SEARCH

SEARCH BY CITATION

Abstract

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

The pituitary tumor transforming (PTTG) gene family comprises PTTG1, 2, and 3. Forced expression of PTTG1 (securin) induces cellular transformation and promotes tumor development in animal models. PTTG1 is overexpressed in various human cancers. However, the expression and pathogenic implications of the PTTG gene family in hepatocellular carcinoma are largely unknown. Gene silencing using short interfering RNA (siRNA) has become an efficient means to study the functions of genes and has been increasingly used for cancer gene therapy approaches. We report that PTTG1, but not PTTG2 and 3, was highly and frequently expressed in liver cancer tissues from patients and highly in SH-J1, SK-Hep1, and Huh-7 hepatoma cell lines. Adenoviral vector encoding siRNA against PTTG1 (Ad.PTTG1-siRNA) depleted PTTG1 specifically and efficiently in SH-J1 hepatoma cells, which resulted in activation of p53 that led to increased p21 expression and induction of apoptosis. The depletion of PTTG1 in HCT116 colorectal cancer cells exhibited a cytotoxic effect in a p53-dependent manner. Ad.PTTG1-siRNA-mediated cytotoxic effect was dependent on expression levels of PTTG1 and p53 in hepatoma cell lines. Huh-7 hepatoma cells, once transduced with Ad.PTTG1-siRNA, displayed markedly attenuated growth potential in nude mice. Intra-tumor delivery of Ad.PTTG1-siRNA led to significant inhibition of tumor growth in SH-J1 tumor xenograft established in nude mice. In conclusion, PTTG1 overexpressed in hepatoma cell lines negatively regulates the ability of p53 to induce apoptosis. PTTG1 gene silencing using siRNA may be an effective modality to treat liver cancer, in which PTTG1 is abundantly expressed. (HEPATOLOGY 2006;43:1042–1052.)

Hepatocellular carcinoma (HCC) is one of the most prevalent human cancers worldwide, with 600,000 estimated new cases annually and almost as many deaths.1 Much is known about the development and causes of HCC.1, 2 Nevertheless, no effective therapy has been found for the vast majority of HCC patients. Therefore, novel approaches to treat liver cancers are needed.2

RNA interference is a sequence-specific posttranscriptional gene silencing mechanism, by which double-stranded RNA inhibits gene expression by degradation of the corresponding mRNA.3 Double-stranded RNA is processed into 21- to 23- nucleotides short interfering RNA (siRNA) that recognizes and cleaves the cognate mRNA.3 Because of the specificity and efficiency of siRNAs in gene silencing, siRNA expressing vectors or synthetic siRNA oligonucleotides targeting oncogenes overexpressed or highly active in tumor cells have been increasingly examined for the development of cancer therapeutics.4

The pituitary tumor transforming gene 1 (PTTG1) is highly expressed in a number of human cancers.5–9 Overexpression of PTTG1 in NIH3T3 cells induces cellular transformation and promotes tumor formation in nude mice.10, 11 Clinical studies show that fibroblast growth factor 2 and vascular endothelial growth factor are elevated in pituitary tumors and mostly correlate with PTTG levels, suggesting the role of PTTG in angiogenesis.12 PTTG1 is one of 17 genes representing metastasis in solid tumors.13 Thus, PTTG1 is considered an oncogene for pituitary tumors and other neoplasia.12

PTTG1 has a variety of functions in multiple cellular processes, such as mitosis, DNA repair, apoptosis, and gene regulation. In particular, PTTG1, associated with separase, prevents sister chromatids from separation until degradation of PTTG1 by anaphase-promoting complex in metaphase to anaphase transition.14 Aberrant expression of PTTG1 disrupts mitosis and causes aneuploidy and genetic instability.14–17 Aneuploidy is a common characteristic of tumors and has also been proposed as a necessary event for tumorigenesis.18 Nevertheless, viability and mild phenotypes of PTTG1 knockout mice and human cells strongly suggest that PTTG1 may be dispensable for regulation of sister chromatid and other normal cellular processes as well.15, 19, 20 PTTG1 associates with p53, and modulates p53-mediated transcriptional activity and apoptosis.21, 22 PTTG1, a p53 target gene,23 also induces p53-independent apoptosis21 and modulates expression of p53.24

Attenuation of PTTG1 expression using synthesized antisense or siRNA oligonucleotides induces apoptosis in HeLa cervical25 and HCT116 colorectal cancer cells in vitro26 and inhibits growth of U87 astrocytoma cells in culture.27

PTTG1 is 91% and 89% identical with PTTG2 and PTTG3 at the amino acid level, respectively.28 PTTG2 expression was detected in a pair of liver tumor and non-tumor tissues.28 Nevertheless, the expression and implication of the PTTG1, 2, and 3 in HCCs remain unclear. Here, we show that PTTG1, but not PTTG2 and 3, associates with HCCs and is highly expressed in some hepatoma cell lines. By using adenoviral vector encoding siRNA targeting PTTG1, we explored the role of PTTG1 in HCC and its usefulness as a potential target for gene therapy of HCC in culture and mice.

Materials and Methods

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

Human Samples.

Primary HCC tissues were obtained from patients who underwent surgical treatment for HCC at Kangnam St. Mary's Hospital (Seoul, South Korea) and Chonbuk National University Hospital (Jeonju, South Korea). Normal liver tissues from patients with other diseases were obtained from Kangnam St. Mary's Hospital. Informed consent was obtained from patients for the use of specimens for research purpose only. Specimens were frozen in liquid nitrogen and stored until use.

Plasmid and Adenoviral Vectors.

Flag-tagged PTTG1 expression vector, pFlag-PTTG1, was constructed by polymerase chain reaction (PCR)-amplifying from pGST-PTTG1 (supplied by 21C Frontier Human Gene Bank, South Korea) and cloning into the EcoRI/BamHI sites of pFlagCMV2 vector (Sigma, St. Louis, MO). pFlag-PTTG2 and -PTTG3 expression vectors were constructed by PCR-amplifying from pPTTG2-GFP and pPTTG3-GFP and cloning into the NheI-NotI sites of PCR259 vector (Q-Biogene, Carlsbad, CA). Oligonucleotides coding for PTTG1-siRNA and its mutants (see Fig. 2A) were synthesized, and inserted into pSuper vector containing H1 promoter and T5 terminal sequences29 by a standard cloning procedure. The resulting plasmids were digested with XbaI and HindIII. The DNA fragments containing H1 promoter, siRNA sequences, and T5, were then inserted into XbaI/HindIII site of pShuttle (Quantum). Adenoviral vectors were generated by recombination of pShuttle plasmids encoding relevant siRNAs and pAdeasy1 in Escherichia coli BJ5183. The resulting DNAs were digested with PacI and transfected into 293 cells. Correct recombinant adenoviruses were plaque-purified. Ad.LacZ encoding β-galactosidase was supplied by Q-Biogene. Viral titer was determined by a 293-plaque assay. Adenoviral vectors for animal experiment were purified as previously reported.30

Cell Culture, Transfections, and Reporter Assay.

Cells were grown in Dulbecco's minimum essential medium with high glucose (4.5 g/L) supplemented with 10% fetal calf serum and antibiotics in a humidified incubator at 37°C in a 5% CO2 atmosphere. Cells were transfected with relevant plasmids by a standard calcium-phosphate method. Luciferase assay was carried out according to the manufacturer's instruction (Promega, Madison, WI).

Immunoblotting, FACS Analysis, and Cytotoxicity Assay.

Cell lysates were prepared by lysing cells in RIPA buffer [50 mmol/L Tris-HCl (pH 8.0), 150 mmol/L NaCl, 1% NP-40, 0.5% sodium deoxycholate, 0.1% sodium dodecyl sulfate, 0.5 mmol/L phenylmethylsulphonylfluoride], and resolved by sodium dodecyl sulfate polyacrylamide gel electrophoresis. Immunoblotting was performed on polyvinylidene fluoride membranes (Millipore, Billerica, MA) according to the manufacturer's instructions. Commercially available antibodies to PTTG1 (Santa Cruz), Flag epitope, β-actin (Sigma), p53, and p21 (Pharmingen, San Jose, CA) were used as recommended by the manufacturers. Signals were developed using Enhanced Chemiluminescence Reagent (Amersham, Castle Hill, New South Wales, Australia).

Cells were transduced with recombinant adenoviruses in serum-free Dulbecco's minimum essential medium for 2 hours with frequent gentle shaking. At 48 hours post-transduction, cells were washed with phosphate-buffered saline (PBS), fixed with 70% ethanol, and stained with PI buffer (10 μg/mL propidium iodide and 1 mg/mL RNase A in 1.1% sodium citrate) at 37°C for 30 minutes. FACS analysis were performed on FACS calibur instrument (Becton Dickinson, Franklin Lakes, NJ).

Cells were seeded on 12-well plates, incubated for 24 hours, and then transduced with adenoviruses at various multiplicities of infection (MOIs). The transduced cells were incubated for 48 hours, before viable cells were fixed with 10% formaldehyde, and stained with 1% crystal violet. The plates were dried, and images were captured with a digital camera.

Animal Experiments.

Five-week-old female BALB/c nude mice were purchased from the animal breeding laboratory at Korea Research Institute of Bioscience and Biotechnology. All mice were fed ad libitum and received humane care in compliance with the Korean NIH guidelines for the care and use of laboratory animals in research. In tumorigenicity experiment, Huh-7 cells were transduced with or without Ad.PTTG1-siRNA or Ad.PTTG1-siRNA1M and washed with PBS. Immediately after being washed, the cells were resuspended in 100 μL PBS for one site and injected subcutaneously into two sites on the right and left abdomen of each mouse. Tumor growth was monitored after injection of tumor cells.

In the in vivo treatment experiments, SH-J1 cells (107) were suspended in 100 μL PBS and injected subcutaneously into the right flank of each mouse. When the tumor size reached approximately 3 to 5 mm in diameter, Ad.PTTG1-siRNA or Ad.PTTG1-siRNA1M in 100 μL PBS, or 100 μL PBS was injected intratumorally into mice. Tumors were monitored for length and width. Tumor volume was calculated according to the following equation: V (mm3) = width2 (mm2) × length (mm)/2.

Statistics.

Statistical analysis was carried out using the unipolar, paired Student t test and the two-sided chi-square test. Data were considered statistically significant when the P value was less than .05.

Results

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

PTTG1 Is Highly and Frequently Expressed in Liver Tumor Tissues From Patients.

We analyzed relative expression levels of PTTG1 between dysplastic nodule and grade II/III samples from microarray data31, 32 (Fig. 1A). PTTG1 expression level was increased 2.9-fold in grades II/III compared with dysplastic nodule stage (P = 3.5 × 10−11), suggesting that PTTG1 expression associates with HCC progression. PTTG2 and PTTG3 were suggested to be associated with tumorigenesis because of their amino acid similarity to PTTG1.28 To determine which PTTG gene associated with HCC, expression levels of PTTG1, 2, and 3 were analyzed in 44 HCCs and 4 normal liver tissues by semi-quantitative reverse transcription polymerase chain reaction (RT-PCR) (Fig. 1B). The PTTG genes are highly homologous at the nucleotide level.28 We therefore tested primers for specific detection of each PTTG gene expression and found that each primer set specifically detected each corresponding PTTG gene (Supplementary Fig. 1A-B). Expression levels of PTTG genes were normalized on the basis of those of β-actin gene and quantified (Supplementary Fig. 1C). The expression of PTTG1 gene in 39 cases of 44 HCCs was increased more than twofold compared with normal liver tissues, whereas PTTG2 and 3 genes were expressed more than twofold in only 7 and 4 cases of 44 HCCs, respectively. Because PTTG1 is susceptible to ubiquitin-mediated proteolysis,14 PTTG1 expression at transcript level would not necessarily represent its functionality in HCC. To this end, we performed tissue array experiment. PTTG1 protein was more intensively detected in seven cases of nine HCCs than in corresponding normal liver tissues (Fig. 1C) and relatively more expressed in 29 cases of 41 primary and metastatic HCCs, including cholangiocarcinoma (Supplementary Fig. 2).

thumbnail image

Figure 1. PTTG1 is expressed frequently and highly in human liver tumor tissues. (A) PTTG1 expression associates with HCC progression. Boxplot of the expression levels of PTTG1 in HCC samples31 is shown. Fifty samples were divided into 3 groups according to Edmonson classification32: 20, 10, and 20 samples in dysplastic nodule, grade I, and grade II/III. The group median and range of the log-ratio (based 2) values were calculated using a statistical package R (www.r-project.org). Each box represents the range that the middle half of the sorted values occupy, with the position of median marked by a thick horizontal line inside the box. The circles represent extreme outliers. (B) PTTG1, but not PTTG2 and PTTG3, is frequently and highly expressed in liver tumor tissues from HCC patients. Total RNAs were isolated by using the RNeasy kit (Qiagen, Hilden, Germany) from four normal liver tissue (NL) and 44 HCC tissues (T). Reverse transcription (RT) was performed with 5 μg RNA, 500 ng oligo-d(T), and 2 μL (400 units) RT-PCR Superscript II (Invitrogen) at 42°C for 1 hour. The RT products (cDNAs) were diluted with distilled water up to 30-fold. PCR amplification was then performed in a volume of 20 μL reaction mixtures containing 2 μL each diluted RT product, and 50 pmol each primer for PTTG1, 2, or 3 genes, or β-actin gene. The PCR conditions were 94°C for 1 minute, 65°C for 1 minute, and 72°C for 1 minute for 32 cycles (PTTG1), 34 cycles (PTTG2 and 3), or 27 cycles (β-actin). The primer sequences for PTTG gene family are shown in Supplementary Fig. 1B. The primer sequences for β-actin are 5′-CTGGAGAAGAGCTACGAGCTGC-3′ (sense) and 5′CTAGAAGCATTTGCGGTGGACG-3′ (antisense). PCR products were separated by electrophoresis on 2% agarose gel and visualized under ultraviolet light after ethidium bromide (EtBr) staining. Beta-actin was used as an internal control to confirm equal amount of templates. (C) PTTG1 protein is expressed highly in human liver tumor tissues. Tissue array (SuperBioChips Lab, South Korea) contains 40 human primary, 10 metastatic liver tumor samples, and 9 normal liver tissues adjacent to or apart from corresponding tumor tissues. Formalin-fixed, paraffin-embedded tissue arrays were dewaxed 3 times with xylene for 5 minutes, hydrated by immersing serially in 100%, 90%, and 80% ethanol solutions for 5 minutes, washed with tap water for 5 minutes, immersed in a citrate buffer (10 mmol/L citrate, pH 6.0), boiled twice in a microwave oven for 5 minutes, washed with PBS, and blocked with 1% serum albumin for 1 hour at room temperature. The slide was incubated with rabbit anti-PTTG1 (1:100) antibody overnight at 4°C, washed with PBS, incubated with anti-rabbit IgG conjugated to fluorescein isothiocyanate (FITC) for 1 hour at room temperature, washed 3 times with PBS, mounted (VECTASHEILD), and photographed under a fluorescence microscope (Zeiss). After hydration, the tissue array was stained with hematoxylin-eosin (HE) (Sigma) and photographed under a light microscope. Immunofluorescence and HE staining of the nine-paired tumor tissues and corresponding normal tissues are shown. The rest is shown in Supplementary Fig. 2.

Download figure to PowerPoint

Gene Silencing Effect of PTTG1-siRNA–Expressing Vector.

We constructed plasmid expression vectors encoding siRNA against PTTG1 or PTTG1-siRNA with two- or four-point mutations as controls (Fig. 2A) and tested whether the PTTG1-siRNA expression vector specifically induced PTTG1 gene silencing (Fig. 2B). pPTTG1-siRNA, but not pPTTG1-siRNA1M or -siRNA2M depleted Flag-PTTG1. Because PTTG1-siRNA could target PTTG2 because of the target-sequence identity to PTTG2, we also examined whether pPTTG1-siRNA targeted PTTG2. pPTTG1-siRNA depleted PTTG2, but not PTTG3 (Fig. 2C). Because PTTG1, but not PTTG2 and 3, was predominantly expressed in SH-J1 cells (Fig. 5A-B), we examined the effect of PTTG1 depletion on growth of invasive sarcoma SH-J1 hepatoma cell line.33 After transfection of equal numbers of SH-J1 cells with pPTTG1-siRNA or -siRNA1M, viable cells were counted at the indicated times. pPTTG1-siRNA, but not pPTTG1-siRNA1M inhibited the cell growth (Fig. 2D), suggesting that PTTG1 may positively regulate cell proliferation.

thumbnail image

Figure 2. Functionality of PTTG1-siRNA expression vector. (A) Schematic illustration of expression vectors expressing siRNA against PTTG1 and its siRNA mutants. pPTTG1-siRNA1M or -siRNA2M contain two- or four-point mutations, in which mutated nucleotides are indicated in boldface. (B) pPTTG1-siRNA, but not pPTTG1-siRNA1M or -siRNA2M, depletes Flag-PTTG1. 293T cells (5 × 106) were transfected with Flag-PTTG1 expression plasmid (10 μg) and incubated for 16 hours. An equivalent number of the cells were then plated in 6-well plates, transfected with the indicated siRNA expression constructs (0, 5, 10 μg), and incubated for 48 hours. Cell lysates were immunoblotted as indicated. (C) pPTTG1-siRNA depletes Flag-PTTG2, but not Flag-PTTG3. Flag-PTTG1, 2, or 3 expression vectors were transfected into 293 cells (20 μg/100-mm plate), and incubated for 16 hours. Equal number of the cells were plated in four 100-mm plates, transfected with the siRNA expression vectors (0, 10, 15 μg), incubated for 32 hours, lysed in RIPA buffer, and immunoblotted as indicated. (D) Transfection of pPTTG1-siRNA inhibits the growth of SH-J1 hepatoma cells. The cells were transfected with or without the indicated siRNA expression vectors (10 μg/100-mm plate), and incubated for 12 hours. Equivalent numbers of SH-J1 cells were plated in 6-well plates (5 × 103 cells/well). Viable cells were counted at the indicated times. Data are the mean ± SD of three independent experiments in duplicate.

Download figure to PowerPoint

Depletion of PTTG1 Activates p53 and Induces Apoptosis in SH-J1 Hepatoma Cells.

Because adenoviral vectors transduce genes efficiently into many types of tumor cells, we constructed Ad.PTTG1-siRNA and Ad.PTTG1-siRNA1M with two-point mutations in the siRNA sequence as a control (Fig. 3A). PTTG1, 2, and 3 proteins are highly homologous at the level.28 We therefore tested the specificity of anti-PTTG1 antibody (Fig. 3B). The anti-PTTG1 antibody reacted much more strongly with Flag-PTTG1 compared with Flag-PTTG2 and 3, suggesting that protein band reacted with anti-PTTG1 antibody mostly represents PTTG1, but not PTTG2 and 3. Transduction of SH-J1 cells with Ad.PTTG1-siRNA, but not Ad.PTTG1-siRNA1M, depleted PTTG1 (Fig. 3C, top panel), indicating that expressed siRNA depleted endogenous PTTG1 specifically, like it depleted overexpressed Flag-PTTG1 (Fig. 2B).

thumbnail image

Figure 3. Adenovirus-mediated transfer of PTTG1-siRNA leads to activation of p53 and induces apoptosis. (A) Schematic illustration of adenoviral vectors encoding PTTG1-siRNA or PTTG1-siRNA1M. The E1 region of Ad type 5 genome was replaced with indicated siRNA expression cassettes. (B) The specificity of anti-PTTG1 antibody. 293 cells were transfected with the indicated vectors (10 μg/100-mm plate), incubated for 48 hours, lysed in RIPA buffer, and immunoblotted as indicated. (C) Ad.PTTG1-siRNA specifically depletes PTTG1 and activates p53, thereby inducing p21. SH-J1 hepatoma cells were transduced with or without the indicated adenoviral vectors at the indicated multiplicity of infection (MOI), incubated for 24 hours, and immunoblotted as indicated. (D) SH-J1 hepatoma cells contain functional p53. The indicated cells were plated in 6-well plates, transfected with the indicated reporter plasmids (0.5 μg/well), and incubated for 24 hours. Luciferase assay was performed with equivalent amounts of cell lysates. p53RE-luc; the reporter plasmid harboring a fire-fly luciferase under the control of p53 response element (RE). p21-luc or pBax-luc; the reporter plasmids encoding the luciferase gene under the control of the promoters of Bax or p21 genes. (E) PTTG1 depletion increases the activity of luciferase reporter gene under the transcriptional control of p53RE or the p21 promoter. SH-J1 hepatoma cells (5 × 106) were transfected with the indicated reporter plasmids (5 μg/100-mm plate), and incubated for 12 hours. An equivalent number of the cells were plated in 6-well plates, transduced with or without the indicated viruses at the indicated MOI, and incubated for 24 hours. Luciferase assay was performed with equivalent amounts of cell lysates. Data described in (D-E) are the means ± SD of three independent experiments in duplicate. (F) PTTG1 depletion induces apoptosis in SH-J1 hepatoma cells. Cells were transduced with or without the indicated viruses at an MOI of 50. After a 24-hour further incubation, cells were stained with propidium iodide and analyzed by using FACS. M1 stands for population of the cells with sub-2n DNA. The experiment was repeated more than twice, and a representative is shown.

Download figure to PowerPoint

PTTG1 induces p53-dependent apoptosis.21, 22 We therefore examined whether PTTG1 depletion-mediated growth inhibition (Fig. 2D) associated with functionality of p53. SH-J1 cells are susceptible to oxidative stress–mediated apoptosis.33 However, whether SH-J1 cells have functional p53 is unknown. To this end, we examined the effect of PTTG1 depletion on endogenous levels of p53 and p21. p21 is one of many genes to be induced by p53 and functions as a cyclin-dependent kinase inhibitor, which induces in cell-cycle arrest at G1 phase.34 p53 level was not markedly affected by PTTG1 depletion, but p21 level was increased in a viral dose-dependent manner (Fig. 3C, second and third panels). We further examined functionality of p53 in SH-J1 cells by assaying the activities of luciferase (luc) reporter constructs such as p53RE-luc, pBax-luc, and p21-luc, in which the reporter gene expression is driven by the p53 response-element (RE), or p21 or Bax gene promoters. The Bax gene is induced by p53 and encodes a proapoptotic member of the Bcl2 gene family involved in the activation of the mitochondria-related death effector complex.35 We transfected HCT116, its p53-null (HCT116Δp53) isogenic colorectal, and SH-J1 cancer cells with the indicated reporter constructs (Fig. 3D). The luciferase assay was performed with equivalent amounts of cell lysates. The reporter gene activities were much higher in HCT116 than in HCT116Δp53 cells, suggesting that the reporter constructs can serve as a means to assess functionality of p53. The reporter activities of p53RE-luc and pBax-luc in SH-J1 cells were slightly lower than those in HCT116 cells, but much higher than those in HCT116Δp53 cells, suggesting that p53 is active in SH-J1 cells. Next, we transfected SH-J1 cells with p53RE-luc or p21-luc reporter constructs. Equivalent number of cells were plated and then transduced with or without adenoviral vectors at the indicated MOI. Luciferase assay was performed with equivalent amounts of cell lysates (Fig. 3E). Ad.PTTG1-siRNA, but not Ad.PTTG1-siRNA1M, increased p53RE- and p21-luc reporter activities in SH-J1 cells in a viral dose-dependent manner. Next, we analyzed the effect of PTTG1 depletion on induction of cell death. SH-J1 cells were transduced with or without Ad.PTTG1-siRNA or -siRNA1M, and population of cells with sub-2n DNA were analyzed by flow cytometry (Fig. 3F). Ad.PTTG1-siRNA transduction induced 31% sub-2n population in the cells, whereas mock- or Ad.PTTG1-siRNA1M transduction induced 1% or 2% sub-2n population. Collectively, the results suggest that PTTG1 depletion activates p53, induces p21 expression, and leads to apoptosis in SH-J1 hepatoma cells.

The Presence of Tumor Suppressor p53 Is Essential for Cytotoxicity by Depletion of PTTG1.

We further examined whether cytotoxic effect by PTTG1 depletion occurred in a p53-dependent manner. To this end, we examined the effect of PTTG1 depletion on HCT116 and HCT116Δp53 cell lines. Lysates of the cells transduced with or without the indicated viruses were immunoblotted with anti-PTTG1 or p53 antibodies (Fig. 4A). Ad.PTTG1-siRNA depleted PTTG1 efficiently in both HCT116 and HCT116Δp53 cells, suggesting that viral transduction occurred at comparable levels. As expected, p53 was detected in HCT116 cells, but not HCT116Δp53. After transduction of HCT116 or HCT116Δp53 cells with Ad.PTTG1-siRNA or -siRNA1M, cell viability was examined by crystal violet-staining (Fig. 4B). Ad.PTTG1-siRNA decreased viability of HCT116 cells in a viral-dose–dependent manner, but not that of HCT116Δp53 cells. Ad.PTTG1-siRNA1M did not affect the viability of both of the cell lines. To examine whether the effect of Ad.PTTG1-siRNA on the cell viability was indeed mediated by PTTG1 depletion, expression levels of PTTG genes were analyzed by RT-PCR (Fig. 4C). PTTG1, but not PTTG2 and 3, was highly expressed in HCT116 and HCT116Δp53 cells. Ad.PTTG1-siRNA efficiently targeted PTTG1. The result indicates that PTTG1 depletion induces cytotoxic effect in a p53-dependent manner.

thumbnail image

Figure 4. p53 is essential for cytotoxicity by the depletion of PTTG1. (A) HCT116 and HCT116Δp53 colorectal cancer cells were transduced with or without the indicated viruses at the indicated MOIs. At a 24-hour further incubation, cell lysates were immunoblotted as indicated. (B) HCT116 and HCT116Δp53 cells were plated in 12-well plates (105 cells/well) and transduced with or without the indicated viruses at various MOIs. After a 48-hour further incubation, cells were stained with crystal violet. The experiment was repeated three times, and a representative is shown. (C) PTTG1, but not PTTG2 and 3, is highly expressed in HCT116 and HCT116Δp53 colorectal cancer cells. The cells were transduced with or without the indicated viruses at an MOI of 50 and incubated for 24 hours and total RNAs were extracted. RT-PCR using one-step RT-PCR premix (Intron Biotech, South Korea) was performed with 2.5 μg RNAs and 10 pmol of each primer set for specific detection of PTTG1, 2, or 3 expression shown in Supplementary Fig. 1B, and for GAPDH as a control. The PCR condition was 94°C for 1 minute, 70°C for 1 minute, and 72°C for 1 minute for 24 cycles. The PCR product was separated by 1.5% agarose gel electrophoresis. EtBr staining is shown.

Download figure to PowerPoint

thumbnail image

Figure 5. Ad.PTTG1-siRNA exerts cytotoxic effects in a manner dependent on expression levels of PTTG1 and p53. (A) Transcript levels of PTTG1, 2, and 3 in hepatoma cell lines. RT-PCR was performed with total RNAs (2.5 μg of each) from the cells and 10 pmol of each primer set for specific detection of PTTG gene family under the conditions described in the Fig. 4C legend. The PCR products were separated by 1.5% agarose gel electrophoresis, and stained with EtBr. Lanes: SH-J1 (lane 1), SK-Hep1 (lane 2), Huh-7 (lane 3), HLK-3 (lane 4), Hep3B (lane 5), SNU368 (lane 6) hepatoma cells, CK-K1 cholangiocarcinoma cells (lane 7), and Chang liver cancer cells (lane 8). (B) Endogenous levels of PTTG1, p53, and p21 in hepatoma cell lines. The cells were lysed in RIPA buffer and immunoblotted as indicated. Lanes are the same as indicated in (A). Lane 9, MRC5 normal fibroblast cells. (C) Transduction efficiency of adenoviral vector. The cells were transduced with Ad.LacZ at an MOI of 50, and stained with X-gal. (D) Cytotoxic effects of Ad.PTTG1-siRNA on hepatoma cell lines. The cells were plated into 12-well plates (105 cells/well), and transduced with or without the indicated viruses at various MOIs. After a 48-hour further incubation, cells were stained with crystal violet. The experiment was repeated three times, and a representative is shown.

Download figure to PowerPoint

The Expression Levels of PTTG1 and p53 Influence Cytotoxicity by PTTG1 Depletion.

We compared expression of PTTG gene family at transcript levels, endogenous protein levels of PTTG1, p53, and p21 in various hepatoma cell lines, and cytotoxic effects by PTTG1 depletion. This was performed to gain an insight into a potential correlation between expression levels of PTTG1 and p53 and PTTG1 depletion-mediated cytotoxicity. Besides, p21 expression in HCCs was found to be predominantly regulated by dependence on p53.36 PTTG1, but not PTTG2 and 3, was highly expressed in five of eight hepatoma cell lines at transcript level (Fig. 5A), which may reflect expression pattern in primary HCCs (Fig. 1B). PTTG1 protein was expressed at high levels in SH-J1, SK-Hep1, and Huh-7 (Fig. 5B, lanes 1-3), and at relatively low levels in HLK3, Hep3B, SNU368, Chang liver, and CK-K1 cholangiocarcinoma cells (lanes 4-8). PTTG1 expression was undetectable in MRC5 normal fibroblast cell line (lane 9). The transcript levels of PTTG1 in Hep3B and SNU368 cells appeared not to reflect protein levels of PTTG1 compared with those in SH-J1, SK-Hep1, and Huh-7 cells (Fig. 5A-B, compare lanes 5 and 6 to 1-3). This discrepancy is currently unclear. This may represent the presence of a posttranslational regulation of PTTG1 expression. p53 expression was detected in all the cell lines except Hep3B (Fig. 5B, second panel, lane 5). Previously, p53 gene was found to be deleted in Hep3B cell line.37 p21 expression appeared to be relatively low in SH-J1, SK-Hep1, and Huh-7 cell lines on the basis of p53 expression levels (compare lanes 1-3 with 4). This might account for high levels of expression of PTTG1 that associates with p53 and modulates its function.22

After we confirmed that adenoviral vector transduced various cell lines at a comparable efficiency (Fig. 5C), the cell lines were transduced with or without Ad.PTTG1-siRNA or -siRNA1M, and cell viability was examined by crystal violet-staining (Fig. 5D). SH-J1, SK-Hep1, and Huh-7 cells, which expressed relatively high levels of PTTG1 and p53, were efficiently killed by transduction of Ad.PTTG1-siRNA (panels 1-3). SH-J1 cells expressing the highest level of PTTG1 among other cell lines were most efficiently killed by Ad.PTTG1-siRNA at an MOI of 10 (panel 1). Cell lines, in which expression levels of PTTG1 are low or undetectable, were weakly killed at an MOI of 200 (panels 4-6) or almost completely resistant to cytotoxic effect by Ad.PTTG1-siRNA (panels 7-9).

Huh-7 Hepatoma Cells Transduced With Ad.PTTG1-siRNA Nearly Lost Their Tumorigenicity In Vivo.

We examined whether PTTG1 depletion reduced tumorigenicity in vivo. Huh-7 cells were transduced with or without Ad.PTTG1-siRNA or -siRNA1M, and then subcutaneously inoculated into mice. Tumor growth was monitored (Fig. 6A). Huh-7 cells mock-transduced (PBS) or transduced with Ad.PTTG1-siRNA1M developed large tumors over 2,000 mm3 within 2 weeks. However, growth of Huh-7 cells transduced with Ad.PTTG1-siRNA was markedly suppressed, suggesting that depletion of PTTG1 leads to significant inhibition of tumor growth in vivo.

thumbnail image

Figure 6. PTTG1 depletion inhibits hepatoma cell growth in nude mice. (A) Huh-7 cells (5 × 106) were mock-transduced (PBS) or transduced with the indicated viruses at an MOI of 100. The tumor cells in PBS were implanted into the two sites of the flank of Balb/c nude mice (n = 3 per group). Mean tumor sizes were monitored for 19 days. Error bars stand for mean ± SD. (B) SH-J1 hepatoma xenograft was established by subcutaneous injection of the tumor cells in dorsal flanks of nude mice. Once the tumor reached approximately 3 to 5 mm in diameter, mice were divided into three groups, and 1 × 109 plaque-forming-units of Ad.PTTG1-siRNA (n = 7 per group), or of Ad.PTTG1-siRNA1M (n = 7) in PBS, or PBS (n = 5) were injected intratumorally once into tumor nodules. Mean tumor volumes were monitored for 19 days. Error bars stand for mean ± SD. (C) SH-J1 tumors excised from nude mice at 19 days described in (B) are shown.

Download figure to PowerPoint

Ad.PTTG1-siRNA Exerts Potent Antitumor Activity in SH-J1 Hepatoma Xenograft Established in Nude Mice.

Mice received subcutaneous injection of SH-J1 hepatoma cells, were randomly divided into three groups, and were injected intratumorally (once) with Ad.PTTG1-siRNA, Ad.PTTG1-siRNA1M, or PBS alone. Tumor growth was monitored (Fig. 6B). Tumors excised from mice 19 days after the virus injection are shown (Fig. 6C). Ad.PTTG1-siRNA treatment significantly decreased tumor growth rate compared with Ad.PTTG1-siRNA1M or PBS controls. One of seven mice treated with Ad.PTTG1-siRNA became tumor-free. The result suggests that PTTG1 gene silencing leads to a significant inhibition of hepatoma cell growth in vivo.

Discussion

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

We demonstrated that PTTG1, but not PTTG2 and 3, is highly and frequently expressed in liver tumor tissues from HCC patients (Fig. 1), suggesting that PTTG1 expression mostly associates with HCC. PTTG1 was expressed at high levels in SH-J1, SK-Hep1, and Huh-7 cell lines, in which expression of p53 and p21 were detected (Fig. 5B). SH-J1 cells contained functional p53 (Fig. 3). SK-Hep1 cells contain rearranged p53 gene but express normal p53 by immunochemical studies.37 p53 is activated in SK-Hep1 cells by doxorubicin treatment,38 suggesting that SK-Hep1 retains functional p53. Huh-7 cells contain p53 with a single point mutation39 but was shown to be susceptible to p53-mediated apoptosis.40 p21 expression is induced by p53 in Huh-7 cells,40 suggesting that mutated p53 retains apoptotic and trans-activating functionality. PTTG1 depletion in SH-J1, SK-Hep1, and Huh-7 cell lines led to loss of viability, most likely apoptotic cell death (Figs. 2D, 3F, and 5D). PTTG1 depletion decreased viability of HCT116 cells in a p53-dependent manner (Fig. 4). Viability of MRC5 cells, in which high levels of p53 and p21 expression but no PTTG1 expression were detected (Fig. 5B, lane 9), was not impaired by PTTG1 depletion (Fig. 5D, panel 9). MRC5 cells are susceptible to adenovirus infection.30 Thus, overexpression of PTTG1 and the presence of functional p53 are likely to be prerequisite for apoptotic cell death by PTTG1 depletion.

p53 gene mutations occur in 30% to 55% of hepatocellular carcinomas,41 indicating that the rest of HCCs contain wt p53. Overexpression of p53 was detected in 19 (31%) HCCs.42, 43 However, whether wt or mutated p53 is overexpressed is unknown. The implication of p53 overexpression in the progression of HCC also remains controversial.42 Because p53 is a potent tumor suppressor, the functionality of overexpressed or wt p53 must have been ablated in HCCs by other alterations. Indeed, proapoptotic proteins, Bax and Bcl-Xs, are down-regulated in p53-overexpressing HCCs.44 Survivin, inhibitor of apoptosis, is highly expressed and associated with p53 dysregulation in HCCs.45 PTTG1 depletion activated p53, evidenced by increased expression of p21 and induction of apoptosis in SH-J1 hepatoma cells (Fig. 3). These findings and p53-dependency of PTTG1 depletion-effect on the cell viability (Fig. 4) suggest that PTTG1 is a factor that negatively regulates overexpressed or functionally intact p53, thereby promoting hepatocellular carcinogenesis. Indeed, PTTG1 expression significantly increased from dysplastic nodule stage to grade II/III (Fig. 1A). In this regard, our findings are consistent with the previous result that the oncogenic effect of increased expression of PTTG1 may result from modulation of p53 function.22

PLC/PRF/5 and Mahlavu cells derived from African HCC patients who share the same mutation at codon 249 of the p53 gene (Arg-Ser substitution).46 This hot spot mutation (codon 249) is related to a high risk of exposure to a chemical carcinogen such as aflatoxin B1.46 The half-life of p53 (Ser-249) in the PLC/PRF/5 cells is short (<30 minutes), whereas the same mutant protein in Mahlavu cells has a longer half-life (>240 min),46 suggesting that the p53 mutation may have nothing to do with the prolonged half-life. Both SK-Hep1 and Huh-7 cells express p53 proteins with prolonged half-lives (>240 min).46 Interestingly, these two cell lines expressed high levels of PTTG1 (Fig. 5B, lanes 2 and 3), which may be responsible for the prolonged half-lives of p53. In this regard, it will be interesting to determine whether overexpression of p53 in HCC patients significantly correlates with increased PTTG1 level.

In culture, HCT116 cells lacking PTTG1 grew somewhat more slowly than parental cells, but other growth characteristics were essentially identical for HCT116 (PTTG1+/+) and (PTTG1−/−) cells, indicating that homozygous loss of PTTG1 is not lethal to human cells.15 PTTG1 also is not absolutely required for animal viability.19, 20 Therefore, an additional mechanism for regulating chromatid separation has been suggested. Indeed, vertebrate separase is regulated by Cdk1-dependent inhibitory phosphorylation in the absence of PTTG1 and inhibited by a PTTG1-independent manner.47 No defect in the viability of the PTTG1−/− cells is consistent with our findings that CK-K1, Chang liver, and MRC5 cells express low or undetectable levels of PTTG1, and their viability is not greatly affected by Ad.PTTG1-siRNA treatment (Fig. 5D). In this context, CK-K1, Chang liver, and MRC5 cells may represent cell lines in which PTTG1 is not essential for chromatid separation and may help to identify cellular factors regulating chromatid separation and underlying mechanism.

Conversely, given that PTTG1 is not essential for cell viability, this is in contrast with our finding that PTTG1 depletion in HCT116 cells markedly led to loss of viability (Fig. 4). Presumably, this might have resulted from characteristics of HCT116 (PTTG1−/−) cells. During the positive selection of HCT116 (PTTG1−/−) cells, other genetic or biochemical alterations might have occurred that permit these cells to survive without PTTG1. Indeed, these cells contain reduced levels of separase protein, impairment of the mitosis-specific processing of the separase, and reduction of substrate-specific activity of separase,15 indicating the occurrence of biochemical alteration. Particularly, the p53 pathway might have been altered directly or indirectly in HCT116 (PTTG1−/−) cells. Thus, speculating that PTTG1 is dispensable for normal cellular processes, but indispensable for survival of certain tumor cells such as SH-J1, SK-Hep1, Huh-7, and HCT116 tumor cell lines in which functionally intact p53 is expressed, is tempting. This possibility is supported by our findings that PTTG1 depletion-mediated cytotoxicity occurs in a p53-dependent manner (Fig. 4) and a manner dependent on expression level of PTTG1 as well (Fig. 5). In this regard, PTTG1 may serve as a molecular target for therapeutic intervention of liver cancer, because down-regulation of PTTG1 would lead to apoptotic cell death in cancer cells, but relatively spare normal cells.

Surgical or locally ablative therapies may offer the prospect of cure for the minority of HCC patients. However, treatment with curative intent is no longer possible for most patients with advanced disease. For some of these patients, with good hepatic reserve and a patent portal venous system, chemoembolization may afford a modest survival benefit. The remainder of patients are frequently treated with systemic therapies with palliative intent. However, no drug treatment has yet clearly proved a significant beneficial effect on survival or quality of life. To overcome the limitations of current therapeutic modalities of liver cancer, gene therapy, and immunotherapy approaches, a variety of strategies including gene-silencing using siRNA have been explored. Indeed, adenoviral vector encoding siRNA targeting p28GANK oncoprotein overexpressed in HCCs induces apoptosis and inhibits the growth of established tumors in nude mice.48 Synthetic siRNA oligonucleotide targeting cyclin E overexpressed in HCCs promotes apoptosis, blocks proliferation of HCC cell lines in culture, and inhibits tumor growth in nude mice.49 Here we show that PTTG1-siRNA expression vector significantly inhibits growth of some hepatoma cell lines in vitro and in vivo. These findings suggest that RNA interference technology would be an effective strategy to treat HCCs.

p53 gene therapy of cancers including HCC has been under intensive investigation.50 Overexpression of antiapoptotic genes in cancer tissues, such as PTTG1 and survivin, would negatively affect the therapeutic outcome of p53 gene transfer. Our findings suggest that the use of PTTG1-siRNA combined with p53 gene may be a novel approach to improve therapeutic outcome of p53 gene transfer. In this regard, developing a means to diagnose HCC patients with high levels of PTTG1 expression and functionally intact p53 will be important. Combination gene therapy approaches may result in a significant beneficial effect for those patients selected on the basis of molecular characteristics of HCC.

In conclusion, we report here that PTTG1 functions as a negative regulator of p53 tumor suppressor in HCC, and attenuation of PTTG1 expression significantly inhibits growth of PTTG1-overexpressing HCC cell lines in vitro and in vivo. Because Ad.PTTG1-siRNA appears to selectively target tumor cells expressing high levels of PTTG1 and functional p53, it may serve as a valuable gene therapeutic approach for treating human cancers, including HCC.

Acknowledgements

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

The authors thank Dr. Sham S Kakar, University of Louisville, KY, for providing PTTG-GFP plasmids, Dr. Dae-Ghon Kim, Chonbuk National University Medical School and Hospital, Republic of Korea, for providing tumor tissues, SH-J1, and HLK3 hepatoma and CK-K1 cholangiocarcinoma cell lines, Dr. Woo Ho Kim, Seoul National University College of Medicine, for providing tissue array, and Dr. Nam-Soon Kim, 21C Frontier Human Gene Bank, Republic of Korea, for providing pGST-PTTG1.

References

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

Supporting Information

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

Supplementary material for this article can be found on the H EPATOLOGY website ( http://interscience.wiley.com/jpages/0270-9139/suppmat/index.html )

FilenameFormatSizeDescription
jws-hep.21137.lgd1.doc26KSupporting Information file jws-hep.21137.lgd1.doc
jws-hep.21137.fig1.tif132KSupporting Information file jws-hep.21137.fig1.tif
jws-hep.21137.fig2.tif360KSupporting Information file jws-hep.21137.fig2.tif

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.