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

J. Biol. Chem., Vol. 275, Issue 27, 20488-20495, July 7, 2000

This Article Abstract Full Text (PDF) All Versions of this Article: 275/27/20488    most recent M909370199v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Janssens, V. Articles by Goris, J. Search for Related Content PubMed PubMed Citation Articles by Janssens, V. Articles by Goris, J. Social Bookmarking                      What's this?

The Phosphotyrosyl Phosphatase Activator Gene Is a Novel p53 Target Gene^*

Veerle Janssens, Christine Van Hoof, Ivo De Baere, Wilfried Merlevede, and Jozef Goris^§

From the Afdeling Biochemie, Faculteit Geneeskunde, Katholieke Universiteit Leuven, Herestraat 49, B-3000 Leuven, Belgium

Received for publication, November 18, 1999, and in revised form, April 27, 2000

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

The minimal promoter of the phosphotyrosyl phosphatase activator (PTPA) gene, encoding a regulator of protein phosphatase 2A contains two yin-yang 1 (YY1)-binding sites, positively regulating promoter activity. We now describe a role for p53 in the regulation of PTPA expression. Luciferase reporter assays in Saos-2 cells revealed that p53 could down-regulate PTPA promoter activity in a dose-dependent manner, whereas four different p53 mutants could not. The p53-responsive region mapped to the minimal promoter. Overexpression of YY1 reverses the repressive effect of p53, suggesting a functional antagonism between p53 and YY1. The latter does not involve competition for YY1 binding, but rather direct control of YY1 function. Inhibition of PTPA expression by endogenous p53 was demonstrated in UVB-irradiated HepG2 cells, both on the mRNA and protein level. Also basal PTPA levels are higher in p53-negative (Saos-2) versus p53-positive (HepG2, U2OS) cells, suggesting "latent" p53 can control PTPA expression as well. The higher PTPA levels in U2OS cells, programmed to overexpress constitutively a dominant-negative p53 mutant, corroborate this finding. Thus, PTPA expression is negatively regulated by p53 in normal conditions and in conditions where p53 is up-regulated, via an as yet unknown mechanism involving the negative control of YY1.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

The most frequent genetic alteration in human cancers is the mutational inactivation of the p53 tumor suppressor gene (1), indicating that loss of p53 activity plays an important role in human carcinogenesis. A number of DNA tumor virus oncoproteins also target p53, including the SV40 large T, adenovirus E1b and papillomavirus E6 proteins, which interact with and inactivate p53 through a variety of mechanisms (2). Moreover, mice deficient for p53 are developmentally normal but susceptible to an array of spontaneous tumors in early adult life (3). Based on these key observations, p53 is considered to be the prototype tumor suppressor.

The p53 protein, normally present at low levels in the cell, can be up-regulated by stimuli such as DNA damage, hypoxia, or ectopic expression of oncogenes (4). Upon activation, p53 diversely affects the expression of a variety of genes, resulting in the induction of pathways leading to either growth arrest or apoptosis (5-7). p53 can activate directly a subset of target genes, such as p21^waf (8), mdm2 (9, 10), and cyclin G (11), through binding to specific sequences located in the promoters of these genes. These specific p53-binding sites typically contain two copies of a 10-base pair motif that is degenerate in eight out of 10 bases and conforms to the sequence, 5'-RRRC(A/T)(T/A)GYYY-3' (12).

However, it has also been reported that p53 can repress the transcription of a large number of genes, which do not have these consensus p53-binding sites within their regulatory regions (13). The exact mechanism underlying this p53-dependent inhibition is presently unclear, but it is thought to be the consequence of p53 inhibiting transcriptional activators (14-18) or components of the basal transcription machinery (19-21). This phenomenon, which seemed to be specific for genes with TATA box elements (22), has recently also been demonstrated in TATA-less promoters (23-25).

Originally, the phosphotyrosyl phosphatase activator (PTPA)¹ was isolated as an in vitro regulator of protein phosphatase 2A (PP2A), one of the major Ser/Thr phosphatases implicated in cell cycle control, responses to signaling, cell differentiation, and cell transformation (26-28). Via an as yet unknown mechanism, PTPA can specifically up-regulate the phosphotyrosyl phosphatase (PTPase) activity of PP2A in the presence of ATP/Mg²⁺ (26, 29, 30). For this activation some highly conserved amino acids seem to be required (31). PTPA has been cloned from rabbit and human cDNA libraries (32) and homologues have been found in Xenopus laevis, Drosophila melanogaster, Saccharomyces cerevisiae, and Schizosaccharomyces pombe (31), indicating a well conserved function for this protein throughout evolution.

Whether the up-regulation of the PTPase activity of PP2A represents the actual in vivo function of PTPA remains elusive. Nevertheless, by a genetic approach in S. cerevisiae, we could identify a genetic interaction between PP2A and PTPA, indicating that PTPA could regulate at least some functions of PP2A.² A recent report by Ramotar et al. (33) shows that in S. cerevisiae, deletion mutants of PTPA display hypersensitivity to oxidative DNA damage, implying a function for PTPA in the pathway(s) signaling the repair or in the repair mechanism itself of this type of DNA damage.

After cloning and characterizing the structure of the human PTPA gene (34), we recently analyzed its promoter region in more detail (35). The minimal promoter, residing in an unmethylated CpG island, was identified between nt 67 and +38 and lacks a TATA box. Within this region, two functional yin-yang 1 (YY1)-binding sites are present. Upon mutation of these sites, promoter activity is completely abolished, implying a positive role for YY1 in the basal and ubiquitous expression of PTPA (35).

Interestingly, in the complete promoter region that was isolated and sequenced (about 3 kilobases), several putative p53 consensus binding sites are present. Given the p53 well defined role in the cellular response to DNA damage (4) and the proposed role, at least in budding yeast, for PTPA in DNA repair, we now evaluated the putative role of p53 on the promoter activity and expression of PTPA. The data obtained classify PTPA as a novel p53 target gene, subject to p53 repression, and open the gate to further investigate the in vivo role of PTPA, particularly in p53-dependent cell cycle arrest and apoptosis.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Plasmids-- The generation of luciferase reporter gene constructs, containing different PTPA promoter fragments is described in detail in Ref. 35. CONp53-pLuc, that was used as a positive control, was generated by ligating two complementary synthetic oligonucleotides 5'-AGCTTAGACATGCCTAGACATGCCT-3' and 5'-AGCTAGGCATGTCTAGGCATGTCTA-3', which were first annealed and phosphorylated by T4 polynucleotide kinase (Amersham Pharmacia Biotech), in the HindIII restriction site of the pGL2-Promoter vector (Promega). Upon sequencing of the insert, two copies of this oligonucleotide (and thus of the p53 consensus sequence) appeared to be present.

To introduce point mutations in the putative p53-binding sites in the context of 102/+38-pLuc, five pairs of complementary and partially overlapping synthetic DNA oligonucleotides, some containing the desired mutations (Table I), were phosphorylated by T4 polynucleotide kinase, denatured, annealed, and ligated. The resulting fragments were then ligated into pGL2-Basic digested with NheI/KpnI and named 102/+38 p53(u)mut-pLuc, 102/+38 p53(d)mut-pLuc, and 102/+38 p53(u,d)mut-pLuc (in which either the upstream, the downstream or both p53 sites are mutated).

View this table: [in this window] [in a new window]   Table I Sequences of oligonucleotides used to generate p53-binding site mutations in the context of the 102/+38 PTPA promoter Combinations of oligonucleotides 1b, 2, 3b, 4b, 5, 6, 7a, 8a, 9, and 10a generated 102/+38 p53(u)mut-pLuc; 1a, 2, 3a, 4a, 5, 6, 7b, 8b, 9, and 10b generated 102/+38 p53(d)mut-pLuc; and 1b, 2, 3b, 4b, 5, 6, 7b, 8b, 9, and 10b generated 102/+38 p53(u,d)mut-pLuc.

To generate YY1(u)-TATA-pLuc, containing the YY1 motif at nt 52/44 of the minimal PTPA promoter (35) upstream of the proximal Tk promoter, two partially overlapping oligonucleotides (5'-CTGACATGGCCG-3' and 5'-CTAGCGGCCATGTCAGAGCT-3') were phosphorylated by T4 polynucleotide kinase, annealed, and ligated into the SacI/NheI sites of a pGL3-Basic-derived plasmid, in which expression of the luciferase gene is driven by the proximal Herpes simplex virus Tk promoter (nt 47/+54) as described (36, 37). Sequencing confirmed the presence of one copy of the YY1(u) motif.

All p53-containing constructs, used for co-transfection, were derived from pSP65hp53, a plasmid containing the complete human p53 cDNA in the pSP65 vector (a kind gift of C. Desaintes (Institut Pasteur, Paris, France)). For the construction of pSG5-p53, the p53 cDNA was PCR amplified from pSP65hp53 with PWO polymerase (Roche Molecular Biochemicals) using the primers 5'-TAGAATTCAAGCTTATGGAGGAGCCGCAGTCAGATC-3' and 5'-AATGGATCCTCTAGATCAGTCTGAGTCAGGCCCTTC-3', and the resulting PCR product cloned in the EcoRI and BamHI restriction sites of pSG5 (Stratagene). pSG5-C p53, encoding amino acids 1-308 of p53, was generated similarly, using 5'-TAGAATTCAAGCTTATGGAGGAGCCGCAGTCAGATC-3' and 5'-ATATGGATCCTCACAGTGCTCGCTTAGTGCTC-3' as primers. pSG5-N p53, comprising amino acids 97-394 of p53, was generated with primers 5'-ATATCATATGGTCCCTTCCCAGAAAACCTA-3' and 5'-AATGGATCCTCTAGATCAGTCTGAGTCAGGCCCTTC-3', followed by cloning in the BamHI and the Klenow polymerase-treated EcoRI sites of pSG5. pSG5-p53 DBD, encoding the DNA-binding domain of p53 (amino acids 97-308), was generated similarly by using primers 5'-ATATCATATGGTCCCTTCCCAGAAAACCTA-3' and 5'-ATATGGATCCTCACAGTGCTCGCTTAGTGCTC-3'.

Site-directed mutagenesis to generate R273Lp53, a dominant negative p53 mutant, was performed using a standard PCR-based method. Two separate PCRs were performed with pSP65hp53 as template: the first with 5'-TAGAATTCAAGCTTATGGAGGAGCCGCAGTCAGATC-3' (start primer) and 5'-GCTTTGAGGTGCTTGTTTGTGCCTGTC-3' (reverse mutated primer), and the second with 5'-GACAGGCACAAACAAGCACCTCAAAGC-3' (forward mutated primer) and 5'-AATGGATCCTCTAGATCAGTCTGAGTCAGGCCCTTC-3' (stop primer). The combined reaction products of these two PCR reactions were then used as template for a second amplification round with the start and stop primers only. The resulting PCR product was cloned into the EcoRV-digested pBluescript vector (Stratagene) and sequenced to confirm the introduction of the mutation. Next, R273Lp53 was cloned into the EcoRI and BamHI sites of pSG5, and into the HindIII and XbaI sites of pRC-CMV (Invitrogen) to generate the pSG5-p53 R273L and pCMV-p53 R273L expression plasmids, respectively.

Cell Cultures and Transient Transfection Assays-- All cell lines were purchased from the American Type Culture Collection (Rockville, MD). HepG2 cells (human liver hepatoma) and U2OS cells (human osteosarcoma), both containing wild-type p53 and Rb (38, 39) were cultured at 37 °C and 5% CO₂ in Dulbecco's modified Eagle's medium (Life Technologies, Inc.), supplemented with 10% fetal calf serum (Life Technologies, Inc.), 2 mM L-glutamine (Life Technologies, Inc.), 100 units/ml penicillin, and 100 mg/ml streptomycin (Life Technologies, Inc.). p53-negative (39) Saos-2 cells (human osteosarcoma) were cultured in McCoy's 5A medium (Life Technologies, Inc.) supplemented with 15% fetal calf serum, 100 units/ml penicillin, and 100 mg/ml streptomycin, at 37 °C and 5% CO₂.

Transient transfections were performed in 24-well dishes using Fugene reagent (Roche Molecular Biochemicals) with 0.4 µg of reporter construct and 50 ng of the p53 expression plasmid (either the wild-type form or the different mutant forms), unless stated otherwise. To evaluate transfection efficiencies, cells were always co-transfected with a reporter vector containing the -galactosidase gene driven by the cytomegalovirus promoter (pCMV--Gal) or by the elongation factor 1 promoter (pEF1--Gal). The amount of this co-transfected DNA was 1/10 of the total DNA used in the transfection. Whenever transfection efficiencies were evaluated between different cell lines, cells were treated with equal amounts of the same DNA-Fugene mixture and from then on treated equally and measured simultaneously. 24-48 h after transfection, cells were harvested in 100 µl of passive lysis buffer (Promega). Protein concentrations were measured by the BCA system (Pierce). After concentration adjustment, luciferase activity was measured with the Luciferase Assay System (Promega) according to the manufacturer's protocol using the MicroLumat LB96P (Berthold) with automated injection. Reagents for measurement of -galactosidase activity were obtained from Tropix (Westburg). For each construct at least two different DNA preparations (QIAGEN) were tested and each transfection was repeated at least three times.

Cell Treatments-- For UV irradiation, three parallel lamps emitting predominantly within the UVB range (280-320 nm) with an emission peak at 313 nm were used. Prior to UVB irradiation, the culture medium was removed, and cells were washed with pre-warmed phosphate-buffered saline (PBS). Under a thin film of PBS the cells were then irradiated for 52 s (which corresponds to a dose of 20-25 J/m²) and the old medium was replaced.

Protein Extract Preparation and Western Blotting-- Cells (10 cm plate) were washed twice with ice-cold PBS and scraped into 200 µl of ice-cold NET buffer (50 mM Tris-HCl, pH 7.4, 150 mM NaCl, 15 mM EDTA, 1% Nonidet P-40, 2 mM benzamidine, 0.5 mM phenylmethylsulfonyl fluoride, 1.5 µg/ml pepstatin, 1 µg/ml leupeptin, 0.1 µM microcystin). After incubation on ice for 10 min, the suspension was vortexed for 20 s and centrifuged in a microcentrifuge at 13,000 rpm for 20 min at 4 °C. Soluble proteins were frozen in liquid nitrogen and stored at 80 °C. Protein concentrations of cell lysates were estimated by BCA (Pierce).

After concentration adjustment protein samples were resolved on 12% SDS-PAGE gels and transferred to polyvinylidene difluoride membranes (Bio-Rad) by semi-dry blotting. The membranes were blocked in 5% non-fat dry milk in PBS-T (0.1% Tween 20 in PBS) for 1 h at room temperature. Incubation with the primary antibody was performed in immunoassay buffer (1% bovine serum albumin, 0.4 mM NaOH, 0.02% NaN₃ in PBS) overnight at 4 °C. The monoclonal anti-p53 antibody (clone DO-7) was purchased from DAKO; anti-PTPA (40), and anti-PP2A_CAT (anti-DYFL) (41) antibodies were generated as described and affinity purified before use. After washing in PBS-T with 5% non-fat dry milk, the membranes were incubated for 1 h with horseradish peroxidase-conjugated secondary antibodies (DAKO) at room temperature. To visualize the bands, the blots were developed using an enhanced chemiluminescence detection system (ECL, Amersham Pharmacia Biotech).

RNA Preparation and Northern Blotting-- Total RNA from treated and non-treated cell lines was isolated using the guanidinium-thiocyanate acid-phenol procedure (42). 20 µg of total RNA was fractionated on 1% agarose gels with 1% formaldehyde in 1 × NBC buffer (50 mM boric acid, 1 mM sodium citrate, and approximately 5 mM NaOH to make pH 7.5). Subsequently, the RNA was transferred to Hybond-N nylon membranes (Amersham Pharmacia Biotech) by capillary blotting. After UV cross-linking, blots were pre-hybridized (at least 4 h) and hybridized (overnight) at 42 °C in the presence of 50% formamide, 0.5% SDS, 50 mM NaH₂PO₄, pH 6.5, 5 × SSC (3 M NaCl, 0.3 M sodium citrate), 5 × Denhardt's solution, and 1 mg/ml denatured salmon sperm DNA. Hybridization was performed with a PTPA-specific probe, containing the full-length PTPA cDNA, or with an 18 S ribosomal RNA probe, at a specific concentration of 1 × 10⁶ cpm/ml. Probes were labeled by random priming (random primed labeling kit, Roche Molecular Biochemicals; [-³²P]dATP, Amersham Pharmacia Biotech). Membranes were washed twice for 15 min at room temperature and once for 5 min at 60 °C in 2 × SSC, 0.1% SDS.

Generation of Cell Lines Constitutively Expressing the R273L p53 Mutant-- For stable transfections, cells were transfected at 80% confluency with 5 µg of either pRC-CMV or pCMV-p53 R273L, using Fugene Reagent (Roche Molecular Biochemicals). Following transfection, the cells were allowed to grow for 48 h. On day 3 cells were split 1:3 and placed in normal growth medium, supplemented with 700 µg/ml of the neomycin analog G418 (Life Technologies). Fresh selection medium was applied every 3 or 4 days. After 2-3 weeks, neomycin-resistant individual colonies could be isolated. To exclude clone-specific differences, experiments were performed with a mixture of >20 stable clones.

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

The PTPA Gene Promoter Is Repressed by Wild-type p53 in Transient Transfection Assays in Saos-2 Cells-- To assess the putative influence of p53 on PTPA promoter activity, we performed transient transfections in p53-negative Saos-2 cells. The luciferase reporter construct, containing the longest PTPA promoter fragment available to us and harboring at least three putative p53-binding sites (2356/+38-pLuc) was co-transfected with pSG5-p53 or with empty pSG5 (which served as a control to keep total amounts of transfected DNA constant). Upon co-transfection of pSG5-p53 a clear and reproducible repression of 2356/+38-pLuc activity could be observed (Fig. 1A, lane 1).

View larger version (20K): [in this window] [in a new window]   Fig. 1.   Wild-type p53 down-regulates PTPA promoter activity in transient transfection assays in Saos-2 cells. A, mapping of the p53-responsive region. Transfections were performed with 0.4 µg of 2356/+38-pLuc, 1565/+38-pLuc, 648/+38-pLuc, 394/+38-pLuc, and 67/+38-pLuc, containing different PTPA promoter fragments, which have their 3' ends at +38 and their 5' ends at 2356, 1565, 648, 394, or 67 nt, relative to the transcription initiation site. CONp53-pLuc is a positive control, containing two copies of a well known p53-binding site in front of an SV40 promoter-driven luciferase gene. Negative controls include the empty luciferase vector (pGL2-Basic) and a CMV promoter-driven luciferase vector (CMV-pLuc). All these luciferase plasmids were co-transfected with 50 ng of pSG5-p53 or with 50 ng of pSG5 alone to keep total amounts of DNA invariable. Transfection efficiencies were evaluated as described under "Experimental Procedures." After adjustment of the luciferase activities to the mean value of all -galactosidase activities, induction factors were calculated as the ratios of the corrected luciferase activities of pSG5-p53-co-transfected over pSG5-co-transfected reporter plasmids, and represented on a logarithmic scale. Standard errors of the mean of at least three independent experiments are indicated. B, dose dependence of the repression of PTPA promoter activity by wild-type p53. 0.4 µg of 67/+38-pLuc, containing the minimal PTPA promoter, or 0.4 µg of CMV-pLuc was co-transfected with varying amounts of empty pSG5 or pSG5-p53, ranging from 250 ng to as little as 80 pg. Induction factors are calculated as in A. Standard errors of the mean of three independent experiments are indicated.

In order to eliminate the possibility that in the transfected cells p53 was having a negative influence on gene transcription in general, we also analyzed the effects of p53 expression on the luciferase expression of two control promoters under the same experimental circumstances (Fig. 1A). No significant difference was seen on the activity of CMV-pLuc (lane 3), whereas introduction of a previously characterized p53 DNA binding sequence (CONp53) (12) upstream of the luciferase gene in the pGL2-Promoter vector, promoted a strong transactivation by p53 (lane 2). In none of the cases was any effect of p53 seen on the empty pGL2-Basic vector (lane 8). Thus, the observed inhibition of PTPA promoter activity by p53 seemed to be specific and did not result from a negative effect of p53 on transcription in general.

The p53-responsive Region Maps to the Minimal PTPA Promoter-- To delineate the p53-responsive region within the PTPA promoter, we performed similar co-transfection assays with several truncated PTPA promoter reporter constructs (Fig. 1A, lanes 4-7). The results indicate that p53 apparently inhibits the minimal PTPA promoter (located between nt 67 and +38) (35), since all reporter constructs still displaying promoter activity, were similarly repressed by expression of p53. This result is not surprising, since for most gene promoters known to be repressed by p53, the repression is situated at the minimal promoter level (43, 44). As such the PTPA promoter may constitute another TATA-less promoter, subject to p53-repression.

p53 Inhibits the Minimal PTPA Promoter in a Dose-dependent Manner-- The degree of repression of the minimal PTPA promoter by p53 strongly depends upon the amount of co-transfected pSG5-p53 DNA (Fig. 1B). Strong inhibition (90%) was reached at 250 ng of pSG5-p53 DNA, whereas 10 ng of pSG5-p53 retained only 40% inhibition. As little as 2 ng of pSG5-p53 failed to inhibit PTPA promoter activity. Similarly, transactivation of CONp53-pLuc was progressively reduced, when co-transfected pSG5-p53 levels were progressively diminished (results not shown). No effect on the activity of CMV-pLuc was observed at any of the pSG5-p53 concentrations used for the co-transfection (Fig. 1B), which indicates that even the p53 concentrations resulting in 90% suppression of PTPA promoter activity were without effect on transcription in general.

p53 Mutants Fail to Repress PTPA Promoter Activity-- We next generated four p53 mutants and examined whether they still retained the ability to repress PTPA promoter activity. p53C, p53N, and p53DBD are deletion mutants that either lack the C terminus (amino acids 309-393), the N terminus (amino acids 1-96), or both. R273Lp53 is a dominant-negative p53 mutant (45), commonly found in human cancers and containing a point mutation in the DNA-binding domain. As shown in Fig. 2, co-transfection of none of these mutants resulted in any down-regulation of minimal PTPA promoter activity, although they were all properly expressed and transcriptionally "active," as measured by their ability to compete with wild-type p53 for transactivation of CONp53-pLuc (not shown). These results suggest that sequences of both the C and the N terminus of p53 are required to mediate repression.

View larger version (41K): [in this window] [in a new window]   Fig. 2.   Loss of PTPA promoter repression by p53 mutants. Saos-2 cells were transfected with 0.4 µg of 67/+38-pLuc, together with 50 ng of pSG5-p53, pSG5-N p53, pSG5-C p53, pSG5-p53 DBD, pSG5-p53 R273L, or the empty pSG5. The generation of the various p53 mutants is described in detail under "Experimental Procedures." Induction factors were calculated as in Fig. 1A. Standard errors of the mean of three independent experiments are indicated.

Exposure of p53-positive HepG2 Cells to UVB Light Induces Accumulation of p53 and Down-regulates PTPA Expression-- HepG2 cells were exposed to UVB irradiation (20-25 J/m²) to activate endogenous p53. Subsequently, the levels of PTPA were monitored by Northern and Western blot analysis over a period of 48 h. The results of the Western blot (Fig. 3A) clearly indicate an inverse relationship between p53 and PTPA protein levels: whereas p53 accumulated upon UVB treatment, PTPA expression was reduced progressively. However, the levels of the PP2A catalytic subunit remained unaffected. The p53 dependence of the observed effect was demonstrated in p53-negative Saos-2 cells, in which PTPA protein levels remained unaltered after UVB irradiation (Fig. 3B).

View larger version (65K): [in this window] [in a new window]   Fig. 3.   UVB irradiation causes a p53-dependent down-regulation of endogenous PTPA protein. A, Western blot of extracts of UVB-irradiated HepG2 cells. Protein extracts were collected 0, 1, 2, 4, 6, 8, 12, 24, and 48 h following treatment of HepG2 cells with 20-25 J/m2 of UVB irradiation. After concentration adjustment, samples of equal protein concentration were resolved on SDS-PAGE gels and subjected to Western analysis. The blot was subsequently developed with anti-PTPA, anti-p53, and anti-PP2ACAT-specific antibodies. B, Western blot of extracts of UVB-irradiated p53-negative Saos-2 cells. The same experiment as in A was now repeated with Saos-2 cells. The blot was developed with anti-PTPA antibodies.

The results of the Northern blot (Fig. 4) show that the decrease in PTPA protein expression is likely mediated by a concomitant down-regulation of PTPA mRNA, since the expression of both the 4.1- and 2.8-kilobase PTPA transcripts was reduced, following UVB irradiation.

View larger version (39K): [in this window] [in a new window]   Fig. 4.   Repression of endogenous PTPA mRNA by activation of p53 following UVB irradiation of HepG2 cells. Total RNA extracts were prepared 0, 1, 2, 4, 6, 8, 12, 24, and 48 h after UVB treatment of HepG2 cells (20-25 J/m2), resolved on a 1% agarose gel, and subjected to Northern analysis. The blot was first developed with a radioactively labeled PTPA-specific probe, encompassing the complete PTPA cDNA. After dehybridization, it was re-probed with an 18 S ribosomal RNA-specific probe, which served as an internal control. kb, kilobase(s).

PTPA Is Differentially Expressed in p53-positive (HepG2 and U2OS) and p53-negative (Saos-2) Cells-- While performing our transfection experiments, we were struck by the relatively high PTPA promoter activities observed in Saos-2 cells. Therefore, we made a quantitative comparison of PTPA promoter activities in Saos-2, U2OS, and HepG2 cells. In order to do so, a CMV promoter-driven or an EF-1 promoter-driven -galactosidase expression plasmid was co-transfected with 67/+38-pLuc and used as an internal control to correct for differences in transfection efficiency between the different cell lines. By doing so, we assumed that CMV promoter and EF-1 promoter activities did not significantly differ between the various cell lines. The results indicate that in p53-negative Saos-2 cells, PTPA promoter activities are indeed significantly higher than in p53-positive U2OS and HepG2 cells (Fig. 5A). The difference is especially striking between both osteosarcoma cell lines (Saos-2 and U2OS). Moreover, these results could be confirmed by Western blotting with PTPA-specific antibodies (Fig. 5B).

View larger version (27K): [in this window] [in a new window]   Fig. 5.   Latent p53 affects basal PTPA expression. A, comparison of luciferase activities of 67/+38-pLuc in U2OS, Saos-2, and HepG2 cells. To correct for differences in transfection efficiencies between the various cell lines, 0.4 µg of 67/+38-pLuc was co-transfected with 0.04 µg of pCMV--Gal. The corrected promoter activities of 67/+38-pLuc in U2OS and HepG2 cells are given relative to the corrected luciferase activity of the same construct in Saos-2 cells, which was set to 100. Similar results were obtained when pEF1--Gal was used as an internal control, instead of pCMV--Gal. B, quantitative comparison of PTPA protein levels in U2OS, Saos-2, and HepG2 cells. Equal amounts of total protein extracts of U2OS, Saos-2, and HepG2 cells were resolved on SDS-PAGE gels and subjected to Western analysis with PTPA-specific antibodies. C, expression of R273Lp53 in wild-type p53-containing U2OS cells results in higher basal PTPA levels. U2OS cells were stably transfected with pRC-CMV (lane 3) or with pCMV-p53 R273L (lane 4). Protein extracts were prepared from these cell lines and from Saos-2 (lane 1) and HepG2 cells (lane 2). Equal protein amounts were subsequently analyzed by Western blotting with PTPA- and p53-specific antibodies.

PTPA Expression Is Enhanced in U2OS Cells Constitutively Overexpressing the R273L Dominant-negative p53 Mutant-- To further substantiate the former results and to check whether the differences in basal PTPA expression could indeed be linked to the p53 status of the cells, we looked for ways to inhibit endogenous p53 expression and/or activity in U2OS cells, in order to examine their effects on basal PTPA expression.

Expression of an antisense p53 did not inhibit p53 expression (not shown). Therefore, we constructed a pool of U2OS cell lines in which the pCMV-p53 R273L plasmid was stably integrated into the genome and thus constitutively overexpressed the dominant-negative R273Lp53 mutant. In these cells PTPA expression was strikingly higher, compared with the PTPA levels in pRC-CMV transfected U2OS cells (Fig. 5C). This result confirms that the dominant-negative R273Lp53 mutant could indeed compete out the effect of wild-type p53, normally present in U2OS cells, resulting in a higher concentration of PTPA, up to the level found in Saos-2 cells where no p53 is present. Moreover, development of the same blot with p53-specific antibodies revealed higher wild-type p53 levels in U2OS, as compared with HepG2 cells (Fig. 5C), which may provide an explanation for the more pronounced difference in PTPA levels between Saos-2 and U2OS cells, than between Saos-2 and HepG2 cells (Fig. 5, B and C).

Functional Antagonism between p53 and YY1-- Since the minimal PTPA promoter is repressed by p53 and knowing from our previous results (35) that YY1 positively regulates basal promoter activity, it seemed logical to assess: 1) whether co-transfection of YY1 could influence the p53-mediated promoter repression and 2) whether p53 could repress the YY1-mediated induction of promoter activity.

Therefore, co-transfections were performed with 67/+38-pLuc and varying amounts of either pSG5-p53 or pCMV-YY1, or both (Fig. 6). The results indicate that co-transfection of pCMV-YY1 (in all the concentrations used) can overcome the p53-mediated repression of 67/+38-pLuc activity (Fig. 6A, compare lane 1 with lanes 2-5). Lanes 6-9 indicate that co-transfection with these specific concentrations of pCMV-YY1 alone, transactivate 67/+38-pLuc. In Fig. 6B the reverse effects are shown. Here, the results indicate that pSG5-p53 can only repress YY1 transactivation of 67/+38-pLuc in those concentrations that are able to repress the basal activity of 67/+38-pLuc. Together, the data suggest an antagonism between YY1 and p53 in the context of the minimal PTPA promoter, consistent with each acting independently in an antagonistic fashion, or with p53 acting by directly affecting YY1.

View larger version (25K): [in this window] [in a new window]   Fig. 6.   Cross-talk between p53 and YY1. A, rescue of p53-mediated repression of PTPA promoter activity by YY1 overexpression. 0.4 µg of 67/+38-pLuc was co-transfected with 50 ng of pSG5-p53, with (lanes 2-5) or without (lane 1) varying amounts of pCMV-YY1 (100, 50, 25, and 12.5 ng, respectively). As a control, 0.4 µg of 67/+38-pLuc was also co-transfected with these same amounts of pCMV-YY1 alone (lanes 6-9). Total amounts of transfected DNA were kept constant by addition of empty pSG5 and/or pCMV. The data of one representative experiment are shown. B, blockage of YY1-induced activation of the PTPA promoter by wild-type p53. 0.4 µg of 67/+38-pLuc was co-transfected with 50 ng of pCMV-YY1, with (lanes 2-5) or without (lane 1) varying amounts of pSG5-p53 (0.4, 2, 10, and 50 ng, respectively). As a control, the same luciferase plasmid was also co-transfected with the same amounts of pSG5-p53 alone (lanes 6-9). Total amounts of transfected DNA were kept constant by addition of empty pCMV and/or pSG5. The results of one representative experiment are shown.

The Mechanism of the p53-mediated Repression of PTPA Promoter Activity Does Not Involve Direct Binding of p53 to the Minimal Promoter-- Simple sequence analysis of the PTPA minimal promoter reveals the presence of two nearly perfect p53-binding sites: one between nt 79 and 44 (5'-GCGCATGCGC (16 base pairs) TGACATGGCC3'), termed p53(u) and another between nt 1 and +35 (5'-AGGCTTGCTC (16 base pairs) CGACATGGCG3'), termed p53(d) (Fig. 7A). Interestingly, the 3' half-site of each p53 consensus site coincides with the formerly identified YY1-binding sites at nt 52/44 and nt +27/+35 (35), suggesting a competition might exist between p53 and YY1 for the same binding sites within the PTPA minimal promoter (Fig. 7A). However, several arguments argue against the functionality of these p53 consensus motifs and therefore exclude the possibility of competition between p53 and YY1 for overlapping binding sites as a mechanism for the observed functional antagonism between these transcription factors.

View larger version (37K): [in this window] [in a new window]   Fig. 7.   Functional analysis of the putative p53 binding motifs in the proximal part of the PTPA promoter. A, detailed part of the PTPA promoter sequence. The sequence of two regions within the proximal part of the PTPA promoter, encompassing the putative upstream p53-binding site, p53(u), and the putative downstream p53-binding site, p53(d), is depicted. Each p53 half-site is boxed and mismatches with the consensus p53 binding motif are depicted by x. The arrows indicate the point mutations introduced in the cores of the 5' half-sites of p53(u) and p53(d). The functional YY1 binding motifs, YY1(u) and YY1(d) (35), are underlined. B, the p53 binding motifs are not required for the p53-mediated repression. Transient transfections were performed in Saos-2 cells with the constructs indicated on the left: 102/+38-pLuc is the wild-type construct, whereas 102/+38 p53(u)mut-pLuc, 102/+38 p53(d)mut-pLuc, and 102/+38 p53(u,d)mut-pLuc contain two point mutations in the 5' half-site of the upstream, the downstream, or both putative p53 binding motifs. Each construct was co-transfected with 50 ng of pSG5-p53 or with 50 ng of empty pSG5. Induction factors are calculated as the ratios of the corrected luciferase activities of pSG5-p53-co-transfected over pSG5-co-transfected reporter plasmids. The standard error of the mean of three independent experiments is indicated.

First, the minimal promoter (67/+38) is missing the 5' half of the potential p53(u) site, which itself overlaps with the YY1 site (at nt 52/44) having the major effect on PTPA promoter activity (35). Nevertheless, the activity of this minimal promoter is equally well suppressed by co-expressing p53 as the 394/+38 promoter (and the longer reporter constructs) which contain the complete p53(u) (Fig. 1A, compare lane 7 with lanes 1, 4, 5, and 6). These data therefore exclude direct binding of p53 to the putative p53(u) site.

Second, we introduced site-directed mutations in either or both of the 5' half-sites of each putative p53 binding motif in the context of 102/+38-pLuc: in p53(u) ⁷⁹GCGCATGCGC⁷⁰ was mutated into ⁷⁹-GCGAATTCGC⁷⁰ and in p53(d) ¹AGGCTTGCTC⁺⁹ was mutated into ¹AGGATTTCTC⁺⁹ (Fig. 7A). These mutations destroy the p53 binding core and would therefore theoretically abolish p53 binding (12). Introducing mutations in the 3' half-sites of each p53 site, which coincide with both functional YY1 motifs (Fig. 7A) was not possible, since these would also drastically affect promoter activity, both in the minimal promoter (35) and a longer promoter context.³ Co-transfection of pSG5-p53 with these mutated promoter constructs indicated that p53 can equally well suppress the expression of the mutated promoter plasmids as the wild-type promoter (Fig. 7B). Therefore, the data suggest that the p53 motifs are not required for the repression.

Third, and consistent with the former data, we could not obtain evidence for direct binding of p53 to different PTPA promoter-specific probes, encompassing p53(u) and/or p53(d), in gel retardation experiments (results not shown). Together, these results indicate that the suppressive effect of p53 on PTPA promoter activity is neither mediated by direct binding to the promoter, nor by direct competition with YY1 binding to this promoter.

p53 Directly Affects YY1 Function-- The functional antagonism observed between p53 and YY1 might suggest that p53 acts directly through YY1. To test this hypothesis, the major YY1 motif present in the minimal PTPA promoter, YY1(u) (35), was inserted upstream of a heterologous promoter (TATA-pLuc, comprising nt 47/+54 of the Tk promoter and a functional TATA box), driving luciferase expression. The resulting plasmid, YY1(u)-TATA-pLuc, was subsequently transfected into Saos-2 cells, either in the absence or the presence of varying amounts of pSG5-p53 (Fig. 8). Luciferase activity measurements indicate that in the absence of p53, YY1(u) clearly stimulates transcription from the Tk-TATA box (lane 1), and this stimulation can be suppressed by p53 co-expression in a dose-dependent manner (lanes 2-5). Whereas the activity of the Tk-TATA box alone is only slightly affected by the highest dose of p53, the activity of YY1(u)-TATA-pLuc is clearly diminished by increasing p53 amounts (lanes 2-5). These data suggest that the sequence requirements for YY1-mediated activation are the same as for p53-mediated suppression. Therefore, the suppressive effect of p53 on PTPA expression likely occurs via direct inhibition of YY1 function, and not via a hypothetical other factor binding to the PTPA minimal promoter. This inhibition of YY1 function could be accomplished through direct interaction with the YY1 protein, or more probably, via independent regulation of transcriptional co-activators or co-integrators involved in YY1-mediated PTPA transcription (see also "Discussion").

View larger version (36K): [in this window] [in a new window]   Fig. 8.   YY1(u)-mediated stimulation of transcription from the heterologous Tk-TATA promoter is suppressed by co-expression of p53. Saos-2 cells were transfected with 0.4 µg of Tk-TATA-pLuc (comprising nt 47/+54 of the herpes simplex virus Tk promoter driving luciferase expression) or of YY1(u)-TATA-pLuc (containing the activatory PTPA YY1(u) motif upstream of Tk-TATA), either in the absence (lane 1) or presence of increasing amounts of pSG5-p53 (30 to 240 ng) (lanes 2-5). To correct for varying transfection efficiencies, luciferase values were normalized to the average -galactosidase value of all samples, and are represented as arbitrary units. The results of one representative experiment are indicated.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

In the present study, we demonstrate that p53 represses transcription from the human PTPA promoter, which contains no consensus TATA box element in its core promoter and is positively regulated by two YY1-binding sites (35). This repression of PTPA expression by p53 is physiologically significant, since induction of functional p53 by UVB irradiation of HepG2 cells leads to down-regulation of endogenous PTPA mRNA and protein levels.

We also provide evidence for the negative regulation of PTPA expression by non-activated p53. Indeed, PTPA expression was strikingly higher in p53-negative Saos-2 cells compared with p53-positive U2OS cells. Both of these cell lines originate from osteosarcoma tissue, so that tissue-specific factors are likely not accounting for these quantitative differences. Moreover, inhibiting the wild-type p53 function in U2OS cells by overexpression of the dominant-negative R273Lp53 mutant results in an obvious up-regulation of PTPA levels. To our knowledge, this is the first report showing that "latent" p53 can negatively control the basal transcription of a target gene. In the same context, Tang et al. (46) already described a role for p53, regulating positively the basal level of expression of p21^waf in the absence of treatment with DNA-damaging agents. Thus, they demonstrated that p53 is capable of sequence-specific DNA binding and transcriptional activation in untreated, proliferating cells. These observations, together with our results, suggest that latent p53 is not a biologically dead protein.

In order to change the level of a protein rapidly by controlling the rate of transcription, a short-lived mRNA or protein is essential. We have no data on the half-life of the PTPA messenger, but the PTPA protein seems to be rather stable: both in HepG2 and Saos-2 cells its half-life is estimated to be about 24 h, as measured through in vivo labeling experiments.³ This has important implications for our results and could be indicative for the existence of additional levels of PTPA regulation, following UVB irradiation, affecting the stability of the protein. So far, however, experiments designed to provide more insight into this topic were not conclusive. Meanwhile, the long half-life of PTPA explains why the protein level of PTPA follows only slowly the change in mRNA. When the time element is eliminated, as in the analysis of PTPA expression in the different cell lines, promoter activity and protein levels correlate nicely and show an inverse correlation of p53 and PTPA.

As for the underlying molecular mechanism of the p53-mediated repression of PTPA transcription, it is unlikely that it involves direct binding of p53 to the minimal promoter, since 1) mutations in the putative p53-binding sites, theoretically abolishing p53 binding, did not affect the ability of p53 to repress promoter activity in transient transfection assays, and 2) no direct p53 binding to the minimal promoter could be demonstrated in band shift assays. As such, the data are in line with the observation that sequence-specific binding by p53 is usually associated with transcriptional activation (4, 8-11), although a sole paper by Lee et al. (47) recently reported p53 repression of the -fetoprotein gene by specific DNA binding. Moreover, the absence of direct p53 binding also excludes the competition hypothesis, in which p53 might compete with YY1 for binding to the same sites in the PTPA promoter.

The functional antagonism observed between p53 and YY1 is most probably the basis for the p53-mediated repression of PTPA, since YY1 binding and intact YY1-binding sites in the minimal promoter are essential for PTPA expression (35). Moreover, p53 suppresses the stimulatory ability of YY1(u) toward a heterologous promoter, suggesting that p53 can directly act upon YY1. This might involve a physical YY1-p53 interaction, resulting in a modified YY1 function. However, so far we could not provide any direct evidence for this hypothesis, since neither by GST pull-down experiments, nor by co-immunoprecipitations of HA-tagged protein versions, such an interaction became evident.

As such, we favor another hypothesis in which a third player may be involved. A good candidate could be the p300/CBP family of transcriptional co-activators. p300 (15) as well as CBP (48-50) can form specific complexes with p53, resulting in the stimulation of transcription from p53-dependent promoters. Moreover, this interaction with p300/CBP is also required for p53-dependent trans-repression, in the sense that association of p53 with p300/CBP can interfere with co-activation of other p300/CBP-dependent factors, such as AP-1 (15), hypoxia-inducible factor 1 (16), and NF-B (18, 51, 52). Interestingly, CBP and p300 also physically interact with YY1, but their role in YY1 function, especially in YY1 transactivation, remains obscure, since it was suggested that the p300-YY1 interaction is merely involved in YY1-mediated repression or initiation, but not transactivation (53-55). Whether the functional interplay between p53 and YY1 in the PTPA promoter context is actually mediated by p300/CBP, remains to be determined.

From a functional point of view, the fact that PTPA expression is actually repressed by p53 might be surprising. Indeed, given the proposed positive role of PTPA in the repair of oxidatively damaged DNA in yeast (33), one would rather expect a positive role for p53 on PTPA expression. However, how exactly p53 is involved in one of the several DNA repair pathways in mammals (56-59) is still not completely clear, and since to our knowledge no p53 homologue is present in yeast, our data could suggest that PTPA might function positively in a p53-independent DNA-repair pathway in mammals.

In conclusion, we have demonstrated that the human PTPA gene is a novel p53 target gene, subject to p53 repression, and that this repressive effect is mediated by the negative control of YY1. As such, these data provide interesting perspectives for the further investigation of the in vivo role of PTPA, particularly in the cellular response to DNA damage.

	ACKNOWLEDGEMENTS

We thank Fabienne Withof and Maria Veeckmans for expert technical assistance. We are grateful to Christian Desaintes (Unité des Virus Oncogènes, Département des Biotechnologies, URA 1644 du CNRS, Institut Pasteur, 75724 Paris Cedex 15, France) for kindly providing both murine and human p53 cDNAs. Thomas Shenk (Howard Hughes Medical Institute and Department of Molecular Biology, Princeton University, NJ) is acknowledged for the gift of the YY1 expression plasmid.

	FOOTNOTES

^* This work was supported in part by grants from the "Fonds voor Wetenschappelijk Onderzoek-Vlaanderen," "Geconcerteerde OnderzoekActies" (GOA) van de Vlaamse Gemeenschap, Human Frontier Science Program, and the European Community Biomed2 Cancer Research Program.The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Post-doctoral fellow of the Fonds voor Wetenschappelijk Onderzoek-Vlaanderen.

^§ To whom correspondence should be addressed. Tel.: 32-16-345-794; Fax: 32-16-345-995; E-mail: jozef.goris@med.kuleuven.ac.be.

Published, JBC Papers in Press, April 27, 2000, DOI 10.1074/jbc.M909370199

² C. Van Hoof et al., submitted for publication.

³ V. Janssens, unpublished observation.

	ABBREVIATIONS

The abbreviations used are: PTPA, phosphotyrosyl phosphatase activator; CBP, cAMP responsive element-binding protein; CMV, cytomegalovirus; DBD, DNA-binding domain; EF-1, elongation factor-1; -Gal, -galactosidase; PP2A, protein phosphatase 2A; Tk, thymidine kinase; YY1, yin-yang 1; nt, nucleotide(s); PCR, polymerase chain reaction; PBS, phosphate-buffered saline: PAGE, polyacrylamide gel electrophoresis.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This Article Abstract Full Text (PDF) All Versions of this Article: 275/27/20488    most recent M909370199v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Art