The tumor suppressor protein p53 (wtp53) can bind to specific target sequences and activate transcription of genes adjacent to these DNA elements. Two p53 binding sites are present in the gene coding for the Fanconi anemia complementation group C (FAC), one in the promoter region (from -1295 to -1266) and one in the coding region of FAC (from +1828 to +1848). Gel shift experiments show that wtp53 binds to the p53 target sequence in the promoter region of the FAC gene. We have investigated whether binding of p53 to these target sites may affect expression of the FAC gene. Transfection experiments show that overexpression of wtp53 in human diploid fibroblasts and lymphoblasts augments transcription of the FAC gene up to three-fold. The transfection efficiency was ~15% for both cell types. The FAC expression activity per transformed cell was stimulated to an estimated level of 18- to 21-fold upon overexpression of p53. The tumor-derived p53 mutants, His175 and His273, that fail to bind DNA showed only a reduced stimulatory activity on FAC transcription. Luciferase assays demonstrated that interaction of p53 with its target site in the FAC promoter does not modulate the promoter activity. We suggest that the p53 binding site contributes to, but may not be an absolute prerequisite for p53- directed transcriptional activation. We conclude that the FAC gene can be added to the list of genes that interact with p53.
Fanconi anemia (FA) is a human genetic disease characterized by chromosomal instability, sensitivity to genotoxic agents, impaired processing of DNA lesions, cell cycle anomalies and cancer predisposition (1 ). Five complementation groups have been described (2 ), but only a single cDNA encoding the FA complementation group C (FAC) polypeptide has been cloned (3 ) and mapped to human chromosome 9. The genes for complementation groups A and D have been assigned to chromosomes 16 and 3 (4 ,5 ), respectively. Fanconi anemia has recently received major attention as a promising candidate for gene therapy (6 ). The FAC gene product has been localized to the cytoplasm (7 ,8 ), a finding that appears to contradict a direct function of FAC in a nuclear repair pathway. Apart from the fact that the FAC protein likely forms a multimeric complex with other proteins (9 ), its function remains enigmatic.
Recently, we have described a p53 binding site within the coding region of the FAC gene (10 ). After the cloning of the FAC promoter (11 ) we also detected a p53 binding site in the promoter region of the FAC gene. While there is no evidence for mutations of p53 itself in FA cells (12 ), there are conflicting reports about the induction of p53 after gamma irradiation and mitomycin C treatment. Accordingly, it has been postulated that the mechanism of the induction of gamma-radiation induced apoptosis is p53-dependent and defective in FA cells (13 ). Even though these findings were disputed (14 ,15 ), it is conceivable that p53 and FAC cooperate in an as yet unknown fashion in the prevention of permanent DNA lesions.
The tumor suppressor protein p53 (wtp53) is able to modulate transcription both in vivo and in vitro (16 -30 ). Although the normal function of p53 still lacks final definition, the results of recent studies suggest that p53 controls cell cycle checkpoints that are important for maintaining the integrity of the genome (31 ,32 ). Thus, loss of p53 function should result in an enhanced frequency of genomic rearrangements or genomic instability (33 ,34 ) and counteract the growth arrest response or programmed cell death (apoptosis) induced by genotoxic insults (35 ,36 ). In accordance with this function, p53 is activated in response to various DNA damaging agents such as UV, gamma irradiation and anticancer drugs (35 ,37 ,38 ). Mechanistically, p53 functions as a transcription factor which recognizes target sites in the promoter and activates transcription of target genes (39 -45 ) including a number of genes involved in cell cycle control, such as the genes for p21 (21 ) and cyclin G (26 ). Other candidate cellular target genes for p53 include genes for GADD 45 (46 ) and Bcl-2 family members (47 ). In the present paper we explore whether p53 interacts with target sites in the FAC promoter and whether such interaction could mediate effects of p53 on the expression of the Fanconi anemia gene FAC.
Close inspection of the FAC promoter sequences (11 ) reveals a p53 binding site in the promoter of FAC (Fig. 1 ). The binding-site is localized at -1295 to -1266 and shows only a single mismatch in the arrangement of the four pentamers described for prototype p53 binding sites (48 ).
To examine whether p53 affects FAC transcription, we overexpressed human p53 in lymphoblastoid and fibroblastoid cells. In addition to wild-type human p53 (Fig. 3 , lane 3) two tumor-derived human p53 mutants were also tested. One of these mutants (His273) (Fig. 3 , lane 1) displays the wild-type conformation of human p53, whereas the other mutant (His175) (Fig. 3 , lane 2) has an altered structural conformation (49 ). Expression of each of the p53 proteins in lymphoblasts (GePa) and fibroblasts (H94-17) was analysed by western blotting using the p53-specific monoclonal antibody PAB421. Cells were transfected with 8 µg of the various p53 expression plasmids. Figure 3 demonstrates that all three different p53 proteins were expressed and could be detected with the antibody PAB421. Each of the three p53 proteins was seven-fold overexpressed, quantified by an Elscript 400 scanner, in p53 transfected cells in comparison to endogeneous p53. Endogeneous p53 of both cell types showed only a modest signal.
To find out whether human p53 modulates the transcriptional activity of the FAC promoter, we studied the activity of a luciferase reporter gene which was under the control of two FAC promoters which differ in length. The vector p1507 contains the p53 binding site while p786 does not. Cotransfection of the luciferase reporter plasmids and the wtp53 expression plasmids into lymphoblastoid cells revealed that increased levels of the wtp53 expression vector (0.1, 1, 8 and 20 µg) do not up-modulate the expression of luciferase (data not shown). There were also no significant differences as a function of the vectors pCMV5 and pC53WT. Furthermore, no significant stimulation of luciferase activity as a function of the presence of the p53 binding site could be observed in four independent transfection experiments (19 142 ± 2834 light units for pC53WT/p1507 cotransfected cells versus 16 757 ± 1417 light units for pC53WT/p786 cotransfected cells from five independent double-transfection experiments). Thus, the FAC promoter-driven FAC gene expression is not activated by overexpression of human wtp53 in this assay.
In this study we present data showing that the expression of the FAC gene product is positively regulated by the transient induction of wtp53 but not by mutant p53. To the best of our knowledge, this is the first demonstration of the ability of p53 to up-regulate the expression of a gene which is directly responsible for a genetic disease.
Because of similarities in the cellular FA phenotype and some p53 functions, i.e., cell cycle defects, genomic instability and DNA repair, it is conceivable that p53 may be relevant to the FA phenotype. This notion received support by the presence of two p53 binding sites, one in the promoter and one in the 3' region, of the FAC gene. The first report about a possible interaction between FAC and p53 was by Rosselli et al. (13 ) who showed that FA(C) cells fail to activate p53 after exposure to gamma irradiation. However, these studies were challenged by the findings of Kruyt et al. (14 ,15 ).
The in vitro growth deficit of FA lymphocytes results from a specific cell cycle lesion, i.e., prolongation of and arrest within the G2 phase segment of the cell cycle (50 ). The latter observation suggests difficulties with maintaining the integrity of the genetic material during replication and preparation for mitosis. For reasons that are not understood, the pattern of congenital malformations and growth disturbance that is seen in FA patients is highly variable. As reported for p53-deficient mice, such a pattern may result from defective protection against developmental oxidative stress (51 ), and FA cells have been shown to be hypersensitive towards oxygen (52 ). It is generally accepted that wild-type p53 is responsible for an ionizing radiation-induced G1 checkpoint control (35 ,36 ). Recently, it has also been reported that p53 induces cell cycle arrest at the G2 boundary of the cell cycle as well (53 -56 ). Both oxygen sensitivity and G2 phase arrest are hallmarks of the FA cellular phenotype and hence could be linked to an impairment of the suggested interaction between p53 and FAC.
In an attempt to elucidate the biological relevance of the putative p53 target site, we transfected human lymphoblasts and fibroblasts with expression vectors encoding wild-type or mutant forms of p53 and monitored expression of the FAC gene. Our transfection experiments show that overexpression of wtp53 in human diploid fibroblasts and lymphoblasts augments the transcription of the FAC gene by 18- to 21-fold.
Stimulatory effects of p53 on gene expression have been reported previously (16 -30 ). For example a 7- to 12-fold stimulation of the CAT gene transcription under the control of the promoter of the muscle-specific creatine kinase (MCK) gene upon p53 overexpression has been reported by Zambetti et al. (30 ). Similarly, Deb et al. (20 ) showed that CAT expression under the control of the promoter of the epidermal growth factor receptor gene was stimulated up to 18-fold in the presence of a p53 expression plasmid. To find out whether the stimulation of FAC expression by human p53 observed in our studies is due to an increased promoter activity, we studied the expression of a luciferase reporter gene which was under the control of the FAC promoter. For these studies two different constructs were used, one containing the p53 binding site of the FAC promoter and the other one lacking this site. However, no significant stimulatory effect could be observed in the presence of the p53 binding site. An explanation for the apparent lack of p53 effects on the FAC promoter activity could be that this promoter construct contains only a single p53 binding site. Previous studies reporting significant stimulatory effects of p53 on gene expression tested promoters containing multiple binding sites for p53 (39 ,54 ). Possibly, stimulation of promoter activity could depend on a synergistic effect between multiple p53 binding sites.
In addition, two p53 binding sites located distantly to each other could exert a cooperative effect on promoter activity as described by Jackson et al. (57 ) for p53-dependent activation of the MCK promoter. The MCK promoter contains two p53 binding sites, one between promoter residues -3182 and -3133 and the other one between residues -177 and -81 relative to the transcription start site. In the presence of both p53 binding sites overexpression of p53 caused a dramatic activation of CAT gene expression under the control of the MCK promoter. However, in the absence of the distal binding site activation was only marginal. This observation was explained by invoking a cooperative interaction between two separate p53 binding sites. These authors propose that a physical proximity of both binding sites could be achieved by looping-out of the intervening DNA. In support of this model, looping-out of DNA and oligomerization of p53 has been shown to be associated with transcriptional activation by p53 (58 ). Therefore, sufficient space between the two p53 response elements seems to be important for a cooperative activation of transcription. It is tempting to speculate that a similar model may also hold true for the FAC gene expression since there is evidence for at least two p53 binding sites in the FAC gene, one in the promoter region and one in the 3' end of the FAC cDNA (9 ). Mutations in the 3' end of the FAC cDNA near the identified p53 binding site lead to a severe clinical phenotype in FA patients (59 ) which highlights the functional significance of this region.
Surprisingly, the tumor-derived p53 mutants His175 and His273 that fail to bind to DNA still showed a measurable stimulatory effect on FAC transcription. This stimulatory effect was only one quarter of that observed during wtp53 activation, but suggests that p53 may also promote gene activity via a direct interaction with the transcriptional machinery. Numerous reports have described interactions between p53 and various regulatory transcription proteins. Both activation and repression of Pol II transcription are thought to involve the interaction of p53 with the TATA box-binding protein (TBP). p53 and TFIID have been shown to cooperate in binding to a p53-responsive promoter and to activate transcription (60 ). Significantly, the interaction between p53 and TBP was not sufficient for transcription stimulation, but required the presence of TBP-associated factors, e.g., TAFII31, TAFII40 and TAFII60 (61 ,62 ). Thus, as most transcription activators, p53 stimulates Pol II transcription via specific coactivators, i.e., TAFs.
Together with the finding that wild-type p53 clearly activates FAC expression, we suggest that the p53 binding site may contribute to, but may not be a prerequisite for p53-directed transcriptional activation. We assume that p53 activates FAC expression only under certain unfavourable conditions, such as DNA damage, which causes p53 to accumulate at higher than normal levels. Elucidation of the mechanism by which p53 activates FAC transcription will be an important part of deciphering the elusive function of the FAC gene.
Epstein-Barr virus-transformed lymphoblastoid cells from a clinically normal individual (GePa) were obtained from H. Joenje (Amsterdam, The Netherlands). Cells were routinely grown in suspension culture with RPMI 1640 medium (GIBCO/BRL) supplemented with 10% fetal calf serum (GIBCO/BRL). High humidity incubators (Heraeus) were equipped with automatic CO2 and O2 sensors. 6% CO2 was used for lymphoblasts.
Primary fibroblastoid cells were obtained from a 20 week male fetus affected by the fragile X syndrome (H94-17). Fibroblast cultures were routinely grown in MEM medium (GIBCO/BRL) supplemented with 10% fetal calf serum (GIBCO/BRL) and incubated at 5% CO2 and 20% O2.
The expression vectors for human wtp53 and two p53 expression vectors with point mutations leading to amino acid exchanges at positions 175 and 273 (pC53wt, pC53mt175 and pCp53mt273) were constructed by B. Vogelstein. The p53 cDNA is under the control of a CMV promoter and based on the vector pCMV5. LUC vectors containing different parts of the FAC promoter were described previously (10 ).
Transient transfections of the plasmids into lymphoblastoid and fibroblastoid cells were achieved by electroporation with a Bio-Rad GenePulser: 15 × 106 cells suspended in serum-free RPMI 1640 or MEM medium were placed in a electroporation cuvette (0.4 cm gap) together with 8 µg of plasmid DNA in a total volume of 400 µl and then pulsed once with 250 V at a capacity setting of 960 µF. After electroporation cells were transferred to 25 ml of prewarmed (37oC) complete medium and allowed to grow in a humidified incubator at 6% CO2/20% O2 for lymphoblasts and 5% CO2/20% O2 for fibroblasts, respectively. After 48 h total RNA was isolated with Trizol (Gibco/BRL).
Cells were transfected with pCMV-nlslacZ. After 48 h cultured cells were rinsed with 150 mM NaCl, 15 mM Na phosphate, pH 7.3 (PBS) and then fixed for 5 min at 4oC in 2% formaldehyde plus 0.2% glutaraldehyde in PBS. The cells were then washed three times with PBS and overlaid with a histochemical reaction mixture containing 1 mg/ml 4-Cl-5-Br-3-indolyl-[beta]-galactosidase (X-gal), 5 mM potassium ferricyanide, 5 mM potassium ferrocyanide and 2 nM MgCl2 in PBS. The X-gal was dissolved in dimethoxysulfoxide (DMSO) at 40 mg/ml, and then diluted into the reaction mixture. Incubation was for 14-18 h at 37oC. [beta]-galactosidase-positive cells were reidentified under the microscope and counted.
Two complementary oligonucleotides containing the putative p53 binding site in the FAC promoter 5'-GAATAGTTGGACATGTTTAAATACTTGAGAGCTATTTTTATTACAA-3' were labeled with [[gamma]-32P]ATP and T4 polynucleotide kinase. After annealing, 1 ng of labeled FAC oligonucleotide was incubated with recombinant human His-p53 that was expressed in Escherichia coli and purified on a Ni++-chelate column. The binding assays (25 ml) contained 500 ng of poly(dA-dT), 100 ng of Pab421 (Dianova) and various amounts of competitor oligonucleotides. After incubation for 30 min at 4oC in binding buffer C (20 mM Tris, pH 7.5, 100 mM NaCl, 0.1% NP-40, 10% glycerol, 5 mM DTT), the products were separated on 4% nondenaturing polyacrylamide gels in 0.5* TEB. As p53-specific competitor an oligonucleotide was used with the sequence of the upper strand 5'-AATTCAGATCTTTGCCTGGACTTGCCTGGCCTTGCCTTTTCG-3' (hRGC). The sequence of the upper strand of the unspecific competitor oligonucleotide is 5'-GATCCTTCGGAGGTCGACCAGTACTCCGGGCGACA-3'.
cDNA encoding wild-type human p53 DNA was cloned into pQE-8 (Quiagen GmbH) which adds six histidine residues to the N-terminus. p53 was overexpressed in E.coli and purified by chromatography on chelating Sepharose under denaturing conditions (8 M urea). The protein was renatured by stepwise dialysis against buffer A (3 M urea, 50 mM MOPS-NaOH, pH 7.8, 500 mM NaCl, 20% glycerol, 5 mM DTT, 100 mM ZnCl2), buffer B (buffer A without urea), and buffer C (buffer B without glycerol).
A 20 µg aliquot of the total RNA purified from different transfected cells was electrophoresed through a 1.2% agarose-formaldehyde gel and transferred to a positively charged nylon membrane (Amersham) according to manufacturer's recommendations. A PCR-amplified fragment of the coding region of FAC was labelled with [[alpha]-32P]dCTP through random priming for use as a probe. The following were used as primers: FACUP (5'-AACCTCAGGAAATCCTCCAGCCAG-3') and FACRP (5'-ACTTGAGTTCGCAGCTCTTTAAGG-3'). After agarose gel electrophoresis the 1042 bp product was isolated using a Qiagen II Gel Extraction Kit and 50 ng was used for the labelling reaction. As a control probe a PCR-amplified fragment of the coding region of [beta]-glucoronidase (GUS) cDNA was used. The following primers yield a 1082 bp fragment: GUSB4: 5'-GCTCTGAATAATGGGCTTCTG-3' and GUSB3: 5'-ACTATCGCCATCAACAACACAC-3'. The blot was probed under high stringency (65oC for hybridisation and washed three times in 2* SSC/1% SDS and once in 0.5* SSC/1% SDS). Quantification of northern blots was performed using the ImageQuant software on a Molecular Dynamics PhosphorImager.
Whole cell protein extracts from transiently transfected cells were solubilized in SDS-sample buffer, sonicated, boiled and separated in 10% SDS-polyacrylamide gel and transferred to nitrocellulose. Filters were probed with the anti-p53 antibody PAB421 and P4oxAb as a peroxidase-coupled second antibody. The visualization was performed with the ECL system following manufacturers recommendations (Amersham). Quantification was performed using an Elscript 400 scanner (Hirschmann). In case of FAC-protein detection also equal amounts of protein extract (measured by Bradford-staining) were separated on a SDS-polyacrylamide gel and transfered to nitrocellulose. Filters were probed with a polyclonal anti-FAC serum (gift of M. Buchwald, Toronto). The visualisation was performed with the alkaline phosphatase system.
Relative levels of FAC and GUS mRNA were measured using a semi-quantitative RT-PCR method. For cDNA synthesis, 5 µg of RNA was incubated in a total volume of 20 µl with 200 U of Moloney Leukemia Virus Reverse Transcriptase (GIBCO/BRL), 1 µl of oligo(dT)12-18 (500 µg/ml), 10 mM dithiothreitol, 20 U of RNASIN (Promega) and 1 mM each of dGTP, dATP, dTTP, dCTP using the reaction buffer provided by the manufacturer. Reactions were performed at 42oC for 50 min. The cDNA product (1 µl) was resuspended in a total volume of 50 µl containing 2.5 U of Taq DNA polymerase (GIBCO/BRL), 50 µM each of dGTP, dATP, dTTP, dCTP, 25 pmol of forward (FACUP and GUSB3) and reverse (FACRP and GUSB4) primers, and 1* reaction buffer provided by the manufacturer. Amplification cycles consisted of 94oC for 45 s, 61oC for 45 s, 72oC for 90 s. Each cDNA was amplified in serial cycles from 25 to 35 with increments of three cycles for FAC and GUS, respectively, to obtain data within the linear-range of the assay. PCR products were size-fractionated by electrophoresis in 1% agarose gels. Forward and reverse primers were designed to span exon-intron junctions so as not to amplify contaminating genomic DNA. Expected sizes of the amplified products for FAC and GUS were 1042 and 1082 bp, respectively.
The assay was performed with the Luciferase Reporter Gene Assay kit from Boehringer-Mannheim following the kit instructions using a Berthold Lumat LB 9501 luminescence reader. Control cells (GePa) were double-transfected with the vector pC53WT and with LUC vectors containing the FAC promoter. p1507 (containing the p53 binding site) and p786 (without p53 binding site) have been described previously (11 ).
We thank H.-W. Zentgraf for providing bacterially expressed human p53, J. Pesch for stimulating discussion and support and Petra Busch and Wilma Hofmann for technical assistance. This work was supported by a grant from the Bundesministerium für Bildung, Wissenschaft, Forschung und Technologie (grant: FKZ 0310721) and by grants from the Deutsche Forschungsgemeinschaft (SFB165, 172).
Human Molecular Genetics Pages
©
Introduction
Results
A p53 binding site is present in the FAC promoter
Overexpression of p53 activates expression of the FAC gene
Effects of human tumor-derived mutant p53 on FAC expression
Binding of wtp53 to the FAC promoter is not directly responsible for activation of FAC transcription
Discussion
Materials And Methods
Cells and cell culture
Plasmids
DNA transfection
[beta]-galactosidase histochemistry
Electrophoretic mobility shift assay
Expression and purification of p53
Northern blot analyses
Western blot analyses
Semi-quantitative RT-PCR
Luciferase reporter gene assay
Acknowledgements
References
REFERENCES
This page is maintained by OUP admin. Last updated Fri Jan 31 11:45:11 GMT 1997. Part of the OUP Journals World Wide Web service.Copyright Oxford University Press, 1996