Targeting NEK2 impairs oncogenesis and radioresistance via inhibiting the Wnt1/β-catenin signaling pathway in cervical cancer

Human cervical cancer cell lines HeLa and SiHa as well as HEK293T cells were purchased from the American Type Culture Collection and grown in DMEM medium supplemented with 10% fetal bovine serum and 100 μg/ml penicillin. All of the above cells were cultured at 37 °C in a humidified atmosphere containing 5% CO_2.

The targeting siRNA sequences in this study were as follows: NEK2 siRNA#1, 5′-GGATCTGGCTAGTGTAATT-3′ and NEK2 siRNA#2, 5′-GCTAGAATATTAAACCATG-3′, which have been described previously [29]. HeLa and SiHa cells were transfected with indicated siRNAs (50 nM) using Lipofectamine RNAiMAX reagent (Invitrogen) according to the manufacturer‘s instructions. Subsequent experiments were performed 48 h post transfection.

Stable NEK2-knockdown cell lines were established as described previously [30, 31]. Briefly, HEK293T cells were transiently transfected with NEK2 shRNAs and pSPAX2 and pMD2G plasmids. Forty-eight hours post transfection, the lentivirus-containing supernatants were filtered and used to infect SiHa cells after mixing with 8 μg/ml polybrene to increase the infection efficiency. Stable cell lines were selected with 2 μg/ml puromycin and confirmed by Western blotting. The shRNA sequences used in our study were as follows:

Control shRNA: 5′-TTCTCCGAACGTGTCACGTTT-3′.

NEK2 shRNA-1: 5′-GGGATCTGAAACCAGCCAATG-3′.

NEK2 shRNA-2: 5′-GCATTAATGCCTCCATTTACA-3′.

HeLa cells were transfected with control or NEK2-targeting siRNAs using Lipofectamine RNAiMAX for 48 h. Total RNA was isolated using TRIzol reagent (Invitrogen) according to the manufacturer’s instructions. Genes meeting the established threshold criteria of a false discovery rate (FDR) of < 5% and a fold change of > 2.0 were considered significantly differentially expressed.

Total cellular RNA was extracted using TRIzol reagent (Invitrogen). Reverse transcription was performed using a Prime RT reagent kit (Toyobo). Real-time PCR was performed using a SYBR® Premix Ex Taq™ Kit (Takara) according to the manufacturer’s instructions. The relative gene expression levels were calculated by the ΔCt method (the Ct of GAPDH minus the Ct of the target gene). Expression of GAPDH was used as the internal control. Primer sequences used for amplification were listed in Additional file 1: Table S1.

Whole cell lysates were prepared in NETN buffer containing 20 mM Tris HCl (pH 8.0), 100 mM NaCl, 1 mM EDTA and 0.5% Nonidet P-40, separated on SDS-PAGE gels, and transferred to PVDF membranes. Western blotting was performed using the appropriate primary antibodies against NEK2 (1:200, sc55601, Santa Cruz Biotechnology), Wnt1 (1:500, ab15251, Abcam), Wnt4 (1:500, sc376279, Santa Cruz Biotechnology), β-catenin (1:1000, #8480, Cell Signaling Technology), Flag (1:1000, F1804, Sigma) and GAPDH (1:1000, #5174, Cell Signaling Technology) overnight at 4 °C. The PVDF membranes were then incubated with secondary antibodies and detected via enhanced chemiluminescence.

Transfected cells were plated in 6-well plates at a density of 1.0 × 10⁴ cells/ml, and the cell numbers in each well were evaluated every other day. Alternatively, cells were plated in 6-well plates at a certain density gradient and grown for two weeks. The colonies comprising more than 50 cells were counted.

2 × 10⁴ cells in logarithmic growth phase were seeded in 96-well plates. The cells were harvested the next day and incubated with a 1/1000 dilution of EdU reagent for 0.5 h. The samples were washed with PBS and then incubated with 4% paraformaldehyde for 30 min. After being washed twice with PBS, the samples were permeabilized with 0.3% Triton X-100 in PBS and stained with reaction solution. Images were acquired via fluorescence microscopy.

After transfection with indicated siRNAs, the cells were collected and washed with cold PBS. Subsequently, the cells were analyzed using an Annexin V-PI Apoptosis Detection Kit I (BA1250, EnoGene, China) according to the manufacturer’s instructions. The apoptosis rates of HeLa and SiHa cells were analyzed by flow cytometry (Beckman, USA).

This assay was performed as described previously [32‐34]. Cells transfected with the indicated siRNAs were seeded in triplicate into six-well plates and irradiated with indicated doses. After two weeks, the colonies comprising more than 50 cells were counted.

The neutral comet assay was performed using Trevigen comet assay kit according to the manufacturer’s instructions. Briefly, SiHa and HeLa cells transfected with indicated siRNAs were immobilized on the comet slide using low melting agarose, lysed overnight before being subjected to electrophoresis at 21 V for 30 min in a neutral unwinding buffer. Gels were then neutralized and stained with SYBR Gold (Invitrogen). Cells were photographed using a fluorescence microscope and the olive tail moment was analyzed by comet score software.

Cells were grown on coverslips and irradiated with 2 Gy. The cells were collected at 4 h after irradiation, fixed with 4% paraformaldehyde and permeabilized with 0.2% Triton X-100 for 5 min. Following three 5-min rinses with PBS, the coverslips were blocked with 5% bovine serum albumin. Finally, the samples were incubated with anti-γ-H2AX (1:500, ab26350, Abcam) or Rad51 antibody (1:500, ab63801, Abcam) overnight at 4 °C. The samples were then washed and incubated with Dylight549-conjugated goat anti-mouse IgG secondary antibody (1:200, A23310, Abbkine) for 1 h at room temperature. After counterstaining with DAPI, immunostained cells were examined with a fluorescence microscope.

A cervical cancer tissue microarray, which contained 41 cervical carcinoma tissues and paired paracarcinoma tissues, was obtained from Shanghai Outdo Biotech (Shanghai, China). IHC analysis was performed as previously described [32‐34]. Briefly, the tissue sections were deparaffinized, rehydrated, and blocked with goat serum. After incubation with the anti-NEK2 antibody (1:100, ab227958, Abcam) overnight at 4 °C, the sections were washed three times with PBS and incubated with an HRP-conjugated secondary antibody (1:300, K8002, Dako) for 30 min at room temperature. Following three 5-min rinses in PBS, the stained sections were reacted with 3,3′-diaminobenzidine for 10 min and then counterstained with 0.1% hematoxylin. The staining in tumor and normal tissues was scored, and the staining percentage was determined. The score calculated by multiplying the values assigned to the staining intensity and percentage was used to evaluate the expression of NEK2.

Animal experiment was approved by the Medical Ethics Committee of Tongji Medical College, Huazhong University of Science and Technology. This assay was performed as previously described [31, 32]. Briefly, 5-week-old female Balb/c mice were randomly grouped and injected subcutaneously with 2 × 10⁷ shControl, shNEK2#1 or shNEK2#1 SiHa cells. Tumors were measured and weighed every three or four days using calipers. The formula used to calculate tumor volumes was as follows: V = (L × W²)/2, where V = volume (mm³), L = length (mm), and W = width (mm).

To investigate the clinical significance of NEK2 in cervical cancer, we first analyzed the transcriptomic profiles of cervical cancer and normal cervical tissues in The Cancer Genome Atlas (TCGA) database and found that NEK2 mRNA levels were significantly upregulated in cervical carcinoma tissue compared with normal tissue (Fig. 1a). In addition, our expression analyses using Gene Expression Omnibus (GEO) data set GSE9750 revealed that the mRNA levels of NEK2 were higher in human cervical cancer cell lines than that in normal cervix epithelial cells (Fig. 1b). To further validate whether NEK2 expression is indeed upregulated, we detected NEK2 protein level in 41 paraffin-embedded cervical cancer tissues and paired paracarcinoma tissues by immunohistochemistry. As shown in Fig. 1c and d, NEK2 was primarily localized in the nucleus and was overexpressed in cervical cancer tissues. NEK2 positivity was dramatically higher in cervical cancer tissues (70.7%) than in the adjacent paracarcinoma tissues (24.4%) (P < 0.001). Notably, a significant correlation was found between NEK2 expression and the tumor stage as well as lymph node metastasis in 123 paraffin-embedded cervical cancer tissues (P < 0.05) (Additional file 2: Table S2). Together, these results indicate that NEK2 upregulation may contribute to oncogenesis in cervical cancer.

Given that NEK2 is overexpressed in cervical cancer tissues, we speculated that it may act as an oncogenic protein in cervical cancer. To test this hypothesis, we first downregulated the expression level of NEK2 in two different cervical cancer cell lines (HeLa and SiHa) by siRNA transfection. As shown in Fig. 2a, NEK2 protein expression was successfully knocked down. As expected, NEK2 silencing suppressed the growth of cervical cancer cells compared with control cells (Fig. 2b and Additional file 3: Figure S1). The colony formation ability of cells was also impaired when NEK2 was knocked down (Fig. 2c). Consistent with this notion, the 5-ethynyl-2′-deoxyuridine (EdU) staining assay showed that the proliferation of cervical cancer cells was inhibited when NEK2 was depleted (Fig. 2d). We then determined the ratio of apoptotic cells, as measured by Annexin V/PI staining and flow cytometry. As shown in Fig. 2e, downregulation of NEK2 resulted in increased proportions of apoptotic cells. These findings demonstrate that NEK2 promotes the growth and proliferation and inhibits the apoptosis of cervical cancer cells.

To further confirm the role of NEK2 in promoting oncogenesis in cervical cancer, we constructed stable NEK2-knockdown SiHa cells using NEK2 shRNAs (Fig. 3a) and found that shRNAs-mediated knockdown of NEK2 significantly inhibits cervical cancer cell growth and proliferation (Fig. 3b and c), which is consistent with above findings. To investigate whether NEK2 drives oncogenesis in vivo, we subcutaneously implanted shcontrol or shNEK2 cells into T-cell-deficient athymic nude mice. As shown in Fig. 3d-f, the ShNEK2 groups displayed a reduction in the tumor size and weight compared to ShControl group. Taking into account the above results, we conclude that loss of NEK2 impairs oncogenesis in cervical cancer in vitro and in vivo.

After investigating the role of NEK2 under physiological conditions, we also sought to determine whether NEK2 participates in cancer radioresistance under radiation-induced DNA damage. To this end, we first performed γ-H2AX foci formation assay since γ-H2AX foci is considered a key marker of DNA damage response [35]. As shown in Fig. 4a, the percentage of γ-H2AX foci-positive cells was dramatically increased in NEK2-depleted cervical cancer cells after irradiation exposure. In addition, the comet tail moment, which also reflects DNA damage, were significantly increased in irradiated NEK2 deficiency cells (Fig. 4b). These results clearly indicate that NEK2 knockdown accelerates DNA damage. It has been well established that Rad51 is an essential modulator of homologous recombination (HR) related to DNA repair [36]. Our results demonstrated that DNA damage-induced Rad51foci formation was significantly decreased upon NEK2 silencing (Fig. 4c), indicating that loss of NEK2 impairs DNA repair. In line with these ideas, NEK2 knockdown enhanced cellular sensitivity to radiation compared with that of control cells (Fig. 4d). Collectively, our results support the notion that NEK2 promotes the radioresistance of cervical cancer cells.

To explore the underlying mechanisms that contribute to the effects of NEK2 in cervical cancer, we performed RNA sequencing (RNA-seq) to compare the genomic expression profiles of HeLa cells transfected with control or NEK2-targeting siRNAs. As shown in Fig. 5a and b, there are many differentially expressed genes with a fold change of greater than two in NEK2 depleted cells. Among these genes, we selected four candidate genes (WNT1, WNT4, MMP9 and DDIT3) that have been reported to be involved in oncogenesis to verify (Fig. 5c). Our real-time PCR results showed that WNT1,WNT4 and MMP9 were significantly downregulated while DDIT3 was upregulated after NEK2 depletion (Fig. 5d), which is consistent with the RNA-seq data. Notably, the protein level of Wnt1 but not Wnt4 was also remarkably decreased when NEK2 was knocked down (Fig. 5e), indicating that NEK2 controls the expression of Wnt1 at the mRNA and protein levels. It has been shown that Wnt1 is the activator of Wnt/β-catenin signaling pathway [37]. To elucidate whether NEK2 is related to this classical oncogenic pathway, we examined the expression of β-catenin and found that NEK2 silencing led to significant reduction in β-catenin at the protein level in two different cervical cancer cell lines (Fig. 5e). Interestingly, bioinformatics analysis with the TCGA database showed that there was a positive correlation between NEK2 and β-catenin mRNA expression in cervical cancer tissues (Additional file 4: Figure S2), further suggesting that NEK2 may be a critical regulator of Wnt/β-catenin signaling in cervical cancer. In addition, we also demonstrated that Cyclin D1, PPAR-δ and c-Myc, which represent downstream target genes of Wnt/β-catenin pathway, were downregulated in NEK2 deficiency cells (Fig. 5f). These results indicate that NEK2 modulates the expression of Wnt1 to activate the Wnt1/β-catenin signaling pathway.

To investigate whether Wnt1 is indeed required for the observed phenotypes caused by NEK2 silencing in cervical cancer cells, we transfected exogenously expressed Wnt1 into NEK2-depleted cells (Fig. 6a). As shown in Fig. 6b and c, the defects in cell growth and proliferation induced by NEK2 knockdown were partially rescued by overexpression of Wnt1. Similarly, restoration of Wnt1 expression partially reversed the increased olive tail moment induced by NEK2 silencing in response to DNA damage (Fig. 6d). Additionally, the enhanced radiosensitivity caused by NEK2 depletion was also partially restored by transfection with a plasmid encoding Wnt1 (Fig. 6e). Accordingly, these results reveal that NEK2 exerts its biological functions mainly via regulating Wnt1 in cervical cancer.