Differential Expression of IQGAP 1 / 2 in Hepatocellular Carcinoma and its Relationship with Clinical Outcomes

Hepatocellular carcinoma (HCC) is a major form of liver cancer and has become the fifth most common malignancy and the third leading cause of cancer related mortality worldwide (Gomaa et al., 2008). Predisposing factors for HCC include chronic hepatitis B and C virus infections, exposure to aflatoxin B1, chronic alcohol consumption, and any hepatic diseases associated with cirrhosis (Thomas, 2009). The current treatments for HCC with established efficacy include: surgical resection/liver transplantation, transcatherter arterial chemoembolization, percutaneous radiofrequency ablation, percutaneous ethanol injection, percutaneous microwave coagulation therapy and molecular targeted therapy (e.g. sorafenib) (Maluccio et al., 2012; Berk et al., 2013; Nishikawa et al., 2013). Although the clinical outcomes of HCC have significantly improved, the survive rate remains poor due to the high rate of recurrence. Therefore, it is crucial to better understand carcinogenesis of HCC and seek new


Introduction
Hepatocellular carcinoma (HCC) is a major form of liver cancer and has become the fifth most common malignancy and the third leading cause of cancer related mortality worldwide (Gomaa et al., 2008).Predisposing factors for HCC include chronic hepatitis B and C virus infections, exposure to aflatoxin B1, chronic alcohol consumption, and any hepatic diseases associated with cirrhosis (Thomas, 2009).The current treatments for HCC with established efficacy include: surgical resection/liver transplantation, transcatherter arterial chemoembolization, percutaneous radiofrequency ablation, percutaneous ethanol injection, percutaneous microwave coagulation therapy and molecular targeted therapy (e.g.sorafenib) (Maluccio et al., 2012;Berk et al., 2013;Nishikawa et al., 2013).Although the clinical outcomes of HCC have significantly improved, the survive rate remains poor due

Differential Expression of IQGAP1/2 in Hepatocellular Carcinoma and its Relationship with Clinical Outcomes
Fa-Da Xia, Zhuo-Lu Wang, Hong-Xi Chen, Yun Huang, Jin-Dong Li, Zhi-Ming Wang, Xin-Ying Li* biomarkers for potential targeted therapy.
IQ-motif containing GTPase-activating proteins (IQGAPs) represents a small subgroup of evolutionally conserved superfamily of GTPase-activating proteins (Bernards, 2003).Mammalian cells contain three IQGAPs, IQGAP1, IQGAP2 and IQGAP3.They share a high degree of sequence homology and a similar domain structure, and differ in tissue distribution (White et al., 2009).All three are large cytoplasmic scaffolding proteins .Their domain structure includes an actin binding calponin homology (CH) domain, a single WW domain capable of binding various proline-rich proteins, four IQ motifs binding calmodulin, a large GTPase binding domain (GBD) known to bind Rho GTPases Rac1 and cdc42, and a RasGAP C-terminus domain (RGCT) (Brown et al., 2006).IQGAP1 is the best characterized one, and its binding partners involve in several cell signal pathways.The function of IQGAP1 protein includes cytoskeltal regulation, coordinating cadeherin mediated cell-cell adhesion, cell polarization and actin reorganization to promote cell migration.Accumulating evidence strongly supports IQGAP1 is an oncogene (Casteel et al., 2012;White et al., 2012).A comparison of structure and function of IQGAP2 with those of IQGAP1 reveals some similarities and several differences.The human IQGAP2 protein exhibits 59% identity to IQGAP1, and their domain structure are well conserved (Briggs et al., 2003).Binding to diverse partners, IQGAP2 and IQGAP1 may play opposing roles in carcinogenesis (Jin et al., 2008;Schmidt et al., 2008;Schmidt, 2012).
It has been demonstrated that IQGAP1 and IQGAP2 are reciprocally altered in HCC (Schmidt et al., 2008;White et al., 2010).The increased IQGAP1 and/or decreased IQGAP2 expression may contribute to the pathogenesis of human HCC.However, the prognostic implication of IQGAP1/2 in HCC remained uncertain.In this study, we investigated the expression patterns of IQGAP1/2 in clinical HCC tissues and evaluated their prognostic roles in 150 HCC patients after curative hepatectomy with long-term follow-up and extensive information on clinicopathologic characteristics.

Patient selection
Fresh tissues were collected from 21 patients with HCC who underwent curative hepatectomy (defined as tumor tissues and para-tumor tissues) and 11 patients who experienced operation due to parenchyma of hepatic hemangioma (normal liver tissues acquired from the incisal edge, defined as normal control tissues) in the department of General Surgery, Xiangya Hospital, Central South University between Mar 2013 and Aug 2013.These 32 tissues were used to detect the mRNA and protein levels of IQGAP1 and IQGAP2.The study was approved by the Institutional Review Board at Xiangya Hospital, Central South University, Changsha, China.
A consecutive series of 150 previously untreated patients who received curative hepatectomy in our department between Jan 2007 to Dec 2007 were enrolled.All patients were histopathological diagnosed and followed-up.In this series, adjacent para-tumor tissues and 11 embedded normal tissues were also included.Their paraffin blocks were used for Immunohistochemistry assay.Curative resection was defined according to BCLC staging system, and patients with a major portal vein tumor thrombus and tumor rupture certified before or during operation were excluded (Bruix et al., 2011).Patients with serious diseases and other malignant tumors were also excluded.The patients' characteristics were shown in Table 1.Tumor stage was defined according to 7th edition tumor-node-metastasis (TNM) classification of American Joint Committee on cancer (AJCC), and tumor differentiation was assessed according to Edmonson and Steiner grading system.

Western blot analysis
Briefly, total proteins were separated on 15% SDS-PAGE gel then transferred to immune-blot PVDF membrane for immunoblot analysis.The membranes were blocked with 5% deffated milk before exposing to mouse monoclonal anti-IQGAP1 antibody against human IQGAP1 and anti-IQGAP2 antibody against human IQGAP2 (ABCAM, Cambridge, MA, USA; dilution 1:1000).Secondary antibody used in this study was goat anti-mouse antibody with a horseradish peroxidaseconjugated.The experiments were performed for at least 3 times.

Immunohistochemistry (IHC)
Immunohistochemical staining was done on formalinfixed and paraffin-embeded tissue using 5-μm section from tissue microarray blocks.Mounted tissue section were baked at 60℃ for 30 min, deparaffinized in xylene and rehydrated through graded alcohols.Antigen was retrieved by heating in 1μM sodium citrate (PH 6.0) in a pressure cooker for 2 min.According to the manufacturer's instruction, anti-IQGAP1 (dilution 1:500) and anti-IQGAP2 antibody (dilution 1:100) were incubated on the section overnight at 4℃ after non-specific staining was blocked, followed by incubation with a horseradish peroxidase conjugated secondary antibody.Staining was visualized using DAB chromogen substrate and counterstained with haematoxylin.

IHC evaluation
All stained sections were assessed independently by two pathologists without knowledge of clinicopathologic features, and any differences in interpretation were resolved by consulting a third pathologist to achieve consensus.Staining for each antibody was considered positive if more than 10% of cells stained strongly in the cytoplasm as described (White et al., 2010;Eom et al., 2011).

Follow up
All the enrolled 150 patients were followed up for 5 years.The disease-free survival was defined as the length of time after hepatectomy for HCC during which a patient survived with no signs of HCC recurrence.Overall survival was defined as the interval between the time of hepatectomy to either death or the last date of follow-up.

Statistical analysis
Group differences were compared with the Student's t-test and Chi-squared test or Fisher's exact test.A spearman's rank correlation was performed to analyze the correlation.Univariate survival analysis was done DOI:http://dx.doi.org/10.7314/APJCP.2014.15.12.4951Differential Expression of IQGAP1/2 in Hepatocellular Carcinoma and its Relationship with Clinical Outcomes by the Kaplan-Meier method, and multivariate analysis was done by the Cox proportional hazards regression model.All statistical analysis was performed using SPSS 19.0 software.P<0.05 was considered to be statistically significant.

Increased IQGAP1 expression and decreased IQGAP2 expression in HCC
IQGAP1/2 mRNA and protein expression were detected and analyzed in all fresh tissues from the 32 patients.The results showed that IQGAP1 mRNA and protein expression level were significantly higher in tumor tissues than para-tumor (p<0.001 and 0.002 respectively) and normal tissues (p<0.001 and 0.030 respectively, Figure 1 and 2).On the contrary, IQGAP2 mRNA and protein expression level were significantly lower in tumor tissues than para-tumor (p<0.001 and 0.007 respectively) and normal tissues (p=0.001 and 0.004 respectively, Figure 1 and 2).
The results demonstrated that positive IQGAP1 expression was correlated with larger tumor size, advanced TNM stage and tumor differentiation ( Ⅲ and Ⅳ) (Table 1).Negative IQGAP2 expression was significantly associated with larger tumor size, muticentirc tumor occurrence, advanced TNM stage and tumor differentiation (Ⅲ and Ⅳ) (Table 1).Moreover, spearman`s rank analysis showed that IQGAP1 expression was negatively correlated with IQGAP2 expression (p<0.001).

Correlation of IQGAP1/2 protein expression with survivals
Survival data were available for all 150 post-surgical  HBV, hepatitis B virus; AFP, alpha fetoprotein (Table 2).Patients with IQGAP1+ or IQGAP2-tumors had significantly reduced disease-free survival (p<0.001 and p=0.006 respectively, Figure 4a and 4b) and overall survival (p<0.001 for both, Figure 4c and 4d).Subsequently, we analyzed the survival curves of 4 groups according to the expressional profile of IQGAP1 and IQGAP2.Patients with IQGAP1+/IQGAP2-tumor had the poorest survival, while the IQGAP-/IQGAP2+ group had the most favorable survival (Figure 4e and  4f).No significant difference were observed between the other two groups.Multivariate analysis indicated that the positive IQGAP1 expression with negative IQGAP2 expression (IQGAP1+/IQGAP2-expression pattern,

Discussion
Recent studies have demonstrated IQGAPs involved in tumorigenesis.IQGAP1 is overexpressed in colorectal carcinoma (Nabeshima et al., 2002), breast cancer (Jadeski et al., 2008), aggressive ovarian adenocarcinomas (Dong et al., 2008) and gastric cancer (Walch et al., 2008), while IQGAP2 expression is lost in gastric carcinoma (Jin et al., 2008), prostate (Xie et al., 2012) and liver cancer (Schmidt, 2012;Gnatenko et al., 2013).It was also reported that IQGAP1 expression is up-regulated in HCC, characterized by the loss of membrane E-cadherin expression, the cytoplasmic translocation of β-catenin, and the overexpression of nuclear target of β-catenin, cyclin D1 (Chen et al., 2010).What's more, tumor suppressor gene Iagap2 is linked to the development of HCC and the activation of Wnt/β-catenin signaling pathway.Inactivation of IQGAP1 in mouse liver impairs tumorigensis caused by IQGAP2 deficiency (Schmidt et al., 2008;Gnatenko et al., 2013).In present study, we demonstrated that IQGAP1 mRNA and protein were up-regulated and IQGAP2 mRNA and protein were down-regulated in human HCC tissues, compared to those in para-tumor and normal liver tissues.These findings suggested that IQGAP1 and IQGAP2 may possess opposing function in the pathogenesis of HCC, and were in accordance with the data observed by other investigators (White et al., 2010).
IQGAP1/2 sharing the same domain structure and significant homology, however, involve in different signaling pathways regulating cell proliferation, transformation, cell motility and invasion (White et al., 2009;Schmidt, 2012).In our study, the expression profiles of IQGAP1 and IQGAP2 were analyzed in a large series of human HCC with direct correlation to clinical and survival data.We demonstrated that positive IQGAP1 and negative IQGAP2 expression were related to some unfavorable clinicpatologic parameters which are associated with advanced tumor stages including larger tumor size, muticentirc tumor occurrence, advanced TNM stage, incomplete tumor encapsulation and tumor differentiation (Ⅲ and Ⅳ).Moreover, positive IQGAP1 expression was significantly associated with negative IQGAP2 expression.Our findings strongly suggested that positive IQGAP1 and negative IQGAP2 were significantly correlated with HCC progression, and provided the evidence that IQGAP1 and IQGAP2 may have distinct functions.Accumulating evidences have elucidated the mechanisms of how IQGAP2 counteracts the effect of IQGAP1 in cancer.Although both IQGAP1 and IQGAP2 binding Rac1 and cdc 42, it appears that IQGAP1 selectively binds to an inacvtive GDP bound form of these GDPases, while IQGAP2 binds both GDP-and GTP-bound forms (Brill et al., 1996;Hart et al., 1996;McCallum et al., 1996;Joyal et al., 1997;Schmidt, 2012).This is particularly relevant to HCC, because targeted ablation of cdc42 in mouse hepatocytes and bile ducts resulted in the developed of HCC (Van Hengel et al., 2008;Schmidt, 2012).IQGAP1 and IQGAP2 are phosphorylated at ser1443 by protein kinase Cε (PKCε) and Thr716 by cAMP-dependent protein kinase (PKA) respectively (Breuhahn et al., 2006;Elliott et al., 2012).PKCε and PKA kinases regulate distinct signaling pathway, regarding cancer progression, and their dissimilar activators and targets may also hold a key to deciphering the mechanisms (Grohmanova et al., 2004;Wang et al., 2009;Schmidt, 2012;).
HCC is a heterogeneous cancer with high mortality.Searching for valuable biomarkers for HCC diagnosis and prognostic prediction has been attracting an increasing number of experts.Plenty of proteins have been shown to have clinical significance for predicting HCC prognosis (Luo et al., 2013;Wang et al., 2014;Yu, 2014).The availability of these clinically applicable biomarkers remains limited, thus there is an urgency to strengthen the role of these biomarkers as prognostic factors for HCC patients.A lack of IQGAP1 related with favorable prognosis has been observed in gastric cancer (Walch et al., 2008).Through this retrospective study of the 150 patients with HCC, we found that positive IQGAP1 and negative IQGAP2 were associated with unfavorable disease-free and overall survival rates.This result suggested that IQGAP1 and IQGAP2 might be important prognostic markers for advanced stage HCC after curative hepatectomy.It is noteworthy that IQGAP1/2 switch acts as an independent prognositic factor in HCC.Further studies are needed to identify which protein plays the predominant role in HCC, thus enhancing their clinical predicative value for the prognosis of HCC patients.
In summary, this study demonstrated for the first time that positive IQGAP1 and negative IQGAP2 expression were correlated with advanced HCC stage and unfavorable prognosis.We recommend that IQGAP1 and IQGAP2 should be used as adjunctive biomarkers to improve prognostication for individual patient.However, there is a fundamental problem in using immunohistochemicl staining in evaluating the expression profile of IQGAP1/2 as subjectively interpreted by histopathologists.
Prospective studies with larger patient populations need further investigating the value of IQGAP1 and IQGAP2 by PCR and western blotting, and the cut-off value of IQGAP1/2 ratio as prognostic predictors.

Figure 1 .Figure 2 .
Figure 1.Relative IQGAP1/2 mRNA Value of HCC, Para-tumor and Normal Tissues.Up-regulation of IQGAP1 and down-regulation of IQGAP2 in HCC tumor compared to para-tumor and normal liver tissues.(* p<0.05)

Figure 3 .
Figure 3. Immunohistochemistry Staining of IQGAP1/2 in HCC, Para-tumor and Normal Tissues.A, positive expression of IQGAP1 in tumor tissue.B, negative expression of IQGAP1 in para-tumor tissue.C, negative expression of IQGAP1 in normal tissue.D, negative expression of IQGAP2 in tumor tissue.E, positive expression of IQGAP2 in para-tumor tissue.F, positive expression of IQGAP2 in normal tissue.(Original magnification: ×400)