Allele-specific deletions in mouse tumors identify Fbxw7 as germline modifier of tumor susceptibility

Genome-wide association studies (GWAS) have been successful in finding associations between specific genetic variants and cancer susceptibility in human populations. These studies have identified a range of highly statistically significant associations between single nucleotide polymorphisms (SNPs) and susceptibility to development of a range of human tumors. However, the effect of each SNP in isolation is very small, and all of the SNPs combined only account for a relatively minor proportion of the total genetic risk (5-10%). There is therefore a major requirement for alternative routes to the discovery of genetic risk factors for cancer. We have previously shown using mouse models that chromosomal regions harboring susceptibility genes identified by linkage analysis frequently exhibit allele-specific genetic alterations in tumors. We demonstrate here that the Fbxw7 gene, a commonly mutated gene in a wide range of mouse and human cancers, shows allele-specific deletions in mouse lymphomas and skin tumors. Lymphomas from three different F1 hybrids show 100% allele-specificity in the patterns of allelic loss. Parental alleles from 129/Sv or Spretus/Gla mice are lost in tumors from F1 hybrids with C57BL/6 animals, due to the presence of a specific non-synonymous coding sequence polymorphism at the N-terminal portion of the gene. A specific genetic test of association between this SNP and lymphoma susceptibility in interspecific backcross mice showed a significant linkage (p = 0.001), but only in animals with a functional p53 gene. These data therefore identify Fbxw7 as a p53-dependent tumor susceptibility gene. Increased p53-dependent tumor susceptibility and allele-specific losses were also seen in a mouse skin model of skin tumor development. We propose that analysis of preferential allelic imbalances in tumors may provide an efficient means of uncovering genetic variants that affect mouse and human tumor susceptibility.     

Probing p53 biological functions through the use of genetically engineered mouse models

The p53 tumor suppressor gene is rendered dysfunctional in the majority of human cancers. To model the effects of p53 dysfunction in an experimentally manipulable organismal context, genetically engineered inbred mice have been the models of choice. Transgenic and knock-out technologies have been utilized to generate an array of different p53 germ line alterations. As expected, many (though not all) of the mutant p53 mouse models are susceptible to enhanced spontaneous and carcinogen-induced tumors of a variety of types. A number of different variables affect the incidence and spectrum of tumors in p53 mutant mice. These include strain background, the nature of the p53 mutation, the presence of wild-type p53 (in addition to mutant p53), exposure to physical and chemical mutagens, or introduction of other cancer-associated genes into the mutant p53 background. In addition to their role in furthering our understanding of the mechanisms of cancer initiation and progression, these models have led to unexpected insights into p53 function in embryogenesis and aging. With the development of ever more sophisticated methods for manipulating the mouse genome, new p53 models are on the horizon, which should deliver advances that will provide not only important mechanistic insights but also discoveries of great clinical relevance.     

Radiation-induced genetic instability in vivo depends on p53 status

In response to ionizing radiation and other agents that damage DNA, the p53 tumor suppressor protein activates multiple cellular processes including cell cycle checkpoints and programmed cell death. Although loss of p53 function is associated with radiation-induced genetic instability in cell lines, it is not clear if this relationship exists in vivo. To study the role of p53 in maintenance of genetic stability in normal tissues following irradiation, we have measured mutant frequencies at the adenine phosphoribosyltransferase (Aprt) and hypothanine-guanine phosphoribosyltransferase (Hprt) loci and examined mechanisms of loss of heterozygosity (LOH) in normal T cells of p53-deficient, Aprt heterozygous mice that were subjected to whole-body irradiation with a single dose of 4Gy X-rays. The radiation-induced mutant frequency at both the Aprt and Hprt loci was elevated in cells from mice with different p53 genotypes. The radiation-induced elevation of p53-/- mice was significantly greater than that of p53+/- or p53+/+ mice and was caused by several different kinds of mutational events at the both chromosomal and intragenic levels. Most significantly, interstitial deletion, which occurs rarely in unirradiated mice, became the most common mechanism leading to LOH in irradiated p53 null mice. These observations support the idea that absence or reduction of p53 expression enhances radiation-induced tumorigenesis by increasing genetic instability at various loci, such as those for tumor suppressor genes.     

p21 protects "Super p53" mice from the radiation-induced gastrointestinal syndrome

Exposure of the gastrointestinal (GI) tract to high doses of radiation can lead to lethality from the GI syndrome. Although the molecular mechanism regulating the GI syndrome remains to be fully defined, we have recently demonstrated that p53 within the GI epithelial cells controls the radiation-induced GI syndrome. Mice lacking p53 in the GI epithelium were sensitized to the GI syndrome, while transgenic mice with one additional copy of p53 called "Super p53" mice were protected from the GI syndrome. Here, we crossed Super p53 mice to p21?/? mice that lack the cyclin-dependent kinase inhibitor p21. Super p53; p21?/? mice were sensitized to the GI syndrome compared to Super p53 mice that retain one p21 allele. In addition, mice lacking p21 were not protected from the GI syndrome with one extra copy of p53. These results suggest that p21 protects Super p53 mice from the GI syndrome.     

The absence of p53 promotes metastasis in a novel somatic mouse model for hepatocellular carcinoma

We have generated a mouse model for hepatocellular carcinoma using somatic delivery of oncogene-bearing avian retroviral vectors to the liver cells of mice expressing the viral receptor TVA under the control of the albumin gene promoter (Alb-TVA mice). Viruses encoding mouse polyoma virus middle T antigen (PyMT) induced tumors, which can be visualized with magnetic resonance imaging, in 65% of TVA-positive animals. While these tumors can exceed 10 mm in diameter, they do not invade locally or metastasize to the lungs. Delivery of PyMT-expressing viruses to Alb-TVA mice lacking an intact p53 gene does not increase tumor incidence. However, the resulting tumors are poorly differentiated, invasive, and metastatic to the lungs. Gene expression microarrays identified over 100 genes that are differentially expressed between tumors found in p53 wild-type and p53 null mice. Some of these genes, such as cathepsin E and Igf2, have been previously implicated in tumor cell migration and invasion. Tumors induced in p53 null, TVA transgenic mice by PyMT mutants with changes in specific tyrosine residues fail to form metastases, indicating that metastasis is dependent on both the oncogene and the absence of p53.     

neu/ERBB2 cooperates with p53-172H during mammary tumorigenesis in transgenic mice.

Thirty percent of human breast cancers have amplification of ERBB2, often in conjunction with mutations in p53. The most common p53 mutation in human breast cancers is an Arg-to-His mutation at codon 175, an allele that functions in a dominant oncogenic manner in tumorigenesis assays and is thus distinct from loss of p53. Transgenic mice expressing mouse mammary tumor virus-driven neu transgene (MMTV-neu) develop clonal mammary tumors with a latency of 234 days, suggesting that other events are necessary for tumor development. We have examined the role of mutations in p53 in tumor development in these mice. We have found that 37% of tumors arising in these mice have a missense mutations in p53. We have directly tested for cooperativity between neu and mutant p53 in mammary tumorigenesis by creating bitransgenic mice carrying MMTV-neu and 172Arg-to-His p53 mutant (p53-172H). In these bitransgenic mice, tumor latency is shortened to 154 days, indicating strong cooperativity. None of the nontransgenic mice or the p53-172H transgenic mice developed tumors within this time period. Tumors arising in the p53-172H/neu bitransgenic mice were anaplastic and aneuploid and exhibited increased apoptosis, in distinction to tumors arising in p53-null mice, in which apoptosis is diminished. Further experiments address potential mechanisms of cooperativity between the two transgenes. In these bitransgenic mice, we have recapitulated two common genetic lesions that occur in human breast cancer and have shown that p53 mutation is an important cooperating event in neu-mediated oncogenesis.     

A mild mutator phenotype arises in a mouse model for malignancies associated with neurofibromatosis type 1

Defects in genes that control DNA repair, proliferation, and apoptosis can increase genomic instability, and thus promote malignant progression. Although most tumors that arise in humans with neurofibromatosis type 1 (NF1) are benign, these individuals are at increased risk for malignant peripheral nerve sheath tumors (MPNST). To characterize additional mutations required for the development of MPNST from benign plexiform neurofibromas, we generated a mouse model for these tumors by combining targeted null mutations in Nf1 and p53, in cis. CisNf1+/-; p53+/- mice spontaneously develop PNST, and these tumors exhibit loss-of-heterozygosity at both the Nf1 and p53 loci. Because p53 has well-characterized roles in the DNA damage response, DNA repair, and apoptosis, and because DNA repair genes have been proposed to act as modifiers in NF1, we used the cisNf1+/-; p53+/- mice to determine whether a mutator phenotype arises in NF1-associated malignancies. To quantitate spontaneous mutant frequencies (MF), we crossed the Big Blue mouse, which harbors a lacI transgene, to the cisNf1+/-; p53+/- mice, and isolated genomic DNA from both tumor and normal tissues in compound heterozygotes and wild-type siblings. Many of the PNST exhibited increased mutant frequencies (MF=4.70) when compared to normal peripheral nerve and brain (MF=2.09); mutations occurred throughout the entire lacI gene, and included base substitutions, insertions, and deletions. Moreover, the brains, spleens, and livers of these cisNf1+/-; p53+/- animals exhibited increased mutant frequencies when compared to tissues from wild-type littermates. We conclude that a mild mutator phenotype arises in the tumors and tissues of cisNf1+/-; p53+/- mice, and propose that genomic instability influences NF1 tumor progression and disease severity.     

Inactivation of heat shock factor Hsf4 induces cellular senescence and suppresses tumorigenesis in vivo

Studies suggest that Hsf4 expression correlates with its role in cell growth and differentiation. However, the role of Hsf4 in tumorigenesis in vivo remains unexplored. In this article, we provide evidence that absence of the Hsf4 gene suppresses evolution of spontaneous tumors arising in p53- or Arf-deficient mice. Furthermore, deletion of hsf4 alters the tumor spectrum by significantly inhibiting development of lymphomas that are normally observed in the majority of mice lacking p53 or Arf tumor suppressor genes. Using mouse embryo fibroblasts deficient in the hsf4 gene, we have found that these cells exhibit reduced proliferation that is associated with induction of senescence and senescence-associated β-galactosidase (SA-β-gal). Cellular senescence in hsf4-deficient cells is associated with the increased expression of the cyclin-dependent kinase inhibitors, p21 and p27 proteins. Consistent with the cellular senescence observed in vitro, specific normal tissues of hsf4(-/-) mice and tumors that arose in mice deficient in both hsf4 and p53 genes exhibit increased SA-β-gal activity and elevated levels of p27 compared with wild-type mice. These results suggest that hsf4 deletion-induced senescence is also present in vivo. Our results therefore indicate that Hsf4 is involved in modulation of cellular senescence, which can be exploited during cancer therapy.     

Small-molecule inhibitor of p53 binding to mitochondria protects mice from gamma radiation

p53-dependent apoptosis contributes to the side effects of cancer treatment, and genetic or pharmacological inhibition of p53 function can increase normal tissue resistance to genotoxic stress. It has recently been shown that p53 can induce apoptosis through a mechanism that does not depend on transactivation but instead involves translocation of p53 to mitochondria. To determine the impact of this p53 activity on normal tissue radiosensitivity, we isolated a small molecule named pifithrin-mu (PFTmu, 1) that inhibits p53 binding to mitochondria by reducing its affinity to antiapoptotic proteins Bcl-xL and Bcl-2 but has no effect on p53-dependent transactivation. PFTmu has a high specificity for p53 and does not protect cells from apoptosis induced by overexpression of proapoptotic protein Bax or by treatment with dexamethasone (2). PFTmu rescues primary mouse thymocytes from p53-mediated apoptosis caused by radiation and protects mice from doses of radiation that cause lethal hematopoietic syndrome. These results indicate that selective inhibition of the mitochondrial branch of the p53 pathway is sufficient for radioprotection in vivo.     

Uncoupling p53 functions in radiation-induced intestinal damage via PUMA and p21

The role of p53 in tissue protection is not well understood. Loss of p53 blocks apoptosis in the intestinal crypts following irradiation but paradoxically accelerates gastrointestinal (GI) damage and death. PUMA and p21 are the major mediators of p53-dependent apoptosis and cell-cycle checkpoints, respectively. To better understand these two arms of p53 response in radiation-induced GI damage, we compared animal survival, as well as apoptosis, proliferation, cell-cycle progression, DNA damage, and regeneration in the crypts of WT, p53 knockout (KO), PUMA KO, p21 KO, and p21/PUMA double KO (DKO) mice in a whole body irradiation model. Deficiency in p53 or p21 led to shortened survival but accelerated crypt regeneration associated with massive nonapoptotic cell death. Nonapoptotic cell death is characterized by aberrant cell-cycle progression, persistent DNA damage, rampant replication stress, and genome instability. PUMA deficiency alone enhanced survival and crypt regeneration by blocking apoptosis but failed to rescue delayed nonapoptotic crypt death or shortened survival in p21 KO mice. These studies help to better understand p53 functions in tissue injury and regeneration and to potentially improve strategies to protect or mitigate intestinal damage induced by radiation.     

TP53 and TP53-related genes associated with protection from apoptosis in the radioadaptive response

We investigated the effect of administering priming low-dose radiation prior to high-dose radiation on the level of apoptosis and on the expression of TP53 and TP53-related genes in mouse splenocytes. The percentage of apoptotic cells was significantly lower in TP53(+/+) mice receiving priming radiation 2 to 168 h before the high-dose irradiation, compared to TP53(+/+) mice exposed to 2 Gy alone. In contrast, TP53(+/-) mice exhibited a reduced level of apoptosis only when priming was performed for 2 or 4 h prior to the high-dose irradiation. In TP53(+/+) mice, primed mice had higher TP53 expression than mice exposed to 2 Gy. Phospho-TP53 (ser15/18) expression was the highest in mice exposed to 2 Gy and intermediate in primed mice. Expression of p21 (CDKN1A) was higher in primed mice compared with mice exposed to 2 Gy. MDM2 expression remained at a high level in all mice receiving 2 Gy. Elevated phospho-ATM expression was observed only in mice exposed to 2 Gy. We conclude that TP53 plays a critical role in the radioadaptive response and that TP53 and TP53-related genes might protect cells from apoptosis through activation of the intracellular repair system.     

Targeting AURKA in treatment of peritoneal tumor dissemination in gastrointestinal cancer

Intraperitoneal (i.p.) tumor dissemination and the consequent malignant ascites remain unpredictable and incurable in patients with gastrointestinal (GI) cancer, and practical advances in diagnosis and treatment are urgently needed in the clinical settings. Here, we explored tumor biological and immunological mechanisms underlying the i.p. tumor progression for establishing more effective treatments.We established mouse tumor ascites models that murine and human colorectal cancer cells were both i.p. and subcutaneously (s.c.) implanted in mice, and analyzed peritoneal exudate cells (PECs) obtained from the mice. We then evaluated anti-tumor efficacy of agents targeting the identified molecular mechanisms using the ascites models. Furthermore, we validated the clinical relevancy of the findings using peritoneal lavage fluids obtained from gastric cancer patients.I.p. tumor cells were giant with large nuclei, and highly express AURKA, but less phosphorylated TP53, as compared to s.c. tumor cells, suggesting polyploidy-like cells. The i.p. tumors impaired phagocytic activity and the consequent T-cell stimulatory activity of CD11b⁺Gr1⁺PD1⁺ myeloid cells by GDF15 that is regulated by AURKA, leading to treatment resistance. Blocking AURKA with MLN8237 or siRNAs, however, abrogated the adverse events, and induced potent anti-tumor immunity in the ascites models. This treatment synergized with anti-PD1 therapy. The CD11b⁺PD1⁺ TAMs are also markedly expanded in the PECs of gastric cancer patients.These suggest AURKA is a determinant of treatment resistance of the i.p. tumors. Targeting the AURKA-GDF15 axis could be a promising strategy for improving clinical outcome in the treatment of GI cancer.     

Loss of p53 protein during radiation transformation of primary human mammary epithelial cells.

The causative factors leading to breast cancer are largely unknown. Increased incidence of breast cancer following diagnostic or therapeutic radiation suggests that radiation may contribute to mammary oncogenesis. This report describes the in vitro neoplastic transformation of a normal human mammary epithelial cell strain, 76N, by fractionated gamma-irradiation at a clinically used dose (30 Gy). The transformed cells (76R-30) were immortal, had reduced growth factor requirements, and produced tumors in nude mice. Remarkably, the 76R-30 cells completely lacked the p53 tumor suppressor protein. Loss of p53 was due to deletion of the gene on one allele and a 26-bp deletion within the third intron on the second allele which resulted in abnormal splicing out of either the third or fourth exon from the mRNA. PCR with a mutation-specific primer showed that intron 3 mutation was present in irradiated cells before selection for immortal phenotype. 76R-30 cells did not exhibit G1 arrest in response to radiation, indicating a loss of p53-mediated function. Expression of the wild-type p53 gene in 76R-30 cells led to their growth inhibition. Thus, loss of p53 protein appears to have contributed to neoplastic transformation of these cells. This unique model should facilitate analyses of molecular mechanisms of radiation-induced breast cancer and allow identification of p53-regulated cellular genes in breast cells.     

Early aging-associated phenotypes in Bub3/Rae1 haploinsufficient mice

Aging is a highly complex biological process that is believed to involve multiple mechanisms. Mice that have small amounts of the mitotic checkpoint protein BubR1 age much faster than normal mice, but whether other mitotic checkpoint genes function to prevent the early onset of aging is unknown. In this study, we show that several aging-associated phenotypes appear early in mice that are double haploinsufficient for the mitotic checkpoint genes Bub3 and Rae1 but not in mice that are single haploinsufficient for these genes. Mouse embryonic fibroblasts (MEFs) from Bub3/Rae1 haploinsufficient mice undergo premature senescence and accumulate high levels of p19, p53, p21, and p16, whereas MEFs from single haploinsufficient mice do not. Furthermore, although BubR1 hypomorphic mice have less aneuploidy than Bub3/Rae1 haploinsufficient mice, they age much faster. Our findings suggest that early onset of aging-associated phenotypes in mice with mitotic checkpoint gene defects is linked to cellular senescence and activation of the p53 and p16 pathways rather than to aneuploidy.