Yin Yang 1 and farnesoid X receptor: a balancing act in non-alcoholic fatty liver disease?
- 1Laboratory of Hepato-Gastroenterology, Institut de Recherche Expérimentale et Clinique, Université catholique de Louvain, Brussels, Belgium
- 2Department of Surgery, NUTRIM School for Nutrition, Toxicology and Metabolism, Maastricht University, Maastricht, The Netherlands
- Correspondence to Professor Isabelle A Leclercq, Laboratoire d'Hépato-Gastro-Entérologie, Institut de Recherche Expérimentale et Clinique, Université catholique de Louvain, Avenue E Mounier 53, Box B1.52.01, Brussels 1200, Belgium;
- Received 7 February 2013
- Revised 18 February 2013
- Accepted 19 February 2013
- Published Online First 8 March 2013
Non-alcoholic fatty liver disease (NAFLD) is the hepatic manifestation of the metabolic syndrome. It comprises a spectrum ranging from bland steatosis or NAFL to non-alcoholic steatohepatitis (NASH) with or without fibrosis. There is a general consensus that patients with NAFLD have a very slow disease progression (if any). By contrast, patients with NASH can exhibit histological progression and can develop fibrosis, cirrhosis and hepatocellular carcinoma. Parenchymal inflammation is an important determinant of the severity and progression of the disease.
Increased fatty acid flux to the liver, from dietary absorption and from the adipose tissue, owing to insulin resistance, is a main contributor to increased hepatic lipid content. In addition, increased de novo lipogenesis, impaired mitochondrial fatty acid oxidation or decreased export of very low density lipoprotein triglyceride all play a part. Ligand-activated nuclear receptors control several key steps in lipid metabolism as well as inflammation and fibrogenesis and thus are potentially crucial players in NAFLD/NASH pathogenesis.1 One of those nuclear receptors is the bile salt sensor farnesoid X receptor (FXR). Besides regulating cholesterol and bile salt homeostasis, FXR is a key regulator of hepatic lipid metabolism. FXR, via induction of a short heterodimer partner, represses de novo lipogenesis. It induces peroxisome proliferator activated receptor (PPAR)α, and thus stimulates fatty acid β oxidation. The role of FXR is emphasised by the development of hepatosteatosis and hyperlipidaemia in FXR−/− mice.2 ,3 In addition to these direct effects on the liver, FXR also affects lipid metabolism and glucose homeostasis through bile salt-mediated induction of fibroblast growth factor (FGF) 15/19 in the enterocyte. FGF 15/19 reaches the liver via the portal blood stream and, upon binding to FGF receptor 4 (FGFR4), represses bile salt synthesis, lipogenesis and gluconeogenesis.4–7 In addition to these pleiotropic effects on lipid metabolism, FXR regulates glucose homeostasis. In conditions of Fxr gene ablation, the failure to suppress gluconeogenesis and reduced clearance of peripheral glucose lead to glucose intolerance.8 This may be explained by a direct effect on the expression of gluconeogenic genes and also by indirect effects on lipid metabolism, lipid accumulation and lipotoxicity, factors known to cause insulin resistance.9
Lu and colleagues provide a convincing example of the role of FXR on hepatosteatosis and glucose homeostasis and add new complexity to the puzzle.10 They showed that hepatosteatosis is associated with upregulation of Yin Yang 1 (YY1) in animal models of NAFLD. YY1 is a transcription factor of the zinc finger class with pleiotropic and dual (complementary and opposite) effects on several cellular events, including cell proliferation, mitochondrial biogenesis and insulin/insulin-like growth factor signalling pathways.11–13 They showed that overexpression of YY1 promoted, while YY1 silencing reduced, steatosis, serum triglycerides, insulin resistance and glucose intolerance in high-fat diet fed or db/db mice. They further showed that those effects were mediated through downregulation of FXR. In models of steatosis and in conditions of loss or gain of YY1 function, YY1 and FXR were oppositely expressed. Also, the effects of YY1 silencing were blunted in FXR−/− mice but reversed by FXR re-expression. These authors identified a YY1 binding site in the first intron of FXR. This mutation abolished the inhibitory effect of YY1 on FXR promoter activity. This means that, not only ligand-binding but also regulation of FXR expression itself determines target gene expression. The relationship between YY1 and FXR levels and the degree of steatosis is shown clearly in animal models as presented in the study by Lu and colleagues. However, translation of those conclusions to the human disease requires further scrutiny.
By exerting a transcriptional control on several target genes that control metabolic pathways relevant to NAFLD, FXR appears an attractive therapeutic target. But there is a down side to FXR stimulation in NAFLD: activation of FXR also modulates the reverse cholesterol transport and appears to be associated with both anti- and proatherogenic effects. It inhibits paraoxonase 1, which inactivates proatherogenic lipids, and impairs high-density lipoprotein formation. This may increase the susceptibility to atherosclerosis if not counterbalanced by anti-atherogenic effects resulting from anti-inflammatory, insulin sensitising, hypolipidaemic and vasoactive consequences of FXR activation.14 Clinical studies using FXR agonists are warranted for clarification as the dominant effect cannot be predicted. Besides, FXR, directly and indirectly through FXR/FGF 19/FGFR4 axis, has mitogenic effects on hepatic cells.15 Here again, this consequence might be beneficial for liver repair but detrimental if it favours unwanted cell proliferation and carcinogenesis.
Despite those concerns, FXR targeting agents remain attractive for NAFLD treatment. Support for this comes from a study of diabetic and steatotic rats treated with the FXR agonist INT-747 (obeticholic acid), an intervention that decreased glycaemia and dyslipidaemia and protected against body weight gain and insulin resistance.16 The results of the INT-747 clinical trial in patients with NASH (FLINT trial) are awaited.17 A challenge might be to separate desired therapeutic effects from unwanted effects (eg, FXR-related effects on cell proliferation) perhaps by the design of organ-specific or target gene-specific ligands. Targeting up-stream regulators such as YY1 may selectively regulate a restricted set of FXR-dependent genes. Although unlikely given the pleiotropic and dual actions of YY1, this remains to be studied experimentally. The paper by Lu et al also highlighted the fact that the resulting effect of YY1 overexpression on lipid accumulation is seen in conditions of energy surplus.10 Thus, the response to modulation of the FXR receptor must be checked in various clinical and metabolic situations.
One possibility just considered briefly by Lu et al10 is that YY1 might be regulated by inflammatory cytokines. Lipopolysaccharide (LPS) administration induced hepatic YY1 expression in mice, and tumour necrosis factor had similar effects in isolated hepatocytes. FXR has anti-inflammatory actions,18 and it will be interesting to discover whether FXR and YY1 reciprocally affect each other's expression. This preliminary evidence, added to other emerging data,3 supports the possibility that the first cause of NAFLD, and possibly other consequences of obesity, might be the triggering of innate/inflammatory effectors, secondarily affecting lipid and glucose homeostasis. Indeed the route to NAFLD/NASH might not be the highroad of liver fat accumulation increasing the susceptibility to a second injurious hit as first proposed, but rather a labyrinth of diverse reciprocally amplifying signals (inflammatory, metabolic, oxidative, etc) and tissue origin (liver, adipose tissue, gut, brain, etc).9 ,19 Up to now, efforts to improve treatment of NASH have focused on improving insulin sensitivity (metformin, PPARγ agonists, etc), on interfering with lipid metabolism to decrease lipid storage (PPARα agonist, olistat, etc) or on protecting against lipotoxicity and oxidative stress (vitamin E, ursodeoxycholic acid, etc).20 Reconsidering inflammation might be timely in the face of disappointing results obtained with current treatment of NAFLD/NASH and the metabolic syndrome.
Contributors VL and IAL wrote the manuscript. FGS (as fibroblast growth factor 19/farnesoid X receptor specialist), BD and YH (as an expert in pharmacology) critically and constructively revised the manuscript for important intellectual content.
Competing interests None.
Provenance and peer review Commissioned; internally peer reviewed.