INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

Transcriptional regulation of specific groups of genes plays a pivotal role in development as well as in differentiation. It has become increasingly clear that regulation of gene expression is driven by the differential activity of transcriptional regulators whose distribution in cells and tissue is often wider than that of their target genes. Furthermore, increasing evidence demonstrates that not only activators but also repressors have a role in the regulation of expression of specific genes (21, 22, 25, 38). A number of transcriptional repressors contain zinc finger and homeodomain motifs (21).

The negative regulator ZEB has recently been identified as the vertebrate homologue of the Drosophila melanogaster zinc finger homeodomain factor, zfh-1 (13, 14, 18, 40, 44). This relies on the high homology at the sequence level, the number of zinc fingers, the location of the zinc finger clusters and homeodomains within the protein, and the genomic organization of the genes (45). Zfh proteins are required for proper differentiation of the central nervous system and for the early process of organogenesis in mesodermal tissues (28, 29, 33). Zfh-1 mutants show imbalanced distribution as well as defects in the segregation of muscle precursors, which are mispositioned in the embryo. ZEB was originally identified as a DNA binding protein. It contains a homeodomain and two zinc finger clusters and binds to a subset of E-box-like sequences, with highest affinity for CACCT and CACCTG, through its N- and C-terminal zinc finger clusters (15, 18, 45). Unlike Twist and members of the Id family, which act by binding and inactivating MRF and MEF-2 proteins, ZEB down-regulates muscle genes by binding to a subset of E boxes and functioning as a transcriptional repressor (2, 5, 39, 40, 48). Overexpression of ZEB in C2C12 cells inhibits myotube formation and expression of specific differentiation markers such as myosin heavy chain and myogenin (39). ZEB appears identical to its chicken and mouse homologues, called EF1, whose overexpression inhibits MyoD-induced expression of some muscle genes in 10T1/2 cells (44). EF1/ZEB null mutant mice develop to term but never survive postnatally. They exhibit severe skeletal defects of various lineages such as craniofacial alterations of neural crest origin, limb and sternum defects, and hypoplasia of intervertebral discs (54). These mice also exhibit defects in hematopoiesis. Their thymi are poorly developed without clear demarcation between cortex and medulla. Furthermore, there is a drastic decrease in the total number of T cells accompanied by impaired thymocyte development (23). These findings are consistent with the fact that ZEB represses interleukin-2 (IL-2) gene expression by binding negative regulatory element NRE-A in the IL-2 promoter (59).

p73, the recently discovered p53 family member, is a nuclear protein that binds to canonical p53 DNA binding sites in gel shift assays and that activates transcription from p53-responsive promoters in transient-transfection experiments (26, 27, 30, 53). Furthermore, overexpression of p73 in p53^+/+ and p53^/ cancer cells induces growth arrest, apoptosis, and differentiation, recapitulating some of the p53 biological activities (1, 10, 26, 27). Unlike p53, p73 is alternatively spliced, giving rise to a family of different isoforms whose individual physiological functions are still unknown (8, 9, 27, 57). p73-deficient mice exhibit severe defects, including hydrocephalus, hippocampal dysgenesis, chronic infections and inflammation, and abnormalities in the pheromone sensory pathway (31, 57). However, they do not develop any spontaneous tumor (57). To date no mutations of p73 have been found in tumors, despite an extensive search (51). In human tumors bearing p53 mutations, p73 biological activities are strongly inhibited by physical interaction with human tumor-derived p53 mutants (11, 17, 35, 50, 51, 58)

The human p73 promoter has recently been characterized (12). It has a TATA-like box and displays a low homology to the p53 promoter (31). Partial characterization of a large region upstream to the first exon has revealed the presence of at least three E2F binding sites, which may account for the recent finding that p73 expression is triggered at both mRNA and protein levels by E2F-1 overexpression (24, 32, 49, 61). It has recently been reported that p73 expression is triggered along neuronal and hematopoietic differentiation, although the molecular mechanisms responsible for this modulation have not been elucidated yet (10, 55). Here we report that p73 mRNA levels are upregulated during muscle and neuronal differentiation of C2C12 and P19 cells, respectively, as well as upon tetradecanoyl phorbol acetate (TPA)-induced monocytic differentiation of HL60 cells. We investigated the transcriptional regulation of p73 expression during differentiation and identified a 1-kb negative regulatory fragment in the first intron of the P73 gene immediately upstream of exon 2. This element functions as a transcriptional silencer in a gene reporter assay and is able to markedly reduce the induction of its own p73 promoter despite strong activation by exogenous E2F-1. Importantly, ZEB binds to six consensus boxes located within this negative regulatory fragment in proliferating cells. This binding is clearly reduced during differentiation of C2C12 cells. Overexpression of dominant-negative ZEB (DB-ZEB) releases the transcriptional repression by ZEB and restores p73 expression in proliferating C2C12 and P19 cells. Thus, we propose a mechanism whereby ZEB regulates p73 gene expression at the crossroad between proliferation and differentiation.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Cloning of 1-kb fragment. A 1-kb fragment of the first intron of p73 was isolated using a PromoterFinder DNA walking kit (Clontech). The fragment was isolated by PCR from human library HDL-4 (catalog no. K1803-1) included in the kit; HDL-4 is a PvuII-digested and adapter-ligated genomic library. The sequences of the gene-specific primers used were complementary to the exon 2 sequence in the p73 coding strand. The sequences of the adapter primers and gene-specific primers used for the primary (AP1, p73A) and secondary (nested) (AP2, p73B) PCR are as follows: AP1, 5'-GTA ATA CGA CTC ACT ATA GGG C; AP2, 5'-ACT ATA GGG CAC GCG TGG T; p73A, 5'-AGA GCT CCA GAG GTG CTC AAA CGT G; p73B, 5'-GTG CTC AAA CGT GGT GCC CCA TCA G. PCRs were performed using the KlenTaq LA DNA polymerase mixture (Sigma). The PCR product was cloned by TA into vector pCR2.1 (Invitrogen) generating vector pCR1000.

Reporter vectors. A 1-kb fragment of the first intron of p73 was amplified by PCR from vector pCR1000 to eliminate the ATG of exon 2 of the p73 gene. This product was cloned in the BamHI site of vector Po-LUC (4), generating p73intr-LUC. A herpes simplex virus (HSV) thymidine kinase promoter obtained by digestion of vector pBLCAT2 at XhoI and HindIII restriction sites was cloned in the SmaI site of Po-LUC, generating TK-LUC, in the SalI site of p73intr-LUC, generating TK-p73intr-LUC, and in the HindIII site of p73intr-LUC, generating p73intr-TK-LUC. The 4-kb and the 370-bp fragments of the p73 promoter derive from BAC clone 190O18 (Research Genetics) of the CITB-978SK-B Hu BAC library (6). A 7.3-kb fragment obtained from EcoRI digestion of this BAC clone was cloned in the EcoRI site of the pBluescript KS vector in the sense orientation, generating vector pBS-7.3-prom. The 4-kb promoter fragment was produced by digestion of pBS-7.3-prom at PvuI sites. This fragment was blunted and cloned in the SmaI site of Po-LUC, generating p73prom-LUC, in the SalI site of p73intr-LUC, generating p73prom-p73intr-LUC, and in the HindIII site of p73intr-LUC, generating p73intr-p73prom-LUC. The 370-bp fragment of the p73 promoter was produced by digestion of pBS-7.3-prom at PstI sites, blunted, and cloned in the SmaI site of PoLUC, generating 370prom-LUC. The 1-kb fragment of the first intron of the p73 gene was cloned in the HindIII site of 370prom-LUC, generating 370prom-p73intr-LUC. The 1-kb fragment of the first intron of the p73 gene carrying mutated boxes 1, 3, and 5 was cloned in the HindIII site of 370prom-LUC, generating 370prom-p73intrM3-LUC Vectors pCI-neo (Promega) carrying Flag-tagged ZEB (pZEB) or the DNA binding domain of ZEB (pZEB-DB) were kindly provided from D. Dean (39). Vectors pCMV-E2F1 and pCMV-E2F1/E132 were kindly provided by K. Helin. (56). MyoD vector was kindly provided by M. Crescenzi (7).

Site-directed mutagenesis. Site-directed mutagenesis was performed by PCR on vector p73intr-LUC using the QuikChange site-directed mutagenesis kit (Stratagene) according to the manufacturer's protocol. Three of the six consensus binding sites for ZEB were mutated, generating vector p73intrM3-LUC. The following oligonucleotides were used to mutagenize, respectively, boxes 1, 3, and 5: mBOX1c, GCC TGG ACA CTG CCG GAT CCT CAT GGG TGT CC; mBox3c, TGA TCC AGG CCC GCC CCG GGA AGG CAG AGC; mBox5c, GCA AGG CGG GGG CTC GAG CTC CAG GGA TGC (mutated sites are underlined).

Cell cultures. C2C12 myoblasts were cultured in Dulbecco modified Eagle medium (DMEM) containing 10% fetal bovine serum (FBS); differentiation was induced by plating the cells onto collagen-coated dishes and switching them to serum-free (SF) medium for 72 h: DMEM supplemented with 5 µg of human insulin, 5 µg of human (holo-) transferrin, and 5 ng of sodium selenite/ml. Fifty micromolar Ara-C was added to the SF medium to eliminate undifferentiated cells. More than 90% of the cells become terminally differentiated under these conditions (43). HL60 and H1299 cells were cultured in RPMI medium containing 10% FBS; differentiation of HL60 cells was induced by the addition of TPA (10⁸ M) to the medium for 48 h. P19 cells (kindly provided by A. M. Salvatori) were cultured in mimimal essential medium (-MEM) containing 7.5% newborn calf serum and 2.5% FBS. Differentiation of P19 cells was induced by plating the cells in bacteriology dishes with -MEM containing 1 µM retinoic acid (RA). After 4 days in RA the cells had formed large embryoid bodies with central necrotic areas; these aggregates were trypsinized, plated on poly-L-lysine-coated plates in medium containing 1 µM Ara-C (36), and harvested at different times.

Transfections and luciferase assays. Transient and stable transfections were performed by CaPO₄-mediated DNA precipitation (20). Stable transfected polyclonal populations overexpressing pZEB-DB or TK-p73intr-LUC were selected in medium containing puromycin (2 µg/ml) for 1 week. For luciferase assays the above-mentioned cell lines were transfected with the reporter plasmid together with various plasmid combinations. An equal number of pCMV -gal plasmids was added to each transfection reaction mixture. Thirty-six hours after the transfection the cells were harvested and luciferase activity was assayed on whole-cell extract, as described previously (4). The values were normalized for -galactosidase and protein contents.

RNA extraction and RT-PCR analysis. Expression of p73 mRNA was analyzed by reverse transcription-PCR (RT-PCR) amplification. Total cellular RNA was extracted with RNAzol B (Biotech, Rome, Italy) according to the manufacturer's protocol from proliferating cells at different times after induction of differentiation. Five micrograms of total RNA was reverse transcribed at 37°C for 45 min in the presence of random hexamers and Moloney murine leukemia virus reverse transcriptase (Gibco-BRL). p73 mRNA analysis was carried out by PCR using oligonucleotides specific for the N-terminus-coding region (42) or for the DNA-binding domain-coding region; the sequences of these two sets of oligonucleotides are, respectively as follows: N-ter/up, 5'-GAG CAC CTG TGG AGT TCT CTA GAG, and N-ter/down, 5'-GGT ATT GGA AGG GAT GAC AGG CG; dbd/up, 5'-CCA AGT CAG CCA CCT GGA CG, and dbd/down, 5'-CTG CTG TTA CAC ATG AAG TTG TAC AGG. The housekeeping aldolase A mRNA, used as an external control, was amplified from each cDNA reaction mixture using the following specific primers: HumAld1, 5'-CGC AGA AGG GGT CCT GGT GA; HumAld2, 5'-CAG CTC CTT CTT CTG CTG CG; MurAld1, 5'-TGG ATG GGC TGT CTG AAC GCT GT; MurAld2, 5'-AGT GAC AGC AGG GGG CAC TGT. Amplified PCR products were electrophoresed on a 2% agarose gel containing ethidium bromide (0.5 µg/ml) and visualized under UV light.

EMSA. Electrophoretic mobility shift assays (EMSAs) were performed on a 25-µl DNA binding reaction mixture which contained 5 to 10 µg of C2C12 whole-cell extract, 4 fmol of labeled duplex oligonucleotides, binding buffer (20 mM Tris-HCl [pH 7.8], 60 mM KCl, 0.5 mM EDTA, 0.1 mM dithiothreitol, 3 mM MgCl₂), 1.5 µg of poly(dI-dC), 10 mM spermidine, and 100 to 400 ng of salmon sperm. The reaction was carried out at room temperature for 15 min, and the protein-DNA complexes were subjected to native electrophoresis on 5% polyacrylamide-0.5× TBE gels. The following oligonucleotides were used as probes (consensus sites are underlined, mutated sites are in boldface): box 1, 5'-TGG ACA CTG CCA CCT CCT CAT GGG T; box 2, 5'-GGG ACC TGA GCC ACC TCC AGG TCC CGG; box 3/4, 5'-TGA TCC AGG CCC GCA CCT CCA AGG CAG AGC TGC CCA CCT GGC CTT CGG TTT CC; box 5, 5'-GCA AGG CGG GGG CAC CTG CTC CAG GGA TGC; mbox 5, 5'-GCA AGG CGG GGG CTC GAG CTC CAG GGA TGC; box 6, 5'-CCA GGG TGC TCA GGT GTC ATT CCT TCC. In supershift experiments antibodies were added to the mixture before the labeled oligonucleotides and the mixture was incubated for 10 min at room temperature. For supershift of ZEB, NF-YB, and ZEB-DB the following amounts of antibodies were used, respectively: 5 µl of crude polyclonal anti-ZEB, purified from rabbits immunized with EF1, the chicken homologue of ZEB, obtained from Funahashi et al. (15); 200 ng of affinity-purified rabbit polyclonal anti NF-YB (kindly provided from R. Mantovani); 2 µg of monoclonal anti-Flag (Sigma). The recombinant ZEB protein used in gel shift assays was produced from a plasmid carrying the ZEB cDNA under the control of the T7 promoter using TnT coupled reticulocyte lysate systems (Promega) according to the manufacturer's protocol.

Western blot analysis. For p73 detection a polyclonal antibody kindly provided by Y. Shaul was used at 1:3,000 (52). For ZEB detection the above-mentioned crude polyclonal anti-ZEB (15) was used. C2C12 and P19 cells were stably transfected with pZEB-DB. For ZEB-DB detection, a monoclonal anti-Flag (Sigma) antibody was used at 1:500. Western blot analysis was performed with the aid of the enhanced chemiluminescence system (Amersham Pharmacia Biotech, Inc.).

Formaldehyde cross-linking and chromatin immunoprecipitation. Formaldehyde cross-linking and chromatin immunoprecipitation were performed as previously described (3). Immunoprecipitation was performed with protein G-agarose (KPL). The chromatin solution was precleared by adding protein G for 1 h at 4°C, aliquoted, and incubated with a mixture containing 3 µg of affinity-purified anti-NF-YB rabbit polyclonal antibody (kindly provided by R. Mantovani), 2 µg of affinity-purified anti-Sp1 rabbit polyclonal antibody (Santa Cruz Biotechnology), and 7 µl of anti-EF1 polyclonal antiserum, preimmune serum, or no antibody overnight at 4°C with mild shaking. Before use, protein G-agarose was blocked with 1 µg of sonicated salmon sperm DNA and 1 µg of bovine serum albumin for 4 h at 4°C and then incubated with chromatin and antibodies for 3 h. Immunoprecipitates were eluted and precipitated with ethanol. The pellets were resuspended in 30 µl of H₂O and analyzed by PCR. The total-input sample was resuspended in 100 µl of H₂O and then diluted 1:100 before PCR. For PCR analysis on cyclin B1, cyclin B2, and thymidine kinase promoters and the p73 gene first intron the following oligonucleotides were used: cycB1-up3, 5'-TGT AGA CAA GGA AAC AAC AAA GCC TGG TGG CC; cycB1-down2, 5'-CAG CCA CTC CGG TCT GCG ACA; cycB2-up3, 5'-TGT AGA CAA GGA AAC AAC AAA GCC TGG TGG CC; cycB2-down2, 5'-CAG CCA CTC CGG TCT GCG ACA; tk-up, GCC CCT TTA AAC TTG GTG GGC; tk-down, GTG AAC TTC CCG GAG GCG CAA; ZEB2C, 5'-GGG ACC TGA GCC ACC TCC AGG TCC CGG; p73h, 5'-CCG GCC TCC GAG GGC AGC T.

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

p73 mRNA level is upregulated during cell differentiation. Recent evidence has clearly shown that p73 mRNA levels increase during neuronal and hematopoietic differentiation (10, 55). Therefore, we analyzed whether p73 expression is modulated during neuronal and muscle differentiation using P19 and C2C12 myoblasts, respectively. As a control we used the monocytic HL-60 cells to monitor p73 expression during hematopoietic differentiation. Total RNA was extracted from C2C12 myoblasts upon serum deprivation (Fig. 1A), from P19 cells upon RA treatment (Fig. 1B), and from monocytic HL60 differentiating cells (Fig. 1C) and was subjected to RT-PCR. We found that p73 mRNA was upregulated during differentiation of the above-mentioned cell lines (Fig. 1). Interestingly, the kinetics of p73 upregulation during C2C12, P19, and HL60 differentiation are quite different. In C2C12 cells p73 mRNA was elevated at 12 h and reduced from 24 to 72 h (Fig. 1A), but in HL60 cells it was induced at 12 to 48 h and remained high due to the TPA treatment (Fig. 1C). On the other hand, in P19 cells p73 mRNA is induced after RA treatment (5 days; Fig. 1B). By contrast, the p73 transcript was undetectable in proliferating C2C12 cells (Fig. 1A) and was expressed at very low levels in undifferentiated P19 and HL60 cells (Fig. 1B and C).

View larger version (34K): [in this window] [in a new window]   FIG. 1.   p73 mRNA is upregulated during muscle, neuronal, and monocytic differentiation of C2C12 (A), P19 (B), and HL60 (C) cells, respectively. Total RNA was extracted from each of the above-mentioned cells and subjected RT-PCR as described in Materials and Methods. Cycle numbers employed in the semiquantitative RT-PCR are shown in panel A. Amplification of aldolase A (Ald-A) was used to normalize equal loading of each RNA sample. The lengths of amplified fragments are shown on the left. Prol., proliferating cells.

Identification of a negative regulatory fragment in the first intron of the p73 gene. It has previously been reported that exon and intron organization of the p53 and p73 genes is highly conserved (27, 34, 47). In the p53 gene, as well as in the p73 gene, the first exon is untranslated and the mRNA derived from this exon might influence translation (37). In particular, the first intron of the p73 gene is very large (nearly 30 kb), within which positive and/or negative regulatory elements may reside (34).

View larger version (42K): [in this window] [in a new window]   FIG. 2.   Identification of 1-kb fragment in the first intron of the p73 gene. (A) Schematic representation of the first intron of the p73 gene flanked by the respective exon 1 and exon 2. The 1-kb fragment located immediately upstream to the starting site of exon 2 is indicated (gray box). (B) Sequence of the 1-kb intronic fragment. The binding sites for ZEB are marked.

View larger version (22K): [in this window] [in a new window]   FIG. 3.   The p73 gene intronic fragment functions as a transcriptional silencer of the thymidine kinase (TK) promoter. (A) Schematic representation of the plasmids used in the transfections listed below. (B to G) C2C12, P19, H1299, and Hela cells (1.5 × 105/60-mm dish) were transiently transfected with the indicated plasmids. An equal amount of CMV--gal was added to each transfection mixture. Cell extracts were prepared 36 h later and subjected to determination of luciferase activity. Results are presented as luciferase activity (Luc) relative to total proteins and -galactosidase (-gal) activity. Histograms show the means of three experiments performed in duplicate.

The p73 gene intronic fragment markedly reduces the activity of its own p73 promoter. To further characterize the silencer activity of the intronic fragment, we first assessed its ability to repress its own p73 promoter. To this end, Po-LUC vectors (4) in which the intronic fragment was inserted downstream of 370-bp or 4-kb fragments of the p73 promoter (370prom-p73intr-LUC and p73prom-p73intr-LUC, respectively) were transiently transfected in H1299, 293, and C2C12 cells. These p73 promoter fragments are similar to those already reported in the literature (12, 49). As shown in Fig. 4 and 5A and B the activities of such promoter fragments were markedly reduced by the presence of the first intron.

View larger version (33K): [in this window] [in a new window]   FIG. 4.   The intronic fragment strongly reduces the transcriptional activity of a short fragment of the p73 promoter. H1299 and 293 cells (1.5 × 105/60-mm dish) were transiently transfected with the indicated plasmids. An equal amount of CMV--gal was added to each transfection mixture. Cell extracts were prepared 36 h later and subjected to determination of luciferase activity. Results are presented as luciferase activity (Luc) relative to total proteins (prot) and -galactosidase (-gal) activity. Histograms show the means of three experiments performed in duplicate.

View larger version (17K): [in this window] [in a new window]   FIG. 5.   The p73 gene intronic fragment markedly reduces the transcriptional activity of its own promoter as well as upon induction of E2F-1. (A and B) C2C12 and H1299 (1.5 × 105/60-mm dish) cells were transiently transfected with the indicated plasmids. Luciferase (Luc) and -galactosidase (-gal) activities were determined 36 h after transfection. (C) H1299 cells were transiently cotransfected with the indicated plasmid combinations and processed as for panels A and B. Results are shown as luciferase activity relative to total proteins (prot) and -galactosidase activity. Histograms show the means of three experiments performed in duplicate.

The transcriptional repressor ZEB binds to the E boxes of the p73 gene intronic fragment in vitro and in vivo. Computer-assisted analysis of the p73 gene intronic sequence revealed the presence of six consensus sites for the ZEB zinc finger/homeodomain repressor. Indeed, it has been reported that ZEB binds to the E box, in particular, to a subset of E box-like sequences, with highest binding affinity for CACCT and CACCTG sequences. As shown in Fig. 2B, the p73 gene intronic fragment contains both types of sequences, indicated by numbers 1 to 6. To investigate whether transcriptional repressor ZEB plays a role in the control of p73 expression, we first mutated three of the above-mentioned binding sites (boxes 1, 3, and 5) that are present in the intronic sequence. As seen in Fig. 4, the mutation of these sites strongly releases the silencer activity of the first intron fragment on its p73 promoter.

View larger version (66K): [in this window] [in a new window]   FIG. 6.   ZEB binds to E boxes of the p73 gene intronic fragment. Gel shift assays were performed with probes resembling each of the E boxes of the p73 gene intronic fragment. (A) Cell extracts derived from proliferating C2C12 cells were incubated with the indicated probes. (B) ZEB binding to E box 5 was competed with a 200-fold molar excess of unlabeled probe but not with a 1,000-fold molar excess of an unrelated consensus (CCAAT). The supershifted complex is present upon addition of the anti-ZEB antibody to the binding reaction. (C and D) Cell extracts derived from differentiated C2C12 cells were incubated with the indicated probes. (E and F) Extracts of P19 and HL60 cells were incubated with the indicated probe. (G) In vitro-translated ZEB was incubated with the indicated probe. Specific and nonspecific competitions are indicated.

View larger version (32K): [in this window] [in a new window]   FIG. 7.   In vivo binding of ZEB to the first intron of the p73 gene in proliferating versus differentiating C2C12 cells. (A and B) Cross-linked chromatin from proliferating (top) and differentiating (12 and 72 h after serum withdrawal) (middle and bottom) C2C12 cells was immunoprecipitated with antibodies (Ab) to ZEB or NF-YB or in the absence of antibodies, and analyzed by PCR with primers specific for the indicated promoters (see Materials and Methods). Input, nonimmunoprecipitated cross-linked chromatin. Preim., preimmune; Prol., proliferating cells. (C) Extracts derived from the indicated cells were subjected to sodium dodecyl sulfate-10% polyacrylamide gel electrophoresis and immunoblotted with anti-ZEB (top) or with anti-p73 (middle) polyclonal sera. Equal loading of protein was measured by Coomassie staining (bottom).

View larger version (36K): [in this window] [in a new window]   FIG. 8.   ZEB binds to the first intron of p73 in vivo. Cross-linked chromatin from proliferating P19 and HL60 cells was immunoprecipitated with antibodies (Ab) to ZEB and NFY-B and crude antiserum or in the absence of antibodies and processed as for Fig. 7. Input, nonimmunoprecipitated cross-linked chromatin. Preim., preimmune; TK, thymidine kinase; CycB2, cyclin B2.

Overexpression of dominant-negative ZEB (ZEB-DB) restores p73 expression in proliferating C2C12 and P19 cells. To further investigate the involvement of ZEB in the regulation of p73 expression, we assessed whether interference with endogenous ZEB releases its negative regulation of p73 promoter activity. To this end we first verified whether increasing amounts of ZEB-DB (39) counteract the transcriptional repression of the p73 gene intronic fragment when transiently cotransfected with TK-p73intr-LUC in C2C12 cells. As shown in Fig. 9A the activity of the thymidine kinase promoter is partially restored with increasing amounts of ZEB-DB. By gel shift assay we verified that ZEB-DB from transiently transfected C2C12 cells was able to bind an oligonucleotide resembling boxes 3 and 4 of the p73 gene intronic fragment (Fig. 9C). The expression level of ZEB-DB from these cells is shown in Fig. 9B.

View larger version (26K): [in this window] [in a new window]   FIG. 9.   Ectopic expression of ZEB-DB releases transcription repression of ZEB and restores expression of p73 mRNA in proliferating C2C12 and P19 cells. (A) C2C12 cells were transiently transfected with the indicated plasmid combinations. Luciferase (Luc) and -galactosidase (gal) activities were measured as for Fig. 3. prot, protein. (B) Total cell extracts (100 µg/lane) derived from C2C12 and its derivatives transiently transfected with ZEB-DB were subjected to immunoblotting with the anti-Flag monoclonal antibody or with the anti- actin antibody for equal loading. Protein molecular sizes are shown on the left. WB, Western blotting. (C) Extracts derived from C2C12 cells transiently transfected with ZEB-DB were incubated with the indicated probe. The binding of ZEB-DB to E boxes 3 and 4 of the p73 gene intronic fragment was competed by a 200-fold molar excess of unlabeled probe but not by a 1,000-fold molar excess of an unrelated consensus probe (CCAAT). (D and E) Total RNA was extracted from the indicated cell lines and subjected to RT-PCR. Two sets of primers encompassing the coding sequence for the N terminus (N-ter) (42) and DNA-binding domain (DBD) of mouse p73 were employed. Amplification of aldolase A (Ald-A) was used to normalize equal loading of each RNA sample. The lengths of the amplified fragments are shown on the left. (F) Total-cell extracts (100 µg/lane) derived from C2C12 and P19 cells stably transfected with ZEB-DB were analyzed as reported for panel B.

View larger version (21K): [in this window] [in a new window]   FIG. 10.   Ectopic expression of MyoD restores p73 mRNA in proliferating C2C12 cells. These cells were transiently transfected with a vector encoding MyoD. The cells were harvested 36 h after transfection. Total RNA was extracted from the indicated cells and subjected to RT-PCR. A set of primers encompassing the coding sequence for the N terminus (N-ter) of mouse p73 was employed. Amplification of aldolase A (Ald-A) was used to normalize equal loading of each RNA sample. The length of amplified fragments is shown on the left.

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

Here we report that the first intron of the p73 gene is important for the regulation of p73 expression at the transcriptional level. We also show that transcriptional repressor ZEB, a vertebrate homologue of the Drosophila Zfh-1 protein (13, 18, 40, 44), is directly involved in the regulation of p73 expression. ZEB was originally isolated as a DNA binding protein. In fact it binds to a subset of E boxes, with highest affinity for CACCT and CACCTG sequences, through its N- and C-terminal zinc finger clusters (18, 45). ZEB exerts its repressor activity by directly binding to the consensus boxes present on target genes, unlike the action of other negative regulators, such as Id proteins and Twist, which bind and inactivate bHLH. It has previously been reported that members of MEF-2 family, which synergize with proteins such as MyoD, myogenin, Myf-5, and MRF-4 to induce muscle differentiation, can be targets for transcriptional repression by ZEB (39, 40). Our findings demonstrate in vivo the binding of ZEB to specific E boxes within the first intron of the p73 gene, and that implies that p73 is a specific target for transcriptional repression by ZEB.

Increasing evidence demonstrates that the p73 mRNA level is upregulated during cell differentiation (10, 55). Thus, p73 expression is tightly regulated at the transcriptional level. Of note, the observation that ZEB binds the first intron of the p73 gene to a greater extent in proliferating cells than in differentiating cells might provide one example of how p73 can be regulated differentially between proliferating and differentiating cells. Further evidence needs to be collected to verify whether the transcriptional repression of ZEB on p73 is cell type specific or altered by the state of the cell or modified in response to different stimuli that activate p73 (1, 19, 51, 60). Our results showing that upregulation of p73 mRNA follows a different kinetic in differentiating C2C12 cells than in P19 and HL60 cells suggest that release of transcriptional repression and maintenance of activation of p73 are tightly regulated. The output of this balance might be affected by cell context and by the type of differentiation.

The recent characterization of the p73 promoter has revealed the presence of at least three E2F binding sites (12). It has clearly been shown that p73 mRNA is strongly induced upon ectopic expression of E2F-1 (24, 32, 49). The E2F-1 transcription factor has also been reported to induce apoptosis through p53-independent pathways. For instance, induction of p73 by E2F-1 is also triggered by T-cell receptor-mediated apoptosis as shown by the reduction of the apoptotic rate upon introduction of a dominant-negative p73 (32). We show that the intronic fragment is able to markedly but not completely reduce the activity of its own p73 promoter upon E2F-1 induction, suggesting that transcriptional repression of ZEB might have a specific impact on p73 promoter activity.

The involvement of ZEB in the regulation of p73 gene expression is supported by the findings that overexpression of ZEB-DB restores p73 mRNA in proliferating C2C12 and P19 cells. This implies that competition with ZEB for the binding to E boxes of the p73 gene intronic fragment contributes in removing transcriptional repression of p73 promoter activity, at least in proliferating cells. It is reasonable to depict a scenario in which bHLH proteins accumulate, as occurs during muscle differentiation, and compete with ZEB in binding to the E boxes present in the first intron of the p73 gene, consequently releasing the transcriptional repression of the p73 promoter. We show that transient overexpression of MyoD restores p73 expression in proliferating C2C12 cells. This raises the possibility that it may act by competing with ZEB for binding to the first intron of the p73 gene, thereby allowing the induction of transcription of p73 at early stages of muscle differentiation. The timing of such competition could be specifically related to each bHLH protein as well as to the type of differentiation. We are currently investigating whether additional bHLH proteins cooperate with MyoD in controlling p73 at the transcriptional level. Moreover, we cannot exclude the possibility that ZEB functions as a direct repressor through its N-terminal repression domain (46). In addition, it has been shown that the impairment of the binding of ZEB to corepressors CtBP1 and CtBP2 strongly reduces its repressor activity (16, 41). ZEB is also widely expressed. This suggests that its control on p73 could be exerted in diverse tissues and in response to diverse differentiation stimuli. To clarify this issue further, evidence needs to be collected.