Notice: Wiley Online Library will be unavailable on Saturday 30th July 2016 from 08:00-11:00 BST / 03:00-06:00 EST / 15:00-18:00 SGT for essential maintenance. Apologies for the inconvenience.
Potential conflict of interest: Nothing to report.
Fibroblast growth factor 19 (FGF19) is an endocrine factor produced by the small intestine in response to uptake of luminal bile salts. In the liver, FGF19 binds to FGF receptor-4, resulting in down-regulation of cytochrome P (CYP) 7A1 and reduced bile salt synthesis. Down-regulation of CYP7A1 under cholestatic conditions has been attributed to bile salt–mediated induction of the transcriptional repressor short heterodimer partner (SHP), because the interrupted enterohepatic cycle of bile salts is thought to abrogate intestinal FGF19 production and thus result in lowering of plasma FGF19 levels. Unexpectedly, we observed marked elevation of plasma FGF19 in patients with extrahepatic cholestasis caused by a pancreatic tumor (2.3 ± 2.3 in cholestatic versus 0.40 ± 0.25 ng/mL and 0.29 ± 0.12 ng/mL in postcholestatic patients who received preoperative drainage by biliary stenting, P = 0.004, and noncholestatic control patients, P = 0.04, respectively). Although FGF19 messenger RNA (mRNA) is virtually absent in normal liver, FGF19 mRNA was strongly increased (31-fold to 374-fold, P < 0.001) in the liver of cholestatic patients in comparison with drained and control patients. In the absence of changes in SHP mRNA, CYP7A1 mRNA was strongly reduced (7.2-fold to 24-fold, P < 0.005) in the liver of cholestatic patients in comparison with drained and control patients, indicating an alternative regulatory pathway. Alterations in transcripts encoding hepatobiliary transporters [adenosine triphosphate–binding cassette, subfamily C, member 3 (ABCC3)/multidrug resistance protein 3 (MRP3), organic solute transporter α/β (OSTα/β), organic anion-transporting polypeptide (OATP1B1)] further suggest that bile salts are secreted via a nonbiliary route in patients with extrahepatic cholestasis. Conclusion: The liver expresses FGF19 under conditions of extrahepatic cholestasis. This is accompanied by a number of adaptations aimed at protecting the liver against bile salt toxicity. FGF19 signaling may be involved in some of these adaptations. (HEPATOLOGY 2009.)
If you can't find a tool you're looking for, please click the link at the top of the page to "Go to old article view". Alternatively, view our Knowledge Base articles for additional help. Your feedback is important to us, so please let us know if you have comments or ideas for improvement.
Because they are potent detergents, synthesis of bile salts is subject to rigorous regulation.1, 2 The principal target for control of bile salt synthesis is the cytochrome P (CYP) 7A1 gene, which encodes the rate-determining enzyme in the dominant biosynthetic pathway. Regulation of CYP7A1 occurs primarily at the transcriptional level and involves several nuclear hormone receptors. Among these ligand-activated transcription factors, the bile salt receptor FXR (farnesoid-X receptor) plays a key role in bile salt–mediated repression of CYP7A1.2–4 Activation of hepatic FXR induces expression of short heterodimer partner (SHP), a transcriptional repressor that diminishes the transactivation potential of several transcription factors required for efficient CYP7A1 expression.3–5 Activation of intestinal FXR by reabsorbed bile salts induces expression and portal release of FGF19 (fibroblast growth factor 19, termed Fgf15 in rodents).6, 7 Binding of FGF19/Fgf15 to the cell surface receptor FGFR4 results in activation of mitogen-activated protein kinase pathways and down-regulation of CYP7A1.7, 8 Studies in mice with intestine-specific or liver-specific disruption of the Fxr gene revealed that administration of a synthetic FXR agonist failed to repress Cyp7a1 in animals deficient in intestinal Fxr.9 This study thus implied an important role for intestinal Fgf15 in regulation of bile salt synthesis.
Impaired bile formation or bile flow can result in intrahepatocytic accumulation of bile salts and concomitant activation of hepatic FXR. This would result in repression of CYP7A1 via transcriptional induction of SHP. Reduced levels of CYP7A1 in liver of patients with primary biliary cirrhosis (PBC) or biliary atresia, however, were not accompanied by changes in SHP messenger RNA (mRNA), suggestive for the involvement of another regulatory pathway.10, 11 Whether FGF19 contributes to regulation of bile salt synthesis under cholestatic conditions is currently unknown. Interruption of the enterohepatic cycle of bile salts likely abrogates intestinal FGF19 expression. Combined with absent expression of FGF19 in normal human liver, plasma FGF19 is thus expected to be lowered in patients with extrahepatic cholestasis.6, 7, 12 In the current study, we sought to determine plasma FGF19 levels in patients with extrahepatic cholestasis, to address the role of FGF19 in regulation of CYP7A1 expression under conditions of extrahepatic cholestasis, and to study adaptive changes in the liver under conditions of extrahepatic cholestasis.
Preoperative liver biopsies (n = 22) were obtained from patients suspected to have a pancreatic or periampullary malignancy. Patients were recruited at the Department of Surgery and were planned to undergo resection (pancreatoduodenectomy) with curative intent. Twenty of 22 patients had obstructive jaundice caused by periampullary tumor growth at initial presentation. Most jaundiced patients participated in an ongoing clinical trial on the effect of preoperative biliary drainage (www.isrctn.org, trial number: ISRCTN31939699).13 Ten jaundiced patients underwent surgery within 1 week without preoperative biliary drainage (cholestatic group). The remaining jaundiced patients received a biliary stent for a mean duration of 12 ± 5 weeks before surgery (drained group, n = 10). Control liver tissue (control group, n = 10) was obtained from nonjaundiced patients with a pancreatic malignancy (n = 2; neuroendocrine tumor = 1, tumor of the pancreatic head = 1) and from patients undergoing liver resection (n = 8; focal nodular hyperplasia = 2, hemangioma = 3, adenoma = 3). In the latter case, normal liver parenchyme was dissected from the resected liver specimen. All surgical specimens were collected in the morning (8:00 AM to 12:00 noon). Liver specimens were collected in RNAlater (Ambion) or Trizol (Invitrogen) and stored at −80°C until RNA isolation. Patients gave their informed consent to the respective protocols of the studies, which were approved by the local Medical Ethical Committee.
Quantification of Hepatic Transcript Levels.
Total RNA was isolated from liver specimens using Trizol reagent (Invitrogen). After DNAse treatment (Promega), 1.25 μg total RNA was reverse transcribed using oligo-dT priming and SuperscriptII (Invitrogen). First-strand complementary DNA equivalent to 12.5 ng total RNA was used as template for real-time polymerase chain reaction analysis employing SybrGreen chemistry (LightCycler FastStartPlus, Roche) and a LightCycler 2.0 System (Roche, Basel, Switzerland). Melting curve analysis was performed after each run to ensure amplification specificity. Transcript levels, determined in two independent complementary DNA preparations, were calculated as described and expressed relative to the 36B4 housekeeping gene.14, 15 Primer sequences and cycling conditions are available on request.
Determination of Plasma FGF19.
Preoperation plasma samples were available from 16 of 22 patients with a pancreatic malignancy, but were not available from the patients undergoing liver resection. Plasma samples were stored at −80°C until analysis. Plasma FGF19 levels were determined using an in-house developed sandwich enzyme-linked immunosorbent assay specific for FGF19, which will be described in detail elsewhere (Schaap et al., manuscript in preparation). Briefly, microtiter plates were coated with goat anti-human FGF19 antibody (AF969, R&D Systems, Minneapolis, MN). Samples and recombinant FGF19 standards (R&D Systems) were diluted in phosphate-buffered saline containing 1.0% casein and 0.05% Tween-20. Captured antigen was detected with biotinylated goat anti-human FGF19 antibody (BAF969, R&D Systems) and streptavidin-horseradish peroxidase, using tetramethylbenzidine as chromogenic substrate. Bilirubin and bile salt levels encountered in cholestatic plasma samples did not interfere with the quantification of FGF19.
Differences between the three patient groups were evaluated by one-way analysis of variance (ANOVA) and Tukey honest significance difference or Games-Howell post-hoc tests. Values were log transformed if normality (Shapiro-Wilk test) or equality of variance (Levene's test) assumptions were rejected. Nonparametric Kruskal-Wallis testing was used when criteria for ANOVA were not met after log transformation. Fischer's exact test was used for analysis of categorical data. SPSS (version 16.0) was used for all statistical analyses. P < 0.05 was considered significant. Data are expressed as mean ± standard deviations.
Plasma FGF19 Is Elevated in Patients with Extrahepatic Cholestasis.
Plasma samples were collected from noncholestatic patients, postcholestatic patients who received a biliary stent, and from cholestatic not-drained patients with a pancreatic or periampullary malignancy. The patient group characteristics are shown in Table 1. As expected, patients receiving preoperative biliary drainage had normal bilirubin and bile salt levels, near normal levels of transaminases, and had significantly lower alkaline phosphatase at the time of surgery. Plasma FGF19 was found to be eightfold and sixfold higher in the cholestatic group in comparison with control (P = 0.04) and drained (P = 0.004) patients, respectively (Fig. 1, Table 1). Plasma FGF19 levels in the control (0.29 ± 0.12 ng/mL) and drained (0.40 ± 0.25 ng/mL) groups were comparable to levels in a group of 28 fasted volunteers (0.28 ± 0.20 ng/mL) (Fig. 1). In six of seven cholestatic patients, plasma FGF19 levels were above the highest value observed in fasted volunteers. Furthermore, plasma FGF19 levels in four of seven cholestatic plasma samples were still above the highest postprandial levels in normal volunteers (Fig. 1).
Table 1. Demographics and Serum Biochemistry of the Patient Groups
Values are expressed as mean ± standard deviation. Significance was evaluated by Fischer's exact test, ANOVA, or Kruskal-Wallis test. Capital letters indicate significance (P < 0.05) for post-hoc comparisons between control and drained (A), control and cholestatic (B), and drained and cholestatic (C) groups.
FGF19 mRNA Is Expressed in Liver of Patients with Extrahepatic Cholestasis.
Elevated plasma FGF19 in the cholestatic patients prompted us to study expression of FGF19 mRNA in the liver, an organ considered to lack FGF19 expression.12 FGF19 mRNA could be detected in 23 of 30 liver specimens with relative expression levels varying over four orders of magnitude. Average threshold cycle (Ct) for detection of FGF19 mRNA by real-time polymerase chain reaction was 43.5, 37.9, and 32.1 for control, drained, and cholestatic groups, respectively. FGF19 mRNA was not detected in seven of 10 control liver specimens, in line with the reported absence of FGF19 in normal adult liver.7, 12 Average expression of FGF19 mRNA was 31-fold and 374-fold higher in the cholestatic group in comparison with drained (P < 0.001) and control groups (P < 0.001), respectively (Fig. 1, Table 2). Although FGF19 mRNA was readily detected in liver biopsy specimens of all drained patients, the expression level appeared biologically trivial because plasma FGF19 levels in the drained group were similar to control levels. Tumoral FGF19 mRNA expression has been observed in cases of primary liver and colon carcinoma, with unknown impact on circulating FGF19 levels.16 The observation of normal plasma FGF19 levels in control and drained patients with a pancreatic malignancy indicates that the presence of a pancreatic tumor per se had no effect on circulating FGF19 levels.
Table 2. Hepatic Expression of Transcripts Engaged in FGF19 Signaling and Aspects of the Adaptive Response to Cholestasis
Normalized transcript levels are expressed relative to the mean level in the control group and are given as mean ± standard deviation (SD). Ct denotes the mean threshold cycle in the control group, the mean Ct value for the 36B4 reference gene in controls is 22.1. Overall significance (PALL) was evaluated by ANOVA or Kruskal-Wallis tests. Capitals indicate significance (P < 0.05) for post-hoc comparisons between control and drained (A), control and cholestatic (B), and drained and cholestatic (C) groups.
FGF19 exerts its effects by binding to a cell surface receptor. Hepatic mRNA level of the FGF19-receptor FGFR4 was slightly elevated (+1.7-fold, P = 0.014) in the cholestatic group in comparison with the drained group (Supporting Fig. 1, Table 2). Messenger RNA level of the FGF19-signaling cofactor βKlotho was similar in all groups (Supporting Fig. 1, Table 2).7, 17
Expression of Bile Salt Synthetic Genes in Liver of Patients with Extrahepatic Cholestasis.
Retention of toxic biliary components under cholestatic conditions initiates a response aimed at minimizing hepatocellular damage.18–20 Hepatic expression of four genes engaged in bile salt synthesis was determined to examine potential alterations in bile salt synthesis or composition. The mRNA level of CYP7A1, the major determinant of bile salt synthesis via the prevailing classical pathway, was 24-fold and 7.2-fold lower in the cholestatic group in comparison with control (P < 0.001) and drained (P = 0.005) groups, respectively (Fig. 2, Table 2). CYP7A1 mRNA was not significantly different between control and drained groups. Considerable, but comparable, variation in CYP7A1 expression was observed in each patient group, causing extreme values to have a relatively large impact on group averages. Using geometrical means, CYP7A1 mRNA expression was 37-fold and 9.8-fold lower in the cholestatic group in comparison with control and drained groups, respectively.
Messenger RNA levels of CYP27A1 and CYP7B1, whose gene products are both engaged in the alternative acidic pathway of bile salt synthesis, were similar in all groups (Supporting Fig. 1, Table 2).2 A trend toward lowered CYP27A1 mRNA in drained and cholestatic groups was apparent (Supporting Fig. 1). The mRNA level of the bile salt hydroxylase CYP8B1 appeared modestly elevated (+1.5-fold, P = 0.014) in the cholestatic group in comparison with the drained group (Fig. 2, Table 2).2
Expression of Hepatobiliary Transporters in Liver of Patients with Extrahepatic Cholestasis.
Messenger RNA levels of the major hepatic transport systems for bile salts, bilirubin, and their conjugates were determined to study possible adaptive changes in hepatobiliary transport of these compounds. Basolateral uptake of bile salts is accomplished via Na+-dependent (Na/Taurocholate cotransporting polypeptide [NTCP]) and Na+-independent (organic anion-transporting polypeptide [OATP]) transporters.18–20 NTCP mRNA level was similar in all groups (Supporting Fig. 1, Table 2). The mRNA level of OATP1B1, which mediates the bulk of Na+-independent bile salt uptake and additionally mediates uptake of unconjugated bilirubin and bilirubin monoglucuronide, was lower in the cholestatic group in comparison with control (−2.2-fold, P = 0.002) and drained (−2.4-fold, P = 0.001) groups (Fig. 3, Table 2).
Both adenosine triphosphate (ATP)–dependent (ATP binding cassette, subfamily C, member 3 [ABCC3]/multidrug resistance-realted protein 3 [MRP3] and ABCC4/MRP4) and ATP–independent (organic solute transporter α/β [OSTα/β]) transporters are engaged in the basolateral efflux of bile salts and their conjugates.18–20 ABCC3/MRP3 mRNA was slightly elevated in the cholestatic group in comparison with control (+1.7 fold, P = 0.002) and drained (+1.5 fold, P = 0.015) groups (Supporting Fig. 1, Table 2). ABCC4/MRP4 mRNA was not significantly different between the groups, although levels tended to be elevated in the cholestatic group in comparison with the control group (P = 0.059, Supporting Fig. 1, Table 2). A notable elevation of OSTα (+2.4-fold to 3.4-fold, P < 0.001) and OSTβ (+22-fold to 67-fold, P < 0.001) mRNA levels was noted in the cholestatic group in comparison with drained and control groups (Fig. 3, Table 2).
Canalicular transporters are engaged in biliary secretion of bile salts (ATP binding cassette, subfamily B, member 11 [ABCB11]/bile salt export pump [BSEP] and ABCC2/MRP2), bilirubin (ABCC2/MRP2), and phospholipids (ABCB4/multidrug resistance 3 [MDR3]).18–20 Whereas ABCB11/BSEP and ABCC2/MRP2 mRNA levels were similar in all groups, ABCB4/MDR3 mRNA was elevated in the cholestatic group in comparison with drained (+2.5-fold, P = 0.005) and control (+3.2-fold, P = 0.001) groups (Fig. 2, Table 2).
Expression of Nuclear Receptors in Liver of Patients with Extrahepatic Cholestasis.
Nuclear receptors bring about many of the transcriptional changes that underlie adaptation of the liver to cholestasis.18, 20, 21 Expression of nine nuclear receptors engaged in different aspects of the hepatic response to cholestasis was examined. Except for a minor reduction of liver X receptor alpha in the drained group in comparison with the control group (−1.4-fold, P = 0.032) and a trend toward lower levels of FXR mRNA in the cholestatic group in comparison with control (−1.4-fold, P = 0.075) and drained (−1.6-fold, P = 0.058) groups (Supporting Fig. 1), mRNA levels of constitutive androstane receptor, hepatocyte nuclear factor 4α, liver-receptor homolog 1, peroxisome proliferator-activated receptor alpha, pregnane X receptor, retinoid X receptor alpha, and SHP were similar in all groups. The mRNA level of the transcriptional co-activator peroxisome proliferator-activated receptor gamma coactivator 1 alpha (PGC1α) was lower (−2.1-fold, P = 0.01) in the cholestatic group in comparison with the drained group.
Modulation of Hepatic Gene Expression Under Conditions of Extrahepatic Cholestasis.
In the preceding sections, ANOVA was used to test for differences in transcript levels among control, drained, and cholestatic groups (Table 2). To scrutinize genes whose expression is altered under the studied conditions of extrahepatic cholestasis, ANOVA with contrast testing was performed. For this analysis, the two noncholestatic groups (in other words, the control and drained groups) were compared with the cholestatic group. Clinical criteria (total bilirubin <17 μmol/L) justify the treatment of the drained group as a noncholestatic group. Equal weights were assigned to the contrast values for the control and drained groups. The results of the analysis are shown in Table 3 and Supporting Table 1. All major findings obtained by noncontrasted ANOVA were recapitulated by contrast tests; in other words, the expression of FGF19, CYP7A1, CYP8B1, ABCB4/MDR3, ABCC3/MRP3, OSTα, OSTβ, and OATP1B1 is significantly altered under the studied conditions of extrahepatic cholestasis. In addition, significant reductions of βKlotho, FXR, PGC1α, and retinoid X receptor alpha mRNA level in the liver of patients with extrahepatic cholestasis were uncovered.
Table 3. Hepatic Transcripts Significantly Modulated Under Conditions of Extrahepatic Cholestasis
Alteration of gene expression under conditions of extrahepatic cholestasis was tested by ANOVA with contrasts between noncholestatic (control and drained groups) and cholestatic groups. The fold change was calculated by dividing mean expression level in the cholestatic group (n = 10) and mean expression level in the noncholestatic group (n = 20).
In patients with malignancy of the pancreatic head or periampullary region, expansion of the tumor mass often leads to bile duct obstruction. The accumulation of biliary constituents in the liver triggers an adaptive response that aims to minimize hepatocellular damage by reducing de novo bile salt synthesis and promoting nonbiliary secretion routes. In the current study, we found indications for decreased bile salt synthesis (decreased CYP7A1 mRNA), decreased bile salt toxicity in the bile ducts (increased ABCB4/MDR3 mRNA), and enhanced hepatocellular bile salt efflux (decreased OATP1B1 mRNA, increased ABCC3/MRP3, OSTα, OSTβ mRNA) in the liver of patients with extrahepatic cholestasis. These findings extend observations in patients with other forms of cholestasis, such as PBC and biliary atresia.10, 11, 18 However, an important novel aspect of the current work is the finding that plasma FGF19 is markedly elevated (six-fold to eight-fold) in patients with extrahepatic cholestasis.
FGF19 is produced by enterocytes in the terminal ileum in response to reabsorbed bile salts. FGF19 is not expressed in normal liver.7, 12 However, we found high levels of FGF19 mRNA in liver of patients with extrahepatic cholestasis (Fig. 1). In view of the interrupted enterohepatic circulation in these patients, the contribution of the ileum to the plasma FGF19 pool is expected to be minimal.6 Hence, it is conceivable that the strongly up-regulated hepatic expression of FGF19 mRNA accounts for the observed increase in plasma FGF19 in these cholestatic patients (Table 1). Patients who received a biliary stent to restore normal bile flow and normal bile salt levels had substantially lower levels of hepatic FGF19 mRNA, and their plasma FGF19 levels appear comparable to those in control patients and fasted volunteers. Although these clinical studies do not allow us to exactly define the mechanism behind the up-regulation of FGF19 mRNA in patients with a chronic obstruction of the biliary tree, it is likely that the increased serum bile salt levels in these patients activate FXR in the liver and that this induces FGF19 expression. Unlike in human subjects with extrahepatic cholestasis, we could not detect expression of the FGF19-orthologue Fgf15 in the liver of mice with extrahepatic cholestasis, that is, bile duct–ligated mice (data not shown). Thus, mice appear inappropriate to delineate the mechanism behind induction of hepatic FGF19 under conditions of extrahepatic cholestasis.
Expression of CYP7A1, a key determinant of bile salt synthesis, is positively regulated by liver-receptor homolog 1, hepatocyte nuclear factor 4α and the transcriptional coactivator PGC1α.2–4 The transcriptional repressor SHP exerts negative regulation of CYP7A1 expression by diminishing the transactivation potential of hepatocyte nuclear factor 4α and liver-receptor homolog 1.3, 4 Although mRNA for FXR and its transcriptional coactivator PGC1α are somewhat lower in the cholestatic group (Table 3, Supporting Fig. 1), the elevated expression of FXR-target genes (FGF19, MDR3/ABCB4, OSTα, OSTβ) indicate effective FXR activation in the liver of cholestatic patients (Supporting Fig. 1, Table 2).22 Nonetheless, hepatic mRNA level of the FXR-target SHP was not elevated. Despite this lack of SHP induction, CYP7A1 mRNA level was drastically lower (7.2-fold to 24-fold) in the cholestatic group in comparison with drained and control groups. FGF19 is a newly recognized negative regulator of CYP7A1 expression.6–9 FGF19 appears to repress CYP7A1 independent of changes in SHP mRNA.6, 8 Elevation of plasma FGF19 in the cholestatic patients may thus have accounted for the observed repression of CYP7A1. Intriguingly, the lack of hepatic Fgf15 mRNA expression in bile duct–ligated mice (data not shown) may underlie the observed up-regulation of Cyp7a1 expression in bile duct–ligated mice.6 Molecular mechanisms for repression of CYP7A1 by FGF19 are still incompletely understood and require additional studies.
Apart from lowered levels of CYP7A1 mRNA, changes in hepatobiliary transporter expression were apparent in livers of the cholestatic patients. Reduced basolateral uptake of bile salts in liver of cholestatic patients is suggested by lowered levels of OATP1B1 mRNA. Although the FXR/SHP pathway has been implicated in repression of OATP1B1, the absence of SHP mRNA induction in the liver of cholestatic patients leaves the possibility that FGF19 signaling is involved in the observed down-regulation of OATP1B1.18, 21 Along with OATP1B1, NTCP is responsible for most bile salts taken up by the liver. No changes in NTCP mRNA level were apparent in the liver of cholestatic patients and in previously described PBC patients, although NTCP mRNA were reported to be lower in patients with inflammatory cholestasis and biliary atresia.10, 23 Combined with reduced basolateral uptake, alterations in transporters promoting basolateral efflux of bile salts (ABCC3/MRP3, OSTα/β) may further reduce the intracellular bile salt load in cholestatic liver and allow bile salts to be secreted via a nonbiliary (renal) route. ABCC3/MRP3 mRNA was modestly elevated in liver of cholestatic patients, whereas induction of OSTα/β mRNA was more pronounced. Similar observations have been reported for patients with PBC and biliary atresia.10, 24, 25
Expression of the phospholipid floppase ABCB4/MDR3 is elevated in the liver of cholestatic patients. This may increase the amount of phosphatidylcholine that is available at the outer canalicular leaflet for mixed micelle formation and thus may represent an adaptive change favoring protection of downstream bile duct epithelium against bile salt toxicity. Similar to observations made in patients with inflammatory cholestasis and PBC, ABCB11/BSEP mRNA was unaltered in the liver of cholestatic patients studied here.25 Maintenance of ABCB11/BSEP expression may reduce bile salt load in the hepatocyte but also may cause a high bile salt concentration in the biliary tree with bile duct proliferation as a possible consequence. No induction of this direct FXR target gene was observed in livers of the cholestatic patients (Supporting Table 1). Further studies are required to shed light on the apparently abrogated up-regulation of some FXR targets (for example, SHP and ABCB11/BSEP) under these cholestatic conditions. Interestingly, lowered levels of FXR mRNA were observed in liver of the cholestatic patients studied here (Table 3) and in patients with biliary atresia.
In summary, we observed strong expression of hepatic FGF19 mRNA and elevated plasma FGF19 in patients with extrahepatic cholestasis. Although prolonged cholestatic conditions appear to limit SHP mRNA induction, persistent up-regulation of hepatic FGF19 expression may allow effective suppression of CYP7A1 by a paracrine mode of action. It will be interesting to learn whether and how FGF19 affects expression of other genes relevant to the adaptive response in cholestasis, and whether FGF19 is involved in the regulation of bile salt homeostasis in chronic cholestatic liver diseases such as PBC and primary sclerosing cholangitis.
The authors thank the surgical teams at our institute and at the centers participating in the DROP trial for collection of liver specimens. We thank Dr. Ruurdtje Hoekstra for generous sharing of resected liver specimens. We thank professor Ulrich Beuers for critical reading of the manuscript.