The human HCC cell line HepG2 was cultured in Dulbecco's modified Eagle's medium (DMEM) containing 10% fetal bovine serum (FBS) (both from Gibco^® Life Technologies, USA) under a humidified condition at 37°C with 5% CO₂.

The Nanog-TALEN plasmid was constructed according to our previous study (17) using a Fast TALE™ TALEN Assembly kit (SiDan-Sai Biotechnology, China). Approximately 10 µg of the Nanog-TALEN plasmid (5 µg each for right and left arm) and control plasmid were mixed with wild-type HepG2 cells (1×10⁶) in 90 µl of Opti-MEM (Gibco^® Life Technologies), and were transferred to electroporation cuvettes for electroblotting. After the cells were transfected with Nanog-TALEN and the control plasmid under 150 V, the cells were exposed to 3 µg/ml puromycin for 3 days. Then, the medium containing puromycin was replaced with growth media.

Genomes derived from the Nanog-targeted and control cells were extracted using a Genomic DNA Extraction Mini kit (Tiangen, China). The DNA fragment surrounding the Nanog-targeted site was amplified and then analyzed by T7E1 (View Solid Biotechnology, China) to evaluate Nanog-mutation efficiency. The Nanog mutation was confirmed by sequencing.

The proliferation of targeted and control cells was monitored using an RTCA SP system, which is a real-time cell-based assay system (ACEA Biosciences, San Diego, CA, USA). According to the manufacturer's instructions, 5×10³ cells were seeded in triplicate in an E-Plate VIEW 16 for real-time monitoring of the proliferative capacities of the cells for 120 h.

Scratch assays were used to determine the migratory ability of HepG2 cells in compliance with our previous description (17). The cells were maintained in triplicate into 6-well plates until completely confluent, and then a scratch was created in the confluent cell monolayers using a 10-µl pipette tip. After the cells were washed 3 times with phosphate-buffered saline (PBS), the cells was cultured in serum-free DMEM. The migratory capacity was assessed at 48 h under a microscope.

We employed a Transwell invasion assay (17) to measure the invasive abilities of the cells in vitro. Cells were trypsinized and resuspended in DMEM at a density of 2×10⁵ cells/ml. Approximately 200 µl of cell suspension was added in triplicate to the upper chamber of the polycarbonate membrane filter, which was embedded with 8-µm pores (Corning, USA), that were pre-coated with Matrigel. The lower chamber was filled with 500 µl DMEM supplemented with 10% FBS. After the cells were incubated for 48 h, non-migrating cells on the surface of the Matrigel in the upper chamber were wiped off by cotton swabs, whereas the cells migrating to the bottom of the membrane were fixed in 95% ethanol for 30 min, and then stained with 0.1% crystal violet. After the stained membrane was washed 3 times with PBS, it was observed under a microscope and the number of migrating cells was counted.

Chemosensitivity assays were performed to evaluate the sensitivity of Nanog-targeted and control cells to chemotherapeutics. Trypsinized cells were washed with PBS and then seeded in triplicate in an E-Plate VIEW 16 at a density of 1×10⁴ cells/well for real-time monitoring using an RTCA SP system. After 6 h, cisplatin (40 µg/ml) was added to the E-Plate VIEW 16 for detection of chemosensitivity at 72 h.

Total RNA from the Nanog-targeted and control cells was extracted using TRIzol reagent (Invitrogen, USA). The corresponding cDNA was synthesized using a reverse transcription kit (Takara, China). SYBR-Green PCR Master Mix was utilized to amplify the corresponding genes of interest. The PCR cycling program was as follows: 95°C for 5 min, followed by 32 cycles of 95°C for 30 sec, 58°C for 30 sec, 72°C for 30 sec with a final 10-min incubation at 72°C. The relative levels of gene expression were analyzed by the 2^−ΔΔCt method. The fold-change of mRNA was normalized to β-actin. Primer sets for corresponding genes were the same as in our previous study (17).

Nanog-targeted and control cells were collected, washed with PBS; and then lysed on ice using protein lysis buffer and protein inhibitors. Protein lysates were mixed with SDS-PAGE protein loading buffer (5:1) and boiled for 5 min, followed by separation by 12% SDS-PAGE. Then, the proteins of interest were transferred onto polyvinylidene fluoride (PVDF) membranes (Millipore, USA). After the membranes were blocked in 5% non-fat milk in tris-buffered saline with Tween-20 (TBST), they were incubated overnight at 4°C with primary antibodies, respectively: anti-Nanog (no. sc-374103), anti-Twist (no. sc-81417) (1:500; Santa Cruz Biotechnology, USA), anti-Sox2 (no. D6D9), anti-Oct4 (no. 2750) (1:500; Cell Signaling, USA), anti-CD133 (no. bs-4770R), anti-CDK2 (no. bs-10726R), anti-cyclin E (no bs-0573), anti-E-cadherin (no bs.1519R), anti-N-cadherin (no. bs-1172R), anti-vimentin (no. bs-0756R), anti-ABCG2 (no. bs-0662R), anti-MDR1 (no. bs-0563R) (1:400; Bioss, China), anti-cyclin D1 (no. AC853), anti-cyclin D3 (no AC856), anti-GAPDH (no. AG019) (1:500; Beyotime Institute of Biotechnology, China). The membranes were then incubated with a goat anti-rabbit (no. A0208) or anti-mouse (no. A0216) IgG-conjugated to alkaline phosphatase secondary antibody for 2 h. Finally, the membranes were washed 3 times with TBST and imaged using a gel imaging system (Bio-Rad, USA).

All data are represented as the mean ± standard deviation (SD) of 3 repeated experiments. The Student's test was used to for comparison between two groups. P<0.05 was considered statistically significant. All data were analyzed with SPSS statistical software 18.0.

To detect whether TALEN successfully achieved cleavage of the target site of the Nanog gene (Fig. 1A), genomic DNA from Nanog-targeted and control cells was extracted and then used to amplify the DNA fragment containing the Nanog-targeted site. The amplified product was analyzed by T7E1 to detect the activity and efficiency of TALEN. Surprisingly, the efficiency of the TALEN-mediated Nanog gene mutation approached 50% after two transfections (Fig. 1B). Subsequently, we selected out single cells for further culture from the Nanog-targeted mixture of cells. After 15 days, single cell-derived subclone genomes were extracted and analyzed by T7E1. Eventually, we identified subclones for further genomic sequencing, and found that subclone 3 and 7 exhibited biallelic Nanog mutations. The sequencing results indicated there were at least 3 Nanog alleles in the HepG2 cells (Fig. 1C).

To determine the effects of Nanog disruption on stemness factors, we detected the expression of the pluripotency factors Oct4, Sox2, Klf4, c-Myc and Lin28, as well as the expression of cancer stem cell (CSC) marker CD133 by quantitative RT-PCR and western blotting. Our results indicated that the expression of Nanog, Oct4, Sox2, Klf4 and CD133 was markedly downregulated in the targeted cells, whereas the expression of c-Myc and Lin28 was not significantly different (Fig. 2A and B).

To determine the effect of Nanog disruption on the apoptosis and proliferation of HepG2 cells, the proliferative capacity of Nanog-targeted and control cells was measured using an RTCA SP system, which demonstrated that the proliferative capacity of Nanog-targeted cells was markedly reduced relative to the control cells (Fig. 3A). We further detected cyclin D1/D3/E1 and cyclin-dependent kinase 2 (CDK2) protein expression by western blotting. In the Nanog-targeted cells, the expression levels of cyclin D1/D3/E1 and CDK2 protein were significantly reduced (Fig. 3B).

The scratch assays indicated that Nanog disruption contributed to the reduced migration of HepG2 cells. As depicted in Fig. 4A, the migration distance of the targeted cells was significantly less than that of the control cells. CXCR4 is a migration-related factor that is able to induce tumor cell migration (14,27,28). Notably, in the present study, we also found that CXCR4 expression was significantly decreased in the Nanog-targeted cells (Fig. 4B). Transwell assays indicated that the number of targeted cells that invaded through the membrane within 48 h was 15±5, which was less than that of the control cells (80±8; P<0.01) (Fig. 4C). Previous studies have reported that matrix metallopeptidase 2 (MMP2) is an invasion-related gene that can regulate the invasive ability of tumor cells. High expression levels of MMP2 are closely related to tumor progression, invasion, metastasis and poor prognosis in various types of cancers (14,29). In the present study, we also found that MMP2 expression was noticeably reduced in the targeted cells compared with expression in the control cells (Fig. 4D; P<0.05).

To elucidate the effect of the disruption of Nanog on EMT, we detected the expression of the epithelial marker E-cadherin, and the mesenchymal markers N-cadherin and vimentin by western blotting. Relative to the control cells, E-cadherin expression was elevated, while N-cadherin and vimentin expression was decreased in the Nanog-targeted cells. We further detected the expression of the EMT regulators Twist and Snail, and found that their expression levels were decreased in the Nanog-targeted cells. (Fig. 5).

To assess the effect of chemotherapeutics on HepG2 cells, both Nanog-targeted and control cells were exposed to cisplatin (40 µg/ml), and chemosensitivity was measured using an RTCA SP system. After 72 h of observation, we noted that the Nanog-targeted cells were more sensitive to cisplatin than the control cells (Fig. 6A). Furthermore, we detected the expression of multi-drug resistant gene 1 (MDR1) and ATP-binding cassette subfamily G member 2 (ABCG2) in the Nanog-targeted and control cells, and found that expression levels of both were significantly downregulated in the Nanog-targeted cells (Fig. 6B).