Cell Biology International (2006) 30, 992-998 - T. Tokiwa and others - Differentiation potential of an immortalized non-tumorigenic human liver epithelial cell line as liver progenitor cells

Brought to you by Portland Press Ltd.

Published on behalf of the International Federation for Cell Biology

Download to Citation Manager

Cell Biology International (2006) 30, 992–998 (Printed in Great Britain)

Abstract

PDF

Citing articles

Differentiation potential of an immortalized non-tumorigenic human liver epithelial cell line as liver progenitor cells

T. Tokiwa^a^*, T. Yamazaki^a, W. Xin^a, N. Sugae^b, M. Noguchi^b, S. Enosawa^c and T. Tsukiyama^d

^aDepartment of Liver Cell Biology, Kohno Clinical Medicine Research Institute, 3-4-4 Kitashinagawa, Shinagawa, 140-0001 Tokyo, Japan
^bDepartment of Pathology, Institute of Basic Medical Science, University of Tsukuba, Ibaraki, Japan
^cDepartment of Regeneration Surgery, Research Institute, National Center for Child Health and Development, Tokyo, Japan
^dKitashinagawa Hospital, Kohno Clinical Medicine Research Institute, Tokyo, Japan

Abstract

We report the differentiation potential of an immortalized non-tumorigenic human liver epithelial cell line, THLE-5b. Under basic culture conditions THLE-5b showed undifferentiated phenotypes. When grown as cell aggregates, THLE-5b exhibited a hepatocyte-like ultrastructure, ammonia metabolic activity and several other indicators that suggest hepatocytic maturation, including up-regulation or induction of liver-specific genes such as albumin and tryptophane 2,3-dioxygenase, and down-regulation of biliary cell markers such as gamma-glutamyl transpeptidase (GGT). Under these conditions, transcriptional factors such as HNF-1 and HNF-4α were also up-regulated or induced. In Matrigel culture, expression of GGT was up-regulated. THLE-5b expressed both albumin and α 1-antitrypsin, but lost expression of CK19 in severe combined immunodeficient mice. Thus, THLE-5b can be aligned with progenitor cells, which are committed to the hepatocytic or biliary epithelial cell lineage. These results imply that bipotent progenitor cell populations similar to THLE-5b cells may exist in adult human liver.

Keywords: Differentiation, Immortalized human cell lines, Liver progenitor cells.

^*Corresponding author. Tel./fax: +81 3 3474 1833.

1 Introduction

Following partial hepatoectomy in adult rat liver, restoration of the lobular parenchyma is due to the rapid growth of remaining hepatocytes. On the other hand, when hepatocytes are unable to grow after injury, a progenitor cell compartment arises and proliferates in the liver; these cells are termed oval cells, which have the potential to differentiate into hepatocytic or biliary cell lineage (Thorgeisson, 1996). Isolation and proliferation of oval cell-like liver progenitor cells (LPC) has been achieved in rodent livers. On the other hand, whether or not LPCs exist in adult human livers is controversial, because isolation, identification and proliferation of human LPCs is difficult (Allain et al., 2002; Selden et al., 2003). Bipotent LPCs or LPC lines could be useful in preclinical applications as well as for the study of the biology of LPCs. An SV40 T antigen-immortalized, non-tumorigenic CK18 positive liver cell line, THLE-5b, was developed from an adult human liver (Lechner et al., 1990). Our preliminary studies show that THLE-5b cells are morphologically and functionally immature, suggesting their potential to differentiate into an hepatocyte phenotype (Tokiwa et al., 1998). In this study we investigated the possibility that THLE-5b cells have the potential to differentiate into hepatic lineage cells in vitro and in vivo.

2 Materials and methods

2.1 Cells

THLE-5b cells, an SV40 large T antigen-immortalized human liver cell line (Lechner et al., 1990), were CK18-positive, but did not show typical epithelial morphology (Fig. 3A). THLE-5b cells grew with about 0.62 population doubling/day. THLE-5b cells were plated in soft agar (10⁵cells/dish) to determine whether the cells were anchorage-dependent, and failed to grow after 4weeks (Tokiwa, personal communication). For tumorigenicity test, 5 million cells were injected subcutaneously into Balb/c nu/nu mice, and no tumors were found after 6months (Lechner et al., 1990). These data suggest that THLE-5b cells are transformed only weakly. THLE-5b cells are certified as being the designated type, free of mycoplasma contamination, and have been used from young stock. Cultures were grown in culture medium RPMI 1640 (Invitrogen Corp., Carlsbad, CA, USA) supplemented with 10% fetal calf serum (FCS, Invitrogen), and antibiotics in a humidified atmosphere of 5% CO2 and 95% air at 37°C.

2.2 Aggregate formation

After reaching confluence, THLE-5b cells were trypsinized and the obtained cell suspensions were filtered through a 150μm nylon filter. Approximately 5×10⁵cells/flask were then inoculated into 25-ml Meyer's flasks containing 3ml of culture medium DMEM/F12 (Invitrogen). Meyer's flasks were precoated with 2.5% poly (2-hydroxyethyl methacrylate) (poly-HEMA, Sigma Chemical Co., St. Louis, MO, USA) in 95% ethanol as described previously (Landry et al., 1985), or with silicon solution (Sigma) as substrates. Cell aggregate formation was carried out under the following conditions: (i) static culture on poly-HEMA substrate (aggregate culture-1 or Agg-1); (ii) rotation culture, in which the Meyer's flasks were rotated at 70rpm on an orbital shaker in a chamber, on poly-HEMA substrate (aggregate culture-2 or Agg-2); and (iii) rotation culture on silicon substrate (aggregate culture-3 or Agg-3). The culture was maintained in a humidified atmosphere of 5% CO2 and 95% air at 37°C. Cell aggregates were removed from the flasks at different time points and prepared for standard histology (24h), immunocytochemistry (24h), transmission electron microscopy (6h, 12h, and 21h) or RT-PCR (24h and 72h) as described below. For the induction study, 3-methylcholanthrene (MCA, Sigma) was added to the culture at 2μmol/L final concentration in dimethylsulfoxide (DMSO; 0.5%) for 48h.

2.3 Matrigel culture

0.05ml of Matrigel (BD Bioscience, San Jose, CA, USA) was placed onto 96-well plates, left to set for 1h, and 6×10⁴cells/well were placed in DMEM/F12 supplemented with 10ng/ml hepatocyte growth factor (HGF). Cells in monolayer culture were also grown in 10ng/ml HGF-containing DMEM/F12 at a density of 6×10⁴cells/well. Cells were recovered after 2 to 5days for RNA extraction.

2.4 Immunocytochemistry

Cells in monolayers were fixed in acetone/methanol (1:1) and cell aggregates were fixed in a 10% buffered formalin solution. The aggregates were then dehydrated, cleared in xylene, and embedded in paraffin. Sections were prepared for immunocytochemistry. The following primary antibodies in this study were used: rabbit anti-human albumin (DAKO, Glostrup, Denmark), rabbit anti-human α1-antitrypsin (Ylem, S.R.L., Roma, Italy), rabbit anti-human transferrin (DAKO), mouse anti-human CK19 (Progen, Biotechnik GmbH, Heidelberg, Germany), mouse anti-human CK18 (Progen), rabbit anti-human α-fetoprotein (AFP, NeoMarkers, Inc., Fremont, CA, USA), mouse anti-human CK14 (Chemicon international, Temecula, CA, USA) and mouse anti-vimentin (DAKO). We used universal quick kit (Vector laboratories, Inc., Burlingame, CA, USA) with diaminobenzidine reagent set (Kirkegaard & Perry Laboratories, Inc., Gaithersburg, MD, USA), or DAKO LSAB2 system, Peroxidase (DAKO) for localization of horseradish peroxidase-labeled reporter reagents.

2.5 RNA extraction, reverse transcription and polymerase chain reaction (RT-PCR)

Total RNAs were isolated as described by Chomczynski and Sacchi (1987). The cDNAs were synthesized by first strand cDNA kit (ReverTraAce-α, Toyobo Co., LTD, Osaka, Japan). PCR amplification was performed with a 5μl aliquot of reverse transcriptase reaction product [except to amplify the CK19 (1μl of 1st round product)], in a total volume of 50μl and in the presence of 0.2mM dNTP, 0.25μM primers and 1.25U Taq DNA polymerase (MBI Fermentas) and using a 25 to 40 cycle program (CK19: 20 cycles for 1st round, 15–17 cycles for 2nd round) consisting of denaturation at 94°C for 1min, annealing 55°C for 1min, and extension at 72°C for 2min. The PCR primers used for amplification were shown in Table 1.

Table 1.

Oligonucleotides used as PCR primers

Target cDNA	Sense	Antisense	bp
Albumin	5′-agcggcacagcacttctctaga-3′	5′-tccacacggaatgctgccatgg-3′	368
AFP	5′-accaaagttaattttactgaaat-3′	5′-gtttgtcctcactgagttggca-3′	644
Transferrin	5′-tctatgggtcaaaagaggatcc-3′	5′-gagagctgaatagttggaattc-3′	612
α1-antitrypsin	5′-tctgtttcttggcctcttcggtg-3′	5′-cgtaacagctaagtgacagggtc-3′	560
CYP1A1	5′-tccagagacaacaggtaa-3′	5′-aggaagggcagaggaatgtgat-3′	371
CYP1A2	5′-ggaagagaaacaagggctgact-3′	5′-aggctatggcgtggtattca-3′	453
CYP3A4	5′-ccaagctatgctcttcaccg-3′	5′-tcaggctccacttacggtgc-3′	378
TO	5′-gggaactacctgcatttgga-3′	5′-gtgcatccgagaaacaacct -3′	222
E-cadherin	5′-tgcccagaaaatgaaaaagg-3′	5′-gtgtatgtggcaatgcgttc-3′	200
GGT	5′-gcccagattgtcaaggacat-3′	5′-ccctttgaggatgttgagga-3′	183
Thy-1	5′-ctagtggaccagagccttcg-3′	5′-tggagtgcacacgtgtaggt-3′	236
c-kit	5′-gtgaatctacttggagcctg-3′	5′-tcattcttgatgtctctggc-3′	500
CK19 outer	5′-aagctaaccatgcagaacctcaacgaccga-3′	5′-ttattggcaggtcaggagaagagcc-3′	1069
CK19 nested	5′-tcccgcgactacagccactactacacgacc-3′	5′-cgcgacttgatgtccatgagccgctggtac-3′	745
HNF-1	5′-ttctaagctgagccagctgcagacg-3′	5′-gctgaggttctccggctctttcaga-3′	275
HNF-4α	5′-tccagcagatccagttcatca-3′	5′-agagagcacctggtgataccgt-3′	372
HNF-6	5′-aaaccaaagacttagctcacctgca-3′	5′-gctagacgagcaaaatcacactcc-3′	387
G3PDH	5′-accacagtccatgccatcac-3′	5′-tccaccaccctgttgctgta-3′	450

2.6 Ammonia metabolic activity

To determine ammonia metabolic activity, ammonium chloride was added to the culture to a final concentration of 0.5mM and the residual ammonia was measured after 6h using a blood ammonia analyzer (Amicheck meter; Daiichi chemical Co. Ltd., Kyoto, Japan).

2.7 Transmission electron microscope (TEM)

Cells in monolayers and aggregates were fixed with 2% glutaraldehyde in 0.1M phosphate buffer (pH 7.2) 6h followed by postfixing with 1% OsO4, dehydrated, graded ethanol, and embedded in Epon 812. Ultrathin sections were prepared by a Bromma 2088 ultratome V (LKB, Schweden, Germany), stained with uranyl acetate and lead citrate. Observation was carried out with a Hitachi H-7000 electron microscope, as described previously (Kano et al., 2000).

2.8 Cell transplantation

Cells were attached to microcarrier beads (Cytodex 3™, Amersham Pharmacia) by incubating 10⁷cells per 1ml of swollen beads overnight at 37°C. These cells were injected intraperitoneally into severe combined immunodeficient mice in the Balb/c background (SCID) as described elsewhere (Malhi et al., 2002). 21days after transplantation, the mass of cells adjacent to beads was submitted to immunocytochemistry testing as mentioned above.

3 Results

3.1 Phenotypic properties of THLE-5b cells under basic culture conditions

By RT-PCR analysis, THLE-5b cells grown as a monolayer expressed hepatocytic genes tryptophane 2,3-dioxygenase (TO), E-cadherin (Fig. 1), α1-antitrypsin (Fig. 2A) and expressed bile duct/oval cell markers gamma-glutamyl transpeptidase (GGT) and CK19 (Fig. 2B). The cells also expressed oval cell/hematopoietic markers Thy-1 and c-kit (Fig. 2C). In addition, THLE-5b cells expressed transcriptional factor HNF-6, but did not express HNF-1 and HNF-4α (Fig. 2D). Immunocytochemical staining analysis revealed that THLE-5b cells expressed CK18 (Fig. 3A), CK19 (Fig. 3F) and CK14 (Fig. 3H), but did not express albumin (Fig. 3B), α1-antitrypsin (Fig. 3D), transferrin, vimentin, and AFP (data not shown). CK18, CK19 and CK14 were positive in 80–90, 30–35, and 25–30%, respectively, of the cells. The undifferentiated phenotype of THLE-5b cells is reminiscent of LPCs.

Fig. 1

RT-PCR analysis of THLE-5b cells grown as aggregates-comparison of culture conditions- From left lane to right lane: cells grown as a confluent monolayer (Mono); cell aggregates in aggregate culture-1 (Agg-1); cell aggregates in aggregate culture-2 (Agg-2); cell aggregates in aggregate culture-3 (Agg-3).

Fig. 2

RT-PCR analysis of THLE-5b cells grown as aggregates. (A) Expression of liver-specific genes at 24h and 72h. From left lane to right lane: cells grown as a confluent monolayer (Mono); cell aggregates in aggregate culture-1 (Agg-1); cell aggregates in aggregate culture-2 (Agg-2). (B) Expression of biliary cell markers at 72h. Left lane: cells grown as a confluent monolayer (Mono). Right lane: cell aggregates in aggregate culture-2 (Agg-2). (C) Expression of oval cell/hematopoietic markers at 72h. Left lane: cells grown as a confluent monolayer (Mono). Right lane: cell aggregates in aggregate culture-2 (Agg-2). (D) Expression of liver-enriched transcription factors at 72h. From left lane to right lane: cells grown as a confluent monolayer (Mono); cell aggregates in aggregate culture-1 (Agg-1); cell aggregates in aggregate culture-2 (Agg-2).

Fig. 3

Immunocytochemistry of THLE-5b cells grown as a monolayer or cell aggregates. (A,B,D,F,H) cells grown as a monolayer; (C,E,G,I) cell aggregates in aggregate culture-2 (Agg-2); (A) CK18; (B,C) albumin; (D,E,) α1-antitrypsin; (F,G) CK19; (H,I) CK14. Cells are not counterstained. Scale bar, 25μm.

3.2 Culture as cell aggregates

3.2.1 Formation of aggregates

THLE-5b cells formed cell aggregates within 1–2days of inoculation. The aggregation pattern in aggregate culture-1 differed substantially from that in aggregate culture-2 and -3 (data not shown). In aggregate culture-1, smaller aggregates were formed than those formed in aggregate culture-2 and -3. THLE-5b cells formed compact aggregates in both aggregate culture-2 and -3.

3.2.2 Gene expression

As shown in Fig. 1, expression of CYP3A4, TO and E-cadherin mRNA of THLE-5b cells in aggregate culture-2 (Agg-2) and/or -3 (Agg-3) was up-regulated or induced at 24 and/or 72h, whereas that of TO and E-cadherin mRNA was slightly up-regulated in aggregate culture-1 (Agg-1) only at 72h. The level of expression of these genes was considerably higher in aggregate culture-2 and -3 than in aggregate culture-1. However, there was no appreciable difference in gene expression pattern between aggregate culture-2 and -3. In the following experiments, therefore, a comparison was made among monolayer culture, aggregate culture-1 and aggregate culture-2 or between monolayer culture and aggregate culture-2 in gene expression.

Expression of albumin and transferrin mRNA was induced in aggregate culture-1 as well as in aggregate culture-2, although that of α1-antitrypsin and CYP1A1 mRNA was up-regulated only in aggregate culture-2 (Fig. 2A). In addition, CYP1A1 expression was found to be inducible with MCA (data not shown). No expression of AFP (Fig. 2A) and CYP1A2 (data not shown) mRNA was found in either of the culture conditions examined. Taken together, these results indicate that hepatocytic differentiation of THLE-5b cells with a few exceptions is promoted by cell aggregation. In aggregate culture-2, expression of GGT mRNA was down-regulated, although the level of CK19 mRNA was only slightly down-regulated (Fig. 2B). Under these conditions, expression of c-kit mRNA was down-regulated, while the level of Thy-1 mRNA remained unchanged (Fig. 2C). On the other hand, expression of HNF-1, HNF-4α and HNF-6 mRNA was up-regulated or induced in aggregate culture-1 and/or aggregate culture-2 (Fig. 2D). The data shown in Figs. 2A–D were obtained at 72h of cell aggregation.

3.2.3 Immunocytochemistry

The results of immunocytochemistry of the aggregates are shown in Fig. 3. The predominant cells expressed albumin (Fig. 3C) and α1-antitrypsin (Fig. 3E) in the aggregates. Meanwhile, CK19-positive cells were rarely found (Fig. 3G) and only a few cells were positive for CK14 (Fig. 3I).

3.2.4 Ammonia-metabolic activity

Ammonia-metabolic activity of THLE-5b cells was higher in aggregate culture-2 (Agg-2) and -3 (Agg-3) than in aggregate culture-1 (Agg-1), as shown in Fig. 4. No activity was found in THLE-5b cells grown as a confluent monolayer (Mono).

Fig. 4

Ammonia-metabolic activity of THLE-5b cells in aggregate culture. From left lane to right lane: cells grown as a confluent monolayer (Mono); cell aggregates in aggregate culture-1 (Agg-1); cell aggregates in aggregate culture-2 (Agg-2); cell aggregates in aggregate culture-3 (Agg-3). Mean±S.D. (n=3).

3.2.5 Transmission electron microscopical analysis

As shown in Figs. 5B–D, ultrastructurally, bile canaliculi-like structures with microvilli were delimited by tight junctions when THLE-5b cells were grown as aggregate culture-2 (Agg-2). These features were noticed within 21h of culture. Under the same culture conditions, the cells were rich in endoplasmic reticulum and ribosomes. Similar results to this were obtained in aggregate culture-3 (data not shown). On the other hand, in aggregate culture-1 (Agg-1) (Fig. 5A) and monolayer culture (Mono) (data not shown), no bile canaliculi-like structures were found during 21h culture, although junction apparatus typical of epithelium was noticed between adjacent cells.

Fig. 5

Electron microscopy of THLE-5b cells in aggregate culture. (A) Cell aggregates in aggregate culture (Agg-1); cell aggregates in aggregate culture-2 (Agg-2) were formed after 6h (B); 12h (A,C); and 21h (D) of culture. Stars indicate tight junction with desomosomes. Agg-2 culture shows bile canaliculi-like structures with microvilli. Original magnification, ×10 000 (A), ×8000 (B,C,D). Scale bar, 1μm.

3.3 Matrigel culture

When grown in Matrigel, THLE-5b cells formed duct-like structures within the aggregates at day 1 of inoculation, but they disappeared after further cultivation (data not shown). RT-PCR analysis of 5-day Matrigel culture revealed that expression of GGT mRNA was up-regulated, although the level of CK19 mRNA was slightly down-regulated (Fig. 6).

Fig. 6

Expression of biliary cell markers in Matrigel. Left lane: cells grown with 10ng HGF/ml as a confluent monolayer (Mono). Right lane: cells grown with 10ng HGF/ml in Matrigel (Matrigel).

3.4 Cell transplantation

THLE-5b cells transplanted in the peritoneal cavity of SCID mice were well localized after 3weeks of experiments. Transplanted THLE-5b cells expressed albumin (Fig. 7A) and α1-antitrypsin (Fig. 7B), but did not express CK19 (Fig. 7C), compared to primary antibody-minus control (Fig. 7D). Scale bar, 25μm.

Fig. 7

Immunocytochemistry of THLE-5b cell mass adjacent to microcarrier beads in SCID mice after 3weeks. (A) albumin; (B) α1-antitrypsin; (C) CK19; and (D) primary antibody-minus control. Cells are not counterstained. Scale bar, 25μm.

4 Discussion

Oncogenic events either target mature hepatocytes or the proposed hepatic stem cells in hepatocarcinogenesis (Zvibel et al., 1998). We believe that transformation of the hepatic stem/progenitor cells resulted in THLE-5b cells for several reasons: (i) mature hepatocytes immortalized with SV40 maintain their epithelial morphology (Woodworth et al., 1986). THLE-5b cells did not exhibit typical epithelial morphology, although they were positive for CK18; (ii) oncogenic transformation of mature hepatocytes often induces expression of fetal genes such as AFP. THLE-5b cells never expressed AFP in any culture conditions examined; (iii) THLE-5b cells grown as a monolayer expressed CK14, Thy-1 and c-kit, which are rarely detected in the mature hepatocytes (Baumann et al., 1999). CK14 is found in LPCs or oval cells and may be a marker for LPCs or oval cells that are capable of differentiating into hepatocytes (Florino et al., 1998). Thy-1 is detected only in oval cells (Petersen et al., 1998); and (iv) oval cells are identified by co-expressing stem cell and hepatic lineage markers (Strain and Crosby, 2000; Vessey and delaHall, 2001). THLE-5b cells expressed hepatocytic and bile duct epithelial cell markers under basic culture conditions.

The potential of cells to differentiate could be tested in three-dimensional configurations of cells formed by aggregation of individual cells. The up-regulation or induction of liver-specific genes was observed in THLE-5b cells when they were grown as aggregates. Significantly, induction of hepatocytic markers by aggregation coincided with down-regulation of LPC marker CK14 and bile duct/oval cell marker GGT. Furthermore, it is noticed that ultrastructural study showed more differentiated structures including the formation of bile canaliculi- and desomosome-like structures, which are thought to be characteristic of mature hepatocytes, when THLE-5b cells were grown as aggregates. Under these conditions, ammonia-metabolic activity was also demonstrated. These results indicate that THLE-5b cells within aggregates have differentiated as hepatocyte-like cells.

A subset of stem cells, termed side population (SP) cells, has been isolated using flow cytometry and the DNA-binding dye Hoechst 33342 in mammalian tissues or cell lines (Goodell et al., 1996). It is known that SP cells overexpress adhesive molecules (Hishikawa et al., 2005). RT-PCR analysis of THLE-5b cells indicated that of 8 hepatocyte-specific genes tested 4, including E-cadherin, and 7 were up-regulated or induced in aggregate culture-1 and -2, respectively. These results indicate that the cells in aggregate culture-1 may have similar phenotypes to those in aggregate culture-2 and suggest that cell aggregation may enrich progenitor cells or SP cells in THLE-5b cells.

Many liver enriched transcriptional factors have been identified, including C/EBP, HNF-1, HNF-3, HNF-4, and HNF-6 families, which constitute the binding activity interacting with hepatocyte-specific genes. HNF-1 and its regulator HNF-4α play a crucial role in differentiation of hepatic cells. In the present study, an increased expression of HNF-1, HNF-4α and HNF-6 as well as hepatocytic genes by THLE-5b cells were shown when they were grown as aggregates. The recent study suggests the presence of a reciplocal cross regulation between HNF-1 and HNF-4α and a synergism between HNF-1 and HNF-6, which may be involved in an activation of HNF-4α as suggested elsewhere (Suaud et al., 1997; Clotman et al., 2002). At present there is no known ligand for HNF-4α (Suaud et al., 1997), but it is likely that compact structures in the aggregates formed in rotation culture are responsible for a modulation of HNF-4α, followed by HNF 1 activation and result in activation of a diverse set of liver genes, after interacting synergistically with adjacent binding factors.

The capacity to differentiate has also been tested by transplanting them into sites in vivo (Grisham and Thorgeirsson, 1996) in which liver-specific differentiation would be expected. THLE-5b cells expressed albumin and α1-antitrypsin but had lost expression of CK19 when they were transplanted intraperitoneally. These results suggest that THLE-5b cells remain functional in vivo and could provide novel tools for use in cell transplantation.

Culture in Matrigel is shown to favor bile duct cell differentiation (Paradis and Sharp, 1989). Bile duct/oval cell marker GGT was up-regulated in Matrigel culture of THLE-5b cells. However, the cells did not form bile duct-like structures and other bile duct/oval cell marker CK19 was slightly down-regulated. These results demonstrate that THLE-5b cells have differentiated into biliary phenotype, but only partially. This may reflect either that the induction to differentiate in Matrigel is not specific as suggested previously (Strick-Marchand and Weiss, 2002) or it may reflect a limited differentiation potential inherent in the cells. In this study, both Matrigel and monolayer cell culture were grown in culture medium with HGF. It will be of interest to explore the effect of HGF on bile duct differentiation potential of THLE-5b cells in these culture conditions.

Normal human somatic cells including mature hepatocytes and fetal liver cells undergo a limited number of divisions in vitro, entering a non-dividing state called cellular senescence (Malhi et al., 2002). The identification of LPCs capable of unlimited expansion and bilineage differentiation is of great interest for cellular therapy (Allain et al., 2002). Along this line non-human primate fetal liver stem cells were immortalized by means of a retroviral vector expressing simian virus 40 large T antigen to evaluate their differentiation potentials (Allain et al., 2002). In this study, we demonstrated that an immortalized human liver epithelial cell line THLE-5b has the potential to differentiate into hepatocytic or biliary cell lineage both in vitro and in vivo. These results imply that bipotent progenitor cell populations similar to THLE-5b cells may exist in adult human liver.

Acknowledgements

We thank Dr. C.C. Harris (Laboratory of Human Carcinogenesis, National Cancer Institute, Bethesda, MD, USA) for providing us with THLE-5b.

References

Allain J-E, Dagher, I, Caputo, DM, Loux, N, Andreoletti, M, Westerman, K. Immortalization of a primate bipotent epithelial liver stem cell. Proc Natl Acad Sci USA 2002:99:3639-44
Crossref Medline 1st Citation 2nd 3rd

Baumann U, Crosby, H, Ramani, P, Kelly, DA, Strain, AJ. Expression of the stem cell factor c-kit in normal and diseased pediatric liver: identification of a human hepatic progenitor cell? Hepatology 1999:30:112-7
Crossref Medline 1st Citation

Chomczynski P, Sacchi, N. Single-step method of RNA isolation by acid guanidium thiocyanate-phenol-chloroform extraction. Anal Biochem 1987:162:156-9
Medline 1st Citation

Clotman F, Lannoy, VJ, Reber, M, Cereghini, S, Cassiman, D, Jacquemin, P. The one cut transcription factor HNF6 is required for normal development of the biliary tract. Development 2002:129:1819-28
Medline 1st Citation

Florino AS, Diehl, AM, Lin, HZ, Lemischka, IR, Reid, LM. Maturation-dependent gene expression in a conditionally transformed liver progenitor cell line. In Vitro Cell Dev Biol 1998:34:247-58
Crossref 1st Citation

Goodell MA, Brose, K, Paradis, G, Conner, AS, Mulligan, RC. Isolation and functional properties of murine hematopoietic stem cells that are replicating in vivo. J Exp Med 1996:183:1797-806
Crossref Medline 1st Citation

Grisham JW, Thorgeirsson, SS. Liver stem cells. Stem cells 1996:233-82
1st Citation

Hishikawa K, Marumo, T, Miura, S, Nakanishi, A, Matsuzaki, Y, Shibata, K. Leukemia inhibitory factor induces multi-lineage differentiation of adult stem-like cells in kidney via kidney-specific cadherin 16. Biochem Biophys Res Commun 2005:328:288-91
Crossref Medline 1st Citation

Kano J, Noguchi, M, Kodama, M, Tokiwa, T. The in vitro differentiating capacity of non-parenchymal epithelial cells derived from adult porcine livers. Am J Pathol 2000:156:2033-43
Medline 1st Citation

Landry J, Bernier, D, Goyette, OR, Marceau, N. Spheroidal aggregate culture of rat liver cells: histotypic reorganization, Biomatrix deposition, and Maintenance of functional activities. J Cell Biol 1985:101:914-23
Crossref Medline 1st Citation

Lechner JF, Smoot, DT, Pffeifer, AM, Tokiwa, T, Harris, CC. A non-tumorigenic human liver epithelial cell culture model for chemical and biological carcinogenesis investigations. Neoplastic transformation in human cell system. Mechanism of carcinogenesis 1990:307-21
1st Citation 2nd 3rd

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 2002:115:2679-88
Medline 1st Citation 2nd

Paradis K, Sharp, IH. In vitro duct-like structure formation after isolation of bile ductular cells from a murine model. J Lab Clin Med 1989:113:689-94
Medline 1st Citation

Petersen BE, Goff, JP, Greenberger, JS, Michalopoulos, GK. Hepatic oval cells express the hematopoietic stem cell marker Thy-1 in the rat. Hepatology 1998:27:433-45
Crossref Medline 1st Citation

Selden C, Chanlmers, S-A, Jones, C, Standish, R, Quaglia, A, Roland, N. Epithelial colonies cultured from human explanted liver in subacute hepatic failure exhibit hepatocytes, biliary epithelial, and stem cell phenotypic markers. Stem cells 2003:21:624-31
Crossref Medline 1st Citation

Strain AJ, Crosby, HA. Hepatic stem cells. Gut 2000:46:743-5
Crossref Medline 1st Citation

Strick-Marchand H, Weiss, MC. Inducible differentiation and morphogenesis of bipotent liver cell lines from wild-type mouse embryo. Hepatology 2002:36:794-804
Medline 1st Citation

Suaud L, Josepha, B, Formstecher, P, Laine, B. mRNA expression of HNF-4 isoforms and of HNF-1/HNF-1 variants and differentiation of human cell lines that mimic highly specialized phenotypes of intestinal epithelium. Biochem Biophys Res Comm 1997:235:820-5
Crossref Medline 1st Citation 2nd

Thorgeisson SS. Hepatic stem cells in liver regeneration. FASEB J 1996:10:1249-56
Medline 1st Citation

Tokiwa T, Kano, J, Noguchi, M, Kodama, M, Tateishi, T. Restoration of differentiated functions in multicellular aggregates of a human liver epithelial cell line. Mat Sci Engineer C 1998:6:249-52
Crossref 1st Citation

Vessey CJ, delaHall, PM. Hepatic stem cells. Pathology 2001:33:130-41
Medline 1st Citation

Woodworth C, Secott, T, Isom, HC. Transformation of rat hepatocytes by transfection with simian virus 40 to yield proliferating differential cells. Cancer Res 1986:46:4018-26
Medline 1st Citation

Zvibel I, Fiorino, AS, Brill, S, Reid, LM. Phenotypic characterization of rat hepatoma cell lines and lineage-specific regulation of gene expression by differentiation agents. Differentiation 1998:63:215-23
Crossref Medline 1st Citation

Received 1 November 2005/7 July 2006; accepted 19 July 2006

doi:10.1016/j.cellbi.2006.07.006

CBI home

Browse archive

ISSN Print: 1065-6995
ISSN Electronic: 1095-8355
Published by Portland Press Limited on behalf of the International Federation for Cell Biology (IFCB)