Elsevier

Journal of Hepatology

Volume 71, Issue 6, December 2019, Pages 1126-1140
Journal of Hepatology

Research Article
FXR modulates the gut-vascular barrier by regulating the entry sites for bacterial translocation in experimental cirrhosis

https://doi.org/10.1016/j.jhep.2019.06.017Get rights and content

Highlights

  • For intestinal bacteria to enter the systemic circulation they must cross the mucus, epithelial and gut-vascular barrier.

  • Cirrhosis, but not portal hypertension per se, grossly impairs the endothelial and muco-epithelial barriers.

  • This promotes pathological bacterial translocation via the portal-venous circulation.

  • Both barriers appear to be FXR-modulated, as FXR-agonists reduce portal-venous bacterial translocation.

Background & Aims

Pathological bacterial translocation (PBT) in cirrhosis is the hallmark of spontaneous bacterial infections, increasing mortality several-fold. Increased intestinal permeability is known to contribute to PBT in cirrhosis, although the role of the mucus layer has not been addressed in detail. A clear route of translocation for luminal intestinal bacteria is yet to be defined, but we hypothesize that the recently described gut-vascular barrier (GVB) is impaired in experimental portal hypertension, leading to increased accessibility of the vascular compartment for translocating bacteria.

Materials

Cirrhosis was induced in mouse models using bile-duct ligation (BDL) and CCl4. Pre-hepatic portal-hypertension was induced by partial portal vein ligation (PPVL). Intestinal permeability was compared in these mice after GFP-Escherichia coli or different sized FITC-dextrans were injected into the intestine.

Results

Healthy and pre-hepatic portal-hypertensive (PPVL) mice lack translocation of FITC-dextran and GFP-E. coli from the small intestine to the liver, whereas BDL and CCl4-induced cirrhotic mice demonstrate pathological translocation, which is not altered by prior thoracic-duct ligation. The mucus layer is reduced in thickness, with loss of goblet cells and Muc2-staining and expression in cirrhotic but not PPVL mice. These changes are associated with bacterial overgrowth in the inner mucus layer and pathological translocation of GFP-E. coli through the ileal epithelium. GVB is profoundly altered in BDL and CCl4-mice with Ileal extravasation of large-sized 150 kDa-FITC-dextran, but only slightly altered in PPVL mice. This pathological endothelial permeability and accessibility in cirrhotic mice is associated with augmented expression of PV1 in intestinal vessels. OCA but not fexaramine stabilizes the GVB, whereas both FXR-agonists ameliorate gut to liver translocation of GFP-E. coli.

Conclusions

Cirrhosis, but not portal hypertension per se, grossly impairs the endothelial and muco-epithelial barriers, promoting PBT to the portal-venous circulation. Both barriers appear to be FXR-modulated, with FXR-agonists reducing PBT via the portal-venous route.

Lay summary

For intestinal bacteria to enter the systemic circulation, they must cross the mucus and epithelial layer, as well as the gut-vascular barrier. Cirrhosis disrupts all 3 of these barriers, giving bacteria access to the portal-venous circulation and thus, the gut-liver axis. Diminished luminal bile acid availability, cirrhosis and the associated reduction in farnesoid x receptor (FXR) signaling seem, at least partly, to mediate these changes, as FXR-agonists reduce bacterial translocation via the portal-venous route to the liver in cirrhosis.

Introduction

The gut-liver axis represents the pathophysiological hallmark for initiation and/or perpetuation of multiple liver diseases1 and has been proposed to be fueled by pathological bacterial translocation (PBT) from the gut.2 In liver cirrhosis, PBT from the gut into the liver and systemic circulation is one of the causes of bacterial infections and the augmented pro-inflammatory response to gut-derived products.[2], [3] In fact, failure to control invading bacteria and bacterial products in concert with host susceptibility determines remote organ injury in liver cirrhosis. This may include acute-on-chronic liver failure, hepatorenal syndrome and hepatic encephalopathy, which are all associated with worsening prognosis.4 PBT in liver cirrhosis has been attributed to small intestinal bacterial overgrowth, increased intestinal permeability and lack of host defense mechanisms.5 Herein, we focused on the first and last barrier separating luminal bacteria and the vascular compartment, namely intestinal mucus and the newly defined gut-vascular barrier (GVB),6 neither of which have been addressed so far in liver cirrhosis and PBT.

Mucus represents the first frontier that commensal microbes in the gut have to cross in order to achieve PBT. The mucus consists of 2 layers with a similar protein composition where mucin-2 (MUC2) is the main component.7 On one hand, the inner mucus layer is firmly attached to the epithelium, is densely packed and is devoid of bacteria.8 On the other hand, the outer mucus layer is much more mobile, looser and is colonized with a distinct bacterial community.9 Goblet cells (GCs) are responsible for the formation of both the inner and outer mucus layer10 but also sense bacteria11 and react accordingly with mucin secretion.[12], [13] After crossing the mucus and epithelial barrier, translocating bacteria reach the lymphatic system, as shown by culture positive mesenteric lymph nodes in experimental cirrhosis in multiple independent studies.[14], [15] In contrast, access to the intestinal microcirculation and portal-venous route has been proposed for PBT,16 but has not been delineated in detail in portal hypertension and liver cirrhosis yet. The splanchnic circulation in portal hypertension presents with multiple vascular abnormalities17 including arterial vasodilation,14 hyporesponsiveness to vasoconstrictors[18], [19] and increased angiogenesis.20 However, accessability of the intestinal microcirculation and thus, portal-venous route has not been investigated in portal hypertension so far. Endothelial barriers are characterized by the presence of junctional complexes which strictly control paracellular trafficking of solutes, fluids and cells.21 In healthy conditions, the endothelial vascular barrier discriminates between differently sized particles of the same nature with 4 kDa-dextran freely diffusing through the endothelium, whereas 70 kDa-dextran does not. Plasmalemma vesicle-associated protein (PV)-1 is an endothelial cell-specific protein that forms the stomatal and fenestral diaphragms of blood vessels22 and regulates basal permeability.23

Liver cirrhosis is characterized by deficient levels of luminal bile acids in the gut.24 Bile acids have been long known for their major effects on the microbiome and the intestinal barrier function. They exert their effects via transcription factors among which the farnesoid X receptor (FXR) is known to be one of the most important. FXR activation has been reported to influence epithelial cell proliferation25 and to exert potent anti-inflammatory actions in the intestine, stabilizing epithelial integrity.[26], [27], [28], [29] Moreover, FXR stimulation in the small intestine exerts antibacterial actions via induction of antimicrobial substances30 and FXR-agonists have been shown to ameliorate chemically induced intestinal inflammation, improving symptoms of colitis and inhibiting epithelial permeability. However, the exact role of bile acids and FXR in controlling intestinal muco-epithelial as well as vascular permeability is still unknown. In addition, the microbiome has been proposed to play a key role in mucus synthesis, release and barrier-function31 but information on its impact on GC density and mucus thickness are limited. Finally, although colonization by microbial commensals is known to promote vascular development32 its impact and modulatory role on the GVB-function is not known.

Taken together, the aims of the current study were i) to characterize changes in the mucus barrier, as well as GVB, in germ-free conditions and in the context of liver cirrhosis or portal hypertension; ii) to delineate PBT from the gut to the liver along the gut-liver axis in liver cirrhosis; iii) to unravel the role(s) of FXR on PBT, the mucus barrier and GVB.

Section snippets

Mice and animal models

Female C57BL/6J mice were purchased from ENVIGO (Horst, The Netherlands) and kept at the Central Animal Facility of the University of Bern under specific pathogen-free (SPF) conditions. Mice were kept in next-generation IVC cages with an enriched environment, 12 h day-night cycle and fed ad libitum. All experiments involving animals were performed in accordance with Swiss Federal regulations and with local institutional approval. Where indicated, animals were kept in either germ-free conditions

Increased gut-liver translocation in intestinal loop experiments in cirrhotic but not pre-hepatic portal-hypertensive mice

In control mice, as well as in PPVL-animals, neither GFP-E. coli nor 4 kDa-FITC-dextran translocated from the ileum to the liver (Fig. 1A). However, 4 kDa-FITC-dextran, as well as GFP-E. coli, were detectable in high numbers and hence, significantly increased in translocation to the liver in cirrhotic (BDL and CCl4-treated) mice (Fig. 1B). This was also confirmed in vivo by applying the dual-band laserendomicroscopy to the liver 1 h after loading the intestinal loop with GFP-E. coli (Fig. S1).

Discussion

In this paper we report changes in mucus- and GVB in relation to microbial modulation and as entry sites for PBT along the gut-liver axis in cirrhosis. Standardized in vivo intestinal loop experiments are utilized to quantify the translocation process from the intestinal lumen to the liver. Pathological increases in bacterial translocation are evidenced in cirrhotic mice, occurring largely independently of the lymphatic route. This does not exclude the well-known PBT along the lymphatic route

Financial support

This work was supported by Swiss-National-Fund SNF 310030_152805 to RW and SNF 31003A_163143 to ADG.

Conflict of interest

The authors declare no conflicts of interest that pertain to this work.

Please refer to the accompanying ICMJE disclosure forms for further details.

Authors’ contributions

M.S. ideated and performed the experiments; M.J., D.S., Y.N., S.M., M.H. helped M.S. in the execution of the mouse experiments; B.Y. and L.H. performed microbial analysis; M.R. provided human investigations and administered the informed consents; A.G., M.R., A.A. participated with ideas, results interpretation, and careful reading of the manuscript; R.W. ideated the study, coordinated the work, and wrote the manuscript.

Acknowledgements

The intestine-specific Fxr-null (FxrΔIE) mice were a kind gift of Prof. B. Schnabl (University of San Diego, CA, USA); Electron microscopy sample preparation and imaging were performed with devices supported by the Microscopy Imaging Center (MIC) of the University of Bern. We greatly appreciate technical support by F. Blank and C. Wotzkow from Department Pneumology and Department for Biomedical Research, University of Bern, Bern, Switzerland.

References (65)

  • V. Massafra et al.

    Splenic dendritic cell involvement in FXR-mediated amelioration of DSS colitis

    BBA

    (2016)
  • B. Renga et al.

    FXR mediates a chromatin looping in the GR promoter thus promoting the resolution of colitis in rodents

    Pharmacol Res

    (2013)
  • M. Domenicali et al.

    A novel model of CCl4-induced cirrhosis with ascites in the mouse

    J Hepatol

    (2009)
  • M. Downes et al.

    A chemical, genetic, and structural analysis of the nuclear bile acid receptor FXR

    Mol Cell

    (2003)
  • R. Wiest et al.

    Bacterial translocation up-regulates GTP-cyclohydrolase I in mesenteric vasculature of cirrhotic rats

    Hepatology

    (2003)
  • D.E. Fouts et al.

    Bacterial translocation and changes in the intestinal microbiome in mouse models of liver disease

    J Hepatol

    (2012)
  • L. Verbeke et al.

    The FXR agonist obeticholic acid prevents gut barrier dysfunction and bacterial translocation in cholestatic rats

    Am J Pathol

    (2015)
  • T. Gustot et al.

    Sepsis in alcohol-related liver disease

    J Hepatol

    (2017)
  • M. Van der Sluis et al.

    Muc2-deficient mice spontaneously develop colitis, indicating that MUC2 is critical for colonic protection

    Gastroenterology

    (2006)
  • J. Wisniewska-Kruk et al.

    Plasmalemma vesicle-associated protein has a key role in blood-retinal barrier loss

    Am J Pathol

    (2016)
  • P. Schwabl et al.

    The FXR agonist PX20606 ameliorates portal hypertension by targeting vascular remodelling and sinusoidal dysfunction

    J Hepatol

    (2017)
  • R. Wiest et al.

    Bacterial translocation (BT) in cirrhosis

    Hepatology

    (2005)
  • Jalan R, Fernández J, Wiest R, et al. Bacterial infections in cirrhosis: a position statement based on the EASL Special...
  • I. Spadoni et al.

    A gut-vascular barrier controls the systemic dissemination of bacteria

    Science (New York, NY)

    (2015)
  • M.E. Johansson et al.

    The gastrointestinal mucus system in health and disease

    Nat Rev Gastroenterol Hepatol

    (2013)
  • M.E. Johansson et al.

    The inner of the two Muc2 mucin-dependent mucus layers in colon is devoid of bacteria

    PNAS

    (2008)
  • H. Li et al.

    The outer mucus layer hosts a distinct intestinal microbial niche

    Nat Commun

    (2015)
  • M.E. Johansson

    Fast renewal of the distal colonic mucus layers by the surface goblet cells as measured by in vivo labeling of mucin glycoproteins

    PLoS ONE

    (2012)
  • G.M. Birchenough et al.

    A sentinel goblet cell guards the colonic crypt by triggering Nlrp6-dependent Muc2 secretion

    Science (New York, NY)

    (2016)
  • R. Wiest et al.

    Bacterial translocation in cirrhotic rats stimulates eNOS-derived NO production and impairs mesenteric vascular contractility

    J Clin Investig

    (1999)
  • R. Schierwagen et al.

    Circulating microbiome in blood of different circulatory compartments

    Gut

    (2018)
  • R. Wiest

    Splanchnic and systemic vasodilation: the experimental models

    J Clin Gastroenterol

    (2007)
  • Cited by (145)

    • Progress in the treatment of malignant ascites

      2024, Critical Reviews in Oncology/Hematology
    • Paneth cell: The missing link between obesity, MASH and portal hypertension

      2024, Clinics and Research in Hepatology and Gastroenterology
    View all citing articles on Scopus
    View full text