidase (MMP)-9 (catalog no. ab76003), mouse monoclonal anti-β-actin (catalog no. ab184220) and rabbit monoclonal anti-NF-κB p65 (catalog no. ab76311) obtained from Abcam, Cambridge, MA, USA; and mouse monoclonal anti-histone H3 (catalog no. 14269), rabbit monoclonal anti-phosphorylated (p)-IκBα (catalog no. 2859), rabbit monoclonal anti-IκBα (catalog no. 4812), mouse monoclonal anti-p-protein kinase B (Akt; catalog no. 12694), mouse monoclonal anti-Akt (catalog no. 2920), rabbit polyclonal anti-p-extracellular signal-regulated kinase (ERK; catalog no. 9101) and rabbit monoclonal anti-ERK (catalog no. 4695) purchased from Cell Signaling Technology, Inc., Danvers, MA, USA).

Data are presented as the mean ± standard deviation. Statistical analyses were performed with SPSS version 16.0 software (SPSS, Inc., Chicago, IL, USA) using one-way analysis of variance followed by the Tukey post-hoc test. P<0.05 was considered to indicate a statistically significant difference.

To determine the role of PROX1 in glioblastoma cell growth, PROX1 was overexpressed in U87MG and LN-229 cells. An MTT assay revealed that overexpression of PROX1 significantly enhanced the proliferation of U87MG cells following a 3-day culture, compared with empty vector-transfected cells (P=0.0342; Fig. 1A). The numbers of colonies were significantly increased in PROX1-overexpressing U87MG cells, as assessed by colony formation assays (P=0.0206; Fig. 1B). Similar results were observed in LN-229 cells (data not shown).

To further confirm the oncogenic role of PROX1 in vivo, tumor xenografts from vector- or PROX1-transfected U87MG cells were generated. Growth of PROX1 tumor xenografts was significantly increased compared with empty vector tumors (P<0.05; Fig. 1C). At 35 days, tumor weight was significantly greater in the PROX1 group compared with the empty vector group (P=0.0379; Fig. 1D).

Following this, the effect of PROX1 overexpression on the invasiveness of glioblastoma cells was assessed. Compared with vector-transfected U87MG cells, PROX1 overexpression significantly increased the number of invading cells following a 24-h incubation (P=0.0074; Fig. 2A). Similarly, PROX1-overexpressing LN-229 cells demonstrated a significantly greater invasive capacity compared with control cells (P=0.0125; Fig. 2B).

The involvement of signaling pathways in the oncogenic activity of PROX1 was subsequently investigated. Western blot analysis revealed that PROX1 overexpression induced the phosphorylation of IκBα and decreased the level of IκBα in U87MG and LN-229 cells (Fig. 3A). Nuclear accumulation of NF-κB p65 was detected in PROX1-overexpressing cells. However, the phosphorylation levels of Akt and ERK were not altered by PROX1 overexpression. To confirm the effect of PROX1 overexpression on NF-κB activation, vector or PROX1 stably-transfected U87MG and LN-229 cells were co-transfected with an NF-κB-dependent reporter gene and a pRL-TK control vector. PROX1-overexpressing U87MG and LN-229 cells demonstrated a 3.5- and 5.8-fold increase in NF-κB-dependent reporter activity, respectively, compared with vector-transfected controls (Fig. 3B). The expression of various NF-κB responsive genes, including cyclin D1 and MMP-9, was subsequently determined. Enforced expression of PROX1 resulted in a marked elevation in the protein expression levels of cyclin D1 and MMP-9, compared with the vector-transfected controls (Fig. 3C).

To determine whether the oncogenic role of PROX1 in glioblastoma cells is mediated via activation of NF-κB, a dominant IκBα mutant was transfected into PROX1-overexpressing U87MG cells, and cell proliferation and invasion were assessed. Western blot analysis confirmed the overexpression of the IκBα mutant in IκBαM-transfected cells, which was accompanied by reduced nuclear accumulation of NF-κB p65 (Fig. 4A). Notably, the delivery of the negative IκBα mutant significantly impaired the proliferation and invasiveness of PROX1-overexpressing cells (P<0.0001 and P=0.0042, respectively; Fig. 4B and C). These results support an important role for NF-κB activation in PROX1-mediated induction of cell proliferation and invasion.