Single-stranded DNA binding protein 2 expression is associated with patient survival in hepatocellular carcinoma

We enrolled a consecutive series of 213 patients with HCC in this study. All patients were diagnosed and underwent surgery at Hanyang University Hospital (Seoul, Korea) between 1991 and 2013. Patients with no complete clinical follow-up data or available paraffin blocks were excluded, leaving 189 patients. The clinicopathologic characteristics of these cases are summarized in Table 1. The mean follow-up period was 57 months. All patients had received curative surgical resections with negative resection margins. No additional treatment was performed in 74 (39.1%) patients. Transarterial chemoembolization was performed in 90 (47.6%) patients, systemic chemotherapy in 8 (4.2%), radiofrequency ablation in 5 (2.6%), and transplantation in 1 (0.5%). According to the seventh edition of the American Joint Committee on Cancer (AJCC) staging system, 164 cases were Stage I or II and 25 were Stage III or IV. In addition, paired non-neoplastic liver tissue adjacent to the HCC, along with 21 normal liver tissue samples resected from patients with traumatic liver injury, were selected to compare SSBP2 expression. We reviewed all hematoxylin and eosin (H&E)-stained slides, pathology reports, and other medical records to confirm the diagnosis. The clinicopathologic parameters assessed were tumor size, tumor focality (number of tumor nodules, single or multiple), histologic grade (4-tier Edmondson-Steiner grading system), small vessel invasion (presence of tumor emboli within central hepatic vein, portal or large capsular vessel, microscopically), large vessel invasion (presence of tumor emboli in major branch of hepatic vein or portal vein, grossly as well as microscopically), AJCC stage, Ki-67 labeling index (eyeball estimation of TMA cores), serum alpha-fetoprotein (AFP), and patient survival. The pathological evaluation was performed by two pathologists (KJ and HK).

We used a manual tissue microarrayer (Unitma, Seoul, Korea) for tissue microarray construction from archival formalin-fixed, paraffin-embedded tissue blocks. The non-necrotic central portion of the carcinoma, spanning 0.5 cm or larger, was selected by light microscopy of H&E-stained sections. Tissue cylinders of 2-mm diameter were punched from a previously marked lesion on each donor block and transferred to the recipient block (Unitma). Each tissue microarray was comprised of 5 × 10 samples.

The immunohistochemical staining for SSBP2 was performed with 4-μm-thick sections from TMA blocks. The sections were deparaffinized in xylene and then rehydrated through graded ethanol. For antigen retrieval, we performed autoclave heating at 100 °C for 20 min in sodium citrate buffer (pH 6.0). Endogenous peroxidase activity was blocked with peroxidase blocking solution (S2023, Dako, Glostrup, Denmark). TMA slides were incubated with primary antibodies at 4 °C overnight and then incubated with labeled polymer (EnVision/HRP, K5007, Dako) for 30 min at room temperature. The primary rabbit monoclonal [EPR11520] antibody was raised against a synthetic peptide corresponding to the human SSBP2 (amino acid sequence 300 to the C-terminus) (ab177944, Abcam, Cambridge, MA). 3, 3′-Diaminobenzidine tetrahydrochloride was used as the chromogen for detection, and Mayer’s hematoxylin counterstain was applied. Immunostaining for Ki-67 was performed using the Bond-Max automated immunostainer (Leica Biosystems, Wetzlar, Germany) with anti-Ki-67 antibody (Dako, M7240).

SSBP2 expression was evaluated semi-quantitatively by two independent pathologists (HK and KJ) who were blind to patient clinical outcome. Positive expression was defined as more than 10% of the tumor cells to show nuclear staining [12]. A representative microscopic image with positive staining in HCC is shown in Fig. 1.

The human HCC cell line (Huh7) was purchased from the Korean Cell Line Bank (Seoul, Korea). Huh7 cells were cultured in RPMI supplemented with 10% FBS at 37 °C with a 5% CO2 concentration. The human siRNA SSBP2 (Bioneer, Daejeon, Korea) and negative control siRNA (Bioneer) were transfected with Lipofectamine 2000 (Invitrogen, Carlsbad, CA, USA) in a 6-well culture plate and harvested after 48 h.

At 48 h after transfection, cells were digested using trypsin for 5 min and then seeded into six 6-well plates, followed by incubation for 5 days. Each plate was harvested daily. The viable cell numbers were manually obtained with a hemocytometer after trypan blue staining. The experiments were performed in triplicate.

The Transwell assay was performed with a Boyden chamber and 10⁵ transfected cells suspended in the upper chamber with serum free media, which was transferred to a 24-well lower chamber (RPMI containing 10% FBS). Then, 24 h after incubation at 37 °C, the upper cells were removed by wiping the surface of the membrane with cotton swabs. The migrated cells on the lower side were fixed with formalin for 10 min and stained with 1% crystal violet. In 10 microscopic fields (× 200), migrated cells were counted, and mean values were compared. The experiments were performed in triplicate.

Huh7 cell lines transfected with SSBP2 siRNA or NC-siRNA were harvested, and total cell lysates were prepared. The cell lysates were separated by SDS-PAGE and transferred onto nitrocellulose membranes. After blocking with 5% skim milk, the membranes were incubated with antibodies against SSBP2 (rabbit monoclonal, ab177944, Abcam), E-cadherin (rabbit monoclonal, #3195, Cell Signaling, Danvers, MA, USA), N-cadherin (rabbit monoclonal, #13116, Cell Signaling), TCF8/Zeb1 (rabbit monoclonal, #3396, Cell Signaling), Twist1 (rabbit monoclonal, #46702, Cell Signaling), and Beta-actin (rabbit monoclonal, #4970, Cell Signaling). The bound antibodies were detected with HRP-conjugated secondary antibodies and visualized using chemiluminescence reagent (Amersham Biosciences, Little Chalfont, UK).

SSBP2 expression was evaluated in normal human liver tissue, non-neoplastic liver parenchyma adjacent to HCC, and HCC tissue (Fig. 1). Normal hepatocytes were negative for SSBP2 in all 21 normal liver tissue samples, whereas the nuclei of normal bile duct epithelium and sinusoidal endothelium were immunoreactive. Positive immunoreactivity was found in one (0.6%) out of 180 non-neoplastic liver parenchyma samples and in 16 (8.5%) out of 189 HCC tissue samples (Table 2). SSBP2 was more frequently expressed in HCC tumor tissues compared to normal liver or adjacent non-neoplastic tissues (P < 0.001, chi-square test for linear trend).

Table 3 shows the correlations between SSBP2 expression and clinicopathologic parameters. Positive SSBP2 expression was significantly correlated with tumor multifocality (P = 0.027, chi-square test), high histologic grade (P = 0.003, chi-square test), and frequent vascular invasion (P = 0.001, chi-square test). There was no statistically significant correlation between SSBP2 expression and tumor size, AJCC stage, or serum AFP. These results demonstrated that SSBP2 expression correlated with an aggressive histologic phenotype of HCC.

The associations between conventional prognostic factors and patient survival were explored. Univariate survival analyses revealed histologic grade (P = 0.008), AJCC stage (P < 0.001) and vessel invasion (P < 0.001) as predictors of poor disease-free survival (DFS) in HCC patients (Table 4). Similarly, histologic grade (P = 0.005), primary tumor size (P = 0.002), AJCC stage (P < 0.001), and vessel invasion (P < 0.001) were negatively correlated with overall survival (OS) (Table 4). SSBP2 expression was demonstrated as a predictor of poor DFS and OS by univariate Cox regression analysis (P = 0.004 and P = 0.027, respectively). Kaplan-Meier survival curves revealed that patients with SSBP2-positive HCC had poor prognosis in both DFS and OS (P = 0.004 and P = 0.026, respectively, log-rank test) (Fig. 2). However, SSBP2 expression was not an independent prognostic factor for disease-free or overall survival in the multivariate analysis with the Cox proportional hazards model (P = 0.117 and P = 0.138, respectively) (Table 5). Subgroup analysis stratified by AJCC and BLCL stage failed to demonstrate the survival difference between SSBP2-positive and negative patients.

SSBP2-positive tumors had a higher Ki-67 proliferation index than SSBP2-negative tumors (P < 0.001, t-test) in 189 clinical HCC samples (Table 3). To determine the role of SSBP2 in cell proliferation, we examined the effect of SSBP2 downregulation on HCC cell proliferation in vitro. The HCC cell line, Huh7, was transfected with siRNA specifically targeting SSBP2. The knockdown efficiency of SSBP2 in cell lines was confirmed by Western blot. However, there was no significant difference in cell proliferation between the negative control and the SSBP2-downregulated groups (data not shown).

SSBP2 expression was correlated with vascular invasion in HCC patients (Table 3). To further evaluate the role of SSBP2 in tumor cell invasion, we performed a Transwell invasion assay in normal HCC cell lines and HCC cell lines (Huh7) transfected with SSBP2 siRNA. The knockdown efficiency of SSBP2 was confirmed by Western blot. SSBP2 downregulation restrained the migratory capacity of Huh7 cells (P = 0.022, t-test) (Fig. 3a and b). Western blots for EMT markers showed that SSBP2 downregulation increased the epithelial maker level (E-cadherin) and decreased the mesenchymal marker level (Twist1) (Fig. 3c). These results support the concept that SSBP2 promotes invasiveness in HCC.