New insights on the pathogenesis of biliary cirrhosis provided by studies in FXR knockout mice

Abstract

The nuclear bile acid receptor, farnesoid X receptor (FXR), may play a pivotal role in liver fibrosis. We tested the impact of genetic FXR ablation in four different mouse models. Hepatic fibrosis was induced in wild-type and FXR knock-out mice (FXR^(−/−)) by CCl(4) intoxication, 3,5-diethoxycarbonyl-1,4-dihydrocollidine feeding, common bile duct ligation, or Schistosoma mansoni (S.m.)-infection. In addition, we determined nuclear receptor expression levels (FXR, pregnane X receptor (PXR), vitamin D receptor, constitutive androstane receptor (CAR), small heterodimer partner (SHP)) in mouse hepatic stellate cells (HSCs), portal myofibroblasts (MFBs), and human HSCs. Cell type-specific FXR protein expression was determined by immunohistochemistry in five mouse models and prototypic human fibrotic liver diseases. Expression of nuclear receptors was much lower in mouse and human HSCs/MFBs compared with total liver expression with the exception of vitamin D receptor. FXR protein was undetectable in mouse and human HSCs and MFBs. FXR loss had no effect in CCl(4)-intoxicated and S.m.-infected mice, but significantly decreased liver fibrosis of the biliary type (common bile duct ligation, 3,5-diethoxycarbonyl-1,4-dihydrocollidine). These data suggest that FXR loss significantly reduces fibrosis of the biliary type, but has no impact on non-cholestatic liver fibrosis. Since there is no FXR expression in HSCs and MFBs in liver fibrosis, our data indicate that these cells may not represent direct therapeutic targets for FXR ligands.

The farsenoid X receptor (FXR/NR1H4) is a member of the ligand-activated nuclear receptor superfamily, which also includes the pregnane X receptor (PXR/NR1I2), the constitutive androstane receptor (CAR/NR1I3), and the vitamin D receptor (VDR/NR1I1) [1]. FXR, which acts as a bile acid sensor, is mainly expressed in the liver, intestine, kidney, and adrenal glands [2]. In the liver, it has been identified and characterized as an important regulator of key metabolic pathways involving bile acids, glucose, cholesterol, and lipids, as well as of homeostatic liver functions such as tissue regeneration and inflammatory response to tissue injury [3,4]. Regarding the involvement of FXR receptor in the inflammatory response, it has been shown that FXR-deficient (FXR^−/−) mice subjected to common bile duct ligation (BDL), to mimic exposure to endogenous bile acids, demonstrate a less prominent ductular reaction and reduced biliary infarcts as compared to their wild-type littermates [5]. A 2004 study demonstrated expression of FXR in quiescent and activated rat hepatic stellate cells (HSC), suggesting that HSC may transduce bile acid-mediated signals into scar production [6]. (Please note that here we are using a strict definition of hepatic stellate cells, meaning pericyte-like cells of the hepatic sinusoid that undergo myofibroblastic differentiation in chronic liver injury.) During the development of liver fibrosis, activated or myofibroblastic HSC are, with portal myofibroblasts (PMF), critically important extra-cellular matrix-producing cells and are, therefore, considered to be important targets in attempts to find new means of preventing or reducing liver fibrosis. Taken together, these observations suggest that FXR could be involved in the modulation of liver fibrogenesis.

The current study was designed to examine the role of FXR in liver fibrosis through a series of complementary experiments, by a group with esteemed expertise in signaling mechanisms in cholestasis. To test whether FXR expression could impact hepatic fibrogenesis, Fickert et al. subjected wild-type and FXR^−/− mice to standard experimental fibrosis models: 3,5-diethoxycarbon-yl-1,4-dihydrocollidine (DDC) feeding, BDL, carbon tetrachloride (CCl₄) intoxication, and Schistosoma mansoni (S. mansoni) infection. The DDC intoxication and BDL models were used as models of biliary cirrhosis, CCl₄ intoxication as a model of non-biliary International Hepatology cirrhosis, and S. mansoni infection as a model of non-biliary liver fibrosis. The authors show that, while having no significant impact on fibrosis in CCl₄ intoxication and S. mansoni infection models, the genetic deletion of FXR is clearly associated with decreased liver fibrosis in cholestatic DDC intoxication and BDL models. This finding is of critical importance, as it supports the emerging notion that biliary cirrhosis may be a pathophysiologically distinct event from non-biliary cirrhosis. Specifically, biliary cirrhosis may be mediated by epithelial–mesenchymal interactions involving bile duct epithelia (BDE) and portal fibroblasts and/or hepatic stellate cells [7]. However, caution is always required when interpreting data obtained using the constitutive gene knock-out approach in mice, because such animals may exhibit an adapted phenotype. This is an important consideration, especially with the well-documented evidence of crosstalk pathways between the targeted FXR receptor and other nuclear receptors, for instance PXR and CAR receptors [8].

The latter set of studies in this paper was designed to determine the liver cell subpopulations expressing FXR and related proteins. Mouse and human fibrotic livers were used for histomorphometric (sirius red staining for collagen and hydroxyproline contents) and immunostaining studies, and studies were made of isolated primary murine HSC and PMF. In congruence with previous reports, the cells found to express FXR immunogenicity in tissue sections from mouse and human were hepatocytes and BDE (with notably weaker staining in BDE). In contrast with previous reports [6], however, no evidence of FXR mRNA was detected in mouse PMF, and FXR mRNA was either absent or present in very low levels in mouse HSC. Furthermore, no evidence of the FXR effector small heterodimer partner (SHP) or the bile acid importer sodium-taurocholate cotransporting polypeptide (Ntcp) was noted in PMF, and Ntcp expression in HSC was absent or minimal. The low/minimal expression of Ntcp in HSC is echoed by our own unpublished studies, in which we were unable to demonstrate influx of bile acids into HSC. These findings suggest that the beneficial effects of FXR gene deletion in experimental biliary cirrhosis are not mediated at the level of liver myofibroblasts. Of note, regarding the methodology employed here, in situ hybridization experiments might have effectively complemented the FXR gene expression studies, as PCR- and antibody-based analyses may occasionally produce inconsistent uneven results, especially when studying primary isolated cells.

Is there a hypothesis that satisfies the observations in this manuscript and related work in the field? We believe that there is. Specifically, it appears that BDE are themselves critical regulators of biliary cirrhosis [9]. The pathologic hallmark of cholestatic conditions leading to biliary cirrhosis is the development of the ductular reaction, characterized by hyperproliferation of BDE and an inflammatory infiltrate. The ductular reaction may indeed indicate the starting point of the fibrogenic process in biliary cirrhosis [7]. In relation to the current work, BDE are known to express FXR [10,11] and, because the loss of FXR has been associated with reduced ductular proliferation in bile ductligated FXR^−/− mice ([5], and Supplementary Fig. 6 from the current manuscript), it is tempting to speculate that genetic FXR ablation could, in fact, alter the phenotypic activation of BDE by blocking their ability to transduce pro-fibrogenic signals to liver myofibroblasts. It is also interesting to note that, when DDC intoxication and BDL models are compared, the mRNA levels of downstream FXR target SHP are regulated in an opposite manner: significantly higher (Fig. 3E, DDC) and lower (Fig. 4E, BDL) than controls, respectively, in wild-type mice, suggesting that the resistance to liver fibrosis conferred by genetic FXR silencing in these models is not dependent on the FXR–SHP axis, and this is supported by the fact that the authors noted no protection from BDL-induced fibrosis in SHP (−/−) mice.

The major controversy to be found in this work is the functional absence of FXR in mouse and human liver myofibroblasts, in stark contrast to the findings of Fiorucci et al. [6], who found clear evidence of FXR mRNA and protein in rat HSC and the related HSC-T6 cell line. A simple explanation would involve an interspecies difference; however, this is not altogether reassuring. Since both investigative groups are known for their experimental expertise, it is difficult to consider technical errors as the primary explanation. Another possibility is the lack of purity of preparations of liver cell subpopulations, even by expert investigators, although this possibility would not account for the finding of FXR in a passaged cell line. This is certainly a situation warranting independent confirmation of both laboratories’ findings by outside investigators, and this is of key importance, since it would strongly influence new therapies to be designed on the FXR axis for the prevention and/or treatment of liver fibrosis.