HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 277, Issue 16, 14177-14185, April 19, 2002

This Article Abstract Full Text (PDF) All Versions of this Article: 277/16/14177    most recent M200480200v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Stiewe, T. Articles by Pützer, B. M. Search for Related Content PubMed PubMed Citation Articles by Stiewe, T. Articles by Pützer, B. M. Social Bookmarking                      What's this?

Transactivation-deficient TA-p73 Inhibits p53 by Direct Competition for DNA Binding

IMPLICATIONS FOR TUMORIGENESIS^*

Thorsten Stiewe, Carmen C. Theseling, and Brigitte M. Pützer

From the Centre for Cancer Research and Cancer Therapy, Institute of Molecular Biology, University of Essen, Medical School, D-45122 Essen, Germany

Received for publication, January 16, 2002, and in revised form, February 13, 2002

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

The p53 family member p73 displays significant structural and functional homology to p53. However, instead of mutational inactivation, overexpression of wild-type p73 has been reported in various tumor types compared with normal tissues, arguing against a classical tumor suppressor function. Recently, N-terminally truncated, transactivation-deficient p73 isoforms (TA-p73) have been identified as a second class of p73 proteins. Because overexpression of p73 in tumors includes TA-p73, we further characterized these novel p73 isoforms. We show that TA-p73 retains DNA-binding competence but lacks transactivation functions, resulting in an inability to induce growth arrest and apoptosis. Importantly, TA-p73 acts as a dominant-negative inhibitor of p53 and full-length p73 (TA-p73). We demonstrate that inhibition of p53 involves competition for DNA binding, whereas TA-p73 can be inhibited by direct protein-protein interaction. Further, we show that up-regulation of endogenous p73 just like ectopic overexpression of TA-p73 confers resistance to p53-mediated apoptosis induced by the chemotherapeutic agent H-7. Because inhibition of p53 is a common theme in human cancer, our data strongly support a role of TA-p73 expression for tumor formation.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

The TP53 gene was the first tumor suppressor gene to be identified and is still considered the prototypical tumor suppressor. In more than half of human tumors the TP53 gene is inactivated directly by mutations, and in many others p53 is functionally compromised epigenetically by various mechanisms (1, 2). In fact, several transforming oncogenes have been shown to be potent inhibitors of p53 (1). Loss of functional p53 therefore appears to be crucial for the development of most, if not all, cancers.

Although p53 was long considered to be unique, recently two p53-related genes were discovered (3-5). TP73 and TP63 encode proteins with remarkable sequence homology to p53, suggesting that they are also involved in the regulation of cell growth and apoptosis. Indeed, in experimental systems, p73 showed many p53-like properties; it could bind to p53 DNA binding sites, transactivate p53-responsive genes, and induce cell cycle arrest or apoptosis (3, 6).

However, apart from the structural and functional similarities between p53 and p73, several pieces of evidence argue against p73 being a classical tumor suppressor. In contrast to p53, p73 is not inactivated by classic viral oncoproteins to allow host cell transformation, indicating that p73 may augment, rather than inhibit, viral and cellular transformation (7). In contrast to mice lacking p53, p73-negative mice are not prone to tumor development (8). Despite initial reports suggesting tumor-associated deletion of p73, many subsequent studies failed to demonstrate mutational inactivation of the TP73 gene in a wide variety of tumors (3).¹ Instead, overexpression of wild-type p73 has been reported in various tumor types compared with normal tissues (9-13). High p73 expression levels were revealed as an independent marker of poor patient survival prognosis in hepatocellular carcinomas and correlated positively with higher risk stages in B-CLLs (B-cell chronic lymphocytic leukemia) (14). Together, these data raise the question about additional activities of p73 in cancer.

The molecular basis for the apparently different functions of p53 and p73 in human tumors is at present unknown but might be related to the differences in genomic organization of the TP53 and TP73 genes. Whereas TP53 does not show much splice variations, the TP73 gene encodes a complex number of isoforms with at least nine different isoforms generated by alternative splicing of the C-terminal exons (Fig. 1) (15). However, apart from a shift toward expression of the shorter C-terminal isoforms in tumor cells, little is known to support their role in tumorigenesis (12, 16, 17). Here, the recent identification of N-terminally truncated, transactivation-deficient p73 isoforms as an additional group of p73 proteins seems to be more promising (8, 18-22). In developing mice these isoforms are predominant (termed N-p73) and are generated from an alternative, cryptic promoter in intron 3 (8). Murine N-p73 has been shown to be a potent anti-apoptotic protein, which rescues sympathetic neurons from apoptosis induced by nerve growth factor withdrawal or p53 overexpression (18). A similar transcript with high sequence homology to murine N-p73 could be identified in human cells (20-22). In addition to these "physiological" N-p73 proteins with a distinct regulation via an independent promoter, aberrantly spliced transcripts (p73ex2, p73ex2/3 and N'-p73) regulated by the TA-promoter are found in human tumor cells (3, 19, 23).² Because the translation start is located in exon 2, these splice variants also encode N-terminally truncated proteins termed TA-p73 (Fig. 1). Interestingly, overexpression of p73 in tumors has been shown to include both full-length p73 (TA-p73) and N-terminally truncated p73 isoform (TA-p73).^1, 2 Moreover, overexpression of TA-p73 results in malignant transformation of NIH3T3 cells supporting a function in tumorigenesis.²

View larger version (20K): [in this window] [in a new window]   Fig. 1.   Genomic organization of the TP73 locus. A, the splicing patterns generating C-terminal isoforms p73 , , , , , and  and the N-terminal isoforms p73ex2, p73ex2/3, N-p73, and N'-p73 are shown. The arrows indicate transcriptional start sites. The N-p73 isoform is generated from a cryptic promoter within intron 3. B, the exon structure of the TA-p73-encoding transcripts is shown in comparison with full-length TA-p73 (exons 1-5 only). Noncoding sequences are depicted in white. C, domain structure of full-length p73. TA, transactivation domain; DBD, DNA-binding domain; OD, oligomerization domain; CT, C terminus.

In this study we have further characterized these novel p73 isoforms. We show that TA-p73 just as full-length p73 (TA-p73) and wild-type p53 is a sequence-specific DNA binding factor. Due to the lack of the N-terminal transactivation domain, however, TA-p73 does not transactivate typical p53-regulated genes, resulting in an inability to induce growth arrest and apoptosis. Moreover, TA-p73 acts as a dominant-negative inhibitor of p53 and full-length p73 (TA-p73). Investigating the mechanism, we demonstrate that inhibition of p53 involves competition for DNA binding, whereas TA-p73 can be inhibited by direct protein-protein interaction. Further, we show that up-regulation of endogenous p73 just like ectopic overexpression of TA-p73 confers resistance to p53-mediated apoptosis induced by the chemotherapeutic agent H-7. Because inhibition of p53 is a common theme in human cancer, our data provide further evidence supporting a role of TA-p73 expression for tumor formation.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Cell Culture and Transfections-- MCF-7 (ATCC, Manassas, VA), H1299 (human bronchoalveolar carcinoma, obtained from B. Opalka, University of Essen), SH-SY5Y (obtained from A. Eggert, University of Essen), and NHDF cells (normal human diploid fibroblasts, obtained from M. Roggendorf, University of Essen) were maintained in Dulbecco's modified Eagle medium supplemented with 10% fetal calf serum. Isoquinoline (1-(5-isoquinolinesulfonyl)-2-methylpiperazine, H-7; Sigma) was used at a final concentration of 50 µM for 24 h and all-trans-retinoic acid (RA; Sigma) at 10 µM for 4 days. For induction of DNA damage, cells were exposed to 3 µM adriamycin (doxorubicin, Sigma) for 16 h, 8 h after infection with adenoviral vectors. Transfections were performed by electroporation.

Plasmids-- Expression plasmids for p53 (pC53), p53R175H (pC53-175), and Gal-p53, HA-p73, and HA-p73 were kindly provided by B. Vogelstein, J. Brady, and G. Melino, respectively. cDNAs encoding untagged p73, p73ex2, p73ex2/3, p73, p73ex2, and p73ex2/3 were amplified by PCR using HA-p73 as a template. The 5' fragment of the cDNA for N-p73 was amplified by RT-PCR³ on cDNA prepared from E2F1-stimulated normal human diploid fibroblasts using the primers 5'-CCGGATCCATGCTGTACGTCGGTGAC-3' and 5'-GTGAATTCCGTCCCCACCTGTGGTGG-3'. The resulting fragment was used to replace the corresponding sequence of TA-p73 and TA-p73. All p73 cDNAs were cloned into pcDNA3.1 and sequence-verified. The TA-p73R292H mutant was generated from pcDNA3.1-p73ex2/3 with the QuikChange^TM Site-Directed Mutagenesis Kit (Stratagene, La Jolla, CA). Unless indicated, TA-p73 or TA-p73 denote the p73ex2/3 or p73ex2/3 construct, respectively. pGL3-p53 and Gal-TK-Luc have been described previously (24, 25).

Antibodies-- The murine anti-p73 monoclonals ER15 and ER13 have been described (7). Goat polyclonal anti-p73 (Ab-7), murine anti-p53 monoclonals DO-1 and PAb421 were obtained from Oncogene Science. Rat monoclonal anti-H-ras (sc-35) and goat polyclonal anti-p21 (sc-397) were obtained from Santa Cruz Biotechnology, Santa Cruz, CA.

Adenoviral Vectors-- cDNAs encoding p53 and various p73 isoforms were cloned into pAdTrack-CMV (kindly provided by B. Vogelstein). Recombinant adenoviruses were produced as described (24, 26).

Luciferase Assay-- Luciferase activities were determined 48 h following transfection using a premanufactured Luciferase Reporter Assay System (Promega) and normalized to the total protein concentration in the cell extract.

Colony Formation Assay, MTT Assay, Flow Cytometry, and DNA Fragmentation Assay-- For colony formation assays, H1299 cells were transfected with 10 µg of pcDNA3.1-based expression plasmids for p53 or p73 cultured in the presence of 500 µg/ml G418 for 3 weeks, fixed, and stained with Giemsa. For MTT viability assays, H1299 cells were plated on 96-well plates, infected with a combination of p53- and TA-p73-expressing adenoviral vectors, and analyzed 48 h post-infection using the CellTiter 96^® AQ_ueous One Solution Cell Proliferation Assay (Promega). MTT assays of SH-SY5Y cells were performed as described (27). DNA analysis by flow cytometry and DNA fragmentation assays of H1299 cells infected with various adenoviral vectors were performed as described (28, 29).

RT-PCR-- Semiquantitative RT-PCR was carried out on total RNA prepared with the RNeasy Mini Kit (Qiagen) essentially as described (24). To obtain a semiquantitative result, we used the minimum number of cycles required to obtain a clear signal in the linear range and labeling of PCR products with [-³²P]dCTP for high sensitivity detection. The amount of PCR product was quantitated on a PhosphorImager. Primer sequences are available upon request.

Electrophoretic Mobility Shift Assay (EMSA), in Vitro Translation-- EMSAs were performed as described (7). For supershift analysis 1 µg of the anti-p73 monoclonal ER15 or the polyclonal anti-p73 antibody Ab-7 were used. In vitro translation was performed with the TNT translation system (Promega) according to the manufacturer's protocol.

Immunoprecipitation, GST-Pull-down Assay, Western Blotting, and Antibodies-- For co-immunoprecipitation experiments, 3 mg of lysate from transfected H1299 cells were precipitated with 1 µg of either p53 antibodies (mixture of DO-1 and PAb421) or HA antibody (F-7, Santa Cruz Biotechnology) in NET buffer (50 mM Tris-HCl, pH 7.4, 150 mM NaCl, 5 mM EDTA, 0.5% Nonidet P-40, 10% glycerol). Immunoprecipitates were subjected to SDS-PAGE, transferred to ECL nitrocellulose (Amersham Biosciences), and immunoblotted with p73-antisera.

For GST-pull-down assays, ³⁵S-labeled in vitro translated proteins were bound with recombinant GST-p53 in GST-binding buffer (20 mM Tris-HCl, pH 7.5, 150 mM NaCl, 2 mM dithiothreitol, 0.1 mM EDTA, 10% glycerol, 0.1% Nonidet P-40) and detected by autoradiography following SDS-PAGE.

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

TA-p73 Isoforms Are Transactivation-defective-- Because TA-p73 proteins share the DNA-binding domain with the full-length isoforms, DNA-binding properties were analyzed by EMSA. As shown in Fig. 2A, full-length proteins as well as TA isoforms all form complexes with a p53 consensus oligonucleotide with comparable affinity, which can be specifically competed and supershifted with appropriate antibodies. Whereas the full-length p73 containing complexes were supershifted by antibodies directed against the C terminus and the N terminus, TA-p73 DNA complexes, which lack the N-terminal epitope, were only supershifted by the C-terminal antibody.

View larger version (66K): [in this window] [in a new window]   Fig. 2.   Characterization of TA-p73 isoforms. A, EMSA demonstrating sequence-specific binding of in vitro translated p73, TA-p73, p73, and TA-p73 to a consensus p53 oligonucleotide. Unlabeled wild-type or scrambled p53 oligonucleotides were used as specific or nonspecific competitors (comp.), respectively. Where indicated, anti-p73 antibodies against a C-terminal (ER15) or N-terminal (Ab-7) epitope were added. The asterisk indicates a nonspecific complex. B-D, transcriptional activity of TA-p73 was analyzed by luciferase assays in H1299 cells cotransfected with 1 µg of luciferase reporter plasmid (pGL3-p53 or pGL3-SV40P) and increasing amounts of p53 or p73 expression plasmids (50, 100, and 200 ng).

Next we analyzed p73-mediated transactivation of p53-regulated reporters by luciferase assay by transient transfection in p53-null H1299 cells (Fig. 2, B and C). Although full-length p73 and p73 activated a p53-responsive luciferase reporter plasmid containing three p53 binding sites upstream of a TATA box, no significant transactivation was observed for TA-p73 and TA-p73, consistent with the lack of the N-terminal transactivation domain. Protein expression of TA-p73 was verified by Western blot (data not shown). Furthermore, wild-type p53 has been shown to inhibit transcription from several viral and cellular promoters without known p53-binding sites such as the SV40 early promoter (30). However, certain transforming p53 mutants or artificial N-terminal deletion mutants lack transrepression activity or even activate such promoters (31, 32). As shown in Fig. 2D, full-length p73 is also a potent repressor of the SV40 promoter, whereas TA-p73 leads to a significant induction of promoter activity.

The effect of TA-p73 on the expression of endogenous p53-regulated target genes (p21^CDKN1A, HDM2, 14-3-3, PIG3, and PIDD) was evaluated by Western blot and semiquantitative RT-PCR in p53-deficient H1299 cells. To obtain high transfection rates we used adenoviral vectors encoding p73, TA-p73, p73, TA-p73, p53, or GFP as a control (expression is shown in Fig. 3A). Whereas p53, p73, and to a lesser extent p73 activated p53 target genes, TA-p73 consistently failed to induce or even repressed the majority of target genes (Fig. 3, A and B).

View larger version (44K): [in this window] [in a new window]   Fig. 3.   Regulation of endogenous target genes by p73 isoforms. A, Western blot analysis of H1299 cells infected adenoviral vectors expressing the indicated transgenes using antibodies against p73, p53, p21, and actin. B, transactivation of endogenous p53/p73 target genes (CDKN1A, HDM2, 14-3-3, PIG3, and PIDD) was analyzed by semiquantitative RT-PCR on total RNA of H1299 cells after infection with adenoviral vectors expressing the indicated transgenes.

TA-p73 Proteins Fail to Induce Cell Cycle Arrest and Apoptosis-- The inability to induce growth arrest and cell death-associated target genes suggests that TA-p73 lacks the cytotoxic activity considered to be the primary mechanism for tumor suppression by p53 and perhaps TA-p73. Indeed, compared with p53 and especially with full-length p73, both TA-p73 and TA-p73 show a significantly reduced ability to suppress colony formation (Fig. 4A). To investigate the underlying causes, we analyzed the cell cycle profile of H1299 cells infected with recombinant adenoviral vectors expressing p53, p73, or TA-p73. As shown in Fig. 4B, both p53 and p73 induced growth arrest and apoptosis, whereas TA-p73 induced no significant cell cycle aberrations. Consistently, only p53, p73, and to a lesser extent p73 induced DNA fragmentation as an indicator of apoptotic cell death (Fig. 4C). These data demonstrate that N-terminally truncated p73 proteins retain the DNA-binding properties of full-length p73 but due to the lack of the transcriptional activation domain in the N terminus fail to activate typical p53 target genes and consequently lack the ability to induce cell cycle arrest and apoptosis.

View larger version (26K): [in this window] [in a new window]   Fig. 4.   TA-p73 is unable to induce growth arrest and apoptosis. A, long-term cytotoxicity of TA-p73 expression was analyzed by colony formation assay in comparison with p53 and full-length p73. The absolute colony number was obtained from duplicate experiments. Bars, S.D. Flow cytometric analysis (B) and DNA fragmentation analysis (C) of H1299 cells after infection with adenoviral vectors expressing the indicated transgenes are shown. The sub-G1 region is marked as M1.

TA-p73 Isoforms Act as Dominant-negative Inhibitors of p53 and TA-p73-- The ability to target p53 DNA binding sites in the absence of transactivation functions suggests that TA-p73 might compete with transactivation-competent isoforms of p73 and wild-type p53. As shown in Fig. 5A, expression of increasing amounts of TA-p73 significantly inhibited transactivation of a p53-regulated luciferase reporter by both p53 and transactivation-competent TA-p73. Consistently, induction of p53 target genes by adenoviral expression of p53 in H1299 cells was completely inhibited by co-infection with TA-p73-expressing vectors (Fig. 5B). In addition, TA-p73 inhibited the transactivation function of endogenous p53 in MCF-7 cells following treatment with adriamycin, despite efficient stabilization of the p53 protein independently of TA-p73 expression (Fig. 4, C and D). As a consequence of the induction of proapoptotic genes, adenoviral expression of p53 led to a rapid loss of cell viability in H1299 cells, which could be significantly inhibited by TA-p73 expression (Fig. 5E). These data indicate that overexpression of TA-p73 inhibits p53- and TA-p73-induced target gene activation, thereby blocking apoptotic cell death as a major tumor suppressor function of p53.

View larger version (45K): [in this window] [in a new window]   Fig. 5.   TA-p73 acts as a dominant-negative inhibitor of p53 and TA-p73. A, transactivation by p53 and p73 is inhibited by coexpression of TA-p73. H1299 cells were cotransfected with 1 µg of the p53-responsive luciferase reporter plasmid pGL3-p53, 100 ng of p53 or p73 expression plasmid, and increasing amounts (50, 100, 200, and 1000 ng) of TA-p73 expression plasmid. RLU, relative luciferase unit(s). B, inhibition of p53-induced activation of endogenous p53 target genes by TA-p73. H1299 cells were infected with 1 multiplicity of infection of Adp53 and 5 multiplicities of infection of AdTA-p73 or AdTA-p73. As a control, uninfected cells, AdGFP-infected cells, and cells infected with Adp53 and AdGFP were included. Expression of HDM2, PIG3, PIDD, and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was analyzed by semiquantitative RT-PCR analysis on total RNA. C, endogenous p53 is stabilized by adriamycin (Doxo) in the absence or presence of TA-p73. Wild-type p53-expressing MCF-7 cells were infected with AdGFP, AdTA-p73, or AdTA-p73 and treated with adriamycin as indicated. p53, p73, and actin protein levels were monitored by Western blot. 293 and Saos-2 cells are shown as positive and negative controls for p53. D, inhibition of adriamycin (Doxo)-induced p53-dependent transactivation by TA-p73. MCF-7 cells were treated as in C, the expression of various p53/p73 target genes was analyzed by semiquantitative RT-PCR analysis. E, expression of TA-p73 rescues cells from p53-induced apoptosis. H1299 cells were infected with 5 multiplicities of infection (moi) of AdGFP, AdTA-p73, or AdTA-p73 and increasing amounts of Adp53. Cell viability was assessed by MTT assay.

Mechanism of Dominant-negative Activity-- In general, two basic mechanisms for inhibition of p53 are conceivable. First, TA-p73 may physically interact with and sequester p53 to form hetero-oligomers that are transactivation-incompetent (33). Second, as TA-p73 retains the core DNA-binding domain and exhibits binding specificity for p53 binding sites, simple competition for DNA sites might prevent p53 or TA-p73 from binding to target gene promoters. To determine whether TA-p73 inhibits p53 function by interfering with sequence-specific DNA binding of p53, we used a fusion protein of p53, the Gal4 DNA-binding domain, and a Gal4-dependent luciferase reporter construct (Gal-TK-Luc). Transfection of increasing amounts of TA-p73 did not lead to substantial inhibition of Gal4-p53-induced reporter activity as was observed for p53 on a p53-dependent reporter (Fig. 5A), suggesting that interference with the sequence-specific DNA binding of p53 might be the primary mechanism of inhibition (Fig. 6A).

View larger version (40K): [in this window] [in a new window]   Fig. 6.   Mechanism of the dominant-negative effect of TA-p73. A, reduced inhibition of Gal-p53 induced transactivation demonstrates that TA-p73 interferes primarily with sequence-specific DNA-binding of p53. H1299 cells were cotransfected with 1 µg of the luciferase reporter plasmid Gal-TK-Luc, 100 ng of Gal-p53 expression plasmid, and increasing amounts (50, 100, 200, and 1000 ng) of TA-p73 expression plasmid. B, EMSA demonstrating competition of sequence-specific binding of in vitro translated p53 by TA-p73 but not TA-p73R292H. p53 binding was activated by PAb421. Anti-p73 (ER15) or anti-p53 (DO-1) antibodies were added as indicated. , supershifted TA-p73 complex; , supershifted p53 complex. C, GST-pull-down assay. 35S-labeled in vitro translated p53 or p73 proteins were subjected to GST-pull-down with recombinant GST-p53. The upper panel shows 10% input. D, co-immunoprecipitation (IP) demonstrating physical interaction of TA-p73 with full-length p73 and p53R175H but not wild-type p53. Lysates of H1299 cells, transfected as indicated, were precipitated with either anti-p53 or anti-HA-tag antibodies. Bound proteins were visualized by Western blot as indicated. Immunoblots (IB) of 1% input are shown in the upper panels. E, inhibition of p73- but not p53-mediated transactivation by TA-p73R292H mutant. H1299 cells were transfected with 1 µg pGL3-p53, 100 ng of p53 or p73 expression plasmid, and 100 ng of TA-p73 or TA-p73R292H, respectively.

This view is further supported by a competition EMSA (Fig. 6B). Both p53 and TA-p73 formed specific DNA complexes that could be supershifted with appropriate antibodies (ER15 for TA-p73 and DO-1 for p53), whereas a DNA binding-defective mutant of TA-p73 (TA-p73R292H) proved unable to bind DNA. The p53 complexes could be efficiently competed by increasing amounts of TA-p73 but not the DNA binding-defective mutant of TA-p73, which underlines the importance of an intact DNA-binding domain of TA-p73 to inhibit p53 function.

To investigate whether TA-p73 interacts with and sequesters p53, we performed in vitro and in vivo protein interaction assays. A GST-pull-down assay with recombinant GST-p53 and various ³⁵S-labeled in vitro-translated p53 and p73 proteins demonstrated strong homotypic interaction between GST-p53 and both wild-type and mutant p53 (Fig. 6C). However, neither full-length TA-p73 nor truncated TA-p73 isoforms significantly bound to GST-p53. These data could be confirmed by in vivo immunoprecipitation. TA-p73 co-precipitated only with the p53 mutant p53R175H, which was previously shown to interact with full-length p73 but not with wild-type p53 (Fig. 6D). The failure of TA-p73 to directly interact with wild-type p53 precludes formation of inactive p53/TA-p73 hetero-oligomers and confines the mechanism of p53 inhibition to competition on the promoter level.

In contrast, TA-p73 specifically co-precipitated with full-length, wild-type p73 (Fig. 6D). Interestingly, the DNA-binding defective TA-p73 mutant TA-p73R292H efficiently inhibited p73- but not p53-mediated transactivation, suggesting that TA-p73R292H inhibits full-length p73 by formation of DNA-binding defective hetero-oligomers (Fig. 6E). Therefore, competition for DNA binding appears to be the major mechanism of p53 inhibition, whereas inhibition of full-length p73 can also be achieved by formation of heteromeric complexes between full-length p73 and TA-p73.

Inhibition of p53-mediated Apoptosis by Up-regulation of Endogenous p73-- To demonstrate the physiological relevance of the dominant-negative activity of TA-p73, we analyzed the effect of endogenous p73 on p53 function. Treatment of wild-type p53 expressing SH-SY5Y neuroblastoma cells with RA resulted in the induction of both endogenous full-length and N-terminally truncated p73 proteins, consistent with recent reports on induction of p73 by RA in other neuroblastoma and myeloid leukemic cells (Fig. 7A) (16, 34). Up-regulation of p73 expression by RA therefore mimics p73 overexpression in cancer cells. Treatment of SH-SY5Y cells with RA conferred resistance to p53-dependent apoptosis induced by the protein kinase inhibitor H-7 (Fig. 7C) without interfering with p53 protein accumulation (Fig. 7B), suggesting that RA inhibits apoptosis signaling via p53, possibly by induction of TA-p73 (27). In fact, overexpression of TA-p73 enhanced resistance to H-7 similar to RA itself (Fig. 7C). These findings strongly suggest that induction of endogenous p73 is able to inhibit apoptosis signaling by endogenous p53.

View larger version (16K): [in this window] [in a new window]   Fig. 7.   Inhibition of p53 by up-regulation of endogenous p73. A, Western blot demonstrating induction of p73 in SH-SY5Y neuroblastoma cells by treatment with RA for 4 days. In vitro translated p73 proteins are shown for comparison. B, Western blot of whole cell extracts showing an increase in p53 protein levels by treatment with H-7 for 24 h in SH-SY5Y cells cultured for 4 days in the absence or presence of RA. C, MTT assay showing H-7 induced loss of cell viability in RA-treated and AdTA-p73-infected SH-SY5Y cells in comparison with mock and AdGFP-infected cells.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

The identification of the p53-homologous TP73 gene on chromosome 1p36, a genomic region frequently deleted in a variety of human cancers, suggested that p73 has tumor suppressor activity similar to that of the classical tumor suppressor, p53. However, p73 is not commonly mutated in all tumor entities analyzed so far. Instead, overexpression of wild-type p73 has been reported frequently and positively correlates with prognostically relevant parameters.¹ Considering that oncogene-induced up-regulation of p73 expression causes apoptosis, sustained overexpression of p73 would therefore require inhibition of its inherent proapoptotic activity (24, 35). On one hand, p53 mutants were demonstrated to inhibit the proapoptotic activity of full-length p73 in a dominant-negative fashion by generating defective hetero-oligomers with wild-type p73. On the other hand, the TP73 gene itself might encode anti-apoptotic isoforms. The first evidence for the latter mechanism was based on the analysis of murine p73, which encodes N-terminally truncated p73 isoforms (N-p73) derived from a cryptic promoter in intron 3 (8). Murine N-p73 is a potent anti-apoptotic protein, which rescues sympathetic neurons from apoptosis induced by nerve growth factor withdrawal or p53 overexpression. Because sympathetic neuron death following nerve growth factor withdrawal is p53-dependent, N-p73 possibly inhibits neuronal apoptosis by acting as a direct antagonist to p53 (18). Recently the human homologue of N-p73 was cloned and shown to possess similar p53- and TA-p73-inhibiting functions, suggesting that increased expression of N-p73 could be involved in tumorigenesis (20-22).

Apart from N-p73, which is a physiological transcript regulated by its own independent promoter, in human tumor cells N-terminally truncated p73 proteins (TA-p73) are also encoded by alternatively (aberrantly) spliced transcripts that lack exon 2 (p73ex2) (3, 19, 23, 36). In addition, we² and others (22) recently described other alternatively spliced transcripts, which either lack exons 2 and 3 (p73ex2/3) or include exon 3B (N'-p73). In an analysis of ovarian cancers, expression of p73ex2 was detected exclusively in cancer cell lines and invasive tumor tissues but not in semi-malignant borderline tumors (23). In an analysis of hepatocellular carcinomas, both p73ex2 and p73ex2/3 were shown to be up-regulated in tumor tissue compared with surrounding normal liver tissue.² Considering that ectopic expression of these isoforms inhibits transactivation and apoptosis induction by p53 and TA-p73 and that many inhibitors of p53 act as transforming oncogenes, they might as well be involved in tumorigenesis (19). This hypothesis was supported by our own findings demonstrating that expression of p73ex2/3 promotes anchorage-independent growth of NIH3T3 cells and tumor growth in nude mice.²

In this study we further characterized the effects of TA-p73 expression and analyzed the mechanism of p53/p73 inhibition in more detail. As sequence-specific DNA binding is a prerequisite for regulation of specific genes, we first confirmed that TA-p73 retains its DNA-binding competence and specificity for p53 binding sites. Due to the lack of the N-terminal transactivation domain, TA-p73 acts as a DNA-binding factor without transactivation function, thereby acting as a dominant-negative inhibitor by blocking the proapoptotic activity of p53 and full-length p73.

In general, two basic mechanisms for inhibition of p53 are conceivable. First, TA-p73 may physically interact with and sequester p53 to form hetero-oligomers that are transactivation-incompetent (33). In support of this theory, Kaghad et al. (3) demonstrated weak interactions between full-length p73 and p53 in a yeast two-hybrid assay. However, others failed to find an interaction between p53 and p73 using purified oligomerization domains or to detect hetero-oligomeric complexes of full-length p73 and wild-type p53 in coimmunoprecipitation and GST pull-down assays (37-40). Consistently, we were unable to detect a physical interaction of TA-p73 and wild-type p53. However, a physical interaction between p73 and p53 has been reported repeatedly for several p53 mutants (38-41). As shown by Gaiddon and colleagues (40), interaction between mutant p53 and wild-type p73 is mediated by the p53 core domain and correlates with recognition of p53 by the conformation-sensitive monoclonal antibody PAb240. These data indicate that physical interaction between p53 and p73 requires a special (mutant) conformation of the p53 core domain and explain specific interaction of TA-p73 with mutant but not wild-type p53.

Second, as TA-p73 retains the core DNA-binding domain and exhibits binding specificity for p53 binding sites, simple competition for DNA sites might prevent p53 or TA-p73 from binding to target gene promoters. This mechanism is supported by our competition EMSA, which demonstrates efficient disruption of p53 complexes by increasing amounts of TA-p73. This competition is absolutely dependent on the DNA binding ability of TA-p73, as shown by experiments with the DNA binding-defective mutant, R292H. Consistently, p53-mediated transactivation was inhibited by TA-p73 but not TA-p73R292H. In addition, inhibition of p53 targeted by fusion to the Gal4 DNA-binding domain to a Gal4-regulated promoter was significantly reduced. We could observe only a 50% reduction with the highest amount of TA-p73. This might hint at additional inhibitory functions of TA-p73 unrelated to interference with sequence-specific DNA-binding or simply be due to unspecific quelching effects. In summary, the data support strongly a model in which TA-p73 exerts a dominant-negative effect by displacing p53 from target gene promoters (Fig. 8A).

View larger version (26K): [in this window] [in a new window]   Fig. 8.   Model for the dominant-negative mechanism of TA-p73. A, inhibition of p53 by direct competition for promoter binding. B, inhibition of full-length TA-p73 by competition for promoter binding and/or formation of transactivation-defective hetero-oligomers.

In contrast, inhibition of full-length p73 appears to be more complex. Because both TA-p73 and TA-p73 efficiently bind to target DNA, simple competition for DNA binding will certainly be involved. However, whereas TA-p73 does not interact with wild-type p53, we observed protein-protein interaction between TA-p73 and TA-p73 by in vivo immunoprecipitation. This interaction, which results in the formation of transactivation-defective hetero-oligomers, appears to be sufficient to inhibit transactivation by TA-p73. In fact, there is no difference in inhibition of TA-p73 by wild-type TA-p73 and the DNA binding-defective mutant TA-p73R292H, which still interacts with TA-p73 (data not shown). Our experiments therefore clearly demonstrate that the dominant-negative effect of TA-p73 involves different mechanisms for inhibition of p53 and TA-p73 (Fig. 8B).

All of the experiments on TA-p73 function by us and others relied on ectopic overexpression of TA-p73. To explain the increased expression level of p73 in tumor tissues with the described dominant-negative effect of TA-p73, it is therefore important to demonstrate that increased expression of endogenous p73 has a similar inhibitory effect on p53 function. It has recently been described that p73 is up-regulated during differentiation of neuroblastoma cells induced by RA (34). Consistently, treatment of wild-type p53-expressing SH-SY5Y cells with RA resulted in increased expression of p73. Interestingly both full-length TA-p73 and truncated TA-p73 levels were elevated simultaneously. Because tumor tissues usually also show concomitant up-regulation of full-length and TA-p73 expression,² this system is an appropriate model to assess the net function of p73 overexpression. SH-SY5Y cells undergo p53-dependent apoptosis when treated with the protein kinase inhibitor H-7 (27). Although p53 protein was stabilized by H-7 in both untreated and RA-treated cells, apoptosis was significantly reduced by RA treatment. Because ectopic expression of TA-p73 had an effect similar to RA treatment on cell survival, it can be assumed that up-regulation of endogenous TA-p73 is responsible for this effect.

Together, our data show that TA-p73 proteins are effective inhibitors of p53 and TA-p73 function. Consistently, increased expression of endogenous p73 including expression of TA-p73 protects cells from p53-dependent apoptosis. These functions of TA-p73 provide a possible explanation for the high level of p73 expression in human cancer cells, even in the absence of p73 mutations, although additional work is needed to further investigate the role of TA-p73 for tumor development and progression.

	ACKNOWLEDGEMENTS

We thank A. Eggert for providing the neuroblastoma cell line and K. Lennarz for support in flow cytometry analysis.

	FOOTNOTES

^* This work was supported by a grant from the Deutsche Krebshilfe, Dr. Mildred Scheel Stiftung (to B. M. P.).The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

To whom correspondence should be addressed: Center for Cancer Research and Cancer Therapy, Institute of Molecular Biology, University of Essen, Medical School, Hufelandstr. 55, D-45122 Essen, Germany. Tel.: 49-201-723-3158; Fax: 49-201-723-5974; E-mail: brigitte.puetzer@uni-essen.de.

Published, JBC Papers in Press, February 13, 2002, DOI 10.1074/jbc.M200480200

¹ Stiewe, T., and Pützer, B. M. (2002) Cell Death & Differ. 9, 237-245

² T. Stiewe, S. Zimmerman, A. Frilling, H. Esche, and B. M. Pützer, submitted for publication.

	ABBREVIATIONS

The abbreviations used are: RT-PCR, reverse transcription PCR; EMSA, electrophoretic mobility shift assay; MTT, 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide; HA, hemagglutinin; GFP, green fluorescent protein; TA, transactivation domain; GST, glutathione S-transferase; RA, retinoic acid.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

<