All patients have provided written informed consent to have their tissues stored and used for future research. The ethics committee of The Affiliated Tumor Hospital of Guangxi Medical University (Guangxi, China) approved this study. A total of 38 HCC tissue samples and corresponding paraneoplastic tissue samples (>2 cm away from the edge of the tumor) were obtained from patients (age, 24–70 years; mean age, 49.5±11.7) who underwent surgical resection at the Affiliated Tumor Hospital of Guangxi Medical University between January 2009 and December 2011. The diagnosis of HCC in each patient was confirmed histopathologically. Variables including tumor stage, presence of portal vein tumor thrombosis, number and size of tumors, serum α-fetoprotein, tumor differentiation and presence of extrahepatic metastases and recurrence were evaluated in the patients with HCC. The tumor stage was determined according to the Barcelona-Clinic Liver Cancer staging classification (12). Tumor differentiation was graded according to the Edmondson grading system (12). Prior to surgical resection, none of the patients with HCC received chemotherapy. In addition, ten normal liver tissue samples surgically resected from patients with benign liver lesions were also collected.

Animal experiments were conducted in accordance with the guidelines for The Care and Use of Laboratory Animals issued by the Ethics Committee of the Affiliated Tumor Hospital of Guangxi Medical University. Rat and tree shrew animal models of HCC induced by aflatoxin B1 (AFB1) have been previously established by our group (13,14). A total of 42 female Wistar rats (age, 4 weeks; weight, 180–210 g) were housed individually under controlled light (12 h light/dark cycle) and temperature conditions. Food and water were available ad libitum. The rats were randomly divided into two groups: An AFB1 group (n=28) and a control group (n=14). The rats in the AFB1 group were intraperitoneally injected with AFB1 (100–200 µg/kg) 1–3 times/week. Liver biopsies were obtained from all rats during the 14 th, 28 th, 42 nd and 55 th week, and all rats were sacrificed on the 64 th week by cervical dislocation. A total of 15 HCC tissue samples and 15 corresponding paraneoplastic tissues samples were collected from the rats in the AFB1 group. A total of 14 normal liver tissue samples were collected from the rats in the control group. Adult tree shrews (n=27; age, 6 months; weight, 100–160 g) were obtained from the Kunming Institute of Zoology, Chinese Science Academy (Yunnan, China), and were allowed to acclimatize for one week to the facilities prior to the experiment. The tree shrews were housed in individual, suspended, stainless steel wire cages, under controlled environmental conditions with a 12 h light/dark photoperiod. They had ad libitum access to tap water and a natural diet containing rice powder (20%), corn powder (20%), beef (20%), wheat bran (10%), soy bean (10%), egg (10%), whole milk powder (5%), and a sugar, salt, vitamin, mineral mix (5%). They were also fed daily with reconstituted powdered milk and fruit. The shrews were randomly divided into an AFB1 group (n=15; 9 males and 6 females) and a control group (n=12; 6 males and 6 females), and as previously reported (15), the animals in the AFB1 group were fed AFB1 (400 mg/kg body weight/day in a small amount of milk; Sigma-Aldrich, St. Louis, MO, USA) from the 1 st to the 90 th week of the experiment, whereas the animals in control group were raised without AFB1 administration. Liver biopsies from each animal were obtained every three months under Ketamine Hydrochloride anesthesia (50 mg/kg; Fujian Gutian Pharmaceutical Co., Ltd., Fujian, China). In the present study, 10 HCC tissue samples and 10 corresponding paraneoplastic tissue samples were obtained when the animals developed cancer (HCC appeared at the 105 th week of experimentation) following sacrifice by cervical dislocation. A total of 10 normal liver tissue samples were collected from tree shrews in control group. After surgical resection, all tissue samples were subjected to rapid freezing in liquid nitrogen and then stored at −80°C. HCC samples were confirmed histopathologically.

Bioconductor (16) 2.10.1 was used to standardize the data. The Robust Multi-array Average (RMA) algorithm (17,18) in the software package, Affy, was used for background correction, standardization and Log2 conversion of raw data on the Affymetrix platform. A t-test was performed to evaluate each probe in each data set. Only genes that were included in the Kyoto Encyclopedia of Genes and Genomes (KEGG) database (19) were subjected to GSEA. Genes with an interquartile range (IQR) <0.5 were excluded. If one gene corresponded to several probes, only the probe with the highest IQR was used. GSEA was performed using the Category package in Bioconductor. The gene sets represented by ≥10 genes were subjected to further analyses, and genes in each pathway were subjected to a t-test. After 1,000 permutations, the P-value of each pathway was obtained. The upregulated and downregulated pathways in all five data sets were compared. The cell cycle pathway was upregulated in all five data sets. All genes in the cell cycle pathway were subjected to meta-analysis in each data set. T-tests were conducted using SAS9.13 software (Cary, NC, USA) to calculate the P-value of each probe in the cell cycle pathway in each data set and the χ² value of each gene was calculated using the following formula:

In which the degree of freedom is two times that of the data set K). Genes with P<0.05 were obtained. Analyses on these gene pathways were performed using the Database for Annotation, Visualization and Integrated Discover (DAVID; http://david.abcc.ncifcrf.gov/) in KEGG.

Total RNA was extracted from tissue samples using TRIzol reagent (Invitrogen; Thermo Fisher Scientific, Waltham, MA, USA), and first-strand cDNA was synthesized by reverse transcription from 1 µg RNA using a RevertAid First Strand cDNA Synthesis kit (Thermo Fisher Scientific, Inc.). RT-qPCR was performed using a standard protocol from the SYBR^® Premix Ex Taq kit (Takara, Dalian, China) on an Applied Biosystems 7300/7500 Real Time PCR system (Applied Biosystems, Grand Island, NY, USA). The cycling conditions were as follows: 95°C for 30 sec, and 40 cycles at 95°C for 5 sec followed by 60°C for 34 sec. The relative mRNA expression levels of cdc25a to control GAPDH were analyzed using ABI PRISM 7300 software v1.3.1 (Applied Biosystems) and calculated with the double standard curves method. The cdc25a forward and reverse primer sequences were 5′-CCAAAGGAACCATTGAGAAC-3′ and 5′-CAGATGCCATAATTTCTGGAG-3′, respectively, and the product length was 138 bp. The forward and reverse primer sequences for the internal control gene, 3-phosphate dehydrogenase (GAPDH), were 5′-AAGAAGGTGGTGAAGCAGGC-3′ and 5′-ACCACCCTGTTGCTGTAGCC-3′, respectively, and the product length was 200 bp.

Western blot analysis to assess cdc25a and GAPDH expression was performed as previously described (14). Protein samples (60 µg) were separated by 10% SDS-PAGE, prior to being transferred onto PVDF membranes (EMD Millipore, Billerica, MA, USA). The membranes were subsequently blocked with Tris-buffered saline with Tween 20 (TBST; Beyotime Institute of Biotechnology, Haimen, China) containing 5% (w/v) skimmed milk powder for c. 2 h at room temperature and then incubated at 4°C overnight with the following primary antibodies: Rabbit polyclonal anti-human Cdc25 (cat. no. ab75743; Abcam, Cambridge, MA, USA) or mouse monoclonal anti-human glyceraldehyde 3-phosphate dehydrogenase (GAPDH; cat. no. TA-08; Beijing Zhongshan Golden Bridge Biotech Co., Ltd., Beijing, China) at dilutions of 1:100 or 1:1000, respectively. The membranes were then washed three times with TBST buffer, and incubated for 1 h at room temperature with IRDye 680LT goat anti-mouse secondary antibody (cat. no. 0926-68020; LI COR Biosciences, Lincoln, NE, USA) or IRDye 680LT goat anti-rabbit secondary antibody (cat. no. 926-68021; LI-COR Biosciences) at 1:7,500 dilution. The membranes were scanned using an Odyssey infrared imaging system (LI-COR Biosciences). Quantification of bands was achieved by ratiometric analysis of the fluorescent intensities of cdc25a and GAPDH using Odyssey Application Software 3.0 (LI-COR Biosciences).

SPSS 13.0 (SPSS, Inc., Chicago, IL, USA) was used for data analysis. Quantitative data are expressed as the mean ± standard deviation. Comparison of mean values between multiple groups was performed using a one-way analysis of variance. P<0.05 was considered to indicate a statistically significant difference.