ELK1-induced up-regulation of KIF26B promotes cell cycle progression in breast cancer

KIF26B is a member of the kinesin superfamily that is up-regulated in various tumors, including breast cancer (BC), which can promote tumor progression. This study aimed to investigate the potential function of KIF26B in BC, and the underlying mechanisms, focusing mainly on cell proliferation. KIF26B expression was examined in BC tissue samples obtained from 99 patients. Then, we performed MTS, EdU and flow cytometry assays to detect cell proliferation, and western blotting to measure the expression of cell cycle-related proteins in MDA-MB-231 and MDA-MB-468 cells following KIF26B knockdown. Promoter analysis was used to study the upstream regulatory mechanism of KIF26B. KIF26B was upregulated in BC tissues. High expression of KIF26B was associated with clinicopathological parameters, such as positive lymph node metastasis, higher tumor grade, and higher proliferative index in BC. Furthermore, knockdown of KIF26B expression inhibited MDA-MB-231 and MDA-MB-468 cell proliferation, arresting cells in the G1 phase of the cell cycle in vitro. Similarly, KIF26B silencing decreased the expression levels of Wnt, β-catenin, and cell cycle-related proteins such as c-Myc, cyclin D1, and cyclin-dependent kinase 4, while increasing the expression of p27. Moreover, ELK1 could bind to the core promoter region of KIF26B and activate its transcription. KIF26B acts as an oncogene in BC by regulating multiple proteins involved in the cell cycle. ELK1 activates KIF26B transcription.


Introduction
Breast cancer (BC) is one of the most common malignant tumors and has become the second risk factor for cancerrelated mortality in women worldwide. Among multiple causes, metastasis and uncontrolled proliferation of tumor cells are two of the most important factors [1,2]. Thus, in this study, we examined the mechanisms associated with the proliferation of BC cells, with the aim of identifying a promising target for the treatment of patients with BC.
KIF26B is a member of the kinesin family (KIF) composed of 2108 amino acids. KIF26B cannot hydrolyze ATP [3]. Accumulating evidence demonstrates that KIF26B can regulate cytoskeleton-driven processes, such as cell migration, polarization, mitosis adhesion, and plays an important role in the development of the kidney and the nervous system [4][5][6]. More recent studies have suggested that KIF26B plays an important role in the oncogenesis or progression of many human cancer types [7,8]. High KIF26B expression is strongly linked to poor prognosis in patients with BC or colorectal cancer [9][10][11]. However, the role of KIF26B in BC, especially the relationship between high KIF26B expression and cell proliferation, has rarely been reported.
The present study aims to clarify the biological function of KIF26B in BC. In this study, we demonstrate the effect KIF26B on BC cell proliferation and cell cycle progression and explore the upstream transcriptional mechanism of KIF26B. 15 Page 2 of 12

Human BC samples
A total of 99 paraffin-embedded BC tissue samples were collected from patients undergoing mammectomy and needle biopsy at Qilu Hospital of Shandong University during 2005-2010. Pathomorphological observation was performed according to the criteria established by the World Health Organization. This study was approved by the Ethical Committee of Shandong University, China (code 2012028).

SiRNA and cell transfection
KIF26B small interfering RNA (si-KIF26B) and negative control (si-NC) were synthesized by RiboBio (Guangzhou, China). The sequence of si-KIF26B was 5′-CGG ACA GCC TCT CCT ATT A-3′ and the si-NC sequence was 5′-TTC TCC GAA CGT GTC ACG T-3′ [10]. After reaching 90% confluence, MDA-MB-231 and MDA-MB-468 cells were seeded in 6-or 12-well plates and transiently transfected with siRNA using the X-treme GENE transfection reagent (Roche Applied Science, Indianapolis, IN, USA) according to the manufacturer's instructions. The examination of KIF26B expression level and subsequent experiments were performed 48 h later.

MTS assay
MDA-MB-231 and MDA-MB-468 cells transfected with si-KIF26B or si-NC for 24 h were seeded in 96-well plates. MTS (5 mg/ml, Promega, Madison, WI, USA) was added into the culture medium at 24, 48, 72, and 96 h, respectively, followed by an incubation of 2 h. The absorbance at 490 nm was then read. Three independent experiments were performed.

EdU assay
MDA-MB-231 andMDA-MB-468 cells transfected with si-KIF26B or si-NC were cultured in 6-well plates for 24 h. The cells were then collected and cultured in 12-well plates for 24 h. The EdU assay was performed in according to the manufacturer's instructions (RiboBio, Guangzhou, China). Cells were visualized under a fluorescent microscope (Olympus, Tokyo, Japan). The experiment was repeated three times.

Flow cytometry
MDA-MB-231 and MDA-MB-468 cells transfected with si-KIF26B or si-NC were cultured in 6-well plates for 48 h. For the cell cycle assay, the harvested cells were stained with propidium iodide (PI, Beyotime). For apoptosis detection, cells were double-stained with annexin V-fluorescein isothiocyanate (FITC) and PI using an FITC annexin V Apoptosis Detection kit (401003, BestBio, China). The cells were immediately examined by flow cytometry (FACScan; BD Biosciences), according to the manufacturer's protocol. Each experiment was performed in triplicate.

Dual-luciferase reporter assays
293 T cells were seeded in 24-well plates overnight, then co-transfected using 0.5 μg/well of the KIF26B promoter regions (− 2000/0, − 1043/0, − 530/0, and − 110/0) overexpression plasmids and pGL3-basic vector and 0.01 μg of pRL-TK plasmid (as the internal control) using Lipofectamine 2000 (11668019, Invitrogen). About 48 h after transfection, the cells were subjected to luciferase activity analysis using a Dual-Luciferase Reporter Assay Systems (Promega) to decide the core promoter of KFI26B gene according to the manufacturer's instructions. Similarly, the plasmids with decided core promoter of KFI26B gene and alternative transcription factor overexpression plasmids (ELK1, STAT4, JUN, pCDNA3.1 basic vector) and 0.01 μg of pRL-TK plasmid were co-transfected to 293 T cells to establish the relevant transcriptional factor of KIF26B. Each experiment was performed in triplicate.

Chromatin immunoprecipitation assay
Chromatin immunoprecipitation (ChIP) assays were carried out using the EZ-Magna ChIP Chromatin Immunoprecipitation Kit (17-10086, Merck). Briefly, the MDA-MB-231 and MDA-MB-468 cells were crosslinked with 1% formaldehyde for 10 min at room temperature before immunoprecipitation reaction. After incubation with 5 μl anti-ELK1 antibody (ab32106, Abcam) or IgG antibody (rabbit anti-IgG, as the negative control) and protein A/G magnetic beads overnight, immunoprecipitation was performed in accordance with the manufacturer's instructions. Co-precipitated DNA was quantified using qPCR. A volume of 10 μl chromatin was used as input control prior to immunoprecipitation. This experiment was repeated three times.

Statistical analysis
Statistical analysis was performed using GraphPad Prism 7 (GraphPad Software, San Diego, CA, USA). Differences between the two groups were analyzed using Student's t test. The χ 2 test and Spearman's correlation were used to evaluate the associations between KIF26B expression levels and clinicopathological parameters in BC. Survival analysis was performed using the Kaplan-Meier method and logrank test. Data are presented as the mean ± SD. p < 0.05 was considered statistically significant.

KIF26B is up-regulated in BC tissues and cell lines
To study the role of KIF26B in BC, IHC was performed to detect KIF26B expression in 99 samples from patients with follow-up data. We found that KIF26B was significantly upregulated in ductal carcinoma in situ (DCIS) and invasive ductal carcinoma (IDC) compared with adjacent normal tissues (p < 0.0001). However, no significant difference was observed between IDC and DCIS (p = 0.3524). In addition, the staining intensity of KIF26B increased with the grade of DCIS and IDC ( Fig. 1A-H). All samples were then assigned to a high or low KIF26B expression group, using median KIF26B expression as cut-off point. High expression of KIF26B was markedly associated with clinicopathological parameters such as positive lymph node metastasis, higher tumor grade, and high proliferative index (Table 1). However, survival analysis showed that high KIF26B expression was not associated with overall survival and disease-free survival (Fig. 1I-J).
We also examined the expression of KIF26B in cell lines using RT-qPCR. KIF26B was expressed at lower levels in an immortalized breast cell line MCF-10A compared with BC cell lines, especially in cells with a higher degree of malignancy, such as MDA-MB-231 and MDA-MB-468 (Fig. 1K). Additionally, Gene Expression Profiling Interactive Analysis (GEPIA) (http:// gepia. cancer-pku. cn/) suggested that KIF26B expression was significantly increased in BC tumor tissue, compared with normal tissue (Fig. 2A). Moreover, the expression level of KIF26B was not associated with overall survival or disease-free survival (Fig. 2B, C). However, high expression of KIF26B was correlated with poorer prognosis in TNBC (Fig. 2D-F). Thus, KIF26B expression was significantly up-regulated in BC tissue samples and cell lines, and KIF26B may be associated with cancer progression.

ELK1 activates KIF26B transcription
To examine the upstream regulatory factor of KIF26B, we performed a promoter analysis. The sequence of the human KIF26B promoter from − 2000 bp to 0 bp (the first base of KIF26B cDNA, i.e., the transcription start site, is designated + 1) was acquired from the UCSC Genome Browser (http:// genome. ucsc. edu/). We constructed a series of truncated fragments from − 2000/− 1043/− 530/− 110 to 0, which were then cloned into the pGL3-basic vector (Fig. 3A). After transient transfection in 293 T cell for 48 h, luciferase activity was decreased distinctly in the − 530 to − 110 bp region (p < 0.001), suggesting that this region may be a core promoter of the human KFI26B gene (Fig. 3B).
Using the online JASPAR program, we identified potential transcription factors that might bind to the -530 to -110 bp region from the ( Figure S1), then validated them using dual-luciferase report assays. ELK1 markedly activated the KIF26B -530 to 0 promoter region, compared with the negative control (p < 0.01, Fig. 3C). RT-qPCR and western blot analysis further indicated that ELK1 could promote KIF26B gene expression (Fig. 3D-F). These findings suggested that ELK1 specifically bound to the core promoter of KIF26B and activated its transcription. To further test this hypothesis, we performed a ChIP assay. Using the online JASPAR program, the ELK1 binding site within the core promoter region of KIF26B was identified as 5′-GTC AGG AAG-3′ (Fig. 3G). PCR products amplified using primers specific for the ELK1 binding sites immunoprecipitated in the input and anti-ELK1 antibody groups, but not in the anti-IgG group (Fig. 3I). Additionally, in the anti-ELK1 group, an approximate threefold and 60-fold enrichment IHC was scored using a semi-quantitative scoring system. Intensity was scored as 1, 2, and 3 for weak, moderate, and strong staining, respectively. P and I were multiplied to generate numerical score  (Fig. 3H). Collectively, these findings demonstrated that ELK1 activated the transcription of KIF26B by directly binding to its core promoter region. Additionally, we performed the experiment on the expression of ELK1 and KIF26B in 30 human BC tissues by immunohistochemistry. A positive correlation was found between the expression of KIF26B and ELK1 (R = 0.131, p = 0.0494) (Fig. 4A-C). GEPIA (http:// gepia. cancer-pku. cn/) also revealed a significant positive correlation between KIF26B and ELK1 in BC (Fig. 4D).

Knockdown of KIF26B expression inhibits cell proliferation in vitro
To examine the biological function of KIF26B in BC, we performed MTS, EdU and flow cytometry in MDA-MB-231 and MDA-MB-468 cells following KIF26B siRNA silencing. First, we used western blot analysis to detect knockdown efficiency of KIF26B at 48 h following siRNA transient transfection. Knockdown of KIF26B reduced KIF26B protein expression levels (Fig. 5A, B).
MTS and EDU assays demonstrated that KIF26B knockdown group inhibited proliferation compared with the control group (p < 0.01, Fig. 5C-F). Flow cytometry analysis suggested that downregulation of KIF26B had no effect on apoptosis of MDA-MB-231 and MDA-MB-468 (p > 0.05, Fig. 5G). Nevertheless, KIF26B knockdown led to a significant increase in the frequency of cells in the G 1 phase of the cell cycle, and a decrease in the S phase relative to the control group, both in MDA-MB-231 and MDA-MB-468 cells (p < 0.05, p < 0.01, Fig. 5H). These findings indicated that KIF26B upregulated cell proliferation in vitro, which was likely achieved by promoting cell cycle progression.

KIF26B promotes cell proliferation through regulating cell cycle-related protein expression in vitro
We then detected the expression of several cycle-related proteins to further explore the mechanism. The results revealed that the levels of c-Myc cyclin D1, CDK4, and p-Rb were decreased, whereas p27 was increased. The levels of other proteins such as CyclinE1, CDK2, CDK6, p53, and p21 remained unchanged. We also examined the expression levels of Wnt6, β-catenin, the key regulatory molecules of c-Myc, which were found to be down-regulated following KIF26B knockdown (Fig. 6A).

c-Myc overexpression rescues the inhibited proliferation of KIF26B downregulation in vitro
We complemented the rescue experiments on phenotypes by restoring c-Myc in KIF26B knockdown MDA-MB-231 and MDA-MB-468 cells. As shown in Fig. 6, MTS (Fig. 7A) and EDU (Fig. 7B, C) assays verified that c-Myc upregulation abolished the inhibited proliferation caused by KIF26B downregulation. In addition, flow cytometry analysis suggested that KIF26B knockdown led to a significant increase  13 9 in the frequency of cells in the G1 phase of the cell cycle, and a decrease in the S phase relative to the control group. However, c-Myc overexpression could rescue the influence of KIF26B downregulation on cell cycle distribution (Fig. 7D).

Discussion
BC is the second highest risk factor for women's cancerrelated mortality following lung cancer and is a highly heterogeneous disease involving various genetic and epigenetic alterations [2]. Numerous gene abnormalities associated with the carcinogenesis and progression of BC have been characterized. For example, kinesin family (KIF) has been reported to play an important role in tumor development, metastasis, and drug resistance for BC patients [12][13][14]. Among these, KIF26B has only recently been reported and is poorly characterized. Our research group has proposed that KIF26B acted as a novel oncogene to promote cell proliferation and metastasis by activating the VEGF pathway in gastric cancer [8]. Other studies reported that KIF26B was abnormally expressed in colorectal and ovarian cancer [7,15]. However, the potential role and underlying regulatory mechanism of KIF26B in human BC remains unclear.
In the present study, we found that KIF26B was significantly upregulated in BC tissues and high-malignancy cell lines. In addition, high KIF26B expression levels were markedly associated with clinicopathological parameters, such as positive lymph node metastasis, high tumor grade, and high proliferative index. This was consistent with a previous report [11]. Survival analysis indicated that patients with high KIF26B expression appeared to have poor prognosis, although this was not statistically significant. This finding requires further validation by expanding the sample size. Gu et al. found that knockdown of KIF26B expression inhibited cell proliferation and migration in the MCF-7 and MDA-MB-231 BC cell lines [10]. Accordingly, our results showed In our study, flow cytometry assays also demonstrated that KIF26B knockdown arrested the cell cycle in the G 1 phase, indicating that KIF26B promoting BC proliferation may be associated with cell cycle transition from the G1 to S phase. In contrast to the study by Gu et al. we found that KIF26B had no effect on apoptosis of BC cells. In addition, we investigated the mechanism through which KIF26B promoted cell cycle progression, which had not been addressed previously.
The mechanism underlying abnormal KIF26B expression has never been reported. In the current study, we carried out a transcriptional regulation analysis. We screened and identified the core promoter region of KIF26B using a dualluciferase reporter assay. Furthermore, ELK1 was identified and found to activate KIF26B transcription by directly binding to the core promoter region of KIF26B. The Ets proteins form a large family of transcription factors with diverse functions. They regulate gene expression by direct or indirect DNA binding. ELK1 along with ELK4/SAP-1 and ELK3/SAP-2/Net make up the ternary complex factor (TCF) subfamily of ETS-domain transcription factors, which bind to a GGAA/T motif and activate target gene transcription [16,17]. Demir et al. reported that Elk-1 constantly moved during different stages of the cell cycle and co-localized with kinesin as well as kinesin-like proteins Eg5 and MKLP-1 The input DNA group showed a specific strong band of the predicted size, while the chromatin complex precipitated by antibody against IgG and ELK1 displayed no band or a very weak band, respectively. The primers were specific for the ELK1 binding sites. Three independent experiments were performed, and data are presented as the mean ± SD. *p < 0.05, **p < 0.01, ***p < 0.001 during mitosis in brain tumor cell lines [18]. However, few studies have evaluated kinesin and ELK1 transcriptional regulation. ELK1 is an element of the Ets family of transcription factors that is associated with malignant progression of BC by integrative bioinformatics analysis of transcriptional regulatory programs [19]. Liu et al. reported that higher ELK1 mRNA expression was also associated with worse recurrence-free survival in patients with triple-negative by survival analysis [20]. Moreover, ELK-1 has been found to co-localize with kinesin as well as kinesin-like proteins Eg5 and MKLP-1 during mitosis and was presumed to play a synergistic role in the cell cycle process [18]. Additionally, it was reported that ELK3 knockdown, similar to ELK1 both bind to variants of the GGAA/T motif, suppressed MDA-MB-231 BC cell proliferation with accumulation at the G 1 cell cycle phase [18,21]. Our results showed that ELK1 bound to KIF26B through the GTC AGG AAG motif, which verified the conservative combination of ELK3 and ELK1 with their target genes. Therefore, we hypothesized ELK1 might promote the proliferation of BC cells through the same mechanism as ELK3 and could cooperate with KIF26B in the cell cycle. This hypothesis requires further validation. Our study described the mechanism underlying transcriptional activation of ELK1 on KIF26B for the first time.
To identify the mechanism underlying cancer cell proliferation, it is necessary to measure the expression of cell cycle-related proteins, such as cyclins, CDKs, and CKIs. Classically, the cell cycle is involved in at least three concomitant mechanisms: activation of cyclins and CDKs, suppression of the CDK inhibitors p15 and p21, and the degradation of p27 [22]. The CDK2-cyclin E and CDK4/ CDK6-cyclin D complexes are required for cell cycle transition from the G 1 to S phase [23]. The activities of CDKs and cyclins in the G 1 -S phase could be inhibited by CDK inhibitors, particularly p27 [24]. In this study, western blot analysis showed that KIF26B silencing not only reduced cyclin D1 and CDK4 and p-Rb levels, but also upregulated p27 expression in vitro.
Myc induces cell proliferation by promoting G 1 to S phase transition during cell cycle progression Myc-p27 antagonism in several human solid tumors and leukemia [22]. Myc expression is regulated by several signaling pathways including Wnt/β-catenin, Notch, and TGF-β pathways, and Myc expression levels are strongly correlated with the activation of Wnt signaling molecules in BC [25].Previous studies have revealed that KIF26B links Wnt5a-Ror signaling to the control of cell and tissue morphogenetic behavior [26,27]. However, the relationship between KIF26B and the classical Wnt/β-catenin signaling pathway is largely unknown. Our previous gene expression profile microarray showed that several Wnt signaling components were abnormally expressed following KIF26B knockdown. The levels of Wnt6 and cell membrane-binding molecules of the Wnt pathway, such as FZD and LRP, decreased following KIF26B knockdown. By Three independent experiments were performed, and data are presented as mean ± SD. *p < 0.05, **p < 0.01 contrast, the Wnt pathway inhibitors DKK2, ANKRD, NKD, and one of the β-catenin degradation complex molecules APC were upregulated (fold change ≥ 2, p < 0.05, accession number GSE72307, supplementary Fig. S2). As a result, the Wnt/β-catenin pathway was inactivated.
In our study, Wnt β-catenin and c-Myc expression levels decreased after KIF26B knockdown, and p27 was upregulated. Hence, we propose that the transcriptional factor ELK1 regulates KIF26B expression by directly binding to its core promoter region. KIF26B, in turn, modulates Wnt/β-catenin signaling to up-regulate c-Myc and regulate the levels of cell cycle-related protein, including cyclin D1, CDK4, and p27 to promote cell cycle progress (Fig. 5B).

Conclusion
Overall, these findings indicate that KIF26B is an important oncogene that can promote the proliferation of BC cells and cycle progression via Wnt/β-catenin signaling. Proposed functional action of KIF26B in regulating BC tumorigenesis. The transcriptional factor ELK1 promoted KIF26B expression by directly binding to its core promoter region, and KIF26B acted on the Wnt/βcatenin signaling pathway to up-regulate c-Myc and regulate the cell cycle-related protein expression levels, including cyclin D1, CDK4, p-Rb, and p27 to promote cell cycle progression