Suppression of Glypican 3 Inhibits Growth of Hepatocellular Carcinoma Cells through Up-Regulation of TGF-β2

Volume 13 Number 8

August 2011

pp. 735–747 735

www.neoplasia.com

Suppression of Glypican 3 Inhibits Growth of Hepatocellular Carcinoma Cells through Up-Regulation of TGF-β21,2

Chris K. Sun, Mei-Sze Chua, Jing He and Samuel K. So Asian Liver Center and Department of Surgery, Stanford University School of Medicine, Stanford University, Stanford, CA, USA

Abstract Glypican 3 (GPC3) is a valuable diagnostic marker and a potential therapeutic target in hepatocellular carcinoma (HCC). To evaluate the efficacy of targeting GPC3 at the translational level, we used RNA interference to examine the biologic and molecular effects of GPC3 suppression in HCC cells in vitro and in vivo. Transfection of Huh7 and HepG2 cells with GPC3-specific small interfering RNA (siRNA) inhibited cell proliferation (P < .001) together with cell cycle arrest at the G1 phase, down-regulation of antiapoptotic protein (Bcl-2, Bcl-xL, and Mcl-1), and replicative senescence. Gene expression analysis revealed that GPC3 suppression significantly correlated with transforming growth factor beta receptor (TGFBR) pathway (P = 4.57e−5) and upregulated TGF-β2 at both RNA and protein levels. The effects of GPC3 suppression by siRNA can be recapitulated by addition of human recombinant TGF-β2 to HCC cells in culture, suggesting the possible involvement of TGF-β2 in growth inhibition of HCC cells. Cotransfection of siRNA-GPC3 with siRNA–TGF-β2 partially attenuated the effects of GPC3 suppression on cell proliferation, cell cycle progression, apoptosis, and replicative senescence, confirming the involvement of TGF-β2 in siRNA-GPC3–mediated growth suppression. In vivo, GPC3 suppression significantly inhibited the growth of orthotopic xenografts of Huh7 and HepG2 cells (P < .05), accompanied by increased TGF-β2 expression, reduced cell proliferation (observed by proliferating cell nuclear antigen staining), and enhanced apoptosis (by TUNEL staining). In conclusion, molecular targeting of GPC3 at the translational level offers an effective option for the clinical management of GPC3-positive HCC patients. Neoplasia (2011) 13, 735–747

Introduction Hepatocellular carcinoma (HCC) is the fifth most common cancer worldwide and the third most common cause of cancer mortality [1]. It is typically aggressive and intrinsically resistant to standard chemotherapeutic agents, underscoring the need for developing more effective therapies for HCC patients [2,3]. Recent studies have implicated glypican 3 (GPC3) as an important protein in HCC. As a histologic marker, GPC3 staining could distinguish malignant HCC from preneoplastic, cirrhotic, or benign liver lesions [4–9]. As a serum marker, the cleaved and secreted form of GPC3 was found to be elevated in the serum of HCC patients, making it a valuable, noninvasive diagnostic marker of HCC [10–12]. Functionally, GPC3 promotes growth of HCC cells through stimulation of the canonical Wnt signaling pathway [13], which regulates the expression of many downstream oncogenic proteins such as c-MYC [14]. Given the importance of Wnt signaling in HCC [15], GPC3 may also have therapeutic potential in the clinical management of HCC.

GPC3 was first identified in patients with Simpson-Golabi-Behmel syndrome, an X-linked disease characterized by prenatal and postnatal overgrowth caused by mutation of the GPC3 gene [16]. It is one of six members of the glypican family, which are heparan sulfate proteoglycans Abbreviations: BMP, bone morphogenic protein; GPC3, glypican 3; GPI, glycosylphosphatidylinositol; HCC, hepatocellular carcinoma; SA-β-Gal, senescence-associatedβ-galactosidase; SMD, Stanford Microarray Database; TGF-β2, transforming growth factor, beta 2 Address all correspondence to: Dr. Mei-Sze Chua, PhD, Department of Surgery, Stanford University School of Medicine, Stanford University, 1201 Welch Rd, MSLS Bldg, P228, Stanford, CA 94305-5655. E-mail: [email protected] 1 This work is supported by grants to the Asian Liver Center at Stanford University from the H. M. Lui Foundation, the C. J. Huang Foundation, and the T. S. Kwok Liver Research Foundation. The authors declare no conflict of interest. 2 This article refers to a supplementary material, which is designated by Figure W1 and is available online at www.neoplasia.com. Received 11 May 2011; Revised 6 June 2011; Accepted 7 June 2011 Copyright © 2011 Neoplasia Press, Inc. All rights reserved 1522-8002/11/$25.00 DOI 10.1593/neo.11664

736

GPC3 Suppression Inhibits HCC Growth via TGF-β2

Sun et al.

that are linked to the exocytoplasmic surface of the plasma membrane by a glycosyl-phosphatidylinositol (GPI) anchor [17]. Glypicans play a critical role in the regulation of cell proliferation and survival, particularly during development and malignant transformation [18–21]. The main function of membrane-attached glypicans is to regulate the signaling of Wnts, Hedgehogs (Hhs), fibroblast growth factors, and bone morphogenic proteins (BMPs) [20,22–24]. Depending on the cellular context, glypicans may have a stimulatory or inhibitory role in these signaling pathways: in tissues where cell proliferation is predominantly controlled by Hh signaling, GPC3 overexpression inhibits proliferation, whereas in tissues where canonical Wnt signaling predominates, GPC3 overexpression stimulates cell proliferation [25]. In HCC, GPC3 interacts with Wnt ligands acting in the canonical pathway to stimulate cell proliferation [15]. Cellular proliferation induced by Wnt3a has recently been attributed to activation of both the extracellular signal–regulated kinase (ERK) and Wnt pathways [26], both of which are implicated in hepatocarcinogenesis associated with hepatitis B or C virus infections, the major risk factors of HCC [27]. In addition, GPC3 has been reported to confer hepatocarcinogenesis through the interaction between insulin-like growth factor II and its receptor, with subsequent activation of the insulin growth factor signaling pathway [28]. The interaction between GPC3 and fibroblast growth factor basic has also been reported to modulate proliferation of HCC cells [23] and could, in part, mediate the oncogenic effect of sulfatase 2 in HCC [29]. Indeed, the overexpression of GPC3 in HCC patients has been positively correlated with fibroblast growth factor receptor 1, insulin-like growth factor 1 receptor, and nuclear localization of β-catenin [30]. The elevated expression of GPC3 in HCC and its potential involvement in multiple signaling pathways contributing to hepatocarcinogenesis together suggest that inhibition of GPC3 may offer a potent and molecularly targeted strategy for intervening with HCC progression. In this study, we confirmed that suppression of GPC3 in HCC cells by RNA interference led to inhibitory effects on cell growth and cell cycle progression. We additionally provide further mechanistic insights into another signaling pathway, the transforming growth factor β (TGF-β) pathway, which might be involved in GPC3-mediated functions in HCC. Taken together, our data provide support that molecular targeting of GPC3 at the translational level is an effective treatment strategy for HCC, which has heterogeneous pathology.

Materials and Methods

Cell Culture The human HCC cell lines Huh7 and HepG2 were cultured in Dulbecco modified Eagle medium and minimum essential medium, respectively (Gibco/Invitrogen, Carlsbad, CA). Media were supplemented with 10% fetal bovine serum (FBS), 100 U/ml penicillin, and 100 μg/ml streptomycin (Invitrogen). Cells were cultured at 37°C in a humidified atmosphere of 5% CO2.

Neoplasia Vol. 13, No. 8, 2011 The transfection reagent Lipofectamine RNAiMAX (Invitrogen) was used to transfect siRNA oligos into Huh7 and HepG2 cells, using the reverse transfection method according to the manufacturer’s instructions.

Cell Proliferation Assay Huh7 and HepG2 cells reverse transfected with siRNA-GPC3 (25 nM) or siRNA-N (25 nM) were seeded onto 96-well plates in triplicate wells (4000 cells/well) in culture medium without antibiotics. At the indicated time points, cell viability was determined by the CellTiter 96 AQueous One Solution Cell Proliferation Assay (Promega, Madison, WI) according to the manufacturer’s protocol. Optical density (OD) was read at 490 nm using a microplate reader (BioTek Instruments, Inc, Winooski, VT). Once cells have attached (approximately 4 hours after transfection), an OD value was obtained as the first time point. The background values (OD values from wells with only the AQueous One Solution) were subtracted from all data. Three independent experiments were done. For experiments with recombinant human (rh) TGF-β2 (R&D Systems, Minneapolis, MN), 1 or 5 ng/ml of TGF-β2 was added to the culture medium, and the effects on proliferation of Huh7 and HepG2 cells were determined as described previously.

Cell Cycle Analysis by Flow Cytometry Huh7 and HepG2 cells were synchronized at the G0/G1 phase by culturing in serum-free medium for 24 hours, then transfected with siRNA-N, siRNA-GPC3 alone, or siRNA-GPC3 with siRNA–TGFβ2 for 48 hours. Cells were then stimulated with 20% FBS for 24 hours and trypsinized, washed, and fixed in 70% ethanol for 30 minutes at 4°C. Fixed cells were stained with propidium iodide with RNaseA in 1× PBS for data acquisition using BD LSRII flow cytometer (BD BioSciences, Franklin Lakes, NJ). Data were analyzed by using FlowJo version 7.5.5 (Tree Star, Ashland, OR). Three independent experiments were done for each experiment set.

Protein Extraction and Western Blot Analysis Transfected cells were harvested and lysed in T-PER Tissue Protein Extraction Reagent (Thermo Scientific, Inc, Waltham, MA) for isolation of total cell lysates. For the isolation of nuclear and cytoplasmic fractions, NE-PER Nuclear and Cytoplasmic Extraction Reagents (Thermo Scientific) were used according to the manufacturer’s instructions. Western blot analysis was performed using standard protocols with specific antibodies against the antigens according to the manufacturer’s recommendations. Monoclonal antibody against human GPC3 (clone 1G12) was from BioMosaics, Inc (Burlington, VT); monoclonal antibody against TGF-β2 and Histone H3 were from AbCam (Cambridge, MA); polyclonal antibodies against PCNA and β-actin were from Santa Cruz Biotechnology (Santa Cruz, CA); and monoclonal antibody against cyclin D1 was from Thermo Scientific. All other antibodies were purchased from Cell Signaling Technology, Inc (Danvers, MA).

Oligonucleotide Microarray Analysis Small Interfering RNA Transfection In Vitro GPC3-specific small interfering RNA (siRNA-GPC3), TGF-β2–specific siRNA (siRNA-TGFβ2), and silencer-negative control siRNA (siRNA-N) were purchased from Ambion, Inc (Austin, TX). Sequences for GPC3 and TGF-β2–specific siRNAs used for the experiments were as follows: siRNA-GPC3: (GGCUCUGAAUCUUGGAAUUtt), s14750 siRNA–TGF-β2: (CACUCGAUAUGGACCAGUUtt), s14059

Genome-wide expression analysis was done using the Agilent 60-mer Oligonucleotide Two-Color Microarray System (Agilent Technologies, Inc, Santa Clara, CA). Total RNA was extracted from siRNA-N– or siRNA-GPC3–treated cells by using RNeasy mini kit (Qiagen, Valencia, CA). The integrity of the RNA was confirmed by the Agilent 2100 Bioanalyzer (Agilent Technologies). RNA samples with RIN number 8 or higher were used for further experiments. The RNA samples were amplified and labeled using the Agilent Quick Amp labeling kit

Neoplasia Vol. 13, No. 8, 2011 according to the manufacturer’s instructions. After fragmentation, 1650 ng of complementary RNA was hybridized onto a 4 × 44K oligonucleotide array at 65°C for 17 hours. The array was then washed and scanned using GenePix 4000A scanner (Agilent Technologies). Data were extracted by using Feature Extraction software and uploaded onto the Stanford Microarray Database (SMD; http://smd.stanford.edu). Supervised analysis using Significance Analysis of Microarrays was used to identify genes that are significantly differentially expressed between siRNA-GPC3–treated cells and siRNA-N–treated cells (a two-fold change cutoff was used). Significantly, differentially regulated pathways were identified using GeneSpring GX11 (Agilent Technologies).

Semiquantitative Real-time Polymerase Chain Reaction Total RNA was extracted using RNeasy Mini kit (Qiagen) and reverse transcribed using TaqMan Reverse Transcription Reagent (Applied Biosystems, Foster City, CA). Semiquantitative real-time polymerase chain reaction (PCR) was carried out, and data were analyzed using a MX3000P Real-time PCR machine (Stratagene, La Jolla, CA). Primers and probe reagents for human GPC3 (assay no. Hs00170471_m1), TGF-β1 (assay no. Hs00998129_m1), TGF-β2 (assay no. Hs00234244_m1), TGF-β3 (assay no. Hs01086000_m1), and 18S rRNA (normalization control; assay no. 4333760F) were purchased as Pre-Developed TaqMan Gene Expression Assay reagents from Applied Biosystems. Transcript quantification was performed in at least duplicate for every sample. The amount of each target gene was normalized with 18S rRNA to control for RNA amount variation.

Senescence-Associated β-Gal Staining

Senescence β-galactosidase (SA-β-Gal) staining was performed by using the staining kit according to the manufacturer’s instructions (Cell Signaling Technology, Inc, Danvers, MA). Briefly, cells were transfected with siRNA-GPC3 or incubated with rhTGF-β2 (1 or 5 ng/ml) for 48 hours before fixing and staining. After SA-β-Gal staining, cells were counterstained with Nuclear Fast Red Solution (Electron Microscopy Sciences, Hatfield, PA), and images were acquired using a fluorescence microscope (Nikon, Melville, NY).

Lentiviral Labeling of Cell Lines and Cell Sorting Huh7 and HepG2 cells were transduced with self-inactivating lentivirus carrying an ubiquitin promoter driving a trifusion reporter gene, which harbors a bioluminescence (firefly luciferase [Fluc]), a fluorescence (mrfp or egfp), and a positron emission tomography reporter gene (ttk) at a multiplicity of infection of 5 as reported previously [31]. Stable expressors were isolated by sorting at 45 pound-force per square inch (BD FACS Aria II; BD BioSciences).

Animal Studies Animal experiments were approved by the Administrative Panel on Laboratory Animal Care of Stanford University. Orthotopic liver tumor models of Huh7 and HepG2 cells (stably expressing the trifusion reporter gene) were established in nude mice as reported previously with some modifications [32]. Briefly, ∼1 × 107 cells in 200 μl of culture medium were injected subcutaneously into the right flank of nu/ nu nude mice (4 weeks old; Charles River Laboratories, Wilmington, MA). Tumor development was monitored daily. Once the subcutaneous tumor reached 1 cm in diameter, it was removed and cut into ∼1- to 2-mm3 cubes, which were then surgically implanted into the left lobe of liver in another group of nude mice (6 weeks old). The mice were given intraperitoneal injections of 25 nM siRNA-GPC3 in 100 μl

GPC3 Suppression Inhibits HCC Growth via TGF-β2

Sun et al.

737

of PBS or siRNA-N as control (five mice each) every 3 days. Tumor growth was monitored once a week for 7 weeks using the Xenogen IVIS in vivo imaging system (Caliper Life Sciences, Hopkinton, CA). Growth curves were plotted using average bioluminescence within each group.

Immunohistochemistry for Cell Proliferation and Apoptosis Immunostaining was performed as reported previously [32]. Briefly, paraffin slides were stained with anti–proliferating cell nuclear antigen (PCNA; Santa Cruz Biotechnology) or anti–TGF-β2 monoclonal antibody (AbCam) at 4°C overnight followed by the avidin-biotinperoxidase protocol. Terminal deoxynucleotidyl transferase biotin– dUTP nick end labeling (TUNEL) assay was done according to the manufacturer’s protocol (Roche Applied Science, Indianapolis, IN).

Statistical Analysis The statistical analyses were done using the SPSS version 15.0 software package (SPSS, Inc, Chicago, IL). Statistical significance was determined by independent-samples t test. P < .05 and P < .01 were considered statistically significant and highly significant, respectively.

Results

Suppression of GPC3 Inhibits In Vitro Proliferation and Cell Cycle Progression in Huh7 and HepG2 Cells We transfected two tumorigenic hepatoma cell lines (Huh7 and HepG2) that overexpress GPC3 protein (based on Western blot results) with 25 nM of a GPC3-specific siRNA (siRNA-GPC3) and observed greatly reduced expression of glycanated and core GPC3 protein in both cell lines compared with cells transfected with vehicle control or with silencer-negative siRNA (siRNA-N; Figure 1A). Transfection with siRNA-GPC3 also reduced proliferation of Huh7 cells (P < .001 from 72 hours) and HepG2 cells (P < .001 from 48 hours) when assessed daily during a 5-day period (Figure 1B). To determine whether the growth-inhibitory effects of GPC3 suppression could result from changes in the cell cycle, we measured cellular DNA content by flow cytometry after cell synchronization by serum starvation, followed by transfection with siRNA-GPC3 for 48 hours and restimulation with 10% FBS for 24 hours. Suppression of GPC3 caused a significant increase in G1 peak in HepG2 cells (P = .04 compared with siRNA-N), together with a decrease in S-phase cells (P = .039; Figure 1C). The expression of cell cycle regulators, cyclin A and cyclin D1, were also downregulated by siRNA-GPC3 in HepG2 cells (Figure 1C). These changes in the cell cycle were less obvious in the Huh7 cells, which showed a small increase in the G0/G1 peak (P = .31 compared with siRNA-N) but an insignificant change in the S peaks (P = .90; Figure 1D). In Huh7 cells, the levels of cyclin A and cyclin D1 were also reduced to a lesser extent than in HepG2 cells (Figure 1D). Thus, the effect of GPC3 suppression on cell growth and cell cycle progression is more marked in HepG2 cells, which have higher levels of GPC3 than Huh7 cells.

Identification of TGF-β2 as an Important Modulator of GPC3-Mediated Signaling To characterize the signaling pathways and molecular changes regulated by GPC3 in HCC cells, we studied the gene expression changes induced by siRNA-GPC3 compared with their siRNA-N–transfected controls. Pathway analysis using GeneSpring GX 11 identified five signaling pathways that are significantly regulated by GPC3 suppression: nuclear

738

GPC3 Suppression Inhibits HCC Growth via TGF-β2

Sun et al.

Neoplasia Vol. 13, No. 8, 2011

Figure 1. Suppression of GPC3 by siRNA-GPC3 inhibits in vitro cell proliferation and cell cycle progression of Huh7 and HepG2 cells. (A) Transfection of GPC3-specific siRNA (siRNA-GPC3) efficiently suppressed expression of core and glycanated forms of GPC3 protein in Huh7 and HepG2 cells. (B) Growth rates of siRNA-GPC3–transfected and control cells were determined using the proliferation assay described under Materials and Methods. Data were obtained from three independent experiments, with triplicates in each experiment. (C, D) Cell cycle analysis by flow cytometry showed that siRNA-GPC3 caused significant cell cycle arrest at the G1 phase in HepG2 cells (C) but only an insignificant change in Huh7 cells (D). Western blots with antibodies against cyclin A and cyclin D1 support these observations. Only P values for comparing G1 cell populations are shown. siRNA-G indicates cells transfected with siRNA-GPC3; siRNA-N, cells transfected with siRNA-N; V, vehicle control. Number of independent experiments = 3.

Neoplasia Vol. 13, No. 8, 2011

GPC3 Suppression Inhibits HCC Growth via TGF-β2

factor κB (P = 3.73e−6), TGFBR (P = 4.57e−5), EGFR1 (P = 1.43e−4), IL-6 (P = 6.43e−4), and Wnt (P = .008). Using Significance Analysis of Microarrays, biologically relevant genes that are commonly, significantly upregulated or downregulated in both HepG2 and Huh7 cell lines were identified using a two-fold change cutoff and P < .0001 (Table 1 and Figure 2A). Biologic functions of encoded proteins were compiled from the SOURCE database available at the SMD. A member of the TGFBR pathway, TGF-β2 (but not TGF-β1 or TGF-β3), was significantly upregulated by GPC3 suppression in both Huh7 and HepG2 cells (P < .05, siRNA-N vs siRNA-GPC3; Figure 2B). The up-regulation of TGF-β2 by siRNA-GPC3 was independently validated using semiquantitative RT-PCR (Figure 2B) and Western blot analysis (Figure 2C) to confirm the negative correlation at both transcript and protein levels, respectively. Two additional GPC3-specific siRNAs (GGGAACCACUUUCUUAUUUtt; GACGUGACCUGAAAGUAUUtt) were used to verify this observation (data not shown). Further analysis of GPC3 and TGF-β2 protein expression in a panel of nine HCC cell lines confirmed this negative correlation: cell lines expressing GPC3 have lower levels of TGF-β2, whereas cell lines with undetectable GPC3 have higher levels of TGF-β2 (Figure 2D). Our results suggest a close association between GPC3 and TGF-β2 in the TGFBR signaling pathway.

TGF-β2 Suppresses Cell Proliferation by Activating R-SMADs in Huh7 and HepG2 Cells To determine whether the growth-inhibitory effects of GPC3 suppression might be partially mediated by TGF-β2, we next looked at the effects of rhTGF-β2 on HCC cell proliferation. Cells were cultured with or without rhTGF-β2 and cell growth was assessed daily for 5 days. We found that rhTGF-β2 significantly inhibited cell proliferation in both Huh7 and HepG2 cells (P < .01 for both cell lines; Figure 3A). Important components of the TGF-β pathway, SMAD2 and SMAD3, were hyperphosphorylated on addition of rhTGF-β2 together with nuclear translocation of SMAD 6 (Figure 3B). Our results imply that TGF-β2 negatively regulates HCC cell proliferation

Sun et al.

739

through the activation of R-SMADs, allowing transduction of extracellular TGF-β2 superfamily ligand signaling from cell membrane– bound TGF-β receptors into the nucleus where they activate the transcription of TGF-β target genes.

TGF-β2 Suppresses Cell Cycle Progression and Induces Replicative Senescence in Huh7 and HepG2 Cells To determine whether the cell cycle effects of GPC3 suppression might be partially mediated by TGF-β2, we looked at the effect of rhTGF-β2 on cell cycle progression of HCC cells. Huh7 and HepG2 cells were grown in culture medium supplemented with rhTGF-β2 (1 or 5 ng/ml) for a period of 48 hours before flow cytometry analyses. Similar to the effects of GPC3 suppression (Figure 1C ), significant increases in the G1 peak (P = .016 for Huh7 cells and P < .001 for HepG2 cells), together with a decrease in S- and G2-phase cells were observed in both cell lines compared with control cells not treated with rhTGF-β2 (P < .001 and P = .034 for S and G2 phases, respectively, for HepG2 cells; P = .017 and P = .004 for S and G2 phases, respectively, for Huh7 cells) (Figure 4A). Western blot analysis further showed that rhTGF-β2 decreased the expression of cell cycle regulators cyclin A and cyclin D1 (Figure 4B), as observed with GPC3 suppression (Figure 1C ). In addition, phosphorylation of retinoblastoma (Rb) protein was decreased, whereas the expression of p15Ink4b and p21Cip1 were increased in both Huh7 and HepG2 cells, indicating a G1 cell cycle arrest (Figure 4B). The addition of TGF-β2 also increased the apoptotic response, demonstrated by the reduced expression of antiapoptotic proteins (Bcl-xL and Mcl-1) in both cell lines, and the reduced expression of Bcl-2 in HepG2 cells only (Huh7 cells do not express Bcl-2) (Figure 4B). Consistent with a recent report that TGF-β1 induces p15Ink4b- and p21Cip1-dependent senescence in HCC cells [33], we also observed that addition of rhTGF-β2 induced cellular senescence in both Huh7 and HepG2 cells as shown by enhanced accumulation of SA-β-Gal (Figure 4C). Taken together, these results suggest that TGF-β2 negatively mediates cell cycle progression through modulation of various

Table 1. Genes Commonly Significantly Upregulated or Downregulated by GPC3 Suppression in HepG2 and Huh7 Cells. Gene Symbol

Gene Name

Biologic Function of Encoded Protein

Fold Change (HepG2)

Fold Change (Huh7)

NUDT15

2.18

16.45

2.65

TGFB2

Transforming growth factor, beta 2

10.37

3.61

XPR1 PANK3

Xenotropic and polytropic retrovirus receptor Pantothenate kinase 3

6.59 4.86

2.77 2.27

GDF11

Growth differentiation factor 11

3.52

2.15

TOP1MT

Topoisomerase (DNA) I, mitochondrial

3.15

2.18

GPC3

Glypican 3

0.11

0.14

RORA

Retinoic acid receptor–related orphan receptor A

May have a role in DNA synthesis and cell cycle progression through interaction with PCNA. Androgen receptor–interacting nuclear protein kinase, to negatively regulate apoptosis by promoting FADD phosphorylation; enhances androgen receptor–mediated transcription. Member of the transforming growth factor beta (TGFB) family of cytokines, which are multifunctional peptides that regulate proliferation, differentiation, adhesion, migration, and other functions in many cell types. Expressed in fetal liver; may function in G protein–coupled signal transduction. Key regulatory enzyme in the biosynthesis of coenzyme A (CoA) in bacteria and mammalian cells; expressed most abundantly in the liver. Member of the bone morphogenetic protein (BMP) family and the TGF-beta superfamily. Members of this family are regulators of cell growth and differentiation in both embryonic and adult tissues. Enzyme that controls and alters the topologic states of DNA during transcription; highly expressed in fetal liver. Member of the glypican-related integral membrane proteoglycan family (GRIPS); may play a role in the control of cell division, growth regulation, and tumor predisposition. Member of the NR1 subfamily of nuclear hormone receptors; enhances expression of hormone response genes. Shown to interact with NM23-2, a nucleoside diphosphate kinase involved in organogenesis and differentiation, as well as with NM23-1, the product of a tumor metastasis suppressor candidate gene.

31.34

HIPK3

Nudix (nucleoside diphosphate linked moiety X)-type motif 15 Homeodomain interacting protein kinase 3

0.28

0.48

Fold change is expressed as the ratio of mean abundance of transcript in siRNA-GPC3–treated cells compared with siRNA-N– (negative control) treated cells.

740

GPC3 Suppression Inhibits HCC Growth via TGF-β2

Sun et al.

Neoplasia Vol. 13, No. 8, 2011

Figure 2. Identification of TGF-β2 as an important mediator in GPC3 signaling by oligonucleotide microarray analysis. (A) Heat map showing nine significantly upregulated or downregulated genes associated with GPC3 suppression in both Huh7 and HepG2 cells. Rows represent individual genes and columns represent individual cell sample. In each sample, the log2 ratio of abundance of transcripts of each gene relative to its mean abundance across all samples is depicted according to the color scale shown at the bottom left. (B) Transfection of Huh7 and HepG2 cells with siRNA-GPC3 upregulated TGF-β2 but not TGF-β1 or TGF-β3 at the transcriptional level as measured by quantitative PCR. Fold difference indicates the ratio of relative RNA quantity (siRNA-GPC3/siRNA-N); *P < .05. Number of independent experiments = 3. (C) Validation of increased protein expression of TGF-β2 in both cell lines transfected with siRNA-GPC3. (D) Protein expression levels of GPC3 and TGF-β2 in a panel of HCC cell lines as detected by Western blot analysis. siRNA-G indicates cells transfected with siRNA-GPC3; siRNA-N, cells transfected with siRNA-N; V, vehicle control.

Neoplasia Vol. 13, No. 8, 2011

GPC3 Suppression Inhibits HCC Growth via TGF-β2

Sun et al.

741

Figure 3. Human recombinant TGF-β2 inhibits cell proliferation and activates R-SMAD in HepG2 and Huh7 cells. (A) TGF-β2 suppressed proliferation of HCC cells as shown by cell proliferation assay. Cells were cultured in media with different concentrations of rhTGF-β2 (1 or 5 ng/ml) as described in Materials and Methods. Data were obtained from three independent experiments, with triplicates in each experiment. (B) Effects of TGF-β2 on phosphorylation of R-SMAD2/3 and nuclear translocation of SMAD6 as shown by Western blot analysis. Histone H3 was used as the loading control for nuclear proteins. Number of independent experiments = 3.

cell cycle regulators, thereby contributing to G1 arrest, apoptosis, and senescence.

TGF-β2 siRNA Partially Reverses Growth Inhibition, Cell Cycle Arrest, and Replicative Senescence Caused by siRNA-GPC3 We next cotransfected siRNA–TGF-β2 with siRNA-GPC3 in HCC cells to interfere with the rise in TGF-β2 levels caused by siRNAGPC3 alone and to observe the resulting effects on cell proliferation and cell cycle progression. The cotransfection of siRNA–TGF-β2 with siRNA-GPC3 partially reversed the growth-inhibitory effects of GPC3 suppression in HepG2 cells but had negligible effect in Huh7 cells (Figure 5A). When the increase in TGF-β2 levels was reduced by siRNA–TGF-β2, cells proliferated faster than when treated with siRNAGPC3 alone. These effects of cell proliferation were also reflected by the levels of SMAD2 and SMAD3 phosphorylation: siRNA-GPC3 alone caused an increase in SMAD2 and SMAD3 phosphorylation in both Huh7 and HepG2 cells; this increase was partially attenuated when cells were cotransfected with siRNA–TGF-β2 (Figure 5B). Thus, TGF-β2 may partially mediate the effect of GPC3 on HCC cell proliferation. Cell cycle arrest at the G1 phase caused by siRNA-GPC3 was partially reversed when cells were cotransfected with siRNA–TGF-β2. In HepG2 cells, cotransfection with siRNA–TGF-β2 decreased the G1 peak when

compared with siRNA-GPC3–treated controls; a similar but less pronounced effect was observed in Huh7 cells (Figure 6A). Cotransfection with siRNA–TGF-β2 also negated the decrease in cyclin A and cyclin D1 levels caused by GPC3 siRNA in HepG2 cells (Figure 6B). In addition, cotransfection with TGF-β2 siRNA decreased the levels of p15Ink4b and p21Cip1 while increasing the levels of antiapoptotic proteins Bcl-2, Bcl-xL, and Mcl-1 that were regulated by siRNA-GPC3 in HepG2 cells (Figure 6B). In Huh7 cells, cotransfection with siRNA– TGF-β2 caused little effect on the levels of cyclin A, cyclin D1, p15Ink4b, or p21Cip1 but instead increased the phosphorylation of Rb and the levels of Bcl-xL and Mcl-1 that were reduced by siRNA-GPC3 (Figure 6B). Transfection with siRNA-GPC3 alone caused marked accumulations of SA-β-Gal in both cell lines, which were significantly reduced when both cell lines were cotransfected with siRNA–TGF-β2 (Figure 6C). These results showed that cotransfection with siRNA–TGF-β2 can partially reverse the cell cycle arrest and replicative senescence induced by GPC3 suppression, implying that TGF-β2 may partially mediate the effects of GPC3 on regulation of cell cycle and senescence.

Suppression of GPC3 Delays HCC Xenograft Growth and Upregulates Cytoplasmic TGF-β2 In Vivo Finally, we generated orthotopic tumors of Huh7 and HepG2 cells in nude mice to confirm the growth-inhibitory effects of siRNA-GPC3

742

GPC3 Suppression Inhibits HCC Growth via TGF-β2

Sun et al.

in vivo. Tumor-bearing mice were given intraperitoneal injections of 25 nM siRNA-GPC3 (n = 5) or siRNA-N (n = 5) in 100 μl of PBS every 3 days until the tumors reached the criteria for euthanasia. Tumor growth was monitored by noninvasive in vivo luciferase imaging. In the HepG2 group, all mice treated with siRNA-GPC3 had a significantly

Neoplasia Vol. 13, No. 8, 2011 smaller tumor size compared with that observed in mice treated with siRNA-N (P < .05 from week 3 onward) (Figure 7A). In the Huh7 group, mice treated with siRNA-GPC3 had smaller tumor volume when compared with mice treated with siRNA-N, although statistical significance was not reached (P > .05; Figure 7B). Concomitantly, tumors treated with

Figure 4. Human recombinant TGF-β2 inhibits cell cycle progression and induces replicative senescence in HCC cells. (A) Addition of rhTGF-β2 to the culture media inhibited cell cycle progression through G1 arrest in both HepG2 and Huh7 cells. Only P values for comparing G1 cell populations are shown. (B) Addition of rhTGF-β2 to culture media caused changes in the expression of cell cycle regulators and antiapoptotic proteins as detected by Western blot analysis. (C) Addition of rhTGF-β2 to culture media also induced accumulation of SA-β-Gal staining in Huh7 and HepG2 cells, indicative of replicative senescence. *P < .05 versus vehicle control group. Number of independent experiments = 3.

Neoplasia Vol. 13, No. 8, 2011

GPC3 Suppression Inhibits HCC Growth via TGF-β2

Sun et al.

743

Figure 5. Transfection of siRNA–TGF-β2 partially reverses the growth inhibition caused by GPC3 suppression. (A) Cotransfection of siRNA–TGF-β2 with siRNA-GPC3 partially reversed the growth inhibition caused by siRNA-GPC3 alone in HepG2 cells, with negligible effect in Huh7 cells. (B) Cotransfection of siRNA–TGF-β2 siRNA with siRNA-GPC3 partially downregulated the activation of SMAD signaling pathway resulting from GPC3 suppression. siRNA G + T indicates cells cotransfected with siRNA-GPC3 and siRNA–TGF-β2; siRNA-G, cells transfected with siRNA-GPC3; siRNA-N, cells transfected with silencer negative control siRNA; siRNA-T, cells transfected with siRNA–TGF-β2.

siRNA-GPC3 had markedly lower expression of GPC3 but enhanced expression of cytoplasmic TGF-β2 in the tumor cells (Figure 7C). In addition, we determined the effects of GPC3 suppression on cell proliferation by PCNA immunohistochemistry and on apoptosis by the TUNEL assay (Figure 7C). The siRNA-GPC3–treated tumors expressed reduced levels of PCNA and contained more apoptotic nuclei in both Huh7 and HepG2 xenografts, suggesting that GPC3 suppression inhibited xenograft growth by reducing cell proliferation while inducing apoptosis in vivo. These effects positively correlated with up-regulation of TGF-β2.

Discussion Our previous gene expression study of human HCC identified GPC3 as the most highly overexpressed membrane-bound protein in HCC compared with nontumor liver and the second most highly overexpressed secretory protein (after AFP) [34]. Until lately, much of the focus on GPC3 has been on its diagnostic potential. In this study, we validated the therapeutic potential of GPC3 in HCC and

reported the involvement of TGF-β2 in GPC3-mediated signaling in HCC cells. Specifically, we showed that suppression of GPC3 in HCC cells enhanced TGF-β2 expression and signaling, which inhibited cell proliferation and cell cycle progression, and induced replicative senescence. GPC3 is a rational target for the treatment of HCC because it is predominantly expressed in HCC tumors compared with its adjacent nontumor or cirrhotic tissues [5], implying specificity. In addition, it potentially regulates multiple pathways involved in hepatocarginogensis [20,22–24], implying a broad spectrum of activity. Several recent studies have investigated different approaches of targeting GPC3, such as the use of anti-GPC3 monoclonal antibody [35] and HLA-A2– and -A24–restricted GPC3-derived peptide for the immunotherapy for HCC [36]. These approaches are based on the induction of antibodydependent cellular cytotoxicity and/or complement-dependent cytotoxicity and on the peptide induction of cytotoxic T lymphocytes, respectively, which, in turn, reduced HCC tumor mass. A mutated, soluble form of GPC3 was also reported to inhibit Wnt signaling in HCC cells, leading to antitumor effects [37]. Our study demonstrates

744

GPC3 Suppression Inhibits HCC Growth via TGF-β2

Sun et al.

Neoplasia Vol. 13, No. 8, 2011

Figure 6. Transfection of siRNA–TGF-β2 partially reverses the cell cycle arrest caused by GPC3 suppression. (A) Cell cycle analysis by flow cytometry indicated that siRNA–TGF-β2 partially reversed the G1 arrest induced by GPC3 suppression in HepG2 cells and, to a lesser extent, in Huh7 cells. (B) Cotransfection of siRNA–TGF-β2 with siRNA-GPC3 correspondingly showed partial reversal of expression levels of cell cycle regulators and antiapoptotic proteins induced by siRNA-GPC3 alone. (C) Cotransfection of siRNA–TGF-β2 and siRNA-GPC3 also partially reduced the accumulation of SA-β-Gal caused by GPC3 suppression. siRNA-G + T indicates cells transfected with siRNAGPC3 and siRNA–TGF-β2; siRNA-G, cells transfected with siRNA-GPC3; siRNA-N, cells transfected with silencer negative control siRNA; siRNA-T, cells transfected with siRNA–TGF-β2; V, vehicle control. Number of independent experiments = 3.

Neoplasia Vol. 13, No. 8, 2011

GPC3 Suppression Inhibits HCC Growth via TGF-β2

Sun et al.

745

Figure 7. Suppression of GPC3 in vivo delays tumor growth and upregulates cytoplasmic TGF-β2. (A, B) Orthotopic liver tumor models of Huh7 and HepG2 cells stably expressing a trifusion reporter gene were given intraperitoneal injections of siRNA-GPC3 (25 nM), and tumor growth was monitored weekly using the Xenogen IVIS 100 imaging system. Representative bioluminescence images of tumors are shown at week 4 after siRNA treatment for HepG2 xenografts and at week 3 after siRNA treatment for Huh7 xenografts. *P < .05, siRNA-N versus siRNA-GPC3 group. (C) Suppression of GPC3 in vivo also enhanced cytoplasmic TGF-β2 expression, reduced tumor proliferation (reduced PCNA levels), and induced apoptosis (TUNEL assay) in both Huh7 and HepG2 xenografts. Representative images are shown. Scale bar, 100 μm.

746

GPC3 Suppression Inhibits HCC Growth via TGF-β2

Sun et al.

for the first time that targeting GPC3 at the translational level in GPC3positive HCC cells activates TGF-β signaling, which, in turn, partially mediates the antitumor effects of GPC3 suppression in HCC cells in vitro and in vivo. The TGF superfamily, including TGF-β, activin, and BMPs, modulates many cellular responses, such as cell division, differentiation, and cell fate decision [38]. Three different isoforms of TGF-β, TGF-β1, TGF-β2, and TGF-β3, have similar but not identical biologic activities [39]. TGF-β signaling is mediated through the binding of TGF-β with TGF-β receptors (TGFBR), which, in turn, recruit and phosphorylate downstream receptor–regulated SMADs (R-SMADs), SMAD2 or SMAD3. Activation of R-SMADs may form signaling complexes with SMAD4 and translocate to the nucleus eliciting tumor suppressive or oncogenic effects [40]. Whereas TGF-β/SMAD signaling can be both promoting and suppressing, its role in HCC progression is not completely understood [41]. Recently, TGF-β1 was reported to induce cellular senescence and inhibit tumor growth in HCC cell lines by a p53-independent and p21Cip1-dependent pathway [33]. Our results indicate that the antiproliferative effects of GPC3 in HCC cells are partially mediated by TGF-β signaling. Suppression of GPC3 in HCC cells overexpressing GPC3 inhibited cell proliferation associated with an increase in phosphorylation of SMAD2/3 and also arrested cell cycle progression at the G1 phase, associated with down-regulation of cyclin A and cyclin D1, accumulation of p15Ink4b and p21Cip1, and down-regulation of Rb phosphorylation. GPC3 suppression also caused an accumulation of SA-β-Gal, an indicator of replicative senescence. These cellular changes were recapitulated by the addition of rhTGF-β2 to the HCC cells in culture, confirming the involvement of TGF-β2 in cell cycle arrest and replicative senescence. Moreover, the cotransfection of siRNAs against GPC3 and TGF-β2 partially reversed these effects on cell proliferation, cell cycle progression, replicative senescence, and the associated molecular changes. We further observed an inverse correlation between the expression of GPC3 and TGF-β2, as GPC3 suppression upregulated TGF-β2 at both transcriptional and translational levels. Thus, GPC3 may inversely regulate TGF-β2, which, in turn, partially mediates the subsequent cellular and molecular changes observed in HCC cells on GPC3 suppression. The two HCC cell lines studied, HepG2 and Huh7, responded to different extents toward GPC3 suppression, probably because of the different genetic contexts between the two cell lines. The poorer response of Huh7 cells may be explained by the following reasons. First, Huh7 cells have shorter cell doubling time (24 hours) than HepG2 cells (33 h) and have a lower percentage of cells in the G1 phase (56.39% ± 2.47%) than HepG2 cells (64.57% ± 2.84%) when cultured in Dulbecco modified Eagle medium supplemented with 10% FBS (P = .01). The shorter doubling time of Huh7 cells may allow the cells to escape faster from the effect of transient siRNA knockdown. Second, we observed that Wnt signaling was inactivated in HepG2 cells but not in Huh7 cells after GPC3 suppression (Figure W1, A-C). Cooperatively, the regulation of Wnt and TGF-β signaling pathways by GPC3 in HepG2 cells may lead to greater inhibitory effects on cell growth and cell cycle progression in this cell line. Third, in Huh7 cells, the ERK pathway might be the predominant pathway regulating proliferation that is affected by GPC3 suppression (Figure W1D), and therefore, we observed a negligible effect of TGF-β2 siRNA on reversing cell growth inhibition caused by GPC3 suppression. Fourth, Huh7 cells have endogenous TGF-β2 expression but HepG2 cells do not. The endogenous expression of TGF-β2 may mask the effect contributed by the low and transient increase in TGF-β2 levels caused by GPC3 suppression.

Neoplasia Vol. 13, No. 8, 2011 In summary, we have shown that suppression of GPC3 activates the TGF-β signaling pathway in HCC cells, eventually inhibiting cell proliferation, arresting cell cycle progression at the G1 phase, and inducing replicative senescence. The extent and biologic effects of GPC3 suppression vary, depending largely on the genetic context of the cells, which reflects the heterogeneity of HCC tumors. Combined with previous reports that interference with GPC3 expression additionally regulates at least two major signaling pathways (Wnt/β-catenin and ERK/ MAPK) in HCC, we suggest that interfering with GPC3 expression and function exert broad-spectrum biologic effects that might be beneficial for the treatment of the heterogeneous subtypes of HCC. Targeting GPC3 in HCC is widely applicable because it is overexpressed in a large percentage (>60%) of HCC patients [4]. Further studies of the mechanisms by which GPC3 regulates multiple signaling pathways may reveal other points of intervention for this important target. Acknowledgments The authors thank Xinrui Yan, Department of Radiology, Stanford University, for labeling the Huh7 and HepG2 cells with trifusion reporter lentivirus. References [1] Parkin DM (2001). Global cancer statistics in the year 2000. Lancet Oncol 2, 533–543. [2] Hayashi PH and Di Bisceglie AM (2006). The progression of hepatitis B- and C-infections to chronic liver disease and hepatocellular carcinoma: presentation, diagnosis, screening, prevention, and treatment of hepatocellular carcinoma. Infect Dis Clin North Am 20, 1–25. [3] Poon RT, Fan ST, Lo CM, Ng IO, Liu CL, Lam CM, and Wong J (2001). Improving survival results after resection of hepatocellular carcinoma: a prospective study of 377 patients over 10 years. Ann Surg 234, 63–70. [4] Baumhoer D, Tornillo L, Stadlmann S, Roncalli M, Diamantis EK, and Terracciano LM (2008). Glypican 3 expression in human nonneoplastic, preneoplastic, and neoplastic tissues: a tissue microarray analysis of 4,387 tissue samples. Am J Clin Pathol 129, 899–906. [5] Libbrecht L, Severi T, Cassiman D, Vander Borght S, Pirenne J, Nevens F, Verslype C, van Pelt J, and Roskams T (2006). Glypican-3 expression distinguishes small hepatocellular carcinomas from cirrhosis, dysplastic nodules, and focal nodular hyperplasia-like nodules. Am J Surg Pathol 30, 1405–1411. [6] Ligato S, Mandich D, and Cartun RW (2008). Utility of glypican-3 in differentiating hepatocellular carcinoma from other primary and metastatic lesions in FNA of the liver: an immunocytochemical study. Mod Pathol 21, 626–631. [7] Man XB, Tang L, Zhang BH, Li SJ, Qiu XH, Wu MC, and Wang HY (2005). Upregulation of glypican-3 expression in hepatocellular carcinoma but downregulation in cholangiocarcinoma indicates its differential diagnosis value in primary liver cancers. Liver Int 25, 962–966. [8] Wang XY, Degos F, Dubois S, Tessiore S, Allegretta M, Guttmann RD, Jothy S, Belghiti J, Bedossa P, and Paradis V (2006). Glypican-3 expression in hepatocellular tumors: diagnostic value for preneoplastic lesions and hepatocellular carcinomas. Hum Pathol 37, 1435–1441. [9] Zhu ZW, Friess H, Wang L, Abou-Shady M, Zimmermann A, Lander AD, Korc M, Kleeff J, and Buchler MW (2001). Enhanced glypican-3 expression differentiates the majority of hepatocellular carcinomas from benign hepatic disorders. Gut 48, 558–564. [10] Capurro M, Wanless IR, Sherman M, Deboer G, Shi W, Miyoshi E, and Filmus J (2003). Glypican-3: a novel serum and histochemical marker for hepatocellular carcinoma. Gastroenterology 125, 89–97. [11] Filmus J and Capurro M (2004). Glypican-3 and alphafetoprotein as diagnostic tests for hepatocellular carcinoma. Mol Diagn 8, 207–212. [12] Hippo Y, Watanabe K, Watanabe A, Midorikawa Y, Yamamoto S, Ihara S, Tokita S, Iwanari H, Ito Y, Nakano K, et al. (2004). Identification of soluble NH2-terminal fragment of glypican-3 as a serological marker for early-stage hepatocellular carcinoma. Cancer Res 64, 2418–2423. [13] Capurro MI, Xiang YY, Lobe C, and Filmus J (2005). Glypican-3 promotes the growth of hepatocellular carcinoma by stimulating canonical wnt signaling. Cancer Res 65, 6245–6254.

Neoplasia Vol. 13, No. 8, 2011 [14] Paul S and Dey A (2008). Wnt signaling and cancer development: therapeutic implication. Neoplasma 55, 165–176. [15] Takigawa Y and Brown AM (2008). Wnt signaling in liver cancer. Curr Drug Targets 9, 1013–1024. [16] Pilia G, Hughes-Benzie RM, MacKenzie A, Baybayan P, Chen EY, Huber R, Neri G, Cao A, Forabosco A, and Schlessinger D (1996). Mutations in GPC3, a glypican gene, cause the Simpson-Golabi-Behmel overgrowth syndrome. Nat Genet 12, 241–247. [17] Filmus J, Shi W, Wong ZM, and Wong MJ (1995). Identification of a new membrane-bound heparan sulphate proteoglycan. Biochem J 311(pt 2), 561–565. [18] Cano-Gauci DF, Song HH, Yang H, McKerlie C, Choo B, Shi W, Pullano R, Piscione TD, Grisaru S, Soon S, et al. (1999). Glypican-3–deficient mice exhibit developmental overgrowth and some of the abnormalities typical of SimpsonGolabi-Behmel syndrome. J Cell Biol 146, 255–264. [19] Filmus J (2001). Glypicans in growth control and cancer. Glycobiology 11, 19R–23R. [20] Filmus J and Capurro M (2008). The role of glypican-3 in the regulation of body size and cancer. Cell Cycle 7, 2787–2790. [21] Gonzalez AD, Kaya M, Shi W, Song H, Testa JR, Penn LZ, and Filmus J (1998). OCI-5/GPC3, a glypican encoded by a gene that is mutated in the SimpsonGolabi-Behmel overgrowth syndrome, induces apoptosis in a cell line–specific manner. J Cell Biol 141, 1407–1414. [22] Capurro MI, Xu P, Shi W, Li F, Jia A, and Filmus J (2008). Glypican-3 inhibits hedgehog signaling during development by competing with patched for hedgehog binding. Dev Cell 14, 700–711. [23] Midorikawa Y, Ishikawa S, Iwanari H, Imamura T, Sakamoto H, Miyazono K, Kodama T, Makuuchi M, and Aburatani H (2003). Glypican-3, overexpressed in hepatocellular carcinoma, modulates FGF2 and BMP-7 signaling. Int J Cancer 103, 455–465. [24] Song HH, Shi W, Xiang YY, and Filmus J (2005). The loss of glypican-3 induces alterations in wnt signaling. J Biol Chem 280, 2116–2125. [25] Filmus J, Capurro M, and Rast J (2008). Glypicans. Genome Biol 9, 224. [26] Yun MS, Kim SE, Jeon SH, Lee JS, and Choi KY (2005). Both ERK and Wnt/ β-catenin pathways are involved in Wnt3a-induced proliferation. J Cell Sci 118, 313–322. [27] Aravalli RN, Steer CJ, and Cressman EN (2008). Molecular mechanisms of hepatocellular carcinoma. Hepatology 48, 2047–2063. [28] Cheng W, Tseng CJ, Lin TT, Cheng I, Pan HW, Hsu HC, and Lee YM (2008). Glypican-3–mediated oncogenesis involves the insulin-like growth factor-signaling pathway. Carcinogenesis 29, 1319–1326.

GPC3 Suppression Inhibits HCC Growth via TGF-β2

Sun et al.

747

[29] Lai JP, Sandhu DS, Yu C, Han T, Moser CD, Jackson KK, Guerrero RB, Aderca I, Isomoto H, Garrity-Park MM, et al. (2008). Sulfatase 2 up-regulates glypican 3, promotes fibroblast growth factor signaling, and decreases survival in hepatocellular carcinoma. Hepatology 47, 1211–1222. [30] Akutsu N, Yamamoto H, Sasaki S, Taniguchi H, Arimura Y, Imai K, and Shinomura Y (2010). Association of glypican-3 expression with growth signaling molecules in hepatocellular carcinoma. World J Gastroenterol 16, 3521–3528. [31] Ray P, De A, Min JJ, Tsien RY, and Gambhir SS (2004). Imaging tri-fusion multimodality reporter gene expression in living subjects. Cancer Res 64, 1323–1330. [32] Sun CK, Man K, Ng KT, Ho JW, Lim ZX, Cheng Q, Lo CM, Poon RT, and Fan ST (2008). Proline-rich tyrosine kinase 2 (Pyk2) promotes proliferation and invasiveness of hepatocellular carcinoma cells through c-Src/ERK activation. Carcinogenesis 29, 2096–2105. [33] Senturk S, Mumcuoglu M, Gursoy-Yuzugullu O, Cingoz B, Akcali KC, and Ozturk M (2010). Transforming growth factor-β induces senescence in hepatocellular carcinoma cells and inhibits tumor growth. Hepatology 52, 966–974. [34] Patil MA, Chua MS, Pan KH, Lin R, Lih CJ, Cheung ST, Ho C, Li R, Fan ST, Cohen SN, et al. (2005). An integrated data analysis approach to characterize genes highly expressed in hepatocellular carcinoma. Oncogene 24, 3737–3747. [35] Ishiguro T, Sugimoto M, Kinoshita Y, Miyazaki Y, Nakano K, Tsunoda H, Sugo I, Ohizumi I, Aburatani H, Hamakubo T, et al. (2008). Anti–glypican 3 antibody as a potential antitumor agent for human liver cancer. Cancer Res 68, 9832–9838. [36] Komori H, Nakatsura T, Senju S, Yoshitake Y, Motomura Y, Ikuta Y, Fukuma D, Yokomine K, Harao M, Beppu T, et al. (2006). Identification of HLA-A2– or HLA-A24–restricted CTL epitopes possibly useful for glypican-3–specific immunotherapy of hepatocellular carcinoma. Clin Cancer Res 12, 2689–2697. [37] Zittermann SI, Capurro MI, Shi W, and Filmus J (2010). Soluble glypican 3 inhibits the growth of hepatocellular carcinoma in vitro and in vivo. Int J Cancer 126, 1291–1301. [38] Ikushima H and Miyazono K (2010). TGFβ signalling: a complex web in cancer progression. Nat Rev Cancer 10, 415–424. [39] Matsuzaki K, Date M, Furukawa F, Tahashi Y, Matsushita M, Sugano Y, Yamashiki N, Nakagawa T, Seki T, Nishizawa M, et al. (2000). Regulatory mechanisms for transforming growth factor beta as an autocrine inhibitor in human hepatocellular carcinoma: implications for roles of SMADs in its growth. Hepatology 32, 218–227. [40] Feng XH and Derynck R (2005). Specificity and versatility in TGF-β signaling through SMADs. Annu Rev Cell Dev Biol 21, 659–693. [41] Giannelli G, Mazzocca A, Fransvea E, Lahn M, and Antonaci S (2011). Inhibiting TGF-β signaling in hepatocellular carcinoma. Biochim Biophys Acta 1815, 214–223.

Figure W1. Effects of GPC3 suppression on the Wnt and ERK/MAPK signaling pathways in Huh7 and HepG2. (A) Suppression of GPC3 by siRNA-GPC3 did not affect the expression of total β-catenin in both Huh7 and HepG2 but inhibited the nuclear localization of β-catenin in HepG2 cells only as shown by Western blot analysis (B) and immunofluorescence (C). (D) Suppression of GPC3 by siRNA-GPC3 reduced phosphorylation of ERK1/2 in Huh7 cells but not in HepG2 cells. siRNA-G indicates cells transfected with siRNA-GPC3; siRNA-N, cells transfected with siRNA-N; V, vehicle control. Histone H3 was used as the loading control for nuclear proteins. Number of independent experiments = 3.