Review

The interaction of hepatic lipid and glucose metabolismin liver diseases

Lars P. Bechmann1, Rebekka A. Hannivoort2, Guido Gerken1, Gökhan S. Hotamisligil3,Michael Trauner4, Ali Canbay1,⇑

1Department of Gastroenterology and Hepatology, University Hospital Essen, University of Duisburg-Essen, Essen, Germany; 2Department ofGastroenterology and Hepatology, University Medical Center Groningen, University of Groningen, Groningen, The Netherlands; 3Departments

of Genetics and Complex Diseases and Nutrition, Harvard School of Public Health, Boston, MA, USA; 4Division of Gastroenterologyand Hepatology, Department of Internal Medicine III, Medical University of Vienna, Austria

Summary autophagy and an imbalance in chemokines and nuclear receptor

It is widely known that the liver is a central organ in lipogenesis,gluconeogenesis and cholesterol metabolism. However, over thelast decades, a variety of pathological conditions highlighted theimportance of metabolic functions within the diseased liver. Asobserved in Western societies, an increase in the prevalence ofobesity and the metabolic syndrome promotes pathophysiologicalchanges that cause non-alcoholic fatty liver disease (NAFLD).NAFLD increases the susceptibility of the liver to acute liver injuryand may lead to cirrhosis and hepatocellular cancer. Alterations ininsulin response, b-oxidation, lipid storage and transport,

Journal of Hepatology 20

Keywords: NAFLD; Fatty acid transporters; HCC; Liver regeneration; Nuclearreceptors; ER stress.Received 13 May 2011; received in revised form 9 August 2011; accepted 10 August2011⇑ Corresponding author. Address: Department of Gastroenterology and Hepatol-ogy, University Hospital Essen, Hufelandstr 55, 45122 Essen, Germany. Tel.: +49201 723 84713; fax: +49 201 723 5719.E-mail address: ali.canbay@uni-due.de (A. Canbay).Abbreviations: ACC, acetyl-CoA carboxylase; ATF6, activating transcription factor-6; AMPK, AMP-activated protein kinase; apoB-48/100, apoprotein B-48/100; BA,bile acids; CD36/FAT, fatty acid translocase; ChREBP, carbohydrate responsiveelement binding protein; CPT-1, carnitine palmitoyltransferase-1; DAG, diacyl-glycerol; DGAT, diacylglycerol-acyltransferase; DNL, de novo lipogenesis; ER, e-ndoplasmatic reticulum; FA, fatty acids; FABP, fatty acid binding protein; FAS,fatty acid synthase; FATPs, fatty acid transport proteins; FFA, free FA; FGF15/19,fibroblast growth factors 15 (mouse) and 19 (human); FoxO, forkhead box proteinO; FXR, farnesoid X receptor; GCKR, glucokinase regulatory protein; GLP-1, glu-cagon like peptide-1; GLUT2, glucose transporter type 2; Got2 or mitochondrialaspartate aminotransferase [mAspAT], glutamate-oxaloacetate-transaminase 2;GPAT, glycerol-3-phosphate-acyltransferase; GS, glycogen synthase; G6Pase, gl-ucose-6-phosphase; HCC, Hepatocellular carcinoma; HNF4a, hepatic nuclear fa-ctor-4-alpha; HNF6, hepatic nuclear factor 6; IGF-1, insulin-like-growth factor 1;IR, insulin resistance; LCFAs, long chain fatty acids; L-GCK, liver glucokinase;LRH1, liver receptor homolog 1; LXR, liver X receptor; MODY, maturity onsetdiabetes of the young; mTOR, mmmalian target of rapamycin; NAFLD, non-alc-oholic fatty liver disease; NASH, non-alcoholic steatohepatitis; NRs, nuclear rec-eptors; PEPCK, phosphoenolpyruvate carboxykinase; PGC1a, peroxisomeproliferator-activated receptor c co-activator-1a; PHx, partial hepatectomy; P-I3K, phosphoinositide-3-kinase; PKA, protein kinase A; PPARs, peroxisome prol-iferator-activated receptors; PYGL, glycogen phosphorylase; ROS, reactive oxygenspecies; SIRT1, sirtuin 1; SREBP-1c, sterol-response-binding-protein-1c; TAGs, t-riacylglycerols; TCA, tricarboxylic acid cycle; TRB3, tribbles-homologue 3; UPR,unfolded protein response; USF1, upstream stimulatory factor 1; VDR, vitamin Dreceptor; VLDL, very low-density lipoproteins.

signaling are held accountable for these changes. Furthermore,recent studies revealed a role for lipid accumulation in inflamma-tion and ER stress in the clinical context of liver regeneration andhepatic carcinogenesis. This review focuses on novel findingsrelated to nuclear receptor signaling – including the vitamin Dreceptor and the liver receptor homolog 1 – in hepatic lipid andglucose uptake, storage and metabolism in the clinical context ofNAFLD, liver regeneration, and cancer.

� 2012 European Association for the Study of the Liver. Publishedby Elsevier B.V. All rights reserved.

Key Points 1

• The prevalence of obesity and the metabolic syndrome is rising dramatically

• Obesity and the metabolic syndrome promote alterations in hepatic lipid and glucose metabolism and are linked to the pathophysiologies of NAFLD and liver cancer

• Lipid accumulation and fatty acid oxidation are important mechanisms in liver damage, repair and regeneration, partially due to induction of autophagy and ROS production

• Insulin and nuclear receptor (including PPAR, LXR, FXR, LRH1 and vitamin D receptor) signaling regulate hepatic lipid and glucose metabolism, and are closely interrelated

• Both metabolic pathways share common regulatory elements as well as metabolites and indistinguishably contribute to NAFLD, liver cancer, and liver regeneration

Introduction

As the main detoxifying organ of the body, the liver also plays acentral role in metabolic homeostasis and is a major site for syn-thesis, metabolism, storage and redistribution of carbohydrates,

12 vol. 56 j 952–964

JOURNAL OF HEPATOLOGY

proteins and lipids. The rapid increase in obesity worldwide isassociated with an increase in the prevalence of non-alcoholic fattyliver disease (NAFLD), making NAFLD the most common liverdisease in Western societies [1,2]. In order to understand thepathogenesis of NAFLD, we discuss the basic physiologicmechanisms of hepatic lipid and glucose metabolism. We alsoaimed at integrating recent clinical and mechanistic data and pointout novel links between basic metabolic pathways and the patho-physiologies of NAFLD, liver regeneration, and carcinogenesis.

NAFLD is characterized by lipid accumulation within hepato-cytes and may progress to non-alcoholic steatohepatitis (NASH).Lipids derive from circulating fatty acids (FA) upon insulin resis-tance (IR)-induced dysregulation of peripheral lipolysis. FAs aretranslocated into the hepatocyte mainly by membrane boundtransport proteins [3]. De novo lipogenesis (DNL) further contrib-utes to hepatic steatosis [4]. Hepatocellular accumulation of lipo-toxic intermediates such as diacylglycerol (DAG) and ceramidscauses hepatic IR [5]. Hepatocytic lipid accumulation predisposesto overproduction of reactive oxygen species (ROS), endoplasmaticreticulum (ER) stress and lipotoxicity [6–8]. Recently, autophagy(especially in the form of macrolipophagy) has been identified toregulate intracellular lipid stores through degradation of lipiddroplets and release of FAs into the cytosol as a rapid response tostarvation [9–11]. Thus, disrupted autophagy might be essentialto the pathogenesis of NASH via lipotoxicity-induced ER stress [12].

The individual steps in hepatic lipid metabolism are orches-trated by a delicate interplay of hormones, nuclear receptors,intracellular signaling pathways and transcription factors. Insulinsignaling plays an important role in the regulation of FA metab-olism, underscoring the close relation between lipid and glucosemetabolism. Insulin affects DNL on multiple levels, via inductionof lipogenic genes, activation of sterol-response-binding-protein-1c (SREBP-1c) and Akt-regulated production of very low-densitylipoproteins (VLDLs) [13]. Insulin, along with other mediators(e.g., calpain-1), further represses autophagy within the hepato-cyte, and thus induces lipogenesis and represses lipid-degrada-tion in the fed state as well [9,12]. On the other hand,glucagon-like peptide-1 (GLP-1) also alters hepatic lipid metabo-lism [14]. FA oxidation and the expression of fatty acid transportproteins (FATPs) are closely regulated by the nuclear receptorperoxisome proliferator-activated receptor (PPAR)a, and in thesteatotic liver also by PPARc [15].

In the following paragraphs, we will discuss the function ofhepatic lipid and glucose metabolism under normal conditionsand their roles in NAFLD, acute liver injury and regeneration as wellas hepatic carcinogenesis. This review focuses on the central role ofnuclear receptor signaling in hepatic glucose and lipid metabolismincluding novel mechanisms like vitamin D receptor (VDR) andliver receptor homolog 1 (LRH1) signaling, autophagy and theinterrelation of ER stress and metabolism Table 1.

Hepatic lipid metabolism

Extrahepatic lipid metabolism

Lipid metabolism starts with the intestinal absorption of dietaryfats. In order to cross the intestinal lumen into the plasma, lipidsare emulsified and hydrolyzed within the lumen. The healthyliver is crucial for intestinal lipid absorption via bile acids (BA)that are synthesized within the hepatocyte and secreted into

Journal of Hepatology 201

the bile duct to emulsify lipid droplets by their amphiphilic prop-erties rendering them accessible to lipase hydrolyzation. Hydro-lyzed lipids are then absorbed by enterocytes, where lipids arere-synthesized and packed into lipoprotein particles (i.e., nascentchylomicrons).

Nascent chylomicrons are secreted into the lymphatic system,where they bypass the liver and enter the circulation within twohours after food intake [16]. During their journey through thevascular system, nascent chylomicrones lose two minor apopro-teins (apoA-I and apoA-IV), which are replaced by apoE andapoC-II that are crucial for their further processing. ApoC-II acti-vates adipocyte lipoprotein lipase (LPL), which facilitates thedigestion of the chylomicron triacylglycerols (TAGs) into FAsand glycerol [17]. FAs are then partially taken up and stored inadipocytes, while chylomicron remnants re-enter the bloodstream. ApoE is then recognized by the hepatocyte LDL receptor,the LDL receptor-related protein (LRP) and scavenger receptorB-1, which facilitate endocytotic uptake of the chylomicron rem-nants. As cholesteryl-ester enriched and triglyceride-depletedproducts of chylomicron metabolism, chylomicron remnants arefinally processed by intracellular lysosomes – and their glycerol,FA, cholesterol, amino acid and phosphate residues are metabo-lized and recycled into new VLDLs (see Fig. 1A for overview).

Hepatic lipogenesis

As mentioned above, hepatic FAs either derive from endogenouslipogenesis, are released from lysosomes by autophagy, or derivefrom the free FA (FFA) plasma pool via active uptake into thehepatocyte. Depending on the metabolic state, FAs are then eitherprocessed to TAGs and stored or rapidly metabolized. Indeed,b-oxidation is the predominant source of energy during the fast-ing state.

Hepatic lipogenesis includes de novo synthesis of FAs fromacetyl-CoA or malonyl-CoA and further processing to TAGs. Inmammals, FA synthesis is catalyzed by acetyl-CoA carboxylase(ACC) and fatty acid synthase (FAS) – an enzyme that is com-plexly regulated by various nuclear receptors (PPARa, PPARcand the bile acid receptor/farnesoid X receptor [FXR]) [18–20].FA elongation requires NADPH as a reducing reagent, which isprovided by the pentose phosphate pathway (Fig. 2B). Remark-ably, PPARa itself is activated by a phospholipid synthesized byFAS, indicating a feedback loop [21].

A close link between glucose and lipid metabolism is indi-cated by the fact that nuclear receptors (NRs) are also importantmediators of insulin signaling and since DNL occurs under ana-bolic condition. The existence of such a link is further supportedby the fact that insulin stimulates FAS expression via the phos-phoinositide-3-kinase (PI3K) pathway [22]. On a transcriptionallevel, SREBP-1c and carbohydrate-responsive element bindingprotein (ChREBP), a glucose dependant transcription factor, syn-ergistically induce expression of FAS and ACC [23].

As FAs and their metabolites are the major cause for lipotox-icity and promote the formation of ROS, FAs are stored for futureuse as TAGs, which are relatively inert and consist of three FAsesterified to a glycerol backbone. TAGs are then either stored inlipid droplets within the hepatocyte or processed to VLDL [7].TAG synthesis is catalyzed by the enzymes mitochondrial glyc-erol-3-phosphate-acyltransferase (mtGPAT) and diacylglycerol-acyltransferase (DGAT) [24]. TAGs are then packaged into VLDLparticles, by conjugation to apoB-100 in a 5:1 TAG/cholesterol

2 vol. 56 j 952–964 953

Table 1. Overview of ligands and function of the most abundant nuclear receptors in hepatocytes.

Nuclear receptor

Natural ligands Chemical ligands/agonists

Function in lipid metabolism

Function in glucose metabolism

Role in NAFLD Role in liver regeneration

Role in HCC

PPARα Fatty acids Fibrates Regulates expression of FAS lipogenesis;CD36/FAT, FATPs FA-Uptake;Acetyl-CoA-synthetase, CPT-1 β-oxidation

Regulates PEPCK, GSK3, Glycogen synthase glycogenmetabolism; insulin sensitivity

Fibrate treatment improves IR

Delayed regeneration in PPARα KO mice

PPARα activation is associated with liver carcinogenesis

PPARγ Prostaglandins Glitazones Regulates expression of CD36/FAT FA-Uptake; SCD-1 FA-metabolism

Regulates GLUT-4 expression insulin sensitivity

Activation;PPARγ KO mice are protected from diet induced steatosis; Glitazones improve IR and TAG accumulation and increase adiponectin levels

Downregulated; glitazones inhibit hepatocyte proliferation

PPARγ KO increases susceptibility to hepatic carcinogens; PPARγ activation induces cell cycle arrest in hepatoma cells

PPARδ Fatty acids Glitazones;GW501516

Regulates SREBP-1c lipogenesis

Induces glycolysis and pentose phosphate pathway shunt

Agonist treatment improves hepatic steatosis in mice

Agonist treatment improves liver regeneration in mice

PPARδ affects COX2 expression in hepatomas

FXR Bile acids Chenodeoxy-cholic acid (CDCA);GW4064

Regulates SREBP-1c lipogenesis;

Regulates VLDL formation via SHP

Regulates PEPCK, glucose-6-phosphatase gluconeogenesis

Induced; SHP is upregulated in NAFLD; FXR KO mice develop steatosis

Mediates liver regeneration after partial hepatectomy; impaired regeneratory capacity in FXR KO mice

FXR KO mice develop hepatic tumors; interaction with Wnt/β-catenin signalingDownstream target SHP reduced in human HCC

LXRα Hydroxysterols T0901317;GW3965

Regulates SREBP-1c, SCD-1, FAS lipogenesis; cholesterol metabolism

Regulates insulin receptor expression, GLUT-4 and IRS expression insulin sensitivity

Induced, LXR promotes hepatic lipogenesis

Reduced activation after partial hepatectomy

Unknown

VDR (1),25-Hydroxy-vitamin D3

Calcitriol-derivates

VDR represses PPARα signaling, direct effects unknown

VDR represses PPARα signaling, direct effects unknown

Downregulated; VDR KO mice develop steatosis; vitamin D protects from diet induced steatosis

leads to impaired hepatic regeneration

VDR polymorphisms are associated with HCC

LRH1 Phospholipids Dilauroyl phosphatidyl-choline (DLPC)

Repression of SHP potential effects on VLDL synthesis; DLPC treatment suppresses SREBP-1c, FAS, ACC-2 and SCD-1 expression in vivo

DLPC treatment reduces hepatic gluconeogenesis and improves insulin response in vivo

DLPC treatment improves hepatic TAG and FA accumulation

Unknown; potential proliferative effects

Unknown; repression of SHP might promote tumor growth

Vitamin D deficiency

Review

ratio. These processes are controlled by SREBP-1c, the liver Xreceptor (LXR), FXR and ChREBP, which again links glucose andlipid metabolism [25].

Hepatic fatty acid uptake

Another source for hepatic FAs is FFA recruitment from theplasma pool. FFAs are derived from lipolysis in adipocytes. Thisoccurs usually in the fasting state, where it is promoted by cate-cholamines, natriuretic peptides and glucagon, while it is usuallyrepressed by insulin [26]. However, the insulin-resistant state(obesity; metabolic syndrome) goes along with increasedadipocyte lipolysis, leading to abundant FFAs in the plasma poolindependently from the nutritional status [27]. FFAs are then

954 Journal of Hepatology 201

taken up by the hepatocytes in a facilitated fashion rather thanby passive processes [28]. FATPs are thus in the focus of NAFLDresearch in which a variety of FATPs have been identified. WhileFATP1 is abundant in muscle and adipose tissue and is barelydetectable in the liver [29], FATP2 and FATP5 are expressed inhepatocytes and most likely facilitate the major amount of FAuptake in the liver [30]. Other transport proteins include fattyacid binding protein (FABP), glutamate–oxaloacetate-transami-nase 2 (Got2; or mitochondrial aspartate aminotransferase [mAs-pAT], a membrane bound protein that mediates the endocytoticuptake of long-chain FAs), and caveolin-1 [31–33]. Fatty acidtranslocase (CD36/FAT) is a membrane glycoprotein present onplatelets, mononuclear phagocytes, adipocytes and hepatocyteswith multiple functions, including thrombospondin-1 receptor

2 vol. 56 j 952–964

LRH1

BA

FGF-19

FXR

Chy

lom

icro

n re

mna

nts

LRP/LDLR

Myocytes Adipocytes

Enterocytes

Dietary lipids

LPL

Lymphatic tract

Blood streamNC

C

A B

C D

HDL

Adipocytokines

CRCholesterol

Cholesterol

TAG

TAG

FA

LDL/HDL

FA mtFA-β-oxidation

Mitochondrialβ-oxidation

LYS

FA uptakeFFA

ATP

Acetyl-CoA

Acetyl-CoA(from PDC)

Malonyl-CoA

Malonyl-CoA

De novolipogenesis

TCA

Ketones

Autophagy

LD

Acetyl-CoA

Acyl-CoA

Dicarboxylic acid

TCA/ketones

Est

erifi

catio

n

VLDL VLDL

Peroximal β-oxidation

ERω-oxidation

SCFAMCFALCFA

LCFAVLCFA

LCFA

CPT1

ROS Hepatocyte stress

Insulin, PPARα

FA

Lipolysis

FA uptake

ERω-oxidation

DNL

mtFA-β-oxidation

LipotoxicityROS

Inflammation Fibrosis

Apoptosis

BA

FFA

Insulin,PPARα, PPARγ

Insulin, LXRFXR, PPARγ

Insulin, mTOR FoxO, Calpain1

Fig. 1. Hepatic lipid metabolism in health and disease. (A) Dietary lipids are emulsified in the intestinal tract by bile acids (BAs). Hydrolyzed lipids are absorbed byenterocytes and packed into nascent chylomicrons (NCs). NCs enter the bloodstream via the thoracic duct where they receive important apoproteins (apoE; apoC-II) from HDL.These apoproteins are important for chylomicrons (C) to deliver TAGs and FAs to adipocytes and myocytes via lipoprotein lipase (LPL) degradation. Chylomicron remnants aretaken up by hepatocytes via LDL-receptor (LDLR) and LDL receptor-related protein (LRP)-mediated endocytosis. BA synthesis is regulated by LRH1 and FXR, which activate BAexport pumps. BA re-uptake by enterocytes stimulates FGF-19 release into the portal blood, which inhibits BA synthesis. (B) Free fatty acids (FFA) derive from lipolysis inadipose tissue and are actively taken up by various FA transporters under the control of insulin (Ins) and nuclear receptor signaling. Under physiologic conditions, the bulk ofFAs is oxidized intramitochondrially and provides ATP and acetyl-CoA for the tricarboxylic acid cycle (TCA). Triglycerides (TAGs) derived from de novo lipogenesis are eitherstored in lipid droplets (LD) or packed into VLDL and exported into the blood stream. Acetyl-CoA for de novo lipogenesis is provided by the pyruvate dehydrogenase complex(PDC), which catalyzes oxidation of pyruvate, the end product of glycolysis. (C) Under physiologic conditions, b-oxidation of short-, medium- and long-chain FAs (SCFA, MCFA,LCFA) are degraded in mitochondria. Therefore, FAs are activated to acyl-CoA and shuttled across the mitochondrial membrane by carnitine palmitoyltransferase-1 (CPT1).Malonyl-CoA, an intermediate of lipogenesis, inhibits CPT1 and thus FA oxidation in the mitochondria. With FA abundance and in the insulin resistant state, LCFA and very-long-chain FAs (VLCFA) are oxidized in peroxisomes and the ER. This leads to an abundance of metabolites that induce formation of ROS and contribute to lipotoxicity. (D) InNASH, increased peripheral lipolysis, upregulation of FA transporters, an increase in DNL, and a switch from mitochondrial b-oxidation to peroxisomal and x-oxidationpromote FA toxicity and the release of ROS. This leads to the induction of hepatocyte apoptosis, the invasion and activation of inflammatory cells, as well as fibrogenesis.

JOURNAL OF HEPATOLOGY

Journal of Hepatology 2012 vol. 56 j 952–964 955

A B

Insulin Akt FoxO,PPARα, FXR, HNF4α,CREBP, Acetyl-CoA

Intermediates,AMP, PI3KCHREBP

Insulin Akt/PI3K GSK3, Glu-6-PAMPK

Insulin, PKA, AMP

Extracellular

GLUT-2

Glucose

Glycogensynthesis

Glycogensynthesis

Glycogeno-lysis

Pyruvate

Pyruvate

Lactate

Acetyl-CoA

Acetyl-CoA

ATP

ADP

Gly

coly

sis

GlycolysisGluconeo-genesis

TCAcycle TCA

cycle

Insulin, SREBP-1c,HNF4α, HNF6, USF1

GCK

GCKGCKG-6-Pase

GCK

GCKR

GCKR

GC

KR

GCK

Glucose

Glucose

Glucose

Glucose

P GlucoseP

GLUT-2

Pentosephosphate

shunt

NADPH

FA oxidation

DNL

GLY

GLY

Fig. 2. Regulation of hepatic glucose metabolism. (A) After intestinal absorption, glucose (Glu) reaches the hepatocyte via the portal vein. The insulin-independentglucose transporter 2 (GLUT2) shuttles Glu across the membrane. Abundance of glucose induces conformational changes of the glucokinase regulatory protein (GCKR),which binds to glucokinase (GCK) and keeps it in the nucleus in the fasting state. GCK is then released into the cytosol and phosphorylates Glu to glucose-6-phosphate (Glu-6-P); depending on the nutritional state, it serves as a substrate for glycolysis or glycogen synthesis, respectively. GCK is transcriptionally regulated by insulin and nuclearreceptor signaling. (B) Glu-6-P is a central intermediate in the hepatic glucose metabolism. It is degraded during glycolysis, which provides energy in the form of two ATPand two NADH molecules per glucose molecule. The product pyruvate is further decarboxylized to acetyl-CoA, which enters the intramitochondrial tricarboxylic acid cycle(TCA). Alternatively, Glu-6-P is degraded in the pentose-phosphate shunt, which provides NADPH, a co-substrate for DNL. Acetyl-CoA is an important product of the TCA,linking glucose and lipid metabolism, as it is the substrate for DNL (Fig. 1). Gluconeogenesis and glycogenolysis provide Glu-6-P as a substrate for glucose synthesis in thefasting state. Glucogenolysis is catalyzed by glycogen phosphorylase, activated by AMP, and repressed by insulin. The key enzyme in gluconeogenesis is PEPCK, which isrepressed by insulin signaling via Akt-mediated FoxO phosphorylation and activated by PPARa.

Review

activity, which has also been identified to facilitate FA uptake[34,35]. Besides the fact that the regulation of FATP activity isgenerally complex, the individual contribution of these FATPs toFA uptake has not been entirely clarified yet. Nevertheless, sig-naling via PPARa again predominantly regulates the transcriptionof these transport proteins as combined with hormonal regula-tion via insulin and leptin [30,36].

Macroautophagy

Autophagy has recently been implied to play a role in hepatic lipidhomeostasis [37]. As a lysosomal pathway, it recycles dispensablecellular constituents into important energy sources during thefasting state [38]. Recent animal studies revealed that autophagyis a key process in hepatic lipolysis and lipid droplet degradation[10,39]. As mentioned above, lysosomes process chylomicronremnants as well as TAGs that accumulate during hepatic lipogen-

956 Journal of Hepatology 201

esis. During starvation, macroautophagy leads to the fusion oflysosomes and lipid droplets into autophagosomes, which arethen degraded; FAs are thus released and can be catabolized viab-oxidation. Starvation leads to repression of the so-called mam-malian target of rapamycin (mTOR), an insulin downstream targetthat inhibits autophagy. Interestingly, rapamycin treatment alsodownregulates SREBP-1c in primary hepatocytes, which suggestsan effect of rapamycin, independent of mTOR, on the activity offorkhead box protein O (FoxO) [40]. Long-term repression ofautophagy is accounted for by insulin action, and Akt mediatedde-phosphorylation of FoxO, a transcriptional activator of autoph-agy related genes (ATGs). Indeed, FoxO simultaneously repressesSREBP-1c activation and thus DNL. Thus, insulin receptor activa-tion induces DNL and represses autophagy-mediated lipid dropletdegradation, both short- and long-term, via two distinct mecha-nisms. However, in the insulin-resistant state and in obesity perse, hepatic mTOR is over-activated and calpain, a repressor of

2 vol. 56 j 952–964

JOURNAL OF HEPATOLOGY

ATGs, is induced [9]. Despite FoxO activation in IR, mTOR and cal-pain activation account for the repression of hepatic autophagy inobese individuals.

Fatty acid oxidation

Oxidation of FAs occurs within mitochondria, peroxisomes and theER, and facilitates degradation of activated FAs to acetyl-CoA. It is arapid and effective way of energy allocation since, for example, theoxidation of one molecule of palmitate produces up to 129 ATPequivalents [41]. FAs are activated by acyl-CoA-synthetase toacyl-CoA in the cytosol. This process is indispensable for enablingFAs to cross membranes and enter organelles. While short- andmedium-chain FAs pass the mitochondrial membrane withoutactivation, activated long-chain fatty acids (LCFAs) are shuttledacross the membrane via carnitine palmitoyltransferase-1(CPT1). Malonyl-CoA, an early intermediate of DNL that accumu-lates upon insulin receptor activation, is an allosteric inhibitor ofCPT1 [42]. Thus, in the fed state, FA oxidation is inhibited andDNL-promoted, allowing for storage and distribution of lipids[43]. In general, short-, medium- and LCFAs are oxidized withinmitochondria (b-oxidation), while toxic, very-long-chain FAs areoxidized within peroxisomes (Fig. 1C). In diabetes or FA overload,cytochrome P450 (a.k.a. CYP4A)-dependent x-oxidation of LCFAsoccurs in the ER and induces ROS and lipid peroxidation [44]. Dur-ing the process of b-oxidation, electrons are indirectly donated tothe electron transport chain to drive ATP synthesis. Acetyl-CoAcan be further processed via the tricarboxylic acid cycle (TCA) or,in the case of FA abundance, be converted into ketone bodies[45]. PPARa and insulin signaling are again involved in the regula-tion of FA oxidation and the formation of ketone bodies via tran-scriptional regulation of mitochondrial HMG-CoA synthase.Interestingly, succinyl-CoA, an intermediate of the TCA, inactivatesmitochondrial HMG-CoA synthase by succinylation, which oncemore intercalates glucose- and lipid metabolism [46]. For a simpli-fied overview of key metabolic pathways in hepatocellular lipidhomeostasis, see Fig. 1B.

Key Points 2

• Insulin activates DNL via PI3K-mediated FAS expression

• SREBP-1c and ChREBP transcriptionally activate FAS and ACC as well as VLDL assembly

• Insulin represses peripheral lipolysis and liberation of FFAs from adipocytes

• Insulin represses autophagy via activation of mTOR and repression of FoxO phosphorylation

• Insulin represses FA oxidation

Hepatic glucose metabolism

Hepatocyte glucose uptake

In the postprandial state, blood glucose is taken up by the hepa-tocyte via the glucose transporter type 2 (GLUT2) – a membrane-bound transporter with high capacity and low affinity for glucose.

Journal of Hepatology 201

In contrast to GLUT4, which is expressed by muscle and adiposetissue, the expression and activity of GLUT2 is independent ofinsulin signaling. In pancreatic islet cells, GLUT2 is thus alsoreferred to as a ‘‘glucose sensor’’ [47]. Once taken up by the hepa-tocyte, glucose is phosphorylated to glucose-6-phosphate by liverglucokinase (L-GCK; Fig. 2), the rate limiting enzyme for hepaticglucose utilization [48]. In contrast to other hexokinases, GCK(syn.: hexokinase IV) is not inhibited by its product, which allowsfor postprandial glycogen storage within the hepatocyte. In thefasting state, L-GCK is inactive and bound to glucokinase regula-tory protein (GCKR) within the nucleus. Post-prandial glucoseabundance and insulin-action synergistically cause rapid dissoci-ation of L-GCK from GCKR and translocation to the cytoplasm[49]. L-GCK is transcriptionally regulated by SREBP-1c, hepaticnuclear factor-4-alpha (HNF4a), hepatic nuclear factor 6(HNF6), FoxO1, and upstream stimulatory factor 1 (USF1) (seeFig. 2A). Indeed, mutations in the GCK gene have been associatedwith IR and the pathogenesis of maturity-onset diabetes of theyoung (MODY) in several studies [50,51].

Glycolysis and glycogen synthesis

Glucose-6-phosphate is either further processed in glycolysis orutilized for glycogen synthesis, depending on the systemic meta-bolic state. Glycolysis, a ten-step process, metabolizes glucose topyruvate with a net gain of two ATP and two NADH molecules perglucose molecule. Glycolysis is regulated by L-GCK, which pro-vides glucose-6-phosphate, phosphofructokinase, which is inhib-ited by its product fructose-1,6-bisphosphate, AMP and pyruvate-kinase (PK), the final step in glycolysis. PK is activated by its sub-strate and inhibited by abundance of ATP. Insulin, epinephrine,and glucagon also regulate PK via the PI3K pathway and ChREBPinduces transcription of PK in the presence of glucose [23]. Pyru-vate is further decarboxylized to acetyl-CoA and then processedin the TCA or utilized for DNL. The pentose phosphate pathwayis an alternative way for degradation of glucose-6-phosphate inhepatocytes, which provides the cell with NADPH, an importantantioxidant and co-substrate for DNL and cholesterol synthesis.In hepatocellular carcinoma (HCC), glycolytic activity is dramati-cally upregulated and associated with increased hexokinase 2activity and expression of GLUT1, leading to altered glucose uti-lization, which has therapeutic and diagnostic implications (forthe so-called Warburg effect, Fig. 3B and C) [52].

Glycogen synthesis is catalyzed by glycogen synthase (GS)after conversion of glucose-6-phosphate to UDP-glucose [53].GS is regulated by the allosteric activator glucose-6-phosphateand is inactive in the phosphorylated state. Glycogen synthasekinase 3 (GSK3) phosphorylates GS and is a downstream targetof Akt/PI3K and thus insulin signaling. GSK3 is a multifunctionalkinase, involved in cell senescence, apoptosis and lipid metabo-lism via phosphorylation of SREBP-1c [54]. Other protein kinasesthat phosphorylate GS are AMP-activated protein kinase (AMPK)and protein kinase A (PKA). Insulin activates glycogen synthesisvia repression of PKA. GS synthesizes the glycogen polymer,which is further branched by a branching enzyme.

Glycogenolysis and gluconeogenesis

In the fasting state, the liver supplies the body with energy bybreaking down glycogen, and following prolonged fasting by glu-coneogenesis [55]. Glycogen breakdown is catalyzed by glycogen

2 vol. 56 j 952–964 957

A

C

B

Pyruvate

LactateAcetyl-CoA

Acetyl-CoA

Glycolysis

TCAcycle

GCK HK2

Glucose

GlucoseGLUT-1

NADPH

DNL

Pentosephosphate

shunt

GlucoseP

Healthy liver

Steatosis

NASH

Cirrhosis

HCC

Obesity

Insulin resistance

Inflammation

Fibrosis

FFA

TNFα

IL-6IL-1βER-stressROS

JNKJAK/STATNFκB

TUM

OR

IGEN

ESIS

GLUT-2 Tumor proliferation

Hepatoma cell

Fig. 3. Hepatic lipid and glucose metabolism in HCC. (A) Mediators of NAFLD progression also contribute to carcinogenesis in the liver. In general, obesity and diabeteshave been identified as risk factors for cancer development as well as inflammation. Cirrhosis is a precancerous condition and, in fact, most HCCs derive from cirrhotic livers.Important mediators of insulin resistance and lipotoxicity also induce dysplasia and carcinogenesis. This figure gives a brief overview of different cytokines and cellsignaling pathways involved in HCC development. (B) Expression of GLUT1 in hepatoma cells leads to increased hepatic glucose utilization. Glucokinase is downregulated,but hexokinase 2 (HK2) is now expressed and phosphorylates glucose with a higher affinity. Aerobic glycolysis leads to a rapid, but rather ineffective energy supply to theproliferating cancer cell. However, this process, referred to as the Warburg effect, facilitates uptake and de novo synthesis of nutrients (nucleotides, amino acids, lipids),available for cell proliferation and tumor growth. (C) Clinically, this effect is utilized in PET diagnostics. An increase in cancer cell glucose uptake, as visualized by FDGPET/CT, acts as an important tool in HCC diagnostics.

Review

958 Journal of Hepatology 2012 vol. 56 j 952–964

JOURNAL OF HEPATOLOGY

phosphorylase (PYGL), which cleaves glucose from the glycogenpolymer and produces glucose-1-phosphate, which is convertedto glucose-6-phosphate by phosphoglucomutase. A debranchingenzyme cleaves the last four glucose monomers. PYGL is regu-lated through allosteric activation by AMP and via phosphoryla-tion by PKA, which is inhibited by insulin.

Key Points 3

• NRs are highly conserved ligand-activated transcription

either endogenous (bile acids; fatty acids; hormones; vitamins) or exogenous (drugs; xenobiotic toxins), and induce conformational changes affecting the transcriptional activation of downstream targets [64, 65]. Most NRs heterodimerize with the retinoid-X-receptor. Table 1 sketches the most abundant NRs in the liver, their ligands, and their roles in hepatic lipid and glucose homeostasis

• PPARα is the most abundant PPAR in the healthy liver. It regulates FA oxidation and uptake as well as gluconeogenesis (PEPCK) and glycogen synthesis (GSK3; GS) (see above). Moreover, PPARα activation

hepatocyte proliferation via induction of microRNA let7c-signaling cascades [66, 67]. In mouse models, but not in humans, activation of PPARα induces hepatic carcinogenesis, which is in part explained by ROS production [68, 69] and induction of microRNA let7c that degrades the c-myc oncogene. Initially an orphan

the therapy of hypertriglyceridemia – and recently a

ligand [21], produced by the enzyme FAS, which in turn is activated by PPARα and other NRs. Emphasizing the cross-talk between NRs, FXR and LXR are known to activate PPARα signaling while VDR appears to repress PPARα signaling [58, 70, 71]. Post-transcriptionally, PPARα protein expression is repressed by microRNA 10b, which is associated with metastasis formation in a variety of cancers [72]

• PPARγ is mainly expressed in adipocytes, where it promotes lipid uptake and TAG storage by upregulation of the LDL receptor and CD36/FAT, induces stearoyl-CoA-desaturase-1 (SCD1), increases insulin sensitivity by induction of GLUT4, and decreases TNFα levels and thus systemic IR [73]. In NAFLD, PPARγ is upregulated

PPARγ KO mice are protected from diet-induced steatosis [74, 75]. PPARγ

signaling [76]. Recently, activation of PPARδ has been shown to reduce hepatic lipogenesis by suppression of SREPB-1c [76, 77] in addition to increasing the hepatic glucose catabolism as well as muscular oxidation

• The PPARγ co-activator-1α (PGC1α) is a transcriptional co-activator that regulates mitochondrial biology and energy homeostasis [78]. PGC1α is induced by fasting and activates FA oxidation and gluconeogenesis via induction of PPARα, FoxO1 and HNF-4α [79]. Sirtuin 1 (SIRT1) deacetylates PGC1α and thus regulates its activity [80]

factors. Specific ligands of most NRs are known, are

has anti-inflammatory effects and (in mice) induces

NR, PPARα is known to be the receptor for fibrates in

diacylglycerol has been identified as an endogeneous

in liver tissue, and liver-specific

has anti-inflammatory features as it represses NFκB

Journal of Hepatology 201

Hepatic gluconeogenesis occurs during prolonged fasting epi-sodes and begins intramitochondrially by the induction of pyru-vate carboxylase in abundance of acetyl-CoA. Interestingly,inhibition of hepatic CPT-1 and thus mitochondrial fatty acid oxi-dation significantly represses hepatic gluconeogenesis in mice[56]. Gluconeogenesis is further regulated via allosteric and

• FXR, the bile acid receptor and its downstream target, small heterodimer partner (SHP), have recently

metabolism as BA administration lowers blood lipids and the TAGs hepatic content. This effect is partially reversed in SHP KO mice and is mediated by transcriptional repression of SREBP-1c [81]. Thus, FXR represses hepatic DNL and VLDL export. FXR KO mice develop hepatic steatosis, and treatment with FXR agonist attenuates hepatic steatosis in mice when fed a

that induces steatosis and oxidative stress, rather than insulin resistance [82, 83]. Other studies also revealed that FXR agonists improve IR [64, 84]. In the intestine,

factors 15 (mouse) and 19 (human) (FGF15/19), which activates hepatic FA oxidation and insulin response via activation of the FGF receptor 4 [85, 86]. BAs have additional effects on lipid and glucose metabolisms via a G-protein-coupled receptor (TGR5/GBAR), which regulates energy expenditure in brown adipose tissue (and possibly skeletal muscle), and promotes intestinal GLP-1 secretion [87]. Thus, BAs may now be viewed as enterohepatic hormones that circulate with absorbed

• Natural ligands of LXR are hydroxysterols. LXRα is expressed primarily in hepatocytes, adipose tissue and macrophages, whereas LXRβ is expressed ubiquitously [88]. LXR regulates SREBP-1c, FAS and SCD-1 expression and LXRα agonist treatment leads to hypertriglyceremia secondary to induction of hepatic lipogenesis [89]. LXR increases hepatic glucose uptake by induction of GLUT4 expression and has

immune response [90, 91]

• LRH1 has recently been implied as a novel NR in BA homeostasis by transcriptional activation of Cyp7A1 [92]. Interestingly, LRH1 activation by dilauroyl phosphatidylcholine improves the hepatic insulin response and hepatic lipid accumulation in high fat diet-fed mice [93]. Accordingly, SREBP-1c, FAS and SCD-1 mRNA levels were downregulated in the treatment group

• The VDRRecent studies have shown that vitamin D supplementation protects mice from diet-induced steatosis and VDR KO mice spontaneously develop hepatic steatosis [94, 95]. Polymorphisms in VDR are associated with HCC development in patients with alcoholic cirrhosis, and VDR activation acts antiproliferative effects [96, 97]. In HCV patients, low

and the VDR regulates T-cell activation [97, 98]

been identified to play a central role in hepatic lipid

methionine- and choline- deficient diet (MCD), a model

FXR activation induces secretion of fibroblast growth

dietary lipids and fine-tune their metabolism

anti-inflammatory attributes as it modulates the innate

is a novel target in the field of hepatology.

vitamin D levels are associated with fibrosis progression,

2 vol. 56 j 952–964 959

Review

transcriptional activation of phosphoenolpyruvate carboxykinase(PEPCK), fructose-1,6-bisphosphatase and glucose-6-phosphatase(G6Pase). Overexpression of PEPCK promotes IR in mice [57].Insulin action represses PEPCK expression via Akt-mediatedFoxO1 phosphorylation. This pathway is repressed by activationof PPARa signaling-mediated induction of the pseudokinase trib-bles-homologue 3 (TRB3), and TRB3 expression is associated withIR in patients [58,59]. Other NRs (PGC1a, FXR), glucagon and glu-cocorticoids also mediate expression of PEPCK and G6Pase[55,60]. FoxO1, a transcriptional activator of PEPCK and G6Pase,is both directly and indirectly activated by PGC1a, HNF4a, CREBPand PPARa [61]. PEPCK knock-out mice not only show decreasedgluconeogenesis, but more importantly a decreased removal ofTCA anions, which causes hepatic TAG accumulation and steato-sis [62]. In mice, PEPCK overexpression in the striated muscleinterestingly leads to longevity and an impressive increase inmuscle strength and endurance; this phenotype is partiallyexplained by an optimized mitochondrial oxidation of TAG [63].These processes again demonstrate the close inter-relationbetween gluconeogenesis and hepatic lipid metabolism (Fig. 2B).

Hepatic lipid and glucose metabolism in liver injury

Serum FFA levels correlate with hepatocyte apoptosis and FAswere found to activate death receptor-mediated apoptosis[64,65]. On the other hand, high glucose concentrations induceapoptosis in hepatoma cell lines, and the HOMA score is associ-ated with hepatocyte apoptosis in NAFLD patients [66,67]. Theseobservations indicate that hepatocyte apoptosis due to an imbal-ance in cell metabolism has clinical implications for NAFLD, liverregeneration, as well as fibrogenesis and carcinogenesis. Theeffects of systemic IR on liver injury via induction of TNFa are dis-cussed in detail below. Intrahepatic fat deposition and obesitydecrease hepatic blood flow by direct compression and systemichypercatecholemia and thus inhibit mitochondrial function andcause formation of ROS, which induces Kupffer cell activation,hepatic inflammation, and hepatocyte apoptosis [68]. Interest-ingly, mitochondrial stress may be partially reversed by treat-ment with insulin-like-growth factor 1 (IGF-1) and PPARcagonists [69,70]. Alcohol-fed PPARa knock-out mice develop aphenotype that mimics alcoholic liver disease in humans, whichis linked to ROS accumulation [71]. Overexpression studies iden-tified that the expression of PEPCK links mitochondrial dysfunc-tion with the ER stress response [72].

As mitochondria are the main organelles in energy combus-tion, the ER is the major site of protein folding and trafficking.Recently, activation of the unfolded protein response (UPR)within the ER has been implied as a key modulator of cellularinflammation and is linked to IR, lipid and glucose metabolism[73]. Three membrane-bound proteins regulate the UPR withinthe ER; PKR-like eukaryotic initiation factor 2a kinase (PERK),inositol-requiring enzyme 1 (IRE1), and activating transcriptionfactor-6 (ATF6). The protein kinase PERK affects transcriptionalregulation of rRNA, which activates NFjB and ATF4 signaling.NFjB regulates inflammatory signaling (IL-6, TNFa), and ATF4regulates glucose metabolism [74,75]. SREBP-1c activation occursduring ER stress and thus affects the lipid metabolism – and theER possibly regulates the number, composition and quality oflipid droplets [76–78]. Furthermore, induction of the mTORpathway in obese individuals activates SREBP-1c, promotes ER

960 Journal of Hepatology 201

stress and inhibits autophagy, a novel pathway in the biologyof hepatic lipid droplets [79,80].

NAFLD, insulin resistance and lipotoxicity

NAFLD represents the most prevalent liver disease in Westernsocieties. It presents with a wide spectrum ranging from simplesteatosis or non-alcoholic fatty liver (NAFL) to fully developedNASH with or without fibrosis. NASH can progress to fibrosis withan increased risk to develop end-stage liver disease or hepatocel-lular carcinoma (HCC) [1]. Hepatic steatosis is defined as an intra-hepatic accumulation of TAGs. In parallel, abundant FAs causelipotoxicity via the induction of ROS release, which causes inflam-mation, apoptosis, and thus the progression to NASH and fibro-genesis [81]. As described above, most FAs derive from thecirculation secondary to increased lipolysis in adipose tissue aswell as DNL [3].

Obesity increases the TNFa production in adipocytes, whichfacilitates adipocyte IR and increases lipolysis rate [82]. Thus,the circulating FFA pool is increased in obese individuals andaccounts for the majority of liver lipids in NAFLD [13]. As men-tioned, uptake of FFAs into the hepatocyte is facilitated by a vari-ety of FATPs; several studies found an upregulation of thesetransporters in NAFLD and NASH as well as a correlation with dis-ease severity [65,83,84]. Fifteen percent of the lipid contentwithin the steatotic liver derives from an increased dietary intakeof lipids [3]. DNL may account for up to 30% of TAGs in steatoticlivers, a mechanism that involves dysregulation in SREBP-1c- andFoxO-mediated hepatic insulin signaling [4,85,86]. Since autoph-agy-related genes are transcriptionally activated by FoxO andinsulin action modulates autophagy, recent studies showed thatmacroautophagy is dysregulated in the metabolic syndrome[87]. Accordingly, in conditional atg7-knockout mice, Singhet al. observed an increase in hepatic lipid accumulation, and ingenetic and dietary mouse models for obesity and hepatic steato-sis, autophagy-related genes were downregulated [10,12].

The long-lasting paradigm claiming TAG accumulation to bethe ‘‘first hit‘‘ that predisposes to further liver damage in thepathogenesis of NASH has recently been replaced by a more com-plex model as emerging evidence points to FAs and their metab-olites as the true lipotoxic agents [7]. Interestingly, lipidaccumulation and altered composition of phospholipids withinER membranes further promotes ER stress and IR in obese mice[8]. Cytosolic TAGs are thus now considered to be inert and, infact, lipid droplet accumulation has recently been found to behepatoprotective [88]. Notably, genetic deletion of DGAT2(responsible for TAG formation) increases hepatocellular injuryin MCD-fed mice despite a reduction in the content of hepatocel-lular TAGs [89]. However, TAG accumulation and lipid dropletformation accompany and parallel pathophysiologic mechanismsin NASH. FFAs are thus now in the focus of basic research – andFAs as well as acyl-CoA and acetyl-CoA have been identified aspotential causes of lipotoxicity [90]. FAs have been found to acti-vate Toll-like receptors and initiate the extrinsic apoptosis cas-cade [91,92]. FAs also interfere with NR signaling, which mightadditionally influence the extent of hepatocyte damage and fur-ther promote IR and ER stress [93,94]. Accordingly, b-oxidationof LCFA within peroxisomes and x-oxidation within the ER areupregulated in NASH and contribute to lipotoxicity and ROSformation [95,96]. This might be secondary to inhibition of

2 vol. 56 j 952–964

JOURNAL OF HEPATOLOGY

mitochondrial b-oxidation due to an accumulation of malonyl-CoA and the inhibition of CPT1 [3,97,98]. In fact, recent studiesindicate that activation of mitochondrial FA oxidation protectsfrom steatosis and IR [99,100]. As mentioned above, DAGs andceramides might as well contribute to hepatocyte damage andIR in NAFLD [7].

These stressors induce a variety of intracellular and paracrinemechanisms that may promote hepatocellular damage. FAsinduce the production of TNFa, and hepatic TNF receptor expres-sion correlates with the disease severity in NAFLD [101]. TNFreceptor activation increases expression of SREBP-1c, whichinduces hepatic lipogenesis and lipid accumulation [102]. AsTNFa-mediated effects are antagonized by adiponectin, adipo-nectin receptors are actually downregulated in NASH [103]. TNFaactivation is further paralleled by death-receptor expression,which facilitates activation of the extrinsic apoptosis cascade.Apoptosis indeed is the predominant form of hepatocellularinjury in NASH [64,104]. In fact, apoptotic activity within the dis-eased liver correlates with disease severity and thus cleaved cyto-keratin-18 fragments in the serum of NAFLD could effectively beutilized as surrogate markers for the progression of NAFLD [105].As previously mentioned, FA accumulation also leads to inductionof ER stress and ROS formation, which again promotes hepaticinjury [1,106].

In summary, while hepatic TAG accumulation seems to be abenign symptom of hepatic steatosis, FA metabolites contributeto the progression of NAFLD to NASH. IR promotes the recruit-ment of FFAs from the serum pool as well as intrahepatic FAaccumulation, which induces apoptosis and ROS formation. FAsthemselves also promote hepatic IR via TNF receptor activation,hence indicating a vicious circle of lipid accumulation and IR asa crucial mechanism in the pathogenesis of NASH (Fig. 1D).

Acute liver injury and regeneration

Injured livers produce cytokine signals that trigger adipose tissueto release FAs into the circulation. Indeed, in the acute response,hepatocytes initiate the transcription of lipogenic genes and accu-mulate TAGs in intracellular lipid droplets. In liver regeneration(e.g., after partial hepatectomy (PHx)), lipid droplet formation isessential to propagate the proliferative response by sufficientlysupplying the organ’s energy household. Droplet-stored lipidsare also used for synthesizing new lipoproteins, bile acids, andentire membranes [107,108]. As previously mentioned, emergingdata supports a close connection between BA and FA metabolism.In this context, FATP-5 has also been suggested to participate in BAmetabolism as a bile acid-CoA ligase [109,110]. FATP-5 isexpressed preferentially towards the space of Disse and closely fol-lows the hepatic sinusoids [109] where FAs and BAs are absorbedfrom the enterohepatic circulation. Intriguingly, it has beenrecently shown that elevated BAs accelerate liver regenerationafter PHx by bile acid receptor (FXR)-dependent signals [107,111].

Autophagy in acute liver injury might to a certain degree acthepatoprotectively as it rapidly supplies the hepatocyte withenergy. However, hyperactivation of autophagy induces celldeath, and the necrosis rate is actually a predictor of liver failure[9,112]. Furthermore, as we previously elucidated key processesin the interplay between adipose tissue and the liver, adiponectinwas identified as an important mediator of STAT3 signaling in theregenerating liver [113,114].

Journal of Hepatology 201

Liver cancer

Emerging data supports a role for lipid and glucose metabolismin hepatocellular carcinogenesis [115]. First of all, the increasein the prevalence of NAFLD is closely associated with the devel-opment of cirrhosis and NASH-related HCC [116]. Accordingly,NASH might account for the majority of HCC in the context of acryptogenic cirrhosis [117]. Obesity and diabetes have been iden-tified as independent risk factors for developing HCC (Fig. 3A)[118,119]. Several studies could recently dissect the effects ofindividual confounders of the metabolic syndrome and clearlydemonstrate a significant risk for NAFLD patients to developHCC [120]. Intriguingly, HCC in NAFLD not only arises from cir-rhotic tissue; this might imply a special role for lipid metabolismand lipotoxicity in the carcinogenesis within this cohort [121].

To date however, the vast majority of HCC cases arise frompatients with cirrhosis secondary to chronic HBV or HCV infection[122]. Wu et al. recently identified various alterations in multiplegenes involved in hepatic lipogenesis, FA oxidation, and TAGmetabolism in tumors derived from HCV patients compared toHCV-positive control patients [123]. Mechanistically, mediatorsand signaling cascades involved in the hepatic IR and the regula-tion of hepatic lipid metabolism are in the focus of HCC research.Several studies identified the insulin receptor downstream targetsPI3-kinase/Akt and IRS to be activated in HCC when compared tosurrounding healthy tissue, which mediates GSK3b phosphoryla-tion and mTOR activation [124]. These mediators not only interactwith b-catenin signaling, but also with autophagy and glycogensynthesis. In fact, knock-out mice with specific deletions inautophagy-related genes show spontaneous development ofhepatocellular dysplasia [125]. As pointed out before, TNFainduces hepatic IR and thus promotes lipid accumulation. TNFand IL-6 signaling is also involved in the activation of the JAK/STATand ERK pathways; both mediators are crucial for the developmentof obesity induced-HCC in a rodent model [126]. On the otherhand, JAK/STAT signaling is closely linked to, and partially modu-lated by, the adipocytokines leptin and adiponectin [127].

NRs play a crucial role in hepatic lipid and glucose homeosta-sis (see also above) and are important mediators of hepatic carci-nogenesis. PPARa mediates hepatic FA-oxidation and transport aswell as Akt phosphorylation via its downstream target TRB-3[15]. PPARa agonists promote carcinogenesis in a rodent modelfor HCV infection [128]. Interestingly, PPARa agonists not onlyinterfere with the binding of PPARa to SREBP-1c, but also withSTAT3 in hepatoma cell lines, thus indicating a potential mecha-nism for its carcinogenic properties [129]. In contrast, PPARc ago-nists prevent NAFLD progression and, more interestingly, were insome studies described to act antineoplastically [130,131]. Nota-bly, the majority of mechanistical data in hepatic carcinogenesisderives from hepatoma cell lines and murine models that areonly of limited value for the elucidation of hepatic carcinogenesisin humans.

Conclusions

Hepatic lipid and glucose metabolism are closely interrelatedwith inflammatory, proliferative and apoptotic signaling withinthe liver. In the liver, these catabolic and anabolic pathwayscan hardly be separated. They share intermediate metabolitesand receptor signaling, and go hand in hand in the pathogenesis

2 vol. 56 j 952–964 961

Review

of the most common liver diseases. Intriguingly, the case thatthese metabolic pathways are also involved in cell proliferation,regeneration and carcinogenesis implies potent future therapeu-tic approaches for life-threatening diseases. The enhanced under-standing of these basic mechanisms is thus imperative as wewitness a rising prevalence of obesity and the metabolicsyndrome.

Conflict of interest

The authors do not have a relationship with the manufacturers ofthe drugs involved either in the past or present and did notreceive funding from the manufacturers to carry out theirresearch. The authors received support from DFG, Wilhelm Lau-pitz Foundation, EASL, and IFORES.

Financial support

AC is supported by the Deutsche Forschungsgemeinschaft (DFG,grant 267/6-1 and 267/8-1), and the Wilhelm Laupitz Foundation.LPB is supported by an EASL Sheila Sherlock short-term fellow-ship and the IFORES Program of the University of Duisburg-Essen.

Acknowledgements

The authors would like to thank Robert Gieseler von der Crone,Jan-Peter Sowa, Judith Ertle and Thomas Schreiter for instructivediscussions and careful review of this manuscript.

References

[1] Feldstein AE. Novel insights into the pathophysiology of nonalcoholic fattyliver disease. Semin Liver Dis 2010;30:391–401.

[2] Browning JD, Szczepaniak LS, Dobbins R, Nuremberg P, Horton JD, Cohen JC,et al. Prevalence of hepatic steatosis in an urban population in the UnitedStates: impact of ethnicity. Hepatology (Baltimore, Md) 2004;40:1387–1395.

[3] Donnelly KL, Smith CI, Schwarzenberg SJ, Jessurun J, Boldt MD, Parks EJ.Sources of fatty acids stored in liver and secreted via lipoproteins inpatients with nonalcoholic fatty liver disease. J Clin Invest2005;115:1343–1351.

[4] Postic C, Girard J. Contribution of de novo fatty acid synthesis to hepaticsteatosis and insulin resistance. lessons from genetically engineered mice. JClin Invest 2008;118:829–838.

[5] Samuel VT, Petersen KF, Shulman GI. Lipid-induced insulin resistance.unravelling the mechanism. Lancet 2010;375:2267–2277.

[6] Sanyal AJ, Campbell-Sargent C, Mirshahi F, Rizzo WB, Contos MJ, SterlingRK, et al. Nonalcoholic steatohepatitis: association of insulin resistance andmitochondrial abnormalities. Gastroenterology 2001;120:1183–1192.

[7] Neuschwander-Tetri BA. Hepatic lipotoxicity and the pathogenesis ofnonalcoholic steatohepatitis: the central role of nontriglyceride fatty acidmetabolites. Hepatology (Baltimore, Md) 2010;52:774–788.

[8] Fu S, Yang L, Li P, Hofmann O, Dicker L, Hide W, et al. Aberrant lipidmetabolism disrupts calcium homeostasis causing liver endoplasmicreticulum stress in obesity. Nature 2011.

[9] Rautou PE, Mansouri A, Lebrec D, Durand F, Valla D, Moreau R. Autophagy inliver diseases. J. Hepatol. 2010;53:1123–1134.

[10] Singh R, Kaushik S, Wang Y, Xiang Y, Novak I, Komatsu M, et al. Autophagyregulates lipid metabolism. Nature 2009;458:1131–1135.

[11] Wang Y, Singh R, Xiang Y, Czaja MJ. Macroautophagy and chaperone-mediated autophagy are required for hepatocyte resistance to oxidantstress. Hepatology (Baltimore, Md) 2010;52:266–277.

[12] Yang L, Li P, Fu S, Calay ES, Hotamisligil GS. Defective hepatic autophagy inobesity promotes ER stress and causes insulin resistance. Cell Metab2010;11:467–478.

962 Journal of Hepatology 201

[13] Savage DB, Semple RK. Recent insights into fatty liver, metabolic dyslip-idaemia and their links to insulin resistance. Curr Opin Lipidol2010;21:329–336.

[14] Gupta NA, Mells J, Dunham RM, Grakoui A, Handy J, Saxena NK, et al.Glucagon-like peptide-1 receptor is present on human hepatocytes and hasa direct role in decreasing hepatic steatosis in vitro by modulating elementsof the insulin signaling pathway. Hepatology (Baltimore, Md)2010;51:1584–1592.

[15] Pyper SR, Viswakarma N, Yu S, Reddy JK. PPARalpha: energy combustion,hypolipidemia, inflammation and cancer. Nucl Recept Signal 2010;8:e002.

[16] Timlin MT, Parks EJ. Temporal pattern of de novo lipogenesis in thepostprandial state in healthy men. Am J Clin Nutr 2005;81:35–42.

[17] Merkel M, Eckel RH, Goldberg IJ. Lipoprotein lipase: genetics, lipid uptake,and regulation. J Lipid Res 2002;43:1997–2006.

[18] Knight BL, Hebbachi A, Hauton D, Brown AM, Wiggins D, Patel DD, et al. Arole for PPARalpha in the control of SREBP activity and lipid synthesis in theliver. Biochem J 2005;389:413–421.

[19] Schadinger SE, Bucher NL, Schreiber BM, Farmer SR. PPARgamma2 regu-lates lipogenesis and lipid accumulation in steatotic hepatocytes. Am JPhysiol Endocrinol Metab 2005;288:E1195–1205.

[20] Shen LL, Liu H, Peng J, Gan L, Lu L, Zhang Q, et al. Effects of farnesoid Xreceptor on the expression of the fatty acid synthetase and hepatic lipase.Mol Biol Rep 2011;38:553–559.

[21] Chakravarthy MV, Lodhi IJ, Yin L, Malapaka RR, Xu HE, Turk J, et al.Identification of a physiologically relevant endogenous ligand for PPARal-pha in liver. Cell 2009;138:476–488.

[22] Sul HS, Latasa MJ, Moon Y, Kim KH. Regulation of the fatty acid synthasepromoter by insulin. J Nutr 2000;130:315S–320S.

[23] Dentin R, Girard J, Postic C. Carbohydrate responsive element bindingprotein (ChREBP) and sterol regulatory element binding protein-1c (SREBP-1c): two key regulators of glucose metabolism and lipid synthesis in liver.Biochimie 2005;87:81–86.

[24] Coleman RA, Lee DP. Enzymes of triacylglycerol synthesis and theirregulation. Prog Lipid Res 2004;43:134–176.

[25] Postic C, Girard J. The role of the lipogenic pathway in the development ofhepatic steatosis. Diabetes Metab 2008;34:643–648.

[26] Arner P. Human fat cell lipolysis: biochemistry, regulation and clinical role.Best Pract Res Clin Endocrinol Metab 2005;19:471–482.

[27] Delarue J, Magnan C. Free fatty acids and insulin resistance. Curr Opin ClinNutr Metab Care 2007;10:142–148.

[28] Berk PD. Regulatable fatty acid transport mechanisms are central to thepathophysiology of obesity, fatty liver, and metabolic syndrome. Hepatol-ogy (Baltimore, Md) 2008;48:1362–1376.

[29] Martin G, Nemoto M, Gelman L, Geffroy S, Najib J, Fruchart JC, et al. Thehuman fatty acid transport protein-1 (SLC27A1; FATP-1) cDNA and gene:organization, chromosomal localization, and expression. Genomics2000;66:296–304.

[30] Ge F, Zhou S, Hu C, Lobdell Ht, Berk PD. Insulin- and leptin-regulated fattyacid uptake plays a key causal role in hepatic steatosis in mice with intactleptin signaling but not in ob/ob or db/db mice. Am J Physiol GastrointestLiver Physiol 2010;299:G855–866.

[31] Zhou SL, Stump D, Sorrentino D, Potter BJ, Berk PD. Adipocytedifferentiation of 3T3–L1 cells involves augmented expression of a43-kDa plasma membrane fatty acid-binding protein. J Biol Chem1992;267:14456–14461.

[32] Zhou SL, Stump D, Kiang CL, Isola LM, Berk PD. Mitochondrial aspartateaminotransferase expressed on the surface of 3T3–L1 adipocytes mediatessaturable fatty acid uptake. Proc Soc Exp Biol Med 1995;208:263–270.

[33] Trigatti BL, Anderson RG, Gerber GE. Identification of caveolin-1 as a fattyacid binding protein. Biochem Biophys Res Commun 1999;255:34–39.

[34] Abumrad NA, el-Maghrabi MR, Amri EZ, Lopez E, Grimaldi PA. Cloning of arat adipocyte membrane protein implicated in binding or transport of long-chain fatty acids that is induced during preadipocyte differentiation.Homology with human CD36. J Biol Chem 1993;268:17665–17668.

[35] Silverstein RL, Febbraio M. CD36, a scavenger receptor involved inimmunity, metabolism, angiogenesis, and behavior. Sci Signal 2009;2:re3.

[36] Wierzbicki M, Chabowski A, Zendzian-Piotrowska M, Gorski J. Differentialeffects of in vivo PPAR alpha and gamma activation on fatty acid transportproteins expression and lipid content in rat liver. J Physiol Pharmacol2009;60:99–106.

[37] Czaja MJ. Autophagy in health and disease. 2. Regulation of lipid metab-olism and storage by autophagy: pathophysiological implications. Am JPhysiol Cell Physiol 2010;298:C973–978.

[38] Finn PF, Dice JF. Proteolytic and lipolytic responses to starvation. Nutrition2006;22:830–844.

2 vol. 56 j 952–964

JOURNAL OF HEPATOLOGY

[39] Shibata M, Yoshimura K, Furuya N, Koike M, Ueno T, Komatsu M, et al. The

MAP1-LC3 conjugation system is involved in lipid droplet formation.Biochem Biophys Res Commun 2009;382:419–423.

[40] Li S, Brown MS, Goldstein JL. Bifurcation of insulin signaling pathway in ratliver: mTORC1 required for stimulation of lipogenesis, but not inhibition ofgluconeogenesis. Proc Natl Acad Sci U S A 2010;107:3441–3446.

[41] Lehninger AL, Nelson DL, Cox MM. Lehninger principles of biochemistry.4th ed. New York: W.H. Freeman; 2005.

[42] McGarry JD, Brown NF. The mitochondrial carnitine palmitoyltransferasesystem. From concept to molecular analysis. Eur J Biochem 1997;244:1–14.

[43] Akkaoui M, Cohen I, Esnous C, Lenoir V, Sournac M, Girard J, et al.Modulation of the hepatic malonyl-CoA-carnitine palmitoyltransferase 1Apartnership creates a metabolic switch allowing oxidation of de novo fattyacids. Biochem J 2009;420:429–438.

[44] Reddy JK. Nonalcoholic steatosis and steatohepatitis. III. Peroxisomal beta-oxidation, PPAR alpha, and steatohepatitis. Am J Physiol Gastrointest LiverPhysiol 2001;281:G1333–1339.

[45] Nguyen P, Leray V, Diez M, Serisier S, Le Bloc’h J, Siliart B, et al. Liver lipidmetabolism. J Anim Physiol Anim Nutr (Berl) 2008;92:272–283.

[46] Hegardt FG. Mitochondrial 3-hydroxy-3-methylglutaryl-CoA synthase: acontrol enzyme in ketogenesis. Biochem J 1999;338:569–582.

[47] Leturque A, Brot-Laroche E, Le Gall M. GLUT2 mutations, translocation, andreceptor function in diet sugar managing. Am J Physiol Endocrinol Metab2009;296:E985–992.

[48] Agius L. Glucokinase and molecular aspects of liver glycogen metabolism.Biochem J 2008;414:1–18.

[49] Chu CA, Fujimoto Y, Igawa K, Grimsby J, Grippo JF, Magnuson MA,et al. Rapid translocation of hepatic glucokinase in response tointraduodenal glucose infusion and changes in plasma glucose andinsulin in conscious rats. Am J Physiol Gastrointest Liver Physiol2004;286:G627–634.

[50] Miller SP, Anand GR, Karschnia EJ, Bell GI, LaPorte DC, Lange AJ. Charac-terization of glucokinase mutations associated with maturity-onset diabe-tes of the young type 2 (MODY-2): different glucokinase defects lead to acommon phenotype. Diabetes 1999;48:1645–1651.

[51] Cuesta-Munoz AL, Tuomi T, Cobo-Vuilleumier N, Koskela H, Odili S,Stride A, et al. Clinical heterogeneity in monogenic diabetes caused bymutations in the glucokinase gene (GCK-MODY). Diabetes Care 2010;33:290–292.

[52] Amann T, Hellerbrand C. GLUT1 as a therapeutic target in hepatocellularcarcinoma. Expert Opin Ther Targets 2009;13:1411–1427.

[53] Roach PJ. Glycogen and its metabolism. Curr Mol Med 2002;2:101–120.[54] Kim YM, Seo YH, Park CB, Yoon SH, Yoon G. Roles of GSK3 in metabolic shift

toward abnormal anabolism in cell senescence. Ann N Y Acad Sci2010;1201:65–71.

[55] Raddatz D, Ramadori G. Carbohydrate metabolism and the liver: actualaspects from physiology and disease. Z Gastroenterol 2007;45:51–62.

[56] Conti R, Mannucci E, Pessotto P, Tassoni E, Carminati P, Giannessi F, et al.Selective reversible inhibition of liver carnitine palmitoyl-transferase 1 byteglicar reduces gluconeogenesis and improves glucose homeostasis.Diabetes 2011;60:644–651.

[57] Valera A, Pujol A, Pelegrin M, Bosch F. Transgenic mice overexpressingphosphoenolpyruvate carboxykinase develop non-insulin-dependent dia-betes mellitus. Proc Natl Acad Sci U S A 1994;91:9151–9154.

[58] Stayrook KR, Bramlett KS, Savkur RS, Ficorilli J, Cook T, Christe ME, et al.Regulation of carbohydrate metabolism by the farnesoid X receptor.Endocrinology 2005;146:984–991.

[59] Oberkofler H, Pfeifenberger A, Soyal S, Felder T, Hahne P, Miller K, et al.Aberrant hepatic TRIB3 gene expression in insulin-resistant obese humans.Diabetologia 2010;53:1971–1975.

[60] Cao R, Cronk ZX, Zha W, Sun L, Wang X, Fang Y, et al. Bile acids regulatehepatic gluconeogenic genes and farnesoid X receptor via G(alpha)i-protein-coupled receptors and the AKT pathway. J Lipid Res2010;51:2234–2244.

[61] Puigserver P, Rhee J, Donovan J, Walkey CJ, Yoon JC, Oriente F, et al. Insulin-regulated hepatic gluconeogenesis through FOXO1-PGC-1alpha interaction.Nature 2003;423:550–555.

[62] Hakimi P, Johnson MT, Yang J, Lepage DF, Conlon RA, Kalhan SC, et al.Phosphoenolpyruvate carboxykinase and the critical role of cataplerosis inthe control of hepatic metabolism. Nutr Metab (Lond) 2005;2:33.

[63] Hanson RW, Hakimi P. Born to run; the story of the PEPCK-Cmus mouse.Biochimie 2008;90:838–842.

[64] Feldstein AE, Canbay A, Guicciardi ME, Higuchi H, Bronk SF, Gores GJ. Dietassociated hepatic steatosis sensitizes to Fas mediated liver injury in mice.J. Hepatol. 2003;39:978–983.

Journal of Hepatology 201

[65] Bechmann LP, Gieseler RK, Sowa JP, Kahraman A, Erhard J, Wedemeyer I,et al. Apoptosis is associated with CD36/fatty acid translocase upregulationin non-alcoholic steatohepatitis. Liver Int 2010;30:850–859.

[66] Chandrasekaran K, Swaminathan K, Chatterjee S, Dey A. Apoptosis inHepG2 cells exposed to high glucose. Toxicol In Vitro 2010;24:387–396.

[67] Civera M, Urios A, Garcia-Torres ML, Ortega J, Martinez-Valls J, Cassinello N,et al. Relationship between insulin resistance, inflammation and liver cellapoptosis in patients with severe obesity. Diabetes Metab Res Rev2010;26:187–192.

[68] Sato N. Central role of mitochondria in metabolic regulation of liverpathophysiology. J Gastroenterol Hepatol 2007;22:S1–6.

[69] Puche JE, Garcia-Fernandez M, Muntane J, Rioja J, Gonzalez-Baron S, CastillaCortazar I. Low doses of insulin-like growth factor-I induce mitochondrialprotection in aging rats. Endocrinology 2008;149:2620–2627.

[70] Wang X, Wang Z, Liu JZ, Hu JX, Chen HL, Li WL, et al. Double antioxidantactivities of rosiglitazone against high glucose-induced oxidative stress inhepatocyte. Toxicol In Vitro 2011.

[71] Okiyama W, Tanaka N, Nakajima T, Tanaka E, Kiyosawa K, Gonzalez FJ, et al.Polyenephosphatidylcholine prevents alcoholic liver disease in PPARalpha-null mice through attenuation of increases in oxidative stress. J Hepatol2009;50:1236–1246.

[72] Lim JH, Lee HJ, Ho Jung M, Song J. Coupling mitochondrial dysfunction toendoplasmic reticulum stress response: a molecular mechanism leading tohepatic insulin resistance. Cell Signal 2009;21:169–177.

[73] Hotamisligil GS. Endoplasmic reticulum stress and the inflammatory basisof metabolic disease. Cell 2010;140:900–917.

[74] Jiang HY, Wek SA, McGrath BC, Scheuner D, Kaufman RJ, Cavener DR, et al.Phosphorylation of the alpha subunit of eukaryotic initiation factor 2 isrequired for activation of NF-kappaB in response to diverse cellularstresses. Mol Cell Biol 2003;23:5651–5663.

[75] Seo J, Fortuno 3rd ES, Suh JM, Stenesen D, Tang W, Parks EJ, et al. Atf4regulates obesity, glucose homeostasis, and energy expenditure. Diabetes2009;58:2565–2573.

[76] Kammoun HL, Chabanon H, Hainault I, Luquet S, Magnan C, Koike T, et al.GRP78 expression inhibits insulin and ER stress-induced SREBP-1c activa-tion and reduces hepatic steatosis in mice. J Clin Invest 2009;119:1201–1215.

[77] Rutkowski DT, Wu J, Back SH, Callaghan MU, Ferris SP, Iqbal J, et al. UPRpathways combine to prevent hepatic steatosis caused by ER stress-mediated suppression of transcriptional master regulators. Dev Cell2008;15:829–840.

[78] Gregor MG, Hotamisligil GS. Adipocyte stress: the endoplasmic reticulumand metabolic disease. J Lipid Res 2007.

[79] Porstmann T, Santos CR, Griffiths B, Cully M, Wu M, Leevers S, et al. SREBPactivity is regulated by mTORC1 and contributes to Akt-dependent cellgrowth. Cell Metab 2008;8:224–236.

[80] Jung CH, Ro SH, Cao J, Otto NM, Kim DH. mTOR regulation of autophagy.FEBS Lett 2010;584:1287–1295.

[81] Cheung O, Sanyal AJ. Abnormalities of lipid metabolism in nonalcoholicfatty liver disease. Semin Liver Dis 2008;28:351–359.

[82] Hotamisligil GS. Inflammation and metabolic disorders. Nature 2006;444:860–867.

[83] Bieghs V, Wouters K, van Gorp PJ, Gijbels MJ, de Winther MP, Binder CJ,et al. Role of scavenger receptor A and CD36 in diet-induced nonalcoholicsteatohepatitis in hyperlipidemic mice. Gastroenterology 2010;138:2477–2486 (2486 e2471-2473).

[84] Berk PD, Zhou S, Bradbury MW. Increased hepatocellular uptake of longchain fatty acids occurs by different mechanisms in fatty livers due toobesity or excess ethanol use, contributing to development of steatohep-atitis in both settings. Trans Am Clin Climatol Assoc 2005;116:335–344(discussion 345).

[85] Sajan MP, Standaert ML, Rivas J, Miura A, Kanoh Y, Soto J, et al. Role ofatypical protein kinase C in activation of sterol regulatory element bindingprotein-1c and nuclear factor kappa B (NFkappaB) in liver of rodents usedas a model of diabetes, and relationships to hyperlipidaemia and insulinresistance. Diabetologia 2009;52:1197–1207.

[86] Taniguchi CM, Kondo T, Sajan M, Luo J, Bronson R, Asano T, et al. Divergentregulation of hepatic glucose and lipid metabolism by phosphoinositide 3-kinase via Akt and PKClambda/zeta. Cell Metab 2006;3:343–353.

[87] Liu HY, Han J, Cao SY, Hong T, Zhuo D, Shi J, et al. Hepatic autophagy issuppressed in the presence of insulin resistance and hyperinsulinemia:inhibition of FoxO1-dependent expression of key autophagy genes byinsulin. J Biol Chem 2009;284:31484–31492.

[88] Ricchi M, Odoardi MR, Carulli L, Anzivino C, Ballestri S, Pinetti A, et al.Differential effect of oleic and palmitic acid on lipid accumulation and

2 vol. 56 j 952–964 963

Review

apoptosis in cultured hepatocytes. J Gastroenterol Hepatol 2009;24:830–840.

[89] Yamaguchi K, Yang L, McCall S, Huang J, Yu XX, Pandey SK, et al. Inhibitingtriglyceride synthesis improves hepatic steatosis but exacerbates liverdamage and fibrosis in obese mice with nonalcoholic steatohepatitis.Hepatology (Baltimore, Md) 2007;45:1366–1374.

[90] Han MS, Park SY, Shinzawa K, Kim S, Chung KW, Lee JH, et al. Lysophos-phatidylcholine as a death effector in the lipoapoptosis of hepatocytes. JLipid Res 2008;49:84–97.

[91] Fessler MB, Rudel LL, Brown JM. Toll-like receptor signaling links dietaryfatty acids to the metabolic syndrome. Curr Opin Lipidol 2009;20:379–385.

[92] Feldstein AE, Werneburg NW, Canbay A, Guicciardi ME, Bronk SF,Rydzewski R, et al. Free fatty acids promote hepatic lipotoxicity bystimulating TNF-alpha expression via a lysosomal pathway. Hepatology(Baltimore, Md) 2004;40:185–194.

[93] Nolan CJ, Larter CZ. Lipotoxicity: why do saturated fatty acids cause andmonounsaturates protect against it? J Gastroenterol Hepatol 2009;24:703–706.

[94] Erbay E, Babaev VR, Mayers JR, Makowski L, Charles KN, Snitow ME, et al.Reducing endoplasmic reticulum stress through a macrophage lipidchaperone alleviates atherosclerosis. Nat Med 2009;15:1383–1391.

[95] Robertson G, Leclercq I, Farrell GC. Nonalcoholic steatosis and steatohep-atitis. II. Cytochrome P-450 enzymes and oxidative stress. Am J PhysiolGastrointest Liver Physiol 2001;281:G1135–1139.

[96] Kohjima M, Enjoji M, Higuchi N, Kato M, Kotoh K, Yoshimoto T, et al. Re-evaluation of fatty acid metabolism-related gene expression in nonalco-holic fatty liver disease. Int J Mol Med 2007;20:351–358.

[97] Dentin R, Benhamed F, Hainault I, Fauveau V, Foufelle F, Dyck JR, et al.Liver-specific inhibition of ChREBP improves hepatic steatosis and insulinresistance in ob/ob mice. Diabetes 2006;55:2159–2170.

[98] Tamura S, Shimomura I. Contribution of adipose tissue and de novolipogenesis to nonalcoholic fatty liver disease. J Clin Invest 2005;115:1139–1142.

[99] Orellana-Gavalda JM, Herrero L, Malandrino MI, Paneda A, Sol Rodriguez-Pena M, Petry H, et al. Molecular therapy for obesity and diabetes based ona long-term increase in hepatic fatty-acid oxidation. Hepatology (Balti-more, Md) 2011;53:821–832.

[100] Rosselli MS, Burgueno AL, Pirola CJ, Sookoian S. Cyclooxygenase inhibitionUp-regulates liver carnitine palmitoyltransferase 1A expression andimproves fatty liver. Hepatology (Baltimore, Md) 2011;53:2143–2144.

[101] Crespo J, Cayon A, Fernandez-Gil P, Hernandez-Guerra M, Mayorga M,Dominguez-Diez A, et al. Gene expression of tumor necrosis factor alphaand TNF-receptors, p55 and p75, in nonalcoholic steatohepatitis patients.Hepatology (Baltimore, Md) 2001;34:1158–1163.

[102] Endo M, Masaki T, Seike M, Yoshimatsu H. TNF-alpha induces hepaticsteatosis in mice by enhancing gene expression of sterol regulatoryelement binding protein-1c (SREBP-1c). Exp Biol Med (Maywood)2007;232:614–621.

[103] Kaser S, Moschen A, Cayon A, Kaser A, Crespo J, Pons-Romero F, et al.Adiponectin and its receptors in non-alcoholic steatohepatitis. Gut2005;54:117–121.

[104] Feldstein AE, Canbay A, Angulo P, Taniai M, Burgart LJ, Lindor KD, et al.Hepatocyte apoptosis and fas expression are prominent features of humannonalcoholic steatohepatitis. Gastroenterology 2003;125:437–443.

[105] Feldstein AE, Wieckowska A, Lopez AR, Liu YC, Zein NN, McCullough AJ.Cytokeratin-18 fragment levels as noninvasive biomarkers for nonalcoholicsteatohepatitis: a multicenter validation study. Hepatology (Baltimore, Md)2009;50:1072–1078.

[106] Horoz M, Bolukbas C, Bolukbas FF, Sabuncu T, Aslan M, Sarifakiogullari S,et al. Measurement of the total antioxidant response using a novelautomated method in subjects with nonalcoholic steatohepatitis. BMCgastroenterology 2005;5:35.

[107] Huang W, Ma K, Zhang J, Qatanani M, Cuvillier J, Liu J, et al. Nuclearreceptor-dependent bile acid signaling is required for normal liverregeneration. Science 2006;312:233–236.

[108] Fernandez MA, Albor C, Ingelmo-Torres M, Nixon SJ, Ferguson C, KurzchaliaT, et al. Caveolin-1 is essential for liver regeneration. Science2006;313:1628–1632.

[109] Doege H, Baillie RA, Ortegon AM, Tsang B, Wu Q, Punreddy S, et al. Targeteddeletion of FATP5 reveals multiple functions in liver metabolism: altera-tions in hepatic lipid homeostasis. Gastroenterology 2006;130:1245–1258.

964 Journal of Hepatology 201

[110] Hubbard B, Doege H, Punreddy S, Wu H, Huang X, Kaushik VK, et al. Micedeleted for fatty acid transport protein 5 have defective bile acidconjugation and are protected from obesity. Gastroenterology 2006;130:1259–1269.

[111] Geier A, Trautwein C. Bile acids are ‘‘homeotrophic’’ sensors of thefunctional hepatic capacity and regulate adaptive growth during liverregeneration. Hepatology 2007;45:251–253.

[112] Bechmann LP, Jochum C, Kocabayoglu P, Sowa JP, Kassalik M, Gieseler RK,et al. Cytokeratin 18-based modification of the MELD score improvesprediction of spontaneous survival after acute liver injury. J. Hepatol.2010;53:639–647.

[113] Shu RZ, Zhang F, Wang F, Feng DC, Li XH, Ren WH, et al. Adiponectindeficiency impairs liver regeneration through attenuating STAT3 phos-phorylation in mice. Lab Invest 2009;89:1043–1052.

[114] Wree A, Kahraman A, Gerken G, Canbay A. Obesity affects the liver - thelink between adipocytes and hepatocytes. Digestion 2011;83:124–133.

[115] Hirsch HA, Iliopoulos D, Joshi A, Zhang Y, Jaeger SA, Bulyk M, et al. Atranscriptional signature and common gene networks link cancer withlipid metabolism and diverse human diseases. Cancer Cell 2010;17:348–361.

[116] Petta S, Craxi A. Hepatocellular carcinoma and non-alcoholic fatty liverdisease: from a clinical to a molecular association. Curr Pharm Des2010;16:741–752.

[117] Bugianesi E. Non-alcoholic steatohepatitis and cancer. Clin Liver Dis2007;11:191–207 (x-xi).

[118] Calle EE, Rodriguez C, Walker-Thurmond K, Thun MJ. Overweight, obesityand mortality from cancer in a prospectively studied cohort of U.S. adults.N Engl J Med 2003;348:1625–1638.

[119] Davila JA, Morgan RO, Shaib Y, McGlynn KA, El-Serag HB. Diabetesincreases the risk of hepatocellular carcinoma in the United States: apopulation based case control study. Gut 2005;54:533–539.

[120] Adams LA, Lymp JF, St Sauver J, Sanderson SO, Lindor KD, Feldstein A, et al.The natural history of nonalcoholic fatty liver disease: a population-basedcohort study. Gastroenterology 2005;129:113–121.

[121] Ertle J, Dechene A, Sowa JP, Penndorf V, Herzer K, Kaiser G, et al.Nonalcoholic fatty liver disease progresses to HCC in the absence ofapparent cirrhosis. Int J Cancer 2010.

[122] Venook AP, Papandreou C, Furuse J, de Guevara LL. The incidence andepidemiology of hepatocellular carcinoma: a global and regional perspec-tive. Oncologist 2010;15:5–13.

[123] Wu JM, Skill NJ, Maluccio MA. Evidence of aberrant lipid metabolismin hepatitis C and hepatocellular carcinoma. HPB (Oxford) 2010;12:625–636.

[124] Villanueva A, Chiang DY, Newell P, Peix J, Thung S, Alsinet C, et al. Pivotalrole of mTOR signaling in hepatocellular carcinoma. Gastroenterology2008;135:1972–1983 (1983 e1971-1911).

[125] Qu X, Yu J, Bhagat G, Furuya N, Hibshoosh H, Troxel A, et al. Promotion oftumorigenesis by heterozygous disruption of the beclin 1 autophagy gene. JClin Invest 2003;112:1809–1820.

[126] Park EJ, Lee JH, Yu GY, He G, Ali SR, Holzer RG, et al. Dietary and geneticobesity promote liver inflammation and tumorigenesis by enhancing IL-6and TNF expression. Cell 2010;140:197–208.

[127] Handy JA, Saxena NK, Fu P, Lin S, Mells JE, Gupta NA, et al. Adiponectinactivation of AMPK disrupts leptin-mediated hepatic fibrosis via suppres-sors of cytokine signaling (SOCS-3). J Cell Biochem 2010;110:1195–1207.

[128] Tanaka N, Moriya K, Kiyosawa K, Koike K, Gonzalez FJ, Aoyama T.PPARalpha activation is essential for HCV core protein-induced hepaticsteatosis and hepatocellular carcinoma in mice. J Clin Invest2008;118:683–694.

[129] van der Meer DL, Degenhardt T, Vaisanen S, de Groot PJ, Heinaniemi M, deVries SC, et al. Profiling of promoter occupancy by PPARalpha in humanhepatoma cells via ChIP-chip analysis. Nucleic Acids Res 2010;38:2839–2850.

[130] Ratziu V, Giral P, Jacqueminet S, Charlotte F, Hartemann-Heurtier A,Serfaty L, et al. Rosiglitazone for nonalcoholic steatohepatitis: one-yearresults of the randomized placebo-controlled Fatty Liver Improvementwith Rosiglitazone Therapy (FLIRT) Trial. Gastroenterology 2008;135:100–110.

[131] Yu J, Qiao L, Zimmermann L, Ebert MP, Zhang H, Lin W, et al. Troglitazoneinhibits tumor growth in hepatocellular carcinoma in vitro and in vivo.Hepatology (Baltimore, Md) 2006;43:134–143.

2 vol. 56 j 952–964