Expression Of Semaphorin-3B, Vascular Endothelial Growth Factor, And Their Common Receptor Neuropilin-1in Hepatocellular Carcinoma

Hepatocellular carcinoma (HCC) is is one of the most frequent carcinomas worldwide and is characterized by progressive development,high postsurgical recurrence,and extremely poor prognosis. The dismal outcome has been attributed to the highly vascular nature of HCC, which increases the propensity to spread and invade into neighboring or distant sites. The possible role of vascular endothelial growth factor (VEGF) in HCC has received much attention because VEGF increases vascular permeability and enhances angiogenesis, represented by microvessel density (MVD). VEGF signalling involves three tyrosine kinase receptors, flt-1(VEGFR-1),KDR/flk-1(VEGFR-2), and flt-4(VEGFR-3), and two non-tyrosine kinase receptors, neuropilin-1(NP1) and-2(NP2), which were initially identified on neuronal cells as receptors for class3semaphorins (SEMA-3).The NP-1and NP-2receptors (NP-2shares47%homology with NP-1) are VEGF165coreceptors that specifically enhance the interaction of VEGF165with VEGFR-2. In several studies,the up-regulation of NP-1is correlated to the aggressiveness in prostate cancer in vivo and increases angiogenesis in colon and pancreatic cancer in vitro.Berge M et.al found that increased NP-1expression in human tumour hepatocytes is significantly associated with primary HCC and enhance disease progression, in both vascular and tumour compartments.The SEMAs were initially identified through their role in axon guidance. It has recently been reported that SEMAs inhibit the angiogenic effect of VEGF165by competition for NP1binding. In addition, lung and breast carcinomas were reported to have frequent allelic loss or loss of heterozygosity (LOH) at3p21.3, where SEMA-3B are located. Thus, SEMA-3B are candidates for tumor suppressor genes. Moreover,a balance between SEMA-3B and VEGF165was reported to induce tumor progression in breast cancer. Nevertheless,there have been no published data on the expression of SEMA-3B and neuropilins in hepatocellular carcinoma.Research ObjectivesWe investigated the expression of SEMA-3B,VEGF, NP1proteins, and MVD in hepatocellular carcinoma. The VEGF/SEMA-3B ratio was also estimated from the VEGF and SEMA-3B staining scores. Finally, we analyzed the prognostic differences according to SEMA-3B, VEGF, NP-1expression, and the VEGF/SEMA ratio in patients with hepatocellular carcinoma.Materials and methods1.Clinical Specimens and Pathological Materials The resections of the carcinoma and tumor-adjacent (> lcm from the cancer tangent) tissues were collected from73patients with hepatocellular carcinoma. These patients (57males and16females, ages32to73) were hospitalized from July2005to December2010in the Department of General Surgery of Qilu Hospital of Shandong University. As controls,14normal liver tissues were also collected. All of these specimens were pathologically confirmed and the patients used in this study had not been treated by either radiofrequency ablation or intervention before surgery. Each of the collected specimens was divided into two parts, one was flash-frozen in liquid nitrogen and the other was paraffin-embedded.2.1mmunohistochemistry To study the expression of SEMA-3B, VEGF and NP1, the sections were stained with SEMA-3B, VEGF and NP1polyclonal antibody (ABCAM,UK,1:200), Immunosignals were detected using streptavidin-biotin-fluorochrome (SABC) supplied within a kit (GeneTex, USA). To measure microvessel density (MVD),mouse anti-human CD34monoclonal antibody (Santa, USA,1:200in PBST) were used and the detection procedure was similar to that used for SEMA-3B. As negative controls, PBS was used instead of primary antibody.For the purpose of quantification, semi-quantitative integration method was adopted.First,five fields on each slice were randomly selected under the magnification of100x, and500cells in each field were counted under a higher magnification (200x).The appearance of yellow to brown particles in the cytoplasm and membranous was considered a positive signal.Staining scores were established by optical analysis using the product of the percentage of positive cells and staining intensity from0to3(no color=0point, yellow=1point, brown=2points and tawny=3points), thus ranging from0to300.The results of immunostaining were classified as negative (-) when the score was0to40, positive(+) when the score was40to300. To analyze the balance between SEMA3and VEGF165, each VEGF staining score was divided by the SEMA score. The resulting VEGF/SEMA-3B ratio was classified as low when it was0to3and high when it exceeded3.To measure MVD, the five highest vascular density areas were selected under100x light microscope, and CD34-positive cells or cell clusters in each area were counted under200x (0.708mm2). The average number of the five regions was used as the MVD value. The results were independently confirmed by two senior pathologists.3.Western blot200mg of each tissue were homogenized in0.6ml of lysis buffer (Bi Yun Tian,China). After incubation for30min at room temperature, the homogenates were centrifuged at12,000rpm for10min at4℃and the supernatant was collected-Protein concentration was determined according to the instruction of the BCA Protein Concentration Quantification Kit (Bi Yun Tian, Beijing, China). For Western blotting,50℃for5min and separated on a10%polyacrylamide gel. The separated proteins were transferred onto a PVDF membrane,which was then blocked in TBST supplemented with5%of dry milk for2-3hours. The blocked PVDF membrane was incubated overnight at4℃with rabbit anti-human SEMA-3B polyclonal antibody (ABCAM,UK,1:5000dilutions in TBST). The membrane was then washed3times with TBST, with each wash lasting for10min. After the last wash, the membrane was incubated for90min at room temperature with goat anti-rabbit secondary antibody Bi Yun Tian Company, China,1:2000dilutions in TBST). The membrane was then washed3times with TBST and immunosignals were detected with the ECL Luminescent Reagent Kit (MILLIPORE, USA).4.Statistics The Chi-square test was used to analyze the association between SEMA-3B,VEGF and NP1expression and clinicopathologic variables and MVD.Spearman rank correlation was used to determine whether there was a positive or negative correlation.Kaplan-Meier analysis was used to calculate the survival curves, and log-rank test was used to compare the survival differences of patient subgroups. Multivariate analysis was used to identify significant independent prognostic factors for RFS. All of the statistical analyses were performed with the SPSS13.0statistical software and P<0.05is considered to be a statistically significant difference.Results1. Immunohistochemistry of SEMA-3B, VEGF, and NP-1in hepatocellular carcinoma1.1. Expression of SEMA-3BExpression of SEMA-3B was observed in the cytoplasm of hepatocellular carcinoma cells, tumor-adjacent, or normal liver tissues. The expression of this protein was significantly down-regulated (P<0.05) in the hepatocellular carcinoma cells when compared to the cells from either tumor-adjacent, or normal liver tissues.The SEMA-3B-positive cells in the normal liver and paraneoplastic tissues were78.6%and85.7%, respectively, while in hepatocellular carcinoma cells only42.9%expressed SEMA-3B protein. In support of this result, our Western blotting analysis showed that the expression level of SEMA-3B protein was much lower in the cancer tissues than that in the tumor-adjacent or normal tissues (85.7%and78.6%, respectively).We analyzed the differences between SEMA-3B-positive and SEMA-3B-negative patients in the clinical and pathological features, these analyses showed that SEMA-3B-negative was significantly associated with larger tumor size (P=0.046), vascular invasion (P=0.045),satellite nodules (P=0.048),Edmondson-Steiner Grades (P=0.010).1.2. Expression of VEGFThe VEGF protein was expressed in the cytoplasm and along the plasma membrane of the tumor cells. The expression of this protein was significantly up-regulated (P<0.05) in the hepatocellular carcinoma cells when compared to the cells from either tumor-adjacent, or normal liver tissues. The VEGF-positive cells in hepatocellular carcinoma cells and paraneoplastic tissues were72.6%and25.1%, respectively, while in the normal liver was not detected. We analyzed the differences between VEGF-positive and VEGF-negative patients in the clinical and pathological features, these analyses showed that VEGF-positive was significantly associated with larger tumor size (P=0.017),Edmondson-Steiner Grades (P=0.034).1.3. Expression of NP-1Expression of NP-1was observed in the endothelial and tumor cells, mostly in the cytoplasm but also along the cell membrane. The expression of this protein was significantly up-regulated (P<0.05) in the hepatocellular carcinoma cells when compared to the cells from either tumor-adjacent, or normal liver tissues. The NP-1-positive cells in hepatocellular carcinoma cells and paraneoplastic tissues were63.0%and4.1%, respectively, while in the normal liver was not detected. We analyzed the differences between NP-1-positive and NP-1-negative patients in the clinical and pathological features, these analyses showed that NP-1-positive was significantly associated with larger tumor size (P=0.002), satellite nodules (P=0.046),Edmondson-Steiner Grades (P=0.019).2Correlations among the expressions of SEMA-3B,VEGF, NP-1and MVD Expression of MVD Microvessel density (MVD) can be measured by the number of CD34-positive endothelial cells. In this study, CD-34positive cells were immunohistochemically stained with the monoclonal anti-CD34antibody. The number of the CD43-positive cells in the area of0.708mm2(full field area under magnification of200) were recorded under light microscopy. To statistically study the MVD in different tissues, we defined92as a median value of MVD in this study.Specifically, the MVD value in SEMA-3B-negative positive cells was greater than that in SEMA-3B-positive negative cells (P=0.003).The negative correlation between the level of SEMA-3B protein and the expression of MVD may reflect that the importance of SEMA-3B in the regulation of tissue vascularization.The MVD value in VEGF-positive cells was greater than that in VEGF-negative cells (P=0.021).The MVD value in NP-1-positive cells was greater than that in NP-1-negative cells (P=0.021).The expression of SEMA-3B showed significant negative correlations with VEGF (P<0.001) and NP-1(P=0.009). There was a tendency toward a positive correlation between VEGF and NP-1, but it was not significant.3Patient survival according to SEMA-3B, VEGF, NP-1and VEGF/SEMA-3B statusAmong the73patients with carcinoma, the prognostic significance of SEMA-3B, VEGF, NP1and NP2immunostaining,and the VEGF/SEMA-3B ratio was analyzed using the Kaplan-Meier method. The results obtained by log-rank test showed that the prognosis was poorer in patients with tumors negative for SEMA-3B (positive:48.4%versus negative:35.7%; P=0.225), although the differences were not statistically significant. There was no prognostic difference according to NP-1immunostaining. Interestingly, patients with a high VEGF/SEMA-3B ratio showed poorer survival than those with a low ratio (high:28.6%versus low:66.7%; P=0.002). Multivariate analysis showed that the VEGF/SEMA-3B ratio was an independent prognostic factor (P=0.031).Conclusions:Low level of SEMA-3B expression, high level of VEGF, NP-1expression in patients with hepatocellular carcinoma was closely associated with tumor progression, angiogenesis. A high VEGF/SEMA-3B ratio has adverse prognostic significance in patients with hepatocellular carcinoma.