The liver cancer cell line HepG2 was purchased from Shanghai Institute of Biochemistry and Cell Biology, Chinese Academy of Sciences (Shanghai, China). HepG2 cells were cultured in Dulbecco's modified Eagle's medium (DMEM)-high glucose containing trypsin (cat no. SH30022.01B) supplemented with 10% fetal bovine serum (FBS; cat no. SH30084.03) (both from HyClone, Logan, UT, USA), 100 U/ml penicillin (cat no. ST488-1; Beyotime Institute of Biotechnology, Shanghai, China) and 100 U/ml streptomycin (cat no. ST488-2; Beyotime Institute of Biotechnology) at 37°C under 95% air and 5% CO₂.

RNA was extracted from the tissue samples using TRIzol^® reagent (Thermo Fisher Scientific, Waltham, MA, USA), according to the manufacturer's instructions. Subsequently, cDNA was synthesized using a TaqMan Reverse Transcription Reagents kit (Thermo Fisher Scientific), according to the manufacturer's protocol. The relative expression levels of mRNA were determined using a Power SYBR-Green PCR Master Mix kit (Thermo Fisher Scientific) and normalized to GAPDH. RT-PCR was performed using the Applied Biosystems 7500 Fast Dx Real-Time PCR instrument (cat no. 4425757; Thermo Fisher Scientific) and the following gene-specific primers (Sangon Biotech Co., Ltd., Shanghai, China): GAPDH: Sense, 5′-TGCCATCAACGACCCCTTCA-3′ and antisense, 5′-TGACCTTGCCCACAGCCTTG-3′; E-cadherin: Sense, 5′-AGCTATCCTTGCACCTCAGC-3′ and antisense, 5′-CCCAGGAGTTTGAG-3′; N-cadherin: Sense, 5′-TCCTGCTCACCACCACTACTT-3′ and antisense, 5′-CTGACAATGACCCCACAGC-3′; Smad: Sense, 5′-ATAAGCAACCGCCTGAACAT-3′ and anti-sense, 5′-TTACCTGCCTCCTGAAGACC-3′; Twist: Sense, 5′-GCTGATTGGCACGACCTCT-3′ and antisense, 5′-CACCATCCTCACACCTCTGC-3′; and vimentin: Sense, 5′-CCAAACTTTTCCTCCCTGAACC-3′ and antisense, 5′-GTGATGCTGAGAAGTTTCGTTGA-3′. A control siRNA specific for the red fluorescent protein, 5′-CCACTACCTGAGCACCCAG-3′, was used as the negative control (sc-37007; Santa Cruz Biotechnology, Inc., Santa Cruz, CA, USA). All primers were designed using the National Center for Biotechnology Information Primer-BLAST tool (http://www.ncbi.nlm.nih.gov/tools/primer-blast/). PCR was performed under the following conditions: Denaturation at 50°C for 2 min, followed by 38 cycles at 95°C for 15 sec and 60°C for 1 min. Gene expression was normalized to internal controls and fold changes were calculated using the relative quantification method (2^−ΔΔCq) (11).

Cells were washed 3 times with ice-cold PBS and then incubated on ice with 250 μl RIPA buffer (cat no. P0013; Beyotime Institute of Biotechnology) with 2.5 μl phenylmethylsulfonyl fluoride (cat no. ST506-2; Beyotime Institute of Biotechnology) for 15–30 min. The cells were collected and centrifuged at 13,000 × g for 10 min at 4°C. The protein concentrations of the cell lysates were measured in duplicate using a bicinchoninic acid protein assay kit (cat no. 23227; Thermo Fisher Scientific). The protein lysates and 6X loading buffer were mixed at a ratio of 4:1 and then boiled for 5 min at 100°C. Equal amounts of total protein were separated by 10% sodium dodecyl sulphate-polyacrylamide gel electrophoresis and transferred onto polyvinylidene difluoride membranes (cat no. FFP39; Beyotime Institute of Biotechnology). The membranes were immunoblotted with monoclonal mouse β-actin (1:1,000; cat. no. sc-8432), goat anti-human p-STAT3 (1:1,000; cat. no. sc-21876), monoclonal mouse anti-human JAK (1:400; cat. no. sc-376996), monoclonal mouse anti-human STAT3 (1:400; cat no. sc-293151), p-JAK (1:400; cat. no. sc-16773), E-cadherin (1:500; cat. no. sc-21791), N-cadherin (1:400; cat. no. sc-393933) and vimentin (1:400; cat. no. sc-373717) antibodies (all from Santa Cruz Biotechnology, Inc.) at 4°C overnight. All antibodies were diluted with 0.5% bovine serum albumin. Following incubation, the corresponding secondary antibody conjugated with peroxidase and enhanced chemiluminescence reagents (Beyo ECL Plus; cat. no. P0018; Beyotime Institute of Biotechnology) were applied, and the blot was visualized (cat. no. 121-2550; Beijing Liuyi Biotechnology Co., Ltd., Beijing, China). The amount of total protein was semiquantified as ratio to β-actin on each gel.

Scattering assay was performed as previously described (7). HepG2 cells (3×10⁵/ml) were seeded into each well of a 24-well plate (cat. no. 662102; Greiner Bio-One GmbH, Frickenhausen, Germany), and incubated overnight at 37°C in an atmosphere of 5% CO₂. The cells were pretreated with 10 μM TGFβ1 for 48 h at 37°C for 48 h in 95% air and 5% CO₂. Representative images were captured at a magnification of ×20 using the Eclipse TE2000-U inverted microscope (Nikon Corporation, Tokyo, Japan).

The invasion assay was performed using Transwell 24-well plates with 8-μm pore polycarbonate membranes (BD Biosciences, Franklin Lakes, NJ, USA). Briefly, the upper side of the membranes was coated with Matrigel (20 μg/well) and the membranes were then air-dried for 1 h at 37°C. The lower side of the membranes was coated with 5 μg fibronectin, and the treated or untreated HepG2 cells (2×10⁵) in 200 μl of DMEM medium with 2.5% FBS were placed in the upper chamber. The lower chamber was filled with DMEM medium with 10% FBS as the chemoattractant. The invasion chamber was incubated for 8 h at 37°C and 5% CO₂. The cells on the upper surface of the membrane were removed by gentle scrubbing with a cotton swab. The membranes were fixed in a stationary liquid of 95% ethanol and 5% acetic acid for 30 min and stained with crystal violet. The number of cells on the lower surface of the membrane in 5 random visual fields (magnification, ×200) was then counted using an Eclipse TE2000-U inverted microscope. Each assay was performed in triplicate.

The wound healing assay was performed as previously described: HepG2 cells (3×10⁵/ml) were seeded into a 6-well plate (cat. no. 657160; Greiner Bio-One GmbH) in serum-containing medium, and incubated at 37°C in an atmosphere of 5% CO₂ in order to form a confluent monolayer. The monolayer was scratched using a sterile plastic pipette tip (cat. no. CLS4860; Sigma-Aldrich, Merck KGaA, St. Louis, MO, USA), and washed with PBS to remove cell debris. Subsequently, fresh medium was added, and 10 μM TGFβ1 or 0.1 ml DMSO was added to each well. The scratched mono-layer was incubated at 37°C in an atmosphere of 5% CO₂ for 48 h. Wound closure was measured in 6 random high-power fields at a magnification of ×200, using Image-Pro^®Express software, version 6 (Media Cybernetics, Inc., Rockville, MD, USA) and an Eclipse TE2000-U inverted microscope (11).

Data were analyzed using SPSS software, version 1.0 (SPSS, Inc., Chicago, IL, USA) and GraphPad Prism software, version 5.0 (GraphPad Software, Inc., La Jolla, CA, USA). Analysis of variance was conducted followed by the Student's t-test. The data are presented as mean ± standard deviation. P<0.05 was considered to indicate statistically significant differences.

To determine the migration and invasion induced by TGFβ1, liver cancer HepG2 cells were treated with different concentrations of TGFβ1 for 48 h, and the migration and invasion of cancer cells were assessed by wound closure assays and Matrigel Transwell chamber invasion assays. The effects of TGFβ1 were observed at concentrations as low as 5 μM. As the TGFβ1 concentration increased, the migration and invasion of HepG2 cells also increased in a concentration-dependent manner, with the most prominent effects observed at a concentration of 10 μM (Fig. 1A and B). These results indicated that TGFβ1 induced HepG2 cell migration and invasion in a concentration-dependent manner. Hence, the concentration of 10 μM was selected for all further mechanistic studies.

TGFβ1 is a factor that promotes EMT in cancer cells, as previously reported (7,8). EMT is an important mechanism of cancer cell invasion and metastasis. The downregulation of E-cadherin expression and the upregulation of vimentin and N-cadherin expression are considered to be markers of EMT. In the present study, TGFβ1 also induced cell scattering (Fig. 2A), indicating that TGFβ1 induces EMT, thereby increasing the migration and invasion of HepG2 cells. To further investigate whether EMT is involved in TGFβ1-induced scattering, migration and invasion of HepG2 cells, the expression of E-cadherin, vimentin and N-cadherin were first detected by qPCR and western blot analysis. As shown in Fig. 2B and C, the expression of vimentin and N-cadherin was upregulated, whereas that of E-cadherin was downregulated following treatment with 10 μM TGFβ1. These results demonstrated that TGFβ1-induced EMT promoted the migration and invasion of HepG2 cells.

Moreover, JAK/STAT3 protein expression was detected by western blot analysis and it was observed that TGFβ1 stimulated the expression of p-JAK and p-STAT3. This finding indicates that TGFβ1-induced EMT may be activated by JAK/STAT3 signaling.

STAT3 is the key transcription factor regulating cell proliferation and survival. STAT3 may be activated by oncostatin M, interferons, interleukin-6 (IL-6) and epidermal growth factor (EGF). It was recently reported that TGFβ1 induced JAK/STAT3 signaling to increase migration and invasion in lung carcinoma cells. Based on these reports, we hypothesized that JAK/STAT3 signaling may be involved in TGFβ1-induced EMT to increase the migration and invasion in HepG2 cells. To confirm this hypothesis, HepG2 cells were incubated with the STAT3 inhibitor AG490 prior to treatment with TGFβ1. As shown in Fig. 3A and C, AG490 significantly suppressed the TGFβ1-induced upregulation of N-cadherin and vimentin expression and the downregulation of E-cadherin expression, compared with the TGFβ1 group. Furthermore, AG490 treatment significantly reduced the number of TGFβ1-induced invasive and migratory cells (Fig. 3E and F), which was consistent with the results obtained from metastasized cell-wound closure (Fig. 3D).

Smad2 and Twist are important factors regulating EMT through TGFβ1. Thus, we hypothesized that TGFβ1 induced EMT by upregulating Twist expression via JAK/STAT3 signaling. To confirm this hypothesis, the expression of pSmad2 and Twist in HepG2 cells treated with TGFβ1 was first detected. The results revealed that TGFβ1 induced the protein expression of pSmad2 and Twist (Fig. 4A–D). By contrast, AG490 treatment reversed the TGFβ1-induced protein expression of pSmad2 and Twist (Fig. 4E–G). Along these lines, TGFβ1 induced the protein expression of pSmad2 and Twist in accordance with the activated JAK/STAT3 signaling.

To further determine whether Twist participated in TGFβ1-induced EMT, HepG2 cells were transfected with siRNA of Twist. As shown in Fig. 5A–C, Twist knockdown significantly suppressed the TGFβ1-induced expression of N-cadherin and vimentin, but reversed the TGFβ1-inhibited expression of E-cadherin. The TGFβ1-induced migration and invasion in HepG2 cells were also reversed through Twist knockdown. Overall, these data strongly suggest that Twist participates in TGFβ1-induced EMT to increase the migration and invasion of HepG2 cells via JAK/STAT3 signaling.