Catálogo de publicaciones - libros

Soon after the discovery of the p53 protein in 1979, the p53 gene was found to be mutated in about half of all human cancers. However, mainly due to the fact that the normally low levels of the p53 protein can be elevated in many cancers, it took some time to realise that p53 was in fact a tumour suppressor gene (Finlay et al., 1989; Levine et al., 1991). This concept was further consolidated by the discovery that the familial cancer predisposition known as Li-Fraumeni syndrome1 is linked to germ-line mutation of the p53 gene (for a review see Varley et al., 1997), and by the clear tumour propensity shown by p53 knock-out mice (for a review see Venkatachalam and Donehower, 1998).

“Guardian of the genome” (Lane, 1992), “Death star” (Vousden, 2000), “Good and bad cop” (Sharpless and DePinho, 2002), “An acrobat in tumorigenesis” (Moll and Schramm, 1998), are just a few of the names that have been attributed to the p53 gene over recent years. However, the cameras (and funding) were certainly not present at the time of the discovery of p53 in 1979 (Crawford, 1983). It was only when the first alterations of the p53 gene in human cancers were discovered 10 years later, in 1989, that p53 started to become really popular, with the title of “molecule of the year” attributed by Science , in 1993 (Harris, 1993). This title was certainly justified, as the observation that more than one half of human cancers expressed a mutant p53 raised extensive clinical possibilities both for diagnosis and treatment.

Somatic mutations are the cornerstone of cancer (Hanahan et al. 2000). The development of cancer involves the contributions of many heritable genetic events as well as of a large number of epigenetic changes, but what makes the turning point between untransformed and transformed cell irreversible is the acquisition of targeted, somatic mutations, conferring to cells a selective advantage for clonal proliferation. These mutations can occur in many different genes, but only a handful of them are frequently mutated in a wide variety of human cancers. They include genes of the RAS family (mainly KRAS ), BRAF1, APC,α–Catenin, p16/INK4a, PTEN and TP53 . After over 20 years of research on mutation detection in cancers, TP53 remains the world champion of somatic mutations, with over 70 % of all the mutations described so far in human cancers (Hainaut et al. 2000).

Finding reliable molecular markers for early diagnosis, prognosis and prediction of response to treatment is a major challenge for cancer management. A marker of prognosis provides information on the risk of relapse and death independently of treatment, whereas a predicitve marker provides information on the potential benefit of a specific treatment (Lonning, 2003). An early diagnostic marker helps to identify lesions at high risk of malignant transformation. Clinical stage, tumor size and morphological grade are the most reliable factors of prognosis. Among numerous molecular markers that have been tested most recently, only a few are used in clinical practice. In breast cancer for example, estrogen and progesterone receptors are used routinely as predictive markers for tumor response to anti-hormone therapy. However, about 30 % of patients with positive receptor status (expected to benefit from anti-hormone treatment) will face a therapeutic failure, showing the limitations of these markers.

Anti-cancer therapy operates on the assumption that the genetic pathways disrupted during tumorigenesis are distinct from those that mediate drug sensitivity. The main objective of this therapy is to present tumor cells with obstacles unrelated to the process of cellular transformation or to exploit vulnerabilities created by tumor development, such as uncontrolled DNA synthesis, checkpoint abnormalities, or an addiction to an oncogenic signal. Cytotoxic therapies, for example, rely on the introduction of DNA damage or the inhibition of chromosome segregation. These lesions, when introduced at high levels, elicit a DNA damage response presumably distinct from any encountered during the early stages of tumor development.

Introducing functional p53 into tumours using adenoviral vectors is leading to success in clinical trials (Edelman et al., 2003; http;//www.introgen.com/infotp.html; http;//www.sibiono.com) and p53 gene therapy gained regulatory approval in China in 2003 where it has been on sale since January 2004. Renewed optimism around the use of p53 gene therapy and increased understanding of p53 function suggest that many more potent and selective variants of p53 may be developed. Reactivating mutant p53 or exploiting specific properties of tumour cells carrying mutations in p53 is a greater challenge. As described elsewhere in this book, tackling this problem has led to a variety of very exciting discoveries. Here we have focused on the current approaches to activate p53 in those cancers that retain wild type p53.

The tumor suppressor p53 is a transcription factor that is at the center of a network of interactions that affect the cell cycle and apoptosis (Vogelstein et al. 2000; Ryan et al. 2001). The protein is induced by a variety of stresses that include oncogene activation and DNA damage caused by chemotherapy and radiotherapy. On induction, it activates a variety of genes whose products lead to G1 and G2 cell cycle arrest and apoptosis (Vogelstein et al. 2000; Ryan et al. 2001). It is such an effective tumor suppressor that it is inactivated in virtually all cancers; in about 50 % of cancers p53 is directly inactivated by mutation and in the remainder its activity is lost by perturbations of its associated pathways and interactions (Hainaut and Hollstein 2000). Reactivating mutant p53 is an important target in the development of novel therapies for cancer (Lane and Lain 2002; Lane and Hupp 2003). To understand how p53 is inactivated, it is necessary to understand its structure and how it responds to mutation. Such knowledge will provide a basis for the rational design of novel therapeutics that may reverse the effects of mutation. In this chapter, we survey the structure of the protein, the effects of mutation and how they may be reversed.

Inactivation of the p53 tumor suppressor by point mutation occurs in a large fraction of human tumors, including almost all tumor types (see p53 Mutation database at http://www.iarc.fr/p53). A majority of p53 mutations are missense mutations that give rise to the expression of mutant p53 proteins with one amino acid substitution. This pattern of mutation stands in sharp contrast to those of most other tumor suppressor genes, e.g. the Rb and p16 genes, which are frequently inactivated by homozygous deletion, smaller deletions or promoter methylation that either results in complete lack of expression of the protein, or expression of a truncated unstable protein. This suggests that p53 mutation not only serves to inactivate p53 but that expression of mutant p53 itself may provide a selective advantage to tumor cells and promote tumor growth. First, point mutant p53 proteins may act in a dominant negative manner, i.e. inhibit the activity of a wild type allele present in the same cell through hetero-oligomerization that forces wild type p53 to adopt a mutant conformation (Milner and Medcalf 1991). In addition, mutant p53 proteins may have acquired novel activites that could support the growth of tumors. These so called gain-of-function (GOF) activities of mutant p53 could involve promiscuous DNA binding and illegitimate activation of target genes, such as the c-Myc oncogene, the multidrug resistance gene (MDR1), VEGF, and the dUTPase gene ((Frazier et al. 1998; Pugacheva et al. 2002; Tsang et al. 2003); www.iarc.fr/p53), whose activation could contribute to tumor development. Moreover, mutant p53 could enhance cell cycle progression and/or cell survival through novel interactions with cellular protein partners, as examplified by the binding of mutant p53 to p73 and other p53 family members (Di Como et al. 1999; Marin and Kaelin 2000; Strano et al. 2002; Monti et al. 2003; Strano and Blandino 2003).

Cancer is caused by gain of function of proteins involved in proliferation and survival, and loss of function of proteins that regulate these processes (Hanahan and Weinberg, 2000). Strategies for treating cancer generally involve development of small molecules that block hyperactive enzymes, or take advantage of abnormal expression of protein targets on the surface of cancer cells. Developing therapies based on loss of function of tumor suppressors presents novel challenges. Loss of the protein phosphates PTEN and loss of the G1/S checkpoint protein pRB occurs frequently in cancer, and offers a number of potential drug targets. Loss of PTEN leads to hyperactivation of downstream enzymes such as AKT and mTOR (McCormick, 2004) whereas loss of pRB leads to hyper-activation of the transcription factor E2F, and increased expression of numerous potential targets, some of which have already been exploited for cancer therapy (dihydrofolate reductase and thymidylate synthase, for example, are the targets of methotrexate and 5-fluorouracil, respectively). Loss of p53, on the other hand, does not appear to offer any direct targets for intervention: in contrast to PTEN and pRB, p53 is a positive regulatory protein, whose targets are obviously lost rather than hyper-activated in cancer cells.

Has the p53-field made a major impact on cancer research? Indeed it has. From its earliest beginnings, cancer research has been looking for “the” fundamental change in cancer cells, the ultimate common denominator. The idea that such a change must exist, was, if not abandoned, substantially mollified by the increasing realization, from the late 1950s, that cancer development is a multistep process, based on the individual reassortment of several unit characteristics (Foulds, 1958) or, as we now say, phenotypic traits.