Elevated oxysterol levels in human and mouse livers reflect nonalcoholic steatohepatitis

Abstract

Nonalcoholic steatohepatitis (NASH), a primary cause of liver disease, leads to complications such as fibrosis, cirrhosis, and carcinoma, but the pathophysiology of NASH is incompletely understood. Epstein-Barr virus-induced G protein-coupled receptor 2 (EBI2) and its oxysterol ligand 7α,25-dihydroxycholesterol (7α,25-diHC) are recently discovered immune regulators. Several lines of evidence suggest a role of oxysterols in NASH pathogenesis, but rigorous testing has not been performed. We measured oxysterol levels in the livers of NASH patients by LC-MS and tested the role of the EBI2-7α,25-diHC system in a murine feeding model of NASH. Free oxysterol profiling in livers from NASH patients revealed a pronounced increase in 24- and 7-hydroxylated oxysterols in NASH compared with controls. Levels of 24- and 7-hydroxylated oxysterols correlated with histological NASH activity. Histological analysis of murine liver samples demonstrated ballooning and liver inflammation. No significant genotype-related differences were observed in Ebi2^-/- mice and mice with defects in the 7α,25-diHC synthesizing enzymes CH25H and CYP7B1 compared with wild-type littermate controls, arguing against an essential role of these genes in NASH pathogenesis. Elevated 24- and 7-hydroxylated oxysterol levels were confirmed in murine NASH liver samples. Our results suggest increased bile acid synthesis in NASH samples, as judged by the enhanced level of 7α-hydroxycholest-4-en-3-one and impaired 24S-hydroxycholesterol metabolism as characteristic biochemical changes in livers affected by NASH.

Keywords: 25-hydroxycholesterol 7α-hydroxylase; Epstein-Barr virus-induced gene 2; cholesterol 25 hydroxylase; mouse feeding model; nonalcoholic fatty liver disease.

Fig. 1.

Liver oxysterol levels in patients with NASH versus controls. Levels of various oxysterols were determined in human liver samples from nine patients with NASH (black squares: with fibrosis; white squares: no fibrosis) and eight controls. Statistical analysis: Mann-Whitney U test. ***P < 0.001, **P < 0.01, and *P < 0.05.

Fig. 2.

Increased body weight and liver inflammation in mice fed an HFD for 20 weeks. A: Weight development in male HFD and STD C57BL/6 mice. B: Epididymal fat weight as a percentage of total body weight after 20-week HFD feeding in C57BL/6 mice. C: Macroscopic aspect of livers from STD and HFD C57BL/6 mice. D: Representative H/E (left), Sirius Red (middle), and Oil Red O staining (right) of STD (upper panel) and HFD (bottom panel) wild-type mice, illustrating steatosis (white arrow), necroinflammation (inserts with magnification), cellular hypertrophy (black arrow), fat accumulation (middle panel), and collagen deposition (right panel). Scale bar: 100 µm. E: Serum liver function tests (AST and ALT) in HFD and STD mice. F: Liver tissue was analyzed for mRNA expression of Tnf, normalized to Hprt. Statistical analysis: Mann-Whitney U test, n ≥ 9. ***P < 0.001, **P < 0.01, and *P < 0.05.

Fig. 3.

EBI2, CH25H, and CYP7B1 are not essential for the induction of NASH by HFD. A: Quantification of liver inflammation by the NAS of Ebi2^−/− (upper panel), Ch25h^−/− (middle panel), and Cyp7b1^−/− (bottom panel). Each column represents one individual mouse. White fractions of bars represent steatosis scoring (including both micro- and macrosteatosis), light blue fractions represent cellular hypertrophy, and dark blue fractions represent necroinflammation scoring. Quantification of (B) NAS and (C) fibrosis score in STD and HFD knockout mice and the respective littermate controls. Each dot represents one mouse, and wild-type mice are shown together with their respective littermate controls; wt vs. Ebi2^−/−: n = 12 + 10, wt vs. Ch25h^−/−: n = 11 + 11, and wt vs. Cyp7b1^−/−: n = 9 + 15.

Fig. 4.

HFD increases liver function tests in mice independent of EBI2, CH25H, and CYP7B1 gene function. Quantification of serum AST and ALT in STD and HFD (A) Ebi2^−/−, (B) Ch25h^−/−, and (C) Cyp7b1^−/− mice and the respective littermate wild-type controls (n_STD = 2–8; n_HFD ≥ 9). Statistical analysis: Mann-Whitney U test.

Fig. 5.

HFD changes the expression levels of genes involved in oxysterol metabolism and increases the expression of Tnf independent from EBI2, CH25H, and CYP7B1 gene function. Quantification of liver tissue mRNA level of Ebi2, Cyp7b1, Ch25h, and Tnf in STD and HFD (A) Ebi2^−/−, (B) Ch25h^−/−, and (C) Cyp7b1^−/− mice and the respective littermate wild-type controls (n_STD ≥ 5; n_HFD ≥ 9). Statistical analysis: Mann-Whitney U test. ***P < 0.001, **P < 0.01, and *P < 0.05.

Fig. 6.

Liver oxysterol levels in a murine model of NAFLD/NASH. Eight-week-old male C57BL/6 mice were fed an HFD or STD for 20 weeks. Levels of the indicated oxysterol in the liver tissue of HFD (black squares: NASH; white squares: NAFLD) and STD controls were measured by LC-MS. Statistical analysis: Mann-Whitney U test (n_STD = 8, ≥3 valid data points; n_HFD = 10, ≥5 valid data points). ***P < 0.001, **P < 0.01, and *P < 0.05.