Human glioblastoma U251 cells were obtained from BeNa Culture Collection (BNCC) (Henan, P.R. China). Cells were cultured in RPMI 1640 Medium (Gibco) with 10% FBS, and maintained in a humidity incubator (Thermo Fisher Scientific, USA) at 37°C with 5% CO2. LipoFectMax™ 3000 Transfection Reagent (ABP Biosciences) was used to transfect the U251 cells.

TPL was purchased from PUSH Bio-technology (Chengdu, P.R. China), dissolved in DMSO to a storage concentration of 10 mM and stored at −20°C. Serum-free DMEM was used to dilute TPL before the experiments. Puromycin aminonucleoside (1.5 μg/ml) was obtained from MedChemExpress.

Total RNA was extracted from the cells using the TRIzol kit (Takara, Dalian, China) as described previously (18). 1 μg RNA was used to transcribe cDNA using the HiScript II Q RT SuperMix (R223-01, Vazyme). Quantitative reverse transcription polymerase chain reaction (qRT-PCR) was performed using a qTOWER³84G(analytikjena). The results were analyzed using the 2^-ΔΔCt method and GAPDH as an internal reference. The primers used in the study were as follows:

PROX1Forward Primer: AAAGGACGGTAGGGACAGCAT, Reverse Primer: CCTTGGGGATTCATGGCACTAA; GAPDH Forward Primer: ATGGGGAAGGTGAAGGTCG, Reverse Primer: GGGGTCATTGATGGCAACAATA.

Western blot analysis was carried out as described previously (19). Briefly, U251 cells were cultured and treated as indicated demand. The total protein was extracted using RIPA lysis buffer (Beyotime), and the cell lysates were subjected to western blot. The polyclonal antibodies (pAb) against PROX1, Bax, P21, and GAPDH were from Proteintech. The anti-MMP2 and MMP9 antibodies were from Affinity Biosciences.

The transcription activation activity of TPL was determined using luciferase assays with a PRXO1-Luc reporter, which was cloned as before (20). U251cells were cultured in 24 wells and co-transfected with PROX1-LUC (0.25mg) and pRL-TK plasmid (10ng, internal control). 24 hours later cells suffered to 0, 50, 100, or 200 nM TPL treatments. 24h later, the cells were lysed, and the lysates were used for luciferase assay.

Cell counting kit-8 (CCK-8) assay (Beyotime) was performed to evaluate the U251cell viability as before (18). Briefly, 1x10⁵ U251cells were seeded in a 96-well plate (NEST, China) for 24 h, and then treated with different concentrations of TPL for another 24h. The CCK-8 solution (10 μl) was added to the culture medium of each well. After that, the cell plate was incubated for 4 hours at 37°C, and the absorbance of each well at 450 nm was recorded using a microplate reader (Spark, TECAN).

The effect of TPL on the cell cycle in glioblastoma was analyzed using the Cell Cycle Staining Kit (Beyotime). And the cell cycle assay was performed as described previously (21). After treatment with TPL for 24h, U251 cells were trypsinized, fixed in 70% ice-cold ethanol for 24 h and washed with PBS, then incubated with 100 mg/mL RNase and 4 mg/mL propidium iodide (PI) in 500 μl PBS. The cell cycle phase was analyzed using a cytoflex (Beckman Coulter).

The effects of TPL on glioma cell apoptosis were measured using the Annexin V-FITC/PI apoptosis kit (Beyotime). U251 cells were treated with TPL or TPL + puromycin (1.5μg/mL) for 24h. Then the cells were washed, digested with trypsin without EDTA, and collected for staining. 200 μl staining buffer with Annexin V-FITC and PI was added to the cells and incubated at room temperature for 15 min, and analyzed with a cytoflex (Beckman Coulter). We calculated the percentage of apoptotic cells by the percentage of cells in the quadrant of AnnexinV-FITC-positive, PI-positive, or PI-negative.

U251cells (80% confluence) were plated into 12-well culture plates (NEST, China), and cultured overnight. A scratch was created artificially using the tip of a 10 μl pipette gun in the middle of the adherent cells. Finally, the medium with different concentrations of TPL was added to the cells. Phase contrast images were taken at 0, 6 and 24h after scratching using an Olympus light microscope with a 10x objective lens. Then the relative migration cells and width were measured using Image J.

For migration experiments, U251cells were treated with different concentrations of TPL for 24hr. Then cells were harvested and re-cultured in the matrigel-coated (Biozellen, rehydrated with medium, 1:50) transwell chamber (8mm, Corning) with 200 µL serum-free DMEM, while the lower part filled with 500 µL DMEM containing 10% FBS. After Incubating for 12 hr, the cells which were attached inside of the filter were wiped with a cotton swab, while the invaded cells on the lower surface were fixed with methyl alcohol for 15 min, and stained with 0.1% crystal violet for 20 min. Then PBS was used to wash the cells until washing out the excess dye. The cells were counted under the Olympus microscope at a magnification of 100x and quantified by Image J.

Cell proliferation was monitored using the iClick™ EdU Andy Fluor™ 647 Imaging Kit (ABP Biosciences) following the manufacturer’s instructions. U251 cells were cultured in 24 wells and treated with 0, 50, 100 nM TPL and 100 nM TPL+pcDNA3.1-PROX1 expression plasmid for 24 h, and incubated with 10 μM EdU for 12 h, fixed with 4% paraformaldehyde for 15 min at 37°C, permeabilized in 0.3% TritonX-100 and washed twice with PBS containing 0.3% BSA. Then, 100 μl of Apollo 567 stain reaction buffer was added for 30 min and washed twice with 0.3% BSA. The cells were stained with 100 μl of DAPI (dilution at 1: 2000) for 20 min at room temperature and imaged under the Olympus microscope at a magnification of 400x. The rate of cell proliferation was calculated with the percentage of EdU-positive red cells over DAPI-stained cells (blue) × 100%.

U251 cells were cultured in 12 wells and treated with 0, 50, and 100 nM TPL, and one well transfected with pcDNA3.1-PROX1 plasmid and 100 nM TPL and harvested as single-cell suspension. We resuspended 1 × 10⁴ cells in 500 μl of a 1:1 mix of medium and Matri-gel (Biozellen), and 100 μl drops were placed on precooled 24-well plates and allowed to solidify for 5 minutes at 4°C. Another 1ml of fixing solution was added for 15 minutes, and then 1ml of the medium was added and cultured in an incubator for 1-2 weeks. Plates were stained with crystal violet (Beyotime) and washed to remove excess crystal violet. The number of colonies was scored with Image J.

All experiments were repeated independently at least three times and GraphPad Prism 6.0 software was used for statistical analyses. All data were presented as mean ± SD. The Student’s t-test and one-way ANOVA were employed to compare differences between groups. p <0.05 was considered to be statistically significant.

PROX1 has been identified as an independent prognostic factor for survival in patients with World Health Organization grade II gliomas (22), In high−grade malignant astrocytic gliomas, PROX1 is highly expressed (15). Additionally, we screened publicly available clinical data of LowerGrade Glioma (LGG) from The Cancer Genome Atlas (TCGA; www.cancergenome.nih.gov), and found that PROX1 expression levels are significantly increased in glioblastoma tumors (n=516) as compared to healthy tissue (n=698) ( Figure 1A ). We also performed the Kaplan-Meier analysis in 516 Lower grade glioma patients from TCGA via the R software and revealed that high relative expression of PROX1 is associated with lower survival rates ( Figure 1B ). These observations are in agreement with the earlier studies that PROX1 acts as a tumor promoter in the pathogenesis of GBM (17).

To initially address the effect of TPL on PROX1, we treated the U251 cells with different concentrations of TPL at 0, 50, 100, or 200 nM, and detected both the mRNA and protein levels of PROX1 using qRT-PCR and Western blot. As presented in Figure 2A , 50, 100, or 200 nM TPL treatments significantly reduced the mRNA levels of PROX1, but 20 nM of TPL had no obvious effect on the expression of PROX1. Figure 2B indicates that TPL markedly inhibited the protein levels of PROX1 in a dose-dependent manner. As reported previously, TPL Inhibits the global gene transcription in cancer cells by degradation the RNA polymerase II (Rpb1) (23). Thus, we hypothesized that TPL might transcriptionally depress the expression of PROX1. To confirm this hypothesis, we performed a dual-luciferase reporter assay in U251 cells using the PROX1 promoter plasmid (PROX1-LUC), which cloned ~ 3100bp of the 5’UTR sequence above the PROX1 transcription start site into the pGL3-basic vector. As shown in Figure 2C , 20, 50, or 100 nM TPL treatments obviously depress the luciferase activity of PROX1 promoter. These results indicate that TPL depresses the PROX1 expression.

To address the role of TPL in Glioblastoma tumors, we detected the cell viability, apoptosis and growth rate using CCK8 assay, FITC-annexin V/PI staining, and soft agar colony formation assays, respectively. For the CCK8 assay, U251 cells were pre-cultured in 96-well plates and treated with TPL at 0, 50, 100, or 200 nM, then calculate the cell viability at 0, 6, 12, and 24h. As shown in Figure 3A , 50, 100, or 200 nM TPL treatments significantly inhibited the viability of the U251 cells in a dose-dependent manner, and its maximal effect was achieved with 100 nM. The soft agar colony assay showed a great degradation of colony-forming ability following 50 or 100 nM TPL treatments ( Figure 3C ). Furthermore, we did the apoptosis assay in U251 cells using TPL at 0, 20, 50, or 100 nM with or without puromycin (1.5μg/ml) for 24 h. As shown in Figures 3D, E , only 50, or 100 nM TPL treatments induced the apoptosis rate of U251 cells at 22% or 33%. However, when combined with puromycin, 30, 50, or 100 nM TPL treatments significantly increased the number of cells undergoing apoptosis at 63%, 79% or 93% ( Figures 3D, F ). Given the critical role of PROX1 in glioma progression and the inhibitory effect of TPL on PROX1 expression, it is possible that PROX1 play an essential role in TPL-mediated glioma inhibition. Then we transfected the U251 cells with PROX1 plasmid for 12 h, and treated with 100 nM TPL for another 24h, then cells were collected for CCK8, apoptosis and soft agar colony assays. And we found that PROX1 partially rescues cell viability ( Figure 3B ), the apoptosis ( Figure 3D ) and growth inhibition ( Figure 3C ) induced by TPL.

To further confirm the effect of TPL on the growth of glioblastoma, we performed the cell cycle and proliferation assay in U251 cells, respectively. The cell cycle assay revealed that 50 or 100 nM TPL treatments obviously arrest the cell cycle at G1/S phase, and the over-expression of PROX1 could partially abolish the cell cycle arrest of TPL ( Figure 4 ). In addition, the proliferation assay showed that 50 or 100 nM TPL treatments markedly inhibit the proliferation rate (EDU⁺ cells) of U251 cells, and PROX1 restore the proliferation inhibition induced by TPL ( Figure 5 ).

Glioblastoma cells are highly aggressive, which is an important factor contributing to the poor prognosis of GBM (24). Therefore, we investigated the effect of TPL on the invasiveness of U251 cells. For migration assay, U251cells were cultured in 12 wells, and treated with TPL at 0, 50, 100 nM or 100 nM+PROX1 for 24 h, then cells were suffered to wound healing assay and graphed at 0, 6 and 24h. As shown in Figure 6 , comparing to control group, 50 or 100 nM TPL significantly depress the migration cells ( Figures 6A, B ) and migration width ( Figures 6A, C ) of Glioblastoma cells. However, PROX1 enables the glioma cells to regain the migratory capacity. For transwell assay, the U251cells were prepared with TPL at 0, 50, 100, 200, 300nM or 100 nM+PROX1 for 24h, and cells were harvested, and re-cultured in Matri-gel coated chamber with 200 µL serum-free DMEM, while the lower part were added 500 µL DMEM containing 10% FBS and indicated concentration of TPL for 12h, then stained with crystal, washed with PBS and graphed. The results suggested that 50, 100, 200 or 300nM TPL dramatically reduce the invasive cells, and PROX1 make the U251 cells regain the invasive capacity ( Figures 7A, B ).

PROX1 was reported to promote the GBM progression via inducing the expression of matrix metallopeptidase 9(MMP9), which were the downstream targets of the NF−κB pathway (17). Our previous results have showed that TPL inhibits the cell cycle, proliferation, migration and invasion, and promote apoptosis of glioma cells. Thus, we detected the expression of the cell cycle related genes P21, and the apoptosis associated gene Bax, the invasive gene MMP2 and MMP9. As shown in Figure 8 , 50, 100, 200 or 300 nm TPL significantly induced the expression of P21( Figures 8A, C ), Bax ( Figures 8A, B ), and depressed the expression of MMP2 ( Figures 8A, D ) and MMP9 ( Figures 8A, E ). In addition, ectopic expression of PROX1 could rescue the effect of TPL on the expression of P21, Bax, MMP2 and MMP9 ( Figures 8A–E ). Those results suggest that TPL blocks the progression of glioma cells via targeting PROX1.