SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. References

Human bone marrow–derived mesenchymal stem cells (BM-MSCs) are expected to be a potential source of cells for transplantation. Although recent reports have shown that isolated MSCs can differentiate into hepatocytes, the efficiency of differentiation is insufficient for therapeutic application. To circumvent this problem, it is necessary to understand the mechanisms of hepatic differentiation of human BM-MSCs. Hepatocyte nuclear factor 3β (HNF3β), a forkhead/winged helix transcription factor, is essential for liver development. In the present study, we established a tetracycline (Tet)-regulated expression system for HNF3β in UE7T-13 BM-MSCs. HNF3β expression significantly enhanced expression of albumin, α-fetoprotein (AFP), tyrosine amino transferase (TAT) and epithelial cell adhesion molecule (EpCAM) genes. The differentiated cells showed hepatocyte-specific functions including glycogen production and urea secretion. During treatment with the Tet-on system for 8 days, over 80% of UE7T-13 cells turned out to express albumin. Furthermore, the combination of Tet with basic fibroblast growth factor (bFGF) efficiently induced the genes such as albumin and TAT, which are associated with maturity of hepatocytes; however, it suppressed genes such as AFP and EpCAM, which are associated with immaturity of hepatocytes, suggesting that Tet-induced HNF3β expression sensitizes BM-MSCs to bFGF signals. Finally, the results of the present study suggest that down-regulation of Wnt/β-catenin signals caused by translocation of β-catenin to cytoplasmic membrane is associated with hepatic differentiation of human BM-MSCs. Conclusion: HNF3β expression induced efficient differentiation of UE7T-13 human BM-MSCs. (HEPATOLOGY 2008;48:597–606.)

Human bone marrow cells can be expanded in vitro and are expected to be a potential source for stem cell therapy without the risk of immune rejection. The bone marrow cells contain mesenchymal stem cells (MSCs) as well as hematopoietic stem cells.1 The multipotency of MSCs has been shown by differentiating into various cell types, such as mesodermal, neuroectodermal, and endodermal cells.1–8 Recently, Lee et al.5 showed that both human bone marrow–derived MSCs (BM-MSCs) and umbilical cord blood–derived MSCs have a potency to differentiate into hepatocytes in vitro.

In our previous reports, the protocol to induce hepatic differentiation of UE7T-13 BM-MSCs has been developed.6 These cells were immortalized by infection with a retrovirus carrying human telomerase reverse transcriptase (hTERT) and one of the early genes of the human papilloma virus, E7. Although hTERT is introduced into UET7T-13 cells, it has been reported that the differentiation potential of the cells is not affected.4 Indeed, it has been reported that human fetal hepatocytes immortalized by hTERT do not disrupt their differentiation potential.9 As a result, we showed that UE7T-13 cells were induced to differentiate into hepatocytes with the treatment of combination of acidic fibroblast growth factor (aFGF), basic fibroblast growth factor (bFGF), and hepatocyte growth factor (HGF) with type IV collagen coating. Moreover, WISP1 and WISP2 have been reported to play an important role in hepatic differentiation of these cells.6 We also have shown that hepatic differentiation of UCBTERT-21 UCB-MSCs is induced by 5-azacytidine/HGF/OSM/bFGF treatment.7 In addition, we also have demonstrated that down-regulation of Wnt/β-catenin signals with small interfering RNA for Fz8 enhances hepatic differentiation of these cells. The results of these studies suggest that Wnt/β-catenin signals play an important role in hepatic differentiation of human MSCs.

Hepatocyte nuclear factors (HNFs) have been shown to be key transcription factors for liver development during mouse embryogenesis.10, 11 Among them, HNF3 is thought as a key player for the hepatogenesis.12, 13 HNF3 is a member of the forkhead box transcription factor family and is reported to regulate expression of more than 100 genes expressed in the liver, pancreas, intestine, and lung during early embryogenesis through a consensus HNF3-binding site.14 However, the analysis of the diverse functions of HNF3β at later stages in embryogenesis was difficult, because HNF3β−/− embryos displayed an early lethal phenotype.15 To circumvent these difficulties, Lee et al.16 established a conditional knockout mouse, HNF3α−/−, HNF3βloxP/loxP, HNF3γ-Cre, whose embryos were completely deficient in hepatic specification as neither liver bud development nor the expression of the earliest hepatic marker gene, α-fetoprotein (AFP). These results suggest that HNF3α and HNF3β are essential for the onset of hepatogenesis.

In the present study, in order to investigate the potency of hepatic differentiation of MSCs by HNF3β, we established a tetracycline (Tet)-regulated expression system for HNF3β in UE7T-13 cells. Using these cells, we tried to examine whether HNF3β expression efficiently induces hepatic differentiation of BM-MSCs and to clarify the underlying mechanism during this process.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. References

Establishment of Tet-on System in UE7T-13.

The human BM-MSCs, UE7T-13 cells,4, 6 the life span of which was prolonged by infecting retrovirus encoding human papillomavirus E7 and hTERT, were used. The cells were maintained as described previously.6 pcDNA6/TR and pcDNA4/TO vectors (T-REx system kit; Invitrogen, Carlsbad, CA) were used. The human HNF3β complementary DNA generated with reverse-transcription polymerase chain reaction (RT-PCR) of total RNA prepared from HuH-7 cells was inserted into multiple cloning sites of pcDNA4/TO. The resulting plasmid was designated pcDNA4/TO-HNF3β. UE7T-13 cells were transfected with the linearized pcDNA6/TR by electroporation and selected with a medium containing 3 μg/mL blastcydin in the first step. The blastcydin-resistant cells were cloned and designated E7-T6-12. In the second step, E7-T6-12 cells were transfected with the linearized pcDNA4/TO-HNF3β by electroporation and selected with a medium containing 100 μg/mL zeocin. Finally, the blastcydin-resistant and zeocin-resistant cells were cloned and designated E7-H-4.

In Vitro Hepatic Differentiation Procedure.

E7-H-4 cells were inoculated at a density of 3.6 × 104 cells/cm2 in six-well plates and maintained in Dulbecco's modified Eagle's medium containing 10% fetal bovine serum, 100 U/mL penicillin, 100 μg/mL streptomycin, and 1 μg/mL Tet (Nakarai Tesque, Kyoto, Japan) with or without cytokines. The cytokines include 10 ng/mL HGF (PeproTech EC, London, UK), 10 ng/mL bFGF (PeproTech EC), and 10 ng/mL oncostatin M (OSM; Diaclone Research, Besançon, France). The culture medium was changed every 2 days, and the cells were cultured for 8 days.

RNA Isolation and Real-Time RT-PCR.

The cells incubated in the presence of 1 μg/mL Tet were taken every 2 days until 8 days. Total RNA was extracted with TRIzol reagent (Invitrogen) and subjected to reverse-transcription using Superscript II (Invitrogen) and oligo(dT) primers. The messenger RNA (mRNA) expression levels were determined by a LightCycler System (Roche Applied Science, Mannheim, Germany) using gene specific primers (Table 1). The mRNA levels of the genes were normalized with that of β-actin.

Table 1. Sequences and Annealing Temperatures of Primers
PrimerSequenceAnnealing Temperature (°C)
  1. Abbreviation: OSMR, oncostatin M receptor.

HNF3bForward: 5′-GCCCGGTCACGAACAAAACG-3′55
 Reverse: 5′-CGTCGTCTTCTTAAGAAG-3′ 
AFPForward: 5′-AGCAGCTTGTTAAATCAACATGCA-3′64
 Reverse: 5′-AAAATTAACTTTGGTAAACTTCTGACTCAGT-3′ 
AlbuminForward: 5′-TGTTGCATGAGAAAACGCCA-3′56
 Reverse: 5′-GTCGCCTGTTCACCAAGGA-3′ 
TATForward: 5′-CTGAAGTTACCCAGGCAATGAAAG-3′60
 Reverse: 5′-TAATAAGAAGCAATCTCCTCCCGAC-3′ 
EpCAMForward: 5′-CTGGCCGTAAACTGCTTTGT-3′62
 Reverse: 5′-AGCCCATCATTGTTCTGGAG-3′ 
FGFR1Forward: 5′-GAGATGGAGGTGCTTCACTTA-3′62
 Reverse: 5′-TACAGGGGCGAGGTCATCA-3′ 
c-MetForward: 5′-GTTTACTTGTTGCAAGGGAGAAGACT-3′62
 Reverse: 5′-TAGGGTGCCAGCATTTTAGCA-3′ 
OSMRForward: 5′-GTGTGGGTGCTTCTCCTGCTTCTGTA-3′58
 Reverse: 5′-TCTGTGCTAATGACTGTGCTTGTGGT-3′ 
b-actinForward: 5′-GACGGCCAGGTCATCACTATTG-3′56
 Reverse: 5′-CCACAGGATTCCATACCCAAGA-3′ 

Western Blot Analysis.

E7-H-4 cells inoculated in 1 μg/mL tetracycline and/or 10 ng/mL bFGF were taken at 6 hours, 12 hours, 1 day, and 2 days. Thirty micrograms of the proteins were subjected to western blot analysis. The rabbit polyclonal antibodies including anti-HNF3β (Santa Cruz Biotechnology, Santa Cruz, CA), anti–β-catenin (Cell Signaling Technology, Danvers, MA), anti-phospho–β-catenin (Ser33/37/Thr41) (Cell Signaling Technology), anti-phospho–β-catenin (Thr41/Ser45) (Cell Signaling Technology), anti–phospho-GSK3β (Santa Cruz Biotechnology) and anti-Actin (Santa Cruz Biotechnology), and mouse monoclonal antibodies including anti-GSK3β (Santa Cruz Biotechnology) and anti-cyclin D1 (Santa Cruz Biotechnology) were used.

Immunofluorescenec Analysis.

E7-H-4 cells inoculated onto cover glasses were cultured in 1 μg/mL Tet and 10 ng/mL bFGF. After incubation for 2, 4, 6, and 8 days, the cells were fixed in phosphate-buffered saline containing 4% paraformaldehyde for 20 minutes and permeabilized with phosphate-buffered saline containing 0.1% Triton X-100 for 10 minutes. The samples were incubated with anti-HNF3β antibody (Santa Cruz Biotechnology, Inc.), anti-human serum albumin antibody (Sigma Aldrich, St. Louis, MO), anti-human AFP antibody (Santa Cruz Biotechnology), anti-human CCAAT enhancer-binding protein α (C/EBPα) antibody (Santa Cruz Biotechnology), anti-CYP1A1/1A2 (Chemicon International, Temecula, CA), and anti-β-catenin antibody (Cell Signaling Technology), followed by incubation with second antibody conjugated with fluorescent phycobilioroteins, Alexa Fluoro 594 goat anti-rabbit immunoglobulin G or Alexa Fluoro 488 goat anti-mouse immunoglobulin G (Molecular Probes, Leiden, Netherlands). DAPI was used for nuclear counterstaining.

Periodic Acid-Schiff Staining for Glycogen.

Periodic acid-Schiff (PAS) staining was performed as described previously.6, 7

Urea Assay.

The E7-H-4 cells were inoculated in the density of 3.6 × 104 cells/cm2 in six-well plates, and then incubated in 1 μg/mL Tet with or without 10 ng/mL bFGF. After 6 days, 3.6 × 104 cells were incubated in the media with 5 mM ammonium chloride, and the amount of urea secreted into the medium was measured according to the previous reports.6, 7

Reporter Assay.

The plasmid TCF4-CMVpro-Luc contains three repeats of the optimal TCF-4 motif CCTTTGATC upstream of the cytomegalovirus promoter, which drives the expression of luciferase.7 The plasmid pRL-TK (Promega) was used as an internal control. The E7-H-4 cells inoculated at the density of 3.6 × 104 cells/cm2 in a 24-well plate were treated in 1 μg/mL Tet with or without 10 ng/mL bFGF for 2 days, were followed by transfection with TCF4-CMVpro-Luc. The E7-H-4 cells cultured without Tet and bFGF were used as a negative control. Transient transfection was performed using FuGENE6 transfection reagent (Roche Applied Science). The plasmid TCF4-CMVpro-Luc worked well when the E7-H-4 cells were treated with 100 ng/mL Wnt-3a (BioDynamics Laboratory, Tokyo, Japan) at 12 hours after transfection as a positive control as previously reported.7 At 24 hours after transfection, the cell lysates were used for reporter assay.

Results

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. References

Establishment of a Tet-on Expression System for HNF3β Using UE7T-13 Cells.

We established a Tet-on expression system for HNF3β in UE7T-13 cells, which are human BM-MSCs immortalized with hTERT and E7 and designated E7-H-4. We verified that the level of mRNA for HNF3β was tightly regulated by addition of Tet in a dose-dependent manner (Fig. 1A). Western blot analysis indicated that the protein levels of HNF3β reached a plateau at 12 hours after addition of Tet (Fig. 1B). Tretement with 1 μg/mL Tet for 2 days induced expression of HNF3β in almost all the cells, as shown by immunofluorescence analysis (Fig. 1F).

thumbnail image

Figure 1. Establishment of a Tet-on expression system for HNF3β using UE7T-13 cells. (A) Real-time RT-PCR analysis showed HNF3β RNA levels in E7-H-4 cells cultured in media containing different concentration of Tet for 4 days. The data are shown as the relative mRNA levels to those of day 0. (B) Western blot analysis showed that HNF3β protein levels in E7-H-4 cells treated with 1 μg/mL of Tet for 6 hours, 12 hours, 1 day, and 2 days. (C-F) Immunohistochemistry showed HNF3β protein in E7-H-4 cells cultured in the absence and present of 1 μg/mL Tet for 2 days (C and E, respectively). Counter-staining was performed with 4',6-diamidino-2-phenylindole (a type of stain) (D and F, respectvely).

Download figure to PowerPoint

Effects of HNF3β Expression on In Vitro Hepatic Differentiation.

The E7-H-4 cells treated with 1 μg/mL Tet for 2 days changed their morphology; the cells became extended and bigger (Fig. 2A,B). To investigate the effect of HNF3β expression on hepatic differentiation of E7-H-4 cells, we measured the mRNA levels of hepatocyte-specific genes with real-time RT-PCR every 2 days. The Tet-on system induced about 60-fold to 90-fold increases in HNF3β mRNA between 2 and 8 days (Fig. 2C). Expression of albumin mRNA was significantly increased at 4 days and dramatically increased at 6 and 8 days (Fig. 2D). Albumin mRNA expression in the cells was decreased when Tet was removed from the medium at 2 or 4 days (data not shown). Hence, this Tet-on system is reversible with respect to hepatic differentiation up to 4 days. AFP mRNA also increased at 4 days and reached a peak at 6 and 8 days (Fig. 2E). Expression of EpCAM reached a plateau at 2 days and was not altered thereafter (Fig. 2F). The mean of CYP1A1 mRNA was not altered (Fig. 2G). Tyrosine aminotransferase (TAT) mRNA was significantly increased 4, 6, and 8 days after treatment (Fig. 2H).

thumbnail image

Figure 2. Effects of HNF3β expression on hepatic differentiation. (A, B) The morphology of E7-H-4 cells. The E7-H-4 cells were cultured in the absence and presence of 1 μg/mL Tet for 2 days (A and B, respectively) and were observed by phase contrast microscopy. (C-K) The mRNA levels of HNF3β (C), albumin (D), AFP (E), EpCAM (F), CYP1A1 (G), TAT (H), FGFR1 (I), c-Met (J), OSMR (K). The E7-H-4 cells treated with 1 μg/ml of Tet for 0, 2, 4, 6, and 8 days were measured with real-time RT-PCR. The data are shown as the relative mRNA levels to those of 0 day. Data are expressed as the mean ± standard error of the mean of three experiments.*P < 0.05, **P < 0.01 compared with the mRNA level at day 0.

Download figure to PowerPoint

The expression of three receptors for important cytokines for hepatic differentiation6, 7 was examined. FGFR1 mRNA was gradually increased from 2 days to 8 days (Fig. 2I). However, expression of c-Met mRNA was significantly decreased at 2, 4, and 6 days (Fig. 2J).The expression level of OSMR was not changed (Fig. 2K).

Functional Analysis of E7-H-4 Cells After Induction of Hepatic Differentiation.

Immunofluorescence analysis was performed for the expression of hepatocyte-specific proteins 8 days after the start of Tet treatment. The sizes of the cells treated with Tet were bigger than those without Tet. The intensities of C/EBPα, AFP, CYP1A/1/2, and albumin mRNAs in the cells treated with Tet were much stronger than those without Tet (Fig. 3A,B). Expression of these hepatocyte-specific proteins was increased with time (data not shown).

thumbnail image

Figure 3. Immunofluorescence analysis of hepatocyte-specific proteins expression and functional analysis of E7-H-4 cells treated with Tet. (A) The expression of C/EBPα and AFP in E7-H-4 cells treated with 1 μg/mL Tet for 0, 2, 4, 6, and 8 days were examined by immunofluorescence analysis. E7-H-4 cells cultured in medium without Tet were used as a control. (B) The expression of CYP1A1/2 and albumin in E7-H-4 cells treated with 1 μg/mL Tet for 0, 2, 4, 6, and 8 days were examined by immunofluorescence analysis. E7-H-4 cells cultured in medium without Tet were used as a control. (C) Urea production in E7-H-4 cells. Cells were treated with Tet for 6 days, followed by further incubation in the presence of 5 mM ammonium chloride for 24 hours. The concentration of secreted urea in the culture media was determined via colorimetric assay. E7-H-4 cells cultured in medium without Tet were used as a control. HepG2 cells were used as a positive control. Data are expressed as the mean ± standard error of the mean of three experiments. *P < 0.05, **P < 0.01, compared with control cells. (D) PAS staining of E7-H-4 cells treated with Tet for 2, 4, 6, and 8 days. Glycogen stored in the cells was stained via PAS staining. E7-H-4 cells cultured in medium without Tet were used as a control. HuH7 cells were used as a positive control. The arrows indicate PAS-positive cells.

Download figure to PowerPoint

We next examined whether E7-H-4 cells treated with Tet have the increased ability of urea synthesis. For urea assay, 5 mM ammonium chloride was added to the culture media of E7-H-4 cells that had been treated with Tet for 6 days, and the level of urea secreted from the cells was measured 24 hours after the addition of ammonium chloride. The urea synthesis was increased by Tet treatment, and the level was similar to that of HepG2 cells (Fig. 3C). The PAS-positive cells were observed in E7-H-4 cells treated with Tet; however, the intensity was not so strong (Fig. 3D).

Effects of Cytokines on the HNF3β-Induced Hepatic Differentiation of E7-H-4 Cells.

Using this system, we attempted to define the cytokine that accelerates hepatic differentiation of E7-H-4 cells in combination with HNF3β. Considering from the expression analysis of FGFR1, c-Met, and OSMR, bFGF was supposed to be required for hepatic differentiation. Treatment with bFGF and Tet additively promoted albumin expression at a dose-dependant manner (Fig. 4A). However, only bFGF had little effect on albumin expression.

thumbnail image

Figure 4. Effects of cytokines on the HNF3β-induced hepatic differentiation. (A) mRNA levels of albumin in E7-H-4 cells treated with a combination of Tet and various concentrations of bFGF for 8 days. The experiment was performed with real-time RT-PCR, and the data are shown as the mRNA levels relative to those of day 0. Data are expressed as the mean ± standard error of the mean of three experiments. *P < 0.05, **P < 0.01. (B, C) mRNA levels of albumin and AFP (B and C, respectively) in E7-H-4 cells treated with Tet, Tet/bFGF, Tet/HGF, and Tet/OSM for 0, 2, 4, 6, and 8 days. The concentration of all cytokines was 10 ng/mL. The experiment was performed with real-time RT-PCR, and the data are shown as the mRNA levels relative to those of day 0. Data are expressed as the mean ± standard error of the mean of three experiments. *P < 0.05, **P < 0.01. (D, E) mRNA levels of TAT and EpCAM in E7-H-4 cells treated with Tet or Tet/bFGF for 0, 2, 4, 6, and 8 days. The experiment was performed with real-time RT-PCR, and the data are shown as the mRNA levels relative to those of day 0. Data are expressed as the mean ± standard error of the mean of three experiments. (F) The expression of albumin in E7-H-4 cells treated with Tet alone or Tet/bFGF for 0, 2, 4, 6, or 8 days was examined by immunofluorescence analysis. The concentration of bFGF was 10 ng/mL. E7-H-4 cells cultured in medium without Tet were used as a control. (G) The percentages of albumin-positive cells in E7-H-4 cells treated with Tet alone or Tet/bFGF. The concentration of bFGF was 10 ng/mL. E7-H-4 cells cultured in medium without Tet were used as a control. Data are expressed as the mean ± standard error of the mean of four fields. *P < 0.05, **P < 0.01 compared with control cells.

Download figure to PowerPoint

Treatment with Tet promoted expression of EpCAM mRNA; however, the combination of Tet and bFGF significantly increased its mRNA at 2 days but not at 4, 6, or 8 days (Fig. 4D). Treatment with Tet induced expression of TAT at 4, 6, and 8 days by three-fold; however, the combination of Tet with bFGF also induced TAT expression at 2 days as well as at 4, 6, and 8 days (Fig. 4E). The expression levels of albumin of the cells treated with Tet and bFGF was similar to those treated with Tet alone (Fig. 4F). The percentages of albumin-positive cells in both groups were almost equal, although the mean of the positive cells treated with Tet and bFGF was a little higher than those with Tet alone (Fig. 4G).

Analysis of Wnt/β-Catenin in E7-H-4 Cells Treated with Tet or Tet/bFGF.

Our previous reports showed that down-regulation of Wnt/β-catenin signaling plays a role in hepatic differentiation of MSCs.6, 7 We focused on Wnt/β-catenin signals as one of the important mechanisms for hepatic differentiation induced by HNF3β expression. First, we analyzed cellular localization of β-catenin in E7-H-4 cells treated with Tet via immunofluorescence analysis. Strong immunofluorescence signal of β-catenin was observed in the cytoplasmic membrane when the E7-H-4 cells were treated with Tet, whereas β-catenin was localized in the nuclei and cytoplasm of the cells when they were not treated with Tet (Fig. 5A). Membraneous translocation of β-catenin was also seen in the cells treated with Tet and bFGF (data not shown). Second, the reporter assay using pTCF4-CMVpro-Luc reporter plasmid was performed to quantitatively evaluate the transcription activity of the Wnt/β-catenin signaling (Fig. 5B). As a result, TCF-4 activity was suppressed after treatment with Tet. These data suggest that down-regulation of Wnt/β-catenin signals plays an important role in hepatic differentiation of MSCs. Almost the same reduction of TCF-4 activity was induced by both Tet and bFGF as seen in Tet alone (data not shown). Third, expression of Wnt/β-catenin signal-related molecules was examined via western blotting (Fig. 5C). Cyclin D1 expression was gradually decreased. Expression levels of β-catenin, phospho-β-catenin, GSK and phospho-GSK were not changed. These data suggest that translocation of β-catenin, not phosphorylation of β-catenin, is responsible for suppressed Wnt/β-catenin signals.

thumbnail image

Figure 5. Analysis for Wnt/β-catenin localization and signals (A) The localization of β-catenin in E7-H-4 cells treated with Tet for 2 days was examined via immunofluorescence analysis. Tet (+) and Tet (−) represent the cells treated in the presence and absence of Tet, respectively. (B) TCF-4 activity in E7-H-4 cells treated with Tet for 2 days via reporter assay. Data are expressed as the mean ± standard error of the mean of three experiments. E7-H-4 cells cultured in medium without Tet were used as a control. *P < 0.05, compared with control cells. (C) The protein expression levels of cyclin-D1, phospho-β-actin (Ser33/Ser37/Thr41), phospho-β-catenin (Thr41/Ser45), β-catenin, phospho-GSK3β, and GSK3β in E7-H-4 cells treated with 1 μg/mL of Tet for 6 hours, 12 hours, 1 day, or 2 days was examined via western blot analysis. Actin was used as an internal control.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. References

Many reports have demonstrated the pluripotency of MSCs; thus, they are expected to be a source of specific cell types for transplantation. The efficiency of hepatic differentiation of MSCs has been improved by modifying culture conditions or by adding various growth factors and cytokines.5–7 However, the efficiencies of hepatic differentiation of MSCs are still insufficient for clinical application. It is necessary to understand the mechanism of hepatic differentiation of MSCs to achieve transdifferentiation with high efficiency. Therefore, we established a Tet-on expression system for HNF3β in UE7T-13. Using this sytem, we demonstrated that Tet-induced expression of HNF3β with bFGF achieved highly efficient differentiation of MSCs into hepatocyte lineage, in which approximately 80% of the cells were albumin-positive after treatment with Tet and bFGF for 8 days.

It has been reported that HNF3β directly regulates expression of a number of hepatocyte-specific genes, including AFP, albumin, TAT, and HNF3β14 In the present study, the morphological change and functional maturity indicated that E7-H-4 cells differentiated into hepatocyte-like cells. In addition, the continuous and enhanced expressions of hepatocyte-specific genes required continuous expression of HNF3β for at least 6 days (data not shown). Therefore, hepatic differentiation of MSCs seem to take at least 6 days in this system. Interestingly, we detected EpCAM and AFP mRNAs that were expressed in fetal liver but not in adult liver.17 Therefore, the E7-H-4 cells treated with Tet for 8 days were not fully mature hepatocytes. Thus, this system provides a useful model, allowing the detailed investigation of an early stage in hepatic differentiation of BM-MSCs.

It has been reported that FGFs are secreted from the cardiac mesoderm and play an important role in the liver development of endoderm in embryogenesis.18, 19 Lee et al.16 demonstrated that the definitive endoderm from conditional knockout mouse for HNF3β did not respond to bFGF and did not express mRNAs of albumin and transthyretin. These results indicate that the expression of HNF3β is essential for definitive endoderm to receive bFGF signals for hepatogenesis. In accordance with this notion, Tet-induced HNF3β expression might sensitize E7-H-4 cells to bFGF signals by elevating expression of FGFR1. Moreover, the addition of bFGF accelerated induction of albumin expression by Tet treatment, and suppressed the immature hepatocyte markers such as AFP and EpCAM. In addition, treatment with bFGF alone was not effective for hepatic gene expression. These results suggest that induction of hepatic differentiation of human BM-MSCs by combination of HNF3β and bFGF mimics hepatogenesis.

Interestingly, the mRNA level of c-MET was down-regulated in E7-H-4 cells treated with Tet. In addition, HGF suppressed hepatic differentiation induced by HNF3β, and promoted mRNA level of platelet endothelial cell adhesion molecule-1 (PECAM1), which is known as an endothelial marker (data not shown). Multipotential functions of HGF have been reported as promoting the proliferation of mature hepatocytes, the motility and invasion of cancer cells, and morphogenesis and survival of epithelial and endothelial cells.20, 21 Moreover, some reports demonstrated that HGF could induce the differentiation of MSCs into not only hepatocytes but also cardiac cells.22 The reason why HGF suppressed hepatic differentiation induced by HNF3β is unknown. HGF is involved in liver regeneration in adults.23 OSM, which is known as a cytokine for hepatic maturation suppressed hepatic differentiation, as well as HGF. These results suggest that HGF and OSM are required at the later stage of hepatogenesis.

Finally, we observed that Wnt/β-catenin signaling was down-regulated in E7-H-4 cells that overexpress HNF3β. Our previous reports demonstrated that the suppression of Wnt/β-catenin signaling led a human MSC line to differentiate into hepatic lineage.6, 7 Wnt signaling is known to be involved in numerous developmental events of animal embryos, including the proliferation of stem cells and the specification of the neural crest.24, 25 Recently, Ke et al.26 showed that the repression of Wnt/β-catenin signaling could promote mouse BM-MSCs to differentiate into hepatocytes in vitro. In the present study, the localization of β-catenin was altered from the nuclei and cytoplasm to the cytoplasmic membrane in association with down-regulation of TCF-4 activity. In addition, although the source of Wnt signals in this system is not known, undifferentiated E7-H-4 cells may secrete them. Taken together, the change of Wnt/β-catenin signals is associated with hepatic differentiation of human MSCs.

In summary, we demonstrated that HNF3β expression could induce UE7T-13 cells to differentiate into hepatocytes and sensitize a response to bFGF. In this process, the down-regulation of Wnt/β-catenin was associated with hepatic differentiation of E7-H-4 cells. These findings provide useful information about stem cell biology that should contribute to the development of regenerative medicine for liver disease.

References

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. References