telomerase gene mutations are associated with cirrhosis formation

Telomerase Gene Mutations Are Associatedwith Cirrhosis Formation

Daniel Hartmann,1,2* Ujala Srivastava,1,3* Michaela Thaler,1,4 Karin N. Kleinhans,1 Gisele N’Kontchou,5

Annika Scheffold,1 Kerstin Bauer,1 Ramona F. Kratzer,1 Natalia Kloos,1 Sarah-Fee Katz,1 Zhangfa Song,1

Yvonne Begus-Nahrmann,1 Alexander Kleger,1,4 Guido von Figura,1,4 Pavel Strnad,4 Andre Lechel,1

Cagatay Gunes,1 Andrej Potthoff,6 Katja Deterding,6 Heiner Wedemeyer,6 Zhenyu Ju,1,7 Ge Song,7

Feng Xiao,8 Sonja Gillen,2 Hubert Schrezenmeier,9 Thomas Mertens,10 Marianne Ziol,5 Helmut Friess,2

Michael Jarek,11 Michael P. Manns,6 Michel Beaugrand,5 and K. Lenhard Rudolph1

Telomere shortening impairs liver regeneration in mice and is associated with cirrhosis forma-tion in humans with chronic liver disease. In humans, telomerase mutations have been associ-ated with familial diseases leading to bone marrow failure or lung fibrosis. It is currentlyunknown whether telomerase mutations associate with cirrhosis induced by chronic liverdisease. The telomerase RNA component (TERC) and the telomerase reverse transcriptase(TERT) were sequenced in 1,121 individuals (521 patients with cirrhosis induced by chronicliver disease and 600 noncirrhosis controls). Telomere length was analyzed in patients carryingtelomerase gene mutations. Functional defects of telomerase gene mutations were investigatedin primary human fibroblasts and patient-derived lymphocytes. An increased incidence of telo-merase mutations was detected in cirrhosis patients (allele frequency 0.017) compared to non-cirrhosis controls (0.003, P value 0.0007; relative risk [RR] 1.859; 95% confidence interval[CI] 1.552-2.227). Cirrhosis patients with TERT mutations showed shortened telomeres inwhite blood cells compared to control patients. Cirrhosis-associated telomerase mutations ledto reduced telomerase activity and defects in maintaining telomere length and the replicativepotential of primary cells in culture. Conclusion: This study provides the first experimental evi-dence that telomerase gene mutations are present in patients developing cirrhosis as a conse-quence of chronic liver disease. These data support the concept that telomere shortening canrepresent a causal factor impairing liver regeneration and accelerating cirrhosis formation inresponse to chronic liver disease. (HEPATOLOGY 2011;53:1608-1617)

See Editorial on 430 Cirrhosis formation is one of the main compli-cations at the endstage of chronic liver disease,such as chronic hepatitis B and C virus (HBV,

Abbreviations: DKC, dyskeratosis congenita; HBV, hepatitis B virus; HCC, hepatocellular carcinoma; HCV, hepatitis C virus; PCR, polymerase chain reaction;TERC, telomerase RNA component; TERT, telomerase reverse transcriptase; TRAP, telomere repeat amplification protocol; TRF telomere restriction fragment.From the 1Institute of Molecular Medicine andMax-Planck-Research Department on Stem Cell Aging, University of Ulm, Ulm, Germany; 2Department of Surgery, Technical

University Munich, Munich, Germany; 3Jackson Laboratory, Bar Harbor, Maine, USA; 4Department of Internal Medicine I, University of Ulm, Ulm, Germany; 5Hopital JeanVerdier, L’Assistance publique - Hopitaux de Paris, Bondy Cedex, France; 6Department of Gastroenterology, Hepatology and Endocrinology, HannoverMedical School, Hannover,Germany; 7Institute of Laboratory Animal Sciences, Chinese Academy of Medical Sciences, Beijing, China; 8Capital Pediatrics Institute, Beijing, China; 9Institute for ClinicalTransfusion Medicine and Immunogenetics, DRK Blood Donor Service Baden-Wurttemberg-Hesse and University of Ulm, Ulm, Germany; 10Institute of Virology, UniversityHospital Ulm, Ulm, Germany; and 11Department of Genome Analysis, Helmholtz Centre for Infection Research, Braunschweig, Germany.Received October 4, 2010; accepted January 13, 2011.Author Contributions: G.N., A.P., K.D., H.W., Z.J., G.S., F.X.,. S.G., H.S., M.Z., H.F., M.P.M., and M.B. recruited and managed patients in the study and

were involved in data collection. D.H., U.S., S.F.K., Z.S., Y.B., A.K., G.F., P.S., and A.L. participated in sample preparation and data interpretation. D.H.,U.S.; M.T., R.F.K., and Z.J. analyzed the sequencing data under the supervision of C.G., M.J., and K.L.R. D.H., U.S., K.N.K., A.S., K.B., R.F.K., N.K., C.G.,and T.M. were involved in functional experiments. K.L.R. designed the experiments and wrote the first draft of the report. All authors were given the opportunityto comment on the drafts of the report and approved the final version before submission.Funding/Support: D.H. is funded by the Else-Kroner-Fresenius Stiftung (Memorial Scholarship of Else Kroner-Fresenius Foundation). U.S. is supported by the

PhD program ‘‘Molecular Medicine’’ of the Hannover Biomedical Research School. K.L.R. is funded by the DFG (Ru745-12) and the European Union(Telomarker, GENINCA 202230). The funding organizations had no role in the design or conduct of the study, collection, management, analysis, or interpretationof the data, or preparation, review, or approval of the article*These authors contributed equally.

1608

HCV) infection, chronic alcoholism, and chronic cho-lestatic disease.1-3 Cirrhosis leads to progressive liverfailure and portal hypertension. In addition, cirrhosisis the main risk factor for the development of livercancer.4 Decompensation of liver function at endstagecirrhosis is associated with a sharp increase in mortal-ity5 and liver transplantation represents the only possi-ble treatment at this stage. An understanding of themolecular causes of cirrhosis formation is essential todevelop new therapeutic strategies for the treatment orprevention of cirrhosis. In addition, this would help topredict individual prognosis and to optimize the tim-ing of final intervention strategies, such as liver trans-plantation. The identification of genetic risk factorsassociated with cirrhosis formation represents one pos-sible way to achieve these goals.6,7

Two models of cirrhosis formation have developed.One hypothesis indicates that cirrhosis develops as aconsequence of a progressive deposition of collagenand scar tissue induced by chronic inflammation andnecrosis. Another, not mutually exclusive, hypothesisindicates that telomere shortening represents an under-lying cause of cirrhosis.8 Telomeres form the ends ofhuman chromosomes. The main function of telomeresis the maintenance of chromosomal stability. However,telomeres shorten as a consequence of cell division dueto the ‘‘end replication problem’’ of DNA polymerase,processing of telomeres during S-phase of cell cycle, andthe absence of telomerase expression in most somatictissues.9 Telomere shortening limits the proliferative lifespan of human cells to 50-70 cell doublings by induc-tion of a permanent cell cycle arrest (replicative senes-cence) in response to telomere dysfunction.10,11 Previ-ous studies have shown that telomere shortening alsolimits the life span of primary human hepatocytes.12

Studies of human cirrhosis have demonstrated thattelomere shortening is a general marker of cirrhosisformation correlating with an accumulation of senes-cent hepatocytes.13,14 In addition, studies on telomer-ase-deficient mice have provided the first experimentalevidence that telomere shortening limits the regenerativeresponse to acute and chronic liver injury, acceleratingthe formation of liver fibrosis and steatosis.15,16 To-gether, these studies have led to the telomere model ofcirrhosis formation, indicating that chronic liver diseases

increase the rate of cell turnover, thus leading to acceler-ated telomere shortening and regenerative exhaus-tion.8,17 In agreement with this hypothesis, it has beenrecognized that proliferative activity declines after longlatencies of chronic liver disease and this decline wasassociated with the progressive formation of disease.18

Genetic studies have proven that mutations in telo-merase are the underlying cause of accelerated telomereshortening and organ failure in some rare human dis-eases, including some forms of dyskeratosis congenita(DKC)19 and aplastic anemia.20 In addition, 10% ofthe cases of familial idiopathic lung fibrosis are associ-ated with telomerase mutations.21,22 In most of thesecases heterozygous mutations were found in either theRNA (TERC) or protein component (TERT) of telo-merase. Interestingly, familial cases of idiopathic lungfibrosis and bone marrow failure also showed anincreased frequency of unexplained liver pathologies,including fibrosis, inflammation, macrovesicular steato-sis, and hepatic nodular regeneration.23-25 Some ofthese patients carried mutations in telomerase genes.25

However, these cases are extremely rare and it remainsan open question whether telomerase mutations occurat increased frequency in patients who develop cirrho-sis as a consequence of chronic liver disease. It is possi-ble that telomerase mutations would impair the regen-erative reserves of hepatocytes in the context ofchronic liver damage. Accordingly, an increased fre-quency of telomerase mutations could be associatedwith cirrhosis induced by chronic liver diseases.Here, we sequenced the TERT and TERC genes in

a cohort of 1,121 individuals, 521 patients with livercirrhosis and 600 controls. The analysis revealed a sig-nificantly increased frequency of telomerase mutationsin cirrhosis patients (14 heterozygous, two homozy-gous allelic variants in 521 individuals; allele frequency0.017) compared to controls (three heterozygoussequencing variants in 600 individuals; 0.003, P value0.0007; Relative risk [RR] 1.859; 95% confidenceinterval [CI] 1.552-2.227). Cirrhosis-associated telo-merase mutations showed functional defects and wereassociated with the evolution of disease complications.Together, these data provide the first demonstration ofa broad involvement of telomerase mutations in theevolution and progression of cirrhosis in response to

Address reprint requests to: K. Lenhard Rudolph, Institute of Molecular Medicine and Max-Planck-Research Group on Stem Cell Aging, University of Ulm,Albert-Einstein-Allee 11, 89081 Ulm, Germany. E-mail: [email protected]; fax: 49-731-5036-1102.CopyrightVC 2011 by the American Association for the Study of Liver Diseases.View this article online at wileyonlinelibrary.com.DOI 10.1002/hep.24217Potential conflict of interest: Nothing to report.Additional Supporting Information may be found in the online version of this article.

HEPATOLOGY, Vol. 53, No. 5, 2011 HARTMANN, SRIVASTAVA, ET AL. 1609

chronic liver injury. The finding could impact on thefuture development of molecular therapies and surveil-lance programs in patient with chronic liver disease.

Patients and Methods

Patients. A total of 1,121 individuals were recruitedfor the current study. Among them, 521 patients werediagnosed with liver cirrhosis; 600 controls were eitherhealthy individuals (n ¼ 473) or patients with chronicHCV infection who did not develop cirrhosis during fol-low-up (average time of follow-up: 21 years, n¼ 127). Weincluded the group of hepatitis C carriers who did not pro-gress towards liver cirrhosis because this cohort provides animportant control indicating that telomerase mutations doassociate with the development of cirrhosis and not withthe occurrence of chronic liver disease per se.Subjects were recruited from (1) the Liver Unit,

Hopital Jean Verdier in Bondy Cedex, France, (2) theDepartment of Gastroenterology, Hepatology and Endo-crinology of Hannover Medical School in Hannover, Ger-many, (3) the Henriettenstiftung Hannover, Germany, (4)the Institute for Clinical Transfusion Medicine and Immu-nogenetics, DRK Blood Donor Service Baden-Wurttem-berg-Hesse, University of Ulm, and (5) the Peking UnionMedical College Hospital, Chinese Academy of MedicalSciences. The study was approved by the local InstitutionalReview Boards and written informed consent was obtainedfrom all subjects. The study was designed in accordancewith the principles of the Declaration of Helsinki andpatient data were evaluated anonymously.

Polymerase Chain Reaction (PCR) and Sequen-cing. We manually amplified and sequenced hTERTand hTERC from genomic DNA prepared from pe-ripheral blood cells and liver samples using PCRamplification.Telomere Restriction Fragment (TRF) Length

Analysis. Telomere length was determined by southernblotting and quantitative PCR. Southern blot was car-ried out as described.13

Cell Cultures. TERT point mutations (hTERTp.P65A, p.P380S, p.G1109R, Supporting Table 2)were generated in the pMSCV retroviral vector con-taining the wildtype hTERT complementary DNA(cDNA) using the QuickChange site-directed muta-genesis Kit (Stratagene). The wildtype and dominant-negative hTERT cDNAs were kindly provided byRobert Weinberg (Whitehead Institute, MIT). HumanBJ fibroblasts (American Tissue Culture Collection,ATCC) were infected with the indicated retroviralpMSCV neo constructs and cultured in Dulbecco’sminimal essential medium (Gibco) supplemented with10% fetal bovine serum (Sigma) and 1% penicillin-streptomycin in 5% CO2 / 20% O2 at 37�C. Lym-phocytes were isolated and immortalized from freshEDTA blood as described.26

Analysis of Telomerase Activity. Telomerase extrac-tion and telomere repeat amplification protocol(TRAP) assays were performed using the TRAPezeTelomerase Detection System (Millipore) according tothe manufacturer’s instructions.

Table 1. Distribution of Telomerase Gene Mutations in the Cirrhosis and Control Groups

Gene

Nucleotide

Variants (mRNA)

Amino Acid

Variants (Codon)

No. of Heterozygotes; No. of Homozygotes

(Allele Frequency)v-Square Test

(P Value)Study Group (n5521) Control Group (n5600)

TERC r.156C>A* — 1;0 (0.001) 0;0

r.244C>U* — 1;0 (0.001) 0;0

r.264G>A* — 0;0 1;0 (0.001)

TERT 1 c.37C>A* p.L13M 1;0 (0.001) 0;0

c.40C>A* p.L14M 1;0 (0.001) 0;0

c.193C>G p.P65A 1;0 (0.001) 0;0

TERT2 c.340A>T* p.E113V 1;0 (0.001) 0;0

c.1138C>T* p.P380S 1;0 (0.001) 0;0

c.1336C>A* p.R446S 0;0 2;0 (0.002)

TERT 10 c.2638G>T* p.A880S 1;0 (0.001) 0;0

c.2645C>T* p.T882I 1;0 (0.001) 0;0

TERT 16 c.3325G>A* p.G1109R 4;2 (0.008) 0;0 0.0072

c.3346G>C* p.E1116Q 1;0 (0.001) 0;0

Total 14;2 (0.017) 3;0 (0.003) 0.0007

The table depicts mutations that lead to amino acid changes in the TERT gene or are below 1% in the control cohort of the current study and in previous studies.

Amino acids are indicated by single-letter abbreviations. Significance (P value) was determined by chi-square test. The cumulative frequency of gene variants was

significantly higher in patients with liver cirrhosis (cumulative allele frequency 0.017) than in noncirrhosis controls (0.003, P ¼ 0.0007). Furthermore, the allele fre-

quency of the p.G1109R mutation in the TERT gene is significantly higher in patients (allele frequency 0.008) than in controls (0; P ¼ 0.0072).

*Highlights novel variants that have not been previously reported.

1610 HARTMANN, SRIVASTAVA, ET AL. HEPATOLOGY, May 2011

Statistical Analysis. Statistical analysis was per-formed using Microsoft Excel and GraphPad Prism soft-ware. A chi-square test was used to calculate P values inTable 1. Linear regression analysis was used in Fig. 3Aand unpaired Student’s t test was used in Fig. 3B,D,E,G.In all assays, P values of less than 0.05 or 0.001 were consid-ered statistically significant or highly significant, respectively.

Results

Telomerase Gene Mutations in Cirrhosis Patients.Sequence analysis was carried out on DNA samplesfrom a total of 1,121 individuals, 521 patients withcirrhosis and a control cohort of 600 individuals(Table 1). In the cirrhosis group, chronic HCV infec-tion was the main cause of cirrhosis followed by HBVinfection and chronic alcoholism (Fig. 1). The controlsamples were derived from healthy individuals (n ¼473) or patients with chronic HCV infection who didnot develop cirrhosis during follow-up (n ¼ 127, aver-age time of follow-up: 21 years).

Sequence analysis was carried out on DNA from pe-ripheral blood cells in most of the cases. The first setof DNA samples (n ¼ 176 cirrhosis patients and n ¼54 controls) was completely sequenced (TERT exons1-16 plus TERC; see Supporting Table 1 for primerdesign). Subsequent sequencing analysis (TERC andTERT exons 1 and 16 completely, TERT exons 2, 10,and 15 partially) was focused on mutated regions thatwere detected in the initially sequenced cohort as wellas on telomerase mutation regions that were publishedin previous studies on other human diseases, such asDKC and idiopathic pulmonary fibrosis.21,22,24,27 Thesequencing analysis identified a set of TERT gene muta-tions leading to amino acid changes in the TERT pro-tein as well as two mutations in the TERC (Table 1,Supporting Figs. 1, 2). These mutations have not beenlisted in the single nucleotide polymorphism database(http://www.ncbi.nlm.gov/projexts/SNP). The overall fre-quency of these gene mutations was significantlyincreased in the cirrhosis group (14 heterozygous, twohomozygous variants in 521 individuals; allele frequency

Fig. 1. Distribution of etiologies of liver cirrhosis. The pie charts show the distribution of etiologies of chronic liver disease in the completestudy cohort (A) and in mutation carriers (B).


0.017, Table 1) compared to the control group (threeheterozygous sequencing variants in 600 individuals, al-lele frequency 0.003, P ¼ 0.0007). Reanalysis of the cir-rhosis-associated gene mutations in frozen liver biopsiesof two patients verified that these telomerase germline

mutations were also detectable in liver (data not shown).Subdividing the control cohort into (1) healthy controlswithout chronic liver disease (n ¼ 473) and (2)chronic liver disease patients without progressiontoward cirrhosis (n ¼ 127) revealed that both

Table 2. Clinical Characteristics of Patients Carrying TERC and TERT Mutations

No. Variant Zygosity Age Sex Race Disease

Initial Dx

of Liver

Cirrhosis

Initial

Dx of

HCC Complications

Time to

Decompensation

Time

to LTX

Child-

Pugh

1 TERC C156A heterozygous 38 F Asian HBV n.a. n.a. Ongoing antiviral treatment n.a. n.a. C

2 TERC C244U heterozygous 65 M White EtOH 2005 n.a. Progression of liver impairment

and major portal hypertension

with iterative bleeding despite

complete abstinence (no LTX

d/t cardiac contraindication)

4 years n.a. B

3 L13M/L14M heterozygous 54 F Black HCV 2007 n.a. Ongoing antiviral treatment n.a. n.a. A

4 P65A heterozygous 58 F Black HCV 2003 2004 Rapid HCC development, LTX for

HCC recurrence in 2006

n.a. 3 years A

5 E113V heterozygous 55 M Asian HBV 1995 n.a. Initial Dx of HBV infection in

1985, rapid progression of

liver impairment, hepatic

decompensation in 1995

10 years n.a. C

6 P380S heterozygous 59 M Asian HCV 2006 2006 HCC occurrence and recurrence

after radiofrequency ablation

in 2006, on waiting list for LTX

n.a. 3 years A

7 A880S/T882I heterozygous 39 F Asian HBV n.a. n.a. Ongoing antiviral treatment n.a. n.a. C

8 G1109R homozygous 52 M White HCV, EtOH 2004 n.a. Rapid progression of liver

cirrhosis despite alcohol

withdrawal with major

atrophy of left lobe liver

5 years n.a. C

9 G1109R homozygous 47 M White HCV, EtOH 2002 2009 Rapid progression of liver

cirrhosis, bifocal HCC

development and progression

of liver impairment in 2009,

on waiting list for LTX

7 years 7 years B

10 G1109R heterozygous 61 F White HBV, HCV 1994 n.a. Rapid progression of liver cirrhosis,

initial Dx of HBV and HCV

infection in 1986, ascitic

decompensation in 1994, kidney

TX 1997 for mesangioproliferative

glomerulonephritis, LTX in 1998

8 years 12 years C

11 G1109R heterozygous 85 F White HCV n.d. 2004 Progression of liver impairment

and HCC development, died

in 2007 of HCC recurrence

n.a. n.a. B

12 G1109R heterozygous 75 M White HCV n.d. 1995 Nonresponsive to interferon

monotherapy in 1989, ascites,

encephalopathy and bleeding

of esophageal varices in 1993,

chemoembolization for

multilocular HCC (3 foci),

LTX & revision d/t graft

failure in 1995

12 years 14 years C

13 G1109R heterozygous 50 F White HCV 2004 n.a. Died in 2004 d/t variceal

bleeding

n.a. n.a. B

14 E1116Q heterozygous 43 M Asian HBV 1997 n.a. Rapid progression of liver

impairment, hepatic

decompensation in 2004

(portal hypertension with

iterative bleeding)

7 years n.a. C

The table shows the indicated clinical characteristics of all cirrhosis patients carrying telomerase gene mutations. Most of the patients showed a rapid progres-

sion towards endstage cirrhosis and 5/14 patients developed liver cancer. Note that the overall frequency of endstage disease characterized by advanced cirrhosis

stage (CHILD B or C), HCC occurrence, conduction or evaluation of liver transplantation was 93% (13/14) in cirrhosis patients harboring telomerase mutations

compared to 62% (327/507) in cirrhosis patients without telomerase mutations (P ¼ 0.042).


subgroups exhibited significantly lower allele fre-quencies of telomerase mutations compared to thecirrhosis group (P ¼ 0.0021 and P ¼ 0.0349,respectively). There was no significant difference inallele frequency of telomerase mutations betweenthe two subgroup control cohorts.One of the TERT gene mutations (c.3325G>A

leading to an amino acid change at positionp.G1109R) was found in six out of the 521 cirrhosispatients (four heterozygous mutations, two homozy-gous, allele frequency 0.008, Table 1) but in none ofthe control samples (0; P ¼ 0.0072).The prevalence of telomerase gene mutations was

not associated with a specific ethnicity of the patients(Supporting Fig. 3) or a specific etiology of cirrhosis(Table 2, Fig. 1). Aside from these gene mutations, anumber of single nucleotide polymorphisms and silentnucleotide mutations (not resulting in amino acidchanges) were identified (Supporting Table 3). Thesegene variants were not present at different frequencies inthe cirrhosis group compared to the control group. Oneexample was the previously described c.58G>A variationin the TERC gene, which has previously been describedto be associated with African ethnic origin28 and wasalso associated with African ethnic origin in our study.Localization of Cirrhosis-Associated Telomerase

Mutations. Together, these results indicated that telo-merase gene mutation, but not polymorphic gene var-iants, were associated with the evolution of cirrhosis.The cirrhosis-associated TERC gene mutation(r.156C>A) was located in the pseudoknot domainand the second cirrhosis-associated TERC gene variant(c.244C>T) was located in the paired P5 region ofthe CR4/CR5 domain of the TERC gene, in closeproximity to the recently identified r.323C>T muta-tion that was associated with bone marrow failure(Fig. 2A).29 Three cirrhosis-associated TERT gene muta-tions were located in Exon 1 (c.37C>A, c.40C>A, andc.193C>G) (Fig. 2B, Table 1, Supporting Fig. 1). Previ-ous studies have shown that alterations at the N-terminusof TERT can affect the ability of TERT to maintaintelomere length in cell culture models.30 The cirrhosis-associated c.37C>A and c.40C>A mutations have notpreviously been identified; the c.193C>G has been iden-tified in a patient with acute myeloid leukemia.31 Inaddition, two cirrhosis-associated mutations in Exon 2and two mutations in Exon 16 of the TERT gene werenew mutations that have not been identified in previousstudies. The most common mutation in cirrhosis patientswas located in the c-terminus of the TERT gene (thec.3325G>A mutation resulting in the amino acid changep.G1109R) (Fig. 2B, Table 1, Supporting Fig. 1G). This

mutation is located next to an amino acid change(p.T1110M), which has previously been associated withpulmonary fibrosis.21

Clinical Characteristics of Cirrhosis Patients withTelomerase Mutations. The mean age of the patientswith telomerase mutation was 55.8 years (Table 2)compared to a mean age of 58.7 years in the group of cir-rhosis patients without telomerase mutations. There wasno obvious overrepresentation of a specific disease etiologyor ethnicity in the mutation carriers compared to the con-trols (see above). However, many of the mutation carriersshowed a rapid progression of chronic liver disease towardcirrhosis of liver cancer development (Table 2). Specifically,the frequency of endstage liver disease (defined as Child Bor C cirrhosis, hepatocellular carcinoma [HCC] occur-rence, and conduction or evaluation for liver transplanta-tion) was significantly higher in the group of cirrhosispatients harboring telomerase mutations (93%, 13 out of14) compared to cirrhosis patients without telomerasemutations (62%, 327 out of 507, P ¼ 0.042), indicatingthat telomerase mutations may induce a more aggressiveprogression of chronic liver disease—a hypothesis thatshould be addressed in future prospective trials.Functional Analysis of Cirrhosis-Associated

Telomerase Mutations. Cirrhosis patients carryingtelomerase gene mutations had significantly shortertelomeres in peripheral blood cells compared to controlpatients without telomerase gene mutations (Fig. 3A,P ¼ 0.0004). A significant reduction of telomeraseactivity was detectable by TRAP assay for three of theinvestigated, cirrhosis-associated telomerase mutations(p.P65A: P < 0.0001, p.G1109R: P ¼ 0.0035, andp.P380S: P < 0.0001), including the most frequentmutation p.G1109R (Fig. 3B,C). Studies on telomer-ase-negative, human fibroblasts revealed that the cirrho-sis-associated TERTmutations (p.P65A and p.G1109R)did not lose immortalization capacity when overex-pressed (Fig. 3D). However, proliferation rates of fibro-blast lines expressing these TERT mutants were signifi-cantly reduced compared to cells expressing wildtypeTERT (Fig. 3D). Similarly, Epstein-Barr virus-immor-talized, primary lymphocytes from two patients with ahomozygous p.G1109R TERT mutation showed animpaired proliferation capacity (Fig. 3E) and shortenedtelomeres (Fig. 3F,G) compared to immortalized lym-phocytes from cirrhosis patients with wildtype TERT.cH2Ax staining of liver sections of six cirrhosis muta-tions carriers and eight liver cirrhosis patients withouttelomerase mutation did not reveal any significant dif-ference in the percentage of cH2Ax-positive hepato-cytes, a marker for the induction of DNA double-strandbreaks (data not shown).


Fig. 2. Localization of cirrhosis-associated telomerase gene mutations. Localization of cirrhosis-associated telomerase gene mutations (A) inthe TERC gene locus (figure adapted from Podlevsky et al.36) (B) in the TERT gene. Amino acid positions of the N-terminal, reverse transcriptase,and C-terminal domains (amino acids are denoted by single-letter abbreviations). The N-terminal region domains (GQ, VSR, CP, QFP), the centralreverse transcriptase motifs (T, 1, 2, A, IFD, B, C, D, E), and the C-terminal region domains (E-I to -IV) are indicated as well. The hypomorphictelomerase reverse transcriptase variants characterized in this study (coding variants only) are displayed in boxes above the corresponding aminoacid sequence. The untranslated region (3-UTR) of the gene with respective variants found in the study cohort is also shown.


Discussion

The current study provides the first evidence that telo-merase mutations are associated with the evolution of cir-rhosis as a consequence of chronic liver disease. The totalallele frequency of telomerase mutations in cirrhosispatients was 0.017, compared to 0.003 in healthy con-

trols or hepatitis C patients without fibrosis progression(P ¼ 0.0007). Furthermore, subgroup analysis (numberof identified mutations in healthy controls compared tothe cirrhosis group: P ¼ 0.0021, number of mutations inchronic liver disease patients without cirrhosis comparedto the cirrhosis group: P ¼ 0.0349) reconfirmed that theidentified telomerase mutations are associated with

Fig. 3. Functional analysis of cirrhosis-associated telomerase gene mutations. (A) The scatterplot shows the telomere length in total white bloodcells from peripheral blood in correlation with the age of liver cirrhosis patients with and without telomerase gene mutations. Linear regression analy-sis revealed a significant difference between the two groups (P ¼ 0.0004). Telomere length of p.G1109R TERT mutation carriers is highlighted (greentriangles). (B-D) Telomerase-negative human fibroblasts (BJ) that were infected with retroviruses carrying (1) wildtype TERT (wt hTERT); (2) a previ-ously described, dominant-negative TERT mutation (DN hTERT)37; and (3) cirrhosis-associated TERT mutations (p.P65A, p.P380S, or p.G1109R).(B,C) Telomerase activity was measured after infection of telomerase-negative BJ-fibroblasts with wildtype TERT or the indicated TERT-mutant. The his-togram shows mean values with error bars indicating standard error of the mean. Note that a significant reduction of telomerase activity was seen forthree cirrhosis-associated TERT mutations relative to wildtype TERT (unpaired Student’s t test, P < 0.0001 for p.P65A, P ¼ 0.0035 for G1109R, andP < 0.0001 for p.P380S). (C) Gel photograph showing a TRAP assay to determine telomerase activity in serially diluted protein samples of fibro-blasts transfected with either wildtype or mutated hTERT in protein concentrations of 50, 100, 150, and 200 ng/lL plus one heat-treated test extractfor each cell line. The arrow points to the internal PCR control, the intensity of the ladder (extension products) corresponds to telomerase activity. (D) The linegraph shows growth curves of telomerase-negative, human fibroblasts (BJ) that were stably infected with the indicated TERT expression constructs. Note that thetwo tested, cirrhosis-associated TERT mutations (p.P65A, p.G1109R) did not lose immortalization potential but showed significantly reduced growth rates com-pared to wildtype TERT infected cells (unpaired Student’s t test, P< 0.005, day 59 and day 157). (E-G) B lymphocytes were isolated from EDTA blood of homo-zygote p.G1109R mutation carriers (age 49.5 6 2.5, n ¼ 2) and healthy controls (age 56.5 6 5.5, n ¼ 2) and immortalized by Epstein-Barr virus infection.(E) The line graph shows growth curves of immortalized polyclonal B cell lines in suspension culture. Note that immortalized lymphocytes of homozygousp.G1109R mutation carriers showed significantly reduced growth rates compared to immortalized lymphocytes of controls (unpaired Student’s t test, P ¼0.004, day 18). (F) Representative southern blot to determine telomere length of immortalized B lymphocytes derived from EDTA blood of homozygousp.G1109R mutation carriers and healthy controls. (G) The histogram shows a reduction in telomere length of immortalized B lymphocytes derived from homozy-gous p.G1109R mutation carriers (3.47 kb6 0.67 kb) compared to healthy controls (unpaired Student’s t test, 8.156 0.08, P¼ 0.0198).


cirrhosis but do not occur in healthy controls or patientswith indolent HCV infection. To our knowledge, thesedata represent the first association of telomerase muta-tions with the evolution and progression of cirrhosis inresponse to chronic liver injury. The ethnic group in ourstudy consisted mainly of whites (70.1%). It remains tobe analyzed whether telomerase mutations occur withsimilar frequency in other ethnic groups.The study shows that cirrhosis-associated TERT

mutations exhibit an impaired function compared towildtype TERT. Cirrhosis-associated telomerase muta-tions result in reduced telomerase activity, impairedtelomere maintenance, and reduced growth rates offibroblasts and lymphocytes. Moreover, a reduction intelomere length was seen in peripheral blood andimmortalized lymphocytes of mutation carriers com-pared to controls. We did not see any significant dif-ference in the percentage of cH2Ax-positive hepato-cytes between liver cirrhosis patients with and withouttelomerase mutations. This, however, does not argueagainst an involvement of telomerase mutations in cir-rhosis. Previous studies have demonstrated that telo-mere shortening and senescence are general signs ofcirrhosis induced by different etiologies.13,14 We pro-pose that telomerase mutations can lead to acceleratedtelomere shortening, thus shortening the time to pro-gression of chronic liver disease toward cirrhosis. Inaddition, telomerase mutations may have extratelomericeffects influencing disease progression. Recent studieshave revealed telomere length-independent effects ofTERT in regulating the transcriptional function of theWnt-signaling pathway and stem cell activity inmice.32 It remains to be seen whether cirrhosis-associ-ated TERT mutations show defects in these noncanon-ical TERT-pathways and whether TERT mutationsaffect the latency of cirrhosis development in patientswith chronic liver disease.Several lines of argument indicate that the current

study likely underestimated the true rate of telomere-related mutations in cirrhosis: (1) other components ofthe telomerase enzyme complex have been shown tobe essential for telomerase activity33 and mutations inone of these components (dyskerin) have been associ-ated with telomere shortening and human disease34;(2) mutations in telomere-binding proteins can impairthe function of telomeres, and a first report has recentlylinked mutation in the telomere-binding protein, TIN2,to the evolution of aplastic anemia35; (3) mutations innoncoding sequences could impair the expression ofboth TERT and TERC. Together, we anticipate thatfull coverage sequencing (coding and noncoding sequen-ces) of all known components of the telomerase enzyme

complex and the shelterin complex of telomere-bindingproteins will reveal an increasing percentage of telo-mere-related mutations in human cirrhosis.It remains to be analyzed whether TERT mutations

influence the development of HCC. Cirrhosis representsone of the main risk factors for HCC formation. There isample evidence that telomere shortening increases the riskof cancer formation in humans by inducing chromosomalinstability. Therefore, TERT mutations that increase therisk of cirrhosis formation may increase the risk of HCC.In contrast, tumor cells need to activate telomere mainte-nance mechanisms in order to gain immortal growthcapacity—a prerequisite for tumor formation. Thus, it ispossible that TERTmutations could also protect individu-als from cancer formation. In the current cohort, 5/14(36%) cirrhosis patients with telomerase mutations(including one patient with a homozygous p.G1109RTERT mutation) developed HCC, indicating that telo-merase mutations did not prevent cancer formation.Together, the current findings represent the first evi-

dence for telomerase mutations in cirrhosis induced bychronic liver disease. The results indicate that telomerasemutations represent confounding factors increasing therisk of cirrhosis formation in the context of chronic liverdisease. The study results improve our understanding onthe molecular causes of cirrhosis. These findings will influ-ence the future development of molecular therapies for cir-rhosis patients and may also have an impact on future sur-veillance programs and decision making in treatment ofpatients with chronic liver disease and cirrhosis.

References1. Friedman SL. Mechanisms of hepatic fibrogenesis. Gastroenterology

2008;134:1655-1669.2. Malhi H, Gores GJ. Cellular and molecular mechanisms of liver. Gas-

troenterology 2008;134:1641-1654.3. Williams EJ, Iredale JP. Liver cirrhosis. Postgrad Med J 1998;74:193-202.4. El-Serag HB, Rudolph KL. Hepatocellular carcinoma: epidemiology

and molecular carcinogenesis. Gastroenterology 2007;132:2557-2576.5. Durand F, Valla D. Assessment of prognosis of cirrhosis. Semin Liver

Dis 2008;28:110-122.6. Ku NO, Gish R, Wright TL, Omary MB. Keratin 8 mutations in patients

with cryptogenic liver disease. N Engl J Med 2001;344:1580-1587.7. Strnad P, Lienau TC, Tao GZ, Lazzeroni LC, Stickel F, Schuppan D,

et al. Keratin variants associate with progression of fibrosis duringchronic hepatitis C infection. HEPATOLOGY 2006;43:1354-1363.

8. Rudolph KL, DePinho RA. Telomeres and telomerase in experimentalliver cirrhosis. In: Chisari FV, Fausto N, Schachter D, Shafritz DA,Arias IM, Boyer JL, eds. The Liver: Biology and Pathobiology. 4th ed.Philadelphia: Lippincott Williams & Wilkins; 2001:1001-1010.

9. Kim NW, Piatyszek MA, Prowse KR, Harley CB, West MD, Ho PL,et al. Specific association of human telomerase activity with immortalcells and cancer. Science 1994;266:2011-2015.

10. Allsopp RC, Chang E, Kashefi-Aazam M, Rogaev EI, Piatyszek MA,Shay JW, et al. Telomere shortening is associated with cell division invitro and in vivo. Exp Cell Res 1995;220:194-200.


11. Brown JP, Wei W, Sedivy JM. Bypass of senescence after disruption ofp21CIP1/WAF1 gene in normal diploid human fibroblasts. Science1997;277:831-834.

12. Wege H, Chui MS, Le HT, Strom SC, Zern MA. In vitro expansionof human hepatocytes is restricted by telomere-dependent replicativeaging. Cell Transplant 2003;12:897-906.

13. Wiemann SU, Satyanarayana A, Tsahuridu M, Tillmann HL, Zender L,Klemnauer J, et al. Hepatocyte telomere shortening and senescence aregeneral markers of human liver cirrhosis. FASEB J 2002;16:935-942.

14. Kojima H, Yokosuka O, Imazeki F, Saisho H, Omata M. Telomeraseactivity and telomere length in hepatocellular carcinoma and chronicliver disease. Gastroenterology 1997;112:493-500.

15. Rudolph KL, Chang S, Millard M, Schreiber-Agus N, DePinho RA.Inhibition of experimental liver cirrhosis in mice by telomerase genedelivery. Science 2000;287:1253-1258.

16. Satyanarayana A, Wiemann SU, Buer J, Lauber J, Dittmar KE, WustefeldT, et al. Telomere shortening impairs organ regeneration by inhibiting cellcycle re-entry of a subpopulation of cells. EMBO J 2003;22:4003-4013.

17. Lechel A, Manns MP, Rudolph KL. Telomeres and telomerase: new tar-gets for the treatment of liver cirrhosis and hepatocellular carcinoma.J Hepatol 2004;41:491-497.

18. Delhaye M, Louis H, Degraef C, Le Moine O, Deviere J, Gulbis B,et al. Relationship between hepatocyte proliferative activity and liverfunctional reserve in human cirrhosis. HEPATOLOGY 1996;23:1003-1011.

19. Vulliamy T, Marrone A, Goldman F, Dearlove A, Bessler M, MasonPJ, et al. The RNA component of telomerase is mutated in autosomaldominant dyskeratosis congenita. Nature 2001;413:432-435.

20. Yamaguchi H, Calado RT, Ly H, Kajigaya S, Baerlocher GM, ChanockSJ, et al. Mutations in TERT, the gene for telomerase reverse transcrip-tase, in aplastic anemia. N Engl J Med 2005;352:1413-1424.

21. Armanios MY, Chen JJ, Cogan JD, Alder JK, Ingersoll RG, Markin C,et al. Telomerase mutations in families with idiopathic pulmonary fi-brosis. N Engl J Med 2007;356:1317-1326.

22. Tsakiri KD, Cronkhite JT, Kuan PJ, Xing C, Raghu G, Weissler JC,et al. Adult-onset pulmonary fibrosis caused by mutations in telomer-ase. Proc Natl Acad Sci U S A 2007;104:7552-7557.

23. Mason PJ, Wilson DB, Bessler M. Dyskeratosis congenita—a disease ofdysfunctional telomere maintenance. Curr Mol Med 2005;5:159-170.

24. Alder JK, Chen JJ, Lancaster L, Danoff S, Su SC, Cogan JD, et al.Short telomeres are a risk factor for idiopathic pulmonary fibrosis. ProcNatl Acad Sci U S A 2008;105:13051-13056.

25. Calado RT, Regal JA, Kleiner DE, Schrump DS, Peterson NR, Pons V,et al. A spectrum of severe familial liver disorders associate with telo-merase mutations. PLoS ONE 2009;11:e7926.

26. Wall FE, Henkel RD, Stern MP, Jenson HB, Moyer MP. An efficientmethod for routine Epstein-Barr virus immortalization of human Blymphocytes. In Vitro Cell Dev Biol Anim 1995;31:156-159.

27. Vulliamy T, Marrone A, Dokal I, Mason PJ. Association betweenaplastic anaemia and mutations in telomerase RNA. Lancet 2002;359:2168-2170.

28. Vulliamy TJ, Dokal I. Dyskeratosis congenita: the diverse clinical pre-sentation of mutations in the telomerase complex. Biochimie 2008;90:122-130.

29. Takeuchi J, Ly H, Yamaguchi H, Carroll KA, Kosaka F, Sawaguchi K,et al. Identification and functional characterization of novel telomerasevariant alleles in Japanese patients with bone-marrow failure syndromes.Blood Cells Mol Dis 2008;40:185-191.

30. Armbruster BN, Banik SS, Guo C, Smith AC, Counter CM. N-termi-nal domains of the human telomerase catalytic subunit required forenzyme activity in vivo. Mol Cell Biol 2001;21:7775-7786.

31. Calado RT, Regal JA, Hills M, Yewdell WT, Dalmazzo LF, ZagoMA, et al. Constitutional hypomorphic telomerase mutations inpatients with acute myeloid leukemia. Proc Natl Acad Sci U S A2009;106:1187-1192.

32. Park JI, Venteicher AS, Hong JY, Choi J, Jun S, Shkreli M, et al. Telo-merase modulates Wnt signalling by association with target gene chro-matin. Nature 2009;460:66-72.

33. Venteicher AS, Abreu EB, Meng Z, McCann KE, Terns RM, Veen-stra TD, et al. A human telomerase holoenzyme protein required forCajal body localization and telomere synthesis. Science 2009;323:644-648.

34. Mitchell JR, Wood E, Collins K. A telomerase component is defective inthe human disease dyskeratosis congenita. Nature 1999;402:551-555.

35. Walne AJ, Vulliamy T, Beswick R, Kirwan M, Dokal I. TINF2 muta-tions result in very short telomeres: analysis of a large cohort of patientswith dyskeratosis congenita and related bone marrow failure syndromes.Blood 2008;112:3594-3600.

36. Podlevsky JD, Bley CJ, Omana RV, Qi X, Chen JJ. The telomerasedatabase. Nucleic Acids Res 2008;36:D339-D343.

37. Hahn WC, Stewart SA, Brooks MW, York SG, Eaton E, Kurachi A,et al. Inhibition of telomerase limits the growth of human cancer cells.Nat Med 1999;5:1164-1170.


telomerase gene mutations are associated with cirrhosis formation

Documents