Construction of shRNA plasmids

KPNA2 cDNA was cloned into the Lenti-OE vector (Genechem, Shanghai, China) to generate KPNA2-overexpressing lentiviral vectors. The shRNAs were cloned into GV492 vector carried by Lenti-OE (Genechem, Shanghai, China), and the knockdown targeted sequences was as follows: KPNA2, 5′ -ATTTACAGTGCCCTGGTTG-3′; c-myc, 5’-ATGTCAAGAGGCGAACACA-3′; E2F1, 5’-TAGATC CGATCCAGCTCAGTGACA-3′, scramble, 5′- TTCTCCGAACGTGTCACGT -3′. The U87 and U251 cells were infected by the lentiviruses, and incubated for 24 h.

Western blot assay

Cells were lysed and sonicated in the buffer containing 1% Triton X-100, 10 mmol/L TrisHCl (pH 7.4) and proteases/phosphates inhibitors (Roche Diagnostics, USA), electrophoresed in 10% SDS-PAGE gel electrophoresis, and transferred to a polyvinylidene difluoride membrane (PVDF) probed with primary antibodies to KPNA2. Subsequently, the membrane was probed with horseradish–peroxidase conjugated secondary antibodies. Then the blots were detected and visualized by a chemiluminescence detection system (Pierce, Rockford, IL, USA). The primary antibodies were used: anti-KPNA2 (Abcam, Cambridge, MA, USA), anti-c-myc(Abcam, Cambridge, MA, USA), anti-histoneH3 (Boster, Pleasanton, CA, USA); anti-E2F1 (Abcam, Cambridge, MA, USA), and anti-GAPDH (Boster, Pleasanton, CA, USA).

Immunohistochemistry (IHC)

Paraffin-embedded glioblastoma blocks were cut into 5 μm thick sections and then subjected to antibody specific for KPNA2 (Abcam, Cambridge, MA, USA) at 4 °C for 12 h, followed by secondary antibody (ZSGB. BIO, Beijing, China) applied to tissues at 37 °C for 30 min. The staining intensity was classified as four grades: 0(no expression or no positive staining), 1(1–10% positive cells or light yellow), 2(11–50% positive cells or brownish yellow), 3(51–75% positive cells or brown), 4(> 75% positive cells or dark brown).

Quantitative real-time PCR (qRT-PCR)

Total RNA was isolated from different cell lines using Trizol reagent (Invitrogen, USA). QRT-PCR was performed using a Real-time PCR System (Applied Biosystems, USA) with a SYBR Green PCR kit (TaKaRa, Japan) according to the manufacturer’s instructions. The primer sequences were as follows:

KPNA2, 5’-ATTGCAGGTGATGGCTCAGT-3′ (forward) and 5’-CTGCTCAACAGCATCTATCG-3′ (reverse);

c-myc, 5’-GGAGCCTATTCTGCCCATTT-3′(forward) and 5’-CGAGGTCATAGTTCCTGTTGGTG-3′(reverse);

HK2, 5’-TCCGTAGTGGGAAAAAGAGAA-3′ (forward) and 5’-GACAATGTGATCAAACAGCTC-3′ (reverse);

PFK1, 5’-GGTGTACAAGCTTCTAGCTC-3′ (forward) and 5’-CAAGTTTAGAGCCACCTTGG-3′ (reverse);

PKM2, 5’-CCACTTGCTGTGCCAAATGGA-3′ (forward) and 5’-GAAGGACTTTACCTTCCAGGA-3′ (reverse);

PDK1, 5’-GAAGCAGTTCCTGGACTTCG-3′ (forward) and 5’-ACCAATTGAACGGATGGTGT-3′ (reverse);

GLS, 5’-CGACACTGGCTAATGGTGGT-3′ (forward) and 5’-TGCAGGAAGACCAACATGGAA-3′ (reverse);

SLC1A5, 5’-CCTGGATCATGTGGTACGCC’ (forward) and 5’-GAAGCGGTAGGGGTTTTTGC-3′ (reverse);

and GAPDH, 5’-GCACCGTCAAGGCTGAGAAC-3′ (forward) and 5’-TGGTGAAGACGCCAGTGGA-3′ (reverse).

Enzymatic activities of PFK, PKM and HK

U87 and U251 cells were seeded in six-well plates at a density of 2.0 × 10⁵ per well, and cultured for 12 h before transfection. The cells were then digested by 0.08% EDTA trypsin after 48 h. The enzymatic activities of PFK, PKM and HK were detected by the respective enzyme activity kit (Jiemei Genetech, Corporation, China) according to the manufacturer’s instructions.

Detection of lactate levels

A lactate assay kit (Jiancheng Corporation Ltd., Nanjing, China) was used to measure the lactate level in the culture medium 24 h after transfection. The lactate was oxidized by lactate dehydrogenase, and generated a product that could interact a probe to produce colors (an absorption maximum at 492 nm). Total viable cell number was used for normalization.

Pyruvate dehydrogenase (PDH) colorimetric assay

Cells were seeded and dissolved by 100 μl ice cold PDH Assay Buffer. Mitochondria from the cultured cells were isolated using a Biovision’s Mitochondria Isolation Kit. According to the protocols of Pyruvate Dehydrogenase Activity Colorimetric Assay Kit(Biovison, San Francisco, USA), 0, 2, 4, 6, 8 and 10 μl of 1.25 mM NADH Standard were added into the wells of 96-well plate and Color densities were measured (OD 450 nm) using a microplate reader (Tecan, Infinite M200, Switzerland) for 10–60 min at 37 °C. Then plot of the NADH Standard Curve used the protein concentration as standardization.

Colorimetric assay on glucose uptake

Cells were seeded in a 96-well plate at a density of 1000 per well and incubated for another 4 days prior to use. The assay was performed using a Glucose Uptake Colorimetric Assay Kit (Biovison, San Francisco, USA) according to the manufacturer’s protocols. The absorbance at 412 nm was measured using a microplate reader (Tecan, Infinite M200, Switzerland) at 37 °C every 5 min until the 100 pmol standard reached 1.8.

Flow cytometry (FCM) analysis

The U87 cells were seeded in a 6-well plate at a density of 2.0 × 10⁵/ well. 48 h after transfection, the cells were digested by 0.08% trypsin and washed with ice-cold PBS for three times. After that, cells were marked with Alexa Flour®647-GLUT1 antibody (Abcam, Cambridge, USA), and the emission was measured using a FACS Calibur cytometer (Beckman coulter, KBB, CA, USA).

Extracellular acidification and oxygen consumption rates (ECAR and OCR)

Cells were seeded in the 24-well plates and incubated for 24 h. The culture medium then was replaced with running buffer 30 min prior to the assay. The mitochondrial inhibitor oligomycin (1 mM) was added into the plates after 20 min and the uncoupler FCCP (300 mM) was added after another 20 min. The ECAR and OCR were then determined by a Seahorse XF24 extracellular Flux analyzer.

Cell viability assay

Cell suspension was incubated in the 96-well plates at a density of 5000 cells per well and placed in a humidified incubator (37 °C, 5% CO2) for 24 h. Then 10 μL of the reageant cell-counting kit8 was added into each well of the plates (CCK-8, BestBio, Shanghai, China). Two hours later, the spectrometric absorbance was measured at 450 nm using a microplate reader (Tecan, Infinite M200, Switzerland).

In vivo tumor formation

The glioblastoma cells (1× 10⁵) were injected into 4 week-old male athymic Nu/Nu mice (Cancer Institute, Medical Science, Chinese Academy) according to the Guide for the Care and Use of Lab Animals. The research was approved by the Animal Care and Use Committee of Shandong University. Transduced cells were subcutaneously injected into the right frontal lobes of the mice. The nude mice were anesthetized and dissected 4 weeks later. Tumor volumes were monitored and overall survival rate was analysed. The tumor volumes were determined as follows: Maximal and minimal diameter of the tumor was measured on the H&E stained sections. Tumor volume was calculated from the formula: Tumor volume = (max. diameter) × (min. diameter) 2 × 0.5 [22, 23].The expressions of PFK1, PKM2 and HK2 were detected by qRT-PCR.

Subcellular fractionation

The cells were harvested and washed with PBS for two times, and subjected to subcellular fractionation by a Minute TM Cytoplasmic and Nuclear Fractionation kit (Invent Biotechnologies, Inc. Eden Prairi. US) referring to the manufacturer’s protocols. The efficacy of fractionation was tested by immunoblotting using anti-Histone H3 antibody (Boster, Pleasanton, CA, USA) as the nuclear protein control, while using anti-MEK1/2 antibody (Abcam, Cambridge, MA, USA) as the cytosolic protein control.

Immunofluorescence

The glioblastoma cells were fixed in 4% paraformaldehyde for 15 min at room temperature,rinsed in 0.3% Trion X-100 for 5 min and were placed on the coated coverslips After blocked with goat serum(ZSGB. BIO, Beijing, China) for 30 min, they were incubated with the primary antibody to c-myc and E2F1 (Abcam, Cambridge, MA, USA)at 37 °Cfor 1 h. Then the coverslips were washed with PBS and stained with Alexa Fluor 594--labelled secondary antibodies followed by the staining with 4′,6-diamidino-2-phenylindole (DAPI). A fluorescent microscope was utilized at last to watch the fluorescence.

Co-IP assay

Firstly, the cells were seeded, transfected and lysed with a gentle detergent, producing a mixture containing the bait–target complexes and other irrelevant proteins. Then a capture antibody together with protein A/G plus agarose was added into the mixture to bind the bait protein specifically after the antibody was crosslinked to the agarose by using the DSS crosslinker, the sample was prepared for SDS-PAGE analysis. All the procedures were referred to a Pierce Crosslink Immunoprecipitation Kit (Thermo Fisher scientific, invitrogen, German).

Luciferase assay

The U87 and U251 cells were seeded in the 24-well plates before transfection. 48 h later, cells were lysed by PLB (1×, 100ul) and put on an shaker (QILINBEIER, China) at 100 rpm for 30 min and then placed in − 80 °C refrigerator for at least 12 h. Firefly luciferase signals and renilla luciferase (internal reference) were detected by a dual-luciferase reporter assay kit (Promega, Madison, WI, USA.) referring to the manufacture’s protocols.

High level of KPNA2 in the glioblastoma cells enhanced glycolytic metabolism

To investigate the role of KPNA2 that plays in the energetic metabolism of gliomas, considering the different P53 status may be an important factor in cellular metabolism. We separately transfected a lentiviral vector of KPNA2 shRNA, wild type KPNA2 or their comparative controls into the U87 (harboring wild-type P53) and U251 (harboring mutant P53) glioblastoma cell lines. The results of quantitative real-time polymerase chain reaction (qRT-PCR) and western blot indicated that KPNA2 was notably upregulated in the KPNA2 transfectants and depleted in the KPNA2-shRNA transfectants (Fig. 2a, Additional file 1: Figure S1A).

Open in a separate window

Fig. 2

KPNA2 promoted the glycolytic metabolism in the glioblastoma cells. a Levels of KPNA2 were analyzed by qRT-PCR and immunoblotting in the U87 glioblastoma cells transfected with the lentiviruses expressing small hairpin RNAs of KPNA2 (shKPNA2), wild-type KPNA2 (KPNA2), or a vector with scrambled nonspecific shRNAs(SC). NC is for the original U87 cells that without any handles, GAPDH served as a loading control. b mRNA levels and c enzymatic activities of HK2, PKM2 and PFK1 were determined in the U87 cells transfected with KPNA2-shRNA(siKPNA2), scrambled shRNA(SC) or wild-type KPNA2(KPNA2). Each bar represented the mean ± s.d. from three independent experiments. *P <  0.05, ***P <  0.001. d Lactate production was measured in the indicated cells. Data were presented as the mean ± s.d. from three independent experiments. **P <  0.01. e MRNA levels of MCT1 and MCT4 in the U87 cells. Data were presented as the mean ± s.d. from three independent experiments. *P < 0.05, **P < 0.01. f Relative deoxyglucose uptake was measured in the indicated cells. Each bar represented the mean ± s.d. from three independent experiments. ***P < 0.001. g The expression of GLUT-1 was detected by flow cytometry in the indicated cells, cells were marked with Alexa Flour®647-GLUT1 antibody and the statistic analyze were followed. Data were presented as the mean ± s.d. from three independent experiments. **P < 0.01. ***P < 0.001

To assess the glycolytic activity of these cells, we first detected the mRNA levels and enzymatic activities of phosphofructokinase (PFK), pyruvate kinase (PKM) and hexokinase (HK), which are key enzymes in the glycolytic pathway. QRT-PCR showed that overexpression of KPNA2 enhanced the expressions of PFK2, PKM1. On the contrary, KPNA2-knockdown cells showed the opposite effects (Fig. 2b, Additional file 1: Figure S1B). Although there was no dramatic change associated with the mRNA of HK2, the enzymatic activities of all three enzymes were significantly enhanced when KPNA2 was up-regulated and decreased when KPNA2 was depleted (Fig. 2c, Additional file 1: Figure S1C). These data evidently suggested that KPNA2 might play an important role in promoting glycolysis of the glioma cells.

Lactic acid production is one of the key features of Warburg effect. We therefor measured the level of lactic acid in the culture medium conditioned by U87 and U251 cells. As the results shown in Fig. 2d, the lactate level in the culture medium was notably increased when KPNA2 was overexpressed and significantly decreased when KPNA2 was knockdown, consistent with the results of the glycolytic enzymes (Fig. 2d, Additional file 1: Figure S1D). To further investigate the lactate metabolism in these cells, we measured the expressions of MCT1 and MCT4 by q-PCR, which were responsible for lactate transport. The results showed that MCT1 was decreased in the KPNA2-knockdown cells, and increased when KPNA2 was overexpressed. In contrast, MCT4 showed differential expressions in the U87 and U251 cells. Specifically, MCT4 was decreased in the U87 cells and increased in the U251 cells respectively when KPNA2 was depleted (Fig. 2e, Additional file 1: Figure S1E).

Glucose uptake is an important biological process of energetic metabolism, especially in cancer cells. 2-deoxyglucose (2-DG) has been widely used because of its structural similarity to glucose. We tested the uptake of 2-DG by colorimetric assay. The results showed that overexpression of KPNA2 led to more absorption of 2-DG than the control cells (Fig. 2f, Additional file 1: Figure S1F). Additionally, we measured the plasma membrane expression of glucose transporter 1 (GLUT-1),which plays a key role in extracellular deoxyglucose uptake. As the data shown in Fig. 2e, in comparison to the control cells, GLUT-1 residing on the plasma membrane was substantially up-regulated in the cells overexpressed with KPNA2 (Fig. 2g, Additional file 1: Figure S1G). On the contrary, KPNA2-depleted cells showed lower levels of both GLUT-1 and 2-DG uptake (Fig. 2f, g; Additional file 1: Figure S1F, G).

KPNA2 regulated the mitochondrial respiration and cell proliferation in the glioblastoma cells

To examine the possible effects of KPNA2 on the oxidative phosphorylation (OXPHOS), we added oligomycin, a pharmacological inhibitor of electron transport, followed by FCCP, a mitochondrial uncoupler, to the U87 and U251 cells (Fig. 3a, Additional file 1: Figure S2A) and tested their oxygen consumption rates (OCRs). The results showed a decrease followed by an increase in the levels of OCRs in all the cell groups. Moreover, both the basal OCR and the calculated reserve capacity were decreased in the cells knock-down of KPNA2, and moderately increased in the KPNA2 up-regulated cells (Fig. 3b, c). On the other hand, the extracellular acidification rate (ECAR) was reduced in the KPNA2-depleted cells and elevated when KPNA2 was overexpressed, unsurprisingly consistent with the lactate production data in Fig. 2d(Fig. 3d, Additional file 1: Figure S2B).

Open in a separate window

Fig. 3

OXPHOS and cell growth were regulated by KPNA2 in the glioblastoma cells. a OCRs were determined in the U87 glioblastoma cells transfected with SC-shRNA, KPNA2-shRNA or wild-type KPNA2. Oligomycin (1 mM) and FCCP (300 mM) were sequentially added at 20th minutes and 40th minutes. b Basal OCR (original OCR minus OCR after addition of oligomycin and c reserve capacity (maximal OCR induced by minus basal OCR) were calculated. Results were presented as the mean ± s.d. from three independent experiments. *P < 0.05, **P < 0.01. d ECARs in the U87 cells were detected using the same methodology with OCRs. e Expressions of PDK1 were detected by immunoblotting in the U87 cells transfected with SC-shRNA, KPNA2-shRNA or wild-type KPNA2. f Enzymatic activities of PDH were determined in the U87 and U251cells transfected with KPNA2-shRNA, scrambled shRNA and wild-type KPNA2. Growth curves of the transduced cells were shown, cultured in the (g—U87, h—U251) glucose-containing (4500 mg/l) or (i—U87, j—251) glucose-free (w/o GLU) medium. Each bar represented the mean ± s.d. from three independent experiments

Pyruvate dehydrogenase (PDH), an important factor in the OXPHOS, stands at the crossroad of glycolysis and mitochondrial respiration. We first examined the expression of pyruvate dehydrogenase kinase 1 (PDK1), which phophorylates and inactivates pyruvate dehydrogenase (PDH). The results showed that the mRNA level of PDK1 was decreased in the KPNA2-overexpressed cells and increased in the cells knock-down of KPNA2 (Fig. 3e, Additional file 1: Figure S2C). We subsequently measured the enzymatic activity of PDH. Indeed, it showed significant reduction in the KPNA2-knockdown cells and concomitant increase in the cells overexpressed with KPNA2 (Fig. 3f).

Since cell growth is directly linked to the balance between oxidative and glycolytic metabolism, we next examined the growth kinetics of the U87 and U251 cells overexpressed or depleted of KPNA2. In comparison to the controls, the cells up-regulated with KPNA2 exhibited significantly increased growth kinetics, indicating a positive role of KPNA2 in cell proliferation (Fig. 3g, h). When we cultured the cells into the glucose-free medium, the growth rate of these cells slowed down dramatically (Fig. 3i, j). On the contrary, level of glucose had a minor effect on the growth kinetics of the glioma cells when KPNA2 was depleted, suggesting that the cells with higher level of KPNA2 relied on glucose as their major energy source.

To further investigate whether the cells with low or no expression of KPNA2 survived glucose restriction most likely through their capacity to metabolize other nutrients, we have investigated the expressions of GLS and SLC1A5, two key factors involved in glutamine uptake and catabolism. The results showed that both SLC1A5 and GLS were indeed increased in the U251 cells when KPNA2 was down-regulated and decreased when KPNA2 was overexpressed, evidently suggesting that KPNA2 might have multiple effects on the metabolic reprogramming to sustain intracellular stability (Additional file 1: Figure S2D).

Targeting KPNA2 in the glioblastoma cells decreased tumorigenic capacity and increases survival of mice bearing xenografts

Considering the in vitro involvement of KPNA2 in glioma cell proliferation, survival and metabolism, we extended our study to determine the impact of KPNA2 depletion on tumorigenic capacity of glioma cells in vivo. When the KPNA2-knockdown U87 cells implanted into the right frontal lobes of immunocompromised mice, apparent decrease in tumor formation (Fig. 4a, b) and subsequent increase in the survival time of the tumor-bearing mice were observed (Fig. 4c).

Open in a separate window

Fig. 4

KPNA2 was associated with the metabolic reprogramming in the glioma xenografts. The U87 cells transfected with small hairpin RNAs of KPNA2 or scrambled nonspecific shRNAs were implanted into the right frontal lobes of the immunocompromised mice. a Representative H&E staining images of mouse brains at 42 days after inoculation were shown. Asterisk indicates tumor (scale bar, 1 mm) b The estimated tumor volumes at that time. Data were presented as the mean ± s.d. from three independent experiments. **P < 0.05. c Curve graph indicated survival time of the xenografted mice. P < 0.05 is considered statistically significant. d The relative expressions and e the activities of key enzymes involved in the glycolytic metabolism of the xenograft tumors were determined. Data were presented as the mean ± s.d. from three independent experiments.*P < 0.05, **P < 0.01

In addition, we tested the mRNA levels of HK2, PFK1 and PKM2 in the xenograft tumors by qRT-PCR analysis. The results showed that except for HK2, depletion of KPNA2 resulted in dramatically lower levels of the respective enzyme (Fig. 4d). Their enzymatic activates were decreased as well (Fig. 4e), indicating that KPNA2-knockdown xenograft tumors had lower level of glycolytic metabolism. In a word, KPNA2 might play an important role in the metabolic transformation of glioblastoma cells, which exerts vital effects on tumor initiation and progression.

C-myc mediated the role of KPNA2 in regulating the glycolytic metabolism of glioma cells

C-myc is one of major oncogenic factors involved in different human cancers. It acts as a transcriptional factor to activate genes that permit metabolic adaptation, such as HK2, PFK1, PKM2 and GLUT1. Considering the broad involvement of KPNA2 in the metabolic reprogramming and tumor growth of glioblastomas, we hypothesized that the role of KPNA2 in glioblastomas may be linked with c-myc.

To verify this hypothesis, we first measured the subcellular localization of c-myc in the KPNA2 depleted or overexpressed cells. Both nuclear/cytosol fractionation and immunofluorescence (IF) analyses (Fig. 5a, b) indicated that more c-myc was retained in the cytoplasm when KPNA2 was depleted. Additionally, co-immunoprecipitation (CO-IP) analysis demonstrated that KPNA2 interacted directly with c-myc in the U87 cells (Fig. 5c). A comparable result was observed in the U251 glioblastoma cells (data not shown), further suggesting that KPNA2 played an important role in mediating the transportation of c-myc into the nucleus.

Open in a separate window

Fig. 5

KPNA2 mediated both the translocation and the transcriptional expression of c-myc. The subcellular distribution of c-myc in the KPNA2-shRNA and control transfected cells by nuclear/cytosol fractionation (a) and immunofluorescence (b) was shown. Error bars indicated the mean ± s.d. of three independent views. MEK1/2 was used as the cytoplasmic control and Histone H3 as the nuclear control. *P < 0.05, ***P < 0.001. c Co-immunoprecipitation analysis was presented between KPNA2 and c-myc in the U87 cells. d Expressions of c-myc in the U87 cells knockdown or overexpressed of KPNA2 were determined by immunoblotting. e The U87 and U251 cells were transfected with pGL2-c-myc, a vector encoding KPNA2-shRNA, wild-type KPNA2 or its comparative control, and pSV-Kerilla. Values in graphs represented the mean of Fluc:Rluc activity±s.d. performed in triplicate. **P < 0.01

We additionally examined the expressions of c-myc in these cells. Interestingly, western bolt analysis indicated that protein level of c-myc was decreased in the cells knockdown of KPNA2 and increased in the cells when KPNA2 was overexpressed (Fig. 5d). When we introduced a luciferase reporter into these cells, a significant increase in the luciferase activity was coupled with up-regulation of KPNA2, whereas knockdown of KPNA2 in the U87 and U251 cells reduced c-myc transcriptional activity (Fig. 5e). These results evidently suggested that KPNA2 might play dual effects on both translocation as well as expression of c-myc.

KPNA2 upregulated the transcriptional expression of c-myc by mediating the translocation of E2F1 into the nucleus

In consideration of KPNA2 role in c-myc transactivation, we applied ChIP assay to the glioma cells. However, no KPNA2 was detected on the promoter region of c-myc compared to the negative controls (data not shown). We therefore speculated that there may be other pathways/factors that mediated the transcriptional regulation of c-myc by KPNA2. E2F1, another cargo protein of KPNA2 [24], is a key factor that participates in regulating cell cycle [25], proliferation [26], apoptosis [27], senescence [28], and DNA-damage response [29]. It has been reported that E2F1 up-regulated c-myc on the transcriptional level. On the other hand, pRb down-regulated c-myc by sequestering E2F1 [30, 31]. In view of this, we hypothesized that in glioma cells, E2F1 may be involved in the transcriptional regulation of c-myc by KPNA2. To test this hypothesis, we first examined the effects of KPNA2 exerting to E2F1 in the glioma cells. The results showed translocation of E2F1 was significantly decreased when KPNA2 was knockdown by both the nuclear/cytosol fractionation and immunofluorescence analyses (Fig. 6a, b). Direct interaction between KPNA2 and E2F1 was further confirmed by the CO-IP assay (Fig. 6c). We subsequently transfected E2F1-shRNA vector into the U87 cells overexpressed with KPNA2. The results showed that the originally up-regulated c-myc on the levels of both mRNA and luciferase activity was significantly decreased when E2F1 was depleted (Fig. 6d, e).

Open in a separate window

Fig. 6

Subcellular redistribution of E2F1 caused by knockdown of KPNA2 contributed to the deregulation of c-myc. The subcellular distribution of E2F1 in the KPNA2-shRNA and control transfected cells by nuclear/cytosol fractionation (a) and immunofluorescence (b) was shown. Error bars indicated the mean ± s.d. of three independent views. MEK1/2 was used as the cytoplasmic control and Histone H3 as the nuclear control. *P < 0.05, **P < 0.01. c Co-immunoprecipitation analysis was presented between KPNA2 and E2F1 in the U87 cells. d Expressions of c-myc were analyzed by immunoblotting in the KPNA2-overexpressed U87 cells silenced with E2F1 or not. e The KPNA2-overexpressed U87 and U251 cells were transfected with pGL2-c-myc, a vector encoding E2F1-shRNA or its comparative control, and pSV-Kerilla. Values in graphs represented the mean of Fluc:Rluc activity±s.d. performed in triplicate. **P < 0.01

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.