Advertisement

Apoptosis

, Volume 13, Issue 9, pp 1135–1147 | Cite as

Downregulation of hepatoma-derived growth factor activates the Bad-mediated apoptotic pathway in human cancer cells

  • Tsun Yee Tsang
  • Wan Yee Tang
  • Wing Pui Tsang
  • Ngai Na Co
  • Siu Kai Kong
  • Tim Tak Kwok
Original Paper

Abstract

Hepatoma-derived growth factor (HDGF) is highly expressed in human cancer and its expression is correlated with poor prognosis of cancer. The growth factor is known to stimulate cell growth while the underlying mechanism is however not clear. Transfection with HDGF cDNA stimulated while its specific antisense oligonucleotides repressed the growth of human hepatocellular carcinoma HepG2 cells. Furthermore, knock-down of HDGF by antisense oligos also induced apoptosis in HepG2 cells and in other human cancer cells, e.g. human squamous carcinoma A431 cells. HDGF knock-down was found to induce the expression of the pro-apoptotic protein Bad and also inactivate ERK and Akt, which in turn led to dephosphorylation of Bad at Ser-112, Ser-136, and activation of the intrinsic apoptotic pathway, i.e. depolarization of the mitochondrial membrane, release of mitochondrial cytochrome c, increase in the processing of caspase 9 and 3. As HDGF knock-down not only suppresses the growth but also induces apoptosis in human cancer cells, HDGF may therefore serve as a survival factor for human cancer cells and a potential target for cancer therapy.

Keywords

HDGF Apoptosis Bad 

Introduction

Hepatoma-derived growth factor (HDGF) is an acidic heparin-binding growth factor originally isolated from the conditioned media of the human hepatoma cell line, HuH-7 [1]. HDGF belongs to the HDGF-related protein family. Members of this family included human HDGF, murine HRP-1, murine HRP-2, murine HRP-3, bovine HRP-4 and lens epithelium-derived growth factor (LEDGF), which was also reported to be highly expressed in human cancer cells. These proteins share a common N-terminal homologous to the HDGF amino terminus (HATH) domain of ~100 amino acid residues with a high level of sequence identity (66–89%), but the C-terminal regions show considerable variation in length and charge [2].

HDGF is expressed in a broad range of tissues, with the highest expression in the liver, lung, kidney, testis and placenta [3]. The growth factor was reported to have mitogenic activity. In addition to hepatoma cells [1], recombinant HDGF was able to stimulate the growth of fibroblasts [4], vascular smooth muscle cells [5] and endothelial cells [6]. HDGF was involved in the regulation of liver development by stimulating the proliferation of fetal hepatocyte [7]. Moreover, the growth factor also facilitated the release of pro-apoptotic factors from mitochondria in TNF treated cells [8]. It was reported that HDGF acts primarily in the nucleus and the translocation of the growth factor into the nucleus is important for its growth-promoting function [9, 10]. This is supported as the 240 amino acid HDGF protein contains a bipartite nuclear localization sequence and a DNA-binding PWWP domain [11].

Over-expression of HDGF can be detected in a number of human cancers including hepatoma [12, 13, 14, 15], gastric cancer [16], colorectal cancer [17], lung cancer [18] and melanoma [19]. Increased expression of HDGF was found to be correlated with high proliferating states of the cancer cells and poor clinical outcome. Therefore, the expression of HDGF was reported to be a potential prognostic factor for hepatoma [12, 13, 14], non-small-cell lung cancer [18], esophageal cancer [20], gastric cancer [21] and pancreatic cancer [22].

HDGF was believed to be involved in the development of cancer. A gradual increase in HDGF expression was detected before the onset of liver tumor development in FLS mice [12]. Over-expression of HDGF in 3T3 fibroblast cells conferred a transformed phenotype and enabled the formation of tumors in nude mice [18]. Moreover, the HDGF level in the poorly differentiated human liver cancer is less as compared with that in the well-differentiated subtype [14]. HDGF was reported to stimulate the growth of vascular endothelial cells [23] and elevated expression of the growth factor may therefore stimulate the angiogenic activity and promote the aggression of tumor. In addition, knock-down of HDGF was reported to suppress the proliferation of hepatoma cells [24] and inhibit the anchorage-independent growth of human lung cancer cells [25]. Therefore, HDGF is suggested to play an important role in cancer development.

Although HDGF is over-expressed in human cancer and its over-expression is correlated with the development, progression and poor prognosis of human cancer, the mechanism of how HDGF may affect the growth of human cancer cells is still not clear. In the present study, the role of HDGF in the regulation of cancer cell proliferation and its underlying mechanism were investigated. It was found that knock-down of HDGF by HDGF antisense oligos not only inhibited the growth but also induced apoptosis in human cancer cells through the Bad mediated intrinsic apoptotic pathway. The results suggest that HDGF may function as a survival factor for human cancer cells and thus a potential target for cancer therapy.

Materials and methods

Cell culture

Human hepatocellular carcinoma HepG2 cells, RHepG2 cells and Hep3B were cultured in Dulbecco’s modified essential medium (DMEM) supplemented with 5% fetal bovine serum (Invitrogen, USA) and L-glutamine (Invitrogen, USA) at 37°C in a humidified atmosphere with 10% CO2. RHepG2 cells were the drug resistant subline of HepG2 cells. Human squamous cell carcinoma A431 cells, cervical carcinoma CaSki cells, SiHa cells, and lung carcinoma Calu-6 cells were cultured in the same condition except that 10% fetal bovine serum was used.

HDGF antisense oligonucleotides transfection

The sequence for the 21-base HDGF antisense oligos is complementary to the coding region of hdgf (nt 1027-1038, 5′-GGTTGGATCGCGACATGGCGG -3′) while the sequence of the sense oligos is the same as the coding sequence of hdgf (5′-CCGCCATGTCGCGATCCAACC-3′). All bases were phosphorothiolated to increase the stability of the nucleotides (TechDragon, Hong Kong). About 2 × 105 cells were seeded in 60 mm Petri dish for 48 h before the transfection. The cells were transfected with 0.2 μM HDGF antisense or sense oligos with the aid of oligofectamine (Invitrogen, USA) for 24 h at 37°C in serum-free DMEM.

HDGF full length cDNA transfection

The full-length HDGF cDNA in pcDNA3 was purchased from ATCC. The cells were seeded for 48 h before the transfection. The cells were transfected with 0.1 μg HDGF cDNA carrying vector (HDGF-pcDNA3) or empty vector (pcDNA3) with the aid of lipofectamine (Invitrogen, USA) for 24 h at 37°C in DMEM with 5% fetal bovine serum.

MTT cell viability assay

After transfection, the cells were allowed to grow for 3 days. After that, the cells were treated with 50 µl of 0.1 mg/ml 3-(4,5-Dimethylthiazole-2-yle)-2,5-diphenyltetrazolium bromide solution (USB, USA) at 37°C for 3 h and then lysed in 150 µl of dimethyl sulfoxide at room temperature for 30 min. The absorbance of each well was measured at 570 nm in a microplate reader.

DNA fragmentation assay

The cells were collected by trypsinization and lysed by lysis solution with 5 mM Tris–HCl buffer (pH 8.0), 100 mM EDTA (pH 8.0), 1% SDS and 0.2 mg/ml proteinase K. After incubation at 45°C for 2 h, the DNA in the crude lysate was extracted by phenol-chloroform and then precipitated by ethanol. The DNA extracted was dissolved in Tris–EDTA buffer with 0.2 mg/ml RNaseA and was incubated in 37°C overnight. Thereafter, 30 µg DNA was loaded into 1.5% EtBr staining agarose gel for electrophoresis. The DNA fragmentation pattern was viewed under UV light box.

Annexin V binding assay

The cells were collected by trypsinization. After washing with PBS, the cells were resuspended in Annexin V binding buffer containing 10 mM Hepes, 140 mM NaCl and 2.5 mM CaCl2 and stained with 5 µl Annexin V-GFP and 1 µg/ml propidium iodide at room temperature for 15 min. The cells were subjected to flow cytometry using the BD FACSCanto flow cytometer (Becton Dickinson, USA).

RT-PCR

The cells were lysed by Tri Reagent (Molecular Research Centre, USA) for RNA extraction. After that, total RNA was subjected to reverse transcription with oligo dT primer (Invitrogen, USA) by MMLV Reverse Transcriptase (Promega, USA) at 42°C for 1 h and heated to 70°C for 15 min. PCR amplification of cDNAs using the pair of primers specific to hdgf within its open reading frame was carried out for 25 cycles of 94°C for 30 s, 55°C for 30 s and 70°C for 1 min; forward primer (5′-CCGCCATGTCGCGATCCAACC-3′) and reverse primer (5′-GGTTGGATCGCGACATGGCGG-3′). RT-PCR assay with primers specific for β-actin was also performed.

JC-1 staining

JC-1 staining was used to determine the mitochondrial membrane potential of cells. The cells were collected by trypsinization, resuspended in PBS, and stained with 10 µM JC-1 dye at 37°C for 15 min. Thereafter, the cells were subjected to flow cytometric analysis using the BD FACSCanto flow cytometer (Becton Dickinson, USA).

Mitochondrial cytochrome c release measurement

The cells were collected by trypsinization and were partially lysed by using 0.5 mg/ml digitonin (Sigma, USA) together with lysis solution containing 30 mM NaCl, 0.4 mM NaH2PO4, 3 mM Na2HPO4, 0.1 M sucrose, 2 µg/ml leupeptin, 8 µg/ml aprotinin and 0.4 mM phenylmethylsulphonyl fluoride (PMSF). The cell lysate was centrifuged and the level of cytochrome c was measured by Western blot analysis.

Western blot analysis

The cells were lysed in lysis solution containing 1% Triton X-100, 4.9 mM MgCl2, 1 mM sodium ortho-vanadate, and protease inhibitors aproptinin (21 µg/ml), leupeptin (0.5 µg/ml), and 1 mM PMSF. The cell lysate was centrifuged and the protein content in the supernatant was determined by the BCA Pierce Assay Kit (Pierce, USA). Protein from each sample was separated by SDS-PAGE according to size and was transferred onto PVDF membrane (Millipore, USA) by electro-blotting. The membrane was incubated with 5% non-fat dry milk in PBS at 4°C overnight and then probed with primary antibody for 2 h and with respective horse-radish peroxidase conjugated secondary antibody for 1 h. The primary antibodies are anti-cytochrome c, caspase 3, Bad, Bax, Bcl-2, Bcl-xL, ERK and pERK (Santa Cruz, USA), caspase 9 (stressgene, USA), pBad S112, pBad S136, pBad S155, Akt and pAkt S473 (Cell signaling, USA) and β-actin (Sigma, USA). The protein bands were detected by the ECL Western blot detection reagent (Amersham pharmacia, USA). The protein band intensity was scanned and analyzed by ImageJ (version 1.38×) software.

Results

HDGF regulated the growth of HepG2 cells

To determine the role of HDGF in the growth of human cancer cells, human hepatocellular carcinoma HepG2 cells were transfected with either HDGF full length cDNA or its specific antisense oligonucleotides. It was found that after HDGF cDNA transfection, the growth of HepG2 cells was increased by ~20% (Fig. 1a). On the other hand, after HDGF knock-down by antisense oligos transfection, the growth of HepG2 cells was suppressed by ~15% (Fig. 1b). It indicates that HDGF may regulate the growth of HepG2 cells.
Fig. 1

The effect of HDGF on the growth of HepG2 cells. (a) The cells were transfected with HDGF cDNA carrying vector (HDGF-pcDNA3) and empty vector (pcDNA3). (b) The cells were transfected with HDGF antisense oligos (HDGF AS) and HDGF sense oligos (HDGF S). After transfection, the cells were allowed to grow for 72 h before MTT assay. Error bar: the standard deviation from at least 3 separate experiments. The protein expression level of HDGF in cells after transfection with vector or oligos was shown at the top of the figure. The number beneath the band of HDGF was the relative expression level of HDGF that was first normalized with the expression level of β-actin and then calculated in relation to that of the control. *P < 0.05

Knock-down of HDGF induced apoptosis in human cancer cells

Suppression of the cell death process is one of the ways that may lead to the increase in cell growth and vice versa. To find out whether the repression of cell growth by HDGF knock-down is associated with apoptosis induction, HepG2 cells were transfected with HDGF antisense oligos, followed by apoptosis detection. It was found that knock-down of HDGF not only suppressed the growth but also induced apoptosis in HepG2 cells as determined by DNA fragmentation assay (Fig. 2a) and Annexin V binding assay (Fig. 2b). Similar results were obtained after knock-down of HDGF by HDGF siRNA in HepG2 cells (Supplementary Fig. 1).
Fig. 2

Apoptosis induction in HepG2 cells after knock-down of HDGF. The cells were transfected with HDGF antisense oligos (AS) or HDGF sense oligos (S). The cells with no transfection (N) were also included as the control. (a) After transfection, the cells were allowed to grow for 24, 48 and 72 h followed by DNA fragmentation assay. (b) The cells were allowed to grow for 48 h after transfection followed by Annexin V/propidium iodide staining assay. The number indicated the percentage of cells in the specific quadrant. (c) The mRNA and the protein expression levels of HDGF in cells after transfection with antisense oligos. The number beneath the band of HDGF was the relative expression level of HDGF. It was first normalized with the expression of β-actin and then calculated in relation to that of the control. M: 100 bp DNA ladder. The experiments have been repeated at least 3 times with similar results. The one shown is the representative one

The induction of apoptosis by knock-down of HDGF was also detected in other hepatocellular carcinoma cells, Hep3B cells, doxorubicin-resistant RHepG2 cells, human squamous carcinoma A431 cells (Fig. 3), human cervical carcinoma CaSki cells, SiHa cells, and human lung carcinoma Calu6 cells (data not shown). It therefore indicates that HDGF may act as a survival factor for human cancer cells.
Fig. 3

Apoptosis induction in RHepG2, Hep3B and A431 cells after knock-down of HDGF. The cells were transfected with HDGF antisense oligos (AS) or HDGF sense oligos (S). (a) After transfection, the cells were allowed to grow for 72 h followed by DNA fragmentation assay. (b) The cells were allowed to grow for 48 h after transfection followed by Annexin V/propidium iodide staining assay. The number indicated the percentage of cells in the specific quadrant. (c) The mRNA expression level of HDGF in cells after transfection with antisense oligos. The number beneath the band of hdgf was the relative expression mRNA level of hdgf. It was first normalized with the expression of β-actin and then calculated in relation to that of the cells with sense oligos transfection. The experiments have been repeated at least 3 times with similar results. The one shown is the representative one

Knock-down of HDGF on the expressions of Bcl-2 family proteins

Bcl-2 family proteins are known to be the major regulators in the process of apoptosis. Since knock-down of HDGF induced apoptosis in human cancer cells, to investigate its underlying mechanism, the expression levels of different Bcl-2 family proteins after knock-down of HDGF were therefore examined. After knock-down of HDGF in HepG2 cells, a significant increase in the expression level of pro-apoptotic protein Bad was observed while there were only a slight or no change in the levels of Bax, Bcl-2 and Bcl-xL (Fig. 4). Therefore, the pro-apoptotic protein Bad is suggested to be involved in apoptosis induced by HDGF knock-down. This is further confirmed as the apoptosis induced by HDGF knock-down is partially overcome by the knock-down of Bad (Supplementary Fig. 2).
Fig. 4

The effect of HDGF knock-down on the expression levels of Bad, Bax, Bcl-2 and Bcl-xL in HepG2 cells. The cells were transfected with HDGF antisense oligos (AS) or HDGF sense oligos (S). The cells with no transfection (N) were also included as the control. Protein was extracted from cells at 0, 4 and 24 h after transfection. Western blot analysis was performed and the expression levels of Bad, Bax, Bcl-2 and Bcl-xL were determined. The number beneath the band of the specific protein was the relative expression level of that protein. It was calculated first by normalizing with the expression level of β-actin and then in relation to the expression at 0 h. The experiments have been repeated at least 3 times with similar results. The one shown is the representative one

Knock-down of HDGF activated the intrinsic apoptotic pathway

As Bad is known to regulate the intrinsic apoptotic pathway, further experiments were therefore performed to find out whether knock-down of HDGF may trigger apoptosis through the intrinsic apoptotic pathway, including the depolarization of mitochondrial membrane, the release of mitochondrial cytochrome c, the processing of caspases 9 and 3. By flow cytometric analysis of cells staining with JC-1, a larger shift of cell population with red fluorescence to green fluorescence, an indication of mitochondrial membrane depolarization, was observed in cells after HDGF knock-down when compared with cells transfected with sense oligos or no transfection (Fig. 5a). Besides this, there is also a larger increase in the release of cytochrome c from the mitochondria (Fig. 5b), processing of caspases 9 and 3 in cells (Fig. 5c) after HDGF knock-down. Furthermore, the inhibitors of caspases 9 and 3 were able to suppress the apoptosis induced by HDGF knock-down as measured by DNA fragmentation and Annexin V binding assays (Fig. 5d). The results therefore indicate that apoptosis induced by knock-down of HDGF may be through the Bad mediated intrinsic apoptotic pathway.
Fig. 5

The effect of HDGF knock-down on the intrinsic apoptotic pathway in HepG2 cells. The cells were transfected with HDGF antisense oligos (AS) or HDGF sense oligos (S). The cells with no transfection (N) were also included as the control. (a) The cells were stained with JC-1 dye at 24 h after transfection for the detection of mitochondrial membrane potential. Increase in cell population stained with JC-1 monomer indicates the depolarization of mitochondrial membrane. (b) The cells were lysed for cytoplasmic protein by using 0.5ug/ml digitonin at 0, 4 and 24 h after transfection for the detection of mitochondrial cytochrome c release. After that, Western blot analysis was performed to determine the level of cytoplasmic cytochrome c. (c) Protein was extracted from cells at 0, 4, 24 h after transfection. Western blot analysis was performed and the expression levels of the pro and the active form of caspases 3 and 9 were determined. The number beneath the band of the specific protein was the relative expression level of that protein. It was calculated first by normalizing with the expression level of β-actin and then in relation to the expression at 0 h. (d) After transfection, the cells were incubated with 30 µM caspase 9 and 3 inhibitors for 72 h followed by DNA fragmentation assay and Annexin V binding assay. Left panel: DNA fragmentation assay. All the experiments have been repeated at least 3 times with similar results. The one shown is the representative one. Right Panel: Annexin V binding assay. The % increase in apoptotic cells is calculated by subtracting the % of apoptotic cells in the group transfected with HDGF antisense oligos from that of the corresponding group transfected with the sense oligos. The data are average from three independent experiments. Mean ± SEM. *P < 0.05, significantly different from cells with no caspase inhibitor

Knock-down of HDGF regulated the phosphorylation status of Bad

The activity of Bad in apoptosis activation is directly related to its phosphorylation status. Dephosphorylation of Bad at Ser-112, Ser-136 and/or Ser-155 sites result in separation of Bad from 14-3-3 proteins, binding to Bcl-2 and Bcl-xL, and eventually triggers apoptosis [26, 27]. Since Bad is up-regulated after HDGF knock-down, the phosphorylation status of Bad was therefore also investigated. After knock-down of HDGF, the total Bad expression was increased by about 2 to 3 fold. At the same time, the levels of Bad phosphorylation at Ser-112 and Ser-136 were decreased by about 30–50% while the level of phosphorylation at Ser-155 remained no change (Fig. 6). Taking into the account of the increase in the total Bad level, therefore, knock-down of HDGF also increased the levels of Bad dephosphorylation at Ser-112, Ser-136, and Ser-155; the degree of dephosphorylation is however greater at Ser-112 and Ser-136.
Fig. 6

The effect of HDGF knock-down on the expression and the phosphorylation levels of Bad at Ser-112, Ser-136 and Ser-155 in HepG2 cells. The cells were transfected with HDGF antisense oligos (AS) or HDGF sense oligos (S). The cells with no transfection (N) were also included as the control. Protein was extracted at 0, 4 and 24 h after transfection. Western blot analysis was performed to determine the expression levels of total Bad and Bad phosphorylated at Ser-112, Ser-136 and Ser-155. The number beneath the band of the specific protein was the relative expression level of that protein. It was calculated first by normalizing with the expression level of β-actin and then in relation to the expression at 0 h. The experiments have been repeated at least 3 times with similar results. The one shown is the representative one. The data for Bad are adopted from Fig. 4

EGF on phosphorylation of Bad upon HDGF knock-down

Phosphorylation of Bad at Ser-112, Ser-136 and Ser-155 was regulated by the Ras/Raf/MEK/ERK, PI3K/Akt and PKA pathways respectively [28, 29]. EGF was reported to stimulate the Ras/Raf/MEK/ERK and PI3K/Akt pathway that lead to phosphorylation of Bad at Ser-112 and Ser-136 [28]. Forskolin, which acts as a PKA activator, was reported to activate the PKA pathway that leads to phosphorylation of Bad at Ser-155 [29]. In the present study, similar situation was demonstrated in HepG2 cells. EGF induced phosphorylation of ERK, Akt, Bad at Ser-112 and Ser-136 but not that at Ser-155. On the contrary, forskolin did not induce phosphorylation of ERK, Akt, Bad at Ser-112 and Ser-136 while it induced phosphorylation of Bad at Ser-155 (Fig. 7). To further determine the importance of Bad dephosphorylation at Ser-112, Ser-136 and Ser-155 in apoptosis induced by HDGF knock-down, EGF and forskolin were therefore included in the subsequent studies. As shown in Fig. 8, EGF partially suppressed the apoptosis induced by HDGF knock-down while forskolin appeared to have no effect. The results therefore suggested that Bad dephosphorylation at Ser-112 and Ser-136 rather than Ser-155 is likely to be important for apoptosis induced by HDGF knock-down. This can further be reflected as the levels of Bad phosphorylation induced by EGF at Ser-112 and Ser-136 were more or less similar in cells even with HDGF knock-down (Fig. 9). On the other hand, since Bad phosphorylation only contributed partly to the apoptosis induction and EGF did not affect the Bad upregulation upon HDGF knock-down, the upregulation of Bad per se by HDGF knock-down may therefore also likely be critical for the apoptosis induction. This is supported further as the apoptosis induced by HDGF knock-down is partially reduced upon the silencing of Bad (Supplementary Fig. 2).
Fig. 7

The effect of EGF and forskolin on the phosphorylation of Bad, ERK and Akt. The cells were incubated with (a) 30 ng/ml EGF and (b) 20 µM forskolin for 30 min or 3 h. After that, protein was extracted and Western blot analysis was performed to determine the expression levels of Bad, ERK and Akt and their respective phosphorylated forms, pBad S112, pBad S136, pBad S155, pERK and pAkt S473. The number beneath the band of the specific protein was the relative expression level of that protein. It was calculated first by normalizing with the expression level of β-actin and then in relation to that of the control. The experiments have been repeated at least three times with similar results. The one shown is the representative one

Fig. 8

The effect of EGF and forskolin on apoptosis induced by HDGF knock-down in HepG2 cells. The cells were transfected with HDGF antisense oligos (AS) or HDGF sense oligos (S). The cells with no transfection (N) were also included as the control. After that, the cells were incubated with (a) 30 ng/ml EGF or (b) 20 µM forskolin for 48 h followed by Annexin V/propidium iodide staining assay. The number indicated the percentage of cells in the specific quadrant. The experiments have been repeated at least 3 times with similar results. The one shown is the representative one. (c) The % increase in apoptotic cells is calculated by subtracting the % of apoptotic cells in the group transfected with HDGF antisense oligos from that of the corresponding group transfected with the sense oligos. In addition to transfection with oligos, the cells were also treated with or without EGF or forskolin. H-AS: transfection with HDGF antisense oligos. The data are average from three independent experiments. Mean ± SEM. *P < 0.05, significantly different from cells with no EGF treatment

Fig. 9

The effect of EGF on the phosphorylation of Bad upon HDGF knock-down in HepG2 cells. The cells were transfected with HDGF antisense oligos (AS) or HDGF sense oligos (S). The cells with no transfection (N) were also included as the control. After that, the cells were incubated with 30 ng/ml EGF for 3 h. Protein was then extracted and Western blot analysis was performed to determine the expression levels of Bad, pBad S112, pBad S136 and pBad S155. The number beneath the band of the specific protein was the relative expression level of that protein. It was calculated first by normalizing with the expression level of β-actin and then in relation to the expression of the control. The experiments have been repeated at least three times with similar results. The one shown is the representative one

EGF on phosphorylation of ERK and Akt upon HDGF knock-down

Phosphorylation of Bad at Ser-112, Ser-136 and Ser-155 was known to be regulated at least by the Ras/Raf/MEK/ERK, PI3K/Akt and PKA pathways respectively [28, 29]. In addition to inactivation of these pathways, the dephosphorylation of Bad may also involve phosphatases such as protein phosphatase 2A, protein phosphatases 1alpha [30, 31]. Knock-down of HDGF not only induced dephosphorylation of Bad, but also induced the dephosphorylation of both ERK and Akt in HepG2 cells (Fig. 10a). In addition to phosphorylation of Bad at Ser-112 and Ser-136, EGF at the same time also induced the phosphorylation of ERK and Akt to a similar extent in HepG2 cells regardless of the HDGF knock-down (Fig. 10b). Therefore, EGF can counteract the HDGF knock-down effect on the dephosphorylation of both ERK and Akt (Fig. 10b) as well as that for the dephosphorylation at Ser-112 and Ser-136 of Bad (Fig. 9). The results support that inactivation of both Ras/Raf/MEK/ERK and PI3K/Akt pathways will at least likely be contributed to the Bad dephosphorylation upon HDGF knock-down.
Fig. 10

The effect of EGF on the expression levels and phosphorylation levels of ERK and Akt upon HDGF knock-down in HepG2 cells. The cells were transfected with HDGF antisense oligos (AS) or HDGF sense oligos (S). (a) The cells with no transfection (N) were also included as the control. (b) After that, cells were incubated with 30 ng/ml EGF for 3 h. Protein was then extracted and Western blot analysis was performed to determine the expression levels of ERK, Akt and their phosphorylated forms, pERK and pAkt S473 site. The number beneath the band of the specific protein was the relative expression level of that protein. It was calculated first by normalizing with the expression level of β-actin and then in relation to the expression of the control. The experiments have been repeated at least 3 times with similar results. The one shown is the representative one

Discussion

HDGF may act as a survival factor for human cancer cells by regulating the process of apoptosis. Knock-down of HDGF was found to suppress the growth (Fig. 1) and trigger apoptotic cell death in human cancer cells (Figs. 2 and 3). The results for cell growth reduction by HDGF knock-down was in agreement with the work of Kishima [24]. Besides this, knock-down of HDGF also reduced the anchorage independent growth of HepG2 cells (data not shown) as that been reported in the lung cancer cells [25]. Although apoptosis induction might be one of the many ways that lead to the growth suppression in cells, the present study is the first to demonstrate that knock-down of HDGF will induce apoptosis in human cancer cells. Similar report was found in the knock-down of midkine (MK), a heparin binding growth factor, which is also highly expressed in a broad range of cancer cells. Knock-down of midkine induced apoptosis in gastric cancer cell lines [32]. Since knock-down of HDGF not only suppressed the growth but also induced apoptosis in cancer cells, HDGF is therefore believed to function as a survival factor for human cancer cells.

The apoptosis induced by knock-down of HDGF is mediated by Bad, a pro-apoptotic Bcl-2 family protein, through the activation of the intrinsic apoptotic pathway. Knock-down of HDGF induced the expression (Fig. 4) and the dephosphorylation of Bad (Fig. 6). Bad is known to trigger apoptotsis by dimerizing with Bcl-2 and Bcl-xL as to neutralize their anti-apoptotic effects and thus promotes the pore formation on the mitochondria membrane, causes mitochondria membrane depolarization, and eventually activated the intrinsic apoptotic pathway [33, 34]. As expected, the apoptosis induced by knock-down of HDGF is associated with the activation of the intrinsic apoptotic pathway as knock-down of HDGF triggered the depolarization of the mitochondrial membrane (Fig. 5a), release of mitochondrial cytochrome c (Fig. 5b) and the processing of caspases 9 and 3 (Fig. 5c, d), which are hallmarks of the intrinsic apoptotic pathway.

Bad dephosphorylation together with Bad upregulation are critical for HDGF effect on apoptosis. The activity of Bad was reported to be regulated by phosphorylation at three serine residues, Ser-112, Ser-136 and Ser-155 [26]. Phosphorylation of Bad at Ser-112 and Ser-136 stimulates the binding of Bad with 14-3-3 proteins and sequesters Bad in the cytosol while the phosphorylation of Bad at Ser-155 inhibits the direct interaction of Bad with Bcl-2 and Bcl-xL [27]. Phosphorylation of Bad at any these three serine residues prevents Bad from triggering apoptosis. On the contrary, dephosphorylation of Bad leads to the release of 14-3-3 protein and allows Bad to gather at the mitochondria where it dimerizes with Bcl-2 and Bcl-xL to trigger the depolarization of the mitochondrial membrane which eventually activates the intrinsic apoptotic pathway [35, 36]. Dephosphorylation of Bad at Ser-112 and Ser-136 rather than Ser-155 was found to be important in mediating apoptosis induced by HDGF knock-down. The levels of Ser-112 and Ser-136 dephosphoryation on Bad upon HDGF knock-down are higher that that at Ser-155. EGF stimulated the phosphorylation of Bad at Ser-112 and Ser-136 but not Ser-155 [28] while forskolin stimulated the phosphorylation of Bad at Ser-155 but not at Ser-112 and Ser-136 (Fig. 7). EGF partially suppressed the apoptosis induced by HDGF knock-down while forskolin cannot (Fig. 8). All these indicate that dephosphorylation of Bad at Ser-112 and Ser-136 rather than Ser-155 is important in mediating apoptosis induced by HDGF knock-down. The significance of Bad upregulation in HDGF knock-down induced apoptosis is hinted as EGF only partially overcomes the HDGF knock-down effect on apoptosis and has no effect on Bad upregulation. The role of Bad is further confirmed as the level of apoptosis induced by HDGF knock-down is partially reduced by co-silencing of Bad (Supplementary Fig. 2).

Knock-down of HDGF, through inactivation of the Ras/Raf/MEK/ERK and PI3K/Akt pathways, induced dephosphorylation of Bad at Ser-112 and Ser-136. According to the literature, phosphorylation of Bad at different serine residues was regulated by different signaling pathways. Phosphorylation of Bad at Ser-112, Ser-136 was at least regulated by the Ras/Raf/MEK/ERK and PI3K/Akt pathways respectively [28, 29]. In addition to Bad phosphorylation, knock-down of HDGF was found to reduce the phosphorylation of ERK and Akt. EGF induced the ERK and Akt phosphorylation and at the same time induced the phosphorylation of Bad at Ser-112 and Ser-136 in HepG2 cells (Fig. 7). Besides this, EGF was found to reverse the decrease in phosphorylation of ERK and Akt upon HDGF knock-down (Fig. 10b). Since EGF but not forskolin can suppress the level of apoptosis induced by HDGF knock-down (Fig. 8), it further supports that inactivation of the Ras/Raf/MEK/ERK and PI3K/Akt may be upstream to Bad dephosphorylation at Ser-112 and Ser-136 that eventually leads to activation of the intrinsic apoptotic pathway upon HDGF knock-down. On the other hand, the role of protein phosphatases that are known to induce dephosphorylation of Bad is not yet defined.

The results of our study indicate that HDGF is important for the growth of human cancer cells probably through the regulation of the intracellular signaling pathways. However, the mechanism of how HDGF regulates the signaling pathways remains to be further investigated. It was suggested that HDGF may act in an autocrine or paracrine manner that it may be secreted out of the cells and bind on the cell surface receptor to activate the intracellular signaling pathways [9]. However, the receptor of HDGF has not yet been identified and this hypothetic mechanism is still needed to be further confirmed. Besides this, knock-down of HDGF may alter the expression of genes that regulate the intracellular signaling pathway. According to Zhang’s study in lung cancer cells [25], knock-down of HDGF was associated with the down-regulation of AXL, a receptor tyrosine kinase, and thus affects the intracellular signaling pathway. On the other hand, HDGF may regulate the gene transcription directly as previously HDGF was detected in the nucleus of cells and the ability of HDGF translocation into the nucleus is important for its growth promoting activity [10]. It was also reported that HDGF may regulate gene expression directly by binding with the promoter region of the target genes [37].

Although the mechanism for HDGF to regulate the intracellular signaling pathways is yet to be defined, our study however indicates that HDGF may act as a survival factor for human cancer cells. This may account for the poor prognosis and low survival rate for those cancer patients with HDGF over-expression. Furthermore, the study also indicates the possible use of HDGF as the potential target in cancer therapy.

Notes

Acknowledgements

This work was supported by a Grant from the Shanghai-Hong Kong Anson Research Foundation, Hong Kong.

Supplementary material

10495_2008_241_MOESM1_ESM.tif (170 kb)
Supplementary Fig. 1 Apoptosis induction in HepG2 cells after knock-down of HDGF with HDGF specific siRNA. The cells were transfected with HDGF siRNA (Hi) and scramble siRNA (S) for 24h. The antisense strand sequence for the HDGF siRNA is GUAUUUGUUGGCUGUUGAU-dTdT. (A) After transfection, the cells were allowed to grow for 72h followed by DNA fragmentation assay. (B) The cells were allowed to grow for 48h after transfection followed by annexin V/propidium iodide staining assay. (C) The expression level of HDGF after transfection with HDGF siRNA was determined by Western blot analysis. The experiments have been repeated at least 3 times with similar results. The one shown is the representative one (TIF 170 kb)
10495_2008_241_MOESM2_ESM.tif (141 kb)
Supplementary Fig. 2 Apoptosis in HepG2 cells with knock-down of HDGF and/or Bad. The cells were transfected with HDGF antisense oligos with or without the Bad antisense oligos. After transfection, the cells were incubated for 48h followed by Annexin V binding assay. The % increase in apoptotic cells is calculated by subtracting the % of apoptotic cells in the group transfected with HDGF antisense oligos from that of the corresponding group transfected with the sense oligos. H-AS: transfection with HDGF antisense oligos only. H+B-AS: co-transfection with HDGF and Bad antisense oligos. The data are average from three independent experiments. Mean±SEM. *p<0.05, significantly different from those transfected with H-AS (TIF 141 kb)

References

  1. 1.
    Nakamura H, Kambe H, Egawa T, Kimura Y, Ito H, Hayashi E et al (1989) Partial purification and characterization of human hepatoma-derived growth factor. Clin Chim Acta 183:273–284. doi: 10.1016/0009-8981(89)90361-6 PubMedCrossRefGoogle Scholar
  2. 2.
    Sue SC, Chen JY, Lee SC, Wu WG, Huang TH (2004) Solution structure and heparin interaction of human hepatoma-derived growth factor. J Mol Biol 343:1365–1377. doi: 10.1016/j.jmb.2004.09.014 PubMedCrossRefGoogle Scholar
  3. 3.
    Abouzied MM, Baader SL, Dietz F, Kappler J, Gieselmann V, Franken S (2004) Expression patterns and different subcellular localization of the growth factors HDGF (hepatoma-derived growth factor) and HRP-3 (HDGF-related protein-3) suggest functions in addition to their mitogenic activity. Biochem J 378:169–176. doi: 10.1042/BJ20030916 PubMedCrossRefGoogle Scholar
  4. 4.
    Nakamura H, Izumoto Y, Kambe H, Kuroda T, Mori T, Kawamura K et al (1994) Molecular cloning of complementary DNA for a novel human hepatoma-derived growth factor. its homology with high mobility group-1 protein. J Biol Chem 269:25143–25149PubMedGoogle Scholar
  5. 5.
    Everett AD, Lobe DR, Matsumura ME, Nakamura H, McNamara CA (2000) Hepatoma-derived growth factor stimulates smooth muscle cell growth and is expressed in vascular development. J Clin Invest 105:567–575. doi: 10.1172/JCI7497 PubMedCrossRefGoogle Scholar
  6. 6.
    Oliver JA, Al-Awqati Q (1998) An endothelial growth factor involved in rat renal development. J Clin Invest 102:1208–1219. doi: 10.1172/JCI785 PubMedCrossRefGoogle Scholar
  7. 7.
    Enomoto H, Yoshida K, Kishima Y, Kinoshita T, Yamamoto M, Everett AD et al (2002) Hepatoma-derived growth factor is highly expressed in developing liver and promotes fetal hepatocyte proliferation. Hepatology 36:1519–1527PubMedGoogle Scholar
  8. 8.
    Machuy N, Thiede B, Rajalingam K, Dimmler C, Thieck O, Meyer TF et al (2005) A global approach combining proteome analysis and phenotypic screening with RNA interference yields novel apoptosis regulators. Mol Cell Proteomics 4:44–55. doi: 10.1074/mcp.M400089-MCP200 PubMedGoogle Scholar
  9. 9.
    Zhou Z, Yamamoto Y, Sugai F, Yoshida K, Kishima Y, Sumi H et al (2004) Hepatoma-derived growth factor is a neurotrophic factor harbored in the nucleus. J Biol Chem 279:27320–27326. doi: 10.1074/jbc.M308650200 PubMedCrossRefGoogle Scholar
  10. 10.
    Kishima Y, Yamamoto H, Izumoto Y, Yoshida K, Enomoto H, Yamamoto M et al (2002) Hepatoma-derived growth factor stimulates cell growth after translocation to the nucleus by nuclear localization signals. J Biol Chem 277:10315–10322. doi: 10.1074/jbc.M111122200 PubMedCrossRefGoogle Scholar
  11. 11.
    Nameki N, Tochio N, Koshiba S, Inoue M, Yabuki T, Aoki M et al (2005) Solution structure of the PWWP domain of the hepatoma-derived growth factor family. Protein Sci 14:756–764. doi: 10.1110/ps.04975305 PubMedCrossRefGoogle Scholar
  12. 12.
    Yoshida K, Nakamura H, Okuda Y, Enomoto H, Kishima Y, Uyama H et al (2003) Expression of hepatoma-derived growth factor in hepatocarcinogenesis. J Gastroenterol Hepatol 18:1293–1301. doi: 10.1046/j.1440-1746.2003.03191.x PubMedCrossRefGoogle Scholar
  13. 13.
    Hu TH, Huang CC, Liu LF, Lin PR, Liu SY, Chang HW et al (2003) Expression of hepatoma-derived growth factor in hepatocellular carcinoma. Cancer 98:1444–1456. doi: 10.1002/cncr.11653 PubMedCrossRefGoogle Scholar
  14. 14.
    Yoshida K, Tomita Y, Okuda Y, Yamamoto S, Enomoto H, Uyama H et al (2006) Hepatoma-derived growth factor is a novel prognostic factor for hepatocellular carcinoma. Ann Surg Oncol 13:159–167. doi: 10.1245/ASO.2006.11.035 PubMedCrossRefGoogle Scholar
  15. 15.
    Huang JS, Chao CC, Su TL, Yeh SH, Chen DS, Chen CT et al (2004) Diverse cellular transformation capability of overexpressed genes in human hepatocellular carcinoma. Biochem Biophys Res Commun 315:950–958. doi: 10.1016/j.bbrc.2004.01.151 PubMedCrossRefGoogle Scholar
  16. 16.
    El-Rifai W, Frierson HF Jr, Harper JC, Powell SM, Knuutila S (2001) Expression profiling of gastric adenocarcinoma using cDNA array. Int J Cancer 92:832–838. doi: 10.1002/ijc.1264 PubMedCrossRefGoogle Scholar
  17. 17.
    Lepourcelet M, Tou L, Cai L, Sawada J, Lazar AJ, Glickman JN et al (2005) Insights into developmental mechanisms and cancers in the mammalian intestine derived from serial analysis of gene expression and study of the hepatoma-derived growth factor (HDGF). Development 132:415–427. doi: 10.1242/dev.01579 PubMedCrossRefGoogle Scholar
  18. 18.
    Ren H, Tang X, Lee JJ, Feng L, Everett AD, Hong WK et al (2004) Expression of hepatoma-derived growth factor is a strong prognostic predictor for patients with early-stage non-small-cell lung cancer. J Clin Oncol 22:3230–3237. doi: 10.1200/JCO.2004.02.080 PubMedCrossRefGoogle Scholar
  19. 19.
    Bernard K, Litman E, Fitzpatrick JL, Shellman YG, Argast G, Polvinen K et al (2003) Functional proteomic analysis of melanoma progression. Cancer Res 63:6716–6725PubMedGoogle Scholar
  20. 20.
    Yamamoto S, Tomita Y, Hoshida Y, Morii E, Yasuda T, Doki Y et al (2007) Expression level of hepatoma-derived growth factor correlates with tumor recurrence of esophageal carcinoma. Ann Surg Oncol 14:2141–2149. doi: 10.1245/s10434-007-9369-9 PubMedCrossRefGoogle Scholar
  21. 21.
    Yamamoto S, Tomita Y, Hoshida Y, Takiguchi S, Fujiwara Y, Yasuda T et al (2006) Expression of hepatoma-derived growth factor is correlated with lymph node metastasis and prognosis of gastric carcinoma. Clin Cancer Res 12:117–122. doi: 10.1158/1078-0432.CCR-05-1347 PubMedCrossRefGoogle Scholar
  22. 22.
    Uyama H, Tomita Y, Nakamura H, Nakamori S, Zhang B, Hoshida Y et al (2006) Hepatoma-derived growth factor is a novel prognostic factor for patients with pancreatic cancer. Clin Cancer Res 12:6043–6048. doi: 10.1158/1078-0432.CCR-06-1064 PubMedCrossRefGoogle Scholar
  23. 23.
    Okuda Y, Nakamura H, Yoshida K, Enomoto H, Uyama H, Hirotani T et al (2003) Hepatoma-derived growth factor induces tumorigenesis in vivo through both direct angiogenic activity and induction of vascular endothelial growth factor. Cancer Sci 94:1034–1041. doi: 10.1111/j.1349-7006.2003.tb01397.x PubMedCrossRefGoogle Scholar
  24. 24.
    Kishima Y, Yoshida K, Enomoto H, Yamamoto M, Kuroda T, Okuda Y et al (2002) Antisense oligonucleotides of hepatoma-derived growth factor (HDGF) suppress the proliferation of hepatoma cells. Hepatogastroenterology 49:1639–1644PubMedGoogle Scholar
  25. 25.
    Zhang J, Ren H, Yuan P, Lang W, Zhang L, Mao L (2006) Down-regulation of hepatoma-derived growth factor inhibits anchorage-independent growth and invasion of non-small cell lung cancer cells. Cancer Res 66:18–23. doi: 10.1158/0008-5472.CAN-04-3905 PubMedCrossRefGoogle Scholar
  26. 26.
    Bae J, Hsu SY, Leo CP, Zell K, Hsueh AJ (2001) Underphosphorylated BAD interacts with diverse antiapoptotic bcl-2 family proteins to regulate apoptosis. Apoptosis 6:319–330. doi: 10.1023/A:1011319901057 PubMedCrossRefGoogle Scholar
  27. 27.
    Tan Y, Demeter MR, Ruan H, Comb MJ (2000) BAD ser-155 phosphorylation regulates BAD/Bcl-XL interaction and cell survival. J Biol Chem 275:25865–25869. doi: 10.1074/jbc.M004199200 PubMedCrossRefGoogle Scholar
  28. 28.
    Fang X, Yu S, Eder A, Mao M, Bast RC Jr, Boyd D et al (1999) Regulation of BAD phosphorylation at serine 112 by the ras-mitogen-activated protein kinase pathway. Oncogene 18:6635–6640. doi: 10.1038/sj.onc.1203076 PubMedCrossRefGoogle Scholar
  29. 29.
    Lizcano JM, Morrice N, Cohen P (2000) Regulation of BAD by cAMP-dependent protein kinase is mediated via phosphorylation of a novel site, Ser155. Biochem J 349:547–557. doi: 10.1042/0264-6021:3490547 PubMedCrossRefGoogle Scholar
  30. 30.
    Ayllon V, Martinez-A C, Garcia A, Cayla X, Rebollo A (2000) Protein phosphatase 1alpha is a ras-activated bad phosphatase that regulates interleukin-2 deprivation-induced apoptosis. EMBO J 19:2237–2246. doi: 10.1093/emboj/19.10.2237 PubMedCrossRefGoogle Scholar
  31. 31.
    Chiang CW, Harris G, Ellig C, Masters SC, Subramanian R, Shenolikar S et al (2001) Protein phosphatase 2A activates the proapoptotic function of BAD in interleukin- 3-dependent lymphoid cells by a mechanism requiring 14–3-3 dissociation. Blood 97:1289–1297. doi: 10.1182/blood.V97.5.1289 PubMedCrossRefGoogle Scholar
  32. 32.
    Wang Q, Huang Y, Ni Y, Wang H, Hou Y (2007) siRNA targeting midkine inhibits gastric cancer cells growth and induces apoptosis involved caspase-3, 8, 9 activation and mitochondrial depolarization. J Biomed Sci 14:783–795. doi: 10.1007/s11373-007-9192-0 PubMedCrossRefGoogle Scholar
  33. 33.
    Schinzel A, Kaufmann T, Borner C (2004) Bcl-2 family members: Integrators of survival and death signals in physiology and pathology. Biochim Biophys Acta 1644:95–105. corrected. doi: 10.1016/j.bbamcr.2003.09.006 Google Scholar
  34. 34.
    Donovan M, Cotter TG (2004) Control of mitochondrial integrity by bcl-2 family members and caspase-independent cell death. Biochim Biophys Acta 1644:133–147. doi: 10.1016/j.bbamcr.2003.08.011 PubMedCrossRefGoogle Scholar
  35. 35.
    Datta SR, Katsov A, Hu L, Petros A, Fesik SW, Yaffe MB et al (2000) 14-3-3 proteins and survival kinases cooperate to inactivate BAD by BH3 domain phosphorylation. Mol Cell 6:41–51. doi: 10.1016/S1097-2765(00)00006-X PubMedCrossRefGoogle Scholar
  36. 36.
    Zha J, Harada H, Yang E, Jockel J, Korsmeyer SJ (1996) Serine phosphorylation of death agonist BAD in response to survival factor results in binding to 14-3-3 not BCL-X(L). Cell 87:619–628. doi: 10.1016/S0092-8674(00)81382-3 PubMedCrossRefGoogle Scholar
  37. 37.
    Yang J, Everett AD (2007) Hepatoma-derived growth factor binds DNA through the N-terminal PWWP domain. BMC Mol Biol 8:101. doi: 10.1186/1471-2199-8-101 PubMedCrossRefGoogle Scholar

Copyright information

© Springer Science+Business Media, LLC 2008

Authors and Affiliations

  • Tsun Yee Tsang
    • 1
  • Wan Yee Tang
    • 1
    • 2
  • Wing Pui Tsang
    • 1
  • Ngai Na Co
    • 1
  • Siu Kai Kong
    • 1
  • Tim Tak Kwok
    • 1
  1. 1.Department of BiochemistryThe Chinese University of Hong KongShatinHong Kong SAR, China
  2. 2.Department of Environmental HealthUniversity of Cincinnati Medical CenterCincinnatiUSA

Personalised recommendations