Original Paper

Oncogene (2005) 24, 8093–8104. doi:10.1038/sj.onc.1208951; published online 25 July 2005

TP53INP1 is a novel p73 target gene that induces cell cycle arrest and cell death by modulating p73 transcriptional activity

Richard Tomasini1, Mylène Seux1, Jonathan Nowak1, Caroline Bontemps1, Alice Carrier1, Jean-Charles Dagorn1, Marie-Josèphe Pébusque1, Juan L Iovanna1 and Nelson J Dusetti1

1INSERM U624, Stress Cellulaire, IFR 137-Institut de Cancérologie et Immunologie de Marseille, Université de la Méditerranée, Marseille, France

Correspondence: NJ Dusetti, INSERM U624, Parc Scientifique et Technologique de Luminy, Case 915, 13288 Cedex 9, Marseille, France. E-mail: dusetti@marseille.inserm.fr

Received 22 October 2004; Revised 15 June 2005; Accepted 22 June 2005; Published online 25 July 2005.



TP53INP1 is an alternatively spliced gene encoding two nuclear protein isoforms (TP53INP1alpha and TP53INP1beta), whose transcription is activated by p53. When overexpressed, both isoforms induce cell cycle arrest in G1 and enhance p53-mediated apoptosis. TP53INP1s also interact with the p53 gene and regulate p53 transcriptional activity. We report here that TP53INP1 expression is induced during experimental acute pancreatitis in p53-/- mice and in cisplatin-treated p53-/- mouse embryo fibroblasts (MEFs). We demonstrate that ectopic expression of p73, a p53 homologue, leads to TP53INP1 induction in p53-deficient cells. In turn, TP53INP1s alters the transactivation capacity of p73 on several p53-target genes, including TP53INP1 itself, demonstrating a functional association between p73 and TP53INP1s. Also, when overexpressed in p53-deficient cells, TP53INP1s inhibit cell growth and promote cell death as assessed by cell cycle analysis and colony formation assays. Finally, we show that TP53INP1s potentiate the capacity of p73 to inhibit cell growth, that effect being prevented when the p53 mutant R175H is expressed or when p73 expression is blocked by a siRNA. These results suggest that TP53INP1s are functionally associated with p73 to regulate cell cycle progression and apoptosis, independently from p53.


TP53INP1, p73, apoptosis, transcriptional activity



Tumor protein 53-induced nuclear protein 1 (TP53INP1) encodes two nuclear protein isoforms (TP53INP1alpha and TP53INP1beta; Tomasini et al., 2001). This gene is transcriptionally induced in the pancreas of mice with acute pancreatitis and in several cell lines upon exposure to many stress agents. Importantly, the TP53INP1 promoter region presents a canonical p53-responsive element, which was shown to be functional (Okamura et al., 2001; Tomasini et al., 2002). Therefore, TP53INP1 appears as a new stress-induced p53 target gene. Moreover, we showed that TP53INP1alpha and TP53INP1beta interact with p53 and the homeodomain-interacting protein kinase-2 (HIPK2) within the promyelocytic leukemia nuclear bodies (PML-NBs). These interactions modify the transcriptional activity of p53 on several target genes such as p21, PIG-3 and mdm2 (Tomasini et al., 2003). Finally, we also found that overexpression of TP53INP1s promotes apoptosis and cell cycle arrest in G1 phase (Tomasini et al., 2003). Therefore, TP53INP1 appears as a key element in p53-mediated cell death and cell cycle arrest, induced by cellular stresses.

p53 shares structural homologies with p73, which encodes a nuclear transcription factor (Kaghad et al., 1997; Strano et al., 2001). p73 produces many isoforms, resulting from alternative splicing and from initiation of transcription by several promoters (Courtois et al., 2004). Some p73 isoforms share functional similarities with p53. p73 overexpression can also activate the expression of p53 target genes and induce cell cycle arrest and/or apoptosis (Kaghad et al., 1997; Jost et al., 1997). It was shown that most of the p73-mediated biological activities are linked to its transactivation function. Whereas p53 is activated and stabilized by a large number of stresses including hypoxia, several oncogenes, DNA-damaging agents and radiation, p73 is known to be stabilized only in response to cisplatin, italic gamma-irradiation and oncogenes (E1A and HPVE6) (Hamer et al., 2001; Das et al., 2003a, 2003b; Shimodaira et al., 2003). Contrary to p53, p73 mutations are infrequently associated with human cancer (reviewed in Ikawa et al., 1999) and p73-decient mice do not develop tumors (Yang et al., 2000) suggesting that p73 is not a classical tumor-suppressor gene.

Previous works reported that p73 is required for p53-dependent induction of apoptosis in response to DNA damage (Flores et al., 2002). Furthermore, ectopic expression of p73 in p53-null mouse embryo fibroblasts (MEFs) can cause growth arrest, apoptosis, cell migration and differentiation (Jost et al., 1997; Fontemaggi et al., 2001; Sablina et al., 2003) similarly to p53. These effects are achieved through the activation of a plethora of specific target genes. Several reports show that p73 can bind p53-responsive elements and activate some p53 target genes (Stiewe and Pützer, 2001; Obad et al., 2004). These data strengthen the notion that these two proteins may cooperate in overlapping but distinct cell growth and apoptosis signalling pathways in response to stress.

The aim of this work is to investigate whether TP53INP1 can be transcriptionally induced by p73 in a p53-independent manner and if the TP53INP1s modulate p73 activity. Here, we show that TP53INP1 expression is upregulated in pancreatic acinar cells during acute pancreatitis in p53-deficient mice and in p53-/- MEFs treated with cisplatin. In addition, we demonstrate that ectopic expression of p73 in p53-deficient cells induces TP53INP1 expression. Moreover, TP53INP1 upregulation modifies in turn the p73 transactivation capacity on some p53 target genes, including TP53INP1 itself. Finally, we also found that TP53INP1alpha and TP53INP1beta overexpressions induce cell growth arrest and promote cell death in a p53-independent manner. Our results provide evidence that p73 regulates TP53INP1 activation which, in turn, modulates p73 transactivation activity leading to cell cycle arrest and cell death in a p53-independent pathway.



TP53INP1 is induced in the pancreas of p53-null mice with acute pancreatitis

We have previously shown that TP53INP1 expression is strongly induced during acute pancreatitis in the pancreas of p53 wild-type (wt) mice (Tomasini et al., 2001) and that TP53INP1 is transcriptionally activated by p53 under different stress conditions (Tomasini et al., 2002). In order to study whether the upregulation of TP53INP1 in pancreas is p53-dependent or not, we induced acute pancreatitis in p53-null mice. Immunohistochemical analyses reveal that TP53INP1 expression is increased during the acute phase of pancreatitis (Figure 1). TP53INP1s staining is only observed in acinar cells and no staining is observed in Langerhans islets or pancreatic duct cells (data not shown). Both RT–PCR (Figure 1a) and Western blot (Figure 1b) data confirm the strong induction of TP53INP1alpha and TP53INP1beta mRNAs and protein expression in the pancreas 3 h from the beginning of the acute phase of pancreatitis (Figure 1c). The peak value of TP53INP1 expression is reached at 24 h, following the evolution of pancreatitis associated protein (PAP) used as marker of acute pancreatitis (Iovanna et al., 1991). Taken together, these results indicate that, during acute pancreatitis in mice, the induction of TP53INP1 in pancreatic acinar cells is p53 independent.

Figure 1.
Figure 1 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Pancreatic induction of TP53INP1 in p53-null mice. Pancreatic tissues from p53-null mice were removed at different times after induction of acute pancreatitis and processed to extract total protein and RNA or mounted on slides for immunohistochemical analysis. (a) RT–PCR analysis. Total RNA was reverse-transcribed and PCR amplified using specific primers. The two isoforms, that is, TP53INP1alpha and TP53INP1beta were detected. RL3 expression was used as internal control. (b) Western blot analysis. Immunoblot analyses using the anti-TP53INP1s antibody were performed using whole protein extracts. Actin was used as control for loading and integrity. (c). Immunohistochemical analysis. TP53INP1 levels in p53-null mouse pancreas were evaluated by immunohistochemistry with anti-TP53INP1 monoclonal antibody (40 times )

Full figure and legend (145K)

TP53INP1s modify the cell cycle independently of p53 cell status

We showed that the transcription of TP53INP1 can be induced in the absence of p53 (see above) and that TP53INP1s overexpression induces the arrest of the cell cycle arrest in G1 and apoptosis, both mediated by p53 (Tomasini et al., 2003). We therefore investigated whether TP53INP1 ectopic expression could affect the cell cycle of p53 mutant cells. Two p53-null cell lines (H358 and H1299) and two cell lines in which TP53 is mutated at codon 273 (Panc-1 and SW480) were transiently transfected with plasmids expressing either TP53INP1 isoform. The percentage of cells in each phase of the cell cycle was quantified by propidium iodide and flow cytometric analyses. In Figure 2a and b, we show that TP53INP1alpha and TP53INP1beta induce G1 cell cycle arrest in all cell lines. The strongest cell cycle arrest is observed for H358 cells whatever the TP53INP1 isoform used (G1/G2 ratio: 2.8 for GFP control cells vs 4.4 for TP53INP1s transfected cells). The cell cycle arrest in G1 is linked to a subsequent decrease in the percentage of cells in S phase. In contrast, the percentage of cells undergoing cell death (sub-G1 phase) is increased to different extents, depending on the cell line: H358 and Panc-1 show very similar percentage of cells in sub-G1 phase, whether TP53INP1 isoforms are overexpressed or not, whereas in H1299 cells their overexpression is followed by significant increase in the sub-G1 DNA content, especially for TP53INP1beta vs 11 and 17% for GFP and TP53INP1alpha or beta, respectively). Finally, significant apoptotic activity in SW480 cells was observed only after TP53INP1beta overexpression (6 vs 14% for GFP and TP53INP1beta, respectively). These results strongly suggest that TP53INP1alpha and TP53INP1beta overexpressions induce G1 cell cycle arrest and cell death independently of p53 activity.

Figure 2.
Figure 2 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

TP53INP1s augment the percentage of cells in sub-G1 phase and induce cell cycle arrest in cell lines, independently from their p53 status. (a) H358, H1299, Panc-1 and SW480 human cell lines were transfected with either pEGFP-N1, pEGFP-N1-TP53INP1alpha or pEGFP-N1-TP53INP1beta. 24 h after transfection cells were treated for FACS, green cells were sorted and cell cycle profile was analysed as described in Materials and methods. Cell count and DNA content are represented by the ordinate and abscissa, respectively. The sub-G1, G1, S and G2/M fractions were shaded and quantitated using the Modfit software package. (b) Detailed values corresponding to the different phases of cell cycle, plus G1/G2 ratio, for each transfection in all cell lines. The results are the meanplusminuss.d. from three independent experiments

Full figure and legend (56K)

Increases in TP53INP1 mRNA levels upon cisplatin treatment of p53-/- cells require p73 activity

Cisplatin is a strong cytostatic and antineoplastic agent able to induce DNA-damage, cell cycle arrest and apoptosis in a p53-independent way (Gong et al., 1999). We used that drug to further characterize the p53-independent induction of TP53INP1 transcription. p53-/- MEFs were treated with cisplatin and TP53INP1 expression was monitored by RT–PCR from 0 to 24 h (Figure 3a). Basal TP53INP1 mRNA levels are low and increase with time after exposure to cisplatin. Interestingly, TP53INP1beta mRNA level increased faster than that of TP53INP1alpha. To confirm these findings, p53+/+ MEFs were also treated with cisplatin and TP53INP1 mRNA expression was measured after 0 and 24 h. Figure 3b shows that cisplatin can also induce TP53INP1 expression in these conditions. It is well documented that p73 is involved in the cellular response to cisplatin (Torigoe et al., 2005), hence the hypothesis that p73 could regulate the induction of TP53INP1 transcription. To check that hypothesis we monitored the influence on TP53INP1 transcriptional activation of knocking down p73 activity by specific siRNA duplex. Figure 3c and d show that if TP53INP1 transcription is induced by cisplatin in p53-/- H1299 cells, that induction is prevented by pretreating the cells with a functional p73 siRNA duplex. Together, these results indicate that cisplatin is able to activate TP53INP1 in the absence of p53, through a p73-dependent pathway.

Figure 3.
Figure 3 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

p73-dependent induction of TP53INP1 mRNA expression upon cisplatin treatment. Cells were treated with cisplatin 25 muM during 1 h. The cisplatin treatment was stopped by medium change. Cells were recovered at different times after treatment. Total RNA was reverse-transcribed and PCR amplified (RT–PCR). Expression of RL3 served as internal control. Total protein extracts were analysed by Western blot. beta tubulin expression was used as a loading and integrity control. (a and b) TP53INP1 expression analysis by RT–PCR on p53-null and p53 wild-type MEFs treated with cisplatin. (c) TP53INP1 expression analysis by RT–PCR after p73 siRNA treatment in H1299 cells. (d) p73 expression analysis by Western blot after p73 siRNA treatment in H1299 cells

Full figure and legend (54K)

p73alpha and p73beta can induce TP53INP1

TP73 undergoes alternative splicing generating several protein isoforms (Irwin and Kaelin, 2001; Strano et al., 2001). Among them, p73alpha and p73beta have the highest transactivation activities (Yang et al., 2002). p73alpha encodes the entire molecule (636 amino acids) while p73beta is shorter, lacking the C-terminal tail. We used cDNAs encoding these two p73 isoforms for subsequent studies. TP53INP1 expression was first assessed by RT–PCR in H1299 cells, transiently transfected with either p73alpha or p53, used as positive control. Quantification was carried out by real-time PCR. Figure 4a shows that both TP53INP1 mRNAs, barely detectable in cells transfected with the empty vector (first row), were induced to similar extents in cells transfected with either p53 ( times 3.5) or p73alpha ( times 1.8). We verified by Western blotting that expression of the encoded proteins was correct (Figure 4b). Figure 4b shows evidence of p21 protein accumulation in transfected cells, confirming that p53 and p73alpha were transcriptionally active. In a second approach, p53-null MEFs were infected at different multiplicity of infection (MOI) with adenovirus-expressing p73beta or GFP, as control, and TP53INP1 expression was monitored by RT–PCR (Figure 4c). Upregulation of TP53INP1 mRNAs in p73beta infected cells was found proportional to the MOI and the two mRNAs increased in parallel. As expected, no induction of TP53INP1 mRNA expression was observed with the GFP adenovirus alone, even at high MOI (Figure 4c). p73beta and GFP protein expressions were confirmed by Western blot analysis (Figure 4d). Taken together, these results demonstrate that TP53INP1 is transcriptionally induced upon ectopic expression of p73alpha and p73beta.

Figure 4.
Figure 4 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

TP53INP1 gene induction by 73alpha and p73beta overexpression. (a) TP53INP1 mRNA expression analysis by RT–PCR in H1299 cells transiently transfected with an empty pcDNA3 vector (Ev) or a recombinant expression vector-expressing p53 or p73alpha. RL3 expression was used as internal control. The same total RNA was used for LightCycler experiments. (b) Western blot analysis of p53, p73alpha and p21 expressions in H1299 cells transfected as in (a). beta tubulin expression was used as a loading and integrity control. (c) RT–PCR analysis of TP53INP1 mRNA expression in H1299 cell line transiently infected with different multiplicity of infection (MOI) with an adenovirus-expressing GFP or p73beta. (d) Western blot analysis of GFP and p73 in H1299 infected as in (c)

Full figure and legend (77K)

TP53INP1 is a transcriptional target of p73

As previously reported (Tomasini et al., 2002) the mouse TP53INP1 promoter region contains a functional p53-responsive element located between -1364 and -1239 bp (Figure 5a). Since p73 shares remarkable sequence homology in DNA binding domain with p53 (Figure 5e), it can recognize p53 binding sites and consequently activate some p53 target genes (Fontemaggi et al., 2002). We investigated whether p73 could activate the TP53INP1 promoter. H1299 cells were transiently cotransfected with TP53INP1 promoter reporter constructs together with plasmids-expressing p73alpha or p53 as positive control. TP53INP1 promoter reporter constructs were the full promoter (p-p53wt/CAT) and a deletion mutant (p-pmut/CAT) lacking the p53-binding element (Figure 5b). CAT activity analysis indicated that, similar to p53, p73alpha activates strongly p-p53wt/CAT (Figure 5c). Deletion of the p53-responsive element (p-pmut/CAT) notably reduced p73 activation, demonstrating that it was indeed functional. In order to confirm these results, H1299 cells were cotransfected with the same TP53INP1 reporter constructs together with a plasmid-expressing p73alpha and a plasmid-encoding p53R175H, a p53 mutant known to markedly reduce the transcriptional activity of the p73 isoforms (Di Como et al., 1999; Strano et al., 2000). As expected, p53R175H decreased the CAT activity generated with p73alpha alone (Figure 5d). Thus, the activity of p73 on the TP53INP1 promoter requires the presence of the p53-responsive element suggesting that p73 overexpression can mediate direct transcriptional regulation of TP53INP1.

Figure 5.
Figure 5 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Functional caracterization of the p73-responsive element in the mouse TP53INP1 promoter. (a) Nucleotide sequence of mouse Trp53inp1 (symbols for mouse and human orthologs are Trp53inp1 and TP53INP1, respectively) 5'-region is given from positions -1535 to +21. The putative p53/p73-binding site and exon 1 are underscored and boxed, respectively. (b) A schematic diagram of TP53INP1 promoter fused to the chloramphenicol acetyl transferase reporter gene used in this study. Numbers refer to nucleotides relative to the first transcription initiation site. (c and d) CAT reporter assays in p53-deficient H1299 cells. Each CAT reporter construct was cotransfected in H1299 cells with expression plasmids for: beta-galactosidase, HA-p73beta, p53, p53R175H or an empty vector. At 24 h after transfection, the CAT activity was measured. The CAT activity of each sample was normalized with beta-galactosidase activity. The empty vector was used to equilibrate the total DNA mass transfected with the different combinations of plasmis. Each value represents the average of at least three independent experiments with standard deviation. (e) Sequence comparison between the p53-responsive element consensus and the p53/p73-binding sites retrieved from different genes and the putative p53/p73-binding site present in the mouse Trp53inp1 promoter region. R represents purine, Y pyrimidine, and W adenine or thymine. Nucleotides mismatching with the consensus sequence for p53 binding are in lower case

Full figure and legend (143K)

The TP53INP1s modify p73 transcriptional activity

In a previous work, we showed that p53 transcriptional activity is modified by TP53INP1s in a promoter-dependent manner (Tomasini et al., 2003). Therefore, we decided to examine the ability of TP53INP1 isoforms to modulate p73 activation on p53 target promoters known to be also activated by p73. p53-/- MEFs were transiently cotransfected with either luciferase reporter plasmids containing p21, Bax or Mdm2 promoters or a CAT reporter plasmid containing the TP53INP1 promoter sequence, together with either: p73alpha, p53 (as positive control) or the empty vector. Cells were harvested 24 h after transfection and the reporter gene activity were measured (Figure 6). TP53INP1 isoforms increased the p73alpha transcriptional activity on the Bax promoter more than twofold (Figure 6a). It is interesting to note that activation by p73alpha is weaker than that of p53. TP53INP1alpha also enhanced the transcriptional activity of p73alpha on the p21 promoter, contrary to TP53INP1beta (Figure 6b). Both isoforms enhanced the activation by p73alpha of the Mdm2 promoter, the effect of TP53INP1beta being stronger (Figure 6c). Surprisingly, both TP53INP1 isoforms inhibited the activation of the TP53INP1 promoter by p53 or p73alpha (Figure 6d). Taken together, these results demonstrate that the TP53INP1s can modulate p73 transcriptional activity, with an efficacy depending on the promoter and on the protein isoform.

Figure 6.
Figure 6 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Effect of TP53INP1alpha and TP53INP1beta on p53 and p73 transcriptional activity in p53-null MEFs. The influence of p53 and p73alpha on the activities of several promoters was assessed in the presence or absence of TP53INP1alpha and TP53INP1beta expression plasmids (a) Bax, (b) p21, (c) Mdm2, (d) TP53INP1. In each experiment, values were normalized to beta-galactosidase activity, which reflects transfection efficiency. Mean values (plusminuss.d.) for at least three independent experiments are shown. Western blot analyses of p73 and p53 proteins, with beta tubulin as loading control, are shown below each graph

Full figure and legend (135K)

TP53INP1alpha and TP53INP1beta overexpressions regulate p73-mediated cell death and cell cycle arrest

p73 was shown in several experimental models to induce cell cycle arrest in G1 and/or G2 and apoptosis (Tsuji et al., 2002; Chau et al., 2004) even in the absence of p53 (Oniscu et al., 2004). Given that i/gene reporter experiments described above demonstrate that TP53INP1s can modulate p73 activity and ii/TP53INP1s induce G1 arrest and cell death in cells lacking p53 activity, we made the hypothesis that, in such cells, TP53INP1s could modulate p73-induced cell cycle arrest and apoptosis. H1299 cells, which express wt p73 but not p53, were co-transfected with plasmids allowing the expression of p73alpha alone or in combination with TP53INP1alpha, TP53INP1beta or p53R175H. Consequences on cell proliferation were compared using colony formation assays. Figure 7a and b show that, as expected, p73alpha alone inhibited colony formation by H1299 cells (Ishida et al., 2000). Also, TP53INP1alpha or TP53INP1beta expressions strongly inhibited colony formation. These results are in agreement with previous studies (Tomasini et al., 2003). In Figure 7a and b (second panels), we show that both TP53INP1 isoforms strongly potentiated the inhibition of colony formation by p73alpha. We noticed that the inhibition of colony formation obtained with the combination of either TP53INP1alpha or TP53INP1beta and p73alpha was stronger than the mere addition of the individual effects, suggesting some sort of synergy. Interestingly, p53R175H cotransfected with p73alpha and TP53INP1alpha or TP53INP1beta restored, in part, colony formation capacity. Nevertheless, this experimental approach was unable to tell whether the observed decrease in cell proliferation was due to cell death or cell cycle arrest. To address that point, we carried out flow cytometry analyses after inhibiting p73 either by the p53R175H mutant or by a specific siRNA. pEGFP-N1-TP53INP1alpha and pEGFP-N1-TP53INP1beta expression plasmids were used to overexpress TP53INP1s-GFP fusion proteins. GFP-expressing plasmid was used as control. p73 knockdown by siRNA was verified by semiquantitative RT–PCR 72 h after transfection (data not shown).

Figure 7.
Figure 7 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Mutant p53R175H counteracts the effect of the TP53INP1s on p73-mediated suppression of cell growth. H1299 cells were transfected with the indicated plasmids prior to selection with Zeocin. (a) Colonies were visualized and counted after staining with crystal violet. (b) The number of colonies generated by cells transfected with the pcDNA4 empty vector was arbitrary settled as 100%. Data presented are the average of at least three independent experimentsplusminuss.d.

Full figure and legend (118K)

Table 1 shows that p73alpha overexpression in combination with GFP-expressing plasmid induced, as expected, a significant arrest of the cell cycle in G1 (increase in G1/G2 ratio from 1.36 to 2.41) while p53R175H overexpression did not alter the cell cycle profile. In contrast, overexpression of p53R175H together with p73alpha inhibited the G1 cell cycle arrest observed with p73alpha alone (G1/G2 ratios of 2.41 and 1.18, respectively). As expected, TP53INP1alpha-GFP and TP53INP1beta-GFP overexpressions induced cell cycle arrest in G1 phase, the percentage of cells in G1 increasing from 49% (control) to 58 and 53%, respectively. The concomitant decrease in the percentage of cells in S phase indicated that the effects of TP53INP1alpha and TP53INP1beta overexpressions were independent from p53. Importantly, cell cycle arrest generated by TP53INP1alpha-GFP and TP53INP1beta-GFP was increased by addition of p73alpha (G1/G2 ratios of 2.23 vs 3.90 and 2.12 vs 2.63, respectively). Addition of p53R175H to the above combination brought back the percentage of cells in G1 to control value reflecting complete suppression of p73alpha and TP53INP1s activities on the cell cycle. TP53INP1alpha and TP53INP1beta were previously reported to induce apoptosis (Tomasini et al., 2001) and this effect is also observed in H1299 cells as judged by the increase in the percentage of cells in sub-G1. Both TP53INP1alpha and TP53INP1beta isoforms increased percentages of cells in sub-G1 phase, TP53INP1beta activity being stronger (3 vs 8% and 3 vs 13%, respectively). Only TP53INP1alpha-mediated cell death was enhanced by p73alpha expression (8 vs 15%) although, under our experimental conditions, transfection of p73alpha alone did not induce significant cell death (3 vs 4%). We also observed that p53R175H decreased cell death induced with the TP53INP1alpha/p73alpha combination. Surprisingly, this mutant was inactive on cell death induction by the TP53INP1beta/p73beta combination. These findings were confirmed by results from Table 2 showing that transfection with a p73 siRNA counteracted the increased percentage of cells in sub-G1 phase occurring upon transfection with TP53INP1alpha or beta (16 vs 8% and 20 vs 13%). Also, p73 siRNA mitigated the cell cycle arrest induced by TP53INP1s (decrease in G1/G2 ratio from 3.05 to 1.66 and 3.00 to 2.08 for TP53INP1alpha-GFP and TP53INP1beta-GFP, respectively). Interestingly, as shown in Tables 1 and 2, TP53INP1alpha activity seems to be more sensitive to the consequences of p73 inhibition than TP53INP1beta activity, especially regarding apoptosis.



Transcription of the TP53INP1 gene is strongly activated by p53 during cell stress (Dusetti et al., 2000; Tomasini et al., 2001, 2003). In previous studies, we identified a p53-responsive element within the TP53INP1 promoter and showed that TP53INP1 is a functional target gene of p53 (Tomasini et al., 2002). In this report, we present evidence that in vivo TP53INP1 can also be induced independently of p53, via p73, a member of the p53 family. We show that TP53INP1 is strongly activated in the pancreas of p53-null mice in the course of experimental acute pancreatitis and in p53-/- MEFs exposed to cisplatin. Cisplatin is a well-known DNA-damaging agent, which induces apoptosis in p53-null cells through activation of a p73 pathway (Gong et al., 1999; Oniscu et al., 2004). Here, we show that ectopic expression of p73 in p53-null cells enabled TP53INP1 transactivation. In addition, we found that the effect of p73 depended, at least in part, on the presence of a p53-responsive element previously described (Tomasini et al., 2002) because deletion of that element reduced markedly but not totally TP53INP1 promoter activity. Altogether our data indicate that TP53INP1 is a direct and functional target of p73. These results are in agreement with several reports showing that, like p53, p73 overexpression transactivates typical p53-responsive genes through a canonical p53/p73 binding site (Figure 5e). The exception to this rule seems to be the human DAN and ADA genes that are specifically activated by p73 (Shinbo et al., 2002; Tullo et al., 2003).

Earlier studies from our laboratory showed that TP53INP1 is a proapoptotic gene induced by p53 (Tomasini et al., 2002). In turn, TP53INP1 expression strongly alters p53 transactivation activity on several p53-dependent promoters such as p21, Bax and Mdm2 with a promoter-specific intensity (Tomasini et al., 2003). This effect could be mediated by the direct interaction between TP53INP1, p53 and the serine-threonine kinase HIPK2 within the PML nuclear bodies (Tomasini et al., 2003). We report here that, similarly, overexpression of p73 transactivates TP53INP1, which in turn modifies p73 activity in a promoter-dependent manner. Consequently, TP53INP1s stimulate p73 cell cycle arrest and pro-apoptotic functions. However, all the attempts to show a TP53INP1-p73 physical interaction failed (Tomasini et al., unpublished results). The ability of TP53INP1s to stimulate the activity of p53 is slightly higher than that seen with p73. Interestingly, the TP53INP1 isoforms modulate differently p73 transcriptional activity. Although both TP53INP1alpha and beta isoforms stimulate the p73 transactivation activity on the Bax promoter, only the alpha isoform can activate p73 activity on the p21 promoter. Together, these data suggest that p53 and p73 might not use an identical set of target genes to induce apoptosis, and that TP53INP1alpha and beta do not increase to the same extent the activities of p53 family members. This observation is particularly important since very little is known about the specific targets of p53 and p73 that are critical in inducing apoptosis. Perhaps cellular regulators of p53 family such as TP53INP1s play an important role in controlling the expression of their target genes in vivo. In addition, both TP53INP1s effectively repress p53 and p73 transcriptional activities on their own promoters, suggesting a negative regulatory feedback loop. Although we do not know yet the molecular mechanism by which TP53INP1s control p53 and p73 transactivation activities, it is clear that TP53INP1s can stimulate their capacity to induce apoptosis and regulate cell cycle. This is of importance since p53 is the gene most frequently mutated in human cancer, in which sensibility of the cells to the induction of apoptosis is drastically diminished. In the absence of functional p53, TP53INP1s could activate p73 to induce cell death (Rodicker and Putzer, 2003). Hence, the activation of TP53INP1s could prevent tumor development. Deciphering the association between the various p53 family members and TP53INP1s might provide important insights into tumorigenesis and cancer treatments.


Materials and methods

Cell lines and cell culture conditions

p53-null MEFs and human carcinoma cell lines, panc-1 (pancreas), SW480 (colon), H358 and H1299 (lung) were grown in Dulbecco's modified Eagle's medium (DMEM) containing 10% fetal bovine serum (FBS), 100 U/ml of penicillin and 100 mug/ml of streptomycin (Gibco/BRL) at 37°C. For drug treatment, cells were exposed to cisplatin (Sigma-Aldrich) at concentration of 25 muM for 1–24 h.

Plasmids and adenovirus

The control plasmid pCMV/betagal was from Promega. pTP53INP1alpha-V5, pTP53INP1beta-V5, pcDNA4, p-p53wt/CAT, p-pmut/CAT, pEGFP-N1, pEGFP-N1-TP53INP1alpha and pEGFP-N1-TP53INP1beta were previously described (Tomasini et al., 2002, 2003). The p53-responsive intronic promoter reporter constructs driving either murine mdm2 (Barak et al., 1993), human bax (Miyashita and Reed, 1995) or human p21WAF1 (el-Deiry et al., 1993) were kindly provided by M Oren and B Vogelstein (Johns Hopkins University, Baltimore, USA). pcDNA-Hu-p53wt was a kind gift from A Sparks (University of Dundee, UK). pcDNA3-p73alpha-HA and pcDNA3-p53R175H were generously provided by G Blandino (Regina Elena Cancer Institute, Italy) and adenovirus Ad-p73beta by S Das (Indian Institute of Science, India).

Adenoviral infections, transient transfections of plasmids and siRNAs

For adenoviral-mediated gene transduction, H1299 cells were plated at a density of 60% per 60 mm dish 1 day prior to infection and grown in DMEM supplemented with 10% FCS, 4 mM glutamine, 50 U/ml penicillin and 50 mug/ml streptomycin. Cells were then incubated 4 h at 37°C in DMEM with 0.1% FCS at different MOI of the adenoviral constructs. Then, the medium was supplemented to 10% FCS and incubated for an additional 20 h.

Cell lines were plated at a 60% density the day before transient transfection of plasmids. Transfections were performed using the FuGENE reagent (Roche Applied Science) according to the manufacturer's protocol.

For siRNA duplex transfection, H1299 cells were plated at a density of 80% 24 h later the siRNA duplex was transfected using OligofectAMINE (Invitrogen) according to the manufacturer's instructions. Cells were then treated with cisplatin 25 muM for 24 h (Figure 3) or, after an additional 24 h period, replated at a density of 60% and the plasmids were transfected the next day as mentioned above (Table 2). The p73 (human specific) siRNA duplex was purchased from cell signaling (catalogue number 6371). A scramble siRNA duplex was used as control.

Experimental acute pancreatitis

Balb/c-mice wt and Balb/c-p53-null mice weighing 30–40 g were maintained in a 12-h light/12-h dark cycle and allowed free access to water and standard rodent chow. Pancreatitis was induced according to Niederau et al. (1985). Mice were intraperitoneally injected with either 50 mug/kg body weight of caerulein dissolved in saline (0.9% NaCl solution) or saline alone every hour over a 6 h-period (seven injections in total). Animals were killed 3, 12 or 24 h after the first injection. Control mice were killed immediately after the first injection. Pancreas were quickly removed, trimmed free of fat and cut into three parts: one part was frozen in liquid nitrogen and stored at -80°C (for protein extraction), another was fixed in PFA 4% and paraffin embedded (for immunhistochemistry experiments) and the last one was immediately used to purify total RNA as previously described by Chirgwin et al. (1979).


Deparaffinized sections (5 mum) of pancreas were hydrated and treated with hydrogen peroxide in PBS to block endogenous peroxidase activity. Sections were preincubated in 1.5% blocking serum for 1 h to reduce nonspecific background signal. They were then incubated overnight at 4°C in a humid chamber with a rat anti-TP53INP1 monoclonal antibody (clone E-12) in 1.5% blocking serum. Thereafter, sections were rinsed in PBS and immunostained by the avidin-biotin complex method (rat ABC staining system, Sc-2019, Santa-Cruz Biotechnology, CA, USA) following the manufacturer's instructions and diaminobenzidine complex was used as chromogen. TP53INP1s labeling was observed by light microscopy.

RNA purification and RT–PCR

Total RNA was extracted from cells using the Trizol reagent (Life Technologies, Inc.) following the manufacturer's instructions. RNA concentrations were determined by spectrophotometry. RNA samples were run in parallel on agarose gel and stained with ethidium bromide to check their quality. RT–PCR was performed in a single step using reagents from the One-Step™ RT–PCR System (Life Technologies, Inc.) following the recommendations of the supplier. Briefly, first-strand cDNA was synthesized from 1.0 mug of total RNA at 45°C for 30 min. After reverse transcription, the RNA/cDNA hybrid was denatured, and the Superscript reverse transcriptase was inactivated by incubating the mix for 2 min at 94°C. Then, amplification cycles were performed as follows: denaturation at 94°C for 15 s, annealing at 55°C for 30 s and extension at 72°C for 1 min. Of the aliquots of RT–PCR reactions 5 mul were subjected to electrophoresis in 1.5% agarose gel and visualized by ethidium bromide staining. Mouse TP53INP1 mRNAs were specifically amplified with forward (5'-CTGCATCTTTGGAATGCTT-3') and reverse (5'-CGACGGAGACCATTTCTGTT-3') primers in positions 464 and 903 of the cDNA, respectively (accession number AY034611). Human TP53INP1 mRNAs were amplified as previously described (Tomasini et al., 2002). TP73 mRNA was specifically amplified with forward (5'-CCCACCACTTTGAGGTCACT-3') and reverse (5'-TCAGCTCCAGGCTCTCTTTC-3') primers in positions 481 and 1223 of the cDNA, respectively (accession number Y11416). Control RL3 mRNA was specifically amplified with sense (5'-GAAAGAAGTCGTGGAGGCTG-3') and antisense (5'-ATCTCATCCTGCCCAAACAC-3') primers in positions 216 and 637 of the cDNA, respectively (accession number NM_013762).

Quantitative real-time PCR

mRNA levels of TP53INPalpha, TP53INP1beta and internal control TBP gene (encoding the TATA binding protein) were quantified using real-time PCR analysis on a LightCycler detection system (Roche Applied Science). First-strand cDNA was synthetized from 2 mug of total RNA using random hexamers and expand reverse transcriptase according to the manufacturer's instructions (Roche Applied Science), subsequently diluted 1 : 10 with water, and stored at -20°C until use. TP53INP1 and TBP PCR products were detected using the Quantitect Probe PCR kit (QIAGEN Operon) following the manufacturer's instructions using the dual-fluorescent Taqman probes (QIAGEN Operon) 5'-FAM-ACCGGCATCTCTTGAGTGCTTGGC-TAMRA-3' (position 759) and 5'-FAM-TCCCAAGCGGTTTGCTGCGGTA-TAMRA-3' (position 811), respectively. The following primers were used: TBP forward 5'-GCCCGAAACGCCGAATATA-3' (position 791); TBP reverse 5'-CGTGGCTCTCTTATCCTCATGA-3' (position 855); TP53INP1 forward 5'-GCACCCTTCAGTCTTTTCCTGTT-3' (position 718) and TP53INP1 reverse 5'-GGAGAAAGCAGGAATCACTTGTATC-3' (position 886). The quantitative real-time PCR was performed in a total volume of 20 mul containing 1 times amplification buffer and 5 mul cDNA template. Samples were heated for 15 min at 95°C and amplified for 45 cycles (denaturation at 95°C for 0 s, annealing and elongation at 60°C for 60 s with a transition rate of 20°C/s). All samples were analysed in duplicate. Data evaluation was performed using the LightCycler data analysis software (version 3.5).

Antibodies and Western blot analyses

Cells stressed with cisplatin, transiently transfected or infected as described above with indicated expression vectors were lysed in lysis buffer (10 mM Hepes (pH 7.9), 10 mM KCL, 400 mM NaCl, 1 mM EDTA, 1 mM EGTA, 1 mM DTT, 0.5% NP40, 1 mM PMSF, 100 mM NaF, 2 mM Na3VO4, 1 mM PMSF, 10 mug/ml leupeptin and 10 mug/ml aprotinin). Protein from frozen mouse pancreas was obtained after tissue homogenization in lysis buffer (50 mM Tris-HCl (pH 8.0), 150 mM NaCl, 1% NP40, 5 mM DTT, 3 mM PMSF, 10 mug/ml leupeptine, 5 mM benzamidine and a cocktail of protease inhibitors from Sigma-Aldrich). An equal amount of total cellular protein per sample was run on a SDS–PAGE and transferred to a nitrocellulose membrane (Hybond ECL from Amersham Pharmacia Biotech). After blockage and first antibody incubation, filters were washed in Tris-buffered saline/Tween 0.1% and then incubated with horseradish peroxidase-conjugated IgG. Antibodies were detected with the ECL chemiluminescence reaction kit (Amersham Pharmacia Biotech) in accordance to the manufacturer's instructions. Antibodies for immunoblotting experiments included: rabbit polyclonal anti-p53 (FL-393 from Santa Cruz Biotechnology Inc.), rat monoclonal anti-TP53INP1 (E-12) and rabbit polyclonal anti-PAP (L8) (the two later were generated in our laboratory), mouse monoclonal anti-p73 (IMG-259 from Imgenex), mouse monoclonal anti-HA (12CA5 from Roche Applied Science), rabbit polyclonal anti-p21WAF1 (sc-397 from Santa Cruz Biotechnology Inc.), rabbit polyclonal anti-beta-actin (ac-15 from Sigma-Aldrich) and mouse monoclonal anti-beta-tubuline (TUB-2.1 from Sigma-Aldrich).

Trans-activation assays

p53-null MEFs and H1299 cells were transiently transfected with various constructs. In CAT activity assays, cell extracts were prepared using the reporter lysis buffer (Promega). 24 h after transfection. CAT activity was determined using the phase extraction procedure. In luciferase assays, cellular extracts were analysed for luciferase activity 24 h after transfection. All assays were carried out for three independent transfections and normalized to beta-galactosidase as previously described (Tomasini et al., 2003). In all transfections the total mass of plasmid DNA was maintained constant by addition of pCDNA4 empty vector.

FACS analyses

p53-null MEFs, H358, H1299, Panc-1 and SW480 cells were seeded at 300 000 cells/100-mm plate and transiently transfected using the indicated expression constructs, as described above. The following day, all cells were harvested and fixed for at least 1 h at 4°C in PBS/70% ethanol. To analyse the cell cycle distribution, fixed cells were resuspended in propidium iodide solution, PBS containing: 100 mug/ml RNase A and 50 mug/ml of propidium iodide (Sigma-Aldrich) and immediately analysed. Flow cytometry was performed using a FACSCalibur (Becton Dickinson Biosciences). Expression of the GFP fluorescence was detected by the FL1 bandpass filter. Data were collected and analysed with the CellQuest™ Pro software (BD Biosciences).

Colony assay

H1299 cells (300 000 cells) were plated into 10-cm Petri dishes and transfected 24 h later with the indicated expression vectors as described above. Transfected cells were selected in Zeocin (0.6 mg/ml) for 10 days. Then, the medium was removed and colonies were washed once with PBS, fixed in 75% methanol and 25% acetic acid for 5 min, and plates were dried. Colonies were stained with Lillie's crystal violet (2 g of crystal violet and 0.8 g of ammonium oxalate in 80% ethanol) for 5 min and subsequently washed with distilled water to remove excess stain. Stained colonies comprising more than 10 cells were scored and counted under inverted microscope. Three independent experiments were carried out and each count was duplicated.



  1. Barak Y, Juven T, Haffner R and Oren M. (1993). EMBO J., 12, 461–468. | PubMed | ISI | ChemPort |
  2. Chau BN, Chen TT, Wan YY, DeGregori J and Wang JY. (2004). Mol. Cell. Biol., 24, 4438–4447. | Article | PubMed | ISI | ChemPort |
  3. Chirgwin JM, Przybyla AE, MacDonald RJ and Rutter WJ. (1979). Biochemistry, 18, 5294–5299. | Article | PubMed | ISI | ChemPort |
  4. Courtois S, de Fromentel CC and Hainaut P. (2004). Oncogene, 23, 631–638. | Article | PubMed | ISI | ChemPort |
  5. Das S, El-Deiry WS and Somasundaram K. (2003a). Oncogene, 22, 8394–8402. | Article | PubMed | ISI | ChemPort |
  6. Das S, El-Deiry WS and Somasundaram K. (2003b). J. Biol. Chem., 278, 18313–18320. | Article | PubMed | ISI | ChemPort |
  7. Di Como CJ, Gaiddon C and Prives C. (1999). Mol. Cell. Biol., 19, 1438–1449. | PubMed | ISI | ChemPort |
  8. Dusetti NJ, Tomasini R, Azizi A, Barthet M, Vaccaro MI, Fiedler F, Dagorn JC and Iovanna JL. (2000). Biochem. Biophys. Res. Commun., 277, 660–667. | Article | PubMed | ISI | ChemPort |
  9. el-Deiry WS, Tokino T, Velculescu VE, Levy DB, Parsons R, Trent JM, Lin D, Mercer WE, Kinzler K and Vogelstein B. (1993). Cell, 75, 817–825. | Article | PubMed | ChemPort |
  10. Flores ER, Tsai KY, Crowley D, Sengupta S, Yang A, McKeon F and Jacks T. (2002). Nature, 416, 560–564. | Article | PubMed | ISI | ChemPort |
  11. Fontemaggi G, Gurtner A, Strano S, Higashi Y, Sacchi A, Piaggio G and Blandino G. (2001). Mol. Cell. Biol., 21, 8461–8470. | Article | PubMed | ISI | ChemPort |
  12. Fontemaggi G, Kela I, Amariglio N, Rechavi G, Krishnamurthy J, Strano S, Sacchi A, Givol D and Blandino G. (2002). J. Biol. Chem., 277, 43359–43368. | Article | PubMed | ISI | ChemPort |
  13. Gong JG, Costanzo A, Yang HQ, Melino G, Kaelin Jr WG, Levrero M and Wang JY. (1999). Nature, 399, 806–809. | Article | PubMed | ISI | ChemPort |
  14. Hamer G, Gademan IS, Kal HB and de Rooij DG. (2001). Oncogene, 20, 4298–4304. | Article | PubMed | ISI | ChemPort |
  15. Ikawa S, Nakagawara A and Ikawa Y. (1999). Cell Death Differ., 6, 1154–1161. | Article | PubMed | ISI | ChemPort |
  16. Iovanna J, Orelle B, Keim V and Dagorn JC. (1991). J. Biol. Chem., 266, 24664–24669. | PubMed | ISI | ChemPort |
  17. Irwin MS and Kaelin WG. (2001). Cell Growth Differ., 12, 337–349. | PubMed | ISI | ChemPort |
  18. Ishida S, Yamashita T, Nakaya U and Tokino T. (2000). Jpn. J. Cancer Res., 91, 174–180. | PubMed | ChemPort |
  19. Jost CA, Marin MC and Kaelin Jr WG. (1997). Nature, 389, 191–194. | Article | PubMed | ISI | ChemPort |
  20. Kaghad M, Bonnet H, Yang A, Creancier L, Biscan JC, Valent A, Minty A, Chalon P, Lelias JM, Dumont X, Ferrara P, McKeon F and Caput D. (1997). Cell, 90, 809–819. | Article | PubMed | ISI | ChemPort |
  21. Miyashita T and Reed JC. (1995). Cell, 80, 293–299. | Article | PubMed | ISI | ChemPort |
  22. Niederau C, Ferrell LD and Grendell JH. (1985). Gastroenterology, 88, 1192–1204. | PubMed | ISI | ChemPort |
  23. Obad S, Brunnstrom H, Vallon-Christersson J, Borg A, Drott K and Gullberg U. (2004). Oncogene, 23, 4050–4059. | Article | PubMed | ISI | ChemPort |
  24. Okamura S, Arakawa H, Tanaka T, Nakanishi H, Ng CC, Taya Y, Monden M and Nakamura Y. (2001). Mol. Cell., 8, 85–94. | Article | PubMed | ISI | ChemPort |
  25. Oniscu A, Sphyris N, Morris RG, Bader S and Harrison DJ. (2004). J. Clin. Pathol., 57, 492–498. | Article | PubMed | ISI | ChemPort |
  26. Rodicker F and Putzer BM. (2003). Cancer Res., 63, 2737–2741. | PubMed | ISI |
  27. Sablina AA, Chumakov PM and Kopnin BP. (2003). J. Biol. Chem., 278, 27362–27371. | Article | PubMed | ISI | ChemPort |
  28. Shimodaira H, Yoshioka-Yamashita A, Kolodner RD and Wang JY. (2003). Proc. Natl. Acad. Sci. USA, 100, 2420–2425. | Article | PubMed | ChemPort |
  29. Shinbo J, Ozaki T, Nakagawa T, Watanabe K, Nakamura Y, Yamazaki M, Moriya H, Nakagawara A and Sakiyama S. (2002). Biochem. Biophys. Res. Commun., 295, 501–507. | Article | PubMed | ISI | ChemPort |
  30. Stiewe T and Pützer B. (2001). Apoptosis, 6, 447–452. | Article | PubMed | ISI | ChemPort |
  31. Strano S, Rossi M, Fontemaggi G, Munarriz E, Soddu S, Sacchi A and Blandino G. (2001). FEBS Lett., 490, 163–170. | Article | PubMed | ISI | ChemPort |
  32. Strano S, Munarriz E, Rossi M, Cristofanelli B, Shaul Y, Castagnoli L, Levine AJ, Sacchi A, Cesareni G, Oren M and Blandino G. (2000). J. Biol. Chem., 275, 29503–29512. | Article | PubMed | ISI | ChemPort |
  33. Tomasini R, Samir AA, Carrier A, Isnardon D, Cecchinelli B, Soddu S, Malissen B, Dagorn JC, Iovanna JL and Dusetti NJ. (2003). J. Biol. Chem., 278, 37722–37729. | Article | PubMed | ISI | ChemPort |
  34. Tomasini R, Samir AA, Pebusque MJ, Calvo EL, Totaro S, Dagorn JC, Dusetti NJ and Iovanna JL. (2002). Eur. J. Cell Biol., 81, 294–301. | Article | PubMed | ISI | ChemPort |
  35. Tomasini R, Samir AA, Vaccaro MI, Pebusque MJ, Dagorn JC, Iovanna JL and Dusetti NJ. (2001). J. Biol. Chem., 276, 44185–44192. | Article | PubMed | ChemPort |
  36. Torigoe T, Izumi H, Ishiguchi H, Yoshida Y, Tanabe M, Yoshida T, Igarashi T, Niina I, Wakasugi T, Imaizumi T, Momii Y, Kuwano M and Kohno K. (2005). Curr. Med. Chem. Anti-Canccer Agents, 5, 15–27. | Article | ChemPort |
  37. Tsuji K, Mizumoto K, Yamochi T, Nishimoto I and Matsuoka M. (2002). J. Biol. Chem., 277, 2951–2957. | Article | PubMed | ISI | ChemPort |
  38. Tullo A, Mastropasqua G, Bourdon JC, Centonce P, Gostissa M, Costanzo A, Levrero M, Del Sal G, Saccone C and Sbisà E. (2003). Oncogene, 22, 8738–8748. | Article | PubMed | ISI | ChemPort |
  39. Yang A, Kaghad M, Caput D and McKeon F. (2002). Trends Genet., 18, 90–95. | Article | PubMed | ISI | ChemPort |
  40. Yang A, Walker N, Bronson R, Kaghad M, Oosterwegel M, Bonnin J, Vagner C, Bonnet H, Dikkes P, Sharpe A, McKeon F and Caput D. (2000). Nature, 404, 99–103. | Article | PubMed | ISI | ChemPort |


We thank Drs S Vasseur and P Soubeyran for helpful comments, V Fontaine for her help in Western blots and monoclonal antibody characterizations, A Giovannetti for siRNA and FACS analysis, R Grimaud and P Berthézène for technical help. We are grateful to G Blandino and S Das for providing us with expression plasmids (pcDNA-p73alpha-HA and pcDNA-p53R175H) and adenovirus (Ad-p73beta), respectively. This work was supported by grants from INSERM R Tomasini is supported by a fellowship from the Ministère de la Recherche et de la Technologie, and J Nowak is supported by a fellowship from the Région Provence-Alpes-Côte-d'Azur and INSERM.