Deletion of p37Ing1 in mice reveals a p53-independent role for Ing1 in the suppression of cell proliferation, apoptosis, and tumorigenesis

Abstract

ING proteins have been proposed to alter chromatin structure and gene transcription to regulate numerous aspects of cell physiology, including cell growth, senescence, stress response, apoptosis, and transformation. ING1, the founding member of the inhibitor of growth family, encodes p37(Ing1), a plant homeodomain (PHD) protein that interacts with the p53 tumor suppressor protein and seems to be a critical cofactor in p53-mediated regulation of cell growth and apoptosis. In this study, we have generated and analyzed p37(Ing1)-deficient mice and primary cells to further explore the role of Ing1 in the regulation of cell growth and p53 activity. The results show that endogenous levels of p37(Ing1) inhibit the proliferation of p53-wild-type and p53-deficient fibroblasts, and that p53 functions are unperturbed in p37(Ing1)-deficient cells. In addition, loss of p37(Ing1) induces Bax expression and increases DNA damage-induced apoptosis in primary cells and mice irrespective of p53 status. Finally, p37(Ing1) suppresses the formation of spontaneous follicular B-cell lymphomas in mice. These results indicate that p53 does not require p37(Ing1) to negatively regulate cell growth and offers genetic proof that Ing1 suppresses cell growth and tumorigenesis. Furthermore, these data reveal that p37(Ing1) can negatively regulate cell growth and apoptosis in a p53-independent manner.

Figure 1

Generation of p37^Ing1-deficient mice. A, schematic of the gene-trapped locus showing the gene trap inserted into exon C. The trap interrupts the Ing1b message isoform and corresponding p37 protein but encodes the shorter Ing1c isoform and p31 protein. B, PCR genotyping of the embryonic stem cells and mouse tail biopsies showing the presence of the gene trap. A 650-bp fragment is generated from Wt Ing1 using primers 1 and 2 and a 700-bp fragment is generated from the targeted Ing1 allele using primers 3 and 4. C, a Southern blot strategy using a BglII digest and a probe to the common exon was used to confirm germ-line transmission of the gene trap. The Wt fragment is ~8 kb and the mutant fragment is ~10 kb in length. A nonspecific (NS) band corresponding to an Ing1 pseudogene present in 129 strain mouse DNA is also observed. D, to confirm that the longer p37^Ing1 form was specifically deleted, a Western blot was done with an Ing1 antibody that recognizes sequences encoded in the common exon. The presence of the shorter form (p31) and the absence of the longer form (p37) were observed. Tubulin was used as a loading control.

Figure 2

Growth regulation byp37^Ing1. A, proliferation of p37^Ing1-deficient cells. Two independent MEF cell lines of p37^Ing1-Wt, p37^Ing1-heterozygous, p37^Ing1-null, or p53-null genotypes were plated in triplicate in 60-mm plates and growth was analyzed over the course of 7 d. Statistical analysis between Wt and p37^Ing1-heterozygous curves shows no significant difference (P < 1.00), but there is statistical difference between Wt and p37^Ing1-null curves (P < 0.005). B, growth of p37^Ing1-deficient cells at low density. Two lines of Wt, p37^Ing1-heterozygous, or p37^Ing1-homozygous genotype were seeded at 10⁴ cells per 10-cm plate and incubated for 8 to 12 d before staining with crystal violet. C, growth inhibition by p37^Ing1 is independent of p53 status. Three MEF cell lines deficient for p53 or for p53 and p37^Ing1 were plated in triplicate in six-well plates and counted over a 7-d period. Wild-type MEFs were also similarly plated as a control. A statistical difference exists between the proliferation rates of p53-null and p37^Ing1/p53-double null MEFs (P < 0.0005). D, immortalization of p37^Ing1-deficient cells. A 3T9 assay was done using two independent MEF cell lines of p37^Ing1-Wt, p37^Ing1-null, or p53-null MEFs.

Figure 3

Cell cycle arrest and ras-induced senescence are normal in p37^Ing1-deficient MEFs, but apoptosis is elevated. A, oncogene-induced senescence in p37^Ing1-deficient cells. Triplicate experiments were done wherein multiple lines each of Wt, p37^Ing1-null, or p53-null MEFs were transduced with recombinant virus with or without H-ras, and cells were plated at equal densities and counted. Columns, ratio of cell number after 6 d in culture versus the cell number at initial plating. B, absence of p37^Ing1 does not alter cell cycle arrest due to DNA-damaging agents. Three lines each of wt, p37^Ing1-null, or p53-null MEFs were plated in duplicate and synchronized in their growth before mock treatment or treatment with Adriamycin, ionizing radiation, or UVC. Cells were stained with BrdUrd and propidium iodide and analyzed by FACS. Columns, ratio of treated cells in S phase versus untreated cells in S phase. C, triplicate experiments of multiple lines of E1A-transduced MEFs were mock treated or treated with Adriamycin for 24 h. Cell viability was determined by staining with propidium iodide before FACS analysis. Representative result and sub-G₁ content derived from three separate experiments. Columns, mean; bars, SD.

Figure 4

Apoptosis is altered in p37^Ing1-deficient thymocytes. A, spontaneous apoptosis is unchanged in p37^Ing1-deficient thymocytes. Thymocytes were isolated from two Wt, p37^Ing1-null, or p53-null mice, plated in triplicate, and cultured for 4 d ex vivo. Cells were harvested at 24-h intervals, fixed in 70% ethanol, and stained with propidium iodide and analyzed by FACS. B, thymocytes were isolated from Wt, p37^Ing1-null, p53-null, or p37^Ing1/p53-double null mice and either mock treated or irradiated with 2.5 Gy of ionizing radiation, fixed in 70% ethanol after 4 h, and stained with propidium iodide for FACS. Columns, average of three separate experiments. The difference in apoptosis between Wt and p37^Ing1-null or between p53-null and p37^Ing1/p53-double null thymocytes is statistically significant (P < 0.005). C and D, real-time PCR was done with primers to Puma or Bax on tissues harvested from mice that were either mock treated or whole-body treated with 10 Gy of ionizing radiation and harvested 4 h later. Columns, average values (relative fold induction) of three mice for each genotype. The difference in Bax expression is statistically significant between Wt and p37^Ing1-null tissues for thymus and spleen (P < 0.05), but not for brain and liver (P < 0.07). The difference in Bax expression between p53-null and p37^Ing1/p53-double null tissues is statistically significant for thymus, spleen, and brain (P < 0.05), but not for liver (P < 0.06).

Figure 5

Tumorigenesis in mice deficient for p37^Ing1. A, Kaplan-Meier survival curves of cohorts of Wt, p37^Ing1-heterozygous, or p37^Ing1-null mice. Mice that were either moribund or reached 22 mo of age were sacrificed for necropsy and fixed in 10% phosphate-buffered formalin. The rate of tumor incidence between Wt and either p37^Ing1-heterozygous or p37^Ing1-null mice is statistically significant (P < 0.005 or P < 0.05, respectively). B, tumors arising in mice were paraffin embedded and H&E stained for pathologic analysis. Tumor sections were also stained with either B220 or CD3 antibodies to determine tissue of tumor origin.