Targeting of miR-150 on Gli1 gene to inhibit proliferation and cell cycle of esophageal carcinoma EC9706

Abstract

OBJECTIVE:

Glioma-associated oncogene homolog 1 (Gli1) in Hedgehog signal pathway regulates Cyclin D1 expression, cell cycle or proliferation modulation. Esophageal cancer patients had significantly elevated Gli1 expression, which is related with survival and prognosis. It has been demonstrated that the level of miR-150 was decreased in esophageal cancer patients compared to normal control. As a complementary relationship exists between miR-150 and 3’-UTR of Gli1, this study investigated if miR-150 played a role in regulating Gli1 expression, and proliferation or cell cycle of esophageal cancer cells.

PATIENTS AND METHODS:

Esophageal squamous cell carcinoma (ESCC) patients from our hospital were recruited to collect tumor and adjacent tissues for miR-150 and Gli1 expression. Esophageal carcinoma cell line EC9706 and normal esophageal epithelial cell line HEEC were compared for expression of miR-150, Gli1 and Cyclin D1. Dual luciferase reporter gene assay examined the targeted relationship between miR-150 and 3’-UTR of Gli1. In vitro cultured EC9706 cells were treated with miR-150 mimic, si-Gli1 or the combination of miR-150 mimic and si-Gli1, respectively, to check their gene expression, cell cycle and proliferation.

RESULTS:

ESCC tissues had significantly higher Gli1 expression and lower miR-150 expression. EC9706 cell also had higher Gli1 expression than that in HEEC, whilst miR-150 was down-regulated. Via targeting 3’-UTR of Gli1 gene, miR-150 inhibited its expression. Transfection of miR-150 mimic, si-Gli1 or the combination of miR-150 mimic and si-Gli1, respectively, remarkably decreased expression of Gli1 and Cyclin D1 expression in EC9706 cells, whose cell cycle arresting at G0/G1 phase was enhanced with weakened proliferation.

CONCLUSIONS:

MiR-150 can induce G0/G1 cell cycle arresting and weaken proliferation of esophageal carcinoma cells via targeted inhibition on Gli1 and downstream expression of Cyclin D1.

Keywords

MicroRNA-150 Gli1 Cyclin D1 cell proliferation cell cycle esophageal carcinoma

1. Introduction

Esophageal cancer (EC) is a type of common malignant tumor that invades digestive system. Esophageal squamous cell carcinoma (ESCC), as a type of EC, was commonly found in China [1]. Glioma-associated oncogene homolog 1 (Gli1) belongs to Gli gene family, which serves as the transcriptional effector of Hedgehog signal pathway [2]. Gli1 contributes to the transduction from extracellular Hedgehog signal pathway into intracellular Gli1 signal, which exerts transcriptional modulation effects on the expression of nuclear target genes [3]. Enhancement of both function and expression of Gli1 can induce the over-activation of Hedgehog signal pathway, and is closely correlated with occurrence, progression and prognosis of various tumors including breast cancer [3], colorectal cancer [2], pancreatic carcinoma [4] and prostate cancer [5]. Previous studies showed significantly elevated expression of Gli1 in tumor tissues of esophageal cancer, and its correlation with patient’s survival and prognosis [6, 7, 8]. MicroRNA (miR) is a group of endogenous non-coding small RNAs, the sequences of which are highly conserved in eukaryotes, and can regulate the expression of target gene via complete or incomplete binding on 3’-untranslated region (3’-UTR) of mRNA, thus mediating cell proliferation, apoptosis and differentiation. Abnormal expression of miR is closely correlated with tumor occurrence, and may exert oncogene or tumor suppressor gene functions depending specific target genes [9, 10]. Previous study showed significantly decreased expression of miR-150 in EC tissues [11], suggesting its possible correlation with EC occurrence. Bioinformatics analysis showed targeted complementary relationship between miR-150 and 3’-UTR of Gli1. This study thus investigated if miR-150 played a role in regulating Gli1 expression and affecting cell proliferation or cell cycle of ES.

2. Materials and methods

2.1 Clinical information

A total of 39 ESCC patients who received treatment in the first affiliated hospital, and college of clinical medicine of Henan university of science and technology from January 2015 to May 2016 were recruited. All patients were diagnosed by pathology examination. There were 22 males and 17 females, aging between 39 and 77 years (average age $=$ 58.9 $\pm$ 13.5 years). All patients did not receive any chemotherapy or radiotherapy before surgery. This study has been pre-approved by the ethical committee of the first affiliated hospital, and college of clinical medicine of Henan University of Science and Technology. All subjects have signed the consent forms before recruitment in this study.

2.2 Histology of the patients

Pairs of fresh normal esophageal tissues and corresponding esophageal carcinoma tissues from patients with pathologically and clinically confirmed esophageal squamous cell carcinoma were collected by the first affiliated hospital, and college of clinical medicine of Henan University of Science and Technology. Primary tumor regions and the corresponding histological normal esophageal mucosa from the same patients were separated by experienced pathologists, and immediately stored at $-$ 70 ${}^{\circ}$ C until use. According to TNM stage criteria stipulated by International Cancer Prevention Union and American Cancer Committee, there were 12, 7, 9 and 11 cases at stage middle I, stage II, stage III and stage IV.

2.3 Cell culture

EC cell line EC9706 (ScienCell Research Laboratories, US) and normal esophageal epithelial cell line HEEC (Jiniou Bio, China) were seeded into RPMI 1640 medium containing 10% FBS (Gibco, US), 100 U/ml penicillin and 100 $\mu$ g/ml streptomycin (Lonza, US). Cells were cultured under 37 ${}^{\circ}$ C with 5% CO ${}_{2}$ .

2.4 Construction of luciferase reporter gene vector and dual luciferase reporter gene assay

Using HEK293T genomic DNA as the template, full length fragment of wild type or mutant forms of 3’-UTR of Gli1 gene was amplified and were cloned into pLUC plasmid (Ambion, US). Recombinant plasmid was then used to transform DH5 $\alpha$ competent cells. Positive clones with correct sequences were screened out by sequencing and were named as pLUC-Gli1-3’-UTR-wt and pLUC-Gli1-3’-UTR-mut. Lipofecctamine 2000 was used to co-transfect pLUC-Gli1-3’-UTR-wt (or pLUC-Gli1-3’-UTR-mut) and miR-150 mimic into HEK293T cells. After 48 h incubation, dual luciferase activity was measured (Promega, US).

2.5 Bioinformatics analysis

Bioinformatics analysis was conducted according to the on-line data bases including MiRanda, TargetScan and PITA. Data were downloaded from the MiRanda database MirBase and $-$ 1.2 was chosen as cut-off mirSVR score to obtain a reasonable empirical probability of target mRNA downregulation. The genes with a probability of conserved targeting (Pct) $>$ 0.1 were enrolled from TargetScan. 10 kcal/mol was set as a general cutoff value among data from PITA.

2.6 EC9706 cell transfection

In vitro cultured EC9706 cells were divided into five groups: miR-NC transfection group; miR-150 mimic transfection group; si-NC transfection group; si-Gli1 group; and miR-150 mimic $+$ si-Gli1 group. 72 h after transfection, cells were collected for assay. Nucleotide fragments used were: si-Gli1 sense: 5’-CCAGG AAUUU GACUC CCAAT T-3’; si-Gli1 anti-sense: 5’-UUGGG AGUCA AAUUC CUGGC T-3’; si-NC sense: 5’-UUCUC CGAAC GUGUC ACGUT T-3’; si-NC anti-sense, 5’-ACGUG ACACG UUCGG AGAAT T-3’ (Ruibo Bio, China).

2.7 qRT-PCR for gene expression assay

Cultured cells were lysed by Trizol method (Invitrogen, US). RNA was obtained by chloroform separation, isopropanol precipitation, washing in 75% ethanol, and was re-suspended in RNAse-free water. PimerScript RT reagent Kit (Takara, Japan) was used to synthesize cDNA from RNA by reverse transcription. cDNA was kept at $-$ 20 ${}^{\circ}$ C for further PCR amplification by using TaqDNA polymerase. Primer sequences were: miR-150 P ${}_{F}$ : 5’-ATAAA GTGCT GACAG TGCAG ATAGT G-3’; miR-150 P ${}_{R}$ : 5’-TCAAG TACCC ACAGT GCGGT-3’; U6 P ${}_{F}$ : 5’-ATTGG AACGA TACAG AGAAG ATT-3’; U6 P ${}_{R}$ : 5’-GGAAC GCTTC ACGAA TTTG-3’; Gli1 P ${}_{F}$ : 5’-AGCGT GAGCC TGAAT CTGTG-3’; Gli1 P ${}_{R}$ : 5’-CAGCA TGTAC TGGGC TTTGA A-3’. Cyclin D1 P ${}_{F}$ : 5’-GCTGC GAAGT GGAAA CCATC-3’; Cyclin D1 P ${}_{R}$ : 5’-CCTCC TTCTG CACAC ATTTG AA-3’; $\beta$ -actin P ${}_{F}$ : 5’-GAACC CTAAG GCCAA C-3’; $\beta$ -actin P ${}_{R}$ : 5’-TGTCA CGCAC GATTT CC-3’. In a 10 $\mu$ L PCR system, one added 5.0 $\mu$ L 2XSYBR Green Mixture (Life Technologies, US), 0.5 $\mu$ L forward and reverse primers (2.5 $\mu$ m/L), and 1 $\mu$ L ddH ${}_{2}$ O, PCR conditions were: 95 ${}^{\circ}$ C for 15 s, followed by 60 ${}^{\circ}$ C 30 s and 74 ${}^{\circ}$ C 30 s. 40 cycles were performed on ABI 7500 fluorescent PCR cycler.

2.8 Western blot

Cells were collected for proteins extraction using routine method. Proteins were added into three times volumes of 4X loading buffer for 5 min boiling. 40 $\mu$ g supernatant was separated in 8% SDS-PAGE for 2.5 $\sim$ 3 h, and was transferred to PVDF membrane in 60 min. The membrane was blocked in 5% defatted milk powder for 60 min at room temperature. Primary antibody (Gli1 at 1:100, Cyclin D1 at 1:200, $\beta$ -actin at 1:500) (Cell Signaling Technology, US) was added for 4 ${}^{\circ}$ C overnight incubation. Unbounded primary antibody was washed away by PBST. HRP conjugated secondary antibody (1:10 000 dilutions) (Jackson ImmunoResearch, US) was added for 60 min room temperature incubation. Unbounded primary antibody was washed away by PBST, followed by ECL development for 1 $\sim$ 3 min at room temperature. The film was exposed and scanned for data collection.

2.9 Flow cytometry for cell cycle

Cells were digested in trypsin and were centrifuged at 1 000 g for 5 min. Cells were re-suspened in pre-cold PBS. After re-centrifugation, the supernatant was discarded. Cells were fixed in 1 ml pre-cold 70% ethanol for 24 h at 4 ${}^{\circ}$ C. Excess ethanol was removed by twice centrifugation in PBS. Staining buffer containing propidium iodine (PI) and RNAse A was added for 30 min dark culture at 37 ${}^{\circ}$ C (Beyotime, China). The mixture was loaded for measuring fluorescence at 488 nm wavelength for DNA content analysis in cells.

Figure 1.

Lower miR-150 expression and higher Gli1 expression. (A) qRT-PCR for Gli1 mRNA expression; (B) qRT-PCR for miR-150 expression; (C) Western Blot for Gli1 protein expression. *, $p<$ 0.05 compared to adjacent tissues.

2.10 CCK-8 assay for cell proliferation activity

All transfected cells were seeded into 96-well plate at 1 $\times$ 10 ${}^{4}$ per well density. 24 h after cell attachment, 48 h incubation was performed. 10 $\mu$ L CCK-8 solution was added 4 h before assay (Beyotime, China). Absorbance value at 450 nm wavelength (A450) was then measured using a microplate reader. Cell proliferation activity $=$ (A450 of treated wells $-$ A450 of blank wells)/(A450 of control wells $=$ A450 of blank wells) $\times$ 100%.

2.11 CFSE staining for cell proliferation

1 mL CFDA SE marker buffer was used to re-suspend all transfected cells (Beyotime, China). After mixing with 1 mL 2XCFDA SE stock buffer, cells were incubated at 37 ${}^{\circ}$ C for 10 min, and were washed in culture medium containing 10% FBS twice. 5 $\sim$ 10 mL culture medium containing 10% FBS was then added for 5 min incubation at 37 ${}^{\circ}$ C to facilitate entry of CFDA SE into cells and removal of unbounded CFDA SE. The supernatant was removed by centrifugation in the last time of washing. Cells after staining were inoculated into 6-well plate. After 48 h culture, fluorescent intensity was measured by flow cytometry to reflect cell proliferation potency.

2.12 Statistical analysis

SPSS18.0 software was used for data analysis. Measurement data were presented as mean $\pm$ standard deviation (SD). SNK was used to compare measurement data between groups. A statistical significance was defined when $p<$ 0.05.

Figure 2.

Lower miR-150 and higher Gli1 expression in EC cells. (A) CFSE staining for cell proliferation; (B) PI staining for cell cycle assay; (C) qRT-PCR for gene expression; (D) Western Blot for protein expression. ( $n=$ 6). *, $p<$ 0.05 compared to HEEC cells. The experiments were repeated three times.

Figure 3.

Gli1 as the target gene of miR-150. (A) Function sites between miR-150 and 3’-UTR of Gli1/Gli1 mut mRNA; (B) Dual luciferase reporter gene assay. *, $p<$ 0.05 comparing between miR-150 mimic and miR-NC. ( $n=$ 6). The experiments were repeated three times.

3. Results

3.1 Reduced miR-150 and elevated Gli1 expression in ECSS tissues

Test results showed that, compared to adjacent tissues, ESCC tumor tissues had significantly elevated mRNA (Fig. 1A) and protein of Gli1 (Fig. 1C) expressions, both of which were gradually elevated with advanced TNM stage, indicating possible correlation between Gli1 up-regulation and ESCC occurrence. Compared to adjacent tissues, ESCC tissues had significantly lower miR-150 expression, which was further decreased with advanced TNM stage (Fig. 1B).

3.2 Lower miR-150 and higher Gli1 expression in ECSS tumor cell line

CFSE staining and flow cytometry results showed significantly lower fluorescent intensity in EC9706 cells compared to HEEC cells (Fig. 2A), indicating significantly higher proliferation potency of EC cells than normal epithelial cells. DNA quantification showed lower S phase ratio in HEEC cells compared to that in EC9706 cells, whilst G0/G1 ratio was significantly higher (Fig. 2B), suggesting higher DNA synthesis and proliferation potency of EC cells than those in HEEC cells, accompanied with less G0/G1 arresting. Further assay showed remarkably lower miR-150 expression in EC9706 cells compared to HEEC cells (Fig. 2C), whilst Gli1 mRNA or protein level was higher (Fig. 2D). In addition, the level of Cyclin D1, which is important for G1/S transition, was also significantly higher in EC9706 cells (Fig. 2D).

3.3 Gli1 is one target gene of miR-150

Bioinformatics analysis showed satisfactory targeted relationship between miR-150 and 3’-UTR of Gli1 mRNA (Fig. 3A). Dual luciferase reporter gene assay showed that the transfection of miR-155 mimic could significantly decreased relative luciferase activity in HEK293T cells transfected by Gli1-3’-UTR-wt plasmid (Fig. 3B), but not on HEK293T cells transfected with mutant form (Gli1-3’-UTR-mut) of plasmids, indicating that miR-150 targeted 3’-UTR of Gli1 mRNA to inhibit its expression.

Figure 4.

MiR-150 up-regulation and G0/G1 phase arresting and proliferation inhibition. (A) qRT-PCR for gene, expression; (B) Western Blot for protein expression; (C) PI staining for cell cycle assay; (D) CCK-8 for cell proliferation activity. Note: a, $p<$ 0.05 comparing between miR-150 mimic and miR-NC group; b, $p<$ 0.05 comparing between si-Gli1 and si-NC; c, $p<$ 0.05 comparing between miR-150 mimic $+$ si-Gli1 and miR-150 mimic; d, $p<$ 0.05 comparing between miR-150 mimic $+$ si-Gli1 and si-Gli1 ( $n=$ 6). The experiments were repeated three times.

3.4 MiR-150 up-regulation induced G0/G1 arresting of EC9706 cells and decreased proliferation potency

Transfection of miR-150 mimic and/or interference of Gli1 significantly down-regulated mRNA and protein expression of Gli1 (Fig. 4A and B). Meanwhile, cyclin D1 expression was also remarkably inhibited (Fig. 4A and B). Flow cytometry results showed that the transfection of miR-150 mimic and/or interference of Gli1 significantly increased G0/G1 phase arresting (Fig. 4C) and decreased cell proliferation potency (Fig. 4D).

4. Discussion

EC represents as a type of malignant tumor with relatively higher incidence and mortality. Its rate of occurrence is 8 ${}^{\rm th}$ highest among all cancers worldwide, and is the 6 ${}^{\rm th}$ deadly in all cancers, or 2 ${}^{\rm nd}$ deadly cancer in digestive tract [12]. The occurrence and development of EC is related to the aberrant activation of signaling pathways, among which the highly conserved pathway, Hedgehog signal pathway is closely related with embryonic development, organ genesis and tissue differentiation, in addition to critical roles during cell proliferation, cell cycle, apoptosis and differentiation regulation [13]. Hedgehog pathway in mammalians includes three extracellular signal proteins, Sonic hedgehog (Shh), Indian hedgehog (Ihh) and Desert hedgehog (Dhh) [14]. Once the Hedgehog signal pathway is activated, the transcriptional effector Gli1 at the terminus of signal pathway is induced, which causes the cascade reaction of oncogene-like genes, leading to malignant transformation and tumor pathogenesis [15]. Gli1 belongs to human Gli homologous genes (Gli1, Gli2 and Gli3), and can regulate the expression of multiple genes including Cyclin D [16], c-myc and Bcl-2 [17]. Previous studies showed that the expression of Gli1 was significantly elevated in EC tumor tissues [15, 18, 19, 20, 21], and miR-150 expression was reduced in ESCC patient tumor tissues compared to adjacent tissues, in addition to the significant correlation between miR-150 expression and tumor size, infiltration depth, lymph node metastasis, vascular metastasis, clinical stage and prognosis [11], suggesting the possible involvement in EC pathogenesis. Consistently, our test results showed remarkably higher Gli1 expression in ESCC tumor tissues, with lower miR-150 expression compared to that in adjacent tissues.

As the regulator for cell cycle protein dependent kinase (CDKs), cyclin D1 can bind and activate G1-specific CDK4 to phosphorylate G1 phase cycle inhibitory protein Rb, which further dissociates from E2F transcriptional factor for initiating cell cycle regulatory gene, thus pushing cell cycle from G1 into S phase, and facilitating cell proliferation [22]. Previous finding showed that Gli1 played an important role in mediating cyclin D1 expression, cell proliferation and cycle progression [16]. Our data confirmed that, compared to HEEC cells, EC9706 cells presented significantly higher proliferation potency, with the increase of cyclin D1 expression and low G0/G1 ratio. Of note, miR-150 increased the level of cyclin D via the targeting inhibition of Gli1 expression, which validated the previous evidence of positive correlation between Gli1 expression and cyclin D1 level in EC cells [6].

Moreover, it has been demonstrated that the over-expression of Gli1 significantly facilitated epithelial mesenchymal transition (EMT) process of EC cells, and potentiated malignancy including invasion and metastasis [18]. In Gli1-mediated cell cycle regulation, Rizvi et al. found that Gli1 could up-regulate CDK-2 expression via functioning on CDK-2, to facilitate G1/S transition, thus promoting proliferation [19]. Our study further showed that the decrease of Gli1 remarkably caused higher G0/G1 ratio and weakening proliferation potency. Additionally, Yokobori et al. found that miR-150 could weaken EMT process and invasion potency of EC cells via targeting the expression of important EMT regulator, ZEB1 [11]. Wang et al. found that miR-150 could enhance the interaction between rpL11 and c-Myc via targeted inhibition on 5S rRNA expression, thus suppressing the transcriptional regulatory role of oncogene c-Myc and inhibiting EC cell proliferation [23]. This study revealed the role of miR-150 in EC pathogenesis, as consistent with Yokobori et al. [11] and Wang et al. [23]. These results indicated that the down-regulation of miR-150 and consequently upregulated Gli1 and its downstream target gene cyclin D1 might be one important mechanism underlying EC pathogenesis. However, the limitation in our study still exists, and the regulatory effect of miR-150 on other G1/S mediators, such as CDK-2/4/6, cyclin E, along with the validation of other CDK inhibitors via regulating Gli1, requires to be further determined.

5. Conclusion

Our study demonstrated that miR-150 played a critical role in promoting cyclin D1 expression and inhibiting the proliferation of EC cells via the targeted inhibition of Gli1, which provides academic basis for the further therapy of esophageal cancer.

Footnotes

Conflict of interest

None.

References

Chen

Zheng

Baade

P.D.

Zhang

Zeng

Bray

Jemal

X.Q.

and He

, Cancer statistics in china 2015, CA Cancer J Clin 66 (2016), 115–132.

Hong

K.D.

Lee

Kim

B.H.

Lee

S.I.

and Moon

H.Y.

, Expression of GLI1 correlates with expression of lymphangiogenesis proteins vascular endothelial growth factor C and vascular endothelial growth factor receptor 3 in colorectal cancer, Am Surg 79 (2013), 198–204.

Takebe

Hunsberger

and Yang

S.X.

, Expression of gli1 in the hedgehog signaling pathway and breast cancer recurrence, Chin J Cancer Res 24 (2012), 257–258.

Sheng

Dong

Zhou

Liu

Dong

and Li

, The clinicopathological significance and relationship of gli1 MDM2 and p53 expression in resectable pancreatic cancer, Histopathology 64 (2014), 523–535.

Chen

Goto

Sakamoto

Tanaka

Matsubara

Nakamura

Zheng

Takayanagi

, and Nomura

, GLI1, a crucial mediator of sonic hedgehog signaling in prostate cancer functions as a negative modulator for androgen receptor, Biochem Biophys Res Commun 404 (2011), 809–815.

Yang

Cui

Kim

and Xuan

, Gli1, a potential regulator of esophageal cancer stem cell is identified as an independent adverse prognostic factor in esophageal squamous cell carcinoma, J Cancer Res Clin Oncol (2016).

Katoh

and Katoh

, Integrative genomic analyses on GLI1: Positive regulation of GLI1 by hedgehog-GLI TGFbeta-Smads and RTK-PI3K-AKT signals and negative regulation of GLI1 by notch-CSL-HES/HEY and GPCR-Gs-PKA signals, Int J Oncol 35 (2009), 187–192.

Mori

Okumura

Tsunoda

Sakai

and Shimada

, Gli-1 expression is associated with lymph node metastasis and tumor progression in esophageal squamous cell carcinoma, Oncology 70 (2006), 378–389.

Zhang

Wei

and Xu

, MiR-150 promotes the proliferation of lung cancer cells by targeting p53, FEBS Lett 587 (2013), 2346–2351.

10.

Wang

W.H.

Chen

Zhao

Zhang

B.R.

H.S.

Jin

H.Y.

and Dai

J.H.

, MiR-150-5p suppresses colorectal cancer cell migration and invasion through targeting MUC4, Asian Pac J Cancer Prev 15 (2014), 6269–6273.

11.

Yokobori

Suzuki

Tanaka

Inose

Sohda

Sano

Sakai

Nakajima

Miyazaki

Kato

and Kuwano

, MiR-150 is associated with poor prognosis in esophageal squamous cell carcinoma via targeting the EMT inducer ZEB1, Cancer Sci 104 (2013), 48–54.

12.

Siegel

R.L.

Miller

K.D.

and Jemal

, Cancer statistics 2016, CA Cancer J Clin 66 (2016), 7–30.

13.

Isohata

Aoyagi

Mabuchi

Daiko

Fukaya

Ohta

Ogawa

Yoshida

and Sasaki

, Hedgehog and epithelial-mesenchymal transition signaling in normal and malignant epithelial cells of the esophagus, Int J Cancer 125 (2009), 1212–1221.

14.

Katoh

and Katoh

, Hedgehog target genes: Mechanisms of carcinogenesis induced by aberrant hedgehog signaling activation, Curr Mol Med 9 (2009), 873–886.

15.

Sheng

Zhang

Huang

Chen

Sultz

Adegboyega

P.A.

Zhang

and Xie

, Hedgehog signaling is activated in subsets of esophageal cancers, Int J Cancer 118 (2006), 139–148.

16.

Furukawa

Miyake

and Fujisawa

, GLI2 expression levels in radical nephrectomy specimens as a predictor of disease progression in patients with metastatic clear cell renal cell carcinoma following treatment with sunitinib, Mol Clin Oncol 5 (2016), 186–192.

17.

Y.Z.

Yan

Y.Y.

Xiao

Q.H.

Yao

W.F.

Zhao

H.Z.

Z.J.

Zhou

M.Y.

M.T.

Zhang

S.S.

Chen

J.J.

and Wei

M.J.

, Salinomycin induces selective cytotoxicity to MCF-7 mammosphere cells through targeting the hedgehog signaling pathway, Oncol Rep 35 (2016), 912–922.

18.

Min

Xiaoyan

Fanghui

Yamei

Xiaoli

and Feng

, The glioma-associated oncogene homolog 1 promotes epithelial-mesenchymal transition in human esophageal squamous cell cancer by inhibiting e-cadherin via snail, Cancer Gene Ther 20 (2013), 379–385.

19.

Rizvi

Demars

C.J.

Comba

Gainullin

V.G.

Rizvi

Almada

L.L.

Wang

Lomberk

Fernandez-Zapico

M.E.

and Buttar

N.S.

, Combinatorial chemoprevention reveals a novel smoothened-independent role of GLI1 in esophageal carcinogenesis, Cancer Res 70 (2010), 6787–6796.

20.

Zhu

You

Tao

and Chen

, Correlation of hedgehog signal activation with chemoradiotherapy sensitivity and survival in esophageal squamous cell carcinomas, Jpn J Clin Oncol 41 (2011), 386–393.

21.

Yoshikawa

Nakano

Tao

Koishi

Matsumoto

Sasako

Tsujimura

Hashimoto-Tamaoki

and Fujiwara

, Hedgehog signal activation in oesophageal cancer patients undergoing neoadjuvant chemoradiotherapy, Br J Cancer 98 (2008), 1670–1674.

22.

Sur

Pal

Roy

Barua

Roy

Saha

and Panda

C.K.

, Tea polyphenols EGCG and TF restrict tongue and liver carcinogenesis simultaneously induced by n-nitrosodie- thylamine in mice, Toxicol Appl Pharmacol 300 (2016), 34–46.

23.

Wang

Ren

Wang

Xiong

Han

Pan

Chen

Zhou

Yuan

and Yang

, Down-regulation of 5S rRNA by miR-150 and miR-383 enhances c-Myc-rpL11 interaction and inhibits proliferation of esophageal squamous carcinoma cells, FEBS Lett 589 (2015), 3989–3997.