To prevent accumulation of genetic lesions, cells subjected to genotoxic stress have two equally important ways to respond. Most commonly, DNA damage elicits a replicative arrest at various cell cycle checkpoints in order to give the cell time to accomplish DNA repair. Alternatively, in the case of intolerable or unrepairable DNA lesions upon excessive insult, apoptosis remains the only possibility to remove genetically damaged material, which may otherwise predispose the cell to genomic instability and cancer.
The ability of p53 tumor suppressor to inhibit malignant transformation and cancer formation has been addressed as far as its capability to control cell cycle progression and induction of apoptosis. A role of p53 as a mediator of growth arrest and apoptosis has been suggested, since it is well established that after g-irradiation, and many other DNA damaging stimuli, p53 is needed for normal cellular response. At the beginning of our study, little was known about the activation pattern and requirement of p53 after UV radiation, which causes DNA damage very different from those generated by ionizing radiation.
To study the action of p53 in UV-damaged cells, a fruitful approach turned out to be UV treatment of mouse fibroblasts synchronized to a particular cell cycle phase. This method allowed us to explore in detail the UV-induced cell cycle responses and cell cycle phase specific activation of p53 as well as other regulators of proliferation. Contrary to expectations, p53 accumulation failed to occur in UV-treated, G1-phase arrested cells, but accumulation was observed when the cells recovered from the G1 arrest and entered the S phase or, alternatively, when the cells were irradiated and subsequently arrested in S. Rather than being dependent on the cell cycle phase as such, p53 accumulation in G1/S and S phases appeared to be dependent on replication of damaged DNA. Furthermore, pRB hypophosphorylation, which was shown to occur independently of functional p53, correlated with the kinetics of both general UV-triggered growth arrest and specific G1-phase arrest. Further studies indicated, however, that despite absent p53 accumulation p53 had a transactivation function in all cell cycle phases. Thus, in contrast to replication-dependent accumulation of p53 its UVC-mediated transactivation seemed independent of cell cycle phase and protein stabilization.
In contrast to the majority of cancers, p53 mutations are seldom detected in human melanomas. Of seven melanoma cell lines studied, three carried a mutation, and cell lines harboring normal p53 expressed high levels of wild-type p53, another special feature of p53 in melanomas. Although all p53 target genes (p21, GADD45, and mdm2) were induced upon UV in cells expressing wild-type p53 - and most of them also in cell lines with mutant p53 cell lines - their induction was dissociated from p53 function. Interestingly, GADD45 induction, but not p53, correlated with growth arrest and induction of apoptosis. In addition to abnormal stabilization of wild-type p53, many aspects of the p53 pathway showed abnormalities, and UV responses did not significantly differ among cells carrying mutant or normal p53 suggesting functional inactivation of wild-type p53 in melanoma cell lines.
Since previous studies were highly suggestive of p53-independent regulation of p21 cyclin-kinase inhibitor, UVC responses and p21 activation were examined in p53-/- mouse fibroblasts. UVC radiation induced rapidly and efficiently both protein and mRNA levels of p21 also in the absence of p53. In contrast to p53+/+ MEFs, however, high UVC doses abrogated p21 protein induction in p53-deficient cells, and secondly, p53-/- cells were more prone to apoptosis, suggesting a requirement of p53 function in these responses. p21 promoter assays confirmed transcriptional activation of p21, and UV-inducibility was mapped to two regions in the p21 promoter, both lacking p53-binding sites.
These studies have particularly shed light on the cell cycle phase-specific regulation of p53 accumulation and transactivation in UVC-damaged cells. In addition, by using several different approaches, p21 induction upon UV treatment was confirmed to occur at the transcriptional level also independently of p53. Finally, despite a wild-type coding sequence of p53 several p53 regulatory steps were proved to be disturbed in melanoma cell lines suggesting that defective function of p53 may, after all, play a role also in this cancer type.