Loss of poly(ADP-ribose) polymerase-1 causes increased tumour latency in p53-deficient mice

Abstract

PARP-1-deficient mice display a severe defect in the base excision repair pathway leading to radiosensitivity and genomic instability. They are protected against necrosis induced by massive oxidative stress in various inflammatory processes. Mice lacking p53 are highly predisposed to malignancy resulting from defective cell cycle checkpoints, resistance to DNA damage-induced apoptosis as well as from upregulation of the iNOS gene resulting in chronic oxidative stress. Here, we report the generation of doubly null mutant mice. We found that tumour-free survival of parp-1(-/-)p53(-/-) mice increased by 50% compared with that of parp- 1(+/+)p53(-/-) mice. Tumour formation in nude mice injected with oncogenic parp-1(-/-)p53(-/-) fibroblasts was significantly delayed compared with parp-1(+/+)p53(-/-) cells. Upon gamma-irradiation, a partial restoration of S-phase radiosensitivity was found in parp-1(-/-)p53(-/-) primary fibroblasts compared with parp-1(+/+)p53(-/-) cells. In addition, iNOS expression and nitrite release were dramatically reduced in the parp-1(-/-)p53(-/-) mice compared with parp-1(+/+)p53(-/-) mice. The abrogation of the oxydated status of p53(-/-) cells, due to the absence of parp-1, may be the cause of the delay in the onset of tumorigenesis in parp-1(-/-)p53(-/-) mice.

Fig. 1. Kaplan–Meier plot of tumour incidence in parp-1^+/+p53^–/– and parp-1^–/–p53^–/– mice. The percentage of mice remaining tumour free versus mice alive at the outset is plotted against age. The difference was highly significant (p <10^–5 by Gehan’s Wilcoxon test and p <10^–5 by log-rank test). Note that during the time of study, no parp-1^+/+p53^+/+ or parp-1^–/–p53^+/+ mice died.

Fig. 2. (A) Western blot analysis of PARP-1 and p53 expression in nuclear (p53) or crude (PARP-1) extracts of MEFs of the indicated genotypes. p53 protein was induced after 5 Gy irradiation. (B) Southern blot of HindIII-digested DNA from parp-1^+/+p53^–/– and parp-1^–/–p53^–/– MEFs infected or not with H-ras V12 retrovirus, using the ras V12 probe. (C) Morphology of primary and transformed parp-1^+/+p53^+/+ and parp-1^–/–p53 fibroblasts. (D) Anchorage-independent growth of transformed parp-1^+/+p53^–/– and parp-1^–/–p53^–/– fibroblasts.

Fig. 3. (A) Tumour formation in BALB/c nude mice. This representative mouse was inoculated with transformed parp-1^+/+p53^–/– fibroblasts in the right thigh and transformed parp-1^–/–p53^–/– fibroblasts in the left thigh. (B) Tumour volume plot after s.c. injection of parp-1^+/+p53^–/– and parp-1^–/–p53^–/– transformed fibroblasts mixed in equal volume of Matrigel into Balb/c nude mice. For each genotype, six sites were injected. Each point represents the mean ± SD of measurements. The difference was significant at all data points (p <0.006, Mann–Whitney U-test).

Fig. 4. (A) Representative flow cytometry scatter plots of MEFs nuclei 18 h after irradiation (5 Gy), plotted as increasing fluorescence of propidium iodide (PI; x-axis) versus increasing FITC fluorescence obtained with an anti-BrdU–FITC-conjugated antibody (y-axis). Gating for S-phase cells is shown in the panel from unirradiated wild-type mice (upper left). (B) Graph representing the mean ± SD change in the percentage of S-phase cells after irradiation (n = 3–6).

Fig. 5. Micronuclei formation in primary MEFs at passage 2 isolated from mice of the indicated genotypes. A total of 700–1000 binucleated cells in each genotype was counted.

Fig. 6. (A) Western blot analysis of iNOS expression in crude extracts of peritoneal macrophages from mice of the genotypes indicated, 18 h after LPS treatment. Expression of actin is shown as a loading control. (B) Nitrite release in primary cultured murine macrophages from mice of the indicated genotype treated or not with 5 µg/ml LPS for 24 h. Each value represents the mean ± SD of three independent experiments. (C) iNOS expression in crude extracts of splenocytes derived from parp-1^+/+p53^–/– and parp-1^–/–p53^–/– mice.