Hepatocyte growth factor signaling pathway inhibits cholesterol 7α-hydroxylase and bile acid synthesis in human hepatocytes

Authors


  • Potential conflict of interest: Nothing to report.

Abstract

Bile acid synthesis in the liver is regulated by the rate-limiting enzyme cholesterol 7α-hydroxylase (CYP7A1). Transcription of the CYP7A1 gene is inhibited by bile acids and cytokines. The rate of bile acid synthesis is reduced immediately after partial hepatectomy and during the early stage of liver regeneration. Hepatocyte growth factor (HGF) released from stellate cells activates a receptor tyrosine kinase c-Met, in hepatocytes and stimulates signaling pathways that regulate cell growth, proliferation, and apoptosis. This study demonstrated that HGF strongly and rapidly repressed CYP7A1 mRNA expression and the rate of bile acid synthesis in primary human hepatocytes. HGF rapidly induced c-Jun and small heterodimer partner mRNA and protein expression and increased phosphorylation of ERK1/2, JNK, and c-Jun. Specific inhibitors of protein kinase C, extracellular signal–regulated kinase 1/2 (ERK1/2), and c-Jun N-terminal kinase (JNK) blocked HGF inhibition of CYP7A1 expression. Knockdown of c-Met by small interfering RNA resulted in a significant increase in CYP7A1 and blocked HGF inhibition of CYP7A1 mRNA expression. Chromatin immunoprecipitation assays showed that HGF induced recruitment of c-Jun and small heterodimer partner (SHP) but reduced recruitment of the coactivators peroxisome proliferators activated receptor ρ coactivator 1α (PGC-1α) and cAMP response element binding protein (CREB)–binding protein (CBP) to chromatin. Conclusion: This study demonstrated that HGF is a novel regulator of CYP7A1 and bile acid synthesis in human hepatocytes and may protect hepatocytes from accumulating toxic bile acids and developing intrahepatic cholestasis during the early stage of liver regeneration. (HEPATOLOGY 2007.)

Bile acids are physiological detergents that facilitate absorption, transport, and distribution of sterols and lipid-soluble vitamins and disposal of toxic metabolites and xenobiotics. Bile acid synthesis and cholesterol 7α-hydroxylase (CYP7A1)2 gene transcription are inhibited by bile acids returning to the liver via enterohepatic circulation of bile acids.1 Because bile acids are amphipathic molecules that function as powerful detergents, their concentrations in hepatocytes have to be tightly regulated to prevent cell damage.2 Bile acids are also signaling molecules that activate nuclear receptors including farnesoid X receptor (FXR), pregnane X receptor, and vitamin D3 receptor and play important roles in the regulation of bile acid synthesis, cholesterol metabolism, and drug metabolism.2 It has been reported that bile acid synthesis and CYP7A1 activity are suppressed after partial hepatectomy and during liver regeneration in animals and human patients.3–5 The increase of bile acids in the remaining hepatocytes after partial hepatectomy and in the early stage of liver regeneration and injury inhibits bile acid synthesis and increases bile acid excretion to protect the liver from accumulating toxic bile acids. A recent study showed that bile acids and FXR are involved in liver regeneration in mice, as FXR-null mice had a lower rate of liver regeneration.6 However, the molecular mechanism underlying the bile acid effect on liver regeneration remains unclear.

Liver rapidly regenerates in response to liver injury. Liver regeneration is orchestrated by external stimuli including cytokines and growth factors.7–9 Hepatocyte growth factor (HGF) was the first identified of the mitogens induced in hepatectomized rat10, 11 HGF is a pleiotropic growth factor that mediates diverse biological processes including development, proliferation, wound healing, and tissue regeneration.12, 13 HGF activates a receptor tyrosine kinase c-Met, which stimulates diverse signaling pathways including Ras, mitogen-activated protein kinase (MAPK), phosphatidylinositol 3′-kinase (PI3-K), and phospholipase C.14–19 The essential role of HGF and c-Met in mammalian development and liver regeneration has been confirmed by genetic disruption of the HGF or c-Met gene in mice.12, 20–23

Despite bile acid synthesis and CYP7A1 expression being suppressed during liver regeneration, the molecular mechanism involved in these processes are not understood. In this study, we examined HGF regulation of CYP7A1 gene expression and defined the signaling pathways and molecular mechanism that mediates HGF regulation of CYP7A1 gene transcription in primary human hepatocytes (PHHs). HGF represses hepatic CYP7A1 gene expression through activation of the c-Jun N-terminal kinase (JNK)/c-Jun, extracellular signal–regulated kinase-1/2 (ERK1/2), and protein kinase C (PKC) pathways. Our study suggests that HGF is an important regulator of bile acid synthesis and may provide new insight into bile acid regulation of lipid homeostasis during liver regeneration and injury.

Abbreviations

CBP, cAMP response element binding protein (CREB)–binding protein; ChIP, chromatin immunoprecipitation; CYP7A1, cholesterol 7α-hydroxylase; ERK, extracellular signal–regulated kinase; FXR, farnesoid X receptor; HDAC, histone deacetylase; HGF, hepatocyte growth factor; HNF4α, hepatocyte nuclear factor 4α; JNK, c-Jun N-terminal kinase; PGC-1α, peroxisome proliferators activated receptor γ coactivator 1α; PHHs, primary human hepatocytes; PKC, protein kinase C; Q-PCR, quantitative real-time PCR; siRNA, small interfering RNA; SHP, small heterodimer partner.

Materials and Methods

Reagents and Plasmids.

The reagents cycloheximide, H89, trichostatin A, and Wortmannin were obtained from Sigma; Ro 31-8220, PD98059, SB203580, SP600125, Akt inhibitor IV, and LY294002 from CalBiochem; and HGF from R&D systems. A series of human CYP7A1 promoter luciferase reporters were constructed previously.24

Cell Culture.

PHHs were obtained from the Liver Tissue Procurement and Distribution System of the National Institutes of Health (S. Strom, University of Pittsburgh, PA). The HepG2 cells were obtained from the American Type Culture Collection (Manassas, VA). Cells were maintained as described previously.24

RNA Isolation and Quantitative Real-Time PCR (Q-PCR).

PHHs were treated with HGF (50 ng/mL) for 1–24 hours or as indicated, and total RNA was isolated using Tri-reagent (Sigma, St. Louis, MO) according to the manufacturer's instructions. Reverse-transcription PCR and Q-PCR were performed to detect CYP7A1, c-Met, c-Jun, HNF4α, PGC-1α, and SHP mRNA as described.24

Bile Acid Analysis.

PHHs were maintained in T-75 flasks for 3–24 hours with or without HGF. Media were collected at the indicated times, and bile acids were extracted for analysis as described previously.25

Immunoblotting Analysis.

Cell lysate preparation and immunoblotting analysis were performed as described.25 The following antibodies were used in immunoblotting: c-Met (C-28, sc-161), c-Jun (H-79, sc-1694), phospho-c-Jun (Ser63/73, sc-16312), HNF4α (H-171, sc-8987), SHP (Q-14, sc-15283), and actin (C-11, sc-1615) were from Santa Cruz Biotechnology (Santa Cruz, CA); phospho-c-Met (Tyr 1230/Tyr 1234/Tyr1235) came from Upstate Biotechnology (07-810); and ERK1/2, phospho-ERK1/2, JNK, and phospho-JNK were from Cell Signaling Technology (#9926 and #9910, Beverly, MA).

Transient Transfection and Luciferase Reporter Assay.

For the luciferase reporter assay, HepG2 cells were plated in 24-well plates and transiently transfected with reporters or expression plasmids as described previously,24 using LipofectAMINE 2000 reagent (Life Technologies Inc., Gaithersburg, MD).

Small Interfering RNA Assays.

The SMART pool of small interfering RNA (siRNA) for human c-Met, small heterodimer partner (SHP), and cyclophilin B were purchased from Dharmacon Research (Lafayette, CO) and transfected into HepG2 cells using LipofectAMINE 2000 reagent. Forty-eight hours after transfection, cells were extracted, and the mRNA levels of c-Met, CYP7A1, SHP, c-Jun, and HNF4α were analyzed by Q-PCR.

Chromatin Immunoprecipitation Assay.

Chromatin immunoprecipitation (ChIP) assays were performed using a ChIP Assay kit (Upstate Cell Signaling Solutions, Lake Placid, NY) according to the manufacturer's instructions. PHHs were treated with HGF (50 ng/mL) for various times, and chromatins were crosslinked in 1% formaldehyde and sonicated.24 Six micrograms each of HNF4α, c-Jun, cAMP response element binding protein (CREB)–binding protein (CBP; A-22, sc-369, Santa Cruz, CA), and peroxisome proliferators activated receptor γ coactivator 1α (PGC-1α; P-19, sc-5815, Santa Cruz, CA) antibodies or 20 μg of anti-SHP antibody was added to precipitate DNA–protein complexes, and nonimmune IgG (sc-2028, Santa Cruz, CA) was used as a control. A 391-bp DNA fragment (−432 to −41) containing BARE-I and BARE-II of the CYP7A1 promoter was PCR-amplified for 30 cycles using 5 μL of the DNA as a template and analyzed on a 1.5% agarose gel. A 394-bp sequence in intron 2 (+2485 to +2879) was amplified as a negative control.

Statistical Analyses.

Statistical analyses were performed by 1-way ANOVA followed by Dunnett's test to determine which groups were significantly different from the control group. Two-group comparisons were performed with the unpaired Student t test. Differences with a P value < 0.05 were considered significant.

Results

HGF Repressed CYP7A1 and Total Bile Acid Synthesis Rate in PHHs.

Treatment of PHHs with HGF (50 ng/mL) caused a rapid decrease of 95% of CYP7A1 mRNA expression within 6 hours (Fig. 1A). HGF dose-response was studied in cells treated with HGF for 6 hours. Figure 1B shows that at a physiological concentration of 5 ng/mL HGF reduced CYP7A1 mRNA expression by about 70%. Cells were treated with HGF (50 ng/mL), and bile acids released into culture medium were analyzed at different times. Figure 1C shows that HGF reduced the rate of bile acid synthesis by about one-third after 6 hours of treatment.

Figure 1.

HGF repressed CYP7A1 mRNA expression and bile acid synthesis in PHHs. (A) PHHs were treated with HGF (50 ng/mL) for the time indicated. (B) PHHs were treated with HGF in the concentrations indicated for 6 hours. Total RNA was isolated for real-time quantitative PCR analysis of CYP7A1 mRNA expression. Data represent mean ± SD of at least 3 individual experiments. (C) PHHs were treated with HGF (50 ng/mL) for the time indicated, and culture media were collected for analysis of the bile acid synthesis rate. Data represent 1 of 3 independent experiments; *P < 0.05, **P < 0.001, statistically significant difference between HGF-treated PHHs and untreated controls.

SHP and c-Jun Were Involved in HGF-Mediated Repression of CYP7A1.

To identify the signaling pathway that mediates HGF inhibition of the CYP7A1 gene, we first studied the effect of HGF on the expression of several transcription factors involved in the regulation of the CYP7A1 gene, including SHP, c-Jun, and HNF4α. Figure 2A shows that HGF (50 ng/mL) rapidly induced the c-Met mRNA transcript in 3 hours (3-fold) and maintained it at a steady-state level to 24 hours. HGF treatment caused a rapid induction of c-Jun (Fig. 2B) and SHP (Fig. 2C) mRNA expression, by 4-fold and 3-fold, respectively, within 1 hour. Expression of c-Jun and SHP declined with longer treatment. However, we did not observe an effect of HGF on HNF4α mRNA expression (Fig. 2D). We also analyzed the effect of HGF on the mRNA levels of coactivators PGC-1α (Fig. 2E) and CBP (Supplementary Fig. 1A). HGF treatment caused a 45% decrease in PGC-1α mRNA expression after 3 hours, but the mRNA levels returned to control levels after 24 hours. However, treatment with HGF had no significant effect on the expression of CBP mRNA.

Figure 2.

mRNA expression of c-Met, SHP, HNF4α, PGC-1α, and c-Jun in PHHs treated with HGF. PHHs were treated with HGF (50 ng/mL) for the time indicated. Total RNA was isolated for Q-PCR analysis of mRNA expression of (A) c-Met, (B) c-Jun, (C) SHP, (D) HNF4α, and (E) PGC-1α. Data represent mean ± SD of at least 3 individual experiments; *P < 0.05, **P < 0.001, statistically significant difference between HGF-treated PHHs and untreated controls.

To study whether HGF induces c-Jun and SHP protein expression, immunoblotting analysis with antibodies was performed. Figure 3A shows that HGF induced c-Met, c-Jun, and SHP protein levels in PHHs. However, HNF4α protein levels were not affected by HGF. These data suggest that HGF may induce c-Jun and SHP to repress CYP7A1 gene expression in liver.

Figure 3.

(A) Western blot analysis of protein expression in PHHs by HGF. PHHs were treated with HGF (50 ng/mL) for the time, and whole-cell extracts (100 μg/lane) were then analyzed by immunoblotting with c-Met, c-Jun, SHP, HNF4α, and actin antibodies. Data are representative of 2 independently performed experiments. (B) PHHs were treated with HGF (50 ng/mL) or CDCA (25 μM) for the time indicated. Total RNA was isolated for Q-PCR analysis of BSEP mRNA expression. (C) PHHs were pretreated for 1 hour with cyclohexamide and then with HGF (50 ng/mL) for 6 hours (CHX/HGF). Cells treated with HGF (50 ng/mL) or CHX alone were controls. Total RNA was isolated for Q-PCR analysis of CYP7A1 mRNA expression. Data represent mean ± SD of at least 3 individual experiments; *P < 0.05, **P < 0.001, statistically significant difference between HGF-treated PHHs and untreated controls.

We tested the effect of HGF on the bile acid/FXR–induced gene bile salt export pump (BSEP) in PHHs. HGF treatment (50 ng/mL) had no significant effect on BSEP mRNA expression in PHHs, whereas CDCA treatment highly induced BSEP mRNA expression as expected (Fig. 3B). We also investigated the effect of HGF on mRNA levels of NTCP and PEPCK, which are known SHP target genes in hepatocytes (Supplementary Fig. 1B,C). HGF gradually decreased NTCP mRNA, suggesting that HGF also inhibits hepatic bile acid uptake to prevent toxic bile acids accumulating. HGF also rapidly decreased PEPCK mRNA expression, by 90%, within 6 hours, but its mRNA level increased to 50% of the control levels after 24 hours.

To determine whether de novo protein synthesis is required for HGF-mediated CYP7A1 repression, PHHs were pretreated with cycloheximide (CHX, 50 μg/mL), a protein synthesis inhibitor, followed by treatment with HGF. At this concentration, CHX had no toxic effect on PHHs, as determined by trypan blue exclusion (data not shown). As shown in Fig. 3C, the addition of CHX before HGF treatment prevented HGF-mediated repression of CYP7A1 mRNA, suggesting that de novo protein synthesis was involved in HGF-mediated repression of CYP7A1 gene expression in PHHs.

HGF Repressed CYP7A1 Expression via PKC, ERK, and JNK Pathways.

To evaluate the potential signal pathways involved in HGF repression of CYP7A1, PHH cells were pretreated with several specific inhibitors of cell-signaling pathways before adding HGF (50 ng/mL). Q-PCR analysis indicated that pretreatment of PD98059 (PD, an inhibitor of MAPK), SP600125 (SP, an inhibitor of JNK), and Ro 31-8220 (Ro, an inhibitor of protein kinase C) prevented HGF repression of CYP7A1 mRNA expression (Fig. 4A). However, there was no significant effect of H89 (an inhibitor of protein kinase A), SB-203580 (SB, an inhibitor of p38 kinase), Wortmannin (WM, an inhibitor of PI3 kinase), LY294002 (LY, an inhibitor of PI3 kinase), and Akt inhibitor IV (AktI, an inhibitor of Akt) on CYP7A1 mRNA expression. These results suggest that HGF inhibits CYP7A1 gene expression via the PKC, ERK1/2, and JNK pathways.

Figure 4.

Effects of protein kinase inhibitors on CYP7A1 mRNA expression in HGF-treated PHHs. (A) PHHs were treated with protein kinase inhibitors H-89 (H89, 10 μM), PD98059 (PD, 20 μM), SB-203580 (SB, 25 μM), SP600125 (SP, 25 μM), Wortmannin (WM, 0.1 μM), LY294002 (LY, 50 μM), Akt inhibitor IV (AktI, 10 μM), or Ro 31-8220 (Ro, 1 μM) for 1 hour and then with HGF (50 ng/mL) for 6 hours. Total RNA was isolated for Q-PCR analysis of CYP7A1 mRNA expression. Data represent mean ± SD of at least 3 individual experiments; *P < 0.05, **P < 0.001, statistically significant difference in the expression of the indicated kinase inhibitor between treated PHHs and untreated controls (−). (B) PHHs were treated with HGF (50 ng/mL) for an indicated period, and whole-cell extracts (50 μg/lane) were then analyzed by immunoblotting with c-Met, phospho-c-Met, JNK, phospho-JNK, ERK1/2, phospho-ERK1/2, c-Jun, phospho-c-Jun, and actin antibodies. Data are representative of 2 independently performed experiments.

Immunoblotting analysis using antiphospho antibodies revealed that HGF significantly induces tyrosine phosphorylation of c-Met after HGF treatment for 5–120 minutes. HGF increased phosphorylation of ERK1/2 (p-ERK1/2) and JNK (p-JNK) in PHHs as early as 5 minutes after treatment (Fig. 4B). Moreover, phosphorylation of c-Jun was also dramatically increased. Phosphorylation of c-Met, JNK, ERK1/2, and c-Jun remain 24 hours after HGF treatment (Supplementary Fig. 2). These data suggest that the ERK and JNK signal transduction pathways are critical components of HGF-mediated repression of CYP7A1 in PHHs.

Identification of Region Conferring HGF Repression of Human CYP7A1 Gene Promoter.

We next attempted to identify potential sequences conferring HGF repression of the CYP7A1 gene promoter. A series of 5′-deletion constructs of human CYP7A1 luciferase reporters were used in transfection assays. Deletion of the human CYP7A1 promoter sequence from −1178 to −150 did not affect HGF inhibition of reporter activity (Fig. 5). However, further deletion to −135 abolished the repressive effect of HGF, suggesting that the region between −150 and −135 conferred the repressive effect of HGF. Interestingly, this region was previously identified as a BARE-II that contains a DR-1 sequence for the binding of HNF4α.26 To further confirm that this DR-1 site mediates the HGF-mediated repression, mutations were introduced into the DR-1 sequence of the −1887 (M1887) and −298 (M298) reporter. This mutant reporter did not respond to HGF. These results suggest that HGF repression of CYP7A1 gene transcription involves the HNF4α binding site.

Figure 5.

Identification of sequence responsive to HGF in the human CYP7A1 gene. A series of deletion constructs of human CYP7A1-luciferase reporters were transfected into HepG2 cells and treated with HGF (50 ng/mL) or vehicle (PBS) under serum-starved conditions. All experiments were done in duplicate, and data represent mean ± SD of 3 individual experiments; *P < 0.05, **P < 0.001, statistically significant difference between HGF-treated HepG2 cells and untreated controls.

Knockdown c-Met Expression Abolished HGF Effect in HepG2 Cells.

To further confirm the role of c-Met in CYP7A1 gene expression, we transfected a SMART Pool siRNA for c-Met to knock down c-Met in HepG2 cells and analyzed its effect on the mRNA expression of c-Met, c-Jun, SHP, HNF4α, and CYP7A1 by Q-PCR. Immunoblotting analysis and Q-PCR assay showed that siRNA to c-Met decreased c-Met protein expression by about 80% (Fig. 6A) and mRNA expression by about 60% (Fig. 6B). In contrast, siRNA c-Met increased CYP7A1 mRNA expression about 2-fold and attenuated HGF repression of CYP7A1 mRNA (Fig. 6C). Interestingly, siRNA to c-Met prevented the induction of c-Jun and SHP mRNA by HGF (Fig. 6D,E). As a negative control, this siRNA had no effect on HNF4α mRNA expression (Fig. 6F). A control SMART Pool siRNA for cyclophilin B did not affect c-Met and CYP7A1 mRNA expression relative to nontransfected cells (data not shown). Knockdown of SHP mRNA by siRNA to SHP increased CYP7A1 mRNA expression by 50% in the absence of HGF but failed to block HGF repression of CYP7A1 mRNA (Supplementary Fig. 3). These data support the hypothesis that endogenous c-Met signaling strongly inhibits CYP7A1 mRNA via induction of SHP expression in HepG2 cells. However, other SHP-independent pathways activated by HGF may be also involved in HGF inhibition of CYP7A1.

Figure 6.

siRNA knockdown of c-Met increased expression of CYP7A1 mRNA in HepG2 cells. (A) HepG2 cells were transfected with the SMART Pool of siRNA-c-Met (20 and 200 pmol) and control siRNA (cyclophilin B, 200 pmol) at the indicated concentration. The effects of siRNA on c-Met expression were measured by western blot analysis. Actin expression was used as a loading control. Data represent 1 of 3 separate experiments and show that reduction of protein expression varied from 40% to 80%. Total RNA was isolated for Q-PCR analysis of mRNA expression of (B) c-Met, (C) CYP7A1, (D) c-Jun, (E) SHP, and (F) HNF4α. Data show mRNA expression of siRNA-c-Met-treated (200 pmol) samples relative to the control-treated (200 pmol) siRNA samples. Data represent mean ± SD of at least 3 individual experiments; *P < 0.05, statistically significant difference between siRNA c-Met and control; #P < 0.05, statistically significant difference between HGF and siRNA c-Met + HGF.

Molecular Mechanism of HGF-Mediated Repression of CYP7A1 Expression.

To investigate the molecular mechanism by which the HGF/SHP pathway represses CYP7A1 gene expression, we first assessed the involvement of histone deacetylases (HDACs) in HGF inhibition of CYP7A1 using an HDAC inhibitor, trichostatin A (TSA). Figure 7 shows that TSA (100 nM) treatment neither affected the basal expression nor the HGF-mediated repression of CYP7A1 mRNA expression in PHHs. These results suggest that HGF-mediated repression of CYP7A1 mRNA does not require recruitment of HDACs.

Figure 7.

Effect of TSA on HGF inhibition of CYP7A1. Primary human hepatocytes were treated for 1 hour with TSA (100 nM) and then with HGF (50 ng/mL) for 6 hours. Total RNA was isolated for Q-PCR analysis of expression of (A) CYP7A1 mRNA and (B) C/EBPα mRNA. Data represent mean ± SD of at least 3 individual experiments; **P < 0.001, statistically significant difference between HGF-treated PHHs and untreated controls.

Next, we analyzed the effect of HGF on the recruitment of HNF4α, c-Jun, SHP, CBP, and PGC-1α to the CYP7A1 promoter by the ChIP assay. PHHs were treated with HGF, and cell extracts were immunoprecipitated with specific antibodies against HNF4α, c-Jun, SHP, CBP, and PGC-1α. ChIP assay results (Fig. 8A) showed that HNF4α was present in the CYP7A1 promoter and that HGF treatment did not affect its binding to CYP7A1 chromatin. Without HGF treatment, CBP and PGC-1α were recruited to CYP7A1 chromatin, apparently by interaction with HNF4α. Interestingly, CBP and PGC-1α were dissociated from the CYP7A1 promoter after 3 and 6 hours and reassociated 24 hours after HGF treatment. Concomitantly, c-Jun and SHP were recruited to the CYP7A1 promoter 3-6 hours after HGF treatment. An intron 2 sequence from +2485 to +2897 of the CYP7A1 gene was amplified, and nonimmune rabbit IgG was used in ChIP assay as a negative control (Fig. 8B). We also performed the ChIP assay using HepG2 cells treated with c-Met siRNA probes. However, our results showed no effect of c-Met siRNA on c-Jun and SHP recruitment to CYP7A1 chromatin (data not shown). These ChIP assay results provide critical in vivo evidence that coactivator competition by c-Jun and SHP may be one of several mechanisms mediating HGF repression of CYP7A1.

Figure 8.

Effect of HGF treatment on recruitment of SHP, c-Jun, CBP, and PGC-1α to CYP7A1 chromatin. PHHs were treated with HGF (50 ng/mL) for the indicated times, and cells were collected for ChIP assays. Chromatin fragments were prepared by sonication and immunoprecipitated with the indicated specific antibodies. (A) DNA fragment (391 bp) covering BARE-I and BARE-II was PCR-amplified as described in the Materials and Methods section. (B) DNA fragment (394 bp) in intron 2 was amplified as a negative control. Data represent 1 of 3 separate experiments.

Discussion

After partial hepatectomy and during the early stage of liver regeneration, CYP7A1 gene expression is repressed and HGF blood level is sharply increased.6, 7, 9 These observations suggest that bile acid synthesis is rapidly inhibited in response to liver injury and that HGF may play an important role in the regulation of bile acid synthesis during liver injury. In the present study, we demonstrated that HGF rapidly and strongly repressed CYP7A1 gene expression in a time- and dose-dependent manner in PHHs. Furthermore, we have revealed that HGF specifically phosphorylates and activates PKC, ERK1/2, JNK, and c-Jun and provided evidence that HGF signaling induces SHP expression and recruitment of c-Jun and SHP to the CYP7A1 promoter. Our results also revealed that the HGF/c-Met-activated signaling pathways including PKC, ERK1/2, and JNK were involved in mediating HGF inhibition of CYP7A1 gene transcription.

A recent study showed that SHP mRNA was induced during liver regeneration.6 However, CYP7A1 mRNA was repressed in FXR-null mice, and cholic acid feeding did not increase HGF levels in mouse livers during liver regeneration, suggesting that HGF inhibits CYP7A1 via FXR-independent pathways. Moreover, we observed an inverse relationship between c-Jun/SHP and CYP7A1 mRNA expression in HGF-treated PHHs. The c-Jun and SHP inhibition of CYP7A1 expression apparently requires de novo protein synthesis. Our results suggest that SHP may be an early-response gene in liver injury.

Diverse signaling pathways have been shown to regulate CYP7A1 gene expression. It has been reported that bile acids mimic phorbol esters by activating PKC, which inhibits CYP7A1.27 Our data that a specific inhibitor of PKC blocked HGF-mediated CYP7A1 repression also support involvement of the PKC pathway in HGF-mediated CYP7A1 repression. These findings are reminiscent of a previous report that HGF increases the concentration of intracellular diacylglycerol, an essential cofactor for PKC, through a phospholipase C–dependent pathway in cultured rat hepatocytes.28 The specific PKC isoform involved in this process remains to be established.

More recently, our laboratory and others have demonstrated that cytokine signaling activates the JNK pathway and inhibits CYP7A1 and NTCP gene transcription.29–32 The role of the ERK1/2 and JNK pathways in mediating the inhibitory effects of HGF on CYP7A1 expression is supported by the observation that HGF rapidly increases phosphorylation of ERK1/2 and JNK. This is consistent with a previous report that HGF stimulates phosphorylation of JNK and ERK1/2 and AP-1 DNA-binding activity in hepatocytes.33 A study using SHP-null mice suggested that SHP itself may be a component of the JNK-signaling cascade34 and that the SHP promoter is transactivated by c-Jun.29, 35 Moreover, our recent study revealed that JNK can phosphorylate HNF4α and that c-Jun interacts with HNF4α to interfere with HNF4α recruitment of PGC-1α to CYP7A1 chromatin, thereby leading to repression of CYP7A1 expression.31

Our ChIP assays clearly showed that c-Jun and SHP induced by HGF were recruited to CYP7A1 chromatin. These 2 factors may block HNF4α recruitment of CBP and PGC-1α to the chromatin and result in inhibition of CYP7A1 gene transcription. Competition for binding and squelching of the limited coactivators could be a common mechanism for the negative regulation of gene transcription by nuclear receptors.36 It has been reported that c-Jun and SHP have the ability to displace coactivators from HNF4α.31, 37 It also has been suggested that SHP functionally interacts with HDAC1 to repress target genes.38 However, we did not observe HDAC1 in the CYP7A1 chromatin with or without HGF treatment (data not shown). Interestingly, HGF also decreased PGC-1α mRNA expression. In accordance with this scenario, increasing the amounts of c-Jun and SHP and decreasing PGC-1α by HGF may interfere with coactivator-mediated HNF4α activity on CYP7A1 transcription. Conversely, decreasing the amounts of c-Jun and SHP may activate HNF4α activity by increasing its recruitment of coactivators. Consistent with our observation, a recent microarray gene-profiling study showed that CYP7A1 gene expression was reduced in HGF-transgenic mice.39 However, in contrast to our results, SHP gene expression also was reduced in this mouse model. It should be noted that these investigators did not perform RNA assays to confirm their microarray data on CYP7A1 and SHP gene expression.

In conclusion, this study has shown that HGF is a novel regulator of CYP7A1 expression and bile acid synthesis in human hepatocytes. HGF increases expression of c-Jun and SHP, which interact with HNF4α to block its recruitment of coactivators, resulting in inhibition of the CYP7A1 gene. Other HGF-activated signaling pathways including PKC, ERK, and JNK also are involved in sustained HGF inhibition of CYP7A1 gene expression and bile acid synthesis during liver injury and regeneration. From the results of this study, we propose that increasing serum HGF levels during liver regeneration or liver injury may reduce bile acid synthesis and bile acid pool size in human liver. The increased HGF levels control the rate of bile acid synthesis to protect hepatocytes from accumulating toxic bile acids and developing intrahepatic cholestasis during liver regeneration and in various hepatic disease states

Ancillary