Recently, we described a p53 mutant mouse model (p53+/m mouse model) that presented a paradox, in which tumor free survival was coupled to reduced lifespan instead of extended lifespan (Tyner et al ., 2002). The p53+/m mice expressed a truncated p53 transcript attached to a leader sequence and showed a moderate increase in p53 activity. Based on this data we hypothesized that enhanced levels of p53 caused the early aging and cancer resistance phenotypes. This hypothesis was attractive as it indirectly implied a 'fine tuning' mechanism of tumor suppressor gene expression to maintain homeostasis at the organismal level (Campisi, 2002;Kirkwood, 2002). Our results also lent support to the evolutionary theory of antagonistic pleiotropy that predicts the deleterious effects of tumor-suppressor genes later in life (Williams, 1957). These observations notwithstanding, the p53+/m mouse model presented a caveat, in which tumor free survival was coupled to reduced lifespan instead of extended lifespan. Studies in model organisms have shown that enhanced resistance to oxidative stress correlates with increased longevity (Johnson et al ., 2002;Murakami et al ., 2003). The tumor suppressor functions of p53 overwhelmingly categorize it as a longevity assurance gene and suggest a synergistic effect on lifespan via tumor free survival of the organism. We show that there are 24 genes deleted in the p53+/m mouse model, complicating the role of p53 in aging and raising the possibility that the phenotypes while resembling 'premature aging-like' symptoms, may not be related to the physiological changes seen during normal aging processes.Along with the partial deletion of one of the p53 alleles, the p53+/m mouse had an upstream deletion that was originally not well defined due to the lack of mouse genome sequence data. This undefined deletion was a major drawback for our hypothesis, leaving open the possibility that the loss of other unidentified genes affected the p53+/m phenotypes. The importance of the upstream deletion was further underlined by the fact that we were unable to generate a p53 m/m mouse from p53+/m crosses even though p53 nullizygosity does not lead to embryonic lethality in mice (Donehower et al ., 1992). To determine the extent of the deletion, we blasted the leader sequence (initially cloned along with the truncated p53 transcript) to the Ensembl mouse genome sequence database and found it to be within the Rho guanine exchange factor 15 ( Arhgef15 ) gene approximately 0.6 Mb upstream of p53.To ascertain the fusion of the Argef15 and p53 alleles, we performed Southern blot assays with multiple restriction enzymes that were common to both the alleles. A schematic representation of the wild-type alleles of p53 and Argef15 is shown in Fig. 1 along with the restriction enzyme sites used for the characterization of the breakpoint present in the p53+/m mouse model. As shown in Fig. 2(a,b), the fusion of the two alleles leads to changes in the restriction fragment length polymorphism (RFLP) patterns of the wt and 'm' alle...