Nucleolar and spindle associated protein 1 promotes metastasis of cervical carcinoma cells by activating Wnt/β-catenin signaling

Human cervical cancer cells HeLa, CaSki, C33A, SiHa, MS751, and ME-180,and the human cervical immortalized cell line ECT1/E6E7, were purchase from the American Type Culture Collection (ATCC, Manassas, VA, USA). HCC94 and HeLa229 cell lines were obtained from Shanghai Chinese Academy of Sciences cell bank (China). HeLa, CaSki, SiHa and MS751 cells were grown in Eagle’s Minimum Essential Medium (ATCC-30-2003) supplemented with 10% fetal bovine serum (FBS, Life Technologies). HCC94 cells were grown in RPMI-1640 medium (Life Technologies, Carlsbad, CA, US) supplemented with penicillin G (100 U/ml), streptomycin (100 mg/ml) and 10% fetal bovine serum (FBS, Life Technologies). HeLa229 and ME-180 cells were grown in Dulbecco’s modified Eagle’s medium (Invitrogen) supplemented with 10% FBS.ECT1/E6E7 cells were cultured in Keratinocyte-Serum Free medium (GIBCO-BRL 17005–042) with 0.1 ng/ml human recombinant EGF, 0.05 mg/ml bovine pituitary extract, and additional calcium chloride 44.1 mg/L (final concentration 0.4 mM). All cell lines were grown under a humidified atmosphere of 5% CO2 at 37 °C.

Our study was performed using 233 paraffin-embedded cervical carcinoma samples, which were took from patients diagnosed at the Sun Yat-Sen University Cancer Center. These specimens were obtained from 140 patients with stage I, 90 patients with stage II, and two patients with stage III cervical cancer who received surgery and postoperative chemotherapy or radiation, according to pathological high-risk factors, from 2005 to 2010. None these patients received any anti-cancer therapy before surgery. Information on the clinical characteristics of the patients with cervical cancer is summarized in the Additional file 1: Table S1. For quantitative real-time reverse transcription polymerase chain reaction (qRT-PCR) and western blotting analysis, fresh cervical cancer and matched noncancerous cervical tissues samples were obtained from sevenpatients, and fresh cervical cancer samples were resected from six patients without metastasis and from eight patients with metastasis. Surgeries were performed at the Sun Yat-sen University Cancer Center between January 2017 and May 2017. Ethics approval and written informed consent for the use of the specimens were provided by the Institutional Research Ethics Committee.

Gene Set Enrichment Analysis (GSEA) is a computational method that determines whether an a prior defined set of genes shows statistically significant, concordant differences between two biological states (e.g. phenotypes). The results presented in the current manuscript use the two phenotypes divided by the median of NUSAP1 mRNA expression of Cervical squamous cell carcinoma and endocervical adenocarcinoma (CESC) TCGA data. GSEA was performed using GSEA 2.0.9 (http://​www.​broadinstitute.​org/​gsea/​) according to its guideline [28, 29]. The gene sets (the curated gene sets C2) were downloaded from GSEA 2.0.9 (http://​www.​broadinstitute.​org/​gsea/​).

RNA extraction, reverse transcription, and real-time PCR were performed as described previously [25]. The primers were designed using primer Express Software v.2.0 and comprised: NUSAP1 forward: 5′-CTGACCAAGACTCCAGCCAGAA-3′ and reverse: 5′-GAGTCTGCGTTGCCTCAGTTGT-3′; SRY-Box 2 (SOX2) forward: 5′-GCTACAGCATGATGCAGGACCA-3′ and reverse: 5′-TCTGCGAGCTGGTCATGGAGTT-3′; octamer-binding protein 4 (OCT4) forward: 5′-CCTGAAGCAGAAGAGGATCACC-3′ and reverse: 5′-AAAGCGGCAGATGGTCGTTTGG-3′; homeobox transcription factor Nanog (NANOG) forward: 5′-CTCCAACATCCTGAACCTCAGC-3′ and reverse: 5′-CGTCACACCATTGCTATTCTTCG-3′; B lymphoma Mo-MLV insertion region 1 homolog (BMI1) forward: 5′-GGTACTTCATTGATGCCACAACC-3′ and reverse: 5′-CTGGTCTTGTGAACTTGGACATC-3′; glyceraldehyde-3-phosphate dehydrogenase (GAPDH) forward: 5′-GTCTCCTCTGACTTCAACAGCG-3′ and reverse: 5′-ACCACCCTGTTGCTGTAGCCAA-3′; MYC forward: 5′-CGTCCTCGGATTCTCTGCTC-3′ and reverse: 5′-GCTGGTGCATTTTCGGTTGT-3′; matrix metalloproteinase 7 (MMP7) forward: 5′-TCGGAGGAGATGCTCACTTCGA-3′ and reverse: 5′-GGATCAGAGGAATGTCCCATACC-3′; matrix metalloproteinase 9 (MMP9) forward: 5′-GCCACTACTGTGCCTTTGAGTC-3′ and reverse: 5′-CCCTCAGAGAATCGCCAGTACT-3′; CD44 forward: 5′-CCAGAAGGAACAGTGGTTTGGC-3′ and reverse: 5′-ACTGTCCTCTGGGCTTGGTGTT-3′.All experiments were conducted three times. Additionally, GAPDH was chose as the internal control to normalize the expression levels of all the genes in the samples, and the fold changes were calculated using the relative quantification 2- [(cycle threshold (Ct) of gene)-(Ct of GAPDH)].

Western blotting assays [30] were conducted using anti-NUSAP1 rabbit polyclonal antibodies (SAB1407458, Sigma-Aldrich, St Louis, MO, USA), antibodies against p84(ab102684, Abcam), GAPDH (D16H11, Cell Signaling), α-tubulin (#3873, Cell Signaling), E-cadherin(#14472, Cell Signaling), vimentin(#5741, Cell Signaling), MYC (#2276, Cell Signaling), MMP7(#71031, Cell Signaling), MMP9(D603H, Cell Signaling), CD44(#37259, Cell Signaling), SUMO-1 (#4930, Cell Signaling),TCF4 (#2569, Cell Signaling), RanBP2 (ab2938, Abcam, Cambridge, MA) and β-catenin(#8480,1:1000, Cell Signaling). GAPDH, α-tubulin, or P84 were used as the endogenous controls.

Lysates were prepared from Siha cells using lysis buffer (150 mM NaCl, 10 mM HEPES, pH 7.4, 1% NP-40). Lysates were then incubated with HA affinity agarose (Sigma-Aldrich), or NUSAP1 or RanBP2 antibody with protein G agarose, overnight at 4 °C. Beads containing affinity-bound proteins were washed 6 times by immunoprecipitation wash buffer (150 mM NaCl, 10 mM HEPES, pH 7.4, 0.1% NP-40), followed by elutions 1 M glycine (pH 3.0). The eluates were then mixed with sample buffer and denatured, and used for the western blot analysis.

Immunohistochemical staining was carried on the 233 paraffin-embedded cervical carcinoma tissues using anti-NUSAP1 antibodies (HPA024904,1:600, Sigma) and antibodies against β-catenin(#8480,1:1000, Cell Signaling). The degree of immunostaining was reviewed and assessed in a blinded manner by two independent pathologists. We defined the score of cell proportions 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 using the following standard: 0, no staining; 1, weak staining (light yellow); 2, moderate staining (yellow brown); 3, strong staining (brown). The staining intensity (SI) was evaluated multiplying the staining intensity score by the proportion of positive tumor cells. During the assessment, we used the SI to evaluate the NUSAP1 and nuclear β-catenin protein expression levels. The possible scores were 0, 1, 2, 3, 4, 6, 8, 9, and 12. Specimens with a score ≥ 6 were defined as having high NUSAP1 or nuclear β-cateninexpression, and samples with a score < 6 were defined as having low NUSAP1 or nuclear β-catenin expression.

The NUSAP1 cDNA was amplified using PCR and subcloned into vector pMSCV-puro-NUSAP1. We cloned two short hairpin RNA (shRNA) oligonucleotides to produce pSuper-puro-NUSAP1-shRNA to knockdown endogenous NUSAP1(RNA#1: 5′-GCACCAAGAAGCTGAGAATGC-3′,and RNA#2:5′- GGAAATGGAGTCCATTGATCA-3′). The Lipofectamine 3000 reagent (Invitrogen) was used to transfect the plasmids into cells, following the manufacturer’s instructions. Stable cell lines expressing NUSAP1 or NUSAP1 shRNA were selected for 10 days by treatment with 0.5 μg/ml of puromycin for 48 h after infection. The sequence of RanBP2 siRNA was GAAUAACUAUCACAGAAUG .

Six-well plates were seeded with cells transfected with vector, NUSAP1, shRNA-vector, or NUSAP1 shRNA and incubated under suitable conditions until 90% confluence was reached. Wounds were induced by scratching the confluent cells using a pipette tip after 48 h of serum starvation. The cells were washed with phosphate-buffered saline (PBS) three times and then incubated in RPMI-1640 medium. At the indicated times (including time 0), the wounds were photographed under an inverted Olympus IX50 microscopeand measured. Each experiment was performed at least three times.

The invasion assay was conducted using aTranswell chamber with an 8-mm membrane filter insert (Corning) with Matrigel (BD,Biosciences). Briefly, the indicated cells were cultured in serum-free medium. The cells were placed into the upper chamber, and the lower chamber was supplied with 1 ml of medium containing 10% FBS. After 48 h of incubation at 37 °C, the cells in the upper chamber were gently removed using a cotton swab. The migratedcells on the lower membrane surface were fixed in 1% paraformaldehyde, stained with hematoxylin, and counted (ten random fields per well; 100× magnification). The count number was represented as the mean number of cells per field of view. All the experiments were conducted in triplicate andthe data are presented as the mean ± standard deviation (SD).

The indicated cells were implanted into six-well ultra-low attachment plates. Cells were incubated in the Dulbecco’s modified Eagle’s medium (DMEM)/F12 serum-free medium (Invitrogen) with 20 ng/ml epidermal growth factor (EGF), 2% B27 (Invitrogen), 5 μg/ml insulin (Sigma-Aldrich), 0.4% bovine serum albumin (Sigma-Aldrich), and 20 ng/ml basic fibroblast growth factor (bFGF; PeproTech). After 10 days of incubation, the number of spheres was calculated and their volume was assessed on a BX-X700 fluorescence microscope (Keyence, Osaka, Japan). The experiment was carried out three times.

To analyze the side population cells proportion, the cell suspensions were labeled with Hoechst 33,342 (Sigma-Aldrich) dye for side population analysis as per standard protocol [31, 32]. Briefly, cells were resuspended at EMEM medium (ATCC-30-2003) containing 2%FBS (Gibco, USA) at a density of 10⁶/mL. Hoechst 33,342 dye was added at a final concentration of 5 Ig/ml in the presence or absence of verapamil (Sigma-Aldrich) and the cells were incubated at 37 °C for 90 min with intermittent shaking. At the end of the incubation, the cells were washed with EMEM medium adding 2%FBS and centrifuged down at 4 °C, and resuspended in ice-cold EMEM medium. Propidium iodide (Sigma, USA) at a final concentration of 2 Ig/mL was added to cells to gate viable cells. The cells were filtered through a 40-lm cell strainer to obtain single cell suspension before sorting. Analysis and sorting was done on a FACS AriaI (Becton Dickinson). The Hoechst 33,342 dye was excited at 355 nm and its dual-wavelength emission at blue and red region was plotted to get the SP scatter.

The indicated cells were placed on 24-well plate and then incubated at 37 °C in 5% CO₂ overnight. The cells were fixed with 4% paraformaldehyde and permeabilized with 0.1% Triton X-100 before being blocked with 1% BSA in PBS buffer for 1 h the next day. Cells were incubated with primary antibodies against E-cadherin (1:400) (24E10, Cell Signaling) and vimentin (1:50) (D21H3,Cell Signaling),whichwere conjugated with Alexa Fluor 488 and Alexa Fluor 555, respectively, at room temperature for 1 h in the dark. Cell nuclei were stained with 4,6 diamidino-2-phenylindole (DAPI, Sigma) for 3 min. The cells were then visualized and photographed under a confocal laser-scanning microscope (Olympus FV1000, Japan).

Cells were implanted in 48-well plates and incubated for 24 h. One nanogram of the pRL-TK Renilla plasmid (Promega) containing the TOP-Flash or FOP-Flash luciferase reporter were transfected into the cells using the Lipofectamine 2000 Reagent (Invitrogen). A Dual-Luciferase Reporter Assay kit (Promega) was used to detect the luciferase activity of the indicated cells according to the manufacturer’s instructions.

BALB/c-nu mice (female, 3–5 weeks old; weighing 11–13 g) were purchased from Hunan SJA Laboratory Animal Co. Ltd. (Changsha, China). All experimental procedures were approved by the Institutional Animal Care and Use Committee of Sun Yat-sen University. As for thelung metastasis colonization mousemodels, the BALB/c-nu mice were randomly separated into four groups (n = 6/group) and injected with 5 × 10⁵cells (SiHa-vector/SiHa-NUSAP1;SiHa-shRNA-Vector/SiHa-NUSAP1-shRNA#1) through their tail veins. The Xenogen IVIS spectrum imagining system (Caliper) was used to assess the bioluminescence of tumor colonization and growth in the lung and popliteal lymph node tissues. At the end of the experiment (45 days), the mice were euthanized and their lungs were excised, fixed in formalin, and paraffin-embedded for the hematoxylin and eosin staining. The numbers of lung surface metastatic nodules were calculated under a dissecting microscope and expressed as the mean ± standard error of the mean (SEM). As for the lymph node metastatic mouse model, the BALB/c-nu mice were randomly divided into four groups (n = 6/group). The NUSAP1, NUSAP1-RNA#1 or vector-transduced SiHa cells (1.5 × 10⁵), which stably express firefly luciferase, were inoculated into the foot-pads of the mice. The mice were sacrificed on day 45, and the primary tumours and popliteal lymph nodes were excisedand paraffin embedded. Serial 4.0 mm sections were taken and analysed by IHC with anti-luciferase antibodies (Abcam). All animal studies were conducted withthe approval of the Medical Experimental Animal CareCommission of Sun Yat-sen University Cancer Center.

NUSAP1 mRNA expression in normal cervix and cervical cancer tissues were examined based on The Cancer Genome Atlas (TCGA; https://​gdc.​cancer.​gov/​) and the Gene Expression Omnibus (GEO) dataset (GSE7803 and GSE9750) (see URL https://​www.​ncbi.​nlm.​nih.​gov/​geo). Wedownloaded the SeriesMatrix File. Gene expression was presented as the mean value of multiple probes for eachgene after log2 transformation. Comparisons (normal cervix tissues versus cervical cancertissues) were analyzed by Mann-Whitney U test. P < 0.05 was considered statistically significant.

The statistical analyses were conducted using the SPSS 16.0 statistical software package (IBM, Armonk, NY, USA). Kaplan–Meier analysis was applied to establish survival curves and the log-rank test was used to evaluate the statistically significant differences. Student’s t-test (two-tailed) was performed to compare the continuous data. In addition, other statistical methods for data analysis included Fisher’s exact test, chi-squared (χ²) test, Mann-Whitney U test, Spearman correlation analysis, and multivariate cox regression analysis. P value of < 0.05 was regarded as statistically significant in all cases.

Microarray analysis (The Cancer Genome Atlas (TCGA) (Cervical squamous cell carcinoma and endocervical adenocarcinoma (CESC) TCGA data), GSE7803 and GSE9750) revealed that NUSAP1 was upregulated in cervical cancer samples compared with that in normal cervix tissues (Fig. 1a-c). To explore the role of NUSAP1 in cervical cancer, we selected seven paired cervical carcinoma tissues and eight cervical cancer cell lines to detect its expression. Western blotting assay and qRT-PCR assays showed that NUSAP1 was upregulated in the seven fresh primary cervical cancer specimens compared with that in the paired normal samples, and NUSAP1 was over expressed in the eight cervical cancer cell lines compared with that in the human cervical immortalized squamous cell line (Ect1/E6E7) (Fig. 1d,e,g,h). Furthermore, the expression of NUSAP1 was evaluatedin those samples that with 5-year metastasis compared with those samples that without metastasis within 5 years (Fig. 1f, i).

The clinical significance of NUSAP1 was further assessed by immunohistochemistry (IHC) staining in 233 archived cervical cancer samples (Additional file 1: Table S1 and Additional file 2: Table S2). The staining intensities of NUSAP1 expression divided into four degrees: negative (0 score, 7%), weak (1 score, 17%), moderate (2 score, 49%), and strong (3 score, 27%) (Fig. 2a). The number of cases with all staining intensities was indicated in Fig. 2b. Correlation analysis showed that high expression of NUSAP1 positively and significantly correlated with lymph node metastasis in patients (Fig. 2c). Importantly, Kaplan–Meier survival curves and log-rank tests revealed that patients with high NUSAP1 expression had shorter 5-year metastasis-free survival (P < 0.001, Fig. 2d). Additionally, multivariate Cox regression analysis found that the NUSAP1 protein expression level and N classification were independent prognostic indicators for cervical cancer (Fig. 2e).

We analyzed the biological role of NUSAP1 in metastasis of cervical carcinoma by gene set enrichment analysis (GSEA) on the basis of mRNA expression data from TCGA of CESC samples. The GSEA results showed that overexpression of NUSAP1 was positively associated with metastasis (Fig. 3a-c) (Additional file 3: Figure S4(A-G) and Additional file 4: Figure S5(A-E)). To further explore the function of NUSAP1 in cervical cancer metastasis, we created stable NUSAP1-overexpressing and -knockdown cell lines using HeLa and SiHa cervical cancer cells (Additional file 5: Figure S1A). As shown in Fig. 3d-e, wound healing assays and the Matrigel-coated transwell assays showed that NUSAP1 upregulation markedly increased the cell migration and invasion ability, while knockdown of NUSAP1caused an apparent decrease in cell migration and invasion. We then evaluated the effect of NUSAP1 on cervical carcinoma metastasis in vivo by lung colonization models and popliteal lymph node metastasis model. SiHa-NUSAP1 and SiHa-NUSAP1-shRNA#1cells were injected into the tail vein or footpads, while SiHa-Vector and SiHa-shRNA-Vector cells were used as controls (n = 6/group). Mice were sacrificed on day 45, and the lung and popliteal lymph nodes were enucleated and analyzed. Strikingly, our results revealed that NUSAP1 strongly enhanced the metastasis of SiHa cells, which was further confirmed by counting the visible metastatic lesions and H&E staining (Fig. 3 h-k). Meanwhile, we found that the lymph nodes in tumours formed from NUSAP1-transducted cells displayed higher numbers of luciferase-positive tumour cells than tumours formed vector-control cells. Conversely, the lymph node formed from NUSAP1-silenced cells had less luciferase-positive tumour cells than vector-control tumours (Fig. 3l,m). Strikingly, the ratio of metastatic to total dissected popliteal lymph nodes were obviously higher in the SiHa-NUSAP1 group (100% (6/6)) than in the vector groups (50% (3/6)). By contrast, there were no metastatic lymph nodes were found in the NUSAP1-silenced groups (Fig. 3n). To investigate whether NUSAP1 affect the proliferation rates of cervical cancer Hela and Siha cell lines. MTT assay shows that overexpression of NUSAP1 or knockdown of NUSAP1 did not significantly affect proliferation rates of cervical cancer Hela and Siha cell lines (Additional file 6: Figure S2A-D). Consistent with the cells proliferation results in vitro, the indicated cells (Siha/Vector, Siha/NUSAP1,Siha/Scramble and Siha/shRNA#1 cells) were also subcutaneously inoculated into the node mouse (n = 6/group). As shown in the Additional file 6: Figure S2E-G the subcutaneous tumors size formed from NUSAP1-ovexpression group and NUSAP1-silenced group has no significant difference. Collectively, these results reveal that NUSAP1 promotes cervical cancer cells metastasis both in vitro and in vivo.

Previous studies indicated that EMT is a critical step for tumor relapse and metastasis [16, 33]. The GSEA plot showed that NUSAP1 expression correlated positively with EMT (P < 0.001; Fig. 4a). Western blotting and immunofluorescent staining demonstrated that high expression of NUSAP1markedly reduced the expression of the epithelial marker E-cadherin and increased the expression level of the mesenchymal marker vimentin compared with that in the control group (Fig. 4b-d). Conversely, E-cadherin expression was upregulated and vimentin expression was downregulated in NUSAP1-silenced cells compared with the vector cells. Taken together, these results showed that NUSAP1 promotes EMT in cervical cancer cells.

Previous studies showed that CSCs have critical roles in the progression and metastasis of multiple types of cancer [22]. To assess whether NUSAP1 promotes the CSC properties of cervical cancer cells, we first analyzed the GSEA based on mRNA expression data from TCGA of CESC samples.. The results showed significant enrichment stem cell gene modules, in samples with high NUSAP1 (P = 0.042; P = 0.037; Fig. 5a, b). Furthermore, we explored the transcription levels of multiple pluripotency-associated markers, including SOX2, NANOG, OCT4, and BMI1 both in HeLa and SiHa cells. In both cell lines, the expression levels of these genes were increased in NUSAP1 overexpressing cells and downregulated in NUSAP1-silenced cells (Fig. 5c, d). Moreover, we also performed tumor sphere formation assays to detect the effect of NUSAP1 on the self-renewal of cervical cancer cells. As expected, we found that NUSAP1 upregulation promoted tumor sphere formation of SiHa and HeLa cells, generating nearly 3-fold more spheres compared with control cells. The tumor sphere formation obtained from NUSAP1-silenced cells generated approximately 3-fold fewer spheres compared with control cells (Fig. 5e). Moreover, flow cytometry assays showed that a smaller proportion of side population (SP) cells in the NUSAP1-silenced SiHa and HeLa cells compared to the control cells, while the proportion of SP cells was significantly increased in NUSAP1-upregulated cells compared to controls (Fig. 5f). These results suggest that NUSAP1 is essential for the maintenance of cervical cancer CSCs properties.

To further explore the mechanism by which NUSAP1 promotes metastasis of cervical cancer cells, we analyzed publicly available gene expression array data for cervical cancer using GSEA. We found that NUSAP1 overexpression was positively related to Wnt/β-catenin signaling (P = 0.002; Fig. 6a). To explore the effect of NUSAP1 on the Wnt/β-catenin pathway, we transfected TOPflash and FOPflash constructs and the RenillaPrl-TK plasmid as an internal controlinto Hela and Siha cells. We found that NUSAP1 elevation remarkably increased β-catenin/TCF transcriptional activity (Fig. 6b). By contrast, knockdown of NUSAP1 significantly repressed the β-catenin/TCF transcriptional activity. Next, we explored the effect of NUSAP1 on the subcellular localization of β-catenin. Consistent with our hypothesis, western blotting analysis of nuclear and cytoplasmic cellular fractions showed that overexpression of NUSAP1 induced, whereas knockdown of NUSAP1 impaired, the translocation of β-catenin into the nucleus (Fig. 6c). Meanwhile, we observed that upregulation of NUSAP1 in Hela and Siha cells markedly enhanced the transcription and translation of representative β-catenin target genes, including MYC, MMP7, MMP9, and CD44. In contrast, silencing of NUSAP1 reduced the expression levels of the representative β-catenin target genes (Fig. 6d, Additional file 7: Figure S3A-B). We next examined whether TCF4 protein is SUMOylated by RanBP2. Siha cells were transfected with TCF4 and SUMO1 cDNA and analyzed by immunoprecipitation and immunobloting. As shown in the Fig. 6e, NUSAP1 overexpression enhanced the SUMOylation of TCF4 compared to the vector group. Further work is needed to elucidate the details of mechanism whereby SUMOylation promotes the functional activity of Wnt/β-catenin signaling. We tested whether SUMOylation E3 ligase (RanBP2) co-IPed with endogenous NUSAP1 in cervical cancer cells. The identity of these proteins was detected by immunoblotting using available antibodies. RanBP2 and TCF4 were present in the immunoprecipitate with the anti-NUSAP1 antibody but not with control IgG (Fig. 6f). Moreover, NUSAP1 and TCF4 protein was detected in the immunoprecipitate with the anti- RanBP2 (Fig. 6g). Furthermore, knockdown of RanBP2 eliminated the SUMO conjugated form of TCF4 in NUSAP1-overexpressed Siha cells (Fig. 6h). Consistent with the previous results, We found that knockdown of RanBP2 remarkably repressed β-catenin/TCF transcriptional activity (Fig. 6i).

To further verify whether β-catenin activation was responsible for the NUSAP1-mediated effects, we detected the impact of blocking Wnt/β-catenin pathway on the metastasis and self-renewal capability of cervical cancer cells using the tankyrase inhibitor XAV-939(which was proved to stabilizes axin, a component of the β-catenin destruction complex, and widely used forblocking Wnt pathway) and a β-catenin-short interfering siRNA (Fig. 7a). As shown in Fig. 7b,both XAV939 and β-catenin siRNA treatments reduced TOP/FOP luciferase activities. Moreover, NUSAP1-induced invasion, EMT, and stemness in cervical cancer cells were repressed by the XAV939 or β-catenin siRNA treatments (Fig. 7c-f). To assess the clinical relevance of NUSAP1 and β-catenin, western blotting analysis was performed to evaluate the protein levels of NUSAP1 and β-catenin in ten fresh cervical carcinoma tissues. As shown in Fig. 7g and h, NUSAP1protein expression was positively associated with nuclear β-catenin protein expression (P = 0.032, r = 0.675). Moreover, immunohistochemical staining showed that NUSAP1 protein expression was positively and closely associated with nuclear β-catenin protein expression in the 233 paraffin-embedded cervical cancer tissues(P < 0.001; Fig. 7i, j). Collectively, these data suggested that the Wnt/β-catenin pathway is required for NUSAP1-mediated metastasis in cervical cancer cells and NUSAP1 enhanced nuclear β-catenin protein expression and then evoked Wnt/β-catenin signaling which resulted in promotion of metastasis in cervical cancer samples.