p53, a tumor suppressor gene, is the most frequently mutated gene in human cancers known to date, carrying mutations in more than half of human tumors (1). Even if the tumors are not mutated in p53, a component in the p53 pathway is found to be disregulated in most tumors. p53 forms a tetramer and functions as a transcription factor. Upon its activation by signals such as DNA damage, aberrant growth signals, UV light, or other cellular stress, it induces expression of genes which results in cell cycle arrest (G1/S and G2/M), apoptosis, maintenance of genomic stability (such as DNA repair), and inhibition of angiogenesis (2). Due to its multiple functions and its nature of integrating divergent pathways, it is referred to as a critical "node" of the cellular signalling network by Vogelstein et al., and that the destruction of the "highly connected node" is a critical step for a cell to enter malignancy (2). Not only is p53 activation controlled at a transcriptional level, but it also undergoes multiple post-translational modifications, such as phosphorylation, acetylation, glycosylation, ribosylation, ubiquitination, or sumoylation (3). The intracellular level of p53 protein is mainly controlled by degradation, mainly by ubiquitination by Mdm2, which targets it for degradation by the proteasome (2).
The majority of the p53 mutations found in tumors are missense mutations, and other mutations such as large deletions are rare (4). Of those missense mutations, approximately 90% are found in the DNA binding domain. The DNA binding domain contain six mutation hotspots, R175, G245, R248, R249, R273, and R282, which account for more than 40% of the missense mutations (1). Germline p53 mutations are seen in 70-80% of patients with Li-Fraumeni syndrome (LFS), a familial cancer predisposition syndrome (5, 6). LFS patients develop multiple tumors, with an approximate 50% incidence by the age 30. The spectra of tumors, unlike other familial cancer syndromes, is broad, including sarcomas, breast cancers, leukemias, brain tumors, and adrenocortical carcinomas (5).
Mice nullizygous for p53 have been generated in an attempt to clarify its role in tumorigenesis in vivo (7, 8). p53-deficient (-/-) mice develop normally, but are prone to a high incidence of spontaneous tumors and have a shorter life span, with ~90% dying by the age of six months. The most common malignancy observed in these mice were lymphomas, a majority of them seen in the thymus (7, 8). p53 heterozygous (+/-) mice are also cancer-prone, although to a lower extent than the -/- mice (8). These mice develop tumors at a later age and have a longer life span compared to the -/- mice, as expected from their requirement for a "second hit". The tumor spectrum differs as well, as the main tumor types seen in these mice are sarcomas (fibrosarcoma, osteosarcoma, etc.). Loss of heterozygosity (LOH) was observed in 9/12 tumors examined. Although the phenotype (including the tumor spectrum) seen in the p53+/- mice was similar to LFS patients, they were not identical, as carcinomas were rarely seen in +/- mice as well as the -/- mice (4, 7, 8).
II. Evidence for gain-of-function p53 mutants *
Although p53 is known to function as a tumor suppressor (1, 2, 9), several lines of evidence suggest that certain mutations of p53 can act in a gain-of-function or a dominant negative fashion (6). Evidence for a dominant negative mutation comes from studies at a cellular level, where the p53 mutant can heterooligomerize with wild-type p53, converts it into a mutant conformation, and inhibit its wild-type function (10, 11). Mutant p53 expression can transform p53-null cell lines (12), indicating that several mutations can act independently of the wild-type, suggesting a gain-of-function. The gain-of-function mutants can elicit altered functions, such as regulation of a distinct set of genes (e.g. MDR1, EGFR, c-myc, BAG-1, fos), increased tumorigenecity, growth, and invasiveness, increased resistance to irradiation, and interaction with different proteins from those of wild-type (e.g. MBP-1) (6; see Figure 3 and Table 1). In vivo evidence also exist, as transgenic mice expressing a missense mutant of p53 shows enhanced tumorigenesis and different tumor spectra on a p53+/+ and +/- background (13). However, the significance of these studies using missense p53 mutants were in question, as the assays or expression levels did not represent true physiological conditions. Clearly, the creation of mice with a "knock-in" of a missense p53 mutant into the endogenous allele were awaited for to find a definitive answer.
III. In vivo evidence of a gain-of-function p53 mutant; study by Olive et al. *
In order to investigate the role of two missense mutations in vivo, Olive and colleagues created mice carrying a R172H or R270H missense mutation on its endogenous allele (4). Both mutations exist in the DNA binding domain and are commonly observed in LFS patients. R172H corresponds to codon 175 in humans and represents a structural mutation, which disrupts the structure of the b-sheet and the loop-sheet helix motif of the DNA binding domain (1). This mutant form of p53 is known to specifically bind the monoclonal antibody (mAb) PAb240. R270H corresponds to codon 273 in humans and represents a contact mutation; the mutation disrupts the contact of p53 to DNA and disables it to act as a transcription factor (1). The mutant alleles were created by introducing a Lox-STOP-Lox (LSL) cassette (14) into intron 1 of the p53 gene, using site-directed mutagenesis to introduce the mutation, and targeting the vector into J1 embryonic stem cells by homologous recombination (Figure 1A and 1B). Heterozygote mice carrying the LSL cassette were crossed to proteamine-Cre transgenic mice, and mutant/+ Cre/+ males were crossed to wild-type 129S4/SvJae females, to generate 172H/+, 270H/+ mice (later referred to as m/+ mice) (Figure 1C and 1D). The mRNA expression level of these mutants were equivalent to that of wild-type, but accumulation of mutant protein was observed (Figure 1E and 1F).
The survival time and tumor spectra of m/+ were compared to +/- mice to assess the role of the mutant alleles. Although the life span of the m/+ mice were similar to +/- mice (Figure 2A), the tumor spectra differed significantly (Figure 2B and Supplementary Figure 1). A significant increase in the incidence of carcinomas were observed (histopathology of tumors shown in Supplementary Figure 2A-F), especially in 270H/+ mice. 7 cases of lung adenocarcinomas were seen in 270H/+ mice, with several of them showing malignant features such as nuclear atypia, desmoplasia, and metastasis, which are commonly seen in human patients. B cell lymphomas were also increased in 270H/+ mice. In the 172H/+ mice, high incidence of osteosarcomas were observed. Many of the tumors seen in these mice were metastatic, with evidence of distal metastasis to sites such as the lung and liver (Supplementary Figure 2E and 2F). Only 4 low-grade carcinomas were reported in the +/- mice. The carcinomas were positive for keratin staining by immunohistochemistry, proving their epithelial origin (Supplementary Figure 3A-C and Supplementary Table 1). Interestingly, loss of heterozygosity was observed in 10/19 tumors analyzed (Figure 2C). These results unquestionably indicate that the 172H/+ and 270H/+ p53 mutants are not loss-of-function alleles, but their tumorigenisity is elicited through a gain-of-function or by a dominant negative function. Furthermore, these mice have strikingly similar phenotypes to LFS patients, validating them as a model for LFS (4).
Additional features of these mice mimic those of LFS patients. Nuclear accumulation of mutant p53 protein is often seen in tumor cells from LFS patients (4), and the same phenomenon was seen in the m/+ mice when immunohistochemistry for p53 was performed on tumors derived from these mice (Supplementary Figure 6A and 6C). Accumulation of the mutant protein was only seen in malignant cells and not in normal or low-grade lung adenocarcinomas (Supplementary Figure 6C). Nuclear accumulation of p53 did not correlate with the genotype of the cell, as accumulation was independent of loss of heterozygosity (Supplementary Table 2), and normal cells of m/- mice did now show accumulation (Supplementary Figure 6B), even after g-irradiation (Supplementary Figure 6D-F).
The dominant negative/gain-of-function of the mutant alleles was further investigated by performing cellular assays on embryonic fibroblasts (MEFs) derived from m/+ mice. MEFs from m/+ mice showed enhanced proliferation by cell cycle analysis (Figure 3A and Supplementary Figure 4A) and by 3T3 passaging assay (Supplementary Figure 4B). Furthermore, the m/+ thymocytes show resistance to g-irradiation-induced apoptosis (Figure 3B. This indicates that the mutant p53 protein can either inhibit the wild-type protein in a dominant negative manner or elicit gain-of-function independent of the wild-type protein. Interestingly, the m/+ MEFs were capable of inducing DNA damage (doxorubicin)-induced G1 arrest (Figure 3A and Supplementary Figure 4A), indicating that the mutant p53 does not interfere with all wild-type functions. The transactivation of a p53 target, p21, in response to doxorubicin treatment was not impaired (Figure 4A), further indicating that the mutant protein is not disfunctional.
Direct evidence for a gain-of-function of the p53 mutants was assesed by comparing mice expressing only the mutant allele (m/- vs. -/-). The m/- and -/- mice exhibit similar life spans(Figure 5A), but their tumor spectra were strikingly different (Figure 5B). The m/- mice had a significantly higher incidence of carcinomas, with most of them showing malignant features such as desmoplasia and stromal invasion (Figure 6A-E). The epithelial origin of these carcinomas were proven by immunohistochemistry (Supplementary Figure 3E and 3F, and Supplementary Table 1). The function of the mutant p53 protein was assessed by performing cellular assays on MEFs derived from these mice. Loss of wild-type function was observed in m/- primary cells, as they exhibit enhanced proliferation capacities (Supplementary Figure 5A and 5B), defective DNA damage-induced G1 arrest (Supplementary Figure 5A), and defective transactivation of p21 (Figure 4B).
To further confirm the gain-of-function of the p53 mutants, the authors performed a knockdown of p53 by lentiviral transduction of p53 shRNA on Os1, a cell line derived from a osteosarcoma metastasis in a 172H/+ mouse which showed loss of heterozygosity. Os1 cells transduced with the shRNA construct (containing a puromycin resistance cassette) were selected by puromycin, and the shRNA was conditionally expressed by infecting these cells with an adenovirus expressing Cre (Figure 7A and 7B). The knockdown of the mutant p53 protein significantly decreased the proliferation rate of Os1 cells (Figure 7C-E). Mutant p53 coimmunoprecipitated with p63 and p73, which are members of the p53 family (Figure 7F), and knockdown of mutant p53 resulted in an increase in p63 and p73 target gene expression (data not shown). These data clearly indicate a gain-of-function by the p53 mutants, and that inhibition of p63 and p73 may be a possible mechanism for the gain-of-function (4).
(Legends for supplementary data could be found here)
IV. Further evidence by Lang et al.; p53 gain-of-function by inactivation of p63 and p73 *
Mice carrying an endogenous mutation at codon 172 (R172H; codon 175 in humans) were also created by Lang and colleagues (15). Similar to results from Olive et al. (4), the life span of -/-, m/-, and m/m mice as well as +/- and m/+ mice were similar (Figure 2A and 2B). The tumor spectra of -/- and m/m mice, as well as +/- and m/+ mice were similar in this study (Table 1); however, mice carrying the mutant allele developed tumors that were highly metastatic (Table 1 and Figure 2C). The differences seen by Olive et al. (4) and this study (15) in tumor spectra may be due to differences in the genetic background of the mutant mice; 129S4/SvJae was used by Olive et al., and C57BL/6 by Lang et al. (4, 15). Accumulation of p53 was seen in tumors but not in normal cells of mice carrying the mutant, and this was not due to loss of heterozygosity (Figure 2C and 2D). Mice nullizygous for mdm2 die during embryogenesis, but can be rescued by deletion of p53 (16, 17). The mdm-/- p53m/+ genotype was embryonic lethal, indicating that the mutant was not acting as a dominant negative over the wild-type p53 (15).
Lang et al. went on to characterize the functions of the mutant p53 protein utilizing cellular assays on MEFs. m/m and m/+ showed enhanced proliferation, even compared to -/- MEFs (Figure 3A and 3B), and increased transformation potential (Figure 3C and 3D). As seen in Olive et al. (4), the mutant p53 interacted with p63 and p73 (Figure 4A). Knockdown of mutant p53 by siRNA increased the ability of p63 and p73 to activate the p21 promoter (Figure 4C and 4D). siRNA knockdown of p63 and p73 enhanced the transformation potential and proliferation capacity of p53-/- cells, but not that of m/m cells (Figure 5E and 5F, and Figure 6). These results demonstrate that the mutant p53 has a gain-of-function phenotype in which one mechanism of action is through the inhibition of p63 and p73 by direct interactions (15).
Comparison and summary of results from the two studies are shown in Table 1 below.
V. Gain-of-function mutation in p53; How does it work?
These two elegant studies clearly demonstrate a gain-of-function in the two p53 missense mutants. However, the molecular basis of how these mutants exhibit their tumorigenic phenotype is still unanswered. One possible mechanism is the inhibition of p53 family members, p63 and p73, by direct contact (4, 15). Furthermore, the inhibition of wild-type p53 function in a dominant negative manner is likely to release cells from tumor suppression (4, 15). Other mechanisms may exist, for example, by regulation of other targets or by binding to other proteins (6, 18; see section II for details). Clearly, the studies by Olive et al. and Lang et al. demonstrate the complexity of tumorigenesis by p53 mutations beyond loss of tumor suppression, and suggests that genotyping of p53 mutations are important for understanding the biology of the tumor, as well as making treatment decisions in the clinic.
VI. References
1. Levine, A.J. 1997. p53, the cellular gatekeeper for growth and division. Cell. 88:323-31.
2. Vogelstein, B., Lane, D., and Levine, A.J. 2000. Surfing the p53 network. Nature. 408:307-10.
3. Bode, A.M., and Dong, Z. 2004. Post-translational modification of p53 in tumorigenesis. Nat Rev Cancer. 4:793-805.
4. Olive, K.P., Tuveson, D.A., Ruhe, Z.C., Yin, B., Willis, N.A., Bronson, R.T., Crowley, D., and Jacks, T. 2004. Mutant p53 gain of function in two mouse models of Li-Fraumeni syndrome. Cell. 119:847-60.
5. Varley, J.M. 2003. Germline TP53 mutations and Li-Fraumeni syndrome. Hum Mutat. 21:313-20.
6. Cadwell, C. and Zambetti, G.P. 2001. The effects of wild-type p53 tumor suppressor activity and mutant p53 gain-of-function on cell growth. Gene. 277:15-30.
7. Donehower, L.A., Harvey, M., Slagle, B.L., McArthur, M.J., Montgomery, C.A. Jr., Butel, J.S., and Bradley, A. 1992. Mice deficient for p53 are developmentally normal but susceptible to spontaneous tumours. Nature. 356:215-21.
8. Jacks, T., Remington, L., Williams, B.O., Schmitt, E.M., Halachmi, S., Bronson, R.T., and Weinberg, R.A. 1994. Tumor spectrum analysis in p53-mutant mice. Curr Biol. 4:1-7.
9. Finlay, C.A., Hinds, P.W., and Levine, A.J. 1989. The p53 proto-oncogene can act as a suppressor of transformation. Cell. 57:1083-93.
10. Milner, J. and Medcalf, E.A. 1991. Cotranslation of activated mutant p53 with wild type drives the wild-type p53 protein into the mutant conformation. Cell. 65:765-74.
11. Jeffrey, P.D., Gorina, S., and Pavletich, N.P. 1995. Crystal structure of the tetramerization domain of the p53 tumor suppressor at 1.7 angstroms. Science. 267:1498-502.
12. Dittmer, D., Pati, S., Zambetti, G., Chu, S., Teresky, A.K., Moore, M., Finlay, C., and Levine, A.J. 1993. Gain of function mutations in p53. Nat Genet. 4:42-6.
13. Harvey, M., Vogel, H., Morris, D., Bradley, A., Bernstein, A., and Donehower, L.A. 1995. A mutant p53 transgene accelerates tumour development in heterozygous but not nullizygous p53-deficient mice. Nat Genet. 9:305-11.
14. Tuveson, D.A., Shaw, A.T., Willis, N.A., Silver, D.P., Jackson, E.L., Chang, S., Mercer, K.L., Grochow, R., Hock, H., Crowley, D., Hingorani, S.R., Zaks, T., King, C., Jacobetz, M.A., Wang, L., Bronson, R.T., Orkin, S.H., DePinho, R.A., and Jacks, T. 2004. Endogenous oncogenic K-ras(G12D) stimulates proliferation and widespread neoplastic and developmental defects. Cancer Cell. 5:375-87.
15. Lang, G.A., Iwakuma, T., Suh, Y.A., Liu, G., Rao, V.A., Parant, J.M., Valentin-Vega, Y.A., Terzian, T., Caldwell, L.C., Strong, L.C., El-Naggar, A.K., and Lozano, G. 2004. Gain of function of a p53 hot spot mutation in a mouse model of Li-Fraumeni syndrome. Cell. 119:861-72.
16. Montes de Oca Luna, R., Wagner, D.S., and Lozano, G. 1995. Rescue of early embryonic lethality in mdm2-deficient mice by deletion of p53. Nature. 378:203-6.
17. Jones, S.N., Roe, A.E., Donehower, L.A., and Bradley, A. 1995. Rescue of embryonic lethality in Mdm2-deficient mice by absence of p53. Nature. 378:206-8.
18. Vousden, K.H. and Prives, C. 2005. P53 and prognosis: new insights and further complexity. Cell. 120:7-10.