The tumor suppressor p53 has an important role in inducing cell-intrinsic responses to DNA damage, including cellular senescence or apoptosis, which act to thwart tumor development. It has been shown, however, that senescent or dying cells are capable of eliciting inflammatory responses, which can have pro-tumorigenic effects. Whether DNA damage-induced p53 activity can contribute to senescence- or apoptosis-associated pro-tumorigenic inflammation is unknown. Recently, we generated a p53 knock-out rat via homologous recombination in rat embryonic stem cells. Here we show that in a rat model of inflammation-associated hepatocarcinogenesis, heterozygous deficiency of p53 resulted in attenuated inflammatory responses and ameliorated hepatic cirrhosis and tumorigenesis. Chronic administration of hepatocarcinogenic compound, diethylnitrosamine, led to persistent DNA damage and sustained induction of p53 protein in the wild-type livers, and much less induction in p53 heterozygous livers. Sustained p53 activation subsequent to DNA damage was accompanied by apoptotic rather than senescent hepatic injury, which gave rise to the hepatic inflammatory responses. In contrast, the non-hepatocarcinogenic agent, carbon tetrachloride, failed to induce p53, and caused a similar degree of chronic hepatic inflammation and cirrhosis in wild type and p53 heterozygous rats. These results suggest that although p53 is usually regarded as a tumor suppressor, its constant activation can promote pro-tumorigenic inflammation, especially in livers exposed to agents that inflict lasting mutagenic DNA damage.
Tumorigenesis has been recognized as not only a process intrinsic to cancer cells, but also a consequence of inflammatory microenvironments.1 Hepatocellular carcinoma (HCC) is a prototype of inflammation-associated cancer, as most human HCC cases have a history of unresolved chronic hepatitis and cirrhosis. Carcinogen exposure and viral infection are major causes of HCC. Genetic damage caused by these agents2 may result in hepatic injury and activation of resident and recruited inflammatory cells, leading to immune surveillance of the damaged cells.3 On the other hand, these inflammatory responses can promote tumorigenesis by creating a microenvironment that stimulates compensatory heopatocyte proliferation, an important adaptive response that maintains liver mass, but which is also an essential factor for HCC development subsequent to genetic damage.4 However, the mechanisms that link genetic damage to induction of hepatic inflammation during liver tumorigenesis remain unknown.
The tumor suppressor p53 provides the primary genetic defense against cancer by mounting DNA damage responses (DDR) to diverse genotoxic stresses.5 Loss or mutation of p53 is common in late stage of liver cancer. However, accumulation of wild-type p53 has been observed in hepatocytes of fibrotic livers as well as in dysplastic liver nodules, particularly in patients with viral hepatitis, but the significance of this is unknown.6, 7, 8 Conditional p53 activation in hepatocytes was shown to induce spontaneous liver fibrosis in mice.9 In contrast, complete deletion of both p53 alleles led to increased formation of fibrotic tissue in another murine model of hepatic fibrosis.10 These apparently contradictory conclusions reflect the complexity of p53 function in the context of inflammation-associated liver diseases.
The rat is the preferred model organism in many fields of biomedicine, including cancer research, owing to its closer physiopathology to humans.11 Recently, we generated the p53 knock-out rat via homologous recombination in rat embryonic stem cells.12 Here we investigated the effect of p53 deficiency in a rat model of carcinogen-induced liver cancer, wherein hepatocarcinogenesis is driven by chronic exposure to diethylnitrosamine (DEN). This model closely recapitulates the inflammation–cirrhosis–HCC axis of its human disease counterpart.13 Our results reveal a possible causal relationship between genetic damage, sustained p53 activation, chronic hepatitis and HCC development.
Results and Discussion
To determine whether the activation of p53 by genetic damage may contribute to HCC development, we subjected wild-type and p53+/− rats to chronic treatment of DEN, which induces hepatocyte DNA damage through DNA adduct formation. This model incorporates persistent genetic damage, chronic injury, unresolved inflammation, progressive cirrhosis and tumorigenesis, and thus shares several features with the microenvironment in which the majority of human HCCs arise (Figure 1a). After 10 weeks of DEN administration, wild-type rats showed evidence of overt hepatic cirrhosis, whereas p53+/− rats had cirrhosis that was far less severe as demonstrated by macroscopic appearance, Sirius red staining and expression of α-smooth muscle actin (Figures 1b and c. In p53+/− rats, strong suppression of hepatic fibrogenesis was already observed 5 weeks after DEN treatment, as indicated by the reduction in messenger RNA expression of early markers of fibrogenesis including TGF-β1 (transforming growth factor β1, encoded by Tgfb1), a-SMA (encoded by Acta2), and collagen-a1(I) (encoded by Col1a1) (Figure 1d). Four weeks after the end of the 10-week DEN treatment, multiple tumors were visible on the livers of wild-type rats, whereas the numbers of detectable tumors were significantly lower on the livers of p53+/− rats (Figure 1a). Histological analysis revealed the presence of more malignant tumors as well as pre-malignant nodular lesions (dysplastic foci, low- and high-grade dysplastic nodules) in wild-type rats than in p53+/− rats, although most tumors in the two groups of rats were of similar grade (Supplementary Figure S1). Notably, inflammatory infiltration of p53+/− rat livers was much less than that of wild-type controls, as evidenced by lower numbers of neutrophils, lymphocytes and macrophages within portal tracts (Figure 2a). Accordingly, the circulating levels of proinflammatory cytokines such as interleukin (IL)-6 and tumor necrosis factor (TNF)α were detected to be lower in p53+/− rats (Figure 2b). To ascertain the inflammatory effects of p53, we administered DEN to wild-type and p53+/− rats for 1 week, and found that the levels of circulating IL-6 and TNFα and their corresponding hepatic mRNAs were also significantly lower in DEN-treated p53+/− rats compared with wild-type controls (Supplementary Figures S2a and b).
Consistent with the observably lower inflammation, p53+/− livers exhibited significantly less hepatic injury and consequent compensatory cell proliferation than wild-type counterparts after DEN administration, as determined by serum alanine aminotransferase (ALT) levels and bromodeoxyrudine (BrdU) uptake, respectively (Figures 3a and b). The difference in hepatic injury is likely unattributable to diminished DEN cytotoxity in p53+/− rats as expression of CYP2E1, which is responsible for DEN bioactivation in hepatocytes,14 did not differ between wild-type and p53+/− rats, whereas the induction of p53 target gene, Mdm2 by DEN, was significantly suppressed in p53+/− and p53−/− rats (Supplementary Figures S2c and d. Although p53 is known to stimulate accumulation of reactive oxygen species (ROS),15 which are potent inducers of hepatic injury and inflammation, the number of cells positive for ROS (superoxide) indicator were similar between DEN-treated wild-type and p53+/− livers (Supplementary Figure S3a). Next we investigated the regenerative status of the liver upon 70% partial hepatectomy, which is a well-established model of liver regeneration. No differences in hepatic injury and compensatory hepatocyte proliferation, as determined by serum ALT levels and Ki67 staining, were seen between the two groups after partial hepatectomy (Supplementary Figures S3c and d, indicating that p53 deficiency does not affect the regenerative capacity of the liver. We then examined whether the observed differences in liver damage are due to p53-mediated apoptosis or senescence, both of which are generally considered as potent tumor suppressor mechanisms that irreversibly prevent tumorigenesis. TUNEL (terminal deoxynucleotidyl transferase (TdT)-mediated dUTP nick-end labeling) staining of the liver tissue revealed a decrease in TUNEL-positive cells in p53+/− rats compared with the control littermates. We also found that the numbers of hepatocyte positive for cleaved caspase-3, an active form of caspase-3, were also significantly lower in p53+/− rats than in wild-type controls (Figures 3c and d). These findings indicate that heterozygous deletion of p53 led to protection against DEN-induced hepatocyte apoptosis. Senescence-associated β-galactosidase staining of the liver sections was also performed, and showed that senescent hepatocytes were not obvious in either group (not shown). These observations are analogous to results seen recently in murine livers, conditionally deleting the gene encoding Mdm2, a protein that promotes p53 degradation, where activation of p53 often causes hepatocytes apoptosis and spontaneous liver fibrosis.9
To corroborate our results from the DEN model indicating p53 promotes hepatic cirrhosis, we used another model of cirrhosis, based on administration of carbon tetrachloride (CCl4). Surprisingly, in contrast to earlier findings in the DEN model, both wild-type and p53+/− rats showed a similar degree of overall cirrhosis, liver injury and inflammatory cytokine production after eight doses of CCl4 (Figure 4a and Supplementary Figure S4). To interrogate the mechanisms underlying the discrepant results between these models, we examined p53 activation in livers of DEN- or CCl4-treated rats. In saline-treated control livers, p53 was inactive or even undetectable. DEN treatment resulted in a marked induction of p53 protein in the wild-type livers and much less induction in p53+/− livers. In contrast, the levels of p53 remained essentially at control levels in CCl4-treated livers. Interestingly, the expression of p53 target gene, p21, was constitutively elevated in both DEN- and CCl4-treated livers, and did not differ between wild-type and p53+/− livers, suggesting p53-independent regulation of p21 during hepatic fibrogenesis (Figure 4b and Supplementary Figure S5). The DDR and p19ARF tumor suppressor are two major activation arms of p53. Immunohistochemical analysis revealed prominent induction of H2A.X phosphorylation in DEN-treated livers analyzed 48 h and 10 weeks after treatment initiation, but not in CCl4-treated livers analyzed 8 weeks after treatment initiation, which indicates that DNA damage was present throughout the DEN treatment (Figure 4c). Furthermore, immnohistochemical analysis of phosphorylated Chk2, a downstream effector of the ATM-dependent DNA damage checkpoint pathway, also showed a comparable degree of DEN-induced DDR in wild-type and p53+/− livers (Supplementary Figures S5b and c. Unlike phospho-H2A.X, p19ARF expression was confined mostly to high-grade dysplastic nodules or advanced HCCs (Figure 4c), suggesting that the p19ARF/p53 pathway is only engaged at later stages of tumor evolution when oncogenic signals are excessively elevated. These data point to robust p53 activation following DEN-induced DNA damage as a probable mechanism for inducing the pro-tumorigenic inflammation in the liver.
Most human HCCs arise in cirrhotic livers exposed to DNA-damaging agents, such as aflatoxin B1, and hepatitis B and C viruses. Virtually all cirrhosis-inducing conditions contribute to HCC development, pointing to important interactions with the host microenvironment.16 The development of well-defined models of liver cancer that recapitulate the human pathological state is essential for elucidating the molecular basis of carcinogenesis and evaluating new antitumor strategies. Although a number of genetically modified mice have been shown to develop spontaneous hepatic tumors, few of the tumors arise from pre-established liver cirrhosis as observed in most patients.17 These models might aid in understanding the aberrant cell-intrinsic signaling that drives oncogenesis, but they do not recapitulate the natural development of cancer in patients, especially in those chronically infected with hepatitis viruses or exposed to carcinogen. For the carcinogen-induced hepatocarcinogenesis model, DEN is most widely used because of its potency and the reproducibility of HCC induction. However, in the murine model of DEN-induced HCC, only one single postnatal injection is administered, so the sequence of events leading to persistent DNA damage, fibrosis, cirrhosis and tumor is completely skipped. In contrast, chronic exposure of rats to DEN gives rise to multistage hepatocarcinogenesis, which can faithfully recapitulate the natural history of persistent DNA damage in liver and chronic hepatic inflammation during human HCC development.13
Using the rat model of DEN-induced HCC, we demonstrate that persistent DNA damage caused by DEN led to sustained p53 activation and enhanced pro-tumorigenic inflammation in liver. Heterozygous deletion of p53 dramatically attenuated the cirrhotic inflammation and hepatic tumorigenesis. By contrast, p53 deficiency in mice was previously shown to fail to significantly change the rate of development of DEN-induced HCC 8 months after the single postnatal injection of DEN, in which no apparent hepatic inflammation was observed.18 This apparent contradiction can be explained by the unique features of the hepatic inflammation, which is mostly triggered by hepatocyte death. Hepatic responses to injury are determined by the interplay between parenchymal and nonparenchymal liver cells, in which hepatocyte death results in activation of Kupffer cells that in turn produce cytokines to promote regeneration.19 Interestingly, p53-deficient mice displayed a lower incidence of chemically induced skin tumors than did wild-type mice. As the skin model also entails a large inflammatory component, our study may shed light on a potentially more general mechanism, linking p53 to inflammation and cancer development.20 Our findings that hepatocyte apoptosis was markedly decreased in p53+/− livers after DEN treatment suggest that protection from apoptosis conferred by p53 deficiency mitigates chronic state of hepatocyte death, inflammatory responses and regenerative proliferation, which together constitutes a risk factor for cancer development.
Among the chemicals that have been used in the induction of liver cirrhosis, CCl4 is most commonly used because of its strong liver-specific cytotoxicity. Although CCl4 has been shown to promote carcinogen-induced hepatocarcinogenesis, it is largely non-hepatocarcinogenic when used alone possibly because of a lack of genotoxic effects. An interesting finding in our study is that no significant difference in liver cirrhosis has been noticed between wild-type and p53+/− rats after repeated administration of CCl4, which is in sharp contrast to the differences observed in DEN-treated animals. Our results show that DEN induced high levels of p53 protein in the rat liver, whereas CCl4 did not increase p53 in this organ. The induction of p53 by DEN is probably unrelated to its cytotoxicity, as p53 induction was already observed at nontoxic doses.21 DEN-induced DDR seem to be mainly responsible for the potent induction of p53, as no apparent phosphorylation of H2A.X was evident after CCl4 treatment. These findings reveal a possible causal relationship between genetic damage and inflammatory responses in the liver, where the activation of p53 by DDR leads to apoptotic cell death, which in turn triggers inflammatory signals in the surrounding microenvironment, which promote the transformation and clonal expansion of genetically mutated cells in the liver.
Hanahan D, Weinberg RA . Hallmarks of cancer: the next generation. Cell 2011; 144: 646–674.
Thorgeirsson SS, Grisham JW . Molecular pathogenesis of human hepatocellular carcinoma. Nat Genet 2002; 31: 339–346.
Kang TW, Yevsa T, Woller N, Hoenicke L, Wuestefeld T, Dauch D et al. Senescence surveillance of pre-malignant hepatocytes limits liver cancer development. Nature 2011; 479: 547–551.
Fausto N . Mouse liver tumorigenesis: models, mechanisms, and relevance to human disease. Semin Liver Dis 1999; 19: 243–252.
Levine AJ, Oren M . The first 30 years of p53: growing ever more complex. Nat Rev Cancer 2009; 9: 749–758.
Livni N, Eid A, Ilan Y, Rivkind A, Rosenmann E, Blendis LM et al. p53 expression in patients with cirrhosis with and without hepatocellular carcinoma. Cancer 1995; 75: 2420–2426.
Sarfraz S, Hamid S, Siddiqui A, Hussain S, Pervez S, Alexander G . Altered expression of cell cycle and apoptotic proteins in chronic hepatitis C virus infection. BMC Microbiol 2008; 8: 133.
Greenblatt MS, Feitelson MA, Zhu M, Bennett WP, Welsh JA, Jones R et al. Integrity of p53 in hepatitis B x antigen-positive and -negative hepatocellular carcinomas. Cancer Res 1997; 57: 426–432.
Kodama T, Takehara T, Hikita H, Shimizu S, Shigekawa M, Tsunematsu H et al. Increases in p53 expression induce CTGF synthesis by mouse and human hepatocytes and result in liver fibrosis in mice. J Clin Invest 2011; 121: 3343–3356.
Krizhanovsky V, Yon M, Dickins RA, Hearn S, Simon J, Miething C et al. Senescence of activated stellate cells limits liver fibrosis. Cell 2008; 134: 657–667.
Abbott A . Laboratory animals: the Renaissance rat. Nature 2004; 428: 464–466.
Tong C, Li P, Wu NL, Yan Y, Ying QL . Production of p53 gene knockout rats by homologous recombination in embryonic stem cells. Nature 2010; 467: 211–213.
Rajewsky MF, Dauber W, Frankenberg H . Liver carcinogenesis by diethylnitrosamine in the rat. Science 1966; 152: 83–85.
Verna L, Whysner J, Williams GM . N-nitrosodiethylamine mechanistic data and risk assessment: bioactivation, DNA-adduct formation, mutagenicity, and tumor initiation. Pharmacol Ther 1996; 71: 57–81.
Polyak K, Xia Y, Zweier JL, Kinzler KW, Vogelstein B . A model for p53-induced apoptosis. Nature 1997; 389: 300–305.
Farazi PA, DePinho RA . Hepatocellular carcinoma pathogenesis: from genes to environment. Nat Rev Cancer 2006; 6: 674–687.
Newell P, Villanueva A, Friedman SL, Koike K, Llovet JM . Experimental models of hepatocellular carcinoma. J Hepatol 2008; 48: 858–879.
Kemp CJ . Hepatocarcinogenesis in p53-deficient mice. Mol Carcinog 1995; 12: 132–136.
Kuraishy A, Karin M, Grivennikov SI . Tumor promotion via injury- and death-induced inflammation. Immunity 2011; 35: 467–477.
Kemp CJ, Donehower LA, Bradley A, Balmain A . Reduction of p53 gene dosage does not increase initiation or promotion but enhances malignant progression of chemically induced skin tumors. Cell 1993; 74: 813–822.
van Gijssel HE, Maassen CB, Mulder GJ, Meerman JH . p53 protein expression by hepatocarcinogens in the rat liver and its potential role in mitoinhibition of normal hepatocytes as a mechanism of hepatic tumour promotion. Carcinogenesis 1997; 18: 1027–1033.
We thank Jiaohong Wang, Shoudong Ye and Zhengmao Lu for technical assistance. This work was supported by a NIH grant to QLY (R01OD010926).
The authors declare no conflict of interest.
Supplementary Information accompanies the paper on the Oncogene website
About this article
Combinatorial treatment of Rhizoma Paridis saponins and sorafenib overcomes the intolerance of sorafenib
The Journal of Steroid Biochemistry and Molecular Biology (2018)
Diethylnitrosamine Increases Proliferation in Early Stages of Hepatic Carcinogenesis in Insulin-Treated Type 1 Diabetic Mice
BioMed Research International (2018)
International Journal of Molecular Sciences (2018)
The matricellular protein CCN1 suppresses hepatocarcinogenesis by inhibiting compensatory proliferation
Low-dose ionizing irradiation triggers a 53BP1 response to DNA double strand breaks in mouse spermatogonial stem cells
Systems Biology in Reproductive Medicine (2016)