SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. References
  7. Supporting Information

Glypican-3 (GPC3) is a heparan sulfate proteoglycan that has an important role in cell growth and differentiation, and its function in tumorigenesis is tissue-dependent. In hepatocellular carcinoma (HCC), the overexpression of GPC3 has been demonstrated to be a reliable diagnostic indicator. However, the mechanisms that regulate the expression and function of GPC3 remain unclear. The oncoprotein c-Myc is a transcription factor that plays a significant role in more than 50% of human tumors. We report here that GPC3 is a transcriptional target of c-Myc and that the expression of c-Myc is also regulated by GPC3, thus forming a positive feedback signaling loop. We found that the overexpression of c-Myc could induce GPC3 promoter-dependent luciferase activity in luciferase reporter experiments. Furthermore, mutational analysis identified c-Myc-binding sites within the GPC3 promoter. The exogenous overexpression of c-Myc increased the endogenous messenger RNA (mRNA) and protein levels of GPC3. Chromatin immunoprecipitation experiments revealed the binding of c-Myc to the endogenous GPC3 promoter, indicating that c-Myc can directly transcriptionally activate GPC3. Interestingly, GPC3 can also elevate c-Myc expression. Overexpression of GPC3 increased c-Myc protein levels, whereas the knockdown of GPC3 reduced c-Myc expression levels. Lastly, the elevated levels of c-Myc correlate with the overexpression of GPC3 in human HCC samples. Conclusion: These data provide new mechanistic insight into the roles of GPC3 and of c-Myc in the development of HCC. (HEPATOLOGY 2012)

Primary hepatocellular carcinoma (HCC) is the world's fifth most common cancer, and the third most common cause of cancer deaths.1 HCC is a common malignant tumor, and causes over 100,000 HCC deaths per year in China.2 Only 10%-20% of total HCCs can be treated in the early stage by surgery. Most HCCs are diagnosed at late stages, and as a result they progress rapidly, are difficult to treat, and exhibit poor prognosis. Early diagnosis and treatment are critical factors for improving the survival of HCC patients.

Recent research has shown that glypican-3 (GPC3) is a specific and sensitive biomarker for the diagnosis of HCC.3 GPC3 is a proteoglycan that is localized on the cell surface. The abnormal expression of GPC3 in HCC was first reported by Hsu et al.4 in 1997. In that study, 75% of HCC patients were found to overexpress GPC3 messenger RNA (mRNA), whereas patients with benign liver disease or normal livers exhibited no expression of GPC3 mRNA or protein. Soluble GPC3 protein can be detected in the serum of 50% of HCC patients, demonstrating its value as a diagnostic marker for HCC.5 The expression of GPC3, its receptor, and of other growth factors coordinate signal transduction pathways that regulate cellular morphology and a variety of cellular behaviors, such as adhesion, proliferation, migration, survival, and differentiation.6, 7 GPC3 is differentially expressed during the invasive growth of liver cancer, suggesting that its expression might be involved in the initiation of liver cancer, and might be a critical step in liver cancer development. Although these data suggest that the modulation of GPC3 might hold therapeutic potential in the treatment of liver cancer, the mechanism of GPC3 activity in HCC remains unclear.

The c-Myc transcription factor regulates cell growth, proliferation, differentiation, and regulation. Because of its significant effect on cell fate, c-Myc expression is tightly controlled at every possible stage of development and exhibits a prominent role in the development of human cancers.8 Although the relationship between c-Myc and cancer is well established both in vivo and in vitro, the molecular mechanisms underlying c-Myc-mediated transformation are poorly understood. To better understand c-Myc function and its role in cancer development, the identification of c-Myc target genes is important.9, 10 In this report, we demonstrate that GPC3 is a direct c-Myc target gene and that c-Myc overexpression increases the level of GPC3. Furthermore, we found that increased GPC3 level also elevates c-Myc expression, suggesting that GPC3 may be an activator of HCC.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. References
  7. Supporting Information

GPC3 Promoter and Plasmids.

The 5′-region of the GPC3 promoter from position −1358 to +293, relative to the transcription start site (GenBank ID: NG_009286.1), was cloned using polymerase chain reaction (PCR) with genomic DNA (forward primer, 5′-GATCCGTCCGCTGGAGTCTCACT-3′; reverse primer, 5′-CGCTCCTTCTTCCAGAGACTGCAGCCC-3′). The resulting PCR product was cloned into the pGL3-Basic luciferase plasmid (Promega) using standard techniques. The pGL3-luciferase constructs containing the GPC3 promoter region with the deletions and site-directed mutations were similarly synthesized. The resulting constructs, which contained different lengths of the GPC3 promoter, were named GPC3-208 and GPC3-1358. Similarly, the series of site-directed mutant plasmids were referred to as Myc1, Myc2, and Myc-all. The c-Myc expression constructs and pRSV β-galactosidase expression plasmids were gifts from Dr. Charles Lopez (Oregon Health and Science University, Portland, OR).

Cell Culture and Luciferase Reporter Assays.

HepG2 cells, 293T cells, Huh7 cells, and HLE cells were either cultured in Dulbecco's modified Eagle's medium or in McCoy's medium (Hyclone) supplemented with 10% heat-treated fetal bovine serum at 37°C in 5% CO2. The detailed methods have been described previously.11

Chromatin Immunoprecipitation.

Chromatin immunoprecipitation analysis was performed as described.11 The PCR of the GPC3 promoter sequence was performed using 35 cycles as follows: 94°C for 15 seconds, 65°C for 45 seconds. The resulting PCR products were analyzed on 2% TAE-agarose gels. The GPC3-specific primer sequences are as follows: forward, 5′-GGAAGGATGAAAAGGAG-3′ and reverse, 5′-CTGGGGCGTTAAAAG-3′. The GPC3 control primer sequences are 5′-CTGGGGAGTAGGACGTCAGTGCTG-3′ (forward) and 5′-GGCTTGCGGCACCACCCCATAGAGC-3′ (reverse).

Patients and Tumor Tissue Samples.

Eighty-nine liver cancer tissue samples were obtained from the Beijing Youan Hospital, Capital Medical University, between January 2008 and December 2010. The clinical and histological diagnoses of HCC in these patients were in accord with the diagnostic criteria recommended by the World Health Organization. Each sample was histologically graded as well differentiated (WD), moderately differentiated (MD), or poorly differentiated (PD) samples. The detailed patient information is shown in Table 1.

Table 1. General Characteristics of the 89 Hepatocellular Carcinoma Patients
Features 
  • *

    No definite reasons for the etiology can gain.

  • W, well differentiation; M, moderate differentiation; P, poor differentiation; HCC, hepatocellular carcinoma; HBV, hepatitis B virus; HCV, hepatitis C virus.

Sex (female/male)25/64
Age (years)Range, 27–83
Etiology 
 HBV59
 HCV19
 Both HBV and HCV9
 Unknown*11
Pathological aspects 
 HCC 
  Differentiation (W/M/P)26/39/24
   Tumor size≤5 cm, 37; >5 cm, 52
   Single/multinodules39/50
  Peritumor samples 
Background (liver cirrhosis/chronic hepatitis)57/32

The standard methods for (1) western blotting; (2) DNA and RNA extraction, PCR, and quantitative-PCR; (3) flow cytometry for cell cycle analysis, apoptosis, and cell growth rate assay; and (4) immunohistochemical and immunofluorescence staining are shown in the Supporting Materials and Methods.

Statistical Analysis.

The data were analyzed using SPSS v. 16 software. Student t test or analysis of variance (ANOVA) was performed to compare the differences between the groups if the data were normally distributed. Alternatively, a Kruskal-Wallis test or a χ2 test was used. Statistical significance was defined as P < 0.05. P-values are annotated in the text and in the figure legends.

Results

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. References
  7. Supporting Information

Functional Characterization of the GPC3 Promoter.

The GPC3 gene was previously shown to contain a minimal promoter sequence ∼2 kb from the 5′-end of the transcriptional starting site. This region contains a number of potential binding sites for common transcription factors, including six sites for Sp1, seven sites for AP2, and two potential CAAT boxes.13 On the basis of these previous findings, we examined the genomic sequence containing the GPC3 promoter (Fig. 1A). Sequence analysis from the −248 to +293 positions relative to the putative transcription start site revealed a CpG island (boxed sequence). To verify that an active promoter was present in this region, we cloned fragments containing different parts of this sequence (Fig. 1B) into the pGL3-Basic luciferase reporter plasmid. After transfection into 293T cells, we found that the +293/−208 region stimulated ∼4-fold higher luciferase activity relative to the empty plasmid (Fig. 1C), indicating that the +293/−208 region of the GPC3 promoter can induce transcription of the luciferase reporter. The +293/−1358 region increased the luciferase reporter activity ∼12-fold (Fig. 1C) compared with the empty plasmid, suggesting that a functional GPC3 promoter is present within this region and further that transcriptional control elements might exist in the −248/−1358 region. A computer-assisted analysis using the TRANSFAC database indicated that this region has multiple putative transcription-factor-binding sites, such as those for CREB, E2F1, and c-Myc.

thumbnail image

Figure 1. The identification of the active promoter region using GPC3 promoter fragments and a luciferase reporter assay. (A) The genomic sequence from position −1358 to +293 relative to the predicted GPC3 transcription start site; cDNA 5′-end indicated as −8. (B) The schematic represents a series of promoter deletion mutants cloned into the pGL3-Basic-luciferase reporter plasmid. The black box represents the transcription initiation site. The open box with [luc] denotes the luciferase reporter. (C) Luciferase activity of the indicated pGL3-GPC3 promoter-reporter constructs after transfection into 293T cells are shown. Relative light unit values were normalized to a β-galactosidase signal. One microgram of each indicated GPC3 promoter-reporter plasmid and pRSV β-galactosidase plasmid was used for all transfections. The standard deviation (SD) of the triplicate experiments is shown.

Download figure to PowerPoint

GPC3 Promoter Has Features Suggestive of a c-Myc Target Gene.

Our luciferase reporter data above suggested that transcriptional control elements exist within the GPC3 promoter. Sequence analysis from position −1358 to +293 relative to the putative transcription start site (boxed sequence) revealed two putative c-Myc binding sites from position −488 to −483 and from position −827 to −822 (red sequence). Thus, we hypothesized that c-Myc would stimulate transcription of the GPC3 promoter luciferase reporter construct. Cotransfection of increasing amounts of a c-Myc expression plasmid with the GPC3 promoter-luciferase reporter induced a dose-dependent increase in luciferase activity (Fig. 2A). To determine whether the c-Myc-binding sites within the −488/−483 and −827/−822 regions are necessary for this regulation, we mutated the c-Myc-binding sites in the GPC3 promoter region (Fig. 2B). Subsequently, the three GPC3-mutant promoter-pGL3 plasmids and the c-Myc expression plasmid were cotransfected into 293T cells; the pGL3-Basic luciferase reporter plasmid was used as a control. We observed that the cotransfection of the c-Myc expression plasmid with the GPC3-mutant promoter-luciferase reporters exhibited nearly 2-fold decreased luciferase activity compared with the GPC3 wildtype-promoter luciferase reporter (Fig. 2B). These results suggested that c-Myc can activate the GPC3 promoter.

thumbnail image

Figure 2. c-Myc targets the GPC3 promoter and up-regulates of GPC3 transcription. (A) Relative light units normalized to the β-galactosidase signal of pGL3-GPC3 (−1358 to +293) (0.1 μg) was cotransfected into 293T cells with increasing amounts of the c-Myc expression plasmid (20-100 ng). (B) Relative luciferase activity normalized to the β-galactosidase signal expressed as the relative fold-change of the GPC3 (−1358 to +293) promoter-reporter plasmid versus a series of Myc binding site-mutated GPC3 promoter-reporter plasmids (0.1 μg) was cotransfected with c-Myc expression plasmids (0.1 μg) in 293T cells. (C) The ChIP assay is shown. An anti-Myc antibody (Lanes 7-12) and nonimmune control IgG (Lanes 3 and 4) were used for the chromatin immunoprecipitation, followed by PCR and ethidium bromide-stained agarose gel electrophoresis. Input chromatin (Lanes 1 and 2) at 1:1,000 and the dilution of PCR products from the IP reactions performed with GPC3-specific primers (Lanes 9-12) or control primers (Lanes 5-8) are shown. (D) The mRNA levels of GPC3 and c-Myc as detected by qRT-PCR after transfection of HepG2 cells with the c-Myc expression plasmid (0.1 μg) are shown. The SD of the triplicate experiments is shown.

Download figure to PowerPoint

To further confirm the transcriptional regulation of the GPC3 gene by c-Myc, we performed chromatin immunoprecipitation (ChIP) assays to test whether c-Myc associated with the endogenous GPC3 promoter (Fig. 2C). We found that GPC3 promoter-specific PCR primers amplified this promoter region from DNA that was immunoprecipitated with the anti-c-Myc antibody but not with the nonimmune IgG. However, the control GPC3 primers, which amplify a promoter region without c-Myc-binding sites, did not generate a PCR product after ChIP with the anti-c-Myc antibody, demonstrating that the interaction between c-Myc and the GPC3 promoter was specific. These findings demonstrated that the c-Myc protein targets and interacts with the GPC3 promoter. To test c-Myc-mediated transactivation of GPC3, we transiently transfected the pcDNA3.1-Myc plasmid into HepG2 cells. The exogenous expression of c-Myc stimulated an increase in GPC3 mRNA as detected by quantitative reverse-transcription PCR (qRT-PCR) (Fig. 2D). These results suggested that c-Myc transcriptionally activates endogenous GPC3 expression.

c-Myc Up-Regulates GPC3 Expression and Contributes to Cell Cycle Progression.

Despite the extensive studies indicating that GPC3 is overexpressed in HCC,14 the upstream pathways that regulate its expression remain poorly understood. Although both the luciferase reporter and the ChIP data indicated that c-Myc can transactivate GPC3 expression, further experiments were needed to confirm the relationship between GPC3 and c-Myc and their association in tumor cells. To test this, the HCC cell line Huh7 or HepG2 was transiently transfected with the c-Myc expression plasmid for 48 hours. The expression of GPC3 and c-Myc was detected by western blotting and immunofluorescence staining. The results showed that there are ∼4-fold (Fig. 3A) or 2-fold (Supporting Fig. S1A) increases in the level of GPC3 expression in cells transfected with the c-Myc expression plasmid compared with the empty pcDNA3.1 plasmid. The induction of GPC3 expression by c-Myc was confirmed by immunofluorescence staining (Fig. 3B; Fig. S1B).

thumbnail image

Figure 3. The expression of c-Myc up-regulates GPC3 and accelerates cellular proliferation rates of HCC. (A) Western blotting showed an increased level of GPC3 after transfection of Huh-7 cells with the c-Myc expression plasmid. β-Actin was used as a loading control. (B) Huh-7 cells were seeded into 6-well plates and then transfected with the pcDNA3.1 empty plasmid or the pcDNA3.1-c-Myc expression plasmid. The cells were subsequently stained with antibodies against GPC3 (green) and c-Myc (red). GPC3 exhibited increased expression both at the cell surface and in the cytoplasm after transfection with the pcDNA3.1-c-Myc expression plasmid. (C) Flow cytometric cell cycle analysis revealed that both the overexpression of GPC3 and c-Myc significantly induced more S phase accumulation of HepG2 cells and increased their cells proliferation rate. Western blotting confirmed that the transient overexpression of c-Myc stimulates GPC3 expression. (D) Flow cytometric cell cycle analysis demonstrated that more HepG2 cells arrested in G1 phase upon cotransfection with the c-Myc overexpression and siRNA-GPC3 plasmids. This treatment decreased HCC cell proliferation. Western blotting confirmed that GPC3 expression was decreased in the cotransfection cells. Three independent experiments were performed. (E) The characterization of the expression of GPC3, c-Myc, or c-Myc/siRNA-GPC3 by HCC cell proliferation assays. HepG2 cells were transfected with the pcDNA3.1 empty plasmid, GPC3, c-Myc, or c-Myc/siRNA-GPC3 plasmids and then subjected to assessment by MTT assay. The cell growth rate was significantly increased in cells that were transfected with the GPC3 or c-Myc expression plasmids compared with those transfected with the pcDNA3.1 empty plasmid and c-Myc/siRNA-GPC3 plasmid; there was no significant difference between the cells transfected with the pcDNA3.1 empty plasmid and c-Myc/siRNA-GPC3 plasmid (*P < 0.05, #P < 0.05 at days 2 and 3). (F) The effects of GPC3 knockdown on the apoptotic rate of c-Myc overexpressing cells. We cotransfected EGFP/c-Myc/siRNA-GPC3 (1:2:2) into HepG2 cells, isolated GFP-positive cells, and analyzed the changes in cell cycle and apoptosis by flow cytometry. The result shows that the cotransfection cells enhanced cellular apoptotic rate compared with the cells transfected with the c-Myc or pcDNA3.1 empty plasmid alone.

Download figure to PowerPoint

To investigate the effects of GPC3 and c-Myc on the cell cycle, we performed flow cytometric cell-cycle analysis of HepG2 or Huh-7 cells that were transfected with the GPC3 and c-Myc expression plasmids for 48 hours. The protein levels of GPC3 and c-Myc are shown in Fig. 3C (upper panel) and in Fig. S1C (upper panel). The data showed that the transient overexpression of c-Myc stimulates the expression of GPC3 (Fig. 3C, lane 2; Fig. S1C lane 2). Interestingly, the overexpression of GPC3 also stimulates the expression of c-Myc (Fig. 3C, lane 3; Fig. S1C lane 3). Figure 3C and Fig. S1C (lower panel) show that the overexpression of either c-Myc or GPC3 significantly increases the fraction of cells in S phase with a concomitant reduction of the G1 phase population.

To determine whether GPC3 mediates the cell-cycle effects of c-Myc, we analyzed alterations in cell-cycle progression in HepG2 or in Huh-7 cells that were cotransfected with small interfering RNA (siRNA)-GPC3 oligos and the c-Myc overexpression plasmid. Western blotting confirmed that the expression of GPC3 was reduced in the cells that were cotransfected with the siRNA-GPC3 oligos and the c-Myc overexpression plasmid (Fig. 3D; Fig. S1D, upper panel). The results displayed in Fig. 3D and Fig. S1D (lower panel) show that compared with the overexpression of c-Myc alone, the knockdown of GPC3 during the overexpression of c-Myc significantly increased the proportion of cells in G1 phase, while decreasing the cells in S phase. No differences were observed between the pcDNA3.1 empty plasmid transfected cells and the cotransfection cells. To further identify that GPC3 affects cellular proliferation via c-Myc, we analyzed the proliferation rates of cells transfected with different plasmids (c-Myc, GPC3, and c-Myc/siRNA-GPC3) by 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay (Fig. 3E; Fig. S1E). The results revealed that in HepG2 or in Huh-7 cells, the overexpression of c-Myc or GPC3 can increase cellular proliferation rates, the cells cotransfected with c-Myc/siRNA-GPC3 plasmid exhibited suppression of cellular proliferation.

To determine whether the siRNA-mediated knockdown of GPC3 in cells with elevated c-Myc levels elicits effects on apoptosis, we cotransfected EGFP/c-Myc/siRNA-GPC3 (1:2:2) into HepG2 or Huh-7 cells, isolated GFP-positive cells, and assessed changes in apoptosis by flow cytometry. The Sub-G0/G1 (apoptosis) and cell cycle (G1, S, G2-M) assay process are shown in Fig. S2. The results showed that the cotransfection of c-Myc and siRNA-GPC3 synergistically enhanced cellular apoptosis compared with the transfection of c-Myc or empty plasmids alone (Fig. 3F; Fig. S1F).

Taken together, these results indicate that c-Myc elevates GPC3 levels and can accelerate the proliferation of liver cancer cells partly through the GPC3. Furthermore, our data indicate that the knockdown of GPC3 results in the reduction of endogenous c-Myc expression levels. The data also indicate that, whereas increased levels of c-Myc or GPC3 result in increased rates of cell growth proliferation, the suppression of GPC3 by siRNA can enhance apoptosis in cells with elevated c-Myc levels.

Relevance of c-Myc and GPC3 Regulation in Human HCC.

Studies have previously shown that more than 50% of all human cancers overexpress c-Myc. The overexpression of c-Myc is attributed to both epigenetic and genetic mechanisms, including genomic amplification.15 To determine whether our in vitro finding that c-Myc transactivates GPC3 is relevant to the development of HCC, we assessed the c-Myc gene copy number and the N-terminal amino acid (S62 and T58) mutation in c-Myc in tumors and their normal adjacent tissues from 89 HCC patients by PCR analysis. The relative copy numbers of the c-Myc DNA in the 89 HCC patient samples as determined by PCR are shown in Fig. 4A. A ratio above 1.5 was defined as a high level of gene amplification.16 On the basis of this definition, 36 of the 89 (40.45%) HCC samples showed increased c-Myc copy numbers (Table 2). The c-Myc copy number was significantly correlated with the histological differentiation of the HCC samples in the comparisons between the PD and the WD group and between the WD and the MD group (P(PD:WD) < 0.01, P(WD:MD) < 0.05).

thumbnail image

Figure 4. The c-Myc gene copy numbers and GPC3 mRNA levels in the primary HCC samples. The c-Myc gene copy number and GPC3 mRNA level in each HCC sample was determined by PCR and qRT-PCR, respectively. (A) The samples are arranged by histological subtype as indicated. The c-Myc copy number was significantly associated with the histological differences that distinguishes between the PD and the WD group and between the WD and the MD group. (PPD:WD < 0.05, PWD:MD < 0.05). (B) The GPC3 mRNA expression levels exhibit similar alteration to c-Myc copy number in the tumor samples, although neither measure was significantly altered between the PD and the MD group. (C) The GPC3 mRNA expression levels in the adjacent tissues surrounding the tumor were analyzed. Compared with the tumor samples, the GPC3 mRNA levels were decreased in the normal adjacent tissue samples. (D) The mRNA levels of c-Myc in the tumor samples were assessed and found that there was a significant association between the levels of GPC3 and c-Myc (P = 0.035).

Download figure to PowerPoint

Table 2. Correlation Between GPC3 and c-Myc Expression
c-Myc DNAGPC3 mRNA Levels 
Copy NumbersOverexpressionNo OverexpressionTotal
Overexpression27936
No overexpression381553
Total652489

To further examine whether increases in c-Myc copy number are associated with increased expression of GPC3, we performed qRT-PCR to measure the levels of GPC3 mRNA transcripts in the 89 HCC samples (Fig. 4B). Upon comparing the tumor tissue with normal adjacent tissue, we found that GPC3 mRNA expression was up-regulated in 65 of the 89 (73.03%) samples of HCC. Of the 65 samples with overexpressed GPC3, 21 (32.3%) cases exhibited amplified c-Myc gene copy number. Increased expression of GPC3 was also detected in 44 tumors that did not exhibit c-Myc amplification, suggesting that the up-regulation of GPC3 may occur via c-Myc amplification-independent mechanisms (Table 2). The GPC3 mRNA expression levels exhibited significant differences between the PD and the WD group and between the WD and the MD group, but did not exhibit significant differences between the PD and the MD group (P(PD: WD) < 0.01, P(PD:MD) > 0.05, P(WD:MD) < 0.01).

We also compared the GPC3 mRNA level between the tumor tissues and the adjacent tissues surrounding the tumor (Fig. 4B,C). We found the values for GPC3 mRNA in the tumor tissues are higher than that in the adjacent tissues. Upon examination of c-Myc mRNA levels, we found that there was a significant association between the levels of GPC3 and c-Myc (Fig. 4B,D; P = 0.035).

Mutations that alter threonine 58 (T58) and serine 62 (S62) of c-Myc have been shown to affect the levels of c-Myc protein, which, as a likely consequence, would affect GPC3 expression. Hence, we investigated all 89 of the primary HCC tumor samples for c-Myc mutations. We found no mutations at the T58 or S62 sites (data not shown).

We next performed immunohistochemistry to further characterize the expression of GPC3 and c-Myc in HCC tissue samples (Fig. 5A). Staining for the GPC3 protein was strong in the cytoplasm and/or plasma membrane of the tumor cells compared with the cells within the normal adjacent tumor tissues. Further analysis revealed that 66 of the 89 (74.2%) and 76 of the 89 (85.4%) tumors exhibited positive immunohistochemical staining for c-Myc and GPC3, respectively (Fig. 5B). The different degrees of positive staining are shown in Supporting Tables 1 and 2. Staining for the c-Myc protein was strongest in the nuclei and weak in the cytoplasm of the tumor cells (Fig. 5A). Staining for both c-Myc and GPC3 did not significantly correlate with any clinicopathological characteristics. In summary, among the 89 samples, 61 samples were positive for both GPC3 and c-Myc, 15 cases were GPC3-positive and c-Myc-negative, five cases were c-Myc-positive and GPC3-negative, and eight cases were negative for both c-Myc and GPC3. Our analysis revealed a significant association between the expression levels of GPC3 and c-Myc (Fig. 5B). A χ2 test was performed using SPSS v. 16.0, P = 0.03.

thumbnail image

Figure 5. Immunohistochemical analysis of GPC3 and c-Myc and classification of their expresses patterns in the HCC samples. (A) Left: HCC and peritumoral tissues are shown. The HCC tissues stained positively for GPC3 and c-Myc (white star), whereas peritumoral tissues stained negatively for GPC3 and moderately for c-Myc. Right: HCC tissues were processed in the absence of primary antibodies as a negative control. The GPC3 protein exhibited cytoplasmic and/or membranous localization in the carcinoma cells. The c-Myc protein exhibited both nuclear and/or cytoplasmic localization in the carcinoma cells. (B) Expression patterns based on immunohistochemistry staining are shown. GPC3 expression was significantly associated with c-Myc levels in the HCC samples (a χ2 test was performed using SPSS v. 16.0, P = 0.03). Magnification: ×400.

Download figure to PowerPoint

Overexpression of GPC3 Increases c-Myc Levels.

To observe the factors that are affected by GPC3, the GPC3 overexpression plasmid was transfected into the HCC cell line HLE. Unexpectedly, the c-Myc level was notably enhanced (Fig. 6A), and this result was also observed with immunofluorescent staining (Fig. 6B). Conversely, to determine whether the down-regulation of GPC3 could also reduce c-Myc expression, we performed qRT-PCR on cells that were transfected with siRNA-GPC3 for 48 hours. The resulting c-Myc mRNA levels were significantly decreased compared with those transfected with control siRNA (siRNA-N) (Fig. 6C). To ascertain the effects of GPC3 knockdown on cell cycle progression, we analyzed the cellular DNA content by flow cytometry as described above. The knockdown of GPC3 in HepG2 cells induced a significant accumulation of cells in G1 phase (P < 0.05, siRNA-GPC3 compared with siRNA-N) and a concomitant reduction in the S phase population (P < 0.05, Fig. 6D) compared with the control cells treated with siRNA-N. GPC3 has been shown to up-regulate Wnt signaling and to enhance the proliferation of HCC cells.17 c-Myc has been demonstrated to be a Wnt target gene.18 To determine whether the GPC3 might regulate c-Myc through Wnt signaling pathway, we evaluated the effect of elevated GPC3 in HepG2 cells in which β-catenin is suppressed. We found that despite the presence of increased GPC3 levels in the HepG2 cells, the concurrent suppression of β-catenin resulted in the down-regulation of c-Myc compared with the control group (Fig. 6E, upper panel). These findings were verified by qRT-PCR (Fig. 6E, lower panel). These findings suggest that GPC3 can elevate c-Myc levels through the Wnt signaling pathway. We conclude, therefore, that the effects of GPC3 on HCC cells are mediated by c-Myc through the Wnt signaling pathway and that the knockdown of GPC3 decreases the growth of the HCC cells.

thumbnail image

Figure 6. Overexpression of GPC3 increases the c-Myc levels through the Wnt signaling pathway, and siRNA-mediated knockdown of GPC3 decreases the proliferation of HCC cells. (A) Western blotting revealed an increase in levels of c-Myc after transfection with the pcDNA3.1-GPC3 expression plasmid; β-actin was used as a loading control. (B) HLE cells were seeded into 6-well plates and then transfected with the pcDNA3.1 empty plasmid or the pcDNA3.1-GPC3 expression plasmid. Next, cells were stained with antibodies against GPC3 (green) and c-Myc (red). c-Myc exhibited increased cytoplasmic localization after transfection with the pcDNA3.1-GPC3 expression plasmid. (C) Relative GPC3 mRNA levels were assessed by qRT-PCR. The mRNA levels of GPC3 were significantly reduced in HepG2 cells after transfection with the siRNA-GPC3 plasmid compared with cells transfected with the siRNA-N plasmid. (D) Flow cytometric cell cycle analysis showed a G1 phase cell cycle arrest exhibited by HepG2 cells transfected with the siRNA-GPC3 plasmid, which is indicative of decreased HCC cell proliferation. Three independent experiments were performed. (E) The overexpression of GPC3 increased c-Myc levels through the Wnt/β-catenin signaling pathway. We investigated the effect of elevated GPC3 on c-Myc levels in HepG2 cells that were suppressed for β-catenin by siRNA. The western blotting results show that c-Myc levels are down-regulated by the silencing of β-catenin in GPC3-overexpressing HepG2 cells compared with GPC3-overexpressing cells and pcDNA3.1 empty plasmid transfected cells (upper panel). This result was verified by qRT-PCR (lower panel). (F) The model shows GPC3 acting on c-Myc through the Wnt/β-catenin signaling pathway and the direct effect of c-Myc on GPC3. The model includes a positive feedback loop involving GPC3 and c-Myc and the induction of tumorigenesis.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. References
  7. Supporting Information

Hepatocarcinogenesis is a multistep process that involves multiple factors including oncogenes. Several studies have focused on GPC3 as a diagnostic marker for liver cancer. However, the mechanisms underlying the activation and regulation of GPC3 in the liver remain poorly understood. Morford et al.19 suggested that the zinc fingers and homeoboxes 2 (Zhx2) gene, which encodes for a transcriptional repressor protein, could inhibit the expression of GPC3 in the adult mouse liver. Furthermore, the Zhx2 gene is always silenced in HCC, and the increased expression of GPC3 may be partly induced by the loss of Zhx2 suppression. Lai et al.20 observed that the expression of GPC3 was dramatically decreased in the Huh7 cell line when the sulfatase-2 (SULF2) gene was knocked down, implying that SULF2 might increase the expression of GPC3, thereby promoting the growth of HCC tumor cells. In our study, we identified c-Myc as a transcription factor that increases GPC3 expression levels. We demonstrated that c-Myc transcriptionally regulates GPC3 expression by direct interaction with the GPC3 promoter. Together, these data indicate that the transcription factor c-Myc is a novel regulator of the GPC3 gene and may accelerate the proliferation of liver cancer cells at least in part by up-regulating the GPC3 pathway.

To further investigate the effects of c-Myc on GPC3 levels, we assessed the mRNA and protein levels of c-Myc and GPC3 in HCC samples. In human malignancies, the overexpression of proto-oncogenes has been found to result from chromosomal translocation, gene amplification, and/or specific point mutations.21 Because increased c-Myc protein levels have been associated with gene amplification, we evaluated the amplification of the c-Myc gene in HCC samples and their respective tissues. In our study, all of the peritumor tissues were c-Myc-negative. This finding indicated either that HCC cells were not present in the peritumor tissues or that the number of such cells could not be detected by the PCR assay. The frequency of c-Myc amplification in our experiments is higher (40.45%) than that observed in a study by Chan et al.,22 which found that c-Myc was amplified in 30% of surgically resected HCC tissues. This discrepancy could be attributed to technical or methodical differences between the studies. Furthermore, we observed that c-Myc copy number was significantly associated with the histological differences that distinguish between the PD and the WD group and between the WD and the MD group. The GPC3 mRNA expression level was similarly altered. Moreover, we found that the c-Myc DNA copy number was amplified in 32.3% of the 65 HCC cases in which GPC3 was overexpressed. This result suggests that the up-regulation of GPC3 in some HCC tumors might be attributed to c-Myc gene amplification. We further compared the GPC3 mRNA levels between the tumor tissues and the normal adjacent tissues. The GPC3 mRNA levels were significantly higher in the tumor tissues than in the normal adjacent tissues, suggesting that GPC3 might be a reliable diagnostic biomarker for HCC. We also assessed the mRNA levels of GPC3 and c-Myc among the HCC tumor samples. We found that there was a significant association between the levels of GPC3 and c-Myc. This result suggests that an intricate relationship exists between GPC3 and c-Myc and that elevated GPC3 levels in tumor cells might facilitate increased c-Myc levels.

Earlier studies demonstrated that mutations at specific N-terminal phosphorylation sites can increase the transforming activity of c-Myc and that these same mutations have critical effects on tumorigenesis and apoptosis. We analyzed HCC samples for these mutations in c-Myc, focusing on the threonine 58 (T58) and serine 62 (S62) mutation sites because these two phosphoacceptor sites have been shown to affect c-Myc stability.23 Point mutations at or adjacent to the c-Myc T58 phosphorylation site are frequently detected in human Burkitt's lymphoma.24 We found no mutations in the 89 HCC samples that we examined. This result was expected because no c-Myc mutations have been reported in solid tumors. However, this finding does not preclude the involvement of the phosphorylation of S62 and T58 in solid tumorigenesis.25 The dysfunction of the c-Myc degradation pathway, which involves T58 and S62 phosphorylation, may constitute a key event that increases c-Myc stability in HCC.

That c-Myc and GPC3 were detected in a majority of the HCC samples examined by immunohistochemistry, together with the finding that GPC3 and c-Myc expression levels were significantly associated, suggest that GPC3 expression is, at least in part, regulated by c-Myc. However, neither GPC3 nor c-Myc protein levels correlated with the differentiation of HCC tissues. Of the HCC tissues, 15 cases were GPC3-positive and c-Myc-negative, five cases were c-Myc-positive and GPC3-negative, and eight cases were negative for both c-Myc and GPC3, suggesting that other pathways and mechanisms are involved in hepatocarcinogenesis.26

Notably, we found that the overexpression of GPC3 increased c-Myc levels. Previous studies have been shown that GPC3 promotes the growth of HCCs by stimulating the canonical Wnt, Hedgehog, fibroblast growth factor, IGF, and bone morphogenetic protein signaling pathways.17, 27–30 Why and how GPC3 increases the level of c-Myc remains to be investigated. He et al.18 identified that c-Myc is a Wnt target gene. Subsequent studies showed in intestinal and colorectal cancers that c-Myc is regulated by β-catenin, which is an effector of the Wnt pathway.31, 32 Cairo et al.33 used siRNA to target β-catenin mRNA in liver cancer cell lines and found that knockdown of β-catenin mRNA resulted in decreased c-Myc levels. However, other studies have shown that c-Myc does not respond to Wnt/β-catenin signaling in the hepatic context.34, 35 Further studies are warranted to clarify these conflicting findings. We verified that GPC3 elevates c-Myc levels through the Wnt signaling pathway (Fig. 6E). We also showed that the knockdown of GPC3 in HepG2 cells induced a significant G1 phase block, suggesting that the reduction of GPC3 perturbed the cell cycle progression of HCC cells and that an anti-GPC3 therapeutic modality might hold potential application for treatment of HCC.

In summary, the current study provides novel evidence indicating that c-Myc is a direct transcriptional regulator of the GPC3 gene. Notably, we found that the overexpression of GPC3 can elevate c-Myc levels through the Wnt signaling pathway (Fig. 6F). Taken together, our data illustrate the interaction between GPC3 and c-Myc, providing a new mechanism that contributes to the initiation and development of HCC.

References

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. References
  7. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. References
  7. Supporting Information

Additional Supporting Information may be found in the online version of this article.

FilenameFormatSizeDescription
HEP_25891_sm_SuppFig1.tif2397KSupporting Information Figure 1. The expression of c-Myc upregulates GPC3 and accelerates HCC cell proliferation. A, Western blotting showed increased levels of GPC3 in HepG2 cells that were transfected with the c-Myc expression plasmid. β-actin was used as a loading control. B, HepG2 cells were seeded into 6-well plates and were then transfected with the pcDNA3.1 empty plasmid or the pcDNA3.1-c-Myc expression plasmid. The cells were next stained with antibodies against GPC3 (green) and c-Myc (red). GPC3 exhibited localization both at the cell surface and within the cytoplasm after transfection with the pcDNA3.1-c-Myc expression plasmid. C, Flow cytometric cell cycle analysis revealed that both the overexpression of GPC3 and c-Myc significantly induced more S phase accumulation of Huh-7 cells, which suggestes increased cell proliferation. Western blotting confirmed that the transient overexpression of c-Myc stimulates GPC3 expression. D, Flow cytometric cell cycle analysis demonstrated more cells G1 phase-arrested Huh-7 cells upon co-transfection with the c-Myc overexpression plasmid and the siRNA-GPC3 plasmid. Consistent with the G1 arrest, this treatment decreased the proliferation rate of the HCC cells. The western blotting results indicated that GPC3 expression was decreased in the co-transfected cells. Three independent experiments were performed. E, The characterization of the expression of GPC3, c-Myc or c-Myc/siRNA-GPC3 by HCC cell proliferation assays. The pcDNA3.1 empty plasmid, GPC3, c-Myc or c-Myc/siRNA-GPC3 plasmids were transfected into Huh-7 cells, and the MTT assay was performed. The cell growth rate was significantly increased in cells that were transfected with the GPC3 or c-Myc expression plasmids compared with those transfected with the pcDNA3.1 empty plasmid and c-Myc/siRNA-GPC3 plasmid, there was no significant difference between the cells transfected with the pcDNA3.1 empty plasmid and c-Myc/siRNA-GPC3 plasmid (*P < 0.05, #P < 0.05 at days 2 and 3). F, The effects of GPC3 knockdown on the apoptotic rate of c-Myc-overexpressing cells. We co-transfected EGFP/c-Myc/siRNA-GPC3 (1:2:2) into Huh-7 cells, isolated for GFP-positive cells, and assessed the alterations in cell cycle and apoptosis. The results show that the co-transfection of c-Myc and siRNA-GPC3 synergistically enhanced apoptosis compared with c-Myc-transfected cells and the empty plasmid -transfected cells alone.
HEP_25891_sm_SuppFig2.tif291KSupporting Information Figure 2. The following plasmid combinations were co-transfected into the Huh-7 or HepG2 cells: EGFP/c-Myc (1:2), EGFP/GPC3 (1:2), EGFP/c-Myc/siRNA-GPC3 (1:2:2). The GFP-positive cells were isolated, and analyzed for sub-G0/G1 (apoptosis peak), G1, S, G2-M. The result shows that cells co-transfected with EGFP/c-Myc/siRNA-GPC3 plasmid exhibited significantly increased apoptosis.
HEP_25891_sm_SuppTab1.doc22KSupporting Information Table 1. Correlation between c-Myc expression and the differentiation of HCC
HEP_25891_sm_SuppTab2.doc22KSupporting Information Table 2. Correlation between GPC3 expression and the differentiation of HCC
HEP_25891_sm_SuppInfo.doc33KSupporting Information

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.