p53

Introduction

When normal mammalian cells are subjected to stress signals (e.g. hypoxia, radiation, DNA damage or chemotherapeutic drugs ...) p53 is activated; additionally to its activation, ubiquitin-dependent degradation of the p53 protein is blocked. The resulting increase in p53-dependent gene transcription leads to the p53-mediated induction of programmed cell death and/or cell cycle arrest. Functional p53 is thought to provide a protective effect against tumorigenesis, and indeed, mutations of p53 have been found in nearly all tumor types and are estimated to contribute to around 50% of all cancers.

Structural and functional aspects of p53

There are four conserved domains in p53:
1. The N-terminal domain is required for transcriptional transactivation
2. A sequence-specific DNA binding domain
3. A tetramerization domain near the C-terminal end
4. The C-terminal domain interacts directly with single stranded DNA.

Activation of p53 can result in cell cycle arrest, presumably to allow DNA repair to occur before replication or mitosis. In some cell types, however, p53 activation results in apoptosis as means of eliminating irreparably damaged cells. The final outcome of p53 activation depends on many factors, and is mediated largely through the action of downstream effector genes transactivated by p53.

By the way: p53 was reported to inhibit RelA-dependent transactivation by competing with RelA for a binding region on the co-activator p300/CBP which is required for coactivation of RelA (Ravi et al., 1998, Cancer Research, 58: 4531-36).

p53 Target Genes

Wild-type p53 binds to specific genomic sites with a consensus binding site 5'-PuPuPuC(A/T)(T/A)GPyPyPy-3'. p53 binds as a tetramer and stimulates expression of downstream genes that negatively control growth and/or invasion or are mediators of apoptosis. It was predicted that the expression of about 200-300 genes might be controlled by p53 transactivation.
  1. p53 target genes involved in p53 control:
    mdm2
  2. p53 target genes involved in cell cycle control:
    p21 WAF1/CIP1, GADD45, WIP1, mdm-2, EGFR, PCNA, CyclinD1, CyclinG, TGFalpha and 14-3-3sigma
  3. p53 target genes involved in DNA repair:
    GADD45, PCNA, and p21 WAF1/CIP1
  4. p53 target genes involved in apoptosis:
    BAX, Bcl-L, FAS1, FASL, IGF-BP3, PAG608 and DR5/KILLER, GML, P2XM
  5. p53 target genes involved in angiogenesis:
    TSP-1, BAI1
  6. p53 target genes involved in cellular stress response:
    TP53TG1, CSR, PIG3
Analysis of p53 binding sites throughout the human genome suggests the existence of many more p53 regulated genes (Review by Tokino and Nakamura, 2000, Crit Rev Onc. Hem., 33:1-6).

Activation of p53

p53 is activated, among others, in response to DNA damage, and many factors interact to signal and modulate this response. There is still controversy over the pathways that lead to the activation of p53. Several mechanisms have been suggested:
One idea is that stress-activated protein kinases phosphorylate p53, protecting it from degradation and activating its function as a transcription factor. Indeed, many phosphorylated forms of p53 are found in cells, and by phosphorylation p53 can be released from a latent state, in which it cannot bind DNA. One attractive candidate for p53 activation by phosphorylation is the DNA-dependent protein kinase (DNA-PK). DNA-PK is activated by DNA damage, and one of its substrates is p53. DNA-PK phosphorylates Ser15 within the critical N-terminal region of p53, which controls the interaction of p53 with the transcriptional apparatus and with the MDM-2 protein. Indeed, recently it was demonstrated, that DNA-PK is required for the p53 response to occur (Woo et al., 1998, Nature, 394: 700-703). Also the ATM kinase, the product of the ATM gene (which is defective in patients with Ataxia Telangiectasia), phosphorylates Ser15 in vivo (Banin et al. and Canman et al., 1998, Science, 281: 1674-1679).

Instead of its phosphorylation, the dephosphorylation of p53 at serine 376 by the ATM-dependent activation of a specific phosphatase might enable DNA binding of p53 and its transcriptional activation. In this process, the so called 14-3-3 proteins bind to the C-terminus of the dephosphorylated p53, and by this possibly activate it.

Another pathway towards activation of p53 involves the mdm-2 gene product. MDM-2 can target p53 for nuclear export and degradation; nonfunctional MDM-2 results in accumulation of p53 and activation of p53-dependent transcription. The mdm-2 gene itself is activated for transcription by p53, so this model implies that p53 is constitutively active, driving transcription of the protein (MDM-2) that targets its own degradation. Blocking the p53 degradation pathway would result in the activation of the p53 response. Indeed, it was shown that the ARF tumor suppressor (also called p14ARF) binds to the complex of p53 and MDM-2, by this stabilizing p53, possibly by inducing degradation of MDM-2 (Zhang et al., 1998, Cell, 92: 725-734). ARF expression itself is regulated by the E2F-1 transcription factor! This connects the Rb pathway to p53: oncogenes like E1A or SV40 T block Rb function, thus activating E2F-1. E2F-1 transcriptional activity leads to the expression of a number of genes required for passage into and through S phase but also to the expression of ARF which stabilizes p53. This would result in either p53 dependent apoptosis or cell cycle arrest unless p53 itself is inhibited, e.g. by the oncogenes E1B and SV 40 T. The (so far unanswered) question arises what happens during normal passage of cells through the cell cycle, when E2F-1 becomes active during transit to S phase but - of course - does not induce cell cycle arrest or apoptosis (review: Prives, C., 1998, Cell, 95: 5-8).

Reviews

Lane, D., 1998, Nature, 394: 616; (Review p53 activation)
Prives, C., 1998, Cell, 95: 5-8; (Review p53 activation)
Amundson, S.A., 1998, Oncogene, 17(25): 3287-99 (Review p53)
Tokino and Nakamura, 2000, Crit Rev Onc. Hem., 33:1-6 (Review p53 targets).