A mouse model of hepatic encephalopathy: bile duct ligation induces brain ammonia overload, glial cell activation and neuroinflammation

Abstract

Hepatic encephalopathy (HE) is a common complication of chronic liver disease, characterized by an altered mental state and hyperammonemia. Insight into the brain pathophysiology of HE is limited due to a paucity of well-characterized HE models beyond the rat bile duct ligation (BDL) model. Here, we assess the presence of HE characteristics in the mouse BDL model. We show that BDL in C57Bl/6j mice induces motor dysfunction, progressive liver fibrosis, liver function failure and hyperammonemia, all hallmarks of HE. Swiss mice however fail to replicate the same phenotype, underscoring the importance of careful strain selection. Next, in-depth characterisation of metabolic disturbances in the cerebrospinal fluid of BDL mice shows glutamine accumulation and transient decreases in taurine and choline, indicative of brain ammonia overload. Moreover, mouse BDL induces glial cell dysfunction, namely microglial morphological changes with neuroinflammation and astrocyte reactivity with blood-brain barrier (BBB) disruption. Finally, we identify putative novel mechanisms involved in central HE pathophysiology, like bile acid accumulation and tryptophan-kynurenine pathway alterations. Our study provides the first comprehensive evaluation of a mouse model of HE in chronic liver disease. Additionally, this study further underscores the importance of neuroinflammation in the central effects of chronic liver disease.

Figure 1

(A) Selected outcome measures to determine underlying liver disease, hyperammonemia and neurological deficits according to ISHEN guidelines. (B) Timeline detailing timing of outcome measure assessment after BDL in both C57Bl/6j and Swiss mice. BDL bile duct ligation, ISHEN International society for hepatic encephalopathy and nitrogen metabolism.

Figure 2

Evolution of liver disease in C57Bl/6j and Swiss mice after chronic BDL. C57Bl/6j mice were subjected to sham or BDL for 7, 14, 21 and 28 days. Swiss mice were subjected to sham or BDL for 6 and 8 weeks. (A) Plasma levels of AST, ALT, bilirubin, albumin and ammonia at different timepoints after sham (C57Bl/6j mice: n = 5–15/timepoint; Swiss mice: n = 13) or BDL (C57Bl/6j mice: n = 6–9/timepoint; Swiss mice: n = 5–13/timepoint) induction. (B) Representative H&E and Sirius Red stained images after sham and BDL induction. Quantification of the fibrotic area on Sirius Red stained liver sections of sham (C57Bl/6j mice: n = 5/timepoint; Swiss mice: n = 7) or BDL (C57Bl/6j mice: n = 5/timepoint; Swiss mice: n = 13) mice at different timepoints. Quantification of biliary infarcts on H&E-stained liver sections of sham (C57Bl/6j mice: n = 5/timepoint; Swiss mice: n = 7) or BDL (C57Bl/6j mice: n = 5/timepoint; Swiss mice: n = 11) mice at different timepoints. Data are represented as means ± SEM. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001 vs. sham control. Arrowheads indicate biliary infarcts. Scale bars represent 200 µm. BDL bile duct ligation.

Figure 3

Evolution of behaviour changes after chronic BDL in C57Bl/6j and Swiss mice. C57Bl/6j mice were subjected to sham (n = 15) and BDL (n = 17) for 28 days. Swiss mice were subjected to sham (n = 13) and BDL for 6 (n = 13) or 8 (n = 6) weeks. (A) Travelling distance, time spent in the center of the open field at different timepoints after sham/BDL inductions in C57Bl/6j (A) and Swiss (C) mice. Average traversal time over the difficult beam at different timepoints after sham/BDL inductions in C57Bl/6j (B) and Swiss (D) mice. Data are represented as means ± SEM. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001 vs. sham control. BDL bile duct ligation.

Figure 4

Plasma cytokine levels post BDL in C57Bl/6j and Swiss mice. Plasma levels of IL-6 and TNF in sham (C57Bl/6j: n = 6; Swiss: n = 5) controls and post-BDL (C57Bl/6j: n = 6/timepoint; Swiss: n = 5). Data are represented as means ± SEM. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001 vs. sham control. BDL bile duct ligation, IL interleukin, TNF tumor necrosis factor.

Figure 5

Ammonia-related metabolites in the cerebrospinal fluid (CSF) of C57Bl/6j mice. (A–C) Relative CSF levels of (A) glutamine, glutamate and (C) osmolytes taurine and creatine in sham controls (n = 8) and BDL (n = 6/timepoint) animals at different timepoints. (B) Spearman correlation between CSF glutamine and plasma ammonia. Spearman correlation coefficient (r) and corresponding p-value are shown. Values are represented as mean ± SEM. *p < 0.05, **p < 0.01. BDL bile duct ligation, CSF cerebrospinal fluid.

Figure 6

Energy markers, bile acids and tryptophan metabolites in CSF post BDL in C57Bl/6j mice. (A–E) Relative CSF levels of (A) glycolysis intermediates and end products, (B) TCA cycle intermediates, (C) AMP, (D) bile acids and (E) tryptophan metabolites in sham controls (n = 8) and BDL animals (n = 6/timepoint) at different timepoints. Values are represented as mean ± SEM. *p < 0.05, **p < 0.01. BDL bile duct ligation, CSF cerebrospinal fluid, TCA tricarboxylic acid.

Figure 7

Astrocyte reactivity and function after BDL in C57Bl/6j mice. (A) Representative confocal maximum intensity projection images of GFAP positive astrocytes and magnified 3D reconstructed astrocytes in the CA3 region of the hippocampus at different timepoints after BDL. (B) Quantification of 3D reconstructed astrocytes, showing branch length, number of branch points, branch segments and endpoints per cell in sham (n = 5) or BDL animals (n = 5/timepoint) at different timepoints after BDL. (C) Expression levels of astrocyte reactivity markers Fkbp5, Cp and Serpina3n in sham (n = 8) and BDL (n = 6/timepoint) mice at different timepoints after BDL. (D) Blood–brain barrier (BBB) permeability assessment through measurement of brain tissue fluorescence after iv injection of 4 kDa FITC-Dextran in sham (n = 5) and BDL (n = 7), 28 days after induction. All data are represented as mean ± SEM. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. Scale bar GFAP = 20 µm. Scale bar 3D reconstruction = 5 µm. BBB blood–brain barrier, BDL bile duct ligation, Cp ceruloplasmin (Cp), Fkbp5 FK506 binding protein 5 (Fkbp5), GFAP glial fibrillary acidic protein, Serpina3n serpin family A member 3.

Figure 8

Microglial activation at different timepoints after BDL in C57Bl/6j mice. (A) Representative confocal maximum intensity projection images of IBA1 positive microglia and magnified 3D reconstructed microglia in prefrontal cortex at different timepoints after BDL. (B) Quantification of 3D reconstructed microglia, showing branch length, number of branch points, branch segments and endpoints per cell in sham (n = 4) or BDL animals (n = 4–5/timepoint) at different timepoints after BDL. (C) Protein levels of MCP-1 and IL-4 in prefrontal cortex in sham (n = 6) or BDL animals (n = 6/timepoint) at different timepoints after BDL. *p < 0.05, **p < 0.01. All data are represented as mean ± SEM. Scale bar IBA1 = 20 µm. Scale bar 3D-reconstruction = 10 µm. BDL bile duct ligation.