Fibroblast growth factor 21 signaling: The liver in focus

Potential conflict of interest: Dr. Marschall consults for Albireo and Intercept. See Article on Page 425 Fibroblast growth factor 21 (FGF21) was first identified as a metabolic factor in a screen for proteins that induce glucose uptake in adipocytes.1 FGF21 is a member of the so‐called hormone‐like subgroup within the FGF superfamily that also contains FGF19 and FGF23. The C‐terminus of FGF21 binds with high affinity to β‐Klotho, which is present in a complex with FGF receptors (FGFRs) on the surface of cells (Fig. 1).2 FGF21 has been shown to promote antidiabetic, lipid‐lowering, and weight‐reducing effects at pharmacological concentrations in rodents.3 In adipose tissue—considered a key target tissue—FGF21 stimulates production of adiponectin, which, in turn, enhances insulin sensitivity in the liver.3 FGF21 has also been shown to elicit many of its metabolic effects through actions on the central nervous system.4 Recently, a liver‐to‐brain hormonal axis was described in which liver‐produced FGF21 functions to regulate the consumption of individual macronutrients and even substances of abuse, such as alcohol.5 Given the numerous metabolic actions of FGF21, FGF21‐based pharmacotherapy is currently being extensively explored for prevention and therapy of a variety of metabolic disorders.3 Recent studies with long‐acting FGF21 analogs in obese patients with type 2 diabetes indeed presented promising perspectives given that they demonstrated improvements of serum lipid profiles and weight loss, although glucose levels basically were unaffected.7Figure 1: Regulation of hepatic insulin sensitivity through inhibition of mTORC1 by FGF21 in hepatocytes. Excessive nutrient‐ and hormonal‐dependent activation of mTORC1 is known to lead to insulin resistance through repression of IRS‐1 by serine phosphorylation. FGF21 represses mTORC1 and thereby reverts its effect on IRS‐1, leading to improved hepatic insulin sensitivity and glucose homeostasis. Abbreviation: PKB, protein kinase B.In this issue of Hepatology, Gong et al.9 significantly expand on the knowledge of molecular mechanisms underlying the effect of FGF21 in the regulation of insulin and glucose homeostasis in the liver. By combining in vitro experiments in human cultured hepatocytes and gain‐ and loss‐of‐function studies in multiple mouse models, the investigators provide several layers of evidence to indicate that FGF21 acts as a novel inhibitor of the nutrient sensor, mammalian target of rapamycin complex 1 (mTORC1), in the liver (Fig. 1). First, administration of FGF21 to mice challenged with a type 2 diabetogenic diet significantly decreased hepatic mTORC1 activity and improved liver insulin sensitivity, as evidenced by reduced serine phosphorylation of insulin receptor substrate 1 (IRS‐1) combined with an increased phosphorylation of Akt (protein kinase B). Consistently, adenovirus‐mediated overexpression of FGF21 in liver caused a potent reduction in hepatic mTORC1 activity and IRS‐1 phosphorylation together with increased phosphorylation of Akt in insulin‐treated mice. In addition, increased levels of FGF21 in mice induced hepatic glycogen synthesis, as evidenced by increased phosphorylation of glycogen synthase kinase 3β and reduced phosphorylation of glycogen synthase. Reciprocal effects on mTOR activity, insulin signaling, and glycogen synthesis were observed in FGF21 knockout mice. Interestingly, the investigators found that the effects of FGF21 overexpression in mice were replicated in human HepG2 and Huh7 hepatocytes with adenovirus‐delivered FGF21, suggesting a hepatocyte‐autonomous manner of FGF21 action. Strikingly, hepatic knockdown of the FGF21 receptor scaffold protein, β‐Klotho, or hepatic hyperactivation of mTORC1/downstream effector S6 kinase 1 (S6K1) abrogated the effect of FGF21 on hepatic insulin sensitivity and glycemic control in mice with diet‐induced insulin resistance, suggesting that inhibition of mTORC1 and augmentation of insulin sensitivity by FGF21 are causally linked. Finally, the investigators demonstrated that FGF21 regulates mTORC1, at least partly, in a tuberous sclerosis complex (TSC)‐dependent manner, although the molecular links between FGF21‐β‐Klotho signaling and TSC1/2 remain elusive. Consistent with a previous investigation,10 the investigators show that FGF21 improves central features of methionine‐ and choline‐deficient diet‐induced nonalcoholic steatohepatitis (NASH) in mice by repressing hepatic steatosis, hepatocellular damage, and perisinusoidal fibrosis. It will be interesting to determine whether FGF21's effects on NASH are mediated by mTORC1 inhibition or by a different pathway. Although accumulating evidence—including the mechanistic studies reported by Gong et al.9 in this issue—points toward beneficial metabolic effects of FGF21, there are many unanswered questions and technical challenges that lie ahead when exploiting the potential therapeutic benefits of FGF21. First, large‐scale recombinant protein production is often technically challenging and costly, and efforts to discover orally bioavailable small molecules to mimic/augment the action of FGF21 or enhance its endogenous production are therefore warranted. Notably, broad ranges of circulating FGF21 levels are reported in humans, and a risk for development of FGF21 resistance has been raised,3 which may compromise treatment outcome. Finally, although rodent studies and first proof‐of‐concept trials in human subjects have provided encouraging improvement of blood‐lipid profile and reduced body weight, the clinically significant effect of FGF21 treatment over time remains to be demonstrated in relevant patient populations, and, certainly, the long‐term safety of treatment with FGF21 must be addressed. Author names in bold designate shared co‐first authorship.