Targeting the overexpressed ROC1 induces G2 cell cycle arrest and apoptosis in esophageal cancer cells

Recent reports showed that regulator of Cullins-1 (ROC1) play an important role in tumor progression in a tumor-specific manner. However, the role and mechanism of ROC1 in esophageal cancer remains elusive. Here we demonstrated that ROC1 was overexpressed in esophageal squamous cell carcinomas, which was positive associated with poor prognosis of esophageal cancer patients. ROC1 knockdown significantly inhibited the growth of esophageal cancer cells in vitro and in vivo. Mechanistically, ROC1 silencing induced G2 cell cycle arrest and triggered apoptosis by accumulating the pro-apoptotic protein NOXA. Consistently, the downregulation of NOXA expression via siRNA substantially attenuated apoptosis induced by ROC1 silencing. These findings suggest that ROC1 is an appealing drug target for esophageal cancer.


INTRODUCTION
Esophageal cancer (EC) is one of the leading causes of cancer related death in the world and there are about 400,200 people died from EC in 2012 [1]. Esophageal squamous cell carcinoma (ESCC) is a major histological subtype among all types of the esophageal tumors, especially in China [2]. Limitations of the traditional chemotherapeutic treatment and an advanced stage at the time of diagnosis account for the worse prognosis of patients with ESCC [3][4][5]. It is necessary to identify new anti-ESCC molecular targets and develop novel therapeutic strategies to ameliorate its outcome.
Regulator of Cullins-1 (ROC1), also known as RING box protein-1(RBX1), interacts with different cullin family members, constitutes the catalytic cores and activates CRL/SCF E3 ligases, which regulate different biological process [6,7]. Recently, ROC1 was reported to be overexpressed in a number of cancers [8][9][10][11], which was associated with poor prognosis of several cancers, such as non-muscle-invasive bladder transitional cell carcinoma (NMIBC) [9], gastric cancer [8,10] and liver cancer [12]. Dysfunction of ROC1 induced embryonic death and abnormal meiosis [13]. These reports suggested that ROC1 could act as an attractive anticancer target. However, there is little known about the expression and role of ROC1 in esophageal cancer.
Here, for the first time, we reported that ROC1 was overexpressed in esophageal squamous cell carcinomas tissues, which is positive correlation with poor prognosis of esophageal cancer patients. Knockdown of ROC1 effectively inhibited cell proliferation in vitro and in vivo, induced G2 cell cycle arrest and triggered apoptosis by accumulating the p21, p27, WEE1 or pro-apoptotic protein NOXA. These findings revealed the detailed mechanism for proliferation-inhibition effect of ROC1 knockdown and suggested that ROC1 was an appealing drug target for esophageal cancer.

ROC1 is overexpressed in ESCC tissues and predicts diminished survival in ESCC patients
To investigate the clinical significance of ROC1 in esophageal cancer, we firstly evaluated the expression levels of ROC1 by immunohistochemistry (IHC) staining of the human ESCC tissue arrays. According to the staining intensity, we classified the samples into five groups with increasing staining intensity from the weakest (±, group 1) to the strongest (++++, group 5) ( Figure 1A). Statistical analysis demonstrated that ROC1 was overexpressed in ESCC tissues compared with their corresponding adjacent normal tissues ( Figure 1B), which was confirmed by Western blotting analysis ( Figure 1D). Furthermore, the overexpression of ROC1 protein was negatively correlated with 5-year overall survival rate of cell carcinoma tissues arrays using ROC1-specific antibodies. According to staining intensity, samples were classified into five groups with increasing staining intensity from the weakest (±, group 1) to the strongest (++++, group 5). (B) Classification of tumor samples according to the staining intensity of ROC1. Mann-Whitney Test was used to evaluate the statistical significance of differences between groups. (C) Kaplan-Meier curves for the overall survival rate of patients with esophageal squamous cell carcinoma according to the expression of ROC1 (P = 0.013, log-rank test). Groups 1-3 was designated as low expression and Groups 4-5 was designated as high expression.  ESCC patients determined by Kaplan-Meier analysis (P = 0.013, log-rank test) ( Figure 1C).

Knockdown of ROC1 inhibits the proliferation of esophageal cancer cells
To further assess the role of ROC1 on cell proliferation of esophageal cancer, ROC1 was knockdown by two specific siRNA oligoes, named siROC1-1 and siROC1-2. Results showed that ROC1 silencing significantly inhibited cell proliferation of both Kyse450 and TE1 cells (Figure 2A-2C). Besides, knockdown of ROC1 also effectively inhibited the cell colony formation in both cell lines ( Figure 2D and 2E). The knockdown efficiency of siRNA targeting ROC1 was confirmed by Western blotting assay ( Figure 2F and 2G).

Knockdown of ROC1 induces G2 cell cycle arrest in esophageal cancer cells
To elucidate the mechanism of ROC1 knockdown for cell growth inhibition, we firstly examined the cell cycle profile of the ROC1-silencing cells. As shown in Figure 3A, knockdown of ROC1 triggered G2/M cell cycle arrest in both Kyse450 and TE1 cells. Furthermore, ROC1 silencing induced significant accumulation of WEE1, an inhibitor of G2-M phase transition [14], while a decrease of p-H3, a hallmark of M phase cells [15], indicating that ROC1-silencing cells were arrested at the G2 phase ( Figure 3B).

Knockdown of ROC1 triggers apoptosis in esophageal cancer cells
Next we examined whether apoptosis was also responsible for the anti-proliferative effects of ROC1 silencing. Results showed that knockdown of ROC1 led to  Figure  4A) and caspase-3-actived cells ( Figure 4B). Similarly, ROC1 silencing significantly induced the cleavage of PARP and caspase-3 ( Figure 4C). Furthermore, we found that knockdown of ROC1 led to the loss of mitochondrial membrane potential (∆Ψm) ( Figure 5A), a classical marker of the activation of intrinsic apoptosis, which suggested that knockdown of ROC1 triggered mitochondrial apoptosis.

NOXA plays a critical role for ROC1 silencinginduced apoptosis
To explore the potential mechanism of apoptosis, we investigated systematically the effect of ROC1 knockdown on the expression of the pro-apoptotic and anti-apoptotic proteins. Among these proteins, pro-apoptotic protein NOXA was significantly up-regulated in both cell lines ( Figure 5B), suggesting that NOXA may be a critical mediator for ROC1-silencing-mediated apoptosis.
To further define the role of NOXA in ROC1silencing-induced apoptosis, the expression of NOXA was down-regulated by siRNA in ROC1-silencing cells. NOXA knockdown significantly reduced the induction of apoptosis ( Figure 6A) and the cleavage of PARP in ROC1silencing cells ( Figure 6B). These findings highlighted a pivotal role of NOXA in ROC1-silencing-induced apoptosis.

ROC1 silencing suppresses the growth of human ESCC tumors in murine model
After demonstrating the inhibition efficacy of ROC1 silencing in vitro, we further investigated the growthsuppressive effect of ROC1 knockdown in subcutaneoustransplantation tumor model of human esophageal cancer in mice. Results showed that ROC1 silencing significantly suppressed tumor growth over time while control tumors grew rapidly, as revealed by the tumor growth curve ( Figure 7A, **P<0.01), tumor size ( Figure 7B, top panel) and tumor weight analysis ( Figure 7B, bottom panel, **P<0.01). Moreover, we further examined whether NOXA and apoptosis were activated in vivo. As shown in Figure 7C, ROC1 was effectively downregulated and the expression of NOXA and cleaved PARP were significantly induced upon ROC1 silencing. These observations indicated that ROC1 silencing inhibited esophageal tumor growth both in vitro and in vivo by accumulating the expression of NOXA.

ROC1 knockdown enhances the cytotoxity of cisplatin (CDDP) to ESCC cells
To investigate whether targeting ROC1 could act as a novel chemosensitizer to increase the anti-ESCC activity of CDDP, we treated ESCC cells TE1 and Kyse450 with DMSO or CDDP after siROC1 transfection. Results showed that ROC1 silencing significantly enhanced the cytotoxity of CDDP and inhibited cell viability ( Figure  8A). Moreover, knockdown of ROC1 significantly enhanced CDDP-induced apoptosis, which is evident by the increased expression of cleaved-PARP ( Figure 8B).

DISCUSSION
Despite the improvement in the therapy for ESCC, the prognosis remains poor and the 5-year survival rate remains lower than 25% [16]. Diagnosis at advanced stage and resistance to chemotherapy still persecute the refractory disease [4]. Therefore, it is urgent to find new therapeutic target. Recently, some reports showed that ROC1 is overexpressed in multiple human tumors and contributed to the tumor progression [8][9][10][11][12]. RBX1/ROC1 disruption results in early embryonic lethality [13,17], cancer cell death [11,17] or the inhibition of tumor cell migration [18]. In the present study, we revealed that ROC1 was hyperexpressed in esophageal squamous cell carcinoma tissues compared to adjacent tissues, and the overexpression of ROC1 was negatively associated with 5-year survival rate of the patients, indicating that ROC1 may play an important role in tumor progression of ESCC. Moreover, knockdown of ROC1 significantly inhibited cell proliferation in vitro and in vivo, and enhanced the cytotoxity of CDDP to ESCC cells. These results indicate that ROC1 serves as an attractive anti-ESCC target.
Previous reports showed that knockdown of ROC1 destroy CRL/SCF complexes and thus disrupt CRL/SCF E3 ligase activity. As a result, a trail of CRL substrate were accumulated and thus triggered DNA damage response [19], lead to G2-M cell cycle arrest [11,19], senescence [11,12], apoptosis [11] or protective autophagy [12] in a cancer cell-specific manner. In terms of apoptosis induction, Jia's finding showed that proapoptotic protein PUMA are involved in RBX1 (ROC1) silencing-induced apoptosis [11], whereas NOXA is associated with RBX2 silencing [20]. Here we found that knockdown of ROC1 led to the accumulation of NOXA and NOXA knockdown afforded significant protection against apoptosis, implying that NOXA also play an important role in ROC1 silencinginduced apoptosis for ESCC cells. NOXA has been proven as a particularly relevant therapeutic target for several tumors, such as chronic lymphocytic leukemia (CLL) [21,22], liver cancer [23], lung cancer [24] and ESCC [25]. As a member of the BH3only proteins, NOXA triggered apoptosis at mitochondria or the endoplasmic reticulum [26], which has an important role for apoptosis induction by cytotoxic drugs, including ubiquitin proteasome system (UPS) inhibitors [27][28][29][30][31][32]. Administration of bortezomib significantly increases the expression of NOXA in human melanoma cell grown and inhibits tumor growth [29,31]. NOXA is an important mediator of bortezomib-induced apoptosis in chronic lymphocytic leukemia cells [27]. Treatment ESCC or acute myelogenous leukemia cells with MLN4924, a neddylation inhibitor, lead to inactivation of CRL/SCF E3 ubiquitin ligase and transactivation of NOXA in a cell-type-specific manner to induce cell apoptosis [25,33]. These studies suggest that NOXA perhaps play an important role in ROC1 silencing-induced apoptosis in esophageal cancer cells.
In summary, our study provides the first piece of appealing evidence supporting the notion that ROC1 is an attractive target for ESCC cancer as well as serves as a potential prognosis marker.

Cell culture
Human esophageal squamous cell carcinoma cell lines Kyse450 and TE1 were cultured in Dulbecco's Modified Eagle's Medium (Hyclone) containing 10% FBS (Biochrom AG) at 37°C with 5% CO 2 . Cisplatin was purchased from MedChem Express and dissolved in DMSO.

Immunohistochemistry (IHC) staining of human esophageal cancer tissue array
Human esophageal cancer tissue array was purchased from Shanghai Outdo Biotech Co. Ltd. The detailed clinicopathological characteristics of esophageal squamous cell carcinoma patients are listed in Table 1 for statistical analysis. IHC staining was done with specific ROC1 antibody (Abcam Trading Company Ltd, Shanghai, China) according to our previous reports [24,25]. Briefly, the tissue array sections (4 μm) were dehydrated and subjected to peroxidase blocking. Primary antibodies were added and incubated at 4°C overnight, followed by staining with a GTVisionTMIII Detection System/ Mo&Rb (Gene Tech Company Limited, Shanghai, China). The slides were counterstained with hematoxylin. The stained slides were observed under microscopy, and images were acquired. Based on staining intensity, samples were classified into five groups from the lowest density (±, group 1) to the highest (++++, group 5) [24,25]. Overall survival was calculated using Kaplan-Meier analysis and compared with the log-rank test. Groups 1-3 designated low expression and Groups 4-5 designated high expression.

Collection of esophageal cancer tissues
Fresh primary esophageal cancer tissues and adjacent esophageal tissues were collected from 20 esophageal squamous cell carcinoma patients undergoing resection at the Linzhou Cancer Hospital (Linzhou, Henan, China) from July 2012 to September 2014. The detailed information was showed in Supplementary Table  1. Histological diagnosis and tumor-node-metastasis (TNM) stages of cancers were determined in accordance with the American Joint Committee on Cancer (AJCC) manual criteria for esophageal cancer.

Cell viability and clonogenic survival assay
Kyse450 and TE1 cells were transfected with siControl or siROC1 for 24 h and then seeded in 96-well plates (2.5×10 3 cells per well) for 120 h. Cell viability was determined by MTT and ATPLite assay according to the manufacturer's protocol.
For the clonogenic assay, cells transfected with siControl or siROC1 were seeded into 6-well plates with 500 cells per well in triplicate and incubated for 9 days. The colonies formed were fixed, stained, and counted. The colonies with more than 50 cells were counted.

Western blotting
Kyse450 and TE1 cells were transfected with siControl or siROC1 for 96 h and then proteins were collected for western blot analysis, using antibodies against p21, p27, WEE1, pH3, cleaved caspase-3, cleaved poly (ADP) ribose polymerase (PARP), Pro-Apoptosis Bcl-2 Family Antibody Sampler Kit, Pro-Survival Bcl-2 Family Antibody Sampler Kit, IAP Family Antibody Sampler Kit and glyceraldehyde-3phosphate dehydrogenase (GAPDH) (Cell Signaling, Boston, MA). Densitometric analysis for the quantification relative to GAPDH was performed using the Image J software.

Cell cycle analysis
Kyse450 and TE1 cells were transfected with siControl or siROC1 for 48 h and then cells were harvested, fixed in 70% ethanol at -20°C, stained with propidium iodide (PI, 50μg/ml, Sigma) containing RNase A (30μg/ml, Sigma) at 37°C for 30 min, and analyzed for cell-cycle profile by flow cytometry (Becton Dickinson FACScan). Data were analyzed with ModFit LT software (Verity).

Evaluation of mitochondrial membrane depolarization
Kyse450 and TE1 cells were transfected with siControl or siROC1 and then Mitochondrial Membrane Depolarization was detected with the mitochondrial membrane potential (MMP) assay kit with JC-1 according to the manufacturer's protocol (YeasenInc, Shanghai, China). The data were acquired and analyzed by flow cytometer as described previously [23].

Tumor formation assay
Tumor formation assay was done according to previous reports [12]. Tumor size was determined by caliper measurement. The tumor volume was calculated using the ellipsoid volume formula (Length×Width 2 /2). Tumor tissues were harvested, photographed, and weighed. Protein expression of tumor tissues was evaluated by western blotting analysis using specific antibodies as indicated. Animal experiments were performed in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals. www.impactjournals.com/oncotarget

Statistical analysis
The statistical significance of differences between groups was assessed using the GraphPad Prism5 software. The unpaired 2-tailed t test was used for the comparison of parameters between groups. The Mann-Whitney test was used for data that are not of normal distribution by SPSS software. The standard deviation (S.D.) value was calculated by Excel software. For all tests, two levels of significance (*P<0.05, **P<0.01) were applied.

Author contributions
Ping Chen and Tao Hu conceived and designed the study; Jingyang Zhang, Shuo Li, Zhaoyang Shang, Shan Lin and Peng Gao performed the experiment and acquired the data; Tao Hu analyzed and interpreted the data; Ping Chen and Tao Hu wrote the manuscript; Saijun Mo and Wenbo Cao collected the esophageal squamous cell carcinoma tissues; Yi zhang and Shuaiheng Hou performed the animal model; Ziming Dong reviewed and edited the manuscript. All authors read and approved the submission of this manuscript.