A dual role of p21 in liver regeneration and hepatocarcinogenesis


  • Potential conflict of interest: Nothing to report.

Willenbring H, Sharma AD, Vogel A, Lee AY, Rothfuss A, Wang Z, Finegold M, Grompe M. Loss of p21 permits carcinogenesis from chronically damaged liver and kidney epithelial cells despite unchecked apoptosis. Cancer Cell 2008;14:59-67 (Reprinted with permission).


Accumulation of toxic metabolites in hereditary tyrosinemia type I (HT1) patients leads to chronic DNA damage and the highest risk for hepatocellular carcinomas (HCCs) of any human disease. Here we show that hepatocytes of HT1 mice exhibit a profound cell-cycle arrest that, despite concomitant apoptosis resistance, causes mortality from impaired liver regeneration. However, additional loss of p21 in HT1 mice restores the proliferative capabilities of hepatocytes and renal proximal tubular cells. This growth response compensates cell loss due to uninhibited apoptosis and enables animal survival but rapidly leads to HCCs, renal cysts, and renal carcinomas. Thus, p21's antiproliferative function is indispensable for the suppression of carcinogenesis from chronically injured liver and renal epithelial cells and cannot be compensated by apoptosis.


Hepatocellular carcinomas (HCCs) represent one of the most common cancers, with growing incidence in several regions around the world. It is well accepted that the major cause for HCC development is chronic liver disease. Although the exact molecular mechanism leading to HCC progression is still not well understood, it is obvious that chronic liver damage is associated with continuous cycles of apoptosis and/or necrosis, followed by compensatory proliferation of hepatocytes. This triggers liver fibrosis and cirrhosis but also results in the accumulation of mutations and transformed hepatocytes due to excessive DNA replication and increased mitosis during liver regeneration.

Chronic liver disease leading to HCC development typically arises from chronic hepatitis B or C virus infection or steatohepatitis, but may also be caused by liver-related genetic disorders such as hereditary tyrosinemia type 1 (HT1). HT1 is an autosomal-recessive disease caused by genetic inactivation of the enzyme fumarylacetoacetate hydrolase (FAH), eventually leading to the accumulation of the highly mutagenic metabolite fumarylacetoacetatein (FAA) in the liver.1 Patients with HT1 suffer from chronic DNA damage in the liver and show an extremely high susceptibility for HCC. The only efficient therapeutic option so far is treatment with the drug nitrisinone (2-[2-nitro-4-(trifluoromethyl)benzoyl] cyclohexane-1,3-dione (NTBC), which blocks the pathway upstream of the formation of FAA. For further analysis of the molecular mechanisms leading to HCC, a murine model of HT has been established earlier by deleting the Fah gene in vivo. Fah−/− mice die shortly after birth from liver failure but can be rescued by earliest application of NTBC.2 In a more recent study it was shown that Fah−/− mice are resistant to apoptotic cell death.3 Although it is well accepted that apoptosis resistance may contribute to hepatocarcinogenesis,4 these results were unexpected in a model of chronic liver disease, which is usually linked to hepatocyte cell death.

In the July issue of Cancer Cell, Willenbring et al. explain in an elegant study the mechanisms underlying liver failure and hepatocarcinogenesis in HT1. It turns out that Fah inactivation does not only confer apoptosis resistance, but also leads to complete cell cycle arrest and failure of liver regeneration. As a single key player for this phenomenons, the tumor suppressor gene p21 was identified, which is strongly up-regulated in Fah−/− mice as a consequence of accumulated DNA damage. Indeed, recent studies have indicated that p21 mediates not only p53-dependent cell cycle arrest, but can also counteract apoptosis in a p53-independent manner.5 Although the mechanism leading to apoptosis resistance through p21 is not fully understood, Willenbring et al. provide some hints that significant down-regulation of the proapoptotic Bcl-2–like protein Noxa through p21 may predominantly block apoptosis in Fah−/− mice.

The p53-dependent cell cycle arrest and apoptosis following DNA damage allows elimination or repair of transformed cells before regeneration starts, and it is believed that thereby apoptosis protects from uncontrolled growth of transformed cells. Concomitant inactivation of Fah and p21, however, restored both the apoptosis capacity of hepatocytes and allowed efficient liver regeneration, but resulted in extraordinarily fast progression to HCC within 8 weeks. This exceeding hepatocyte proliferation was a unique consequence of p21 depletion, because other cell cycle inhibitors such as p53, p15, p16, and p19 were still active but could not compensate for p21 loss. Based on the fact that the cell cycle regulatory machinery in mammals—especially concerning positive regulators—is highly redundant, with only one cyclin-dependent kinase (Cdk16) and only two cyclins (cyclin A2, B17) being absolutely essential, it is extremely surprising but very interesting that the negative regulation of the cell cycle that protects from HCC development relies exclusively on p21.

The promising idea behind the inactivation of p21 in Fah−/− mice was to reactivate apoptosis in order to first eliminate DNA-damaged hepatocytes followed by balanced liver regeneration, thus overcoming the consequences of HT1. However, the study of Willenbring et al. clearly demonstrates that the antiproliferative function of p21 is essential for preventing uncontrolled hepatocyte growth, whereas apoptosis alone is not sufficient to protect from liver carcinogenesis in HT1.

Thus the dilemma of HT1 disease progression is now explained (Fig. 1) but still unsolved: apoptosis resistance leading to accumulation of transformed hepatocytes as well as lack of cell cycle progression and liver regeneration with the consequence of liver failure are both caused exclusively by inappropriate high levels of p21. Simply antagonizing p21, however, is not helpful; rather, it turns out to be carcinogenic in chronic liver diseases. Therefore, p21 alone does not seem to be a promising therapeutic target, especially as the antiproliferative effect of p21 is dose-dependent and fine regulation of p21 expression levels might be problematic. However, the study of Willenbring et al. re-evaluates the important functions of p21 and highlights the high potential of p21 as a diagnostic marker in chronic liver disease.

Figure 1.

Role of p21 in hereditary tyrosinemia type 1 disease. The accumulation of FAA eventually results in accumulation of DNA damage in hepatocytes, leading to activation of p53 and further up-regulation of p21. The cell cycle inhibitor p21 in turn is responsible for strict cell cycle arrest in HT1, leading to failure of liver regeneration but also strong protection from hepatocarcinogenesis. In parallel, p21 also inhibits apoptosis, presumably via down-regulation of Noxa, which may result in accumulation of cancer-prone hepatocytes. The reason why p53 cannot induce apoptosis in a p21-independent manner in HT1 (grey arrow) needs further investigation.