Fig. 2: Divergent effects of p15 loss or p16 loss in HRas-mediated urothelial tumorigenesis.

a Mouse breeding scheme. Two sets of intercrosses were carried out (i) between Upk2-HRas transgenic mice (HRas*) and p16Ink4a (p16) knockout (p16^−/−) in which exon 1α of the gene was replaced by the neo gene, thus leaving an independent transcript of the gene, p19Arf, intact; and (ii) between HRas* transgenic mice and p15Ink4b (p15) knockout (p15^−/−) mice. b Representative genotyping results by genomic PCR of major experimental groups and controls (1 mouse/lane; 2 mice/genotype). The genotyping was performed on every mouse with highly reproducible results. c Western blotting of p16, p15, p19, ERK1/2 and phosphorylated ERK1/2, and β-actin (loading control) in urothelial cells of specific genotypes. Each lane represents one mouse and two independent mice were used per genotype. The experiment was repeated twice with similar results. Source data are provided as a Source Data file. d Divergent tumorigenesis in urothelial cells expressing HRas* and lacking p16 versus those expressing HRas* and lacking p15 (n = 10/time point/genotype). P value shown resulted from the comparison between HRas^WT/*/p16^−/− and HRas^WT/*/p15^−/− mice, using the Kaplan–Meier method followed by log-rank statistical test by software SPSS v17.0. Note the markedly earlier onset and higher frequency of tumorigenesis in HRas^WT/*/p15^−/− mice than in HRas^WT/*/p16^−/− mice. Also note that, while the significant induction of p16 in p15-lacking mice did not prevent bladder tumor formation, the significant induction of p15 in p16-lacking mice did exert a strong tumor-suppressive effect (Fig. 1d). e Bladder tumorigenic status in HRas^WT/*/p16^−/− versus HRas^WT/*/p15^−/− mice. Representative images of H&E-stained cross-sections of bladders from mice (age 2–10 months) expressing oncogenic HRas and lacking p16 or p15 as shown in the cohort in Fig. 1d. Note that the HRas^WT/*/p15^−/− mice developed nodular urothelial hyperplasia at 2 months of age and superficial papillary bladder tumors from 4 months onward, and that in contrast, the bladders of the HRas^WT/*/p16^−/− mice only contained mucosa folds from simple urothelial hyperplasia, and were largely devoid of tumors (see corresponding high-powered images for red-boxed areas in Supplementary Fig. 1). All panels are of the same magnification and the scale bar in the upper-left panel equals 500 μm. n = 10 mice per genotype per time point.