Nucleolar and spindle associated protein 1 promotes the aggressiveness of astrocytoma by activating the Hedgehog signaling pathway

Primary normal human astrocytes (NHAs) were purchased from Sciencell Research Laboratories. The glioma cell lines U-118MG, U-87MG, A-172, SW 1088, SW 1788 and LN-18 were purchased from American Type Culture Collection (ATCC). These cells were grown in Dulbecco’s modified Eagle medium (DMEM) supplemented with 10% fetal bovine serum (FBS). U138MG was also purchased from ATCC and cultured in DMEM supplemented with 10% FBS. Fresh brain tumor tissues that were clinically histopathologically diagnosed at the Sun Yat-sen University-Affiliated First Hospital were used. Patient’s consent and approval from the Institutional Research Ethics Committee were acquired for use of data for the research. All the tissues were collected and processed within 30 min after resection. The primary cultured tumor cells were obtained after mechanical dissociation, as previously described [39].

A total of 221 paraffin-embedded glioma samples, including WHO grade II–IV tumors, were used in this study. All of them were both clinically and pathologically diagnosed at the Sun Yat-sen University-Affiliated First Hospital between 2000 and 2010. The clinicopathological characteristics of the specimens are shown in Additional file 1: Table S1. Normal brain tissues were obtained from individuals who had died from traffic accidents and confirmed to be free of any preexisting pathologically detectable conditions. Prior donor consent and the approval of the Institutional Research Ethics Committee were obtained for use of the data. Clinical and pathological classification was conducted according to the seventh edition of the classification system of the American Joint Committee on Cancer (AJCC).

Total RNA was extracted from cell lines and freshly frozen samples with TRIzol reagent (Invitrogen, Carlsbad, CA, USA) and was reverse-transcribed with the first-strand cDNA synthesis kit (Invitrogen). Real-time PCR reactions were conducted using Platinum SYBR Green qPCR SuperMix-UDG reagents (Invitrogen). Reverse transcriptase was used as the negative control, and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as the endogenous control. All experiments were repeated twice. The 2^-ΔΔCT equation was used to calculate the relative expression levels. The PCR primers used in this study were as follows: NUSAP1 (5′: GAAGCGCGGCATTCTTCATT, 3′: CGGCGATACTCGGAAGATGG), GAPDH (5′: CACCATCTTCCAGGAGCGAG, 3′: GACTCCACGACGTACTCAGC), GLI1 (5′: GCTCTGGACATACCCCACCT,3′: GCAGCTCCCCCAATTTTTCTG), PTCH1 (5′: TCGCTCTGGAGCAGATTTCC, 3′: TCTCGAGGTTCGCTGCTTTT), HIP1 (5′: GGCGACATGGATCGGATGG, 3′: ACAGCCACTTCCTGCGTATT), CCND1 (5′: AAAGAATTTGCACCCCGCTG, 3′: GACAGACAAAGCGTCCCTCA), CCNE1 (5′, GCAGGATCCAGATGAAGAAATG, 3′: TAATCCGAGGCTTGCACGTT), HDAC1 (5′: TGCAAAGAAGTCCGAGGCAT, 3′: ACCCTCTGGTGATACTTTAGCA).

PMSCV/NUSAP1 was generated by subcloning the PCR-amplified human NUSAP1 coding sequence into the pMSCV vector (Clontech). Human NUSAP1 targeting shRNA oligonucleotide sequences (RNA#1: 5-GCACCAAGAAGCTGAGAATGC-3, and RNA#2: 5′-GGAAATGGAGTCCATTGATCA-3) were cloned to generate pSuperretro-NUSAP1-shRNA(s). The Lipofectamine 3000 reagent (Invitrogen) was used for transfecting plasmids. Retroviral production and infection were performed as described previously [40]. Stable cell lines expressing NUSAP1 or NUSAP1 shRNA were selected for 10 days by treatment with 0.5 μg/ml puromycin for 48 h after infection. In the same way, stable cell lines expressing Nusap1 shRNA (RNAi#1, 5-GCATGTTAAGGAAACTCAGCC-3, and RNAi#2, 5-GCAGCGCCTCATCAAGAAAGT-3) were established and selected. Human GLI1 targeting shRNA oligonucleotides sequences were as follows: RNA#1: 5-GCCACCAAGCTAACCTCATGT-3, and RNA#2: 5-GCCTGAATCTGTGTATGAAAC-3.

Western blot analysis was conducted using anti-NUSAP1, anti-GAPDH, anti-α-tubulin, anti-MMP2, anti-MMP9, anti-Ki67, anti-β-actin, and anti-elongation factor 1 alpha (EF1α) antibodies (Abcam, Cambridge, MA, USA). Human GAPDH, α-tubulin, β-actin, or EF1-α were used as the endogenous reference.

Immunohistochemistry was performed in 221 clinical glioma tissue sections using a previously described method [41]. The degree of immunostaining was reviewed and scored separately by two independent pathologists blindly. The scores were determined by combining the proportion of positively-stained tumor or normal pancreatic epithelial cells and the intensity of staining. Cell proportions were scored as follows: 0, no positive cells; 1, < 10% positive cells; 2, 10%–35% positive cells; 3, 35%–75% positive cells; 4, > 75% positive cells. Staining intensity was graded according to the following standard: 1, no staining; 2, weak staining (light yellow); 3, moderate staining (yellow brown); 4, strong staining (brown). The staining index (SI) was calculated as the product of the staining intensity score and the proportion of positive cells. Using this method of assessment, we evaluated protein expression of NUSAP1 in glioma specimens by determining the SI, with possible scores of 0, 2, 3, 4, 6, 8, 9, 12, and 16. Sample with a score index ≥ 8 were determined as high expression and samples with a score index < 8 were determined as low expression.

Cells (5 × 10³ per well) were seeded in 96-well culture plates and stained with 100 μl of sterile MTT dye (0.5 mg/ml; Sigma, St. Louis, Missouri, USA) at 1, 2, 3, 4 and 5 days; this was followed by additional incubation for 4 h at 37 °C. After removal of the culture medium from each well, 150 μl of dimethyl sulfoxide (Sigma, St. Louis, MO, USA) was added and thoroughly mixed for 15 min. Following this, a microplate reader (Bio-Rad 3500; Hercules, California, USA) was used to determine the optical density, and absorbance was measured at a wavelength of 570 nm with a reference wavelength of 655 nm. All the experiments were repeated three times.

The indicated cells were plated in 6-well plates (1 × 10³ cells per well) and cultured for 2 weeks. The colonies were fixed with methanol for 10 min and stained with 1% crystal violet for 1 min. All the experiments were repeated three times.

For the Transwell assay or Transwell matrix penetration assay, the indicated cells (1 × 10⁴) were plated on the upper side of a polycarbonate Transwell filter with or without Matrigel in the upper chamber of the BioCoat™ invasion chambers (BD, Bedford, MA). After 22 h of incubation at 37 °C, the cells in the upper chamber were removed with cotton swabs, and the migrated and invaded cells on the lower membrane surface were fixed in 1% paraformaldehyde and stained with hematoxylin. The cells were counted (ten random 100× fields per well) and expressed as the mean number of cells per field of view. All the experiments were repeated three times, and the data were expressed as mean ± standard deviation (SD) values.

The indicated cells were cultured on 6-well plates with DMEM containing 10% FBS. Until the cells become confluence, we made a 500-μm wide cell-free gap by scratching the bottom of the plate with a pipette tip, and the cells were further incubated for 24 h. Phase-contrast images of the wound healing process were obtained digitally using an inverted Olympus IX50 microscope with a 10× objective lens at 0 and 24 h after the scratching. Then, the length of the healed wound was compared with the length of the initial wound.

The indicated cells were trypsinized and suspended in complete medium containing 0.3% agar. The cell-agar mixture was plated on the top of a bottom layer with 1% agar-containing medium. About 10 days later, viable colonies that were larger than 0.5 mm in diameter were counted. All the experiments were repeated three times.

The indicated cells were co-transfected with the indicated plasmids and luciferase reporter plasmids in 6-well plates and culture for 48 h, after which the cells were harvested and lysed for luminescence detection. The procedure and detection were performed with a luciferase assay kit according to the manufacturer’s protocol. Renilla luciferase was activated for emission of primary luminescence. All the experiments were repeated three times.

The animal studies were approved by the Ethics Committee of Sun Yat-Sen University, and all the experiments conform to the relevant regulatory standards. SW 1088 cells (5 × 10⁵), SW 1088/NUSAP1 cells (5 × 10⁵), U-87 MG/Scramble cells (5 × 10⁵), or U-87 MG/NUSAP1 shRNA#1 cells (5 × 10⁵) were stereotactically implanted into the brain of nude mice (five mice per group). The tumor-bearing mice were sacrificed 5 weeks after implantation, and the whole brain was resected. The brain specimens were cut to about 4-μm sections and embedded in paraffin for immunohistochemistry and hematoxylin-eosin (H&E) staining. After deparaffinization, immunohistochemistry was conducted using an anti-NUSAP1 antibody. Deparaffinized tumor sections were stained with Mayer’s hematoxylin solution for H&E staining. Images were captured using the AxioVision Rel.4.6 computerized image analysis system (Carl Zeiss).

We first analyzed the expression levels of NUSAP1 in astrocytoma, including anaplastic astrocytoma (n = 19) and glioblastoma (n = 81), using data deposited in the Oncomine database. NUSAP1 expression was found to be upregulated in astrocytoma compared to normal brain tissues (Fig. 1a). Further, analysis of NUSAP1 expression from data deposited in The Cancer Genome Atlas (TCGA) database revealed that the increase in the mRNA expression of NUSAP1 in astrocytoma tissues was in tandem with the increase in the WHO grade of the tumor (Fig. 1b). To further confirm these results obtained from public datasets, we examined the expression of NUSAP1 in two normal brain tissues, eight astrocytoma tissues, one primary normal human astrocyte (NHA) cell line and seven astrocytoma cell lines by real-time PCR and western blotting. Real-time PCR and western blotting analyses revealed that the mRNA and protein expression of NUSAP1 was higher in the eight astrocytoma tissues than in the two normal brain tissues (Fig. 1c and d). Compared with the NHA cell line, the expression of NUSAP1 was also upregulated in all seven astrocytoma cell lines, as confirmed by RT-PCR analyses and western blotting (Fig. 1e and f). Thus, our data indicated that the expression of NUSAP1 was upregulated in astrocytoma cell lines and tissues.

As the expression of NUSAP1 was upregulated in astrocytoma, we further investigated whether there was a correlation between NUSAP1 expression and survival. We analyzed the expression of NUSAP1 in astrocytoma samples which were selected from low grade glioma (LGG) data of the TCGA databases to perform the survival analysis and found that astrocytoma patients with high NUSAP1 expression had poorer overall survival than patients with low NUSAP1 expression (Fig. 2a). Moreover, upregulation of NUSAP1 was closely related to shorter relapse-free survival (Fig. 2b). To investigate the association between NUSAP1 and clinical characteristics of the patients, we examined NUSAP1 expression by immunohistochemical analysis of 221 paraffin-embedded astrocytoma tissues (Additional file 1: Table S1). Among the NUSAP1-positive cases, 69 (31.22%) had low NUSAP1 expression, while 152 (68.78%) had high NUSAP1 expression (Additional file 1: Table S1). Furthermore, NUSAP1 expression increased markedly with human glioma WHO grade and quantitative IHC analysis revealed that the mean optical density (MOD) of NUSAP1 staining in glioma cells increased significantly with the WHO grade, consistently (Fig. 2, C-E), suggesting that high NUSAP1 protein expression contributes to glioma progression.

Moreover, the statistical analysis indicated that upregulated NUSAP1 protein was associated with WHO tumor grade (P < 0.001) and vital status (P < 0.001) (Additional file 1: Table S2). Kaplan–Meier analysis and log-rank testing revealed that NUSAP1 expression levels were inversely correlated with survival time, whether at WHO grade I–II or at WHO grade III–IV (Fig. 2F; Additional file 2: Figure S1). Moreover, univariate and multivariate survival analyses revealed that NUSAP1 expression was an independent prognostic factor of glioma (P < 0.001) similar to the WHO grade (P < 0.001) (Additional file 1: Table S3). Taken together, NUSAP1 protein upregulation in glioma contributes to glioma progression and correlates with poor prognosis of the disease.

To investigate whether NUSAP1 has an effect on the pathogenesis of astrocytoma, we used gene set enrichment analysis (GSEA) to predict the possible biological functions of NUSAP1 in the cancer. The findings indicated that NUSAP1 might be involved in proliferation of the cancer cells (Fig. 3a). To verify this, several stable cell lines were established: the SW 1008 and A172 cells stably expressed ectopic NUSAP1, while NUSAP1 expression was silenced in the U-87 MG and A172 cells by small hairpin RNA (shRNA) (NUSAP1 expression was confirmed in these cell lines by western blotting) (Fig. 3b and Additional file 3: Figure S2A). The effect of NUSAP1 on proliferation of astrocytoma cells was evaluated by overexpression and silencing of NUSAP1 transcripts. The 3-(4, 5-dimethyl-2-thiazolyl)-2,5-diphenyl-2H-tetrazolium bromide (MTT) assay showed that the NUSAP1-transduced astrocytoma cells (SW 1088 and A172 cells) displayed a significant increase in viability, while the NUSAP1-silenced U-87 MG and A172 cells displayed contrasting results (Fig. 3c and Additional file 3: Figure S2B). These results were further confirmed by the colony formation assay (Fig. 3d and Additional file 3: Figure S2C). Furthermore, in the Transwell matrix penetration assay, overexpression of NUSAP1 was found to enhance invasion in the SW 1088 and A172 cells, while knockdown of NUSAP1 inhibited cell invasion in the U-87 MG and A172 cells (Fig. 3e and Additional file 3: Figure S2D). We also found that the NUSAP1-transduced astrocytoma cells displayed an increase in migration ability, while the NUSAP1-silenced cells displayed a decrease in their migration ability in the wound healing assay (Fig. 3f and Additional file 3: Figure S2E). These observations indicated that upregulation of NUSAP1 promoted the proliferation, invasion and migration of astrocytoma cells.

As indicated by the above results, NUSAP1 promoted the aggressiveness of astrocytoma. Therefore, we tried to confirm whether NUSAP1 had the same effect in vivo. Before conducting the in vivo experiment, we simulated the internal environment to evaluate the growth ability of astrocytoma cells via an anchorage-independent growth assay. As demonstrated by the anchorage-independent growth assay, overexpression of NUSAP1 markedly increased the anchorage-independent growth ability of SW 1088 and A172 cells, while suppression of NUSAP1 reduced that growth ability of U-87 MG and A172 cells (Fig. 4a and Additional file 3: Figure S2F). Subsequently, the SW 1088 and U-87 MG cell lines, which stably overexpressed and suppressed NUSAP1 expression, respectively, were stereotactically implanted into the brain of nude mice; there was also a control group in which a control astrocytoma cell line was implanted. Upregulation of NUSAP1 significant shortened the survival of mice, while silencing of NUSAP1 increased the survival of mice (Fig. 4b). Immunohistochemical staining with an anti-NUSAP1 antibody showed that there was a significant increase in the positive staining densities of NUSAP1-overexpressing tumor samples as compared to the control tumor samples, while the positive staining density was decreased significantly in NUSAP1-suppressed samples (Fig. 4c). The findings also consistently showed that the tumor margins of the NUSAP1-overexpressing tumor samples were more incomplete, while the tumor margins of NUSAP1-suppressed tumor specimens were more complete. Thus, the findings indicate that NUSAP1 promotes aggressiveness in astrocytoma in vivo. To understand the underlying mechanism, we examined whether NUSAP1 could alter the extracellular matrix around the tumor cells. As revealed by western blotting, MMP2, MMP9 and Ki67 showed strong expression in the NUSAP1-transduced cell line, while MMP2, MMP9 and Ki67 expression was weak in the NUSAP1-knockdown cell lines (Fig. 4d and Additional file 3: Figure S2G). Collectively, these data indicated that overexpression of NUSAP1 promoted aggressiveness in astrocytoma.

Previously, Peterson KA et al. defined a prioritized set of 841 enriched Gli1-binding regions (GBRs), including NUSAP1 [42] which prompted us guess to whether there was any association between HH pathway and NUSAP1. Interestingly, GSEA analysis revealed that the expression level of NUSAP1 in TCGA astrocytoma dataset was positively correlated with the HH pathway gene signature (Fig. 5a), further suggested that NUSAP1 may play a role in activation of HH pathway. Furthermore, we found that upregulation of NUSAP1 resulted in an increase in the protein level of GLI1 in the nucleus, while the protein level of GLI1 in cytoplasm was decreased (Fig. 5b and Additional file 4: Figure S3A). Further, downregulation of NUSAP1 could decrease the protein level of GLI1 in the nucleus, while slightly altering GLI1 expression in the cytoplasm. Consistent with the results of western blotting, the Luciferase reporter assays showed that the activity of GLI1 was upregulated in NUSAP1-transduced cells and downregulated in NUSAP1-knockdown cells (Fig. 5c and Additional file 4: Figure S3B). Subsequently, we detected several target genes downstream of the HH signaling pathway by real-time PCR, including PTCH1, HIP1, CCND1, CCNE1 and HDAC1 (Fig. 5d and Additional file 4: Figure S3C-D). Upregulation of NUSAP1 increased, while downregulation NUSAP1 decreased, the expression of PTCH1, HIP1, CCND1, CCNE1 and HDAC1. Our data indicated that NUSAP1 promoted GLI1 translocation to the nucleus from the cytoplasm and subsequently led to activation of the HH signaling pathway in astrocytoma cells.

To confirm whether NUSAP1 promotes tumor aggressiveness by activating the HH pathway, we induced knockdown of the expression of GLI1 by small interfering RNA in NUSAP1-transduced SW1008 cells. As confirmed by RT-PCR, the mRNA level of GLI1 was significantly decreased in NUSAP1-siGLI1 cells (Fig. 6a). Further, the colony formation assay showed that the colony-forming ability of NUSAP1-siGLI1 cells was decreased in comparison with NUSAP1-transduced SW 1088 cells. When the NUSAP1-tranduced SW 1088 cells were treated with GANT61—a small-molecule inhibitor of GLI1- and GLI2-mediated transcription at the nuclear level—the colony formation ability of the cells was decreased (Fig. 6b). Furthermore, NUSAP1-siGLI1 significantly inhibited the cell invasion ability of SW 1088 cells, while treatment of the NUSAP1-transduced cells with GANT61 caused a greater decrease in their cell invasion ability (Fig. 6c). Moreover, NUSAP1-siGLI1 cells showed a decrease in their anchorage-independent growth ability, and treatment of the NUSAP1-transduced cells with GANT61 also caused a significant decrease in their growth ability (Fig. 6d). Moreover, we found that, in response to SMO agonist purmorphamine treatment, the activity of GLI1 was only slightly increased in SW 1088 glioma cells, which displays low NUSAP1 expression, but significantly increased in U-87 MG glioma cells, which exhibits high NUSAP1 expression. Importantly, the simulative effect of SMO agonist purmorphamine on GLI1 activity in U-87 MG glioma cells was dramatically arrogated by NUSAP1 depletion (Fig. 6e). Therefore, these results further supported the notion that NUSAP1 plays critical role in activation of the Hedgehog signaling pathway in glioma cells.