Hook1 inhibits malignancy and epithelial–mesenchymal transition in hepatocellular carcinoma

Abstract

Hook1 is a member of the hook family of coiled-coil proteins, which is recently found to be associated with malignant tumors. However, its biological function in hepatocellular carcinoma is yet unknown. Here, we evaluated the Hook1 levels in human hepatocellular carcinoma samples and matched peritumoral tissues by real-time polymerase chain reaction. Small interfering RNA knockdown and a transforming growth factor-β-induced epithelial–mesenchymal transition model were employed to investigate the biological effects of Hook1 in hepatocellular carcinoma. Our results indicated that Hook1 levels were significantly lower in hepatocellular carcinoma tissues than in the peritumoral tissues. In addition, Hook1 expression was significantly associated with hepatocellular carcinoma malignancy. Hook1 was downregulated after transforming growth factor-β-induced epithelial–mesenchymal transition. Moreover, Hook1 knockdown promoted epithelial–mesenchymal transition and attenuated the sensitivity of hepatocellular carcinoma cells to doxorubicin. In summary, our results indicate that downregulation of Hook1 plays a pivotal role in hepatocellular carcinoma progression via epithelial–mesenchymal transition. Hook1 may be used as a novel marker and therapeutic molecular target in hepatocellular carcinoma.

Keywords

Hook1 hepatocellular carcinoma epithelial–mesenchymal transition drug resistance transforming growth factor-β

Introduction

Hepatocellular carcinoma (HCC) is the fifth most commonly diagnosed cancer and the second most frequent cause of cancer-related deaths worldwide.¹ Currently, various strategies including liver resection,² liver transplantation,³ radiofrequency ablation,⁴ transcatheter arterial chemoembolization,⁵ molecular-targeted therapy,⁶ and systemic chemotherapy⁷ are used to treat HCC; however, the prognosis is still unsatisfied.

Epithelial–mesenchymal transition (EMT) plays essential role in progression and metastasis of various types of cancer.⁸ It is believed that EMT is a critical step that facilitates the invasiveness of the cells to achieve intravasation, extravasation, and consequent remote metastases formation.⁹ In addition, EMT contributes to the acquired drug resistance in chemotherapy.¹⁰ Thus, the acquired invasiveness during EMT promotes recurrence and metastasis of various malignant tumors including HCC, resulting in a poor prognosis.^11–13

The Hook gene was first reported by Mohr¹⁴ nearly a century ago. In recent decades, Hook1, as well as its homologs Hook2 and Hook3, have been found to play important roles in endocytic trafficking.¹⁵ Additionally, it is thought to be involved in the dynamic balance of microtubule cytoskeleton.¹⁶ Since cytoskeleton is important in cancer cell biology such as proliferation and migration,¹⁷ hook family members may regulate cancer development. Recently, Li et al.¹⁸ first reported that Hook1 negatively modulated EMT in non–small cell lung cancer. However, the role of Hook1 in human HCC remains unknown. In this study, we explored the role of Hook1 in HCC using clinical specimens and HCC cell lines.

Materials and methods

Patients and surgical specimens

A total of 70 frozen HCC tissues and corresponding peritumoral tissues were obtained from patients underwent liver resection at the Second Affiliated Hospital of Zhejiang University School of Medicine between 2010 and 2013. The specimens were immediately frozen in liquid nitrogen following surgical resection and stored at −80°C before total RNA extraction. A normal liver tissue specimen was obtained from a patient without viral hepatitis, cirrhosis, or any other lesions, who underwent liver resection for giant hepatic hemangioma. Clinical data were collected in a prospective database. This study was conducted with the approval of the Institutional Review Board and Ethics Committee of our hospital and according to the Declaration of Helsinki. Informed consent was obtained from all the patients included.

Cell lines and cell culture

HCC cell lines were obtained from the Shanghai Institute for Biological Sciences (Shanghai, China). Huh-7, HepG2, and MMC-LM3 cells were cultured in high-glucose Dulbecco’s modified Eagle’s medium (DMEM; HyClone, Logan, UT). Hep3B cells were cultured in Minimum Essential Medium (MEM; HyClone). SNU-387, SNU-449, QGY-7703, Bel-7402, and HL7702 cells were cultured in RPMI-1640 medium (HyClone). All media were supplemented with 10% fetal bovine serum (FBS; HyClone) and 1% penicillin/streptomycin (Sigma, St. Louis, MO). The cells were maintained at 37°C in a humidified incubator with 5% CO₂ in air and used within 3 months after resuscitation.

Induction of EMT by transforming growth factor-β

HCC cell lines were seeded in six-well plates at a density of 2 × 10⁵ cells/well. After 24 h, cells were treated for another 24 h with FBS-free medium containing 5 ng/mL transforming growth factor (TGF)-β (Cell Signaling Technology, Danvers, MA). Induction of EMT in HCC cells was confirmed via real-time polymerase chain reaction (PCR), western blot, and immunofluorescence by assessing the expression of E-cadherin and Vimentin. Huh-7 and HepG2 cells after TGF-β treatment were used for cell viability and wound healing assays.

Cell viability assay

To evaluate the cell viability, HCC cells were seeded into 96-well plates at a density of 5000 cells per well and were incubated overnight. The culture medium was then replaced with complete medium containing indicated concentrations of doxorubicin (0.125, 0.25, 0.5, and 1.0 µg/mL) and the cells were cultured for another 48 h. Cell viability was tested using a Cell Counting Kit-8 (CCK-8; Dojindo Molecular Technologies, Gaithersburg, MD), according to the manufacturer’s instructions and was presented as a ratio to the untreated cells. The half-maximal inhibitory concentration (IC₅₀) was determined by fitting data to the equation as previously described.¹⁹ V% = 100%/[1 + ([doxorubicin]/IC₅₀)^p], where V% is the percentage viability and [doxorubicin] is the concentration (µg/mL) of doxorubicin.

RNA interference and quantitative real-time PCR

Human Hook1 small interfering RNA (siRNA) and negative control siRNA were purchased from Santa Cruz Biotechnology (Dallas, TX, USA). For RNA interference, cells were transfected with siRNA using Lipofectamine 3000 Transfection Reagent (Thermo Fisher Scientific, Waltham, MA, USA) following the manufacturer’s instructions. The transfection medium was replaced with complete medium after 6 h, and the efficiency of siRNA knockdown was confirmed by immunoblotting 48 h after the addition of complete medium.

Total RNA was extracted using TRIzol LS Reagent (Ambion) and was reverse transcribed into complementary DNA (cDNA) using the Prime Script Reagent RT Kit (Takara Biotechnology, Kusatsu, Japan). Totally, 200 ng cDNA was employed for quantitative real-time polymerase chain reaction (qPCR) with 10 µL reaction per well. qPCR was performed on an ABI Prism 7500 Fast Real-Time System (Applied Biosystems, Foster City, CA, USA). All qPCR assays were performed in triplicate, and the data were obtained from at least three independent experiments. The messenger RNA (mRNA) level of target genes was normolized to the mRNA level of GAPDH (glyceraldehyde 3-phosphate dehydrogenase) using the comparative 2^−ΔΔCt method.²⁰ Primers were designed and purchased from Takara. Primers used include GAPDH (forward: ATCATCAGCAATGCCTCC, reverse: TCCTTCCACGATACCAAAG), Vimentin (forward: TGAGTACCGGAGACAGGTGCAG, reverse: TAGCAGCTTCAACGGCAAAGTTC), E-cadherin (forward: TACACTGCCCAGGAGCCAGA, reverse: TGGCACCAGTGTCCGGATTA), and Hook1 (forward: TGCTGCTGAGATTATGCCAGTGGA, reverse: TCAGCCTCTGCTCAGTTTCCAGTT).

Immunoblotting

Immunoblotting was performed as previously reported.²¹ Briefly, 20 mg of protein lysates was fractionated on 8%–10% Tris–glycine polyacrylamide gels, transferred to polyvinylidene difluoride membranes, blocked, and incubated overnight at 4°C with primary antibodies against Hook1 (Santa Cruz), E-cadherin (Cell Signaling Technology), Vimentin (Cell Signaling Technology), or GAPDH (Cell Signaling Technology) as indicated. All antibodies were diluted at 1:1000 except for anti-GAPDH antibodies, which were diluted at 1:2000. The membrane was then treated with horseradish peroxidase–conjugated secondary antibodies (1:2000 dilution; Beyotime Biotechnology, Shanghai, China) and developed using Western Chemiluminescent HRP Substrate (ECL; EMD Millipore Immobilon). All the primary antibodies were purchased from Cell Signaling Technology.

Wound healing assay

Huh-7 and HepG2 cells were cultured in six-well plates until 100% confluence was achieved. The monolayer of cells was wounded by scratching with a sterile pipette tip, and the cells were incubated for another 48 h. Phase-contrast images were captured at 0, 24, and 48 h, and the width of wound was tested to reflect the wound healing ability of cells.

Immunofluorescence

For immunofluorescence analysis, Huh-7 and HepG2 cells were seeded in 24-well plates at a density of 100,000 cells per well after transfected with siRNA as described above. The cells were fixed with 4% paraformaldehyde for 15 min, washed with phosphate buffered saline (PBS) thrice, blocked with 5% bovine serum albumin (Sigma) for 30 min at room temperature, and incubated with anti-Vimentin or anti-E-cadherin antibodies (1:100 dilution; Cell Signaling Technology) at 4°C overnight. The cells were then incubated with fluorescein isothiocyanate–labeled secondary antibodies (1:200 dilution; Santa) for 2 h and washed with PBS thrice. The cells were then incubated with 4′,6-diamidino-2-phenylindole (DAPI; 1:10,000 dilution; Sigma) for 10 min at room temperature, washed twice with PBS, and observed using an inverted fluorescence IX81 microscope (Olympus, Tokyo, Japan). For F-actin staining, phalloidin-iFluor (Abcam) and DAPI were used as a similar protocol mentioned above, and cells were observed using a confocal laser scanning microscopy platform TCS SP8 (Leica, Wetzlar, Germany).

Statistical analysis

Quantitative data were presented as the mean ± standard deviation (SD) or standard error of the mean (SEM). Comparisons were performed using unpaired Student’s t tests. Data were analyzed using Prism 6 (GraphPad). A p value of <0.05 was considered statistically significant.

Results

Hook1 expression is inhibited in HCC

We first detected the expression level of Hook1 mRNA in HCC tumor tissues and matched peritumoral tissues in 70 HCC patients. In 13 patients, the Ct value of Hook1 in either HCC or non-cancerous tissues exceeded 30 when the Ct value of GAPDH ranged from 15 to 20; data of these patients were excluded from the analysis. In the left 57 paired samples (Table 1), Hook1 expression significantly decreased in HCC tumor tissues compared to that in peritumoral tissues (p < 0.001; Figure 1(a)). In addition, in the samples with metastatic inclination (with hepatic, portal, or microvascular tumor thrombosis), the relative Hook1 expression (tumor/peritumoral tissue) was even lower compared to that in the samples without thrombosis (p < 0.05, Figure 1(b)).

Table 1.

Demographic and clinical characteristics of the patients.

Variables	TT (15)	Non-TT (42)	p value
Age^a (years)	53.7 (12.9)	57.5 (12.5)	0.328
Gender (F/M)	0/15	5/37	0.386
HBsAg (negative/positive)	2/13	6/36	1.000
Liver cirrhosis (no/yes)	6/9	19/23	0.962
Preoperative AFP^a (ng/mL)	21346.3 (33788.0)	2761.4 (9406.6)	0.001^b
Tumor number (1/>1)	13/2	38/4	1.000
Tumor size^a (cm)	8.7 (5.3)	6.3 (3.9)	0.135^b
Vessel invasion	7	–	–
Microvessel invasion	1	–	–
PV/HV/IVC invasion	7	–	–
TNM stage (III–IV)	15/15	24/42	0.006
Differentiation (well/moderate or poor)	5/10	21/21	0.418
Hook1 ratio^a	5.41 (3.65)	2.76 (2.07)	0.001

TT: tumor thrombosis, including tumor emboli in portal vein (PV), hepatic vein (HV), inferior vein cava (IVC), microvessel, and other vessels; SD: standard deviation; AFP: alpha-fetoprotein.

Data are presented as mean (SD).

Statistical analysis was calculated using log values of the original data.

Figure 1.

The Hook1 mRNA level in HCC specimens and cell lines. (a) Hook1 was significantly downregulated in HCC tumor tissues than in the corresponding peritumoral tissues. (b) The ratio of Hook1 mRNA expression in HCC tumor tissue to the matched peritumoral tissue was measured. The TT group (n = 15) exhibited a less expression of Hook1 than the non-TT group (n = 42). Bar represents SEM. (c) Hook1 expression in the liver and HCC cell lines indicated that the more mesenchymal-like cells exhibited lower Hook1 expression (from green to red).

Based on the observation of these surgical specimens, we next investigated whether Hook1 was involved in the modulation of the biological behavior of HCC cells. To this end, we detected the basal levels of Hook1 in different HCC cell lines, including epithelial-type (QGY-7703, HepG2, Hep3B, and Huh-7) and mesenchymal-type (HCC-387, SMMC-7721, Bel-7402, and HCC-LM3) cell lines. In most cases, mesenchymal-type cells have higher capacity of migration and invasion, and epithelial-type cells can acquire such ability by undergoing EMT.²² Consistent with the observations in clinical specimens, Hook1 expression appeared lower in mesenchymal-type cell lines (Figure 1(c)). Since mesenchymal-type HCC is usually more invasive, these results suggested that lower expression of Hook1 was associated with increased invasiveness in HCC cells.

Hook1 expression is downregulated in a TGF-β-induced EMT model of HCC cells

To confirm the association between Hook1 and biological features of HCC cells, two epithelial-type HCC cell lines, Huh-7 and HepG2, were chosen. Both cell lines are most frequently used in HCC study and are typical models for TGF-β-induced EMT.²³ The morphological features of Huh-7 showed remarkable changes following treatment with 5 ng/mL TGF-β for 24 h, changing from slab-like cells to spindle-shaped cells (Figure 2(a)). However, there were no significant morphological changes in HepG2 cells. Consistent with the previous study,²⁴ F-actin reorganization was observed in both Huh-7 and HepG2 cells with TGF-β treatment (Figure 2(b)). In addition, the EMT markers E-cadherin and Vimentin were detected using both qPCR and immunoblotting. As expected, E-cadherin was downregulated and Vimentin was upregulated obviously following TGF-β treatment (Figure 2(c)–(e)), consistent with the findings of a previous study.²³ We also noted that TGF-β significantly decreased the expression of Hook1 (Figure 2(f)), suggesting that Hook1 was involved in TGF-β-induced EMT in HCC cells.

Figure 2.

Hook1 expression was reduced in TGF-β induced EMT. After stimulation with 5 ng/mL TGF-β, EMT occurred in the epithelial-type HCC cell lines. Phalloidin staining showed reorganization of F-actin in both (a, b) Huh-7 and HepG2 cells. (c) Downregulation of E-cadherin and (d) upregulation of Vimentin were observed in real-time qPCR, as well as in (e) immunoblotting analysis. (f) Hook1 expression decreased after TGF-β treatment.

Hook1 negatively regulates TGF-β-induced EMT in HCC cells

To investigate whether Hook1 downregulation is an etiology or a consequence of EMT induction, we used siRNA to knockdown Hook1 expression in Huh-7 and HepG2 cells and detected their EMT characteristics. It turned out that decreased E-cadherin and increased Vimentin were detected after Hook1 knockdown in both mRNA (Figure 3(a) and (b)) and protein levels (Figure 3(c) and (d)). Consistently, the migration capacity of Huh-7 and HepG2 cells was enhanced when Hook1 expression was inhibited (Figure 3(e) and (f)). These findings indicated that Hook1 suppression was sufficient to induce EMT in HCC cells.

Figure 3.

Hook1 RNA interference promoted HCC cell EMT. Huh-7 and HepG2 cells were transfected with Hook1 siRNA or negative control siRNA. (a, b) qPCR showed decreased E-cadherin expression and increased Vimentin expression in the mRNA levels after Hook1 knockdown. (c) Immunoblotting showed decreased E-cadherin and increased Vimentin and Hook1 expression after Hook1 knockdown. (d) Immunofluorescence showed decreased E-cadherin and increased Vimentin after Hook1 knockdown. (e, f) Wound healing assays showed enhanced migration capacity after Hook1 knockdown.

Hook1 knockdown decreases the sensitivity of HCC cells to doxorubicin

It has been frequently reported that HCC cells undergoing EMT are more resistant to chemotherapeutic agents.²⁵ Doxorubicin has been widely applied in numerous chemotherapy studies for various types of cancers including HCC.²⁶ We detected the cell viability after treating Huh-7 and HepG2 cells with 0–1 ng/mL doxorubicin. The ICs₅₀ of doxorubicin to both Huh-7 (0.44 and 6.03 µg/mL for control and Hook1 knockdown, respectively) and HepG2 (0.77 and 5.01 µg/mL for control and Hook1 knockdown, respectively) were significantly increased after Hook1 knockdown (Figure 4). These results showed that decreased Hook1 expression could lead to drug resistance in HCC cells.

Figure 4.

Knockdown of Hook1 increased the drug resistance of HCC cell lines. After Hook1 knockdown by siRNA, the sensitivity of (a) Huh-7 and (b) HepG2 cells to doxorubicin significantly decreased.

Discussion

The role of Hook1 in malignant tumors is quite novel. Traditionally, Hook1 was reported to play a role in endocytosis and maintenance of cell shape. In the past decade, Hook1 was reported to be downregulated in breast cancer and ovarian cancer.^27,28 Previous studies also implicated that Hook1 might be involved in EMT.^28,29 Hook1 is known as a protective factor in lung cancer, and we found that Hook1 expression was much lower in HCC tumor tissues than in the corresponding peritumoral tissues. Furthermore, we found that Hook1 downregulation was associated with the malignancy of HCC specimens and cell lines. Hook1 expression was the highest in primary hepatocytes, and the decrease of Hook1 was observed in HCC cell lines, especially those with mesenchymal features. Surprisingly, Hook1 expression in HL7702, a cell line frequently used as normal liver cell succedaneum, is also low. HL7702 is a genetically modified cell line; therefore, it is probably that the decreased Hook1 expression may be due to its immortal characteristics. More importantly, Hook1 expression was even lower in HCC samples with thrombi, suggesting that Hook1 is negatively associated with invasiveness of tumor cells. By using human specimens and multiple HCC cell lines, we showed that Hook1 could function as a suppressor in HCC.

EMT is considered an important step for metastasis of solid cancers. It is a complicated process involving intravasation, transportation in circulation, and extravasation.⁹ By undergoing EMT, cancer cells acquire mesenchymal characteristics such as weakened cell-to-cell adhesion, lost polarity, augmented invasion, and changes in biomarkers, resulting in biologically aggressive behavior. The role of EMT in hepatic fibrosis and HCC progression has been thoroughly investigated and discussed.^30,31 TGF-β-induced EMT is a classic model in HCC. In our study, TGF-β treatment successfully initiated EMT in Huh-7 and HepG2 cells, showing changes of morphology, migration, F-actin reorganization, and expression of E-cadherin and Vimentin. However, we failed to observe significant morphological change in HepG2 cells after stimulation with TGF-β. We assumed that the special growth pattern of HepG2 cells stacking together with no free stretching might be the reason for the observation. The phenomenon has also been reported previously and does not impede the affirmation of EMT in HepG2 cells.³² Furthermore, although HepG2 and Huh-7 are two frequently used cell lines with obvious epithelial traits, the basal expression of E-cadherin and Vimentin is distinct between the two cell lines. HepG2 normally shows very weak expression of Vimentin, indicating a weaker ability of migration. However, this will not influence the affirmation of EMT if other expected changes exist.

Given that Hook1 was inhibited in TGF-β-induced EMT, we further confirmed that Hook1 knockdown was able to induce EMT in the absence of TGF-β. This means that reduction of Hook1 expression probably contributes to TGF-β-induced EMT but not merely a consequence or an epiphenomenon. Although EMT has been extensively studied, its mechanisms are still not fully elucidated. The association between Hook1 and EMT in cancer is only reported by Li et al.¹⁸ very recently, showing that Hook1 suppresses TGF-β-induced EMT by blocking the phosphatase activity of Src homology 2–domain containing protein-tyrosine phosphatase-2 (SHP2). Consistent with these findings, our study revealed that interference of Hook1 expression facilitated EMT in HCC cells. Surprisingly, however, no significant changes in SHP2 expression were observed after TGF-β treatment in our settings (data not shown), implying that the interaction with SHP2 is not the only way that Hook1 functions as an EMT suppressor. More complicated mechanisms may underlie this phenomenon that Hook1 negatively regulates EMT.

We also showed the decreased sensitivity of HCC cells to doxorubicin in case of Hook1 knockdown. Since EMT plays a critical role in cancer multidrug resistance in HCC,^26,33 it is currently unknown whether Hook1 affects drug resistance directly or indirectly via EMT. This needs further investigation.

In conclusion, we found that Hook1 was downregulated in HCC patients, and its expression level was associated with malignancy. In addition, Hook1 knockdown facilitated EMT and increased resistance to doxorubicin in HCC cells. However, the mechanism by which of Hook1 regulates EMT is complicated and requires further investigation. Our study suggests that Hook1 contributes to EMT and can be a promising approach to HCC therapy.

Footnotes

Acknowledgements

X.S. and Q.Z. contributed equally to this work.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was financially supported by National Natural Science Foundation of China (Nos. 81472212 and 81401954), Key Program of Medical Scientific Research Foundation of Zhejiang Province, China (No. WKJ-ZJ-1410), Key Program of Administration of Traditional Chinese Medicine of Zhejiang Province, China (No. 2014ZZ007), and Zhejiang Provincial Program for the Cultivation of High-level Innovative Health talents.

References

Siegel

Miller

Jemal

Cancer statistics, 2016. CA Cancer J Clin 2016; 66: 7–30.

Kluger

Salceda

Laurent

. Liver resection for hepatocellular carcinoma in 313 Western patients: tumor biology and underlying liver rather than tumor size drive prognosis. J Hepatol 2015; 62: 1131–1140.

Yang

Aqel

Pungpapong

. Direct acting antiviral therapy and tumor recurrence after liver transplantation for hepatitis C-associated hepatocellular carcinoma. J Hepatol 2016; 65: 859–860.

Lee

Lin

Zeng

. Association between nucleos(t)ide analog and tumor recurrence in hepatitis B virus-related hepatocellular carcinoma after radiofrequency ablation. Hepatology 2016; 63: 1517–1527.

Peng

Chen

. Prognostic nomogram for patients with unresectable hepatocellular carcinoma after transcatheter arterial chemoembolization. J Hepatol 2015; 63: 122–130.

Parikh

Marshall

Singal

. Survival and cost-effectiveness of sorafenib therapy in advanced hepatocellular carcinoma: an analysis of the SEER-Medicare database. Hepatology 2017; 65: 122–133.

Terazawa

Kondo

Hosoi

. Transarterial infusion chemotherapy with cisplatin plus S-1 for hepatocellular carcinoma treatment: a phase I trial. BMC Cancer 2014; 14: 301.

Przybyla

Muncie

Weaver

VM.

Mechanical control of epithelial-to-mesenchymal transitions in development and cancer. Annu Rev Cell Dev Biol 2016; 32: 527–554.

Diepenbruck

Christofori

Epithelial-mesenchymal transition (EMT) and metastasis: yes, no, maybe?

Curr Opin Cell Biol 2016; 43: 7–13.

10.

Jayachandran

Dhungel

Steel

JC.

Epithelial-to-mesenchymal plasticity of cancer stem cells: therapeutic targets in hepatocellular carcinoma. J Hematol Oncol 2016; 9: 74.

11.

David

Huang

Chen

. TGF-beta tumor suppression through a lethal EMT. Cell 2016; 164: 1015–1030.

12.

Yuan

Yang

Wang

. A long noncoding RNA activated by TGF-β promotes the invasion-metastasis cascade in hepatocellular carcinoma. Cancer Cell 2014; 25: 666–681.

13.

Iwahashi

Shimada

Utsunomiya

. Epithelial-mesenchymal transition-related genes are linked to aggressive local recurrence of hepatocellular carcinoma after radiofrequency ablation. Cancer Lett 2016; 375: 47–50.

14.

Mohr

OL.

The second chromosome recessive hook bristles in

Drosophila melanogaster. Hereditas 1927; 9: 169–179.

15.

Kramer

Phistry

Genetic analysis of hook, a gene required for endocytic trafficking in drosophila. Genetics 1999; 151(2): 675–684.

16.

Maldonado-Baez

Cole

Krämer

. Microtubule-dependent endosomal sorting of clathrin-independent cargo by Hook1. J Cell Biol 2013; 201(2): 233–247.

17.

Hall

The cytoskeleton and cancer. Cancer Metastasis Rev 2009; 28(1–2): 5–14.

18.

Wang

Zhao

. SHP2 positively regulates TGFbeta1-induced epithelial-mesenchymal transition modulated by its novel interacting protein Hook1. J Biol Chem 2014; 289: 34152–34160.

19.

Chen

Xia

. Rapamycin enhances cetuximab cytotoxicity by inhibiting mTOR-mediated drug resistance in mesenchymal hepatoma cells. Cancer Biol Ther 2014; 15: 992–999.

20.

Livak

Schmittgen

TD.

Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) method. Methods 2001; 25: 402–408.

21.

Zhang

Bai

Chen

. Wnt/β-catenin signaling enhances hypoxia-induced epithelial–mesenchymal transition in hepatocellular carcinoma via crosstalk with hif-1α signaling. Carcinogenesis 2013; 34: 962–973.

22.

Son

Moon

Epithelial-mesenchymal transition and cell invasion. Toxicol Res 2010; 26: 245–252.

23.

Fan

Shiau

Tai

. SHP-1 is a negative regulator of epithelial–mesenchymal transition in hepatocellular carcinoma. Oncogene 2015; 34: 5252–5263.

24.

Wang

. TMEPAI regulates EMT in lung cancer cells by modulating the ROS and IRS-1 signaling pathways. Carcinogenesis 2013; 34: 1764–1772.

25.

Chen

Yan

. NSC 74859 enhances doxorubicin cytotoxicity via inhibition of epithelial–mesenchymal transition in hepatocellular carcinoma cells. Cancer Lett 2012; 325: 207–213.

26.

Dai

Ahn

Wang

. Ascochlorin enhances the sensitivity of doxorubicin leading to the reversal of epithelial-to-mesenchymal transition in hepatocellular carcinoma. Mol Cancer Ther 2016; 15: 2966–2976.

27.

Taube

Herschkowitz

Komurov

. Core epithelial-to-mesenchymal transition interactome gene-expression signature is associated with claudin-low and metaplastic breast cancer subtypes. Proc Natl Acad Sci U S A 2010; 107: 15449–15454.

28.

Kurrey

Jalgaonkar

Joglekar

. Snail and slug mediate radioresistance and chemoresistance by antagonizing p53-mediated apoptosis and acquiring a stem-like phenotype in ovarian cancer cells. Stem Cells 2009; 27: 2059–2068.

29.

De Craene

Gilbert

Stove

. The transcription factor snail induces tumor cell invasion through modulation of the epithelial cell differentiation program. Cancer Res 2005; 65: 6237–6244.

30.

Kim

Choi

Hur

. RAR-related orphan receptor gamma (ROR-γ) mediates epithelial-mesenchymal transition of hepatocytes during hepatic fibrosis. J Cell Biochem. Epub ahead of print 28 October 2016. DOI: 10.1002/jcb.25776.

31.

Liu

. PCAF inhibits hepatocellular carcinoma metastasis by inhibition of epithelial-mesenchymal transition by targeting Gli-1. Cancer Lett 2016; 375: 190–198.

32.

Dituri

Mazzocca

Fernando

. Differential inhibition of the TGF-β signaling pathway in HCC cells using the small molecule inhibitor LY2157299 and the D10 monoclonal antibody against TGF-β receptor type II. PLoS ONE 2013; 8(6): e67109.

33.

Chen

Liang

. Inhibition of mTORC2 induces cell-cycle arrest and enhances the cytotoxicity of doxorubicin by suppressing MDR1 expression in HCC cells. Mol Cancer Ther 2015; 14: 1805–1815.