Transcription factor OCT4 promotes cell cycle progression by regulating CCND1 expression in esophageal carcinoma

Cancer Letters 354 (2014) 77–86

Contents lists available at ScienceDirect

Cancer Letters j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / c a n l e t

Transcription factor OCT4 promotes cell cycle progression by regulating CCND1 expression in esophageal carcinoma Zhigang Li a, Xinxin Li b, Chunguang Li c, Yinghan Su d, Wentao Fang a, Chenxi Zhong a, Weidan Ji e, Qian Zhang e, Changqing Su e,* a

Department of Thoracic Surgery, Shanghai Chest Hospital Esophageal Disease Center, Shanghai Jiao-Tong University, Shanghai 200030, China School of Medicine, Shanghai Jiao-Tong University, Shanghai 200025, China c Department of Cardiothoracic Surgery, Changhai Hospital, The Second Military Medical University, Shanghai 200168, China d Department of Biology, Xi’an Jiaotong-Liverpool University, Suzhou 215123, China e Department of Molecular Oncology, Eastern Hepatobiliary Surgical Hospital & National Center of Liver Cancer, The Second Military Medical University, Shanghai 200438, China b

A R T I C L E

I N F O

Article history: Received 23 May 2014 Received in revised form 26 July 2014 Accepted 30 July 2014 Keywords: Cyclin Transcription factor Cell cycle Gene expression Esophageal cancer

A B S T R A C T

The CCND1 gene is overexpressed in esophageal cancer and accelerates cell cycle progression. However, the mechanism whereby the upstream genes or factors directly regulate CCND1 expression remains unknown. By analyzing the 5′-UTR region of the CCND1 gene, we found that this region contains an octamer motif (ATTTTGCAT), which suggests that the expression of CCND1 might be directly associated with octamerbinding transcription factor 4 (OCT4). In this study, the wild-type and the octamer motif-mutanted CCND1 promoters were cloned, and their corresponding luciferase reporter vectors were then constructed to study the molecular mechanism by which OCT4 regulates the expression of CCND1 and influences the biological behaviors of esophageal cancer cells. The results indicated that suppressing the expression of CCND1 and OCT4 in esophageal cancer cells reduced cell proliferative and invasive abilities, induced cell cycle G1-phase arrest, and slowed the growth of xenografts in nude mice. Suppression of OCT4 expression significantly decreased the wild-type CCND1 promoter activity and down-regulated the expression of CCND1, but did not affect the activity of the mutant promoter. Whereas, suppression of CCND1 did not affect OCT4 expression, suggesting that OCT4 regulates CCND1 expression by activating the CCND1 promoter and subsequently promoting cell cycle progression. The results revealed and confirmed that OCT4 is the upstream factor that directly binds to the CCND1 promoter to regulate CCND1 expression, then to promote cell cycle progression and accelerate the proliferation and invasion of esophageal cancer cells. This finding may significantly contribute to elucidating the regulatory mechanism involved in the cell cycle progression of esophageal cancer cells and may aid in screening potential gene targets for the biological therapy of esophageal cancer. © 2014 Elsevier Ireland Ltd. All rights reserved.

Introduction The classical feature of malignant tumors is cell cycle dysregulation, which leads to a state of uncontrolled proliferation, inhibited differentiation and apoptosis of the cells. The underlying mechanism is an imbalance in the expression and secretion of cell cycle-related regulatory factors. Among these factors, Cyclin D1, encoded by the CCND1 gene, is an initiator of the cell cycle. CCND1 expression is up-regulated at an early stage when cells are stimulated by growth factors. When CCND1 is abnormally overexpressed, the result is the continuous stimulation of cell division, leading to

* Corresponding author. Tel.: +86 21 8187 5351; fax: +86 21 8187 5351. E-mail address: [email protected] (C. Su). http://dx.doi.org/10.1016/j.canlet.2014.07.049 0304-3835/© 2014 Elsevier Ireland Ltd. All rights reserved.

uncontrolled cell proliferation and eventually tumor formation [1,2]. The phenomena of CCND1 gene amplification and overexpression are found in a variety of tumors, and there is a certain degree of correlation between the level of CCND1 overexpression and the malignant behaviors of the tumor. Therefore, CCND1 is considered as an oncogene. CCND1 overexpression occurs frequently in esophageal cancer cells, which may be the major reason that esophageal cancer cells have a high proliferative activity. The overexpressed CCND1 acts at the G1/S checkpoint to promote the rapid transit from the G1 phase into the S phase and accelerate the progress of cell division [3]. The downstream mechanism of CCND1 action is clearly understood. The CCND1 gene product binds to cyclin-dependent kinases 4 and 6 (CDK4/6), activates CDK4/6 kinase activity, and enhances the phosphorylation of the tumor suppressor protein Rb. The phosphorylation

78

Z. Li et al./Cancer Letters 354 (2014) 77–86

of Rb diminishes the ability of Rb to bind the transcription factor E2F. E2F is thus released, leading to the expression of a series of downstream target genes and accelerating cell cycle progression [4]. In this process, as an important member of the cyclin-dependent kinase inhibitors (CKIs), the p16 protein can competitively bind to CDK4/6, inhibit CDK4/6 activity, antagonize CCND1 function, and inhibit the progression of the cell cycle. However, the p16 gene is inactivated in most tumor tissues [5]. Our preliminary study demonstrated that the reactivation of p16 expression can induce the cell cycle arrest, promote the apoptosis of cancer cells, and inhibit the growth of tumors [6]. However, the upstream mechanism for the overexpression of CCND1 in esophageal cancer remains unclear. The study of the Eca109 esophageal cancer cell line showed that the activation of NF-kappa B (NF-κB) induced by the degenerative spermatocyte homolog 1 (DEGS1) is one of the causes of CCND1 overexpression in esophageal cancer cells [7]. Enhancing the expression of lactate dehydrogenase A (LDHA) in esophageal cancer cells promotes CCND1 expression by activating the AKT signaling pathway, leading to cancer cell proliferation [8]. The increased expression of CCND1 in esophageal cancer cells is closely associated with the activation of signaling pathways that involve the epidermal growth factor receptor (EGFR), Notch, c-myc, and Wnt2/βcatenin [9,10]. In summary, there are numerous factors that affect CCND1 expression in esophageal cancer. However, studies on the upstream genes or factors that directly regulate CCND1 expression are rare. We analyzed the 5′-UTR region of the CCND1 gene and found that this region contains an octamer motif (ATTTTGCAT), which is a sequence that potentially binds to OCT4. We hypothesized that CCND1 expression may be related to octamer-binding transcription factor 4 (OCT4). OCT4 is a transcription factor of the POU (Pit-Oct-Unc) family, which contains 324 amino acids and has a molecular weight of approximately 38 kD. OCT4 is a type of sequence-specific DNA-binding protein that binds to a target promoter or octamer motif binding site in an enhancer region to activate target gene transcription and expression. OCT4 plays a role in maintaining the stemness and regulating the differentiation of stem cells. This factor participates in cell differentiation, embryonic development, and other important developmental stages of life. OCT4 is primarily expressed in embryonic stem cells. From cell differentiation to maturation, OCT4 expression is gradually lowered until the capacity for expression is essentially lost in normal tissues [11]. Studies have proven that approximately 623 protein-encoding genes and the regulation of the promoters for five mRNAs are associated with OCT4 in embryonic stem cells; most of these genes are common targets of OCT4, SOX2, and Nanog [12]. Therefore, the transcription factors OCT4, SOX2, and Nanog play important roles in the signaling pathways within embryonic stem cells, with OCT4 at the top of the regulatory hierarchy of these three transcription factors. In the process of embryogenesis, abnormal OCT4 gene expression may ultimately lead to fetal abnormalities or death. Recent studies have found that OCT4 also plays an important role in the carcinogenesis and development of many malignant tumors. OCT4 expression has been found in various tumors, including esophageal cancer. Increasing evidence indicates that OCT4 may be a type of oncogene involved in carcinogenesis and cancer development that plays a critical role in maintaining the activity of tumor stem cells [13,14]. In this study, we detected CCND1 and OCT4 expression in esophageal cancer tissues and analyzed the relationship of the expression of these genes to the clinical pathological features and postoperative survival. Furthermore, we investigated the mechanism whereby OCT4 regulates CCND1 expression by conducting experiments with esophageal cancer cells. The results contribute to understanding the mechanisms leading to the biological characteristics of malignant esophageal cancer and may also be relevant

to screening gene targets for the biological therapy of esophageal cancer. Materials and methods Construction of the experimental vectors The CCND1-shRNA plasmid (pGenesil-shCCND1), the OCT4-shRNA plasmid (pGenesil-shOCT4), and the negative control plasmid [pGenesil-shMC, with the enhanced green fluorescent protein reporter gene (EGFP)] were provided by Wuhan Genesil Biotechnology Co., Ltd. (Wuhan, China). The 19-nt sense DNA of CCND1shRNA (5′-GCC CTG CTG GAG TCA AGC C-3′) targets the base pairs 901–919 of the CCND1 gene (M64349.1), and the 19-nt sense DNA of OCT4-shRNA (5′-CCC TCA CTT CAC TGC ACT G-3′) targets the base pairs 1233–1253 of the OCT4 gene (DQ486513.1). The mock control shRNA vector (shMC: 5′-GAC TTC ATA AGG CGC ATG C-3′) was concomitantly constructed. An adenovirus carrying the full-length OCT4 cDNA (Ad5-OCT4) and the corresponding control adenovirus (Ad5-EGFP) were constructed and stored by the Department of Molecular Oncology, Eastern Hepatobiliary Surgery Hospital, Second Military Medical University (Shanghai, China) [15]. The luciferase reporter vectors (pGL3WP-Luc, pGL3MP-Luc) controlled by the wild-type CCND1 gene promoter (WP; nucleotides 2501–3178, GenBank Z29078.1) and the mutant promoter (MP: The base pairs “ATTTGCAT” in WP from −252 to −245 were replaced by base pairs “ATCTGTAT”) were constructed and stored by the Department of Molecular Oncology, Eastern Hepatobiliary Surgery Hospital, Second Military Medical university (Shanghai, China) [15]. Cell culture and the establishment of genetically modified cell sublines The esophageal squamous cell carcinoma cell lines (Eca109, TE1) and the normal esophageal squamous cell line (HET-1A) were purchased from the Cell Bank of Shanghai Institutes for Biological Sciences and Cell Biochemistry, Chinese Academy of Sciences. The cells were cultured according to the product instructions. The shRNA plasmids were transfected into the Eca109 and TE1 cancer cell lines with Lipofectamine 2000 (Invitrogen Corporation Shanghai Representative Office, Shanghai, China) according to the manufacturer’s instructions. Briefly, cells were seeded into six-well plates at a density of 1 × 105 cells/well and transfected with 20 μg of shRNA plasmids/well. After a 48-hour incubation and selection with G418 (400 μg/ml), the cells were harvested for the experiment. The normal esophageal squamous cell line HET-1A was infected with the adenovirus Ad5-OCT4 and the control adenovirus Ad5-EGFP, respectively, at the multiplicity of infection (MOI) of 50 or 100 pfu/cell. Fluorescent microscopy was used to observe the EGFPpositive cells 48 h later after infection. The cells were then harvested for further experiments. Detection of CCND1 and OCT4 expression The parental cell lines and their genetically modified cell sublines aforementioned were seeded in 24-well plates at a density of 5 × 104 cells/well. After culturing for 48 h, the cells were collected and lysed by repeated freeze–thaw cycles at −80 °C, and the proteins were isolated. Western blot analysis was used to detect the expression of CCND1 and OCT4 by using the mouse anti-OCT4 (Santa Cruz Biotechnology, Inc., Santa Cruz, CA, USA) and mouse anti-CCND1 (Cell Signaling Technology, Inc., Beverly, MA, USA). Determination of CCND1 promoter activity The cells were seeded into 24-well plates at a density of 5 × 105 cells/well. After culturing for 24 h, the cells were transfected separately with 200 ng/well pGL3WPLuc and pGL3MP-Luc using Lipofectamine 2000. The cells were co-transfected with 20 ng/well of pRL-TK, and pGL3-Control was used as a positive control. After a 48hour transfection, the Dual-Luciferase Reporter Assay System (Promega Corporation, Shanghai, China) was used according to the kit instructions to perform the determination of luciferase activity in the tested cells on an LB-9506 Luminometer (Berthold Technologies, Germany). Chromatin immunoprecipitation (ChIP) assay ChIP assay was performed according to the protocols provided by the kit manufacturer (Merck Millipore, Shanghai, China). Cells were seeded in six-well plates at 1 × 106 cells/well and incubated with 1% formaldehyde for 10 min at room temperature, and then subjected to prepare the chromatin samples. The prepared samples were managed to immunoprecipitation with anti-OCT4 antibody (Santa Cruz Biotechnology, Inc.) or negative control rabbit IgG (Cell Signaling Technology, Inc.). The immunoprecipitated DNA samples were analyzed by PCR using the primers (forward: 5′-AGA TTC TTT GGC CGT CTG TC-3′; reverse: 5′-GCA GCG AGG GGC AGA GCC CA3′). PCR was carried out for 35 cycles by using a step cycle of 95 °C for 30 s, 56 °C

Z. Li et al./Cancer Letters 354 (2014) 77–86

for 50 s, 72 °C for 1 min, and followed by 72 °C for 5 min. A 366-bp product was analyzed by electrophoresis on 5% agarose gel. Distilled water was used instead of the immunoprecipitated DNA samples as a negative control, and amplification of soluble chromatin prior to immunoprecipitation was used as an input positive control.

79

Statistical analysis The cell-based experiments were performed three times each. All experimental data were represented as “the mean ± standard deviation”. The PASW Statistics 18 software was used for the t test analysis; P < 0.05 was considered to be statistically significant.

Cell proliferation experiment

Results A cell proliferation Kit I (Roche Diagnostics GmbH, Germany) was used to determine the cell viability of the cells with different gene expression status. The abovedescribed parental cell lines and their genetically modified cells were seeded into 96-well plates at a density of 5 × 103 cells/well. The initial confluency was approximately 50%. The cells were cultured for 24, 48, 72, and 96 h, with eight duplicate wells for each time point. After discarding the medium, 100 μl/well of 0.1 mol/l phosphate-buffered saline (PBS) solution was added; 10 μl/well 3-(4,5-dimethylthiazol2-yl)-2,5-diphenyl tetrazolium bromide (MTT) labeling reagent was then added to a final concentration of 0.5 mg/ml. After incubating for 4 h, 100 μl/well of solubilization solution (10% SDS in 0.01 mol/l HCl) was added, and the plates were then incubated overnight. A microplate reader (Model 550, Bio-Rad Laboratories Inc., Shanghai, China) was used to measure the absorbance at a wavelength of 570 nm, with a reference of wavelength 655 nm. The cell viability was calculated (A570 of the cells in experimental wells/A570 of the cells in control wells × 100%), and the cell survival curve was plotted.

Cell invasion assay Cell invasion was evaluated by transwell chamber assay (Millipore, Billerica, MA, USA). Totally 5 × 104 cells were seeded on an 8-μm pore size transwell coated with extracellular matrix (ECM) (1:6) (BD Biosciences, Shanghai, China). After incubation at 37 °C for 72 h, the cells adherent to the upper surface of the filter were removed and stained with crystal violet, and the cell number was counted.

Cell cycle analysis via flow cytometry All the cells were seeded into six-well plates at a density of 5 × 105 per well and cultured to an exponential growth phase. The cells were collected and fixed with pre-cooled 75% ethanol at 4 °C overnight. After washing two times with PBS, the propidium iodide (PI) solution containing RNase was added, and the cells were stained for 30 min in the dark and detected by flow cytometry.

Animal model experiments The experiments of esophageal carcinoma xenograft mouse models were approved by the Animal Ethics Committee of the Second Military Medical University. Forty healthy purebred 4-week-old male BALB/C mice were provided by the Shanghai SLAC Animal Center, Chinese Academy of Sciences, Shanghai, China. Model I: Logarithmic phase cells, including the parental Eca109 cells and Eca109 cells transfected with the shRNA plasmids (pGenesil-shCCND1, pGenesil-shOCT4, and pGenesilshMC) were injected subcutaneously into the right side of 28 nude mice at a dose of 1 × 106 cells/100 μl for each mouse. Tumor formation was observed weekly after cell inoculation. After the tumors appeared, mice with either the largest or smallest tumors in each group were excluded from the study, and the tumor volume in the remaining five mice was routinely measured. Model II: Eca109 cells transfected with pGenesil-shOCT4 were injected subcutaneously into the right side of 22 nude mice at a dose of 1 × 106 cells/100 μl for each mouse. After the tumors were formed, mice with either the largest or smallest tumors were excluded from the study. The remaining 20 mice were randomly divided into four groups. Among these four groups, the mice in two groups received multiple injections of Ad5-EGFP or Ad5OCT4 adenoviruses at a dose of 2 × 108 pfu/100 μl for each mouse, administered every other day with a total of five injections. The same volume of PBS was injected synchronously in another control group. The other remaining group received no additional treatment. The tumor volume was routinely measured and was calculated using the formula “maximum diameter × minimum diameter2 × 0.5”. When a mouse in any group presented with an oversized tumor, with a tumor volume greater than 2000 mm3 (the maximum volume approved by the Ethics Committee), the experiment was terminated, and the mice were sacrificed under ether anesthesia and the tumors were collected. A digestion solution (0.1% collagenase with 0.01% hyaluronidase and 0.002% DNase) was used to digest a portion of the tumor tissue for 2 h at 37 °C to prepare a single-cell suspension for the cell cycle analysis via flow cytometry (as previously described). Another portion of the tumor tissue was fixed with 10% neutral buffered formalin for 6 h, and paraffin-embedded sections were prepared. The expression of CCND1, OCT4 and proliferating cell nuclear antigen Ki67 (with the rabbit anti-Ki67 polyclonal antibody, Abcam, Cambridge, MA, USA) in the tumor tissue was determined by immunohistochemical assay. The percentage of positive cells in each section was assessed within five high-power fields (40 × objective).

Relationship between CCND1 and OCT4 expression in esophageal cancer cell lines The esophageal squamous cell carcinoma cell lines Eca109 and TE1 were transfected with the negative control plasmid pGenesil-shMC. Under fluorescent microscopy, we observed five medium-power fields (20 × objective lens) and counted the EGFP-positive cells to calculate the transfection efficiency of the plasmid. At 48 h later after transfection, the transfection efficiencies of the Ec109 and TE1 cells were 36.46% and 29.82%, respectively. After the cells were selected with G418, the transfection efficiencies of the Ec109 and TE1 cells were increased to 89.64% and 96.38%, respectively (Fig. 1A). The normal esophageal squamous cell line HET-1A was infected with the control adenovirus Ad5-EGFP, and the infection efficiency was 76.92% and 92.46% when the MOI was 50 and 100 pfu/cell, respectively (Fig. 1B). The Western blot analysis revealed that Eca109 and TE1 cells were positive for the expression of CCND1 and OCT4, but no expression was found in HET-1A cells. CCND1 expression was significantly decreased in the Eca109 and TE1 cells after they were transfected with pGenesil-shCCND1, while the expression of OCT4 was not significantly changed. Both CCND1 and OCT4 were significantly down-regulated after Eca109 and TE1 cells were transfected with pGenesil-shOCT4. HET-1A cells transfected with Ad5-OCT4 (MOI = 100 pfu/cell) showed positive expression of CCND1 and OCT4 (Fig. 1C). The results suggested that OCT4 increases CCND1 expression and is an upstream regulatory factor for CCND1. OCT4 increases CCND1 expression by regulating CCND1 promoter activity The gene sequence analysis revealed that the CCND1 promoter contains an octamer motif sequence (ATTTTGCAT) (Fig. 2A). We constructed a wild-type CCND1 promoter-luciferase reporter plasmid (WP) and a plasmid containing the promoter with an octamer motif mutant sequence (MP)-Luciferase reporter to examine the CCND1 promoter activity in esophageal cancer cells and normal esophageal squamous cells. Both the CCND1 wild-type and mutant promoter activities were significantly higher in the esophageal cancer cell lines Eca109 and TE1 compared with the normal cell line HET-1A. The wild-type CCND1 promoter activity was significantly higher than the mutant promoter in the Eca109 and TE1 cells. The wild-type promoter activity was approximately the same as that of the mutant promoter in the HET-1A cells. After transfection with pGenesilshCCND1 to knock down the expression of CCND1, both the wildtype and mutant promoter activities were decreased, but this decrease was not statistically significant. After transfection with pGenesil-shOCT4 to knock down OCT4 expression, the wild-type promoter activity was significantly decreased, but there was no significant decrease in the activity of the mutant promoter (Fig. 2B). To identify the binding of OCT4 to the putative overlapping binding site within CCND1 promoter, ChIP assay was performed. The results showed that a corresponding band could be amplified from Eca109 cellular DNA samples immunoprecipitated by anti-OCT4 antibody, as that of amplification with input, while no band was

80

Z. Li et al./Cancer Letters 354 (2014) 77–86

Fig. 1. Relationship between CCND1 and OCT4 expression in esophageal cancer cell lines. (A) The esophageal squamous cell carcinoma cell lines (Eca109, TE1) were seeded into six-well plates at a density of 1 × 105 cells/well and transfected with 20 μg/well of pGenesil-shMC plasmid, then selected with G418. At 48 h later after transfection, we counted the EGFP-positive cells within five medium-power fields (20 × objective lens), and calculated the transfection efficiency of the plasmid under fluorescent microscopy, with or without selection of G418. (B) The normal esophageal squamous cell line (HET-1A) was seeded into six-well plates at a density of 1 × 105 cells/well and infected with the control adenovirus Ad5-EGFP at the multiplicities of infection (MOI) of 50 and 100 pfu/cell, and the EGFP-positive cells at 48 h later after infection were observed under fluorescent microscopy. (C) Eca109 and TE1 cells were seeded into six-well plates at a density of 1 × 105 cells/well, transfected with 20 μg/well of the indicated shRNA vectors and then selected with G418. HET-1A was infected with the adenovirus Ad5-OCT4 or Ad5-EGFP at the multiplicity of infection (MOI) of 100 pfu/cell. All the parental cell lines and their genetically modified cell sublines were harvested and detected the expression of CCND1 and OCT4 by Western blot. The relative expression levels of CCND1 and OCT4 were normalized with glyceraldehyde-3-phosphate dehydrogenase (GAPDH); **P < 0.01, ***P < 0.001.

amplified in the negative controls (rabbit IgG and H2O), suggesting that OCT4 specifically binds to the predicted motif of CCND1 promoter to directly regulate the CCND1 transcription and expression (Fig. 2C). Influence of CCND1 and OCT4 expression on proliferation and invasion of esophageal cancer cells Cells were seeded in 96-well plates at a density of 5 × 103 cells/ well. The initial cell confluency was set at 50% (0 h). The cell viability was examined at 24, 48, 72, and 96 h. The results indicated that cell viabilities of Eca109 and TE1 cells gradually increased with extended culture periods and reached 105.63% and 99.05% at 96 h, respectively. Using the shRNA plasmid to knock down CCND1 expression, cell viabilities were significantly reduced and

reached 81.70% (P = 0.0060) and 78.61% (P = 0.0208) in the Eca109 and TE1 cells, respectively, at 96 h. After knocking down OCT4 expression, cell viabilities were further decreased to 72.79% (P = 0.0029) and 68.82% (P = 0.0056) in the Eca109 and TE1 cells, respectively, at 96 h. Cell viability of the normal esophageal squamous cell line HET-1A was lower than that of the cancer cells, being 86.14% after a 96-hour incubation, but significantly increased to 98.04% (P = 0.0376) after the cells were infected with Ad5-OCT4 with an MOI of 100 pfu/cell (Fig. 3A). To investigate the effect of OCT4 on cancer cell invasive ability, Eca109 cells were measured by transwell invasion assay. After silenced OCT4 expression, the relative invasive cell number of Eca109 cells was significantly decreased compared with the parental cells (P = 0.0024; Fig. 3B).

Z. Li et al./Cancer Letters 354 (2014) 77–86

81

Fig. 2. OCT4 increases CCND1 expression by regulating CCND1 promoter activity. (A) The CCND1 promoter contains an octamer motif sequence (ATTTTGCAT) from −252 to −245. According to the octamer motif sequence, a wild-type CCND1 promoter-luciferase reporter plasmid (pGL3WP-Luc) and a plasmid containing the promoter with an octamer motif mutant sequence-Luciferase reporter (pGL3MP-Luc) were constructed. (B) The parental cell lines and their genetically modified cell sublines were seeded into 24-well plates at a density of 5 × 105 cells/well and co-transfected with 200 ng/well pGL3WP-Luc or pGL3MP-Luc together with 20 ng/well of pRL-TK using Lipofectamine 2000. At 48 h later after transfection, cells were harvested and used to detect the relative activity of the promoters with the Dual-Luciferase Reporter Assay, normalized with the activity of pGL3-Control in every cell line; *P < 0.05, **P < 0.01, ***P < 0.001. (C) ChIP assay in Eca109 cells showed that OCT4 binds to the CCND1 promoter overlapping octamer motif sequence. Immunoprecipitation of chromatin DNA fragments was carried out with anti-OCT4 antibody. The rabbit IgG and H2O were used as negative controls; the input chromatin sample was used as a positive control for PCR.

82

Z. Li et al./Cancer Letters 354 (2014) 77–86

Fig. 3. CCND1 and OCT4 promote the proliferation of esophageal cancer cells. (A) The parental cell lines and their genetically modified cell sublines were seeded into 96well plates at a density of 5 × 103 cells/well. The initial confluency was approximately 50%. The cells were cultured for 24, 48, 72, and 96 h, and used to determine the cell viability by a cell proliferation Kit I (MTT assay), with eight duplicate wells for each time point; *P < 0.05, **P < 0.01. (B) Eca109 cells were seeded on an 8-μm pore size transwell coated with extracellular matrix (ECM), and the invasive cell number were stained with crystal violet and counted after incubation at 37 °C for 72 h; **P < 0.01.

Relationship between CCND1 and OCT4 expression and the regulation of cell cycle in esophageal cancer cells Flow cytometry was used to examine the changes of cell cycle. Compared to the parental esophageal squamous cells, knocking down the expression of CCND1 and OCT4 decreased the percentages of cells in S phase and increased the percentages of cells in G1 phase in both Eca109 and TE1 cells. However, HET-1A cells infected with Ad5-OCT4 had an opposite pattern, i.e., a decrease of G1 phase percentage and an increase of S phase percentage (Fig. 4). Effect of OCT4-regulated CCND1 expression in esophageal cancer xenograft mouse models Esophageal cancer cells were injected subcutaneously into BALB/C nude mice to establish the xenograft models. The formation and growth speed of tumors were different among the mice treated with the Eca109 cells that were transfected with the different modified genes. The tumor formation and growth speed were faster with the parental Eca109 cells and the Eca109 cells transfected with pGenesilshMC, followed by the Eca109 cells transfected with pGenesilshCCND1, while the cells transfected with pGenesil-shOCT4 had the lowest tumor formation and slowest growth speed (Fig. 5A, left panel). This result indicated that the gene in the upstream of cell cycle regulation play more significant role in cancer cell proliferation. For the xenograft tumors established by Eca109 cells transfected with pGenesil-shOCT4, the tumor grew significantly faster after the administration of restoration of OCT4 expression mediated by adenovirus (Fig. 5A, right panel). The xenograft tumor specimens were collected to perform an immunohistochemical assay for gene expression and flow cytometry for cell cycle analysis. Compared with the parental Eca109-established xenografts, CCND1 expression was

decreased in the Eca109 xenografts with knockdown of CCND1 expression. In addition, the number of Ki67-positive cells was decreased, the percentage of cells in S phase was decreased, and the percentage of cells in G1 phase was increased. The expression of both OCT4 and CCND1 was decreased in the Eca109 xenografts with knockdown of OCT4 expression. In addition, the number of Ki67-positive cells was decreased, and the cell cycle presented the same pattern of a decreased S phase ratio and an increased G1 phase ratio. For the Eca109 xenografts with knockdown of OCT4 that were administered the Ad5-OCT4 virus, the expression of OCT4 and CCND1 and the number of Ki67-positive cells were increased, whereas the cell cycle presented a completely opposite pattern with an increased ratio in the S phase and a decreased ratio in the G1 phase (Fig. 5B, C and D). Discussion Cell cycle is divided into the G1, S, G2, and M phases. Some cells are in a quiescent stage outside of the cell division cycle, which are in the G0 phase. Each phase has a series of regulatory factors involved in controlling the cell cycle. The balance between these activators and inhibitors is the basis for ensuring the normal progression of cell cycle. The key regulatory factors are cyclin-dependent kinases (CDKs), which can be activated by cyclins to induce the expression of transcription factor E2F and promote the progression of cell cycle. Therefore, CDKs, cyclins, and E2F are considered to be the positive regulatory factors of cell cycle [16,17]. Another class of factors, the cyclin-dependent kinase inhibitors (CKIs), is able to inhibit CDK activity and delay cell cycle progression. Therefore, CKIs are considered to be the negative regulatory factors of cell cycle. CKIs include the proteins of p15, p16, p18, p19, p21, p27, and p57 [18–20]. Under normal circumstances, the expression of cyclins is

Z. Li et al./Cancer Letters 354 (2014) 77–86

Fig. 4. CCND1 and OCT4 promote the cell cycle progression of esophageal cancer cells. The parental cell lines and their genetically modified cell sublines were seeded into six-well plates at a density of 5 × 105 per well and collected, fixed with pre-cooled 75% ethanol at 4 °C overnight, stained with propidium iodide (PI) solution, and detected by flow cytometry; *P < 0.05, **P < 0.01, ***P < 0.001.

83

84

Z. Li et al./Cancer Letters 354 (2014) 77–86

Fig. 5. CCND1 and OCT4 promote the growth of esophageal cancer cell xenografts in nude mice. (A) The parental Eca109 cells and their genetically modified cell sublines were injected subcutaneously into the right side of nude mice at a dose of 1 × 106 cells/100 μl for each mouse, and then without (left panel, Model I) or with (right panel, Model II) treated with the Ad5-OCT4 and Ad5-EGFP adenoviruses at a dose of 2 × 108 pfu/100 μl for each mouse, administered every other day with a total of five injections. Tumor volume was routinely measured for 35 days after cell inoculation; n = 5 in each group, *P < 0.05, **P < 0.01. (B) The xenograft tumors were collected, fixed with 10% neutral buffered formalin for 6 h, and the paraffin-embedded sections were prepared for examining the expression of CCND1, OCT4 and Ki67 by immunohistochemistry, original magnification ×400; bar: 10 μm. (C) The percentage of positive cells for CCND1, OCT4 and Ki67 expression in each section was assessed within five highpower fields (40 × objective); **P < 0.01, ***P < 0.001. (D) The xenograft tumors were prepared a single-cell suspension for the cell cycle analysis via flow cytometry; *P < 0.05, **P < 0.01.

Z. Li et al./Cancer Letters 354 (2014) 77–86

at a low level during the time-dependent cell cycle phases. Cyclins, together with CKIs, constitute the regulatory systems for cell cycle. Different subtypes of cyclins activate different subtypes of CDKs and then regulate different phases of cell cycle. For example, the CCND1 gene, encoding cyclin D1 protein, plays a role in promoting the progression from the G1 phase to the S phase. Most human cancers have a cell cycle dysfunction that is closely related to the occurrence and development of the tumors. Numerous studies have shown that, in esophageal cancer cells, CCND1 expression is up-regulated and activates CDK4/6, whereas the p16, p53 and Rb are deactivated, resulting in the dysregulation of DNA-damage repair system, blockage of apoptotic pathway, and acceleration of cell cycle [21]. The typical feature of malignant tumor is the dysregulation of cell cycle with cells undergoing uncontrolled cell division and proliferation. DNA is the basic substance for cell proliferation and division. DNA synthesis is accelerated during cellular carcinogenesis, in which a large number of cells enter cell cycle, and the cells in the G1 phase rapidly enter into the S phase. In the process of culturing esophageal squamous epithelial cells, the dominant negative mastermind-like 1 (DNMAML1), an inhibitor of Notch signaling pathway, activates CCND1 expression and promotes the development of cell hyperplasia [10]. A study of 165 esophageal cancer specimens showed that the increased phosphorylation level of the mammalian target of rapamycin (mTOR) signal protein was related to CCND1 overexpression, which affected cancer cell proliferation, metastasis and invasion [22]. Another study of 181 esophageal cancer surgical specimens revealed that the positive expression rate for CCND1 was 37.6% and gradually increased with the decreased degree of cancer cell differentiation, suggesting that there is a certain correlation between CCND1 and cancer cell differentiation [23]. An immunohistochemical analysis of multiple parameters was performed in 100 cases of esophageal cancer. The results indicated that the positive expression rate of CCND1 was as high as 67.0%, which was accompanied by the upregulated expression of mutant type p53 and MDM2, and the downregulated expression of p21, p16 and pRb. CCND1 expression was associated with the postoperative disease-free survival and overall survival of the patients. Therefore, CCND1 expression can be used as a prognostic indicator for esophageal cancer patients [24]. In summary, the disorder of cell cycle regulation is one of the important molecular events in the development of esophageal cancer, which causes cancer cells to have more proliferative and metastatic capacity. This characteristic provides a basis for the molecular diagnosis and biological therapy of this type of cancer. CCND1 overexpression can be found in various types of human tumors, suggesting that it is a common feature of malignant tumors [1,25]. There have been many literatures about the downstream signaling pathways and mechanisms for CCND1, which have mostly focused on the fact that the product of CCND1 gene can activate CDKs to accelerate cell cycle progression. However, studies of the upstream regulatory factors and the direct activating mechanism for CCND1 overexpression in cancer cells have rarely been reported and many questions remain to be elucidated. It is already known that the activation of the EGFR, Notch, c-myc, Wnt2/β-catenin, AKT, and mTOR signaling pathways is associated with CCND1 expression in cancer cells [9,10,22,26]. There are many regulatory factors involved in these pathways; most of the factors have an indirect effect on CCND1 expression, and the upstream genes or factors that directly regulate CCND1 expression remain unclear. We performed a gene sequence analysis of the full-length of CCND1 gene and found that there is an octamer motif sequence (ATTTTGCAT) in the CCND1 promoter, which is in accordance with the bonding-site sequence for OCT4. We hypothesized that CCND1 expression may be related to OCT4. To confirm whether OCT4 plays a role in enhancing CCND1 expression through the octamer motif in esophageal cancer cells, we applied a series of gene engineering techniques and promoter

85

activity experiments to prove the hypothesis on a cellular level and with animal experiments. Our study showed that CCND1 and OCT4 are both overexpressed in two esophageal cancer cell lines, Eca109 and TE1, and are negative in normal esophageal squamous cell line HET-1A. Knocking down OCT4 expression in Eca109 and TE1 cells can decrease the expression of CCND1, but knocking down CCND1 expression in Eca109 and TE1 cells cannot affect the expression of OCT4. The expression of OCT4 mediated by adenovirus vector in the normal esophageal squamous cell line HET-1A can result in the positive expression of CCND1. These results indicated that OCT4 is an upstream regulatory factor for CCND1 expression. We cloned the CCND1 promoter and a promoter with a mutant octamer motif sequence and constructed their luciferase reporter plasmids to examine the relative promoter activity of CCND1 in the experimental cell lines. The results demonstrated that CCND1 promoter activity was consistent with OCT4 expression in both the esophageal squamous and normal cell lines. The CCND1 wild-type promoter and mutant promoter activities were significantly higher in the OCT4-positive esophageal cancer cell lines compared with the OCT4-negative normal cell line. Knocking down OCT4 expression decreased the promoter activity of CCND1 but no significant decrease in activity was found with the mutant promoter. The ChIP assay further demonstrated that OCT4 specifically binds to the predicted motif of CCND1 promoter to directly regulate the CCND1 transcription and expression. OCT4 is a transcription factor of the POU family of factors. It is a type of sequence-specific DNA-binding protein that binds to the octamer motif binding site in the promoter or enhancer region of the target genes to activate gene transcription and expression. OCT4 has an effect on maintaining the stemness and regulating the differentiation of stem cells, and participates in cell differentiation, embryonic development, and other important life developmental stages. Recent studies have shown that OCT4 also plays an important role in carcinogenesis and the development of many malignant tumors [14,27]. OCT4 expression can be found in various tumors, including esophageal cancer [14,28]. Increasing evidence has proven that OCT4 is an oncogene involved in carcinogenesis and cancer development and plays a critical role in maintaining tumor stem cell activity. This study showed that there is a subpopulation of OCT4positive cells within esophageal cancer cell lines. This subpopulation of tumor cells may be the progenitor cells with stem cell features in esophageal cancers and are associated with the proliferation, invasion, recurrence, metastasis, and other malignant features of esophageal cancer. OCT4 induces CCND1 expression and promotes cell cycle progression, causing esophageal cancer cells to rapidly proliferate, readily relapse, and metastasize early. In the in vitro experiments and the esophageal carcinoma xenograft mouse model, we further confirmed that knocking down CCND1 or OCT4 expression can decrease the proliferation rate of esophageal cancer cells, induce a cell cycle arrest, and slow the growth of xenograft tumors. The knockdown of OCT4 expression had a significant effect on these changes. Re-expression of OCT4 can quickly reverse the effect of CCND1 or OCT4 silence and promote the growth of xenograft tumors. In summary, we found and confirmed that OCT4 is an upstream regulatory factor that directly up-regulates CCND1 expression in esophageal cancer cells. Our results will help to elucidate the regulatory mechanism involved in the cell cycle progression of esophageal cancer cells and aid in screening potential gene targets for the biological therapy of esophageal cancer. Acknowledgments This work was supported by the Plan Project of Shanghai Outstanding Academic Leaders (13XD1400300 to C. Su), the National Science and Technology Major Significant Projects of New Drugs Cre-

86

Z. Li et al./Cancer Letters 354 (2014) 77–86

ation (2014ZX09101003 to C. Su), and the National Natural Science Foundation of China (81370552 to C. Su, 81372472 to Z. Li, 81301829 to C. Li). Conflict of interest The authors have no conflicts of interest to disclose. References [1] J. Chung, H. Noh, K.H. Park, E. Choi, A. Han, Longer survival in patients with breast cancer with cyclin d1 over-expression after tumor recurrence: longer, but occupied with disease, J. Breast Cancer 17 (2014) 47–53. [2] R.G. Pestell, New roles of cyclin D1, Am. J. Pathol. 183 (2013) 3–9. [3] S. Ye, K.B. Lee, M.H. Park, J.S. Lee, S.M. Kim, p63 regulates growth of esophageal squamous carcinoma cells via the Akt signaling pathway, Int. J. Oncol. 44 (2014) 2153–2159. [4] D.B. Rivadeneira, C.N. Mayhew, C. Thangavel, E. Sotillo, C.A. Reed, X. Graña, et al., Proliferative suppression by CDK4/6 inhibition: complex function of the retinoblastoma pathway in liver tissue and hepatoma cells, Gastroenterology 138 (2010) 1920–1930. [5] V. Srivastava, B. Patel, M. Kumar, M. Shukla, M. Pandey, Cyclin D1, retinoblastoma and p16 protein expression in carcinoma of the gallbladder, Asian Pac. J. Cancer Prev. 14 (2013) 2711–2715. [6] H. Hu, Z. Li, J. Chen, D. Wang, J. Ma, W. Wang, et al., P16 reactivation induces anoikis and exhibits antitumour potency by downregulating Akt/survivin signalling in hepatocellular carcinoma cells, Gut 60 (2011) 710–721. [7] W. Zhou, X.L. Ye, Z.J. Sun, X.D. Ji, H.X. Chen, D. Xie, Overexpression of degenerative spermatocyte homolog 1 up-regulates the expression of cyclin D1 and enhances metastatic efficiency in esophageal carcinoma Eca109 cells, Mol. Carcinog. 48 (2009) 886–894. [8] F. Yao, T. Zhao, C. Zhong, J. Zhu, H. Zhao, LDHA is necessary for the tumorigenicity of esophageal squamous cell carcinoma, Tumour Biol. 34 (2013) 25–31. [9] W. Wang, L. Xue, P. Wang, Prognostic value of β-catenin, c-myc, and cyclin D1 expressions in patients with esophageal squamous cell carcinoma, Med. Oncol. 28 (2011) 163–169. [10] S. Naganuma, K.A. Whelan, M. Natsuizaka, S. Kagawa, H. Kinugasa, S. Chang, et al., Notch receptor inhibition reveals the importance of cyclin D1 and Wnt signaling in invasive esophageal squamous cell carcinoma, Am. J. Cancer Res. 2 (2012) 459–475. [11] F. Emhemmed, S. Ali Azouaou, F. Thuaud, V. Schini-Kerth, L. Désaubry, C.D. Muller, et al., Selective anticancer effects of a synthetic flavagline on human Oct4-expressing cancer stem-like cells via a p38 MAPK-dependent caspase-3dependent pathway, Biochem. Pharmacol. 89 (2014) 185–196. [12] L.A. Boyer, T.I. Lee, M.F. Cole, S.E. Johnstone, S.S. Levine, J.P. Zucker, et al., Core transcriptional regulatory circuitry in human embryonic stem cells, Cell 122 (2005) 947–956.

[13] L.H. Looijenga, H. Stoop, H.P. de Leeuw, C.A. de Gouveia Brazao, A.J. Gillis, K.E. van Roozendaal, et al., POU5F1 (OCT3/4) identifies cells with pluripotent potential in human germ cell tumors, Cancer Res. 63 (2003) 2244–2250. [14] S. Amini, F. Fathi, J. Mobalegi, H. Sofimajidpour, T. Ghadimi, The expressions of stem cell markers: Oct4, Nanog, Sox2, nucleostemin, Bmi, Zfx, Tcl1, Tbx3, Dppa4, and Esrrb in bladder, colon, and prostate cancer, and certain cancer cell lines, Anat. Cell. Biol. 47 (2014) 1–11. [15] L. Cao, C. Li, S. Shen, Y. Yan, W. Ji, J. Wang, et al., OCT4 increases BIRC5 and CCND1 expression and promotes cancer progression in hepatocellular carcinoma, BMC Cancer 13 (2013) 82. [16] G. Bretones, M.D. Delgado, J. León, Myc and cell cycle control, Biochim. Biophys. Acta (2014) doi:10.1016/j.bbagrm.2014.03.013. [17] R.J. Duronio, Y. Xiong, Signaling pathways that control cell proliferation, Cold Spring Harb. Perspect. Biol. 5 (2013) a008904. [18] S. Lim, P. Kaldis, Cdks, cyclins and CKIs: roles beyond cell cycle regulation, Development 140 (2013) 3079–3093. [19] S. Tane, A. Ikenishi, H. Okayama, N. Iwamoto, K.I. Nakayama, T. Takeuchi, CDK inhibitors, p21(Cip1) and p27(Kip1), participate in cell cycle exit of mammalian cardiomyocytes, Biochem. Biophys. Res. Commun. 443 (2014) 1105–1109. [20] C.Y. Zhang, W. Bao, L.H. Wang, Downregulation of p16ink4a inhibits cell proliferation and induces G1 cell cycle arrest in cervical cancer cells, Int. J. Mol. Med. 33 (2014) 1577–1585. [21] S. Benzeno, F. Lu, M. Guo, O. Barbash, F. Zhang, J.G. Herman, et al., Identification of mutations that disrupt phosphorylation-dependent nuclear export of cyclin D1, Oncogene 25 (2006) 6291–6303. [22] S.H. Kim, G.C. Chau, Y.H. Jang, S.I. Lee, S. Pyo, S.H. Um, Clinicopathologic significance and function of mammalian target of rapamycin activation in esophageal squamous cell carcinoma, Hum. Pathol. 44 (2013) 226–236. [23] H. Shiozaki, Y. Doki, H. Yamana, K. Isono, A multi-institutional study of immunohistochemical investigation for the roles of cyclin D1 and E-cadherin in superficial squamous cell carcinoma of the esophagus, J. Surg. Oncol. 79 (2002) 166–173. [24] R. Mathew, S. Arora, R. Khanna, M. Mathur, N.K. Shukla, R. Ralhan, Alterations in p53 and pRb pathways and their prognostic significance in oesophageal cancer, Eur. J. Cancer 38 (2002) 832–841. [25] M.C. Casimiro, M. Velasco-Velázquez, C. Aguirre-Alvarado, R.G. Pestell, Overview of cyclins D1 function in cancer and the CDK inhibitor landscape: past and present, Expert Opin. Investig. Drugs 23 (2014) 295–304. [26] Y. Gu, S. Zhang, Q. Wu, S. Xu, Y. Cui, Z. Yang, et al., Differential expression of decorin, EGFR and cyclin D1 during mammary gland carcinogenesis in TA2 mice with spontaneous breast cancer, J. Exp. Clin. Cancer Res. 29 (2010) 6. [27] L.L. Tsai, F.W. Hu, S. Lee, C.H. Yu, C.C. Yu, Y.C. Chang, Oct4 mediates tumor initiating properties in oral squamous cell carcinomas through the regulation of epithelial-mesenchymal transition, PLoS ONE 9 (2014) e87207. [28] C. Li, Y. Yan, W. Ji, L. Bao, H. Qian, L. Chen, et al., OCT4 positively regulates Survivin expression to promote cancer cell proliferation and leads to poor prognosis in esophageal squamous cell carcinoma, PLoS ONE 7 (2012) e49693.