Telomerase reconstitution immortalizes human fetal hepatocytes without disrupting their differentiation potential

GASTROENTEROLOGY 2003;124:432– 444

Telomerase Reconstitution Immortalizes Human Fetal Hepatocytes Without Disrupting Their Differentiation Potential HENNING WEGE,* HAI T. LE,* MICHAEL S. CHUI,* LI LIU,* JIAN WU,* RANJIT GIRI,‡ HARMEET MALHI,‡ BALJIT S. SAPPAL,‡ VINAY KUMARAN,‡ SANJEEV GUPTA,‡ and MARK A. ZERN* *Transplant Research Institute, University of California, Davis Medical Center, Sacramento, California; and ‡Marion Bessin Liver Research Center, Albert Einstein College of Medicine, Bronx, New York

See editorial on page 568. Background & Aims: The availability of in vitro expandable human hepatocytes would greatly advance liverdirected cell therapies. Therefore, we examined whether human fetal hepatocytes are amenable to telomerase-mediated immortalization without inducing a transformed phenotype and disrupting their differentiation potential. Telomerase is a ribonucleoprotein that plays a pivotal role in maintaining telomere length and chromosome stability. Human somatic cells, including hepatocytes, exhibit no telomerase activity. Consequently, their telomeres progressively shorten with each cell cycle until critically short telomeres trigger replicative senescence. Methods: The catalytic subunit, telomerase reverse transcriptase, was expressed in human fetal hepatocytes. Transduced cells were characterized for telomerase activity, telomere length, proliferative capacity, hepatocellular functions, oncogenicity, and their in vivo maturation potential. Results: The expression of human telomerase reverse transcriptase restored telomerase activity in human fetal hepatocytes. Telomerase-reconstituted cells were capable of preserving elongated telomeres, propagated in culture beyond replicative senescence for more than 300 cell doublings (to date), and maintained their liver-specific nature, as analyzed by a panel of hepatic growth factors, growth factor receptors, and transcription factors as well as albumin, glucose-6-phosphatase, glycogen synthesis, cytochrome P450 (CYP) expression profiles, and urea production. Moreover, the immortalized cells exhibited no oncogenicity, and no upregulation of c-Myc was detected. The cells engrafted and survived in the liver of immunodeficient mice with hepatocellular gene expression. Conclusions: Reconstitution of telomerase activity induces indefinite replication in human fetal hepatocytes and offers unique opportunities for examining basic biologic mechanisms and for considering development of stable cell lines for liver-directed therapies.

iver failure constitutes a major health problem. Liverdirected cell therapies, such as hepatocyte transplantation1 and bioartificial liver support,2,3 offer alternatives to orthotopic liver transplantation, which, as the only currently available effective treatment for liver failure, is restricted by the shortage of donor organs. Although use of primary human hepatocytes would be ideal for liverdirected cell therapies, the availability of these cells is limited, and expansion of human hepatocytes in culture is difficult. Using malignant cell lines, e.g., HepG2 or C3A cells,2 or xenogenic (porcine) hepatocytes3 for liverdirected cell therapies carries potential risks to recipients, e.g., inoculation of tumor cells4 or zoonotic diseases, notably transmission of porcine endogenous retrovirus.5,6 This situation has prompted intensive research to generate human hepatocyte-derived cell lines and to explore various strategies for immortalizing and expanding the number of human hepatocytes. Telomeres consist of kilobases (kb) of reiterated hexanucleotides with the sequence 5⬘-TTAGGG-3⬘ and constitute the specialized ends of eukaryotic chromosomes.7 An adequate telomere length is required for chromosome stability because telomere looping is necessary to provide capping and protection of chromosome ends against degradation and ligation.7 Normal somatic cells lose telomeric DNA at a rate of 50 to 200 base pairs (bp) per population doubling (PD) because of replication-associated attrition.8 Eventually, this telomeric loss leads to critically short telomeres that are incapable of looping, and replicative senescence is activated with loss of cell

L

Abbreviations used in this manuscript: bp, base pair; BrdU, 5-bromo2ⴕ-deoxyuridine; C/EBP, CCAAT/enhancer binding protein; CYP, cytochrome P450; FH, fetal hepatocytes; FH-hTERT, hTERT-transduced fetal hepatocytes; HGFR, hepatocyte growth factor receptor; HNF, hepatocyte nuclear factor; hTERT, human telomerase reverse transcriptase; kb, kilobase; PCR, polymerase chain reaction; PD, population doublings; RT-PCR, reverse transcription-polymerase chain reaction; TRF, terminal restriction fragment. © 2003 by the American Gastroenterological Association 0016-5085/03/$35.00 doi:10.1053/gast.2003.50064

February 2003

division capacity.9 Consequently, it has been suggested that telomere attrition governs the proliferative lifespan of human somatic cells (mitotic clock) and represents the molecular correlate for the Hayflick limit of cell division.10 In germline cells, stem cells, and cancer cells, telomerase, a ribonucleoprotein complex with reverse transcriptase activity, counterbalances telomere attrition and retains a telomere equilibrium by de novo synthesis of telomeric repeats onto the eroding chromosome ends.11 Human somatic cells exhibit no telomerase activity because of the down-regulation of the catalytic telomerase subunit telomerase reverse transcriptase (hTERT). However, other elements of the telomerase holoenzyme have been detected in various cell types, including human hepatocytes.12 Therefore, hTERT expression is considered rate limiting for telomerase activity. In contrast with mature human hepatocytes, hepatocytes isolated from the fetal human liver possess significant spontaneous proliferative capacity, although the proliferative activity of cultured human fetal hepatocytes (FH) begins to decrease over several months,13 which is similar to other fetal cells.9 It has been demonstrated that telomerase reconstitution, via hTERT-expression, enables proliferating human somatic cells, such as foreskin fibroblasts,14 mammary epithelial cells,15 microvascular endothelial cells,16 and keratinocytes,17 to preserve their telomeres and to bypass replicative senescence while maintaining phenotypic functions. Moreover, telomerase-immortalized cells do not exhibit oncogenicity and are responsive to various types of DNA damage with cell-cycle arrest.18,19 These findings suggested to us that human hepatocytes should be amenable to telomerasemediated immortalization without disrupting their differentiation potential to generate untransformed hepatocyte-derived cell lines.

Materials and Methods Cell Isolation and Culture FH were isolated and cultured according to methods described elsewhere.13 In brief, tissues from elective terminations at 22–24 weeks of gestation were obtained from the Fetal Tissue Repository at Albert Einstein College of Medicine under approval from the institutional Committee on Clinical Investigations. Fetal livers were digested in 0.05% collagenase (Worthington Biochemical Corporation, Lakewood, NJ) for 30 – 40 minutes at 37°C. Dissociated cells were passed through 80-␮m dacron mesh, washed, and pelleted at 500g for 4 minutes at 4°C. The cell pellet was resuspended in Dulbecco’s modified Eagle’s medium (Invitrogen, Carlsbad, CA), containing 5 ␮g/mL insulin (Sigma, St. Louis, MO), 2.4 ␮g/mL hydrocortisone (Sigma), 10% inactivated fetal bovine serum

TELOMERASE–IMMORTALIZED FETAL HEPATOCYTES

433

(Invitrogen), and standard antibiotics. Primary cultures were established in tissue culture dishes at 4 ⫻ 103 cells per cm2 in a humidified 5% CO2 atmosphere. Cultures were passaged at a ratio of 1:4 thereafter.

Cell Transduction and Selection To produce retroviral vectors, 293T cells were cotransfected with the amphotropic packaging plasmid pCL-Ampho20 (Imgenex, San Diego, CA) and the expression plasmid pBabe-puro-hTERT or the control plasmid pBabe-puro21 in 100-mm tissue culture dishes. The pBabe-puro plasmids and the 293T cell line were kindly provided by Dr. R. A. Weinberg, Whitehead Institute for Biomedical Research, Cambridge, Massachusetts. The expression plasmid pBabe-purohTERT contains hTERT and puromycin acetyltransferase as a selectable marker under the control of a Moloney murine leukemia virus-derived promoter. FuGene 6 (Roche Molecular Biochemicals, Indianapolis, IN) was used for cotransfection according to the manufacturer. After 48 and 72 hours, viral supernatants were purified by passing through a 0.45-␮m filter and used for immediate transduction in the presence of 8 ␮g/mL hexadimethrine bromide (Sigma). The titers of the supernatants were 2– 4 ⫻ 104 transducing units/mL, as determined by bioassay. On 4 consecutive days, passaged FH were transduced for 3 hours daily, corresponding to a total multiplicity of infection of 0.5. Following transduction and recovery of transduced cells for 24 hours, cells were selected for 7 days with 0.75 ␮mol/L puromycin dihydrochloride (Sigma).

Analysis of Transgene Expression and Telomerase Reconstitution hTERT expression was evaluated by real-time quantitative reverse transcription-polymerase chain reaction (RTPCR). Total RNA (150 ng), extracted with TriReagent-LS (Molecular Research Center, Cincinnati, OH), was analyzed with the LightCycler TeloTAGGG hTERT Quantification Kit (Roche Molecular Biochemicals). hTERT expression was normalized against the housekeeping gene porphobilinogen deaminase. The TeloTAGGG Telomerase PCR ELISAPLUS (Roche Molecular Biochemicals) was used to measure telomerase activity in cellular extracts in comparison with a control template of 0.1 amole telomeric repeats. The telomerase activity assay uses the PCR-based telomeric repeat amplification protocol11 and utilizes an enzyme-linked immunosorbent assay for quantification. The results were expressed as total product generated per microgram of cellular protein over a telomereprimer elongation time of 20 minutes.22

Southern Blot Analysis of Mean Telomere Length Genomic DNA was extracted with DNAzol (Invitrogen), digested with HinfI and RsaI, electrophoresed, blotted, and transferred to positively charged Magnacharge membranes (Osmonics, Minnetonka, MN) by alkaline blotting as described.23 Membranes were hybridized with 32P-(TTAGGG)3

434

WEGE ET AL.

as a telomeric probe using Hybrisol II (Intergen, Purchase, NY) and washed following published protocols.23 Mean terminal restriction fragment (TRF) length was determined from scanned autoradiographs by integrating the signal intensity above background over the entire TRF distribution using ImageQuaNT software (Molecular Dynamics, Sunnyvale, CA). In brief, TRF smears were divided into 30 equally sized boxes, and the signal intensity within each box (ODb) was used together with the molecular weight at the midpoint of the box (MWb) to compute mean TRF length (L), in which L ⫽ ⌺ (ODb ⫻ MWb)/⌺ (ODb).24 A subtelomeric portion of 2.5 kb was assumed for converting mean TRF length to mean telomere length.25

Proliferation Analyses Cells were passaged when confluent, and the number of population doublings (PD) was calculated as PD ⫽ log (Nf/ Ni)/log 2, in which Nf is the final number of cells and Ni the number of cells initially seeded.26 No correction was made for dead cells or for cells that failed to reinitiate growth at subculture. To measure the percentage of cells in S-phase (DNA synthesis), 106 cells were cultured overnight in 25-cm2 tissue culture flasks and labeled for 2 hours with 10 ␮mol/L 5-bromo-2⬘-deoxyuridine (BrdU) on the following day. BrdUincorporation was detected with the BrdU Flow Kit (BD PharMingen, San Diego, CA) using a FACScan flow cytometer and CellQuest software (BD Immunocytometry Systems, San Jose, CA). To visualize senescent cells, subconfluent cultures were fixed with 0.2% glutaraldehyde in 2% formaldehyde and stained in a humidified chamber at 37°C for 12–16 hours. The staining solution contained 1 mg/mL 5-bromo-4-chloro-3indolyl ␤-D-galactopyranoside (Sigma) in staining buffer described previously.27 Stained cell populations were evaluated by light microscopy and considered senescent if more than 90% of cells exhibited the characteristic blue senescenceassociated ␤-galactosidase stain.

Phenotype Characterization For the evaluation of hepatic growth factors, growth factor receptors, and transcription factors, total RNA (1 ␮g), extracted with TRIzol (Invitrogen), was analyzed by 1-step RT-PCR employing the GeneAmp Gold RNA PCR Kit (Applied Biosystems, Foster City, CA) with 0.2 mmol/L forward and reverse primers. Cytochrome P450 (CYP) expression profiles were investigated by reverse transcription of DNase I-digested RNA employing ThermoScript RT-PCR Systems (Invitrogen). Amplification was performed with Platinum PCR Super Mix (Invitrogen) and 200 nmol/L forward and reverse primers. Primer sequences and RT-PCR parameters are listed in Table 1. All samples, including no template controls, were subjected to 35 PCR cycles and resolved by standard agarose gel electrophoresis. To detect albumin synthesis, ethanol-fixed cells were incubated with 1:100 diluted monoclonal anti-human serum albumin antibody clone HSA-11 (Sigma) or IgG2a isotype control (Sigma) for 4 hours at 4°C. Antibody binding was

GASTROENTEROLOGY Vol. 124, No. 2

detected with 1:5000 diluted anti-mouse immunoglobulin conjugated with fluorescein isothiocyanate (Sigma), which was incubated with cells for 45 minutes at 4°C before flow cytometry. Glucose-6-phosphatase activity was visualized in unfixed cells by incubating with glucose-6-phosphate and postfixation with ethanol essentially as described previously.28 For glycogen staining, cells were fixed in cold ethanol for 10 minutes at room temperature. After hydration, cells were incubated with 1% aqueous periodic acid and Schiff’s reagent for 5 minutes each, followed by washes in 0.5% sodium bisulphite water.28 Urea synthesis was measured in confluent cell cultures. In brief, cells were cultured in 12-well tissue culture plates, washed multiple times with phosphate buffered saline, and incubated in 1 mL serum-free Hepatocyte Maintenance Medium (BioWittaker, Walkersville, MD) to avoid interference from serum. Urea concentrations were determined in centrifuged aliquots of culture medium with colorimetric Infinity BUN Reagent (Sigma) and compared with standards with 2–200 ␮g urea/mL. Medium without urea and medium from 293T cell cultures served as negative controls. Cell counts were obtained to calculate urea production in pg/cell.

Oncogenicity Assays To determine serum dependence and contact inhibition, cells were cultured in serum-free medium for 5 days or in regular serum-supplemented growth medium until confluency for 15 days. The percentage of cells in S-phase was measured with the BrdU-incorporation assay specified above. Soft agar colony assays were performed to evaluate anchorage-independent growth as described.29 Briefly, dilutions of 50, 500, and 5000 cells in 0.33% agar were overlaid onto 0.5% base agar in 60-mm tissue culture dishes. Malignant HepG2 cells served as a positive control. After 15 days, colony formation was scored by counting cell colonies microscopically. Tumorigenicity was assessed by inoculating 106 cells subcutaneously into the dorsal flanks, left and right of the midline, of 6 – 8-week-old male athymic nu/nu mice (Charles River Laboratories, Wilmington, MA). Tumorigenic 293T cells were inoculated as a positive control and tumor formation was monitored twice weekly for 15 weeks. Tumor development in the liver was investigated by transplanting 1.5 ⫻ 106 cells through the portal vein into 8to 10-week-old male athymic nude and NOD.CB17-Prkdcscid/J mice (Jackson Laboratories, Bar Harbor, ME), also known as NOD-SCID mice. After 10 –12 weeks, the livers were dissected and macroscopically examined for tumors.30 All animal experiments were approved by the institutional Animal Care and Use Administrative Advisory Committee of the University of California, Davis, California.

Analysis of c-Myc Expression Protein was extracted, quantitated, and Western blot analysis of c-Myc was performed following published procedures,31 using a 1:100 dilution of mouse anti-c-Myc monoclonal antibody 9E10 (Santa Cruz Biotechnology, Santa Cruz, CA). To ensure equivalent amounts of protein loading, membranes were stripped with 0.2 mol/L NaOH for 5 minutes and

February 2003

TELOMERASE–IMMORTALIZED FETAL HEPATOCYTES

435

Table 1. Primer Sequences and RT-PCR Parameters Gene HGF TGF␣ TGF␤1 TGF␤2 HGFR EGFR TGF␤1R TGF␤2R FGFR IGF-1R HNF1␣ HNF1␤ HNF3␤ HNF4 C/EBP␣ C/EBP␤ GAPDH CYP 1B1 CYP 2C9 CYP 2B CYP 3A4 CYP 2E1 CYP 1A1

Primer sequence 5⬘–3⬘

PCR parametersa

Ampliconb

F: AGGAGCCAGCCTGAATGATGA R: CCCTCTGATGTCCCAAGATTAGC F: ATGGTCCCCTCGGCTGGA R: GGCCTGCTTCTTCTGGCTGGCA F: GCCCTGGACACCAACTATTGCT R: AGGCTCCAAATGTAGGGGCAGG F: GATTTCCATCTACAAGACCACGAGGGACTTGC R: CAGCATCAGTTACATCGAAGGAGAGCCATTCG F: TGGTCCTTGGCGTCGTCCTC R: CTCATCATCAGCGTTATCTTC F: CTACCACCACTCTTTGAACTGGACCAAGG R: TCTATGCTCTCACCCCGTTCCAAGTATCG F: CGTGCTGACATCTATGCAAT R: AGCTGCTCCATTGGCATAC F: TGCACATCGTCCTGTGGAC R: GTCTCAAACTGCTCTGAAGTGTTC F: ATGTGGAGCTGGAAGTGCCTC R: GGTGTTATCTGTTTCTTTCTCC F: ACCCGGAGTACTTCAGCGCT R: CACAGAAGCTTCGTTGAGAA F: GTGTCTACAACTGGTTTGCC R: TGTAGACACTGTCACTAAGG F: GAAACAATGAGATCACTTCCTCC R: CTTTGTGCAATTGCCATGACTCC F: CACCCTACGCCTTAACCAC R: GGTAGTAGGAGGTATCTGCGG F: CTGCTCGGAGCCACAAAGAGATCCATG R: ATCATCTGCCACGTGATGCTCTGCA F: CAAGAAGTCGGTGGACAAGAAC R: CCTCATCTTAGACGCACCAAGT F: GCAACGCCGADAGAGGA R: TGTCCTGCATTGTCGCC F: CCATGGAGAAGGCTGGGG R: CAAAGTTGTCATGGATGACC F: CACCAAGGCTGAGACAGTGA R: GCCAGGTAAACTCCAAGCAC F: GGACAGAGACGACAAGCACA R: TGGTGGGGAGAAGGTCAAT F: GGCACACAGCCAAGTTTACA R: CCAGCAAAGAAGAGCGAGAG F: TGTGCCTGAGAACACCAGAG R: GCAGAGGAGCCAAATCTACC F: CCGCAAGCATTTTGACTACA R: GCTCCTTCACCCTTTCAGAC F: AGGCTTTTACATCCCCAAGG R: GCAATGGTCTCACCGATACA

95, 56, 72 1 min, 45 s, 1 min 95, 59, 72 45 s, 30 s, 1 min 95, 58, 72 45 s, 30 s, 1 min 95, 58, 72 45 s, 30 s, 1 min 95, 54, 72 30 s, 45 s, 1 min 95, 58, 72 45 s, 30 s, 1 min 95 s, 54, 72 30 s, 45 s, 1 min 95, 58, 72 45 s, 30 s, 1 min 95, 54, 72 30 s, 45 s, 1 min 95, 54, 72 30 s, 45 s, 1 min 95, 52, 72 45 s, 30 s, 1 min 95, 52, 72 1 m, 45 s, 1 min 95, 56, 72 1 m, 45 s, 1 min 95, 58, 72 45 s, 30 s, 1 min 95, 58, 72 45 s, 30 s, 1 min 95, 58, 72 45 s, 30 s, 1 min 95, 8, 72 45 s, 30 s, 1 min 94, 57, 72 30 s, 30 s, 1 min 94, 57, 72 30 s, 30 s, 1 min 94, 57, 72 30 s, 30 s, 1 min 94, 57, 72 30 s, 30 s, 1 min 94, 57, 72 30 s, 30 s, 1 min 94, 57, 72 30 s, 30 s, 1 min

360 297 161 503 342 675 251 784 165 229 251 374 235 363 450 252 196 230 200 209 201 202 197

C/EBP, CCAAT/enhancer binding protein; CYP, cytochrome P450; EGFR, epidermal growth factor receptor; F, forward primer; FGFR, fibroblast growth factor receptor; GAPDH, glyceraldehyde phosphate dehydrogenase; HGF, hepatocyte growth factor; HGFR, hepatocyte growth factor receptor; HNF, hepatocyte nuclear factor; IGF-1R, insulin-like growth factor-type I receptor; R, reverse primer; TGF, transforming growth factor; TGFR, transforming growth factor receptor. aTemperatures are tabulated in the first lane in degrees celsius and the corresponding times in the second lane. Performing one-step RT-PCR, reverse transcription was carried out at 42°C for 20 minutes with a pre-PCR denaturation at 95°C for 10 minutes. bAmplicon size in bp.

reprobed with goat anti-actin polyclonal antibody I-19 (Santa Cruz Biotechnology) at a dilution of 1:200. Peroxidase-conjugated anti-mouse and anti-goat immunoglobulins (Santa Cruz Biotechnology) were diluted as recommended by the manufacturer, followed by chemiluminiscence with ECL Western blotting detection reagents (Amersham Biosciences, Piscataway, NJ). Scanned Western blot images were quantitated

densitometrically using ImageQuaNT (Molecular Dynamics), and c-Myc signals were normalized against actin signals and further compared with passaged control FH.

Differentiation Bioassays To investigate differentiation potential, 2 ⫻ 106 cells were transplanted into the livers of partially hepatectomized

436

WEGE ET AL.

GASTROENTEROLOGY Vol. 124, No. 2

Table 2. TaqMan Primer-Probe Sets 5⬘-Labela

Gene

Primer-probe sequence 5⬘–3⬘

HSA

F: AGTTTGCAGAAGTTTCCAAGTTAGTG T: ACATTCAAGCAGATCTCCATGGCAGCA R: AGGTCCGCCCTGTCATCAG F: TCGCTACAGCCTTTGCAATG T: AGCCTTCATGGATCTGAGCCTCCGG R: TTGAGGGTACGGAGGAGTTCC F: GTGTATCAGCAGAGACCACCGA T: TTCTCCATTCATGATCTTGGCGATGCA R: CATCCAAGCTCATGGCATCA

HAT

HTF

FAM

FAM

FAM

Concentrationb 300 nmol/L 300 nmol/L 50 nmol/L 900 nmol/L 300 nmol/L 300 nmol/L 300 nmol/L 300 nmol/L 300 nmol/L

F, forward primer; FAM, 6-carboxyfluorescein; HAT, human ␣1-antitrypsin; HSA, human serum albumin; HTF, human transferrin; R, reverse primer; T, hybridization probe. aAll hybridization probes were quenched with 6-carboxytetramethylrhodamine. bPrimer concentrations were optimized for an annealing temperature of 60°C.

NOD-SCID mice through the portal vein.30 In brief, the liver was exposed, and the middle and/or right lobe, corresponding to 20%–30% of the liver mass, were ligated with 2-0 silk. After hepatectomy, 200 ␮L of cell suspension (10 ⫻ 106 cells/mL) were injected intraportally through a 31-gauge needle. Mice undergoing transplantation were killed 4 weeks after transplantation, and cell engraftment was visualized in cryosections using a human centromere probe as described previously.32 Human liver tissue and liver tissue samples from mice not undergoing transplantation served as controls. Total RNA was extracted from liver tissue to generate cDNA as described above. Real-time quantitative PCR was employed to quantitate the expression of albumin, ␣1-antitrypsin, and transferrin in comparison with glyceraldehyde phosphate dehydrogenase as housekeeping control. Primer-probe sets (Table 2) were specific for human targets and without cross-reactivity to murine cDNA. TaqMan Pre-Developed Assay Reagents (Applied Biosystems) were used to amplify glyceraldehyde phosphate dehydrogenase. Following amplification with TaqMan Universal PCR Master Mix (Applied Biosystems) and the ABI Prism 7700 thermal cycler (Applied Biosystems), semilog amplification curves were evaluated by comparative quantification (⌬⌬CT). Expression levels were normalized to the housekeeping control and compared with human hepatocytes in primary culture that were obtained from the Liver Tissue Procurement and Distribution System, Minneapolis, Minnesota. For an intraperitoneal bioassay, cells were attached to collagen-coated Cytodex 3 microcarrier beads (Amersham Biosciences), injected into the peritoneal cavity of NOD-SCID mice, and recovered after 4 weeks for histochemical analysis of frozen sections as described.13,32 To verify the presence of human cells on the microcarrier beads, genomic DNA was extracted and subjected to Alu-sx PCR.32

Statistical Analysis Data are presented as means ⫾ SEM. The number of observations is stated in the text and the Figures. The unpaired Student t test was used for statistical analysis, and P ⬍ 0.05 was considered significant.

Results Telomerase Activity and Telomere Length Following hTERT-Transduction Isolated hepatocytes from a fetal human liver were cultured and transduced with hTERT or a control vector without hTERT at passage 4. Following puromycin selection, high levels of hTERT messenger RNA (mRNA), 18.96 ⫾ 3.80 hTERT copies per copy of porphobilinogen deaminase (n ⫽ 5), were detected in hTERT-transduced cells (FH-hTERT) using real-time quantitative RT-PCR. In comparison, no hTERT message was amplified in either untransduced FH (n ⫽ 4) or cells transduced with the control vector (n ⫽ 2). Furthermore, telomerase activity, as assessed by the telomeric repeat amplification protocol, increased significantly from 0.17 ⫾ 0.05 (n ⫽ 4) to 65.75 ⫾ 4.49 total product generated 䡠 ␮g⫺1 䡠 20 min⫺1 (n ⫽ 5; P ⬍ 0.001), levels that were similar to HepG2 and other telomerase-positive cell lines. The low telomerase activity in control FH was not consistently detected and was presumably due to occasional telomerase-positive cells. No increase in telomerase activity was detected in FH transduced with the control vector (n ⫽ 2). Analysis of telomerase activity after 50 PD demonstrated that telomerase activity did not decrease in FH-hTERT, despite continuous cell proliferation during long-term culture. Having established that hTERT-transduction successfully reconstituted telomerase activity in FH, we assessed telomere length by Southern blot analysis to investigate whether telomerase reconstitution resulted in telomere stabilization. As expected, FH-hTERT preserved elongated telomeres, whereas the mean telomere length in untransduced control cells shortened progressively to less than 6 kb after 25 PD. This is illustrated by representative TRF smears (Figure 1A) and the linear trendlines

February 2003

TELOMERASE–IMMORTALIZED FETAL HEPATOCYTES

437

Figure 1. Telomere length as determined by Southern blot analysis. Genomic DNA was digested with HinfI and RsaI, electrophoresed, blotted, and hybridized with 32P-(TTAGGG)3 as a telomeric probe. (A) Shown are representative TRF smears of control FH and FHhTERT at different PD. Arrow indicates gel loading and the ruler DNA size markers in kb. (B) Mean telomere length was estimated based on weighted densitometric calculations, assuming a subtelomeric portion of 2.5 kb. Each data point represents the mean of 2 or more observations. Linear trendlines were added to visualize telomere length development. The first data point of control FH and FH-hTERT at 14 PD is 7.22 ⫾ 0.27 kb (mean ⫾ SEM; n ⫽ 3).

of mean telomere length development (Figure 1B). In comparison, the mean telomere length of senescent human fibroblasts in vitro was reported to be 6 kb.8 –10 According to our data, the telomere loss in control FH was 50 to 150 bp per PD (Figure 1B), comparable with reported telomere attrition rates for human fibroblasts.8 –10

Figure 2C). After 30 PD, which represents a time close to senescence, or near senescence with less than 1 PD per week,14 the fraction of S-phase cells was significantly lower in control FH than FH-hTERT (9.62% ⫾ 0.03%; n ⫽ 3 vs. 34.06% ⫾ 1.69%; n ⫽ 5; P ⬍ 0.001; Figure 2C).

Proliferative Capacity Following Telomerase Reconstitution

To address whether FH-hTERT maintained hepatocellular functions, transduced cells were examined in comparison with primary and passaged control FH. First, the expression profiles of various hepatic growth factors, growth factor receptors, and transcription factors were established. It was noteworthy that most genes examined were expressed in passaged FH (passage 1), as well as in FH-hTERT after 50 PD (Figure 3). Passaged FH showed a decrease in hepatocyte growth factor receptor (HGFR) expression (Figure 3). Moreover, in these cells, expression of CCAAT/enhancer binding protein (C/EBP) ␣ and hepatocyte nuclear factor (HNF) 4 was virtually undetectable in the fourth and subsequent cell passages (data not shown). In contrast, HGFR expression appeared to have been up-regulated in FH-hTERT at 50 PD, and these cells continued to express liver-enriched differentiation factors, especially C/EBP␣ and HNF-4 (Figure 3). Members of the C/EBP gene family regulate both metabolism and cell growth and are predominantly expressed in tissues with active gluconeogenesis.33 Therefore, we investigated glucose metabolism in FH-hTERT.

To examine the ability of FH-hTERT to bypass replicative senescence, we cultured control FH and FHhTERT and determined cell number and cell proliferation rates with each subpassage. Altogether, 5 FHhTERT and 6 control FH (4 untransduced and 2 control vector-transduced) cultures were established and studied. After 30 –35 PD, cell proliferation virtually ceased in control FH (Figure 2A) with entry into replicative senescence, as suggested by senescence-associated ␤-galactosidase stain (Figure 2B). However, in FH-hTERT, cell proliferation accelerated at this stage of expected replicative senescence and continued for more than 300 PD (to date). The immortality threshold, set at twice the number of PD at replicative senescence, was reached 200 –250 days after transduction with hTERT (Figure 2A). To further ascertain the observed changes in the cell division potential, the percentage of cells in S-phase was measured. In control FH, S-phase cell fraction decreased in cells approaching senescence (27 PD and 30 PD;

Characterization of TelomeraseImmortalized FH

438

WEGE ET AL.

GASTROENTEROLOGY Vol. 124, No. 2

single-cell level. It was noteworthy that the culture conditions used in these studies were designed at supporting cell proliferation, and conditions have not been optimized for inducing differentiated hepatocellular functions. Additional studies concerning hepatocellular functions in these cells examined CYP gene expression and ureagenesis. FH-hTERT expressed CYP 1 and 2 families of genes that are involved in hepatic drug metabolism. In particular, dioxin-inducible CYP 1B1, ethanol-inducible CYP 2E1, aromatic compound-inducible CYP 1A1, and a marker gene for the CYP 2B subfamily were detected by RT-PCR in 2 different pools of uninduced FHhTERT after 50 PD (Figure 6A). Similar CYP expression profiles were obtained for later cultures, and CYP 3A4 was also detectable at 35 PD. Over a culture period of 3 days, steady rates of urea production (57.4 ⫾ 4.4 pg 䡠 cell⫺1 䡠 day⫺1; n ⫽ 6) were observed in FH-hTERT at 50 and 55 PD (Figure 6B). Urea production doubled after prolonged confluent culture (2– 4 weeks) of FH-hTERT at 150 PD, indicating the potential of FH-hTERT to maintain metabolic hepatic functions (Figure 6B). In comparison, monolayers of primary rat hepatocytes are known to produce 210 –310 pg urea 䡠 cell⫺1 䡠 day⫺1.34 Oncogenicity of Telomerase-Immortalized Cells Figure 2. Proliferation and replicative senescence. (A) Representative proliferation curves of control FH and FH-hTERT pools were generated from cell counts and plotted as PD vs. days posttransduction. Each point represents the mean of 2 independent cell counts. The immortality threshold was set at twice the number of PD at replicative senescence. (B) Onset of senescence was visualized in subconfluent, nonproliferating control FH cultures (30 –35 PD) by senescence-associated ␤-galactosidase stain (blue). No blue cells were observed in FH-hTERT cultures. Note the large cytoplasm of senescent cells (original magnification 40⫻). (C) DNA synthesis in control FH and FHhTERT was analyzed by BrdU incorporation and flow cytometry. Bars show percentages of cells undergoing DNA synthesis (mean ⫾ SEM). The number of observations at various PD are shown at the top of the bars (*FH-Control 30 PD vs. FH-hTERT 30 PD, P ⬍ 0.001).

Histochemistry revealed glycogen storage and glucose6-phosphatase activity in confluent FH-hTERT at 50 PD, in a pattern similar to primary FH (Figure 4). Next, we analyzed albumin production as a marker for hepatocellular protein synthesis by flow cytometry. At 50 PD, FH-hTERT exhibited a clear increase in green fluorescence following albumin-specific fluorescein staining. The magnitude of albumin expression in these cells was equivalent to HepG2 cells (Figure 5). This experiment did not permit precise quantification of albumin synthesis levels in our cells. However, the findings were significant because albumin synthesis was demonstrated in the majority of FH-hTERT (more than 65%) on a

Cell culture assays were performed to investigate the oncogenicity of FH-hTERT. As expected for untransformed cells, DNA-synthesis rates of FH-hTERT significantly decreased from above 50% to less than 8% at confluency (contact inhibition) and during serum starvation (Table 3). Anchorage-independent growth is also an excellent in vitro indicator of a malignant growth potential.29 Thus,

Figure 3. Hepatocellular gene expression. Passaged FH (passage 1) and FH-hTERT at 50 PD were investigated by RT-PCR and standard agarose gel electrophoresis. Shown are the specific bands of the indicated targets (abbreviations as in Table 1) for representative reactions. RT-PCR reactions were repeated 2– 4 times for verifying results.

February 2003

TELOMERASE–IMMORTALIZED FETAL HEPATOCYTES

439

Figure 4. Glucose-6-phosphatase activity and glycogen storage of primary FH and FH-hTERT at 50 PD. Phase contrast micrographs were taken from primary FH (A) and FH-hTERT (B). Glucose-6-phosphatase activity was visualized in primary FH (C) and FH-hTERT (D) by incubating with glucose-6-phosphate. For glycogen staining, ethanol-fixed primary FH (E) and FH-hTERT (F ) were incubated with 1% aqueous periodic acid and Schiff’s reagent, each followed by washes in 0.5% sodium bisulphite water (original magnification 40⫻).

we performed soft agar colony assays with FH-hTERT and HepG2 cells. After 15 days, no colonies were observed in plates seeded with FH-hTERT (2 different cell pools with 3 plates per seeding density), whereas a seeding cell number-dependent colony formation was confirmed for HepG2 cells with 14 ⫾ 4 colonies for 50 seeded cells (n ⫽ 4), 125 ⫾ 9 colonies for 500 seeded cells (n ⫽ 4), and 830 ⫾ 71 colonies for 5000 seeded cells (n ⫽ 4). None of the mice inoculated subcutaneously with FH-hTERT (2 different cell pools at 120 and 150 PD; 3 mice per cell pool) showed tumor formation within the observation period of 15 weeks. In the positive control group, 3 of 4 mice had detectable tumors after a latency period of 5.1 ⫾ 1.0 weeks reaching tumor diameters of more than 10 mm, 7.7 to 11.6 weeks after 293T cell inoculation. In addition, no intrahepatic tumors were detected in immunodeficient athymic nude

and NOD-SCID mice 10 –12 weeks after FH-hTERT injection (different cell pools at 150 –170 PD; n ⫽ 5 mice). Furthermore, despite more than 160 PD, we observed no up-regulation of c-Myc by Western blot analysis (Figure 7). Densitometry of scanned c-Myc and actin bands revealed no changes for FH-hTERT after 160 PD. An increase in c-Myc was a feature of telomerase-immortalized human mammary epithelial cells cultured for less time by other investigators, raising concerns of potential telomerase immortalization-related genetic changes.35 In Vivo Differentiation of TelomeraseImmortalized Cells To investigate whether telomerase immortalization and long-term culture with extensive proliferation disrupted the ability of FH to respond to the hepatic

440

WEGE ET AL.

GASTROENTEROLOGY Vol. 124, No. 2

traperitoneally transplanted FH-hTERT displayed a hepatocyte-like morphology with glucose-6-phosphatase activity and glycogen synthesis revealing the hepatocellular phenotype of recovered cells (Figure 9A–C). The identity of the cells covering the beads was confirmed to be human by demonstrating the presence of Alu-sx sequences (Figure 9D).

Discussion The present study demonstrates that ectopic expression of the catalytic telomerase subunit hTERT reconstitutes telomerase activity in FH. Moreover, hTERT-transduced FH maintain their telomeres and bypassed replicative senescence. We used fetal cells for our studies because of their growth factor-independent proliferation,36,37 which enabled us to investigate telomere-dependent replicative aging in human hepatocytes without the interference of growth factor-related mechanisms. Our findings reveal for the first time that the limited proliferation in cultured FH is telomere dependent and that telomerase-negative FH enter replicative senescence following a definite number of cell divisions. We believe that the proliferative capacity in FH-hTERT was enhanced by telomerase reconstitution because this inactivated telomere-dependent replicative

Figure 5. Flow cytometric demonstration of albumin synthesis in FH-hTERT at 50 PD. The human embryonic kidney cell line 293T served as a negative control and HepG2 cells as positive control. The representative histograms show green fluorescence intensity in the FL1 channel of the cytometer vs. cell counts for samples incubated with isotype control (bold graph) or albumin antibody (shaded graph).

microenvironment with phenotype-specific maturation, FH-hTERT at 140 –160 PD were transplanted into the livers of immunodeficient mice (n ⫽ 3). In situ hybridization demonstrated engraftment and survival of FHhTERT 4 weeks after transplantation. As expected, transplanted human cells were localized in portal areas as single cells or small cell clusters (Figure 8A). Expression of albumin, ␣1-antitrypsin, and transferrin reached levels that were comparable with primary human hepatocytes (Figure 8B), thus illustrating the ability of FH-hTERT to differentiate into mature hepatocytes and to display significant hepatocellular gene expression. In addition to the intrahepatic transplantation studies, an intraperitoneal bioassay was conducted to recover FH-hTERT from the peritoneal cavity for histochemical analysis (n ⫽ 4). Within 4 weeks after transplantation, vascularized conglomerates developed and FH-hTERT formed confluent masses on the transplanted beads. In-

Figure 6. Evidence of biotransformation activity. (A) CYP gene expression profiles in primary FH and FH-hTERT at 50 PD were obtained by RT-PCR with agarose gel electrophoresis. Shown are the specific bands of the indicated genes. For verifying results, RT-PCR was repeated twice for 2 different cell pools and at various replicative stages. (B) Ureagenesis in FH-hTERT at 50 and 55 PD, and prolonged confluent cultures of FH-hTERT at 150 PD, was quantitated in aliquots of the culture medium. Urea production rates are graphed as pg/cell over a culture period of 3 days. The data points are means ⫾ SEM of triplicate wells. The experiment was repeated twice.

February 2003

TELOMERASE–IMMORTALIZED FETAL HEPATOCYTES

441

Table 3. Serum Dependence and Contact Inhibition Percentage of cells in S-phase Culture condition

FH-Control (24 PD)

FH-hTERT 1 (100 PD)

FH-hTERT 2 (120 PD)

10% FBS, 1 daya 10% FBS, confluentb 0% FBS, 5 daysc

36.79 ⫾ 0.53 (n ⫽ 3) ND ND

52.93 ⫾ 1.76 (n ⫽ 3) 7.83 ⫾ 0.85 (n ⫽ 4)d 6.09 ⫾ 0.74 (n ⫽ 4)d

50.71 ⫾ 2.23 (n ⫽ 4) 6.15 ⫾ 1.13 (n ⫽ 4)d 5.43 ⫾ 0.54 (n ⫽ 4)d

NOTE. Values are means ⫾ SEM. Data for 2 different FH-hTERT pools. FBS, fetal bovine serum; ND, not determined; PD, population doubling. aOvernight semiconfluent culture (1 day) in medium containing 10% FBS. bProlonged confluent culture (15 days) in medium containing 10% FBS. cShort-term culture (5 days) in medium without FBS. dSignificant difference from 10% FBS, 1 day (P ⬍ 0.001).

aging, resulting in an extended proliferative lifespan of cells in culture. Therefore, our studies verify the significance of telomere stabilization for immortalizing human hepatocytes. It is generally thought that 25–50 billion hepatocytes, corresponding to 10%–20% of the liver mass, are necessary to sustain human life.38 Considering the large quantities of cells required for hepatocyte-based therapies, using cells from the fetal human liver should be an intriguing possibility because these cells exhibit significant proliferative capacity with the ability to differentiate into mature hepatocytes. Previous studies have showed that fetal rat hepatocytes generated mature hepatocytes in the hepatic microenvironment39,40 and that hepatocellular functions were induced in mouse embryonic hepatoblasts when supplementing the culture medium with differentiating signals, such as sodium butyrate or dimethyl sulfoxide.41 Similarly, signals emanating from extracellular matrix of rat adult hepatocytes

Figure 7. Western blot analysis of c-Myc in passaged FH and FHhTERT at different PD. Using Western blot analysis, c-Myc content was evaluated using passaged FH (Pass. FH) as calibrator and HeLa extract as a positive control. Shown are representative bands and their size in kilodaltons together with the densitometric values normalized to actin and relative to Pass. FH. The experiment was repeated twice to confirm the results.

increased differentiated liver functions in rat fetal hepatocytes.42 Hepatocytes from fetal human livers were used in early studies in patients with fulminant hepatic failure.43 In this context, the continued expression of hepatic growth factors and growth factor receptors in our telomerase-immortalized FH suggests that FH-hTERT maintain responsiveness to paracrine signals capable of inducing cell differentiation. This conclusion is further supported by the uninterrupted expression of HNF-4, a major transcription factor, which is required for hepatocyte differentiation and liver gene expression.44 HNF-4 acts upstream in a transcription factor cascade, including HNF-1␣, that drives hepatocyte differentiation.44 Maintenance of hepatocellular functions was demonstrated in our cells by glycogen storage, glucose-6-phosphatase activity, and albumin synthesis, especially in confluent cells. For potential applications in liver-directed cell therapies, biotransformation and detoxification activities are of considerable importance. Despite extensive proliferation in our culture conditions, which were not optimized for drug metabolism and ammonia detoxification, FH-hTERT expressed CYP genes and synthesized urea, which are characteristic properties of mature hepatocytes.34 Transplantation studies were conducted to elucidate the in vivo maturation potential of telomerase-immortalized FH. Intrahepatic transplantation demonstrated integration and hepatocyte-specific differentiation of FH-hTERT in the hepatic parenchyma, signifying that FH-hTERT will potentially be suitable for cell transplantation and liver repopulation. Our findings support previous reports that telomerase-immortalized cells retain their ability to display phenotypic functions in vivo45,46 and that cells from the fetal human liver have the ability to differentiate in the liver microenvironment.13,40 An infinite supply of immortalized human fetal liver cells with physiologic in vivo maturation potential provides opportunities to study hepatocyte maturation.

442

WEGE ET AL.

GASTROENTEROLOGY Vol. 124, No. 2

Figure 8. Intrahepatic engraftment and function of transplanted cells. FH-hTERT cells were transplanted after 140 –160 PD into the liver of NOD-SCID mice. (A) Transplanted cells were identified 4 weeks later with a human centromere probe showing red signals (arrows). Insets show higher magnification views of transplanted cells seen adjacent to portal areas (Pa). Tissue sections were lightly counterstained with toluidine blue (original magnification, 400⫻; insets 1000⫻). (B) Transplanted FHhTERT cells showing hepatocellular gene expression with realtime quantitative PCR using human-specific primer-probe sets. The level of gene expression is shown relative to that in cultured primary human hepatocytes (mean ⫾ SEM). Replicate RNA extracts (n ⫽ 3) were analyzed from 3 NOD-SCID mouse recipients (SCID 1, SCID 2, SCID 3).

Figure 9. Intraperitoneal bioassay. FH-hTERT at 140 –160 PD were attached to microcarrier beads, transplanted into the peritoneal cavity of NOD-SCID mice, and recovered after 4 weeks. (A) Recovered microcarrier (mc) conglomerates contained cells with hepatocyte-like morphology (H & E stain). Histochemistry demonstrated presence of glucose-6-phosphatase activity (B) as well as glycogen storage (C) (original magnification, 400⫻). (D) Genomic DNA was extracted from multiple microcarrier beads and subjected to Alu-sx PCR to verify presence of transplanted human cells (lane 1, negative control; lanes 2–7, recovered microcarrier beads).

February 2003

We have not found any evidence of oncogenicity in FH following telomerase reconstitution. FH-hTERT proliferation was serum and anchorage dependent and was inhibited by cell-cell contact. McCullough et al.47 showed that the liver microenvironment represses the expansion of tumorigenic liver-derived cells. Therefore, absence of tumor formation in heterologous subcutaneous sites provided strong evidence in favor of an untransformed phenotype in our FH-hTERT. Furthermore, tumors were not formed following transplantation of FH-hTERT into livers. Untransformed, immortalized FH with hepatocyte differentiation potential that respond appropriately to proliferative stimuli should be an ideal cell source for liver-directed cell therapies, especially if immortalization is not associated with oncogene up-regulation. Recently, Wang et al.35 reported that c-Myc expression was up-regulated in hTERT-immortalized human mammary epithelial cells between 107 and 135 PD. These investigators concluded that extension of lifespan by telomerase reconstitution might select for c-Myc overexpressing cells. In our study, telomerase reconstitution did not result in c-Myc up-regulation after 160 PD. Thus, we believe that c-Myc overexpression should neither be a requirement nor an invariant consequence of hTERT-transduction. In contrast, our findings indicate that telomerase reconstitution does not induce a transformed phenotype and is likely to be genoprotective throughout the expanded proliferative lifespan of immortalized cells.18,19 The in vitro proliferative capacity of mature human hepatocytes was enhanced following the introduction of simian virus 40 large T antigen.48 Large T antigen neutralizes essential negative cell cycle check points that are frequently inactivated in tumor cells, for instance p53 and retinoblastoma protein.49 Because it is uncertain whether inactivation of p53 and retinoblastoma protein would increase the risk of malignant cell transformation, Kobayashi et al.48 engineered LoxP sites into the transgene to permit removal of the large T antigen before cell transplantation. Nevertheless, somatic cells expressing the large T antigen continue to lose telomeric repeats with each cell doubling and eventually enter crisis leading to extensive cell death and selection of further transformed cells.25,49 Thus, large T-antigen expression by itself does not result in an immortalized phenotype and might activate a variety of deleterious cellular events during cell immortalization. On the other hand, telomerase reconstitution in human fetal cells should represent a more physiologic approach, although additional studies will obviously be necessary.

TELOMERASE–IMMORTALIZED FETAL HEPATOCYTES

443

In conclusion, we established that hepatocytes from the fetal human liver are amenable to hTERT-mediated telomerase reconstitution and immortalization without deleterious consequences, such as inducing a transformed phenotype or disrupting the ability to differentiate in vivo. Telomerase-immortalized hepatocytes could potentially play roles in liver-directed cell therapies and provide a novel system for studying basic biologic mechanisms, including in vitro and in vivo maturation of progenitor cells, cell-cycle regulation, and others. Such telomerase-reconstituted immortalized cells may also serve as model systems for studies of liver cell injury, hepatitis virus replication, and pharmacologic investigations.

References 1. Fox IJ, Roy Chowdhury J, Kaufman SS, Goertzen TC, Roy Chowdhury N, Warkentin PI, Dorko K, Sauter BV, Strom SC. Treatment of the Crigler-Najjar syndrome type I with hepatocyte transplantation. N Engl J Med 1998;338:1422–1426. 2. Ellis AJ, Hughes RD, Wendon JA, Dunne J, Langley PG, Kelly JH, Gislason GT, Sussman NL, Williams R. Pilot-controlled trial of the Extracorporeal Liver Assist Device in acute liver failure. Hepatology 1996;24:1446 –1451. 3. Watanabe FD, Mullon CJP, Hewitt WR, Arkadopoulos N, Kahaku E, Eguchi S, Khalili T, Arnaout W, Shackelton CR, Rozga J, Solomon B, Demetriou AA. Clinical experience with a bioartificial liver in the treatment of severe liver failure. Ann Surg 1997;225:484 – 494. 4. Nyberg SL, Remmel RP, Mann HJ, Peshwa MV, Hu WS, Cerra FB. Primary hepatocytes outperform Hep G2 cells as the source of biotransformation functions in a bioartificial liver. Ann Surg 1994; 220:59 – 67. 5. Patience C, Takeuchi Y, Weiss RA. Infection of human cells by an endogenous retrovirus of pigs. Nat Med 1997;3:282–286. 6. Nyberg SL, Hibbs JR, Germer JJ, Persing DH. Transfer of porcine endogenous retrovirus across hollow fiber membranes: significance to a bioartificial liver. Transplantation 1999;67: 1251–1255. 7. Collins K. Mammalian telomeres and telomerase. Curr Opin Cell Biol 2000;12:378 –383. 8. Allsopp RC, Vaziri H, Patterson C, Goldstein S, Younglai EV, Futcher AB, Greider CW, Harley CB. Telomere length predicts replicative capacity of human fibroblasts. Proc Natl Acad Sci U S A 1992;89:10114 –10118. 9. Campisi J. The biology of replicative senescence. Eur J Cancer 1997;33:703–709. 10. Harley CB. Telomere loss: mitotic clock or genetic time bomb? Mutat Res 1991;256:271–282. 11. Kim NW, Piatyszek MA, Prowse KR, Harley CB, West MD, Ho PLC, Coviello GM, Wright WE, Weinrich SL, Shay JW. Specific association of human telomerase activity with immortal cells and cancer. Science 1994;266:2011–2015. 12. Avilion AA, Piatyszek MA, Gupta J, Shay JW, Bacchetti S, Greider CW. Human telomerase RNA and telomerase activity in immortal cell lines and tumor tissue. Cancer Res 1996;56:645– 650. 13. Malhi H, Irani AN, Gagandeep S, Gupta S. Isolation of human progenitor liver epithelial cells with extensive replication capacity and differentiation into mature hepatocytes. J Cell Sci 2002;115: 2679 –2688. 14. Bodnar AG, Ouellette M, Frolkis M, Holt SE, Chiu CP, Morin GB, Harley CB, Shay JW, Lichtsteiner S, Wright WE. Extension of life-span by introduction of telomerase into normal human cells. Science 1998;279:349 –352.

444

WEGE ET AL.

15. Wang J, Xie LY, Allan S, Beach D, Hannon GJ. Myc activates telomerase. Genes Dev 1998;12:1769 –1774. 16. Yang J, Chang E, Cherry AM, Bangs CD, Oei Y, Bodnar A, Bronstein A, Chiu CP, Herron GS. Human endothelial cell life extension by telomerase expression. J Biol Chem 1999;274:26141–26148. 17. Dickson MA, Hahn WC, Ino Y, Ronfard V, Wu JY, Weinberg RA, Louis DN, Li FP, Rheinwald JG. Human keratinocytes that express hTERT and also bypass a p16INK4a-enforced mechanism that limits life span become immortal yet retain normal growth and differentiation characteristics. Mol Cell Biol 2000;20:1436–1447. 18. Jiang XR, Jimenez G, Chang E, Frolkis M, Kusler B, Sage M, Beeche M, Bodnar AG, Wahl GM, Tlsty TD, Chiu CP. Telomerase expression in human somatic cells does not induce changes associated with a transformed phenotype. Nat Genet 1999;21:111–114. 19. Morales CP, Holt SE, Ouellette M, Kaur KJ, Yan Y, Wilson KS, White MA, Wright WE, Shay JW. Absence of cancer-associated changes in human fibroblasts immortalized with telomerase. Nat Genet 1999;21:115–118. 20. Naviaux RK, Costanzi E, Haas M, Verma IM. The pCL vector system: rapid production of helper-free, high titer, recombinant retroviruses. J Virol 1996;70:5701–5705. 21. Counter CM, Hahn WC, Wei W, Dickinson MA, Caddle S, Beijersbergen RL, Lansdorp PM, Sedivy JM, Weinberg RA. Dissociation among in vivo telomerase activity, telomere maintenance, and cellular immortalization. Proc Natl Acad Sci U S A 1998;95: 14723–14728. 22. Kim NW, Wu F. Advances in quantification and characterization of telomerase activity by the telomeric repeat amplification protocol (TRAP). Nucleic Acids Res 1997;25:2595–2597. 23. Nichols WS, Schneider S, Chan RCK, Farthing CF, Daar ES. Increased CD4⫹ T-lymphocyte senescence fraction in advanced human immunodeficiency virus type 1 infection. Scand J Immunol 1999;49:302–306. 24. Vaziri H, Scha¨chter F, Uchida I, Wei L, Zhu X, Effros R, Cohen D, Harley CB. Loss of telomeric DNA during aging of normal and trisomy 21 human lymphocytes. Am J Hum Genet 1993;52:661– 667. 25. Counter CM, Avilion AA, LeFeuvre CE, Stewart NG, Greider CW, Harley CB, Bacchetti S. Telomere shortening associated with chromosome instability is arrested in immortal cells which express telomerase activity. EMBO J 1992;11:1921–1929. 26. Vaziri H, Benchimol S. Reconstitution of telomerase activity in normal human cells leads to elongation of telomeres and extended replicative life span. Curr Biol 1998;8:279 –282. 27. Dimri GP, Lee X, Basile G, Acosta M, Scott G, Roskelley C, Medranos EF, Linskens M, Rubelj I, Pereira-Smith O, Peacocke M, Campisi J. A biomarker that identifies senescent human cells in culture and aging skin in vitro. Proc Natl Acad Sci U S A 1995;92:9363–9367. 28. Ott M, Rajvanshi P, Sokhi R, Alpini G, Aragona E, Dabeva M, Shafritz DA, Gupta S. Differentiation-specific regulation of transgene expression in a diploid epithelial cell line derived from the normal F344 rat liver. J Pathol 1999;187:365–373. 29. Clark GJ, Cox AD, Graham SM, Der CJ. Biological assays for Ras transformation. Methods Enzymol 1995;255:395– 412. 30. Ostrowska A, Karrer FM, Bilir BM. Histological identification of purified and cryopreserved allogenic hepatocytes following transplantation in a murine model without host immunosuppression. Transpl Int 1999;12:188 –194. 31. Liu H, Lo CR, Jones BE, Pradhan Z, Srinivasan A, Valentino KL, Stockert RJ, Czaja MJ. Inhibition of c-Myc expression sensitizes hepatocytes to tumor necrosis factor-induced apoptosis and necrosis. J Biol Chem 2000;51:40155– 40162. 32. Cho JJ, Malhi H, Wang RY, Joseph B, Ludlow J, Susick R, Gupta S. Enzymatically labeled chromosomal probes for in situ localization of human cells xenografted into mice. Nat Med 2002;8: 1033–1036. 33. Bucher NLR, Farmer SR. Liver regeneration following partial hepa-

GASTROENTEROLOGY Vol. 124, No. 2

34.

35. 36.

37.

38.

39.

40.

41. 42.

43.

44.

45.

46.

47.

48.

49.

tectomy: genes and metabolism. In: Strain AJ, Diehl AM, eds. Liver growth and repair. 1st ed. London: Chapman & Hall, 1998:3–27. Schwartz RE, Reyes M, Koodie L, Jiang Y, Blackstad M, Lund T, Lenvik T, Johnson S, Hu WS, Verfaille CM. Multipotent adult progenitor cells from bone marrow differentiate into functional hepatocyte-like cells. J Clin Invest 2002;109:1291–1302. Wang J, Hannon GJ, Beach DH. Risky immortalization by telomerase. Nature 2000;405:755–756. Lilja H, Blanc P, Demetriou AA, Rozga J. Response of cultured fetal and adult rat hepatocytes to growth factors and cyclosporine. Cell Transplant 1998;7:257–266. Lilja H, Arkadopoulos N, Blanc P, Eguchi S, Middleton Y, Meurling S, Demetriou AA, Rozga J. Fetal rat hepatocytes: isolation, characterization, and transplantation in the Nagase analbuminemic rats. Transplantation 1997;64:1240 –1248. Sussman NL, Gislason GT, Kelly JH. Extracorporeal liver support. Application to fulminant hepatic failure. J Clin Gastroenterol 1994;18:320 –324. Sigal S, Rajvanshi P, Reid LM, Gupta S. Demonstration of differentiation in hepatocyte progenitor cells using dipeptidyl peptidase IV deficient mutant rats. Cell Mol Biol Res 1995;41:39 – 47. Dabeva MD, Petkov PM, Sandhu J, Oren R, Laconi E, Hurston E, Shafritz DA. Proliferation and differentiation of fetal liver epithelial progenitor cells after transplantation into adult rat liver. Am J Pathol 2000;156:2017–2031. Rogler LE. Selective bipotential differentiation of mouse embryonic hepatoblasts in vitro. Am J Pathol 1997;150:591– 602. Brill S, Zvibel I, Halpern Z, Oren R. The role of fetal and adult hepatocyte extracellular matrix in the regulation of tissue-specific gene expression in fetal and adult hepatocytes. Eur J Cell Biol 2002;81:43–50. Habibullah CM, Syed IH, Qamar A, Taher-Uz Z. Human fetal hepatocyte transplantation in patients with fulminant hepatic failure. Transplantation 1994;58:951–977. Li J, Ning G, Duncan SA. Mammalian hepatocyte differentiation requires the transcription factor HNF-4␣. Genes Dev 2000;14: 464 – 474. Yang J, Nagavarapu U, Relloma K, Sjaastad MD, Moss WC, Passaniti A, Herron GS. Telomerized human microvasculature is functional in vivo. Nat Biotechnol 2001;19:219 –224. Thomas M, Yang L, Hornsby PJ. Formation of functional tissue from transplanted adrenocortical cells expressing telomerase reverse transcriptase. Nat Biotechnol 2000;18:39 – 42. McCullough KD, Coleman WB, Ricketts SL, Wilson JW, Smith GJ, Grisham JW. Plasticity of the neoplastic phenotype in vivo is regulated by epigenetic factors. Proc Natl Acad Sci U S A 1998; 95:15333–15338. Kobayashi N, Fujiwara T, Westerman KA, Inoue Y, Sakaguchi M, Noguchi H, Miyazaki M, Cai J, Tanaka N, Fox IJ, Leboulch P. Prevention of acute liver failure in rats with reversibly immortalized human hepatocytes. Science 2000;287:1258 –1262. Ozer HL. SV 40-mediated immortalization. Prog Mol Subcell Biol 2000;24121–24153.

Received April 17, 2002. Accepted October 17, 2002. Address requests for reprints to: Mark A. Zern, M.D., University of California, Davis Medical Center, 4635 Second Avenue, Suite 1001, Sacramento, California 95817. e-mail: [email protected]; fax: (916) 734-8097. Supported by grants from the Studienstiftung des deutschen Volkes, Germany (to H.W.), NIH grants R01 DK46952 and P30-DK-41296 (to S.G.), a grant from the Alpha One Foundation (to M.A.Z.), and a grant from Transplant Hope, University of California, Davis Medical Center (to M.A.Z.). H.W. received a research award from the American Liver Foundation. J.W. is a recipient of the Liver Scholar Award by the American Liver Foundation.