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

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in ISI Web of Science 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 ISI Web of Science (6) Citing Articles via Google Scholar Google Scholar Articles by Malaguarnera, R. Articles by Frasca, F. Search for Related Content PubMed PubMed Citation Articles by Malaguarnera, R. Articles by Frasca, F.

Dipartimento di Medicina Interna e di Medicina Specialistica — Endocrinologia, Università di Catania, 95123 Catania, Italy
¹ Centro di Biomorfologia, Università di Messina, 98122 Messina, Italy
² Servizio di Anatomia Patologica, Ospedale Vittorio Emanuele, 95124 Catania, Italy
³ Malattie Apparato Respiratorio, Università di Catania, 95123 Catania, Italy
⁴ Dipartimento di Scienze Biomediche, Università di Catania, 95124 Catania, Italy

(Requests for offprints should be addressed to F Frasca; Email: f.frasca{at}unict.it)

	Abstract

Top Abstract Introduction Materials and methods Results Discussion References

 
Inactivation of p53 and p73 is known to promote thyroid cancer progression. We now describe p63 expression and function in human thyroid cancer. TAp63 is expressed in most thyroid cancer specimens and cell lines, but not in normal thyrocytes. However, in thyroid cancer cells TAp63 fails to induce the target genes (p21Cip1, Bax, MDM2) and, as a consequence, cell cycle arrest and apoptosis occur. Moreover, TAp63 antagonizes the effect of p53 on target genes, cell viability and foci formation, and p63 gene silencing by small interfering (si) RNA results in improved p53 activity. This unusual effect of TAp63 depends on the protein C-terminus, since TAp63ß and TAp63 isoforms, which have a different arrangement of their C-terminus, are still able to induce the target genes and to exert tumour-restraining effects in thyroid cancer cells. Our data outline the existence of a complex network among p53 family members, where TAp63 may promote thyroid tumour progression by inactivating the tumour suppressor activity of p53.

	Introduction

Top Abstract Introduction Materials and methods Results Discussion References

 
Thyroid cancer is a common endocrine malignancy and several genetic abnormalities have been identified in the different thyroid cancer histotypes, involving both oncogenes and tumour suppressor genes. Mutations in proto-oncogenes (Ret, BRAF, Ras) are often observed in well-differentiated thyroid tumours (papillary and follicular) (Fagin 2002). Indeed, Ret/PTC rearrangements and BRAF mutations are observed in approximately 10–50% of papillary thyroid cancer, whereas Ras mutations are observed in approximately 20–50% of follicular thyroid cancer (Gimm 2001).

Tumour suppressor gene abnormalities, responsible for thyroid tumour progression, involve PTEN, ß-catenin and p53. Decreased PTEN expression, presumably by loss of heterozygosity, has been observed in papillary (10%), follicular (15%) and poorly differentiated thyroid carcinomas (60%), whereas ß-catenin mutations have been found in approximately 60% of anaplastic thyroid carcinomas (Gimm 2001). More specifically, p53 mutations are found in more than 80% of the poorly differentiated (anaplastic) thyroid carcinomas (Fagin et al. 1993). As a consequence, loss of p53 function is believed to play an important role in thyroid tumour progression from well (papillary and follicular) to poorly (anaplastic) differentiated thyroid cancer (Fagin et al. 1993). Moreover, several reports have shown that, even in the absence of inactivating mutations, the p53 protein is inactive in certain thyroid tumours and cell lines (Wyllie et al. 1995, Nishida et al. 1996), suggesting that other mechanisms may be responsible for p53 inactivation in these tumours. Finally, up-regulation of non-mutated p53 protein has been related to a poor clinical outcome in thyroid cancer (Dobashi et al. 1993, Nishida et al. 1996, Ruter et al. 1996).

Two novel members have been added to the p53 family: p63 (Yang et al. 1998) and p73 (Kaghad et al. 1997). These proteins have remarkable similarities in both structure and function to p53, since they can transactivate p53-responsive genes including p21Cip1, Bax and MDM2, and induce cell cycle arrest and apoptosis (Jost et al. 1997, Kaghad et al. 1997, Yang et al. 1998).

In addition to the transcriptionally active (full length) TAp63 and TAp73 isoforms, p63 and p73 genes, by the use of an inner promoter located in intron 3, may generate the Np63 and Np73 variants, which are N-terminally truncated and exert a dominant negative effect towards p53, TAp63 and TAp73 (Yang et al. 1998, 2000, Pozniak et al. 2000). Furthermore, p63 and p73 may undergo multiple C-terminal splicing, generating at least six isoforms for p73 (, ß, , , , ) (Kaghad et al. 1997, De Laurenzi et al. 1998, Zaika et al. 1999), and three isoforms for p63 (, ß, ) (Yang et al. 1998). At variance with p53 null mice, however (Donehower et al. 1992), p63 and p73 knockout mice do not develop spontaneous tumours (Mills et al. 1999, Yang et al. 2000), suggesting that p63 and p73 function is not strictly related to tumour suppressor activity. Moreover, attempts to identify mutations in the p63 and p73 gene in human cancers have been largely unsuccessful. More interestingly, several cancers overexpress the dominant negative isoforms Np63 and Np73 (Zaika et al. 1999, 2002) and data obtained from fibroblasts in vitro suggest that Np63 and Np73 may display an oncogenic potential (Hibi et al. 2000, Petrenko et al. 2003).

We have shown that thyroid cancer cells express Np73 and TAp73 (Frasca et al. 2003). In these cells, TAp73 tumour suppressor activity is kept latent by several mechanisms including the cytoplasmic entrapment of c-Abl (Vella et al. 2003), interaction with p53 mutants and Np73 (Frasca et al. 2003).

In the present study, we explored the role of p63 in thyroid tumours and found that TAp63 is expressed in most thyroid cancers but not in the normal thyroid or in follicular adenomas, and it may represent, therefore, a marker of malignancy. In thyroid cancer cells, TAp63 does not elicit p53-like responses. In contrast, TAp63 exerts an unexpected inhibitory effect on the tumour suppressor activity of p53. The absence of TAp63 tumour suppressor activity and the presence of an anti-p53 effect suggest a role for p63 in thyroid tumour progression.

	Materials and methods

Top Abstract Introduction Materials and methods Results Discussion References

 
Cells

The human thyroid cancer cell lines (see Table 2) BC-PAP (papillary) and FRO (follicular) were provided by Drs A Fusco and M Santoro (Naples, Italy); SW-1736 (anaplastic), Hth-74 (anaplastic) and C-643 (anaplastic) cells were provided by Dr N E Heldin (Uppsala, Sweden); FF-1 (anaplastic) and AM-1 (anaplastic) cells were established in our laboratory; 8505-C (papillary) cells were purchased from DMSZ (Braunschweig, Germany); FTC-133 (follicular) and 8305-C (anaplastic) cells were purchased from ECACC (Salisbury, Wilts, UK); C-98 cells, a clone harbouring a mutation in the p53 gene, were established from TPC-1 (a papillary thyroid cancer cell line provided by Dr A Fusco, Naples, Italy). All thyroid cancer cell lines (see Table 2) were grown in RPMI 1640 (Sigma, St Louis, MO, USA) supplemented with 2 mM glutamine, 10% FBS and 100 µg penicillin and streptomycin/ml. Normal thyroid cells in primary culture were obtained from surgical specimens after treatment with 1 mg collagenase IV/ml (Sigma). The human osteosarcoma cell line, Saos-2, and the simian kidney cell line, COS-1, were provided by Dr J Y Wang (La Jolla, CA, USA) and cultured in DMEM (Sigma) supplemented with 10% FBS, and 100 µg penicillin and streptomycin/ml. The human breast cancer cell line, MCF-7 (ATCC, Manassas, VA, USA), and the human oesophagus carcinoma cell line, A431 (Dr Weir, Boston, MA, USA), were grown in MEM supplemented with FBS and antibiotics as described above.

Human thyroid cancer specimens were obtained at surgery and stored in liquid nitrogen until processing.

Immunohistochemistry

In order to set up this technique, p63-positive (FTC-133 and C-643) and p63-negative (C-98) (see Fig. 2) thyroid cancer cells were grown in monolayers, harvested by trypsinization, and centrifuged at 270 g for 10 min at 4°C. As a positive control, we used the A431 oesophagus cancer cell line, which expresses p63 at a higher level than thyroid cancer cells (see Fig. 2B). Cell pellets were immediately frozen in liquid nitrogen or, alternatively, fixed with paraformaldehyde and paraffin embedded. From these pellets were obtained 7 µm-thick sections of both fixed and unfixed cells, which were subjected to immunohistochemical staining for p63 and p53. In these experiments in A431 cells, p63 was detected in both frozen and paraffin-embedded sections. In contrast, in p63-positive thyroid cancer cells, p63 immunostaining was observed only in frozen sections, but not in paraffin-embedded sections. Thyroid tissue sections were cut with a cryostat at –30°C, fixed with acetone at –20°C for 10 min and hydrated with PBS at room temperature for 45 min. After blocking in 2% normal serum for 20 min, sections were incubated overnight with the anti-pan-p63 monoclonal antibody 4A4 (1 : 100) (Santa Cruz Biotechnology, Santa Cruz, CA, USA), the anti-p63 goat polyclonal antibody C-18 (1 : 100) (Santa Cruz Biotechnology) or the anti-p53 monoclonal antibody DO-1 against the N-terminus of p53 (1 : 200) (Santa Cruz Biotechnology). Specific labelling was detected with biotin-conjugated anti-mouse/anti-rabbit/anti-goat IgG and avidin–biotin peroxidase complex. Sections were counterstained with either haematoxylin QS or Nuclear Fast Red (NFR), examined and photographed using an Olympus BH-2 microscope. In every experiment, sections were incubated with secondary antibody alone to further verify the specificity of the reaction.

View larger version (54K): [in this window] [in a new window]   Figure 2 Expression and localization of p63 in human thyroid cancer cells. (A) Three primary cultures of normal thyrocytes, 3 papillary (C-98, BC-PAP, 8505C), 2 follicular (FRO, FTC-133), and 6 anaplastic (FF1, SW-1736, C-643, Hth-74, 8305C, AM-1) thyroid cancer cell lines were screened by RT-PCR for p63 isoform expression. The A431 cell line was used as a positive control. (B) Lysates from the same cell lines were subjected to immunoprecipitation (IP) with anti-pan-p63 polyclonal antibody (H137) and then blotted (Blot) with an anti-pan-p63 monoclonal antibody (4A4). COS-1 cells, transiently transfected with TAp63, TAp63 or Np63, were used as a positive control. (C) Cellular localization of p63 by immunofluorescence. Primary thyrocytes, C-98 (papillary, p63 negative), FTC-133 (follicular, p63 positive) and C-643 (anaplastic, p63 positive) thyroid cancer cell lines were plated onto cover slips, fixed and stained for p63 (4A4 antibody, red) and filamentous actin (phalloidin, green). Nuclei were visualized with Hoechst (blue). A431 cells were used as a positive control.

Cells were fixed in 3.7% formaldehyde, permeabilized with PBS/0.3% Triton X-100, blocked with PBS/10% normal goat serum and incubated with primary antibodies for 1 h. To detect endogenous p63 we used the anti-pan-p63 monoclonal antibody 4A4 (Santa Cruz Biotechnology). To detect transfected p63 we used the anti-MycTag monoclonal antibody 9E10 (Santa Cruz Biotechnology). Cells were then incubated with Alexa-conjugated (Alexa Fluor 594 or 488) secondary antibodies (Molecular Probes, Leiden, The Netherlands) for 1 h. To visualize the cytoplasm, the cells were also incubated with Alexa-conjugated phalloidin (Molecular Probes) for an additional 30 min. The cells were finally counterstained with Hoechst (Sigma) to colour the nuclei. Epifluorescence microscopy was performed with an Olympus microscope. The images were digitally acquired with an Orca CCD camera (Hamamatsu, Hamamatsu City, Japan) and processed with Image-Pro Plus 4.0 software (Media Cybernetics, Silver Spring, MD, USA).

Immunoprecipitation and immunoblot analysis

Cell lysates were prepared in RIPA buffer containing 0.1% SDS and protease inhibitor cocktail (Roche Biochemical Inc., Basel, Switzerland). For immuno-precipitation experiments, 1 mg cell lysate was incubated for 2 h with 2 µg antibody. After incubation with protein A-Sepharose (Amersham Biosciences, Uppsala, Sweden), samples were resuspended in loading buffer, separated by SDS-PAGE and transferred to PVDF membranes (Millipore, Bedford, MA, USA). The membranes were blocked with 5% milk–Tris-buffered saline plus Tween (TBST) and then immunoblotted with primary antibodies (1 µg/ml). Appropriate horseradish-peroxidase-conjugated secondary antibodies were added at 1 : 2000 (Amersham Biosciences), and proteins were visualized by enhanced chemiluminescence (Pierce, Rockford, IL, USA).

The following antibodies were used for immuno-precipitation: polyclonal anti-p63 antibody H129, polyclonal anti-pan-p63 H137 and monoclonal antibody DO-1 against the N-terminus of p53 (all from Santa Cruz Biotechnology). The following antibodies were used for western blotting: anti-pan-p63 monoclonal antibody 4A4 (Santa Cruz Biotechnology), anti-p53 monoclonal antibody DO-1 (Santa Cruz Biotechnology), anti-p21Cip1 polyclonal antibody (Santa Cruz Biotechnology), anti-ß-actin monoclonal antibody (Sigma), anti-GFP monoclonal antibody (Covance Research Product Inc. Princeton, New Jersey, USA) and anti-Myc monoclonal antibody 9E10 (Santa Cruz Biotechnology).

Transcript analysis by RT-PCR

Total RNA (1 µg) was reverse transcribed with Superscript II (Invitrogen, Paisley, Strathclyde, UK) and oligo(dT) primers. Of the synthesized cDNA 2 µl were then combined in a PCR specific for p63 and ß, using primers 5' CGT ACA GGC AAC AGC AAC A 3' (forward) and 5' CGT TGC GCT GCT GAG GGT TGA 3' (reverse) spanning exons 10 and 14 of the p63 gene (fragment size 632 bp). The TAp63 transcript was detected using primers 5' CCC AGA GCA CAC AGA CAA A 3' (forward) and 5' CAC AGA TCC GGG CCT CAA A 3' (reverse) spanning exons 2 and 8 (fragment size 896 bp). The p63 transcript was detected using primers 5' ATG CCC AGT ATG TAG AAG A 3' (forward) and 5' GGG CTT GGA ATG TCT AAA G 3' (reverse) spanning exons 6 and 15 of the p63 gene (fragment size 697 bp), and the Np63 transcript was detected using primers 5' AAC AAT GCC CAG ACT CAA 3' (forward) and 5' ACA GGC ATG GCG CGG ATA 3' (reverse) spanning intron 3 and exon 5 (fragment size 392 bp).

Plasmids and transfections

pBOS-H2B-GFP and pCDNA3.1-p53 were provided by Dr J Y Wang (La Jolla, CA, USA); pCDNA3.1-p53-GFP was a gift from Drs G Wahl and JM Stommel (La Jolla, CA, USA); pCDNA3.0-Myc-TAp63, pCDNA3.0-Myc-TAp63, pCDNA3.0-Myc-Np63, p21Luc and BaxLuc were donated by Dr G Blandino (Rome, Italy); pCDNA3.0-Myc-TAp63ß was kindly provided by Dr L Guerrini (Milan, Italy).

All transfections were performed in 6-well plates with Fugene 6 (Roche Biochemical Inc., Basel, Switzerland) according to the manufacturer’s instructions (DNA : Fugene ratio 1 : 3), and cells were processed 24 h after transfection.

We tested the onco-suppressor effect of p53 and p63 constructs by evaluating the reduction of the number of transfected cells (apoptosis plus inhibition of cell growth) as previously reported (Ozaki et al. 2003). We transfected p53 and p63 constructs (2 µg/well) together with H2B-GFP (0.2 µg/well). Forty-eight hours after transfection, the GFP-positive cells were scored under a fluorescence microscope and numbers obtained were expressed as a percentage of GFP-positive cells among the total population and were compared with the empty-transfected cells.

Luciferase assay

The p21Luc, BaxLuc and MDM2Luc constructs were co-transfected with pCDNA3.1, pCDNA3.0-Myc-TAp63, pCDNA3.0-Myc-TAp63ß, pCDNA3.0-Myc-TAp63, pCDNA3.0-Myc-Np63 and pCDNA3.1-p53 (DNA ratio 1 : 1). A vector coding for the Renilla luciferase (provided by Dr E Conte, Catania, Italy) was also co-transfected in all conditions (DNA ratio 1 : 20). Twenty-four hours after transfec-tion, the cells were processed with the Dual Luciferase assay (Promega Corp., Madison, WI, USA) according to the manufacturer’s instructions. Luciferase activity was normalized for transfection efficiency (Renilla activity).

Gene silencing by siRNA

Cells were plated onto 6-well plates (10⁵ cells/well) and kept in antibiotic-free medium for 24 h. Cells were then transfected with a mixture containing OptiMEM, 8 µl lipofectamine/well (Lipofectamine 2000, Invitrogen) and either 0.5 µg GFP-small interfering (si) RNA or 0.5 µg TAp63-siRNA/well (Dharmacon Research Inc., Lafayette, CO, USA) for 5 h. The sequence of these siRNAs is available from the manufacturer. Cells were then incubated with fresh medium for 48 h and transfected with p53 (0.5–1.0 µg/well) together with p21-luc (1.0 µg) and Renilla (0.2 µg), using Fugene 6 reagent (Roche). Twenty-four hours after transfection, cell extracts were analysed for p21 activity by the luciferase assay. Aliquots of these samples were also subjected to western blot for the assessment of p63 status.

Cell cycle evaluation

Cells were synchronized for 36 h in serum/leucine-free medium and released in complete medium for 12 h (cell cycle) or 72 h (apoptosis) in the presence or absence of 2 µM doxorubicin. Adherent and floating cells were harvested and resuspended in 70% ethanol and stored at –20°C. Permeabilized cells were centrifuged and resuspended in PBS containing 20 µg Propidium Iodine (PI)/ml plus 40 µg RNAse/ml (Sigma) for 30 min in the dark. Cells were then subjected to FACS analysis (Coulter Elite flow cytometer, Beckman Coulter, Milan, Italy) and gated for PI (X axis = FL2; Y axis = events). To evaluate the subG1 population, a log scale was applied to the X axis (PI, FL2). Cells transfected with GFP-tagged constructs were not treated with 70% ethanol so as to avoid GFP denaturation and, as a consequence, loss of fluorescent emission. GFP-containing cells were fixed instead with 1% paraformaldehyde in PBS for 2 h at 4°C and permeabilized with 0.3% Triton in PBS for 20 min at room temperature or, alternatively, treated with citrate hypotonic buffer. Triton-permeabilized cells were incubated with PI and RNAse overnight at 4°C, whereas citrate-permeabilized cells were incubated with PI and RNAse for 30 min at room temperature in the dark. Cells were than subjected to FACS analysis and gated for PI (FL2) and GFP (FL1). Cell cycle analysis was performed by placing separate gates in both the transfected (GFP positive), and untransfected (GFP negative) population. The percentage of GFP-positive cells ranged from 2 to 20% of total cells. For GFP-positive cells at least 10⁴ events were counted.

Foci formation

C-98, BC-PAP, SW-1736 and C-643 thyroid cancer cells were plated onto 6-well plates (10⁵ cells/well) in complete 10% FCS medium. After 24 h, each well was transfected using the Fugene 6 method (Roche) with either 2 µg empty vector, or a mixture containing 1 µg empty + 1 µg p53, 1 µg empty + 1 µg TAp63, 1 µg empty + 1 µg Np63, 1 µg p53 + 1 µg TAp63, or 1 µg p53 + 1 µg Np63. Cells were allowed to grow until 90% confluent and were split onto 100 mm Petri dishes. Cells were then grown in complete medium containing G418 (Gibco) (concentration range 0.5–1.0 mg/ml, depending on the cell line) to allow antibiotic (geneticin) selection. The foci obtained with this procedure were fixed in 11% glutaraldehyde, stained with Crystal Violet (BDH, Poole, Dorset, UK) and counted.

Statistical analysis

FACS analysis results were compared by two-way analysis of variance. Significance was obtained by Student’s t-test (*P<0.05, **P<0.01, ***P<0.001). Statistical analysis was carried out with Microsoft Excel software.

	Results

Top Abstract Introduction Materials and methods Results Discussion References

 
TAp63 is expressed in human thyroid cancer tissue

Previous studies performed in paraffin-embedded specimens indicated that p63 is expressed in a small subset of papillary and anaplastic thyroid carcinomas (Preto et al. 2002). We examined 23 frozen thyroid specimens by immunohistochemistry (IHC) (Fig. 1A, Table 1) using the anti-pan-p63 4A4 antibody. In these experiments, p63 immunoreactivity was present in most thyroid cancers (9/9 papillary carcinomas, 7/7 follicular carcinomas, 9/11 anaplastic carcinomas) (Table 1). In contrast, p63 was not detected in normal thyroid (n = 8), whereas it was detected in 1 out of 7 benign adenomas (Table 1). p63 was also detected in vessels of both normal and neoplastic thyroid tissue (Fig. 1A, right panel). In thyroid cancer cells, p63 immunoreactivity was mostly localized in the nucleus (Table 1). Immunohistochemistry with a specific anti-p63 antibody gave negative results (not shown) and immunohistochemistry in paraffin-embedded tissues provided a weak signal only in a small percentage of thyroid specimens, in accordance with the previous observations (Preto et al. 2002). Since the 4A4 antibody used in these experiments recognizes all p63 isoforms, we studied p63 isoform expression in frozen thyroid tissue specimens by RT-PCR (Fig. 1B). As a negative control we used C-98 thyroid cancer cells (p63 negative, see Table 2), while as a positive control we used the oesophagus cancer cell line A431 (Kaelin 1999). In thyroid cancer specimens, we detected the TAp63 transcript but found no mRNA for TAp63ß, TAp63 and N-p63 (Fig. 1B). TAp63 transcript was also found in normal thyroid tissues; its expression, in accordance with the IHC results, must be attributed to the presence of p63 immunoreactivity in blood vessels (Fig. 1A). In the same samples, we then evaluated p63 protein expression by immuno-precipitation with the anti-pan-p63 antibody (H137) and by western blot analysis with the anti-pan-p63 antibody (4A4) (Fig. 1C). As a positive control we used lysates of COS-1 cells transfected with TAp63, TAp63 or Np63 (Fig. 1C). In accordance with the RT-PCR results, we found that in thyroid cancer specimens only the TAp63 protein was expressed (Fig. 1C). In contrast to malignant tissue, TAp63 protein was not detected in normal thyroid tissues (Fig. 1C, left panel). Taken together, these data indicate that the TAp63 protein is expressed in most thyroid cancer cells but not in normal thyroid or in follicular adenoma cells. The apparent contrast of the TAp63 transcript presence in normal thyroid is due to the TAp63 expression in endothelial cells. It is also interesting to note that the dominant negative isoform, Np63, which has been reported to be upregulated in several p63-positive malignancies, is not expressed in thyroid tumours.

View larger version (59K): [in this window] [in a new window]   Figure 1 p63 expression in human thyroid cancer specimens. (A) Immunohistochemistry for p63 in human thyroid samples was performed on frozen thyroid specimens using anti-pan-p63 antibody (4A4) (see also Table 1 and the Methods section). Eosin staining (H/E; top panels) and p63 staining (bottom panels) are shown for normal thyroid, follicular adenoma, papillary carcinoma, follicular carcinoma and anaplastic carcinoma at x20 magnification. A representative p63 immunostaining obtained in blood vessels of normal thyroid tissue is shown (right). (B) Specimens obtained from three normal thyroids, two papillary, two follicular and three anapalstic thyroid carcinomas were analysed by RT-PCR for the presence of TAp63, p63/ß, TAp63 and Np63 transcripts. The human oesophagus cancer cell line A431 (p63 positive) and the thyroid cancer cell line C-98 (p63 negative) were used as controls. RT-PCR for the ubiquitous gene Ele-1 was also performed (lower panel). (C) Lysates from the same specimens were immunoprecipitated (IP) with an anti-pan-p63 polyclonal antibody (H137) and subjected to Western blot analysis (Blot) with anti-pan-p63 monoclonal antibody (4A4). C-98 cells were used as a negative control; COS-1 cells, transfected with TAp63, TAp63 or Np63, and A431 cells were used as positive controls.

TAp63 is expressed in thyroid cancer cell lines

To identify an in vitro model to study the role of TAp63 in thyroid cancer biology, we explored by RT-PCR the expression of p63 in a panel of thyroid cancer cell lines, representative of the three thyroid cancer histotypes (3 papillary, 2 follicular and 6 anaplastic) (Fig. 2A). Normal thyrocytes in primary culture were also studied (Fig. 2A, B, C, on the left). The TAp63 transcript was present in 2 out of 3 papillary thyroid cancer cell lines (BC-PAP and 8505C), 1 out of 2 follicular thyroid cancer cell lines (FTC-133) and all (6/6) anaplastic thyroid cancer cell lines (FF-1, SW-1736, C-643, Hth-74, 8305C and AM-1) (Fig. 2A). Moreover, 3 cell lines (FTC-133, SW-1736 and C-643) were also faintly positive for the TAp63 transcript (Fig. 2A). In contrast, the N-p63 transcript was not found in the thyroid cancer cell lines tested (Fig. 2A). In contrast, normal thyrocytes were negative for p63 transcript (Fig. 2A, on the left).

In the same cell lines, we then evaluated p63 protein expression by western blot (Fig. 2B) and found that the TAp63 protein was present in BC-PAP, 8505C, FTC-133, C643 and Hth-74 cells (Fig. 2B). In contrast, in FF-1, SW-1736, 8305C and AM-1 cells, which were positive for TAp63 mRNA, the protein was not detected (Fig. 2B). C-98 and FRO cells expressed neither TAp63 mRNA nor the protein (Fig. 2A, B). Data obtained are summarized in Table 2. The TAp63 protein was not detected in any of the cell lines expressing the mRNA. A high expression of all three p63 protein isoforms (TAp63, Np63 and TAp63) was observed in A431 cells, used as a positive control (Fig. 2B on the left). In primary normal thyrocytes, p63 was always absent at western blot analysis (Fig. 1B, on the left). However, when AM-1 and 8305-C cells, which were positive for the TAp63 transcript and negative for the protein (Fig. 2 and Table 2), were incubated with the proteasomal inhibitor MG132, a faint band corresponding to TAp63 protein was observed (not shown). These findings suggest that in these cells the TAp63 protein is present, but its expression is decreased below detectable levels by proteasomal degradation.

In these thyroid cancer cells, we then evaluated p63 localization by immunofluorescence using the 4A4 antibody (Fig. 2C). In C-98 cells (p63 negative, see Fig. 2A, B and Table 2) we did not detect any p63 immunoreactivity (Fig. 2C), whereas in FTC-133 and C-643 cells (p63 positive, see Fig. 2A, B and Table 2), the p63 signal was localized in the nucleus, in a manner similar to that observed in the p63-positive cells A431 (Fig. 2C). In accordance with the results obtained by immunohistochemistry, normal thyrocytes in primary culture did not display any p63 immunoreactivity (Fig. 2C, on the left).

These data indicate that TAp63 is present in most thyroid cancer cells in permanent culture, but not in normal thyrocytes, and confirm the data obtained in thyroid tissue specimens.

Endogenous p63 does not exert p53-like functions in thyroid cancer

p63 is able to activate a pool of genes, which are also common targets of p53 and p73 (Sasaki et al. 2001), including p21Cip1, Bax and MDM2 (Zhu et al. 1998, Lee & La Thangue 1999, Nakano et al. 2000). Doxorubicin is an effective DNA damaging agent, which leads to the activation of p53 and p73 and, as a consequence, to the transactivation of target genes. Hence, to explore the onco-suppressor activity of TAp63, we exposed thyroid cancer cells to doxorubicin. To this end, we used cell lines with a genetic background suitable for our experiments, i.e. we selected FTC-133 and C-643 cells, which express TAp63 and p53 inactive mutants but not p73 (Fig. 2 and Table 2). Hence, in these cells doxorubicin is expected to induce p21Cip1/Bax expression exclusively via TAp63. As a positive control we employed the human breast cancer cell line MCF-7, which has a wild-type p53 and maintains a normal response to doxorubicin. As a negative control we used FF-1 thyroid cancer cells, harbouring mutated p53 (Frasca et al. 2003) and lacking both p73 (Frasca et al. 2003) and p63 proteins (Fig. 2 and Table 2).

In MCF-7 cells, exposure to doxorubicin increased the expression level of p53 and, as a consequence, led to the induction of the target gene p21 (Fig. 3A). In contrast, doxorubicin neither significantly affected the TAp63 level in C-643 and FTC-133 cells (Fig. 3A), nor caused the appearance of p63 protein in FF-1 cells (Fig. 3A) and C-98 cells (not shown). As a consequence, doxorubicin failed to increase p21Cip1 content in FF-1, C643, C-98 and FTC-133 thyroid cancer cell lines (Fig. 3A). In accord with the failure of p21Cip1 induction (Fig. 3A), doxorubicin failed to cause a significant G1 arrest in thyroid cancer cells (Fig. 3B), independently of p63 status, whereas it did do so in MCF-7 cells, as expected. Moreover, doxorubicin treatment failed to induce the proapoptotic gene Bax and apoptosis in thyroid cancer cells (data not shown). In accordance with the abrogation of p53 function (Blagosklonny 2002), exposure to doxorubicin for 72 h induced a G2/M arrest in all four thyroid cancer cell lines (not shown), indicating that doxorubicin was used at an effective concentration.

View larger version (50K): [in this window] [in a new window]   Figure 3 Activity of endogenous TAp63 in thyroid cancer cells. (A) FF-1 (p63 negative), C-643 (p63 positive) and FTC-133 (p63 positive) thyroid cancer cells were incubated with 2 µM doxorubicin (Dox) for the indicated times. MCF-7 cells, which, unlike thyroid cancer cells, express wild-type (wt) p53, were used as a positive control. Cells were lysed and blotted with anti-pan-p63 monoclonal antibody 4A4, anti-p53 (DO-1), anti-p21, and anti-ß-actin as a loading control. COS-1 cells, transiently transfected with TAp63, TAp63 or Np63, were used as a positive control. mut, mutant. (B) C-98, FF-1 (p63 negative), C-643, FTC-133 (p63 positive) thyroid cancer cell lines were analysed for variations in their cell-cycle profiles before and after treatment with 2 µM doxorubicin. The bars represent average and standard deviation of the cell-cycle distribution from FACS analysis (G1, solid bars; S, open bars; G2/M, hatched bars) of three separate experiments. The MCF-7 cell line was used as a positive control. **P<0.01.

TA63 function is not restored by ectopic expression

Since DNA damage, following exposure to doxorubicin, was not able to increase the TAp63 level in thyroid cancer cells, we suspected a defect upstream of p63 induction. To address this issue, we induced TAp63 overexpression by transient transfection in C-98 (p63 negative, see Table 2) and FTC-133 (p63 positive, see Table 2) thyroid cancer cells. As a control, p53/p63 null human osteosarcoma cell line Saos-2 cells, which are commonly used as a model to study p53 family member functions (Yang et al. 1999, Ghioni et al. 2002), were also included. In all cell lines p53 was also transfected as a positive control (Fig. 4A).

View larger version (33K): [in this window] [in a new window]   Figure 4 Activity of ectopic TAp63 in thyroid cancer cells. (A) TAp63-Myc and p53-GFP were transfected in C-98 (p63 negative) and C-643 (p63 positive) thyroid cancer cells. The human osteosarcoma cell line, Saos-2 (p53 and p63 negative) was used as a control. Twenty-four hours after transfection, cells were lysed and subjected to Western blot analysis with either anti-Myc or anti-GFP monoclonal antibodies (two upper panels). Aliquots of the same samples were analysed for the expression of the downstream target protein p21Cip1 and reprobed with an anti-ß-actin antibody (two lower panels). (B) The same cells were transfected with an empty vector, TAp63-Myc, or p53 together with H2B-GFP to mark the transfected population. Twenty-four hours after transfection, the GFP-positive cells were analysed by FACS analysis to determine their cell-cycle profile. The bars shown (G0/G1, solid bars; S, open bars; G2/M, hatched bars) represent the average plus standard deviation from three separate experiments. *P<0.05, **P<0.01. (C) The same cells were transfected with an empty vector, TAp63-Myc, or p53 together with H2B-GFP to mark the transfected population. Twenty-four hours after transfection, the GFP-positive cells were analysed by FACS analysis to determine the apoptotic population (subG1). The bars shown (empty, open bars; TAp63, hatched bars; p53, solid bars) represent the average and standard deviation from three separate experiments. *P<0.05, **P<0.01.

We then tested the effect of overexpressed TAp63 on cell cycle distribution in C-98 (p63 negative, see Table 2), FTC-133 (p63 positive, see Table 2) and Saos-2 (control) cells (Fig. 4B). To this end, TAp63 or p53 were transiently transfected together with H2B-GFP to mark the transfected population. FACS analysis performed in the GFP-positive cells showed that TAp63 did not induce G1 arrest in thyroid cancer cells, whereas p53 did, in accordance with p21Cip1 induction (Fig. 4B). In contrast, both TAp63 and p53 were able to elicit G1 arrest in control Saos-2 cells (Fig. 4B). In addition, FACS analysis revealed that TAp63 did not increase the apoptotic population (subG1) in thyroid cancer cells (Fig. 4C), whereas p53 did (Fig. 4C). This increase, however, was less than 10% of the total population, in accordance with previous reports indicating that thyroid cancer cells are refractory to p53-induced apoptosis (Moretti et al. 1997, 2000). In control Saos-2 cells, both TAp63 and p53 significantly increased apoptosis (subG1 was approximately 50% of the total population) (Fig. 4C).

These data indicate that even ectopic overexpression of TAp63 is not able to elicit p53-like responses in thyroid cancer cells and raises the question why this protein does not exert any tumour suppressor activity in these cells.

TAp63 does not interact with p53 mutants in thyroid cancer cells

Since transient transfection experiments suggested a possible defect intrinsic to TAp63, we tested whether TAp63 protein in thyroid cancer cells is inhibited by the direct interaction with p53 mutants (Gaiddon et al. 2001, Strano et al. 2002). To this end, we performed co-immunoprecipitation experiments in p63 positive (C-643 and FTC-133), and p63 negative (C98) thyroid cancer cell lines, and in the A431 oesophageal cancer cell line (Park et al. 1994) as a positive control (Fig. 5). In A431 cells we were able to detect p63 in anti-p53 immunoprecipitates (Fig. 5), whereas we did not find p53/p63 co-immunoprecipitation in any thyroid cancer cell line (Fig. 5).

View larger version (17K): [in this window] [in a new window]   Figure 5 Co-immunoprecipitation experiments for p53 and p63. Lysates from p63-positive 8505C, C-643, Hth-74 and FTC-133 thyroid cancer cells were immunoprecipitated (IP) with either anti-p63 (H129) or anti-p53 (DO-1) antibody and blotted with an anti-p63 monoclonal antibody (4A4). A431 cells were used as a positive control (on the left), whereas COS-1 cells transfected with TAp63 were used as a standard (on the right). No co-immunoprecipitation is evident in thyroid cancer cell lines.

p63 function is selectively abrogated in the TAp63 isoform

The p63 protein exists in various C-terminal splicing isoforms (, ß and ) and, at variance with p53, contains a C-terminal domain, which is a protein interaction module endowed with inhibitory function (Moll et al. 2001). Such a domain is present in TAp63 but absent in TAp63ß and TAp63. Moreover, it has been reported that exon 13 alone, located at the p63 C-terminus and missing in TAp63ß, may inhibit TAp63 transcriptional activity by interaction with unknown proteins (Ghioni et al. 2002). To test whether the C-terminal inhibitory domain is involved in the inactivation of TAp63 in thyroid cancer cells, we compared the transactivation activity of ectopic TAp63, TAp63ß (lacking exon 13), TAp63 (lacking exons 11–14), Np63 and p53 on the target genes p21Cip1, Bax and MDM2 in thyroid cancer cells and p53/p63-null Saos-2 cells (Table 3). Luciferase assays (expressed as a percent of maximal induction) revealed that TAp63, TAp63ß, TAp63 and p53 displayed comparable transactivation activity in Saos-2 cells, as previously reported (Table 3) (Ghioni et al. 2002). In contrast, TAp63 displayed a weak transactivation activity in thyroid cancer cells (C-98, FF-1, C-643 and FTC-133), similar to that observed with the dominant negative isoform Np63 (Table 3).

View larger version (45K): [in this window] [in a new window]   Figure 6 Comparison of TAp63 and TAp63 activity in thyroid cancer cells. (A) TAp63 and TAp63 were transfected in C-98 (p63 negative) and FTC-133 (p63 positive) thyroid cancer cells and Saos-2 (p53 and p63 negative) osteosarcoma cells. Twenty-four hours after transfection, cells were lysed and subjected to western blot analysis with anti-p21Cip1 (upper panel), anti-Myc (middle panel) and anti-ß-actin antibody (lower panel). (B) C-98 (p63 negative) and FTC-133 (p63 positive) thyroid cancer cells were transfected with the indicated p53 and p63 constructs together with H2B-GFP to mark the transfected population. Twenty-four hours after transfection, the GFP-positive cells were analysed by FACS to determine their cell-cycle profile. Bars show the profile of transfected cells expressed as a percentage of the total population (G1, solid bars; S, open bars; G2/M, hatched bars) and represent the average ± S.E. from three separate experiments. *P<0.05, **P<0.01.

These results suggest that TAp63 does not act as a typical p53 family member in thyroid cancer cells, which maintain a high responsiveness to the tumour suppressor activity of TAp63ß and TAp63.

TAp63 exerts a dominant negative effect on p53 activities

It is known that Np63, the N-terminal truncated p63 isoform, is devoid of transactivation activity and is able to exert a dominant negative effect towards p53 family members (Yang et al. 1998). Since TAp63 displayed very poor transactivation activity in thyroid cancer cells, comparable to that of Np63, we tested whether TAp63 could antagonize p53 activities.

We transfected p21Cip1, Bax and MDM2 promoters together with p53, TAp63 and Np63 in C-98, FF-1 (p63 negative, see Table 2), C-643, FTC-133 (p63 positive, see Table 2) thyroid cancer cells and in p53-null Saos-2 cells. The ability of TAp63 to antagonize the effect of p53 was evaluated by luciferase assay (Table 4). In Saos-2 cells, TAp63 did not significantly affect the activity of p53 on reporter genes (Table 4), whereas Np63 exerted an antagonistic effect, as expected (Table 4). However, it was interesting to note that in Saos-2 cells the effect of TAp63 was not additive or synergistic with p53 (Table 4). Surprisingly, TAp63 antagonized p53 transactivation activity to a variable extent in thyroid cancer cells (C-98, FF-1, C-643, FTC-133; Table 4), in a manner similar to that of Np63 (Table 4).

View this table: [in this window] [in a new window]   Table 4 Inhibition of p53 transcriptional activity on p21, Bax and Mdm2 promoters by TAp63 and Np63 in thyroid cancer cells

View larger version (49K): [in this window] [in a new window]   Figure 7 Effect of TAp63 on p53-mediated tumour suppression in thyroid cancer cells. (A) Saos-2 osteosarcoma cancer cells (p53 and p63 negative), C-98, FF-1 (p63 negative), and C-643, FTC-133 (p63 positive) thyroid cancer cells were transfected with the indicated constructs together with H2B-GFP. Forty-eight hours after transfection, GFP-positive cells were counted under a fluorescence microscope and expressed as a percentage of total cells. Values represent means ± S.D. of three separate experiments performed in duplicate. (B) C-98 (p63 negative) thyroid cancer cells were transfected with the indicated constructs. Cells were than split onto 100 mm Petri dishes, subjected to antibiotic (geneticin) selection for 2–4 weeks, and foci were stained with Crystal Violet. (C) C-98, FF-1 (p63 negative), C-643 and FCT-133 (p63 positive) thyroid cancer cells were transfected with the indicated constructs. Cells were than split onto 100 mm Petri dishes, subjected to antibiotic (geneticin) selection for 2–4 weeks, and foci were stained with Crystal Violet, as above. Bars represent the number of foci contained in each plate (mean ± S.D. from eight different plates).

Since these data were obtained by p63 overexpression, we tried an alternative approach aimed at reducing the expression of TAp63 (Fig. 8). To this end, we subjected the p63-positive FTC-133 cells, which display the highest level of TAp63 (Fig. 3A), and the p63-negative C-98 cells (as a negative control) to p63 silencing by the siRNA technique (Fig. 8). Forty-eight hours after incubation with siRNAs, we transfected p53 together with the p21Cip1 reporter. The luciferase assays revealed that p63 silencing in FTC-133 cells enhanced the p21Cip1 promoter response to increasing doses of p53 cDNA, whereas no difference was observed in the p63-negative C-98 cells, as expected (Fig. 8). Taken together, these data suggest that in thyroid cancer cells TAp63 may have a tumour-promoting role by antagonizing p53 in a manner similar to that of the dominant negative isoform, Np63.

View larger version (20K): [in this window] [in a new window]   Figure 8 Effect of TAp63 silencing on p53 transcriptional activity in thyroid cancer cells. C-98 (p63 negative) and FTC-133 (p63 positive) thyroid cancer cells were transfected with the indicated siRNAs. Forty-eight hours after transfection, cells were transfected with the indicated doses of pCDNA3.1p53 together with p21Cip1 gene reporter. After 24 h cells were lysed and samples were subjected to luciferase assay. Values represent means ± S.D. of three separate experiments performed in triplicate and are expressed as fold induction over empty-transfected cells arbitrarily set at 1. A representative blot for p63 in FTC-133 cells treated with the indicated siRNAs is shown on the right.

	Discussion

Top Abstract Introduction Materials and methods Results Discussion References

Evidence of p63 expression in human cancers is not novel and has already been reported in a variety of human malignancies, including thymomas, non-Hodgkin’s lymphomas, bladder, breast, prostate and lung cancers (Moll et al. 2001). It is noteworthy that p63 has also been detected in a small subset of papillary and anaplastic thyroid carcinomas (Preto et al. 2002). Immunohistochemistry experiments performed in those studies, however, might have underestimated the real prevalence of p63 expression in thyroid tumours, because of the use of paraffin-embedded specimens, in which the antigenic properties of p63 can be altered. In contrast, by examining frozen samples of thyroid cancer we found that the large majority was positive for p63 expression by immunohistochemistry, a probable effect of the better preservation of the p63 epitopes with the freezing process.

p63 up-regulation in human tumours (Okada et al. 2002) is often concomitant with the overexpression of the dominant negative isoform Np63 (Crook et al. 2000, Hibi et al. 2000, Park et al. 2000). In some tumours, Np63 is preferentially expressed, suggesting that this p63 isoform may act as an oncogene (Crook et al. 2000, Hibi et al. 2000, Park et al. 2000). However, a previous report showing TAp63 expression in gastric cancer suggests that TAp63 isoforms may also be involved in tumour progression (Tannapfel et al. 2001). This might be the case for thyroid tumours, where Np63 is not expressed.

Our studies indicate that in thyroid cancer TAp63 is devoid of any onco-suppressor activity. In fact, genotoxic stress caused by doxorubicin was not able to activate TAp63 in malignant thyroid cancer cells, suggesting that this p63 isoform was not involved in p53-like activities. Since no increase in TAp63 expression was observed after doxorubicin treatment, we first hypothesized an impaired signalling upstream of TAp63 itself. However, TAp63 overexpression, obtained by transient transfection, also failed to induce the target genes p21Cip1, Bax and MDM2, suggesting that the defect was either downstream or intrinsic to TAp63 itself. A defect downstream of p63 was ruled out with the ectopic expression of p53, TAp63ß and TAp63, which share similar downstream pathways with TAp63. All these onco-suppressors were still able to transactivate target genes and to exert tumour suppressor activity in these thyroid cancer cells.

One possible inactivation mechanism intrinsic to the p63 protein is the interaction with p53 mutants (Gaiddon et al. 2001, Strano et al. 2002). In our hands, however, co-immunoprecipitation experiments performed in thyroid cancer cells excluded such interaction. Moreover, experiments performed in the p53-null thyroid cancer cell line SW1736 also ruled out the possibility of a p53 interference with TAp63 (not shown).

In thyroid cancer cells, only TAp63 is transcriptionally weak, since TAp63ß and TAp63 display transcriptional activity similar to that of p53. Therefore, the onco-suppressor activity defect is restricted to the TAp63 isoform. One difference between p53 and p63 resides in the carboxyl tail: p63, in a similar manner to p73, contains a carboxyl terminus that undergoes alternative splicing and gives rise to different isoforms. Deletion studies have shown that the last 71 amino acids at the C-terminal domain (TI domain), which are missing in both TAp63ß and TAp63, are endowed with inhibitory properties towards p63 transcriptional activity (Serber et al. 2002). This inhibition occurs by an intramolecular interaction between the TI domain and the transactivation domain (TA) located at the N-terminus of p63 (Serber et al. 2002). This interaction is responsible for the occupancy of the TA domain, which, consequently, is no longer available for trans-activation (Serber et al. 2002). Moreover, previous evidence established a dominant negative capability of the TI domain, since Np63 (endowed with TI), but not Np63 (devoid of TI) is able to inhibit TAp63 activity (Yang et al. 1998). These results outline, therefore, the importance of the TI domain in the dominant negative activity of Np63 isoforms. A previous report has also shown that the p63 C-terminal domain encoded by exon 13 (which is missing in TAp73ß) exerts an inhibitory effect on TAp63 activity, possibly by interacting with various proteins (Ghioni et al. 2002). It is also known that, although endowed with a very weak transcriptional activity, TAp63 is still able to bind DNA in a manner similar to that of p53 and the other more active p63 isoforms (Yang et al. 1998). Taken together, this evidence suggests that in some cell contexts, such as in thyroid cancer cells, TAp63 may occupy the DNA binding sites of p53 responsive elements, thereby preventing occupancy by more transcriptionally active p53 family members. This interpretation could partially explain why TAp63 expression may antagonize p53-mediated tumour suppression in thyroid cancer cells, thus establishing an oncogenic role similar to that of Np63.

Although this is the first report dealing with direct evidence of a dominant negative/pro-tumourigenic role of TAp63, indirect evidence of this unsuspected TAp63 role is already present in the literature. Indeed, patients with mutations that introduce a premature stop codon in the TAp63 C-terminus show defects in hands and feet similar to those with mutations in the DNA binding domain (DBD) (Celli et al. 1999), suggesting that the loss of DNA binding capability by TAp63 (due to mutations in the DBD) has effects similar to those observed with TAp63 inappropriate activation due to the loss of the TI domain (stop codon). Therefore, both DNA binding and weak transactivation activities may be required for TAp63 to allow proper skeletal development, in order to avoid the premature apoptosis in skeletal precursors. In the light of these considerations, TAp63 should be able to antagonize either homologue (TAp63ß and TAp63) or paralogue (p53, TAp73) p53 family members, and TAp63 expression in thyroid cancer could be regarded as a mechanism aimed at inhibiting p53-mediated apoptosis.

The TI domain of p63 is very similar to that of p73, but very different from that of p53 (Moll & Slade 2004). Indeed, a similar antagonistic role against p53-mediated onco-suppression has also been reported for TAp73 in ovarian cancer (Vikhanskaya et al. 2000) and leukaemia cells (Freebern et al. 2003). It is reasonable to suppose, therefore, that TAp63 and TAp73, which are endowed with weaker transcriptional activity than p53, may acquire antagonistic properties against p53 in the presence of unknown co-repressor(s) present in some cancer types. This could partially explain why the dominant negative effect of TAp63 is not observed in Saos-2 cells, suggesting that this phenomenon may be cell context-dependent. In this situation, well-differentiated thyroid tumours, rarely harbouring p53 mutations, may take advantage of the expression of TAp63 and Np73 (Frasca et al. 2003), which may antagonize p53 and, at variance with p53, are resistant to MDM2-mediated degradation (Okada et al. 2002).

The complexity of the p53 family protein network must be taken into account when considering gene therapy in thyroid cancer aimed at restoring the wild-type p53 status. More specifically: (a) unlike TAp63, TAp63ß and TAp63 are still effective in transactivating target genes and providing tumour suppressor functions in poorly differentiated thyroid cancer cells; (b) p63 proteins are strictly homophilic and are refractory to tetramerization with different members of the family (Moll & Slade 2004) and, therefore, at variance with wild-type p53, ectopic TAp63ß and TAp63 are more resistant to the dominant negative effect of p53 mutants. Hence, adenoviral vectors carrying TAp63ß and TAp63 should be tested in the anaplastic thyroid cancer cells harbouring p53 mutations since they may prove useful for designing possible anti-cancer therapies by gene delivery.

	Acknowledgements

 
This work was supported by a grant from AIRC (Associazione Italiana Ricerca sul Cancro) and MURST (Ministero dell’Università e della Ricerca Scientifica e Tecnologica) to R V. F F and P V are fellows of the AICF (American Italian Cancer Foundation). V V and A A are supported by a fellowship from AIRC. The authors declare that there is no conflict of interest that would prejudice the impartiality of this scientific work.

	Footnotes

 
^* (R Malaguarnera and A Mandarino contributed equally to this work)

	References

Top Abstract Introduction Materials and methods Results Discussion References

 
Blagosklonny MV 2002 Sequential activation and inactivation of G2 checkpoints for selective killing of p53-deficient cells by microtubule-active drugs. Oncogene 21 6249–6254.[CrossRef][Medline]

Celli J, Duijf P, Hamel BC, Bamshad M, Kramer B, Smits AP, Newbury-Ecob R, Hennekam RC, Van Buggenhout G, van Haeringen A et al. 1999 Heterozygous germline mutations in the p53 homolog p63 are the cause of EEC syndrome. Cell 99 143–153.[CrossRef][ISI][Medline]

Crook T, Nicholls JM, Brooks L, O’Nions J & Allday MJ 2000 High level expression of deltaN-p63: a mechanism for the inactivation of p53 in undifferentiated nasopharyngeal carcinoma (NPC)? Oncogene 19 3439–3444.[CrossRef][ISI][Medline]</