The p53 and Mdm2 families in cancer

Abstract

Cells within an organism are occasionally exposed to either intracellular or environmental stress. Such stress often has genotoxic potential that enhances the probability of cancer. Two gene families, the p53 family (p53, p63 and p73) and the Mdm2 family (Mdm2 and MdmX), serve as major integrators of the signals generated by genotoxic and oncogenic stress. Their co-ordinated modulation ensures an optimal response to stress and decreases the likelihood of cancer. Work over the past year has provided better understanding of the p53–Mdm2 module that lies in the heart of this regulatory network, and of the intricate interplay between the various members of the network.

Introduction

The p53 tumor suppressor gene encodes a sequence-specific transcription factor whose activity is either disabled or attenuated in the vast majority of human cancers 1., 2., 3., 4., 5.. p53 inactivation is often achieved through mutations affecting the p53 locus directly. In other cases, p53 functionality is compromised through excessive activity of its major negative regulator, Mdm2, as well as of the Mdm2-related protein MdmX.

Several years ago, two p53-related genes were identified, and were eventually dubbed p63 and p73, respectively. Although both proteins bear significant similarity to p53, the physiological functions of each member of the p53 family appear to be rather distinct [6]. Thus, p53-null mice exhibit only relatively limited developmental defects, particularly affecting neural tube closure and spermatogenesis. In contrast, mice lacking p63 display severe developmental defects in epithelial cell differentiation and die shortly after birth, whereas p73-deficient mice suffer from a variety of other developmental aberrations, affecting primarily the nervous system and the respiratory tract [6]. The differences among the individual p53 family members also extend to the effect of their ablation on cancer development. Mice null for p53 are highly susceptible to cancer with a relatively early onset, and heterozygotes retaining only a single wild-type (wt) p53 allele are also cancer prone, albeit with a later onset and usually contingent upon somatic loss of the remaining p53 allele. In contrast, neither loss of p73 nor viable loss of a p63 allele seem to predispose mice to cancer. Thus, at least in these mouse models, neither p63 nor p73 behaves as a tumor suppressor. Nevertheless, numerous studies utilizing human tumor material as well as cultured cells have indicated non-random relationships between altered expression of p63 and p73 and cancer. Attempts to establish unequivocally the relevance of p63 and p73 to cancer are complicated by the fact that, unlike p53, these proteins exist in multiple forms as a result of alternative splicing and usage of multiple transcription start sites. In fact, it is now becoming clear that these different variants often exert opposing effects on cell fate and cell behavior. Thus, the transactivation competent (TA) variants of p63 and p73 can activate to varying degrees many canonical p53 target genes, and may in some circumstances double for p53 in its anti-proliferative action. In contrast, variants lacking the transactivation domain (ΔN forms) actually antagonize TA family members, including p53 itself. Consequently, their hyperactivation may in fact exert oncogenic rather than tumor-suppressor effects.

The Mdm2 and MdmX proteins 7., 8., 9. maintain yet another level of regulation on the p53 family. Mdm2 possesses E3 ubiquitin ligase activity towards p53. Through its ability to ubiquitinate p53 and target if for proteasomal degradation, Mdm2 plays a key role in retaining p53 at very low concentrations under non-stressed conditions. At the same time, the Mdm2 gene is a positive transcriptional target of p53, whose expression is often elevated subsequent to induction of p53 activity. This defines a negative feedback loop wherein p53 upregulates Mdm2, whereas Mdm2 downregulates p53. This loop can be viewed as a regulatory module, into which a plethora of incoming signals feed and thereby modulate p53 levels and activity in accordance with intracellular and extracellular cues. Unlike Mdm2, its ‘cousin’ MdmX does not drive the destruction of p53. Hence, the relevance of MdmX as a physiological regulator of p53 function has been questioned in the past. However, MdmX is also emerging now as part of the intricate network that controls the cellular activity of p53.

In this review, we focus on some of the advances made during the past year towards elucidating the complex interplay between Mdm2, MdmX and the p53 family members, with particular emphasis on cancer-related issues. New papers addressing these proteins keep appearing at an enormous rate, with >4700 publications on p53 published in 2000 alone. Hence, it is impossible to cover comprehensively all the recent literature in such a short review. In many instances review articles, rather than primary reports, are cited. We apologize to all authors whose important work has not been cited.p53: optimizer of cell fate?

Recent advances in the cloning of p53 orthologs from Drosophila and Caenorhabditis elegans have highlighted the evolutionary conservation of p53 functions. As in higher organisms, these orthologs are required for apoptosis following ionizing irradiation 10., 11., 12••.. Likewise, the requirement for p53 in assuring proper meiotic chromosome segregation in C. elegans [12••] is reminiscent of the role of p53 in mouse spermatogenesis [13].

The p53-dependent apoptotic response is a well-documented anti-cancer mechanism 3., 14., as are p53-dependent growth arrest and DNA repair. The antiproliferative effects of p53 are called into action in response to a variety of stress signals, many of which are implicated in neoplastic processes [1]. A recently reported novel inducer of the p53 response is massive genomic DNA demethylation, as is caused by inactivation of the DNA-methyltransferase gene Dnmt1 [15]; this response culminates in apoptosis, eliminating cells whose gene expression patterns have been distorted via aberrant demethylation. Similarly, p53-dependent replicative senescence also serves to curtail the proliferation of potentially cancer-prone cells. p53 can also address oncogenic stress more directly, by counteracting the oncogenic forces that drive neoplastic cell behavior. For instance, p53 activation accelerates degradation of β-catenin, whose oncogenic action contributes to a variety of epithelial tumors 16., 17., 18..

The molecular mechanisms through which p53 elicits apoptosis have been explored extensively [3]. The past year has yielded an impressive array of new pro-apoptotic transcriptional targets of p53. These include the gene encoding apoptotic protease-activating factor-1 (Apaf-1) 19•., 20•., 21•., the Bcl-2-binding protein PUMA 22•., 23•., the death-domain protein Pidd [24], the tumor-suppressor phosphatase PTEN [25•] and p53DINP1 [26]. Other genes that can modulate cell fate, shown to be targets of p53, include HB-EGF, which activates the MAPK pathway [27] and c-Ha-Ras [28]. Interestingly, the apoptotic activity of p53 is modulated in a selective manner by LKB1, a serine/threonine kinase the malfunction of which underlies the Peutz–Jegher syndrome (PJS); this disease involves the formation of multiple intestinal polyps, some eventually progressing to malignancy [29•]. Defective p53-dependent apoptosis may underlie the high tumor susceptibility of PJS patients.

Whereas pivotal pro-apoptotic functions of p53 are being unraveled, it is noteworthy that p53 may also exert anti-apoptotic effects under certain circumstances. In some cases, functional p53 may even interfere with cell killing by chemotherapy [30]. Interestingly, p53 dictates the resistance of non-transformed cells to parvoviruses, which kill efficiently tumor cells lacking functional p53 31., 32.. p53 may, thus, adjust the cell-fate decisions to the particular signals that drive the p53 response. A remarkable case in point is the contribution of p53 to the extended longevity of C. elegans under stressful conditions [12••]. Hence, the role of p53 as ‘optimizer of cell fate’ most probably extends well beyond the mere prevention of cancer.