Oncogene (1999) 18, 7621 7636 1999 Stockton Press All rights reserved 0950 9232/99 $15.00 http://www.stockton-press.co.

uk/onc

Twenty years of p53 research: structural and functional aspects of the p53 protein
Pierre May*,1 and Evelyne May1
1

Laboratoire de Cancerogenese Moleculaire, UMR 217 CEA-CNRS, DRR, DSV, CEA 60-68 Av. Division Leclerc B.P. no 6-92265 Fontenay Aux Roses Cedex, France

Keywords: p53 protein; tumour suppressor; transactivator; DNA damage; cell growth arrest; apoptosis

Introduction From its modest beginnings in 1979, as a transformation-associated protein, to the discoveries that p53 plays a key role in tumour suppression and in the cellular response to DNA damage, p53 has risen to molecular superstardom both in the research community and in the large public. The aim of this review is to relate how the p53 history has unfolded until now, and to underscore our present knowledge of this paradigmatic protein. To attempt coverage of all aspects of p53 would be unrealistic. Rather, we restrict our considerations to the properties of p53 as tumour suppressor and as cell cycle regulator activated by DNA damage, emphasizing the relationship between structure and function of the p53 protein. Early studies indicate a role of p53 in cell transformation The p53 protein was originally identied by the observation that antibodies against SV40 large T antigen from animal bearing tumours produced by SV40-transformed cells coimmunoprecipitated a cellular protein of apparent molecular mass 53 kDa, called p53. From these observations, it was concluded that p53 interacted with SV40 large T antigen (Chang et al., 1979; Kress et al., 1979; Lane and Crawford, 1979; Linzer and Levine, 1979). Quite independently, p53 protein was also identied by its apparently high expression in chemically induced tumours (Deleo et al., 1979). It was also noted that a high level of p53 was found in a large variety of transformed cells regardless of how the cells were transformed, whereas the p53 level in non-transformed cells was generally very low (Kress et al., 1979; Lane and Crawford, 1979; Linzer and Levine, 1979; Rotter et al., 1980). Later, it was discovered that viral oncoproteins from other DNA tumour viruses have the properties to bind p53. Thus the adenovirus E1B (Sarnow et al., 1982) and human papillomavirus HPV E6 protein (Schener et al., 1990), Epstein Barr virus nuclear antigen (Szekely et al., 1993), hepatitis B virus X protein (Wang et al., 1994) and human cytomegalovirus IE84 protein (Speir et al., 1994), form complexes with the p53 protein.
*Correspondence: P May

Moreover it was found that many transformed cell lines, as well as primary human cells from patients with various types of tumours, contained an elevated level of p53, whereas non transformed cells contained only minute amounts of p53. This dierence was attributed to stabilization in transformed cells of the normally short-lived protein (Oren et al., 1982). p53 was also identied by its ability to be immunoprecipitated from a variety of transformed cell lines with antibodies produced by certain tumour-bearing animals (Kress et al., 1979; Melero et al., 1979; Rotter et al., 1980). Similarly, anti-p53 antibodies were also identied in sera from cancer patients (Crawford et al., 1982; Caron de Fromentel et al., 1987; Angelopoulou et al., 1994). During the 1980s, several studies provided grounds for believing that p53 is an essential cell cycledependent protein (Mora et al., 1980; Reznikov et al., 1989; Milner, 1984; Milner and Milner, 1981; Mercer et al., 1982; Reich and Levine, 1984; Kaczmarek et al., 1986; Shohat et al., 1987). Moreover, suggestive evidence for a role of the p53 protein in DNA replication was obtained by the ndings that mouse p53 is able both to compete with DNA polymerase a for binding to the SV40 T antigen (Gannon and Lane, 1987) and to inhibit SV40 origindependent DNA replication in monkey COS cells (Braithwaite et al., 1987). The idea that p53 was a growth regulatory protein was compatible with its nuclear localization and short half-life of 5 10 min in normal (non transformed) cells (Reich et al., 1983, 1984; Rogel et al., 1985). In 1984, it was observed that exposure of non transformed cells to ultraviolet radiation stimulated the accumulation of p53, due to post-translational stabilization (Maltzman and Czyzyk, 1984). As early as 1986, evidence was obtained that a particular conformational form of p53 interacts selectively with heat shock protein hsp70, demonstrating variations in anity of mutant proteins p53 for hsp70, whereas wild-type p53 protein does not bind to hsp70 protein (Pinhasi-Kimhi et al., 1986; Hinds et al., 1987; Sturzbecher et al., 1987; Ehrhart et al., 1988). From oncogene to tumour suppressor gene The molecular cloning of the p53 cDNA (Zakut-Houri et al., 1985) allowed a variety of studies to increase our insight into the function of p53 in cell transformation in vitro and in tumorigenesis in vivo. In fact, a number of properties initially attributed to p53 protein, including the ability to immortalize cells (Jenkins et al., 1984) and to transform primary rat embryo broblasts in cooperation with ras (Eliyahu et al.,

Structure and function of p53 protein P May and E May

7622

1984; Parada et al., 1984) suggested that it was a dominantly acting oncogene. Mutation of p53 gene was reported to enhance its transformation eciency, a nding which tted with the hypothesis that p53 was a cellular oncogene that could be activated by mutation (Jenkins et al., 1985). However, in marked contrast with the notion of p53 acting as an oncogene, there was conclusive evidence that the loss of normal p53 expression and function may constitute an important step in the transformation process. This conclusion was based on the nding that rearrangement and functional inactivation of p53 gene was observed at high frequency in mouse spleen tumours induced by the Friend erythroleukaemia virus (Mowat et al., 1985; Chow et al., 1987; Rovinsky et al., 1987; Ben David et al., 1988; Munroe et al., 1988). The paradox was resolved when it was shown that the originally cloned p53 cDNAs used in the early experiments unwittingly contained dominant negative missense mutations within a conserved region of p53 important for both the conformation and biological activity of the protein. Subsequent studies in the late 1980s clearly established that genuine wild-type p53 acts as an authentic tumour suppressor gene instead of an oncogene. p53 is a tumour suppressor gene In direct contrast with the mutant forms, wt p53 cannot cooperate with ras, and indeed can actually suppress transformation by mutant p53 gene and ras (Finlay et al., 1989; Eliyahu et al., 1989). The wild-type form of protein acts as tumour suppressor, as for example the transfection of the wild-type gene into human osteosarcoma cells lacking endogenous p53 abrogates the neoplasticity of the cells (Chen et al., 1990). Furthermore p53 was shown to be lost or to contain mutations that inactivate p53 in about half of almost all human cancer types arising from a wide spectrum of tissues (Baker et al., 1989, 1990; Caron de Fromentel and Soussi, 1992; Hollstein et al., 1994; Marshall, 1991; Nigro et al., 1989) and is now known to be the most commonly altered gene in human tumour (Caron de Fromentel and Soussi, 1992; Hollstein et al., 1994, 1996). Some mutant p53 proteins may transform by acting as inhibitory proteins competing with and blocking the activity of the endogenous wild-type p53 protein. Such mutants possess dominant negative mutations. It is known that the p53 protein forms tetramers, which are the functional forms (Kraiss et al., 1988; Clore et al., 1994, 1995). Dominant negative p53 mutants have a wrong conformation and produce with wt-p53 inactive mixed tetramers of altered shape, which express an unusual epitope recognized by the monoclonal antibody PAb240 (Gannon et al., 1990). A large number of tumour point mutations within the gene have been reported and many of these are so-called strong mutations that have a dominant-negative eect (Milner et al., 1991). In addition to losing p53 function in general, some mutant proteins may acquire new functions that contribute to transformation and tumorigenic potential (gain of functions mutants) (Dittmer et al., 1993). Mutations in the p53 gene have also been linked to Li Fraumeni syndrome, an inherited susceptibility

disorder in which aected individuals are at increased risk of developing a variety of cancers, including soft tissue sarcoma and cancers of the bone, breast, brain and genito-urinary tract (Malkin et al., 1990). Aected individuals also tend to develop specic tumour at an earlier age than usual. The importance of these observations was underscored by the nding that mice transgenic for expressing mutant p53 alleles (Lavigueur et al., 1989) or homozygously defective for p53 (Donehower et al., 1992; Harvey et al., 1993; Purdie et al., 1994; Jacks et al., 1994) develop normally but they have a high incidence of tumours, especially of lymphoid tissue. In recent years, the p53 gene has been shown to be a key regulator in a wide range of cellular processes, including cell cycle control, DNA repair, genome stability, programmed cell death, dierentiation (Rotter et al., 1994), senescence (Wynford-Thomas, 1999) and angiogenesis (Vojta and Barrett, 1995; Bouck, 1996). In addition p53 is a major component of DNA damage response pathway in mammalian cells. Some of these normal cellular functions of p53 can be modulated and sometimes inhibited by interaction with certain viral oncoproteins (SV40 large T antigen, Ad E1B HPV E6, human hepatitis B virus X protein, etc.), or with the cellular protein MDM-2. This protein has been identied as the product of mdm-2 oncogene, which is amplied in some types of tumour (Momand et al., 1992; Oliner et al., 1992). This is an alternative mechanism of p53 inactivation which is common in many tumour types (including soft tissue sarcomas, neural tumours, bladder, cervical and breast carcinomas as well as leukaemias). These protein protein interactions inactivate p53 function by abrogating specic DNA binding and/or transactivation activity, sequestring wild-type p53 in the cytoplasm or promoting its degradation (Beaudry et al., 1996). Recently a potential mutation-independent mechanism of p53 inactivation was described which involves abnormal cytoplasmic sequestration of wild-type p53 with concomitant nuclear exclusion (Moll et al., 1996). This phenotype was present in 37% of inammatory breast carcinomas and greater than 95% of undifferentiated neuroblastomas but never in dierentiated benign derivatives of NB (Moll et al., 1992, 1995, 1996). This naturally occurring translocation defect compromises the suppressor function of p53 and likely plays a role in the tumorigenesis of these tumours previously thought to be unaected by p53 alterations. Immunochemical analysis of p53 conformation During the 1980s Milner's and Lane's groups developed an approach using antibodies as real conformational probes aimed at analysing the conformational changes in the p53 protein. Studies using conformation-specic monoclonal antibodies show that single point mutations in p53 can alter the conformation of the entire polypeptide. The wild-type p53 is characteristically reactive with PAb246 (murine specic) and PAb1620 but not with PAb240 whereas the reverse is true for a great number of mutants of p53 associated with cancer (Gannon et al., 1990; Cook and Milner, 1990). PAb246 and PAb1620 recognize topologically related epitopes

Structure and function of p53 protein P May and E May

which are disrupted by protein denaturation. In contrast, PAb240 recognizes a conserved amino acid motif, (residues 213 217 in human p53) which is cryptic in wild-type p53 but commonly exposed in many mutants (Stephen and Lane, 1992). p53, a sequence specic transactivator The biochemical mechanism through which p53 exerts its growth-suppressive activity was actively pursued. Several convergent lines of observation have increased our insight into the biochemical activities of p53 protein. First, it was reported that p53 contained an acidic domain near its N-terminus resembling those previously described in well-characterized transcription factors. By using hybrid proteins containing the DNA binding domain of yeast GAL4 and portions of p53 (Fields and Jang, 1990; Raycroft et al., 1990), it was shown that p53 protein contained a transcriptionactivating sequence that functions in both yeast and mammalian cells. The NH2-terminal 73 residues of p53 activated transcription as eciently as the herpes virus protein VP16 which contains one of the strongest known activation domain. The activation domain has been shown later to require only the region lying between codons 1 and 42 (Unger et al., 1993). The strength of the activation and nuclear localization of p53 suggested that p53 is involved in transcriptional control and the inability of p53 mutant proteins to activate transcription might enable them to be transformation-competent. This transactivating property of p53 was clearly substantiated when it was discovered that p53 protein has the property to interact with specic double-strand DNA sequences (Kern et al., 1991; Bargonetti et al., 1991) from which a consensus p53 DNA-binding site has been derived (El-Deiry et al., 1992). After having selected 18 independent human genomic fragments that could bind to wt p53 protein in vitro, the binding sites within these fragments were analysed. Precise mapping of the binding sequences within these clones revealed a consensus binding site with a striking internal symmetry consisting of two copies of the 10 base-pair motif 5'-PuPuPu-C(A/T)(T/A)GPyPyPy-3' separated by 0 13 nucleotides. One copy of the motif is insucient for binding p53 and subtle alterations of the motif, even when present in multiple copies, result in loss of anity for p53. Mutants of p53, representing each of the four `hot spots' frequently altered in human cancer, failed to bind to the consensus site. The symmetry of the four half sites within the consensus sequence suggests that p53 interacts with DNA as a tetrameric protein, which is consistent with the known structure of p53 (Kraiss et al., 1988; Clore et al., 1994). Furthermore, based on experimental data (Bourdon et al., 1997) it was proposed an extended consensus sequence composed of the two decamers of the ElDiery consensus sequence anked by two additional ones. Other studies have directly supported the notion that p53 can specically bind to the consensus sequence. From a large pool of random oligonucleotides, a small subset was selected by the properties of binding to p53 (Funk et al., 1992). The oligomers shared a 20-bp sequence very similar to the consensus sequence, with no intervening sequence between the 10-bp motifs.

When this sequence was placed upstream of a reporter gene, it conferred p53-dependent activation in transient expression assays. Several additional DNA sequences specically recognized by p53 have been identied. While many of these sequences showed similarity with the El-Deiry consensus sequence, others did not. For instance the p53 binding sites identied in the SV40 origin of replication (Bargonetti et al., 1991) markedly deviates from the consensus sequence. The fact that p53 bound DNA specically and contained an acidic activation domain at its Nterminus prompted the conclusion that p53 has the characteristics of a transcriptional regulator. This conclusion was substantiated when it was reported that cotransfection of a p53 expression vector with a plasmid containing a wild-type p53-binding site placed upstream of a reporter gene results in a high level of reporter gene expression in mammalian cells. The consensus sequence is present either in the regulatory region or in intron of a variety of target genes and is thought to activate transcription via an interaction between its N-terminal acidic domain (rst 42 amino acids) and the basal transcription apparatus (possibly TATA box binding protein associated factors such as human TAFII40 and TAFII60 (Thut et al., 1995), and the p62 subunit of the dual transcription/repair factor TFIIH (Wang et al., 1995a). Functional inhibition of the p53 activation domain also appears to play a role in human tumorigenesis. For instance the product of the mdm2 oncogene, which binds and is believed to mask the activation domain of p53, is overexpressed in certain tumours, resulting in the abrogation of p53 function, accompanied by the proteolytic degradation of p53, mediated by the ubiquitin pathway (Momand et al., 1992; Oliner et al., 1992, 1993; Midgley and Lane, 1997). Additional data showed that p53 may down-regulate the expression from a number of other promoters including those of the c-fos, c-jun, c-myc, b-actin, hsc-70, IL-6 genes (Ginsberg et al., 1991; Lechner et al., 1992; Ko and Prives, 1996). In contrast to the p53-mediated transactivation, the p53mediated transrepression does not involve an interaction between p53 and DNA. Instead this repression may be the result of an association between p53 and the cellular TBP (TATA-binding protein) which is a component of the TFIID complex (Seto et al., 1992; Ragimov et al., 1993; Truant et al., 1993). All the modular regions of the transcriptional activator p53 were nally identied when the central region of p53 protein containing the four conserved regions and the major mutation hot spots (residues 102 292) (Pavletich et al., 1993; Bargonetti et al., 1993; Wang et al., 1993) was recognized as the specic DNA-binding domain. Recently it was shown that interacting with coactivators such as p300/CBP, PCAF as well as posttranslational phosphorylation, glycosylation and acetylation, appear to be necessary for p53 to function with maximal activity as a transcription factor (Wadgaonkar and Collins, 1999; Scolnick et al., 1997). Other biochemical activities of p53 Recent research suggests that p53 may participate in the repair machinery (Smith et al., 1994; Wang et al., 1995a; Ford and Hanawalt, 1995; Li et al., 1996). For

7623

Structure and function of p53 protein P May and E May

7624

example (Wang et al., 1994, 1995a), it was shown that p53 binds to and modulates the repair activity of the nucleotide excision repair factors XPB and XPD. More recently (Buchlop et al., 1997) it has been demonstrated that p53 interacts with human Rad51 protein, a proposed member of the mammalian recombination machinery. Mapping studies using a panel of deletion mutants indicate that human (h) Rad51 could bind to two regions of p53: one between amino acids 94 and 160 and a second between 264 and 315. Addition of anti-p53 antibody PAb 421 (epitope 372 381 amino acids) inhibited the interaction with h Rad51. On the basis of their results, the authors suggest that p53 interaction with Rad51 may inuence DNA recombination and repair and that additional modications of p53 by mutations and protein binding may aect this interaction. In addition, p53 has been shown to bind single stranded DNA ends (Bakalkin et al., 1994) and to possess strong DNA-DNA and RNA-RNA strand annealing activity (Bakalkin et al., 1994; Oberosler et al., 1993; Brain and Jenkins, 1994). A link between p53 mutations and some carcinogens involved in cancer development p53 is mutated in a large fraction of many dierent types of human cancer. The nature, type and location of mutations depend upon both the tissue origin of the cancer and particular carcinogens, endogenous or exogenous, responsible for the genetic alterations. In carcinomas, most of the mutations are missense mutations producing an altered protein, and the second allele is then lost, reducing the mutant allele to homozygosity. The point mutations that alter p53 function are distributed over a large region of the molecule, especially the highly conserved central region. Frameshift and nonsense mutations that truncate the proteins are located outside of these regions. A considerable number of p53 mutations have been reported, a few of them display much higher frequencies than the other, including those at codons 175, 249, 273 and 282. Mutations in these residues account for 40% of the total missense mutations reported in human cancers. It should be noted that there is a tissue-specic distribution of `hot spot' mutations. The rst correlation between a specic carcinogen and p53 was suggested by Harris, Ozturk and their colleagues (Hsu et al., 1991; Bressac et al., 1991) who identied with a high frequency the presence of G4T transversion at codon 249 (R4S) in hepatocellular carcinomas (HCC) from southern Africa and eastern Asia (Bressac et al., 1991). Another example was provided by Brash et al. studying the involvement of p53 in human squamous cell carcinomas (Brash et al., 1991). They found that 14/24 (58%) of invasive squamous cell carcinomas of the skin contained mutations in the p53 gene, each altering the amino acid sequence. Involvement of UV light in these mutations was indicated by the presence in three of the tumours of a CC?TT double-base change which is known to be induced only by UV. The analysis of p53 in other tumour types has also increased our insight into the link between p53 mutations and risk factors identied by epidemiological studies. For example it has been established that

exposure to cigarette smoke is correlated with G:C to T:A transversion with a DNA coding strand bias and a reduced frequency of G:C4A:T. The DNA coding strand bias is consistent with preferential repair of the transcribed strand (Hanawalt, 1987). Moreover G:C?A:T transitions are common in colon cancer but less frequent in hepatocellular carcinomas (Harris and Hollstein, 1993; Bennett et al., 1999; Hollstein et al., 1996). The p53 gene and its product: structure and function The human p53 gene was shown to span about 20 kb of DNA and to be localized to the short arm of chromosome 17 (17p13). This gene is composed of 11 exons, the rst of which is non coding and localized 8 10 kb away from exons 2 through 11 (Benchimol et al., 1985; Isobe et al., 1986; McBride et al., 1986; Miller et al., 1986; Oren, 1985). The p53 has been conserved during evolution (Soussi et al., 1987). Through cross-species comparison of amino acid sequences, the p53 proteins showed the existence of ve highly conserved regions within the amino acid residues 13 23, 117 142, 171 181, 234 250 and 270 286 (Soussi et al., 1990; Soussi and May, 1996). These regions, termed domains I V, were expected to be crucial for the p53 functions. The more recent data of the literature conrm this view (Soussi and May, 1996). The human p53 protein consists of 393 amino acids and contains four major functional domains. At the Nterminus is a transcriptional activation domain (amino acids 1 42) and within the central part of p53 is the sequence-specic DNA-binding domain (amino acids 102 292). The C-terminal portion contains an oligomerization domain (amino acids 323 356) and a regulatory domain (amino acids 360 393). The anatomy of the human p53 protein in relation to its function will be now described in more details. A schematic representation of the structure of human p53 protein is shown in Figure 1. Amino-terminal region of p53 The N-terminal region contains the activation domain (amino acid 1 42) (Fields and Jang, 1990; Unger et al., 1992) although neighbouring sequences are probably needed for the activation domain to function optimally. The acidic Nterminal transcriptional domain allows p53 to recruit the basal transcriptional machinery, including the TATA box binding protein (TBP) and TBP-associated factors (TAF) components of TFIID (Lu and Levine, 1995; Thut et al., 1995). It should be noted that TBP interacts with the carboxyl terminus of p53 (amino acids 318 393) as well as with its activation domain (amino acids 20 57) (Horikoshi et al., 1995). The amino terminus of p53 interacts also with the single stranded DNA-binding protein RP-A and the p62 subunit of the transcription/ repair factor TFIIH (Dutta et al., 1993; He et al., 1993; Li and Botchan, 1993). The adenovirus E1B-55 kDa protein, the human MDM2 protein, and hepatitis B virus X protein, bind to the amino-terminal region of p53 and inhibit its transactivation function (Yew and Berk, 1992; Oliner et al., 1993; Momand et al., 1992; Levine, 1997). Amino acid residues 22 and 23 play a key role in the

Structure and function of p53 protein P May and E May

7625

Figure 1 Schematic representation of the structure of p53 protein. At the top are shown the viral and cellular proteins known to interact with particular regions of p53. Then, from top to bottom, are represented functional domains, conserved regions, posttranslational modications, respectively. The sites of phosphorylation and acetylation are shown together with the kinases and acetylases that have been identied. The tertiary structure of the site-specic DNA-binding domain is shown (bottom). L1, L2, L3 indicate loops, and LSH indicates a loop-sheet-helix structure demonstrated by Cho et al. (1994). The tetrahedrally coordinated zinc atom is shown. The numbers refer to the amino acid residues. More details are given in the text

binding of p53 to E1B-55 kDa and MDM2 (Lin et al., 1994; Levine, 1997). In addition, the amino acid residues 19, 22 and 23 are required for transcriptional activation by the p53 protein in vivo (Lin et al., 1994). The same amino acids are involved in the in vitro binding of the TATA-associated factors TAFII60 and TAFII40 (Lu and Levine, 1995; Thut et al., 1995). Interestingly these amino acid residues are highly conserved, strongly suggesting that the highly conserved region I (Soussi et al., 1990) must be slightly modied and actually extends from amino acid residues 13 23 (Soussi and May, 1996). The negative regulation of p53 through MDM2 functions in two ways. Firstly, MDM2 binds to the activation domain of p53 and thereby inhibits the ability of p53 to stimulate transcription (Momand et al., 1992; Oliner et al., 1993). Secondly, MDM2 also plays a central role in the regulation of p53 levels since the oncoprotein targets p53 for rapid degradation (Haupt et al., 1997; Kubbutat et al., 1998). The mdm2 gene itself is activated by p53, which gives the opportunity for feedback negative control of p53 activity (Wu et al., 1993). These observations lead to an explanation as to why wild-type p53 is less stable than tumour-derived mutants of p53 incapable of inducing the expression MDM2 (Midgley and Lane, 1997). In addition to its ability to function as a transactivator p53 can repress transcription of several cellular and viral genes (Ginsberg et al., 1991; Seto et al., 1992; Subler et al., 1992; Ago et al., 1993; Mack et al., 1993; Murphy et al., 1996). The N-terminal region contains also a proline rich region (amino acid 63 97) with striking similarity to

SH3-binding proteins and which is required for p53mediated apoptosis in some experimental systems (Sakamuro et al., 1997), for suppressing tumour cell growth (Walker and Levine, 1996) and for human papilloma E6 to target degradation (Li and Cono, 1996). The fact that this region is proline-rich and contains ve repeats of the SH3 binding motif PXXP suggests that it may be involved in physical interaction with elements of signal transduction pathways, for example c-abl. Although deletion of the proline-rich domain does not aect transactivation of several promoters, such as WAF1, mdm2 and BAX, it does alter transcriptional repression (Venot et al., 1998). A chief point of interest is the fact that p53 gene is polymorphic at amino acid 72 which results in either a proline (p53 Pro) or a arginine (p53 Arg) at this position (Storey et al., 1998). The replacement of proline with arginine abolishes one of the ve PXXP motifs. Recently it has been reported that the p53 Arg is signicantly more susceptible than p53 Pro to E6mediated degradation to human papilloma virus (HPV)-induced tumours in homozygous p53 Arg (Storey et al., 1998). This correlates with the clinical nding that individuals homozygous for Arginine 72 are more susceptible to human papillomavirus associated tumorigenesis than heterozygotes, at least in certain populations (Storey et al., 1998). Central region of p53 The central part of the protein, in an area bounded approximately by amino acids 102 to 292 contains the four conserved regions II V. It is within this central part that 80 90% of the tumour mutations are found. Moreover, the central part contains

Structure and function of p53 protein P May and E May

7626

the DNA binding domain which recognizes and binds a consensus target sequence (El-Deiry et al., 1992). It corresponds also to the core domain that is resistant to proteolysis by enzymes such as thermolysin or subtilisin (Bargonetti et al., 1993; Pavletich et al., 1993). It is noteworthy that the central region also functions as a protein binding domain interacting with simian virus 40 (SV40) large T antigen (Jenkins et al., 1988) and the cellular proteins 53BP1 and 53BP2 (Ruppert and Stillman, 1993; Gorina and Pavletich, 1996; Iwabuchi et al., 1994). X-ray crystallography analysis by Cho and colleagues (Cho et al., 1994) of the crystal structure of the core DNA-binding domain bound to its cognate site has been immensely informative. Under these conditions, the structure of the core domain consists of a large b sandwich that acts as a scaold for three loop-based elements. The sandwich is made up of two antiparallel bsheets containing four and ve b-strands, respectively. The scaold anchors the loops and participates in headto-tail dimerization. The rst loop L1 (LSH for loop sheet helix) binds to DNA within the major groove. The second loop L2 binds to DNA within the minor groove. The third loop L3 packs against L1 and stabilizes it. The L2 and L3 loops are connected by a zinc atom tetrahedrally coordinated on amino acids Cys 176, His 179, Cys 238 and Cys 242 (Figure 1). This zinc atom stabilizes the structure of the loops. Such a role of zinc is substantiated by the fact that metal chelating agents inhibit the sequence-specic DNA binding and change its conformation as detected by antibodies (Hainaut and Milner, 1993a,b). One of the most notable features of the three dimensional structure is its correlation with the data on mutations. A major observation is that the loop domains overlap with highly conserved domains and that the majority of mutations reside within the three loops (Figure 1). Two classes of mutations are predicted from the knowledge of this structure. Class I aects residues that directly contact DNA. These include Arg248 and Arg273 which are the most frequently mutated residues in p53. Class II mutants aect residues that do not contact DNA directly but which made extensive contacts with other residues in the polypeptide and have a strong stabilizing eect on the protein structure. The DNA contact mutants of p53 retain a wild-type conformation and form stable complexes with authentic wild-type p53. Moreover the mutant/wild-type p53 mixed complexes can bind a specic DNA target, indicating that DNA contact mutants may not interfere with wild-type p53 function. In contrast structural mutants can exert a dominant negative eect over wildtype protein. The structural mutations alter the conformation of the p53 protein and produce a protein that now react with the monoclonal antibody PAb 240 whose epitope (residues 212 217) is not accessible to the binding of this antibody in the wildtype structure (Gannon et al., 1990; Stephen and Lane, 1992). The loss of the wild-type conformation can also be revealed by the loss of reactivity both to the monoclonal antibodies, PAb1620 specic for murine and human p53 and to the mouse specic PAb246 (Yewdell et al., 1986). Carboxy-terminal region of p53 The C-terminal region (residues 300 393) includes: (1) a exible

linker region (appr. amino acids 300 318) connecting the central core domain and the C-terminal region; and (2) a tetramerization domain from amino acids 323 356. It is well established that the p53 forms tetramers (Kraiss et al., 1988) via an oligomerization domain (amino acids 323 356). The structure of this tetramerization domain has been deduced from nuclear magnetic resonance (Clore et al., 1994) and X-ray crystallography (Jerey et al., 1995). It forms a highly symmetrical tetramer. Each monomer is composed of a turn (Asp324 Gly325), a b-strand (Glu326 Arg333), a second turn (Gly334) and an a-helix (Arg335 Gly356). Two monomer peptides form a dimer in which the a-helices and b-strands are antiparallel. Two dimers interact to form a tetramer through their a-helices. The b-strands lie on the outside of the tetramer on opposite faces. Thus the tetramer is best described as a dimer of dimers, with two dimers interacting through their a-helices to form a four-helix bundle. Tetramerization appears to be required for ecient transactivation in vivo and for p53-mediated suppression of growth of carcinoma cell lines (Pietenpol et al., 1994). It should be noted that three nuclear localization signals (NLS) have been identied in the C-terminal region. Mutagenesis of the most N-terminal signal (NLS1, amino acids 316 325) induces the synthesis of a totally cytoplasmic p53 protein, while alteration of the NLS2 (amino acids 369 375) and NLS3 (amino acids 379 384) leads to both cytoplasmic and nuclear localization (Dang and Lee, 1989; Shaulsky et al., 1990b). Adjacent to the oligomerization domain (323 356), there is a basic region (amino acids 363 393) which has also been referred to as an apoptotic domain (Wang et al., 1996), a transcriptional regulatory domain (Wang and Prives, 1995) or a DNA damage recognition domain. This extreme C-terminal domain (amino acids 363 393) seems to act as a negative regulator of p53 sequence specic binding. This idea is further substantiated by the nding that binding of p53 to the anti-p53 monoclonal antibody PAb421 (which recognizes a C-terminal epitope) can activate specic DNA binding (Abarzua et al., 1996; Hupp and Lane, 1995; Selivanova et al., 1997, 1999; Shaw et al., 1996). According to the allosteric model for regulation of p53 activity (Hupp and Lane, 1994; Hupp et al., 1995; Halazonetis and Kandil, 1993), an interaction between the C-terminal domain and another region in a p53 tetramer locks the tetramer in a DNA binding incompetent state. The activation of p53 in cultured mammalian cells has been correlated with phosphorylation (Ko and Prives, 1996), acetylation (Sakaguchi et al., 1998) glycosylation (Shaw et al., 1996) and proteolytic removal (Okorokov et al., 1997) of the C-terminal domain. These modications are thought to activate p53 by causing a conformational change of the protein, regulated by an allosteric eect. A number of viral and cellular proteins have been shown to bind to C-terminal region of p53, including the hepatitis B virus X protein, the TFIIH subunits XPD and XPB which are two helicases in vitro, and TBP (Wang et al., 1994, 1995b; Horikoshi et al., 1995). A third helicase, designated CSB, involved in a DNA repair

Structure and function of p53 protein P May and E May

pathway, binds also to the C-terminal region of p53 (Ko and Prives, 1996; Levine, 1997). p53 and the cellular response to DNA damage As we mentioned above, transgenesis experiments demonstrated that mice which lack functional p53 genes mature normally in utero, although they are predisposed to an early development of tumours. Although p53 has been shown to be involved in dierentiation and development (Almog and Rotter, 1997) as well as in senescence (Wynford-Thomas, 1999) and angiogenesis (Vojta and Barrett, 1995; Bouck, 1996), no clear role for p53 has been identied in the normal growth of most cell types. This apparent contradiction was resolved when it was discovered that wild-type (but not mutant) p53, plays a critical role in mediating cell cycle arrest or apoptosis in response to DNA damage or other kinds of stress such as hypoxia, nucleotide depletion, hyperoxia, activated oncogenes, etc. An early clue was the observation (Maltzman and Czyzyk, 1984) that wt p53 is stabilized in cells exposed to UV radiation, and a considerable amount of research has since conrmed the central role of DNA damage in mediating the p53-induced growth arrest or apoptosis. Following the DNA damage the p53 protein rapidly accumulates and becomes activated (Kastan et al., 1991; Lu and Lane, 1993), probably in response of double strand breaks induced by the genotoxic agents. p53 binds strongly to strand breaks by its C-terminal domain (Nelson and Kastan, 1994) and this leads to the stabilization and activation of the protein. Several processes might be implicated in activating p53, including phosphorylation, glycosylation, binding to regulatory proteins, alternative splicing and acetylation (Gu and Roeder, 1997; Siliciano et al., 1997; Giaccia and Kastan, 1998). Perhaps the best studied activator of p53 is DNA damage following ionizing irradiation. Activation of p53 has two outcomes: cell growth arrest or apoptosis. The overexpression of p53 protein that is observed in irradiated cells can also be produced by ectopic p53 expression after p53 gene transfer into non irradiated control cells. Non irradiated cells expressing thereby high levels of wildtype p53 have been shown to undergo either cell cycle arrest in G1 (Diller et al., 1990; Lin et al., 1992) or apoptosis (Yonish-Rouach et al., 1991, 1994; Shaw et al., 1992). Since radiation and chemotherapeutic agents either directly or indirectly damage DNA, it is reasonable to suppose that p53 status has a strong impact on the ecacy of cancer therapy. In point of fact, a number of studies raise the possibility that cells lacking p53 activity might more readily survive certain forms of cancer therapy (Lowe et al., 1994). Role of p53 in growth arrest In certain cells, induction of p53 in response to ionizing radiation (IR) results in transient arrest in the G1 phase of the cell cycle. For example, 5-Gy ionizing irradiation produces p53dependent G1 growth arrest in proliferating fibroblasts (Kastan et al., 1991). It is well established that wild-type p53 is required for the induction of G1 arrest in response to IR because cell lines engineered to lack p53 activity display an attenuated response (Kuerbitz

et al., 1992; Kessis et al., 1993). In addition, embryonic broblasts from p53-null mice lose the G1 arrest response to IR (Kastan et al., 1992). It should be noted that in normal diploid broblasts, exposure to ionizing irradiation results in long term p53-dependent arrest with prolonged induction of p21WAF1 (Di Leonardo et al., 1994). The irreversibility of this arrest can be attributed to the inability of these cells to repair even a small number of double strand DNA breaks, so that the activating signal persists (Huang et al., 1996). The ability to optimally induce p53 in response to IR requires the ATM gene product, which is mutated in the cancer-prone disease ataxia telangectasia (AT) (Kastan et al., 1992; Savitsky et al., 1995). The Cterminal half of the ATM gene contains sequence similarity to several DNA checkpoint genes and a domain homologous to phosphatidylinositol 3-kinase, consistent with the notion that the ATM protein functions as a protein kinase that phosphorylates p53 on Ser15 and thereby contributes to p53 activation (Banin et al., 1998; Canman et al., 1998). It is noteworthy that the phenotypic defects in AT are not merely the result of insucient p53 induction after IR. Cultured cells from AT patients are defective in all known cell cycle checkpoints, which otherwise remain intact in cells that have only lost p53 function. Furthermore, AT cells only appear to be defective in the p53 induction pathway in response to certain types of damage, such as that created by IR and radiomimetic agents (e.g. bleomycin) whereas this pathway remains intact following exposure to ultraviolet radiation or topoisomerase inhibitors (Canman and Kastan, 1997). Recently several ndings have greatly contributed to the understanding of how p53 mediates the G1 arrest. These discoveries were obtained from the search for genes that can be transcriptionally activated by p53. The rst p53-regulated gene identied was gadd45, whose expression is induced in response to a wide variety of DNA-damaging agents (Papathanasiou et al., 1991) but whose induction following ionizing radiation is dependent on normal p53 function (Kastan et al., 1992; Zhan et al., 1994). Although the mechanism by which gadd45 induces growth arrest is currently unknown, GADD45 has been found to associate with proliferating cell nuclear antigen (PCNA) a protein involved in both DNA replication and repair. Gene targeting strategies have recently established p21WAF1 as a critical mediator of the p53-mediated G1 arrest response (El-Deiry et al., 1993). Dierence in the requirement for p21WAF1 in establishing growth arrest following DNA damage, however, has been revealed. Two independent studies have shown that embryonic broblasts obtained from p217/7 mice are only partially defective in their ability to undergo G1 arrest following exposure to IR (Deng et al., 1995; Brugarolas et al., 1995) suggesting that another gene product transactivated by p53 contributes to a complete response. In contrast, the G1 arrest response was completly abrogated in a human tumour cell line expressing wild-type p53 in the absence of p21WAF1 (Waldman et al., 1995). The p21WAF1 was identied as a potent inhibitor of several CDKs, including cyclin D-CDK4/6, cyclin E-CDK2 and cyclin A-CDK2.

7627

Structure and function of p53 protein P May and E May

7628

One potential downstream target of p21 inhibitory activity in G1 is the cell cycle-dependent phosphorylation of the retinoblastoma (RB) protein. RB is in a hypophorylated form during G1. In this state it binds to and sequesters the S phase promoting E2F family of transcription factors (Sherr, 1998; Giaccia and Kastan, 1998). Advancement through the cell cycle is thought to be mediated by sequential phosphorylation of RB by G1 cyclin-dependent kinases, resulting in the release of active E2F, which then leads to the transcriptional activation of genes required for S phase progression (Sherr, 1998; Giaccia and Kastan, 1998). In this pathway p21WAF1 leads to the inhibition of cyclin D-CDK4/6 complexes and subsequent accumulation of the unphosphorylated form of RB, which arrests cell in G1. This G1 arrest is supposed to prevent the replication of damaged DNA and to allow DNA repair before entry into S phase. Rb7/7 mouse embryonic broblasts, however, have only partially lost the G1 arrest response when compared to p537/7 broblasts (Slebos et al., 1998). This suggests that RB alone is not the sole downstream target of p53. Instead p107 and/or p130, both of which form complexes with E2F family members, might also play a role in eliciting p53-mediated growth arrest. Overall, E2F appears to be a critical downstream target within the p53 pathway because overexpression of E2F-1 can overcome IR-induced G1 arrest and p21mediated inhibition of cdk activity (Degregori et al., 1995a,b). Alteration of any of these downstream components may have an eect similar to that of inactivating p53 itself in preventing the pathway from functioning. Activation of cyclin G (Okamoto and Beach, 1994; Zauberman et al., 1995) is also induced by activated p53 protein and might contribute to the control of normal cell cycle progression. After irradiation, wild-type p53 also up-regulates the mdm2 gene. MDM2 inhibits the transcriptional activity of p53 by blocking the association of the transactivation domain of p53 with the transcriptional machinery (Momand et al., 1992; Oliner et al., 1993). However, MDM2 also plays a central role in the regulation of p53 levels in the cells since this oncoprotein targets p53 for rapid degradation (Haupt et al., 1997; Kubbutat et al., 1997), implying an autoregulatory feedback between p53 and MDM2. This regulation may allow cells to recover from G1 arrest. p53 may also act as a spindle-checkpoint or a G2 checkpoint It has been observed that mouse embryo broblast cells develop tetraploid populations after treatment with spindle inhibitors. Most of the cells arrest without re-entering S phase. In contrast, cells that are p53-null and similarly treated continue to cycle and a much larger proportion of cells become tetraploid or octaploid. In this same set of experiments, it has been also noted that about half of the p53-null cell population become tetraploid. These data suggest that p53 is a component of a spindle checkpoint in mice and that the spontaneous development of tetraploid cell populations further indicates that this checkpoint is required during normal cell division (Cross et al., 1995). Moreover, exposure of colorectal cancer (CRC) cells to ionizing radiation results in a cell-cycle arrest in G1

and G2. The G1 block is in part mediated by p21 WAF1 that is transcriptionally controlled by p53. Several studies have suggested that the G2/M block following DNA damage is also p53-dependent (Agarwal et al., 1995; Stewart et al., 1995). Through a quantitative analysis of gene expression patterns in CRC cell lines, it has been shown that 14-3-3s gene is strongly induced by ionizing irradiation. The induction of 14-3-3s is mediated by a p53 element located 1.8 kb upstream of its transcription start site. Exogenous introduction of 14-3-3s into cycling cells results in a G2 arrest (Hermeking et al., 1997). p21WAF1 has also been implicated in regulating DNA replication. p21WAF1 directly interacts with PCNA and inhibits PCNA-dependent DNA replication (but not DNA repair) in in vitro systems (Flores-Rozas et al., 1994; Li et al., 1994; Waga et al., 1994). p53 protein has also been involved in directly inhibiting DNA replication through its association with the single-stranded DNA-binding protein complex RPA (Dutta et al., 1993) that is required for DNA unwinding during replication. p53-mediated apoptosis In certain cell types (e.g. cells of hematopoietic origin), activation of the p53 pathway leads to apoptosis. Stimuli such as DNA damage, withdrawal of growth factors, and ectopic expression of the nuclear oncogenes c-myc, E2F, adenovirus E1A, SV40 large T or HPV E7 stimulates p53-dependent apoptosis. If p53 is mutated or inactivated in some manner, the apoptotic response is not activated and cell proliferation is allowed. Analysis of cells and tissues of p53-null mice has shown that functional p53 is necessary for DNA damage-induced apoptosis of cortical thymocytes (Clarke et al., 1993), myeloid progenitor cells (Lotem and Sachs, 1993), marrow preB cells (Strasser et al., 1994), quiescent peripheral B and T lymphocytes (Strasser et al., 1994), keratinocytes (Ziegler et al., 1994) and small intestine cells (Clarke et al., 1994). The rst indication of the role of p53 in apoptosis was obtained by using a clone of the mouse myeloid cell line M1 lacking endogenous expression of p53. Cells of this clone were stably transfected with a temperature-sensitive mutant which acquires the conformation of wild-type p53 at permissive temperature (328C). In these experiments, it was observed that upon downshift to the permissive temperature, the transfectants underwent a rapid loss of viability with all the characteristics of apoptosis (Yonish-Rouach et al., 1991). Interestingly, the same mutant induces a growth arrest in broblast at the permissive temperature (Michalovitz et al., 1990), whereas the M1 cells continue to divide as p53 induces cell death (YonishRouach et al., 1993). This demonstrates that at least in this cell type the two pathways are separable. Similarly, it was shown (Lin and Benchimol, 1995) that apoptosis and G1 arrest can be uncoupled by addition of a cytokine in culture medium in murine erythroleukaemia cells transfected with a temperature sensitive p53 allele. At the permissive temperature, in response to p53, these cells undergo G1 arrest, apoptosis, and dierentiation. However, addition of a cytokine (erythropoietin, KIT-ligand or interleukin-3) to the culture medium blocks p53-mediated apoptosis and

Structure and function of p53 protein P May and E May

dierentiation but not the p53-mediated G1 arrest. These observations are consistent with the notion that the G1 arrest and apoptosis mediated by p53 represent divergent biological pathways. The importance of p53-mediated apoptosis to protection from transformation has been exemplied in tissue culture where p537/7 cells fail to undergo cell death following introduction of a variety of oncogenes with dramatic enhancement of malignant transformation. Gene mediated by p53 appears to be generally necessary for inducing apoptosis (Sabbatini et al., 1995; Attardi et al., 1996; Yonish-Rouach et al., 1995). While induction of the p21 WAF1 gene has been shown not to be essential for p53-mediated apoptosis, p53 induces the expression of Fas/APO1 (Owen-Schaub et al., 1995), member of the death receptor family, BAX (Miyashita et al., 1994; Selvakumaran et al., 1994; Miyashita and Reed, 1995; Han et al., 1996) the product of which can antagonize the anti-apoptotic activity of Bcl-2 (White, 1996) and IGF-BP3 (Buckbinder et al., 1995), which sequesters the cell survival factor insulin-like growth factor-1 (Baserga, 1994); the gene encoding the cathepsin-D protease (Wu et al., 1998) PAG608 which encodes a nuclear zinc nger protein (Israeli et al., 1997) and the human homolog of the Drosophila sina gene (Nemani et al., 1996). Recently a series of p53-induced genes (PIG genes) have been identied that are predicted to encode proteins that could generate or respond to oxidative stress. These results suggest that p53-mediated apoptosis involves activation of redox-controlling target followed by increase in ROS, oxidative damage to mitochondria and caspase activation (Polyak et al., 1997). Moreover, p53 down-regulates Bcl2 gene whose product acts as an anti-apoptotic agent (White, 1996). In some situations, transcriptional activation by p53 is dispensable for the induction of apoptosis, since mutants in p53 which fail to activate transcription can still induce apoptosis (Chen et al., 1996; Haupt et al., 1995a) and p53-dependent apoptosis can occur in the presence of inhibitors of transcriptional and translation (Caelles et al., 1994). It is well established that induction of the p53 tumour suppressor protein can lead to either cell cycle arrest or apoptosis. How p53 might regulate growth arrest versus apoptosis? Although the mechanism governing the decision of the cell is not elucidated, deletion of p21WAF1 can cause cells that would otherwise undergo p53-dependent cell cycle arrest to undergo apoptosis, instead. Several factors, including the cell type, the presence or absence of survival factors in the external environment, the extent of DNA damage and the levels of p53 are involved in the choice between cell cycle arrest and apoptosis (Chen et al., 1996). Also cooperation between the p53 and RB pathways may be of major importance in determining the biological response to DNA damage. For example, inactivation of RB results in the loss of G1 arrest and induction of apoptosis after DNA damage. This might be explained by the release of E2F (Figure 2), which when overexpressed on its own can induce apoptosis (Haupt et al., 1995b). Thus, modulation of RB and E2F through p53 signalling in response to DNA

damage may play a central role in deciding the choice between cell cycle and arrest. Moreover, it has been shown that the Wilms tumor suppressor gene product, WT1, binds p53 and inhibits p53-dependent apoptosis without aecting p53-dependent growth arrest, also revealing a determining mechanism for p53 downstream events (Maheswaran et al., 1995). p19ARF links the tumour suppressors RB and p53 From the analysis of transgenic mice defective for either p53 or RB genes and from the studies of p53regulated genes has emerged the nding that both p53 and RB are the components of two cooperating pathways. This link between p53 and RB is also illustrated by the fact that oncoproteins encoded by the DNA tumour viruses both cancel RB function to drive cells into S phase and neutralize p53 to prevent host cell apoptosis (White, 1996). The cooperating eects between mutations in RB and p53 for the induction of cancer is illustrated by the following observations. Lens cells in transgenic mice with lens-specic deciency in RB function are found to undergo apoptosis. When both RB and p53 functions are inactivated in lens cells, apoptosis is inhibited and lens tumours are developed (Pan and Griep, 1994; Howes et al., 1994; Morgenbesser et al., 1994). A new link between RB and p53 has been recently obtained by the functional study of murine p19ARF (or its human homolog p14ARF) (Stott et al., 1998). The INK4a-ARF locus encodes two distinct tumour suppressors p16 INK4a (or human p15) and p19 ARF (or human homolog p14). Whereas p16 INK4a restrains cell growth through preventing phosphorylation of the retinoblastoma protein, p19 ARF (or human homolog p14) acts by attenuating MDM2mediated degradation of p53 (Bates et al., 1998; Stott et al., 1998) (Figure 2). Recent data indicate that MDM2 shuttles between the nucleus and the cytoplasm and that nucleo-cytoplasmic shuttling of MDM2 is essential for MDM2's ability to promote p53 degradation. Therefore, MDM2 must export p53 from the nucleus to the cytoplasm where it targets p53 for degradation. Tao and Levine recently showed that expression of p19 ARF blocks the nucleo-cytoplasmic shuttling of MDM2 (Tao and Levine, 1999). Taken together, it appears that p19 ARF could stabilize p53 by protecting it from the MDM2-mediated degradation (Chin et al., 1998). Regulation of p53 function Since activation of p53 (that is its conversion from a latent to an active state) is a central event in response to dierent types of stress or to ectopic surexpression, it is clear that the triggering of p53 activation must be tightly regulated. In fact, both p53 stabilization and activation involve a variety of molecular mechanisms. Phosphorylation and acetylation In addition to causing p53 to accumulate, DNA damage is widely believed to activate p53 as a transcription factor through posttranslational mechanisms (Figure 1). Several protein kinases can phosphorylate distinct domains of p53 in

7629

Structure and function of p53 protein P May and E May

7630

Figure 2 Schematic representation of the events occurring after ionizing irradiation of cells with a wt p53 gene. The outcome of the cellular response is either cell growth arrest or apoptosis, depending on several determinants, including the cell type, the extent of DNA damage, the level of p53 and the genetic background. Ionizing irradiation produces DNA strand breaks that result in accumulation and activation of p53. Optimal induction of p53 after irradiation appears to require a wild type ATM product. Activation of p53 results in the transcriptional transactivation of many genes (GADD45, mdm2, p21 WAF1, BAX, Cyclin G, IGF BP3, PIG3) and in the down-regulation of several genes including BCL2. In cells in which the outcome is G1 arrest, an increase in p21 WAF1 protein contributes to the arrest of cells in G1, through inhibition of cyclin-cdk complexes, with a subsequent accumulation of unphosphorylated form of RB which arrests cells in G1 by preventing the release of E2F from its complex with RB. E2F is required for the G1-to-S transition. In certain cells, the outcome is apoptosis. It is likely to be mediated by increased levels of Fas/APO1/CD95, BAX, IGF BP3 and PIG3. In addition the products of INK4a/ARF locus, namely p16INK4a and p19 ARF are involved at the junction of p53 and RB pathways. p16INK4a is an inhibitor of cyclin-dependent kinase that functions upstream of RB. p19 ARF blocks MDM2 inhibition of p53 activity. (*) p53* stands for activated p53. (**) **p19 ARF is from mouse origin. The human homolog is p14 ARF

vitro. For example the DNA-PK (DNA-activated protein kinase), ATM kinase, (Banin et al., 1998; Canman et al., 1998) and CK1 (casein kinase I-like protein kinase), CDK-activating kinase (CAK) each phosphorylate dierent sites in the N-terminal transactivation domain of p53 (reviewed by Steegenga et al., 1996; Giaccia and Kastan, 1998; Chernov et al., 1998) (Figure 1). These phosphorylation events eect the transactivation function of p53. For example mutation of the three N-terminal serine residues that are kinase targets to alanine in mouse results in a signicant decrease in suppressor function and reduces transactivation (Mayr et al., 1995). Mutations of various serine residues has a dierential eect on the ability of rat p53 to bind to and induce transcription from various promoters (Lohrum and Scheidtmann, 1996) suggesting that specic patterns of serine phosphorylation are likely to occur in response to dierent types of stress, leading in turn to stress-specic patterns of gene expression. The C-terminal regulatory domain of p53 is a target for phosphorylation by casein kinase 2 (CK2), protein kinase C (PKC) (Herrmann et al., 1991) cdk

(Wang and Prives, 1995), and p34cdc2 (Milner et al., 1990) p34cdc2 can bind to p53 in vivo (Milner et al., 1990) and phosphorylate it in vitro (Wang and Prives, 1995) (Figure 1). Phosphorylation by cell cycle-dependent protein kinases suggests that the activity of p53 is regulated dierentially during the cell cycle. Phosphorylation by PKC and casein kinase2 (CK2) in vitro stimulate p53 to bind to DNA (Hupp et al., 1993) probably through a conformational change of the protein. However, the activation of PKC by phorbol ester does not cause a change in the phosphorylation of the C-terminal of mouse (Milne et al., 1996), suggesting that the PKC site may be phosphorylated constitutively. Experiments with the human p53 mutant Ser392 Ala showed that phosphorylation of C-terminal domain by casein kinase 2 is not required for p53 to transactivate target genes (Fiscella et al., 1994). Taken together, the data suggest that, in vivo the C-terminal domain of p53 may have functions in addition to regulating DNA binding and that constitutive phosphorylation of this domain may be required to elicit such functions.

Structure and function of p53 protein P May and E May

Regulation of p53 exerted by the C-terminal tail of p53 Modications to the carboxy-terminal, including deletion of the last 30 amino-acids (Hupp et al., 1992), protein-binding (Hupp et al., 1992), phosphorylation of Ser315, Ser318 and Ser393 (Hupp et al., 1995; Takenaka et al., 1995; Wang and Prives, 1995) and recently acetylation by CREB-binding protein (CBP)/p300 (Gu and Roeder, 1997) were shown to enhance sequencespecic DNA-binding of wild-type p53, possibly by inhibition of p53's non-sequence-specic DNA-binding activity (Anderson et al., 1997). CBP and p300 are closely related histone acetyl-transferases (HATs) that interact with p53 through its amino terminus and function as coactivators for p53-mediated transcription (Avantaggiati et al., 1997; Lill et al., 1997). CBP/p300 are complexed with another HAT, p300/CBP-associated factor PCAF (Yang et al., 1996). These recent ndings suggest that activation of sequence-specic DNAbinding by p53 in response to DNA damage may be mediated through CBP/p300 and/or PCAF. Gu and Roeder showed that p300 acetylates the carboxyl terminus of p53 (Gu and Roeder, 1997). Recently Sakaguchi et al. showed that PCAF acetylated p53 in vitro on Lys320 (Sakaguchi et al., 1998). These posttranslational modications are shown in Figure 1. Large DNA-mediated negative regulation of specic DNA binding by p53 It was recently reported (Anderson et al., 1997) that the interactions of the C terminus with genomic DNA in vivo would prevent p53 binding to specic promoters and that activation of specic DNA binding results primarily from blocking the negative regulation associated with the binding of large nucleic acids. It was suggested that cells may have very specic activator molecules that are amplied in response to appropriate signals. Independent regulation of stabilization and activation of p53 An increase in transactivation due to p53 with no increase in the level of the protein was observed in cells with low doses of UV radiation (Bellamy et al., 1997) and microinjection of an antibody to the C-terminal domain also stimulated p53-dependent transcription, even in the absence of DNA damage (Hupp et al., 1995). Chernov and Stark (1997) found that sodium salicylate, which inhibits protein kinases, inhibits the activation of p53, with no signicant eect on the accumulation of the protein. In addition, it was reported that the lifetime and activities of p53 are likely to be regulated (Chernov et al., 1998) independently by dierent phosphorylation events. Redox modulation The DNA-binding activity and conformation of p53 are also regulated by the redox state of the protein such that oxidation inhibits DNA binding, whereas reduction favors it (Hainaut et al., 1993a,b; Hupp et al., 1993; Rainwater et al., 1995). Several cysteine residues in the core DNA-binding domain have been involved in zinc coordination, and mutational analysis has allowed to identify cysteines at positions 173, 235 and 239 which participate in DNA binding and are also critical for transcriptional activation and suppression of transformation (Rainwater et al., 1995). These results have profound implications with respect to the regulation of p53

after genotoxic stress conditions, including ionizing radiation, and p53 might be regulated by both the presence of oxygen intermediates and the antioxydant defense mechanism of the cell (Hainaut et al., 1993a,b). In addition, studies from several groups have also suggested that redox-sensitive proteins such as HIF-1a (An et al., 1998) and Ref-1 (Jayaraman et al., 1997) can interact with p53 and alter its activity or level. Mutation-independent mechanism of p53 inactivation Besides acquiring a mutation within the p53 gene, there are a number of mutation independent mechanisms of functional inactivation of p53. A rst example is the sequestration of wild-type p53 protein by another viral or cellular protein which abrogates its suppressive action. Another mutation independent mechanism of p53 inactivation is nuclear exclusion of wild-type p53, which is often observed in certain types of tumour (Stommel et al., 1999; and references therein). This reects the existence of a cellular mechanism responsible for p53 nuclear translocation and cytoplasmic sequestration. In fact p53 was shown to be actively transported through the nuclear membrane in both directions. Nuclear localization (Shaulsky et al., 1990b) and nuclear export signals have been identied (Stommel et al., 1999). Dierences in the intracellular localization of wild-type protein during the cell cycle under normal growth conditions suggest that such a control mechanism could be cell cycle regulated (Schaulsky et al., 1990a; David-Pfeuty et al., 1996). The emerging p53 family A further level of p53 complexity emerged when the presence was reported of at least two new families of protein (p73 and p63), very similar to p53 and displaying striking homology concerning the transactivation, DNA binding, and oligomerization domains. They dier considerably at the C-terminus, with both p73 (Kaghad et al., 1997) and p63 (Schmale and Bamberger, 1997) having long extensions whereas p63 has in addition a 40 amino acid N-terminal extension. The nding of p53 homologs may explain the fact that p53-null mice develop mostly normally. There is extreme conservation of critical residues in MDM2 binding region. Some isoforms of p63 and p73 can transactivate many of p53's targets and some forms also induce cell cycle arrest. Nonetheless, data on dierences between these proteins and p53 are emerging (Kaelin, 1999). For example p73 is not induced by DNA damage and is not targeted by viral oncoproteins such as SV40 T antigen, adenovirus E1B-55 kDa and human papilloma virus E6. In marked contrast to p53, neither p73 nor p63 appears to be frequently mutated in human cancer. Additional studies are required to further elucidate the role, if any, that p73 and p63 play in cell growth control and carcinogenesis. Conclusions Over the past 20 years, there has been tremendous progress in elucidating the biochemical and biological properties of the p53 protein. The key role of p53 in the process that leads to cancer as well as its central

7631

Structure and function of p53 protein P May and E May

7632

role as cell cycle regulator activated by DNA damage are now conclusive. It may be of value to elucidate the global conformation of the entire structure of p53. In this connection, it is of interest to discuss some recent ndings concerning p53 as a paradigmatic multidomain protein. p53 appears to be composed of at least four functional domains which are all involved in modulating the protein's function as transcription factor. At least two domains, the site-specic DNA-binding and the tetramerization domain, are independent folding regions. The question arises whether the folding of these domains is the same when these fragments are isolated or when they are integrated into the entire chain. In fact it has been shown that the p53 protein in its tetrameric form can be converted between latent and activated forms by a concerted allosteric activation, showing that distinct, independent domains interact
References Abarzua P, Lo Sardo JE, Gubler ML, Spathis R, Lu YA, Felix A and Neri A. (1996). Oncogene, 13, 2477 2482. Agarwal ML, Agarwal A, Taylor WR and Stark GR. (1995). Proc. Natl. Acad. Sci. USA, 92, 8493 8497. Ago SN, Hou J, Linzer DI and Wu B. (1993). Science, 259, 84 87. Almog N and Rotter V. (1997). Biochim. Biophys. Acta, 1333, F1 F27. An WG, Kanekal M, Simon MC, Maltepe E, Blagosklonny MV and Neckeres LM. (1998). Nature, 392, 405 408. Anderson ME, Woelker B, Reed M, Wang P and Tegtmeyer P. (1997). Mol. Cell. Biol., 17, 6255 6264. Angelopoulou K, Diamandis EP, Sutherland DJ, Kellen JA and Bunting PS. (1994). Int. J. Cancer, 58, 480 487. Attardi LD, Lowe SW, Brugarolas J and Jacks T. (1996). EMBO J., 15, 3693 3701. Avantaggiati ML, Ogryzko V, Gardner K, Giordano A, Levine AS and Kelly K. (1997). Cell, 89, 1175 1184. Bakalkin G, Yakovleva T, Selivanova G, Magnusson KP, Szekely L, Kiseleva E, Klein G, Terenius L and Wiman KG. (1994). Proc. Natl. Acad. Sci. USA, 91, 413 417. Baker SJ, Fearon ER, Nigro JM, Hamilton SR, Preisinger AC, Jessup JM, vanTuinen P, Ledbetter DH, Barker DF and Nakamura Y. (1989). Science, 244, 217 221. Baker SJ, Markowitz S, Fearon ER, Willson JK and Vogelstein B. (1990). Science, 249, 912 915. Banin S, Moyal L, Shiel SY, Taya Y, Anderson CW, Chessa L, Smorodinsky NI, Prives C, Reiss Y, Shiloh Y and Ziv Y. (1998). Science, 281, 1674 1677. Bargonetti J, Friedman PN, Kern SE, Vogelstein B and Prives C. (1991). Cell, 65, 1083 1091. Bargonetti J, Manfredi JJ, Chen X, Marshak DR and Prives C. (1993). Genes Dev., 7, 2565 2574. Baserga R. (1994). Cell, 79, 927 930. Bates S, Phillips AC, Clark PA, Stott F, Peters G, Ludwig RL and Vousden KH. (1998). Nature, 395, 124 125. Beaudry GA, Bertelsen AH and Sherman MI. (1996). Curr. Opin. Biotechnol., 7, 592 600. Bellamy COC, Prost S, Wyllie AH and Harrison DJ. (1997). J. Pathol., 193, 177 181. Ben David Y, Prideaux VR, Chow V, Benchimol S and Bernstein A. (1988). Oncogene, 3, 179 185. Benchimol S, Lamb P, Crawford LV, Sheer D, Shows TB, Bruns GA and Peacock J. (1985). Somat. Cell Mol. Genet., 11, 505 510. Bennett WP, Hussain SP, Vahakangas KH, Khan MA, Shields PG and Harris CC. (1999). J. Pathol., 187, 8 18.

with each other through a global conformational change (Halazonetis and Kandil, 1993; Hupp and Lane, 1994). Another point of interest is the fact that the presently known sites of post-translational modifications are all located on the outside of the specic DNA binding and tetramerization domains, the most structured domains. The sites of post-translational modications appear to be essentially located in the regulatory regions of the p53 molecule (see Figure 1). Finally, a fundamental aspect of p53 is that, although not essential for the cell cycle per se, it has been shown to participate in cell cycle checkpoint and apoptosis functions that regulate homeostatic tissue renewal. As such p53 appears to be at the head of key cellular pathways which provide many critical targets for the rational design of anti-cancer agents to neutralize the consequences of alterations in these pathways.

Bouck N. (1996). Biochim. Biophys. Acta, 1287, 63 66. Bourdon JC, Deguin-Chambon V, Lelong JC, Dessen P, May P, Debuire B and May E. (1997). Oncogene, 14, 85 94. Brain R and Jenkins JR. (1994). Oncogene, 9, 1775 1780. Braithwaite AW, Sturzbecher HW, Addison C, Palmer C, Rudge K and Jenkins JR. (1987). Nature, 329, 458 460. Brash DE, Rudolph JA, Simon JA, Lin A, McKenna GJ, Baden HP, Halperin AJ and Ponten J. (1991). Proc. Natl. Acad. Sci. USA, 88, 10124 10128. Bressac B, Kew M, Wands J and Ozturk M. (1991). Nature, 350, 429 431. Brugarolas J, Chandrasekaran C, Gordon JI, Beach D, Jacks T and Hannon GI. (1995). Nature, 377, 552 557. Buchhop S, Gibson MK, Wang XW, Wagner P, Sturzbecher HW and Harris CC. (1997). Nucleic Acids Res., 25, 3868 3874. Buckbinder L, Talbott R, Velascomiguel S, Takenaka I, Faha B, Seizinger BR and Kley N. (1995). Nature, 377, 646 649. Caelles C, Helmberg A and Karin M. (1994). Nature, 370, 220 223. Canman CE and Kastan MB. (1997). Adv. Pharmacol., 41, 429 460. Canman CE, Lim DS, Cimprich KA, Yaya Y, Tamai K, Sakaguchi K, Appella E, Kastan M and Siliciano JD. (1998). Science, 281, 1677 1679. Caron de Fromentel C, May-Levin F, Mouriesse H, Lemerle J, Chandrasekaran K and May P. (1987). Int. J. Cancer, 39, 185 189. Caron de Fromentel C and Soussi T. (1992). Genes Chromosomes Cancer, 4, 1 15. Chang C, Simmons DT, Martin MA and Mora PT. (1979). J. Virol., 31, 463 471. Chen PL, Chen YM, Bookstein R and Lee WH. (1990). Science, 250, 1576 1580. Chen X, Ko LJ, Jayaraman L and Prives C. (1996). Genes Dev., 10, 2438 2451. Chernov MV, Ramana CV, Adler VV and Stark GR. (1998). Proc. Natl. Acad. Sci. USA, 95, 2284 2289. Chernov MV and Stark GR. (1997). Oncogene, 14, 2503 2510. Chin L, Pomerantz J and Depinho RA. (1998). Trends Biochem. Sci., 23, 291 296. Cho YJ, Gorina S, Jerey PD and Pavletich NP. (1994). Science, 265, 346 355.

Structure and function of p53 protein P May and E May

Chow V, Ben David Y, Bernstein A, Benchimol S and Mowat M. (1987). J. Virol., 61, 2777 2781. Clarke AR, Gledhill S, Hooper ML, Bird CC and Wyllie AH. (1994). Oncogene, 9, 1767 1773. Clarke AR, Purdie CA, Harrison DJ, Morris RG, Bird CC, Hooper ML and Willie AH. (1993). Nature, 362, 849 852. Clore GM, Omichinski JG, Sakaguchi K, Zambrano N, Sakamoto H, Appella E and Gronenborn AM. (1994). Science, 265, 386 391. Clore GM, Omichinski JG, Sakaguchi K, Zambrano N, Sakamoto H, Appella E and Gronenborn AM. (1995). Science, 267, 1515 1516. Cook A and Milner J. (1990). Br. J. Cancer, 61, 548 552. Crawford L, Leppard K, Lane D and Harlow E. (1982). J. Virol., 42, 612 620. Cross SM, Sanchez CA, Morgan CA, Schmike MK, Ramel S, Idzerda RL, Raskind WH and Reid BJ. (1995). Science, 267, 1353 1356. Dang CV and Lee WM. (1989). J. Biol. Chem., 264, 18019 18023. David-Pfeuty T, Chakrani F, Ory K and Nouvian Dooghe Y. (1996). Cell Growth Dier., 7, 1211 1225. Degregori J, Kowalik T and Nevins JR. (1995a). Mol. Cell. Biol., 15, 4215 4224. Degregori J, Leone G, Ohtani K, Miron A and Nevins JR. (1995b). Genes Dev., 9, 2873 2887. Deleo AB, Jay G, Appella E, Dubois GC, Law LW and Old LJ. (1979). Proc. Natl. Acad. Sci. USA, 76, 2420 2424. Deng CX, Zhang PM, Harper JW, Elledge SJ and Leder P. (1995). Cell, 82, 675 684. Di Leonardo A, Linke SP, Clarkin K and Wahl GM. (1994). Genes Dev., 8, 2540 2551. Diller L, Kassel J, Nelson CE, Gryka MA, Litwak M, Gebhardt M, Bressac B, Ozturk M, Baker SJ, Vogelstein B and Friend SH. (1990). Mol. Cell. Biol., 10, 5772 5782. Dittmer D, Pati S, Zambetti G, Chu S, Teresky AK, Moore AM, Finlay C and Levine AJ. (1993). Nat. Genet., 4, 42 46. Donehower LA, Harvey M, Slagle BL, Mcarthur MJ, Mongomery Jr CA, Butel JS and Bradley A. (1992). Nature, 356, 215 221. Dutta A, Ruppert JM, Aster JC and Winchester E. (1993). Nature, 365, 79 82. Ehrhart JC, Duthu A, Ullrich S, Appella E and May P. (1988). Oncogene, 3, 595 603. El-Deiry WS, Kern SE, Pietenpol JA, Kinzler KW and Vogelstein B. (1992). Nat. Genet., 1, 45 49. El-Deiry WS, Tokino T, Velculescu VE, Levy DL, Parsons R, Trent JM, Lin D, Mercer WE, Kinzer KW and Vogelstein B. (1993). Cell, 75, 817 825. Eliyahu D, Michalovitz D, Eliyahu S, Pinhasi-Kimhi O and Oren M. (1989). Proc. Natl. Acad. Sci. USA, 86, 8763 8767. Eliyahu D, Raz A, Gruss P, Givol D and Oren M. (1984). Nature, 312, 646 649. Fields S and Jang JA. (1990). Science, 249, 1046 1049. Finlay CA, Hinds PW and Levine AJ. (1989). Cell, 57, 1083 1093. Fiscella M, Zambrano N, Ullrich SJ, Unger T, Lin D, Cho B, Mercer WE, Anderson CW and Appella E. (1994). Oncogene, 9, 3249 3257. Flores-Rozas H, Kelman Z, Dean FB, Pan ZQ, Harper JW, Elledge SJ, O'Donnell M and Hurwitz J. (1994). Proc. Natl. Acad. Sci. USA, 91, 8655 8659. Ford JM and Hanawalt PC. (1995). Proc. Natl. Acad. Sci. USA, 92, 8876 8880. Funk WD, Pak DT, Karas RH, Wright WE and Shay JW. (1992). Mol. Cell. Biol., 12, 2866 2871. Gannon JV, Greaves R, Iggo R and Lane DP. (1990). EMBO J., 9, 1595 1602. Gannon JV and Lane DP. (1987). Nature, 329, 456 458.

Giaccia AJ and Kastan MB. (1998). Genes Dev., 12, 2973 2983. Ginsberg D, Mechta F, Yaniv M and Oren M. (1991). Proc. Natl. Acad. Sci. USA, 88, 9979 9983. Gorina S and Pavletich NP. (1996). Science, 274, 1001 1005. Gu W and Roeder RG. (1997). Cell, 90, 595 606. Hainaut P and Milner J. (1993a). Cancer Res., 53, 4469 4473. Hainaut P and Milner J. (1993b). Cancer Res., 53, 1739 1742. Halazonetis TD and Kandil AN. (1993). EMBO J., 12, 5057 5064. Han JH, Sabbatini P, Perez D, Rao L, Modha D and White E. (1996). Genes Dev., 10, 461 477. Hanawalt PC. (1987). Environ. Health Perspect., 76, 9 14. Harris CC and Hollstein M. (1993). N. Engl. J. Med., 329, 1318 1327. Harvey M, McArthur MJ, Montgomery CA, Butel JS, Bradley A and Donehower LA. (1993). Nat. Genet., 5, 225 229. Haupt Y, Maya R, Kazaz A and Oren M. (1997). Nature, 387, 296 299. Haupt Y, Rowan S and Oren M. (1995b). Oncogene, 10, 1563 1571. Haupt Y, Rowan S, Shaulian E, Vousden KH and Oren M. (1995a). Genes Dev., 9, 2170 2183. He Z, Brinton BT, Greenblatt J, Hassell JA and Ingles CJ. (1993). Cell, 73, 1223 1232. Hermeking H, Lengauer C, Polyak K, He TC, Zhang L, Thiagalingam S and Kinzler KW. (1997). Mol. Cell, 1, 3 11. Herrmann CP, Kraiss S and Montenarh M. (1991). Oncogene, 6, 877 884. Hinds PW, Finlay CA, Frey AB and Levine AJ. (1987). Mol. Cell. Biol., 7, 2863 2869. Hollstein M, Rice K, Greenblatt MS, Soussi T, Fuchs R, Sorlie T, Hovig E, Smithsorensen B, Montesano R and Harris CC. (1994). Nucleic Acids Res., 22, 3551 3555. Hollstein M, Shomer B, Greenblatt M, Soussi T, Hovig E, Montesano R and Harris CC. (1996). Nucleic Acids Res., 24, 141 146. Horikoshi N, Usheva A, Chen JD, Levine AJ, Weinmann R and Shenk T. (1995). Mol. Cell. Biol., 15, 227 234. Howes KA, Ransom LN, Papermaster DS, Lasudry JGH, Albert DM and Windle JJ. (1994). Genes Dev., 8, 1300 1310. Hsu IC, Metcalf RA, Sun T, Welsh JA, Wang NJ and Harris CC. (1991). Nature, 350, 427 428. Huang LC, Clarkin KC and Wahl GM. (1996). Proc. Natl. Acad. Sci. USA, 93, 4827 4832. Hupp TR and Lane DP. (1994). Curr. Biol., 4, 865 875. Hupp TR and Lane DP. (1995). J. Biol. Chem., 270, 18165 18174. Hupp TR, Meek DW, Midgley CA and Lane DP. (1992). Cell, 71, 875 886. Hupp TR, Meek DW, Midgley CA and Lane DP. (1993). Nucleic Acids Res., 21, 3167 3174. Hupp TR, Sparks A and Land DP. (1995). Cell, 83, 237 245. Isobe M, Emanuel BS, Givol D, Oren M and Croce CM. (1986). Nature, 320, 84 85. Israeli D, Tessler E, Haupt Y, Elkeles A, Wilder S, Amson R, Telerman A and Oren M. (1997). EMBO J., 16, 4384 4392. Iwabuchi K, Bartel PL, Li B, Marraccino R and Fields S. (1994). Proc. Natl. Acad. Sci. USA, 91, 6098 6102. Jacks T, Remington L, Williams BO, Schmitt EM, Halachmi S, Bronson R and Weinberg RA. (1994). Curr. Biol., 4, 1 7. Jayaraman L, Murthy KG, Zhu C, Curran T, Xanthoudakis S and Prives C. (1997). Genes Dev., 11, 558 570.

7633

Structure and function of p53 protein P May and E May

7634

Jerey PD, Gorina S and Pavletich NP. (1995). Science, 267, 1498 1502. Jenkins JR, Chumakov P, Addison C and Sturzbecher HW. (1988). J. Virol., 62, 3903 3906. Jenkins JR, Rudge K, Chumakov P and Currie GA. (1985). Nature, 317, 816 818. Jenkins JR, Rudge K and Currie GA. (1984). Nature, 312, 651 654. Kaczmarek L, Oren M and Baserga R. (1986). Exp. Cell Res., 162, 268 272. Kaelin WG. (1999). J. Natl. Cancer Inst., 91, 594 598. Kaghad M, Bonnet H, Yang A, Creancier L, Biscan JC, Valent A, Minty A, Chalon P, Lelias JM, Dumont X, Ferrara P, McKeon F and Caput D. (1997). Cell, 90, 809 819. Kastan MB, Onyekwere O, Sidransky D, Vogelstein B and Craig RW. (1991). Cancer Res., 51, 6304 6311. Kastan MB, Zhan Q, El-Deiry WS, Carrier F, Jacks T, Walsh WV, Plunkett BS, Vogelstein B and Fornace AJ. (1992). Cell, 71, 587 597. Kern SE, Kinzler KW, Bruskin A, Jarosz D, Friedman P, Prives C and Vogelstein B. (1991). Science, 252, 1708 1711. Kessis TD, Slebos RJ, Nelson WG, Kastan MB, Plunkett BS, Han SM, Lorincz AT, Hedrick L and Cho KR. (1993). Proc. Natl. Acad. Sci. USA, 90, 3988 3992. Ko LJ and Prives C. (1996). Genes Dev., 10, 1054 1072. Kraiss S, Quaiser A, Oren M and Montenarh M. (1988). J. Virol., 62, 4737 4744. Kress M, May E, Cassingena R and May P. (1979). J. Virol., 31, 472 483. Kubbutat MH, Jones SN and Vousden KH. (1997). Nature, 387, 299 303. Kubbutat MH, Ludwig RL, Ashcroft M and Vousden KH. (1998). Mol. Cell. Biol., 18, 5690 5698. Kuerbitz SJ, Plunkett BS, Walsh WV and Kastan MB. (1992). Proc. Natl. Acad. Sci. USA, 89, 7491 7495. Lane D and Crawford L. (1979). Nature, 278, 261 263. Lavigueur A, Maltby V, Mock D, Rossant J, Pawson T and Bernstein A. (1989). Mol. Cell. Biol., 9, 3982 3991. Lechner MS, Mack DH, Finicle AB, Crook T, Vousden KH and Laimins LA. (1992). EMBO J., 11, 3045 3052. Levine AJ. (1997). Cell, 88, 323 331. Li G, Mitchell DL, Ho VC, Reed JC and Tron VA. (1996). Am. J. Pathol., 148, 1113 1123. Li R and Botchan MR. (1993). Cell, 73, 1207 1221. Li R, Waga S, Hannon GJ, Beach D and Stillman B. (1994). Nature, 371, 534 537. Li X and Cono P. (1996). J. Virol., 70, 4509 4516. Lill NL, Grossman SR, Ginsberg D, DeCaprio J and Livingston DM. (1997). Nature, 387, 823 827. Lin D, Shields MT, Ullrich SJ, Appella E and Mercer WE. (1992). Proc. Natl. Acad. Sci. USA, 89, 9210 9214. Lin JY, Chen Jd, Elenbaas B and Levine AJ. (1994). Genes Dev., 8, 1235 1246. Lin YP and Benchimol S. (1995). Mol. Cell. Biol., 15, 6045 6054. Linzer DIH and Levine AJ. (1979). Cell, 17, 43 52. Lohrum M and Scheidtmann KH. (1996). Oncogene, 13, 2527 2539. Lotem J and Sachs L. (1993). Blood, 82, 1092 1096. Lowe SW, Bodis S, McClatchey A, Remington L, Ruley HE, Fisher DE, Housman DE and Jacks T. (1994). Science, 266, 807 810. Lu H and Levine AJ. (1995). Proc. Natl. Acad. Sci. USA, 92, 5154 5158. Lu X and Lane DP. (1993). Cell, 75, 765 778. Mack DH, Vartikar J, Pipas JM and Laimins LA. (1993). Nature, 363, 281 283. Maheswaran S, Englert C, Benett P, Heinrich G and Haber DA. (1995). Genes Dev., 9, 2143 2146.

Malkin D, Li FP, Strong LC, Fraumeni Jr JF, Nelson CE, Kim DH, Kassel J, Gryka MA, Bischo FZ and Tainsky MA. (1990). Science, 250, 1233 1238. Maltzman W and Czyzyk L. (1984). Mol. Cell. Biol., 4, 1689 1694. Marshall CJ. (1991). Cell, 64, 313 326. Mayr GA, Reed M, Wang P, Wang Y, Schwedes JF and Tegtmeyer P. (1995). Cancer Res., 55, 2410 2417. McBride OW, Merry D and Givol D. (1986). Proc. Natl. Acad. Sci. USA, 83, 130 134. Melero JA, Stitt DT, Mangel WF and Carroll RB. (1979). Virology, 93, 466 480. Mercer WE, Nelson D, Deleo AB, Old LJ and Baserga R. (1982). Proc. Natl. Acad. Sci. USA, 79, 6309 6312. Michalovitz D, Halevy O and Oren M. (1990). Cell, 62, 671 680. Midgley CA and Lane DP. (1997). Oncogene, 15, 1179 1189. Miller C, Mohandas T, Wolf D, Prokocimer M, Rotter V and Koeer HP. (1986). Nature, 319, 783 784. Milne DM, McKendrick L, Jardine LJ, Deacon E, Lord JM and Meek DW. (1996). Oncogene, 13, 205 211. Milner J. (1984). Nature, 310, 143 145. Milner J, Cook A and Mason J. (1990). EMBO J., 9, 2885 2889. Milner J, Medcalf EA and Cook AC. (1991). Mol. Cell. Biol., 11, 12 19. Milner J and Milner S. (1981). Virology, 112, 785 788. Miyashita T, Krajewski S, Krajewska M, Wang HG, Lin HK, Liebermann DA, Homan B and Reed JC. (1994). Oncogene, 9, 1799 1805. Miyashita T and Reed JC. (1995). Cell, 80, 293 299. Moll U, Laquaglia M, Benard J and Riou G. (1995). Proc. Natl. Acad. Sci. USA, 92, 4407 4411. Moll U, Ostermeyer AG, Haladay R, Winkeld B, Frazier M and Zambetti G. (1996). Mol. Cell. Biol., 16, 1126 1137. Moll U, Riou G and Levine AJ. (1992). Proc. Natl. Acad. Sci. USA, 89, 7262 7266. Momand J, Zambetti GP, Olson DC, George D and Levine AJ. (1992). Cell, 69, 1237 1245. Mora PT, Chandrasekaran K and McFarland VW. (1980). Nature, 288, 722 724. Morgenbesser SD, Williams BO, Jacks T and Depinho RA. (1994). Nature, 371, 72 74. Mowat M, Cheng A, Kimura N, Bernstein A and Benchimol S. (1985). Nature, 314, 633 636. Munroe DG, Rovinski B, Bernstein A and Benchimol S. (1988). Oncogene, 2, 621 624. Murphy M, Hinman A and Levine AJ. (1996). Genes Dev., 10, 2971 2980. Nelson WG and Kastan MB. (1994). Mol. Cell. Biol., 14, 1815 1823. Nemani M, Linares-Cruz G, Bruzzoni-Giovanelli H, Roperch JP, Tuynder M, Bougueleret L, Cherif D, Medhioub M, Pasturaud P, Alvaro V, der Sarkissan H, Cazes L, Le Paslier D, Le G, Israeli D, Dausset J, Sigaux F, Chumakov I, Oren M, Calvo F, Amson RB, Cohen D and Telerman A. (1996). Proc. Natl. Acad. Sci. USA, 93, 9039 9042. Nigro JM, Baker SJ, Preisinger AC, Jessup JM, Hostetter R, Cleary K, Bigner SH, Davidson N, Baylin S and Devilee P. (1989). Nature, 342, 705 708. Oberosler P, Hloch P, Ramsperger U and Stahl H. (1993). EMBO J., 12, 2389 2396. Okamoto K and Beach D. (1994). EMBO J., 13, 4816 4822. Okorokov AL, Ponchel F and Milner J. (1997). EMBO J., 16, 6008 6017. Oliner JD, Kinzler KW, Meltzer PS, George DL and Vogelstein B. (1992). Nature, 358, 80 83. Oliner JD, Pietenpol JA, Thiagalingam S, Gyuris J, Kinzler KW and Vogelstein B. (1993). Nature, 362, 857 860.

Structure and function of p53 protein P May and E May

Oren M. (1985). Biochim. Biophys. Acta, 823, 67 78. Oren M, Reich NC and Levine AJ. (1982). Mol. Cell. Biol., 2, 443 449. Owen-Schaub LB, Zhang W, Cusack JC, Angelo LS, Santee SM, Fujiwara T, Roth JA, Deisseroth AB, Zhang WW, Kruzel E and Radinsky R. (1995). Mol. Cell. Biol., 15, 3032 3040. Pan H and Griep AE. (1994). Genes Dev., 8, 1285 1299. Papathanasiou MA, Kerr NC, Robbins JH, McBride OW, Alamo Jr I, Barrett SF, Hickson ID and Fornace Jr AJ. (1991). Mol. Cell. Biol., 11, 1009 1016. Parada LF, Land H, Weinberg RA, Wolf D and Rotter V. (1984). Nature, 312, 649 651. Pavletich NP, Chambers KA and Pabo CO. (1993). Genes Dev., 7, 2556 2564. Pietenpol JA, Tokino T, Thiagalingam S, Eldeiry WS, Kinzler KW and Vogelstein B. (1994). Proc. Natl. Acad. Sci. USA, 91, 1998 2002. Pinhasi-Kimhi O, Michalovitz D, Ben Zeev A and Oren M. (1986). Nature, 320, 182 184. Polyak K, Xia Y, Zweier JL, Kinzler KW and Vogelstein B. (1997). Nature, 389, 300 305. Purdie CA, Harrison DJ, Peter A, Dobbie L, White S, Howie SEM, Salter DM, Bird CC, Wyllie AH, Hooper ML and Clarke AR. (1994). Oncogene, 9, 603 609. Ragimov N, Krauskopf A, Navot N, Rotter V, Oren M and Aloni Y. (1993). Oncogene, 8, 1183 1193. Rainwater R, Parks D, Anderson ME, Tegtmeyer P and Mann K. (1995). Mol. Cell. Biol., 15, 3892 3903. Raycroft L, Wu HY and Lozano G. (1990). Science, 249, 1049 1051. Reich NC and Levine AJ. (1984). Nature, 308, 199 201. Reich NC, Oren M and Levine AJ. (1983). Mol. Cell. Biol., 3, 2143 2150. Reznikov MV, Fidler R, Rubtsov PM, Skriabin KG, Chumakov PM, Prasolov VS and Baev AA. (1989). Mol. Biol. (Mosk), 23, 1692 1699. Rogel A, Popliker M, Webb CG and Oren M. (1985). Mol. Cell. Biol., 5, 2851 2855. Rotter V, Aloni-Grinstein R, Schwartz D, Elkind NB, Simons A, Wolkowicz R, Lavigne M, Beserman P, Kapon A and Goldnger N. (1994). Semin. Cancer Biol., 5, 229 236. Rotter V, Witte ON, Coman R and Baltimore D. (1980). J. Virol., 36, 547 555. Rovinsky B, Munroe D, Peacock J, Mawat M, Berstein A and Benchimol S. (1987). Mol. Cell. Biol., 7, 853. Ruppert JM and Stillman B. (1993). Mol. Cell. Biol., 13, 3811 3820. Sabbatini P, Lin JY, Levine AJ and White E. (1995). Genes Dev., 9, 2184 2192. Sakaguchi K, Herrera JE, Saito S, Miki T, Bustin M, Vassilev A, Anderson CW and Appella E. (1998). Genes Dev., 12, 2831 2841. Sakamuro D, Sabbatini P, White E and Prendergast GC. (1997). Oncogene, 15, 887 898. Sarnow P, Ho YS, Williams J and Levine AJ. (1982). Cell, 28, 387 394. Savitsky K, Bar-Shira A, Gilad S, Rotman G, Ziv Y, Vanagaite L, Tagle DA, Smith S, Uziel T and Sfez S. (1995). Science, 268, 1749 1753. Schener M, Werness BA, Huibregtse JM, Levine AJ and Howley PM. (1990). Cell, 63, 1129 1136. Schmale H and Bamberger C. (1997). Oncogene, 15, 1363 1367. Scolnick DM, Chehab NH, Stavridi ES, Lien MC, Caruso L, Moran E, Berger SL and Halazonetis TD. (1997). Cancer Res., 57, 3693 3696. Selivanova G, Iotsova V, Okan I, Fritsche M, Strom M, Groner B, Grafstrom RC and Wiman KG. (1997). Nat. Med., 3, 632 638.

Selivanova G, Ryabchenko L, Jansson E, Iotsova V and Wiman KG. (1999). Mol. Cell. Biol., 19, 3395 3402. Selvakumaran M, Lin HK, Miyashita T, Wang HG, Krajewski S, Reed JC, Homan B and Liebermann D. (1994). Oncogene, 9, 1791 1798. Seto E, Usheva A, Zambetti GP, Momand J, Horikoshi N, Weinmann R, Levine AJ and Shenk T. (1992). Proc. Natl. Acad. Sci. USA, 89, 12028 12032. Shaulsky G, Ben Ze'ev A and Rotter V. (1990a). Oncogene, 5, 1707 1711. Shaulsky G, Goldnger N, Ben Ze'ev A and Rotter V. (1990b). Mol. Cell. Biol., 10, 6565 6577. Shaw P, Bovey R, Tardy S, Sahli R, Sordat B and Costa J. (1992). Proc. Natl. Acad. Sci. USA, 89, 4495 4499. Shaw P, Freeman J, Bovey R and Iggo R. (1996). Oncogene, 12, 921 930. Sherr CJ. (1998). Genes Dev., 12, 2984 2991. Shohat O, Greenberg M, Reisman D, Oren M and Rotter V. (1987). Oncogene, 1, 277 283. Siliciano JD, Canman CE, Taya Y, Sakaguchi K, Appella E and Kastan MB. (1997). Genes Dev., 11, 3471 3481. Slebos RJ, Resnick MA and Taylor JA. (1998). Cancer Res., 58, 5333 5336. Smith ML, Chen IT, Zhan Q, Bae I, Chen CY, Gilmer TM, Kastan MB, O'Connor PM and Fornace Jr AJ. (1994). Science, 266, 1376 1380. Soussi T, Caron de Fromentel C and May P. (1990). Oncogene, 5, 945 952. Soussi T, Caron DF, Mechali M, May P and Kress M. (1987). Oncogene, 1, 71 78. Soussi T and May P. (1996). J. Mol. Biol., 260, 623 637. Speir E, Modali R, Huang ES, Leon MB, Shawl F, Finkel T and Epstein SE. (1994). Science, 265, 391 394. Steegenga WT, van der Eb AJ and Jochemsen AG. (1996). J. Mol. Biol., 263, 103 113. Stephen CW and Lane DP. (1992). J. Mol. Biol., 225, 577 583. Stewart N, Hicks GG, Paraskevas F and Mowat M. (1995). Oncogene, 10, 109 115. Stommel JM, Marchenko ND, Jimenez GS, Moll U, Hope TJ and Wahl GM. (1999). EMBO J., 18, 1660 1672. Storey A, Thomas M, Kalita A, Harwood C, Gardiol D, Mantovani F, Breuer J, Leigh IM, Matlashewski G and Banks L. (1998). Nature, 393, 229 234. Stott FJ, Bates S, James MC, McConnell BB, Starborg M, Brookes S, Palmero I, Ryan K, Vousden KH and Peters G. (1998). EMBO J., 17, 5001 5014. Strasser A, Harris AW, Corcoran LM and Cory S. (1994). Nature, 368, 457 460. Sturzbecher HW, Chumakov P, Welch WJ and Jenkins JR. (1987). Oncogene, 1, 201 211. Subler MA, Martin DW and Deb S. (1992). J. Virol., 66, 4757 4762. Szekely L, Selivanova G, Magnusson KP, Klein G and Wiman KG. (1993). Proc. Natl. Acad. Sci. USA, 90, 5455 5459. Takenaka I, Morin F, Seizinger BR and Kley N. (1995). J. Biochem., 270, 5405 5411. Tao WK and Levine AJ. (1999). Proc. Natl. Acad. Sci. USA, 96, 6937 6941. Thut CJ, Chen JL, Klemm R and Tjian R. (1995). Science, 267, 100 104. Truant R, Xiao H, Ingles CJ and Greenblatt J. (1993). J. Biol. Chem., 268, 2284 2287. Unger T, Mietz JA, Schener M, Yee CL and Howley PM. (1993). Mol. Cell. Biol., 13, 5186 5194. Unger T, Nau MM, Segal S and Minna JD. (1992). EMBO J., 11, 1383 1390. Venot C, Maratrat M, Dureuil C, Conseiller E, Bracco L and Debussche L. (1998). EMBO J., 17, 4668 4679.

7635

Structure and function of p53 protein P May and E May

7636

Votja PJ and Barrett JC. (1995). Bba-Rev. Cancer, 1242, 29 41. Wadgaonkar R and Collins T. (1999). J. Biol. Chem., 274, 13760 13767. Waga S, Hannon GJ, Beach D and Stillman B. (1994). Nature, 369, 574 578. Waldman T, Kinzler KW and Vogelstein B. (1995). Cancer Res., 55, 5187 5190. Walker KK and Levine AJ. (1996). Proc. Natl. Acad. Sci. USA, 93, 15335 15340. Wang XW, Forrester K, Yeh H, Feitelson MA, Gu JR and Harris CC. (1994). Proc. Natl. Acad. Sci. USA, 91, 2230 2234. Wang XW, Vermeulen W, Coursen JD, Gibson M, Lupold SE, Forrester K, Xu G, Elmore L, Yeh H, Hoeijmakers JH and Harris CC. (1996). Genes Dev., 10, 1219 1232. Wang XW, Yeh H, Schaeer L, Roy R, Moncollin V, Egly JM, Wang Z, Freidberg EC, Evans MK and Tae BG. (1995a). Nat. Genet., 10, 188 195. Wang XW, Yeh H, Schaeer L, Roy R, Moncollin V, Egly JM, Wang Z, Friedberg EC, Evans MK, Tae BG, Bohr VA, Weeda G, Hoeijmakers JHJ, Forrester K and Harris C. (1995b). Nat. Genet., 10, 188 195. Wang Y and Prives C. (1995). Nature, 376, 88 91. Wang Y, Reed M, Wang P, Stenger JE, Mayr G, Anderson ME, Schwedes JF and Tegtmeyer P. (1993). Genes Dev., 7, 2575 2586. White E. (1996). Genes Dev., 10, 1 15. Wu GS, Saftig P, Peters C and El-Deiry WS. (1998). Oncogene, 16, 2177 2183.

Wu X, Bayle JH, Olson D and Levine AJ. (1993). Genes Dev., 7, 1126 1132. Wynford-Thomas D. (1999). J. Pathol., 187, 100 111. Yang XJ, Ogryzko VV, Nishikawa J, Howard BH and Nakatini Y. (1996). Nature, 382, 319 324. Yew PR and Berk AJ. (1992). Nature, 357, 82 85. Yewdell JW, Gannon JV and Lane DP. (1986). J. Virol., 59, 444 452. J, Gotteland M, Mishal Z, Viron A Yonish-Rouach E, Borde and May E. (1994). Cell Death Dier., 1, 39 47. Yonish-Rouach E, Deguin V, Zaitchouk T, Breugnot C, Mishal Z, Jenkins JR and May E. (1995). Oncogene, 11, 2197 2205. Yonish-Rouach E, Grunwald D, Wilder S, Kimchi A, May E, Lawrence JJ, May P and Oren M. (1993). Mol. Cell. Biol., 13, 1415 1423. Yonish-Rouach E, Resnitzky D, Lotem J, Sachs L, Kimchi A and Oren M. (1991). Nature, 352, 345 347. Zakut-Houri R, Bienz-Tadmor B, Givol D and Oren M. (1985). EMBO J., 4, 1251 1255. Zauberman A, Lupo A and Oren M. (1995). Oncogene, 10, 2361 2366. Zhan QM, Bae I, Kastan MB and Fornace AJ. (1994). Cancer Res., 54, 2755 2760. Ziegler A, Jonason AS, Leell DJ, Simon JA, Sharma HW, Kimmelman J, Remington L, Jacks T and Brash DE. (1994). Nature, 372, 773 776.