Telomeres, p53, Hepatocyte Nuclear Factor 4α, and Liver Disease

Dyskeratosis congenita is an inherited disorder characterized by defects in the ability of cells to maintain telomeres. Patients commonly present with pediatric bone marrow failure and liver and/or pulmonary fibrosis in the fourth to fifth decade of life. The prevalence of dyskeratosis congenita is estimated as 1 in 1 million; however, due to the incomplete penetrance of disease, the true prevalence is unknown.⁽¹⁾ Defects in at least 10 distinct telomere-associated genes lead to the development of premature stem cell and tissue failure. Patients with dyskeratosis congenita also have a predisposition to develop cancer; approximately 5% develop hepatocellular carcinoma.⁽²⁾ Interestingly, in the general population, hepatocellular carcinoma often arises in the setting of cirrhosis and shortened hepatic telomeres and is associated with reactivation of hepatic telomerase in >90% of patients.⁽³⁾ In Munroe et al.’s article featured in this issue of Hepatology, human embryonic stem cells are used as an in vitro model system to study the effects of telomere shortening in dyskeratosis congenita, and the authors present novel findings that further our understanding of telomeres in liver disease.⁽⁴⁾

Our understanding of telomeres dates back to Hayflick and Moorhead, who cultured human fibroblasts and discovered that cells did not grow indefinitely in culture but stopped dividing after 60-80 doublings.⁽⁵⁾ This number of cell divisions before cells undergo senescence later became known as the “Hayflick limit.” In 1971, Olovnikov discovered that, due to the mechanics of DNA replication, DNA polymerase did not replicate the entire lagging strand and that chromosomal DNA progressively shortened with each cell division.⁽⁶⁾ Because genetic information was potentially lost from the ends of each chromosome with each DNA replication, he termed this the “end-replication problem.” Following intensive investigation, Blackburn, Grieder, and Szostak solved the end-replication problem with their discovery of telomeres, for which they were awarded the 2009 Nobel Prize in Physiology or Medicine.⁽⁷⁾ Telomeres consist of tandem TTAGGG repeats that maintain the integrity of chromosomal ends. They serve several crucial functions, including preventing loss of genetic information, abnormal fusion between chromosomes, and activation of ataxia telangiectasia mutated and ataxia telangiectasia and Rad3-related DNA damage response pathways that occur in response to free chromosomal ends.⁽⁸⁾

At birth, telomeres range from 8 to 12 kb; and with each cell division, telomeric DNA is lost.⁽⁸⁾ Once telomere length is below a certain threshold, cells activate DNA damage response pathways, leading to p53/p21 activation, and undergo senescence or apoptosis.^{(3, 8)} In most cells, telomere length declines with age. For example, in healthy human livers, one study found that hepatic telomeres decrease from a mean of 9.7 kb at age 20 to 8.1 kb by age 60.⁽⁹⁾ However, certain cell types, such as fetal tissue, stem cells, bone marrow, and malignant tumors, require constant proliferation and can restore and maintain telomere length over many divisions through the expression of telomerase.

Telomerase is a ribonucleoprotein complex that extends telomeres in a templated fashion. It is composed of telomerase reverse transcriptase (TERT), an RNA-dependent DNA polymerase; telomerase RNA component (TERC), an RNA template; and dyskerin, a protein that stabilizes the telomerase complex. TERT is the key subunit by which telomerase is regulated, and it is suppressed in most cells, including mature hepatocytes in adults.⁽⁸⁾ The most common mutations in dyskeratosis congenita are missense mutations in dyskerin.⁽¹⁾ Mutations in this gene destabilize the telomerase complex, preventing the maintenance of telomeres and leading to shortened telomeres over time.

In this issue of Hepatology, Munroe et al. used clustered regularly interspaced short palindromic repeats (CRISPR)/CRISPR-associated 9 to introduce a dyskerin A353V mutation in human embryonic stem cells and assess its impact on hepatic differentiation. This mutation is common in patients with dyskeratosis congenita.⁽¹⁾ Although telomere disease has been studied in mice, this approach has been limited because telomerase-deficient mice fail to fully recapitulate human disease. This study is the first to examine the impact of telomere disease on hepatocytes using human embryonic stem cells as an in vitro model. The authors found that cells with dyskerin A353V developed shortened telomeres after repeated passaging and, as a result, showed p53 stabilization. Surprisingly, however, p53 stabilization did not lead to senescence and apoptosis. Instead, the expression of p53 prevented hepatic differentiation and facilitated ongoing proliferation. This contrasts with the known role for p53 as a tumor suppressor in response to DNA damage and indicates that p53 may be acting on these cells through some other mechanisms. The authors then show that impairment of differentiation first occurs in the hepatic endoderm stage and that cells with shortened telomeres failed to express hepatocyte nuclear factor 4α (HNF4α). This novel finding is the first time HNF4α has been linked to telomere disease. The authors further demonstrate that by overexpressing TERC (to counteract impaired stabilization of the telomerase complex by dyskerin A353V), knocking out p53, or conditionally expressing HNF4α, they could partially rescue cell differentiation. They conclude that p53 stabilization in response to shortened telomeres inhibits expression of HNF4α, prevents differentiation of human embryonic stem cells into hepatocyte-like cells, and allows for ongoing proliferation. Their findings represent a novel and unexpected result that advances our understanding of telomeres in liver disease.

Telomere length is typically shorter in liver disease than in healthy liver. Chronic injury leads to cell death, and in response, hepatocytes divide and proliferate to replace injured tissue; with successive injury, hepatic telomeres may become progressively shortened.^{(3, 8)} Studies have shown that the rate of telomere shortening in hepatocytes may be associated with the rate of fibrosis progression.⁽⁸⁾ In cirrhosis, it may be associated with an increased risk for hepatocellular carcinoma.^{(3, 8)} The authors’ results demonstrate that p53 stabilization in response to shortened telomeres leads to decreased HNF4α expression, impaired differentiation, and ongoing proliferation. It would be interesting to examine whether this novel mechanism contributes to the dedifferentiation and malignant transformation of hepatocytes in cirrhosis and dyskeratosis congenita–related liver disease. In line with the authors’ findings, hepatocellular carcinoma may be associated with overexpression of p53, decreased HNF4α levels, and ongoing proliferation.^{(3, 10)} The authors go on to postulate that the decreased expression of HNF4α may be due to the binding of p53 to the HNF4α P1 promoter.⁽¹¹⁾ It would be interesting to determine whether other p53 regulatory mechanisms are also involved. Although their findings represent an interesting new mechanism in vitro, future in vivo studies are warranted to further examine the relevance of these findings in cirrhosis and hepatocellular carcinoma.

Munroe et al. provide insight into possible mechanisms of liver disease in dyskeratosis congenita. Most significantly, by using the differentiation of human embryonic stem cells into hepatocytes as an in vitro model to examine the connection of telomeres to liver disease, the authors demonstrate the novel and important finding that shortened telomeres led to p53 stabilization, decreased expression of HNF4α, impaired differentiation, and ongoing proliferation. Given the limitations of rodent models in studying telomere disease, human embryonic stem cells represent an important complementary model. These findings provide justification for future investigations examining the role of telomere function and signaling in human liver disease, with potentially important therapeutic implications.