See Article on Page 1600
The pathogenesis of liver cirrhosis is a complex process for which the mechanisms are not completely elucidated. Although exogenous or environmental risk factors leading to chronic liver injury, such as viral hepatitis B and C infection, alcohol intake, or fatty liver disease, can be identified in most patients with cirrhosis, approximately 5% of patients with cirrhosis have no apparent risk factors.1 In addition, patients with identical risk factors have a diverse spectrum of clinical manifestations. The reasons why some patients with identifiable risk factors progress to cirrhosis whereas others have a benign course remain unclear and cannot be completely explained by known environmental and/or obvious host factors (i.e., age and sex). Therefore, a genetic predisposition may contribute to the development of cirrhosis.
Evidence supporting the role of genetic factors as a risk for cirrhosis has been accumulating during the last decade. Data from epidemiologic studies reported the prevalence of cryptogenic cirrhosis as 3.1-fold higher in Hispanic Americans, but 3.9-fold lower in African Americans than in European Americans despite the same prevalence of diabetes in Hispanics and African Americans.2 Studies of the National Academy of Sciences–National Research Council Twin Registry reported that concordance rates for developing alcoholic cirrhosis were significantly higher in monozygotic twins than in dizygotic twins (16.9% versus 5.3%, P < 0.001).3 Recently, Huang et al. proposed a cirrhosis risk score based on a genetic marker panel (seven single-nucleotide polymorphisms [SNPs] in six genes: AP3S2 [adaptor-related protein complex 3, sigma 2 subunit], AQP2 [aquaporin 2], AZIN1 [antizyme inhibitor 1], STXBP5L [syntaxin binding protein 5-like], TLR4 [Toll-like receptor 4], and TRPM5 [transient receptor potential cation channel, subfamily M, member 5, and in the intergenic region between DEGS1 [degenerative spermatocyte homolog 1], NVL [nuclear valosin-containing protein-like]) for identifying the risk of developing cirrhosis in Caucasian patients with chronic hepatitis C infection.4 This score had a higher area under the receiver operating characteristics curve compared to clinical factors (age, sex, alcohol) (0.73 versus 0.53) and has been validated in another group of Caucasian patients with mild chronic hepatitis C infection (METAVIR stage F0-F2 at initial liver biopsy).5 Furthermore, the association of an SNP in the PNPLA3 (patatin-like phospholipase domain-containing protein 3) gene (rs738409) with fibrosis in patients with nonalcoholic steatohepatitis and with alcoholic cirrhosis has also been reported.6, 7 These data suggest that genetic risk factors influence the progression to cirrhosis. The functional bases for the predispositions due to these SNPs have not been completely characterized.
In addition to the results from SNP studies, the concept of telomere shortening as a genetic risk factor for cirrhosis has been proposed. Telomeres consist of repeat DNA sequences (TTAGGG) and a specialized protein complex named the telosome or shelterin. They are located at the ends of linear chromosomes, the so-called “tips of the chromosomes”, and function as a “cap” to protect the chromosome from end-to-end fusion and destruction by nuclease and/or ligase enzymes.8 Telomerase is an enzymatic protein complex, comprising two essential components: telomerase reverse transcriptase (hTERT) and a telomerase RNA component (hTERC). This enzymatic complex is responsible for maintaining telomere length by synthesizing new DNA sequences and adding them to the end of the chromosome. Nevertheless, during each cell division, telomere length inevitably reduces due to the inability of DNA polymerase to fully replicate the terminal chromosomal segment. Once telomeres are critically short, a DNA-damage program is activated that leads either to apoptosis or cell senescence. Because continuous DNA turnover accelerates telomere shortening, this process is accentuated in conditions with high cell turnover such as chronic liver injury. The resulting cellular growth arrest and/or senescence appears to be profibrogenic by as-yet undefined mechanisms.
Kitada et al. were the first to demonstrate the relationship between telomere shortening and cirrhosis in 1995.9 Telomere length in tissue from cirrhotic liver was shorter than in liver with chronic hepatitis and both were shorter than telomere lengths in normal liver tissue. Subsequent studies confirmed that a shortened telomere length was correlated with the degree of fibrosis, suggesting that telomere shortening may be an important cause or marker of cirrhosis.10-12
In 2000, Rudolph et al. tested this hypothesis in telomerase knockout murine models. Mice with shortened telomeres had less capacity than did wild-type mice for liver regeneration after partial hepatectomy. Mice with dysfunctional telomeres also displayed accelerated development of cirrhosis after liver injury. Restoration of telomerase by the delivery of the telomerase RNA gene resulted in reduced fibrosis and improved liver function.13
In this issue of HEPATOLOGY, Calado et al.14 and Hartmann et al.15 both report on the association between telomerase TERT and TERC gene mutations and cirrhosis in patient populations with various etiologies including hepatitis C virus (HCV)-induced cirrhosis (37% and 42%), alcohol-induced cirrhosis (25% and 13%), mixed HCV- and alcohol-induced cirrhosis (8% and 12%), hepatitis B virus–induced cirrhosis (3% and 16%), and others (27% and 17%).14, 15 Telomere length and telomerase activity were also investigated in these reports.
Calado et al. studied gene mutations in DNA from buccal mucosa tissue or peripheral blood in patients with cirrhosis and controls. They found missense mutations in the TERT and TERC genes in nine patients and one patient, respectively, of 134 patients with cirrhosis. The most frequent variant was in exon 15 of the TERT gene at codon Ala1062Thr (found in six patients with cirrhosis). Telomere length in peripheral blood cells of patients with cirrhosis was significantly shorter than in controls. Telomerase activity in vitro was shown to be reduced in most TERT variants.
Similarly, Hartmann et al. studied gene mutations in DNA from peripheral blood cells of patients with cirrhosis and controls. They report a significant increase in the frequency of TERT and TERC gene mutations in patients with cirrhosis (16 of 521 patients) compared to controls. Patients with TERT mutations had shorter telomeres in peripheral white blood cells and a significant reduction in telomerase activity in skin fibroblasts and lymphocytes.
Taken together, these results indicate that telomerase mutations result in a decrease in telomerase activity. This accelerates telomere shortening, leading to impaired hepatic regeneration and more rapid progression to fibrosis. Patients with telomerase mutations are at risk for developing cirrhosis, and telomere shortening may be one of the key steps in cirrhosis formation (Fig. 1).
What are the potential clinical implications of these findings? Cirrhosis is a major precursor phenotype to the development of hepatocellular carcinoma (HCC), and telomerase activity is typically reactivated during liver carcinogenesis. Are patients with these TERT and TERC mutations more or less likely to develop HCC after developing cirrhosis? The prevalence of these TERT and TERC mutations is relatively low, representing 7.5% of patients in the Calado et al. study and 3.1% of patients in the Hartmann et al. study. Although their prevalence is low and they, therefore, may not be a major contributing factor to cirrhosis at the population level, the identification of these mutations raises important questions about our clinical approach to patients with cirrhosis and our conceptual view of risk of cancer. For example, are there predisposing mutations for cirrhosis in other genes involved in the maintenance of telomere function, such as the genes for the other telosome components, including POT1 (protection of telomeres 1 homolog), ACD/TPP1 (adrenocortical dysplasia homolog), TINF2/TIN2 (TERF1-interacting nuclear factor 2), TERF1/TRF1 (telomeric repeat binding factor [NDMA-interacting]1), TERF2/TRF2, and TERF2IP/RAP1 (telomeric repeat binging factor 2, interacting protein), and interacting proteins such as DKC1, NOLA1, NOLA2, and NOLA3? Should assays of telomerase gene mutations be used as a stratification factor for selecting patients for treatment of their liver disease, given the presumption that they will be more likely to develop progressive fibrosis? Or, should these assays be used for stratifying patients in clinical trials of antifibrotic agents to reduce unrecognized bias? It has been recognized for a number of years that there is a familial predisposition to HCC; could this be related to germline transmission of telomerase gene mutations? There is also the clinical observation that a subgroup of patients with cirrhosis will develop HCC relatively early in the natural history of cirrhosis, when they still have Child-Pugh class A liver dysfunction, whereas others will develop HCC at more advanced stages of liver dysfunction. Intriguingly, many individuals progress through the natural history to advanced end-stage liver disease without developing HCC; therefore, are they in some way protected from or less susceptible to carcinogenesis? The findings of the studies by Calado et al. and Hartmann et al. are important because they provide a new perspective on these questions and raise further questions that should be elucidated through future research.