Definition of a novel growth factor-dependent signal cascade for the suppression of bile acid biosynthesis

Abstract

The nuclear bile acid receptor FXR has been proposed to play a central role in the feedback repression of the gene encoding cholesterol 7 alpha-hydroxylase (CYP7A1), the first and rate-limiting step in the biosynthesis of bile acids. We demonstrate that FXR directly regulates expression of fibroblast growth factor-19 (FGF-19), a secreted growth factor that signals through the FGFR4 cell-surface receptor tyrosine kinase. In turn, FGF-19 strongly suppresses expression of CYP7A1 in primary cultures of human hepatocytes and mouse liver through a c-Jun N-terminal kinase (JNK)-dependent pathway. This signaling cascade defines a novel mechanism for feedback repression of bile acid biosynthesis and underscores the vital role of FXR in the regulation of multiple pathways of cholesterol catabolism in the liver.

Figure 1.

FXR-dependent induction of FGF-19. Primary cultures of human hepatocytes were treated for 48 h with the indicated concentrations of CDCA (1–100 μM) or GW4064 (0.1–5 μM). Control cultures received vehicle (0.1% DMSO) alone. Total RNA was isolated and Northern analysis performed using cDNA probes for FGF-19, SHP, CYP7A1, and G3PDH.

Figure 2.

Identification of an FXR response element in the FGF-19 gene. (A) Structure of the FGF-19 gene showing the position of the intron–exon boundaries, restriction endonuclease sites used in the generation of reporter gene constructs, and the sequence of the FXR response element located in intron 2. Exons are shown as solid bars and all coordinates are relative to the FGF-19 transcription initiation site. (B) Delineation of an FXR response element. Three BamHI fragments encompassing the entire FGF-19 structural gene were inserted downstreamof the reporter gene in the pGL3-tk-Luc, which contains a minimal thymidine kinase (tk) promoter (bases –105 to +51, solid box) linked to luciferase reporter gene. Site-directed mutagenesis of the putative FXRE was performed as described in Materials and Methods to produce pGL3-tk-FGF19mut(7899–848). All constructs were transfected in HuH7 cells and exposed to GW4064 (1 μM) for 24 h prior to harvest and determination of luciferase and SPAP activities. Luciferase values are normalized to SPAP and expressed as fold activation compared with control cells transfected with the same reporter, but not receptor, and which received vehicle (0.1% DMSO) alone. Data represent the mean ± S.E.M. from eight individual transfections.

Figure 3.

The FXR response element in Intron 2 binds FXR–RXRα. The ability of FXR–RXRα-heterodimers to bind the IR-1 motif in intron 2 of the FGF-19 gene was examined by EMSA as outlined in Materials and Methods. (A) Sequence of the oligonucleotide probes (sense strand only) corresponding to a previously characterized FXR-binding site in the human I-BABP promoter (hI-BABP IR1), the putative FXR response element from intron 2 of FGF-19 (FGF-19 IR1), and a mutated derivative of this site (FGF-19mut IR1, mutated bases are shown in lowercase). (B) EMSA was performed with recombinant human FXR and/or RXRα as indicated and radiolabeled h-IBABP IR1 or FGF-19 IR1 (lanes 1–3,4–6, respectively). Competition EMSA (lanes 7–15) was performed using in vitro synthesized FXR and RXRα, radiolabeled hI-BABP IR1, and the indicated molar excess of unlabeled hI-BABP IR1 (lanes 7–9), FGF-19 IR1 (lanes 10–12), or FGF-19mut IR1 (lanes 13–15). The position of the shifted FXR–RXRα complex and free probes are indicated.

Figure 4.

FGF-19 suppresses expression of CYP7A1 in primary cultures of human hepatocytes and in vivo. (A) Primary cultures of human hepatocytes were treated with the indicated concentration of recombinant FGF-19 for 6 h prior to harvest and isolation of total RNA. Northern blot analysis of CYP7A1, SHP, and β-actin expression was performed as described in Materials and Methods. Results froma single hepatocyte donor are shown, but are typical of three separate preparations. (B) Total RNA was isolated from the livers of adult male FVB mice injected intravenously with saline (200 μL), or 2 × 10¹¹ units of either null adenovirus or FGF-19-expressing adenovirus as outlined in Materials and Methods. Northern analysis was performed sequentially with cDNA probes corresponding to FGF-19, mouse CYP7A1, mouse SHP, and G3PDH.

Figure 5.

Activation of the JNK signaling cascade by FGF-19 causes suppression of CYP7A1. (A) Primary cultures of human hepatocytes were treated with FGF-19 (80 ng/mL) alone or in combination with the JNK inhibitor SP600125 (10 μM). SP600125 was added to the culture medium as a 1000× stock in DMSO 30 min before the addition of recombinant FGF-19. Control cultures received vehicle (0.1% DMSO) alone. Cells were cultured for a further 6 h prior to harvest, and Northern blot analysis of CYP7A1 and β-actin expression. (B) FGF-19 activates the JNK-signaling cascade. Total cell lysates were prepared from hepatocytes treated for 3 h as described above. The level of activated c-Jun was analyzed by immunoblotting using antibodies specific for Ser 63-phosphorylated c-Jun. Data are representative of a least three individual hepatocyte preparations.

Figure. 6.

Role of FXR in the suppression of CYP7A1. Activation of FXR by bile acids leads to induction of the FGF-19 gene. FGF-19 is secreted fromthe hepatocyte and can signal to other hepatocytes through its cell-surface receptor FGFR4. This results in suppression of CYP7A1 expression through a JNK-dependent signaling cascade. FXR can also activate expression of the SHP gene, which can directly suppress the CYP7A1 gene through an interaction with LRH-1 (Goodwin et al. 2000; Lu et al. 2000). SHP itself may be a component of the JNK-signaling pathway (Wang et al. 2002).