Involvement of the NF-κB signaling pathway in proliferation and invasion inhibited by Zwint-1 deficiency in Pancreatic Cancer Cells

Pancreatic cancer (PC) is an intractable cancer that is difficult to diagnose early and has a 5-year survival rate of less than 8%. ZW10-interacting kinetochore protein (ZWINT) is a crucial gene that contributes to chromosome instability and is essential for spindle assembly and kinetochore-microtubule attachment during meiosis and mitosis. However, the mechanism through which Zwint-1 promotes PC progression is yet to be elucidated. Here, we report that Zwint-1 is highly expressed in clinical PC specimens (based on analysis of the Gene Expression Profiling Interactive Analysis database) and various PC cell lines. Importantly, Zwint-1-deficient PC cells showed reduced nuclear factor-kappa B (NF-κB) (Ser536) phosphorylation along with inhibited proliferation and colony formation due to downregulation of NF-κB-regulated genes such as CCND1, cIAP1/2, and XIAP. In addition, Zwint-1-deficient PC cells showed reduced invasion and migration abilities, and decreased expression levels of the metalloproteinases MMP2 and MMP9. Furthermore, Zwint-1 deficiency arrested the PC cell cycle at the G2/M phase because the chromosomes failed to segregate properly, and the apoptosis rate in these cells gradually increased, accompanied by increased caspase-3 activation and anti-poly (ADP ribose) polymerase cleavage. Apoptosis caused by Zwint-1 deficiency was demonstrated to occur through caspase-dependent pathways based on experiments involving treatment with a pan-caspase inhibitor (Z-VAD-Fmk). Thus, Zwint-1 contributes to cell growth, invasion, and survival through NF-κB signaling pathways, suggesting that it could serve as a PC biomarker and new therapeutic target.


Introduction
All eukaryotic cells must precisely replicate their genomes during mitosis and then distribute them to newly forming daughter cells. For proper chromosome segregation, sister chromosomes must bind to the microtubules extending from opposing spindle poles through the kinetochore, a dynamic multi-protein assembly on the centromere [1][2][3][4]. Kinetochores consist of hundreds of conserved proteins, which are divided into major structural sub-complexes such as the constitutive centromereassociated network (CCAN), KNL1-Mis12-Ndc80 (KMN) network, and ROD-Zwilch-ZW10 (RZZ) complexes [1-3, 5, 6]. The CCAN, which has several kinetochore subunits, is further assembled into a CENP-A nucleosome [7] and is constitutively localized to the centromere throughout the cell cycle [6,8]. The CCAN acts as a foundation for kinetochore assembly and is a connector for centromeric chromatin, while the KMN network forms a complex with the microtubule-binding site of the kinetochore [8]. CENP-C binds to Ndc80 complexes through interactions with KNL1 and Mis12 complexes, and CENP-T is directly linked to Ndc80, contributing to KMN network localization [9]. Ndc80 complexes also Ivyspring International Publisher directly bind microtubules [2]. RZZ is required for dynein/dynactin recruitment [10][11][12] and spindle assembly checkpoint activation at the kinetochore [13].
ZW10-interacting kinetochore protein 1 (Zwint-1) was initially identified as a protein that interacts with Zeste White 10 (ZW10) by a yeast two-hybrid screen, and was subsequently demonstrated to be a kinetochore component that plays an important role in spindle assembly and kinetochore-microtubule attachment during meiosis and mitosis [14][15][16]. Zwint-1 can directly interact with components of the KMN complex, specifically Ndc80 and Mis12, and acts as a bridge between the RZZ and KMN complexes required for kinetochore formation and spindle checkpoint activity [17]. Zwint-1 is also a mitotic checkpoint component required for the stable association of CENP-F and dynamitin with the kinetochore to ensure accurate chromosome segregation [15]. Recent studies have suggested that Zwint-1 could be a potential cancer biomarker [18,19] given its high expression levels in several human malignancies such as prostate cancer, ovarian cancer, bladder cancer, lung cancer, hepatocellular carcinoma, and pulmonary adenocarcinoma [18,[20][21][22][23][24][25]. Moreover, knockdown of Zwint-1 has been shown to inhibit the proliferation, migration, invasion, and colony formation of lung cancer cell, and to enhance cell apoptosis [23].
However, little is known about the role of Zwint-1 in pancreatic cancer (PC). In this study, we investigated the expression of Zwint-1 in tissues from patients with PC in The Cancer Genome Atlas (TCGA) database and in PC cell lines, and evaluated the effects of Zwint-1 on PC cell tumorigenesis and progression using a Zwint-1 knockdown system in vitro.
The cutoff values were |log 2 fold change (FC)| of 1 and P value < 0.01. The overall survival (OS) of patients with PC was extracted from TCGA data and compared with the expression levels of the Zwint-1related genes.

Cell proliferation and viability assay
After adhesion for 24 h, the cells were transfected with Zwint-1 siRNA or control siRNA and incubated for 3 or 7 d. For cell proliferation and viability assays, MIA PaCa-2 and PANC-1 cells were seeded into 12well plates at a density of 1 × 104 cells/well or 8 × 104 cells/well and tested using 3-(4,5-dimethylthiazol-2yl)-2,5-diphenyltetrazolium bromide (MTT) solution. For colony formation assays, the cells were seeded into 6-well plates at a density of 500-1000 cells/well and incubated for 14 d. The colonies were then fixed in 100% methanol, stained with 10% crystal violet, and counted. Each assay was performed in triplicate.

Wound-healing assay
The motility of Zwint-1 siRNA and control siRNA-transfected cells was examined by a wound-healing assay. The cells were seeded in 6-well plates at 5 × 10 5 cells per well. After 24 h, wounds were created in cell monolayers in each well using a P1000 pipette tip. The cells were then rinsed once with phosphate-buffered saline (PBS) and cultured for 24 h or 72 h. Cell motility was expressed as the cell migration rate.

Cell migration and invasion assays
Migration and invasion abilities under Zwint-1 deficiency were assessed using Transwell plates (Corning, Corning, NY, USA). For the cell migration assay, 5 × 10 4 cells in 200 μL of 1% serum DMEM were seeded directly into each well of the Transwell chambers with 8-μm pore membranes. For the cell invasion assay, the cells were seeded in the Transwell chambers coated with Matrigel (Corning), and medium containing 10% FBS was added to the lower chamber. After incubation for 24 h, the medium in the upper chamber was removed, and the cells were fixed and stained using the Differential Quik Stain Kit. The cells adhering to the upper surface of the membrane were removed using a cotton applicator. The number of cells on the lower side of the membrane was counted. Each experiment was performed in triplicate.

Fluorescence-activated cell sorting (FACS) analysis
For cell cycle and DNA content analysis, cultured cells were incubated in trypsin-ethylenediaminetetraacetic acid at 37°C in an atmosphere containing 5% CO 2 , collected by centrifugation, and washed once with 1× PBS. The cells were centrifuged, supernatants were removed, and then the cells were stained with 50 μg/mL propidium iodide (PI; Sigma-Aldrich, St. Louis, MO, USA), along with 100 U ribonuclease A from the bovine pancreas (Sigma-Aldrich). FACs analysis was performed with a FACSCalibur instrument (BD Biosciences, San Diego, CA, USA) according to standard protocols.
Apoptosis was also analyzed by FACS using fluorescein isothiocyanate (FITC)-conjugated annexin V (BD Biosciences, San Diego, CA, USA) and PI (Sigma-Aldrich) staining. The analysis was performed with a FACSCalibur (BD Biosciences) instrument according to the standard protocol. Apoptosis was blocked with the pan-caspase inhibitor Z-VAD-FMK (20 µM; R&D Systems, Minneapolis, MN, USA).

Immunofluorescence
Cells were grown on Thermo Scientific Nunc Lab-Tek II Chamber Slides, permeabilized with 0.5% Triton X-100 for 1 min, and fixed with 4% paraformaldehyde for 10 min. The fixed cells were incubated for 1 h at room temperature with blocking solution (1% bovine serum albumin) and then incubated overnight at 4°C with anti-CREST (Immuno Vision Technologies, Brisbane, CA) and anti-α-tubulin (Abcam, Cambridge, MA, USA) primary antibodies. The cells were then incubated with secondary antibodies and 100 ng/mL DAPI for 3 h. Samples were mounted in Prolong Gold Antifade reagent (Invitrogen) and viewed under a confocal microscope (Zeiss LSM710 with ZEN software).

Statistical analysis
All data are presented as the mean ± standard error of the mean (SEM) values from three independent experiments. Results were statistically analyzed using GraphPad Prism (version 5.0, GraphPad Software Inc., San Diego, CA, USA) using one-way or two-way analysis of variance followed by Bonferroni's multiple comparison tests.

Zwint-1 expression is upregulated in PC patient tissues and PC cell lines
We first confirmed the mRNA expression of the CCAN (CENP-A, CENP-C, CENP-T), KMN (KNL1, Mis12, Ndc80), and RZZ (ROD, Zwilch, ZW10) complexes, as well as Zwint-1, in clinical PC specimens from TCGA (n = 179) and normal tissues from GTEx data (n = 171) using the publicly available GEPIA database. Zwint-1 expression was elevated in tumors, whereas expression of the genes involved in the CCAN, KMN, and RZZ complexes was not ( Fig.  1A and Supplementary Fig. 1A). In addition, the OS of PC patients was inversely proportion to the Zwint-1 expression level (Fig. 1A and B). Other clinical and demographic information about the patients is shown in Supplementary Fig. 2A-I, which was obtained from the cBioportal database. We also measured the level of Zwint-1 protein in various PC lines, including AsPC-1, PANC-1, MIA PaCa-2, and Capan-1 cells, by western blotting, which demonstrated higher levels in AsPC-1, PANC-1, MIA-PaCa-2, and Capan-1 cells compared to those in normal HPDE cells (Fig. 1C and  1D).

Zwint-1 deficiency inhibits the colony formation and proliferation of PC cells via the nuclear factor kappa B (NF-κB) signaling pathway
Between the two siRNAs tested, Zwint-1_2 had a greater effect on inhibiting Zwint-1 protein expression better ( Fig. 2A), resulting in reduced colony formation in MIA PaCa-2 (Fig. 2B), PANC-1, and Capan-1 cells (Fig. 2C). Therefore, Zwint-1_2 siRNA was selected for use in subsequent experiments. The MTT assay showed that Zwint-1 deficiency significantly inhibited the proliferation of MIA PaCa-2 and PANC-1 cells for 3-7 d after siRNA transfection (Fig. 2D). Western blotting showed that the expression level of p-NF-κB (p65) (Ser536) was reduced due to Zwint-1 deficiency, whereas there was no difference in the expression level of NF-κB (p65) between Zwint-1-expressing and Zwint-1-deficient cells. In addition, the expression level of cyclin D1 (a target of NF-κB) was significantly reduced due to Zwint-1 deficiency in PC cells (Fig.  2E). Finally, the expression levels of cIAP1/2 and XIAP, targets of NF-κB involved in cell survival, were also markedly reduced in Zwint-1-deficient PC cells (Fig. 2F). These results indicated that Zwint-1 is involved in the proliferation and survival of PC cells at least partially through NF-κB signaling.

Zwint-1 deficiency reduces the migration and invasion capacity of PC cells
After inhibiting PC cell growth by culturing under low FBS conditions, the number of migrating Zwint-1-deficient MIA-PaCa-2 cells was markedly reduced compared to that of control cells (Fig. 3A). Transwell assays further showed that the number of invading Zwint-1-deficient MIA-PaCa-2 cells was reduced compared to that of the control cells (Fig. 3B), and the wound-healing ability was also decreased in Zwint-1-deficient PC cells (Fig. 3C) further confirming reduced migration potential, particularly after 48 h. These results were supported by the reduced expression levels of MMP-2 and MMP-9 under Zwint-1 deficiency (Fig. 3D), indicating that Zwint-1 plays essential roles in migration and invasion by activating MMP2 and MMP9 expression in PC cells.

Zwint-1-deficient PC cells are arrested in the G 2 /M phase leading to high rates of apoptosis
Flow cytometry to determine the proportion of cells in the subG 1 , G 1 , S, and G 2 /M phases of the cycle demonstrated an increased proportion of Zwint-1deficient PC cells in G 2 /M compared to that in controls. In addition, subG 1 populations (apoptotic cells) increased compared to those in the controls at 24 to 72 h, whereas the G 1 cell population was decreased in Zwint-1-deficient PC cells in a concentrationdependent manner (Fig. 4A, 4B). Using western blotting, we also measured the expression of cyclin A2 (a regulator of the G 1 phase) and p-histone H3 (Ser10) (a regulator of the G 2 /M phase transition) to determine the molecular mechanisms altered by Zwint-1 deficiency. After 24 h of Zwint-1 siRNA transfection, cyclin A2 and p-histone H3 (Ser10) levels were significantly increased in the Zwint-1-deficient PC cells (Fig. 4C). Immunofluorescence observations of the centromere markers CREST and alpha-tubulin demonstrated well-aligned chromatids separated into two poles with a centromere in control cells. By contrast, although the tubulin of Zwint-1-deficient PC cells was attached to the centromere, no normally dividing cells were observed (Fig. 4D-4F). These results indicated that Zwint-1-deficient PC cells were arrested in the G2/M phase because they could not divide.

Apoptosis is increased in Zwint-1-deficient PC cells through caspase-dependent signaling
To determine whether Zwint-1-deficient PC cells were lost through apoptosis or necrosis, we double-stained the cells with annexin V-FITC and PI and quantified phenotypic changes in apoptotic cells. FACS analysis showed that the proportion of the late-apoptosis (annexin V-FITC-positive and Annexin V-FITC/PI-positive) population was significantly higher in Zwint-1-deficient cells compared to that in control cells at 72 h (Fig. 5A, upper right quadrant). The levels of caspase-3 and cleaved PARP proteins increased in Zwint-1-deficient PC cells compared to those of controls (Fig. 5B), further confirming a role of apoptosis in the observed cell death. These results are consistent with the increase in the subG1 population in Zwint-1-deficient PC cells. Treatment of Zwint-1-deficient PC cells with 20 μM of the pan-caspase inhibitor Z-VAD-Fmk reduced the rate of apoptosis (Fig. 5C), and also reduced the levels of caspase-3 activation and PARP cleavage increased underZwint-1 deficiency (Fig. 5D). These results indicated that increased apoptosis due to Zwint-1 deficiency occurs through caspase-dependent signaling pathways.

Discussion
Zwint-1 is not only necessary for normal cell division during mitotic metaphase but is also highly expressed in various carcinomas. In this study, we showed that Zwint-1 is highly expressed in PC cells and tissues, and promotes the proliferation and invasion of PC cells through NF-κB signaling. In line with previous studies in patients with other cancer types [18,[20][21][22][23][24][25], Zwint-1 was found to be upregulated in PC tissues and cell lines, and patients with high Zwint-1 expression had lower OS than those with low Zwint-1 expression. Zwint-1-deficient PC cells exhibited suppressed colony formation, proliferation, migration, and invasion due to a reduction in NF-κB phosphorylation and the expression of related genes. These data support evidence that Zwint-1 knockdown inhibits lung cancer cell proliferation, migration, and invasion, as well as colony formation [23]. Furthermore, Zwint-1deficient PC cells showed induced G2/M arrest due to abnormal cell division, which promoted apoptosis through caspase-dependent pathways. These findings suggest that Zwint-1 might play an important role in the promotion and progression of PC. To the best of our knowledge, this is the first report showing that Zwint-1 contributes to PC cell progression through NF-κB signaling.    The NF-κB family of transcription factors includes RelA (p65), RelB, c-Rel, NF-κB1 (p50/p105), and NF-κB2 (p52/p100), which play important roles in inflammation, cell proliferation and differentiation, immune responses, and cancer [31][32][33]. They share a Rel homology domain that regulates the development and progression of cancer by allowing the binding of NF-κB-specific DNA motifs [34]. Masking nuclear position signals prevents NF-κB transcription factors from being translocated into the cell nuclei by NF-κB inhibitor (IκB); thus, NF-κB remains dormant and inactive in the cytoplasm [35]. However, in most cancers, NF-κB is activated and is considered to be a major signal mediator that contributes to cancer development and progression by promoting cell proliferation, regulating apoptosis, stimulating angiogenesis, and increasing invasion and metastasis [36][37][38]. In particular, NF-κB signaling pathways play an important role in the development and progression of PC and drug resistance [39]. Indeed, we observed that Zwint-1-deficient PC cells inhibited temozolomide resistance ( Supplementary Fig. 3A).
NF-κB also plays an essential role in controlling the transcription of genes such as cyclooxygenase-2 (COX2) and cyclin D1, which are important in the early and late stages of aggressive cancers; cIAP-1/2, XIAP, and cellular FLICE inhibitory protein (FLIP), which are important genes that encode apoptosis suppressor proteins; and MMP2, MMP9, and vascular endothelial growth factor (VEGF), which are important genes in invasion and angiogenesis [40][41][42][43][44]. In our study, various changes in proliferation, invasion, migration, and apoptosis were observed in Zwint-1-deficient PC cells, suggesting a role for NF-κB signaling. Zwint-1-deficient PC cells showed reduced phosphorylation of NF-κB and significantly reduced the protein expression levels of NF-κB target genes, including cyclin D1, cIAP1/2, XIAP, MMP2, and MMP9. Consistently, Zwint-1-deficient PC cells also showed reduced proliferation ability, invasion, and migration, as well as increased apoptosis.

Conclusion
Zwint-1 deficiency inhibits PC cell proliferation, invasion, and migration through the NF-κB signaling pathway, in addition to promoting cellular apoptosis through a caspase-dependent pathway.