Skip Navigation


Carcinogenesis Advance Access originally published online on November 4, 2007
Carcinogenesis 2008 29(1):211-218; doi:10.1093/carcin/bgm236
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
29/1/211    most recent
bgm236v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Tannapfel, A.
Right arrow Articles by Pützer, B. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Tannapfel, A.
Right arrow Articles by Pützer, B. M.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© The Author 2007. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oxfordjournals.org

Autonomous growth and hepatocarcinogenesis in transgenic mice expressing the p53 family inhibitor DNp73

Andrea Tannapfel2,{dagger}, Katja John1,{dagger}, Nikica Mise1,4, Anke Schmidt1, Sven Buhlmann1, Saleh M. Ibrahim3 and Brigitte M. Pützer1,*

1 Department of Vectorology and Experimental Gene Therapy, Biomedical Research Center, University of Rostock Medical School, Schillingallee 69, Rostock D-18055, Germany
2 Department of Pathology, Ruhr University, Bürkle-de-la Camp-Platz 1, Bochum D-44789, Germany
3 Immunogenetics Research Group, University of Rostock Medical School, Schillingallee 70, Rostock D-18055, Germany
4 Present address: Renal Section, Division of Medicine, Imperial College London, Du Cane Road, London W12 ONN, UK

* To whom correspondence should be addressed. Tel: +49 381 494 5066/5068; Fax: +49 381 494 5062; Email: brigitte.puetzer{at}med.uni-rostock.de


   Abstract
Top
 Abstract
Introduction
 Materials and methods
 Results
 Discussion
 Funding
 References
 
p53 family proteins carry on a wide spectrum of biological functions from differentiation, cell cycle arrest, apoptosis and chemosensitivity of tumors. Conversely, N-terminally truncated p73 (DNp73) functions as a potent inhibitor of all these tumor suppressor properties, implicating its participation in malignant transformation and oncogenesis. Several reports indicated considerable up-regulation of DNp73 in hepatocellular carcinoma (HCC) that correlates with reduced survival of patients, but little is known about the functional significance of DNp73 to tumorigenesis in vivo due to the lack of an appropriate model. To address this, we generated transgenic mice in which DNp73 expression is directed to the liver by the albumin promoter. Gene expression was tested by mRNA and protein analyses. Transgenic mice exhibited prominent hepatic histological abnormalities including increased hepatocyte proliferation resulting in preneoplastic lesions (liver cell adenomas) at 3–4 months. Among 12- to 20-month-old mice, 83% of animals developed hepatic carcinoma. HCC displayed a significant increase of hyperphosphorylated inactive retinoblastoma, whereas p53-regulated inhibitors of cell cycle progression were down-regulated in the tumors. Our data firmly establish the unique oncogenic capability of DNp73 to drive hepatocarcinogenesis in vivo, supporting its significance as a marker for disease severity in patients and as target for cancer prevention. This model offers new opportunities to further delineate DNp73-mediated liver oncogenesis but may also enable the development of more effective cancer therapies.

Abbreviations: HCC, hepatocellular carcinoma; Rb, retinoblastoma; TA, transactivation


   Introduction
Top
 Abstract
 Introduction
Materials and methods
 Results
 Discussion
 Funding
 References
 
The tumor suppressor gene TP53 and its homolog TP73 share substantial structural and functional homology. Both proteins contain an acidic amino terminal transactivation (TA) domain, a central core DNA-binding domain and a carboxyterminal oligomerization domain. Consistently, p73 binds to p53 DNA target sites, transactivates p53-responsive genes and is capable of inducing cell cycle arrest and apoptosis in mammalian cells in a p53-like manner (1). However, unlike TP53, TP73 gives rise to a complex number of protein isoforms due to alternative promoter usage and differential mRNA splicing.

A detailed analysis of p73 in human cancer cells revealed the presence of several N-terminally truncated p73 isoforms ({Delta}N, {Delta}N', {Delta}Ex2, {Delta}Ex2/3) lacking the TA domain that accounts for the tumor suppressor function of the full-length TAp73 protein (referred to as DNp73). Thereof, the {Delta}Np73 transcript is generated from the P2p73 intragenic promoter, whereas all the other N-terminal variants of p73 are produced by alternative exon splicing from the E2F1-responsive P1p73 promoter in the 5'untranslated region upstream of the non-coding exon 1 (2). The transcripts {Delta}Np73 and {Delta}N'p73 encode the same protein.

All DNp73 isoforms fail to induce cell cycle arrest and apoptosis by acting as dominant-negative (DN) competitors for DNA binding and/or heteroduplex formation with p53 and full-length TAp73, thereby efficiently counteracting their growth-suppressive properties (3,4). By antagonizing the proapoptotic p53 family members, the DNp73 isoforms serve as antiapoptotic proteins. Moreover, they confer drug resistance to tumor cells harboring wild-type p53 and/or TAp73 (3,4). Conversely, down-regulation of endogenous DNp73 enhances p53- and TAp73-mediated apoptosis in cancer cells (4). In addition to the DNp73-mediated inhibition of p53/p73-dependent signaling pathways, N-terminally truncated p73 isoforms also interfere with the activation of the retinoblastoma (Rb) protein by increased phosphorylation, resulting in enhanced E2F activity and proliferation of fibroblasts (5). According to the current model, Rb is sequentially phosphorylated at the G1 to S phase transition by the action of cyclin D-Cdk4/6 and cyclin E-Cdk2 (6). The cyclin D-Cdk4/6 complex is specifically inhibited by the Cdk inhibitor p16, whereas p21 is a ubiquitous inhibitor of both cyclin D- and cyclin E-dependent kinases. Activation of E2F by DNp73 was efficiently abrogated by both kinase inhibitors, underscoring that cyclin E-Cdk2 and cyclin D-Cdk4/6 activities are required for DNp73 to phosphorylate Rb and activate E2F. By inactivating the two major tumor suppressor pathways, DNp73 acts functionally analogous to other oncoproteins.

Indeed, there is growing evidence that DNp73 might act as a biological relevant oncogene in primary human cancers. Overexpression of the p73{Delta}Ex2/3 isoform results in malignant transformation of immortalized NIH3T3 fibroblasts in cell culture that produce tumors in nude mice (7). {Delta}Np73 promotes immortalization of primary mouse embryo fibroblasts in vitro and cooperates with Ras in driving their transformation in vivo (8). More recently it was shown that DNp73 plays an essential role in the control of myogenic differentiation (9). DNp73 inhibits myogenic differentiation and enables cooperating oncogenes to transform myoblasts to tumorigenicity in nude mice.

N-terminally truncated p73 isoforms were found frequently up-regulated in a variety of human cancers, but not in normal tissues. This includes tumors of the breast, ovary, endometrium, cervix, vulva, vagina (4,10), prostate cancer (11), melanoma (12), neuroblastoma (13,14) and hepatocellular carcinoma (HCC) (15,16). In ovarian cancers, expression of the p73 transcript lacking exon 2 was exclusively detected in cancer cell lines and invasive tumor tissues, but not in semimalignant borderline tumors (17). The resulting tumor-associated change in p73 subunit composition could account for a shift in the net function of p73 from proapoptotic to antiapoptotic and allow the growth promoting potentially oncogenic functions of DNp73 to outweigh the growth-inhibiting, tumor-suppressive activities. With regard to human cancer, it is therefore still an eminent and unanswered question, whether the DNp73 protein acts as an oncogene during tumorigenesis in vivo.

In this report, we examined the functional significance of prolonged DNp73 expression to cell growth and tumorigenesis in mice transgenic for the p73{Delta}Ex2/3 isoform, in which transgene expression is driven by the liver-specific murine albumin promoter (18). We found that N-terminally truncated p73 is a bona fide oncogene, which is by itself capable of driving liver carcinogenesis in vivo. Both transgenic lines maintain tissue-specific DNp73 overexpression over the 20-month time period studied and developed progressive liver disease. The oncogenic activity of DNp73 was mediated by the up-regulation of hepatocyte proliferation resulting in multifocal preneoplastic and neoplastic lesions. Moreover, a striking increase of hyperphosphorylated Rb in hepatic carcinomas together with a down-regulation of p53-regulated inhibitors of cell cycle progression suggests that DNp73-mediated proliferation is operative during liver oncogenesis.


   Materials and methods
Top
 Abstract
 Introduction
 Materials and methods
Results
 Discussion
 Funding
 References
 
Mouse strains
C57BL/6J mice were obtained from Charles River Laboratories (Sulzfeld, Germany). Animals were housed in a dedicated pathogen-free animal facility under a 12-h light/dark cycle with free access to food and water. All animal procedures were conducted in adherence to ethical standards and with approval of the local Animal Care Committee.

Construction of the transgene
The human p73{Delta}Ex2/3β cDNA was isolated from the pcDNA3.1 plasmid by BamHI digestion and cloned into the EcoRV restriction site of the Alb e/p pBS plasmid (a gift from Carl A. Pinkert) between the mouse albumin promoter and the SV40 polyadenylation signal (see Figure 1A). Plasmid DNA was prepared by alkaline lysis and purified by CsCl-ethidium bromide density gradient centrifugation. The transgene was linearized with KpnI/SpeI, and the 5-kb fragment isolated was used for pronuclear injection.


Figure 1
View larger version (52K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 1. Expression pattern in transgenic mice. (A) Schematic diagram of the DNp73 construct. The liver-specific 2.2-kb murine albumin enhancer/promoter was used to drive expression of p73{Delta}Ex2/3β cDNA. Expression is terminated by the SV40 polyadenylation signal. (B and C) Analysis of transgene expression in various tissues (liver, kidney, lung, heart, brain) by RT–PCR on total RNA (B) and western blot from whole-cell lysates (C) using primers specific for p73{Delta}Ex2/3 (389-bp fragment) and mouse anti-human p73 monoclonal antibody (ER15, BD Bioscience). GAPDH (B) and actin (anti-β-actin antibody; Santa Cruz Biotechnology) (C) were used as loading controls in all tissues.

 
Generation and screening of transgenic mice
The linerized DNA was injected into the pronuclei of fertilized oocytes from C57BL/6J mice. After overnight incubation (37°C/5% CO2), embryos surviving to two-cell stage were transferred into the oviducts of pseudopregnant (C3HeJ F1) females using strandard protocols (19). Tail biopsies were taken from pubs at weaning (3–4 weeks), DNA extracted and analyzed by RT–PCR using primers p73_4S (5'-GACGGAATTCACCACCATCCT-3') and p73_5AS (5'-CCAGGCTCTCTTTCAGCTTCA-3') under the following conditions: denaturation at 95°C for 5 min, followed by 35 cycles of 95°C for 45 s, 64°C for 45 s and 72°C for 1 min, and a final extension step at 72°C for 10 min. This amplifies a 389-bp product from TP73 cDNA. Founders were propagated as hemizygous mutants by backcrossing to C57BL/6J mice. Transgenic mice were maintained for periods of up to 2 years under specific pathogen conditions and were routinely screened for the transgene by PCR.

Semiquantitative RT–PCR and western blotting
Adult mice were killed and tissues were dissected and snap frozen in liquid nitrogen. Total RNA was prepared using the RNeasy Mini Kit (Qiagen, Germany) according to the manufacturers' instructions. RNAs were reverse transcribed using oligo-dT primers (Applied Biosystems, Darmstadt, Germany) and the Omniscript Reverse Transcription Kit (Qiagen, Hilden, Germany). PCR amplification of the resulting cDNA was performed using HotMaster Taq DNA Polymerase (Eppendorf, Hamburg, Germany) with specific primer pairs for murine GAPDH: sense (5'-ACATCAAATGGGCTGAGGCCGCTG-3') and antisense (5'-TTCTGGGTGGCAGTGATGGCATGG-3'); cyclin D1: sense (5'-CCATGGAACACCAGCTCCTG-3') and antisense (5'-AGAGGCCACGAACATGCAGG-3'); cyclin E: sense (5'-TCGGGTCTGAGTTCCCAAGCC-3') and antisense (5'-GGCTGAAATGCAGTCCTGGG-3'); β-catenin: sense (5'-GGTACCTGAAGCTCAGCGCA-3') and antisense (5'-CGTCTCAGGGAACATGGCAG-3'); PIN1: sense (5'-ATTGCAGCTCTGCCAAAGCC-3') and antisense (5'-CCACTGCCCTTCTGAAGTCA-3'); p21: sense (5'- ATTGCGATGCGCTCATGG-3') and antisense (5'- CGTTTTCGGCCCTGAGAT-3'); 14-3-3{sigma}: sense (5'-TGCTGGACTCGCACCTCATC-3') and antisense (5'-TCTTGGCCAGCGAGATGG-3'); MDM2: sense (5'-ACGTCAGCCTGAAGGTGGGA-3') and antisense (5'-ATCCTGATGCGAGGGCGT-3'); Fas/CD95: sense (5'-TGCGCCTCGTGTGAACATG-3') and antisense (5'-AGCACAGGAGCAGCTGGA-3') and PIDD: sense (5'-AACACTCGGGTCCCTGTCCA-3') and antisense (5'-TGCAGGCGCACAAAGGGA-3'). To obtain a semiquantitative result within the linear range, PCR was performed using the minimum number of cycles necessary to acquire a clear signal. PCR products were stained with SYBR® Gold nucleic acid gel stain (Molecular Probes, Leiden, Netherlands) and quantitated in relative software units by the Bio-Imaging-Analyzer (Fuji, Düsseldorf, Germany) using TINA software. For western blot analysis, whole-cell extracts were prepared from tumor tissues. Cells were lysed in ice-cold RadioImmuno Precipitation Assay buffer (10 mM Tris–HCl, 150 mM NaCl, 0.1% sodium dodecyl sulfate, 0.1% Triton X-100, 1% deoxycholate, 5 mM EDTA), supplemented with protease inhibitor cocktail Complete Mini (Roche, Mannheim, Germany) and total protein concentration was quantified by a modified Bradford assay (Bio-Rad, München, Germany). Equal amounts of protein were separated by sodium dodecyl sulfate–polyacrylamide gel electrophoresis and transferred to nitrocellulose membranes (Amersham Pharmacia, Freiburg, Germany). Membranes were probed with anti-human p73 antibody (ER15; BD Bioscience, Heidelberg, Germany) or phospho-Rb (Ser807/811; Cell Signaling Technology, Frankfurt, Germany) according to the manufacturers’ guidelines. Primary antibody was detected using the appropriate horseradish peroxidase-conjugated secondary antibody (Amersham Pharmacia). Anti-actin (Santa Cruz Biotechnology, Heidelberg, Germany) was used as loading control.

Liver histology and immunohistochemistry
After anesthesia with diethylether, liver tissue from mice at various ages was immediately removed, fixed in 3.7% paraformaldehyde (in 0.1 M PBS, pH 7.4) overnight at 4°C and embedded in paraffin. Serial sections (5 µm) were cut through the entire liver, mounted on poly-L-lysine microscope slides, and processed for either histology by haemotoxylin/eosin staining or immunohistochemistry by deparaffination for 10 min in xylol as described before (20). After passing through decreasing concentration of ethanol, sections were stained with antibody against Ki-67 (MIB-1, DAKO, Germany), DNp73 (ER15; BD Bioscience), cyclin E (Novocastra Laboratories, Newcastle, UK), cyclin D1 (Zytomed Systems, Berlin, Germany) or phospho-Rb (Ser807/811) from Cell Signaling Technology at 4°C overnight. Immunostaining was visualized by using appropriate biotin-conjugated secondary antibodies followed by immunoperoxidase detection with the Vectastain ABC Elite kit (Linaris, Wertheim, Germany) and diaminobenzidine substrate (Linaris). Counterstaining was performed with hematoxyline (Linaris) (20).

In accordance to the literature and own experience (21,22), in this animal model, hepatocellular adenomas were diagnosed, when the portal structures and acinar architecture were lost and the proliferating hepatocytes showed a sinuoidal growth pattern with compression of the adjacent liver parenchyma. In contrast, HCCs were diagnosed when the neoplastic hepatocytes exhibited a significant degree of abnormal trabecular growth pattern or pseudoglandular structures were formed. The hepatocyte trabeculae were thicker than two cells. The enlarged atypical HCC cells had an eosinophilic cytoplasma with distended and dysplastic nuclei with an increased number of mitosis and also atypical mitosis (examples in figures).

p53 status
Exons 2–11 of the p53 gene were amplified by PCR and sequenced as described before (23).

Statistical analysis
MIB-1 and pRb positivity was assessed by counting an average number of 800 cells in pieces of 200 cells in four different fields of each focus. Two slides were counted in every case, leading to a total of 1600 evaluated cells for each lesion. Stained cell nuclei were considered to be positive, using a light microscope (40 times magnified). We calculated the MIB-1 index as the percentage of cells with positive nuclear staining in the total number of cells counted. Areas within each section showing maximum of reactivity were identified and confirmed by preliminary counting of 200 cells. The intraobserver error was calculated in a preliminary examination using the same material. It showed that at least 200 cell nuclei should be assessed to have the results fall within the 5% of the estimated real mean with a probability of 95%. To minimize the interobserver error, all countings were performed separately. In eight cases, in which conflicting numbers were evaluated, recounting was performed to obtain concordance of opinion.

Differences in frequencies between subgroups were analyzed with the Mann–Whitney U-test for unpaired samples. Statistical significance was set at P < 0.05.


   Results
Top
 Abstract
 Introduction
 Materials and methods
 Results
Discussion
 Funding
 References
 
Generation of transgenic lines and expression pattern of the transgene
We first attempted to create transgenic mice using the ubiquitously active HMG-CoA reductase promoter, but repeatedly failed to obtain founder mice that expressed the transgene although other transgenic lines were readily generated with this promoter. Considering that DNp73β might interfere with some essential developmental processes, we cloned a tissue-specific DNp73β construct, consisting of the liver-specific murine albumin enhancer/promoter [1.9-kb NheI-BamHI/0.3-kb region; (18)] followed by the p73{Delta}Ex2/3β cDNA and the SV40 polyadenylation signal (Figure 1A). The human p73{Delta}Ex2/3β cDNA was isolated from the pcDNA3.1 plasmid by BamHI digestion and cloned into the EcoRV restriction site of the Alb e/p pBS plasmid between the albumin promoter and the polyA signal (see Figure 1A). The transgene was linearized with KpnI/SpeI, and the 5-kb isolated fragment was injected into the pronuclei of fertilized oocytes from C57BL/6J mice. After overnight incubation, embryos surviving to two-cell stage were transferred into the oviducts of pseudopregnant (C3HeJ F1) females (19). Five transgenic founder lines were developed carrying DNp73 with a deletion of exon 2/3. All founders were mated on the C57BL/6J strain to propagate the lines. The transgene was inherited in a typical Mendelian fashion at a percentage of ~50% positive transgenic animals per litter. To determine in which organs the transgene is being expressed, the mice were killed and DNp73 was measured by RT–PCR and western blot analysis in a variety of organs using primers specific for p73{Delta}Ex2/3 (15) and anti-human p73 antibody, respectively. DNp73 mRNA was detected in liver and kidney of each of the founder mice, but not in any other tissue that was assayed including lung, spleen, heart and brain (Figure 1B). On protein level, a strong transgene expression was observed exclusively in the liver from transgenic mice, whereas all other tissues from transgenic (Figure 1C) and non-transgenic control animals (data not shown) examined were negative. Two transgenic lines each one overexpressing DNp73 at comparable high levels were chosen for further characterization.

DNp73 transgenic mice display increased proliferation of hepatocytes and liver cell adenoma
In order to determine if DNp73 contributes to aberrant cell growth regulation, we analyzed 25 mice from two trangenic lines (male and female) for pathological abnormalities. Compared with normal wild-type liver tissue with regular portal tracts and hepatocytes arranged as cords without any signs of inflammation, bleeding or necrosis (Figure 2A and B), animals of both DNp73 transgenic lines showed a complete acinar disarrangement with loss of portal tracts and central veins, an increase in cellular basophily and mild atypia. The adjacent liver parenchyma was compressed by trabecular-arranged hepatocytes forming a ‘pseudocapsule’ (Figure 2C and D). Histologically, these liver nodules were made up of groups of cells that were smaller than normal hepatocytes and were well circumscribed. In contrast to carcinomas, these lesions showed only mild cellular atypia, atypical mitosis did not occur. Nodular growth pattern as well as the absence of a trabecular or pseudoglandular structure was used to classify these early lesions as preneoplastic nodules, the so-called liver cell adenomas in mice by the Frith and Ward (21) criteria.


Figure 2
View larger version (138K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 2. Histological abnormalities in mice overexpressing DNp73 in the liver. Liver tissue from mice at various ages was fixed and embedded in paraffin. Serial sections (5 µm) were cut through the entire liver and processed for histology. (A and B) Normal liver parenchyma from non-transgenic mice with regular portal tracts (A, center) and hepatocytes arranged as cords. The sinusoids are small, not dilated. Very few fat drops are visible. (C and D) Liver of transgenic animals with proliferating hepatocytes forming adenomatous nodules (asterisk) of trabecular-arranged hepatocytes without portal tracts with a sharp border to normal hepatocytes (indicated by arrows). (D) Hyperplastic, proliferating hepatocytes arranged within cords consisting of more than two layers without clearly visible sinusoids. Inset: double-layered enlarged hepatocytes with mild atypia. Hematoxylin and eosin-stained liver sections; original magnification x5 (A and C) and x 20 (B and D).

 
These preneoplastic lesions were found at a frequency of 45% in 3- to 4-month-old mice (2–5 nodules per liver) and 90–100% at 8 months or older (7–15/8–12 nodules per liver) independent of sex (Table I). Normal or preneoplastic transgenic livers showed a strong nuclear staining by immunohistochemistry for Ki-67, indicating an increased proliferation of hepatocytes (Figure 3C and D) compared with the livers of non-transgenic mice with only a few positive nuclei visible (Figure 3A and B). Ki-67-positive cells appeared also within the portal tracts, indicating a hyperproliferative status of cholangiocytes as well (Figure 3D). Quantification of Ki-67-positive cells (MIB-1 index) revealed a median percentage of positive cells in the transgenic livers of 8% (4–12, SD 3.2) versus 2% in the wild-type group (0–6, SD 2.4, P < 0.05).


View this table:
[in this window]
[in a new window]

  		Table I. Hepatic lesions in DNp73 transgenic mice

 


Figure 3
View larger version (109K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 3. Proliferation index and DNp73 protein expression in preneoplastic adenoma of transgenic mice. (A) Normal liver parenchyma from wild-type mice with regular portal tracts with portal vein and bile ducts. (B) Corresponding immunohistochemical staining of normal wild-type liver tissue for Ki-67 (MIB-1): nuclei of proliferating cells are indicated by a brown reaction product. Only very few brown nuclei are visible in the wild-type liver. (C) Proliferating hepatocytes with few fat droplets in paraffin-embedded liver tissue from transgenic mice expressing DNp73. Hepatocytes are arranged in double layer cords. (D) Immunostaining of liver tissue from transgenic mice. Arrowheads indicate Ki-67. Ki-67-positive cells are also within portal tracts (inset). Quantification of Ki-67-positive cells (MIB-1 index) revealed a median percentage of positive cells in transgenic animals of 8% (4–12, SD 3.2) versus 2% in the wild-type group (0–6, SD 2.4); P ≤ 0.05. Anti-human p73 immunohistochemistry of livers from wild-type (E and F) and transgenic mice (G and H). (E) Normal liver parenchyma with normal sized portal tracts, without the expression of DNp73 (F). (G) Small, basophilic hepatocytes (left) and ‘clear cells’ with enlarged cytoplasma (right). (H) Immunoreactivity of DNp73 within the cell nuclei of adenomatous-arranged hyperplastic hepatocytes. Sections were stained with antibody against Ki-67 (MIB-1; DAKO) or ER15 antibody (BD Bioscience); hematoxylin–eosin); original magnification x20 (A, B and E), x25 (C and D), x 50 (F, G left and H) and x 55 (G right).

 
We have shown previously that overexpression of p73 isoforms lacking the N-terminus induces proliferation of normal human diploid fibroblasts that is associated with an increase in the activity of the cell cycle promoting transcription factor E2F (5). To substantiate the interpretation that hyperproliferation in liver tissues of transgenic mice is the result from overexpression of the DNp73 transgene, DNp73 immunoreactivity was analyzed in areas of hepatic hyperplasia (Figure 3G) using anti-human p73-specific antibody. Immunohistochemistry revealed nuclear localization and homogenous staining of DNp73 protein in proliferating hepatocytes within identifiable preneoplastic nodules (Figure 3H). In contrast, there was no detectable p73-specific immunoreactivity in the same tissue of wild-type age-matched control animals (Figure 3F). The DNp73 transgenic phenotype was restricted to the liver as no histological abnormalities were observed in the kidneys or other organs that is consistent with the weak or complete lack of trangene expression by the albumin promoter in these tissues.

Development of hepatic carcinoma in DNp73 transgenic mice
Beside the observed adenomatous preneoplastic liver lesions, DNp73 mice developed HCC with a high degree of penetrance. Neoplastic nodules were first noted in the livers of DNp73 transgenic mice at 12 months. By 12–20 months of age, all animals had adenomatous nodules and in 83% disease had progressed to HCC (Table I), whereas no tumor mass was found in the livers of non-transgenic age-matched controls (data not shown). Tumor formation occured gender independent. Small neoplastic nodules were microscopically found in all the mice that had grossly identified HCCs. The tumors were well circumscribed, without capsules (Figure 4Aa–c) and most of them had outgrown the liver in terms of infiltrative growth. Grossly identified HCC had a white–gray color. Measurable tumors ranged in size from 0.6 mm to 1.5 cm in a 15-month-old mouse. There was no significant difference in the size of lesions measured in male as compared with female mice. Microscopic examination showed that these neoplastic nodules were made up of atypical proliferating hepatocytes with trabecular or also pseudoglandular structures. The cells were increased in size and occasionally exhibited a clear cell cytoplasm. The number of mitosis was increased. In analogy to human HCC, these tumors were well-differentiated (G1) tumors. Overall, there was no evidence of invasion or metastasis. Liver sections from transgenic mice at 12–20 months of age showed DNp73 immunoreactivity in the cell nuclei of adenomatous-arranged hyperplastic hepatocytes and grossly identified HCCs. Moreover, RT–PCR and western blot analysis confirmed a persistent equally high expression of the transgene in the tumor as well as in the adjacent tissue (data not shown). Sequence analysis of the endogenous p53 gene in transgenic liver tissues revealed wild-type p53 in the tumors.


Figure 4
View larger version (68K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 4. Neoplastic tumor formation in DNp73 transgenic mice, accumulation of phosphorylated Rb and regulation of endogenous p53 target genes in HCC. (A, left panel) Liver sections from DNp73 transgenic mice: HCC as a circumscribed tumor nodule (a–c). Tumor cells partially with fatty changes. (d–i) Immunostaining of phospho-Rb (pRb) in (d and e) HCC cells, (f and g) adenoma and (h and i) non-tumoral liver tissue from transgenic mice using phospho-Rb-specific (Ser807/811) antibody, hematoxylin–eosin. Original magnification x5 (a), x 10 (b), x 20 (c) and x 25 (d–i). (A, right panel) western blot analysis of phospho-Rb (Ser807/811) in three tumors (T1–T3) versus normal transgenic livers (N1–N3). (B) Evaluation of E2F activity in normal and neoplastic transgenic liver cells compared with non-transgenic (wild-type) livers by immunohistochemical analysis (left panel) and semiquantitative RT–PCR (right panel) of E2F targets (Cyclin E, Cyclin D1, β-catenin and PIN1); Tg, transgenic, hematoxylin–eosin; original magnification x20 (a and b), x 25 (c, d and f) and x 40 (e). (C) TA of p53 target genes (p21Cip, 14-3-3{sigma}, MDM2, Fas/CD95 and PIDD) was analyzed by semiquantitative RT–PCR on total RNA from non-tumorous and tumorous liver tissues of DNp73 transgenic mice. GAPDH was used as a control. PCR products were quantitated in relative software units by the Bio-Imaging-Analyzer (Fuji) using TINA software (shown as fold induction or reduction, respectively). Data were normalized to GAPDH values (set as 1, indicated by red interrupted line) (B, C, right panel).

 
Effect of DNp73 on phosphorylation of Rb- and p53-regulated gene expression
The ability of tumor cell populations to expand in number is determined not only by net cell proliferation but also by the rate of cell attrition (i.e. apoptosis) (24). As demonstrated before (5), DNp73-mediated proliferation is attributed to enhanced phosphorylation and thereby inactivation of the Rb protein. Since this proliferative function of DNp73 provides a possible explanation for the selective advantage of p73 overexpressing cells during human tumorigenesis, we analyzed sections from normal liver, adenomas and hepatic carcinomas for Rb phosphorylation. As shown in Figure 4Ae, HCC cells stained clearly positive for phospho-Rb (67.0 ± 8.6%). A significantly lower amount of phospho-Rb-positive cells was found in preneoplatic adenomas (23.7 ± 5.1%; Figure 4Ag), whereas pRb was detected only in 3.2 ± 1.4% of hepatocytes from normal transgenic liver tissue (Figure 4Ai). Consistent with the higher proliferation rate in normal livers of transgenic versus wild-type mice, normal transgenic hepatocytes showed a slightly higher expression of the E2F targets cyclin E (c) and cyclin D1 (d) on protein level [cyclin E: 0.5% (0.1–0.9, SD 0.02, P < 0.05); and cyclin D1: 1.4% (0.9–1.9, SD 0.03, P < 0.05)] (Figure 4B, left panel), suggesting that E2F activity is enhanced in normal transgenic hepatocytes. In HCC tumors, cyclin D1 and cyclin E was 2- and 5-fold overexpressed at mRNA level, as compared with the non-tumorous transgenic tissue (Figure 4B, right panel). On protein level, cyclin D1 (f) and cyclin E (e) was detected in 14% (9–21) and 20% (17–28) of the tumor cell nuclei, respectively (Figure 4B, left panel). In case of cyclin D1, expression was heterogeneously within the tumors with a predominance at the infiltrating margins. These data strongly support that DNp73 actively contributes to the development of the malignant phenotype by a stepwise progression from normal or preneoplastic liver cells to neoplastic lesions associated with the accumulation of inactive Rb protein.

Pang et al. (25) have shown that the peptidyl-prolyl-isomerase PIN1, which is a downstream target of E2F (26), is critically involved in the pathogenesis of HCC via deregulation of β-catenin and cyclin D1. Therefore, the transcript levels of PIN1 and β-catenin were determined in the tumors versus normal transgenic liver cells. We showed that PIN1 was ~2-fold overexpressed in HCC cells versus normal transgenic hepatocytes. In contrast, the level of β-catenin mRNA in tumors was shown not to exceed that in non-tumor tissues of transgenic mice (Figure 4B, right panel).

DNp73 isoforms act as a dominant-negative inhibitor of p53 and TAp73. We previously demonstrated that inhibition of p53 by p73{Delta}Ex2/3β involves competition for DNA binding, whereas TAp73 can be inhibited by direct protein–protein interaction (3). Therefore, another conceivable mechanism whereby DNp73 promotes hepatocarcinogenesis in transgenic mice is by inhibiting wild-type p53 and full-length TAp73, thereby blocking their growth arresting and proapoptotic function. The effect of DNp73 on the expression of endogenous p53-regulated target genes (p21Cip, 14-3-3{sigma}, MDM2, Fas/CD95 and PIDD) (3,16) was evaluated by semiquantitative RT–PCR. Figure 4C shows a significant repression of the p53-regulated cell cycle inhibitors p21 and 14-3-3{sigma} in HCC cells compared with non-tumoral liver from trangenic mice. Reduced transcript levels were also evident for the p53 target MDM2, whereas no difference in expression was observed for the p53-regulated proapoptotic genes Fas/CD95 and PIDD. We also addressed the issue of whether DNp73-mediated inhibition of apoptosis contributes to hepatocarcinogenesis in transgenic mice by TUNEL staining. Although a trend toward increased apoptosis occurred in adenomas as compared with HCC (data not shown), the difference was not statistically significant. In every case, only very few apoptotic cells were observed in transgenic and wild-type mice as expected from human liver, adenoma or HCC (20,27,28). Moreover, consistent with the lack of down-regulation of apoptotic gene expression in transgenic liver tumors, we did not observe a difference in resistance to apoptosis induced by chemotherapeutic agents in normal or preneoplastic transgenic liver cells when compared with non-transgenic cells (data not shown).


   Discussion
Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
Funding
 References
 
Amino terminus-truncated p73 isoforms show dominant-negative behavior toward TAp73 and wild-type p53. In human tumor cell lines, DNp73 is an efficient transdominant inhibitor of p53/TAp73-mediated transcriptional activity, growth suppression, apoptosis and confers resistance to chemotherapy. Independent of p53, DNp73 negatively affects the Rb protein tumor suppressor pathway. Therefore, it is of particular interest that DNp73 is frequently overexpressed in a variety of human cancers (4,10,12–15). Intriguingly, a strong correlation was shown between deregulated DNp73 and the p53 status in tumors. p53 wild-type cancers exhibited a significantly higher deregulation of the N-terminally truncated p73 isoforms than p53 mutant cancers, supporting the hypothesis that overexpression of transdominant p73 isoforms can function as epigenetic inhibitors of p53 in vivo, thereby alleviating selection pressure for p53 mutations in tumors (29). In addition, high DNp73 expression levels were found to correlate with the response failure to chemotherapy and worse recurrence-free and overall survival of patients with p53 mutant ovarian cancer (30), suggesting an additional oncogenic mechanism independent of its known p53-inhibitory activity. All these correlations between high level expression of DNp73 variants and various prognostic parameters make it unlikely that DNp73 expression is only secondary to other transforming events and suggest that truncated p73 is implicated in tumorigenesis and possibly functions as a dominant oncogene to enhance tumor progression and therapeutic resistance.

Transgenic mice have been instrumental in demonstrating the requirement for oncogenic events in cell transformation. Our results indicate that the increased expression of DNp73 is capable and sufficient to induce hepatocarcinogenesis in vivo. Both transgenic lines expressing DNp73 by the liver-specific Alb promoter developed the same phenotype with stepwise progression from preneoplastic liver cell adenoma by 3–4 months in 45% of cases to neoplastic lesions by 12–20 months in 83% of cases. Overexpression of the DNp73 protein was found in the cell nuclei of adenomatous-arranged hyperplastic hepatocytes, small neoplastic nodules and grossly identified HCCs from transgenic mice. Consistent with the weak or complete lack of trangene expression by the Alb promoter in non-liver tissues, the transgenic phenotype was restricted to the liver, and no histological abnormalities were observed in other organs.

The relatively high penetrance of preneoplastic and neoplastic tumor formation in our mice could imply a high activity of the DNp73 isoform in dysplastic areas of the liver. In this regard, the adenomatous nodules, which were strongly positive for Ki-67, clearly expressed DNp73 in proliferating hepatocytes of the foci. In addition, according to our previous findings demonstrating that DNp73 mediates enforced cell proliferation through inactivation of Rb (5), preneoplastic adenoma and to a much higher extent HCC lesions of the transgenic mice stained positive for phosphorylated Rb, suggesting that phospho-Rb accumulation contributes to the malignant phenotype. That this increase in Rb phosphorylation is due to the DNp73 expression rather than a consequence of tumor progression arises from data indicating that the proliferation rate and E2F activity is higher in transgenic hepatocytes than in normal non-transgenic liver cells. These results may support that DNp73 expression is not simply a consequence of malignant transformation but actively promotes hepatocyte transformation and the development of the tumorigenic phenotype by interfering with Rb activity. Moreover, our results show that down-regulation of p53-dependent negative regulators of cell cycle progression is a second conceivable mechanism, whereby the DNp73 transgene leads to hepatic tumors. In our transgenic mice, no statistically significant correlation was found between the degree of apoptosis occurring in the livers and disease progression. This might be due—at least in part—to the fact that apoptotic cell death is a rare event in liver tumors. A reciprocal relationship of proliferation and apoptosis could be a hypothesis, but had not been proven in a complex tumor situation. Another explanation is that DNp73-induced cellular transformation in this environment is not linked to transgene-mediated protection from apoptosis shown in other studies (3,4,7,16). The appearance of discrete multifocal lesions rather than generalized hyperplasia of hepatocytes in the absence of an apparent variability in expression intensity with higher expression of the transgene in these lesions may suggest that overexpression of DNp73 as an initiating event is conducive to the development of additional genetic lesions brought about by multiple rounds of proliferation that enable progression from hyperplasia to HCC.

The importance of the transgenic mouse data is evidenced by previous results from patients with HCC. HCC is frequently seen as a unifocal as well as a multifocal disease. Lesions appear as foci of well-differentiated, poorly differentiated or clear cell type hepatocytes. In HCCs, high p73 expressions levels were revealed as an independent marker of poor patient survival prognosis (20,31,32). Immunohistochemistry demonstrated that p73 preferentially accumulates in the nuclei of HCC cells (31). In agreement with our data, it was shown that both the induction of p73 and the inactivation of the E2F/Rb pathway are common events in human HCC. This was accompanied by the activation of E2F1 target genes including cyclin E and p14ARF in most liver tumors (32). Moreover, it has been shown that cyclin D1 is strikingly overexpressed in 58% of HCC (33), and that its expression closely correlates with the expression level of PIN1 in tumors (26,34). PIN1 is an E2F target that is essential for Ras-induced tumorigenesis via up-regulation of cyclin D1 (35). Importantly, cyclin D1 and PIN1 were up-regulated at similar levels in the DNp73 transgenic HCC tumors, suggesting that DNp73-mediated deregulation of the E2F/Rb pathway may be responsible for PIN1 and cyclin D1 overexpression in HCC. However, PIN1 also up-regulates cyclin D1 via accumulation of β-catenin (26), which seems to be another critical event in hepatocarcinogenesis (25,36). Since β-catenin was not up-regulated in the liver tumors of transgenic mice, the Wnt/β-catenin signaling pathway may not contribute to DNp73-related liver carcinogenesis. Nevertheless, additional studies are needed to fully understand the mechanisms of NH2-truncated p73 as a positive regulator of hepatocarcinogenesis.

A detailed analysis of p73 in hepatic tumors from patients indicated the presence of high DNp73 levels in p73 overexpressing tumor cells (15,16). In these studies, aberrant DNp73 expression was identified as the causal factor for unfavorable outcome. Patients with tumors overexpressing DNp73 exhibited a significantly shorter survival time than those whose tumors were negative for DNp73 (16). Our evidence that overexpression of N-terminally truncated p73 in the liver of transgenic animals results in progressive disease culminating in HCC development clearly establishes DNp73 as a dominant oncogene in liver oncogenesis. The persistence of transgene expression throughout the long latency period of disease progression coupled with the high degree of penetrance of the phenotype argues for its pivotal role in establishing and maintaining these tumors, and thus should be considered as a potential target for preventive and therapeutic strategies. The DNp73 transgenic mice provide a unique model not only for defining the molecular events by which DNp73 isoforms deregulate hepatocyte growth control in vivo but also for the development of such therapeutic approaches against liver tumors overexpressing the DNp73 protein.


   Funding
Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Funding
References
 
Deutsche Krebshilfe to B.M.P; FORUN program of the Medical Faculty.


   Footnotes
 
{dagger} These authors contributed equally to this work. Back


   Acknowledgments
 
The authors thank Ilona Klamfuß for technical assistance in pronuclear injection and maintenance of transgenic mice and Carl Pinkert for providing the pBS plasmid.

Conflict of Interest Statement: None declared.


   References
Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Funding
 References
 

  1. Jost CA, et al. p73 is a simian [correction of human] p53-related protein that can induce apoptosis. Nature (1997) 89:191–194.
  2. Stiewe T, et al. Role of p73 in malignancy: tumor suppressor or oncogene? Cell Death Differ. (2002) 9:237–245.[CrossRef][Web of Science][Medline]
  3. Stiewe T, et al. Transactivation-deficient Delta TA-p73 inhibits p53 by direct competition for DNA binding: implications for tumorigenesis. J. Biol. Chem. (2002) 277:14177–14185.[Abstract/Free Full Text]
  4. Zaika AI, et al. DeltaNp73, a dominant-negative inhibitor of wild-type p53 and TAp73, is up-regulated in human tumors. J. Exp. Med. (2002) 196:765–780.[Abstract/Free Full Text]
  5. Stiewe T, et al. Inactivation of retinoblastoma (RB) tumor suppressor by oncogenic isoforms of the p53 family member p73. J. Biol. Chem. (2003) 278:14230–14236.[Abstract/Free Full Text]
  6. Harbour JW, et al. Cdk phosphorylation triggers sequential intramolecular interactions that progressively block Rb functions as cells move through G1. Cell (1999) 98:859–869.[CrossRef][Web of Science][Medline]
  7. Stiewe T, et al. Transactivation-deficient DeltaTA-p73 acts as an oncogene. Cancer Res. (2002) 62:3598–3602.[Abstract/Free Full Text]
  8. Petrenko O, et al. DNp73 facilitates cell immortalization and cooperates with oncogenic Ras in cellular transformation in vivo. Mol. Cell. Biol. (2003) 23:5540–5555.[Abstract/Free Full Text]
  9. Cam H, et al. p53 family members in myogenic differentiation and rhabdomyosarcoma development. Cancer Cell (2006) 10:281–293.[CrossRef][Web of Science][Medline]
  10. Becker K, et al. Patterns of p73 N-terminal isoform expression and p53 status have prognostic value in gynecological cancers. Int. J. Oncol. (2006) 29:889–902.[Web of Science][Medline]
  11. Guan M, et al. Aberrant expression of DeltaNp73 in benign and malignant tumours of the prostate: correlation with Gleason score. J. Clin. Pathol. (2005) 58:1175–1179.[Abstract/Free Full Text]
  12. Tuve S, et al. Alterations of DeltaTA-p73 splice transcripts during melanoma development and progression. Int. J. Cancer (2004) 108:162–166.[CrossRef][Web of Science][Medline]
  13. Casciano I, et al. Expression of DeltaNp73 is a molecular marker for adverse outcome in neuroblastoma patients. Cell Death Differ. (2002) 9:246–251.[CrossRef][Web of Science][Medline]
  14. Douc-Rasy S, et al. DeltaN-p73alpha accumulates in human neuroblastic tumors. Am. J. Pathol. (2002) 160:631–639.[Abstract/Free Full Text]
  15. Stiewe T, et al. Quantitative TP73 transcript analysis in hepatocellular carcinomas. Clin. Cancer Res. (2004) 10:626–633.[Abstract/Free Full Text]
  16. Müller M, et al. TAp73/DeltaNp73 influences apoptotic response, chemosensitivity and prognosis in hepatocellular carcinoma. Cell Death Differ. (2005) 12:1564–1577.[CrossRef][Web of Science][Medline]
  17. Ng SW, et al. Primary pulmonary sarcomatoid carcinoma with intracardiac extension. Jpn. J. Thorac. Cardiovasc. Surg. (2006) 54:416–419.[CrossRef][Medline]
  18. Pinkert CA, et al. An albumin enhancer located 10 kb upstream functions along with its promoter to direct efficient, liver-specific expression in transgenic mice. Genes Dev. (1987) 1:268–276.[Abstract/Free Full Text]
  19. Hoang TR, et al. A proinflammatory role for Fas in joints of mice with collagen-induced arthritis. Arthritis Res. Ther. (2004) 6:404–414.[CrossRef]
  20. Tannapfel A, et al. Expression of p73 and its relation to histopathology and prognosis in hepatocellular carcinoma. J. Natl. Cancer Inst. (1999) 91:1154–1158.[Abstract/Free Full Text]
  21. Frith C, et al. A morphologic classification of proiferative, and neoplastic hepatic lesions in mice. J. Environ. Pathol. Toxicol. (1979) 3:329–351.[Web of Science][Medline]
  22. Dombrowski F, et al. Hepatocellular neoplasms induced by low-number pancreatic islet transplants in autoimmune diabetic BB/Pfd rats. Cancer Res. (2006) 66:1833–1843.[Abstract/Free Full Text]
  23. Venkatachalan S, et al. Retention of wild-type p53 in tumors from p53 heterozygous mice: reduction of p53 dosage can promote cancer formation. The EMBO J. (1998) 17:4657–4667.[CrossRef][Web of Science][Medline]
  24. Hanahan D, et al. The hallmarks of cancer. Cell (2000) 100:57–70.[CrossRef][Web of Science][Medline]
  25. Pang R, et al. PIN1 overexpression and β-catenin gene mutations are distinct oncogenic events in human hepatocellular carcinoma. Oncogene (2004) 23:4182–4186.[CrossRef][Web of Science][Medline]
  26. Ryo A, et al. PIN1 is an E2F target gene essential for Neu/Ras-induced transformation of mammary epithelial cells. Mol. Cell. Biol. (2002) 22:5281–5295.[Abstract/Free Full Text]
  27. Tannapfel A, et al. Genes involved in hepatocellular carcinoma; deregulation in cell cycling and apoptosis. Virchows Arch. (2002) 440:345–352.[CrossRef][Web of Science][Medline]
  28. Tannapfel A, et al. Apoptosis and proliferation in relation to histopathological variables and prognosis in hepatocellular carcinoma. J. Pathol. (1999) 187:439–445.[CrossRef][Web of Science][Medline]
  29. Concin N, et al. Transdominant DeltaTAp73 isoforms are frequently up-regulated in ovarian cancer. Evidence for their role as epigenetic p53 inhibitors in vivo. Cancer Res. (2004) 64:2449–2460.[Abstract/Free Full Text]
  30. Concin N, et al. Clinical relevance of dominant-negative p73 isoforms for responsiveness to chemotherapy and survival in ovarian cancer: evidence for a crucial p53-p73 cross-talk in vivo. Clin. Cancer Res. (2005) 11:8372–8383.[Abstract/Free Full Text]
  31. Zemel R, et al. p73 overexpression and nuclear accumulation in hepatitis C virus-associated hepatocellular carcinoma. Dig. Dis. Sci. (2002) 47:716–722.[CrossRef][Web of Science][Medline]
  32. Sayan AE, et al. Acquired expression of transcriptionally active p73 in hepatocellular carcinoma cells. Oncogene (2001) 20:5111–5117.[CrossRef][Web of Science][Medline]
  33. Joo M, et al. Cyclin D1 overexpression in hepatocellular carcinoma. Liver (2001) 21:89–95.[CrossRef][Web of Science][Medline]
  34. Bao L, et al. Prevalent overexpression of prolyl isomerase Pin1 in human cancer. Am. J. Pathol. (2004) 164:1727–1737.[Abstract/Free Full Text]
  35. Yu O, et al. Specific protection against breast cancers by cyclin D1 ablation. Nature (2001) 411:1017–1021.[CrossRef][Medline]
  36. Thompson MD, et al. WNT/β-catenin signaling in liver health and disease. Hepatology (2007) 45:1298–1305.[CrossRef][Web of Science][Medline]
Received August 22, 2007; revised October 10, 2007; accepted October 18, 2007.


Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?



This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
29/1/211    most recent
bgm236v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Tannapfel, A.
Right arrow Articles by Pützer, B. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Tannapfel, A.
Right arrow Articles by Pützer, B. M.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?