The present study was approved by the ethics committee of Guilin Medical University (Guilin, China), and written informed consent was obtained from each patient involved in the study. A total of 20 paired cancerous and matched adjacent normal tissues were collected from patients with HCC undergoing hepatectomy at the Affiliated Hospital of Guilin Medical University between 2012 and 2014. The tissues were snap-frozen in liquid nitrogen and stored at −80°C following surgery for reverse transcription-quantitative polymerase chain reaction (RT-qPCR) and western blot analyses. Another 216 paired paraffin-embedded HCC samples for use in immunohistochemical analysis, were collected between 2010 and 2014 and obtained from the Affiliated Hospital of Guilin Medical University and Zhengzhou People's Hospital (Zhengzhou, China). The tissues were prepared into a tissue microarray chip by Guilin Fanpu Biological Technology Co., Ltd. (Guilin, China). The survival rates were calculated from the date of surgery to the date the patient succumbed to morality or the last follow-up. Medical details, including age, tumor size and serum level of α-fetoprotein, were collected from the medical records of each patient. Tumor staging was performed according to the World Health Organization standards (26), and histological tumor grading was based on Edmondson-Steiner classification (27).

Fozen tissue samples were pulverized by mortar and pestle in liquid nitrogen. Then, ice-cold TRIzol (Invitrogen; Thermo Fisher Scientific, Inc., Waltham, MA, USA) was added to the powdered tissues, which were subsequently transferred to Eppendorf tubes (Eppendorf, Hamburg, Germany) on ice for RNA extraction. Total RNA was extracted and purified from 20 pairs of fresh frozen HCC tissues and corresponding noncancerous tissues. The RT step was carried out using PrimeScript™ II 1st strand cDNA Synthesis kit (Takara Biotechnology Co., Ltd., Dalian, China). The levels of messenger RNA (mRNA) were quantitated using SYBR^® Green Realtime PCR Master Mix (Toyobo Co., Ltd., Osaka, Japan) and analyzed in a ViiA™ 7 Real-Time PCR System (Applied Biosystems; Thermo Fisher Scientific, Inc.). The primers used for RT-qPCR analysis were purchased from Invitrogen (Thermo Fisher Scientific, Inc.) and were as follows: RYBP, forward 5′-TCGCAACTTCCATTGATT-3′ and reverse 5′-TCACACTCAGTCATACCT-3′; GAPDH, forward 5′-TCGCAACTTCCATTGATT-3′ and reverse 5′-TCACACTCAGTCATACCT-3′. The amplification conditions consisted of the following: 30 min at 42°C for reverse trancsription and 2 min at 94°C for Taq activation, followed by 35 cycles of 94°C for 20 sec, 58°C for 20 sec and elongation at 72°C for 30 sec. GAPDH was used as the internal control for determining the mRNA expression of RYBP. Fluorescent data were converted into quantification cycle (Cq). ΔCq values of each sample were calculated as Cq gene of interest - Cq GAPDH. A ΔCq value of 3.33 corresponds to a magnitude lower of gene expression compared with that of GAPDH. The experiment was performed in triplicate. The results were normalized with respective internal controls.

The homogenized HCC samples were lysed in radioimmunoprecipitation assay lysis buffer, containing 150 mmol/l NaCl, 1% Triton X-100, 0.5% deoxycholate, 0.1% SDS and 50 mmol/l Tris (pH 7.4), and the lysates were harvested by centrifugation at 16,000 × g at 4°C for 30 min. The concentration of protein was determined by BCA Protein Assay kit (Beyotime Institute of Biotechnology, Haimen, China). Subsequently, 20 µg of protein was separated by electrophoresis on a 12% sodium dodecyl sulfate polyacrylamide gel and transferred onto a polyvinylidene fluoride membrane. Following blocking of nonspecific binding sites for 60 min with 5% nonfat milk, the membranes were incubated overnight at 4°C with anti-rabbit polyclonal antibody against RYBP (GR104527-2; Abcam, Cambridge, MA, USA) at a 1:1,000 dilution. The membranes were then washed three times with Tris-buffered saline with Tween-20 (TBST) for 10 min and were probed with an anti-rabbit immunoglobulin G (IgG) antibody (GGHL-15PXSPP; Immunology Consultants Laboratory, Portland, OR, USA) at a 1:1,000 dilution at room temperature for 1 h. Following three washes with TBST, the membranes were developed using an enhanced chemiluminescence system (Cell Signaling Technology, Inc., Danvers, MA, USA). The band intensity was measured by densitometry using Quantity One software version 4.62 (Bio-Rad Laboratories Inc., Hercules, CA, USA). The protein level of RYBP was normalized to the level of β-actin, detected using a mouse monoclonal antibody against β-actin (KC5A08; Kangchen Biotech, Shanghai, China) overnight at 4°C at a 1:10,000 dilution.

Formalin-fixed, paraffin-embedded tissue blocks were used for immunohistochemical analysis to detect the expression of RYBP, ZEB1 and ZEB2 using standard methods. The tissues were first fixed with 10% formalin at 4°C for 24 h. The paraffin-embedded tissue sections (6 µm thick) previously constructed into a tissue microarray chip were deparaffinized and quenched for endogenous peroxidase activity with methanol and 3% hydrogen peroxide for 15 min. The sections were then processed in 10 mmol/l citrate buffer (pH 6.0) and heated at 120°C for 5 min to retrieve the antigen. The sections were incubated for at 37°C 1 h with anti-RYBP (goat anti-rabbit polyclonal antibody; 1:100 dilution), anti-ZEB1 (goat anti-rabbit polyclonal antibody; ab124512; 1:100 dilution; Abcam) and anti-ZEB2 antibodies (goat anti-rabbit polyclonal antibody; 1:100 dilution; ab138222; Abcam) diluted 1:100 in 1% bovine serum albumin (BSA) (Beyotime Institute of Biotechnology). As a negative control, sections were incubated with 1% BSA/PBS without primary antibody. The sections were then washed with PBS for 5 min and incubated with the anti-rabbit IgG antibody (1:1,000 dilution; Immunology Consultants Laboratory) at 37°C for 15 min. Following rinsing in PBS, the reaction was visualized under a light microscope by incubating the sections with diaminobenzidine solution for 15 min, following which the sections were weakly counterstained with hematoxylin. All the immunostained sections were evaluated in a blinded-manner with no knowledge of the clinicopathological information. For the assessment of RYBP, ZEB1 and ZEB2, five fields in each specimen were randomly selected and >500 cells were counted under a microscope (Olympus Corporation, Tokyo, Japan) to determine the mean percentage of immunostained cells relative to the total number of cells. The positive cell staining percentages were scored into four categories: 0 for 0%, 1 for 1–33%, 2 for 34–66% and 3 for 67–100% staining. The immunohistochemical staining intensities were also scored into four grades (0, 1, 2 and 3), according to the brown color intensity of the cells: 0 for no color, 1 for light color, 2 for medium color and 3 for dark brown color. The sum of the percentage and intensity scores was used as the final RYBP, ZEB1 and ZEB2 staining score. The staining scores were defined as low expression for scores of 0–2 and high expression for scores of 3–6.

All statistical analyses were performed using SPSS version 19.0 (IMB SPSS, Armonk, NY, USA). The paired-samples-test and one-way analysis of variance were used to compare the mRNA expression of RYBP normalized to GAPDH between the HCC and corresponding adjacent non-tumor tissues, whereas the χ² test was applied for the comparison of dichotomous variables. The Kaplan-Meier estimate was used for survival analysis, and the log-rank test was selected to compare the cumulative survival durations in the patients. P<0.05 was considered to indicate a statistically significant difference.

To investigate the expression profile of RYBP in HCC, the present study performed RT-qPCR and western blot analyses in 20 pairs of HCC tissues and matched adjacent non-tumor tissues. The mRNA expression levels of RYBP was markedly lower in the HCC samples, compared with the high levels of expression in the matched adjacent tissues (P=0.012; Fig. 1A).

Consistent with the RT-qPCR data, the protein expression of RYBP was also low in 16 of the 20 (80.0%) paired tissue samples, determined using western blot analysis (Fig. 1B). The protein level of RYBP was significantly lower in the HCC tissues, compared with that in the adjacent non-tumor tissues (P<0.01).

Immunohistochemical analysis was then performed in all 216 archival paraffin-embedded HCC samples. In total, 60 of the 216 (27.8%) cases were positive for the expression of RYBP in the cancerous tissues, whereas 156 of the 216 (72.2%) cases were negative for the expression of RYBP (Fig. 2A; P<0.01). The association between the protein expression levels of RYBP and the clinicopathological characteristics of RYBP was determined using the χ² test. As shown in Table I, the negative expression of RYBP was significantly associated with tumor size (P=0.046) and metastasis (P=0.028), which suggested that the expression of RYBP was correlated with the diagnosis and prognosis of HCC.

To determine the prognostic value of RYBP in HCC, the present study also assessed the association between the expression of RYBP and survival rates using Kaplan-Meier analysis with a log-rank test. As shown in Fig. 2B, the survival rates of the patients with HCC were significantly different between cases positive for the expression of RYBP and cases negative for the expression of RYBP (P=0.027). In patients with HCC, a low expression level of RYBP indicated a poorer prognosis, compared with those with a high expression level of RYBP.

As there was a significant correlation between RYBP and HCC metastasis, the present study aimed to investigate the role of RYBP in EMT. A number of well-known EMT markers were selected and their expression was detected in HCC tissues using immunohistochemistry. ZEB1 was the first EMT marker assessed, which represses the E-cadherin promoter and induces EMT by recruiting SMARCA4/BRG1 (28). In all 216 archival paraffin-embedded HCC samples, 159 of 216 (73.6%) cases were positive for the expression of ZEB1, whereas 57 of 216 (26.4%) cases were negative for the expression of ZEB1 (Fig. 3A; P<0.01).

To determine the correlation between ZEB1 and the clinicopathological characteristics and prognosis of HCC, a χ² test and Kaplan-Meier analysis with a log-rank test were used. As shown in Table II, the expression of ZEB1 was significantly associated with clinical stage (P=0.035) and metastasis (P=0.008). As shown in Fig. 3B, the survival rates of patients with HCC were significantly different between those positive for the expression of ZEB1 and those negative for the expression of ZEB1 (P<0.001). In patients with HCC, the group with high expression levels of ZEB1 had shorter survival rates, compared with the group expressing low levels of ZEB1. These results indicated that ZEB1 is a suitable EMT marker for HCC.

Statistical analysis was also performed to examine the association between the expression of RYBP and ZEB1 in the HCC tissues. As shown in Table III, the expression of RYBP was negatively correlated with the expression of ZEB1 in HCC tissues (r=−0.473; P<0.001). These results suggested that a low level of RYBP may promote EMT in HCC.

As with ZEB1, ZEB2 is a transcription inhibitor of E-cadherin implicated in gastric cancer (29,30). In the 216 archival paraffin-embedded HCC samples, 135 of the 216 (62.5%) cases were positive for the expression of ZEB2 in cancerous tissues, whereas 81 of the 216 (37.5%) cases showed low expression levels of ZEB2 (Fig. 4A; P<0.01). The correlation between the expression of ZEB2 and the clinicopathological characteristics and prognosis of HCC were examined using a χ² test and Kaplan-Meier analysis. As shown in Table IV, the positive expression of ZEB2 was significantly associated with metastasis (P=0.008). The log-rank test showed that the survival rate of patients with HCC in the ZEB2-positive expression group was significantly shorter, compared with that in the ZEB2-negative expression group (Fig. 4B; P<0.001). As with ZEB1, a significant correlation was found between the expression of ZEB2 and the metastasis and prognosis of HCC.

Statistical analysis of the association between the expression of RYBP and ZEB2 also showed a negative correlation (r=−0.416; P<0.001; Table V). Taken together, RYBP was negatively correlated with ZEB1 and ZEB2, suggesting the importance of RYBP in the EMT process in HCC.